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 |
|---|---|---|---|---|---|
Enterococcus faecalis BM4518 is resistant to vancomycin by synthesis of peptidoglycan precursors ending in D-alanyl-D-serine . In the chromosomal vanG locus , transcription of the resistance genes from the PYG resistance promoter is inducible and , upstream from these genes , there is an unusual three-component regulatory system encoded by the vanURSG operon from the PUG regulatory promoter . In contrast to the other van operons in enterococci , the vanG operon possesses the additional vanUG gene which encodes a transcriptional regulator whose role remains unknown . We show by DNase I footprinting , RT-qPCR , and reporter proteins activities that VanUG , but not VanRG , binds to PUG and negatively autoregulates the vanURSG operon and that it also represses PYG where it overlaps with VanRG for binding . In clinical isolate BM4518 , the transcription level of the resistance genes was dependent on vancomycin concentration whereas , in a ΔvanUG mutant , resistance was expressed at a maximum level even at low concentrations of the inducer . The binding competition between VanUG and VanRG on the PYG resistance promoter allowed rheostatic activation of the resistance operon depending likely on the level of VanRG phosphorylation by the VanSG sensor . In addition , there was cross-talk between VanSG and VanR'G , a VanRG homolog , encoded elsewhere in the chromosome indicating a sophisticated and subtle regulation of vancomycin resistance expression by a complex two-component system .
Vancomycin-resistant enterococci are a major cause of nosocomial infections and an important public health problem because the treatment options for the infections they cause are very limited [1] . Vancomycin , which can be the only antibiotic effective against multiresistant clinical isolates , acts by binding to the C-terminal D-alanyl-D-alanine ( D-Ala-D-Ala ) residues of peptidoglycan precursors blocking the extracellular steps in peptidoglycan synthesis [2] . Resistance in Enterococcus is mediated by nine types of operons that produce modified peptidoglycan precursors ending in D-Ala-D-Lac ( vanA , -B , -D , and-M ) or D-Ala-D-Ser ( vanC , -E , -G , -L , and-N ) to which vancomycin bind with a low affinity and from the elimination of the high affinity precursors ending in D-Ala-D-Ala [3–6] . Expression of the vancomycin resistance operons is regulated by VanS/VanR-type two-component signal transduction systems composed of a membrane-bound histidine kinase ( VanS-type ) and a cytoplasmic response regulator ( VanR-type ) that acts as a transcriptional activator [3] . The sensors modulate the levels of phosphorylation of the regulators . In the presence of vancomycin , VanS acts primarily as a kinase that autophosphorylates and transfers its phosphate to VanR . Phosphorylated VanR binds to the promoters upstream from the vanRS regulatory and resistance operons leading to increased transcription of the regulatory and resistance genes [7–9] . The phosphatase activity of VanS-type sensors is required for negative regulation of the resistance genes in the absence of vancomycin preventing accumulation of VanR-type regulators phosphorylated by acetylphosphate or by kinases encoded by the host chromosome [7 , 10] . VanG-type Enterococcusfaecalis clinical isolates from Australia and Canada are distinct from other Van-type enterococci . The chromosomal vanG cluster ( Fig 1 ) confers resistance to vancomycin ( MICs , 16 μg/ml ) by inducible synthesis of precursors ending in D-Ala-D-Ser [11] . It contains the vanYG , WG , G , XYG , TG resistance genes , the last three strictly required for resistance encode , respectively , a VanG ligase to synthesize D-Ala-D-Ser , a VanXYG D , D-carboxypeptidase to hydrolyse D-Ala-D-Ala , and a VanTG membrane bound serine racemase to produce D-Ser ( Fig 1 ) . As opposed to the other van gene clusters , the vanG regulatory operon contains three genes , vanUG , vanRG , and vanSG , encoding a "three component" regulatory system ( Fig 1 ) . Additional gene vanUG encodes a transcriptional regulator belonging to the Xre protein family and of unknown function . The vanURSG genes are co-transcribed , even in the absence of vancomycin , from the PUG regulatory promoter , whereas transcription of the resistance genes is inducible and initiated from the PYG resistance promoter [11] . Cryptic vanG-like operons are common in Clostridium difficile , a major human pathogen which is a target for vancomycin , and a vanUG gene encoding a protein identical to VanUG was found in a clinical isolate ( GenBank N° AVLW01000050 ) . A VanUG-like protein ( GenBank N° YP002939420 ) , 79% identical with VanUG , was detected in an Eubacterium associated with a two-component system controlling an ABC-type transporter and a protein ( GenBank N°YP007781704 ) with 76% identity was reported in Ruminococcus bromii associated with a CheY related regulator and a partial vanG operon . These regulators have not been studied . We report the role of VanUG in the transcription of the vanG operon in E . faecalis . We show that VanUG binds to the PUG regulatory and PYG resistance promoters and negatively regulates the vanURSG regulatory and resistance operons . In contrast , VanRG binds only to PYG . It thus appears that , upon induction by vancomycin , the VanSG sensor phosphorylates VanRG which competes and displaces VanUG from PYG leading to transcription of the resistance operon in a dose dependent manner . Thus , rheostatic regulation of resistance gene expression results from binding of a repressor and an activator encoded in a single operon to the same promoter .
Primer extension of the region upstream from vanUG indicated that , irrespective of induction , the transcriptional start site for vanURSG was located 22 bp upstream from the translation initiation codon of vanUG [11] . The PUG promoter consists of -35 and -10 regions corresponding to δ70 recognition sequences separated by 17 bp ( Fig 2A ) . To determine if VanUG and VanRG bind to the PUG regulatory promoter region and to identify putative specific binding sites , DNaseI footprinting experiments were carried out . A radiolabeled PCR probe corresponding to positions -247 to +110 relative to the transcription initiation site of PUG was incubated with increasing amounts of purified VanUG , VanRG , and VanRG phosphorylated ( VanRG-P ) by acetyl phosphate . The PUG region protected by VanUG depended on the protein concentration , extending from -70 to -20 ( positions relative to the transcription initiation site ) overlapping the -35 sequence at a low concentration ( Fig 2B , lane 6 ) and from -70 to +10 at higher concentrations ( Fig 2B , lanes 7 and 8 ) . The region ( -70 to -20 ) contained two adjacent imperfect palindromic sequences likely corresponding to the binding motifs of VanUG ( Fig 2A ) . As opposed to the wild-type fragment , two DNA fragments containing double mutations in the imperfect dyad symmetry operator of PUG were not retarded by VanUG , indicating a key role in VanUG binding ( S1 Fig ) . The appearance of several DNase I hypersensitive sites ( Fig 2B ) corresponding to bending of the DNA duplex suggested binding of two VanUG monomers or dimers . This is consistent with the presence of two inverted repeats in the PUG region ( Fig 2A ) and with the two-step gel retardation ( S1 Fig ) . In contrast to VanUG , VanRG and VanRG-P did not bind to the PUG promoter . The vanG operon is part of a large genetic element and is transferable from E . faecalis BM4518 to E . faecalis JH2-2 from chromosome to chromosome [11] . Since clinical isolate BM4518 is not transformable , we studied the VanURSG system in transconjugant BM4522 ( JH2-2::vanG ) ( S1 Table ) . To determine the role of VanUG on PUG , the vanUG , vanRG , and vanSG genes of BM4522 were inactivated individually by in-frame deletions leading to BM4720 ( ΔvanUG ) , BM4721 ( ΔvanRG ) , and BM4722 ( ΔvanSG ) . Transcription of the regulatory genes was quantified by RT-qPCR . In BM4522 , low level transcription occured at similar levels without and with various concentrations of vancomycin indicating that the PUG promoter was not inducible by vancomycin ( Fig 2C ) . In the absence of vanUG , vanRG and vanSG were transcribed in the absence or presence of vancomycin at higher level ( ≈ 5-fold ) from PUG indicating that VanUG acted as a repressor on this promoter region ( Fig 2D ) . In the absence of vanRG or vanSG , transcription of the regulatory genes remained unchanged even in the presence of vancomycin . To confirm regulation of PUG by VanUG , the vanURSG genes were cloned into vancomycin susceptible Escherichia coli NR698 [12] under the control of promoter Pspank upstream from PUG fused to a chloramphenicol acetyltransferase ( CAT ) reporter gene , the two promoters being separated by a transcription terminator ( term ) ( Table 1 ) . Subsequently , each of the three genes was inactivated . E . coli RNA polymerase bound to the PUG promoter ( S2A Fig ) which was active in the new host , in the presence or in the absence of vancomycin ( Table 1 ) . CAT was produced at a maximum level in the absence of vanUG by plasmids pAT952 ( PspanktermPUGcat ) , pAT966 ( PspankvanRGtermPUGcat ) , and pAT969 ( PspankvanRSGtermPUGcat ) ( Table 1 ) . In contrast , in the presence of VanUG , CAT production was decreased to similar basal levels by plasmids pAT965 ( PspankvanUGtermPUGcat ) , pAT967 ( PspankvanURGtermPUGcat ) , and pAT968 ( PspankvanURSGtermPUGcat ) ( Table 1 ) . These results confirmed that VanUG acts as a strong repressor on the PUG promoter . Transcription of the resistance genes is under the control of VanURSG and , as discussed above , VanUG negatively autoregulates vanURSG transcription from the PUG regulatory promoter . To determine if VanRG and VanSG acted as a two-component system and to study the putative interaction of VanUG with these proteins , VanUG , VanRG , and the cytoplasmic histidine kinase domain of VanSG were purified as C-terminal His-tag proteins ( S1 Table ) . VanSG autophosphorylated in the presence of [γ-32P]-ATP ( Fig 3A ) . When incubated with purified VanUG or VanRG , phosphorylated VanSG transferred its phosphate group to VanRG ( Fig 3B ) but not to VanUG ( Fig 3E ) . Phosphorylation of VanRG was fast and efficient , occurring in less than a minute . To test the phosphatase activity of VanSG , hydrolysis of VanRG-P over time was analysed in the absence or in the presence of VanSG . Purified [32P]-VanRG was stable in vitro for at least 30min and then dephosphorylated slowly ( Fig 3C ) ; addition of purified VanSG increased dephosphorylation only slightly ( Fig 3D–3G ) . These results indicate that VanRSG was functional and had characteristics similar to those of other VanRS-type two-component systems [7 , 9] and that VanUG did not affect phosphorylation nor dephosphorylation of VanRG and VanSG ( Fig 3E and 3F ) . To study the putative binding of VanUG and VanRG to the PYG region and to identify specific binding sites , DNaseI footprinting experiments were carried out . The inducible PYG promoter is composed of -35 ( AAAACA ) and -10 ( TACAAT ) regions separated by 16 bp which have similarity with δ70 recognition sequences , although the -35 sequence is not conserved consistent with the fact that the promoter is positively regulated ( Fig 4B ) . Analysis of the PYG region revealed three 12-bp directly repeated VanRG binding motifs and a deduced consensus sequence ( T/C ) CGTANGAAA ( T/A ) T was analogous to that in the PR and PH vanA operon promoters [13] . In the PUG region , similar sequences were not found ( Fig 2A ) which could explain lack of VanRG binding . The radiolabeled probe corresponding to positions -163 to +69 relative to the transcription initiation point of the PYG promoter and containing the three conserved sequences was incubated with increasing amounts of purified VanUG , VanRG , and VanRG-P ( Fig 4 ) . The three proteins protected in a concentration-dependent manner an overlapping DNA region that included the three direct repeats . The PYG region protected by VanUG was much larger than that by VanRG and VanRG-P extending from -110 to -3 and overlapped the -35 sequence at 0 . 2 and 1μM ( Fig 4A , lanes 17 and 18 ) . The PYG region protected by VanRG and VanRG-P extended from -100 to -56 at low concentration ( Fig 4A , bracket I , lanes 3 and 8 ) and from -100 to -43 at higher concentrations ( Fig 4A , bracket II , lanes 4 and 5 , and 9 and 10 ) . There were three binding motifs a , b , and c with different affinities for VanRG and VanRG-P in the PYG promoter region ( Fig 4 ) . Only a slight difference in affinity in favor of VanRG-P at 0 . 2μM was noted for the "a" site ( Fig 4A , lane 2 ) compared with VanRG which could be due to inefficient phosphorylation of VanRG by acetylphosphate . VanRG and VanRG-P bound to the a and b sites ( Fig 4A , lanes 2 , 3 , and 8 ) with higher affinity than to the c site ( Fig 4A , lanes 4 and 5 , and 9 and 10 ) , whereas VanUG bound to this DNA region with the same affinity ( Fig 4A ) . To study the consequences of the binding of VanUG and VanRG to overlapping regions of PYG on the expression of the resistance genes , the VanTG serine racemase was used as a reporter ( Fig 5 ) . In clinical isolate BM4518 and transconjugant BM4522 , synthesis of the serine racemase was dependent on the concentration of vancomycin ( Fig 5 ) . In contrast , in BM4720 ( ΔvanUG ) , the resistance operon was expressed at its maximum even at low concentrations of vancomycin . These results suggested that VanUG acts as a repressor of PYG and that , in its absence , there is no fine-tuning of resistance expression from this promoter . Thus , modulation of transcription by vancomycin was due to the phosphorylation level of VanRG mediated by VanSG provided that VanUG was present . Surprisingly , as in the wild-type strain , induction was dependent on the concentration of the inducer in BM4721 ( ΔvanRG ) ( Fig 5 ) . This could be accounted for by the presence of a VanR homolog in the host . In fact , we found , in both E . faecalis BM4518 and transconjugant BM4522 which were entirely sequenced ( GenBank N°PRJNA245745 ) , a gene specifying a VanR'G protein with 65% identity with VanRG ( S3A Fig ) . In BM4722 ( ΔvanSG ) there was no synthesis of VanTG in the presence of vancomycin indicating that VanRG and VanR'G are not phosphorylated in the absence of VanSG . Double mutant BM4723 ( ΔvanRG , ΔvanR'G ) derived from E . faecalis BM4721 ( ΔvanRG ) was susceptible to vancomycin ( MIC , 1μg/ml ) and VanTG production was no longer inducible by vancomycin , indicating cross-talk between VanSG and VanR'G ( Fig 5 ) . To avoid interference by this regulator , transcription from the PYG promoter was studied in E . coli NR698 since E . coli RNA polymerase was able to bind to this promoter ( S2B Fig ) . The vanURSG , vanRSG , and vanUSG genes were cloned under the control of Pspank upstream from the PYG transcriptionally fused to a cat gene generating pAT970 ( PspankvanURSGtermPYGcat ) , pAT971 ( PspankvanRSGtermPYGcat ) , and pAT972 ( PspankvanUSGtermPYGcat ) . In the absence of VanUG , induction by vancomycin led to similar levels of CAT synthesis in the strain harboring pAT971 ( PspankvanRSGtermPYGcat ) whatever the concentration of the inducer , whereas with pAT970 ( PspankvanURSGtermPYGcat ) CAT production depended on the vancomycin concentration ( Table 2 ) . These results confirmed that , in the presence of vancomycin , VanUG is required for rheostatic gene transcription from PYG and that VanRG phosphorylation is essential for expression of the resistance genes since , in the absence of this regulator in pAT972 ( PspankvanUSGtermPYGcat ) , the level of CAT activity was low , both without ( 74U±9 ) and with ( 104 U ± 13 ) vancomycin ( 0 . 30 μg/ml ) . In the absence of vancomycin , CAT activity was lower in E . coli producing vanUG encoded by pAT970 ( PspankvanURSGtermPYGcat ) than in its counterpart harboring pAT971 ( PspankvanRSGtermPYGcat ) . This confirms that VanUG acts as a repressor on the PYG resistance promoter ( Table 2 ) . Since VanUG and VanRG bound at overlapping sites of PYG , to assess a possible effect of VanRG on the binding of VanUG , we performed DNaseI footprinting assays on the labeled PYG probe with purified VanRG and VanUG ( Fig 6 ) . Low and medium concentrations ( 64 nM and 128 nM ) of VanUG which allow binding to PYG were tested with increasing concentrations of VanRG . Upon addition of VanRG , the binding profile of VanUG faded while that of VanRG appeared and increased in a dose dependent manner ( Fig 6A ) . In the reverse experiment two approriate concentrations of VanRG were challenged by increasing concentrations of VanUG and the binding of VanRG decreased also in the presence of VanUG ( S4 Fig ) . In summary , VanUG alone did not allow transcription of the resistance genes ( Fig 6B ) . It thus appears that at a low concentration of vancomycin there was competition between VanUG and VanRG , the latter being partially phosphorylated , transcription of vanYGWGGXYGTG was low . In contrast , at high concentrations of vancomycin , VanRG was efficiently phosphorylated and able to displace VanUG leading to maximal transcription of the resistance genes from the PYG promoter . To study the role of VanUG in this sophisticated resistance mechanism , the fitness cost of BM4720 ( ΔvanUG ) compared with that of BM4522 in monocultures in the absence and in the presence of vancomycin ( 1 μg/ml ) was analysed by determination of the growth rates ( Table 3 ) . The results showed that the growth rates of both strains were indistinguishable in the absence of vancomycin indicating that non induced VanG-type resistance is not costly for the host . In contrast , in the presence of vancomycin , the relative growth rate of BM4720 ( ΔvanUG ) ( 0 . 74 ) was significantly reduced when compared with that of BM4522 ( 0 . 93 ) indicating that increased expression of resistance was significantly more costly in the absence of vanUG .
Among the ubiquitous two-component regulators , VanR/VanS-type systems are one of the rare to control expression of genes mediating antibiotic resistance [3] . In the VanG-type strains , a membrane associated sensor kinase ( VanSG ) which detects a signal associated with the presence of vancomycin in the environment and a cytoplasmic response regulator ( VanRG ) that acts as a transcriptional activator are also present ( Fig 1 ) and functional ( Fig 3 ) but there is , in addition , a VanUG transcriptional regulator ( Fig 1 ) . In the two main VanA- and VanB-type systems , the regulatory genes ( vanRS ) and the resistance genes are transcribed from independent and coordinately regulated promoters , but VanR is the only known direct regulator of the resistance genes [3 , 8 , 13] . In VanG-type strains , co-transcription of vanURSG is repressed from PUG by VanUG ( Fig 2 and Table 1 ) and expression of the resistance genes from PYG is activated by VanRG and repressed by VanUG ( Fig 5 and Table 2 ) . Thus , VanUG regulates the resistance genes both directly , by binding to the PYG promoter region ( Fig 4 ) , and indirectly by repressing synthesis of VanRGSG ( Fig 5 ) . Like other members of the XRE protein family ( S3B Fig ) [14–16] , VanUG binds to short repeated sequences which span the promoters ( Fig 2A and 2B ) . Unlike the VanR and VanRB proteins which bind to their own promoters [8 , 13] , VanRG does not regulate its own expression ( Fig 2 ) . No sequences similar to the VanRG consensus binding site are found in PUG ( Figs 2 and 4 ) . VanRG , as VanR and VanRB , belongs to the OmpR-PhoB subclass of response regulators that have the peculiarity to bind to their target promoters in the unphosphorylated or phosphorylated form [8 , 13 , 17 , 18] . Phosphorylation of VanR and VanRB enhances the affinity of the proteins for their respective regulatory PR or PRB and resistance PH or PYB promoter regions allowing increased transcription of the regulatory and resistance genes [8 , 13] . In VanA-type strains , VanR and VanR-P bind to PR and PH regions which contain a single or two 12-bp conserved sites , respectively [13] . Comparison of the sequences of the PUG and PYG regions with the 12-bp consensus sequence spanned by VanR and VanR-P revealed three binding sites in the PYG region with a consensus sequence ( Fig 4B ) similar to that in VanA-type resistance [13] . As for the regulatory PR and resistance PH promoters , the positioning of these sites in PYG was upstream from the -35 motif . VanUG , VanRG , and VanRG-P protected overlapping regions , the two latter binding to PYG a and b sites with a higher affinity than to the c site ( Fig 4 ) . There are only two sites in the PH promoter but VanR generated a more extensive footprint ( 80 bp for PH ) than VanRG ( 42bp for PYG ) likely due to higher cooperativity of VanR . Although not essential for binding in vitro , phosphorylation of VanRG increased its affinity for the PYG resistance promoter ( Fig 4 ) . In the PUG promoter region no sequences similar to the consensus were found ( Fig 2A ) which could explain the absence of binding of VanRG and low-level transcription from the regulatory promoter . In many instances , regulation of gene transcription in E . coli occurs essentially through control of the phosphatase activity of the sensor [19 , 20] . In VanA- and VanB-type strains , the level of phosphorylation of VanR and VanRB is modulated by the kinase and phosphatase activities of the VanS and VanSB sensors [7 , 10 , 21] . Phosphatase activity is critical for response regulators , such as VanR and VanRB , whose phosphorylated form is highly stable , to ensure that the protein is not permanently activated . In VanG-type strains , in the absence of VanUG , induction by vancomycin led to maximal VanTG serine racemase ( Fig 5 ) or CAT synthesis ( Table 2 ) even at low concentrations of the inducer . Since in the absence of VanUG there was no modulation of resistance genes transcription from the PYG promoter , this suggests that a low amount of VanRG-P is sufficient to induce the resistance operon . VanUG did not modulate VanRG and VanSG phosphorylation ( Fig 4F ) and was not phosphorylated by VanSG ( Fig 4E ) . Surprisingly , at least in vitro , the phosphatase activity of VanSG was not very efficient ( Fig 4D ) in comparison with those of VanS or VanSB [7 , 9] . Expression of VanG-type resistance was thus inducible by vancomycin due to the presence of VanUG as opposed to direct modulation of VanR activity by VanS in the other van operons . In the absence of vancomycin only VanUG bound to the PYG promoter; however when the concentration of vancomycin increased , VanRG being more efficiently phosphorylated by VanSG , displaced progressively VanUG allowing gradual transcription of the resistance genes ( Fig 6 ) as it is likely the case with VanR'G , the VanRG homolog encoded elsewhere in the chromosome . In B . subtilis , when both repressors SinR and SlrR are bound to the degU promoter , they can be displaced by the response regulator DegU leading to activation of the degU gene [22] . Also in B . subtilis , CcpC activates aconitase gene citB expression whereas CodY binds to its promoter and represses citB transcription [23]; PutR which is an activator essential for transcription of the putBCP operon for proline utilization is displaced by the CodY repressor [24] . VanUG does not possess the characteristics of auxiliary regulators which can interact with histidine kinases , influencing signal perception and transduction . Nor does it interact with the response regulator to alter its phosphorylation status or its DNA binding ability , the recruitement of RNA polymerase on the promoter , or to sequester it through protein:protein interaction [25 , 26] . The results presented here show that competition between the VanUG repressor and the VanRG activator for binding to the PYG promoter may be responsible for the complex regulation of the resistance genes ( Fig 6 ) . This is an unusual example of rheostatic regulation of gene transcription due to binding competition between two regulators encoded in the same operon . It also elucidates an unsuspected strategy by which enterococcal clinical isolates regulate transcription of acquired genes for vancomycin resistance . In previous work , we showed in VanB-type resistance that , despite the complex dual biochemical mechanism of resistance to vancomycin , its biological cost in enterococci is negligible when non induced , whereas a significant fitness reduction is observed when resistance is expressed in the presence of the inducer , the antibiotic itself [27] . Thus resistance is expressed exclusively when needed for bacterial survival . In VanG-type strains , tight regulation of resistance expression involves VanUG which can thus be considered as a compensatory component , drastically reducing the biological cost associated with vancomycin resistance in the presence of antibiotic .
The origin and properties of the strains and plasmids are described in S1 Table . Escherichia coli TOP10 ( Invitrogen , Groningen , The Netherlands ) and NR698 ( susceptible to vancomycin ) [12] were used as a host for recombinant plasmids . E . coli BL21λDE3 [28] , in which the T7 RNA polymerase gene is under the control of the inducible lacUV5 promoter carries the pREP4 plasmid allowing co-expression of the GroESL chaperonin to optimize recombinant protein solubility [29] . E . coli TG1 RepA [30] was used as a host for constructions in the pAT944 ( pGhost9Ωcat ) vector ( S1 Table ) . Kanamycin ( 50μg/mL ) was used as a selective agent for cloning PCR products in the pCR-Blunt vector ( Invitrogen ) . Ampicillin was used to select pUC1813 [31] . pDR111 ( gift from David Rudner , Harvard University ) , which harbors the Pspank promoter between two fragments of the B . subtilis amyE gene , is a derivative of the Pspac-hy plasmid pJQ43 containing an additional lacO binding site to achieve a better repression in the absence of the IPTG inducer . Pspank is a lacI repressible IPTG inducible-promoter for gene overexpression . Spectinomycin ( 60μg/mL ) and chloramphenicol ( 10μg/mL ) were added to the medium to prevent loss of plasmids derived from pDR111 ( Pspank ) and pAT944 ( pGhost9Ωcat ) , respectively . Enterococcus faecalis JH2-2 is a derivative of strain JH2 that is resistant to fusidic acid and rifampin [32] . In all experiments , strains were grown in brain heart infusion ( BHI ) at 37°C with shaking at 110 rpm . Labeled PUG ( 357 bp ) and PYG ( 233 bp ) fragments were generated by PCR with BM4518 total DNA as a template and primer pairs VanG12-VanG126 and VanSG6-YG10 ( S2 Table ) , respectively , using a combination of an unlabeled primer with an end-labeled primer ( 625nM ) with T4 polynucleotide kinase ( 0 . 075 U/μl ) ( New England Biolabs ) and [γ32P]-ATP ( 3000 Ci/mmol ) ( Perkin Elmer ) . The PCR reactions were carried out in a 50-μl volume and the products purified as described [8] . Purified labeled PCR products corresponding to wild-type and mutated PUG promoter region fragments were recovered from a 6% polyacrylamide gel and used as a probe for the gel shift assay after addition of 100 μl of ammonium acetate ( 0 . 5 M ) diluted in Tris buffer ( 10 mM , pH8 . 5 ) overnight at 37°C . The PUG and mutated PUG probes ( 10 , 000cpm each ) were incubated with various concentrations of purified VanUG regulator at 30°C for 20min in 20 μl of 50mM Tris-HCl ( pH7 . 8 ) containing 20 mM MgCl2 and 0 . 1 mM dithiothreitol ( DTT ) . After addition of the DNA dye solution ( 40% glycerol , 0 . 025% bromophenol blue and 0 . 025 xylene cyanol ) , the mixture was loaded on a 7 . 5% polyacrylamide gel in the absence of protein denaturants . The gels were dried and analysed by autoradiography . Complexes with the labeled promoter regions ( 5nM ) were formed for 30 min at 30°C in 15 μl of buffer C ( 20 mM Hepes pH 8 . 0 , 5 mM MgCl2 , 50 mM potassium glutamate , 5 mM DTT , and 500μg/ml bovine serum albumin ) using RNA polymerase of E . coli at 50 nM or VanUG , VanRG , or VanRG-P at increasing concentrations . For DNase I experiments , 1 . 5 μl of DNase I solution ( 1 μg ml-1 in 10 mM Tris-HCl , 10 mM MgCl2 , 10 mM CaCl2 , 125 mM KCl ) were added and incubated at 30°C for 10s when the labeled promoter regions were alone , or for 20 s when when RNA polymerase or VanUG , VanRG or VanRG-P were present in the mixture . The reaction was stopped and all the samples were extracted , precipitated , washed , resuspended , and loaded on a sequencing gel as described [8] . Protected bands were identified by comparing the migration with that of the same fragment treated for the A+G sequencing reaction [33] . The gels were analysed by autoradiography . Enterococci grown in 100 ml of brain heart infusion in 250-ml bottles , with and without vancomycin , at 37°C with shaking at 110 rpm to OD600 = 0 . 8 were harvested . RNA was prepared using the Fast RNA ProBlue kit ( MBP Biomedicals ) according to the manufacturer's protocol , treated with DNase ( Turbo DNA-free , Invitrogen ) , and checked for the absence of contaminant DNA in a standard PCR , using the same primers as for the RT-PCR . RNA concentrations were determined by measuring absorbance with a NanoDrop2000 ( ThermoScientific ) . cDNA synthesis and RT-qPCR were performed with a Light Cycler RNA amplification kit SYBR greenI ( Roche Diagnostic GmbH ) in a total reaction volume of 19μl with 0 . 5 μM gene-specific primers ( VanG129-VanG102 for vanUG , VanRG2-VanRG10 for vanRG , VanSG2-VanSG10 for vanSG , and rpoB5-rpoB12 for rpoB ) ( S2 Table ) according to the manufacturer's instructions . Amplification and detection of specific products were performed using the LightCycler sequence detection system ( Roche ) with the following cycle profile: 1cycle at 55°C for 20 min for the reverse transcription step , followed by 1 cycle at 95°C for 30 s , 45 cycles at 95°C for 5 s , 52°C for 15 s , and 72°C for 15 s . The level of every gene transcript was normalized relative to rpoB transcript levels . Plasmids pAT940 ( pET28ΩvanUG ) , pAT941 ( pET28ΩvanRG ) , and pAT942 ( pET28ΩvanSG ) ( S1 Table ) were introduced into E . coli BL21λDE3/pREP4 [29] . The transformants were grown in 1 liter of LB medium in Fernbach flasks with shaking at 110 rpm at 28°C until OD600 = 0 . 8 , IPTG ( 1 mM ) was added to induce protein production , and incubation was pursued for 4 h . E . coli crude protein extracts were loaded on 1-ml His-Trap fast-flow columns ( GE , Healthcare ) equilibrated with buffer A ( 50mM NaH2PO4 pH 7 . 5 , 300 mM NaCl , 30 mM imidazole ) and the proteins were eluted with an imidazole gradient ( 30mM-500mM ) . Fractions were dialysed against buffer B ( 50mM NaH2PO4 pH 7 . 5 , 300 mM NaCl , 50% glycerol ) . Protein concentration was determined using the Bio-Rad protein assay [34] . Autophosphorylation of VanSG ( 40 μg ) was performed in a final volume of 100 μl of buffer A ( final concentrations: 50 mM Tris-HCl , 50mM KCl and 1 mM MgCl2 , pH7 . 5 ) . The reaction was initiated by the addition of 5 μl of ATP ( 1mM final ) containing 200 μCi of [γ-32P]ATP and incubated at room temperature for 1 h . ATP was removed using 500 μl Sephadex G-50 spin column equilibrated with buffer A . The reaction was quenched by the addition of 5 μl of β-mercaptoethanol-stop solution ( Sigma ) , followed by electrophoresis on 12% NuPAGE Novex Bis-Tris gels ( Invitrogen ) in MOPS buffer ( 1X ) , and autoradiography . Phosphotransfer to purified VanUG and VanRG were carried out in buffer A . The reaction was initiated by the addition of 10 μl of the purified autophosphorylation reaction mixture of VanSG ( 40 μg ) described above to a 15 μl reaction mixture containing VanUG or VanRG ( 55 μg each ) . After incubation for various periods of times at room temperature , the phosphotransfer reactions were quenched by the addition of stop solution ( Sigma ) followed by electrophoresis on 12% NuPAGE Novex Bis-Tris gels ( Invitrogen ) in MOPS buffer ( 1X ) and autoradiography . VanUG ( 220 μg ) or VanRG ( 225 μg ) were incubated in 100 μl of buffer B ( 50 mM Tris-HCl , pH7 . 8 , 20 mM MgCl2 , 0 . 1 mM dithiothreitol ) containing 178 pmol ( 3 . 3 μCi ) of acetyl[32P]phosphate ( Hartmann Analytical , Germany ) at room temperature for 60 min . Excess acetyl[32P]phosphate was removed using Sephadex G-50 spin columns equilibrated with buffer B . Aliquots ( 10 μl ) were withdrawn at designated time points , and the phosphorylation reactions were quenched with β-mercaptoethanol-stop solution followed by electrophoresis on 15% SDS-polyacrylamide gels and autoradiography . The VanUG ( 220 μg ) and VanRG ( 225 μg ) response regulators were labelled with acetyl[32P]phosphate for 1 h at room temperature as described above , and 52 μg of VanSG histidine kinase were added , and incubation was pursued for various periods of times . Aliquots ( 10 μl ) were withdrawn at designated time points and the reactions were stopped , followed by electrophoresis on 15% SDS-polyacrylamide gels and autoradiography . The plasmids were constructed as follows . Plasmids pDR111 , pAT949 , pAT950 , pAT952 , pAT964 , pAT965 , pAT966 , pAT967 , pAT968 , pAT969 , pAT970 , pAT971 , and pAT972 were introduced by transformation into vancomycin susceptible E . coli NR698 and transformants were selected on agar containing chloramphenicol ( 10 g/ml ) or ampicillin ( 100 μg/ml , for pDR111 ) ( Tables 1 and 2 ) . In Gram-positive bacteria , pGhost9 [36] which replicates at 28°C but is lost above 37°C , allowed construction of E . faecalis BM4522 derivatives by insertional inactivation . Plasmids pAT945 ( pGhost9CmΩΔvanUG ) , pAT946 ( pGhost9CmΩΔvanRG ) , and pAT947 ( pGhost9CmΩΔvanSG ) were electrotransformed into E . faecalis BM4522 [11] to generate , respectively , BM4720 ( ΔvanUG ) , BM4721 ( ΔvanRG ) , and BM4722 ( ΔvanSG ) ( S1 Table ) . Plasmid pAT973 ( pGhost9CmΩΔvanR'G ) was electrotransformed into E . faecalis BM4721 ( ΔvanRG ) to generate the double mutant BM4723 ( ΔvanRG , ΔvanR'G ) . Transformants were selected at the permissive temperature ( 28°C ) on M17 plates containing 10g/ml of chloramphenicol and 0 . 5% glucose . A colony of each transformant was inoculated into 50 ml of M17 broth containing 0 . 5% glucose and incubated for 2h at 28°C . The culture was then shifted to a non-permissive temperature ( 42°C ) for 2 h and integrants , following a first recombination event , were selected at 42°C on M17 agar containing chloramphenicol ( 10g/ml ) . Plasmid excision , by a second recombination event , was favored by subculturing at 28°C in the absence of chloramphenicol and plasmid loss was screened for by plating at 42°C on M17-glucose followed by replica plating on chloramphenicol . The integration locus was determined by PCR following digestion with SmaI and sequencing . For preparation of extracts , 8 ml of an overnight culture were added to 100 ml of broth in the absence or in the presence of vancomycin and strains were grown until OD600 = 0 . 8 in 250 ml bottles with shaking at 110 rpm . The cells were harvested by centrifugation , washed in 0 . 1M phosphate buffer pH 7 . 0 , resuspended in the same buffer , lysed by sonication , followed by centrifugation at 10 , 000 g during 45 min . The resuspended pellet for VanTG racemase [11] and supernatant for CAT activity , were assayed as described [38] . Total DNA from BM4518 and BM4522 strains was purified and sequencing library preparation was carried out using the Nextera DNA Sample Preparation kit ( Illumina , San Diego , CA ) , according to manufacturer’s specifications . Quality and quantity of each sample library was measured on an Agilent Technologies 2100 Bioanalyzer ( Santa Clara , CA ) . Libraries were normalized to 2nM , multiplexed and subjected to 250-bp paired end sequencing ( Illumina MiSeq ) . On average , 5 million high-quality paired-end reads were collected for each strain , representing >220-fold coverage of the ~2 . 9 Mb genomes . Reads were assembled de novo utilizing CLC Genomics Workbench ( CLC bio , Cambridge , MA ) . Functional annotations were performed using a custom pipeline as described previously [39] . Growth rates were determined in microplates coupled to a spectrophotometer iEMS reader ( Labsystems ) . Strains were grown overnight at 37°C without or with 1 μg/ml of vancomycin . The cultures were diluted at OD 0 . 15 into 10 ml of broth without or with vancomycin ( 1μg/ml ) and grown at 37°C with shaking until the beginning of the stationary phase . The cultures were diluted 1/1 , 000 to inoculate 105 bacteria into 200 μl of broth in a 96-well microplate that was incubated overnight at 37°C with shaking . Absorbance was measured at 600 nm every 3 min . Each culture was replicated three times in the same microplate . Growth rates performed in three independent experiments were determined at the beginning of the exponential phase and the relative growth rates were calculated as the ratio of the growth rate of the strain induced by vancomycin versus that of the non induced strain . | Various modes of gene regulation coexist in cells . One corresponds to the “switch on/ off” mechanism in which the regulator induces the promoter to a defined level . In another mechanism , the regulator activates the promoter to various levels according to the intensity or the nature of an input signal . In this study , we show that in VanG-type vancomycin resistant Enterococcus faecalis a repressor ( VanUG ) allows rheostatic expression of a target resistance promoter by competing with a response regulator ( VanRG ) which otherwise acts together with a sensor ( VanSG ) by a "switch on/off" mechanism as part of a two-component regulatory system . Unusually , both regulators are encoded in the same operon . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Competition between VanUG Repressor and VanRG Activator Leads to Rheostatic Control of vanG Vancomycin Resistance Operon Expression |
Although invasive cytomegalovirus ( CMV ) disease is uncommon in the era of antiretroviral therapy ( ART ) , asymptomatic CMV coinfection is nearly ubiquitous in HIV infected individuals . While microbial translocation and gut epithelial barrier dysfunction may promote persistent immune activation in treated HIV infection , potentially contributing to morbidity and mortality , it has been unclear whether CMV replication in individuals with no symptoms of CMV disease might play a role in this process . We hypothesized that persistent CMV replication in the intestinal epithelium of HIV/CMV-coinfected individuals impairs gut epithelial barrier function . Using a combination of state-of-the-art in situ hybridization technology ( RNAscope ) and immunohistochemistry , we detected CMV DNA and proteins and evidence of intestinal damage in rectosigmoid samples from CMV-positive individuals with both untreated and ART-suppressed HIV infection . Two different model systems , primary human intestinal cells differentiated in vitro to form polarized monolayers and a humanized mouse model of human gut , together demonstrated that intestinal epithelial cells are fully permissive to CMV replication . Independent of HIV , CMV disrupted tight junctions of polarized intestinal cells , significantly reducing transepithelial electrical resistance , a measure of monolayer integrity , and enhancing transepithelial permeability . The effect of CMV infection on the intestinal epithelium is mediated , at least in part , by the CMV-induced proinflammatory cytokine IL-6 . Furthermore , letermovir , a novel anti-CMV drug , dampened the effects of CMV on the epithelium . Together , our data strongly suggest that CMV can disrupt epithelial junctions , leading to bacterial translocation and chronic inflammation in the gut and that CMV could serve as a target for therapeutic intervention to prevent or treat gut epithelial barrier dysfunction during HIV infection .
Immune activation and intestinal epithelial barrier dysfunction are major hallmarks of HIV infection that persist in spite of potent combination antiretroviral therapy ( ART ) [1–6] . The barrier properties of mucosal intestinal epithelium are maintained by a monolayer of columnar epithelial cells that are firmly connected by intercellular tight junctions [7] . Disruption of these junctions was proposed as a mechanism for the increased colonic permeability in ART-suppressed patients [8] . In HIV infection , impaired integrity of the intestinal epithelial barrier facilitates bacterial translocation , a major contributor to chronic immune activation [4 , 6 , 9–11] . The precise mechanisms by which HIV perturbs the intercellular tight junctions of intestinal epithelia remain an active area of investigation [12–14] . Currently , gut epithelial barrier dysfunction during HIV infection is attributed to the increased production of inflammatory cytokines by activated mucosal T cells [15 , 16] and by mucosal epithelial cells directly responding to HIV-1 gp120 [17 , 18] . However , the cellular sources of inflammatory cytokines that lead to gut barrier dysfunction in HIV-infected individuals remain controversial [16] , and the mechanisms described above do not include the presence in the gut of a variety of opportunistic pathogens , specifically cytomegalovirus ( CMV ) , which has repeatedly been suggested as a cofactor for HIV disease progression and persistence of inflammation despite suppressive ART [19–21] . Importantly , the gastrointestinal tract experiences severe CD4+ T cell depletion at all stages of HIV disease [22] , which is not completely restored in chronically HIV-infected individuals following ART [23–25] and could thus be an important site for local and asymptomatic CMV reactivation . CMV was a common opportunistic pathogen in HIV-infected individuals before the introduction of ART , accounting for significant morbidity and mortality [26 , 27] . While end-organ disease from CMV ( e . g . , retinitis , colitis , and encephalitis ) is rare during ART-mediated viral suppression , CMV appears to contribute to persistent immune activation and inflammation in this setting [28] and is responsible for a sizable proportion of the entire memory T-cell response [29] . CMV often reactivates asymptomatically in ART-suppressed HIV-infected individuals , who are almost universally coinfected with CMV [30] , and the virus has a strong physiologic association with aging [29 , 31 , 32] , cardiovascular diseases [20 , 33–36] , and mortality [37 , 38] . The gastrointestinal tract is a major site of CMV disease in people with immune deficiency including individuals with HIV infection [39–43] . CMV targets endothelial , stromal , and intestinal epithelial cells and can cause erosive and ulcerative processes [40 , 44–46] . CMV lesions in the gut may compromise mucosal epithelial barrier function , which is maintained by intercellular tight junctions , multiprotein complexes that seal the space between adjacent cells [47] . We and others have previously reported that CMV disrupts the tight junctions of polarized retinal pigment epithelial cells and the adherens junctions of endothelial cells [48–52] . In this study , we used state-of the-art techniques in dual immunohistochemistry ( IHC ) and in situ hybridization ( ISH ) to show that CMV persistence in the rectosigmoid tissues of asymptomatic CMV-positive individuals with both untreated and ART-suppressed HIV infection was associated with gut epithelial barrier dysfunction . We used two different model systems to investigate the permissiveness of intestinal epithelial cells to CMV replication: primary human intestinal cells differentiated in vitro to form polarized monolayers and a mouse model of the human gut . We found that independent of HIV , CMV disrupts tight junctions of polarized intestinal cells , significantly reducing transepithelial electrical resistance ( TER ) , a measure of epithelial monolayer integrity , and enhancing epithelial barrier permeability . Furthermore , CMV-associated disruption of the intestinal epithelium integrity could be attributed , at least in part , to the CMV-induced proinflammatory cytokine interleukin-6 ( IL-6 ) . Importantly , letermovir , a novel anti-CMV drug currently in clinical development , preserved epithelial polarity in this system . Our results support a potentially active role for CMV in driving epithelial barrier dysfunction and microbial translocation . Taken together , these observations underscore a novel way to prevent and treat gut epithelial barrier dysfunction in HIV infection .
We analyzed rectosigmoid biopsies from CMV/HIV-coinfected individuals for CMV replication by IHC and RNAscope ISH . All these individuals had no symptoms or suspicion of CMV disease . To determine whether HIV-status is associated with an increase of CMV detection in the gut , a HIV-negative CMV-positive control group was included in the study . Clinical characteristics of the study cohort and the results of CMV detection are individually described in Table 1 . First , we identified regions of HIV-1 replication in biopsy samples from untreated CMV/HIV-coinfected individuals by RNAscope . Thymic organoid tissue from a HIV-infected humanized BLT mouse was used as a positive control for HIV detection [53] ( S1 Fig ) . HIV RNA was detected in numerous cells residing in gut-associated lymphoid tissue ( GALT ) ( Fig 1A ) , and a majority of these cells were CD3-positive T cells ( Fig 1B ) . Numerous CD163+CD68+ macrophages , often containing HIV RNA , were detected near intestinal crypts ( Fig 1C ) , and some of these macrophages revealed a single HIV transcript ( red dot ) in the nucleus ( Fig 1C , inset ) , indicating potential productive HIV replication . HIV RNA-positive cells were detected in rectosigmoid biopsies from all 3 asymptomatic CMV-positive individuals with untreated HIV infection that were examined , but not in biopsy samples from 5 HIV-negative CMV-negative individuals ( S1F and S1G Fig ) . Next , we assessed whether CMV infection coexisted in rectosigmoid biopsies from individuals with untreated HIV infection . CMV structural proteins were detected in the cytoplasm of intestinal epithelial cells and in leukocytes within the intestinal epithelium ( Fig 1D–1G ) . Immunostaining for CMV proteins was verified by both chromogenic ( Fig 1D ) and fluorescent ( Fig 1E–1G ) IHC . The specificity of detection of CMV proteins was verified by the lack of immunostaining with an isotype-control IgG ( Fig 1D & 1H , matching regions of interest of adjacent sections ) . CMV was detected in rectosigmoid biopsies from all 3 CMV-positive individuals with untreated HIV infection that were examined ( Table 1 ) , and CMV proteins were not detected in biopsy samples from 5 HIV-negative , CMV-negative individuals ( Fig 2B ) . We were surprised to observe clear evidence of CMV infection in the rectosigmoid biopsies from ART-suppressed HIV/CMV-coinfected individuals , although CMV was not present in all samples . CMV IE proteins , which are essential for CMV replication , were evident in the nuclei of numerous intestinal epithelial cells in 9 out of 19 samples ( Fig 2A ) . Intestinal epithelial cells that were identified by cytokeratin costaining frequently contained CMV proteins in the cytoplasm ( Fig 2E , inset ) . The specificity of CMV detection was verified by the absence of CMV staining in gut biopsies from HIV-negative CMV-negative individuals ( Fig 2B ) and by the lack of immunostaining of the adjacent section with an isotype control IgG ( Fig 1D & 1H , matching regions of interest of adjacent sections ) . We further assessed the extent of CMV infection in rectosigmoid biopsies from ART-suppressed individuals by DNAscope ISH ( Fig 2C ) . A human fetal lung explant inoculated ex vivo with CMV was used as a positive control for CMV detection [54] ( S2 Fig ) . Abundant CMV DNA was detected in numerous intestinal epithelial cells outlined by cytokeratin immunostaining ( Fig 2C and 2D ) . Importantly , the pattern of CMV DNA often had remarkable similarities to those we observed for CMV IE expression detected by IHC ( Fig 2C and 2A , yellow stars ) . Similar results were observed using probes targeting CMV IE ( S2A Fig ) and CMV pp65 coding sequences ( S2B Fig ) . Using two methods , IHC and DNAscope ISH , CMV was detected in biopsies from 9 of 19 ( 47% ) asymptomatic ART-suppressed HIV/CMV-coinfected individuals ( Table 1 ) . We also examined the level of HIV replication in the rectosigmoid biopsies from ART-suppressed HIV/CMV-coinfected individuals by RNAscope ISH . Single HIV RNA-positive cells were occasionally detected in these biopsies ( Fig 2F ) in contrast to rectosigmoid biopsies from CMV-positive individuals with untreated HIV infection ( Fig 1A–1C ) . CD68/CD163-positive macrophages containing HIV RNA ( Fig 2F , insets ) were occasionally detected in regions adjacent to those containing CMV DNA ( Fig 2C ) . Among ART-suppressed individuals , 64% had evidence of HIV replication in samples with detectable CMV activity , whereas 33% had evidence of HIV replication in samples without detectable CMV , although this difference was not statistically significant ( P = 0 . 18 ) . CMV infection in the intestine of individuals with inflammatory bowel diseases has been frequently correlated with an increased production of IL-6 [55] , and CMV shedding was associated with higher IL-6 levels in vaginal swabs of ART-suppressed HIV-infected women [56] . Therefore , we measured IL-6 in plasma samples obtained from the study participants immediately prior to biopsy . A trend toward higher IL-6 levels was found in the plasma of those ART-suppressed HIV/CMV-coinfected individuals whose rectosigmoid biopsies had evidence of CMV activity compared to gut samples from those without evidence of CMV activity ( median 2 . 6 versus 1 . 0 pg/ml , Mann Whitney , one-tailed P = 0 . 16 ) or those who were HIV- and CMV-seronegative ( median 2 . 6 versus 1 . 4 pg/ml , Mann Whitney , one-tailed P = 0 . 15 ) , although the differences were not statistically significant . IL-6 is a pleiotropic cytokine that can undermine the integrity of the intestinal barrier [57] . We next assessed whether CMV infection in the gut coexists with disrupted epithelial barriers . Tissue sections of the rectosigmoid biopsies described in Table 1 were coimmunostained for CMV proteins and zonula occludens-1 ( ZO-1 ) , a marker of tight junctions ( Fig 3 ) . ZO-1 was confined to the luminal apical surface of the intestinal epithelium and appeared uniformly distributed in the rectosigmoid biopsies from HIV-negative CMV-negative individuals ( Fig 3A ) . In contrast , in rectosigmoid biopsies from 3 untreated HIV/CMV-coinfected individuals , ZO-1 appeared discontinuous and vague in the crypt containing CMV-infected epithelial cells ( Fig 3B ) . In 9 rectosigmoid biopsies from ART-suppressed HIV/CMV-coinfected individuals , CMV detection in intestinal epithelial cells was accompanied by disruption of the continuity of ZO-1 ( Fig 3C ) . Interestingly , ZO-1 staining was intact and continuous in the crypts of rectosigmoid biopsies from 4 HIV-negative CMV-positive individuals ( Fig 3D ) . Although CMV proteins were detected in those biopsies in cells surrounding intestinal crypts , no CMV proteins were detected in the intestinal epithelial cells ( Table 1 ) , and epithelial integrity was not disturbed . To confirm our finding that disruption of intestinal tight junctions is specific to CMV , we examined multiple gut biopsies from an individual with clinically diagnosed CMV enteritis and colitis before and after valganciclovir treatment . Numerous CMV IE-positive cytomegalic cells were detected in the intestinal crypts in all 4 pretreatment biopsies ( Table 1 ) . Notably , those cytomegalic cells were often lifted from the level of the intestinal monolayer into the lumen , exhibiting damage of the epithelial barrier ( Fig 3E ) . The staining for ZO-1 in those crypts had decreased intensity in the tight junction regions and often was discontinuous . ZO-1 localization was diffuse or lost in the detaching cytomegalic cells and had remarkable similarity to the CMV-infected crypts observed in the rectosigmoid biopsies from untreated and ART-suppressed HIV/CMV-coinfected individuals ( Fig 3B and 3C ) . Importantly , after valganciclovir treatment , ZO-1 immunolabeling was intact , continuous , and strictly limited to the tight junction regions of the intestinal crypts , and no CMV was detected ( Fig 3F ) . Taken together , these data indicate that CMV persists in the intestinal epithelium of CMV-positive individuals with both untreated and ART-suppressed HIV infection and is associated with the disruption of epithelial tight junctions . To further investigate CMV pathogenesis in the intestinal epithelium , we developed a SCID-hu gut mouse model . Human fetal gut subcutaneously implanted into severe-combined immunodeficient ( C . B17 scid ) ( SCID ) mice developed a lumen with a morphologically precise mucosal layer containing villi and crypts , lamina propria , and muscularis layers during 4 weeks in vivo ( Fig 4A ) . The mucosal epithelium of the villi and crypts was composed of a single layer of intestinal epithelial cells expressing human cytokeratin ( Fig 4B ) . We analyzed 34 intestinal xenografts originating from 4 fetal donors and in all cases , the fetal intestine had matured into differentiated human intestine by 4 weeks after implantation . At this stage of development , gut implants were intraluminally inoculated with 5 . 7–6 . 4 log10 infectious units ( IU ) of CMV VR1814 . Using immunofluorescence techniques , we observed substantial CMV infection with marked mucosal damage ( Fig 4C ) 7 days after inoculation . During this period , CMV spread throughout the entire intestinal epithelial layer , depleting cytokeratin-positive epithelial cells ( Fig 4C ) . Cytomegalic IE-positive epithelial cells appeared to be detaching into the lumen of the crypts ( Fig 4D ) , exhibiting notable similarity to the detaching cytomegalic cells detected in the gut of an individual with active CMV infection ( Fig 3E ) . These data indicate that the intestinal epithelium is highly susceptible to CMV infection . To recapitulate CMV infection in rectosigmoid tissues , we devised an in vitro polarized cell model with primary human colon epithelial cells ( HCoEpiC ) . These cells retain the morphological and functional properties of intestinal epithelial cells during the first four passages . They expressed cytokeratin , an epithelial cell marker ( S3A Fig ) , and were negative for vimentin , a mesenchymal cell marker ( S3C Fig ) . When HCoEpiC were inoculated with CMV at a multiplicity of infection ( MOI ) of 1 . 0 , approximately 30% of the cells expressed CMV IE at day 1 ( Fig 5A and 5C ) and at day 3 , some infected cells expressed the CMV envelope glycoprotein gB , indicating productive infection ( Fig 5B ) . The epithelial origin of the infected cells was verified at each passage by coimmunostaining for cytokeratin ( S3B Fig , Fig 5A and 5B ) . Once HCoEpiC cells were plated on collagen-coated transwells for 6 days , they differentiated forming polarized monolayers and expressed ZO-1 in a ring-like pattern ( Fig 5D ) . When polarized HCoEpiC were inoculated with CMV at a MOI of 1 . 0 , only an average of 3 . 7% of the cells expressed CMV IE at day 1 ( Fig 5C and 5E ) . Thus , the susceptibility of polarized HCoEpiC to CMV infection was significantly lower as compared to nonpolarized cells ( P<0 . 0001 ) . Furthermore , we found that polarized HCoEpiC sustained a persistently low level of CMV replication as indicated by infectious virus released into the culture medium at approximately 2 . 5 log10 IU per ml during 3–9 days after virus inoculation ( Fig 5F ) . Although this titer of infectious virus is very low relative to the high level of virus in the inoculum ( day 0 ) , it was significantly higher than residual input virus detected at day 1 . Moreover , no infectious virus was detected when virus replication was suppressed with the reference compound ganciclovir ( GCV ) , a widely used polymerase inhibitor for the treatment of CMV infection and the novel small-molecule CMV viral terminase inhibitor AIC246 ( letermovir ) [58] . The intriguing observation of persistent low-level CMV replication in polarized HCoEpiC could result from the distinctive feature of intestinal epithelial cells to detach into the crypt lumen with the loss of polarity , thus exposing additional target cells to CMV infection ( Fig 3E , Fig 4D ) . Indeed , 9 days after inoculation , we detected only about 7% of IE-positive cells ( Fig 5G and 5I ) , but cytomegalic cells abundantly expressing CMV gB were detected above the polarized monolayer showing their detaching capacity . Despite the low number of CMV IE-positive cells at day 9 after inoculation , it was significantly higher than in the presence of GCV or letermovir ( Fig 5I ) . These data indicate that although CMV preferentially infects nonpolarized intestinal cells , it is capable of maintaining a low level of replication in highly differentiated polarized intestinal cells . Taken together , these in vivo and in vitro experiments showed that human intestinal epithelial cells are fully permissive for CMV replication . The association of CMV with the disruption of intercellular junctions of retinal pigment epithelial cells [48–50] and endothelial cells [51 , 52] prompted us to examine the effect of CMV infection on the integrity of the intestinal epithelium . When HCoEpiC were grown on collagen-coated porous membranes of transwells that allow access to the culture medium from both the apical and basolateral plasma membranes , they differentiated into a highly polarized epithelial monolayer . Cells formed a sealed epithelium with closed tight junctions ( Fig 5D ) and developed a considerable TER ( >100 Ohm*cm2 ) over 6 days ( Fig 6A ) , mirroring intestinal epithelium in humans . Although TER of the HCoEpiC monolayers continued to increase , we observed the formation of multilayered crypt-like structures and invaginations after 12 days , which hampered further study of epithelial permeability . We therefore performed all experiments within the first 12 days when the cells formed a single polarized monolayer . To examine the role of polarity in the infection process , we applied the virus inoculum either to the apical surface of polarized cells or to their basolateral surface . For this reason , HCoEpiC were grown on the upper or lower surface of the transwell membrane to orient upward the apical or basolateral side of the monolayer , respectively ( Fig 6 ) . For apical infection ( Fig 6A ) , TER of mock-infected HCoEpiC cells continuously increased after formation of a dense monolayer with well-established tight junctions 6 days after seeding ( Fig 5D ) . In contrast , TER of CMV-infected cells declined almost 3-fold ( P<0 . 01 ) over the 6 days after virus inoculation ( Fig 6A ) and was accompanied by the disappearance of a ring-like pattern of ZO-1 ( Fig 5E ) . For the study of basolateral infection , we used a 10-fold higher virus inoculum than applied to the apical surface and measured the infectivity of virus passed over blank collagen-coated transwell membranes to ensure that comparable numbers of virions reached the basolateral surface of the HCoEpiC monolayers , as we reported earlier [51] . Infection of HCoEpiC from the basolateral side also significantly reduced TER by almost half ( P<0 . 05 ) over 6 days of infection ( Fig 6B ) . Thus , we found that CMV VR1814 infected polarized HCoEpiC from both sides , significantly decreasing their TER . However , HCoEpiC developed monolayers with higher TER when they were seeded on the upper surface for apical infection ( Fig 6A ) , so apical infection was chosen for all following experiments . To determine whether virus replication is required for the disruption of intestinal epithelial tight junctions , we inhibited CMV replication with GCV and letermovir ( Fig 7A ) . Polarized HCoEpiC were mock or CMV infected at a MOI of 1 . 0 , and TER was monitored during the course of infection . The TER of CMV-infected untreated cells decreased gradually over the course of infection starting from 1 h after virus inoculation , although differences relative to the mock-infected control reached statistical significance only at day 1 ( P<0 . 05 ) . Treatment of polarized HCoEpiC after CMV inoculation with an inhibitory concentration of GCV and letermovir remarkably prevented significant declines in monolayer resistance over the 9-day period ( P<0 . 05 ) . The effect of both drugs was sufficient to maintain epithelial cell polarity at TER >100 Ohm*cm2 . Although both drugs had no effect on the TER of mock-infected cells during the first 3 days after treatment , the TER of mock-infected cells treated with GCV significantly declined at day 9 ( P<0 . 05 ) but not of cells treated with letermovir ( Fig 7B ) . This effect of GCV on the TER of mock-infected cells was reflected by the slight decline in TER at day 9 in infected cells that were otherwise protected by GCV treatment ( Fig 7A ) . No changes in monolayer resistance were observed when cells were infected with virus inoculum filtered through Millex syringe filter units with 0 . 1 μm pore size to remove CMV virions . These results indicated that the virus inoculum contains no factors that could affect monolayer integrity and that CMV itself induced a decline in TER ( Fig 7B ) . We further examined the effect of CMV on epithelial permeability by monitoring the ability of a fluorescently labeled , inert dye ( FITC-dextran ) to migrate from the apical transwell compartment to the lower compartment . Consistent with the results of the TER studies ( Fig 7C ) , a significant increase in FITC-dextran concentration in the lower chamber ( P<0 . 0001 ) was observed in CMV-infected cells compared with mock-infected cells ( Fig 7D ) . Treatment with letermovir moderately prevented CMV-induced FITC-dextran leakage across polarized HCoEpiC monolayers ( P<0 . 05 ) but did not block it to the level of mock-infected controls . Mock-infected cells displayed well-defined epithelial junctions 9 days after inoculation as indicated by ZO-1 expression in a ring-like pattern ( Fig 7E ) . CMV-infected polarized HCoEpiC lacked ZO-1 expression indicating the entire disassembly of epithelial junctions ( Fig 7F ) . Notably , CMV-infected polarized HCoEpiC treated with letermovir occasionally expressed CMV IE but retained tight junctions , although a fraction of ZO-1 was redistributed in the cytoplasm ( Fig 7G ) . Altogether , these results showed that CMV infection disrupted the tight junctions of polarized colon epithelial cells , reducing their TER and increasing epithelial barrier permeability . CMV is an inducer of TNF-α and IL-1β in the monocyte cell line THP-1 [59 , 60] and of IL-6 in peripheral blood mononuclear cells , endothelial cells , and lung fibroblasts [61–63] . On the other hand , it is well known that epithelial tight junction integrity and barrier permeability could be compromised by proinflammatory cytokines TNF-α , IL-6 , and IL-1β [64–66] . Therefore , we investigated whether CMV infection induced expression of these proinflammatory cytokines in intestinal epithelial cells . Polarized HCoEpiC were mock or CMV infected; TER was monitored every 3 days; and IL-6 , TNF-α , and IL-1β in the culture medium were measured by ELISA ( Fig 8 ) . Over 9 days after inoculation , we observed a gradual decline of TER that was significantly lower compared with mock-infected cells ( P<0 . 05 ) and indicated CMV-induced impairment of polarized monolayer barrier function ( Fig 8A ) . In parallel , we observed increased production of IL-6 , but not TNF-α or IL-1β ( Fig 8B–8D ) . IL-6 was detected 1 day after inoculation , and peak levels were 2 . 7-fold higher than in mock-infected controls ( P<0 . 05 ) . A significant increase in IL-6 relative to mock-infected controls was also observed at days 3 and 9 ( P<0 . 05 ) and although IL-1β showed a trend toward an increase at day 3 , the differences were not statistically significant . To assess donor-to-donor variation in proinflammatory cytokine production , we further compared HCoEpiC derived from two donors ( Fig 9A–9D ) . As shown for day 9 after inoculation in Fig 9A , TER was decreased by 2 . 7-fold in cells derived from donor 1 and 3-fold in cells from donor 2 ( P<0 . 05 compared to mock infected ) . In cells derived from donor 1 , CMV significantly elevated TNF-α by 2 . 1-fold ( P<0 . 05 ) and IL-6 by 2 . 5-fold ( P<0 . 05 ) but did not elevate IL-1β ( Fig 9B–9D ) . In cells derived from donor 2 , CMV significantly induced IL-6 by 1 . 4-fold ( P<0 . 05 ) , but not TNF-α nor IL-1β . Since IL-6 was induced by CMV in cells derived from both donors , we further investigated its role in function-inhibiting experiments ( Fig 9E ) . To determine whether IL-6 is required for CMV-induced disruption of the intestinal epithelial barrier , we used anti-IL-6 neutralizing antibody to block IL-6 function . To stimulate a higher level of IL-6 production , polarized HCoEpiC cells were inoculated with a higher dose of virus at a MOI of 5 . 0 and were treated before and during CMV infection with mouse IgG anti-human IL-6 or isotype control IgG . Although anti-IL-6 neutralizing antibody were able to prevent , at least partially , CMV-induced disruption of polarized HCoEpiC integrity over 6 days after inoculation ( P<0 . 05 ) , the most significant effect was observed the day after inoculation when TER of the treated HCoEpiC monolayers was 1 . 4-fold higher than in cells treated with isotype control IgG ( P<0 . 01 ) . On the other hand , the TER of CMV-infected cells treated with anti-IL-6 IgG at this time point was still significantly lower than TER of mock-infected cells ( P<0 . 05 ) . These results indicate that CMV-associated disruption of colonic epithelium integrity could be attributed only to some extent to CMV-induced IL-6 . Taken together , these data indicate that infection of polarized colon epithelial cells with CMV VR1814 induces rapid ( within 24 h ) release of IL-6 , which is able to significantly reduce the TER of the monolayer , thus diminishing epithelial barrier function . Importantly , the novel drug letermovir itself had no effect on the integrity of polarized colon epithelial cells and was able to largely protect colonic epithelial barriers from CMV-induced dysfunction .
During HIV infection , immune activation linked to intestinal epithelial barrier dysfunction persists despite potent suppressive ART [5 , 6 , 10 , 67–69] . The underlying mechanisms are complex and remain unclear , but the role of opportunistic viral pathogens in the gut has yet to be fully appreciated . Here we report that CMV persists in the rectosigmoid tissues of asymptomatic CMV-positive individuals with both untreated and ART-suppressed HIV infection and that CMV infection coexists with discontinuous epithelial tight junctions . Independent of HIV , CMV impairs the integrity of polarized human intestinal cells , significantly reducing transepithelial electrical resistance and enhancing epithelial barrier permeability . CMV-associated disruption of intestinal epithelium integrity was mediated , at least in part , by CMV-induced IL-6 . These observations suggest that CMV reactivation in the gastrointestinal epithelium of HIV-infected individuals could be a potent cofactor that stimulates release of proinflammatory cytokines from intestinal epithelial cells , compromising barrier function and locally initiating bacterial translocation that leads to chronic inflammation in the gut ( Fig 10 ) . Cytomegalovirus , as an opportunistic pathogen in HIV-infected individuals [70] , was the cause of significant morbidity and mortality before the introduction of ART [26 , 27] . In the era of effective ART , CMV remains an important cofactor in HIV disease progression [30 , 71–74] [37 , 38] , showing strong association with systemic inflammation [28 , 29] , aging physiology [29 , 31 , 32] , and cardiovascular disease [20 , 33–35] . Importantly , in ART-suppressed individuals , T-cell reconstitution in the gastrointestinal tract never reaches the levels observed in uninfected healthy persons [22–25] and could thus be an important site of CMV reactivation [75] . Indeed , the persistent T-cell activation in HIV-infected individuals on ART that is associated with decreased gains in CD4+ T-cell counts [76] and mortality in some , but not all , studies [4 , 77 , 78] was reduced by valganciclovir , a potent anti-CMV nucleoside analog [28] . In our study , we detected numerous HIV RNA-positive T cells and macrophages in the GALT ( Fig 1A–1C ) of untreated HIV/CMV-coinfected participants , suggesting ongoing elimination of CD4+ cells that allows CMV reactivation . In rectosigmoid biopsies from ART-suppressed HIV/CMV-coinfected individuals , we also observed occasional HIV RNA-positive T cells and macrophages , indicating continued low levels of HIV expression in the gut that may promote continued T-cell depletion ( Fig 2F ) . Active CMV infection ( determined by CMV antigenemia and a tissue biopsy specimen that is positive by either culture or IHC staining ) is a well-recognized cause of symptomatic colitis in untreated HIV-infected individuals with advanced AIDS . We show here that CMV also persists in the rectosigmoid tissues of asymptomatic CMV-positive participants with both untreated and ART-suppressed HIV infection ( Table 1 ) . In untreated HIV infection , CMV proteins were detected in intestinal epithelial cells and in leukocytes located within the mucosal epithelium ( Fig 1D–1G ) . In ART-suppressed HIV infection , both CMV proteins and viral DNA were detected in the intestinal epithelial cells in 9 of 19 ( 47% ) biopsies , suggesting that the intestinal epithelium supports CMV replication ( Fig 2A–2E ) . Importantly , CMV caused a patched pattern of infected cells , a finding that suggests that the lack of CMV proteins or DNA in the other 10 biopsies may represent a sampling error , especially because such small pieces of tissue are taken from one of the largest organs in the body . Thus , we cannot exclude the possibility that CMV was indeed present in the gut of the ART-suppressed HIV/CMV-coinfected individuals who had CMV-negative biopsies in our experiments . The small pinch biopsies may not be fully representative of all gut areas , and CMV shedding may be intermittent , limiting our tissue analyses . On the other hand , the detection of even a few infected cells in these limited areas could reflect the presence of a very large number of CMV-infected cells throughout the intestine . Our finding that CMV infection in the gut biopsies of asymptomatic CMV-positive participants with both untreated and ART-suppressed HIV infection coexists with discontinuous epithelial tight junctions was especially intriguing . Tight junctions are a distinguishing structural feature of polarized epithelial cells and are crucial for the formation their barrier function [47] . We found here that the typically very distinct apical localization of the junctional adaptor protein ZO-1 in epithelial cells of the intestinal crypts was impaired in the regions where CMV proteins were detected ( Fig 3 ) . Importantly , when CMV replication was suppressed with valganciclovir , all intestinal crypts had well-developed junctions with ZO-1 immunolabeling strictly limited to the tight junction regions ( Fig 3F ) . These observations were notably similar to our data for CMV infection in polarized HCoEpiC ( Fig 5E and Fig 7 ) . HIV has well-established cofactor relationships with CMV [70] . Asymptomatic shedding of CMV in the male and female genital tract of ART-suppressed HIV-infected individuals is associated with increased systemic immune activation and higher levels of HIV DNA in peripheral CD4+ cells [56 , 72 , 79] . In accordance with that observation , we noticed a trend toward association between the presence of CMV DNA/proteins and HIV RNA in the rectosigmoid tissues of asymptomatic CMV-positive participants with ART-suppressed HIV infection ( Table 1 ) . Although the differences did not reach statistical significance in this small set of participants , we anticipate that future study with larger numbers of participants will better define the link between CMV and HIV replication in the gut . Nevertheless , our finding that ZO-1 staining was intact in the crypts of rectosigmoid biopsies from 4 HIV-negative CMV-positive individuals ( Fig 3D ) supports this link . Despite detection of numerous CD68/CD163-positive macrophages containing CMV proteins in these biopsies , no spread of CMV infection to intestinal epithelial cells was detected . These observations suggest that HIV indeed is associated with an increase of CMV infection concurrent with damage to the intestinal epithelium . To define the role of intestinal epithelial cells during gastrointestinal CMV disease independent of HIV , we further investigated their susceptibility to CMV infection using a SCID-hu gut mouse model reported previously [80–83] . We observed extensive damage of the mucosal epithelium 7 days after inoculation with CMV ( Fig 4C ) , and infection was indicated by the abundant expression of CMV IE proteins . The lumen of damaged intestinal crypts contained numerous IE-positive cytomegalic cells ( Fig 4D ) , mirroring the shedding of cytomegalic cells observed in the gut biopsies with clinically diagnosed CMV colitis ( Fig 3E ) . We are intrigued by the high level of susceptibility of the intestinal epithelium to CMV in vivo and our finding that CMV persists in the intestinal epithelium of HIV/CMV-coinfected individuals . To focus further on the study of CMV pathogenesis exclusively in intestinal epithelial cells , we modeled CMV infection in vitro by the use of primary colon epithelial cells . Although CMV disease can involve any region of the gastrointestinal tract in AIDS patients , the colon is the site most frequently infected [40–42 , 84] . Much of our knowledge about CMV pathogenesis in the intestinal epithelium comes from studies conducted in the Caco-2 cell line of heterogeneous human epithelial colorectal adenocarcinoma cells [85 , 86] . The permissiveness of Caco-2 cells to various laboratory strains of CMV is controversial , so we performed our study in primary colon epithelial cells ( HCoEpiC ) using VR1814 , a clinical isolate of CMV that was adapted for growth in human umbilical vein endothelial cells ( HUVEC ) [87] and maintained at low passages 27–29 . These cells formed polarized monolayers when plated on collagen-coated transwell inserts for 6 days , expressing ZO-1 in a ring-like pattern ( Fig 5D ) , and they were permissive to CMV replication producing infectious viral progeny ( Fig 5E–5H ) . Experiments with CMV Towne in polarized Caco-2 cells inoculated at a MOI of 25 showed that this strain preferentially infects from the basolateral surface . In contrast , we found that the CMV strain VR1814 impaired integrity of polarized HCoEpiC monolayers to a greater extent when the cells were inoculated at a MOI of 1 . 0 from the apical surface ( Fig 6 ) . Basolateral entry of CMV into polarized HCoEpiC recapitulates hematogenous virus spread because the basolateral surface of the mucosal epithelium is oriented toward the underlying blood vessels of the lamina propria . Apical entry of CMV into polarized HCoEpiC may resemble both homosexual virus transmission via semen [88–91] and postnatal CMV transmission via breast milk in premature infants [92–94] . The latter is very intriguing because HCoEpiC cells were derived from the fetal intestine , therefore CMV entry through the apical surface could be specific to fetal/neonatal cells . We and others have previously reported that CMV disrupts the tight junctions of polarized epithelial cells and the adherens junctions of polarized endothelial cells [48–52] . In Caco-2 cells , CMV disrupted polarized monolayers and decreased TER beginning 12 days after inoculation [85] . In our experiments , CMV disrupted tight junctions of polarized HCoEpiC and significantly reduced TER by 1 day after inoculation , demonstrating the high degree of virulence of VR1814 in primary colon epithelial cells ( Fig 7A ) . Notably , the novel anti-CMV drug letermovir , which exhibits potent antiviral activity against clinical isolates [58 , 95] , prevented damage of the HCoEpiC monolayer to a level that was sufficient to maintain cell polarity ( Fig 7A and 7C ) and to inhibit CMV-induced “leakage” of FITC-dextran ( Fig 7D ) . Importantly , in contrast to GCV , letermovir itself had no effect on the integrity of mock-infected polarized HCoEpiC ( Fig 7B ) . Of note , both drugs were not able to prevent the initial small TER declines at day 1 after CMV inoculation ( Fig 7A ) . GCV targeting viral polymerase UL54 acts at early steps of the CMV replication cycle [96] . Letermovir is known to act at a late stage of CMV replication that involves viral DNA processing and/or packaging without inhibiting de novo synthesis of CMV DNA and protein expression [95] . We suggested earlier that increased barrier permeability of polarized retinal pigment epithelial cells was mediated by the CMV protein US9 [49 , 97] . Since CMV US9 gene expression peaks 24 h after infection [98] , it would be interesting to determine whether US9 or other CMV early genes could be directly involved in the disruption of intercellular tight junctions of the intestinal epithelium . CMV is a well-known inducer of TNF-α and IL-1β in the monocyte cell line THP-1 and of IL-6 in peripheral blood mononuclear cells , lung fibroblasts , and endothelial cells [59–63] . Moreover , TNF-α transcripts are abundantly expressed in colonic mucosa from untreated AIDS patients with CMV colitis and are associated with macrophage-like cells containing cytomegalic inclusions [99] . Importantly , upregulation of those proinflammatory cytokines has been attributed to CMV IE gene expression , and these cytokines increase paracellular permeability of epithelial and endothelial cells [57 , 100–109] . It is interesting that despite the induction of proinflammatory cytokines early after infection , CMV suppresses TNF-α and IL-1β signaling pathways at late times , demonstrating unique adaptation capacities that enable virus persistence within the host [110] . Altogether , these published observations suggest that CMV-induced disruption of intestinal tight junctions could be mediated by these proinflammatory cytokine pathways . Indeed , in our experiments we found that the initial drop in HCoEpiC TER at 24 h after CMV inoculation was accompanied by a significant increase ( P<0 . 05 ) in secretion of IL-6 ( Fig 8A and 8B ) . The role of CMV-induced IL-6 in this process was confirmed by experiments demonstrating an increase in TER when cells were treated before and during infection with IL-6-neutralizing antibodies ( Fig 9E ) . At 24 h , the TER of mock-infected cells was still higher than CMV-infected cells in the presence of anti-IL-6 antibodies ( P<0 . 05 ) , suggesting that other CMV-induced factors or viral genes could be involved in the modulation of intestinal epithelium permeability . We anticipate that future studies will better define those additional factors . Nevertheless , our results showed that CMV-associated disruption of colon epithelial integrity could be attributed , at least in part , to CMV-induced IL-6 . Notably , HIV infection does not induce IL-6 expression in vitro [111] , although increased plasma IL-6 levels have been associated with HIV disease progression risk [1 , 112–116] . The exact cause of elevated plasma IL-6 in chronic HIV disease remains unclear , but it could result from a combination of factors . Our data indicate that persistent CMV replication in various sites of HIV-infected individuals could be one of these factors . Others have reported that active CMV infection could be often detected in the intestine of individuals with inflammatory bowel diseases and may contribute to the inflammatory process through virus-induced IL-6 [55] . It is interesting to note that intestinal tissue samples in that prior study included 10 gut biopsies from CMV-infected AIDS participants , and CMV-specific antigens were detected in all 10 samples , 4 of which were double-positive for both CMV and IL-6 . Furthermore , CMV shedding was associated with higher IL-6 levels in vaginal swabs after initiation of ART in HIV-infected women [56] . We also found a trend toward higher IL-6 plasma level in samples obtained from participants with CMV-positive gut biopsies compared to CMV-negative biopsies , but the differences did not reach statistical significance . These results may reflect the fluctuating levels of IL-6 that we observed in our in vitro HCoEpiC time-course experiments ( Fig 8B ) and indicate variable levels of IL-6 in various tissues containing CMV foci . Nevertheless , the modest changes in IL-6 plasma levels in this small set of participants are highly intriguing , and we anticipate that future study with larger numbers of participants will better define the relationship between plasma IL-6 and the presence of CMV in the gut . It is equally interesting to consider the role of CMV-induced IL-6 in increased epithelial proliferation in ART-treated and untreated individuals with intestinal epithelial barrier dysfunction [6] . We reported earlier that disruption of adherens junctions of polarized endothelial cells by CMV triggered proliferation of bystander endothelial cells [51] , suggesting a balance between lysis of infected endothelial cells and replacement by uninfected ones [117] . It has also been reported that the CMV-encoded chemokine receptor US28 mediates cell proliferation through activation of the IL-6-STAT3 signaling axis [118] and that the CMV secretome promotes angiogenesis and lymphangiogenesis through IL-6 signaling [119 , 120] . Based on these published reports , we speculate that CMV infection of the intestinal epithelium may trigger IL-6 production , resulting in enhanced cell proliferation to replace infected intestinal cells . Experiments with polarized HCoEpiC revealed cytomegalic gB-expressing cells positioned above the cell monolayer ( Fig 5H ) , mimicking the shedding of cytomegalic cells in the gut of an individual with CMV colitis ( Fig 3E ) and in the SCID-hu gut infected in vivo with CMV ( Fig 4D ) . Furthermore , CMV slowly replicated in polarized HCoEpiC releasing low levels of viral progeny ( Fig 5F ) . These findings suggest that CMV does not lyse all the cells in foci of polarized intestinal cells . Detached infected cells could be replaced by more permissive nonpolarized proliferating cells ( Fig 5C ) , creating small foci of CMV infection with compromised tight junctions that may locally facilitate bacterial translocation leading to persistent inflammation [10] . AIDS-associated gastrointestinal CMV disease typically responds well to ganciclovir and foscarnet [121 , 122] . However , the impact of CMV on the junctions of the intestinal epithelium in the presence of currently approved or novel HCMV drugs has never before been studied . Here we found that CMV-induced disruption of polarized HCoEpiC monolayer integrity could be prevented by the novel small-molecule CMV terminase inhibitor letermovir [58] ( Fig 7A and 7C ) . Despite the observation that the TER of CMV-infected cells in the presence of letermovir was still below the level of mock-infected cells , it exceeded the 100 Ohm*cm2 threshold , indicating that cells maintained their intact junctions and cell polarity . Letermovir is a new , anti-CMV agent currently being developed for the prevention of CMV disease in immunocompromised transplant patients [123 , 124] . Importantly , letermovir has no antagonism across a broad panel of commonly used anti-HIV drugs and was suggested for treatment of active CMV infection in HIV/CMV-coinfected individuals [125] . Our finding that attenuation of CMV infection by letermovir was sufficient to prevent significant CMV-induced declines in TER supports development of novel interventional strategies to increase intestinal epithelial barrier function in HIV infection . In summary , in this paper we addressed the role of an important opportunistic pathogen , CMV , as a cofactor of intestinal barrier dysfunction in asymptomatic HIV infection by several approaches . First , we used a combination of state-of-the-art ISH technology ( RNAscope ) and immunohistochemistry to demonstrate that CMV persists in the gut of CMV-positive participants with both untreated and ART-suppressed HIV infection and is associated with the disruption of epithelial tight junctions . We then used two different model systems: polarized primary human intestinal cells and a mouse model of human gut to assess the impact of CMV on the intestinal epithelium independently of HIV . We found that CMV impairs the integrity of intestinal epithelial cells , reducing transepithelial electrical resistance and enhancing epithelial barrier permeability , and that this effect is mediated , at least in part , by CMV-induced IL-6 . These results support our hypothesis that CMV reactivation in the gastrointestinal epithelium of HIV-infected individuals is a potent cofactor that stimulates release of proinflammatory cytokines from intestinal epithelial cells , compromising their barrier function , and initiating localized bacterial translocation leading to chronic inflammation in the gut . Moreover , we showed that CMV-associated disruption of intestinal epithelium integrity could be largely prevented by letermovir , a novel anti-CMV drug currently in clinical development . Thus , the addition of letermovir to a suppressive ART regimen might be explored as a strategy to reduce microbial translocation and systemic immune activation in future trials . Altogether , our results provide further evidence that CMV remains an important cofactor in HIV pathogenesis even during ART-mediated HIV suppression and suggest new antiviral interventions to prevent gut epithelial barrier dysfunction in treated HIV infection .
Archived rectosigmoid biopsy samples were from the SCOPE cohort at the University of California , San Francisco ( UCSF ) . The SCOPE cohort is an ongoing longitudinal study of over 1 , 500 HIV-infected and uninfected adults followed for research purposes . The UCSF Committee on Human Research reviewed and approved the SCOPE study ( IRB# 10–01218 ) , and all participants provided written informed consent . Protocols involving animals were approved by the UCSF Institutional Animal Care and Use Committee ( #AN111327-02 ) . This protocol adheres to the National Institutes of Health’s Public Health Service Policy on Humane Care and Use of Laboratory Animals . Archived rectosigmoid biopsy samples were retrieved from the SCOPE cohort at UCSF as previously described [6] . For SCOPE studies , participants were recruited based on CD4 count , duration of ART , and HIV infection and treatment status , and they had no symptoms or suspicion of CMV disease . Using this cohort , archived gut biopsies and plasma samples were selected from 5 HIV-negative , CMV-negative controls; 3 CMV-positive , HIV-viremic untreated individuals; 19 CMV-positive , ART-suppressed individuals; and 4 HIV-negative , CMV-positive individuals ( Table 1 ) . Two samples were from an individual with clinically diagnosed active CMV infection before and after valganciclovir treatment . Prior to sigmoidoscopy and gastrointestinal biopsy , study participants underwent a blood draw and received a Fleet enema . CD4+ T-cell counts ( cells/mm3 ) and CMV serology ( CMV IgG ) were performed by a Clinical Laboratory Improvement Amendments ( CLIA ) -certified clinical laboratory located at Zuckerberg San Francisco General . Rectosigmoid biopsies ( ~3 mm in diameter ) were obtained 10–20 cm from the anus using jumbo forceps , and four biopsies were formalin fixed and paraffin embedded for IHC and ISH . Fetal gut tissues ( 18–24 g . w . ) were obtained from women with normal pregnancies before elective termination for nonmedical reasons with informed consent according to local , state , and federal regulations . Single intact segments of the human fetal intestine ( 2−3 cm in length ) were transplanted subcutaneously on the back of 6–8-week-old male C . B17 scid mice ( C . B-Igh-1b/IcrTac-Prkdcscid Taconic ) [83] . Mice were maintained under specific pathogen-free conditions . Four weeks after gut implantation , mice were anesthetized , and implants were inoculated with 5 . 7–6 . 4 log10 IU of CMV by direct injection into the gut lumen . At 7 and 14 days after inoculation , mice were euthanized , and implants were dissected and fixed in 3 . 7% formaldehyde solution ( Sigma-Aldrich ) for immunohistochemistry . VR1814 , a clinical isolate of CMV that was adapted for growth in HUVEC [87] , was a generous gift ( at passage 23 ) from Dr . Maria Grazia Revello . Virus was propagated in HUVEC ( Lonza ) , and viral stocks at passages 27–29 were prepared from supernatant virus [126 , 127] . The titers of infectious virus were determined by a rapid method of immunological detection and quantification of CMV IE proteins that has been shown to correlate with the conventional plaque assay [126 , 128] . Neonatal human dermal fibroblasts ( NHDF-Neo ) ( Lonza ) were used as the indicator monolayer for the assay , and virus titers are expressed as IU [126] . HCoEpiC isolated from human fetal colonic tissue were purchased from ScienCell at passage 1 . HCoEpiC were cultured in colonic epithelial cell medium ( CoEpiCM , ScienCell ) , and all experiments were performed at passage 2 or 3 . The purity of colonic epithelial cells was verified at each passage by immunofluorescence staining of formaldehyde-fixed cytospin preparations with a rabbit monoclonal antibody to human cytokeratin 19 ( Abcam ) ( S3 Fig ) . To form polarized monolayers , HCoEpiC at passage 2 were seeded at 3 x 104 cells/cm2 on either the upper or lower surface of 6-mm-diameter transwell permeable membranes ( 0 . 45 μm pore size , Costar ) coated with 30 μg/ml rat-tail type 1 collagen ( BD Biosciences ) for 30 min at 37°C . To seed the cells on the lower surface , the transwell insert was inverted for 2 h to allow the cells to attach . The monolayers were grown for at least 6 more days with fresh medium added every 2 days . TER of cells grown on transwell permeable membranes was measured with a Millicell ERS-2 voltohmmeter ( Millipore ) . The value obtained from a blank insert coated with collagen was subtracted to obtain net resistance , which was multiplied by the membrane area to yield the resistance in area-corrected units . TER values exceeding the 100 Ohm*cm2 threshold indicated that cells had developed tight junctions and were polarized . The polarity of HCoEpiC monolayers was verified in parallel by fluorescence immunolocalization of ZO-1 , a marker of tight junctions ( Fig 5D ) . The functional integrity of the tight epithelial monolayer was further confirmed by the lack of FITC-dextran leakage in the transepithelial permeability assay . The permeability of the polarized monolayers was quantified by measuring the transepithelial flux of 4-kDa FITC labelled dextran ( Sigma-Aldrich ) . Briefly , following polarization , 150 μl of Hank’s balanced salt solution ( HBSS ) containing FITC-dextran ( 1 mg/ml ) was applied to the upper chamber of the transwell , and 450 μl of HBSS was added to the lower chamber . As a positive control ( permeable monolayer ) , 1% Triton X-100 ( Sigma-Aldrich ) in HBSS was used to destroy epithelial integrity . Following incubation at 37°C for 2 h , 100 μl of the HBSS from the lower chamber was transferred into a translucent 96-well plate to measure relative fluorescence units ( RFU ) in triplicate . A fluorescence standard curve was prepared with fixed concentrations of FITC-dextran in HBSS from 200–3 . 125 μg/ml for subsequent extrapolation of unknown FITC-dextran concentrations recovered from the basolateral transwell compartment . RFU was measured with a SpectraMax M2 multimode microplate reader ( Molecular Devices ) with excitation wavelength of 490 nm and emission wavelength of 520 nm . To suppress CMV replication after virus adsorption , polarized HCoEpiC were treated with inhibitory concentrations ( 10 times the IC50 ) of letermovir ( 50 nM ) ( MedchemExpress , LLC ) or GCV ( 20 μM ) ( Invivogen , Inc . ) as described [58] . Since letermovir was dissolved in DMSO , a vehicle control ( 0 . 001% DMSO ) was included . The effect of IL-6 was studied in function-inhibiting experiments using a neutralizing anti-human IL-6 mouse monoclonal antibody ( Abcam ) . Before and during infection , polarized HCoEpiC were incubated with anti-human IL-6 at 0 . 1 mg/ml ( 5 times the 50% neutralization dose ) as described [129] . Low-endotoxin , azide-free isotype control mouse IgG ( Abcam ) at the same concentration was used as a control . NHDF-Neo used as the indicator monolayer for titration of infectious CMV , and HCoEpiC cytospins were fixed with ice-cold 70% methanol ( Sigma-Aldrich ) for 20 min . Polarized HCoEpiC in transwells were fixed in 3 . 2% paraformaldehyde ( Electron Microscopy Sciences ) for 15 min at room temperature and then permeabilized for 5 min in 0 . 1% Triton-X-100 ( Sigma-Aldrich ) . Gut implants were fixed in 3 . 7% formaldehyde ( Sigma-Aldrich ) and infiltrated with 5–15% sucrose followed by embedding in optimal-cutting-temperature compound and freezing in liquid nitrogen [130] . Tissue sections of archived FFPE biopsy samples were deparaffinized in xylene and rehydrated in graded alcohols . Endogenous peroxidase was quenched with 3% H2O2 in methanol for 10 min , and heat-induced epitope-retrieval using 10 mM sodium citrate buffer ( pH 6 . 0 ) was performed in a DC2002 decloaking chamber ( Biocare Medical ) . For ZO-1 detection , antigen retrieval was followed by permeabilization with 0 . 2% Triton X-100 for 45 min . Primary antibodies included a mouse monoclonal to CMV IE and a blend ( Millipore ) of anti-CMV monoclonal antibody clone 8B1 . 2 reacting with an IE protein of 68–72 kDa , clone 1G5 . 2 reacting with an unspecified late CMV protein , and clone 2D4 . 2 reacting with a late structural protein of 47–55 kDa; a rabbit monoclonal specific for human cytokeratin 19 ( ab76539 ) from Abcam; and rabbit polyclonals to ZO-1 from Invitrogen . Primary antibody binding to its target antigen was detected by chromogenic or fluorescent methods . For chromogenic detection , secondary antibodies were labeled with horseradish peroxidase ( HRP ) or alkaline phosphatase ( AP ) using ImmPRESS reagents ( Vector Laboratories ) . Chromogenic ImmPACT DAB and SG HRP substrates and ImmPACT Red AP substrate were also from Vector Laboratories . Counterstaining for chromogenic detection was performed with hematoxylin ( Vector Laboratories ) . For fluorescence detection , secondary antibodies labeled with Alexa Fluor 488 or Alexa Fluor 594 were obtained from Jackson ImmunoResearch , and diamidino-2-phenylindole ( DAPI ) counterstain was obtained from Vector Laboratories . For double immunolabeling , cells were simultaneously incubated with primary antibodies from various species , with secondary antibodies labeled with HRP or AP for chromogenic detection , and Alexa Fluor 488 or Alexa Fluor 594 for fluorescent detection . The specificity of each immunohistochemical reaction was verified with nonimmune rabbit or mouse IgG ( Vector Laboratories ) as the primary antibody . A new generation state-of-the-art ISH technology ( RNAscope ) developed by Advanced Cell Diagnostics employs a unique “double Z” probe design that greatly increases signal-to-noise ratio enabling the visualization of single transcripts . RNAscope was performed using a 2 . 0HD Reagent Kit-RED kit ( Advanced Cell Diagnostics ) according to the manufacturer's instructions . HIV-1 RNA was detected using a HIV-1-gagpol probe , which targets gag-pol coding sequence region 587–4601 . CMV DNA was detected using HHV5-IE and HHV5-pp65 probes , which target the noncoding sequence regions of CMV strain Merlin UL123 ( 172678–173852 ) and UL83 ( 120742–122152 ) respectively , all from Advanced Cell Diagnostics . Human peptidyl-prolyl cis-trans isomerase B ( PPIB ) , encoded by the PPIB gene , was detected with the Hs-PPIB probe in HeLa cell line control slides ( Advanced Cell Diagnostics ) and served as an RNAscope positive control . The RNAscope assay was followed by standard IHC with chromogenic detection . Tissue analyses were performed blinded to the clinical information and required examination of at least 5 tissue sections . ISH and IHC tissue sections and cells were analyzed with a Leica DM6000 B bright field/fluorescence microscope equipped with a Leica DFC 500 camera . Images were acquired with Leica LAS v4 . 3 software and analyzed with ImageJ software . CMV-infected and mock-infected culture media from polarized HCoEpiC were collected from both the upper and lower chambers of the transwells at various times after inoculation , centrifuged to remove cell debris , aliquoted , and stored at –70°C . Concentrations of TNF-α , IL-6 , and IL-1β in the culture media and plasma samples were measured by Quantikine ELISA ( R&D Systems ) . Statistical analysis was performed by GraphPad Prism version 5 . All values are expressed as the mean ± standard error of the mean ( SEM ) . Assay results were compared using the nonparametric Mann Whitney U-test and chi-square test , and differences were considered significant if P<0 . 05 . | Intestinal epithelial barrier dysfunction is a well-known consequence of HIV infection that persists in spite of ART . The underlying mechanisms by which HIV perturbs intestinal epithelial junctions remain unclear , and the impact of opportunistic viral pathogens in the gut has not been fully appreciated . HIV-infected individuals are almost universally coinfected with CMV . While ART has resulted in a dramatic decline in the occurrence of end-organ CMV diseases , CMV remains an independent contributor to systemic inflammation in HIV-infected people . In our analysis of rectosigmoid biopsies from CMV/HIV-coinfected individuals , we found active CMV replication associated with intestinal damage in the gut of ART-suppressed HIV-infected individuals with no symptoms of CMV disease . We demonstrated that CMV productively infects intestinal epithelial cells and , independent of HIV , disrupts their tight junctions and compromises epithelial barrier function . Furthermore , the CMV-induced proinflammatory cytokine IL-6 is a key factor in this process , and attenuation of CMV replication by letermovir , a new anti-CMV agent currently in clinical development , was sufficient to prevent CMV-induced loss of epithelial integrity . Our data highlight the role of CMV as a cofactor in intestinal epithelial barrier dysfunction in asymptomatic HIV infection and suggest a novel treatment strategy to prevent intestinal epithelial barrier dysfunction and inflammation in HIV infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"physiology",
"hiv",
"infections",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"biopsy",
"antiviral",
"therapy",
"pathogens",
"immunology",
"junctional",
"complexes",
"microbiology",
"surgical",
"and",
"invasive",
"medi... | 2017 | Replication of CMV in the gut of HIV-infected individuals and epithelial barrier dysfunction |
Membraneless organelles important to intracellular compartmentalization have recently been shown to comprise assemblies of proteins which undergo liquid-liquid phase separation ( LLPS ) . However , many proteins involved in this phase separation are at least partially disordered . The molecular mechanism and the sequence determinants of this process are challenging to determine experimentally owing to the disordered nature of the assemblies , motivating the use of theoretical and simulation methods . This work advances a computational framework for conducting simulations of LLPS with residue-level detail , and allows for the determination of phase diagrams and coexistence densities of proteins in the two phases . The model includes a short-range contact potential as well as a simplified treatment of electrostatic energy . Interaction parameters are optimized against experimentally determined radius of gyration data for multiple unfolded or intrinsically disordered proteins ( IDPs ) . These models are applied to two systems which undergo LLPS: the low complexity domain of the RNA-binding protein FUS and the DEAD-box helicase protein LAF-1 . We develop a novel simulation method to determine thermodynamic phase diagrams as a function of the total protein concentration and temperature . We show that the model is capable of capturing qualitative changes in the phase diagram due to phosphomimetic mutations of FUS and to the presence or absence of the large folded domain in LAF-1 . We also explore the effects of chain-length , or multivalency , on the phase diagram , and obtain results consistent with Flory-Huggins theory for polymers . Most importantly , the methodology presented here is flexible so that it can be easily extended to other pair potentials , be used with other enhanced sampling methods , and may incorporate additional features for biological systems of interest .
Intracellular compartmentalization is essential for normal physiological activity . This is commonly accomplished through isolation by lipid membranes or vesicles , but can also be achieved without the use of a membrane via membraneless organelles [1–3] . These organelles include processing bodies [4] , stress granules [3 , 5–7] and germ granules [8 , 9] in the cytoplasm , and nucleoli [10] and nuclear speckles [11] in the nucleus . It has recently been established that many of these membraneless organelles can be described as phase separated liquid-like droplets [8 , 12] . The process of liquid-liquid phase separation ( LLPS ) allows these organelles to spontaneously coalesce and disperse , and is important for many biological functions , such as response to heat shock and other forms of stress [6 , 13 , 14] , DNA repair [15 , 16] , regulation of gene expression [17 , 18] , cellular signaling [3 , 19] , and many other functions requiring spatial organization and biochemical regulation [10 , 20–22] . LLPS has also been implicated as a precursor to the formation of hydrogels [23] and fibrillar aggregates [7 , 15] , suggesting possible relevance to the pathogenesis of many diseases including Amyotrophic Lateral Sclerosis ( ALS ) and Frontotemporal Dementia ( FTD ) [15 , 24] . Experimental studies have characterized different properties of biological LLPS , and have shown that many systems share several common characteristics . First , the formation and dissolution processes can be tuned by the cellular environment such as changes in temperature , pH and salt concentration [25] , by post-translational modification such as phosphorylation [19 , 26] , and by mixing with other biomolecules such as proteins [27] , RNA [28–30] , and ATP [2 , 30] . Second , the concentrated phase has liquid-like properties , including fusion , dripping , wetting [25] and Ostwald ripening [28] , and its viscosity is typically several orders of magnitude higher than that of water [2 , 8 , 25] . Third , LLPS is commonly driven or modulated by low complexity ( LC ) intrinsically disordered regions ( IDRs ) of the protein sequence [6 , 25 , 31] , suggesting similarities to the well-characterized LLPS of polymer mixtures [32] . It should be noted that a disordered domain is not necessary for LLPS to occur [14] , and indeed LLPS is known to occur for folded proteins during crystallization or purification [33] . Folded domains along with IDRs have also been shown to modulate LLPS properties [34] . Lastly , some proteins involved in LLPS process are also able to form fibril structures [7 , 15] , suggesting a possible connection between the liquid-like droplet and solid fibril states . However , the molecular level understanding of LLPS cannot be easily obtained by experimental methods due to the difficulty of obtaining structural properties even in the concentrated phase [6] , and the cumbersome process of screening mutations [35] . A number of recent theoretical and simulation studies have addressed protein phase separation . Jacobs and Frenkel used Monte Carlo simulations to study multiple-component phase separation and found that the phase boundary is very sensitive to intermolecular interactions , but less dependent on the number of components in the system [36] . Lin and Chan applied the random phase approximation to treat electrostatic interactions [37] and Flory-Huggins theory for mixing entropy and other interactions . They were able to capture the sequence specificity of charged amino acids and found that the dependence of the phase boundary of the IDP Ddx4 on salt concentration can be explained by considering only electrostatic screening in their model [38] . It was also found that the monomer radius of gyration ( Rg ) is correlated with the corresponding critical temperature in both theoretical work [39] and experiment [14] . This supports the hypothesis that fundamental polymer physics principles can be used to understand LLPS [40] . However , a computational framework capable of capturing the general sequence specificity including both hydrophobic and electrostatic interactions and molecular details on both intra- and inter-molecular interactions is still missing . All-atom simulation has the potential of fulfilling both tasks [41 , 42] with the use of force fields suitable for intrinsically disordered proteins ( IDPs ) [43 , 44] . Such a force field has been recently applied to study the monomer properties of TDP-43 which is known to undergo LLPS [45] . However , computational efficiency imposes limits on the use of all-atom representation for simulating LLPS directly . Even the use of coarse-grained simulations requires well-designed sampling methods to overcome the enthalpy gap between the two phases [46 , 47] . In this work , we introduce a general computational framework for studying LLPS , combining a residue based potential capable of capturing the sequence specific interactions and the slab simulation method capable of achieving convergence for phase transition properties including critical temperature , and protein concentration in dilute and concentrated phases . To demonstrate the capabilities of the model , we have selected two model proteins: the LC domain of RNA-binding protein , Fused in Sarcoma ( FUS ) , and the DEAD-box helicase protein , LAF-1 , both of which are able to phase separate in vitro and in vivo [6 , 15 , 25] . Mutations of FUS have been shown to be highly relevant to the pathogenesis of ALS [48 , 49] and display the ability to alter the kinetics of both droplet formation and aggregation into fibrils [15] . In addition , both the full length and disordered domain of LAF-1 phase separate in vitro [25] , allowing us to explore the impact of a large , rigid domain on the LLPS behavior . The manuscript is organized as follows . First , we introduce our computational framework including the coarse-grained potential , the sampling method and the treatment of folded proteins in the simulations . We then present the application of the method using two model systems . For the first system , we show the comparison of phase diagrams for wild-type ( WT ) FUS and a set of mutants , and that they are qualitatively consistent with recent experimental measurements . For the second , we demonstrate how inclusion of the folded domain alters the LAF-1 phase diagram . In both FUS and LAF-1 , we show the flexibility of the framework by providing the results for two different coarse-grained potentials . Lastly , we investigate the phase diagram dependence on chain length , closely related to the “multivalency” effect often discussed in the context of LLPS .
All-atom simulations are unable to reach the time scales needed to study phase separation with current state-of-the-art computational hardware resources and sampling methods . We therefore introduce a coarse-grained representation of the protein , in which each residue is represented as a single particle ( Fig 1a ) . The model takes into account the chemical properties of the 20 naturally occurring amino acids , listed in S1 Table , thus making it sequence specific . The potential energy function contains bonded , electrostatic , and short-range pairwise interaction terms . Bonded interactions are modelled by a harmonic potential with a spring constant of 10 kJ/Å2 and a bond length of 3 . 8 Å . Electrostatic interactions are modeled using a Coulombic term with Debye-Hückel [50] electrostatic screening to account for salt concentration , having the functional form: E i j ( r ) = q i q j 4 π D r exp ( - r / κ ) , ( 1 ) in which κ is the Debye screening length and D = 80 , the dielectric constant of the solvent medium ( water ) . For all the simulations for which phase diagrams are generated , a Debye screening length of 1 nm , corresponding to an ionic strength of approximately 100 mM , is used . When determining Rg for IDPs from the literature , ionic strength is set to match that from the experimental results , as listed in ( S2 Table ) . The short-range pairwise potential accounts for both protein-protein and protein-solvent interactions . Here we have introduced two different models: the first is based on amino acid hydrophobicity [51 , 52] and uses a functional form introduced by Ashbaugh and Hatch [53]; the second is based on the Miyazawa-Jerningan potential [54] with the parameterized functional form taken from Kim and Hummer [55] .
As a first application of our model to LLPS , we use the prion-like LC domain of the protein FUS ( FUS-LC ) which is sufficient to induce LLPS in vitro in the absence of other biomolecules [15] . FUS-LC is an ideal system to test our model as it is fully disordered and displays very low secondary structure content [6] . The sequence is largely uncharged , with only 2 anionic aspartate residues within its 163 amino acid sequence . To test for sequence-specific effects , we conducted simulations for several different variants of the FUS-LC peptide , wild-type and four phosphomimetic mutants where a set of the 12 naturally phosphorylated threonine or serine residues are mutated to glutamate . [78] The first of these mutants is the 12E mutant , which contains all 12 glutamate substitutions , and does not undergo LLPS under similar conditions to FUS WT [79] . We additionally test the 6E mutant reported in the same work [79] , and two designed variations of the 6E mutant , termed 6E’ and 6E* which maximize and minimize , respectively , the clustering of charged residues within the sequence under the constraints of preserving the amino acid composition of 6E , and only mutating naturally occurring phosphorylation sites . [80] Utilizing the slab method , we determine the range of temperatures at which the simulated FUS chains separate into two phases , and calculate the coexistence curve using both the HPS and KH models . The concentration of the dilute phase gives the predicted critical/saturation concentration of the protein , the concentration above which it will begin to form droplets in solution . The concentration of the dilute phase is on the order of 0 . 1-10 mg/mL over the tested temperature range , consistent with typical concentrations used to observe phase separation of FUS WT in vitro [6] ( ∼1-5 mg/mL ) . We find that the critical temperature differs between the two models for FUS WT . However the coexistence curves and the phase diagrams are qualitatively similar ( S8a Fig ) , as are the intermolecular contact maps ( S9 Fig ) . To evaluate the impact of the phosphomimetic mutations , we determine the phase diagram for FUS WT , 6E , 6E’ , 6E* , and 12E using the HPS model ( Fig 5 ) . The 12E mutant phase separates at a much lower temperature , with the critical temperature smaller than even the lowest temperature at which we can observe coexistence between two phases for FUS WT ( due to the prohibitively small concentration of the low-density phase ) . This is consistent with the experimental observation that FUS 12E does not phase separate in contrast to FUS WT at similar conditions [79] . The 6E mutants all lie between the two extreme cases , and have nearly identical phase diagrams . While the difference of just 6 amino acids results in a greatly altered phase-separation ability from wild type , the rearrangement of these mutations does not have such an effect . However , these mutations were done under very strict constraints which do not allow for much change in the degree of charge clustering . We also calculate the inter-chain contacts , defined as two amino acids of different chains within 21/6 σij of each other . There are no specific contacts formed in either of the cases ( S9 Fig ) , suggesting that LLPS of FUS WT is not driven by a specific region within the protein sequence . However , when comparing the different 6E mutants at the same temperature , the degree to which different regions of the peptide interact are greatly affected ( S10 Fig ) . This shows that despite having nearly identical phase diagrams , the interactions involved in phase separation can vary . The average interaction strength per residue χ can also be obtained by fitting the phase diagram to Flory-Huggins theory [72 , 73] as shown in S11 Fig . We find that there is a clear decreasing trend of χ from 0 . 410 to 0 . 325 kcal/mol with increasing number of phosphomimetic mutations ( S6 Table ) . To further check the liquid-like nature of the concentrated phase , we calculate the mean squared displacement ( MSD ) as a function of time using NVT simulations for WT , 6E and 12E at 500mg/mL ( S12 Fig , S3 Movie ) . For each , there is a linear region with non-zero slope suggesting that the concentrated phase is liquid-like , and not a solid aggregate . The diffusion coefficient from fitting the linear region is ∼3x10−6 cm2/s , three orders of magnitude larger than measured in the experiment ( 4x10−9 cm2/s [6] ) as can be expected from a coarse-grained simulation and as we are using low friction Langevin dynamics . Finally , we check monomer radius of gyration in both the dilute and concentrated phase and find that chains in the concentrated phase are generally more extended than those in the dilute phase ( S13 Fig ) . Next , we apply our model to DEAD-box helicase protein , LAF-1 , which has been shown to phase separate as both its IDR and as full length , including a 437-residue folded domain , in vitro [25] . To test the effect of inclusion of folded domains , three variants of LAF-1 sequences have been simulated , including the N-terminal IDR of LAF-1 , the helicase domain , and full length LAF-1 with both the IDR and folded domain as well as the prion-like C-terminal domain which is also disordered . The IDR sequence is of similar length to FUS , but contains a larger fraction of charged amino acids , ( ∼26% ) compared to FUS WT ( ∼1% ) , and FUS 12E ( ∼9% ) , and includes both attractive and repulsive electrostatic interactions . For LAF-1 IDR , we simulated the phase diagram with both KH and HPS models . As was the case for FUS WT , the phase diagrams are qualitatively similar between the two models ( S8b Fig ) . In Fig 6 we compare the phase diagrams of the full length and IDR regions of LAF-1 . The phase diagram for the full length protein is shifted toward higher temperatures , and suggests a smaller saturation concentration as compared to the LAF-1 IDR alone at the same temperature . The results for the helicase domain alone also clearly show phase separation ( S14 Fig ) . The experimental phase boundary in ∼120 mM NaCl is ∼0 . 05 mg/mL for full length LAF-1 , but ∼0 . 4 mg/mL for the isolated IDR [25] . Even though we cannot accurately estimate the low protein concentrations in the dilute phase so as to quantitatively compare with the experimental values , we do see an increase in the saturation concentration when adding the folded domain as has been seen by experiment . We note that the concentrations obtained from the high density phase are much higher than recently estimated by Wei et al . [81] , however , they are quite comparable with those measured by Brady et al . for the similar DEAD-Box Helicase protein Ddx4 [74] . Fitting the phase diagram to Flory-Huggins theory , we obtain the average interaction strength per residue , χ , of LAF-1 IDR ( S8 Fig ) . The χ is 0 . 270 kcal/mol for KH model and is 0 . 298 kcal/mol for HPS model , both comparable to 0 . 3 kcal/mol obtained from experimental Ddx4 data [74] . The reason for the change of critical temperature upon inclusion of the folded domain is likely two-fold . First , the folded domain contains more hydrophobic residues with an average hydrophobicity of 0 . 664 ( 0 . 579 for the surface residues ) in contrast to 0 . 520 for LAF-1 IDR ( S1 Table ) , therefore strengthening the intermolecular attraction . In addition , providing more interaction sites per chain generally favors a higher critical temperature , because more interactions can be formed with a smaller loss of entropy , an effect commonly referred to as multivalency [35] . The impact of multivalency on the phase coexistence will be investigated explicitly in the next section . In the concentrated phase , we also investigate the intermolecular contacts in Fig 7 . Unlike the case of FUS , there are regions along the sequence where there is a relatively high propensity to form contacts , ( residue 21 to 28 , RYVPPHLR ) and ( residue 13 to 18 , NAALNR ) . These regions are present in both the IDR with the KH and HPS model ( Fig 7a and 7b ) and in the full length protein ( Fig 7c ) . The central region of these two segments is composed of uncharged amino acids , suggesting the importance of hydrophobic patches in the sequence even with a large fraction of charged residues . As is shown in both 1D and 2D contact maps ( Fig 7a , 7c and 7d ) , the pattern , and number of contacts within the IDR look similar in both the IDR and the full length LAF-1 simulations . This observation also applies to the helicase domain in the helicase only , and full length LAF-1 simulations ( S16 Fig ) . This suggests that the key residues contributing to the droplet formation are the same for the disordered peptide with and without the folded domain ( Fig 7d ) . Additionally , the disordered part of the protein ( including both the N-terminal and C-terminal disordered regions ) contributes more contacts than the folded domain in the simulation of full length LAF-1 , consistent with the experimental observations that the disordered region of LAF-1 is the driving force for the LLPS [25] . The intramolecular contact map in the two phases ( S17 Fig ) supports the change of Rg ( S13 Fig ) in that the peptide has fewer long range contacts in the concentrated phase than in the dilute phase . We additionally calculate the mean squared displacement ( MSD ) as a function of time for all the three variants of LAF-1 ( i . e . , IDR , helicase and full length ) using NVT simulations at concentrations predicted for the condensed phase at 210K ( S4 , S5 and S6 Movies ) , to see how the different regions affect the diffusion of the protein within the concentrated phase . There is a linear region with non-zero slope for all the variants ( S12 Fig ) suggesting liquid-like behavior . The IDR has a much larger diffusion coefficient than both the full length and the helicase domain of LAF-1 making it the most mobile of the three . This is likely due to its flexibility as well as the lower concentration . The diffusion coefficient for full length LAF-1 is an order of magnitude larger than that of just the helicase domain , further supporting the importance of the flexible region for maintaining liquid-like behavior of proteins inside the droplet . Multivalency has shown to be important in driving LLPS in experimental studies [20 , 35] where proteins with a higher number of repeated units begin to form droplets at lower concentrations . Usually multivalency is used to describe a certain number of specific interaction sites per molecule . For polymers , there is inherently a large number of possible interactions between molecules , so for well-mixed sequences specific residue-residue interactions are less likely to play a role in assembly . Nonetheless , increasing the chain length will ( for a given sequence composition ) increase the number of available interaction sites per chain , and thus , increase multivalency of the system . In order to investigate the mechanism of such behavior , we use a model system where we take the first 40 residues from FUS LC and make several repeated units in the form of [FUS40]n , in which n = 1 , 2 , 3 , 4 and 5 . We then conduct multiple slab simulations for each of these sequences , keeping the total number of atoms constant ( see detailed system size in S6 Table ) . The phase diagrams of [FUS40]n in Fig 8a show that the phase boundary shifts to higher temperatures and lower concentrations with increasing chain length . To understand the mechanism of such dependence , we apply Flory-Huggins theory [72 , 73] , which has previously been used to understand IDP phase separation [9 , 37 , 38 , 74] , to fit the phase transition properties obtained by molecular dynamics simulations when varying the chain length N . If we assume that each solvent molecule occupies one lattice position , we can fit all five phase diagrams from different chain lengths to the binodal of Flory-Huggins theory using the same average interaction strength per residue χ and protein density ρ ( Fig 8a and S6 Table ) . Since there is analytic solution for the critical temperature and concentration from Flory-Huggins theory: the critical temperature T c ∝ N / ( N + 1 ) 2 and the critical concentration ρ c ∝ 1 / ( N + 1 ) , we can also fit our simulated Tc and ρc as a function of the chain length with these approximate equations ( assuming prefactor as the fitting parameter ) , as shown in Fig 8b and 8c . These results suggest that the phase diagram dependence on the chain length can be described by Flory-Huggins theory . The term that is sensitive to changes in chain length is the mixing entropy per segment . With increasing the chain length , the mixing entropy per segment decreases , and therefore the critical temperature increases . It would then be easier to observe LLPS with a longer chain at the same temperature , in the sense that the dilute-phase concentration is smaller , consistent with experimental observations [20] . This factor should be considered when making mutations to protein sequences with the aim of understanding the molecular origin of LLPS: in general , chain truncation or extension will disfavor or favor LLPS , respectively , regardless of the sequence-specific effects . Similarly , when cutting a larger protein into fragments in order to evaluate the contribution of each to driving LLPS , it is expected in general that longer fragments will be able to phase separate at a higher temperature .
We have introduced a general framework for conducting molecular dynamics simulations of LLPS leading to protein assemblies constituting many membraneless organelles . Coarse-graining to amino-acid-resolution gives access to length and time scales needed to observe this phenomenon , and to achieve convergence of thermodynamic observables ( i . e . , phase diagram , critical temperature and protein concentration in the dilute and concentrated phases ) while preserving sequence-level information , thus allowing observance of changes induced by mutations to the protein sequence . The force fields utilized in this work are based on previously determined , knowledge-based potentials , parameterized to accurately represent the radius of gyration of disordered proteins , but the framework is also flexible to incorporate other residue-based pairwise interaction potentials . The two force fields generate similar intermolecular contact maps within the concentrated phase , suggesting that description of the weak nonspecific interactions in IDPs can be captured easily with different models . We have tested the framework and the two force fields with two model systems , which undergo phase separation in vitro , yielding phase diagrams , thus giving the critical temperature , and saturation concentration at the tested temperatures . Despite the simplicity of the currently used potentials , and the fact that they were exclusively optimized based on the properties of monomeric proteins , we demonstrate the ability to predict how various perturbations to the system can change the LLPS . In the case of FUS LC , the model is able to capture the experimentally observed variation of phase diagram when introducing mutations . In LAF-1 , the model is able to capture the experimentally observed difference between the phase separation of full length and truncated disordered-only sequences . We also show that the inclusion of the disordered parts function to increase the diffusion of LAF-1 within the condensed phase . We have also investigated an important feature of LLPS regarding the dependence of phase behavior on chain length , which is well established in polymer physics and was previously observed in experiment [20] . We show that there is an upward shift in the phase diagram ( temperature-concentration ) with increasing chain length . At a given temperature , the saturation concentration will be higher for shorter chain lengths . Both the critical temperature and concentration are in good agreement with Flory-Huggins theory and therefore suggest the behavior can be explained by relative loss of entropy . With this in mind , if the phase behavior of a protein of interest cannot be observed in vitro , making repeated units might be a convenient way to shift the phase diagram enough that LLPS will be observable under more reasonable experimental conditions . One must also consider this effect when making changes to protein length , such as His tags , or cleavage of a certain section of residues , and how just the change in chain length may affect the coexistence . Additionally , we measure certain important properties of proteins within the concentrated phase for the two model systems such as intermolecular contact propensities , which are quite difficult to resolve experimentally . With FUS LC , the intermolecular contacts are evenly distributed throughout the length of the peptide , suggesting that non-specific hydrophobic interactions are largely responsible for driving the phase-separation . For LAF-1 , we observe enhanced intermolecular contacts within a specific region ( residue 21-28 ) , largely composed of hydrophobic amino acids , suggesting that even though LAF-1 contains 26% charged residues , hydrophobic interactions are still an important driving force for LLPS . There are some features that cannot be captured in the presented model , but can be added in future work . First the absolute temperature of the simulation is not comparable to experiment . The phase behavior at the lower critical solution temperature , which is observed in some disordered peptides experimentally [31] , cannot be captured , either . Both require the addition of a temperature dependent solvation energy term into the framework , and more comparison with experimental Rg data ( or other relevant data ) . Second , we have not fully tested the ionic strength dependence of the current model because of the breakdown of Debye-Hückle electrostatic screening at high ionic strength , even though the trend of LAF-1 experiments when varying salt concentration is captured in the current model . However , we do not see any ionic strength dependence for FUS LC , which is inconsistent with experiment [6] . To capture salt dependence in proteins with negligible charged amino acid content , it may be necessary to include a description of “salting-out” effects , i . e . , the change of solubility with salt concentration as captured by the Hofmeister series . In S19 Fig , we show that the literature-known amino acid specific salting-out coefficients [82–85] are strongly correlated with the hydrophobicity scale and therefore it may be possible to model the salting-out effect with an additional energy term using the same hydrophobicity scale . In the future , we would also like to introduce additional handles ( such as a structure-based potential for intramolecular interactions ) to allow for conformational changes within the folded parts of a chain . This will allow us to study LLPS of proteins with small populations of folded regions that are important for self-assembly . | Liquid liquid phase separation ( LLPS ) of low-complexity protein sequences has emerged as an important research topic due to its relevance to membraneless organelles and intracellular compartmentalization . However a molecular level understanding of LLPS cannot be easily obtained by experimental methods due to difficulty of determining structural properties of phase separated protein assemblies , and of choosing appropriate mutations . Here we advance a coarse-grained computational framework for accessing the long time scale phase separation process and for obtaining molecular details of LLPS , in conjunction with state of the art enhanced sampling methods . We are able to qualitatively capture the changes of the phase diagram due to specific mutations , inclusion of a folded domain , and variation of chain length . The model is flexible and can be used with different knowledge-based potential energy functions , as we demonstrate . We expect a wide application of the presented framework for advancing our understanding of the formation of liquid-like protein assemblies . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"sequencing",
"techniques",
"protein",
"interactions",
"enzymes",
"electricity",
"enzymology",
"simulation",
"and",
"modeling",
"electrostatics",
"phase",
"diagrams",
"molecular",
"biology",
"techniques",
"intrinsically",
"disordered",
"proteins",
"research",
"and",
"analys... | 2018 | Sequence determinants of protein phase behavior from a coarse-grained model |
The Kaposi sarcoma associated herpesvirus ( KSHV ) genome encodes more than 85 open reading frames ( ORFs ) . Serological evaluation of KSHV infection now generally relies on reactivity to just one latent and/or one lytic protein ( commonly ORF73 and K8 . 1 ) . Most of the other polypeptides encoded by the virus have unknown antigenic profiles . We have systematically expressed and purified products from 72 KSHV ORFs in recombinant systems and analyzed seroreactivity in US patients with KSHV-associated malignancies , and US blood donors ( low KSHV seroprevalence population ) . We identified several KSHV proteins ( ORF38 , ORF61 , ORF59 and K5 ) that elicited significant responses in individuals with KSHV-associated diseases . In these patients , patterns of reactivity were heterogeneous; however , HIV infection appeared to be associated with breadth and intensity of serological responses . Improved antigenic characterization of additional ORFs may increase the sensitivity of serologic assays , lead to more rapid progresses in understanding immune responses to KSHV , and allow for better comprehension of the natural history of KSHV infection . To this end , we have developed a bead-based multiplex assay detecting antibodies to six KSHV antigens .
Kaposi sarcoma-associated herpesvirus ( KSHV ) is the causative agent of Kaposi sarcoma ( KS ) , primary effusion lymphoma ( PEL ) and a type of multicentric Castleman's disease ( MCD ) [1]–[3] . Unlike other human herpesviruses , KSHV is not ubiquitous in human populations . The prevalence of KSHV infection generally parallels the incidence of KS , and varies strikingly according to geography , ethnicity , and certain behavioral risk factors [4] . Prevalence is very high in sub-Saharan Africa , ranging from 35 to 60% [4]–[7] and elevated in Mediterranean regions , from 10 to 30% [8] , [9] . In South America , prevalence is high in Amerindians but not in non-Amerindians living in adjoining areas in comparable conditions [10] , [11] . In the US and Western Europe , prevalence is generally low but is elevated in men who have sex with men ( MSM ) [12] , [13] and in those born in certain areas of elevated KSHV prevalence [14] . These observations are built on more than fifteen years of sero-epidemiological studies based on various assays for detecting KSHV antibodies [15] . Commonly used tests include immunofluorescence assays ( IFA ) using either latent or lytic PEL cell lines [16] and ELISAs based on a small number of recombinant antigens or peptides [17] . We have developed and extensively used ELISAs based on recombinant K8 . 1 ( a lytic gene ) and ORF73 ( a latent gene ) [18] . More recently , luciferase immunoprecipitation system ( LIPS ) technology had been applied to KSHV serology with considerable success [19] , [20] . While current assays are useful and reasonably reliable for understanding KSHV epidemiology , none have been specifically developed for diagnosis of KSHV infection in individual subjects , particularly asymptomatic persons [21]–[23] . In our epidemiological studies we usually classify subjects as seropositive when they are reactive to either ORF73 or K8 . 1 . We have shown that a subject may be reactive to ORF73 for many years prior to showing reactivity to K8 . 1 and vice versa . The pattern of reactivity is not predictable [24] , [25] . Longitudinal studies utilizing other assays have also demonstrated that seroreactivity to individual antigens fluctuates over time [26]–[28] , suggesting that evaluation of additional KSHV antigens may increase test sensitivity . The KSHV genome encodes more than 85 proteins , and most have yet to be studied in terms of seroreactivity . A recent study described KSHV and EBV protein arrays , but very few subjects were examined using this novel tool [29] . More sophisticated serologic assays are necessary to further investigate the natural history of KSHV infection and KSHV-associated malignancies . In order to address these issues , we have expressed 72 KSHV ORFs and used the purified recombinant proteins to systematically screen for serological reactivity by ELISA . We tested individuals with either previous diagnoses of KSHV-associated diseases , or a low likelihood of KSHV infection . Such systematic analysis of the KSHV proteome allows a more complete investigation of serologic responses arising during asymptomatic KSHV infection and KSHV associated diseases . This will facilitate a better characterization of the natural history of KSHV infection and the immune responses to the virus , and may provide possible clues to further investigate disease pathogenesis . Improvement of KSHV serodiagnosis would be an immediately applicable outcome of this process .
Eighty two individuals were identified from two cohorts of subjects ( Table 1 ) . Forty-three healthy donors were chosen from amongst participants of the Research Donor Program of the Occupational Health Service at the Frederick National Laboratory for Cancer Research , Frederick , MD ( RDP group ) . Healthy US blood donors have a low ( <5% ) seroprevalence of KSHV infection [30] , and subjects in this group were therefore considered KSHV uninfected; moreover , all RDP subjects are HIV seronegative . Thirty nine KSHV infected subjects with pathologically confirmed KSHV-associated malignancies were selected from HIV and AIDS Malignancy Branch ( HAMB ) protocols at the National Cancer Institute , Bethesda , MD , including 36 HIV-infected and 3 HIV-uninfected subjects ( HAMB group ) . The group included 25 patients with KS , 19 with a history of MCD ( not symptomatic at time of evaluation ) , and 3 with PEL treated with immunochemotherapy; some subjects had more than one KSHV-associated malignancy . Samples were collected June 2010–March 2011 from HAMB subjects and February 2010–December 2010 from RDP subjects . All subjects were enrolled on study protocols approved by the relevant National Cancer Institute Institutional Review Board . All patients gave written informed consent in accordance with the Declaration of Helsinki . The assay was carried on as previously described [18] with several modifications , as detailed below . Either plasma or serum samples were used; assay characteristics and performances had been previously determined to be equivalent in testing the two type of specimen ( data not shown ) . KSHV-viral load was measured using previously described methods [33] . Briefly , DNA was extracted from peripheral blood mononuclear cells ( PBMCs ) and KSHV DNA was detected using primers for the K6 gene region . The number of cellular equivalents was determined using a quantitative assay for human endogenous retrovirus; KSHV viral load ( VL ) was reported as viral DNA copies per million PBMCs . Antigens were covalently attached to Bio-Plex Pro Magnetic COOH beads ( Biorad , Hercules , CA ) via a sulfo-N-hydroxysulfosuccinimide mediated ester according to the manufacturer's protocol . To each well were added 2500 beads in 50 µL assay buffer and 50 µL of diluted serum ( 1∶50 , 1∶100 or 1∶200 ) , which were incubated for 1 hour at room temperature and washed . The secondary antibody , a goat F ( Ab′ ) 2 Anti-Human IgG ( γ ) , R-PE conjugate , ( Life Technologies , Grand Island , NY ) was then incubated for 30 minutes; samples were washed , resuspended in 100 µL of assay buffer and analyzed on the Bio-Plex 200 system ( Biorad , Hercules , CA ) . The median fluorescence intensity ( MFI ) across all counted beads was computed for each sample , and recorded after subtracting the background fluorescence . Seroreactivity was evaluated by comparing median ODs and MFIs using unpaired Mann-Whitney test between two groups; p-values of the U statistic were computed asymptotically ( 100 permutations ) and Benjamini-Hochberg ( B–H ) corrected for multiple comparisons . For receiver operator characteristic ( ROC ) comparisons , the area under the curve ( AUC ) was computed using the trapezoidal rule , asymptotic normal p-values were Bonferroni corrected for multiple comparisons . ODs comparisons were performed using GeneSpring v11 . 5 ( Agilent Technologies , Santa Clara , CA ) . All other statistical analyses were performed using Prism v6 ( GraphPad , La Jolla , CA ) or STATA v11 . 2 ( Statacorp , College Station , TX ) .
Seventy three recombinant His6-MBP fusion proteins from 72 of the KSHV ORFs were expressed and purified ( Table S1 ) . The recombinant proteins included 69 expressed in baculovirus and 15 expressed in E Coli; 26 were expressed in both . The variable genes K1 and K15 were not cloned as part of this study . Of the 12 remaining ORFs , 4 were not successfully expressed and 8 could not be purified intact or in sufficient amount . To assess the antigenicity of the recombinant proteins generated , we measured antibody responses by ELISA in a panel of specimens from our KSHV-infected subjects ( HAMB ) and healthy blood donors from the Frederick National Laboratory for Cancer Research ( FNLCR ) Research Donor Program donor ( RDP ) groups . Subject characteristics are noted in Table 1 . Seroreactivity against KSHV encoded antigens are presented as a heatmap in Figure 1 , where each recombinant protein is shown in a column and each subject in a row . The optical density ( OD ) is represented by the colour intensity . Panel A shows reactivity in subjects with documented KSHV disease ( HAMB subjects ) and panel B shows reactivity in healthy donors ( RDP subjects ) . Intensity and breadth of reactivity was substantially greater in samples from HAMB subjects . One of the healthy donors ( RDP1 ) showed reactivity to K8 . 1 and ORF73 , as well as to 3 other antigens , consistent with KSHV infection . This observation is in accordance with estimates of KSHV prevalence in US blood donors [30] . To determine which responses differentiated infected and uninfected individuals , the ratio of median reactivity in HAMB and RDP subjects was calculated for each antigen . A comparison between median ODs in HAMB vs . RDP samples is shown as a volcano plot in Figure 2 . The values obtained are indicated on the horizontal axis , whereas the Benjamini-Hochberg ( B–H ) corrected p-value of the test statistic for the difference in median reactivity between the two groups is shown on the vertical axis . Eight proteins showed a median OD ratio greater than 1 . 5 with a corrected p-value lower than 0 . 05 . These were ORF73 and K8 . 1 ( and the isoform K8 . 1B ) , ORF65 , ORF38 , ORF61 , ORF59 and K5 . The pattern of seroreactivity observed in subjects with documented KSHV disease was highly variable . Subjects showed reactivity to as few as 2 to greater than 15 antigens . The intensity of the responses , as indicated by OD measurements , was also variable for individual antigens and subjects . Using individual OD measurements , we constructed variables summarizing the breath and the intensity of the serologic reactivity in each HAMB subject . For each antigen , OD values greater than those in the upper 5% of the distribution in the RDP donors were considered indicative of a significant reactivity . The number of such responses , and a measure of average reactivity ( mean OD ) was recorded for each subject . As expected from previous reports , responses to K8 . 1 , ORF73 , and ORF65 , were observed in a majority of HAMB subjects ( 81–100% ) ; responses to K8 . 1 and to the K8 . 1B splice variant [34] were quite similar . Reactivity to ORF38 , ORF61 ORF59 and K5 was also observed in a high proportion of subjects ( 74–86% ) . Some proteins , like ORF23 and ORF32 , did not appear to be antigenic in this system , or in these HAMB subjects , and ORF57 and K2 appeared to elicit responses in both HAMB and RDP subjects , suggesting that such reactivity is non-specific . To explore the association between serologic responses and KSHV DNA detection in patients with a history of KSHV associated malignancies , KSHV viral load ( VL ) in peripheral blood mononuclear cells ( PBMCs ) was determined in HAMB subjects at the time of sampling for serology; 7 ( 26% ) had detectable KSHV VL . Additionally , in 35 of the subjects , KSHV VL had also been measured longitudinally prior to sampling . The number of measurements varied between 3 and 25 ( mean 7 . 7 , Standard Deviation 6 . 12 ) and the length of retrospective follow up varied between 77 days and 13 years ( mean 1615 , SD 1356 days ) . Thirty-three subjects ( 87% ) had detectable KSHV during at least one visit . Neither detectable KSHV VL at the time of serum collection nor history of KSHV VL ( expressed as mean VL over the entire follow up , peak VL , or ever-positive VL ) were associated with the breadth or intensity of responses . The same was observed when individual KSHV VL measurements were introduced in a longitudinal model . Results did not change when the analyses were restricted to lytic antigens , or after dropping the 3 PEL patients , who were receiving immunochemotherapy potentially affecting antibody response . . In contrast , history of HIV infection appeared to affect reactivity in subjects with documented KSHV-associated malignancies . Three HAMB subjects in this panel were HIV negative ( noted in Figure 1 ) ; they showed reactivity to fewer antigens ( 5 . 3 , 95%CI 0–11 . 6 , vs . 26 . 6 , 95%CI 21 . 1–32 . 14 ) and with lower mean ODs ( 0 . 2 , 95%CI 0 . 0–0 . 4 , vs . 0 . 6 , 95%CI 0 . 5–0 . 7 ) . In order to build upon the findings of our screening assays in a practical , translational manner , we next developed a multiplex bead-based assay that allows concurrent testing of the six antigens that demonstrated median OD ratio between RDP and HAMB subjects greater than 2: K8 . 1 , ORF73 , ORF65 , ORF38 , ORF59 and ORF61 . K8 . 1B was not included because K8 . 1 and K8 . 1B provided nearly identical data . The assay requires only one microliter of serum , and this approach has been successful for other pathogens in both epidemiological and diagnostic settings [35] , [36] . Single-plex and then multiplex assay parameters were optimized ( data not shown ) . HAMB and RDP samples were re-analyzed in multiplex for the 6 antigens . In Figure 3 , panel A shows the distributions of MFI values in the HAMB and RDP group for each antigen . Differences in reactivity between RDP and HAMB subjects were confirmed to be statistically significant using the multiplex assay ( corrected p-value less than 0 . 0001 for each comparison ) . Panel B displays the receiver operating characteristics [37] ( ROC ) for each of the antigens , demonstrating that all assays discriminate between KSHV infected subjects and healthy donors . Sensitivity and specificity were 87% and 95% for K8 . 1 , 92% and 95% ( ORF73 ) , 87% and 82% ( ORF65 ) , 72% and 95% ( ORF38 ) , 69% and 86% ( ORF61 ) , 62% and 95% ( ORF59 ) . The single KSHV infected healthy donor was retained in the RDP group for all analyses . Had the individual been excluded or included in the HAMB group , sensitivity and specificity estimates would have not changed appreciably . Analytic specificity was demonstrated by pretreating samples with the appropriate antigen or an irrelevant antigen , prior to adding antigen-conjugated beads . Specific IgG antibodies in serum bind to the soluble antigen , preventing adsorption to the antigen-conjugated beads and leading to a loss of signal ( Panel C ) . In order to determine if the signal was due to non-specific binding to the MBP purification solubility tag fused to the antigen , reactivity to ORF38 either fused to MBP or cleaved from MBP was measured; the two were shown to be equivalent ( Figure 4 ) . Analytical specificity testing , performed as detailed above , further indicated that seroreactivity derived from the specific portion of the protein , not from the solubility tag . Having ascertained that the sensitivity and specificity of the newly developed multiplex assay for each of the 6 KSHV antigens were satisfactory , we examined its quantitative features compared to those of individual ELISAs . Specifically , we evaluated the dynamic range of the new assay format , as the relatively narrow dynamic range of ELISA limits the use of ODs as a marker of antibody levels . The MFIs recorded for each antigen in the bead-based assay were plotted against the ODs obtained in the corresponding ELISA ( Figure 5 ) . For each antigen , the two sets of measurement correlated well ( Spearman's rank coefficient varied between 0 . 76 and 0 . 86 , p<0 . 00001 for all comparisons ) . The ELISA generally presented a smaller dynamic range , here defined as the ratio between the largest and smallest detected values ( mean , 61; SD , 59 vs . mean , 560; SD , 197 ) , which compresses the highest OD values . The difference is particularly evident for reactivity to KSHV capsid proteins K8 . 1 and ORF65 . These findings did not vary when the bead-based assay was repeated using a different serum dilution , confirming the quantitative robustness of the assay . These data demonstrate the greater dynamic range of the bead-based assay format , compared to ELISA , which is likely to facilitate semi-quantitative analysis of antibody data , an important feature of this assay that will be useful for future KSHV research .
In order to systematically evaluate the antigenic characteristics of KSHV-encoded proteins , we sought to express the entire KSHV proteome . We obtained recombinant proteins from 72 of the more than 85 described ORFs . The K1 and K15 genes were excluded from this study because of their high variability . Expression and/or purification of the few remaining proteins were not possible due to various technical issues . The initial ELISA screening of these recombinant proteins with a panel of well characterized subjects with KSHV-related disease and healthy donors revealed a defined pattern of reactivity . A low background was detected in the healthy donors , as compared to strong and diverse reactivity in the KSHV infected subjects . Seven antigens appeared to strongly discriminate the two populations . These were the previously described ORF73 , K8 . 1 ( including the splice variant K8 . 1b ) and ORF65 , and the newly identified ORF38 , ORF61 , ORF59 and K5 . However , individual patients showed different response patterns to the various antigens . One of the 44 healthy donors , who was seroreactive to K8 . 1 , ORF73 , ORF65 as well as to ORF38 and ORF61 , was deemed to be KSHV infected . KSHV seroprevalence in the RDP group as determined by this assay was 2 . 3% , consistent with published estimates for US blood donor populations [30] . However , source population and selection criteria for RDP subjects differ somewhat from those of transfusion products donors , and our convenience sample cannot be considered representative of any population . We demonstrate a diversity of antibody responses in subjects with documented KSHV-associated malignancies , and these findings are similar to what is observed when examining immune responses to other herpesviruses [38] , [39] , but the reasons for such diversity are not clear . It can be hypothesized that diverse responses to KSHV antigens may result from the variability in the natural history of the infection . Important factors may include age of primary infection , frequency of reactivation and level of viral replication , co-infection with HIV or other pathogens , immune status and development of particular KSHV-associated diseases . Our sample size and composition were not designed to allow us to ascertain an association between particular patterns of seroreactivity and factors such as age , gender , diagnosis of specific KSHV malignancies or disease duration . A previous study using luceriferase immunoprecipitation systems has shown different patterns of reactivity to a lytic antigen ( K8 . 1 ) and latent antigens ( LANA and v-Cyclin ) in patients with KS as compared to patients with MCD [20] , and it will be of interest to see how the multiplex assay discriminates among different KSHV disease states . Prior or concurrent KSHV viral load did not show an association with breadth or strength of antibody reactivity in this study , in contrast with what observed in other investigations , but , again , our sample was relatively small and not intended to detect such . Further evaluation of specifically designed study samples is required . Only a few HIV negative individuals were included amongst our KSHV infected subjects . Although we were able to observe an association between HIV infection and broader , stronger responses to KSHV antigens , we remain cautious regarding the generalizability of such finding . Further studies examining the clinico-pathological correlates of the diversity and intensity of antibody responses to KSHV are warranted . In particular , it will be necessary to consider both HIV infected and uninfected populations from areas of low or high endemicity , and to investigate subjects with KSHV associated diseases , as well as asymptomatic individuals . More generally , it is imperative that findings uncovered in the present discovery cohort be externally validated on independent cohorts . It is worth at this point considering the KSHV antigens that we have newly found to discriminate infected from non-infected persons and their role in the KSHV life cycle . The newly described antigens include several interesting KSHV-encoded proteins . ORF38 is a poorly characterized tegument protein . Serological responses against ORF38 were recently observed in a handful of AIDS-KS patients in a study utilizing a protein array platform [29] . ORF61 encodes the large subunit of the KSHV ribonucleotide reductase , which catalyzes the conversion of ribonucleoside diphosphates to the corresponding deoxyribonucleotides , controlling the cellular concentration of the latter . The ORF59 protein ( PF-8 ) , which is transcribed from the same polycistronic transcript as the ORF61 protein , is a processivity factor for the viral DNA polymerase [40] . Both are likely to be critical for viral replication . Evidence for seroreactivity to ORF59 had been identified in AIDS-KS patients in a small study [41] but no information is available on antibodies against ORF61 . K5 encodes a RING-CH E3 ubiquitin ligase which downregulates human tetherin expression on the cell surfaces and targets it for endosomal degradation , thereby promoting the release of viral progeny from infected cells [42]; furthermore , it is involved in multiple immune modulation mechanisms [43]–[46] . Although its role in KSHV lytic infection and transmission is likely to be significant , its antigenic properties had not been thus far described . Recent studies of immune responses in herpesvirus infections have consistently identified significant antigens that are orthologs of ORF38 , ORF61 and ORF59 . These include UL39 in HSV1 [39] and HSV2 [47] , UL45 in HCMV [38] , as well as BORF2 and BMFR1 in EBV [29] . Further investigations on the humoral and cell mediated responses to these antigens is warranted , to determine their significance in in the natural history of herpesvirus infection and pathogenesis of associated diseases . To this end , we developed and validated a multiplex bead based assay , which can test simultaneously for reactivity to 6 KSHV antigens with performances comparable to those obtained using individual ELISAs . However , bead based assays have a number of advantages over ELISAs . The dynamic range is wider , reducing the need for serial dilution of samples with very high antibody levels . Moreover , bead based assays can be developed to allow for semi-quantitative determination of antibody levels [48] in lieu of determining antibody titers . This is a significant characteristic , since antibody titers have been shown to be related to KSHV replication [49] and to be associated with the development of KS [50] , [51] . Finally , the technology requires a much smaller sample volume , and the cost per antigen , per sample , is lower . Additional antigens can be further multiplexed into the assay as needed for specific epidemiologic studies . In summary , this study provides a systematic antigenic analysis of the KSHV proteome . Several new antigens have been identified and relevant assays have been developed as a tool that can be evaluated for serodiagnosis and that will be useful in epidemiologic studies and characterization of KSHV associated conditions . Furthermore , cloning , expression and transduction vectors as well as recombinant proteins have been produced , which will constitute a significant resource for the KSHV research community . | Kaposi sarcoma-associated herpesvirus ( KSHV ) is the cause of Kaposi sarcoma , primary effusion lymphoma and a form of multicentric Castleman's disease , affecting mainly persons with HIV , other immunosuppressed patients and elderly men . Such diseases are most common where KSHV prevalence is high , in sub-Saharan Africa and the Mediterranean , and amongst men who have sex with men . Various assays for the serodiagnosis of KSHV have been developed to investigate global KSHV epidemiology . ELISAs utilizing one lytic ( K8 . 1 ) and one latent antigen ( ORF73 ) are often used . However , a more complete characterization of immune responses to all KSHV antigens may have important epidemiologic and clinical applications . We systematically expressed and purified 73 of the 85 proteins encoded by KSHV and analysed serologic responses in patients with KSHV-associated malignancies compared to healthy subjects . We identified significant reactivity to ORF38 , ORF61 , ORF59 and K5 , in addition to the known K8 . 1 , ORF73 and ORF65 . Reactivity patterns were varied; however , HIV infected individuals were reactive to more antigens , with greater intensity . Next , we developed a bead-based assay that can test a small sample simultaneously for six KSHV antigens . The new tool can improve the detection of KSHV and the characterization of asymptomatic infection and KSHV associated diseases . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"immunodeficiency",
"viruses",
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"medicine",
"pathology",
"and",
"laboratory",
"medicine",
"immunity",
"medical",
"microbiology",
"hiv",
"viral",
"pathogens",
"virology",
"microbial",
"pathogen... | 2014 | Heterogeneity and Breadth of Host Antibody Response to KSHV Infection Demonstrated by Systematic Analysis of the KSHV Proteome |
Modelling ionic channels represents a fundamental step towards developing biologically detailed neuron models . Until recently , the voltage-gated ion channels have been mainly modelled according to the formalism introduced by the seminal works of Hodgkin and Huxley ( HH ) . However , following the continuing achievements in the biophysical and molecular comprehension of these pore-forming transmembrane proteins , the HH formalism turned out to carry limitations and inconsistencies in reproducing the ion-channels electrophysiological behaviour . At the same time , Markov-type kinetic models have been increasingly proven to successfully replicate both the electrophysiological and biophysical features of different ion channels . However , in order to model even the finest non-conducting molecular conformational change , they are often equipped with a considerable number of states and related transitions , which make them computationally heavy and less suitable for implementation in conductance-based neurons and large networks of those . In this purely modelling study we develop a Markov-type kinetic model for all human voltage-gated sodium channels ( VGSCs ) . The model framework is detailed , unifying ( i . e . , it accounts for all ion-channel isoforms ) and computationally efficient ( i . e . with a minimal set of states and transitions ) . The electrophysiological data to be modelled are gathered from previously published studies on whole-cell patch-clamp experiments in mammalian cell lines heterologously expressing the human VGSC subtypes ( from NaV1 . 1 to NaV1 . 9 ) . By adopting a minimum sequence of states , and using the same state diagram for all the distinct isoforms , the model ensures the lightest computational load when used in neuron models and neural networks of increasing complexity . The transitions between the states are described by original ordinary differential equations , which represent the rate of the state transitions as a function of voltage ( i . e . , membrane potential ) . The kinetic model , developed in the NEURON simulation environment , appears to be the simplest and most parsimonious way for a detailed phenomenological description of the human VGSCs electrophysiological behaviour .
In computational neuroscience , modelling of ionic channel behaviour represents a fundamental step to develop biophysically detailed neuron models . As key players in the mechanisms underlying excitability , impulse conduction and signal transduction , both the voltage-gated and ligand-gated ion channels are essential components of the electrophysiological behaviour of each neuronal cell and , consequently , of the neural networks these cells make up [1–2] . Until recently the phenomenological behaviours of the voltage-gated ionic channels have been mainly modelled according to the formalism introduced by the seminal and forward looking work of Hodgkin and Huxley [3] . By exploiting their substantially fair approximation to the macroscopic currents of the voltage-gated ionic channels , the models derived by the Hodgkin-Huxley ( HH ) equations have been instantiated , even recently , in a multiplicity of realistic cellular and network models [4–7] . The overall simplicity and the relative light computational load of the HH formalism especially make them particularly well suited in modelling biologically detailed neural networks . However , following the continuing achievements in the biophysical and molecular comprehension of these pore-forming transmembrane proteins , the HH formalism turned out to carry limitations and inconsistencies in reproducing in detail the ion-channels electrophysiological behaviour [8–12] . More detailed insights into the single channel kinetics provided by patch-clamp techniques [12–13] and into their molecular structure by means of x-ray crystallography [14] have greatly advanced our comprehension of ion channels to a degree difficult even to conceive when Hodgkin and Huxley developed their impressive and seminal research . The more information about ion-channel gating has been achieved , the clearer is the need for models with explicit representation of single ion-channel states . In the HH formalism the gating parameters do not represent specific kinetic states of ion channels , and the HH model is sometimes not sufficient to capture various aspects of the channel behavior [15–16] . For the aforementioned reasons , Markov-type kinetic models have been developed to accurately represent an ionic channel as a collection of states and a set of transitions between them . In recent years , many kinetic models have been developed , specifically focused on single isoforms of different channels or on specific details of them , which are derived from both functional and structural studies ( e . g . , [17–19] ) . The developed models , biophysically detailed and aimed at representing the molecular conformational changes of the transitions between states , usually carry a considerable number of states and related transitions . They are well suited to describe the detailed microscopic behaviour of single ion channels , but their computational load makes them much less suitable for the implementation in multi-compartmental conductance-based biologically detailed neurons and neural networks models . As a result , although a variety of Markov-type kinetic models have been used to analyze the functional biophysical properties of single ion channels , yet very little of this information is used to develop conductance-based models of neural structures [20] . Thus , nowadays the two mutually exclusive options in modelling ionic channels rely on the HH formalism , which is global and computationally light , or the Markov-type kinetic models , which are specific , detailed and computationally heavy . Among the ionic channels , voltage-gated sodium channels ( VGSC ) are probably the most studied and modelled voltage-gated ionic channels . They are directly involved in the cellular excitation and in the onset of the spike . In humans , nine subtypes or isomers ( NaV1 . 1 to NaV1 . 9 ) of VGSC exist , each of them with peculiar kinetics and tissue distribution [1–2] . They exhibit a diversified and complex , membrane potential-dependent gating behavior [8] , and even slight modifications of their gating kinetics by genetic mutations give rise to a number of severe human diseases in peripheral nerves , skeletal muscles , the heart and the central nervous system [2 , 21] . This purely modelling study is aimed at developing a Markov-type kinetic model for VGSCs , which is detailed ( accounting for different features of the VGSCs macroscopic current ) , unifying ( accounting for all ion-channel isoforms ) and computationally efficient ( with a minimal set of states and transitions ) . By exploiting experimental data gathered from previously published electrophysiological studies from different laboratories investigating single isomers of the VGSC , we derived a simplified common kinetic model for VGSCs , suitable to be adopted in biologically detailed simulation of neural structures . The obtained results show that it is possible to develop a unifying kinetic model for VGSCs macroscopic currents by adopting a new simplified state diagram and novel equations describing the voltage dependence of the state transitions .
As we were mainly interested in modelling the macroscopic electrophysiologic behaviour of single isoforms of the sodium channel , we searched the literature for experimental data from single human VGSC ( NaV1 . 1 to NaV1 . 9 ) α-subunits , heterologously expressed in mammalian cell line ( usually Human Embrionic Kidney 293 cells ) [22–29] . Studies on NaV1 . 8 and NaV1 . 9 , whose transfection in non-nervous cell line is practically challenging , were performed in homozygous NaV1 . 8-cre mice Dorsal Root Ganglia neurons lacking endogenous NaV1 . 8 [30] , and , respectively , in ND7/23 cells ( a hybrid from mouse neuroblastoma and rat neurons ) [31] . In some of the considered studies one or two β-subunits were also co-transfected: β1-subunit was coexpressed with NaV1 . 7 [29] , and β3-subunit with NaV1 . 3 [25]; β1- and β2-subunits were both coexpressed in NaV1 . 1 [22] and in NaV1 . 2 experiments [23] . The electrophysiological experiments were conducted by means of the whole-cell patch-clamp method , usually at room temperature and our model corrected for the experimental temperature ( by means of the temperature coefficient , Q10 ) . Modelled graphics of every VGSC with reference to the experimental ones are displayed in a Supporting Information file ( S1 Appendix . Modelled graphics with reference to the original ones ) . For each VGSC isomers , the following electrophysiological data were gathered and in turn reproduced by modelling: When available , we also compared our simulations with the following electrophysiological data: The ionic channel current is governed by Ohm's law , wherein conductance is determined by the fraction of channels in the open states , xO ( 0 to 1 ) : INa ( t ) =g¯Na⋅xO ( t ) ⋅ ( V ( t ) −VNa ) , ( 5 ) where g¯Na is the sodium maximal conductance and VNa is the reversal potential of that ion . Transitions between the O ( open ) , I ( inactivated ) , and C ( closed ) states are described by conventional Markovian model equations [32] written for the fractions of channel , xO , xI , xC , to be in these states: dxOdt=ACOxC− ( AOC+AOI ) xO ( 6 ) dxIdt=AOIxO−AICxI ( 7 ) xO+xC+xI=1 ( 8 ) where AXY is the transition rate between the state X and the state Y . The topology of the single , general model and the transitions between its states have been searched for through progressive appoximations using heuristic optimization . All simulated experiments were performed by means of NEURON version 7 . 4 simulation environment [33] . The kinetic equations were written and solved directly using KINETIC methods of NMODL language of NEURON , which is a derivative of the MODL description language of the SCoP package [34] . All virtual experiments were performed on an one-compartmental cylindrical 'soma' 50 μm long with a diameter of 63 . 66 μm , so that the membrane area was set to 10'000 μm2 . The membrane capacitance was set at 1 μF/cm2 [35] . The maximal conductance density for each VGCS isomer inserted into the soma was arbitrarily set to 0 . 1 S/cm2 , and the resulting ionic current density was measured in mA/cm2 . The capacitive currents were subtracted from the total current in all the simulations . The time for single integration step ( dt ) was set to 0 . 025 ms . At every step , the rate constants of each transition were multiplied by the temperature coefficient , Q10 , calculated as follows: Q10=3 ( T°−20°10° ) ( 9 ) Original NEURON source code was developed to simulate the protocols needed to yield the electrophysiological features of the channels . The simulated voltage-clamp protocols are depicted in Fig 1 . The source code along with the virtual experimental procedures is provided and available as a ModelDB [36] entry ( http://modeldb . yale . edu/230137 ) . All simulations were performed on an iMac desktop computer running a MacOS version 10 . 12 . 5 ( ™ and © 1983–2017 , Apple Inc , Cupertino , CA , USA ) . The developed code automatically supplied the appropriate graphics , which replicated the macroscopic currents and the electrophysiological relationships found in the experimental studies ( see S1 Appendix . Modelled graphics with reference to the original ones ) . A both empirical and quantitative curve fitting method was then adopted to reconcile experimental and modelled data . Firstly , the curves and relationships obtained by the simulations were compared by visual inspection to the experimental ones . Then , the modelled curves were fitted for the Eqs ( 2 ) to ( 4 ) , as appropriate , by using a nonlinear least-squares minimization method included in NEURON ( Multiple Run Fitter subroutine ) , which in turn derives from the PRAXIS ( principal axis ) method described by Brent [37] . Finally , the parameters of the Eqs ( 2 ) to ( 4 ) of the modelled curves were compared to the experimental ones ( Table 1 ) . The agreement of the modelled data with the experimental ones was considered acceptable when the former were within two standard deviations of the latter . In order to merely test the suitability of our model to be implemented in cell models , and to compare the features of the spikes it provides to those carried out by other channel models ( built according to both HH or Markov-type formalisms ) , we inserted the developed channel model in three previously published cell models [11 , 38–39] and performed a series of voltage-clamp and current stimulation simulations . The previously published cell models were downloaded from the ModelDB [36] repository and are accessible , respectively , with the accession numbers: 3805 , 98005 , 180370 . For model specifications see the corresponding papers [11 , 38–39] . The first implementation sample , where our model is compared to a sodium channel model developed according to the HH formalism , is reported in the Result section . The remaining two samples are provided in a Supporting Information file ( S2 Appendix . Examples of model channel implementation ) . The second sample compares the performances of our model with a more complex Markov-type model , the third sample shows how our model behaves in a morphologically detailed neuron model .
Table 1 shows the values of the main electrophysiological features reproduced by the model , alongside the available corresponding experimental values for comparison , in all VGSC subtypes . The displayed values are the parameters of the fitting of experimental and simulated curves with the Eqs ( 4 ) to ( 6 ) . It can be noted that the most of the modelled data are within one standard deviation of the experimental values . Graphics from a single isomer , namely the Nav1 . 5 VGSC , comparing experimental and modelled data , are shown in Fig 2 . The complete set of graphics for each VGSC isomer , showing the curves obtained during the simulations , with reference to the corresponding experimental data in previously published electrophysiological studies , is provided as a Supporting Information file ( S1 Appendix . Modelled graphics with reference to the original ones ) . The most parsimonious state diagram able to account for the phenomenological behaviour of all the VGSC isomers was found to be a six-state one ( Fig 3 ) . It is arranged in two closed , two open and two inactivated states . The second inactivated state ( I2 ) was considered as a deeper inactivated state than I1 , only connected to I1 . This topology was found to be the simplest to consistently account for both the slow and fast inactivations , as well as for the two time constants of the recovery from inactivation . The second open state ( O2 ) , only linked to C2 , was added to reproduce in detail a second slower constant of decay from activation , usually detectable on the current-voltage curves . An alternative solution , which considered O2 as only linked to O1 ( by analogy with I1 and I2 ) , was discharged because it provided not realistic tail currents of deactivation . Two closed states ( C1 and C2 ) were found to be sufficient to faithfully reproduce the activation kinetics and the tail currents after a brisk repolarization . All transitions between two consecutive states were considered reversible ( with one exception , see below ) , and the paired forward and backward transitions were computed by equations carrying numerical values ( coefficients ) of the same order of magnitude . The only exception was the O1 to I1 transition , where the backward transition ( I1 to O1 ) was described by an infinitesimal value , so that the O1 to I1 transition could be considered irreversible . The dynamics of the different states of the channels are described by the following set of coupled ordinary differential equations: dC1dt=I1C1*I1+C2C1*C2− ( C1C2+C1I1 ) *C1 ( 10 ) dC2dt=C1C2*C1+O1C2*O1+O2C2*O2− ( C2C1+C2O1+C2O2 ) *C2 ( 11 ) dO1dt=C2O1*C2+I1O1*I1− ( O1C2+O1I1 ) *O1 ( 12 ) dO2dt=C2O2*C2−O2C2*O2 ( 13 ) dI1dt=I2I1*I2+C1I1*C1+O1I1*O1− ( I1C1+I1I2+I1O1 ) *I1 ( 14 ) dI2dt=I1I2*I1−I2I1*I2 ( 15 ) Moreover , the states obey the law of mass conservation: O1+O2+I1+I2+C1+C2=1 ( 16 ) Since the studies by Hodgkin and Huxley [3] , the voltage dependence of the rate transitions has been mathematically modelled ( Fig 4A ) as an exponential equation ( black line ) , or as a sigmoid ( blue line ) , or as a combined linear and exponential equation ( red line ) . In other cases [32 , 40] a sigmoid curve with minimum and maximum asymptote has been adopted ( Fig 4B inset ) , described by equations as below , Aω=τminω+τmaxω⋅[1+exp ( V−V1/2ωkω ) ]−1 ( 17 ) where ω is the transition between two states , τminω and τmaxω are the two asymptotes , V1/2ω the hemiactivation voltage , and kω the slope which describes the voltage sensitivity of the transition rate . By progressive optimizations , we found the most suitable general equation to be adopted in all the transitions was a sigmoid one . For most of the transitions , the minimum asymptote was conveniently set to zero , while in a few cases , notably for the O1 to I1 transition , it needed a non-zero value . Furthermore , to accommodate in detail the time course of the current-voltage curves , it was found appropriate to slightly modify the sigmoid , adding a bending at the beginning of the rising slope of the curve ( Fig 4B ) . This way , the modified sigmoid could be mathematically described as the combination of two sigmoids with opposite slope: the first one placed towards the hyperpolarized side and carrying a positive slope factor , the second one toward the depolarized side and carrying a negative slope factor . As a result , the general equation adopted to describe this double sigmoid was set as follows: Aω=Bhypω⋅[1+e ( V−Vhypωkhypω ) ]−1+Bdepω⋅[1+e ( V−Vdepωkdepω ) ]−1 ( 18 ) where Bhypω , Vhypω and khypω are the magnitude , the hemiactivation and the slope factor , respectively , of the voltage dependence of the transition rate ω in the hyperpolarized region , and Bdepω , Vdepω and kdepω the corresponding values in the depolarized region . With this formalism , the slope factor ( k ) in the hyperpolarized region assumes a positive value and a negative value in the depolarized one . In addition , when the transition rate is better described by a simple sigmoid , which happens in most of the transitions , one of the two terms of the general equation can be conveniently dropped . The complete set of parameters values for the simulation of all the sodium channel isomers is provided in Table 2 . As a general rule , according to previously proposed Markov-type channel models [e . g . , 27 , 41] , the backward and forward transitions between two consecutive states are described by equations with opposite slopes . This arrangement is usually adopted to account for the wide differences in state occupancy of the channel at different voltage values . Yet , in modelling the voltage-intensity curves and relations , we obtained more realistic results by adopting for the activation sequence ( C1 to C2 to O1 and reverse ) equations with identical slopes of the main sigmoid . Moreover , in paired forward and backward transitions an identical hemivoltage point was found to consistently fit the experimental data , as well as a shift of the transitions between C1 and C2 towards more depolarized values than the transitions between C2 and O1 . As an example , Fig 5 shows the voltage-intensity curves of NaV1 . 2 ( Fig 5A ) and NaV1 . 9 ( Fig 5B ) . This is an instance of the most divergent electrophysiological behaviour in two channel isomers , yet the model is able to reproduce in detail the real data in both the cases . The transition rates dependences from voltage in C1 to C2 transition ( green ) , C2 to C1 ( yellow ) , C2 to O1 ( red ) , O1 to C2 ( purple ) , and O1 to I1 ( black ) are depicted in Fig 5C and 5D . The forward and backward transitions between two consecutive states have identical , not opposite , slope values and identical hemiactivation points . The transitions between C1 and C2 are shifted to more depolarized values compared to the transition between C2 and O1 . In addition , the backward transitions ( C2 to C1 , and O1 to C2 ) are described by a double sigmoid , whereas the corresponding forward transitions by simple sigmoids . Here the double sigmoid of the backward transitions accounts for the short time constants of deactivation recorded as tail currents ( right end of the curves in Fig 5A and 5B ) . The higher values , indeed , of the backward transition rates at more hyperpolarized voltages drive the channel into more closed states with short latency during brisk repolarizations . Fig 6 shows the plot of the simulated results obtained during recovery from inactivation ( repriming ) in NaV1 . 2 . A relatively long depolarizing conditioning pulse sets all the channels up into inactivated states ( Fig 1C ) . A following repolarizing step , variable in duration , enables the transition from inactivated to closed states ( recovery from inactivation ) in a proportion of channels , which increases with the duration of the repolarizing step . The subsequent depolarizing test pulse probes the proportion of the channels having recovered from the inactivation . The progressive recovery with increasing duration of the repolarization is shown in Fig 6A . In the graphic of Fig 6B , the relative amplitude of the transient current following the test pulse is drawn against the logarithm of the duration of the repolarizing step . The fast and slow time constants of recovery depend on the interplay between the two inactivated states I1 and I2 during the first depolarizing step and the following repolarizing phase . As can be seen in Fig 6C , where the fractions of I1 ( red line ) and I2 ( blue line ) states are plotted against time , a step depolarization ( inset ) suddenly drives almost all channels into I1 state . As the depolarization lasts ( 100 ms in this simulation ) an increasing fraction of channels slowly moves to the I2 state . When repolarizing , the exit from the inactivated states follows a fast ( I1 , red ) and a slow ( I2 , blue ) course , which together account for the two time constants of recovery . By fine-tuning the parameters of I1 to I2 transition , and those of I2 to I1 transition , the relative fractions of inactivated states and the slow time constant of recovery , respectively , can be adjusted to consistently reproduce the experimental data . This subsection is only intended as a not exhaustive proof of concept of the feasibility and suitability of the proposed channel model to be implemented in different types of computational models . As such , no in-depth exploration of the implementations here presented has been performed .
In their pioneering and seminal works , Hodgkin and Huxley combined the voltage-clamp techniques and quantitative modelling to provide a deterministic and continuous description of the macroscopic ionic currents [3] . They were able to clarify the nonlinear behavior of ions permeation through the cellular membrane in response to membrane depolarizations , and to disclose the relationship between these ions fluxes and the axonal spike . As a consequence , a wealth of data about the electrophysiological behaviour of the excitable membranes were provided with a fairly accurate mathematical description: from the form , amplitude and velocity of a propagated action potential to the subthreshold depolarizations , to the refractory period after a spike , to the inward sodium and outward potassium movements associated to an impulse . However , thanks to the patch-clamp techniques [13] , and the discovery of the molecular structure of the pore-forming proteins [14] , it became clear that excitable membranes are studded with discrete ion channels undergoing random fluctuations between open and closed stable states [8] . In this respect , the HH formalism appeared as a simple macroscopic and deterministic description of a phenomenon that ultimately arises from the microscopic and stochastic behaviour of the system [9] . Markov-type models have been proposed as efficient kinetics scheme , suitable to capture the essential properties of a number of neural structures , like voltage-gated channels , transmitter-gated channels and second messenger-activated channels [40] . A Markov model represents an ion channel as a collection of states and a set of transition probabilities between them , and rely on the assumptions that: a ) the configuration of a channel protein can be operationally grouped into a set of distinct states separated by large energy barriers , and b ) the probability of state transitions is dependent only on the present state occupied [1 , 40] . The various states represent a sequence of protein conformations that underlies the gating of the channel . The time evolution of the probability of state Si is described by the Master equation [48]: dP ( Si , t ) dt=∑j=1nP ( Sj , t ) P ( Sj→Si ) −∑i=1nP ( Si , t ) P ( Si→Sj ) ( 25 ) where P ( Si , t ) is the probability of being in a state Si at the time t , and P ( Si→Sj ) is the transition probability from state Si to state Sj . In the limit of large numbers of identical channels , the quantities given in the master equation can be reinterpreted . The probability of being in a state Si become the fraction of channels in state Si , noted si , and the transition probabilities from state Si to state Sj becomes the rate constants , rij , of the reactions rijSi⇄Sjrji . ( 26 ) In this case , the master equation can be rewritten as: dsidt=∑j=1nsjrji−∑i=1nsirij ( 27 ) which is a conventional kinetic equation for the various states of the system [49] . Stochastic Markov models , as in Eq ( 25 ) , are adequate to describe the stochastic behaviour of ion channels as recorded using single-channel recording techniques [50] . In other cases , where a large number of ion channels are involved , as in the whole-cell patch-clamp recordings here considered , the macroscopic currents are continuous and more adequately described by conventional kinetic equations , as in Eq ( 27 ) . In the former case , derived from single-channel recordings , a Markov model is usually designated starting from ion-channel molecular representation , with each state of the model corresponding to a different configuration of the molecule . This approach gives a better understanding of the biophysical structure and functioning of the channel , and , by taking into account also the smaller gating currents , it details even the minimal , non-conducting molecular displacement . Stochastic Markov models derived from single-channel recordings in ligand-gated ion channels have proven to be able to solve the inverse problem , that is the direct fitting of the models with raw data , with provision of estimates for rate constants and estimation of the errors for those estimates [51–52] . However , it is also possible to take a signal-processing approach to the design of Markov models , [20 , 49 , 53]: the required model is the minimal model that represents with sufficient accuracy the response of the channel to the stimulation protocols . This approach leads to more economical models , more suitable for numerical simulations of large collections of channels and of neurons , and was followed in the present paper . In particular , this phenomenological approach is more reminiscent of the empirical and classical fitting of ionic macroscopic currents developed by Hodgkin and Huxley [3] . It could be argued that such approach is more similar to the fitting of a curve and hardly suitable to reveal the finest details of the biophysical features of the channel . Yet , this is specifically in agreement with the main goal of the present study , which was to develop a model without the structural HH limitations and able to include the most recent experimental data on the macroscopic currents of VGSCs . The phenomenological approach is intended to develop the smallest number of states and transitions necessary to replicate the electrophysiological VGSCs behaviour . Consequently , there is no exact correspondence of the model states with the physical states of the channel . In other words , in our phenomenological model the states do not represent single physical events , but each of them should be considered as an aggregate of molecular configurations suitable to be treated as a functional entity . For example , while a series of four ( or more ) closed states are commonly hypothesized ( usually in adherence with the tetrameric structure of the proteic channel ) before an open state can develop following a depolarizing step , our model collapses them all in only two . Two closed states , indeed , are necessary and sufficient to deterministically reproduce: a ) the tail currents after a brisk repolaritation , b ) the kinetics of the activation sequence , c ) the kinetics of fast inactivation . The aim to develop the computationally lightest model allows us to make one transition practically irreversible ( from O1 to I1 ) , and releases the phenomenological model from the principle of microscopic reversibility , like other kinetic schemes based on macroscopic currents ( see [54] , for a recent example ) . Microscopic reversibility , indeed , only holds when the states are elementary processes ( collisions , molecules , elementary reactions , etc ) . On the other hand , most , but not all , ion channels obey the law of microscopic reversibility [55] , and the law is only true at genuine equilibrium , which could not be the case when some sort of external energy supply is involved ( in this case , an ionic gradient ) [55] . Kuo and Bean [41] proposed a Markov-type model of VGSC incorporating the results of their study on the kinetics of the recovery from inactivation of sodium channel in rat hippocampal CA1 neurons . Since then , the model or its variants have been adopted as a more detailed alternative compared to the HH models [11 , 56–59] . Yet , being provided with 12 states and 32 transitions , its computational load makes it quite heavy to be implemented in multi-compartmental neuron models and networks of those . Mimicking the open probability of the HH model , an 8-state Markov-type kinetic model of voltage-gated ion channels has also been proposed [59] . More recently , aimed at modelling the slow inactivation of VGSC , a new version of the model has been proposed [27] for the isomer NaV1 . 5 , derived from the 8-state model by Milescu et al [59] . To account for the slow inactivation , 4 inactivated states were added , and all the transitions between states ( similarly to the present model ) were made voltage-dependent . The resulting model ( Figure 3B in [27] ) fits very smoothly the complex electrophysiological behaviour of NaV1 . 5 , yet it is equipped with 12 states and 34 transitions . Compared to the model by Zhang [27] , the one here proposed is able to simulate in detail the phenomenological behavior of NaV1 . 5 as well as of each other VGSC isomers with a significantly lower number of states and transitions ( 6 and 12 , respectively ) . In recent years , the increasing availability to the scientific community of powerful computational systems [44 , 60] able to process , even in parallel , huge amount of data in relatively short time , has promoted the development of simulations of complex neural structures [4–7] , equipped with large amounts of neural cells and synaptic connections . These simulations tried to be realistic and biologically inspired as much as possible , according to a bottom-up plan , and they succeeded , indeed , in replicating a number of electrophysiological experimental features of the cells . In such bottom-up approaches , however , the value of detailed kinetics of ionic channels , as building blocks of cells electrophysiology , cannot be ignored . Taking examples from the VGSC here described , the presence of complex kinetics with fast and slow inactivations must be considered , as they have direct effects on the recovery from inactivation and , consequently , on the refractory time of the cells which , in turn , affects the ability of the cell to fire repetitive action potentials . In addition , it should be considered that biologically inspired models have to consistently deal with the diversity and variety of the different channel isoforms . A differential topographic clustering of distinct ionic channels isomers , indeed , has been described also in subcellular compartments [61] . In real neurons , indeed , also subtle differences in ion-channels kinetics do have relevance . This is mainly showed with the clearest evidence by the effects of genetic mutations affecting ion-channels genes , considered pathogenic in a number of channelopathies . As regard to the VGSC , indeed , even slight differences in kinetics sustained by the mutation can give rise to severe diseases [21–22 , 30] . The present work is a purely modelling study , and the proposed model relies on previously published experimental data . On one hand , this limits the genuine interpretation the modeller can actively draw from the raw experimental data . On the other hand , it is probably not affordable for a single laboratory to conduct research on the electrophysiological behavior of all the VGSC isomers , and data have necessarily to be collected from different studies . Lack of interaction with experimenters also acts in the opposite way , as some suggestion of not canonical experimental paradigm ( e . g . the adoption of different levels of repolarization during inactivation protocols ) , arisen during modelling studies , cannot be performed . Moreover , different laboratories often perform experiments with not exactly similar protocols , and in some cases not all the useful experimental data are available . One example is the slow component of the recovery from inactivation ( Table 1 ) , where the second slower time constant of recovery can be only discernible by electrophysiological protocols carrying a long conditioning pulse , followed by longer repolarizing intervals , before the test pulse . A further example is the two time constants of fast inactivation or the ultra-slow inactivation , which have been quantitatively reported on in only few studies . A further limitation is that some less investigated VGSC electrophysiological features , like the resurgent currents , have not been accounted for in this study . An ensuing follow-up of the present study will be to evaluate the suitability of the proposed model as a general kinetic model for all voltage-gated ionic channels with similar molecular structure ( four-metameric pore-forming proteins with six transmembrane domains in each metamere ) , calcium and potassium ion channels in primis . | A unifying novel kinetic model of human voltage-gated sodium channels is proposed , which is able to reproduce in detail the macroscopic currents of all the ion-channel isomers , from NaV1 . 1 to NaV1 . 9 . Its topology consists of six states ( two closed , two open , two inactivated ) and twelve transitions , and it is particularly well suited to be implemented in biologically inspired multi-compartmental neural cells and neural network models . It represents the most parsimonious kinetic model able to account for the most recently described electrophysiological features , and it has been developed by taking into account the experimental data gathered by published work reporting on each different isomer heterologously expressed in mammalian cell lines . Equipped with original differential equations , the model reproduces in detail the ion-channel macroscopic electrophysiological features with the minimal computational load . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"chemical",
"compounds",
"markov",
"models",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"isomerism",
"ion",
"channels",
"mathematics",
"sodium",
"channels",
"isomers",
"animal",
"cells",
... | 2017 | A single Markov-type kinetic model accounting for the macroscopic currents of all human voltage-gated sodium channel isoforms |
Methods to determine blood-meal sources of hematophagous Triatominae bugs ( Chagas disease vectors ) are serological or based on PCR employing species-specific primers or heteroduplex analysis , but these are expensive , inaccurate , or problematic when the insect has fed on more than one species . To solve those problems , we developed a technique based on HRM analysis of the mitochondrial gene cytochrome B ( Cyt b ) . This technique recognized 14 species involved in several ecoepidemiological cycles of the transmission of Trypanosoma cruzi and it was suitable with DNA extracted from intestinal content and feces 30 days after feeding , revealing a resolution power that can display mixed feedings . Field samples were analyzed showing blood meal sources corresponding to domestic , peridomiciliary and sylvatic cycles . The technique only requires a single pair of primers that amplify the Cyt b gene in vertebrates and no other standardization , making it quick , easy , relatively inexpensive , and highly accurate .
Trypanosoma cruzi , the etiological agent of Chagas disease , is transmitted by hematophagous Reduviidae insects belonging to Triatominae subfamily , which are mostly distributed on the American continent and comprises 141 species in 15 genera [1] , [2] . T . cruzi can infect more than 180 species of terrestrial and arboreal mammals belonging to nine orders and 25 subfamilies , which are natural blood-meal sources of triatomine bugs [3] . Depending on the habitats of triatomines , at least two transmission cycles of T . cruzi have been identified: domestic and sylvatic . The first one includes humans , domestic animals , and domiciliated vectors . The second one involves sylvatic insects and wild animals [4] . A third cycle associated with anthropic environments can occur in several places where domestic animals are kept in sheds near dwellings , called by some authors a peridomestic cycle [5] . However , the epidemiological scenario of Chagas disease is more complex because it can include overlapping cycles [6]–[10] , intrusions of wild vectors into human dwellings [11] , [12] , colonization and re-infestation of new sylvatic vector species after spraying houses [4] , [13]–[15] , the regular intrusion of wild mammals inside houses [16] , house infestation by two vector species [17] , etc . The identification of T . cruzi's transmission cycle has been considered an essential issue to establish and design vector control and surveillance strategies [18] . In this sense , the determination of blood-meal sources in triatomine insects is essential to identify infection foci , risk factors , and new vector and mammal species involved in the transmission cycles [19] . In triatomines , so far , the methods used for this purpose include serologic and PCR-based molecular techniques . ELISA-based serological techniques are the most frequently used in blood-meal determination of Chagas disease vectors [20]–[24] . However , this method lacks specificity because it cross-reacts with species within the same family and is time-consuming and very expensive because it requires the preparation of specific antisera for each vertebrate host involved [25] . Unlike serological tools , PCR-based molecular methods are more specific and usually easier to perform , although the design of specific primers for each species involved in transmission scenarios makes it difficult to identify the vector and mammals from the wild cycle where a wide variety of species exist . Additionally , this technique is wasteful because each sample has to be amplified by PCR many times with different molecular markers [26]–[28] . A more powerful approach [29]–[31] is the heteroduplex assay of the mitochondrial cytochrome B ( Cyt b ) gene ( reviewed in [19] ) . This technique can detect differences in the DNA sequences of this gene among species by differences on electrophoretic profiles [29] . Although it is increasingly used in triatomines , standardization is often difficult , and when insects have fed on more than one species the application and the analysis could be complex . Recently , high-resolution melting ( HRM ) has become a sensitive genotyping method , based on the characteristics of thermal denaturation of the amplicons . This method has a much higher performance information never before achieved by classical DNA melting curve analysis [32] . HRM is performed using a fluorescent double-stranded DNA dye that can be used in fully saturating conditions [33] , [34] . The amplicon is analyzed by gradual denaturation through increasing temperature and decreased fluorescence caused by the release of intercalating dye from DNA . The melting temperature ( Tm ) and specific shape of the melting curve result from the DNA sequence , GC content , and amplicon length [35] , [36] , so it is a useful technique to obtain species-specific genotypes . Due to the increased demand for rapid , economic , easy , and high-throughput genotyping analyses , there has been a considerable focus on HRM , which can detect sequence variants without sequencing or hybridization procedures [37]–[40] . Here we present a fast , inexpensive , accurate , and sensitive technique to identify blood-meal sources of triatomines based on Cyt b gene HRM analysis of host-specific genotypes and its application in field studies .
All Animals were handled in strict accordance with good animal practice as defined by the Colombian code of practice for the care and use of animals for scientific purposes . Ethical approval ( Act N° 53 , 30/06/2009 ) for analyzing animal specimens was obtained from the animal ethics Committee of the University of Antioquia , Medellin , Colombia . To differentiate among species-specific genotypes of natural blood-meal sources of triatomine bugs , a 383-bp fragment from the Cyt b gene was amplified from DNA samples extracted from tissue or blood of 36 individuals corresponding to 14 species . Species were chosen according to their epidemiological relevance as reservoirs of T . cruzi or by their proximity to anthropic environments . The species , number of individuals per species , and type of sample used to obtain DNA are shown in Table 1 . To validate HRM-Cyt b profiles , six fifth-stage nymphs per species of Rhodnius prolixus , R . colombiensis , and Triatoma maculata maintained under controlled laboratory conditions of temperature ( 28°C ) and relative humidity ( 70% ) were fed with chicken blood until satiated . Feces from three individuals of each species and intestinal content from three other individuals were collected 5 days post-feeding for posterior DNA extraction and amplification of the Cyt b gene . Additionally , the sensitivity of HRM Cyt b over time was analyzed in two groups of five R . prolixus adults fed with chicken or human blood . The intestinal content and feces were collected 1 , 5 , 15 , and 30 days post-feeding and processed as below . To evaluate the capability of this technique to detect HRM-Cyt b profiles in mixed feeds , five fifth-instar R . prolixus individuals were fed with both mice and chicken blood and the intestinal content and feces were collected at 5 days post-feeding for subsequent HRM analysis . To apply the HRM technique to triatomines from the field , a total of 20 insects were collected in four localities of Colombia's Caribbean region , which display different eco-epidemiological transmission cycles of T . cruzi , as follows: ( 1 ) six domiciliary R . prolixus and three peridomiciliary T . dimidiata were collected by timed manual capture in an indigenous community with high epidemiological risk from the Sierra Nevada de Santa Marta area [41]; ( 2 ) three peridomiciliary T . maculata and one sylvatic Eratyrus cuspidatus ( light-trap captured ) were collected in the town of Talaigua Nuevo , located in the Bolivar department where a moderate epidemiological risk exists [42]; ( 3 ) three sylvatic T . dimidiata were collected manually in a hole in a fallen tree in Turbo , near the border between Colombia and Panama in the Antioquia department; this region is considered a nonendemic zone and a few reports on triatomine visiting houses have been released [43]; ( 4 ) finally , four sylvatic R . pallescens were collected in a palm tree in Aguachica in the Cesar department , where nondomiciliary triatomines have been reported [43]; they were caught using live chicken-bait traps ( Table S1 ) . All insects were placed in plastic bottles , marked , transported to the laboratory , and identified according to the classification proposed by Lent and Wygodzinsky [44] . Feces samples from insects were collected and analyzed by HRM of the Cyt b gene to identify the blood-meal sources of each insect . DNA was extracted using the phenol-chloroform method [45] , with some modifications . Briefly , 200 µL of serum , 25 mg of tissue , 25 µL of intestinal content , or 30 µL of feces were incubated at 37°C for 4 h with 200 µg of proteinase K , 1% sodium dodecyl sulfate , 2 . 5 mM disodium EDTA , and 25 mM sodium acetate . DNA was extracted with phenol and chloroform , precipitated with 0 . 3 M sodium acetate and absolute ethanol , washed with 70% ethanol , vacuum dried , and then dissolved in 50 µL of distilled water . Quality and DNA concentration were measured using a Nanodrop 2000 spectrophotometer ( NanoDrop Technologies , Wilmington , DE , USA ) . A 383-bp fragment from the Cyt b gene was real time PCR-amplified from genomic DNA samples using the 5′-CCCCTCAGAATGATATTTGTCCTCA-3′ and 5′- CCATCCAACATCTCAGCATGATGAAA-3′ primers [29] . Real time PCR was performed at a final volume of 10 µL , containing 0 . 5 µM of each primer , 0 . 25 mM of each dNTP , 3 . 5 mM MgCl2 , 0 . 5 U TrueStart Hot Start polymerase , 1 . 5× SYBR green dye ( Invitrogen ) , and 10 ng of genomic DNA . Each sample was amplified in triplicate . The thermal cycling conditions were as follows: pre-heating at 95°C for 5 min , 35 cycles at 95°C for 30 s , 60°C for 30 s , and 72°C for 30 s in a thermal cycler Rotor Gene Q system ( Qiagen ) . Following the real time PCR , melting-curve analysis of amplicons was conducted in the thermocycler ( Rotor-Gene Q ) by increasing the temperature from 72°C to 92°C at ramping increments from 0 . 1°C/s , recording changes in fluorescence with changes in temperature ( dF/dT ) , and plotting against changes in temperature . DNA extracted from triatomine legs and T . cruzi were used as negative control . HRM analysis was carried out using the Rotor Gene Q software v2 . 2 with normalization regions between 76 . 15–78 . 65°C and 89 . 50–91 . 00°C . Genotypes were defined by selecting a sample from each standard species as a reference control to identify unknown samples . The software then auto-called the genotype and melting temperatures of each amplicon and provided a confidence percentage based on the square root of the correlation coefficient between samples and the reference genotypes . Any specimen generating melt curves but having a dF/dT less than 1 . 0 was considered not subjected to genotyping . Averages of melting temperatures and the confidence percentage of the specimen replicates were assigned to a representative genotype , and the standard deviation , confidence intervals , and variation coefficient were calculated using the Prism GraphPad v4 . 0 software ( GraphPad software , Inc ) . To confirm the sequence identity of those samples tested by HRM analysis , a 358-bp fragment of Cyt b was sequenced for the standard samples and ten test samples chosen randomly ( 2 , 3 , 4 , 5 , 10 , 12 , 15 , 16 , 19 , and 20 ( Table S1 ) ) . For each specimen , both forward and reverse sequences were used to generate a consensus sequence using Bioedit v . 7 . 0 . 5 [46] , and then positional nucleotide sites were compared after multiple alignment done using the ClustalW algorithm [47] implemented in Bioedit v . 7 . 0 . 5 [48] . Additionally , a nucleotide Basic Local Alignment Search ( Blastn ) was performed to estimate the matched hit scores , identity percentages and e-values of the Cyt b tested . Finally , the distance tree based on net genetic distances ( p-distances ) was performed using the neighbor-joining algorithm with 1000 bootstrap replicates using MEGA 5 . 05 [49] .
All samples analyzed amplified a 383-bp product , as previously reported to the Cyt b gene [29] ( data not shown ) and no product was obtained when the insect DNA or T . cruzi DNA were used . The amplicon melt curve and Tm showed low intraspecies variability ( Table 2 , Figure 1 ) . A specific profile for each species analyzed was observed ( Figures 2 and S1 ) , with confidence percentages ( %C ) ranging from 78 . 81% and 98 . 69% ( Table 2 ) , indicating that samples were correctly identified in each of the experiments conducted in triplicate . Although some species exhibited the same Tm value , HRM profiles in these species were clearly discriminated and recognized as different genotypes , such as dog and sheep or pig and horse ( Table 2 , Figure S1 ) . On the other hand , although the cow melt curve showed two peaks , the genotype was recognized with a high confidence percentage ( Table 2 ) . These two peaks were found in different types of samples ( muscle and blood ) . The HRM analysis of Cyt b in R . prolixus collected inside dwellings indicated that five of the six individuals evaluated had at least a human blood-meal source , and the remaining one a dog blood-meal source ( Table 4 ) . Two peridomiciliary T . dimidiata showed mixed blood-meal sources ( human and dog; human and a nonidentified source ) ( Table 4 ) . Sylvatic T . dimidiata showed a mixed blood-meal source of opossum , human , and a nonidentified source . Of two T . maculata caught in a house , one was found to have fed on a human blood-meal source , while the other had a mixed human and nonidentified source . Finally , the sylvatic E . cuspidatus and R . pallescens were identified as being blood-fed from human , chicken , and nonidentified sources ( Table 4 , Figure 3 ) . As mentioned above , four mixed samples , including one intradomiciliary insect , showed a peak with a Tm = 84 . 83±0 . 05 , which was inconsistent with any of our standards ( Table 4 ) , indicating another additional blood-meal source that was not considered in this study . Five samples did not amplify . Blastn results showed significant identity values greater than 95% for all samples except for sample number 20 ( Table 5 ) . Multiple alignment of the test sample and standards showed great nucleotide similarity and coherence with HRM results previously obtained . The percentage of identity among the field-collected samples and their respective standards ranged between 95% and 99% in samples from human and opossum , and 57% in the sample from chicken ( Table 5 ) . Finally , the distance tree showed well-supported groups according to HRM results ( Figure S3 ) . It is worth mentioning that the field-collected samples sequenced included a sample ( number 12 ) that was positive for mixed feeding for HRM analysis . In this case , the sequence analysis also identified the two species ( human ( 12_1 ) and opossum ( 12_2 ) ) ( Table 5 ) .
The epidemiological scenario of Chagas disease has become increasingly complex over the years . The natural habitats of some human populations within the forest and deforestation caused by humans are but two of the reasons that may complicate this scenario . Thus , the classical separation of transmission cycles defined for this disease could be different for many places , making it difficult to determine the epidemiological characteristics of particular regions . With this unclear scenario , the determination of blood-meal sources in hematophagous vectors has become essential to surveillance and prevention of potential infection foci . In this study , HRM analysis of the Cyt b gene made it possible to identify 14 species successfully even when some of them had the same Tm values . Each species was well recognized under a variety of species reaching high confidence percentage values , thus showing the power of this classification using gene amplification and HRM analysis . It is worth noting that this recognition is possible with a single PCR , making it a quick and inexpensive technique . It is important to highlight that no studies with this number of species standards have been conducted with Chagas disease vectors . Pizarro and Stevens [28] designed primers against different targets from 11 species and Mota et al . [26] also identified 11 species in their study . ELISA-based techniques could identify the same number of species , but it is very expensive and , in most cases , anti-serum for many species is not commercially available . On the other hand , studies based on PCR methods may also increase the number of species , but it is necessary to standardize PCR conditions for each species . Additionally , every DNA sample must be submitted to a number of PCRs to identify the blood sources , making the technique very expensive . Instead , the technique used herein only needs a single pair of primers , making it easy to include other species according to every epidemiological scenario with no other standardizations of PCR conditions . A similar power is reached when DNA extracted from intestinal content and feces is analyzed with this technique , demonstrating that both can be used for the identification of the blood source from insects . Although no misidentification was recorded with DNA from insect feces , low confidence percentages were reached and other samples did not amplify , showing that the quality of the sample is a limitation of the technique , as reported for real-time PCR [50] . Nonetheless , work with feces has substantial advantages because it is sometimes hard to extract intestinal content from insects , especially when they have been starved for a long time . Another benefit of feces is the possibility of making a diagnosis of T . cruzi infection and the determination of the blood-meal source; both tests that can be done with a single sample of DNA without sacrificing the insect . The HRM technique proved to work in both types of samples ( intestinal and fecal ) at least until 30 days after insects were fed . This showed the applicability of the technique even when the blood-meal sources were limited to the triatomines . Pizarro and Stevens [28] reported on the feasibility of their technique , based on the primer design for each species , for 2 months . It is probable that our technique has identical results , but this was not verified . However , many factors have an influence on the presence of blood in the intestinal tract over time , as described by Lee et al . [30] . Chicken blood has nucleated erythrocytes , so a large amount of DNA should remain in the intestine for a long time . This could explain the presence of unspecific peaks in the melt curve when insects were fed with human blood because the insects are usually fed with chicken blood in laboratory conditions . Thus , the present results suggest that chicken blood remains in the intestinal tract longer than mouse and human blood ( data not shown ) . On the other hand , the cow profile showed two melting peaks , suggesting the amplification of nonspecific fragments . This could stem from nuclear copies of mtDNA ( Numts ) . It is well known that Numts can be amplified in genetic studies based on mtDNA [51] , [52] . Some of these Numts have been described in cattle [53] . However , this phenomenon does not produce mistakes in classification because the shape remains the same across different specimens included . Unspecific peaks led to testing mixed feedings , showing that the technique detects both Tm sources . This allows identification of many species from which the insects have been fed , even when there are no standards for those species . This is impossible with methods such as ELISA or techniques based on PCR-specific primers . Heteroduplex analysis identifies mixed feeding as well , but the electrophoretic profile becomes very complex , making it difficult to analyze , and sometimes it does not recognize even a single species within the mixture [30] . Sequencing of DNA fragments can resolve mixed feedings well [54] , but is quite difficult to analyze and time-consuming . However , we suggest using sequencing of amplicons only when unidentified samples by HRM analysis are present . We processed 20 samples from the field but only 15 amplified . The other five samples were probably of insufficient quality , which is not surprising because it was feces DNA . In general , the expected genotypes were recognized by the analysis . Those insects captured inside houses showed human feeding and one dog feeding , suggesting a domestic transmission cycle involving humans and to a lesser extent dogs , which has been reported to have epidemiological relevance [55]–[58] . Samples from the peridomestic or sylvatic cycle also showed human feeding , in agreement with other studies showing that R . prolixus and T . dimidiata fed mostly on human blood [23] . Sometimes , mixed feeding with human blood showed a low peak in the melt curve ( Figure 3 ) . It is possible that the other techniques were not sensitive enough to detect human blood because of the small amount of DNA in the sample [31] . We included R . pallescens caught with live-bait trap , which showed chicken-blood feeding as expected by the type of bait used . This control from the field confirms the capacity and accuracy of the technique even when field samples are analyzed , which is the aim of technique . E . cuspidatus showed human feeding , which is surprising because is a strictly sylvatic species . However , Dib et al . [59] incriminated E . cuspidatus in transmission of parasite when comparing RAPD profiles from T . cruzi strain isolates from human and vector . These two results showed that although sylvatic this species probably visits dwellings looking for feeding sources . Additionally , Cortés and Suárez [42] reported E . cuspidatus inside houses in the study area , suggesting that the insect occasionally colonized intra- and peridomestic environments , which is confirmed by the present result . More studies must be conducted to determine the role of this species in transmission of T . cruzi in this area . There was only one genotype that could not be identified with a Tm = 84 . 83°C±0 . 05 , indicating that the number of species standards included in the study was adequate . However , this could be improved knowing the fauna and diversity of species of a particular study area . When the sample is not recognized , the next step is sequencing , as reported by other authors [29]–[31] . Based on these results , we recommend the use of Tm values as a first approach to identify a sample from insects collected in natural populations . Then the species selected based on the Tm could be used as a standard sample for HRM analysis , and finally the species not identified could be sequenced . In conclusion , we believe that epidemiological studies involving vectorial incrimination and transmission dynamics must identify the blood-meal source to cover the entire panorama of transmission . HRM analysis of the Cyt b gene is the most powerful technique in this type of study because it can accurately identify the species even when the vector has mixed feeding , it has a high resolution power , and it is fast , easy , and inexpensive . However , it is important to obtain high-quality DNA and be mindful of the fauna of the study area to have an adequate number of standard species . | Chagas disease is one of the most important tropical diseases in America . This disease is caused by the parasite Trypanosoma cruzi and transmitted through the feces of blood-sucking insects known as triatomines . Different species of insects have different habits and food sources that confer variable degrees of epidemiological importance . In this paper , we propose the use of High Resolution Melting ( HRM ) analysis of cytochrome b ( cyt b ) gene PCR products to identify blood-food sources in triatomines . This tool can effectively differentiate blood-meal sources of insects collected from the field . Such data allows for targeted investigations of insect species that are likely to be involved in the transmission of the parasite to humans in different regions . This knowledge is very important because it allows establishing and designing vector control and surveillance strategies according to each geographical area and to stop the transmission of the parasite to human populations by insects . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"neglected",
"tropical",
"diseases"
] | 2012 | High-Resolution Melting (HRM) of the Cytochrome B Gene: A Powerful Approach to Identify Blood-Meal Sources in Chagas Disease Vectors |
Proteases have been implicated in a variety of developmental processes during the malaria parasite lifecycle . In particular , invasion and egress of the parasite from the infected hepatocyte and erythrocyte , critically depend on protease activity . Although falcipain-1 was the first cysteine protease to be characterized in P . falciparum , its role in the lifecycle of the parasite has been the subject of some controversy . While an inhibitor of falcipain-1 blocked erythrocyte invasion by merozoites , two independent studies showed that falcipain-1 disruption did not affect growth of blood stage parasites . To shed light on the role of this protease over the entire Plasmodium lifecycle , we disrupted berghepain-1 , its ortholog in the rodent parasite P . berghei . We found that this mutant parasite displays a pronounced delay in blood stage infection after inoculation of sporozoites . Experiments designed to pinpoint the defect of berghepain-1 knockout parasites found that it was not due to alterations in gliding motility , hepatocyte invasion or liver stage development and that injection of berghepain-1 knockout merosomes replicated the phenotype of delayed blood stage growth after sporozoite inoculation . We identified an additional role for berghepain-1 in preparing blood stage merozoites for infection of erythrocytes and observed that berghepain-1 knockout parasites exhibit a reticulocyte restriction , suggesting that berghepain-1 activity broadens the erythrocyte repertoire of the parasite . The lack of berghepain-1 expression resulted in a greater reduction in erythrocyte infectivity in hepatocyte-derived merozoites than it did in erythrocyte-derived merozoites . These observations indicate a role for berghepain-1 in processing ligands important for merozoite infectivity and provide evidence supporting the notion that hepatic and erythrocytic merozoites , though structurally similar , are not identical .
Malaria , caused by parasites of the genus Plasmodium , continues to be a global health problem , causing significant morbidity and mortality particularly in resource poor settings [1] . Human infection begins with the injection of sporozoites into the skin where , using gliding motility , they find and enter the blood stream , carrying the parasites to the liver [2] . Here , they invade hepatocytes and develop into exo-erythrocytic forms ( EEFs ) , replicating to produce thousands of hepatic stage merozoites . Once mature , these liver stage merozoites bud from the infected hepatocytes and enter the blood stream in packets termed merosomes [3] . Hepatic merozoites are released from merosomes and invade erythrocytes , where they develop and divide to produce daughter blood stage merozoites . Infected red blood cells eventually rupture to release the newly formed merozoites that can then go on to invade new red blood cells . Thus , an iterative cycle of parasite replication is established , leading to high numbers of parasite-infected erythrocytes in the host blood stream and clinical symptoms of malaria . A proportion of erythrocytic stage parasites differentiate to sexual stage parasites , which are transmitted to the mosquito as it takes a blood meal . Fertilization occurs in the mosquito midgut and the parasite migrates across the midgut wall to form oocysts containing sporozoites that are released and invade the salivary glands to be injected during the next blood meal . With the emergence of insecticide-resistant mosquitoes and parasites that are increasingly resistant to available antimalarial drugs , there is an urgent need for the identification of new drug targets . Cysteine proteases play key roles at multiple stages of the Plasmodium life cycle , including functions in host cell invasion [4–6] , hemoglobin degradation [7 , 8] and facilitation of parasite egress from hepatocytes [3] and erythrocytes through cleavage of both parasite proteins [9 , 10] and erythrocyte ankyrin [11] . Of the 33 putative cysteine proteases encoded in the P . falciparum genome [12] , the falcipain family of papain-like cysteine proteases contains four members , with falcipain-2 and -3 having well-established roles in the degradation of host erythrocyte hemoglobin in the parasite food vacuole [13–17] . While falcipain-1 was the first cysteine protease to be characterized in P . falciparum [18] , its physiological role in the lifecycle of the parasite still remains poorly understood . Compared to falcipain-2 and -3 , which are similar in sequence ( 68% of sequence identity ) , falcipain-1 shares only 38–40% of sequence identity to the other falcipains . Falcipain-1 was detected in the transcriptome [19] and proteomes of asexual and sexual erythrocytic stages of the parasite , as well as in the sporozoite stage [20–22] . Based on the generation of an inhibitor for falcipain-1 , it was suggested that this protease plays an important role in merozoite invasion of erythrocytes [23] . These data are consistent with many lines of evidence showing that proteases are required for host cell invasion by Apicomplexan parasites , specifically for processing of surface proteins to expose adhesive domains and to release adhesive interactions ( [24–28] , reviewed in [9] ) and with a previous study that found deletion of the rodent ortholog of falcipain-1 resulted in a blood stage growth defect [13] . Interpretation of these results has been complicated by two subsequent studies which found that deletion of falcipain-1 in P . falciparum lines 3D7 and D10 did not impact growth of erythrocytic stages of the parasite [17 , 29 , 30] . Given the controversy surrounding the role of falcipain-1 in merozoite invasion , and the possibility that it might function in mosquito stages , we proposed to shed light on the role of falcipain-1 by studying its ortholog in the rodent parasite P . berghei . Rodent malaria parasites have a smaller repertoire of falcipain-family proteases compared to P . falciparum with only one ortholog with similarity to falcipain-2/-3 identified . However , orthologs of falcipain-1 exist in all Plasmodium species studied to date [13 , 15] In the present study , we generated a deletion mutant of berghepain-1 , ( PBANKA_132170 ) as well as an epitope-tagged berghepain-1 parasite . Our results support a role for berghepain-1 in hepatic and erythrocytic merozoite infection of erythrocytes and in particular point to a function of this protease in infection of mature red blood cells . Importantly , we find that hepatic merozoite infectivity is impaired more than infectivity of blood stage merozoites , indicative of differences between these two types of merozoites .
To investigate whether berghepain-1 is critical P . berghei development we generated berghepain-1 deletion or ‘knockout’ ( BP1-KO ) parasites by replacing the berghepain-1 open reading frame with a drug selection cassette , using the flanking untranslated regions to target the locus for homologous recombination ( S1 Fig ) . Two independent transfections were performed and verification of berghepain-1 deletion was performed by PCR and protein deletion was verified using a probe that binds to the berghepain-1 protein ( S2 Fig ) . One clone from each transfection , designated BP1-KO clone 1 and BP1-KO clone 2 , were characterized . Data shown displays combined results obtained using both clones , when indicated . To control for effects due to manipulation of the genetic locus , a berghepain-1 control parasite ( BP1-CON ) was generated using a targeting plasmid containing the entire berghepain-1 gene with its endogenous 5' and 3' regulatory elements , as well as the selectable marker cassette ( S1 Fig ) . Each construct was transfected into GFP-expressing P . berghei ANKA clone 507cl1 [31] , a strain lethal to mice . A . stephensi mosquitoes were infected with both berghepain-1 knockout clones , as well as the control line BP1-CON and development from oocyst to salivary gland sporozoites was followed . Both BP1-KO clone 1 and BP1-KO clone 2 parasites were compared to the control line BP1-CON and were found to generate normal numbers of oocysts ( Fig 1A ) , and salivary gland sporozoites ( Fig 1B ) , indicating that berghepain-1 does not have a crucial role at these stages of the parasite life cycle . Gliding motility is key for sporozoite migration out of the inoculation site and for infection of hepatocytes upon arrival in the liver [[32 , 33] , reviewed in [2]] . To test whether berghepain-1 knockout sporozoites are motile and viable , we performed an in vitro gliding motility assay , quantifying the proportion of motile salivary gland sporozoites and the trails they leave behind . As sporozoites glide , they deposit trails of surface proteins , such as the circumsporozoite protein ( CSP ) , which can be stained and trails can be manually counted [34] . Motility of BP1-KO sporozoites was comparable to that of control parasites , indicating that berghepain-1 does not have a critical role in sporozoite gliding motility ( Fig 1C ) . To investigate the infectivity of berghepain-1 knockout sporozoites , we inoculated mutant and control sporozoites intravenously ( i . v . ) into mice and determined the time to detectable blood stage infection , termed the prepatent period , by Giemsa-stained blood smears . Typically in mice , an i . v . inoculum of 100 or 1 , 000 wild-type sporozoites results in detectable blood stage parasites with a prepatent period of three and four days , respectively , with each day of delay indicating an approximate 10-fold reduction in the infectious inoculum or in the downstream events ultimately leading to detectable blood-stage parasites [35] . Following inoculation of 100 to 10 , 000 sporozoites into C57BL/6 and Swiss Webster mice , many of the mice inoculated with BP1-KO clone 1 or clone 2 sporozoites failed to develop blood stage parasitemia ( Table 1 ) . Of those mice that developed parasitemia , the prepatent period of BP1-KO sporozoites was consistently delayed by approximately four days compared to mice inoculated with the same number of BP1-CON sporozoites ( Table 1 ) . Given the delay in prepatent period after injection of BP1-KO sporozoites compared to controls , we set out to systematically investigate sporozoite infection and development in the liver , to identify the stage at which berghepain-1 is required . To assess the liver infectivity of BP1-CON and BP1-KO parasites in vivo , 10 , 000 BP1-KO or BP1-CON salivary gland sporozoites were injected i . v . or intradermally ( i . d . ) and 40 h post infection livers were harvested for quantification of Pb18S rRNA . No significant reduction of BP1-KO liver stage growth compared to BP1-CON was found by either route of inoculation ( Fig 2A and S3 Fig , respectively ) , suggesting that berghepain-1 is not required for sporozoite exit from the dermis or infection and development in the liver . To further investigate EEF development , BP1-CON and BP1-KO sporozoites were allowed to invade HepG2 cells in vitro . At 60 h post infection , cells were fixed and EEFs were manually counted and their diameter was measured . BP1-KO parasites showed robust development in vitro and no differences in the number or size of EEFs were observed ( Fig 2B and 2C ) . Following this , we imaged the different stages of EEF development in vitro . After an initial growth phase , nuclear division occurs and the parasite membrane invaginates to form the cytomere stage [36] . Subsequently , individual hepatic merozoites bud from each cytomere to form a mature EEF full of hepatic merozoites . MSP1 , the major surface protein of merozoites [37] is observed lining the cytomeres and then localizes to individual hepatic merozoites [38] . At 56 h and 72 h post infection , BP1-CON and BP1-KO EEFs were stained for MSP1 , which showed that the cytomere and late schizont stages in BP1-KO parasites are morphologically indistinguishable to control EEFs ( Fig 2D ) . The numbers of cytomere stage and fully mature EEFs were also counted in these experiments and there were no differences between controls and BP1-KO parasites . Overall , these data suggest that berghepain-1 is not required for liver stage growth or maturation of P . berghei . In order to successfully release infective merozoites into the blood , the parasitophorous vacuole membrane ( PVM ) enclosing the EEF ruptures to release hepatic merozoites into the cytoplasm of the hepatocyte . These then bud from the hepatocyte in packets termed merosomes to enter the blood stream [3 , 39] . Given the involvement of cysteine proteases in both PVM rupture and the release of merosomes [3] , we speculated that berghepain-1 might be involved in this process . We quantified the number of merosomes and detached cells , which contain ruptured EEFs , released into culture supernatants at 65 h post infection and normalized this number to the number of EEFs present at 48 h post infection . For clarity , we will refer to merosomes and detached cells as merosomes throughout the manuscript . After 65 h of in vitro culture , we found that compared to BP1-CON parasites , similar numbers of merosomes were produced by BP1-KO clone 2 ( Fig 3A ) and BP1-KO clone 1 ( S4 Fig ) . To evaluate merosome morphology and loss of the PVM in berghepain-1 knockout merosomes we fixed and stained merosomes with antibodies to MSP1 , to visualize individual merozoites , and UIS4 [up-regulated in infective sporozoites gene 4 [40]] , a marker for the PVM [41] . A previous study demonstrated that PV rupture occurs in the infected hepatocyte and is immediately followed by merosome formation [39] so we would not expect to see UIS4 staining on properly developed merosomes . Staining for MSP-1 showed normal segregation of merozoite membranes ( Fig 3B ) and staining for UIS4 , showed normal loss of the parasitophorous membrane ( Fig 3C ) , suggesting that berghepain-1 knockout merosomes have normal morphology . Together , these data demonstrate that BP1-KO parasites develop into EEFs and form morphologically normal merozoites and merosomes . To test whether the BP1-KO merosomes produce infectious hepatic merozoites , we collected merosomes from in vitro cultures of BP1-CON and BP1-KO parasites , 60–65 h post-infection of HepG2 cells . Upon i . v . injection of five BP1-CON merosomes , mice became positive for blood stage parasites on day 4 after inoculation . In contrast , mice injected with BP1-KO exhibited a significant delay in prepatent period , with mice developing detectable parasitemia on day 9 after injection ( Fig 4A ) . Given that egress of Plasmodium blood- , liver- and mosquito-stages relies on protease activity [42] one possible explanation for this delay is that berghepain-1 may participate in the rupture of the merosome membrane and subsequent egress of the hepatic merozoites . Since infection by Plasmodium is known to alter the stiffness of hepatocyes , reducing their deformability [43] , we hypothesized that the delay in prepatent period of BP1-KO parasites after injection of merosomes could be due to an altered elasticity of the merosome membrane , which would change its ability to rupture . To investigate this , we tested the elasticity of the membrane surrounding BP1-CON and BP1-KO merosomes by atomic force microscopy ( AFM ) . No significant difference in merosome rigidity , represented by the Young’s modulus , a mechanical property of elastic solid materials [43] , was found between populations of BP1-KO clone 1 and clone 2 and BP1-CON merosomes ( Fig 4B ) , suggesting normal elasticity of the merosome membrane of BP1-KO parasites . To further investigate whether reduced infectivity of BP1-KO merosomes was due to impaired merozoite release from the merosomes , we injected 5000 unruptured or mechanically ruptured merosomes into mice . Merosomes were mechanically disrupted with 10 strokes through a 30 gauge needle and immediately injected i . v . into mice . Microscopy of both BP1-CON and BP1-KO merosomes subject to this procedure confirmed that merozoites from 98% of merosomes were released . Upon injection of 5000 ruptured merosomes of the control parasite BP1-CON , parasites were detectable by Giemsa-stained blood smear after an average of 1 . 9 days ( Fig 4C ) . In contrast , injection of the same number of ruptured merosomes of the BP1-KO parasite resulted in detectable blood stage parasitemia at an average of 7 . 7 days after injection , similar to the prepatent period of unruptured BP1-KO merosomes ( Fig 4C ) . Thus , BP1-KO hepatic merozoites , when mechanically released from merosomes , have the same impairment in their ability to establish a blood stage infection as their unruptured counterparts . These data demonstrate that berghepain-1 knockout hepatic merozoites are not adequately primed for erythrocyte invasion , and suggest a critical role for berghepain-1 in preparing hepatic merozoites for the successful infection of red blood cells either during the development of hepatic merozoites within the EEF or at the time of merozoite invasion of the red blood cell , or both . Given the previous finding that falcipain-1 functions during blood stage merozoite invasion of erythrocytes [23] , and our current finding that berghepain-1 , the ortholog of falcipain-1 , likely functions during hepatic merozoite infection of erythrocytes , we set out to characterize the berghepain-1 knockout parasite in the blood stage . After infection with P . berghei ANKA blood stage parasites , parasitemia of susceptible mice typically rises rapidly and mice die within 7–8 days of experimental cerebral malaria , a syndrome characterized by inflammation in the brain and other organs [44–46] . We compared the growth of BP1-CON and BP1-KO parasites by inoculating equal numbers of infected red blood cells i . v . into Swiss Webster mice and monitoring parasitemia and survival of infected mice ( Fig 5 ) . All BP1-CON infected mice showed a rapid increase in parasitemia and death by day 8 post-infection . In contrast , in BP1-KO-infected mice , parasitemia initially increased but then plateaued for a few days ( Fig 5B ) , after which it began to rise more rapidly . Mice inoculated with BP1-KO parasites died between days 14 and 19 post infection with high parasitemias . This is consistent with previous studies demonstrating that attenuation of P . berghei ANKA parasites or manipulation of the host immune response , prevents death from severe malaria [47–51] . Since the mice do not die an early death , the parasites continue to grow until high parasitemias ultimately kill the animal , likely from severe anemia . These data suggest that in addition to the role in priming hepatic merozoites for invasion , berghepain-1 has functions in the erythrocytic stage of the life cycle , consistent with a previous study which found a reduced blood stage growth rate for P . berghei berghepain-1 knockout parasites [13] . The observed lag in parasite growth of BP1-KO asexual stages was reminiscent of the growth pattern observed in the non-lethal strains of P . berghei and P . yoelii , which have a marked preference for invading reticulocytes [52–54] , young erythrocytes newly released from the bone marrow , which account for 1–3% of circulating erythrocytes in a non-anemic animal . For parasites with a reticulocyte preference , the plateau in parasitemia during the acute phase of infection reflects the depletion of available reticulocytes . This is followed by a rise in parasitemia as a consequence of the ensuing anemia , which induces a reticulocytemia [52–54] . We hypothesized that BP1-KO parasites are restricted to reticulocytes in manner similar to the non-lethal rodent malaria parasites . To test this experimentally , we induced a transient reticulocytemia in mice by pretreatment with the hemolytic agent phenylhydrazine ( PHZ ) [54] . Treated mice had more than 48% reticulocytes compared to 1% - 4% reticulocytes in PBS-treated control mice . Infecting PHZ-treated mice with BP1-KO parasites resulted in a rapid increase in parasitemia comparable to BP1-CON during the early stages of infection , achieving high parasite burdens and eliminating the plateau phase observed in PBS-treated mice infected with BP1-KO parasites ( Fig 6A and 6B ) . Nonetheless , the effect of PHZ is temporary and by day 9 post-treatment ( day 6 of infection ) , reticuolocyte counts return to baseline [54 , 55] and parasitemia once again plateaus . Of note , the parasitemia of PHZ-treated mice infected with BP1-CON parasites also rose more rapidly than in PBS-treated mice , likely due to a reticulocyte preference of wild-type P . berghei ANKA parasites . Furthermore , lethality , which is reduced in the BP1-KO parasite , is enhanced by PHZ-treatment: while only 15% of PBS-treated mice infected with BP1-KO had died by day 14 , upon PHZ-treatment , 70% mice infected with BP1-KO parasites had died by day 14 ( Fig 6C ) . These data demonstrate that the growth delay and lethality of BP1-KO parasites can be restored by increasing the reticulocytes available for invasion , suggesting that berghepain-1 is involved in erythrocyte tropism , specifically in mediating infection of mature erythrocytes . To further characterize the reticulocyte preference of berghepain-1 knockout parasites , we inoculated 10 , 000 synchronized BP1-CON and BP1-KO blood stage schizont parasites and counted the number of parasites developing in reticulocytes versus normocytes from days 4 to 7 post infection . Infected reticulocytes were identified by simultaneous staining of blood smears with Giemsa-stain and a stain for reticulin , which is specific for reticulocytes [56] . Parasite infectivity for each of the two erythrocyte populations was determined by expressing the number of infected reticulocytes as a percentage of total infected red cells over days 4 to 7 post infection ( Fig 7A ) . While BP1-CON parasites had a preference for reticulocytes at days 4 and 5 , with over 60% of parasites invading reticulocytes , this percentage dropped to below 20% as the total parasitemia increased ( Fig 7A ) . During early stages of infection , BP1-KO parasites behaved similarly , with 75% of BP1-KO parasites developing in reticulocytes at day 4 , however in contrast to the BP1-CON parasite , this preference only marginally dropped over the following days . These data suggest that reticulocytes are the preferred target cells for both control and BP1-KO P . berghei parasites . However , as reticulocytes are consumed by the infection , only control parasites are readily able to invade normocytes , supporting a role for berghepain-1 in infection of mature erythrocytes . To further confirm that the growth pattern of BP1-KO parasites was due to their reticulocyte restriction , we inoculated BP1-KO and BP1-CON infected red blood cells i . v . into Swiss Webster mice and monitored both parasitemia and reticulocyte counts over time . As expected , control parasites grew rapidly despite low reticulocyte numbers and quickly killed the mice ( Fig 7B ) . In contrast , BP1-KO parasite growth followed the expansion of the reticulocyte pool , initially growing to parasitemias of ~ 2 to 3% , then plateauing and only increasing after the induction of a reticulocytosis ( Fig 7B ) . Since an overall longer cell cycle can also produce a slow-growing phenotype , we set out to investigate whether the slow growth of berghepain-1 knockout blood stage parasites could be due to a change in length of cell cycle . Synchronized blood stage schizonts were injected into a mouse and hourly Giemsa-stained blood smears were used to monitor the transition from G1 to S-phase , which occurs between early and mid-trophozoite stage [57] . Using percent of total parasites that were rings/early trophozoites versus mid/late trophozoites as a readout for cell cycle duration , we found that BP1-KO parasites cell cycle mirrored that of BP1-CON parasites ( Fig 7C ) . It should be noted that the schizont stage of P . berghei parasites adhere to endothelium and are not circulating; thus , after 20 h , numbers of circulating parasites decreased as mature trophozoites developed into schizonts . Our data suggest that the slower growth in BP1-KO parasites is due to a deficiency in infecting erythrocytes rather than slower development . Our data demonstrate that deletion of berghepain-1 gives rise to a phenotype in both blood stage and hepatic stage merozoites . We hypothesized that the role of berghepain-1 in these two distinct populations of merozoites is not equivalent since BP1-KO merosome inoculation results in a prepatent period delay of 5 days compared to control parasites ( Fig 4 ) whereas BP1-KO blood stage parasites are less attenuated ( Figs 5–7 ) . To directly compare the infectivity of BP1-KO hepatic and blood stage merozoites , we compared the onset of detectable parasitemia in Swiss Webster mice inoculated with synchronized schizonts and merosomes . Mice were injected i . v . with 1 , 000 or 10 , 000 purified blood stage schizonts derived from in vitro overnight culture of blood stage BP1-CON and BP1-KO parasites or with 1 , 000 BP1-CON and BP1-KO hepatic merosomes isolated from in vitro liver stage cultures . While schizonts contain between 10 to 14 individual merozoites [58] , merosomes are more variable , harboring between 100 and 1000 hepatic merozoites [3] . Mice injected with 1 , 000 or 10 , 000 BP1-CON blood stage schizonts developed detectable parasitemia by day 4 and 3 , respectively . After injection of the same number of BP1-KO blood stage schizonts , a delay of 0 . 6 days and 1 day , respectively , was seen compared to BP1-CON ( Table 2 ) . This was in stark contrast to mice injected with merosomes: while injection of BP1-CON merosomes led to detectable blood stage parasitemia within one day , we did not detect parasites until 5 . 6 days after BP1-KO merosome inoculation . Thus , the delay to patency of BP1-KO merosomes was greater by ~ 4 days compared to the delay observed with BP1-KO blood stage schizonts . Since the delay in prepatent period after inoculation of BP1-KO hepatic merozoites may in part result from reduced invasion capacity of the erythrocytic merozoites in the ensuing blood stage infection , we attempted to restore growth by inoculating merosomes into mice pretreated with phenylhydrazine . This improved the infectivity of BP1-KO merosomes by ~ 1 . 5 days , suggesting that the patency delay of BP1-KO merosomes reflects the combined delay in both populations of merozoites . However , in contrast to blood stage merozoites , reticulocytosis does not fully restore BPI-KO merozoite infectivity . Overall , these experiments demonstrate that hepatic merozoites of the berghepain knockout parasite are significantly more impaired than erythrocytic merozoites and highlight that while both hepatic and erythrocytic merozoites invade red blood cells , clear differences exist between the merozoites released from the liver and those released from infected red blood cells . To investigate the timing and localization of berghepain-1 expression , we generated a parasite line in which the endogenous berghepain-1 gene was fused to a triple myc tag , a short sequence derived from the c-myc gene ( S5 Fig ) . Using this line , we investigated whether berghepain-1 is expressed at the protein level in blood stage parasites , performing immunofluorescence microscopy of early and late blood stage schizonts . As shown in Fig 8 , berghepain-1-myc is expressed in early schizonts and the staining localizes to the individual merozoites in segmented mature schizonts . We then investigated berghepain-1 expression during liver stage development . Immunofluorescence microscopy of HepG2 cells infected with myc-tagged berghepain-1 parasites showed low levels of berghepain-1-myc expression at 24 h post infection , which increased at 36 h in late hepatic trophozoite stages ( Fig 9A ) . At 48 h post infection , berghepain-1-myc surrounded the individual nuclei and this perinuclear pattern was still present at 56 h ( Fig 9A ) . At 33 h post infection , berghepain-1-myc was also found in larger sub-compartments of the parasites , which co-localized with the staining of the ER marker BiP [59 , 60] ( Fig 9B ) . Berghepain-1-myc did not co-localize with cytosolic marker HSP70 [61] , or the membrane marker MSP1 ( Fig 9A and 9C ) or with CSP ( S6 Fig ) . The specificity of the anti-c-myc staining is demonstrated by the lack of staining of liver stages of the parental control line ( S7 Fig ) . Though these data demonstrate that berghepain-1 is expressed during liver stage development , we did not , however , detect expression of berghepain-1-myc in merosomes , the stage that is attenuated in berghepain-1 deletion mutants ( Fig 9C ) . Thus , it is possible that berghepain-1 functions prior to the formation of merosomes , priming merozoites that will then be packaged into merosomes for exit from the liver . We cannot , however , eliminate a role for berghepain-1 in hepatic merozoites as it’s possible that expression levels in hepatic merozoites are too low to be detected by our methodology . The latter possibility is consistent with expression data of other proteases whose low abundance makes it difficult to detect [62 , 63] .
The function of falcipain-1 , the most highly conserved member of the falcipain family of proteases , has been the subject of some controversy . While an inhibitor of falcipain-1 blocked erythrocyte invasion by merozoites [23] , two independent studies showed that falcipain-1 disruption did not affect growth of blood stage parasites [29 , 30] . Since the rodent model affords a more in-depth analysis of protein function across all life cycle stages of Plasmodium , we disrupted berghepain-1 , the falcipain-1 ortholog of the rodent parasite P . berghei , in an attempt to better understand the role of this protease . Our study revealed that berghepain-1 has a role in erythrocyte infection by both hepatic and erythrocytic merozoites . Furthermore , the impact of berghepain-1 deletion is significantly more pronounced in hepatic merozoites , indicating that hepatic merozoites are not identical to their blood stage counterparts . An important role for berghepain-1 in erythrocyte infectivity by blood stage rodent malaria parasites is supported by several lines of evidence: Berghepain-1 knockouts have a growth delay , consistent with previous work [13] , and reduced lethality , which can be restored , at least temporarily , by increasing the pool of young erythrocytes . The reticulocyte tropism of berghepain-1 knockout parasites was further confirmed by reticulin staining of infected cells , and analysis of cell cycle duration showed normal development of the mutant parasite following invasion , indicating that the defect in berghepain-1 knockout merozoites is specific to one or more steps in the entry process rather than growth . Indeed , previous work showed that asexual blood stages of berghepain-1 deletion mutants produce wild-type levels of hemozoin , suggesting that unlike berghepain-2 , the function of berghepain-1 is not associated with hemoglobin digestion [13] . Our findings are supported by a previous study demonstrating that an inhibitor of falcipain-1 impacts erythrocyte invasion by P . falciparum and the localization of falcipain-1 to the apical end of merozoites [23] . However , two subsequent studies showed that deletion of falcipain-1 did not result in a blood stage growth phenotype [29 , 30] , raising the possibility that there are essential differences between the host cell invasion pathways used by rodent and human parasites . Another possibility is that the selective pressure generated by many rounds of replication during in vitro culture of P . falciparum could select for parasites that are able to compensate invasion defects . Indeed , P . falciparum merozoites can invade erythrocytes using multiple pathways and some of these may not rely on the activity of falcipain-1 . This is supported by our observation that berghepain-1 knockout parasites are not dramatically inhibited in erythrocytic stage growth , indicating that in the rodent parasites as well , alternate invasion pathways are utilized by the BP1-KO parasites . Thus , taken together these data raise the possibility that falcipain-1 and its orthologs have a conserved role across species . Given the reticulocyte tropism of the BP1-KO parasite , we hypothesize that berghepain-1 is involved in infection of mature erythrocytes , possibly by cleaving a parasite ligand required for this process . This could function in initial adhesion to the host cell or in the invasion process , either of which would be consistent with our data . Supporting this hypothesis is evidence that cathepsin L , a falcipain-like protease in the related apicomplexan parasite Toxoplasma gondii , was found to proteolytically mature adhesins as they traffic to micronemes , the specialized secretory organelles whose regulated secretion is essential for invasion [64] . Though the substrate ( s ) for berghepain-1 remain unknown , possible candidates include the rodent malaria 235 kDa rhoptry proteins [65 , 66] , members of the reticulocyte-binding-like ( RBL ) protein family found in all Plasmodium species and known to be involved in erythrocyte invasion . In P . yoelii , Py235 proteins influence host erythrocyte preference and are associated with virulence , with more virulent parasites invading a wider range of erythrocytes [37 , 67 , 68] . Interestingly , distinct subsets of Py235 proteins are expressed in liver and blood stage parasites [69] . Future work involving mass spectrometric approaches that probe for potential cognate substrates of berghepain-1 will shed additional light on the function of this protease . We also found a critical role for berghepain-1 in the pre-erythrocytic stage of infection . The pronounced delay in blood stage infection after sporozoite inoculation suggested that berghepain-1 functions at one or more steps between sporozoite localization to the liver and initiation of blood stage infection . Experiments designed to test each stage of this process revealed that injection of berghepain-1 knockout merosomes could replicate the pronounced delay in blood stage infection after berghepain-1 knockout sporozoite inoculation . Additional experiments with mechanically ruptured merosomes pinpointed the defect to a decreased infectivity of merozoites arising from mature liver stage parasites . Given the expression of BP1-myc during EEF development , this suggests a role for berghepain-1 in preparing hepatic merozoites for infection of red blood cells . Intriguingly , while these two distinct merozoite populations appear morphologically identical and are functionally similar in that both must invade red blood cells , we observed a more significant attenuation of BKO-1 hepatic merozoites compared to their blood stage counterparts . Though we do not yet understand how the same protease differentially impacts these distinct merozoite populations , there are two possible scenarios . One possibility is that the same ligand is processed in blood stage and hepatic merozoites , with this event having a more critical role in hepatic merozoite infectivity . Alternatively , berghepain-1 could have a different role , possibly processing a different substrate , in each of these merozoite populations . Studies with endogenously-tagged berghepain-1 showed that it localizes to merozoites of blood and liver stage schizonts , but is not found in merosomes . These localization data are consistent with either possibility since parasite egress from the mother cell differs in these two merozoite populations . In the blood stage , PV rupture and erythrocyte membrane rupture occur in rapid succession whereas in the liver , the formation of merosomes is an additional step , occurring after PV rupture , and enabling hepatic merozoites to exit the liver sinusoid . Thus , if berghepain-1 acts at some point prior to PV rupture in both merozoite populations , it is not surprising that the merosomes , with already primed merozoites , contain little berghepain-1 . Unfortunately , the small amount of hepatic merozoite material that can be collected combined with the lack of an in vitro infectivity assay for hepatic merozoites , have made it difficult to more precisely determine the hepatic merozoite defect . Future work focusing on identification of the berghepain-1 substrate ( s ) will be critical to elucidating its role in hepatic and blood stage merozoite infectivity . Little is known about whether there are finer-scale differences between blood stage and hepatic merozoites , with only one previous study addressing this topic . This elegant work demonstrated that different Py235 family members are expressed in hepatic versus erythrocytic merozoites [69] . Based on their data , these authors suggested that hepatic and blood stage merozoites differentially rely on distinct invasion pathways , a hypothesis that is supported by our data . This makes sense in light of the different biological niches of each merozoite population . Given the bottleneck of sporozoite transmission , hepatic merozoites likely originate from 1 to 5 infected hepatocytes and as a result , their numbers are 3 to 5 logs lower than their blood stage counterparts [70] . Despite their low numbers , it is essential that hepatic merozoites succeed in invading erythrocytes for if they fail , gametocytes will not be produced for transmission to the mosquito . This is in contrast to blood stage merozoites which are present in large numbers and thus risk killing the host , a scenario which would also jeopardize transmission to the mosquito . Thus , while it is imperative for hepatic merozoites to maximize their infectivity for erythrocytes , blood stage merozoites must walk a line between maintaining infection and not killing the host . Therefore it is plausible that these two populations of merozoites differ in their invasion pathways in ways that are more complex than we can currently appreciate . Future work comparing liver and blood stage merozoites to better understand their differences will help inform the search for suitable drug targets for prophylactic or dual stage drug interventions .
All animal work was conducted in accordance with the recommendations by New York University and Johns Hopkins University Animal Care and Use Committees ( ACUC ) , under the ACUC-approved protocols 110608 , M011H467 and M014H363 . All animal experiments performed at the LUMC were approved by the Animal Experiments Committee of the Leiden University Medical Center ( 12042 ) . The Dutch Experiments on Animal Act were established under European guidelines ( EU directive no . 86/609/EEC regarding the Protection of Animals used for Experimental and Other Scientific Purposes ) . All efforts were made to minimize suffering . Experiments were performed in male and/or female 4- to 6-week-old Swiss Webster or NMRI mice and C57BL/6 mice , purchased from Taconic and Charles River . Male Wistar-Kyoto rats were also used for transfection experiments . Recombinant P . berghei BP1-CON and BP1-KO parasites were generated by double homologous recombination in which the native berghepain-1 locus was replaced with a selection cassette ( BP1-KO ) or a wildtype copy of the berghepain-1 with the selection cassette ( BP1-CON ) . Targeting plasmid pBP1-KO was generated by flanking the human dihydrofolate reductase ( hDHFR ) cassette in plasmid pDEF-hDHFR-flirte [71] with 1 . 6 kb of the berghepain-1 5’ UTR and 1 . 3 kb of berghepain-1 3’ UTR , both cloned from gDNA ( S1 Fig ) . For the control construct pBP1-CON , 1 . 84 kb of the berghepain-1 5’ UTR , 1 . 56 kb of the berghepain-1 ORF and 1 . 33 kb of the berghepain-1 3’ UTR were cloned from gDNA and inserted into pDEF-hDHFR-flirte as outlined in S1 Fig . Transfection was performed using plasmid digested with EcoR1 to liberate the DNA fragment , containing sequence from the 5’ and 3’ UTRs of berghepain-1 , to drive double homologous recombination . P . berghei ANKA parasites clone 507cl1 [31] , were electroporated with 5 μg of digested plasmid DNA , injected into mice , selected with pyrimethamine and cloned by limiting dilution in mice , following standard procedures [72] . The reporter line , PbGFP-Lucschz ( line 1037cl1; www . pberghei . eu mutant RMgm-32; ) was used to generate the transgenic berghepain-1-myc line . In this line , the gfp-luc expression cassette is stably integrated into the 230p locus without introduction of a drug-selectable marker and is under the control of the blood stage schizont-specific ama1 promoter [50] . The berghepain-1 ORF ( without its stop codon ) was PCR-amplified from wild type P . berghei ANKA genomic DNA with primer sets L7424/L7425 ( see S1 Table ) . This PCR product was digested with SpeI and BamHI , and C-terminally fused to a triple c-myc tag by ligation into the SpeI/BamHI digested vector pL1612 , resulting in construct pL2018 ( S5 Fig ) . Prior to transfection , pL2018 was linearized with AflII . Transfection , selection and cloning of transgenic parasites with pyrimethamine were carried out as described previously [72] , generating the transgenic line berghepain-1-myc ( line 2338 ) , expressing endogenously C-terminally tagged berghepain-1 . Anopheles stephensi mosquitoes were reared using standard procedures and fed on Swiss-Webster mice infected with the indicated parasite line . On day 13 after infective blood meal , mosquitoes were dissected and the midguts were observed for oocyst counts using an upright Nikon E600 microscope with a phase contrast PlanApo 10x objective . For salivary gland sporozoite numbers , salivary glands were harvested on day 19 after infective blood meal from 20 mosquitoes and counted on a hemocytometer . Sporozoite gliding motility was assayed as previously described [25] . Glass 8-chambered Lab-tek wells ( ThermoScientific ) were coated with 10 μg/μl mAb 3D11 , specific for the repeat region of the P . berghei circumsporozoite protein [73] , in PBS overnight at 25°C . Salivary gland sporozoites in 3% BSA in Dulbecco's Modified Eagle Medium ( DMEM ) were added to each well and incubated for 1 h at 37°C . Wells were fixed in 4% paraformaldehyde and trails were visualized by staining with biotinylated mAb 3D11 , followed by detection with strepativin conjugated to FITC ( Amersham ) . Trails associated with sporozoites and the number of circles per trail were counted using fluorescence microscopy on an upright Nikon E600 and 40x objective . To examine sporozoite infectivity in vivo , 4- to 6-week-old Swiss Webster or C57BL/6 mice were inoculated i . v . with the indicated number of sporozoites in DMEM . The onset of blood stage infection was determined by daily observation of Giemsa-stained blood smears , beginning on day 3 after inoculation . For intradermal inoculation , mice were lightly anesthetized by intraperitoneal injection of ketamine/xylazine ( 35–100 μg ketamine/g body weight ) and maintained at 37°C on a slide warmer . Sporozoites were injected into the ear pinna , in a total volume of 0 . 2 μl DMEM , with a Flexifill microsyringe ( World Precision Instruments ) . To examine in vivo sporozoite development in the liver , 4- to 6-week-old Swiss Webster or C57BL/6 mice were inoculated i . v . with 10 , 000 sporozoites in 200 μl of DMEM . 40 h later , livers were harvested for total RNA isolation and infection was quantified using reverse transcription followed by real-time PCR , using primers that recognize P . berghei–specific sequences within the 18S rRNA , as outlined previously [74] . Copy number was ascertained by comparison with a plasmid standard curve . Cells of the human hepatoma cell line HepG2 ( ATCC , HB-8065 ) were maintained in DMEM supplemented with 1 mM L-glutamine , 10% Fetal Calf Serum and 5 mg⁄mL penicillin/streptomycin ( complete medium ) at 37°C and 5% CO2 , as previously described [75] . 2 . 5 x 105 HepG2 cells per well were plated onto coverslips coated with collagen I ( BD Biosciences #354236 ) and grown for 8–12 h in 24-well plates . Sporozoites were dissected in DMEM and 4–6 x 104 sporozoites were added per well . After sporozoites were allowed to invade for 2 h at 37°C , free sporozoites were removed by washing with complete medium containing 5μg/mL Fungizone ( Cellgro 30-003-CF ) and 10X penicillin/streptomycin ( wash medium ) , and then maintained in complete medium . Cells were washed twice per day with wash medium until the indicated timepoint , when they were fixed with 4% paraformaldehyde ( PFA ) and mounted . EEFs were observable due to their GFP-expression , and total EEFs per coverslip were manually counted . For immunofluorescence assays , EEFs were stained as outlined below . To quantify the formation of merosomes , HepG2 cells were grown at a density of 50 , 000 cells per well of a 24-well plate and infected as described above . At 50 h post infection , culture supernatant volume was reduced to 0 . 5 ml medium/well . Culture supernatant containing merosomes was collected between 60 and 65 h post infection using a pasteur pipette and counted using a hemocytometer . Merosomes ( numbers depending on the experiment ) were injected i . v . into Swiss Webster mice for prepatent period experiments . For mechanical rupture assays , 25 merosomes per μl in a total volume of 1 ml were sheared by 10 strokes through a 30 gauge needle using a 1 ml syringe and within 5 min were injected i . v . into mice . Samples from ruptured and control merosomes were fixed in 0 . 4% PFA and nuclei were stained with DAPI to allow microscopic analysis of rupture , which revealed that 98% of merosomes were ruptured by the procedure . For atomic force microscopy experiments , infection with BP1-CON and BP1-KO clones 1 and 2 was allowed to proceed in HepG2 cells until 65 h post infection , when merosomes were collected from the culture supernatant . Total medium from two infected wells of a 24-well plate was collected in a 1 . 5 ml tube and merosomes were allowed to settle for 15 min at room temperature . Medium was then carefully removed to leave ~30–50 μl containing the merosomes and 500 μl of 1% PFA was added to fix the merosomes , in order to stop their movement . After 3 min , 1 ml of PBS was added and merosomes were again allowed to settle for 15 min at room temperature . The paraformaldehyde solution was removed , leaving 30–50 μl and 150 μl DMEM was added to the merosomes . Nanoindentation experiments were carried out at 25°C using an atomic force microscope NanoWizard II ( JPK Instruments , Berlin , Germany ) mounted on the top of an Axiovert 200 inverted microscope ( Carl Zeiss , Jena , Germany ) . Measurements were made using non-functionalized OMCL TR-400-type silicon nitride tips ( Olympus , Japan ) . Tip spring constants were calibrated by the thermal fluctuation method , having a nominal value of 0 . 02 N/m . For cell contact , the distance between the cantilever and the cell was adjusted to maintain a maximum applied force of 800 pN before retraction . Data collection for each AFM force-distance cycle was performed at 1 . 5 Hz and with a z-displacement range of 8 μm . The acquired force curves were analyzed using JPK Image Processing v . 4 . 2 . 53 , by the application of the Hertzian model , to obtain the cells Young’s modulus ( E ) . The AFM probe was modeled as a quadratic pyramid , with a tip angle of 35° ( half-angle to face ) and a Poisson ratio of 0 . 50 . For data analysis , multiple readouts of Young’s modulus from a single cell were averaged and the mean used to represent the value for that cell . Rare outlier values above 300 Pa were discarded , though statistical significance of variance between BP1-CON and BP1-KO populations did not change if they were included . For IFAs of EEFs , wells with infected HepG2 were washed and fixed in 4% paraformaldehyde/PBS for 1 h at room temperature . For IFAs of merosomes , supernatants were collected and spun at 50xg onto poly-L-lysine-coated coverslips and fixed for 20 min with 4% PFA at room temperature . Both EEF and merosomes were permeabilized in methanol overnight at -20°C and blocked with 1% BSA/PBS for 1 h at room temperature before incubation with primary and secondary antibodies for 1 h each at room temperature . The following antibodies were used , diluted in 1% BSA/PBS: mouse anti-MSP 25 . 1 diluted 1:500 [37] , polyclonal rabbit anti-UIS4 diluted 1:5000 [41] , mouse anti-CSP at 1 μg/ml [clone 3D11; [73]] , mouse anti-Plasmodium HSP-70 diluted 1:500 [clone 2E6; [61]] , mouse anti-BiP diluted 1:200 [60] , and rabbit anti-c-myc diluted 1:400 ( C3956 , Sigma ) . Secondary antibodies used were anti-mouse Alexa Fluor 488 conjugate ( A11029 , ThermoFisher ) and anti-rabbit Alexa Fluor 594 conjugate ( A11012 , ThermoFisher ) , each diluted 1:500 . Samples were preserved in Prolong Gold mounting medium containing DAPI ( Life Technologies ) . Images for Figs 2 and 3 were acquired using an upright Nikon 90i fluorescence microscope and a 40x objective . Images for Fig 5 , S5 and S6 Figs were acquired using a LSM700 laser scanning confocal microscope ( Zeiss AxioObserver ) with a 63x/1 . 4 PlanApo oil objective using Zen software . Thin blood smears were air-dried , fixed with 4% paraformaldehyde/PBS , permeabilized with 0 . 1% Triton X-100/PBS and blocked with 3% BSA/PBS before incubation with primary and secondary antibodies . Rabbit anti-c-myc antibody ( C3956 , Sigma ) was diluted 1:400 and mouse anti-MSP 25 . 1 [37] was diluted 1:2000 in 1% BSA/PBS . Secondary detection was with anti-mouse Alexa Fluor 488 conjugate ( A11029 , ThermoFisher ) and anti-rabbit Alexa Fluor 594 conjugate ( A11012 , ThermoFisher ) , each diluted 1:500 . Samples were preserved in Prolong Gold mounting medium containing DAPI ( Life Technologies ) and imaged using a LSM700 laser scanning confocal microscope ( Zeiss AxioObserver ) with a 63x/1 . 4 PlanApo oil objective and images were acquired using Zen software . Mice were treated with phenylhydrazine ( PHZ; Sigma-Aldrich , P26252 ) dissolved in PBS pH 7 . 4 , delivered intraperitoneally at 100 μg/g of body weight . Three total doses were given , administered every other day , and mice were injected with infected red blood cells three days after the final dose . To monitor reticulocyte numbers , blood smears were Giemsa-stained and reticulocytes , which stain blue due to residual RNA , were counted as a percentage of total erythrocytes . For culture of P . berghei schizonts , cardiac blood at 2–3% blood stage parasitemia was collected from 2 to 4 Swiss Webster mice and was incubated in RPMI-1640 ( Invitrogen ) supplemented with 10% FCS and gentamycin for 16–23 h , gently shaking at 80 rpm in culture flasks that were flushed with 5% CO2 , 5% O2 , 90% N2 as described previously [72] . To quantify the number of parasites developing in reticulocytes versus normocytes , mice were inoculated i . v . with 10 , 000 BP1-CON and BP1-KO schizont stage parasites , obtained from in vitro culture as described above . Blood smears of infected mice were stained using a Brilliant cresyl blue and Giemsa double staining technique described previously [56] . Briefly , microscope slides were coated with 0 . 3% Brilliant Cresyl Blue ( BCB ) in 95% ethanol and dried overnight . These slides were incubated for 15 min at room temperature to allow absorbance of BCB , followed by methanol fixation and standard Giemsa staining [56] . Parasites in BCB-staining cells and cells not stained with BCB were counted . Growth assays for cell cycle determination were started with i . v . injection of 108 synchronized schizonts , obtained from overnight culture of BP1-CON and BP1-KO blood stage parasites as above . Giemsa-stained blood smears were performed hourly for the next 30 h and scored as to the percent of total infected cells that were rings/early trophozoites versus mid/late trophozoites . Since the G1-S transition in blood stage Plasmodium parasites occurs as the parasite transitions from early and mid-stage trophozoite [57] , these counts are indicative of the time it takes for the parasite to go through its cell cycle . At ~ 21 hours , schizonts begin to develop and their sequestration meant we could only follow parasite growth up to this time [76] . Though P . berghei is synchronous for up to 2 cycles [77] , we found that the synchronicity of the second cycle was not as tight , making it difficult to obtain accurate cell cycle data after the first cycle . | Malaria affects hundreds of millions of people and is the cause of hundreds of thousands of deaths each year . Infection begins with the inoculation of sporozoites into the skin during the bite of an infected mosquito . Sporozoites subsequently travel to the liver , where they invade and replicate in hepatocytes , eventually releasing the stage of the parasite that is infectious for red blood cells , termed merozoites . Hepatic merozoites initiate blood stage infection , the stage that is responsible for the clinical symptoms of malaria . The blood stage of the parasite grows through repeated rounds of invasion , development and egress of blood stage merozoites , which then continue the cycle . Proteases are among the enzymes that are essential for parasite survival and their functions range from invasion of red blood cells , to the breakdown of red cell hemoglobin , to the release of parasites from red cells . As the function of the cysteine protease falcipain-1 in the lifecycle of the human malaria parasite Plasmodium falciparum remains poorly understood , we decided to study berghepain-1 , the orthologue of the rodent malaria parasite P . berghei by generating a berghepain-1 deletion parasite . Using this mutant , we demonstrate that berghepain-1 has a critical role in both hepatic and erythrocytic merozoite infectivity . Little is known about differences between these two types of merozoites and our data leads us to conclude that these merozoites are not identical . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"body",
"fluids",
"cloning",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"parasitemia",
"apicomplexa",
"protozoans",
"red",
"blood",
"cells",
"molecular",
"biology",... | 2017 | Deletion of the rodent malaria ortholog for falcipain-1 highlights differences between hepatic and blood stage merozoites |
Mirtrons are microRNA ( miRNA ) substrates that utilize the splicing machinery to bypass the necessity of Drosha cleavage for their biogenesis . Expanding our recent efforts for mammalian mirtron annotation , we use meta-analysis of aggregate datasets to identify ~500 novel mouse and human introns that confidently generate diced small RNA duplexes . These comprise nearly 1000 total loci distributed in four splicing-mediated biogenesis subclasses , with 5'-tailed mirtrons as , by far , the dominant subtype . Thus , mirtrons surprisingly comprise a substantial fraction of endogenous Dicer substrates in mammalian genomes . Although mirtron-derived small RNAs exhibit overall expression correlation with their host mRNAs , we observe a subset with substantial differences that suggest regulated processing or accumulation . We identify characteristic sequence , length , and structural features of mirtron loci that distinguish them from bulk introns , and find that mirtrons preferentially emerge from genes with larger numbers of introns . While mirtrons generate miRNA-class regulatory RNAs , we also find that mirtrons exhibit many features that distinguish them from canonical miRNAs . We observe that conventional mirtron hairpins are substantially longer than Drosha-generated pre-miRNAs , indicating that the characteristic length of canonical pre-miRNAs is not a general feature of Dicer substrate hairpins . In addition , mammalian mirtrons exhibit unique patterns of ordered 5' and 3' heterogeneity , which reveal hidden complexity in miRNA processing pathways . These include broad 3'-uridylation of mirtron hairpins , atypically heterogeneous 5' termini that may result from exonucleolytic processing , and occasionally robust decapitation of the 5' guanine ( G ) of mirtron-5p species defined by splicing . Altogether , this study reveals that this extensive class of non-canonical miRNA bears a multitude of characteristic properties , many of which raise general mechanistic questions regarding the processing of endogenous hairpin transcripts .
MicroRNAs ( miRNAs ) are an extensive class of ~22 nucleotide ( nt ) regulatory RNAs , generated from hairpin precursors , that figure prominently into post-transcriptional regulatory networks in diverse eukaryotes [1 , 2] . Biochemical studies of the initially recognized set of miRNAs revealed their biogenesis by a common mechanism [3] . For example , miRNAs in animals are typically generated via stepwise cleavage by two RNase III enzymes: nuclear Drosha cleaves primary miRNA transcripts into pre-miRNA hairpins , while cytoplasmic Dicer cleaves pre-miRNAs into small RNA duplexes [4] . Following duplex loading into an Argonaute ( Ago ) protein , one strand of the duplex ( the mature miRNA ) is preferentially retained relative to its partner strand ( the miRNA* or "star" species ) , and guides the Ago complex to target transcripts [5] . Following the delineation of the canonical miRNA biogenesis pathway , non-canonical loci were recognized to be independent of enzyme ( s ) believed obligate for the production of functional miRNAs [6] . The major alternative pathway utilizes splicing to generate pre-miRNA hairpin mimics termed mirtrons , thereby bypassing Drosha [7 , 8] . Mirtrons come in several flavors ( Fig 1A ) , depending on how directly the pre-miRNA hairpin ends are defined by splicing [9] . With conventional mirtrons , both hairpin ends are generated by splicing and form a short 3' overhang that is appropriate for nuclear export and dicing . With 3'-tailed mirtrons , the 5' hairpin end corresponds to the 5' splice donor , but an unstructured region follows to the 3' splice site . Where studied in Drosophila , the 3' tail is resected by the RNA exosome to generate the pre-miRNA substrate [10]; this mechanism has not yet explicitly been shown in vertebrates . Mirtrons with an opposite polarity also exist , namely 5'-tailed mirtrons where an unstructured region precedes a hairpin that resides precisely at the 3' end of the intron [11–13] . The enzyme ( s ) responsible for resection of 5' tails has not yet been identified . Recently , two intronic miRNAs ( mir-3062 and mir-7661 ) were proposed to be dependent on splicing , even though neither of their hairpin termini directly abut splice sites [14] . For such "two-tailed mirtrons" , neither the nucleases nor the order of tail removal is known . Non-canonical miRNA loci are often collectively considered as a minor class , as bulk miRNA reads in most cells and tissues derive from canonical loci . Nevertheless , there are situations where reads from non-canonical loci are not only substantial , but can even be functionally predominant . For example , the unusual Dicer-independent locus mir-451 obligately depends on slicing by Ago2 for its maturation [15–17] , and this miRNA is one of the top-expressed miRNAs in erythrocytes [18] . As another example , while miRNAs are present in mouse oocytes , they appear to be functionally inactive in this setting [19 , 20] , leaving endo-siRNAs as the main functional Dicer-substrate small RNA in mouse oocytes [21–23] . The net functional contribution of non-canonical miRNA pathways is further hinted at by phenotypic discrepancies amongst mutants of different core components of the canonical miRNA biogenesis pathway [24] . Another way to evaluate the contribution of non-canonical miRNA loci is by the numbers of confidently annotated loci . The number of miRNA genes in any species is under constant revision , and depends on the types of data and stringency of annotation applied . Some studies of mammalian miRNAs have varied from ~500 [25] to several thousand [26 , 27]; these studies have typically focused on canonical miRNA loci . Although these estimates range over many fold , they are in any case only a small fraction of the 100 , 000s to millions of plausible miRNA-like hairpins that can be computationally predicted in mammalian genomes . Since the majority of these genomes is transcribed [28 , 29] , this suggests that the general limitation of genomic hairpins to generate miRNAs is not at the level of transcription , but instead reflects our insufficient knowledge of features required for miRNA biogenesis . Perhaps surprisingly , then , we recently used strigent annotation criteria to report well over 400 novel loci that generate splicing-derived miRNAs in mouse and human [13] . Therefore , even though the levels of small RNAs generated by most mirtrons is modest , they actually comprise a substantial fraction of clear Dicer-substrate loci in mammals . Of note , the levels of bulk mirtrons are comparable to many hundreds of other recently-emerged mammalian miRNAs [13] . These observations suggest that non-canonical pathways may have substantial impact on the dynamics of miRNA evolution and species-specific regulation [13 , 30] . In this study , we mined additional deep sequencing data to identify ~500 novel mammalian mirtrons using stringent criteria . We use this expanded set of ~1000 loci to identify the characteristic sequence , structural , and genomic features of mammalian mirtrons . Although mirtrons generate miRNA-class regulatory RNAs via pre-miRNA dicing , as with canonical miRNAs , we find that mirtron hairpins and their small RNAs have a multitude of distinctive properties compared with canonical miRNAs . These reveal novel aspects of endogenous Dicer substrates that are recognized when Drosha is not the initiating nuclease , and implicate unexpected pathways that process the 5' and 3' ends of pre-miRNAs .
We recently analyzed a large collection of mouse and human small RNA data to identify nearly 500 mirtrons , i . e . loci that generate splicing-derived miRNAs [13] . We defined these as intronic hairpins for which one or both hairpin termini abut intron junctions , and that are associated with confident evidence for endogenous dicing into small RNA duplexes . These predominantly comprised 5'-tailed mirtrons , with smaller numbers of conventional mirtrons and a minor set of 3'-tailed mirtrons . Our previous study also classified many candidate splicing-dependent hairpins with more limited expression evidence , and these followed a similar distribution of mirtron subtypes as the confident loci [13] . Given the compelling read evidence of many of these candidates , which nonetheless did not meet our stringent annotation criteria , we inferred many of them might be considered genuine mirtrons with additional small RNA data . Notably , even with recent state-of-the-art efforts for mammalian miRNA annotation [26] , the distinct structural features of mirtrons ( i . e . short hairpins ) means that they have generally escaped efforts for miRNA annotation to date . Moreover , the recognition of "two-tailed mirtrons" [14] and mirtron trimming pathways [31] suggested that it is necessary to consider broader annotation criteria to fully capture the diversity of splicing-dependent small RNAs ( Fig 1A ) . Therefore , we also paid attention to intronic read pileups in proximity to , but that did not directly abut , splice junctions . Many additional small RNA datasets have since become available , covering a broader range of tissue and cell types than we considered previously . We processed 189 additional human and 125 mouse datasets , and aggregated these with other data ( S1 Table ) ; in total over 1000 deeply sequenced libraries . The newly-analyzed data include many Ago-IP libraries , which have particular power to increase annotation confidence . This was particularly relevant to mouse , for which we incorporated substantial Ago-IP datasets from brain [32] , T cells [33] , skin cells [34] and NIH-3T3 cells [35] . We then re-ran our pipeline to identify candidate mirtrons , and vetted these extensively by individual inspections of their read evidence . Because of the greater read depth and library variety now analyzed , we were able to increase the stringency of our annotation criteria from our previous study [13] ( see Materials and Methods ) . Some loci were supported by ancillary layers of evidence for their biogenesis . For example , for the novel mirtron located in human HSPA8 , we recovered not only precise miRNA and star reads , but also the phased loop read representing the Dicer byproduct ( Fig 1B ) . Notably , while both its miRNA and star reads accumulated in Ago2-IP data , its cloned loop species did not . Although special hairpin loops associate with Ago proteins [36 , 37] , pre-miRNA loop species are typically rejected from Argonaute complexes . We also recognized needs for flexible annotation criteria . In particular , the existence of substantial reads in Ago-IP libraries seemed appropriate to classify miRNA-generating loci , even when certain other features were suboptimal . For example , we typically demand that both miRNA and star species be multiply cloned to annotate mirtrons and canonical miRNAs with confidence [13 , 38] . However , some highly cloned reads from intron terminal hairpins exhibited few companion star species , and therefore did not meet our previous duplex criterion . Although it seemed prudent to consider these only as "candidates" , strongly asymmetric strand selection might preclude accumulation of star species [39 , 40] . We elevated the status of a small set of candidate loci with modest numbers of star species , if they generated ≥100 total mature arm reads of which ≥20 reads were from Ago-IP datasets ( see Materials and Methods ) . Fig 1C depicts a novel mirtron hairpin within human U2AF2 with hundreds of mature reads in aggregate data , including >100 Ago2-IP reads , but only one star read . In total , the greater library depth and Ago-IP datasets allowed us to "upgrade" many mirtrons from our previous lists of candidate loci [13] . As well , we identified some mirtrons with exquisite tissue-specificity , which required the analysis of new libraries ( e . g . Fig 1D and 1E ) . Reciprocally , our refreshed efforts barely necessitated that we "downgrade" any previously annotated mirtrons; that is , nearly all of them looked even more robust with larger data . The vast majority of novel mirtrons in both human and mouse are 5'-tailed loci , with modest numbers of conventional mirtrons and minor sets of 3'-tailed loci and two-tailed loci . In particular , we expand the latter class from two reported earlier to 24 . A robust example of a novel two-tailed mirtron is shown in Fig 1F , for which small RNAs from ECEL1 dominantly accumulated in Ago-IP data . In total , we confidently identify 478 human ( 242 novel to this study ) and 488 mouse ( 248 novel to this study ) loci as generating splicing-derived , diced small RNA duplexes . We summarize the tallies of the various classes of mirtron loci in human and mouse in Fig 2A . Additional information regarding the evidence supporting these loci is provided in S2 and S3 Tables , and extensive information on the reads and libraries that map to all human and mouse mirtrons can be explored in the Supplementary Websites . Curiously , hardly any of these stringently-annotated mirtrons are conserved between human and mouse ( Fig 2B and S1 Fig ) . Even amongst the modest number of positionally conserved mammalian mirtrons that generate small RNAs , only in a few of these cases do the mirtron hairpins exhibit a classic "saddle-shaped" pattern of evolution indicative of evolutionary constraint as trans-regulatory species [41 , 42] ( S2 Fig ) . We emphasize our conservative procedures in annotating novel mirtrons . Beyond incorporation of strict duplex evidence , the minimum numbers of reads supporting our mirtron annotations are comparable to or exceed those of many extant canonical miRNAs ( Fig 2C and 2D ) . In fact , 320 canonical human and mouse miRNA loci in the miRBase repository have less read support than our set of ~1000 mirtrons . As well , the vast majority of our mirtrons contribute reads to Ago-IP libraries ( Fig 2C and 2D ) . To investigate this in more detail , we selected human and mouse datasets with rigorous control and multiple Ago-IP libraries , and compared the behavior of canonical miRNAs and mirtrons . Meister and colleagues analyzed endogenous human Ago1-4 contents from HeLaS3 cells , and compared these to total RNA as input and control IP using rat RmC antibody [43] . Since Ago4 is not expressed in these cells , Ago4-IP and RmC-IP show the same background levels of miRNA accumulation , whereas input RNA and Ago1/2/3-IP datasets all show enrichment for canonical miRNAs ( Fig 2E ) . The similar representation of miRNAs in the input and Ago-IP datasets is likely due to the fact that canonical miRNAs are the predominant small RNA species in these cells , regardless of Ago purification . Although mirtrons are overall lower-expressed , we find that they actually exhibit more informative segregation across these datasets . That is , RmC-IP and Ago4-IP showed almost no mirtron reads , whereas the input library had intermediate levels relative to Ago1/2/3-IP libraries ( Fig 2E ) . Therefore , mirtron-derived small RNAs are genuinely enriched in Ago complexes . We also examined data from Corey and colleagues , who generated IgG- , Ago1- and Ago2-IP datasets from mouse CD4+ T cells [33] . Although mirtrons accumulated to much lower levels than canonical miRNAs , both classes of miRNAs were significantly enriched in both Ago1-IP and Ago2-IP datasets relative to IgG-IP library ( Fig 2F ) . The behavior of individual canonical miRNA and mirtron loci in these control and Ago-IP datasets is provided in S4 and S5 Tables . Together , these analyses provide evidence that mirtrons generally generate small RNAs with the biochemical properties of genuine miRNAs . Still , as our effort relied upon a large meta-analysis , a question arises how many loci are due to aggregation of a few reads in each of many libraries . To address this , we plotted the maximum number of reads mapped to mouse and human mirtrons across individual libraries ( S3A and S3B Fig ) . These analyses showed a few dozen mirtrons fit the bill of having only single-digit reads in any particular dataset , even though all passed a 50 read minimum cutoff . It might be that some mirtrons are expressed lowly , but tissue-specifically . We attempted to address this scenario by grouping similar libraries-of-origin and repeating the above analysis . This treatment of data substantially reduced the numbers of these most modestly-represented mirtrons ( S3C and S3D Fig ) , consistent with the notion that many of them might be spatially restricted . We explored this further by comparing the library distribution of mirtrons and canonical miRNAs . We observed that mouse and human mirtrons were biased to present their dominant expression in a small number of libraries , whereas canonical miRNAs tended to accumulate more broadly across many libraries ( S3E and S3F Fig ) . The interpretation of this analysis may be biased by the overall lower accumulation of mirtrons relative to canonical miRNAs . Nevertheless , we observed many mirtrons with exquisite tissue restriction . For example , even though we analyzed many hundreds of human and mouse libraries , small RNA reads from mirtrons in mouse Pkp1 ( Fig 1D ) and human Titin ( Fig 1E ) were recovered nearly exclusively from skin Ago-IP or heart total RNA data , respectively . In summary , the splicing pathway contributes strongly to the cellular pool of endogenous Dicer substrates in different mammals . The overall view is that the many hundreds of mirtrons are expressed in a modest range , but a range that encompasses a substantial set of modestly-expressed mammalian miRNAs . We subsequently performed all of our analyses independently on these mostly non-overlapping sets of mouse and human mirtrons , from which we could derive general features of mammalian mirtrons and compare them to Drosha-substrate miRNAs . The majority of intronic miRNAs are found on the sense transcribed strands of host protein-coding transcripts [44] , and the accumulation of intronic miRNAs and host mRNAs is overall correlated [45] . Nevertheless , intronic miRNAs are not necessarily produced in the course of transcribing host protein-coding genes , since miRNA biogenesis can be regulated post-transcriptionally . Conversely , some miRNAs might be transcribed from intronic promoters , and thus their expression might be decoupled from their inferred host genes [46 , 47] . Mirtron biogenesis is presumably more intimately coupled with the maturation of their host genes . Nevertheless , their accumulation might not necessarily mirror each other , for the same reasons that apply to canonical miRNAs . To test the expression correlation of mirtrons with their host genes , we utilized publicly available datasets of tissue-specific mouse and human mRNA-seq data ( see Materials and Methods ) . We began by comparing the correlations of mRNA and small RNA accumulation across tissues . When plotted as cumulative frequency distributions , we observed strong positive correlations between mRNA and miRNA in both mouse ( p<5 . 05E-9 ) and human ( p<1 . 62E-5 ) data , compared to the background distributions obtained when cognate tissue identities were shuffled ( S4 Fig ) . Given that these analyses involved tissue samples and expression data prepared independently and that may not be directly comparable , the extent of correlations is likely an underestimate . One way the true correlation might be underestimated in the above analysis , utilizing the mRNA expression of all exons , would be if mirtrons were generated from specific mRNA isoforms . We attempted to remedy this by performing correlation analysis only with spliced mRNA reads that span exon-exon junctions across mirtrons . This analysis generated a positive correlation in the human data , but it was less significant than with the gene level analysis . We can easily rationalize this , however , due to undersampling , since we observed some human loci with no spliced RNA-seq reads across a given mirtron locus ( S5 Fig ) . However , the available mouse RNA-seq data ( ~3 . 8 billion from 7 tissues; ~550 million reads/tissue ) , were much deeper than the human data ( 370 million reads from 6 tissues; ~61 million reads/tissue ) . Indeed , the superior depth of the mouse data for spliced reads across mirtrons substantially improved the correlation of host gene-mirtron expression . As shown in the CDF plot ( Fig 3A ) , there is a strong bias for well-correlated pairs ( p<9 . 97E-12 ) . A similar conclusion can be seen via the binned data plot ( Fig 3B ) . Interestingly , this visualization makes apparent a subset of loci for which the accumulation of spliced flanking exon reads and mirtron-derived miRNAs are particularly uncorrelated ( Fig 3B ) . We present some individual cases of highly correlated and uncorrelated host mRNA/mirtron mouse expression profiles in Fig 3C and show additional mouse and human examples in S5 Fig . In many of these instances , we observe that the host gene is expressed more broadly than where the mirtron-derived small RNAs accumulated . Such cases are suggestive of post-transcriptional regulation of mirtron biogenesis . In analyzing the structural features of mirtrons , we sought properties that might distinguish them from other mammalian introns ( i . e . , "bulk" introns ) . We compared the length distribution of introns that hosted mirtrons and tailed mirtrons relative to all other introns , plotting them in bins of 100 nt and pooling remaining introns >15 kb and >10 kb in length in human and mouse , respectively . In both the human and mouse genomes , the number of introns is greatest for the 100–200 nt window , gradually tailing off from there with increasing lengths ( Fig 4A and S6A Fig ) . Conventional mirtrons were mostly restricted to the <100 nt window , in both mouse and human . Although the certain 3'-tailed mirtron loci ranged into larger sizes , these mostly derived from similar sizes as conventional mirtrons . Therefore , although the number of 3'-tailed loci was modest , their similar length properties in both mammalian species suggests that the capacity for 3' trimming does not dramatically increase the range of introns that mirtrons can occupy . The 5'-tailed mirtrons exhibited distinct properties from the other mirtron classes . On the one hand , they also exhibited a preference for relatively smaller introns in both human and mouse . Amongst all 100nt bins of intron sizes , the largest numbers of 5'-tailed mirtrons derived from <100nt and 100–200 nt introns . Still , there were significant populations of 5'-tailed mirtrons >1 kb , with ~90% of these loci cumulatively derived from introns up to 3 kb in length ( Fig 4A and S6A Fig ) . Despite this flexibility in length , the frequency of 5'-tailed mirtrons in progressively larger introns decreased more quickly than the overall length profile of bulk introns , with 90% of all introns are contained within the <10 kb range ( Fig 4A and S6A Fig ) . This implies that 5'-tailed mirtrons arise more readily in introns of relatively smaller length , mostly <3 kb . The length bias of 5’-tailed mirtrons suggests that the 5' resection pathway is progressively less effective on tails longer than a few kb . Based on our identification of several examples of potentially alternative unannotated splicing of mirtron precursors , it is possible that some instances of 5'-tailed mirtrons in long annotated introns may actually derive from alternative splicing of shorter introns . The longest annotated host intron of a 5'-tailed mirtron belongs to mir-5129 , located in an 87 kb intron of mouse Zeb2 [48] . Given that the next largest host intron of a tailed mirtron is ~36 kb ( mouse Klf7 ) , and the vast majority of 5'-tailed mirtrons reside in introns of up to a couple kb , one wonders whether mir-5129 is generated directly from splicing of such a long intron . Examination of the UCSC reveals a population of uncharacterized Zeb2 ESTs that utilize an internal 5' splice site joining the mir-5129 3' splice site , resulting in a ~32 kb intron . While this is still rather extreme , it supports the possibility that "long" introns bearing 5’-tailed mirtrons may be subject to alternative splicing of shorter unannotated introns , or perhaps , some type of recursive splicing [49 , 50] . We next examined the primary sequence properties of mirtrons . We were again drawn to the 5'-tailed mirtrons because of their abundance . We generated sequence logos of 422 human and 429 mouse 5'-tailed mirtron hairpins that were anchored at three different locations: by the 5' end of the 5p species , by the 5' end of the 3p species , and by the 3' end of the 3p species ( i . e . , by the AG splice acceptor ) . We compared these to the nucleotide content of bulk introns 50–3000 nt in length , which comprise the strong majority of 5'-tailed mirtron host intron lengths . To provide a proxy for the start of the mirtron hairpin in bulk introns , we aligned these introns from the -65 position from the end of the intron , corresponding to the average distance of the start of 5'-tailed mirtrons from their intron ends in human/mouse . Several characteristics of 5'-tailed mirtrons emerged from this analysis . First , the 5' ends of both 5p and 3p species exhibit 5' uridine for a majority of loci in both human ( Fig 4B ) and mouse ( S6B Fig ) . Such 5' U bias is typical for canonical miRNA species , but was not observed in comparable regions of bulk introns . These data support the notion that these processed small RNAs of 5'-tailed mirtrons have been selected according to similar rules as canonical miRNAs , likely including the preference of the MID domain of mammalian Ago2 to select small RNAs with 5'-U [51] . Second , the duplex regions of 5'-tailed mirtrons are characterized by high guanine content on their 5p arms and high cytidine content on their 3’ arms , resulting in a high degree of G:C pairing ( Figs 4B and 4C and S6B and S6C ) . Although the 3’ ends of introns generally contain polypyridimine tracts , they exhibit somewhat greater uridine bias , whereas the 3' ends of 5'-tailed mirtrons exhibit substantial cytidine enrichment in human ( Fig 4C ) and mouse ( S6C Fig ) . The region of bulk introns from -65 to -35 of the splice acceptor exhibits little nucleotide bias , making the enrichment of G residues in 5'-tailed mirtron-5p arms particularly striking . The high GC content of all classes of mirtrons suggested that they adopted much more stable structures than bulk introns . We compared all classes of mirtron hairpins with a length-matched distribution of control introns ( see Materials and Methods ) . These analysis showed that mirtrons indeed collectively exhibit far lower minimum free energy ( MFE ) per base than do non-mirtronic introns in human ( Fig 4D ) and mouse ( S6D Fig ) . Altogether , these observations provide evidence that mammalian mirtrons and tailed mirtrons emerge from an intron subpopulation with length and sequence characteristics that are distinct from bulk mammalian introns . We next studied the numbers of introns in genes that host mirtrons . The null hypothesis might be that mirtrons are equally eligible to arise in most genes , with some expectation that genes with more introns might have more chances to harbor an intronic non-coding RNA locus . Directed expression tests in Drosophila [7 , 8 , 52] and mammalian cells [13 , 53 , 54] demonstrate that functional mirtrons can be generated from single intron constructs , and the average mammalian genes have ~7 introns , which might provide reasonably abundant opportunities for their emergence . However , we observed that the average numbers of introns in mirtron-hosting genes far exceeded the genomewide averages in both human ( Fig 4E ) and mouse ( S6E Fig ) . In fact , these trends were true for all three classes of splicing-derived miRNAs , as conventional mirtrons , 5'-tailed mirtrons , 3'-tailed mirtrons , and two-tailed mirtrons all derived from host genes that bore 2 . 5–3 times the average numbers of introns ( p = 1 . 9E-116 and p = 6 . 4E-118 in human and mouse , respectively ) . This may imply an evolutionary advantage for the potential of mirtron emergence in genes with larger numbers of introns . We analyzed in detail the distribution of intron numbers for protein-coding host genes for the different mirtron classes and for a selection of other non-coding RNAs ( ncRNAs ) , including canonical miRNAs , snoRNAs and tRNAs . In particular , since mirtrons are generally recently-emerged , we split up the canonical miRNAs into ones that are well-conserved across mammals and ones that are specific to rodents or to primates . Both classes of miRNAs and the other ncRNAs derive from genes with intermediate numbers of introns , relative to the different classes of mirtrons and bulk genes in human ( Fig 4E ) . The same trends were also observed in mouse ( S6E Fig ) . We gain additional information by binning each class of non-coding element . We clearly observe that the distribution of host gene intron numbers for all classes of mirtrons is larger and broader than for bulk genes . In contrast , while the distribution of intron numbers for genes that host various types of ncRNAs is broader than that of bulk genes , the preferred host intron numbers for tRNAs , snoRNAs and conserved miRNAs overlaps substantially with bulk genes ( Fig 4F and S6F Fig ) . Therefore , all classes of mirtrons appear to emerge preferentially from a population of protein-coding genes that is distinct from the habitat of many other types of intronic ncRNAs . Plant miRNAs exhibit great heterogeneity in their hairpin lengths , with some spanning hundreds of nucleotides [3] . Select canonical miRNA hairpins in Drosophila are also exceptionally long [55] . For example , the pre-mir-989 hairpin is 143 nt in length , inclusive of the cloned miRNA/star species . We emphasize that mir-989 is a canonical miRNA that generates a typical AGO1-loaded miRNA , since some other "long miRNAs" in Drosophila are actually members of the hpRNA class that generates endo-siRNAs [56] . In mammals , pre-miRNAs for confidently annotated loci rarely exceed 80 nt ( http://www . mirbase . org/ ) . Potential explanations for this restriction are that it reflects preferred substrate requirements of Dicer , or perhaps a need to avoid inadvertent activation of the interferon pathway , which recognizes dsRNA non-specifically . Amongst miRNAs annotated prior to 2012 , the only pre-miRNA longer than 100 nt ( i . e . the hairpin length inclusive of the cloned miRNA/star species ) were the human loci mir-548o and mir-1236 , and the mouse loci mir-3102 , mir-702 and mir-1983 . The veracity of mir-548o is potentially questionable since all reads ascribed to this locus map to multiple genomic locations , many of which exhibit more compact hairpins ( http://www . mirbase . org/ ) . Curiously , the remaining >100 nt pre-miRNAs are all Drosha-independent , with mir-1983 deriving from a tRNA and the others deriving from splicing [11 , 25 , 42] . These observations might imply that alternative sources of pre-miRNAs may have more flexible hairpin lengths than does the canonical biogenesis pathway . With this in mind , we were struck by the number of recently-annotated mouse and human mirtrons ( this study and [13] ) with pre-miRNA hairpins in excess of 100 nt . Fig 5A and 5B highlight cases of exceptionally long pre-miRNA hairpins in conventional and two-tailed mirtrons . Although the hairpin structure usually deteriorates distal to the miRNA/star duplex , we still observed specific small RNA duplexes that link genomically distant miRNA/star species . These loci add to the catalog of non-canonical loci in the upper echelon of mammalian pre-miRNA lengths , and suggested that splicing can generate a broader distribution of hairpin lengths than can Drosha . We analyzed this systematically by comparing the length distributions of canonical pre-miRNAs and mirtrons . As expected , the distribution of Drosha-generated pre-miRNAs in mouse ( Fig 5C ) and human ( Fig 5D ) is tightly centered around 60 nt ( mean±SD: 60±4 . 7 nt and 60±4 . 3 nt for human and mouse , respectively ) . Curiously , although we observe examples of atypically long pre-miRNAs in diverse mirtron biogenesis subtypes ( e . g . , Fig 5A and 5B ) , the dominant signal for long pre-miRNAs was contributed by conventional mirtrons . We do see that the distribution of 5'-tailed ( mean±SD: 63±9 . 9 nt for human and 62±8 . 9 nt for mouse ) , 3'-tailed mirtron ( mean±SD: 61±7 . 8 nt for human and 64±10 . 2 nt for mouse ) , and two-tailed mirtron ( mean±SD: 69±24 . 4 nt for human and 64±3 . 7 nt for mouse ) pre-miRNA hairpins is somewhat broader and more extreme , and includes members that extend into an exceptionally long range ( >100 nt , Fig 5C and 5D ) . The length distribution of 5'-tailed mirtrons is statistically significant from canonical pre-miRNAs ( p = 2 . 4E-7 and p = 8 . 5E-4 for human and mouse , respectively ) since the set of canonical miRNAs is several fold larger , but lacks any extreme pre-miRNA lengths . Still , the average lengths of pre-miRNAs from tailed mirtrons are ~60 nt in both species . By contrast , the lengths of conventional mirtron pre-miRNA were noticeably increased in both mouse ( Fig 5C ) and human ( Fig 5D ) datasets . Their average lengths were 80–85 nts , a range almost never broached by canonical miRNA hairpins . Moreover , their length distribution is much more heterogenous than any other miRNA classes ( mean±SD: 83±9 . 7 nt and 87±18 . 5 nt for human and mouse , respectively ) . Comparison of conventional mirtron and canonical miRNA pre-miRNA are highly statistically significant ( p = 1 . 1E-14 for human and p = 3 . 8E-11 for mouse ) . Therefore , the long-standing observation of a typical length restriction of canonical pre-miRNA hairpins is not a universal feature of miRNA substrates . Instead , we infer that the conventional mirtron class of Drosha-independent hairpins accesses greater structural freedom relative to other miRNA substrates , including other classes of splicing-derived miRNAs , whilst maintaining capacities for Dicer cleavage . Confident annotation of miRNAs rests heavily on the terminal specificity and precision of the mature species , and stringent criteria are necessary to distinguish genuine miRNAs from amongst a sea of genomic hairpins whose incidental transcription and random degradation generate cloned short RNAs [3 , 30] . However , the recognition of abundant genomic substrates of mirtron resection pathways raises new complexity in defining the precision of “confident” miRNAs . In particular , the implied involvement of exonucleases in the biogenesis of tailed mirtrons may induce additional heterogeneity than is typical for canonical miRNAs . We analyzed terminal heterogeneities at the 5' ends and 3' ends of 5p-arm and 3p-arm reads from the three classes of mirtrons , and compared these to canonical miRNAs ( Fig 6 ) . We indeed observed that specific mirtron classes exhibit substantially increased terminal heterogeneity . For example , we observe that the 5' ends of 5p-arm reads of 3'-tailed mirtrons exhibited relatively diffuse start positions upstream and downstream of the dominant 5' end ( Fig 6A and 6B , poundsigns ) , whereas conventional and 5' tailed mirtrons exhibited directional heterogeneity 1-nt downstream of the dominant 5' start of 5p-arm reads defined by the splice donor reference , particularly in human data ( Fig 6A , asterisks ) . On the other hairpin side , we observed clear signatures in which the 3' termini of 3p-arm reads from conventional and 5'-tailed mirtrons were dominantly extended by one nucleotide from the splice acceptor reference ( Fig 6C and 6D , plus-signs ) . The 3' tailing of the 3p-arms of conventional and 5'-tailed mirtrons are mostly due to untemplated uridylation , and to a lesser extent adenylation ( S7 Fig ) . We have described this phenomenon previously in invertebrate and vertebrate mirtrons [31] , and our data here extend this more broadly . Indeed , simple inspection of many loci highlighted in main Figs show that mirtrons with 3p reads defined by splicing frequently accumulate high levels of untemplated uridylation and/or adenylation ( e . g . , Figs 1B and1D and 1E and 5 and 7A and 7B; see also Supplementary Websites for the full read patterns of all annotated mirtrons ) . Indeed , the presence of high levels of 3'-modified read is a characteristic feature that we can now use to positively evaluate novel mirtron annotations . However , as the 5' heterogeneity patterns of mirtrons had not been analyzed , we sought to characterize their bases more deeply . Since the alteration of miRNA 5'-termini will redirect their targeting function , it is commonly thought that most "useful" miRNAs should not have heterogenous 5' ends . This is perhaps relevant to the abundant 5'-tailed mirtron pool . Although Giardia Dicer measures substrate cleavage from the 3' end [57] , it was more recently shown that mammalian Dicer measures its cleavage position from the 5’ end of the hairpin [58] . Heterogeneous definition of the 5' ends of 5'-tailed mirtron hairpins may therefore induce corresponding heterogeneity in the 5' ends of their 3p species . Fig 7 illustrates contrasting loci with respective 5'-end heterogeneity . A dramatic example of alternative 5'-termini is provided with mouse mir-5132 , located in Irak1 ( Fig 7A ) . It exhibits alternative 5' ends of both 5p and 3p species , differing in registers over 3–4 nt each , an isomiR range that is very unusual for canonical miRNAs . Because the 3' end of the mirtron hairpin is fixed , in contrast to alternative Drosha cleavage which coordinately affects both sides of the hairpin , we infer that the degree to which the 5' end of the pre-miRNA hairpin exonucleolytically trimmed will create different small RNA duplexes with different outcomes . The 5p “CGUGG…” terminus likely generates a hairpin that can undergo some alternative dicing to yield alternative 5' ends of corresponding 3p reads , and these are dominantly uridylated . However , further 5' trimming to generate the 5p "UGG…" terminus may be expected to shift the Dicer cleavage site so as to generate a 25-nt uridylated species . Since this is beyond the typical length of Ago cargoes , this seems to be associated with 3' trimming of 3p reads generated from the strongly alternative dicing reaction , as observed in Ago-IP data ( Fig 7A ) . Many 5'-tailed mirtrons exhibit increased 5' heterogeneity of reads from both hairpin arms ( Fig 6A and 6B ) . However , this is not universally the case . For example , a mirtron in mouse Abca2 ( mir-3087 ) exhibits extremely heterogeneous 5' ends of its 5p species , with three different starts being represented equally . However , the vast majority of 3p reads are defined precisely . The disparate behavior of 5' terminal precision of 5p and 3p reads are represented in Ago-IP data ( Fig 7B ) . Such behavior suggests that strategies for strand selection are more complex than currently envisioned . Overall , it appears that definition of 5' termini of miRNA-5p species by a nuclease other than Drosha ( i . e . , with 5' tailed mirtrons ) can influence the register of Dicer cleavage in ways that are not currently predictable . Moreover , the processing of many mirtrons lacks 5' precision , which is usually taken as a hallmark for genuine miRNAs , even though such mirtrons still exhibit clear evidence for dicing and are frequently captured in Ago-IP data . This data indicate the complexity and challenges of annotating Dicer-substrate miRNAs . Unlike the 5' termini of 5p species from 5'-tailed mirtrons , the 5p species from conventional mirtrons and 3'-tailed mirtrons are directly defined by splicing . Therefore , a base expectation is for them to initiate relatively homogenously with the 5' splice donor sequence "GU" . While this is the case for many loci , we unexpectedly observed several conventional mirtrons and 3'-tailed mirtrons whose dominant 5p species did not initiate with 5' GU . Instead , these loci exhibited frequent loss of 5'-G of 5p reads , which we refer to as "xU" reads . Some prominent cases with the highest frequency of "xU" species had hairpins with short 5' overhangs . For example , hsa-mir-4745 generates >90% "xU" reads and its pre-miRNA exhibits 4:3 nt ( 5':3' ) overhangs ( Fig 7C ) . In essence , loss of 5'-G might be interpreted as removal of a single-nt 5' tail . However , loci such as hsa-mir-1236 exhibit a typical pre-miRNAs with a 3' overhang , but still generate high frequency "xU" reads ( Fig 7D ) . We sought the breadth of this phenomenon by plotting the "xU" frequency of all splice donor-derived mirtron-5p loci . Considering loci with ≥10 total 5p reads , we observed that 27/37 human and 11/31 mouse mirtron-5p loci that accumulated at least 10% 5' G-decapitated species ( Fig 7E and S8 Fig ) . Moreover , there were 12 human and 2 mouse 5p mirtrons for which xU reads were the majority of cloned species . Such 5'-directed trimming converts mirtron-5p species from having the least-favored ( 5'-G ) to the most-preferred ( 5'-U ) nucleotide for binding to the Ago2 MID domain [51] , and we confirm the accumulation of abundant "xU" species in Ago-IP datasets ( e . g . Fig 7C and 7D ) . Conceivably , these phenomenon may be related to the fact that mammals exhibit seemingly robust pathways for 5' mirtron tail removal , as suggested by the >800 such loci ( Fig 2A ) . Notably , the exonucleotic removal of single 5' nucleotides from the 5p species of canonical pre-miRNA hairpins would not be distinguishable from alternative Drosha processing . However , this process can utilize conventional pre-miRNA structures ( Fig 7D ) . Overall , the observation of substantial 5' decapitation of nucleotides from the 5' end suggests a mechanism to alter seed identity via an unidentified enzyme .
In spite of a still-growing appreciation of non-canonical miRNA biogenesis pathways [6] , the bulk of miRNA reads in most cell types are generated from canonical loci . In fact , the majority of miRNA reads in individual celltypes can often be accounted for by a dozen or so miRNAs , and sometimes fewer [59] . On the other hand , our recent [13] and current efforts to annotate mirtrons in mouse and human genomes yield the unexpected conclusion that a substantial fraction of confident miRNA loci in mammals are actually non-canonical . As with our efforts with canonical miRNAs , we utilized strict criteria to evaluate small RNA evidence for specific dicing of precursor hairpins , and the vast majority of mirtrons generate reads present in Ago complexes ( S2 and S3 Tables ) . The regulatory impact of small RNAs is concentration-dependent [60 , 61] . As the bulk of mirtrons generate small RNAs that accumulate modestly , it is unclear what impact individual mirtrons have on endogenous gene expression . In fact , the massive turnover of mirtrons between mouse and human ( Fig 2B ) indicates they only infrequently incorporated into conserved regulatory programs . Despite this , mirtrons do generate miRNA-class molecules and generally incorporate into Ago complexes ( Fig 2C–2F ) . One way for mirtron-derived small RNAs to gain enhanced impact is to "piggyback" onto existing conserved miRNA target networks that are already selected for function . Such a strategy was proposed for certain mammalian non-canonical miRNAs [62] , and we observe that the Drosophila 3'-tailed mirtron mir-1017 has converged onto the pre-existing seed of the canonical Brd box family [10] . Although mouse and human mirtrons do not seem to be enriched for seed similarity to conserved mammalian miRNAs , relative to other portions of their mature sequences , we do observe a set of potential mirtron "seed mimics" . We summarize information on these relationships in S6 Table , which may serve to prioritize a subset of mirtrons for functional studies . Mirtrons are advantageous for characterizing post-primary processing of miRNA species , because alterations to hairpin termini that are generated by splicing can be easily distinguished [31] . In general , terminal modifications of canonical miRNAs can be confidently inferred only when the resultant reads fail to match the genome ( e . g . with additions of untemplated nucleotides ) . Otherwise , 5' trimming , 3' trimming and 3' tailing events that fortuitously match the genome are difficult to distinguish from alternative RNase III processing events . In contrast , such modifications to mirtrons can be categorized as reads that lack 5'-GU or AG-3' nucleotides at splicing-derived termini . In this study , we found that mirtron-5p species initially defined by splice donor sequences can exhibit high frequency 5' trimming of guanine residues . In essence , "xU" reads may represent removal of a single 5' nucleotide tail . This modification may increase the capacity for loading mirtron-5p reads , since it alters their 5' ends from being disfavored ( 5'-G ) to optimal ( 5'-U ) binding to the Ago MID domain [51] . Such trimming is uniquely noticeable with mirtrons whose 5' pre-miRNA hairpin end is defined directly by splicing , and would not be distinguished from alternate Drosha cleavage of canonical pre-miRNAs . Since 5' trimming of even a single nucleotide is expected to induce a radical shift in miRNA target networks , it will be interesting to know if such post-primary processing events occur in a "hidden" fashion with any canonical pre-miRNAs . On the other hairpin end , close inspection of mirtron-derived read patterns raises questions regarding the strategy for selection of cleavage position ( s ) by Dicer . An initial model from Giardia studies posited that the PAZ domain allows Dicer to measures substrate cleavage from the 3' end [57] . However , it was more recently shown that mammalian Dicer contains a 5' phosphate-binding pocket , which allows it to measure its cleavage position from the 5' end of the hairpin [58] . In vitro experiments showed that variation of pre-miRNA 3' overhang length by several nts did not adjust the register of Dicer cleavage . Consistent with this , we observe that abundant 3' untemplated uridylation is often associated with a common dominant 5' end of mirtron-3p species , suggesting that that this modification may not adjust Dicer cleavage . Reciprocally , we also observe many 5'-tailed mirtron hairpins that exhibit variable 5' ends of 5p species , presumably due to exoribonucleolytic trimming that is less precise than Drosha cleavage . While we find many cases in which alternative 5' hairpin ends likely induce alternative Dicer cleavages , in keeping with a 5' measuring rule , we also find cases in which highly variable 5' hairpin ends are associated with precise definition of partner 3p species . These observations hint at further complexity in defining the positions of Dicer cleavage . Overall , our deep annotation of mammalian mirtrons reveals multiple aspects of small RNA processing that are not apparent with the general study of canonical miRNAs , and delineates informative sets of test substrates with which to examine these further . Although some mirtrons are well-conserved as with canonical miRNAs [13 , 42] , these are clearly the exception . We previously documented increased evolutionary flux of mirtrons relative to canonical miRNAs across Drosophila species [30] . This notion is bolstered by our current appreciation that of ~500 mouse mirtrons and ~500 human mirtrons that we annotated according to stringent criteria , only a few percent are shared between these species . Indeed , the paucity of conserved mirtrons may suggest they are actively prevented from incorporation into conserved regulatory networks . We speculate that the Drosha/DGCR8 complex effectively filters a subpopulation of hairpin substrates for further processing by Dicer , which is more indiscriminate in substrate selection . On the other hand , the splicing pathway may fortuitously generate a diversity of "dice-able" hairpins . In support of this scenario , we showed the structural properties and small RNAs patterns from mirtrons are collectively much more heterogeneous than canonical miRNAs , and in some cases , embody characteristics that would normally make their annotation as miRNAs somewhat suspect . If we consider that mirtrons are generally adventitious substrates that access Dicer , then this could have facilitated the emergence of mechanisms that generally suppress their biogenesis . Conceivably , this may be related to the high rates of mirtron uridyation [31] . We observe that uridylation of mirtron-3p termini defined by splicing is a pervasive feature in mammalian mirtrons ( Fig 6 and S7 Fig ) , which indeed mostly fall into the 5' tailed or conventional classes ( Fig 2A ) . Since hairpin uridyation has been linked to the turnover of certain pre-miRNAs [63] , it is conceivable that mirtron tailing is linked to their differential evolutionary lability . Evidence for such a mechanism has recently been presented in the Drosophila system [64 , 65] , and this scenario merits investigation in mammals .
We collected 690 human and 348 mouse small RNA data sets from various tissues/cell lines from NCBI GEO/SRA . Among them , 189 human and 125 mouse new small RNA data sets were added since our previous meta-analysis [13] . The accession numbers of all data sets analyzed in this study are summarized in S1 Table . After removing 3’ adaptor sequences , we mapped human and mouse small RNA reads to human ( hg19 ) and mouse ( mm9 ) genome assemblies , respectively . Unmapped reads were iteratively trimmed one nucleotide each iteration retaining a read length of ≥17nt , and then mapped to the genome using Bowtie with no mismatches , up to 4 iterations . We obtained intron annotations from UCSC Genome Browser via table browser and used the “knownCanonical” table including the canonical splice variant of genes . We re-annotated mirtrons using our annotation pipeline for small RNA mapping to intron termini and analysis of hairpin and duplex properties [13] , with modifications to increase read depth stringency . In the main pipeline , we required at least 50 total duplex reads ( mature and star strand reads ) , with at least 5 star strand reads from which 3' duplex overhang characteristics were assessed . Because we included a number of Ago-IP libraries in this analysis , we also considered loci with <5 star strand reads to be confident if there were at least 100 reads from the mature arm , at least 20 of which were recovered from Ago-IP datasets . However , only a relatively small number of loci necessitated this second criterion ( see S2 and S3 Tables ) . As we analyzed greater small RNA library variety and read depth than before [13] , we were able to identify novel 242 human and 248 mouse mirtrons even though our criteria were more stringent . All of the loci were manually vetted to ensure confident inference that their progenitor hairpins were both generated by splicing , and subject to dicing to yield the mapped small RNAs . There are 9 loci that failed these criteria , but we believe should be recorded as confident mirtron loci ( listed in S2 and S3 Tables ) . 5 loci had ≤ 3 star reads but ≥ 20 reads were recovered in Ago-IP libraries ( human: uc002okf . 4 , uc009vxw . 2 , uc003bmk . 3 , uc001uej . 1; mouse: uc007nzl . 1 ) , and 4 loci had 3–4 star reads but ≥10 Ago-IP reads and ≥100 total reads ( human: uc001bvb . 1 , uc003yzv . 2 , uc002ojz . 2; mouse: uc008hbc . 17 ) . We also downgraded 4 mirtrons in mirBase which failed the 50 total reads criteria ( hsa-mir-1178 , hsa-mir-7107 , hsa-mir-1238 ) , or lacked both star reads and Ago-IP reads ( hsa-mir-4641 ) . Finally , we note that while the reads assigned to the vast majority of mirtron loci map uniquely , in rare cases some reads were mapped to multiple locations . The quality of dicing features amongst the aggregate reads had to be satisfactory at each genomic locus considered as a mirtron . Detailed information on the mapped reads , their modified reads , the libraries of origin , and other genomic information on all the mirtrons annotated in this study can be browsed at these web links: http://compgen . cshl . edu/mirna/mam_mirtrons/hg19_candidate/ ( human ) and http://compgen . cshl . edu/mirna/mam_mirtrons/mm9_candidate/ ( mouse ) . Mirtron hairpin boundaries were defined on the basis of intron splice donor ( GT ) or acceptor ( AG ) sequences . The derivation of mirtron-3p reads from splicing permitted us to utilize the "AG" splice acceptor as an external reference for the primary-processed species . Thus , reads that extended past the "AG" likely bear untemplated nucleotides , and were called as such regardless of whether they match the genome or not . The boundaries of canonical pre-miRNA hairpins were based on miRBase v19 with sequences defined by the most abundant matching read from all aggregated small RNA libraries analyzed in this study . We also obtained mRNA-seq data for human ( body map II; SRA accession: ERP000546 ) and mouse ( SRA accession: SRP016501 ) across different tissues . We mapped these mRNA-seq data using TopHat to corresponding genomes hg19 and mm9 , respectively . For each mirtron:host pair an expression value was calculated for both the mirtron and the host gene . The mirtron expression level was quantified as a proportion of small RNA reads per million mapped miRNA reads . Since mRNAs are subject to various types of alternative processing , which could potentially affect the observed correlation levels , the host gene expression was estimated by multiple measures to enable as robust assessments as possible . Host gene expression was alternately estimated at the ( 1 ) gene-level ( RPKM , union of all gene exons ) , ( 2 ) mirtron-flanking exons ( RPKM ) , and ( 3 ) mirtron spanning spliced reads ( RPM ) . While this final category provides the most precise quantification of host intron usage , junction spanning reads are considerably more rare , motivating the use of gene and flanking exon level estimates . For each mirtron:host mRNA pair the Pearson correlation was calculated between expression levels in matched samples ( human: [brain , breast , heart , kidney , liver , white blood cells] , mouse: [brain , heart , kidney , liver , lung , skeletal muscle , spleen] ) . The distribution of observed correlation values was represented using histogram and cumulative distribution plots . An alpha = 0 . 99 empirical confidence envelope ( the gray shaded area on the cumulative plot ) was generated by repeating the analysis 100 times with shuffled tissue labels . A two-tailed 1-sample Wilcoxon test was used to assess the significance of the observed deviation of the distribution median from zero . | Bulk miRNAs in most animal cells derive from cleavage of hairpin precursor transcripts by the Drosha and Dicer RNase III enzymes . A variety of non-canonical miRNA biogenesis strategies are known , including "mirtrons" for which splicing substitutes for Drosha . However , as non-canonical miRNAs usually account for a minor fraction of total miRNA reads , the alternate pathways are often considered of minor impact . In this study , we describe meta-analysis of >1000 mouse and human small RNA datasets , and surprisingly , find evidence of nearly 1000 mirtron using stringent criteria , half of which were novel annotations . Therefore , mirtrons comprise a substantial fraction of total Dicer substrates . We use this catalog to perform diverse sequence and structural analyses of mirtron loci that differentiate them from bulk introns and also to perform diverse comparisons of mirtrons and canonical miRNAs . Surprisingly , we find that many seemingly fundamental features of miRNA genes , which have been well-studied over the years , are not actually general features of miRNAs , but specifically those of hairpins that transit the Drosha-Dicer pathway . Instead , the nearly 1000 splicing-derived miRNAs reveal expanded and/or distinct properties encompassed by endogenous Dicer substrates in mammalian cells . These features reveal hidden impacts that shape miRNA processing , function , and evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Analysis of Nearly One Thousand Mammalian Mirtrons Reveals Novel Features of Dicer Substrates |
Persistent infection of basal keratinocytes with high-risk human papillomavirus ( hrHPV ) may cause cancer . Keratinocytes are equipped with different pattern recognition receptors ( PRRs ) but hrHPV has developed ways to dampen their signals resulting in minimal inflammation and evasion of host immunity for sustained periods of time . To understand the mechanisms underlying hrHPV's capacity to evade immunity , we studied PRR signaling in non , newly , and persistently hrHPV-infected keratinocytes . We found that active infection with hrHPV hampered the relay of signals downstream of the PRRs to the nucleus , thereby affecting the production of type-I interferon and pro-inflammatory cytokines and chemokines . This suppression was shown to depend on hrHPV-induced expression of the cellular protein ubiquitin carboxyl-terminal hydrolase L1 ( UCHL1 ) in keratinocytes . UCHL1 accomplished this by inhibiting tumor necrosis factor receptor-associated factor 3 ( TRAF3 ) K63 poly-ubiquitination which lead to lower levels of TRAF3 bound to TANK-binding kinase 1 and a reduced phosphorylation of interferon regulatory factor 3 . Furthermore , UCHL1 mediated the degradation of the NF-kappa-B essential modulator with as result the suppression of p65 phosphorylation and canonical NF-κB signaling . We conclude that hrHPV exploits the cellular protein UCHL1 to evade host innate immunity by suppressing PRR-induced keratinocyte-mediated production of interferons , cytokines and chemokines , which normally results in the attraction and activation of an adaptive immune response . This identifies UCHL1 as a negative regulator of PRR-induced immune responses and consequently its virus-increased expression as a strategy for hrHPV to persist .
Human papillomaviruses ( HPVs ) are absolutely species-specific small double-stranded DNA viruses . Persistent infections with a number of HPVs , predominantly types 16 and 18 , can induce cancers of the anogenitalia as well as of the head and neck region . These so-called high-risk HPVs ( hrHPVs ) are widespread within all human populations where they are commonly transmitted by sexual contact [1] . The undifferentiated keratinocytes of the squamous epithelia are the primary target for hrHPV [2] where it establishes an infection that can last for up to 2 years , indicating that hrHPV has evolved mechanisms to effectively evade the innate and adaptive immune mechanisms protecting the majority of immunocompetent hosts [3] , [4] . Viruses and microbes contain pathogen-associated molecular patterns that are recognized by the host's pattern recognition receptors ( PRRs ) , comprising the Toll-like receptors ( TLRs ) , nucleotide oligomerization domain-like receptors and retinoic acid-inducible gene I ( RIG-I ) -like receptors ( RLRs ) [5] . While all of these receptors activate signaling cascades that lead to activation of NF-κB via the canonical route , only RLRs and some TLRs activate interferon regulatory factors ( IRFs ) which induce the production of type I interferons ( IFN ) and other effector molecules [6] . The signals from the PRR to the cell nucleus are coordinated via ubiquitination , including that of the different tumor-necrosis factor receptor-associated factors ( TRAFs ) and the NF-κB essential modulator ( NEMO ) . Poly-ubiquitination of TRAF and NEMO allows downstream signaling whereas disassembly of the formed poly-ubiquitin chains by deubiquitinating enzymes provides a mechanism for downregulating immune responses [6] , [7] . Keratinocytes ( KCs ) express TLRs 1–3 , TLR5 , TLR6 , TLR10 , RIG-I , protein kinase R ( PKR ) , and MDA5 independent of their differentiation status and gain the expression of TLR9 upon full differentiation indicating that these cells may respond to pathogenic challenges [8] , [9] , [10] . Thus , KCs should be able to sense the presence of hrHPV genomic DNA directly via TLR9 or indirectly via RIG-I [5] , [11] , [12] . The expression levels of these PRR were not altered in hrHPV+ KCs [10] . However , via genome-wide expression profiling of keratinocytes activated through TLR3 , PKR , RIG-I and MDA-5 we found that the presence of hrHPV dampens a network of genes encoding chemotactic , pro-inflammatory and antimicrobial cytokines suggesting that HPV's immune evasion strategy may rely on countering PRR-mediated cell signaling [10] . To understand the mechanisms underlying hrHPV's capacity to dampen PRR signaling we utilized a system that resembles the natural infection with HPV as closely as possible . It comprises the use of primary KCs that stably maintain the hrHPV genome as episomes following transfection . These hrHPV+ KCs grow at similar rates as non-transfected KC and have been shown to mimic HPV infection in vivo as they undergo the entire differentiation-dependent HPV life cycle documented by genome amplification , late gene expression , and virus production , upon culture of hrHPV+ KCs in organotypic raft cultures [13] , [14] , [15] . In addition , we used non-infected primary KC cultures and primary KCs newly infected with authentic HPV16 virions . These primary KCs were compared with respect to PRR signaling under different conditions and resulted in the identification of the cellular enzyme ubiquitin carboxyl-terminal hydrolase L1 ( UCHL1 ) that was specifically upregulated by hrHPV in primary keratinocytes to dampen innate immunity . UCHL1 acted on the PRR-signaling pathway adaptor molecules TRAF3 and NEMO and its inhibition restored PRR-induced production of IFNβ and pro-inflammatory and chemotactic cytokines .
Undifferentiated uninfected primary KCs and hrHPV+ KCs were tested for their capacity to respond to triggers of innate immunity by incubation with Pam3CSK4 ( TLR1/2 ) , poly ( I∶C ) ( TLR3 , RIG-I , PKR and MDA-5 ) [9] , lipopolysaccharide ( LPS , TLR4 ) , flagellin ( TLR5 ) , R848 ( TLR7/8 ) , or CpG ( TLR9 ) . The supernatant of non-infected keratinocytes contained higher levels of MIP3α and IL-8 but not MIP1α than hrHPV+ KCs at the basal level . Activation with poly ( I∶C ) induced the production of high amounts of MIP3α , IL-8 and MIP1α in KCs but not in hrHPV+ KCs . Flagellin especially triggered the production of MIP3α by KCs but not in hrHPV+ KCs , although IL-8 was still produced ( Figure 1A ) . The function of TLR9 , expressed only at high protein levels in differentiated keratinocytes as measured by immunohistochemistry [10] and by RT-qPCR ( Figure 1B ) , was tested by the capacity of CpG oligodeoxynucleotides ( CpG ODN ) to trigger the expression of mRNAs of pro-inflammatory cytokines and chemokines . Because suspension in methyl cellulose – to differentiate keratinocytes – does not allow the harvest of supernatant , secreted protein levels could not be measured . However , the experiments clearly showed that CpG ODN-stimulation resulted in the gene expression of IFNB1 ( IFNβ ) , IL-8 and CCL20 ( MIP3α ) in differentiated KCs but not in undifferentiated KC cultures ( Figure 1C ) . As a control , KCs were also stimulated with poly ( I∶C ) as TLR3 , RIG-I and MDA-5 expression is independent of KC differentiation [10] and this resulted in the induction of pro-inflammatory cytokine expression in both undifferentiated and differentiated KCs ( Figure S1 ) . In contrast to differentiated uninfected KCs , the hrHPV+ KCs that expressed TLR9 after differentiation , failed to induce the expression of IFNβ , IL-8 and MIP3α upon incubation with CpG ( Figure 1C ) , indicating that PRR-signaling can be suppressed in undifferentiated and differentiated hrHPV+ KCs . As the basal KCs are the target for hrHPV and TLR9 is not functionally expressed in basal KCs and hrHPV+ KCs displayed an impaired production of cytokines in response to poly ( I∶C ) , subsequent studies were performed in the context of poly ( I∶C ) stimulation . In addition to the secretion of cytokines , also the gene expression levels of MIP3α , CCL5 ( RANTES ) and IFNβ in hrHPV+ KCs were lower when compared to uninfected KCs upon 3 or 24 hours of poly ( I∶C ) stimulation ( Figure 2A ) . The production of pro-inflammatory cytokines and chemokines upon activation of the NF-κB pathway requires the phosphorylation and nuclear translocation of the subunit p65 [6] . The levels of phosphorylated p65 were lower in poly ( I∶C ) stimulated hrHPV+ KCs than in non-infected KCs ( Figure 2B ) , suggesting that the functional impairment of PRR signaling occurs upstream of this molecule . The IKK complex is a key component of the poly ( I∶C ) -induced NF-κB pathway , with NEMO ( IKKγ ) functioning as a scaffold . The degradation of NEMO may form a mechanism for viruses to avoid innate immune signaling [16] , [17] . Therefore , the effect of hrHPV on the protein levels of NEMO was analyzed . Following treatment of non-infected KCs and hrHPV+ KCs with cycloheximide ( CHX ) – to prevent new protein synthesis – it became clear that NEMO degradation was enhanced in hrHPV+ KCs ( Figure 2C and Figure S2 ) , thereby explaining the decreased phosphorylation of p65 observed . The production of type I IFN ( e . g . IFNβ ) requires the activation of cytosolic IRF3 by phosphorylation and subsequent translocation to the nucleus . Analysis of poly ( I∶C ) stimulated KCs and hrHPV+ KCs suggested that also the levels of phosphorylated IFR3 levels were decreased in HPV+ KCs ( Figure 2D ) . To confirm that the impairment in the production of IFNβ and pro-inflammatory cytokines did not simply reflect biological differences between the different primary KCs used but indeed was caused by hrHPV , we infected primary keratinocytes with infectious HPV16 virions ( Figure 3A ) for 24 hours and then stimulated the non-infected and newly infected KCs with poly ( I∶C ) for another 24 hours after which the levels of IFNβ , RANTES and MIP3α transcripts were measured ( Figure 3B ) . After 24 hours of infection there was a small but discernible increase in the levels of these genes indicating that the keratinocytes initially react to the presence of the virus . However , the levels already dropped at 48 hours post-infection indicating that the virus rapidly exerted its PRR-signaling inhibitory effects . In addition , at the same time point these newly hrHPV-infected keratinocytes displayed a hampered activation of IFNβ , RANTES and MIP3α following 24 hours of stimulation with poly ( I∶C ) ( Figure 3B ) . Moreover , we repressed the polycistronic viral mRNA transcript [18] , [19] in hrHPV+ KCs by the use of siRNA targeting HPV16 E2 as this allows the destruction of the whole RNA chain . Indeed the suppression of HPV early gene E2 expression translated into an overall decrease in viral early gene expression ( Figure 3C ) and an increase in the transcription of IFNβ , RANTES and MIP3α following poly ( I∶C ) stimulation ( Figure 3D ) . Together these data demonstrate that the innate immune response to viral and bacterial-derived PRR stimuli of both undifferentiated and differentiated hrHPV+ keratinocytes is suppressed by HPV at a point downstream of the PRR receptors but upstream of the transcription factors that relay the PRR signals to the nucleus . Our data suggest that hrHPV+ keratinocytes manifest a generalized inability to respond to stimulation through interference at , or downstream of the cytosolic part of the PRR signaling pathways . We therefore re-analyzed the genome-wide expression profiles ( Gene Expression Omnibus accession number GSE21260 ) of several different uninfected KC cultures and hrHPV+ KC cultures reported previously [10] by Ingenuity Pathways Analysis ( IPA ) and found a highly significant enrichment of genes belonging to the protein ubiquitination pathway ( Table S1; p = 6 . 69×10−5 ) . In this pathway , the gene for the enzyme ubiquitin carboxyl-terminal hydrolase L1 ( UCHL1 ) was the most upregulated gene in hrHPV+ KCs compared to uninfected KCs ( Figure 4A and B ) . The upregulation of UCHL1 in hrHPV+ KCs was confirmed by RT-qPCR in both foreskin and vaginal epithelial hrHPV+ KC cultures and expression was not influenced by poly ( I∶C ) activation ( Figure 4C ) . Furthermore , UCHL1 upregulation at the protein level was tested and shown for three different hrHPV+ KCs by western blotting ( Figure 4D ) . Moreover , expression of UCHL1 was upregulated 2 days post-infection of HPV16 in primary keratinocytes when compared to mock-infected primary keratinocytes ( Figure 4E ) , whereas knock-down of the polycistronic viral mRNA transcript in hrHPV+ KCs by siRNA for HPV16 E2 resulted in a decreased UCHL1 expression ( Figure 4F ) . Thus , the cellular deubiquitinase UCHL1 is upregulated by hrHPV . Although UCHL1 had not been associated with the inhibition of PRR signaling , its enhanced expression in hrHPV+ KCs fits well with the general role of deubiquitinases in controlling PRR signaling [6] . To test whether hrHPV-induced UCHL1 inhibits PRR signaling , we used lentiviral vectors expressing short-hairpin RNA ( shRNA ) against UCHL1 and this resulted in a downregulated expression of UCHL1 transcripts and protein levels in hrHPV+ KCs ( Figure 5A and B ) . Upon stimulation with poly ( I∶C ) , hrHPV+ KCs expressing shRNA against UCHL1 ( shUCHL1 ) but not hrHPV+ KCs expressing a control shRNA ( shControl ) restored poly ( I∶C ) -mediated induction of type I interferon and proinflammatory cytokines ( Figure 5C ) . Similar results were obtained using transiently transfected RNA interference ( RNAi ) oligos targeting UCHL1 but not with control RNAi oligos ( Figure S3 ) . An increase in the expression levels of IL8 and MIP3α was detected in hrHPV+ KCs in which UCHL1 was downregulated . Gene expression increased to the same levels found in UCHL1-non silenced hrHPV+ KCs cells stimulated with poly ( I∶C ) ( Figure S3 ) . This suggests that downregulation of UCHL1 increases the gene expression of IL-8 and MIP3α in hrHPV+ KCs . Conversely , transfection of uninfected KCs to overexpress UCHL1 resulted in a decreased expression of MIP3α , RANTES and IFNβ upon poly ( I∶C ) stimulation ( Figure 5D and E ) . Based on control experiments in which KCs were transfected with green fluorescent protein expressing plasmids , the transfection efficiency of keratinocytes was 30–40% ( not shown ) , indicating that in a large part of the keratinocytes the activation of cytokine-encoding genes is not impaired and explaining the expression levels of these cytokine-encoding genes that are still detected . All together , these data clearly demonstrate that UCHL1 can downregulate the PRR-mediated activation of both the type I IFN and proinflammatory cytokine and chemokine pathways . We then asked whether the restoration of PRR signaling , as indicated by an increased induction of type I interferon and proinflammatory cytokines by the knock down of UCHL1 in hrHPV+ KCs would also be reflected in the levels of phosphorylated p65 ( p65-p ) and IRF3 ( IRF3-p ) upon poly ( I∶C ) stimulation . Therefore , the p65-p and IRF3-p levels were analyzed in whole cell extracts of HPV16+ KCs stably expressing shRNA against UCHL1 or control shRNA and following 3 h or 24 h of stimulation with poly ( I∶C ) . Knock down of UCHL1 in hrHPV+ KCs resulted in increased p65 phosphorylation at 3 and 24 hours after poly ( I∶C ) stimulation ( Figure 6A ) coinciding with enhanced cyto- and chemokine production ( Figure 5C ) . In addition , analysis of hrHPV+ KCs treated with cycloheximide revealed that NEMO degradation was alleviated when UCHL1 was knocked down by shUCHL1 as compared to the shControl hrHPV+ KCs ( Figure 6B ) . Furthermore , higher levels of phosphorylated IRF3 were detected in hrHPV+ KCs in which UCHL1 was knocked down as compared to hrHPV+ KCs expressing the shControl after 3 hours of poly ( I∶C ) stimulation ( Figure 6C ) . TRAF3 ubiquitination is critical for type I IFN production and is a likely target for ubiquitin-modifying enzymes such as UCHL1 . As the biochemical experiments to understand the nature of this interaction would require substantial amounts of primary KCs , which can only grow for a few passages thereby restricting their use in biochemical studies , we switched to the HEK293T cell system that is widely used for these purposes . To investigate the interaction between UCHL1 and TRAF3 we overexpressed UCHL1 and Flag-tagged TRAF3 in HEK293T cells . After FLAG immunoprecipitation , we confirmed that UCHL1 co-immunoprecipitated with TRAF3 ( Figure 7A ) . TRAFs are activated by oligomerization and auto-ubiquitination , a process that results in lysine 63 ( K63 ) -linked poly-ubiquination of TRAF , and this event can be induced by either their overexpression or by receptor activation . In contrast K48-linked poly-ubiquitination results in proteasome-mediated degradation of ubiquitinated TRAFs [6] . To test whether UCHL1 modified TRAF3 ubiquitination status , Flag-tagged TRAF3 and haemagglutinin A ( HA ) -tagged ubiquitin were overexpressed in control or UCHL1 overexpressing HEK293T cells . Poly-ubiquitination of TRAF3 was clearly visible by immunoblot analysis but strongly reduced when UCHL1 was also overexpressed ( Figure 7B , Figure S4 ) . No reduction in poly-ubiquitination was detected when as a control the growth regulated ubiquitin-specific protease 8 ( USP8 ) , which similar to UCHL1 displays carboxyl-terminal hydrolase activity , was overexpressed ( Figure 7B ) . The UCHL1-associated decreased detection of poly-ubiquitinated TRAF3 was not the result of increased TRAF3 degradation as blocking the proteasomal degradation pathway by the inhibitor MG132 did not result in a reappearance of poly-ubiquitinated TRAF3 ( Figure 7C ) . Instead , experiments in which HA-tagged ubiquitin mutants ‘K63 Only’ and ‘K48 Only’ ( where all lysine residues , except at position K63 and K48 , respectively , were mutated to arginine ) showed that UCHL1 removed K63-linked poly-ubiquitins but not K48-linked poly-ubiquitins ( Figure 7D ) , consistent with the known deubiquitinating capacity of UCHL1 [20] . K63-linked ubiquitination is required for TRAF3 to bind its partner TBK1 to activate the downstream type I IFN-signaling pathway . As expected , UCHL1-mediated deubiquitination of TRAF3 resulted in less TRAF3 bound to TBK1 in UCHL1 overexpressing cells when compared to control cells ( Figure 7E ) . These data clearly show that UCHL1 binds and deubiquitinates TRAF3 resulting in a decreased TRAF3-TBK1 complex formation . Poly-ubiquitination of TRAF6 and its downstream partner NEMO is critical for the PRR-induced activation of proinflammatory cytokine genes [6] . Since the overexpression of UCHL1 clearly affected proinflammatory cytokine synthesis ( Figure 5 ) the interaction of UCHL1 with TRAF6 and NEMO was tested . Co-expression and immunoprecipitation experiments in HEK293T cells showed that UCHL1 bound to TRAF6 but not to NEMO ( Figure 7A ) . In contrast to what we observed for TRAF3 , UCHL1 displayed a modest effect on the poly-ubiquitination of TRAF6 ( Figure 8A ) . However , poly-ubiquitination of NEMO was reduced in UCHL1 overexpressing cells ( Figure 8B , Figure S4 ) but not in USP8 overexpressing cells ( Figure 8D ) . Inhibition of proteasome function by MG132 suggested that the reduced poly-ubiquitination of NEMO was the result of enhanced degradation of NEMO in cells overexpressing UCHL1 ( Figure 8C , compare lanes 2 and 4 ) , albeit that the total protein levels of NEMO in these transfected cells remained unaffected . This is not unexpected as also in the endogenous setting ( Figures 2 & 6 ) the degradation of NEMO could only be visualized when the hrHPV+ KCs where pretreated with cycloheximide to prevent new protein synthesis . Collectively , these data support the notion that UCHL1 can suppress the PRR-signaling pathways necessary for type I IFN and pro-inflammatory cytokine production by the removal of the activating K63 ubiquitins from TRAF3 and the forced degradation of NEMO .
We have employed a unique model for hrHPV infection to examine the potential mechanisms underlying the capacity of hrHPV to evade host immunity by suppression of the innate immune response [10] . We utilized primary KC cultures that were newly infected with HPV16 virions or primary KCs stably maintaining the episomal hrHPV genome to show that despite the expression of multiple PRRs the production of IFNβ and pro-inflammatory cytokines and chemokines is suppressed by hrHPV as a consequence of reduced PRR signaling . We provided firm evidence that this suppression depends on the hrHPV-induced upregulation of the cellular ubiquitin-modifying enzyme UCHL1 in infected primary KCs . Finally , classical biochemical studies in HEK293T cells [11] , [21] , [22] performed to understand how UCHL1 mechanistically could suppress the production of type I interferons and pro-inflammatory cytokines revealed that UCHL1 regulated the ubiquitination of the PRR-signaling pathway adaptor molecules TRAF3 and NEMO . UCHL1 removes activating K63-linked ubiquitin molecules from TRAF3 resulting in a lower amount of the downstream signaling complex TRAF3-TBK-1 to suppress the type I IFN pathway . This puts UCHL1 within the family of other deubiquinating enzymes that regulate the PRR pathways by selectively cleaving lysine-63 ( K63 ) -linked ubiquitin chains from TRAFs ( e . g . DUBA , OTUB1 , OTUB2 , A20 ) [21] , [22] , [23] , [24] , [25] , [26] . Furthermore , we showed that UCHL1 bound to TRAF6 and mediated the enhanced degradation of NEMO as a mechanism to suppress the proinflammatory cytokine NF-κB pathway . Notably , the ubiquitin-modifying enzyme A20 , a known negative regulator of the TLR pathway , has two ubiquitin-editing domains allowing it to remove and to add ubiquitin chains ( 22 , 26 ) . UCHL1 has also been reported to have these two opposing functions ( 20 ) . The ligase activity of UCHL1 may explain the ubiquitination of TRAF6 observed in our study . Although UCHL1 did not bind to NEMO , it is known that other deubiquitinating enzymes ( e . g . CYLD , A20 ) bind to TRAFs in order to dock on the IKK complex and to associate with NEMO [21] , [27] . TRAF6-dependent poly-ubiquitination of NEMO is well known [28] . It is highly likely that UCH-L1 acts in a similar fashion and this would fit with TRAF6-NEMO interaction and our observations that NEMO is degraded . Our data on the suppression of NF-κB signaling via the degradation of NEMO by UCHL1 fits well with earlier observations concerning the overexpression of UCHL1 in vascular cells . Here UCHL1 attenuated TNF-α induced NF-κB signaling and this was associated with stabilization of IκBα and a decrease in its basal ubiquitination [29] . The activation of NF-κB signaling requires IκBα to become degraded following an interaction with the IκB kinase complex ( IKK ) which comprises NEMO . Hence , the degradation of NEMO may explain previous observations on UCHL1-associated stabilization of IκBα . UCHL1 is not found to be central in the network of genes affected by hrHPV , suggesting that it is not part of the cellular genes affected in order to assist in HPV genome replication and viral protein production [10] . This indicates that UCHL1 is not directly involved in viral propagation but rather recruited by hrHPV to suppress keratinocyte-mediated production of cytokines and chemokines that would result in the attraction and activation of an adaptive immune response , thereby enabling the virus to persist and propagate . Many viruses utilize multifunctional viral proteins in order to evade NF-κB- and IRF-mediated immune responses , to favor viral replication and/or to modulate cellular apoptosis and growth pathways [30] . The group of pox viruses have evolved to inhibit NF-κB-signaling by targeting one or more of the many different molecules of this signaling cascade [31] . The vaccinia virus B14 protein is known to inhibit NF-κB signaling by a variety of toll-like receptor agonists at the level of the IKK complex , of which NEMO is a member [32] . The vaccinia virus A64R protein inhibits TRIF-TRAF3-IRF signaling [33] . The pathogenic NY-1 hantavirus Gn protein inhibits TRAF3 signaling by blocking the formation of TBK1-TRAF3 complexes [34] whereas the LMP1 protein of Epstein-Barr virus directly binds to TRAF3 [35] . Furthermore , foot-and-mouth disease virus 3c protease cleaves NEMO [16] and cytomegalovirus M54 protein induces the proteasome-independent degradation of NEMO [17] . In contrast , human papillomaviruses , with a rather limited coding capacity in their genomes , rely for many aspects of their life cycle on the utilization of cellular proteins [36] and this includes the recruitment of different cellular E3 ligases to mediate degradation of cellular proteins through the ubiquitin-proteasomal pathway [37] . UCHL1 is one of the most abundant proteins in the mammalian nervous system and is involved in regulating synaptic transmission at the neuromuscular junctions [38] . Aberrant expression is related to Parkinson's disease [20] and is also implicated in oncogenesis [39] . In hrHPV+ keratinocytes UCHL1 is expressed and redirected to adopt a new function that is to serve as a negative regulator of the PRR-signaling pathway . As such it mimics the ubiquitin-modifying enzyme A20 which is the natural negative regulator of the TLR pathway [22] , [26] , [40] . UCHL1 interferes with the adaptor molecules TRAF3 , TRAF6 and NEMO which all function at junctions for the immune stimulating signals from different PRR and type I IFNR to activate NF-κB- and IRF-mediated immune responses . Therefore , the utilization of UCHL1 represents a truly effective use of a cellular protein as it may suppress the immunostimulatory signals initiated through recognition of HPV genomic DNA by TLR9 [5] and RIG-I [11] , [12] as well as those obtained via the cell surface receptors for type I IFN [41] . The high expression of UCHL1 in primary keratinocytes carrying infectious hrHPV [13] , [14] is generally lost after transformation of these keratinocytes to tumor cells . Although transformed keratinocytes expressing un-physiologically high levels of E6 and E7 via retroviral transduction still may express UCHL1 , only a minority of spontaneously HPV-transformed cervical carcinoma's and none of the well known HPV-induced cancer cell lines overexpress UCHL1 [42] , indicating that under normal conditions UCHL1 overexpression in HPV transformed cells is not a common event . The expression of the hrHPV oncoproteins E6 and E7 is required to maintain the transformed state of keratinocytes [2] , [43] suggesting that it is not E6 or E7 , but one or more of the other viral proteins responsible for upregulation of UCHL1 ( currently under investigation ) . Previous studies on the innate immune response to hrHPV relied on the overexpression of hrHPV E6 and/or E7 proteins , showing that the viral DNA-sensing TLR9 was altered [8] and that overexpressed HPV E6 or E7 could bind to IRF3 [44] and/or the co-activator CPB [45] . Furthermore , overexpressed hrHPV E6 and/or E7 attenuated IκB kinase signaling [46] , and interfered with the nuclear translocation of the interferon-stimulated gene factor 3 ( ISGF3 ) transcription complex [47] . The fact that these studies were performed with only HPV E6- and E7 transfected or transformed cells may explain why the central role of UCHL1 in dampening immunity towards hrHPV+ keratinocytes was not discovered before . In addition , the loss of UCHL1 mediated suppression of the NF-κB pathway in hrHPV E6/E7-induced cancer cells fits well with the notion that solid tumors require the NF-κB-mediated expression of proteins that promote survival , proliferation , invasion and metastasis [48] which is acquired through the E6-mediated deactivation of CYLD [49] , a negative regulator of TRAF2 and TRAF6-mediated activation of NF-κB [21] , [24] . All together , our data implicate UCHL1 as a negative regulator of the PRR pathways helping hrHPV to evade host immunity and allowing it to persist in keratinocytes .
Primary cultures of human epithelial keratinocytes were established from foreskin [50] and vaginal tissues and grown in serum-free medium ( Defined KSFM , Invitrogen , Breda , The Netherlands ) . Keratinocyte lines stably maintaining the full episomal HPV genome following electroporation were grown in monolayer culture using E medium in the presence of mitomycin C treated J2 3T3 feeder cells [13] , [14] for two passages and were then adapted to Defined K-SFM for one passage before experimentation . None of the cell cultures were used after passage 15 and the non-transformed state of the cells used was confirmed by the expression of both E1 and E2 so that the cells used truly represent the preneoplastic state in which the HPV genomes remained episomal and were capable of the complete viral life cycle . Keratinocytes were terminally differentiated by placing them into serum-free medium containing 1 . 75% methyl cellulose and 1 . 8 mM Ca2+ for 24 hours [50] . Cells were harvested by washing out the methyl cellulose three times . HEK293T cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum , 2 mM l-glutamine and 1% penicillin-streptomycin ( Gibco-BRL , Invitrogen ) . Transient transfections were performed using calcium phosphate or Lipofectamine 2000 ( Invitrogen ) . Primary basal layer human foreskin keratinocytes were seeded at 7 . 5×104 cells per well of a 24-wells plate in K-SFM and then allowed to attach for 48 hours . Cells received fresh medium ( Mock infected ) or medium containing native HPV16 isolated from raft cultures at a MOI 100 for 24 hours . Cells were stimulated with or without 25 ug/ml poly ( I∶C ) in K-SFM for 0 or 24 hours and harvested at the indicated time-points . Full length human cDNA clones for UCHL1 , TRAF3 , TRAF6 and TBK1 were obtained from Open Biosystems ( Surrey , UK ) . The cDNA clones were PCR amplified and subcloned either into pcDNA3 . 1 expression vector or into Flag-tagged pcDNA3 . 1 vector . Full-length Flag-NEMO construct was kindly provided by Dr . C . Sasakawa , University of Tokyo , Japan [51] . HA-tagged wild-type and mutant ubiquitin constructs were kindly provided by Dr . A . Iavarone , Columbia University , USA . Total RNA was isolated using TRIzol ( Invitrogen ) according to manufacturer's instructions . RNA was purified using RNeasy Mini Protocol ( Qiagen , Venlo , The Netherlands ) . Total RNA ( 0 . 2 µg ) was reverse transcribed using SuperScript III reverse transcriptase ( Invitrogen ) and oligo dT primers ( Promega , Madison , USA ) . TaqMan PCR was performed using TaqMan Universal PCR Master Mix and pre-designed , pre-optimized primers and probe mix for IL-8 , MIP-1α , MIP-3α , RANTES , IL-1β , IFNβ , UCHL1 and GAPDH ( Applied Biosystems , Foster City , USA ) . Threshold cycle numbers ( Ct ) were determined using the 7900HT Fast Real-Time PCR System ( Applied Biosystems ) and the relative quantities of mRNA per sample were calculated using the ΔΔCt method as described by the manufacturer using GAPDH as the calibrator gene . 5×105 cells were plated in 1 ml in each well of 24-well flat bottom plate . Cells were left unstimulated or stimulated with Pam3CSK4 ( 5 µg/ml ) , Poly ( I∶C ) ( 25 µg/ml ) , LPS ( 3 . 33 µg/ml ) , flagellin ( 150 ng/ml ) , R848 ( 1 µg/ml ) , CpG ( 1 µM ) or TNFα ( 50 ng/ml ) for 24 hours . Flagellin was a kind gift from Jean-Claude Sirard ( Institut Pasteur , Lille , France ) . TLR ligands were purchased from Invivogen ( San Diego , USA ) . The supernatants were harvested and IL-8 , MIP-3α , and MIP-1α concentrations were determined using corresponding Quantikine ELISA kits ( R&D Systems , Oxon , UK ) . Non-targeting RNAi oligos ( ON-TARGETplus Non-targeting Pool , catalogue D-001810-10-20 ) and oligos targeting UCHL1 ( ON-TARGETplus SMARTpool , catalogue L-004309-00 ) were purchased from Dharmacon ( Chicago , IL ) . Cells were transfected with RNAi using N-TER Nanoparticle siRNA Transfection System ( Sigma-Aldrich , St . Louis , MO ) according to manufacturer's instructions . 24 hours after transfection , cells were stimulated with poly ( I∶C ) ( 25 µg/ml ) for another 24 hours and experiments were performed . The shRNA's used were obtained from the MISSION TRC-library of Sigma-Aldrich ( Zwijndrecht , The Netherlands ) . The MISSION shRNA clones are sequence-verified shRNA lentiviral plasmids ( pLKO . 1-puro ) provided as frozen bacterial glycerol stocks ( Luria Broth , carbenicillin at 100 µg/ml and 10% glycerol ) in Escherichia coli for propagation and downstream purification of the shRNA clones . pLKO . 1 contains the puromycin selection marker for transient or stable transfection . The construct against UCHL1 ( NM_004181 ) was TRCN0000011079 ( LV079 ) : CCGGCAGTTCTGAAACAGTTTCTTTCTCGAGAAAGAAACTGTTTCAGAACTGTTTTT and the control was: SHC004 ( MISSION TRC2-pLKO puro TurboGFP shRNA Control vector ) : CCGGCGTGATCTTCACCGACAAGATCTCGAGATCTT GTCGGTGAAGATCACGTTTTT . HPV16+ KCs were seeded 7 . 5×104 cells per well to a 12-wells plate in K-SFM and were allowed to attach over night . Medium was replaced by infection medium ( K-SFM+30% virus supernatant; MOI = 5 ) , containing either the lentivirus LV079 in IMDM 5% FCS or as control SHC004 . HPV16+ KCs were infected over night after which infection medium was replaced by K-SFM containing 1000 ng/ml puromycin for 48 hours to select for successfully infected HPV16+ KCs . Then the medium was replaced by K-SFM without puromycin and cells were grown for 24 hours . To stimulate the PRR pathways lentivirus-infected HPV16+ KCs were given K-SFM containing either no poly ( I∶C ) ( two wells ) or 25 ug/ml poly ( I∶C ) and were cultured for 21 hours . Then one of the two non-stimulated wells received 25 ug/ml poly ( I∶C ) and all cells were cultured for another 3 hours . Cells were harvested and total RNA was isolated . Silencer Select siRNA against HPV16 E2 ( AACACUACACCCAUAGUACAUtt ) was designed using siRNA Target Finder software ( Ambion , Invitrogen ) . Blast search revealed that the designed E2 siRNA does not match with the known human transcriptome . E2 and Negative control #2 ( NC2 ) siRNA ( sequence not provided by manufacturer ) were purchased from Ambion . HPV16+ KCs were transfected with 50 nM siRNA E2 or NC2 using Lipofectamine 2000 ( Invitrogen ) according to the manufacturer's instructions . 48 hours post-transfection cells received K-SFM containing no Poly ( I∶C ) or 25 ug/ml Poly ( I∶C ) and were cultured for 24 hours after which target gene expression was assayed by qRT-PCR . For Western blotting , polypeptides were resolved by SDS–polyacrylamide gel electrophoresis ( SDS–PAGE ) and transferred to a PVDF membrane ( Bio-Rad , Veenendaal , The Netherlands ) . Immunodetection was achieved with anti-Flag ( 1∶2000 , Sigma-Aldrich ) , anti-HA ( 1∶1000 , Covance ) , anti-TRAF3 , anti-TRAF6 ( both 1∶500 , Santa Cruz , CA ) , anti-ubiquitin lysine 48-specific ( 1∶1000 , Millipore , Amsterdam , The Netherlands ) , anti-poly-ubiquitin lysine 63 specific ( 1∶1000 , Millipore ) , anti-TBK1 ( 1∶400 , Santa Cruz ) , anti-NEMO ( FL-419 , Santa Cruz ) , anti-UCHL1 ( 1∶1000 Millipore , 1∶100 Abcam or 1∶1000 Santa Cruz ) , anti-USP8 ( #8728 , Cell Signaling Technology , Danvers , MA , USA ) , anti-phospho-p65 ( Ser538; 1∶1000 , #3033 Cell Signaling Technology ) and anti-phospho-IRF3 ( Ser396; 1∶2000 , #4947 , Cell Signaling Technology ) or β-actin ( 1∶10 , 000 , Sigma-Aldrich ) antibodies . The proteins were visualized by a chemoluminescence reagent ( Thermo Scientific , Etten-Leur , The Netherlands ) . X-Ray films were scanned using a GS-800 calibrated densitometer and Quantity One software ( Bio-Rad , Veenendaal , The Netherlands ) to quantify the intensity of the bands as a measure of the amount of protein of interest in the blot . The relative amount was determined by calculating the ratio of each protein over that of the density measured for the household protein β-Actin . For immunoprecipitation , cells were collected after 48 h and then lysed in NP40 buffer supplemented with a complete protease inhibitor cocktail ( Roche , Almere , The Netherlands ) . After pre-clearing with protein A/G agarose beads for 1 h at 4°C , whole-cell lysates were used for immunoprecipitation with either mouse or rabbit anti-Flag antibodies ( Sigma-Aldrich ) , or rabbit anti-TRAF3 or rabbit anti-TRAF6 . One to two µg of the antibody was added to 1 ml of cell lysate , which was incubated at 4°C for 2–3 h . After addition of protein A/G agarose beads , the incubation was continued for 1 h . Immunoprecipitates were extensively washed with lysis buffer and eluted with SDS loading buffer and boiled for 5 min . For immunoprecipitation under denaturing conditions , proteins were extracted using regular immunoprecipitation buffer plus 1% SDS and heated at 95°C for 5 min . The samples were diluted ( 10-fold ) in regular immunoprecipitation buffer before immunoprecipitation . | A persistent infection with high-risk human papillomavirus ( hrHPV ) may cause cancer . Whereas keratinocytes – the cells infected by hrHPV – are equipped with different receptors allowing them to recognize invading pathogens and to activate the immune system , hrHPV has developed ways to evade the host's immune response for sustained periods of time . We showed that hrHPV accomplishes this by interfering with the signaling of the pathogen receptors , thereby hampering the production of cytokines that are known to attract and activate the immune system . HrHPV accomplishes this by upregulating the expression of a cellular protein called ubiquitin carboxyl-terminal hydrolase L1 ( UCHL1 ) . This protein suppresses the activation of signals downstream of the pathogen receptor leading to reduced transcription factor activation and downstream gene expression , in particular that of type I interferon and pro-inflammatory cytokines . This lowers the attraction of immune cells and thereby the chance of hrHPV-infected cells to be recognized and eliminated and as such enables hrHPV to persist . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"signal",
"transduction",
"viral",
"immune",
"evasion",
"immunity",
"viral",
"persistence",
"and",
"latency",
"virology",
"innate",
"immunity",
"gene",
"expression",
"immunology",
"immune",
"suppression",
"biology",
"microbiology",
"molecular",
"cell",
"biology",
"immun... | 2013 | Human Papillomavirus (HPV) Upregulates the Cellular Deubiquitinase UCHL1 to Suppress the Keratinocyte's Innate Immune Response |
Cox regression is commonly used to predict the outcome by the time to an event of interest and in addition , identify relevant features for survival analysis in cancer genomics . Due to the high-dimensionality of high-throughput genomic data , existing Cox models trained on any particular dataset usually generalize poorly to other independent datasets . In this paper , we propose a network-based Cox regression model called Net-Cox and applied Net-Cox for a large-scale survival analysis across multiple ovarian cancer datasets . Net-Cox integrates gene network information into the Cox's proportional hazard model to explore the co-expression or functional relation among high-dimensional gene expression features in the gene network . Net-Cox was applied to analyze three independent gene expression datasets including the TCGA ovarian cancer dataset and two other public ovarian cancer datasets . Net-Cox with the network information from gene co-expression or functional relations identified highly consistent signature genes across the three datasets , and because of the better generalization across the datasets , Net-Cox also consistently improved the accuracy of survival prediction over the Cox models regularized by or . This study focused on analyzing the death and recurrence outcomes in the treatment of ovarian carcinoma to identify signature genes that can more reliably predict the events . The signature genes comprise dense protein-protein interaction subnetworks , enriched by extracellular matrix receptors and modulators or by nuclear signaling components downstream of extracellular signal-regulated kinases . In the laboratory validation of the signature genes , a tumor array experiment by protein staining on an independent patient cohort from Mayo Clinic showed that the protein expression of the signature gene FBN1 is a biomarker significantly associated with the early recurrence after 12 months of the treatment in the ovarian cancer patients who are initially sensitive to chemotherapy . Net-Cox toolbox is available at http://compbio . cs . umn . edu/Net-Cox/ .
Survival analysis is routinely applied to analyzing microarray gene expressions to assess cancer outcomes by the time to an event of interest [1]–[3] . By uncovering the relationship between gene expression profiles and time to an event such as recurrence or death , a good survival model is expected to achieve more accurate prognoses or diagnoses , and in addition , to identify genes that are relevant to or predictive of the events [4] , [5] . The Cox proportional hazard model [6] is widely used in survival analysis because of its intuitive likelihood modeling with both uncensored patient samples and censored patient samples who are event-free by the last follow-up . Due to the high dimensionality of typical microarray gene expressions , the Cox regression model is usually regularized with penalties such as penalty in ridge regression [7]–[10] , Lasso regularization [11]–[16] and regularization in Hilbert space [17] . While those penalties were designed as a statistical or algorithmic treatment for the high-dimensionality problem , these Cox models are still prone to noise and overfitting to the low sample size . An important prior information that has been largely ignored in survival analysis is the modular relations among gene expressions . Groups of genes are co-expressed under certain conditions or their protein products interact with each other to carry out a biological function . It has been shown that protein-protein interaction network or co-expressions can provide useful prior knowledge to remove statistical randomness and confounding factors from high-dimensional data for several classification and regression models [18]–[21] . The major advantage of these network-based models is the better generalization across independent studies since the network information is consistent with the conserved patterns in the gene expression data . For example , previous studies in [18] , [20] discovered that more consistent signature genes of breast cancer metastasis can be identified from independent gene expression datasets by network-based classification models . The observations also motivated several graph algorithms for detecting cancer causal genes in protein-protein interaction network [22] , [23] . In this article , we propose a network-based Cox proportional hazard model called Net-Cox to explore the co-expression or functional relation among gene expression features for survival analysis . The relation between gene expressions are modeled by a gene relation network constructed by co-expression analysis or prior knowledge of gene functional relations . In the Net-Cox model , a graph Laplacian constraint is introduced as a smoothness requirement on the gene features linked in the gene relation network . Figure 1 illustrates the general framework of Net-Cox for utilizing gene network information in survival analysis . In the framework , the cost function of Net-Cox , shown in the box , combines the total likelihood of Cox regression with a network regularization . The total log-likelihood is a function of the linear regression coefficients and the base hazard on each followup time , represented by the likelihood ratios with the patient gene expression data and the survival information specified by followup times and event indicators . The gene network is either constructed with gene co-expression information or a given gene functional linkage network . The gene network is modeled as a constraint to encourage smoothness among correlated genes , for example gene and in the network , such that the coefficients of the genes connected with edges of large weights are similarly weighted . The cost function of Net-Cox can be solved by alternating optimization of and by iterations . An algorithm that solves the Net-Cox model in its dual representation is also introduced to improve the efficiency . The complete model is explained in detail in Section Materials and Methods . In this study , we applied Net-Cox to identify gene expression signatures associated with the outcomes of death and recurrence in the treatment of ovarian carcinoma . Ovarian cancer is the fifth-leading cause of cancer death in US women [3] . Identifying molecular signatures for patient survival or tumor recurrence can potentially improve diagnosis and prognosis of ovarian cancer . Net-Cox was applied on three large-scale ovarian cancer gene expression datasets [3] , [24] , [25] to predict survivals or recurrences and to identify the genes that may be relevant to the events . Our study is fundamentally different from previous survival analysis on ovarian cancer [3] , [24]–[26] , which are based on univariate Cox regression . For example , in [3] , gene expression profiles from 215 stage II–IV ovarian tumors from TCGA were used to identify a prognostic gene signature ( univariate Cox ) for overall survival , including 108 genes correlated with poor ( worse ) prognosis and 85 genes correlated with good ( better ) prognosis . In [24] , a Cox score is defined to measure the correlation between gene expression and survival . The genes with a Cox score that exceeds an empirically optimized threshold in leave-one-out cross-validation were reported as signature genes . Similarly , in [25] and [26] , a univariate Cox model was applied to identify association between gene expressions and survival ( univariate Cox ) . Our study is based on gene networks enriched by co-expression and functional information and thus identifies subnetwork signatures for predicting survival or recurrence in ovarian cancer treatment .
To evaluate the generalization of the models , we first measured the consistency among the signature genes selected from the three independent datasets by each method . Specifically , we report the percentage of common genes in the three rank lists identified by a method . This measurement assumes that even under the presence of biological variability in gene expressions and patient heterogeneities in each dataset , genes that are selected in multiple datasets are more likely to be true signature genes . Thus , higher consistency across the datasets might indicate higher quality in gene selection . In Figure 2 , we plot the number of common genes among the first ( up to 300 ) genes in the gene ranking lists from all of the three datasets for the death event and two datasets ( TCGA and Tothill ) for the recurrence event . For the parameter setting of Net-Cox , we fixed to be the optimal parameter in the five-fold cross-validation ( see Section Materials and Methods and report the results with and . Since the ranking lists of Net-Cox with are nearly identical to those of , they are not reported for better clarity in the figure . The first observation is that the gene rankings by Net-Cox are more consistent than those by and at all the cutoffs . Moreover , Net-Cox with identified more common signature genes than Net-Cox with . For example , for the tumor recurrence outcome , Net-Cox ( Co-expression ) with and identified 36 and 29 common genes among the first 100 genes in the gene ranking lists , Net-Cox ( Functional linkage ) with and identified 49 and 23 common genes , and and only identified 19 and 6 common genes , respectively . In general , variable selection by is not stable from high-dimensional gene expression data , and thus , the overlaps in the gene lists by are significantly lower than the other methods . It is also interesting to see the gradient of the overlap ratio from to , and then to ( ) , which indicates that , when a gene network plays more an important role in gene selection , the gene rankings tend to be more consistent . This observation is consistent with previous studies with protein-protein interaction network or gene co-expression network [18] , [20] , [21] . Note that since the overlaps are across three datasets for the death event and across two datasets for the recurrence event , the overlaps for the death event is expected to be lower than those for the recurrence event . Another important difference is that the same functional linkage network is always used while the co-expression network is dataset-specific . Thus , it is also expected that the overlaps by Net-Cox with the functional linkage network is higher than those by Net-Cox with the co-expression network . Together , the results demonstrate that Net-Cox effectively utilized the network information to improve gene selection and accordingly , the generalization of the model to independent data . Five-fold cross-validation was first conducted for parameter tuning for Net-Cox , and on each dataset . The optimal parameters of Net-Cox are reported in Table S1 . To test how well the models generalize across the datasets , we trained Net-Cox model , model , and model with the TCGA dataset , and then predicted the survival of the patients in the other two datasets with the TCGA-trained models . In training , we used the optimal and from the five-fold cross-validation to train the models with the whole TCGA dataset . The results are given in Table 2 . In all the cases , Net-Cox obtained more significant in the log-rank test than and . To further compare the results , we show the Kaplan-Meier survival curves and the ROC curves in Figure 3 . The first four columns of plots in the figure show the Kaplan-Meier survival curves for the two risk groups defined by Net-Cox with co-expression network and functional linkage network , , and . The fifth column of plots compare the time-dependent area under the ROC curves based on the estimated risk scores ( ) . In Figure 3 , in many regions , Net-Cox achieved large improvement over both and while the improvement is less obvious in several other regions . Overall , Net-Cox achieved better or similar AUCs in all the time points in the three plots . To evaluate the statistical significance of the differences between the time-dependent AUCs generated by Net-Cox and the other two methods , in Table S2 we report at each event time with the null hypothesis that the two time-dependent AUCs estimated by two models are equal . At many points of the event time , the time-dependent AUCs generated from Net-Cox are significant higher . The cross-validation log-partial likelihood ( CVPLs ) for the combinations of in the five-fold cross-validation are also reported in Table S3 . In all the cases , the optimal CVPLs of Net-Cox are higher than those of . was fine-tuned with 1000 choices of parameters with a very small bin size . In one of the cases ( TCGA: Recurrence ) , the optimal CVPL of is higher but in the other cases , the optimal CVPLs of Net-Cox are higher . Interestingly , the optimal is often or , indicating the optimal CVPL is a balance of the information from gene expressions and the network . The observations prove that the network information is useful for improving survival analysis . The left column of Figure S1 shows the average time-dependent area under the ROC curves based on the estimated risk scores ( ) of the patients in the fifth fold of the five repeats , and Table S4A and S4B show log-rank of the fifth fold of the five repeats . Net-Cox achieved the best overall survival prediction although the results are less obvious than those of the cross-dataset analysis . To understand the role of the gene network on the consistency in gene selection and the contribution to the log-partial likelihood , we tested Net-Cox with randomized co-expression networks . In each randomization , the weighted edges between genes were shuffled . We report the mean and the standard deviation of the percentage of overlapping genes of 50 randomizations in Figure 4 . Compared with the consistency plots with the true networks , the overlaps by Net-Cox on the randomized networks are much lower . We also report the boxplot of the log-partial likelihood in the same 50 randomized co-expression network with in Figure 5 . Compare with the log-partial likelihood with the real co-expression network , the range of the likelihood generated with the randomized networks is again lower by a large margin , which provides clear evidence that the co-expression network is informative for survival analysis . To further understand the role of the network information in cross-validation , we fixed the optimal parameter and conducted the same five-fold cross-validation with randomized co-expression networks to compute the CVPL with different in {0 . 01 , 0 . 1 , 0 . 5 . 0 . 95} . We repeated the process on 20 random networks for each . The boxplots of CVPLs with different s are shown in Figure 6 . In all measures , the CVPL with the true gene network is well above the mean of the 20 random cases . Another important observation is that , in both plots , when the randomized network information is more trusted with a smaller , the variance of the CVPLs is also getting larger; and the case with gives the worst CVPL mean and the largest variance . The result indicates that the randomized networks did not provide any valuable information in survival prediction . In contrast , with the true gene network , CVPLs generated from and are much higher than the ones from and ( ) . Again , these results convincingly support the importance of using the network information in survival prediction . Besides the 2647 Sloan-Kettering genes , all the 7562 mappable genes were also tested to evaluate Net-Cox , and by consistency of signature gene selection across the three datasets and accuracy of survival prediction in similar experiments . For the signature gene consistency , Figure S2 reports the percentage of common genes identified by each method in the ranking lists from the datasets . For the cross-dataset validation , Table S5 shows the log-rank test by training the TCGA datasets and test on the other two datasets , and Figure S3 shows the Kaplan-Meier survival curves for the two risk groups defined by Net-Cox , and and compares the time-dependent area under the ROC curves . For the five-fold cross-validation , the right column of Figure S1 shows the average time-dependent area under the ROC curves based on the estimated risk scores ( ) of the patients in the fifth fold of the five repeats , and Table S4C and S4D report log-rank test of the fifth fold of the five repeats . Overall , similar observations are made in experimenting with all the genes , though the improvements are less significant compared with the results by experimenting with the Sloan-Kettering cancer genes . One possible explanation is that , since the genes in the Sloan-Kettering gene list are more cancer relevant , the gene expressions may be more readily integrated with the network information . To analyze the signature genes identified by Net-Cox and , we created consensus rankings across the three datasets by re-ranking the genes with the lowest rank by Net-Cox and in the three datasets . Specifically , for each gene , a new ranking score is assigned as the lowest of its ranks in the three datasets , and then , all the genes were re-ranked by the new ranking score . The top-15 genes selected by Net-Cox and in the consensus rankings are shown in Table 3 . For the death outcome , nine signature genes , FBN1 , VCAN , SPARC , ADIPOQ , CNN1 , DCN , LOX , EDNRA , LPL , known to be related to ovarian cancer [27]–[35] are only discovered by Net-Cox . Among the ten common genes highly ranked by both Net-Cox and , three are collagen genes , and MFAP5 , TIMP3 , THBS2 , and CXCL12 are previously known to be relevant to ovarian cancer [36]–[39] . For the recurrence outcome , there are eleven common signature genes detected by both Net-Cox and . Net-Cox identified six additional ovarian cancer related signature genes [27]–[29] , [40]–[42] . The intersection of the 60 genes identified by Net-Cox in Table 3 contains 41 unique genes . We performed a literature survey of the 41 genes , out of which eighteen are supported by literature to be related to ovarian cancer shown in Table 4 . Most of the genes whose over-expression is associated with poor outcome are stromal or extracellular-related proteins . The genes such as VCAN , TIMP3 , THBS2 , ADIPOQ , PARC , NPY , MFAP5 , DCN , LOX , FBN1 , EDNRA , and CXCL12 are either components or modulators of extracellular matrix . In particular , LOX protein is involved in extracellular matrix remodeling by cross-linking collagens . Extracellular matrix remodeling through over-expression of collagens has been shown to contribute to platinum resistance , and platinum resistance is the main factor in chemotherapy failure and poor survival of ovarian cancer patients . Therefore , the identification of these extracellular matrix proteins as biomarkers of early recurrence and poor survival outcome in patients with ovarian cancer is consistent with the suggested pathobiological role of some of these proteins in platinum resistance . The top-100 signature genes with the largest regression coefficients by Net-Cox and learned from the TCGA dataset were mapped to the human protein-protein interaction ( PPI ) network obtained from HPRD [43] and also analyzed with DAVID functional annotation tool [44] . We report the densely connected PPI subnetworks constructed from the 100 genes selected by Net-Cox in Figure 7 . Compared with the PPI subnetworks generated from the 100 genes selected by , which contain 10 genes in the death subnetwork and 6 genes in the recurrence subnetworks ( shown in Figure S4 ) , the subnetworks are both larger and denser . The subnetworks identified from the co-expression networks in Figure 7 ( A ) are also larger than the subnetworks identified by the functional linkage network in Figure 7 ( B ) although many genes are shared . In the recurrence subnetworks , DCN , THBS1 , and THBS2 are members of the signaling KEGG pathway , and FBN1 controls the bioactivity of TGFs and relates to polycystic ovary syndrome [27] . In addition , ten genes are members of the focal adhesion KEGG pathway . These results point to a possibility that extracellular matrix signaling through focal adhesion complexes may constitute a pathway by which tumor cells escape chemotherapy and produce recurrence in chemotherapy [45] . Nine genes in the death subnetworks are members of the extracellular matrix ( ECM ) -receptor interaction KEGG pathway , and eighteen genes are annotated as ECM component . It was shown that ECM acts as a model substratum for the preferential attachment of human ovarian tumor cells in vitro [46] . FOS and JUN constitutes a nuclear signaling components downstream of extracellular signal-regulated kinases ( ERK1/2 ) that are mediators of growth factor and adhesion-related signaling pathways [47] . In addition , the genes are also enriched by regulation of gene expression , positive regulation of cellular process , developmental process , transcription regulator activity , and growth factor binding , all of which are well-known cancer relevant functions . The significantly enriched GO functions are listed in Table S6 and Table S7 . Extracellular matrix , extracellular region , and extracellular structure organization are consistently the most significantly enriched in the analysis . FBN1 was ranked 1st and 8th by Net-Cox with co-expression network in death and recurrence outcomes while only ranked FBN1 at 27th and 42nd , respectively . It is interesting to note that in the PPI subnetworks in Figure 7 ( A ) , FBN1 is connected with VCAN and DCN , both of which bear the annotation of extracellular matrix . The dense subnetwork boosted the ranking of FBN1 when Net-Cox was applied . We further validated the role of FBN1 in ovarian cancer recurrence using tumor microarrays ( TMAs ) consisting of a cohort of 78 independent patients ( see Section Materials and Methods ) . The expression level of FBN1 in ovarian cancer was scored by one observer who is blinded to the clinical outcome and described as: absent ( 0 ) , moderate ( 1 ) , and high ( 2 ) as illustrated by Figure 8 . In Figure S5A , the Kaplan-Meier survival curve shows the recurrence for groups by the FBN1 staining scores . At the initial 12 month , there is no difference in the recurrence rate between the groups with high and low FBN1 staining . After 12 month , the recurrence rate is lower in the low staining group . The similar patterns are also observed in the re-examination of the gene expression datasets in Figure S5B–E . Except the TCGA dataset on the Affymetrix platform ( Figure S5E ) , the pattern is clearly observed on the other two platforms , exon arrays and Agilent arrays . The discrepancy in the Affymetrix data could be related to data pre-processing or experimental noise . The plots suggest that FBN1 plays a role on platinum-sensitive ovarian cancer , and it could be developed as a target for platinum-sensitive patients with high FBN1 expression after about 12 months of the treatment . In the context of ovarian cancer treatment , a platinum-sensitive patient group can be defined as the group of patients who was free of recurrence or developed a recurrence after month of the treatment , where depends on the treatment plan and the follow-up . To better evaluate the role of FBN1 , we plot the Kaplan-Meier survival curve only for the platinum-sensitive patients in Figure 9 , i . e . we removed all the patients who developed recurrence before month and considered the follow-ups up to 72 months after the treatment . Due to the small sample size of the Mayo Clinic data , we set while for the gene expression datasets . In Figure 9A , the difference between the survival curves of low FBN1 staining and high staining patient groups is more significant . Similarly , Figure 9B–E show the survival curves for the platinum-sensitive patients for groups by the expression value of FBN1 in gene expression datasets . Compare to the matched curves in Figure S5 , the log-rank test are more significant except the TCGA dataset on the Affymetrix platform . Overall , the observations strongly support the hypothesized role of FBN1 in platinum-sensitive ovarian cancer patients .
Many methods were proposed for survival analysis on high-dimensional gene expression data with highly correlated variates [4] , [5] . In this paper , we propose Net-Cox , a network-based survival model , which to our knowledge is among the first models that directly incorporate network information in survival analysis . The graph Laplacian constraint introduced in Net-Cox is positive definite and thus , the Net-Cox model can be solved as efficiently as solving the model . In the dual form of Net-Cox , the model is scalable to genomic data with . Net-Cox not only makes survival predictions but also generate densely connected subnetworks enriched by genes with large regression coefficients . Net-Cox is most related to the shrinkage-based Cox models typically with ( Lasso ) and ( ridge ) penalties [5] . The purpose of applying regularization is to obtain a sparse estimate of the linear coefficients for solving the high-dimensionality problem . A Ridge penalty results in small regression coefficients to avoid overfitting problem with the small sample size . Compared with Net-Cox , neither Lasso nor ridge regularized Cox regression models are designed to incorporate any prior information among genes in the objective function for survival analysis . Another alternative solution in the literature is to apply dimension reduction methods to obtain a small number of features for subsequent survival analysis such as principal components analysis ( PCA ) [48]–[50] and partial least squares ( PLS ) [51]–[54] . These methods first compute the principle components to capture the maximal covariance with the outcomes or the maximal variance in the gene expression data , and then project the original high-dimensional gene expressions into a space of the directions of the principle components . Typically , these methods do not utilize any prior information . It is also usually difficult to interpret the results since the features in the project space are not directly mappable to any particular gene expression . There are also tree-based ensemble methods for survival analysis such as bagging of survival trees and random forests [55] , [56] . The tree-based methods usually also require a variable selection step to reduce the dimensionality . Multiple trees are then built from different samplings of training data and the results of the individual trees are aggregated for making predictions . Since the trees are built from random sampling , the resulted forests consist of different trees . Thus , the interpretation of the trees can be very difficult [4] . In [57] , a supervised group Lasso approach ( SGLasso ) is proposed to account for the cluster structure in gene expression data as prior information in survival analysis . In this approach , gene clusters are first identified with clustering . Important genes are then identified with Lasso model within each cluster and finally , the clusters are selected with group Lasso . More recently , the method in [58] combined a group Lasso constraint with Lasso Cox regression ( sparse-group Lasso ) . An additional parameter is introduced to balance between Lasso and group Lasso constraints . There are two major discrepancies between Net-Cox and the graph Lasso methods . First , while group Lasso assumes non-overlapping cluster structures among gene expressions , the gene network introduced in Net-Cox captures more global relation among all the genes . Specifically , beyond the cluster partition of genes into co-expression groups , a gene network represents pair-wise relationships between genes , which contain information of modularities , subgraph structures and other global properties such as centralities and closenesses . Second , while SGLasso adopts an unsupervised strategy to cluster genes as predefined groups for selection , Net-Cox identifies subnetwork signatures in a supervised manner , in which the selected subnetworks are enriched by genes with large regression coefficients by the design of the network constraint . In Table S3 ( g ) , we reported the results of group Lasso and sparse-group Lasso in the five-fold cross-validation with the R package “SGL” [58] . Compared with the CVPLs by the other methods in Table S3 ( a ) – ( f ) , the CVPLs in Table S3 ( g ) for group Lasso and sparse-group Lasso are consistently lowest when 25 or 100 gene clusters are used as groups . Thus , we did not further compare and analyze other results by the group Lasso models . The experiments in this paper clearly demonstrated that the network information is useful for improving the accuracy of survival prediction as well as increasing the consistency in discovering signature genes across independent datasets . Since the signature genes were discovered based on their relation in the networks , they enrich dense PPI subnetworks , which are useful for pathway analysis . It is also interesting to note that the PPI subnetworks of signature genes identified by Net-Cox on the TCGA dataset is enriched by extracellular matrix proteins such as collagens , fibronectin , and decorin . Previous gene expression studies had identified stromal gene signatures in ovarian tumors to be associated with poor survival outcome [24] . Therefore , our observation that the stromal subnetwork enriched by extracellular matrix proteins and stromal-related proteins is consistent with the role of stromal gene signature in poor prognosis . Finally , collagen matrix remodelling has been linked to platinum resistance , and ovarian cancer cells grown on collagens are more resistant to platinum agents than their counterpart grown on non-collagen substratum [59] . The tumor array validation indicates that FBN1 can serve as a biomarker for predicting recurrence of platinum-sensitive ovarian cancer .
We denote gene relation network by , where is the vertex set , each element of which represents a gene , and is a positively weighted adjacency matrix . is a diagonal matrix with and is the normalized weighted adjacency matrix by dividing the square root of the column sum and the row sum . Two gene relation networks were used with Net-Cox , the gene co-expression network and the gene functional linkage network . Three independent microarray gene expression datasets for studying ovarian carcinoma were used in the experiments [3] , [24] , [25] . The information of patient samples in each dataset is given in Table 1 . All the three datasets were generated by the Affymetrix HG-U133A platform . The raw . CEL files of two datasets were downloaded from GEO website ( Tothill: GSE9899 ) and ( Bonome: GSE26712 ) [24] , [25] . The TCGA dataset was downloaded from The Cancer Genome Atlas data portal [3] . The raw files were normalized by RMA [62] . After merging probes by gene symbols and removing probes with no gene symbol , a total of 7562 unique genes were derived from the 22 , 283 probes and overlapped with the functional linkage network for this study . Note that the Bonome dataset does not provide information on recurrence . Thus , only TCGA and Tothill datasets were used for studying recurrence while all the three datasets were used for studying death . In cross-dataset validation , the batch effects among the three datasets were removed by applying ComBat [63] . Besides testing all the genes , for a better focus on genes that are more likely to be cancer relevant , we derived a set of 2647 genes from the cancer gene list compiled by Sloan-Kettering Cancer Center ( SKCC ) [64] . The TCGA datasets with AgilentG4502A platform ( gene expression array ) and HuEx-1_0-st-v2 ( exon expression array ) were used to evaluate the signature gene FBN1 in Figure 9 . The processed level 3 data with expression calls for gene/exon were downloaded from the TCGA data portal . Consider the Cox regression model proposed in [6] . Given , the gene expression profile of patients over genes , the instantaneous risk of an event at time for the patient with gene expressions is given by ( 1 ) where is a vector of regression coefficients , and is an unspecified baseline hazard function . In the classical setting with , the regression coefficients are estimated by maximizing the Cox's log-partial likelihood: ( 2 ) where is the observed or censored survival time for the patient , and is an indicator of whether the survival time is observed ( ) or censored ( ) . is the risk set at time , i . e . the set of all patients who still survived prior to time . The commonly used Breslow estimator [65] to estimate the baseline hazard is given by ( 3 ) The partial likelihood and the Breslow estimator are induced by the total log-likelihood ( 4 ) with ( 5 ) The optimal regression coefficients is estimated based on the maximization of the total log-likelihood by alternating between maximization with respect to ( with Newton-Raphson ) and ( by equation ( 3 ) ) . In the analysis of microarray gene expressions , the number of gene features is larger than the number of subjects by several magnitudes ( ) . Fitting the Cox regression model will lead to large regression coefficients , which are not reliable . One possible solution is to introduce a constraint to shrink regression coefficients estimates towards zero [7] , [10] . In the model , the regression coefficients are estimated by maximizing the penalized total log-likelihood: ( 6 ) where is the penalty term and is the parameter controlling the amount of shrinkage . Another possibility is to introduce a constraint for variable selection [11] , [13] . The model penalizes the log-partial likelihood ( equation ( 2 ) ) by leading to: ( 7 ) In our experiments , R package “glmnet” [66] was used in the implementation of . We introduce a network-constraint to the Cox model as follows , ( 8 ) where is a positive semidefinite matrix derived from network information , is an identity matrix , and is the parameter controlling the weighting between the total likelihood and the network constraint . is another parameter weighting the network matrix and the identity matrix in the network constraint . For convenience , we define and rewrite the object function as ( 9 ) The term in equation ( 8 ) is a network Laplacian constraint to encode prior knowledge from a network . Given a normalized graph weight matrix , we assume that co-expressed ( related ) genes should be assigned similar coefficients by defining the following cost term over the coefficients , ( 10 ) As illustrated in Figure 1 , the Laplacian constraint encourages a smoothness among the regression coefficients in the network . Specifically , for any pair of genes connected by an edge , there is a cost proportional to both the difference in the coefficients and the edge weight . Large difference between coefficients on two genes connected with a highly weighted edge will result in a large cost in the objective function . Thus , the objective function encourages assigning similar weights to genes connected by edges of larger weights . By adding an additional constraint to weighted by , we obtain the network constraint = in equation ( 8 ) and ( 9 ) . The of similarly regularizes the uncertainty in the network constraint , which could have a singular Hessian matrix , and the parameter balances between the and the “Laplacian-norm” . The smaller the parameter , the more importance put on the network information . The objective function defined by equation ( 9 ) can be solved by alternating optimization of and . The maximization with respect to is done by Newton-Raphson method . The derivative of equation ( 9 ) is ( 11 ) where , and the second derivative is ( 12 ) where is the diagonal matrix with . Thus , the full algorithm to solve the Net-Cox model is given below . Using Newton-Raphson method to update requires inverting the Hessian matrix , which is time consuming and often inaccurate . An alternative approach is to reduce the covariant space from p to n , which relates to singular value decomposition that exploits the low rank of the gene expression matrix [10] . The equation ( 13 ) implies that for some . Thus , the dual form of equation ( 9 ) with respect to is ( 14 ) with and . In its dual form , it is clear that the new object function ( 14 ) is equivalent to equation ( 9 ) but the problem dimension is reduced from to . To determine the optimal tuning parameters and , we performed five-fold cross-validation following the procedure proposed by [10] on each of the three datasets . In the cross-validation , four folds of data are used to build a model for validation on the fifth fold , cycling through each of the five folds in turn , and then the pair that maximizes the cross-validation log-partial likelihood ( CVPL ) are chosen as the optimal parameters . CVPL is defined as ( 15 ) where is the optimal learned from the data without the fold . In the equation , denotes the log-partial likelihood on all the samples and denotes the log-partial likelihood on samples excluding the fold . We performed a grid search for the optimal maximizing the sum of the contributions of each fold to the log-partial likelihood in CVPL . In particular , was chosen from 1e-5 , 1e-4 , 1e-3 , 1e-2 , 1e-1 , 1 ( larger than 1 do not change the ranking of anymore ) , and was chosen from . Note that , when , Net-Cox ignores the network information and is reduced to . For , the optimal was chosen from 1000 by the “glmnet” parameter setting with the largest CVPL . The Log-rank test [67] and time-dependent ROC [68] were used to evaluate measurements of the prediction performance by a survival model . For the gene expression profile in the test set , the prognostic indexes is computed , where is the regression coefficients of the survival model , to rank the patients by descending order . We assigned the top 40% of the patients as the group and the bottom 40% as the group . The Log-rank test is a statistical hypothesis test for comparison of two Kaplan-Meier survival curves with the null hypothesis that there is no difference between the population survival curves , i . e . the probability of an event occurring at any time point is the same for each population . The test statistic is compared with a distribution with one degree of freedom to derive the significance reflecting the difference between two survival curves . The log-rank test only evaluates whether the patients are assigned to the “right group” but not how well the patients are ranked within the group by examining the . A more refined approach is afforded by the time-dependent ROC curves [68] , [69] . Time-dependent ROC curves evaluate how well the classifies the patients into and prognosis groups . Letting , we can define time-dependent sensitivity and specificity functions at a cutoff point aswith being the event indicator at time [69] . The corresponding ROC curve for any time , , is the plot of versus with different cutoff point . is denoted as the area under the curve . A larger indicates better prediction of time to event at time , as measured by sensitivity and specificity evaluated at time . We plot the AUCs at each time to compare the methods . To select gene variables in the multi-variate scenario by Net-Cox and , we ranked the genes by the magnitude of the coefficients . To justify this simple ranking method , we examined the relation between the magnitude of the coefficients for each gene and the contribution of the gene to the log-partial likelihood in Figure S6 . It is clear in the plot that the genes towards the two tails of the ranking list contributes most of the likelihood , and the proportion of the contributions are consistent with the ranking . For , we ranked the genes by the first-time jump into the active set when decreasing the tuning parameter in the solution path . With approval by the Mayo Clinic Institutional Review Board , archived ovarian epithelial tumor specimens from patients with advanced-stage , high-grade serous , or endometrioid tumors obtained prior to exposure to any chemotherapy were utilized to construct the TMA array . The array was constructed using a custom-fabricated device that utilizes a 0 . 6-mm tissue corer and a 240-capacity recipient block . Triplicate cores from each tumor were included , as were cores of liver as fiducial markers and controls for immunohistochemistry reactions . Five-micrometer-thick sections were cut from the TMA blocks . Immunohistochemistry was performed essentially as described in [70] . Sections of tissue arrays were deparaffinized , rehydrated , and submitted to antigen retrieval by a steamer for 25 minutes in target retrieval solution ( Dako , Carpinteria , CA , USA ) . Endogenous peroxide was diminished with 3% for 30 min . Slides were blocked in protein block solution for 30 min and then blocked with avidin and biotin for 10 min each , followed by overnight incubation with 1∶1000 diluted Anti-FBN1 antibody ( HPA021057 , Sigma-Aldrich ) at . The sections were then incubated with biotinylated universal link for 15 min and streptavidin for 25 min at . Slides were developed in diaminobenzine and counterstained with hematoxylin . | Network-based computational models are attracting increasing attention in studying cancer genomics because molecular networks provide valuable information on the functional organizations of molecules in cells . Survival analysis mostly with the Cox proportional hazard model is widely used to predict or correlate gene expressions with time to an event of interest ( outcome ) in cancer genomics . Surprisingly , network-based survival analysis has not received enough attention . In this paper , we studied resistance to chemotherapy in ovarian cancer with a network-based Cox model , called Net-Cox . The experiments confirm that networks representing gene co-expression or functional relations can be used to improve the accuracy and the robustness of survival prediction of outcome in ovarian cancer treatment . The study also revealed subnetwork signatures that are enriched by extracellular matrix receptors and modulators and the downstream nuclear signaling components of extracellular signal-regulators , respectively . In particular , FBN1 , which was detected as a signature gene of high confidence by Net-Cox with network information , was validated as a biomarker for predicting early recurrence in platinum-sensitive ovarian cancer patients in laboratory . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microarrays",
"systems",
"biology",
"genomics",
"functional",
"genomics",
"mathematics",
"statistics",
"biology",
"computational",
"biology",
"statistical",
"methods"
] | 2013 | Network-based Survival Analysis Reveals Subnetwork Signatures for Predicting Outcomes of Ovarian Cancer Treatment |
It is well established that the variability of the neural activity across trials , as measured by the Fano factor , is elevated . This fact poses limits on information encoding by the neural activity . However , a series of recent neurophysiological experiments have changed this traditional view . Single cell recordings across a variety of species , brain areas , brain states and stimulus conditions demonstrate a remarkable reduction of the neural variability when an external stimulation is applied and when attention is allocated towards a stimulus within a neuron's receptive field , suggesting an enhancement of information encoding . Using an heterogeneously connected neural network model whose dynamics exhibits multiple attractors , we demonstrate here how this variability reduction can arise from a network effect . In the spontaneous state , we show that the high degree of neural variability is mainly due to fluctuation-driven excursions from attractor to attractor . This occurs when , in the parameter space , the network working point is around the bifurcation allowing multistable attractors . The application of an external excitatory drive by stimulation or attention stabilizes one specific attractor , eliminating in this way the transitions between the different attractors and resulting in a net decrease in neural variability over trials . Importantly , non-responsive neurons also exhibit a reduction of variability . Finally , this reduced variability is found to arise from an increased regularity of the neural spike trains . In conclusion , these results suggest that the variability reduction under stimulation and attention is a property of neural circuits .
Traditionally , neuroscience aims to discover the neural mechanisms underlying perceptual , cognitive and motor functions by examining neural responses as subjects repeatedly perform a behavioral task . Typically , neural responses are extracted by averaging over those trials and the obtained firing rates are often the only information retained . This approach discards the high firing irregularity and the high variability across trials that individual neurons activity exhibit [1] , [2] , fluctuations that a priori limit information encoding . At different scales , high fluctuations are also observed in the so-called ongoing activity , and have been shown to play a role on the task-induced activity [3]–[6] . Therefore , the challenging question is: on a single-trial basis , how and in which conditions these a priori detrimental fluctuations allow an efficient information encoding ? Recent experimental studies have examined the neural variability across a variety of species , cortical areas , brain states and stimulus conditions [7]–[10] . Measuring the neural variability with the Fano factor , the mean-normalized variance of the neural spike counts over trials , these studies have found that stimuli generally reduced neural variability [10] , in line with previous results in the visual system [11] . Additionally , neural variability has been found to decrease in an attentional paradigm [7] , [9] . Theoretically , using a rate model , a recent study [12] , [13] has proposed that variability reduction arises from a stimulus induced suppression of an otherwise chaotic ongoing state . Using a spiking network model , we demonstrate here that the variability reduction can arise from an alternative network effect presented in the framework of attractor networks . The formalism of attractor dynamics offers a unifying principle for representation and processing of information [14]–[18] . Co-activation of neurons induces stronger mutual synaptic connections , leading to assembly formation . Reverberatory activity between assembly members can then lead to memory by the persistence of neural activation . The concept of neural assemblies was later formalized in the framework of statistical physics [14]–[16] , where these co-activated neurons lead to attractors in the phase space of the recurrent neural dynamics: patterns of co-activation can represent fixed points from which the dynamical system evolves . In this framework , we show that during spontaneous activity , as measured by the mean-normalized variance of the spike count ( the Fano factor ) , neural variability is high when the network exhibits noise-driven excursions between multiple attractors . The application of an external stimulation stabilizes one specific attractor and suppresses the excursions between different attractors , leading to a reduction of neural variability . After an exhaustive study of the Fano factor changes in the network , we conclude that variability reduction is associated with one fundamental condition , namely that in the spontaneous condition , the network working point is around the edge of the bifurcation above which multiple stable ( multistable ) activated attractors appear . Moreover , we show that the reduced variability can be attributed to an increased regularity of the spike trains , as measured by the coefficient of variation ( CV ) of the interspike interval ( ISI ) distribution .
In the attractor network considered here ( see Figure 1A and Materials and Methods ) , each excitatory population is selective for a specific external stimulus . In each of these populations , the recurrent weights are therefore assumed to have increased by Hebbian learning mechanisms to a value , called the cohesion level . The inhibition level is regulated by the GABA synaptic connections provided by the inhibitory population . The stationary and stable states ( attractors ) of this network can be studied using a standard mean-field approximation [16] , [19] , applied here to the case when input rate fluctuations are absent . Figure 1B plots the obtained bifurcation diagram in the spontaneous condition . The diagram shows the firing rate difference between the possible stationary states as a function of the cohesion and inhibition levels . Two regions can be distinguished . First , a large region where there is a unique stationary stable state ( large dark blue region ) : this state corresponds to a low activation state where excitatory and inhibitory neurons fire at a low mean rate ( approximately 3 Hz and 9 Hz when , respectively ) . Second , above a bifurcation , in a region of intermediate inhibition level and sufficiently large cohesion level , the network exhibits multistability with the coexistence of 6 stable states , namely: 5 equivalent states where only one of the 5 specific populations is highly activated , and a low activation state as described above . This last attractor remains stable just above the bifurcation , because of a so-called subcritical bifurcation . Below the bifurcation , although the activated attractors are unstable , fluctuations can induce transient excursions of the network state towards them , the dynamics around these attractors being partially stable even if not globally . First , we focus on spontaneous activity . For a network working point around but below the bifurcation , for example for a cohesion level and an inhibition level , Figure 2A show the network spontaneous activity ( for times less than 10 s ) . The activity is irregular , not only from the timing of spikes but at the rate level , where abrupt changes occur from time to time . Each neural population has different rate fluctuations . The rate distribution for all selective populations ( see Figure 2B ) is large ( mean rate: 3 . 02 Hz; standard deviation: 4 . 53 Hz ) , with a unique peak at zero rate and with a long tail . These rate fluctuations cannot be explained solely by the input rate fluctuations: they reveal the excursions of the activity between the different network attractors . This type of network dynamics may be at the origin of the similar large firing rate fluctuations observed in the cortex of behaving monkeys ( Reynolds and Mitchell , personal communication ) . When an external stimulus is applied to a given population ( Hz to population 1 at time s in Figure 2A ) , not only the activity of this population increases but the neural activity fluctuations due to the wanderings between the different attractors is sharply reduced . Actually , under stimulation , only the attractor corresponding to high activity in this population and low activity in all others is stable . To study the changes in the neural variability in the network when a specific external stimulus is applied , we investigate the Fano factor reduction corresponding to the difference between the spontaneous condition and when one selective excitatory population is externally stimulated by a Poisson spike train with rate I . The Fano factor was calculated from scatter plots of the neural spike count variance versus mean by linear regression fits constrained to pass through the origin . The spike counts variance and mean were calculated for each neuron separated in windows of 100 ms and averaging over 1000 trials . Applying the mean-matched procedure of Churchland et al . [10] , neurons in the stimulated population define the non-matched case ( as stimulation increases their rate ) , whereas neurons in the non-stimulated populations define the matched case ( as stimulation changes only slightly their rates ) . Figure 3 plots the Fano factor without external stimulation , with external stimulation ( Hz ) and their difference as a function of the cohesion and inhibition levels . Figure 3A and 3B show the results for the matched condition and for the non-matched condition , respectively . Both cases show that a reduction of the Fano factor consistent with the experimental findings occurs around the bifurcation line ( black line ) . More precisely , the matched rate case shows that multistability is required for the Fano factor to change . The Fano factor is reduced on the right part of the bifurcation line , for , requiring then sufficient inhibition . For the non-matched rate case , the Fano factor is well reduced around the whole multistability region . Note that , due to the presence of noise , multiple attractors manifest themselves outside the region calculated with the mean-field approach ( see Figure 1B ) . Furthermore , the Fano factor level for those regions is consistent with the observed experimental values , namely: around 1 . 4 for the non-stimulated case and 1 for the stimulated case . Beyond the changes observed over trials , the spiking statistics , like the neural ISI distribution , is also likely to change under stimulation , although it has not been reported experimentally . To compute the ISI distribution , we have simulated the network activity over long time intervals ( s ) , and have characterized the ISI distribution by its coefficient of variation ( ) . We have compared the and the Fano factor for two values of the cohesion level , a low value for which there is no multistability ( ) and a high value for which there is ( ) . As shown in Figure 4 , the CV and the Fano factor are always very similar , and vary similarly across all conditions . Therefore , our model results suggest that the Fano factor change is due to an underlying change in the spiking statistics . For a fixed cohesion level allowing multistable states , Figure 5 shows for the non-matched and matched conditions the Fano factor reduction as a function of the external stimulation and inhibition levels . For the stimulated neurons , the neural variability reduction appears in regions where the spontaneous state is around the bifurcation and increases with the external stimulation . For the non-stimulated neurons , the reduction of neural variability emerges also in the same region . In the spontaneous state , neural variability is high because of the network state excursions between different attractors . The application of an external stimulation stabilizes one specific attractor and suppresses the wanderings between different attractors ( see Figure 2A ) , leading to a neural variability reduction . Figure 6 shows in more detail the evolution of the spike count mean , variance and Fano factor as a function of time for a network with a cohesion level and an inhibition level , as in Figure 2 . The first 1000 ms were simulated in the spontaneous condition . A specific external stimulation was applied from 1000 to 2000 ms to the selective neural population 1 . The top part of Figure 6 corresponds to the averaged results obtained from the 80 neurons of the stimulated population 1 , whereas the bottom part corresponds to the averaged results from the 320 neurons in the other non-stimulated populations . For the stimulated neurons , the spike count mean and variance increase when the stimulus is applied ( at 1000 ms ) , but in such a way that the Fano factor is reduced . For the non-stimulated neurons , the stimulation effect is to reduce both spike count mean and variance , in a way that reduces the Fano factor . The Fano factor was calculated from the spike count variance versus mean scatter plots , where each point represents one neuron , the Fano factor being deduced using a linear regression fit ( see Materials and methods ) . Recent experiments have studied the effect of attention on the neural variability over trials [7] , [9] , Single V4 cells were recorded in awake behaving monkeys when one stimulus in the neuron's receptive field was behaviorally attended or not . Both studies reported a relatively small but significant decrease of the Fano factor in the attended condition with respect to the non-attended one . In our simulations , we modeled the attentional bias by increasing the level of exogenous input to the stimulus-specific population when this stimulus was attended , similar to an increase of the stimulus-related input . In this sense , it can be regarded as a baseline shift mechanism , or an increase in contrast ( However , it does not accommodate any attentional gain mechanism that would be better modeled through changes in postsynaptic sensitivity , possibly through NMDA receptor dynamics ) . As the neural variability reduction increases with stimulus strength ( see Figure 5 ) , the application of an attentional bias therefore decreases the Fano factor compared to the non-attended case . The effect of attention on the Fano factor evolution in shown in Figure 7 by comparing with the non-attended case . These results show that the model reproduces the range of Fano factor reductions observed experimentally [7] , [9] . As a model prediction , we consider now the case where two stimuli are presented simultaneously in a neuron's receptive field and when attention is allocated to only one of them , a situation referred to lead to “biased competition” [20] , [21] . In the network , selective populations 1 and 2 encode each one of the two simultaneously presented visual stimuli: the target which should be attended and the distractor which should be ignored . After 500 ms of spontaneous activity , the two external visual stimuli are applied to these two populations for 1000 ms . In the attended condition , the selective population corresponding to the target receives the attentional bias . Figure 8 plots the neural variability reduction for the cohesion level . The variability reduction is obtained as the difference of the average Fano factor for the 80 neurons in the attended ( Figure 8A ) versus in the non-attended stimulated populations ( Figure 8B ) . For both populations , variability is reduced by attention around the region where the system is multistable without attention ( around ) . The mechanism responsible for the reduction of neural variability is identical to the one described above . In the condition without attention noise-driven excursions between attractors generates a high neural variability . Allocation of attention stabilizes one of the attractors , namely the one corresponding to higher activation of the attended target population and lower activation of the ignored distractor population .
In the spontaneous or undriven state , why do cortical circuits exhibit a relatively high degree of neural variability across trials ? Why does this variability decrease when a stimulus is presented or when attention is paid ? Here , we investigated what could underlie these phenomena in a realistic neural network . Our results show that , under spontaneous conditions , the high degree of neural variability in a neural circuit could essentially be due to fluctuation-driven excursions between the different attractors of the circuit dynamics . This is possible if , in the parameter space , the spontaneous state of the circuit resides around the edge of a bifurcation above which multistable attractors appear . The application of an external excitatory drive , either mediated by a sensory stimulus or by attention , stabilizes one specific attractor and suppresses in this way the transitions between the different attractors . This results in a net decrease in neural variability as measured as a by the Fano factor . More precisely , the matched rate case shows that multistability is required for the Fano factor to change . The Fano factor is reduced on the right part of the bifurcation line , requiring then sufficient inhibition ( ) . For the non-matched rate case , the Fano factor is well reduced around the whole multistability region . In conclusion our results show that , in the model parameter space , there exists a region where the Fano factor is reduced , both in the non-matched and the matched rate case . Because the spike count signal-to-noise ratio is increased , this reduction suggests an improved encoding of the external signal . Beyond the variability over trials , we have also shown that the of the neural ISI distributions varies similarly to the Fano factor across all conditions , meaning that the variability reduction is due to a concomitant increase of the spike trains regularity . It would be interesting to verify this model prediction experimentally . However , due to relatively short recorded time intervals , this quantity may be difficult to measure and the [22] , which requires only the knowledge of two consecutive ISIs , could be employed instead . The above conclusions have been obtained for the present heterogeneously connected network and rely on the existence of multiple attractors . However , the present scenario does not depend on the specific network structure , provided the network exhibits multistability . In this case , there is a region of the parameter space where there is strict multistability , meaning the co-existence of multiple stable states . From dynamical systems theory , it is known that close to this region , the unstable attractors can still transiently attract the dynamics , a behavior which will be favored by fluctuations . Note that , because multistability is needed here to reproduce the experimental observations , this excludes a priori single neuron mechanisms . In a recent study [12] , [13] , the authors have proposed that the ongoing spontaneous activity is chaotic , and that stimulation suppresses this chaos . They use a phenomenological firing rate model which allows a theoretical understanding of the network behavior . The realism of the present model allows a quantitative comparison with experimental results but prevents at the same time such a theoretical understanding . Only could we predict the stationary states of the network using a mean-field approach . Beyond the naïve analogy that stimulation suppresses ongoing fluctuations in the two models , a number of differences between the two models ( in their case: rate model , no noise , temporal input , phase transition ) indicate that the two scenarios are different . Further experimental and theoretical evidence supports the present scenario . At the microscopic level and using optical imaging , Arieli et al . [4] ( see also [5] ) first showed that spontaneous activity is highly coordinated across large neural assemblies in the primary visual cortex ( V1 ) of an anesthetized cat . Furthermore , the pattern of co-activation is feature-specific in the discharge of individual neurons and is temporally locked to the activation of other cells with similar orientation preferences whose spatial organization is described by orientation maps . Finally , the variability of such ongoing activity can explain much of the variability in subsequent sensory-evoked responses , indicating a potential link with perception . Blumenfeld et al . [23] accounted for this type of cellular ongoing activity by assuming that this activity resulted from noise-driven transitions between multistable attractors of the intracortical network . They suggested a rate model endowed with a simple local connectivity rule , and showed that it yields attractor states that are highly similar to the orientation maps alternatively activated in the absence of stimulation . They also considered the case where the activity is evoked by a visual stimulus and showed how a structured afferent input can select the orientation map that matches the orientation of the stimulus . Their model therefore suggests that orientation maps are encoded in the lateral connections , and that these connections can generate an orientation map both when the activity is spontaneous and when it is evoked by a visual stimulus . In a recent work , where the same biophysical spiking neural network was used , it was shown that , using the Fisher information , the network neural activity best encoded a small external input , or modulatory input ( like it is believed to be the case for attention ) , in the region of the parameter space where excitatory and inhibitory input currents almost balance each other , and which correspond to ( Deco and Hugues , PLoS One , in press ) . This regime of balanced input is supported by experimental evidence in vitro [24] and in vivo [25] , [26] . Taken together with the present results , we can conclude that , in the region of the parameter space around the bifurcation and where input currents almost balance , our model agrees with all the experimental findings , which suggests that these observations may also correspond to a region of best stimulus encoding . At the global level of large-scale neural systems , a broad body of experimental work , mainly using the BOLD fMRI hemodynamic response , has suggested that brain activity during resting state is not random but has a spatiotemporal structure ( for a review see [27] ) . From these findings , ongoing neural activity may therefore be organized in a series of functional networks , so-called resting state networks . Suggested by a modeling study , these networks may also emerge from noise-induced transitions between multiple oscillatory brain states [28] . In this sense , the model of Blumenfeld et al . [23] at the microscopic cellular level , and the model of Deco et al . [28] at the global neuroanatomical level , propose that spontaneous ongoing activity is built up with multiple attractors , each one related with different specific stimulations or tasks , and that this activity fluctuates during spontaneous activity ( or rest ) due to transitions between those attractor states , induced by noise and unstructured input .
The spiking activity of neurons in the network is described by an integrate-and-fire model . Integrate-and-fire ( IF ) neurons are point-like elements , whose dynamical state is described by their membrane potential . An IF neuron can be described by a basic circuit consisting of a cell membrane capacitance in parallel with a membrane resistance , driven by input currents coming from connected neurons . Hence , the subthreshold dynamics of the membrane potential of each neuron in the network is given by the following equation: ( 1 ) where is the membrane leak conductance , is the resting potential , and is the synaptic current . The membrane time constant is defined by . When the voltage across the membrane reaches a given threshold , the neuron generates a spike which is then transmitted to other neurons and the membrane potential is instantaneously reset to and maintained there for a refractory time during which the neuron is unable to produce further spikes . The spikes arriving to a given neural synapse produce an input to the neuron which induce post-synaptic excitatory or inhibitory potentials ( through a low-pass filter formed by the membrane and synaptic time constants ) . In Equation 1 , , , , and are the synaptic conductances , and , the excitatory and inhibitory reversal potentials , respectively . The dimensionless parameters of the connections are the synaptic weights . The NMDA currents are voltage dependent and they are modulated by intracellular magnesium concentration . The gating variables are the fractions of open channels of neurons and they are given by: ( 2 ) ( 3 ) ( 4 ) ( 5 ) ( 6 ) The sums over the index represent all the spikes emitted by the presynaptic neuron j ( at times ) . In Equations 2–6 , and are the rise and decays times for the NMDA synapses , and and the decay times for AMPA and GABA synapses . The rise times of both AMPA and GABA synaptic currents are neglected because they are short ( <1 ms ) . The values of the constant parameters and default values of the free parameters used in the simulations are displayed in Table 1 . We use the same network to study both situations , namely the effects of stimulation and attention on the neural variability over trials . The network consists of ( in our simulations ) interacting neurons , where are excitatory ( pyramidal ) cells and are inhibitory cells ( interneurons ) , consistent with the neurophysiologically observed proportions [29] . We use an attractor network where neurons are organized into a discrete set of populations ( see Figure 1A ) . There are three different population types , namely: 1 ) the inhibitory population , 2 ) the excitatory non-selective population and 3 ) the excitatory selective population . The inhibitory population is made of the inhibitory neurons in the modeled brain area and mediates competition in the attractor network by distributing a global inhibitory signal . The non-selective population is composed of all excitatory neurons that are not receiving any specific external input and which therefore provides a background level of excitation . The remaining excitatory neurons are clustered in different populations , 5 in the simulations reported here . Each contains neurons ( in our simulations ) which are sensitive to a specific external stimulus . The network is fully connected , meaning that each neuron in the network receives excitatory and inhibitory synaptic contacts . The connection strengths between and within the populations are determined by dimensionless weights . We assume that the connections are already formed , e . g . by earlier self-organization mechanisms , as if they were established by Hebbian learning , with the coupling between two neurons being strong if their activities are correlated and weak if they are anticorrelated . The recurrent self-excitation within each selective population is given by the weight ( ) , which is called the cohesion level , and the weaker connection between them by the weight ( ) . The synaptic efficacy depends on by the relation . This serves to ensure that the average excitatory synaptic efficacy will remain constant as is varied across conditions . Neurons in the inhibitory population are mutually connected with an intermediate weight . These neurons are also connected with all excitatory neurons with the same intermediate weight , which for excitatory-to-inhibitory connections is and , for inhibitory-to-excitatory connections , is denoted and called the inhibition level . Neurons in each excitatory population are connected to neurons in the population with a feedforward and feedback synaptic weights and , respectively . The remaining connections are all set to the baseline value , i . e . to 1 . All neurons in the network always receive an external background input from external neurons emitting uncorrelated Poisson spike trains at rate . The resulting spike train is still a Poisson spike train , with rate . More specifically , and for all neurons inside a given population p , the resulting spike train is assumed to have a time-varying rate , governed by ( 12 ) where ms , kHz , kHz is the standard deviation of and is a normalized Gaussian white noise . Due to noise , negative values of that could arise are rectified to zero . These input rate fluctuations represent the noisy fluctuations that are typically observed in vivo . Additionally , neurons in a specific selective population could receive other inputs when an external stimulus is applied or when attention is allocated to that population . These inputs are specified by adding a corresponding rate to the rate of the background Poissonian input spike train . Without stimulation , all neurons only receive the background input . In the stimulation case , the first selective population is stimulated , receiving an extra input whose rate is . Without external stimulation , the spontaneous activity of the network consists in a noise-driven wandering between different attractors ( See Results for more detailed explanations ) , each corresponding to higher activation in one selective population and lower activation in all the others . Stimulation onset stabilizes one attractor , corresponding to the high activation of the stimulated selective population . The spiking activity for one trial is simulated for 500 ms without stimulation , allowing the network activity to stabilize , and the stimulus is then presented during 100 ms . Results during the stimulus period are averaged over 1000 trials initialized with different random seeds . We analyze the effect of attention on the neural variability and study the encoding of an attended stimulus . For this , we first address the recent experimental results when one stimulus is presented in a neuron's receptive field [7] , [9] . In a second part , as a prediction , we analyze the case where two stimuli are presented in a neuron's receptive field , a case used to elicit “biased competition” [20] , [21] . In the model , when attention is applied to a given stimulus , a biasing input corresponding to a rate is added to the input of the corresponding stimulus specific population [30] , [31] . In the case of one stimulus , the effect of attention is consequently assimilable to an increase of the stimulus strength or contrast , and is essentially a particular case of the stimulation case . In the case of two stimuli , each simulation started with a period of 500 ms ( for network activity stabilization ) . Then , during a period of 1000 ms , an identical stimulus was presented to selective populations 1 and 2 , represented by the corresponding extra rates Hz , respectively . Two cases were compared: with and without attention . In the case with attention , an extra attentional bias was added to the population 1 , corresponding to the attended spatial location ( i . e . Hz ) . In the case without attention , no bias was applied . The spiking activity was averaged over 2000 trials initialized with different random seeds . In the spiking simulations , we characterized neural variability using the mean-normalized variance of the spike counts , i . e . the Fano factor . It is defined as , where is the variance and is the mean of the spike counts of a neuron in a time window W . In all cases , we used a time window of 100 ms . The Fano factor measures the noise-to-signal ratio and therefore characterizes the neural variability over trials . For example , for a Poisson process , the variance equals the mean spike count ( ) for any length of the time window . We calculated the Fano factor by fitting a linear regression ( constrained to pass through the origin ) to scatter plots of the spike-count variance versus mean for each of the 80 neurons in the analyzed population . The variance and mean of the spike counts were calculated for each individual neuron by averaging over 1000 trials . In conditions for which the rate of some populations was not significantly affected , which is the case in practice for the non-stimulated neuronal populations , we used the mean-matching procedure for the Fano factor described in Churchland et al . [10] whose aim is to have the same distribution of mean firing rates and therefore factor out the rate for the Fano factor changes . In brief , the mean matching procedure consists of selecting neurons such that the distribution of the spike count for each condition ( with or without stimulation ) is the same . We took the greatest common distribution of the computed spike count for each condition . Individual points are then randomly removed until the actual distribution matched the common distribution . The mean-matched Fano factor is based on the remaining points . | To understand how neurons encode information , neuroscientists record their firing activity while the animal executes a given task for many trials . Surprisingly , it has been found that the neural response is highly variable , which a priori limits the encoding of information by these neurons . However , recent experiments have shown that this variability is reduced when the animal receives a stimulus or attends to a particular one , suggesting an enhancement of information encoding . It is known that a cause of neural variability resides in the fact that individual neurons receive an input which fluctuates around their firing threshold . We demonstrate here that all the experimental results can naturally arise from the dynamics of a neural network . Using a realistic model , we show that the neural variability during spontaneous activity is particularly high because input noise induces large fluctuations between multiple –but unstable- network states . With stimulation or attention , one particular network state is stabilized and fluctuations decrease , leading to a neural variability reduction . In conclusion , our results suggest that the observed variability reduction is a property of the neural circuits of the brain . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology",
"neuroscience"
] | 2012 | Neural Network Mechanisms Underlying Stimulus Driven Variability Reduction |
The matrix ( M ) proteins of rhabdoviruses are multifunctional proteins essential for virus maturation and budding that also regulate the expression of viral and host proteins . We have solved the structures of M from the vesicular stomatitis virus serotype New Jersey ( genus: Vesiculovirus ) and from Lagos bat virus ( genus: Lyssavirus ) , revealing that both share a common fold despite sharing no identifiable sequence homology . Strikingly , in both structures a stretch of residues from the otherwise-disordered N terminus of a crystallographically adjacent molecule is observed binding to a hydrophobic cavity on the surface of the protein , thereby forming non-covalent linear polymers of M in the crystals . While the overall topology of the interaction is conserved between the two structures , the molecular details of the interactions are completely different . The observed interactions provide a compelling model for the flexible self-assembly of the matrix protein during virion morphogenesis and may also modulate interactions with host proteins .
Rhabdoviruses are single-stranded RNA viruses that possess non-segmented negative-sense genomes encoding five open reading frames and form enveloped , bullet-shaped virions [1] . Dimarhabdoviruses , the supergroup of rhabdoviruses that infect mammals and mosquitoes [2] , are of considerable economic and social importance . Members of the genus Lyssavirus such as rabies virus cause lethal meningoencephalitis in humans and animals [3] while vesicular stomatitis virus ( VSV; genus Vesiculovirus ) causes symptoms clinically identical to those of foot-and-mouth disease in cattle and occasional , limited infections in humans [1] . The rhabdovirus matrix ( M ) protein is small ( ∼20–25 kDa ) but plays a number of roles during the replication cycle of the virus . The M protein is an important structural component of rhabdovirus virions , forming a layer between the glycoprotein- ( G- ) containing outer membrane and the nucleocapsid core composed of the virus nucleoprotein ( N ) , polymerase ( L ) , phosphoprotein ( P ) and RNA genome [4]–[6] . M condenses the nucleocapsid core into the ‘skeletons’ seen in mature virions [7] , [8] , recent evidence suggesting that M does so in association with pre-formed nucleocapsid–G plasma membrane microdomains [9] . M aggregates in vitro [10]–[12] to form long fibers [11] , the N terminus of the protein and a region between residues 121–124 being important for this self-association [11] , [13] . In addition to being spread through the cytosol of infected cells , M is targeted to mitochondria [14] , [15] , to nuclei [16] , and to plasma membranes [9] , [17] . M has been shown to interact directly with negatively-charged membranes [17]–[19] and can induce their deformation [20] . This interaction is mediated primarily by the N terminus of M , which contains several positively-charged amino acid residues , although in VSV residues 121–124 may also be involved [19]–[22] . In addition to its structural roles , M has been implicated in controlling the balance between transcription and replication of the viral genome [21] , [23] , in promoting budding [24] , [25] , and in modulating host-cell transcription [26] , -translation [27] and -apoptosis [28]–[31] . The specific protein∶protein interactions that mediate a number of these functions have been identified . A ‘late domain’ ( sequence PPXY ) located toward the N terminus of M promotes budding by interacting with the WW-domain of NEDD4 , a ubiquitin ligase that interacts with the vesicle formation and cargo sorting ESCRT complexes , although the precise mechanism by which this would facilitate budding remains unclear [24] , [25] , [32] , [33] . VSV M has been shown to inhibit the production of host proteins by binding directly to Rae1 and blocking the export of host mRNA from the nucleus , residues near the N terminus of M being essential for this interaction [27] , [34] . M also regulates the translation of host mRNA by binding to and/or modulating the phosphorylation state of translation initiation factors [26] , [35] , [36] , as well as inducing the production of viral proteins through an unknown mechanism [21] . To date , the only structural information available on rhabdovirus M proteins is the structure of a thermolysin-stable M core ( Mth ) of VSV Indiana ( VSVInd ) [19] . The proteolytic treatment removed the N-terminal 47 residues and cleaved the surface-exposed hydrophobic loop between residues 121–124 , and only residues 58–121 and 128–227 were visible in the structure . To further investigate the role of the N terminus and hydrophobic surface-exposed loop and to investigate the structural conservation of M across Rhabdoviridae , we solved the structures of full-length M from VSV serotype New Jersey ( VSVNJ ) and from the lyssavirus Lagos bat virus ( LBV ) . These structures reveal that rhabdovirus M proteins share a similar overall fold and self-associate via a stretch of amino acids ( in the otherwise-disordered N terminus ) that bind to a similar region on the globular domain , although the molecular details of the interaction interface differ dramatically between VSVNJ and LBV M . This inter-molecular interaction provides a plausible mechanism for the self-assembly of M , leading to enhanced affinity for membranes . Further , the differences in this interaction provide a structural framework for understanding the distinct cytopathic effects of vesiculoviruses and lyssaviruses .
The structure of VSVNJ M was solved by SeMet single-wavelength anomalous dispersion phasing and refined to 1 . 83 Å resolution with residuals R/Rfree = 0 . 157/0 . 179 ( Table 1 ) . VSVNJ M forms a globular domain with a central α helix sandwiched on one side by an extensive 5-stranded β sheet and on the other by two α helices and a smaller two-stranded β sheet ( Figure 1A ) that is very similar to the thermolysin-resistant core of M ( Mth ) from VSV serotype Indiana ( VSVInd ) , with 0 . 9 Å root-mean-squared displacement ( rmsd ) over 156 Cα atoms . Residues 121–128 , disordered in the VSVInd Mth structure , are observed in the structure of VSVNJ M with residues 121–124 forming a short stretch of α helix ( α2 . 5 , Figure 1A ) . The N-terminal 57 residues of VSVNJ M do not form part of this globular domain . While residues 1–40 and 53–57 are not well ordered and could not be modeled in electron density , residues 41–52 are located in strong electron density bound in a deep hydrophobic pocket formed by the loops between sheet β1 and helix α1 , the region between helices α2 and α2 . 5 ( including sheet β2 ) , and the stretch of 3–10 helix immediately preceding helix α3 . F46 is central to the interaction: the side chain of F46 sits deep in the hydrophobic pocket lined by the side chains of residues Y81 , V84 , L116 , Y131 , Y197 and the backbone between residues 114 and 116 ( Figure 2 ) . The backbone of F46 forms hydrogen bonds with the backbone of F78 and with a water molecule that bridges the backbone of F78 and side chain of Y131 . Residues 45–51 of the bound ligand interact with residues flanking the deep pocket into which the side chain of F46 binds . F45 sits in a shallow hydrophobic pocket lined by the hydrophobic side of the backbone peptide planes between residues 77–79 and by the side chains of P77 and R79 . The backbone of G47 forms H-bonds with the backbone of A118 and with a solvent atom that bridges the backbones of residues G47 , A118 and M51 . M48 sits in a shallow groove formed by the side chains of P77 , A118 , V122 and Y131 . E49 forms a hydrogen bond with the side chain of Q117 and with a water molecule that also hydrogen bonds with the side chain of Y197 . The backbone of D50 forms a hydrogen bond with the side chain of Q117 . M51 sits in a shallow hydrophobic pocket formed by the hydrophobic side chains of A118 , P120 , V122 , L123 and the hydrophobic peptide plane between residues 118 and 120 . Overall , these interactions bury 1050 Å2 of surface area . The co-localization of strong anomalous scattering with the two SeMet residues ( 48 and 51 ) allowed unambiguous identification of the bound peptide ( Figure S1A ) and SDS-PAGE analysis confirmed that the crystallized VSVNJ M was intact ( Figure S2A ) . Residue 52 of the bound peptide is 46 Å from the first ordered residue of the globular domain to which it binds ( S58 ) , a distance too great to be spanned by the missing 5 residues . Distance constraints dictate that this bound peptide derives from an adjacent molecule in the crystal , related by the crystallographic symmetry operator [−x−½ , −y−½ , z−½] with residue 58 lying 11 Å from residue 52 of the bound peptide ( Cα–Cα distance ) . While electron density linking residue 52 to residue 58 of the adjacent monomer is observed in maps calculated using data to 4 Å resolution , these residues could not be modeled because in higher resolution maps this density is significantly reduced , presumably due to disorder . The interaction of the globular domain with the N-terminal peptide of an adjacent molecule in the crystal gives rise to non-covalently linked linear polymers of VSVNJ M monomers ( Figure 3A ) . Asides from the bound N terminus , the most striking structural difference between VSVNJ M and VSVInd Mth in the globular domain is in the orientation of residues 191–202 , which form part of helix α3 and of the loop that precedes it ( Figure 4 ) . In VSVNJ M this loop has shifted toward the interacting N-terminal residues: the Cα atom of K196 moves 9 . 5 Å from its position in VSVInd Mth and residues 195–200 form a stretch of 3–10 helix . Y197 on this helix forms part of the deep hydrophobic pocket in which F46 resides , in addition to forming a water-mediated H-bonds with the side chain of E49 ( Figure 4 ) . Alignment of vesiculovirus M sequences ( Figure 5 ) shows that F46 and the residues that form the deep binding pocket in which it sits are highly conserved amongst VSV serotypes , although there are some conservative substitutions of flanking residues . In viruses Isfahan , Piry , Alagoa , Cocal and Chandipura the core F46 residue of the N-terminal interacting motif is conserved , but the flanking residues are significantly changed . However , side chains that form the hydrophobic pocket in which F46 is buried are conserved ( Y131 , F78 ) or conservatively substituted ( Y81F/C , V84A , L116M ) . In spring viremia of carp virus , a dimarhabdovirus [2] not assigned to the vesiculovirus genus that infects fish rather than mammals , residue F46 is not conserved and it is unclear whether M from this virus would be able to self-associate in a manner similar to that observed for VSVNJ M . The structure of LBV M was solved by SeMet multi-wavelength anomalous dispersion phasing and refined to 2 . 75 Å resolution with residuals R/Rfree = 0 . 207/0 . 255 ( Table 1 ) . Residues 48–202 of LBV M form a globular domain with an overall fold that closely resembles those of VSVInd Mth and VSVNJ M , with 2 . 8 and 3 . 1 Å root-mean-squared deviation between 138 and 139 equivalent Cα positions , respectively , despite the aligned residues sharing less than 10% sequence identity ( Figure 1 ) . While the central α helix and back 5-stranded β sheet overlay very well , the loop between sheets β2 and β3 is much shorter in LBV M than in VSVNJ M and no stretch of helix is present between these sheets . Helices α1 and α3 , sheets β6 and β7 and the loop between sheets β4 and β5 are also shifted between the LBV M and VSVNJ M structures . Strikingly , in the structure of LBV M a stretch of peptide is again observed bound in a shallow hydrophobic groove formed by the β1–α1 and β2–β3 loops of the globular domain ( Figure 1B ) . Anomalous difference density co-located with the Se atom of SeMet33 in the SeMet-labelled protein unambiguously identifies the bound peptide as LBV M residues 30–37 ( Figure S1B ) , no electron density being evident for residues 1–29 or 38–47 . This interaction is centred on residues 32–36 , which form a short stretch of left-handed polyproline-II helix ( Figure 2B ) . The backbone of residues 33–35 packs against the backbone of the β1–α1 loop of the globular domain , M33 forming two hydrogen bonds with Y67 . W112 in the β2–β3 loop sits under residues 33–35 , its indole nitrogen forming a hydrogen bond with the carbonyl oxygen of P34 ( Figure 6 ) . The side chain of P36 sits in a hydrophobic pocket formed by the P107 and W112 side chains and the backbone of M110 and N111 ( Figure 2B ) . Overall , these interactions bury 830 Å2 of surface area . SDS-PAGE confirmed that the crystals contained full-length LBV M ( Figure S2B ) and gel filtration analyses of full-length and truncated ( Δ1–45 ) LBV M were consistent with the N-terminal portion of M adopting an extended/disordered conformation in solution ( Figure S3 ) as has been observed previously for VSV M [13] . As in VSVNJ , the distance between the last residue of the bound peptide and the first residue of the LBV M globular domain is too great to be spanned by the intervening residues ( 49 Å from E37 Cα to E48 Cα ) , and the bound peptide must come from a neighbouring monomer . The interaction with the most likely monomer , related by the crystallographic symmetry operator [1+x−y , 1−y , 1−z] ( 23 Å from E37 Cα to E48 Cα; Figure S4 ) , generates linear polymers of non-covalently linked molecules in the crystal ( Figure 3B ) . The sequence of the interacting N-terminal region of LBV M is conserved in lyssaviruses , the only exception being the conservative substitution of M33 with leucine ( Figure 7 ) . The residues of the globular domain to which they bind are also conserved , with just two non-disruptive exceptions ( M110L in strain SADB-19 and N111S in strain ZAMRAV51 , Figure 7 ) .
In the structures of both VSVNJ and LBV M , a peptide from the otherwise-disordered N terminus of the protein is observed bound to the globular domain near the β1–α1 and β2–β3 loops ( Figures 1A & 1B ) . The overall location of this interacting region in the sequence of the proteins is similar , the interacting residues being less than 20 residues from the start of the globular domain ( Figure 1C ) , and in both the interaction is between adjacent molecules , thereby forming non-covalently linked linear polymers of M in crystallo ( Figure 3 ) . It is therefore particularly striking that the nature of the interfaces formed by the N-terminal interacting residues and the globular domains differ so significantly ( Figure 2 ) . The similar overall nature of the self-association , despite large differences in the molecular details of the interaction interfaces , is compelling evidence that the self-interaction is biologically relevant rather than being an artefact of crystallization . The self-association of VSVNJ M is centered on F46 , which binds into a deep hydrophobic pocket on the surface of the globular domain . Residues 45–47 adopt an extended conformation , flanked on either side by single turns of α helix . To the best of our knowledge this self-association interface shares no homology with previously identified protein∶protein interaction interfaces . In contrast , the peptide recognition cleft of LBV M is quite shallow and residues 33–36 ( sequence MPPP ) , which bind into this hydrophobic pocket on LBV M , form a short stretch of polyproline-II helix . The recognition of polyproline-II helices formed by proline-rich motifs ( PRMs ) is an important theme in protein∶protein interactions . Six classes of PRM-binding domains have previously been described: SH3 , WW , EVH1 , UEV and GYF domains and profilins [37] . These are generally characterized by the presence of a central tryptophan residue , around which the polyproline helix wraps , with shallow hydrophobic pockets accommodating the proline side chains [37] . LBV M exhibits this generic mode of binding ( Figure 6 ) , but details of the interaction differ from the known classes of PRM interactions . The structure of LBV M therefore reveals both a novel 7th family of PRM-binding domain and defines the interaction of this domain with its cognate ligand . The M proteins of rhabdoviruses play important roles in virus assembly . They condense the nucleocapsid cores into a tightly-coiled nucleocapsid-M complex ( termed ‘skeletons’ ) [7] , [8] , form a layer between the nucleocapsid and the surrounding lipid bilayer [4]–[6] , and promote virus budding [6] , [24] , [38]–[40] . An obvious functional implication of the self-association in the structures of VSVNJ and LBV M is in virus assembly , by facilitating long-range organization of M molecules and thereby enhancing the local concentration of M . Experiments investigating the self-association of VSV M support this hypothesis . M assembly is a two-stage process involving the sequential addition of M monomers to small pre-formed M nuclei to form fibres [10] , [11] . The N-terminal portion of M plays a critical role in the second step , polymerisation [11] , [13] . Treatment of M with trypsin gives rise to a stable fragment ( Mt ) spanning residues 44–229 [41] that retains the ability to form fibres but can only nucleate M aggregation in the presence of added Zn2+ [10] , [11] . However , M treated with therymolysin ( Mth ) , comprising residues 48–121 and ( 122 , 123 or 124 ) –229 , does not aggregate [13] , [19] . This is consistent with removal in Mth of F46 , the residue central to the interaction , and with the ‘untethering’ of the β2–β3 loop , which forms part of the interaction surface on the globular domain ( Figures 2 & S5 ) . We propose that the association between the N-terminal portion of M and the globular domain of an adjacent M molecule observed in our structures is the same as that which promotes addition of M monomers to pre-formed nuclei to yield large M fibers in vitro [11] and presumably promotes virus assembly in vivo . While it is possible that the movement of the loop that links sheet β5 to helix α3 in VSVNJ M relative to VSVInd M ( Figure 4 ) represents a conformational switch that facilitates nucleation or polymerization , the mechanism by which such a switch would be induced remains unclear . Mutational analysis of the β2–β3 loop supports a role for the observed interaction between the N-terminal portion of M and the globular domain of an adjacent molecule in virus morphogenesis . Substitution of VSVInd residues 121–124 ( sequence AVLA ) with DKQQ gives rise to a mutant M protein that shows reduced capacity to recruit free M or MDKQQ into pre-formed nucleocapsid-MDKQQ complexes , although the general ability to self-associate is maintained [21] . Mapping this mutation onto the structure of VSVNJ M ( Figure S5 ) reveals that the mutated residues surround the site of interaction of the N-terminal peptide , but they do not interfere with the burial of F46 into the deep hydrophobic pocket and would thus presumably not abolish the interaction entirely . A double mutation of M48 and M51 to arginine in VSVNJ yields a virus phenotype competent for assembly but unable to inhibit host-cell gene expression [42]–[44] . The side chain of M48 is not required for self-association , since in VSVInd it is replaced with valine ( Figure 5 ) . Further , while both M48 and M51 form part of the VSVNJ self-association interface , neither is completely buried ( Figures 2 & S5 ) and the observed interaction could most likely be maintained in the mutated M protein with little energetic penalty . Mutation of residues 35 and 36 in the N-terminal interacting region of lyssavirus M to serine and alanine , respectively , reduces viral fitness [45] . This is consistent with our proposed model , although further experiments are required to distinguish mutations that modulate self-association from those which interrupt the interaction of the PPXY ‘late domain’ with the host-cell budding machinery . Assembly of rhabdoviruses may require the M protein to interact with a circular or helical scaffold formed by the nucleocapsid [46]–[49] . As the non-covalent polymers of M observed in the crystal lattices are straight , the relative orientations of adjacent globular domains observed in the crystals might not reflect the packing of globular domains in the final assembled virion . However , the flexible nature of the tether that links one globular domain to the next , via interaction with the flexible N-terminal segment , could accommodate major rearrangements of adjacent globular domains . This would allow the required curvature of the M polymers and facilitate reorientation of M to interact with other components of the virion . Based on the dimensions of ‘shaved’ VSV virus particles and number of M molecules in such particles [50] , assuming a model where M lies immediately below the plasma membrane , the mean distance between the centres of adjacent M proteins would be ∼45 Å . This is significantly larger than the M-to-M distances observed in the non-covalent linear polymers formed within the crystals ( 35 Å for VSVNJ and 28 Å for LBV , Figure 3 ) and is consistent with a loose tethering of M proteins in the assembled virions . Such flexibility would allow for higher concentrations of M at points of higher membrane curvature , as has been observed recently for VSV M [9] . A similar beads-on-a-string arrangement in virions has been postulated for the influenza virus matrix protein [51] . The self-association of M identified in our structure informs previous experiments on the association of VSV M with membranes . M associates with membranes both in vitro [18]–[20] and in vivo [9] , [22] . M is thought to link the nucleocapsid and the envelope of the virus [4]–[6] , although recent evidence suggests that M might actually be recruited to pre-formed nucleocapsid–G plasma membrane microdomains [9] . Mt , in which the N-terminal 43 residues are removed by trypsin proteolytic cleavage , maintains its ability to interact with membranes [19] , [20] . The interaction is significantly weaker than for wild-type M , confirming that the positive-charge of the lysine-rich N terminus is important for membrane association , but the fact that some association is maintained suggests the presence of other , potentially weaker membrane-interaction interfaces on M [22] . In contrast , Mth is almost entirely unable to interact with membranes [19] , [20] . Mth has only 4 residues fewer at the N terminus than Mt [13] . Since none of these are positively-charged , the decrease in affinity can't be due to a loss of charge-mediated affinity for membranes . Furthermore , substitution of the hydrophobic side chains in the surface loop cleaved by thermolysin ( residues 121–124 ) for charged residues does not abolish membrane association [21] , indicating that an additional membrane-attachment interface has not been excised by the thermolysin treatment . We propose that the observed decrease in membrane affinity arises instead from the inability of Mth to self-associate . Polymerisation at the membrane , mediated by the self-association observed in our structure , would provide avidity enhancement of binding thus overcoming the lack of N-terminal charge in Mt . In addition to their role in virus assembly and budding , rhabdovirus M proteins are important for subverting the host immune response by suppressing the production of host genes . It has previously been observed that VSV M blocks host gene translation by binding directly and specifically to Rae1 , a protein involved in nuclear export of mRNA [34] . Substitution in M of residues 52–54 with alanine completely abolishes this interaction [27] , [34] . A second substitution in this area , M51R has the same effect , although a direct loss of interaction with Rae1 has not been shown for that mutant [52] . Our structure reveals that the Rae-1 binding site on VSV M partially overlaps with the N-terminal self-association motif ( Figure 1 ) . Steric considerations make it likely that self-association of VSV M and Rae-1 binding would be mutually exclusive . A direct interaction between a cellular protein and a sequence overlapping the N-terminal portion of M involved in self-association has also been observed in lyssaviruses . In this case the interaction is between the ‘late domain’ ( sequence PPEY , residues 35–38 ) and the WW domain of NEDD4 [24] , a ubiquitin ligase that interacts with proteins in the ESCRT pathway and promotes virus budding ( Figure 1 ) [25] . The polyproline-II helix conformation adopted by residues 32–36 of LBV M is entirely compatible with binding of this ‘late domain’ to the NEDD4 WW domain . As above , steric clashes would prevent simultaneous interaction of these residues with WW domains and with the globular domain of LBV M . Since PRMs and their binding motifs are such a common theme in cellular protein∶protein interactions it is likely that both the PRM motif and PRM-binding groove of LBV M also mediate specific interactions with other cellular proteins , although such binding partners are yet to be identified . Such an interaction between the self-association pocket on the globular domain of M and an unknown cellular protein has recently been identified for VSV . While mutation of the VSVInd M β2–β3 loop residues 121–124 to DKQQ ( MDKQQ ) interferes only modestly with the self-association of M ( see above ) , it produces a marked reduction in the amount of viral mRNA translated in infected cells [21] . As this phenotype can be rescued by co-infection with wild-type VSV it probably arises from a loss of function rather than a gain of inhibition . This phenotype was mapped specifically to the β2–β3 loop , reversion of residues 121–122 to the wild-type sequence ( AV ) restoring wild-type levels of viral mRNA translation [21] . It is likely that the yet-unidentified factor required to promote efficient viral mRNA translation binds in a manner similar to that observed for the N-terminal segment in our structure . As discussed above , V122 forms part of the binding pocket for M48 ( V48 in VSVInd ) . While mutation of V122 to lysine doesn't significantly impair self-association of VSVInd it is possible that the effect would be greater on ligands with a larger hydrophobic residue in a position equivalent to residue 48 . It is equally likely that V122 would be more buried in an interaction with this ( unknown ) cellular partner , reducing its ability to ‘swing away’ from the binding cleft . To summarise , both VSV and lyssavirus M have known cellular interaction motifs that overlap with the self-interaction motifs revealed by the structures presented here . However , the molecular details of the self-interaction motifs differ significantly and it is likely that the cellular binding partners of VSV and LBV M proteins are also distinct , consistent with the different host-range and cytopathogenicity of vesiculoviruses and lyssaviruses . The region between the β1–α1 and β2–β3 loops and the fragments of the otherwise-disordered N-terminal tails to which they bind are clearly hot-spots for rhabdovirus M protein∶protein interactions . This suggests a tempting evolutionary hypothesis to explain the similarity in overall interaction topology but difference in molecular interfaces between vesiculoviruses and lyssaviruses . The self-association grooves on the globular surfaces of the proteins and their cognate ligands might have evolved to mimic desirable protein∶protein interactions within the host cells , the functional constraint imposed by needing to remain competent for self-assembly ( and thus enable viral morphogenesis ) having maintained the overall topology of the interaction . The maintenance of interaction with cellular partners and self-association at the same locus on the protein raises a second interesting possibility , that the observed self-associations also play a role in regulating interaction of M with cellular partners by ( partially ) sequestering the binding interfaces , as has been suggested for the HIV M protein [53] . This would provide a raison d'être for the shortened M gene product observed in VSV ( residues 51–229 ) [43]; it possesses the deep hydrophobic peptide-binding groove but not the N-terminal peptide , thereby providing the virus with a pool of M protein ( with unoccupied binding grooves on their globular surfaces ) able to interact with cellular binding partners .
The matrix ( M ) protein from Lagos bat virus ( LBV ) was cloned , expressed , purified and diffraction data were collected as described previously [54] . M from VSV serotype New Jersey ( VSVNJ ) was cloned into pOPINS , encoding an N-terminal His6-SUMO fusion tag , and selenomethionine-labelled ( SeMet ) protein was expressed and purified as described for LBV M [54] . Purified VSVNJ M was concentrated to 1 . 2 mg/mL and crystallization trials were attempted at 20 . 5°C in sitting drops containing 100 nL protein and 100 nL precipitant solution equilibrated against 95 µL reservoirs in 96-well plates . Crystals of SeMet VSVNJ M grew in 20% v/v isopropanol , 20% w/v PEG 4000 and 0 . 1 M sodium citrate ( pH 5 . 6 ) and were cryoprotected by a quick pass through reservoir solution supplemented with 20% v/v glycerol before flash cryocooling in a cold ( 100 K ) stream of nitrogen gas . Diffraction data were recorded from a single crystal of SeMet VSVNJ M at a wavelength of 0 . 9803 Å , to maximize the selenium anomalous signal , on ESRF beamline ID23EH1 . Diffraction data were processed using XDS [55] and SCALA [56] as implemented by the xia2 automated data processing package ( Winter et al . , in preparation ) . The structures of VSVNJ and LBV M were solved at 1 . 83 Å and 3 . 0 Å resolution by single- and multiple-wavelength anomalous dispersion analysis using the diffraction data described above and elsewhere [54] , respectively . For both , selenium atoms were located and their positions refined using SHELXD [57] and SHARP [58] as implemented by autoSHARP [59] . For VSVNJ M , electron density maps were solvent-flatted using SOLOMON [60] and DM [61] . The structure was traced by manually placing VSVInd Mth ( PDB ID 1LG7; [19] ) into electron density and subsequently rebuilding in COOT [62] . For LBV M , the experimental map was solvent flattened and an initial partial model traced using cycles of automatic building in RESOLVE and restrained refinement in REFMAC5 [63]–[65] followed by manual rebuilding in COOT . The initial LBV M model was placed into the high-resolution ( 2 . 75 Å ) native data by rigid-body refinement in REFMAC5 . Final TLS+restrained refinement of both structures was performed in REFMAC5 [66] , earlier refinement of LBV M having been performed using BUSTER/TNT [67] . The MolProbity server [68] and the validation tools present in COOT informed the refinement of both structures . Refinement statistics are shown in Table 1 and final refined coordinates and structure factors have been deposited with the PDB with accession IDs 2w2r ( VSVNJ M ) and 2w2s ( LBV M ) . Superpositions and structure-based alignment of VSV and LBV M were performed using SSM [69] and MUSTANG [70] . Vesiculovirus and lyssavirus M protein sequences were aligned using MUSCLE [71] and sequence alignment figures were produced with the assistance of JalView [72] and Inkscape ( http://www . inkscape . org ) . Molecular graphics were produced using PyMOL ( DeLano Scientific LLC ) . | Rhabdoviruses are of considerable socioeconomic importance . For example , rabies virus causes lethal encephalitis resulting in approximately 50 , 000 human deaths per year . Rhabdoviruses infect cells and propagate despite having small genomes that encode only five multifunctional proteins . One of these , the matrix protein , plays a structural role in virus assembly in addition to modulating the production of host and virus proteins , promoting viral egress from the host cell and modulating cell death . We have solved the 3-dimensional crystal structures of matrix proteins from two distantly related rhabdoviruses: Lagos bat virus and vesicular stomatitis virus . The two proteins have very similar structures despite having dissimilar amino acid sequences . Surprisingly , for both we observe self-association between a pocket on the main globular domain and one extremity of an adjacent molecule in the crystal . Repetition of this interaction gives rise to non-covalent polymers of matrix proteins , adjacent proteins being tethered by a flexible linker . This provides a compelling molecular mechanism for the self-association of matrix molecules required for virus assembly . While the general mode of polymerization is conserved between the two structures , the precise molecular details of the interactions differ , consistent with these matrix proteins binding different cellular factors during infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology/virion",
"structure,",
"assembly,",
"and",
"egress",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"biophysics/macromolecular",
"assemblies",
"and",
"machines"
] | 2008 | Rhabdovirus Matrix Protein Structures Reveal a Novel Mode of Self-Association |
The human pathogen Haemophilus influenzae has the ability to quickly adapt to different host environments through phase variation of multiple structures on its lipooligosaccharide ( LPS ) , including phosphorylcholine ( ChoP ) . During colonization with H . influenzae , there is a selection for ChoP+ phase variants . In a murine model of nasopharyngeal colonization , this selection is lost in the absence of adaptive immunity . Based on previous data highlighting the importance of natural antibody in limiting H . influenzae colonization , the effect of ChoP expression on antibody binding and its bactericidal activity was investigated . Flow cytometric analysis revealed that ChoP+ phase variants had decreased binding of antibody to LPS epitopes compared to ChoP− phase variants . This difference in antibody binding correlated with increased survival of ChoP+ phase variants in the presence of antibody-dependent , complement-mediated killing . ChoP+ phase variants were also more resistant to trypsin digestion , suggesting a general effect on the physical properties of the outer membrane . Moreover , ChoP-mediated protection against antibody binding correlated with increased resilience of outer membrane integrity . Collectively , these data suggest that ChoP expression provides a selective advantage during colonization through ChoP-mediated effects on the accessibility of bactericidal antibody to the cell surface .
Haemophilus influenzae is an extracellular , gram-negative pathogen that is a primary causative agent of otitis media in children and is also frequently isolated from adults with pneumonia and exacerbations of chronic obstructive pulmonary disease ( COPD ) [1]–[5] . Colonization of the upper respiratory tract with H . influenzae is common and is the first step in disease development , as H . influenzae carriage is associated with recurrent otitis media episodes in children [6] , [7] . While the Hib conjugate vaccine has greatly reduced the burden of disease caused by type b H . influenzae [8] , [9] , non-typeable H . influenzae ( NTHi ) strains , which are unencapsulated , remain a common source of respiratory tract infections . Vaccine strategies targeting NTHi strains are complicated by the high variability of outer membrane antigens [10] , [11] . One of the structurally diverse molecules on the surface of H . influenzae is the lipopolysaccharide ( LPS ) . The LPS of H . influenzae is truncated compared to the LPS of other gram-negative bacteria . It contains no repetitive O antigen side chains and is also referred to as lipooligosaccharide ( LOS ) [12] , [13] . H . influenzae LPS consists of lipid A attached to 3-deoxy-D-manno-oct-2-ulosonic acid ( KDO ) , with three conserved inner core heptoses to which various oligosaccharide extensions , and other non-carbohydrate molecules , can be attached [14] . Mass spectrometry ( MS ) analysis of different H . influenzae isolates has revealed a significant level of diversity in LPS structures [15]–[21] . For example , the length and composition of the hexose extensions from the inner core heptoses , as well as the attachment of molecules such as sialic acid and glycine , varies both between different strains and within glycoforms of the same isolate . A major source of LPS variability in H . influenzae is on-off switching , or phase variation , involving LPS biosynthesis genes [22] , [23] . One of the phase variable molecules expressed on H . influenzae LPS is phosphorylcholine . Phosphorylcholine [ ( CH3 ) 3N+CH2CH2PO4−] , or ChoP , is a small , zwitterionic molecule that is covalently attached to the LPS through its phosphate group . ChoP is a surface structure of a number of bacteria in addition to H . influenzae , particularly those found in the respiratory tract , including Streptococcus pneumoniae , Pseudomonas aeruginosa , and Neisseria species [24]–[26] . ChoP is also a component of eukaryotic membrane lipids in the form of phosphatidylcholine . H . influenzae must acquire choline from the environment , and turnover of host lipids can be a major source of choline during colonization [27] , [28] . Choline import , phosphorylation , and attachment to H . influenzae LPS is controlled by genes in the lic1 locus . The choline kinase gene lic1A contains a tetranucleotide repeat that is responsible for ChoP phase variation . Slipped-strand mispairing within the repeat region of lic1A creates a translational on-off switch controlling ChoP expression [29] . As a result , the control of ChoP attachment to the LPS is stochastic , and phase variation occurs at a high frequency [30] . Phase variation of ChoP expression may provide a mechanism for Haemophilus to display a variety of phenotypes , allowing rapid adaptation to different host environments . ChoP attachment to the LPS enables recognition by C-reactive protein ( CRP ) , which binds to ChoP and initiates classical pathway complement-mediated killing [31] . In host environments with high levels of CRP , such as in the blood , there is a selective advantage for ChoP− phase variants [32] . In addition , an antibody response can be initiated against LPS epitopes containing ChoP [33] , [34] . However , the maintenance of phase variable ChoP expression predicts that there are also advantages for ChoP+ bacteria in select host environments . During H . influenzae colonization there is a strong selection for ChoP+ phase variants . This selection has been observed in several animal models of Haemophilus colonization , as well as during human carriage [31] , [35]–[37] . ChoP expression increases adherence to epithelial cells through interaction with platelet-activating factor receptor ( rPAF ) , which normally binds the ChoP-containing molecule PAF . While in vitro experiments have demonstrated the capacity of ChoP+ bacteria to bind rPAF , mice deficient in rPAF have no colonization defect [38] , [39] . These data suggest that there are additional host factors involved in the selection for ChoP+ bacteria during colonization . Here , we show that ChoP attachment to the LPS alters the physical properties of the outer membrane and reduces antibody binding to the surface of H . influenzae .
The role of adaptive immunity in the selection of ChoP+ phase variants was examined in vivo using a murine model of nasopharyngeal colonization . The percentage of ChoP+ phase variants was determined by colony immunoblotting following colonization with a mixture of ChoP+/− variants of strain H632 , an NTHi strain that is able to colonize mice [40] . The proportion of ChoP+ phase variants in the output ( colonizing ) population was substantially greater than that in the input ( inoculum ) after challenge of immune competent , ( wild-type ) BALB/c mice ( Figure 1 ) . In contrast , there was no evidence for a selection of ChoP+ phase variants following colonization in BALB/c mice lacking an adaptive immune system ( severe combined immune deficiency , or SCID ) [41] . These results demonstrate that adaptive immunity is important for the selection of ChoP+ phase variants during colonization . The importance of natural , or pre-existing , antibody in limiting H . influenzae colonization has been demonstrated previously [40] . These data led to an examination of the impact of ChoP expression on antibody binding by flow cytometry . It was found that ChoP+ variants had reduced antibody binding to their surface compared to ChoP− variants ( Figure 2 ) . The ChoP+ phase variant of NTHi strain H233 had decreased binding of natural IgG using both serum from ( wild-type ) BALB/c mice ( Figure 2A ) and normal human serum ( NHS; Figure 2C ) . Revertants of the originally selected phase variants of strain H233 , as well as ChoP+/− phase variants in another NTHi strain , H729 , showed a similar effect ( Figure 2A , C ) . The ChoP-mediated reduction of the binding of natural antibody in serum ( naïve control ) was also observed in the serum of mice previously intranasally immunized with H . influenzae ( Figure 2B ) . To confirm the difference in antibody binding was not dependent on other serum components , the effect of ChoP on binding of IgG purified from NHS was also examined , with the same result ( Figure 2D ) . This binding assay was also conducted using a mutant strain of Rd with constitutive ChoP+ expression ( phase locked ChoP+ ) , H491 . This strain was grown in chemically defined medium ( CDM ) with or without choline . The binding of IgG in NHS to H491 was reduced when choline was added to the CDM , compared to CDM without choline ( Figure 2E ) . ChoP expression also reduced binding of IgA from NHS ( Figure 2F ) , from human nasal secretions ( Figure 2G ) , and IgM purified from NHS ( Figure 2H ) . To examine whether there was an additional effect of ChoP expression on binding of complement component C3 , baby rabbit serum ( BRS ) was used as a source of complement without natural antibody to H . influenzae . While there was no difference C3 binding to ChoP+/− phase variants in BRS alone , the ChoP+ phase variant of strain H233 had reduced C3 binding in the presence of BRS with purified IgG from NHS ( Figure 2I ) . Collectively , these data demonstrate that ChoP expression results in decreased antibody binding , which limits complement deposition , on the surface of H . influenzae . The classical pathway of complement-mediated killing can be initiated by binding of CRP or bactericidal antibody to H . influenzae . In order to determine if ChoP expression affects classical pathway complement-mediated killing following antibody binding , NHS depleted of C-reactive protein ( CRP ) was used as an antibody and complement source for bactericidal assays . ChoP+ phase variants had increased survival in CRP-depleted NHS compared to ChoP− variants of strain H233 , as well as revertants of the originally selected phase variants ( Figure 3A ) . There was also increased survival of the ChoP+ phase variant of H233 using IgG purified from NHS ( in the same concentration as that used for flow cytometry in Figure 2D ) with BRS as a complement source ( Figure 3B ) . The increased survival of ChoP+ bacteria in the presence of CRP-depleted NHS was observed for multiple other NTHi strains , constitutive ChoP+ and ChoP− mutants in Rd , and an H . influenzae type b strain ( Eagan ) ( Figure 3C ) . Serum was IgG-depleted to determine whether the difference in survival of ChoP variants was dependent on antibody . In IgG-depleted NHS there was a recovery of survival for the ChoP− variant of H233 , while the addition of purified IgG restored killing ( Figure 3D ) . No differences in survival were detected for bactericidal assays conducted in BRS as a source of complement without antibody , or in NHS with MgEGTA buffer , which allows alternative pathway complement-mediated killing only ( not shown ) . Collectively , these results suggest that ChoP expression increases survival in the presence of bactericidal antibody . The potential targets on the surface of H . influenzae that are protected from antibody binding by ChoP expression were examined using antibody-depleted NHS and mAbs . Purified LPS was used to absorb LPS–specific antibodies from NHS prior to conducting bactericidal assays . LPS antibody pre-absorption with wt , constitutive ChoP− LPS resulted in increased survival of the constitutive ChoP− mutant in Rd , H446 ( Figure 4B ) . In contrast , there was no increase in survival of the constitutive ChoP− mutant after pre-absorption with ChoP− LPS from the Rd opsX mutant strain , which is highly truncated with no KDO heptose extensions ( Figure 4A , B ) . To confirm that the LPS pre-absorbed serum retained complement activity , purified IgG was added back to the LPS pre-absorbed serum , and killing was observed ( not shown ) . These results indicate that the majority of the bactericidal antibody affected by ChoP expression in NHS is LPS oligosaccharide-specific . Of note , that there was a greater increase in the survival of the Rd constitutive ChoP− mutant H446 following NHS pre-absorption with ChoP− LPS , compared to pre-absorption with ChoP+ LPS , a result that correlated with the effect of ChoP on antibody binding ( Figure 4B ) . While purified LPS may not accurately mimic the environment of the outer membrane , this result demonstrates that the same decrease in antibody binding observed in ChoP+ whole bacteria is also observed for ChoP+ LPS . The mAb 6E4 , which binds H . influenzae LPS [42] , was used to examine the effect of ChoP on antibody binding to a specific LPS epitope . ChoP expression reduced 6E4 binding and increased survival following incubation in 6E4 with BRS as a complement source ( Figure 4C , D ) . This was observed for NTHi strain H729 and for the constitutive ChoP+ mutant in Rd , H491 . These results confirmed that ChoP expression provides protection against bactericidal antibody binding to LPS oligosaccharide epitopes . A selection of LPS mutants in Rd was used to investigate the importance of the LPS molecular environment for ChoP-mediated protection against antibody binding ( Figure 4A ) . The ChoP+ phase variant of the lpsA mutant strain , which no longer has HepIII hexose extensions , maintained reduced antibody binding and increased survival in the presence of CRP-depleted NHS ( Figure 5 ) . In contrast , there was no longer a protective effect of ChoP expression on antibody binding and complement-mediated killing for the orfH and lpt6 mutant strains . The orfH mutation results in a lack of HepIII , while the lpt6 mutation prevents attachment of a conserved phosphoethanolamine molecule to HepII . ChoP can be attached to hexose extensions from multiple heptose residues , most commonly HepI and HepIII [20] , [43] , [44] . The importance of the position of ChoP for its effect on antibody binding was examined using lic1D exchange mutants , as the lic1D allele dictates the position of ChoP attachment [45] . In Rd , ChoP is attached to a hexose extension on HepI ( HI ) . Alteration of the position of ChoP in Rd from H1 to a hexose extension on HepIII ( H3 ) resulted in the loss of ChoP-mediated protection against antibody binding and antibody-dependent bactericidal activity ( Figure 6 ) . Collectively , these data demonstrate that the constituents of the oligosaccharide and the molecular environment of ChoP are both important for the effect of ChoP on antibody binding and resistance to antibody-dependent , complement-mediated killing . We next investigated the mechanism for ChoP-mediated protection against antibody binding , considering two possibilities; 1 ) steric hindrance , where ChoP obscures key epitope ( s ) and 2 ) ChoP alteration of the physical properties of the outer membrane , resulting in decreased membrane accessibility . In order to test the effect of ChoP expression on the ability of molecules other than antibodies to access the membrane , trypsin sensitivity was compared for ChoP+ and ChoP− phase variants . The fluorescent dye Cy5 , which labels lysine residues , was used to quantify the exposure of outer membrane surface proteins by flow cytometry . Following trypsin digestion , there was decreased Cy5 binding to the ChoP− phase variant of strain H233 , as well as the ChoP− revertant of the original ChoP+ phase variant ( Figure 7A ) . The same effect was observed for the constitutive ChoP− mutant in Rd , H446 ( Figure 7A ) . The reduction in Cy5 binding following trypsin digestion suggests that ChoP expression affects the general accessibility of molecules , including antibodies , to outer membrane targets . In order to further examine the effect of ChoP on the physical properties of the outer membrane , membrane barrier function was compared in ChoP+ and ChoP− phase variants . Bacterial uptake of the dye ethidium bromide ( EtBr ) was used to measure the effect of ChoP on the permeability of the outer membrane . The rate-limiting step for EtBr uptake is transversal of the outer membrane [46] . In the presence of a low concentration of polymyxin B , there was an increased rate of EtBr uptake in the ChoP− , compared to the ChoP+ , phase variants of H233 ( Figure 7B ) . The same effect was observed for the revertants of the original variants of strain H233 as well as the constitutive ChoP+ and ChoP− mutants in Rd ( Figure 7B ) . It was necessary to include polymyxin B to cause initial membrane destabilization for dye uptake . While ChoP expression does not affect killing by polymyxin B alone ( not shown ) , the difference in EtBr uptake rates may reflect a difference in outer membrane susceptibility to polymyxin B . The alteration of polymyxin B-induced EtBr uptake demonstrates that ChoP expression strengthens the barrier function of the outer membrane . Differences in membrane barrier function often correlate with changes in the gel-to-liquid crystalline phase transition temperature , or Tm , which can be determined by differential scanning calorimetry ( DSC ) [47] . In order to test the effect of ChoP expression on the Tm of H . influenzae LPS , DSC was performed on LPS purified from ChoP+ and ChoP− bacteria . The Tm of LPS from the Rd constitutive ChoP+ strain H491 was determined to be 29 . 8± . 2°C , while the Tm of the constitutive ChoP− mutant strain H446 was significantly higher , at 34 . 3± . 1°C ( p< . 0001 ) . Phase transition temperatures were independent of Mg2+ concentration . The effect of ChoP on the integrity of the outer membrane was examined by comparing EDTA sensitivity in ChoP+ and ChoP− phase variants . EDTA treatment chelates the divalent cations that are important for maintaining LPS interactions and membrane stability [48] . The ChoP+ variant of strain H233 had increased resistance to EDTA , compared to the ChoP− variant ( Figure 8A ) . ChoP revertants of these variants showed the same trend ( not shown ) . Growth of the Rd constitutive ChoP+ strain H491 in CDM with choline also resulted in increased EDTA resistance , compared to survival following growth in CDM without choline ( Figure 8B ) . In contrast , the expression of a digalactoside residue ( Galα1-4Gal ) in strain H233 , detected by the mAb 4C4 , had no impact on EDTA resistance ( Figure 8A ) . The importance of the position of ChoP for its effect on outer membrane integrity was also investigated . For the Rd lic1D exchange mutant strains , only ChoP in the H1 , but not H3 , position resulted in increased EDTA resistance ( Figure 8C , D ) . These results correlate with the effect of ChoP position on antibody binding , suggesting that the same structural requirements for ChoP-mediated reduction of antibody binding are necessary for its effect on the integrity of the outer membrane . Next , the effect of divalent cation concentration on antibody binding was explored . Increasing the Mg2+ concentration ( up to 50 mM ) resulted in reduced antibody binding to the ChoP− phase variant of H233 ( Figure 8E ) . These results demonstrate that excess Mg2+ , which increases the stability of the outer membrane [49]–[51] , can correct for the difference in antibody binding between ChoP+ and ChoP− variants . Together , these data indicate that ChoP expression alters the physical properties of the outer membrane , and that these effects correlate with the reduction of antibody binding in ChoP+ phase variants . ChoP is a zwitterionic molecule with a positively charged quaternary amine group , which may be important for its effect on antibody binding and the physical properties of the outer membrane . The structural components of ChoP that are required for its effect on antibody binding and the integrity of the outer membrane were examined using a strain that incorporated a ChoP analog , dimethylethanolamine phosphate [ ( CH3 ) 2NCH2CH2PO4−] . This ChoP analog differs from ChoP by a single methyl group , which reduces the positive charge on the amine group with minimal alteration of the overall structure . Incorporation of the ChoP analog into the LPS of the Rd constitutive ChoP+ strain H491 grown in CDM+dimethylethanolamine was confirmed by MS ( not shown ) . H491 grown in CDM+dimethylethanolamine had a similar level of antibody binding as when the strain was grown in CDM alone ( Figure 2E ) . In addition , while H491 grown in CDM+choline had increased EDTA resistance compared to when it was grown in CDM alone , H491 grown in CDM+dimethylethanolamine showed the same percent survival in EDTA as when it was grown in CDM alone ( Figure 8B ) . While it cannot be ruled out that the single methyl group impacts the steric inhibition of antibody binding , these results suggest that the positively charged quaternary amine on ChoP is important for the reduced antibody binding and increased outer membrane integrity observed in ChoP+ phase variants .
Understanding the requirements for H . influenzae colonization is integral to the effort to reduce the burden of NTHi-associated disease . H . influenzae is susceptible to antibody-dependent , classical pathway complement-mediated killing in vitro , and this may be an important mechanism for host control of H . influenzae in vivo . For example , human patients with primary antibody deficiencies have persistent colonization and higher rates of disease from NTHi strains [52] . In this light , bacterial factors that affect antibody recognition could play a major role in H . influenzae survival during colonization . In this study , it was found that ChoP+ phase variants have reduced binding of antibody , including antibody binding to LPS epitopes , as well as increased survival in the presence of antibody-dependent , complement-mediated killing . While there was no difference in C3 deposition or bacterial survival for ChoP+/− phase variants in the absence of antibody , the experiments conducted in the present study do no exclude the possibility that ChoP expression affects binding of other classical pathway complement components . The major bactericidal antibody from the serum sources in this study was IgG , which can reach the site of colonization through transcytosis [53] , [54] . Indeed , nasal lavage fluid from BALB/c mice contains IgG that binds H . influenzae targets , including LPS , and a role for complement in limiting H . influenzae colonization has also been demonstrated [40] . Increased resistance to bactericidal antibody was observed for ChoP+ phase variants of multiple H . influenzae strains , despite the heterogeneity observed in this bacteria [55] . These data demonstrate a novel mechanism for evasion of antibody recognition by H . influenzae during colonization . It was also shown that adaptive immunity is required for the increase in the selection of ChoP+ phase variants during colonization . While the short-term colonization model used for the in vivo experiments demonstrates a role for ChoP expression in protection against natural antibodies , it was also shown that ChoP+ phase variants have reduced binding of antibody from pre-exposed , immune hosts . Taken together , these results suggest that ChoP expression provides a selective advantage at the mucosal surface during colonization , as ChoP+ bacteria are better protected against antibody binding and antibody-dependent clearance . ChoP is one of several LPS structural determinants whose attachment to the LPS is controlled by stochastic phase variation [29] , [56] . Other phase variable decorations to the LPS have been shown to have an effect on serum resistance . For example , loss of O-acetylation , sialylation , or the digalactoside residue Galα1 , 4Gal results in increased serum sensitivity , attributed to different mechanisms [57]–[59] . The LPS of H . influenzae is highly heterogeneous [60] , and phase variation of LPS epitopes may allow bacteria to quickly adapt to the repertoire of antibodies present in different host environments . While it was shown in the current study that ChoP+ phase variants have reduced binding of the mAb 6E4 , the full scope of the LPS , or non-LPS , epitopes protected from antibody recognition by ChoP expression remains unknown . As mentioned previously , ChoP attachment to surface structures has been observed in several bacterial species . In addition to the LPS , ChoP has been found in bacteria on teichoic acid , pili , and an elongation factor protein [61]–[63] . It has also recently been shown that an effector protein injected by Legionella pneumophila modifies host regulatory factors with ChoP [64] . The current study supplies another example of how the attachment of this ubiquitous molecule modulates the properties of its target . The importance of the molecular environment of the LPS for ChoP-mediated protection against antibody binding was examined using a set of LPS mutants . It was found that two conserved inner core LPS structures ( PEtn and HepIII ) are required for the reduction of antibody binding in ChoP+ phase variants . This result could be due to the direct loss of epitopes that are normally protected against antibody binding by ChoP expression , or through an indirect effect of these structures on binding of antibody to other LPS epitopes . Previously , it was shown that changing the location of ChoP attachment affects sensitivity to CRP [45] . In accordance with this data , we determined that the position of ChoP attachment to the LPS is important for the effect of ChoP on antibody binding in Rd . While the inner core structure of LPS is conserved among H . influenzae strains , ChoP can be attached to hexose extensions off of any of the three inner core heptoses or to a fourth heptose present in some NTHi strains [17] , [18] , [20] . There are also NTHi strains with a partial duplication of the lic locus , resulting in the attachment of two ChoP residues to the LPS in ChoP+ phase variants [65] . Each H . influenzae strain may have optimized its LPS structural arrangement to enable ChoP-mediated protection against antibody binding . The variable position of ChoP also argues against its main function being sterically hindering antibody binding to other LPS epitopes . The finding that ChoP+ phase variants of H . influenzae are also less sensitive to trypsin digestion of outer membrane proteins led to the investigation of the effect of ChoP on the physical properties of the outer membrane . The reduced access of a non-antibody molecule to the membrane suggests that the effect of ChoP on antibody binding is due to a general effect on the outer membrane , rather than direct steric hindrance . This concept is supported by a recent study from this lab demonstrating that mutations in H . influenzae that change the distribution of phospholipids in the outer membrane result in decreased outer membrane stability and increased antibody binding to LPS epitopes [66] . In support of the hypothesis that ChoP expression affects the physical properties of the outer membrane , it was shown that ChoP+ phase variants have reduced sensitization to treatments that may compromise the outer membrane , such as polymyxin B-induced EtBr permeability . Polymyxin B is a cationic peptide that targets negatively charged residues in the LPS [67] . A previous study demonstrated that ChoP expression reduces sensitivity to killing by the cationic antimicrobial peptide LL-37 [68] , further supporting a ChoP-mediated impact on the integrity of the outer membrane . While studies in several bacterial species have shown that lipid A modifications can affect antimicrobial resistance and membrane permeability [46] , [49] , [69] , [70] , the data presented in the current paper demonstrate that modifications outside of the lipid A-KDO inner core region can also impact membrane integrity and barrier function . DSC has been used to determine the phase transition temperatures of phospholipid membrane systems as well as purified LPS from various bacterial strains [71] , [72] . In the current study , the phase transition temperature for purified LPS from a constitutive ChoP+ strain was found to be reduced compared to that for ChoP− LPS . A similar trend is observed in phospholipid membrane systems , where the addition of lipids containing ChoP reduce membrane permeability and the Tm [47] . In DSC experiments using LPS isolated from Salmonella minnesota , it was found that mutants with reduced oligosaccharide extensions had a lower Tm , demonstrating that changes to the oligosaccharide , in addition to lipid A alterations , can affect the phase transition temperature [51] , [73] . It was shown in the current study that ChoP+ phase variants of H . influenzae have increased resistance to membrane disruption by EDTA . EDTA chelates the divalent cations that are important for the stabilization of the outer membrane through association with multiple negatively charged phosphate groups on the LPS [50] , [74] . It has been shown that EDTA treatment causes loss of outer membrane organization and shedding of LPS molecules , resulting in reduced membrane integrity [74] , [75] . The modification of outer membrane integrity correlated with an effect on antibody binding , as the addition of Mg2+ alone resulted in decreased antibody binding . While ChoP is a relatively small structural addition to H . influenzae LPS , the presence of the positively charged quaternary amine group may impact the outer membrane by altering charge interactions . It was shown in this study that bacteria that incorporated dimethylethanolamine phosphate instead of ChoP did not have reduced antibody binding or increased EDTA resistance . These data indicate that the positively charged amine group , rather than the negatively charged phosphate group , is required for the effect of ChoP on outer membrane integrity and antibody binding . The expression of molecules with amine groups on lipid A , as well as KDO , has been shown to increase outer membrane stability in other bacteria [76] , [77] . Our study suggests a similar effect through modification of the outer core of the oligosaccharide with ChoP . This also supports the notion that a large part of the effect of ChoP on antibody binding is indirect , through alteration of outer membrane accessibility , rather than direct steric inhibition . In summary , these data indicate that ChoP expression increases the barrier function and integrity of the outer membrane , and these alterations correlate with a reduction in the accessibility and binding of antibody to ChoP+ phase variants . Phase variation of ChoP may be an important consideration in the design of NTHi vaccines targeting LPS epitopes , as selection for ChoP+ phase variants could abrogate vaccine effectiveness . In contrast , vaccines targeting ChoP would select for ChoP− phase variants , and thereby increase effective immune responses to other cell surface epitopes . While phase variation of LPS structures can directly alter the presence of specific epitope ( s ) , these data suggest a novel mechanism whereby ChoP expression affects the ability of antibody to recognize bacterial targets by altering access to outer membrane antigens .
This study was conducted according to the guidelines outlined by National Science Foundation Animal Welfare Requirements and the Public Health Service Policy on the Humane Care and Use of Laboratory Animals . The protocol was approved by the Institutional Animal Care and Use Committee , University of Pennsylvania Animal Welfare Assurance Number A3079-01 . All strains are listed in Table 1 . Strains were grown in brain heart infusion media ( Becton Dickinson Biosciences , Franklin Lake , NJ ) supplemented with Fildes enrichment ( Remel , Lenexa , KS ) and 20 µg/mL β-Nicotinamide adenine dinucleotide hydrate ( Sigma , St . Louis , MO ) , referred to as sBHI . When specified , strains were grown in CDM , prepared as previously described [78] . CDM was supplemented with 300 µM of choline chloride or the choline analog N , N-Dimethylethanolamine ( Sigma ) where indicated . Selection of ChoP and Galα1-4Gal containing phase variants was performed by colony immunoblotting as previously described [79] . Revertants of ChoP variants were selected in the same manner . Each ChoP phase variant population was determined to be over 98% ChoP+ or ChoP− by colony immunoblotting , and constitutive ChoP mutants were 100% ChoP+ ( H491 ) or ChoP− ( H446 ) . The percentage of ChoP+ and ChoP− bacteria in each phase variant population remained constant during growth to log phase , as there is no selective pressure on ChoP expression in vitro . ChoP+ colonies transferred to a nitrocellulose membrane were detected using a 1∶10 , 000 dilution of the monoclonal antibody TEPC-15 ( Sigma ) followed by a 1∶10 , 000 dilution of alkaline phosphatase-conjugated anti-mouse IgA ( Sigma ) . Colonies expressing the Galα1-4Gal structure were selected using a 1∶10 , 000 dilution of the monoclonal antibody 4C4 [80] followed by a 1∶10 , 000 dilution of alkaline phosphatase-conjugated anti-mouse IgG ( Sigma ) . Colonization studies were conducted as described previously [40] . Briefly , mice were intranasally inoculated with 107 CFU/mL of bacteria that were first washed and diluted in phosphate-buffered saline ( PBS ) . ChoP variants were grown separately , followed by combination at a 3∶1 ratio of ChoP− ∶ ChoP+ bacteria by volume prior to inoculation . Nasal lavage fluid was collected in 200 µl of PBS and plated onto sBHI containing 50 µg/mL streptomycin following three days of colonization . Bacterial counts obtained by nasal lavage were comparable to those collected by plating nasopharyngeal tissue homogenates . The percentage of ChoP+ colonies was determined through detection of ChoP by colony immunoblotting . Antibody binding was detected by flow cytometry as previously described [66] . Briefly , 200 µl reactions containing mid-logarithmic phase bacterial cells in Hank's buffer without Ca2+ or Mg2+ ( Gibco , San Diego , CA ) supplemented with 5% fetal calf serum ( HyClone , Logan , UT ) were incubated with primary antibody for 60 min at 37°C . Primary antibody sources included naïve BALB/c serum ( 1∶200 dilution ) , NHS ( 1∶200 dilution ) , IgG purified from NHS ( Sigma , 4 . 8 µg ) , IgM purified from NHS ( Sigma , 3 . 7 µg ) , normal human nasal secretions ( 1∶200 dilution ) , mAb 6E4 ( 1∶100 dilution for H729 , 1∶500 dilution for Rd ) , and BRS ( 1∶50 dilution ) . Serum collected from BALB/c mice that had been intranasally inoculated at day 0 , 7 , and 14 with either PBS ( naïve ) or 107 CFU/mL of constitutive ChoP− type b strain H445 ( immune ) was also used as a source of primary antibody ( 1∶20 dilution ) . Reactions mixtures were then washed and re-suspended in 1∶200 dilutions of secondary antibody , followed by incubation at 4°C for 60 min . Secondary antibodies included goat anti-mouse IgG-FITC , goat anti-human IgG-FITC , goat anti-human IgA-FITC , goat anti-human IgM-FITC ( Sigma ) , and goat anti-rabbit polyclonal C3-FITC ( MP Biomedical Chappel , Irvine , CA ) . Reaction mixtures were then washed and re-suspended in PBS with 1% bovine serum albumin and 0 . 5% paraformaldehyde prior to flow cytometric analysis on a BD FACS Calibur flow cytometer ( Becton Dickinson Biosciences ) . A total of 50 , 000 cells were collected from each reaction mixture , and the MFI of antibody binding was determined using FlowJo software ( Tree Star , Ashland , OR ) . Assays were conducted in 200 µl reaction mixtures containing 20 µl of mid-logarithmic phase bacterial cells ( OD620 0 . 5 ) diluted to 105 CFU/mL in Hank's buffer with Ca2+ and Mg2+ ( Gibco ) . Following the addition of serum , reaction mixtures were incubated for 45 min at 37°C . One serum source was CRP-depleted NHS , used at a 1∶10 dilution for H233 and Eagan , a 1∶5–1∶10 dilution for other NTHi strains , and a 1∶50 dilution for Rd . CRP was depleted from NHS using immobilized p-aminophenyl phosphoryl choline gel , which also results in depletion of anti-ChoP antibodies , according to the manufacturer's protocol ( Thermo Scientific , Rockford , IL ) . Another serum source was a 1∶20 dilution of BRS , to which purified IgG from NHS was added ( Sigma , 4 . 8 µg ) . IgG depletion of NHS was performed using a Protein G column ( GE Healthcare Bio-Sciences , Uppsala , Sweden ) . IgG eluted from the Protein G column according to manufacturer's instructions was also used as a source of purified IgG ( 0 . 25 µg/mL ) for bactericidal assays . Survival due to alternative pathway mediated killing alone was determined by comparing survival in BRS alone ( no H . influenzae antibodies ) and by chelating NHS with gelatin veronal buffer containing MgEGTA ( Boston Bioproducts , Worcester , MA ) [81] . Serum pre-absorption of anti-LPS antibodies was conducted by incubation of NHS ( 1∶50 dilution ) with 1 µg LPS overnight at 4°C . Baby rabbit serum was used as a source of complement without antibody at a 1∶20 dilution for H233 , a 1∶10 dilution for H729 , and a 1∶25 dilution for Rd . Bactericidal assays with baby rabbit serum were supplemented with IgG purified from NHS ( 4 . 8 µg ) for H233 , or mAb 6E4 at a 1∶10 dilution for H729 and a 1∶100 dilution for Rd . Percent survival was determined relative to complement-inactivated serum , which was incubated for 30 min at 56°C prior to use . Assays to determine EDTA sensitivity were performed by addition of EDTA ( 1–4 mM ) to bacterial cells in sBHI diluted in Hank's buffer with Ca2+ and Mg2+ , followed by incubation for four hours at 37°C . Percent survival was determined relative to no-EDTA controls . LPS extractions were performed using the phenol-chloroform-petroleum ether method as previously described [82] , with modifications [83] . Briefly , bacterial pellets were washed sequentially in ethanol , acetone , and petroleum ether prior to lyophilization . The lyophilized samples were re-suspended and mixed overnight in a 2∶5∶8 extraction mixture of phenol∶ chloroform∶ petroleum ether . Following filtration and evaporation of chloroform and petroleum ether , LPS was precipitated from phenol in 5∶1 mixture of acetone∶ diethyl ether . Ultracentrifugation was used to further purify LPS re-suspended in water , followed by lyophilization . Partial digestion of outer membrane proteins with trypsin was performed as previously described [84] . Briefly , 200 µl of mid-logarithmic phase bacterial cells ( OD620 0 . 5 ) were washed and re-suspended in 10 mM Tris-HCL , pH 7 . 5 . Following addition of 1 mg/mL trypsin , bacteria were incubated for 2 hrs at 37°C . The cells were then washed and re-suspended in 10 mM carbonate buffer prior to staining with Cy5 according to manufacturer's instructions ( GE Biosciences Amersham , Buckinghamshire , UK ) . Surface proteins with exposed lysines were labeled with 10 µl of Cy5 ( 40 pmol ) in the dark for 20 min , and reactions were stopped with 20 µl of 10 mM lysine . Cells were washed with 10 mM carbonate buffer and re-suspended in PBS with 1% bovine serum albumin prior to flow cytometric analysis on a BD FACS Calibur flow cytometer ( BD Biosciences ) . A total of 50 , 000 cells were collected from each reaction mixture , and the MFI of antibody binding was determined using FlowJo software ( Tree Star ) . Ethidium bromide was used as a measure of outer membrane permeability as previously described [46] . Bacterial cells grown to stationary phase ( OD620 0 . 8 ) were re-suspended in PBS , with 15 µg/mL polymyxin B and 6 µM EtBr added directly prior to each measurement . Fluorescence was measured at an excitation wavelength of 544 nm , an emission wavelength of 610 nm using FLUOstar OPTIMA ( BMG Labtech , Ortenberg , Germany ) . EtBr uptake was expressed by RFU/s . Purified LPS samples for each strain were diluted to 2 mg/mL in PBS , sonicated , and subjected to three cycles of incubation at 56°C for five min , vortexing 1 min , and cooling to 4°C . Where specified , MgCl2 was added to the PBS at 1∶1 and 5∶1 [MgCl2]∶ [LPS] molar ratios . Following preparation , samples were stored at 4°C for several hours before running on the DSC instrument . Heat capacity profiles were determined at a scan rate of 60°C/hr over a temperature range of 10–60°C in a high-resolution differential scanning calorimeter ( MCS , MicroCal , Amherst , MA ) . Three consecutive heating and cooling scans were measured per sample . PBS buffer-buffer references were subtracted from sample data , and concentration was normalized based on sample concentration . The Microcal ORIGIN software package was used to progress baselines . The gel-to-liquid crystalline phase transition temperature , Tm , was determined by integration from baseline to calculate the midpoint of the transition . Purified LPS samples were subjected to mild acid hydrolysis and electrospray ionization-mass spectrometry ( ESI-MS ) was performed as previously described [85] . Differences between groups were assessed for statistical significance using an unpaired Student's t-test ( GraphPad PRISM4 , GraphPad Software , La Jolla , CA ) . The reference numbers for genes mentioned in the text include: opsX ( 0261 ) , lpsA ( 0765 ) , orfH ( 0523 ) and lpt6 ( 0258 , 0259 ) from the Rd database for the published genome sequence [86] , as well as lic1A ( 950399 ) and lic1D ( 950403 ) GeneID from the NCBI GenBank database . | The bacterial pathogen Haemophilus influenzae evades immune responses during colonization in its human host . Decoration of the bacterial surface with different structures is one way that Haemophilus avoids host recognition . In this study , we show that the attachment of the small molecule phosphorylcholine , or ChoP , to the lipopolysaccharide covering the bacterial surface allows H . influenzae to avoid the immune response by inhibiting antibody binding . The presence of ChoP alters the bacterial surface to reduce its accessibility . The ability of ChoP to affect antibody binding is dependent on the positive charge of the molecule , which changes the physical properties of the bacterial membrane . The increased survival of bacteria with ChoP attached to their surface enriches ChoP+ bacteria during colonization . This study reveals a novel mechanism for bacterial evasion of a host immune response . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunity",
"immunology",
"biology",
"microbiology",
"bacterial",
"pathogens"
] | 2012 | Phosphorylcholine Allows for Evasion of Bactericidal Antibody by Haemophilus influenzae |
One of the major challenges that developing organs face is scaling , that is , the adjustment of physical proportions during the massive increase in size . Although organ scaling is fundamental for development and function , little is known about the mechanisms that regulate it . Bone superstructures are projections that typically serve for tendon and ligament insertion or articulation and , therefore , their position along the bone is crucial for musculoskeletal functionality . As bones are rigid structures that elongate only from their ends , it is unclear how superstructure positions are regulated during growth to end up in the right locations . Here , we document the process of longitudinal scaling in developing mouse long bones and uncover the mechanism that regulates it . To that end , we performed a computational analysis of hundreds of three-dimensional micro-CT images , using a newly developed method for recovering the morphogenetic sequence of developing bones . Strikingly , analysis revealed that the relative position of all superstructures along the bone is highly preserved during more than a 5-fold increase in length , indicating isometric scaling . It has been suggested that during development , bone superstructures are continuously reconstructed and relocated along the shaft , a process known as drift . Surprisingly , our results showed that most superstructures did not drift at all . Instead , we identified a novel mechanism for bone scaling , whereby each bone exhibits a specific and unique balance between proximal and distal growth rates , which accurately maintains the relative position of its superstructures . Moreover , we show mathematically that this mechanism minimizes the cumulative drift of all superstructures , thereby optimizing the scaling process . Our study reveals a general mechanism for the scaling of developing bones . More broadly , these findings suggest an evolutionary mechanism that facilitates variability in bone morphology by controlling the activity of individual epiphyseal plates .
The three-dimensional ( 3D ) morphology of bones is fundamental to the ability of an organism to move , feed , and protect itself . Yet , we know little about the mechanisms that regulate bone shaping during development . Most bones develop by endochondral ossification , a process whereby an anlage of cartilage roughly in the shape of the future bone is formed and then gradually replaced by mineralized tissue [1–5] . The replacement of cartilage by ossified tissue is regulated by the growth plates , which are composed of chondrocytes that maintain a tightly controlled dynamic balance between proliferation and differentiation . The process begins by the formation of a primary ossification center approximately in the middle of the cartilaginous anlagen , which is composed of two growth plates facing in opposite directions . During development , the two growth plates continuously replace cartilage with ossified tissue while moving away from each other . This causes a gradual increase in the length of the bone [6] until maturity is reached and elongation ceases . In recent years , a vast amount of effort has been invested in deciphering the molecular mechanisms that control the growth plate . These efforts have resulted in the identification of molecular pathways that regulate chondrocyte proliferation and differentiation [2 , 5 , 7] . Interestingly , although these molecular mechanisms seem to be generic for all growth plates , variations in elongation rates between growth plates have been reported [8–14] . To date , the mechanism that controls the specific growth rate of each growth plate is still missing . Although bidirectional elongation is a universal mechanism for bone growth , it nevertheless introduces a major challenge to bone morphogenesis . A fundamental characteristic of the unique morphology of each long bone is a set of protrusions of varying shapes and sizes , which are scattered along the exterior of the bone and thus break its morphological symmetry . These superstructures , known as bone ridges , tuberosities , condyles , etc . , are necessary for the attachment of tendons and ligament as well as for articulation . To perform these functions they are located at specific positions along the bone [15–17] . Bone superstructures emerge during early skeletogenesis [18 , 19] . During growth , bones elongate extensively by advancement of the two growth plates away from the superstructures . It is therefore expected that during elongation , superstructures would remain at their original position near the center of the bone . Nevertheless , the end result is proper spreading of superstructures along the mature bone , which clearly implies the existence of a morphogenetic mechanism that corrects their locations . There are two possible morphogenetic routes that can lead to the proper positioning of superstructures . The first is that superstructures emerge at their final relative position along the bone and then maintain this position throughout the entire growth period . This strategy is known as isometric scaling . Another possibility , known as allometric scaling , is that superstructures gradually converge from their initial to their final relative positions during growth . The question of isometric versus allometric scaling is particularly interesting in bones . An ossified bone is a rigid object and so are the superstructures protruding from it , implying that they cannot be relocated by means of cell migration or proliferation . Therefore , any scaling mechanism must be adapted to overcome these physical restrictions . Although this aspect of bone morphogenesis has been largely neglected , Bateman [20] has suggested a mechanism for modifying the position of superstructures along the bone . This mechanism is based on gradual drift along the shaft achieved either by mineral deposition at either the proximal or the distal surface of the superstructure , or by synchronous deposition and absorption of mineral at opposite surfaces , collectively known as bone modeling . This results in drifting of the superstructure towards the deposition side or away from the absorption side . Although this hypothesis offers a reasonable strategy for the scaling of developing bones , its contribution has not been examined yet . In this study , we show that long bones are scaled isometrically from early embryonic stages to adulthood . Moreover , we identify a new morphogenetic mechanism that , combined with bone modeling , maintains the relative position of superstructures by controlling the balance between epiphyseal growth rates . Finally , we show that the growth balance of each bone is optimized for minimizing the drifting activity of its elements , thus facilitating isometric scaling in long bones .
During skeletogenesis , bone superstructures emerge at an early stage [18 , 19] , after which the long bones increase considerably in length . To gain understanding on how the relative position of superstructures is regulated during elongation , it is first necessary to document the distribution of superstructures along the bone throughout the growth period . For that , we established a database of 3D micro-CT images of the six long bones of the limbs , namely humerus , radius , ulna , femur , tibia and fibula , between embryonic day ( E ) 16 . 5 and postnatal day ( P ) 40 . In each bone , anatomical points that mark the longitudinal position of superstructures , collectively referred to in the following as symmetry-breaking elements , were identified . These included the tip; the proximal and the distal margins of tuberosities , trochanters , crests , condyles , and processes; and the fusion zone of the tibiofibular complex ( Fig 1 ) . Then , the relative position of each element between the proximal and the distal ends of the bone was measured ( see Materials and Methods and S1 Fig ) . Because the focus of this work was on longitudinal scaling , in all the analyses described in the following element positions were calculated as a function of the total bone length . As evident in Fig 2 , the relative position of all elements in all bones was preserved throughout the entire growth , as the average range of deviation was 4 . 4% , with the exception of the tibiofibular fusion point ( 14 . 3% ) . These results show that longitudinal scaling in growing long bones is highly isometric . Drift by bone modeling was previously suggested to preserve the relative position of superstructures in mineralized bones [20] . As , to our knowledge , this hypothesis has not been tested thoroughly during development , we proceeded to analyze the drifting patterns of different bone elements . To extract drifting patterns from a temporal sequence of images , it is first necessary to standardize the spatial positions of all imaged bones of the same type by applying an image registration algorithm [21–23] . This would allow defining a standard longitudinal axis along which drifting can be quantified . Since bones are rigid structures that cannot undergo deformation and change only by means of mineral deposition and absorption , an anatomically correct matching between two bone images should be estimated by rigid registration , namely by pure translation and rotation of the bone . Registration based on matching salient and anatomically similar geometrical features such as points , surfaces , lines , or volume segments is a common approach for the registration of biological images [22] . As the bone images depicting the morphogenetic sequence are acquired at different times during development , a valid salient feature must be temporally preserved . Moreover , for population-based analysis , it is also required that the morphology of the salient feature be similar in different animals . To identify such features , we first scrutinized fluorochrome-labeled developing bones for conspicuous mineralized structures that are temporally preserved in individual mice ( Fig 3A and 3B ) . Mice were labeled twice during development , first with calcein ( green ) , and then one or two days later with alizarin complexone ( red ) . Hence , regions labeled only with calcein would indicate mineralized structures that had been deposited before the alizarin injection and maintained until the day of examination . To cover the entire period of embryonic and early postnatal development effectively , we applied three marginally overlapping injection regimes . Analysis showed that during the entire developmental period ( E16 . 5-P6 . 5 ) in both humerus and femur , a thin and continuous layer of dense mineral exhibited high temporal preservation . This layer , referred herein as the cortical core , stretched along the inner side of the bone cortex around the mid-length of the shaft . Next , we assessed the morphological preservation of the cortical core between different mice of the same age . For that , we manually registered all bone images to each other by matching the cortical cores and generated statistical maps that indicate the fraction of bones that are mineralized at each anatomical point ( see Materials and Methods ) . High fraction values ( ≥0 . 7 ) would indicate that the morphology of the mineralized structure is preserved in bones of mice of that age group . As seen in Fig 3C ( top row ) , the cortical core is highly preserved across the population in all age groups . To further validate our results , we jointly and directly examined the temporal and the morphological preservation of the cortical core . To that end , we generated statistical maps that included all bones from either pairs or triplets of consecutive days ( Fig 3C , middle and bottom rows , respectively ) . In all the maps , extensive regions belonging to the bone collar exhibited high levels of preservation . Taken together , these results demonstrate the morphological preservation of the cortical core , both over time within individual bones and between bones of different mice of the same age group ( Fig 3D ) . This preservation establishes the cortical core as a reliable salient structure for cross-stage registration of both prenatal and early postnatal bones of different mice that differ by up to 3–4 developmental days . To recover the morphogenetic sequence of developing long bones , we derived an automated algorithm for rigid registration of multiple bone images ( Fig 4 ) . To compute an accurate transformation between the extracted cortical cores of two bone images , we start by coarsely estimating the registration by extracting a cylindrical shape descriptor from each image and matching the descriptors by custom-designed affinity function , based on the L2-norm . The registration is refined by applying normalized cross correlation ( NCC ) as a similarity measure , and downhill descent as an optimization scheme [24–27] . In order to align multiple images from different developmental stages , we apply a graph theoretic approach , in which we first compute the pairwise transformation between pairs of bones that differ in length by 20% at most , and retain the resulting NCC score of each pair in a distance matrix DN×N . We then compute the maximum spanning tree ( MST ) of the graph encoded by D with a preselected image serving as the root , and infer the final transformation of each image based on the path from the corresponding vertex in the MST to the root . To evaluate the results of the algorithm , we conducted both visual and quantitative analyses ( Fig 5 and S1 File , respectively ) . The results demonstrate the high accuracy and reliability of the registration algorithm as a method for recovering the morphogenetic sequence of long bone development in general and , in particular , for the extraction of drifting patterns of bone superstructures during development . The application of the algorithm on our database of ex vivo images was complemented by manual registration of all in vivo postnatal bone images , performed by matching irregular and conspicuous mineralization patterns in the cortical and trabecular regions . To study the drifting patterns of various anatomical elements , we defined a variable termed “physical position” as the distance ( in mm ) of an element along the longitudinal axis of the bone from a predefined point of origin ( S2 Fig ) . Changes in the physical position of an ossified element directly reflect its drift along the bone . As seen in Fig 6 , the results showed both elements that drifted proximally , such as the distal margin of the humeral deltoid tuberosity ( green marker ) , and distally , as the distal margin of the femoral third trochanter ( yellow marker ) during later developmental stages . In addition , variation in drifting rates between elements on the same bone were observed , such as the tip of the supinator crest in the humerus ( yellow marker ) compared with the distal margin of the deltoid tuberosity during late developmental stages . Interestingly , our analysis also revealed that other elements remained stationary for periods ranging from several days to most of the developmental process ( for definition of stationary elements , see Materials and Methods ) . These stationary elements were observed in different relative positions along the bone . For example , in the radius and ulna they were found near the proximal end , in the tibia and fibula at the center of the bone and , during late developmental stages of the humerus , such an element was found near the distal end . Moreover , during development some elements shifted modes between drifting and stationary . The observed drifting of elements provides strong support for the contribution of this mechanism to long bone scaling . However , the identification of stationary elements in all bones suggests the existence of an additional mechanism that preserves the relative position of elements along the shaft , which is active throughout the development of long bones . During longitudinal growth , the distance of elements from both ends of the bone increases , which is likely to change their relative positions . However , at any given time during the process , there is a single transverse plane along the shaft , referred to in the following as the fixed plane ( FP ) , whose relative position remains unchanged . This plane is found where the ratio between its distances to the proximal and distal ends of the bone equals the ratio of growth rates at the two ends ( Fig 7 ) , as in the equation: Distance ( FP , ProximalEnd ) Distance ( FP , DistalEnd ) =ProximalGrowthRateDistalGrowthRate ( see Materials and Methods for the mathematical formulation of the FP model ) . Interestingly , according to this model , the relative position of elements located in close proximity to the FP will be preserved as well , thereby rendering any drifting activity unnecessary . Conversely , elements located far from the FP would have to drift away from the FP in order to compensate for the change in relative position . Moreover , the farther from the FP an element is located , the faster it will have to drift . Based on this notion , we hypothesized that preservation of the relative position of stationary elements is achieved by regulation of the balance between proximal and distal growth rates . An obvious implication of this hypothesis is that passive preservation of the relative position of a stationary element over time necessitates that the specific balance between proximal and distal growth rates remains constant during that period . To date , a comprehensive analysis of proximal and distal growth patterns in each bone throughout the elongation process has not been performed . Therefore , to validate this implication , we extracted this information from the registered images of all long bones . When growth data is represented as a function of the total length of the bone , a constant balance between the growth rates is indicated by alignment of all data points along a straight line ( Fig 8 ) . Strikingly , statistical analysis of each bone showed that during all periods in which a stationary element was identified , a straight line provided a good fit to the data points ( R2 ranging between 0 . 87 and 0 . 99 for different bones and periods , p-value < 10e-05 ) . Another implication of this hypothesis is that for two different bone types , the closer the relative positions of their stationary elements are , the more similar their growth balance will be . Indeed , in both radius and ulna , where stationary elements are located adjacently and close to the proximal end of the bone , the relative contributions of distal growth were similar and significantly higher than those of proximal growth . In contrast , during late development of the humerus , where the stationary element is located near the distal end of the bone , the relative contribution of distal growth is significantly lower than that of proximal growth . Taken together , these results demonstrate the high agreement of the specific growth patterns of each bone with our model and , thus , support a potential role for differential longitudinal growth in regulating long bone isometric scaling . In order to provide direct evidence for the role of the mechanism that regulates growth balance in preservation of the relative position of stationary elements , it is necessary to demonstrate high spatial proximity between the FP and stationary elements . We therefore proceeded to determine the physical position of the FP in each bone during elongation based on the balance between proximal and distal growth rates ( for the calculations , see Materials and Methods ) . As seen in Fig 9 , during most stages at which a stationary element has been identified , the FP was found in extremely high proximity to it . For example , throughout the period of E18 . 5–P32 , the radial tuberosity ( blue marker ) is highly stationary ( range of relative position , 84%–89% ) and is also in close proximity to the FP ( relative position , 87% ) . Conversely , all elements located at a distance from the FP drifted in a rate and direction that corresponded with their position relative to the FP , as predicted by the FP model . For example , during the last stages of humerus development , the distal margin of the deltoid tuberosity ( green marker ) drifted proximally at a higher average rate ( 48 . 3 μm/day ) than the drift exhibited by the tip of the supinator crest ( 8 . 2 μm/day; yellow marker ) , which was closer to the FP . Taken together , these results provide direct evidence that the relative position of stationary elements in developing long bones is protected by a mechanism that maintains a unique and dynamic balance between proximal and distal growth rates of each bone . Moreover , they show that element drift serves as a complementary mechanism for preservation of the position of elements residing away from the FP , as both direction and rate of the drift correspond closely to the position of the element relative to the FP . Our results show that isometric scaling in long bones is achieved by coordination between the growth balance mechanism and element drift . However , the ability of the drifting mechanism to compensate for changes in the relative positions of elements raises the question of the necessity of a growth balance mechanism for the scaling process , as an alternative to symmetric growth in all bones . The process of element drift involves deposition and absorption of mineral by osteoblasts and osteoclasts , respectively [20] . Therefore , setting the growth balance to form a FP in high proximity to elements would reduce the required level of drifting and is thus likely to facilitate bone scaling . In light of this notion , we hypothesized that the growth balance is optimized for minimizing the drifting activity of symmetry-breaking elements in each bone . To examine this hypothesis , we first inspected the tendency of the FP to be located in high proximity to symmetry-breaking elements during growth in different bones . As can be seen in Fig 9 , in each bone and at almost any time point during development , there is at least one element whose relative position is protected by the FP . Moreover , in bones where the FP is relocated , it shifts from one element to another . These results provide initial evidence for the bias of the growth balance mechanism towards minimizing drifting activity . In order to test our hypothesis quantitatively , we designed a “drifting cost” assay for calculating the growth balance that minimizes the total distance ( in mm ) drifted by all elements of a bone at each time interval during development ( Fig 10; see Materials and Methods ) . Results showed that throughout the entire development of all long bones , FP position either overlaps or is in high proximity to the range that minimizes drifting activity . To evaluate statistically the likelihood of this proximity , we performed a permutation test under the null hypothesis that the position of the FP is not affected by the position of symmetry-breaking elements ( see Materials and Methods ) . The results ( p-value = 0 . 0013 ) clearly rejected the null hypothesis . Taken together , these results provide strong evidence that the growth balance mechanism is optimized for minimizing the drifting activity of symmetry-breaking elements .
In this work , we show that long bones are scaled isometrically throughout development . From the cellular level to the entire organism , growth in size typically necessitates changes in the physical proportions between the constituent parts in order for the system to remain functional [28 , 29] . Therefore , allometry is the prevailing mode of growth in both development and evolution [30–36] . During mouse development and growth , long bones elongate on average by more than five times , while the body mass increases by more than 20 times . Given the magnitude of these changes , one would expect that bones undergo allometric growth . However , our striking results show that apart from a minor deviation of the tibiofibular junction , longitudinal scaling of all long bones is clearly isometric throughout elongation . This persistent preservation implies that maintaining physical proportions of elongating bones has been highly significant in the evolutionary success of vertebrates . To the best of our knowledge , no comprehensive investigation of the implications of isometry during skeletal growth has ever been published . Yet , it can be speculated that isometry has important biomechanical and/or locomotor advantages . One of the main challenges that developing bones face during growth is the necessity to withstand increasing mechanical forces applied both by the body mass of the organism and by muscle activity [37 , 38] . Recently published results of finite element analysis on bones of xenarthrans [16 , 31] suggest that the relative position of bony superstructures , such as the femoral third trochanter , is regulated in order to mitigate the biomechanical consequences of a 100-fold increase in body mass during phylogenesis . This suggests that maintenance of the relative position of bony superstructures may have been selected to optimize the ability of bones to withstand the increase in bending loads during development as well . In this aspect it is interesting to note that the shape and relative proportions of skeletal elements , which are key determinants of the mechanical integrity of bones , scale close to isometrically when considered over nearly the entire size range of mammalians [39] . Another hypothesis is that isometry allows the maintenance of complex muscle firing patterns involving many muscles by preserving the positions of muscle attachment sites during growth . This strategy was proposed for the swallowing action in humans [40] , in which despite extraordinary growth , the relative positions of muscle insertions in the neck remains constant during ontogeny to prevent aspirating food and liquids . In the context of the axial skeleton , maintaining firing patterns directly relates to the evolutionary importance of highly coordinated motility activities such as running and climbing , which are essential for survival . A third possibility relates to the connection between skeletal proportions and the unique gait of each animal [41–43] . Alteration in skeletal proportions has long been known as a major driving force in the generation of new and diverse phyla , each optimized to a different motoric repertoire [42 , 44] . In view of this , it would be reasonable to interpret the isometric nature of developing long bone scaling as a means to maintain certain locomotor properties throughout the lifetime of the organism . Alterations in the relative position of such elements would most likely result in different force/speed ratios , which would in turn modify motoric function and environmental fitness . Thus far , the lack of reliable methods for quantitative recovery of the morphogenetic changes bones undergo during development has hampered the study of mechanisms that regulate long bones scaling . Our newly developed algorithm for rigid registration of multiple images enables accurate assessment of the morphogenetic sequence of each growing bone , allowing us to directly address this question . One mechanism that has been implicated in positional regulation of elements is drifting by mineral deposition and/or absorption at opposite surfaces of the element [20] . Utilizing our image registration algorithm , we provide here empirical evidence for the specific contribution of drifting to isometric scaling in each long bone . Notwithstanding the advantages of this mechanism , drifting of an ossified element requires massive and continuous activity of osteoblasts and osteoclasts , taking a high energetic toll on the morphogenetic process . Moreover , when the element serves as an anchoring point for tendons and ligaments , these tissues must be repositioned too [45–47] . This toll can be reduced by maintaining relative position of elements during elongation , thereby reducing the need for drifting . For that to happen , the fixed plane in each bone needs to be in close proximity to symmetry-breaking elements . Indeed , our data clearly demonstrate that in all bones and throughout elongation , the location of the FP protects specific elements from the need to drift . We therefore conclude that the ratio between proximal and distal longitudinal growth rates , which is reflected by the location of the FP , is a key mechanism in bone scaling . Moreover , we show that the distance from the FP determines which elements are to remain stationary at each time interval , as well as the rate and direction of drift of non-stationary elements . Altogether , these findings establish growth balance between the plates as the mechanism that orchestrates long bone scaling by synchronizing between drifting and passive protection of stationary elements . As mentioned , the rationale behind the existence of stationary elements is postulated to be energy saving during scaling . Indeed , our drifting cost assay shows that in all bones and throughout development , the location of the FP is either optimal or near-optimal for minimizing drifting activity . This implies that the optimization of the scaling process involves an accurate and dynamic balance between proximal and distal growth rates , which requires tight regulation of the activity of each growth plate throughout the elongation process . Generic mechanisms that regulate growth plate activity by synchronizing chondrocyte proliferation and differentiation , such as the PTHrP–IHH ( Gene ID: 19227 , 16147 , respectively ) feedback loop , have been extensively studied [1 , 2 , 48–52] . However , our analysis demonstrates that each growth plate exhibits a distinctive activity pattern , resulting in a unique ratio between elongation rates at the two ends of each bone . Along the same line , previous reports have shown differential growth plate activity , which leads to asymmetric longitudinal growth [8 , 13 , 53] . Typically , forelimb bones tend to grow away from the elbow joint , whereas bones in hind limbs tend to grow toward the knee joint . These findings and ours clearly imply the existence of additional mechanisms that control the specific activity of each growth plate . Interestingly , some of these works were performed on other model animals such as rat [54] , pig [11] , rabbit [55–60] , chick [61] , and humans [10] , suggesting that asymmetric growth of long bones is evolutionarily conserved across species . To date , little is known about mechanisms underlying differential activity between individual growth plates [13 , 53 , 62 , 63] . Our results indicate that the balance between proximal and distal growth rates of each bone is constantly optimized to minimize bone modeling . This suggests the existence of a feedback mechanism that incorporates data relating to the relative positions of superstructures into the molecular and cellular mechanisms that control growth plate activity . The regulatory role of the perichondrium/periosteum in growth plate activity has long been recognized and has been the subject of considerable research [2 , 62–75] . Damage to the integrity of the periosteum , as a result of bone fracture [64] or circumferential division with or without stripping of the periosteum [65–67 , 74] , was shown to lead to temporal acceleration of bone growth . This acceleration is typically accompanied by alteration in the balance between proximal and distal growth rates . Because the periosteal sheath is stretched over the entire external surface of the bone , including both the superstructures and the growth plates , it can pass to the growth plates signals concerning the relative position of superstructures . Interestingly , Crilly [64] previously hypothesized that periosteal tension down-regulates growth plate activity , as the higher the tension level , the more inhibited growth plate activity is . Moreover , he postulated that the damaged periosteum forms a scar tissue at the site of destruction . This scar tissue , which anchors the periosteum into the bone , creates an independent tension level near each growth plate . As a result , a new growth balance is formed , which equals the ratio between the distances from the site of the scar to the two ends of the bone , therefore maintaining the relative position of the scar site . This hypothesis can be generalized to account for normal growth conditions as well . Superstructures can be considered as natural anchoring points for the periosteum into the ossified bone , either due to the insertion of tendons through them into the bone cortex , or by means of steric interference , such as in the tibiofibular junction . This results in a regulatory loop whereby the superstructures determine the tension levels of the two periosteal segments , which control the ratio of growth rates by inhibiting growth plate activity , which in turn maintains the relative position of the superstructure . Since the perichondrium/periosteum is well known to express many signaling molecules that regulate growth plate activity [75–77] , it is plausible to hypothesize that these molecules are involved in mediating the mechanical signals . For instance , mice homozygous for a targeted disruption of fibroblast growth factor 18 ( FGF18; Gene ID: 14172 ) , synthesized by perichondrial cells , exhibit increased chondrocyte proliferation , differentiation to hypertrophic chondrocytes and IHH signaling [76] . Other signaling molecules synthesized in the perichondrium , such as FGF9 ( Gene ID: 14180 ) [76] , multiple bone morphogenetic proteins ( BMPs ) and WNTs [77] have been shown to operate on receptors expressed on chondrocytes , further demonstrating the regulatory role of the perichondrium/periosteum on growth plate activity . In this work , we uncover the isometric nature of longitudinal scaling of long bones during growth . Using a newly developed algorithm , we recover for the first time , to our knowledge , the morphogenetic sequence of developing long bones from early embryonic stages to maturity . These data enabled us to provide accurate assessments of both the specific activity of the different growth plates and the drifting patterns of symmetry-breaking elements along the bone shaft . Based on these analyses , we conclude that longitudinal growth patterns in each bone are adjusted to preserve isometry . The constant tendency of the growth balance to protect element positions strongly suggest that symmetry-breaking elements are involved in the mechanism that regulates the differential activity of growth plates .
For harvesting of embryos , timed-pregnant females were sacrificed by CO2 intoxication or cervical dislocation . Embryos and postnatal mice were sacrificed by decapitation with surgical scissors . All experiments were approved by the Institutional Animal Care and Use Committee ( IACUC ) at the Weizmann Institute of Science . For in vivo micro-CT imaging , mice were anesthetized with isoflurane ( 2-chloro-2-[difluoromethoxy]-1 , 1 , 1-trifluoro-ethane ) . C57/Bl6 mice ( Harlan Laboratories , Jerusalem , Israel ) were used for all analyses . For the generation of statistical maps and for validation of the automated registration algorithm only males were analyzed . Sex was determined in utero by PCR [78] and by external examination postnatally . At every developmental stage until P4 , six to eight mice from at least three different litters were evaluated . At P6 , two to four mice from two different litters were examined . Plug date was defined as E0 . 5 . The gravid uterus was dissected out and suspended in a bath of cold phosphate-buffered saline ( PBS ) , and the embryos were harvested after amnionectomy and removal of the placenta . Before scanning , harvested limbs were fixated overnight in 4% PFA/PBS and gradually dehydrated to 100% ethanol ( dehydration sequence: 25% , 50% , 70% , and twice in 100% for 30 min each ) . Then , according to [79] tissue samples were soaked in 2% iodine/ethanol solution ( Sigma ) for 48–72 h , depending on limb volume , at 4°C . Lastly , tissue samples were washed twice in 100% ethanol for 30 min and kept until scan at -20°C . For assessment of the distribution of symmetry-breaking elements along the shaft , bones at stages between E16 . 5 and P10 were scanned ex vivo using iodine contrast agent to allow visualization of cartilaginous tissue ( in each age group , n ≥ 3 ) . The rest of the images were obtained by sequential in vivo scans of four mice . The tibia and fibula were measured separately although they fuse early during development because their growth plates remain separated . For the generation of statistical maps and for validation of the registration algorithm , samples were scanned ex vivo in PBS solution by an eXplore Locus SP micro-CT scanner ( GE Healthcare , London , Ontario , Canada ) at 45kVp and 120μA . For all scans , 900 projections over 360 degrees , with 4 frames averaged for each projection at an exposure time of 2 , 850 ms per frame , resulted in an isotropic voxel size of 7 . 139 μm . Voxel intensity was represented by data type int16 . Calibration hydroxyapatite phantoms ( GE Medical ) were used to facilitate conversion of the linear attenuation of a given voxel to mgHA/cm³ . For measurements of relative and physical positions of element ( described in the following ) and of epiphyseal growth rates at E16 . 5 to P10 , bone samples from three to six mice were scanned ex vivo in 100% ethanol by a Zeiss Xradia Micro XCT-400 x-ray microscope ( Pleasanton , CA , United States ) at 40kV and 200μA with a linear magnification of x0 . 5 or x4 . Between 350 and 1 , 650 projection images were taken over 180 degrees at an exposure time of 500–7 , 500 ms per projection , providing a final isotropic voxel size of 4–18 μm . Optimal beam hardening was used in every reconstruction . Voxel intensity was represented by data type uint8 . Following reconstruction , voxel size of all images was standardized to 7 . 139 μm using trilinear image interpolation . For the same measurements in P20–P40 bones , bones of four mice were scanned in vivo by TomoScope 30S Duo scanner ( CT Imaging , Germany ) equipped with two source-detector systems in air medium at 41kV and 1mA . For all scan , 720 projections over 360 degrees at an exposure time of 125 ms per projection resulted in an isotropic voxel size of 36 μm . Voxel intensity was represented by data type int16 . Calibration to HU was performed using a factory default module . To calculate element position relative to the bone ends , all bones were reoriented such that the longitudinal axis of the bone was parallel to the vertical axis ( z ) of the image grid . Then , the vertical coordinate of the documented element ( Ez ) and of both the proximal ( Pz ) and distal ( Dz ) ends of the bone were measured in voxel units . Finally , the relative position of the element ( ERelativePosition ) was calculated as: ERelativePosition=Ez−DzPz−Dz×100 Calculation of the physical position of elements was done after image registration and determination of the origin of the vertical axis ( Oz ) . The vertical coordinate of the documented element ( Ez ) was measured in voxel units . Finally , the physical position of the element ( EPhysicalPosition; in mm ) was calculated as: EPhysicalPosition= ( Ez−Oz ) ×dVz where dVz is the physical size ( mm ) of the voxel in the vertical dimension . Mineral deposition was evaluated by intraperitoneal injections of calcein ( Sigma # C0875; 2 . 5 mg/kg body weight ) and alizarin complexone ( Sigma # a3882; 7 . 5 mg/kg ) into pregnant females and , postnatally , into cubs . Two mice were used for each injection regime . Prenatally harvested limbs were fixated overnight in 4% PFA/PBS , dehydrated to 100% ethanol , embedded in paraffin and sectioned at a thickness of 7 μm . Postnatally harvested limbs were fixated 24 h in 4% PFA/PBS and gradually dehydrated from 70% ethanol to 100% ethanol twice for 48 h each time . Then , samples were infiltrated and embedded in JB-4 Embedding Kit ( Electron Microscopy Science #14270–00 ) and sectioned longitudinally at a thickness of 7 μm . Fluorescence was visualized by confocal microscopy . Confocal imaging was performed using a Zeiss LSM 510 upright confocal microscope ( Carl Zeiss , Jena , Germany ) with an EC Plan-Neofluar 10x/0 . 3 objective , NA 1 . 0 . Calcein fluorochrome was excited with a 488 nm argon laser and alizarin with 561 nm argon laser . Following imaging , all images of the same section were stitched using Microsoft Image Composite Editor ( version 1 . 4 . 4 . 0 ) . Contrast was increased by using the “auto contrast” tool of Google Picasa ( version 3 . 9 . 137 ) and Matlab’s “imadjust” function . To segment bone from background voxels , we calculated for each image a private global threshold using Otsu’s method [80] and binarized it to bone ( 1 ) and background ( 0 ) . Threshold values , measured in milligrams hydroxyapatite per cubic centimeter , ranged between 128 . 5 mg HA/cm3 in early developing bones and 317 . 1 mg HA/cm3 in older bones . Then , to filter out background voxels that exceeded the threshold , all voxels that did not belong to the largest connected component ( i . e . the bone ) were zeroed out . Lastly , each image was manually inspected to assure the quality of the binarization . First , micro-CT bone images were manually registered by matching between the cortical cores . Then , each image was segmented and binarized as described above . To allow application of arithmetic operations between binarized images , the data type of all images was casted from Boolean to floating point . To generate statistical maps of morphological preservation over all bones belonging to the same age group , all binarized images of the following age group were averaged: E16 . 5 ( n = 7 ) , E17 . 5 ( n = 8 ) , E18 . 5 ( n = 8 ) , P1 . 5 ( n = 7 ) , P2 . 5 ( n = 7 ) , P4 . 5 ( n = 8 ) and P6 . 5 ( n = 3 ) . To generate statistical maps of morphological preservation over all bones belonging to pairs ( e . g . E16 . 5 , E17 . 5 ) or triples ( e . g . E16 . 5 , E17 . 5 , E18 . 5 ) of consecutive days , all binarized images of each group were averaged . Lastly , statistical maps were color-coded such that highly preserved regions are shown red and little preserved regions in blue . We start by addressing the particular case of pairwise image registration . To avoid the potential interference of non-overlapping image segments while matching the cortical cores , we zero-out the trabecular bone segments ( see S2 File ) in both source ( static ) and target ( transformed ) images using an in-house software , as well as all background segments by applying Otsu’s global thresholding method [80] . Although the remaining segments include cortical regions other than the core , this approximation provides registration results of comparable quality to those estimated using the cortical core alone . In order to estimate the spatial transformation that anatomically aligns the target and source images , we apply a volume based registration approach that utilizes normalized cross correlation ( NCC ) as a similarity measure and downhill descent as the optimization method [24–26] . Since the shape of both registered objects is cylinder-like , initiating the registration using an arbitrary transformation might result in local optima , such as matching the proximal end of one bone to the distal end of the other . Thus , in order to initiate the optimization in the vicinity of the global NCC optimum , we compute a coarse estimate of the registration using a cylindrical shape descriptor of the bone cortex . The descriptor , referred to as the circumferential profile ( CP ) : ψ ∊ ℝ+9 , 360 , is a polar representation of the external contour of the cortex at nine equidistant transverse slices along its length ( i . e . at 10% , 20% , … , 90% length between the proximal and distal ends of the cortex ) , at 360 equi-angular sample points along the contour of each slice . The radii are measured from the geometric center of that section . We first apply principal component analysis ( PCA ) on the point cloud consisting of all nonzero voxels , to allow direct extraction of the descriptor from transverse slices . Thus , the longitudinal dimension of the bone ( PC1 ) is aligned with the x-axis , and all transverse slices are aligned with YZ planes . A descriptor of a target image ( ψt ) is matched to a descriptor of a source image ( ψs ) using an affinity function based on the L2-norm: Ψs , t ( θ ) =−19∑l=19‖ψt[l , ( 1 , … , 360 ) +θ]−ψs[l , ( 1 , … , 360 ) ]‖2 where for a given angle θ , the affinity value is the average Euclidian distance between the circumferential profile of It rotated by θ about PC1 , and that of Is . Therefore , the set of local optima values of Ψs , t is likely to include the vicinity of the global NCC optimum as well . We also consider right/left inversions ( mirroring ) , and proximal/distal inversions of It , to identify all local optima of Ψs , t . In order to improve the localization of the vicinity of the global NCC optimum , we apply several volume-based registration steps , initialized at each identified point of local affinity optimum . The path resulting in the highest NCC score is further optimized by additional volume-based registration refinement steps , until convergence is reached . A pair of images that fails to achieve an NCC score of 0 . 7 or higher is marked as an erroneous match and undergoes manual validation and , possibly , correction by the operator . To generalize the registration procedure from two images to a larger dataset , we utilize a graph theoretic approach whereby we first compute all pairwise registrations between pairs of bones that differ in length by 20% at most , while storing the obtained NCC score of each pair in a distance matrix DN×N . We compute the maximum spanning tree ( MST ) of the graph encoded by D with a preselected image serving as the root . To guarantee that the graph is connected for the extraction of the MST , we register all length-wise consecutive pairs of bones , regardless of their length differences . Last , we align all of the images to the root MST image , by aggregating the transformations along the path and connecting them over the MST graph . The origin of the longitudinal axis of each bone ( Z = 0 ) was set at the transverse section in which the diameter of the bone collar in the average E16 . 5 image was minimal . This section was presumed to be the location of initial chondrocyte hypertrophy at the primary ossification center from which longitudinal growth progresses bidirectionally and , therefore , it provided a natural choice for the longitudinal origin of the bone . Positive values represent elements located proximally to the origin , whereas negative values represent distal locations . Notably , although the existence of a longitudinal origin is necessary for all further calculations , its specific location along the axis does not influence any of the obtained results . By definition , a stationary element is one for which the physical position remains constant over time . However , to account for random fluctuations and noise , we define the speed of drift of an element between time points t0 and t1 as the absolute change in physical position divided by the change in total bone length: EDriftSpeed ( t0 , t1 ) =|EPhysicalPosition ( t1 ) −EPhysicalPosition ( t0 ) |BoneLength ( t1 ) −BoneLength ( t0 ) We also define a level of tolerance ( 0 < ε < 1 ) to separate between stationary ( EDriftSpeed ( t0 , t1 ) < ε ) and drifting elements ( ε ≤ EDriftSpeed ( t0 , t1 ) ) . In order to set an appropriate ε value , we sought to identify the minimal speed of an element that is guaranteed to be drifting . To this end , we relied on the identification of drifting elements reported in [20] . We then calculated the rate of drift of these elements and the rates of proximal and distal growth based on the data extracted from the image analyses ( Fig 5 and Fig 7 ) . This analysis showed that drifting speeds range between 0 . 1409 ( the distal end of the femoral third trochanter between days P20 and P24 ) and 0 . 5546 ( the distal end of the humeral deltoid tuberosity between days P24 and P28 ) . These results suggests 0 . 1409 as the drifting speed below which changes in physical position of elements are likely to result from random fluctuations or noise and , therefore , the elements should be classified as stationary . In order to further minimize the possibility of false rejection of drifting elements , we used the lower value of ε = 0 . 1 , i . e . , when the change in physical position of an element is less than 10% the total elongation of the bone during the specified time interval . Assume a bone with a total length l ∊ ℝ+ , in which the physical positions of the proximal and distal ends of the bone are represented as a function of l: Dist ( l ) < 0 < Prox ( l ) , respectively , such that: Prox ( l ) − Dist ( l ) = l . Note that since Dist ( l ) is negative and monotonically decreasing , the growth curve of the distal end is: −Dist ( l ) . According to our statistical regression analysis , during each time period in which a stationary element has been identified a linear line provided a good fit to the data points . Therefore , the total elongation of the distal end ( Dist ) as a function of the total length of the bone ( l ) can be expressed as: Dist ( l ) = a ⋅ l + b , where a and b are the slope and intercept of the line , respectively . Since growth is represented as a function of total bone length , the rate of elongation of the entire bone is constantly 1 . In addition , the rate of elongation of the distal end is a and , therefore , the relative contribution of the distal end to the elongation of the bone is constantly a1=a , which by definition is equal to the relative position of the FP: FPRelative = a . Taken together , we get that: Dist ( l ) = FPRelative ⋅ l + b , thus allowing us to extract the relative position of the FP from the parameters obtained by the regression . By definition , the physical position of the FP is: FPPhysical = −Dist ( l ) + FPRelative ⋅ l = −b , which leads to: Dist ( l ) = FPRelative ⋅ l − FPPhysical , thus allowing us to extract the physical position of the FP as well . Assume a bone with n symmetry-breaking elements located at constant relative positions: ERelativei ( i=1 , … , n ) , such that ERelativei≤ERelativei+1∀i∈{1 , … , n−1} , and a FP with a relative position FPRelative ∊ [0 , 1] . We define a loss function that quantifies the sum of the rates of drift over all n elements , and we search for the relative position of the FP that minimizes the loss: argminFPRelative∈[0 , 1]∑i=1n|EPhysicali ( l ) ′| . According to the FP model: EPhysicali ( l ) ′=ERelativei−FPRelative and therefore: ∑i=1n|EPhysicali ( l ) ′|=∑i=1n|ERelativei−FPRelative| , which is the sum of distances of all elements from the FP in terms of relative positions . It can be shown that for n∈ℕodd:argminFPRelative∈[0 , 1]=median{ERelativei} and for n∈ℕeven:argminFPRelative∈[0 , 1]={x|ERelativen2≤x≤ERelativen2+1} . That is to say , when there are an odd number of elements , the optimal relative position of the FP overlaps that of the middle element along the bone , whereas when the number of elements is even , any relative position between the two middle elements is optimal . Based on this mathematical term , to calculate the optimal relative position of the FP through the development of each bone we first calculated the average relative position of each element . Then , for each value of total bone length argminFPRelative∈[0 , 1] was calculated , while considering only ossified elements . Statistical analysis was performed using the R language for statistical computing [81] . Regression analysis for linearity of growth data was performed using ordinary least square and evaluated based on scatterplots , Pearson’s correlation and p-values of the t-scores . The statistical significance of the proximity between FPs and symmetry-breaking elements was determined by permutation analysis . The null hypothesis was that the distribution of FP positions is not affected by the positions of elements and , therefore , the average distance of a FP from all elements would not have changed if it had been located on another bone . First , we measured for all FPs on all bones the distances to all elements , then calculated the average distance for each bone and , finally , the average distance for all bones . Then , the FPs and the elements of the bones were mixed and permutation distribution was calculated by a one- ( left- ) sided test . Significant difference ( p-value < 0 . 01 ) would imply rejection of the null hypothesis . All computations were conducted on an Intel Core i7-3930K CPU at 3 . 2 GHz with 16 GB of RAM , with Windows 7 64-bit operating system and on Matlab platform ( MatLab and Image Processing toolbox R2014a ( 8 . 3 . 0 . 532 ) 64-bit , The MathWorks , Inc . , Natick , Massachusetts , United States ) . The volume-based registration module for rigid image registration algorithm is from the freely available Image Registration Toolkit ( IRTK ) , used under license from Ixico Ltd [25–27] . Manual image registration for the assessment of preservation of mineralized structures and for spatial standardization of all in vivo images was performed using the free and open-source software package 3DSlicer v . 3 . 6 . Visualization of 3D isosurface of micro-CT images was performed using the free software MicroView 2 . 1 . 2 ( GE Healthcare ) . | One of the major challenges that developing organs face is scaling , that is , the adjustment of physical proportions during the massive increase in size . Bone superstructures are projections that typically serve for tendon and ligament insertion or articulation . Therefore , superstructure position along the bone is crucial for musculoskeletal functionality . As bones are rigid structures that elongate only from their ends , it is unclear how superstructure positions are regulated during growth to end up in the right locations . Here , by analyzing a massive database of micro-CT images of developing mouse long bones , we show that all superstructures maintain their relative positions throughout development . It has been suggested that during development , superstructures are continuously reconstructed and relocated along the shaft , a process known as drift . However , our analysis reveals that most superstructures did not drift at all , implying the involvement of another mechanism . Indeed , we identify a novel mechanism for bone scaling , whereby each bone exhibits a specific and unique balance between the growth rates from its two ends , which accurately maintains the relative position of its superstructures . Moreover , we show mathematically that this mechanism minimizes the cumulative drift of all superstructures , thereby optimizing the scaling process . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Isometric Scaling in Developing Long Bones Is Achieved by an Optimal Epiphyseal Growth Balance |
The Human T-Lymphotropic Virus type 1c subtype ( HTLV-1c ) is highly endemic to central Australia where the most frequent complication of HTLV-1 infection in Indigenous Australians is bronchiectasis . We carried out a prospective study to quantify the prognosis of HTLV-1c infection and chronic lung disease and the risk of death according to the HTLV-1c proviral load ( pVL ) . 840 Indigenous adults ( discharge diagnosis of bronchiectasis , 154 ) were recruited to a hospital-based prospective cohort . Baseline HTLV-1c pVL were determined and the results of chest computed tomography and clinical details reviewed . The odds of an association between HTLV-1 infection and bronchiectasis or bronchitis/bronchiolitis were calculated , and the impact of HTLV-1c pVL on the risk of death was measured . Radiologically defined bronchiectasis and bronchitis/bronchiolitis were significantly more common among HTLV-1-infected subjects ( adjusted odds ratio = 2 . 9; 95% CI , 2 . 0 , 4 . 3 ) . Median HTLV-1c pVL for subjects with airways inflammation was 16-fold higher than that of asymptomatic subjects . There were 151 deaths during 2 , 140 person-years of follow-up ( maximum follow-up 8 . 13 years ) . Mortality rates were higher among subjects with HTLV-1c pVL ≥1000 copies per 105 peripheral blood leukocytes ( log-rank χ2 ( 2df ) = 6 . 63 , p = 0 . 036 ) compared to those with lower HTLV-1c pVL or uninfected subjects . Excess mortality was largely due to bronchiectasis-related deaths ( adjusted HR 4 . 31; 95% CI , 1 . 78 , 10 . 42 versus uninfected ) . Higher HTLV-1c pVL was strongly associated with radiologically defined airways inflammation and with death due to complications of bronchiectasis . An increased risk of death due to an HTLV-1 associated inflammatory disease has not been demonstrated previously . Our findings indicate that mortality associated with HTLV-1c infection may be higher than has been previously appreciated . Further prospective studies are needed to determine whether these results can be generalized to other HTLV-1 endemic areas .
The Human T-Lymphotropic Virus type 1 ( HTLV-1 ) is an oncogenic retrovirus that preferentially infects CD4+ T cells[1] . Worldwide , HTLV-1 infects as many as 20 million people who predominantly dwell in areas of high endemicity in south-western Japan and developing countries of the Caribbean basin , South America and sub-Saharan Africa[2] . An endemic focus is present in central Australia[3] where more than 40% of Indigenous adults are HTLV-1c-infected in some remote communities[4] . Clinically significant sequelae of HTLV-1 infection include a haematological malignancy , Adult T cell Leukemia/Lymphoma ( ATL ) , and inflammatory diseases , such as HTLV-1 associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) [1] . In Japan and the Caribbean , life-time risks of HAM/TSP and ATL range between 0 . 3–4% and 1–5% , respectively[1] . Bronchiectasis is the most common clinical manifestation of HTLV-1 infection in Indigenous Australians , amongst whom the adult prevalence of this condition is the highest reported worldwide ( >1% ) [5 , 6] . Chest computed tomography has also revealed bronchiectasis in Japanese adults infected with HTLV-1; however , the most frequently reported radiological pattern of HTLV-1 associated pulmonary disease in this population is bronchitis/bronchiolitis[7 , 8] , which has not been described in Indigenous Australians . In endemic areas in Japan and Africa , HTLV-1 seropositivity is associated with increased mortality[9–12] , which has been attributed to non-neoplastic conditions[9 , 10] . The interpretation of these studies is limited by their inability to control for clinically defined comorbid conditions that might independently increase mortality[9 , 11 , 12] [10] . For example , HTLV-1 seropositivity had no effect on mortality in a large hospital-based cohort of Indigenous Australian adults after adjusting for other medical conditions[13] . Given the close association between the number of HTLV-1-infected cells in peripheral blood ( the HTLV-1 proviral load , pVL ) and serious HTLV-1 associated complications[1 , 14] , any influence of HTLV-1 infection on mortality might be revealed by stratifying outcomes according to HTLV-1 pVL . In central Australia , Indigenous adults with higher HTLV-1c pVL have more extensive , radiologically defined pulmonary injury[6] and are more likely to present with life-threatening bacterial infections[15] . A single , small study in Guinea-Bissau , where causes of death could not be ascertained , found that mortality increased with HTLV-1 pVL[16] . The present study was therefore commenced to quantify the prognosis of HTLV-1c infection and chronic lung disease and the risk of death according to the HTLV-1c pVL in a hospital-based cohort of Indigenous adults who were well characterized with regard to comorbid conditions and for whom causes of death could be accurately determined in nearly all cases .
Alice Springs Hospital ( ASH ) is the only medical facility serving central Australia , an area of >1 , 000 , 000 km2 . Critically ill patients are transferred by air to ASH , which has sophisticated diagnostic capabilities . All Indigenous patients aged >15 years with a discharge diagnosis of bronchiectasis , 1st June 2008 to 31st December 2013 , were identified from the ASH patient management database , which coordinates all in-patient and out-patient hospital activities . Indigenous status was determined from self-reported data obtained at admission , as recorded in the patient information database . Potential subjects were offered enrolment when next admitted for >48 hours . Among 165 eligible cases , 154 were recruited ( eleven subjects left hospital before recruitment was possible ) . Written reports for chest high-resolution computed tomography ( cHRCT ) were reviewed for all subjects , confirming bronchiectasis in 104 cases and bronchitis with or without bronchiolitis in 33 cases ( bronchitis alone , 20; bronchitis and bronchiolitis , 12; bronchiolitis alone , 1 ) ( Fig 1 ) . Patients with chronic pulmonary disease were treated according to local guidelines which includes antibiotic therapy for infective exacerbations [17] . A further 686 Indigenous patients aged >15 years who were admitted for >48 hours were prospectively recruited during the same period . These control subjects had no evidence of lower respiratory tract infection at the time of recruitment , no recorded discharge diagnosis of bronchiectasis , and no clinical or radiological evidence of bronchiectasis , Research team members who were unaware of HTLV-1 serostatus were responsible for recruitment ( Fig 1 ) . Demographic and clinical details were extracted from medical records at the time of recruitment using a standardized data-collection form . HTLV-1 associated conditions were identified from medical records at baseline and study end . No control patient developed chronic pulmonary disease during the study period . Mortality data was obtained at study end from the ASH patient management database , and the cause of death was determined from death certificates held in Registries in the Northern Territory of Australia and South Australia . Death certificates were not available for four subjects who died in remote communities in Western Australia , for whom a cause of death was sought from the responsible remote clinic . Bronchitis was diagnosed where cHRCT revealed bronchial wall thickening or dilatation not fulfilling criteria for bronchiectasis , and bronchiolitis where cHRCT revealed multiple centrilobular nodules or a ‘tree-in-bud’ pattern[8] . Chronic obstructive pulmonary disease ( COPD ) required a clinical diagnosis in the medical record and appropriate chest X-ray findings . Emphysema without bronchial wall injury or bronchiolitis was recorded for 18 subjects with COPD examined by cHRCT . Chest HRCT was not performed on 12 subjects who did not meet criteria for such imaging[17] . No subject with symptoms consistent with HAM/TSP received lumbar puncture; the diagnosis was therefore considered ‘probable’ in all cases . Asymptomatic HTLV-1-infected subjects were those without radiological evidence of airway inflammation or recognized HTLV-1 associated conditions [1] . Residence >80 km from the township of Alice Springs was defined as remote . The study was approved by the Central Australian Human Research Ethics Committee . All patients , and their parents/guardians if aged <18 years , gave written informed consent in primary languages . Whole blood samples were collected from each participant at the time of recruitment . Peripheral blood buffy coats ( PBBC ) were prepared , and plasma and PBBCs were stored at ASH at -80° C until transfer to the National Serology Reference Laboratory , Melbourne . Samples were screened for antibodies to HTLV-1 using both an enzyme immunoassay ( Murex HTLV-I + II , DiaSorin , Italy ) and a particle agglutination assay ( Serodia HTLV-1 , Fujirebio , Tokyo , Japan ) . Any sample reactive on either screening assay was tested by Western blot ( HTLV-I/II Blot2 . 4 , MP Biomedicals Asia Pacific Pte . Ltd . , Singapore ) and HTLV-1c PCR . Primers and fluorescently labelled hydrolysis probes were designed to target a highly conserved 88 bp fragment of the gag gene in the p19 coding region of the Australo-Melanesian HTLV-1 subtype C[18] and multiplexed with primers and probes to the albumin gene[19] . SP cells were used to generate a standard curve from which HTLV-1 pVL ( copies per 105 peripheral blood leukocytes; PBL ) was calculated . Samples and standards were extracted using the Qiagen QIA blood Mini Extraction kit and the extracts amplified on a Stratagene Mx3000p Real Time PCR Instrument ( Integrated Sciences ) . The extract ( 5 μL ) was added to 20 μL of Master mix containing 2 x Brilliant Multiplex QPCR Master Mix ( Agilent Technologies ) 0 . 3 μM of each primer ( Gene works ) and 0 . 16 μM of each probe ( Sigma-Aldrich ) and amplified at 95°C for 10 minutes , 45 cycles at 95°C for 30 seconds , 65°C for 60 seconds and 72°C for 60 seconds . The clonality of HTLV-1-infected PBLs was determined by high-throughput sequencing of PBBC cell genomic DNA . The oligoclonality index ( OCI ) was calculated as previously described[20] , and then adjusted to limit underestimation of the OCI due to the small observed number of proviruses[21] . The OCI provides a measure of the non-uniformity of the clone abundance distribution of the infected cell population: OCI = 1 indicates perfect monoclonality ( only one clone constitutes the total proviral load ) ; OCI = 0 indicates perfect polyclonality ( all clones have the same abundance ) [20] . Samples were selected for clonality analysis if subjects had HTLV-1 pVL >100 copies per 105 PBL , were HBsAg negative and strongyloides seronegative . Although 53 subjects met these criteria , technical difficulties prevented analysis for nine subjects ( inadequate number of unique integration sites to accurately determine OCI , 7; unable to sequence integration site , 2 ) . The OCI was therefore compared between 29 asymptomatic and 15 symptomatic subjects ( bronchiectasis , 10; bronchitis/bronchiolitis , 3; uveitis , 2 ) . All analysis was performed using Stata version 14 . 2 ( StataCorp , College Station , USA ) . HTLV-1 pVL was log-transformed and also categorized as low if <1000 and high if ≥1000 per 105 PBL , a cut-off that has been associated with an increased risk of HAM/TSP[22] . Differences between subjects who were HTLV-1 uninfected , those with low HTLV-1 pVL , and those with high HTLV-1 pVL were assessed using ANOVA for continuous variables and chi-squared tests for categorical variables . For statistical purposes , causes of death were grouped into six non-overlapping categories: bronchiectasis , sepsis , cardiovascular disease , malignancy , chronic kidney disease and chronic liver disease . We used survival analysis to determine the association between HTLV-1 pVL and both overall and cause-specific mortality . Subjects were followed until either date of death or 30th March 2015 . The association between overall mortality and HTLV-1 pVL was assessed using log-rank tests and Kaplan-Meier curves in univariate analysis and using Cox regression for multivariate analysis . Associations with cause-specific mortality were assessed using competing risks analysis with all causes except the specific cause of interest treated as a competing risk . Where HTLV-1 pVL was treated as a categorical variable we also tested for a trend by creating a continuous variable with value zero for those uninfected , and with the median value of HTLV-1 pVL for those in the low and high pVL categories . Predictors of chronic airways inflammation were assessed using multivariate binary logistic regression . A 2-sided Type 1 error rate of p<0 . 05 was regarded as indicating statistical significance in each analysis .
Radiologically defined airways inflammation was more common among HTLV-1c-infected subjects ( Table 1 ) . Bronchiectasis was confirmed in 59/307 ( 19 . 2% ) HTLV-1c-infected subjects and 45/533 ( 8 . 4% ) HTLV-1c uninfected subjects . Similarly , cHRCT revealed bronchitis/bronchiolitis in 21/307 ( 6 . 8% ) HTLV-1c-infected subjects and 12/533 ( 2 . 3% ) who were HTLV-1c uninfected ( Table 1 ) . Compared to the median HTLV-1c pVL of asymptomatic subjects ( n = 208 , 30 . 4 copies per 105 PBL ( min , 0 . 01; max , 18600 ) , those for subjects with bronchiectasis ( n = 59 , 494 copies per 105 PBL; min , 0 . 01; max , 87900 ) ( p = 0 . 001 ) and bronchitis/bronchiolitis ( n = 21 , 486 copies per 105 PBL; min , 0 . 01; max , 70200 ) ( p = 0 . 042 ) were 16-fold higher , ( Fig 2 ) . Median HTLV-1c pVL of subjects with any airways inflammation ( bronchiectasis and bronchitis/bronchiolitis; 490 copies per 105 PBL; min , 0 . 01; max 87900 ) was 16-fold higher than that of asymptomatic subjects ( p<0 . 001 ) . Few other recognised causes of bronchiectasis were found ( Table 2 ) . In a multivariate model that controlled for demographic factors , smoking and harmful alcohol consumption , HTLV-1 infection increased the risk of any airways inflammation 2 . 9-fold ( p<0 . 001 ) ( Table 3 ) , while HTLV-1c pVL ≥1000 copies per 105 PBL increased the risk 2 . 2-fold among HTLV-1-infected subjects ( p = 0 . 006 ) ( Table 4 ) . Other HTLV-1 associated conditions included infective dermatitis ( 4 ) , crusted scabies ( 4 ) , probable HAM/TSP ( 2 ) and uveitis ( 2 ) . Four subjects with HTLV-1-associated bronchiectasis had other sequelae of HTLV-1 infection ( infective dermatitis , 2; HAM/TSP , 1; uveitis , 1 ) . There was no difference in strongyloides seropositivity between groups ( Table 1 ) , nor was there any difference in log-transformed HTLV-1c pVL using linear regression ( p = 0 . 409 ) ; median ( IQR ) HTLV-1c pVL copies per 105 PBL for subjects who were strongyloides seronegative ( 133; IQR 2 . 1 , 1387 ) , seropositive ( 61; IQR 1 , 1039 ) and those with equivocal strongyloides serological results ( 190; IQR 2 . 7 , 2473 ) . Although the risk of malignancy was not increased among HTLV-1c infected subjects , one subject ( HTLV-1c pVL , 5500 copies per 105 PBL ) developed ATL during follow-up , one developed penile cancer ( HTLV-1c pVL , 70200 copies per 105 PBL ) and another metastatic anal cancer ( HTLV-1c pVL , 28000 copies per 105 PBL ) . The median OCI did not differ between asymptomatic ( 0 . 401; IQR 0 . 350 , 0 . 485 ) and symptomatic groups ( 0 . 411; IQR 0 . 355 , 0 . 552 ) ( p = 0 . 51 ) or when symptomatic subjects with uveitis were excluded from the analysis ( symptomatic group median OCI 0 . 429; IQR 0 . 369 , 0 . 552 ) ( p = 0 . 341 ) . Median log10 HTLV-1c pVL copies per 105 PBL in subjects selected for clonality analysis did not differ between asymptomatic subjects ( 0 . 947; IQR 0 . 342 , 1 . 964 ) and those with airways inflammation ( 1 . 50; IQR 0 . 669 , 4 . 074 ) ( p = 0 . 20 ) . During 2140 person-years of follow-up , 155 deaths were recorded ( HTLV-1 uninfected , 85; HTLV-1 infected , 70 ) . Non-bronchiectasis causes of death were infections ( 37 ) ( lower respiratory tract infections , 15 ) , cardiovascular disease ( 37 ) , malignancy ( 15 ) , end-stage kidney disease ( ESKD ) ( 15 ) , chronic liver disease ( 10 ) , intracerebral haemorrhage ( 4 ) , primary pulmonary hypertension ( 2 ) and amyloidosis ( 1 ) . Subjects with high HTLV-1c pVL were more likely to die during the study period ( Fig 3 ) . Mortality rates for high HTLV-1c pVL , low HTLV-1c pVL and HTLV-1 uninfected subjects were 28 . 4% ( 27/95 ) , 20 . 2% ( 43/212 ) and 15 . 9% ( 85/533 ) , respectively ( p = 0 . 011 ) . The unadjusted HR for death among subjects with low and high HTLV-1c pVL were 1 . 24 ( 95% CI , 0 . 85 , 1 . 81 ) and 1 . 75 ( 95% CI , 1 . 13 , 2 . 670 ) , respectively ( p = 0 . 021 for trend ) . The statistical significance was diminished after adjusting for age , gender , place of residence and harmful alcohol consumption ( p = 0 . 084 ) . The effect of HTLV-1c pVL on overall mortality was lost in a multivariate model that included bronchiectasis ( aHR , 1 . 045; 95% CI , 0 . 658–1 . 660 ) ( Table 5 ) . Other predictors of death were age at test , male gender and comorbid conditions ( Table 5 ) .
In a large hospital-based cohort of Indigenous Australian adults , a higher baseline HTLV-1c pVL prospectively predicted a bronchiectasis-related death , which occurred at a mean age of only 49 . 5 years . In addition to confirming a previously reported association between HTLV-1c infection and bronchiectasis[5] , the present study also revealed an association with bronchitis/bronchiolitis . Airways inflammation was strongly associated with higher HTLV-1c pVL[6 , 15] . The median HTLV-1c pVL of subjects with radiologically defined airways inflammation was 16-fold higher than that for asymptomatic HTLV-1-infected subjects , and risk of airway inflammation increased three-fold among subjects with higher HTLV-1c pVL in an adjusted model . HTLV-1 associated inflammatory diseases are thought to result from a genetically determined , inefficient cytotoxic T lymphocyte response , permitting widespread dissemination of the virus in a large number of HTLV-1-infected T-cell clones , which is reflected in a high HTLV-1 pVL[23] . Organ infiltration by HTLV-1-infected lymphocytes then leads to high local HTLV-1 antigen levels , provoking an immune response and tissue injury following the release of pro-inflammatory cytokines and chemokines[23] . Although this has been best studied for the prototypical HTLV-1 associated disease , HAM/TSP , HTLV-1 infection is also associated with inflammation in other organs[14] including the lungs [24] . Consistent with the presumed mechanism of pathogenesis of HAM/TSP [23] , pulmonary involvement is associated with infiltration of HTLV-1-infected lymphocytes[25 , 26] , increased tax/rex mRNA[27] expression , and an inflammatory cytokine milieu in bronchoalveolar lavage fluid[27] . In large Japanese case series , cHRCT was abnormal in 30–61% of HTLV-1 infected subjects of whom 23 . 6–29 . 5% had a bronchitis/bronchiolitis pattern of disease and 15 . 6–22 . 5% had frank bronchiectasis[7 , 8] . The pathological correlate of these observations is lymphocyte infiltration in bronchiole walls[7] . Persistent HTLV-1-mediated airways inflammation may therefore lead to progressive bronchial wall dilatation , and bronchiectasis . High rates of bronchiectasis among HTLV-1-infected Japanese adults[7 , 8] , and associations between HAM/TSP and bronchiectasis in UK [28] and Brazilian cohorts [29] suggest that HTLV-1-associated bronchiectasis affects individuals of diverse genetic backgrounds infected with HTLV-1 strains other than HTLV-1c . Although HTLV-1 associated pulmonary disease in Japan is thought to be largely sub-clinical[14] , published clinical details are limited and prospective survival studies have not been performed . Consistent with other HTLV-1 associated inflammatory diseases[1 , 14] , the median baseline HTLV-1c pVL of subjects with airways inflammation was substantially higher than that of asymptomatic subjects . For example , the median HTLV-1c pVL in subjects with HAM/TSP is between 7-fold and 16-fold greater than that of asymptomatic subjects[22][28][30–32] . Among subjects with HAM/TSP , higher HTLV-1 pVL correlates with more severe motor weakness[31] and more rapid neurological progression[32] . We previously demonstrated that HTLV-1-infected Indigenous adults have more diffuse bronchiectasis[5] and that a higher HTLV-1c pVL correlates with more extensive pulmonary injury[6] . In contrast to HTLV-1-mediated inflammation in other tissues , pulmonary parenchymal injury can result in directly life-threatening complications , including respiratory failure[5] . Among subjects with airways disease in whom HTLV-1 oligoclonality could be studied , there was no difference in the median OCI when compared to that of asymptomatic subjects . This suggests that higher HTLV-1c pVL were due to an increased number of infected clones rather than clonal expansion , which is consistent with the conclusion previously reported for subjects with HAM/TSP[20] Increased mortality due to a specific HTLV-1-associated inflammatory disease has not been prospectively demonstrated previously . However , an excess mortality that is not attributable to ATL or currently recognized HTLV-1-associated inflammatory diseases has been reported in other endemic areas . Adjusted hazard ratios of death are 1 . 3[10] to 1 . 77–1 . 87[9] in Japanese outpatient cohorts , and 3 . 8 and 2 . 3 for young and middle-aged adults , respectively , in a community-based cohort in Guinea-Bissau[11 , 12] . In a study that included only 48 HTLV-1-infected subjects in Guinea-Bissau , mortality was associated with higher HTLV-1 pVL[16] . In Japan , excess mortality was attributed to non-neoplastic diseases , most commonly unspecified kidney and cardiac conditions[9 , 10] . Although the ASH cohort included subjects with established ESKD and heart disease , HTLV-1 infection was only associated with bronchiectasis-related deaths , and this effect was only revealed after stratifying by HTLV-1c pVL . The difference between studies in the clinical conditions associated with excess mortality may reflect the high burden of illness in our hospital-based cohort , the inability to control for comorbid conditions in other studies , and differences in the social circumstances of the various study populations . The strengths of this prospective cohort study include the recruitment of nearly all eligible subjects with a discharge diagnosis of bronchiectasis , the use of cHRCT for diagnosis and the blinding of ASH researchers to the HTLV-1 serostatus of subjects and of those who performed HTLV-1 studies to their clinical state . Nevertheless , some design limitations must be recognized . First , investigations to exclude other causes of airways disease could not be performed in all cases . However , consistent with previous studies[5 , 6] , a specific aetiology was rarely found among >70% of subjects who were screened for conditions generally associated with bronchiectasis[17] . Although an effect of childhood respiratory infections cannot be excluded in the present study , we previously found HTLV-1 infection to be the major predictor of adult bronchiectasis in a case-control study that controlled for such infections[6] . Second , only subjects with a discharge diagnosis of bronchiectasis were specifically targeted for recruitment . Twelve subjects with COPD ( HTLV-1 infected , 4; HTLV-1 uninfected , 8 ) were incidentally recruited but not examined by cHRCT because they did not clinically warrant further imaging[17] . The contribution of HTLV-1c infection to less severe respiratory disease than that associated with a discharge diagnosis of bronchiectasis , and the validity of our conclusions in a community setting , require further study . Finally , the absence of an association between strongyloides seropositivity and higher HTLV-1c pVL may be due to the fact that strongyloides serology was assayed in subjects without symptomatic strongyloidiasis . In summary , HTLV-1c infection and higher HTLV-1c pVL were strongly linked to airways inflammation in a hospital-based cohort of Indigenous Australian adults . Furthermore , higher baseline HTLV-1c pVL prospectively predicted death due to bronchiectasis , which may result from more extensive disease[5 , 6] , predisposing to life-threatening complications[5] among subjects who are unable to control HTLV-1 replication[6] . Elucidating the causes of higher mortality among people infected with HTLV-1 is relevant to an estimated 20 million people living with HTLV-1 infection in resource-poor areas[2] where the impact of HTLV-1 infection has been little studied . | The Human T-Lymphotropic Virus type 1 ( HTLV-1 ) infects up to 20 million people worldwide who predominantly reside in resource-limited areas . The virus is associated with a haematological malignancy ( adult T-cell leukaemia/lymphoma , ATL ) , and inflammatory diseases involving organ systems including the spinal cord , eyes and lungs . Determining the outcomes of infection in most HTLV-1 endemic areas is extremely difficult; however , the virus is highly endemic to central Australia where the Indigenous population has access to sophisticated medical facilities . We prospectively followed a large hospital-based cohort of Indigenous Australian adults that was well characterized with regard to base-line comorbid conditions , HTLV-1 serostatus and HTLV-1 proviral load ( pVL ) . A higher baseline HTLV-1 pVL was strongly associated with an increased risk of airway inflammation ( bronchitis/bronchiolitis and bronchiectasis ) and death , which most often resulted from complications of bronchiectasis . Increased mortality due to an HTLV-1-associated inflammatory condition has not been demonstrated previously . The morbidity and mortality associated with HTLV-1 infection may therefore be substantially higher than has been assumed from an analysis of cohorts of subjects with adult T-cell leukaemia or HTLV-1-associated myelopathy . These findings have important implications for epidemiological research and for determining health care priorities in resource-limited settings . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"death",
"rates",
"blood",
"cells",
"inflammatory",
"diseases",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"respiratory",
"infections",
"pathogens",
"immunology",
"microbiology",
"diet",
... | 2018 | Human T-Lymphotropic Virus type 1c subtype proviral loads, chronic lung disease and survival in a prospective cohort of Indigenous Australians |
INT6/eIF3e is a highly conserved component of the translation initiation complex that interacts with both the 26S proteasome and the COP9 signalosome , two complexes implicated in ubiquitin-mediated protein degradation . The INT6 gene was originally identified as the insertion site of the mouse mammary tumor virus ( MMTV ) , and later shown to be involved in human tumorigenesis . Here we show that depletion of the Drosophila orthologue of INT6 ( Int6 ) results in short mitotic spindles and deformed centromeres and kinetochores with low intra-kinetochore distance . Poleward flux of microtubule subunits during metaphase is reduced , although fluorescence recovery after photobleaching ( FRAP ) demonstrates that microtubules remain dynamic both near the kinetochores and at spindle poles . Mitotic progression is delayed during metaphase due to the activity of the spindle assembly checkpoint ( SAC ) . Interestingly , a deubiquitinated form of the kinesin Klp67A ( a putative orthologue of human Kif18A ) accumulates near the kinetochores in Int6-depleted cells . Consistent with this finding , Klp67A overexpression mimics the Int6 RNAi phenotype . Furthermore , simultaneous depletion of Int6 and Klp67A results in a phenotype identical to RNAi of just Klp67A , which indicates that Klp67A deficiency is epistatic over Int6 deficiency . We propose that Int6-mediated ubiquitination is required to control the activity of Klp67A . In the absence of this control , excess of Klp67A at the kinetochore suppresses microtubule plus-end polymerization , which in turn results in reduced microtubule flux , spindle shortening , and centromere/kinetochore deformation .
The INT6 gene was originally identified as the insertion site of the mouse mammary tumor virus ( MMTV ) [1] . MMTV integration into the INT6 gene causes the production of a C-terminally truncated Int6 protein ( INT6ΔC ) . Ectopic expression of INT6ΔC in mouse mammary glands leads to tumor formation [2] . In addition , INT6ΔC can induce malignant transformation of human tissue culture cells , which produce tumors when injected into immunodeficient mice [2–4] . However , the examination of several breast cancer cell lines did not detect INT6ΔC expression [2 , 5] . Moreover , many human breast cancers are characterized by INT6 deregulation; some tumors show low levels of INT6 [6–9] , while others exhibit an upregulation of the protein [10] . Thus , even if in most cases INT6 acts as a tumor suppressor , it can also have an oncogenic role . INT6 is a highly conserved protein that has been also identified as a subunit ( eIF3e ) of the eukaryotic translation initiation factor eIF3 [11] . INT6/eIF3e interacts with subunits of the COP9 signalosome ( CSN ) and 26S proteasome , which are involved in protein ubiquitination and degradation of polyubiquitinated proteins , respectively [12–14] . Consistent with these biochemical data , studies carried out in diverse systems have implicated INT6 in the regulation of the three complexes . In contrast to other eIF3 subunits , INT6/eIF3e is not essential for global translation and appears to mediate the translation of a limited subset of mRNAs [5 , 15–17] . In both fission yeast and humans , INT6 promotes proteasome assembly via its interaction with the Rpn5 proteasomal subunit , and INT6-depleted cells accumulate polyubiquitinated proteins [18] . There is also evidence that INT6 is functionally related with the CSN complex . For example , the Drosophila orthologue of INT6 ( Int6 ) regulates CSN-mediated cullin neddylation [19] . INT6 has been implicated in mitotic division in budding yeast , Drosophila and human cells . Studies in S . pombe , have shown that Yin6 , the yeast orthologue of INT6 , cooperates with Ras1 to ensure proper chromosome segregation . Defective chromosome segregation was rescued by human INT6 , highlighting the functional conservation of the gene [18 , 20] . RNAi mediated depletion of INT6 in human cells resulted in abnormal spindles , defective chromosome alignment at metaphase , and failure in cytokinesis , a phenotype attributed to reduced activity of the Cdk1 kinase [21] . Int6-depleted cells have been shown to delay during metaphase with short spindles [22] . Here we demonstrate that in Int6 RNAi cells spindle shortening is accompanied by a deformation of both centromeres and kinetochores , a reduction of the intra-kinetochore distance , and a persistent inability to satisfy the spindle checkpoint ( SAC ) . Our results suggest that these phenotypic traits are the consequence of an accumulation at kinetochores of a non-ubiquitinated form of Klp67A , a conserved plus-end-directed kinesin-like protein that suppresses microtubule ( MT ) polymerization at plus ends [23–27] .
Previous studies showed that Int6-depleted S2 cells exhibit short spindles and are delayed in metaphase [22] . To further define the mitotic phenotype elicited by Int6 depletion we re-examined S2 cells treated for 5 days with Int6 dsRNA , a treatment that resulted in a drastic reduction of Int6 ( Fig 1A ) . We chose a 5-day RNAi treatment because at 4 days Int6 was not sufficiently depleted; we only examined dividing cells with a minimal karyotype ( ~ 12 chromosomes;[28] ) . Thus , we limited our observations to cells that were unlikely to carry mitotic defects generated by reduction of Int6 during the previous cell cycles . Staining for both tubulin and DNA revealed that most dividing Int6 RNAi cells are arrested in metaphase and exhibit short and compact spindles ( Fig 1B–1D ) . Notably , approximately 70% of these metaphases displayed a tight chromosome alignment comparable to that observed in live metaphases just before anaphase . Anaphase and telophase figures of Int6-depleted cells were also shorter than their normal counterparts , but did not exhibit gross defects in chromosome segregation ( S1 Fig ) . We also examined cell division in live Int6 RNAi cells that express mCherry-tubulin and histone-GFP . Here again , we limited our observation to cells with a minimal karyotype . We time-lapse recorded mitosis of these RNAi cells starting from prometaphase; they remained in metaphase for much longer times ( up to 3 hours ) than control cells , which entered anaphase within 35 min after metaphase plate formation ( Fig 1E and 1F; S1 and S2 Movies ) . Interestingly , early prometaphase spindles of control and Int6 RNAi cells were similar in length , but approximately 10 minutes before formation of the metaphase plate the spindles of Int6-depleted cells started to shorten , and after 30 minutes spent in metaphase they were 35% shorter than those of untreated controls ( Fig 1H ) . We also video-recorded chromosome movement during anaphase; we found that the chromosomes of control cells ( n = 12 ) and Int6 RNAi cells ( n = 16 ) move at 1 . 10 ± 0 . 09 and 0 . 35 ± 0 . 04 μm/min , respectively ( Fig 1E and 1G ) . Thus , Int6 deficiency slows down chromosome movement during anaphase A . Immunostaining for Cid ( the Drosophila CenpA homolog ) revealed that in metaphases of Int6-depleted cells many Cid signals are abnormally shaped compared to those of control cells . These signals were ellipsoid or cylindrical in shape and had their major axis oriented orthogonally with respect to the longitudinal spindle axis ( Fig 2A ) . To quantify the effect of Int6 depletion on Cid signals we focused on those that were distinct from other signals and measured the ratio between their major and minor axis . For all signals we considered as major axis the one orthogonal to the spindle axis ( see Materials and methods for details ) . We found that in Int6-depleted metaphases these ratios were significantly higher than in controls , suggesting that loss of Int6 leads to a deformation of metaphase centromeres ( Fig 2B ) . We next asked whether the centromere deformation reflected an increase in the Cid amount ( possibly due to its reduced degradation; see below ) . Western blotting showed Int6 RNAi and mock RNAi cells exhibit very similar Cid levels ( Fig 2C ) , suggesting that the centromeres of Int6-depeleted cells are morphologically abnormal and not simply larger than those of control cells . We also co-stained Int6 RNAi metaphases for both Cid and the outer kinetochore component Ndc80 [29] . We found a high degree of coincidence between the two fluorescent signals ( Fig 2D ) . Consistent with this observation , in Int6-depleted metaphases the ratios between the major and minor axes of the Ndc80 signals were significantly higher than in controls ( Fig 2E ) . Thus , in Int6-depleted cells both the centromere chromatin and the outer kinetochore are similarly deformed . To ascertain whether the kinetochore deformation phenotypes observed in Int6 RNAi cells was due to their prolonged arrest in metaphase , we examined cells treated with the proteasome inhibitor MG132 , which is known to block S2 cells in metaphase [30] . Cells were treated with MG132 for 6 hours , fixed and then stained for tubulin , Cid and DNA . As expected , in MG132 treated cultures all dividing cells were blocked in metaphase with well-aligned chromosomes and morphologically normal centromeres ( Fig 2B ) . Thus , the centromere phenotype seen in Int6 RNAi cells cannot be a direct consequence of the metaphase arrest suffered by these cells . We next examined Cid-stained centromere regions in Int6-depleted and control cells incubated for 2 hours with colchicine ( Fig 2F ) . To measure the shape of these regions we used the fit-ellipse function of the ImageJ software , which provides the length of the major and minor axis of the fluorescent signal ( see Materials and methods for details ) . In colchicine-treated Int6-depleted cells , the average axial ratio of Cid signals was higher than in colchicinized controls , but the difference in ratios was reduced compared to that observed in non-colchicinized cells ( compare Fig 2B and 2F ) . These results indicate that the presence of the spindle MTs contributes to centromere/kinetochore deformation in Int6-deficient cells . Although the Int6-deficient metaphases show tightly aligned chromosomes , they exhibit a strong delay in anaphase entry , suggesting that the SAC is not satisfied . In Drosophila , satisfaction of the SAC requires axial stretching of the kinetochores that is manifested as an increase in the intra-kinetochore distance ( intra-KD ) [31]; namely , the distance between the outer corona marked by proteins such as Ndc80 and the inner kinetochore marked by Cid/CenpA ( Fig 3A ) . SAC is satisfied when the intra-KD is elevated , while it remains active when the intra-KD is relatively low . In contrast , the inter-kinetochore distance ( inter-KD; the distance between the Cid/CenpA signals associated with sister chromatids ) does not affect the SAC activity [31] . To assess the inter- and intra-KDs we examined Int6-depleted and control metaphases stained for both Cid and Ndc80; metaphases from colchicine-treated cells or cells treated with the proteasome inhibitor MG132 served as negative and positive controls for intra-KD , respectively [32–34] . In Int6 RNAi metaphases , the inter-KD was significantly higher than the control value but the intra-KD was significantly lower than that seen in control metaphases ( Fig 3B and 3C ) . The relatively high inter-KD indicates that the spindle MTs exert tension on the bioriented sister kinetochores leading to the formation of a compact metaphase plate with tightly aligned chromosomes . However , the low intra-KD of these chromosomes is likely to prevent SAC silencing and anaphase onset [31] . To obtain additional insight into the kinetochore structure of Int6-depleted cells , we performed a transmission electron microscopy ( TEM ) analysis by examining single ultrathin sections ( of approximately 30 nm ) of kinetochores displaying end-on attached MTs ( Fig 4A and 4B ) . In control cells , the mean length of kinetochore plates ( 24 metaphases , 66 kinetochores ) was 279 ± 9 nm , while the mean number of MTs emanating from the kinetochores was 4 . 6 ± 0 . 1 ( Fig 4C and 4D ) . In Int6-depleted cells ( 25 metaphases , 68 kinetochores ) , both the mean kinetochore length ( 433 ± 13 nm ) and MT number ( 7 . 1 ± 0 . 2 ) were significantly higher than in controls ( Fig 4C and 4D ) . Interestingly , the kinetochore length ( 433/279 = 1 . 55 ) and the MT number ( 7 . 1/4 . 6 = 1 . 54 ) ratios between Int6 RNAi and control cells are virtually identical , indicating that the MT capturing ability of deformed kinetochores is the same as that of normal kinetochores . It should be noted that the single ultrathin sections we examined contain only a fraction of the MTs that are normally attached to a Drosophila kinetochore . Analyses of 100-nm-thick serial sections have previously suggested that S2 cell kinetochores are associated with an average of 11 MTs [35] . We examined cross-sections through 5 kinetochores of control cells and found 10–15 end-on attached MTs , consistent with previous results . TEM analysis also revealed a 50% decrease in the mean distance between the edge of chromatin and the outer edge of the kinetochore plate in metaphases of Int6-depleted cells . ( Fig 4E and 4F; 22 kinetochores in 14 cells for controls; 30 kinetochores in 13 cells for Int6 RNAi cells ) . This decrease is consistent with the short intra-kinetochore distance between the inner ( Cid ) and outer ( Ndc80 ) kinetochore components observed by light microscopy ( Fig 3C ) . To obtain further insight into the mechanisms underlying the mitotic phenotype caused by loss of Int6 we performed double RNAi experiments . We first carried out RNAi against Int6 and mad2 , which encodes a component of the SAC machinery that mediates metaphase arrest [36] . Double RNAi cells for mad2 and Int6 displayed a relief from the partial metaphase arrest observed in Int6 RNAi cells , showing an anaphase frequency comparable to that found in cells depleted of Mad2 only ( Fig 5A and 5B ) . These cells also showed metaphase spindles that were significantly longer than those of cell depleted of Int6 only , but still significantly shorter than control spindles ( Fig 5C ) . Finally , while in mad2 RNAi cells the centromere shape was normal , in mad2 Int6 double RNAi cells the centromeres were deformed showing an average axial ratio comparable to that observed in Int6-depleted cells ( Fig 5D; see also Fig 2B; see Materials and methods for the procedure used for Cid signal measurement ) . These findings suggest that the metaphase arrest phenotype caused by Int6 depletion depends on a persistent SAC activity , and also demonstrates that prolongation of mitosis is not required for centromere deformation in Int6-depleted cells . Because Int6-depleted cells exhibit a limited centromere deformation after colchicine treatment , we sought to confirm the role of spindle MTs in the genesis of this phenotype . We thus performed double RNAi against Int6 and Ndc80 , which encodes a protein that mediates MT-kinetochore attachment and is required for proper SAC activity [29 , 37] . Double RNAi cells showed the same phenotype as cells depleted of Ndc80 only; namely , they displayed morphologically normal centromeres/kinetochores , metaphase spindles of regular length , and scattered chromosomes ( Fig 6A–6D ) . Ndc80 RNAi cells and double RNAi cells also showed many cells with elongated spindles associated with chromosomes with unseparated sister chromatids ( Fig 6A and 6B ) . These peculiar cells , show high levels of Cyclin B , suggesting that they are metabolically in metaphase [22] . Similar mitotic figures have been previously observed in Cid ( the Drosophila homologue of CenpA ) -depleted cells and have been named pseudo ana-telophases ( PATs , [22] ) because they often show central spindle-like structures associated with irregular contractile rings . Here , to avoid possible confusion , we designate them as prometaphase-like cells with elongated spindles ( PMLES ) . The observations on mad2 Int6 and Int6 Ndc80 double RNAi cells indicate that the short spindle phenotype depends on both kinetochore-MT attachment and SAC activity . They also suggest that the centromere deformation phenotype depends on kinetochore-MT attachment but not on SAC activity . This conclusion is consistent with the observation that colchicine treated Int6-depleted cells exhibit fewer misshapen centromeres when compared to colchicine-treated controls . It is indeed likely that at the time of colchicine treatment a fraction of Int6-deficient cells was in metaphase and had already experienced kinetochore-MT interaction and undergone centromere deformation . Shortening of the spindle suggests a change in the dynamic of spindle microtubules in Int6-depleted cells . To reveal the nature of this change we measured the rate of microtubule flux and microtubule turnover both within the K-fibers and near the spindle poles . The MT poleward flux is the continuous flow of the MT subunits towards the spindle poles driven by tubulin addition at their plus ends and tubulin disassembly at their minus ends [23 , 38] . Depletion of Int6 resulted in a significant decrease in the poleward velocity of photobleached marks during metaphase: from 0 . 91 ± 0 . 24 μm/min in control cells ( n = 20 ) to 0 . 30 ± 0 . 12 in Int6-depleted cells , ( n = 17 ) ( Fig 7A and 7B ) . The reduction in flux rate upon Int6 RNAi is consistent with the slow movement of chromosomes during anaphase observed in these cells . Because poleward flux involves continuous polymerization of MTs at the kinetochores and balanced depolymerization at the spindle poles , spindle shortening observed in Int6-depleted cells strongly suggests that incorporation of new subunits into the plus ends of kinetochore-attached MTs occurs at a lower rate compared to control . In contrast , all parameters of Fluorescence Recovery After Photobleaching ( FRAP ) within K-fibers near the kinetochores were similar in Int6-depleted and control cells ( Fig 7C and 7D ) . This suggests that detachment and reattachment of K-fiber microtubules to the kinetochore are not affected by Int6 depletion . Control and Int6-depleted cells also showed comparable FRAP parameters when photobleaching was performed near the spindle poles ( S2 Fig ) . Int6 has been implicated in the ubiquitin-mediated protein degradation pathway [14 , 16] and may therefore mediate proteolysis of factors that regulate MT behavior . We reasoned that loss of Int6 could result in an accumulation of a protein that would reduce net growth of the MT plus ends embedded in the kinetochore . A good candidate for this role was the plus end-directed Klp67A kinesin-like protein , which localizes at kinetochores and represses MT plus end growth [23–26] . We thus envisaged that loss of Int6 could result in failure of Klp67A degradation and that the consequent accumulation of this protein could affect MT behavior in an opposite fashion to Klp67A depletion . To test this possibility , we performed immunostaining experiments to determine whether Int6-depleted cells accumulate Klp67A . In control cells , Klp67A was associated with the spindle MTs and accumulated on the kinetochores . In Int6 RNAi cells , Klp67A displayed an accumulation on both the kinetochores and the spindle MTs . Measurements of fluorescence intensity revealed that in Int6-depleted metaphases there is a significant increase in spindle- and kinetochore- associated Klp67A compared to controls ( Fig 8A and 8B ) . A small but significant increase in Klp67A was also detected by Western blotting ( Fig 8C and 8D ) . Given that the mitotic index in S2 cells is 3% , the limited Klp67A increase observed in Western blots is consistent with the hypothesis that Klp67A undergoes ubiquitin/proteasome-mediated degradation mainly during mitosis . We next asked whether overexpression of Klp67A could mimic the phenotype elicited by Int6 depletion . We overexpressed Klp67A-GFP by placing the gene under the control of the Drosophila metallothionein ( Mt ) inducible promoter . After addition of copper sulfate ( 100 μM CuSO4 ) to the culture medium , dividing cells overexpressing Klp67A-GFP were easily recognizable for their fluorescent spindles ( Fig 8E ) . Fixed cells expressing Klp67A-GFP displayed different degrees of fluorescence; to define their mitotic phenotype we examined only cells that were in the top 20% for fluorescence intensity . An analysis of these cells revealed that they have short spindles and are delayed in their progression through metaphase just like Int6-depleted cells ( Fig 8F and 8G ) . In addition , they displayed a significant increase in the long/short axis ratio of the Cid ( CenpA ) signals compared to controls ( Fig 8H ) , but this increase was lower than that observed in Int6-depleted cells ( see Fig 2 ) . RNAi against mad2 in Klp67A-GFP overexpressing cells rescued the metaphase arrest phenotype , led to a partial recovery of spindle length but did not rescue centromere deformation ( Fig 8F–8H ) . Thus , the defects caused by Klp67A-GFP overexpression are similar to those caused by Int6 depletion . To extend the comparison between Int6-depleted and Klp67A-GFP overexpressing cells , we generated a cell line constitutively expressing mCherry-tubulin and carrying Mt-Klp67A-GFP . Here again , we induced Klp67A-GFP expression by addition of 100 μM of CuSO4 and focused on spindles that appeared to be in the top 30% as to green fluorescence . In the cells examined , chromosomes were not marked but metaphases were recognizable because they displayed at dark stripe at the spindle equator , in correspondence of the congressed chromosomes . We analyzed spindle behavior by filming newly formed bipolar spindles from prometaphase to metaphase; we found that they exhibit a considerable shortening as they proceed to and through metaphase , just like those of Int6-depleted cells ( S3 Fig ) . In addition , we measured FRAP of kinetochore-associated mCherry-marked MT bundles in metaphases overexpressing Klp67A-GFP ( green spindles ) ; non-green metaphases from the same cultures were used as controls . We found that Klp67A-GFP overexpression does not alter the fluorescence recovery time compared to cells expressing normal Klp67A levels ( S4 Fig ) . Thus , also the in vivo analysis indicates Klp67A-GFP overexpressing cells and Int6-depleted cells exhibit similarities in their mitotic behavior . These results suggest that the mitotic defects caused by Int6 depletion are largely due to an accumulation of Klp67A at kinetochores . To test this hypothesis we compared Int6 and Klp67A double RNAi cells with cells lacking Klp67A only . We observed identical phenotypes with both cell samples showing normal kinetochores , misaligned metaphase chromosomes and spindles longer than those of mock-treated cells ( Fig 9A–9E ) . In addition , both cells samples did not show anaphases and displayed only PMLES ( Fig 9A–9C ) , suggesting that they are both defective in kinetochore-MT interaction [22] . We also compared Int6 and Klp10A double RNAi cells with cells lacking Klp10A only . We performed this experiment because Klp10A is known to destabilize MT minus ends at the spindle poles [26 , 39] . We found that in Klp10 RNAi cells the spindles are substantially longer that in control cells , consistent with previous results [26 , 39] ( S5 Fig ) . Cells deficient for both Int6 and Klp10 displayed spindles significantly longer than those of Int6-depleted cells but still shorter than control spindles ( S5 Fig ) . These results indicate that Int6 and Klp10A play antagonistic roles during spindle assembly , and suggest that Int6 deficiency affects MT plus ends at kinetochores , consistent with the flux and FRAP results ( see Fig 7 ) . We finally investigated whether the subcellular localization of Int6 correlates with Klp67A localization . We generated an anti-Int6 antibody , which specifically recognized the Int6 protein in Western blots ( Figs 1A and S6 ) . Immunostaining of S2 cells with this antibody revealed that Int6 is uniformly distributed in both interphase and mitotic cells; the staining was strongly reduced in Int6 RNAi cells , demonstrating the antibody specificity ( S6 Fig ) . The diffuse Int6 distribution does not conform to Klp67 localization at kinetochores and spindle MTs . This suggests that the role of Int6 is not restricted to a specific cellular compartment , consistent with its association with different protein complexes involved in diverse functions [12–14] . Given that Int6 interacts with both the proteasome and the signalosome , we asked whether Int6 has a role in protein ubiquitination . Because synchronization of Drosophila cells it is virtually impossible , we carried out our analyses using asynchronous cells populations . We performed an IP analysis using S2 cells expressing Ubiquitin-FLAG ( Ub-FLAG ) treated with either Int6 dsRNA or a mock dsRNA ( control ) . Co-IP performed using anti-FLAG agarose beads showed that precipitates from control cells exhibit a clear Klp67A band that was not detected in Int6-RNAi cells ( Fig 10A–10C ) . We also performed an IP analysis using S2 cells expressing Klp67A-GFP and Ub-FLAG , and control cells expressing the GFP protein and Ub-FLAG . Western blotting analysis of precipitates obtained with anti-FLAG beads showed that Int6 RNAi cells exhibit a substantial reduction in general protein ubiquitination compared to mock-treated cells . In addition , these precipitates displayed a very strong reduction of the ubiquitin-conjugated Klp67A-GFP band compared to precipitates from non-RNAi cells ( Fig 10D and 10E ) . Collectively , these results indicate that Int6 mediates Klp67A ubiquitination in S2 cells .
Our results suggest that loss of Int6 leads to an accumulation of non-ubiquitinated Klp67A near the kinetochores , and that the phenotypic traits seen in Int6-deficient cells could be a consequence of this accumulation . Klp67A belongs to the kinesin 8 family and has putative orthologues in both yeasts ( Kip3 in S . cerevisiae and Klp5/6 in S . pombe ) and humans ( Kif18A ) . Although these kinesins affect MT plus ends growth , they appear to act through different mechanisms . S . cerevisiae Kip3p acts as MT depolymerase and removes tubulin subunits from MTs in a length-dependent manner [40] . The role of mammalian Kif18A is somewhat controversial . Earlier work suggested that Kif18A , like its yeast homologue , has MT depolymerizing activity and preferentially destabilizes long microtubules [41] . However , more recent studies indicated that Kif18A suppresses addition of new tubulin subunits at MT plus ends , without directly destabilizing them [42–45] . Although the molecular mechanisms underlying Klp67A activity are currently unknown there is abundant evidence that this kinesin represses MT plus end growth . It has been reported that RNAi against Klp67A leads to long spindles in S2 cells [23–26] , and that the kinetochore bound pool of Klp67A regulates spindle length [27] . In vivo analyses in S2 cells have shown that Klp67A associates with MTs and accumulates near the kinetochores; while overexpression of Klp67A leads to dose-dependent spindle shortening [26 , 46] . Finally , it has been reported that loss of Klp67A results in a dramatic MT elongation also in Drosophila embryos [47] . We have shown that Int6 depletion does not alter turnover of kinetochore MTs but leads to spindle shortening and reduction of the flux rate . Our results suggest that these phenotypes are dependent on the presence of an excess of Klp67A at the kinetochores . However , envisaging a molecular model that reconciles the three phenotypes is not straightforward . The reduction in the flux rate is consistent with a reduction in MT polymerization at plus ends , which could also lead to spindle shortening . Thus , we propose that Klp67A accumulation at the plus ends of kinetochore MTs blocks their growth without affecting their attachment and detachment rates . It is therefore possible that Klp67A caps MT plus ends and suppresses their growth just as Kif18A in human cells . We also note that the depletion and overexpression phenotypes of Kif18A and Klp67A are rather similar . Depletion of either protein leads to long spindles , and increases anaphase chromosome velocity ( in Kif18A depleted cells ) or the speed MT poleward flux ( in Klp67A depleted cells ) , which is positively correlated to the velocity of anaphase chromosome movement [23] . Overexpression of either protein leads to short spindles , reduced anaphase chromosome velocity , and increases inter-KD [42] . Thus , regardless their mechanism of action , Klp67A and Kif18A appear to have similar effects on MT plus ends during mitosis . We have shown that in Int6-depleted cells the SAC is not satisfied even in metaphases that exhibit tightly aligned chromosomes . The precise reason for this persistent SAC activity is unclear . Studies carried out in Drosophila S2 cells have shown that the SAC cannot be turned off until cells achieve a sufficient intra-KD [31] . In budding yeast , SAC is active when the Mps1 Kinase phosphorylates the KNL1 orthologue Spc105 , and is satisfied when Mps1 and Spc105 are separated by an internal change in kinetochore structure caused by end-on MT attachment [48] . Earlier studies in human cells suggested that SAC satisfaction requires an intra-kinetochore stretch [49] . In contrast , recent work in human cells has shown that a hyperstable kinetochore-MT attachment mediated by a non-phosphorylable form of Hec1 can silence SAC independently of the intra-KD [50 , 51] . In addition , it has been recently shown that taxol-treated human cells with low intra-KD can progress through mitosis if unattached kinetochores are not present [52] . Our results suggest that Int6-dependent Klp67A accumulation at kinetochores locally suppress MT growth . This would conceivably lead to a shortening of the MTs embedded into the kinetochore and low intra-KD . However , we also found that in metaphases of Int6-depleted cells the inter-KD is increased , suggesting that kinetochores are stably attached to MTs and under tension . We note that a low intra-KD accompanied by an elevated inter-KD is not an unprecedented result , as a similar effect has been observed in earlier studies . Examination of cells exposed to different treatments showed that inter-KD and intra-KD are not always correlated . For example , depletion of the condensin I subunit CAP-D2 causes a marked increase in the inter-KD and suppresses intra-KD [49] . It has been thus concluded that the intra-KD is more related to structural rearrangements within the kinetochore than to a mechanical pulling force [31 , 49] . We observed that Int6 depleted cells are strongly delayed in progression through metaphase by the SAC activity but eventually undergo anaphase . Thus , it appears that they can satisfy the SAC even if this takes much longer time than in control cells . In agreement with previous studies in Drosophila [31] , it is possible that Int6-depleted cells are strongly delayed in SAC satisfaction because of their low intra-KD . However , we cannot exclude that SAC satisfaction is prevented by the centromere/kinetochore deformation , which is likely to cause internal changes in kinetochore structure that might affect SAC signaling . The centromere/kinetochore deformation observed in Int6-depleted metaphases is both unexpected and novel . To best of our knowledge this phenotype has never been observed in any cell type . Recent work has described variations in kinetochore morphology during normal mitosis of human cells . However , these variations pertain only to proteins of the outer kinetochore domain and do not apply to metaphase kinetochores with end-on attached MTs , which are morphologically stable [53] . What is then the stimulus that induces centromere/ kinetochore deformation ? Here again , we only speculate that the reduction of kinetochores MT growth caused by Klp67A accumulation could induce morphological changes in centromere/kinetochore structure . Our Klp67A accumulation-based model for the Int6-dependent phenotype raises the question of the mechanism leading to reduced Klp67A ubiquitination and degradation . Int6/eIF3e interacts with both the proteasome and the COP9 signalosome ( CSN ) , which regulates the activity of the Cullin-Ring ubiquitin Ligases ( CRLs ) [54] . CRLs are activated by neddylation of their cullin subunits , a process regulated by the CSN complex [54 , 55] . Previous studies in yeast , humans and plants have shown that depletion of Int6 homologues affects both proteasome and CSN activity [16 , 56] . In contrast , studies in Drosophila revealed that Int6 is an essential gene required for cullin neddylation but not for proteasome function [19] . Indeed , Int6 mutant larvae accumulate high levels of non-neddylated Cul1 , while Int6 overexpression leads to accumulation of neddylated cullins [19] . Consistent with these results , we found that Int6 depletion does not result in accumulation of ubiquitinated proteins but it is instead leading to a general reduction of protein ubiquitination and to an accumulation of non-ubiquitinated Klp67A . In this respect , we would like to mention that ubiquitination is not only a way to target proteins for degradation , but it is also a widespread mechanism for conformational and functional regulation of proteins [57] . Thus , it is possible that the non-ubiquitinated Klp67A that accumulates at kinetochores of Int6-depleted cells has a slightly different activity compared to ubiquinated Klp67A or Klp67A-GFP . As mentioned earlier , our results do not exclude the possibility that Int6-deficient cells accumulate other mitotic proteins in addition of Klp67A . However , most phenotypic traits observed in Int6-depleted cells are also seen in Klp67A overexpressing cells . Thus , if loss of Int6 results in the accumulation of another unknown protein ( s ) on the mitotic apparatus , this protein is unlikely to cause an appreciable mitotic defect . The mitotic phenotype caused by Int6 depletion in Drosophila cells is quite different from the phenotype of INT6-deficient human cells , which exhibit defective spindle morphology , failure to align the chromosomes in metaphase plate and defects in chromosome segregation and cytokinesis [21] . This is not a surprise because , as mentioned previously , loss of INT6 in human cells mainly impairs proteasome activity whereas in Drosophila it primarily affects CSN function . However , it should be noted that Kif18A overexpression in human cells results in short spindles , compact metaphase plates and slow chromosome movement during anaphase [42 , 58 , 59] , a phenotype similar to that elicited by Klp67A overexpression/Int6 downregulation . These findings suggest two hypotheses to explain why Int6-depleted Drosophila and human cells exhibit different mitotic phenotypes . It is possible that human INT6 does not control ubiquitin-mediated Kif18A degradation [60] . Alternatively , one might speculate that INT6-depleted human cells fail to degrade a protein ( s ) whose accumulation masks the phenotype caused by an excess of Kif18A .
S2 cells were cultured at 25°C in Schneider’s medium ( Sigma ) supplemented with 10% fetal bovine serum ( FBS , Gibco ) . dsRNA production and RNAi treatments were carried out according to [22] . dsRNA-treated cells were grown for 5 days at 25°C , and then processed for cytological and biochemical analyses . To depolymerize MTs , cells were treated with 25 μM colchicine ( SIGMA ) for 2 h . Proteasome inhibitor MG132 ( 10 μM; SIGMA ) was added to cell cultures for 6 h . PCR products and dsRNAs were synthesized as described in [22] . Individual Drosophila gene sequences were amplified by PCR from a pool of cDNAs obtained from 5 different libraries: 4 libraries from 0–4 , 4–8 , 8–12 and 12–24 h embryos and an imaginal disc library , all kindly provided by Nicholas H . Brown [61] . The primers used in the PCR reactions were 35 nt long and all contained a 5’ T7 RNA polymerase binding site ( 5’-TAATACGACTCACTATAGGGAGG-3’ ) joined to a gene-specific sequence . The sense and antisense gene-specific sequences were as follows: Int6 , sense CCACCGACATTC , antisense TTGACGATCCAG; mad2 , sense CTCTCGAAGAAC , antisense TCTATCTCGCAG; Ndc80 , sense ATGGCAGCTTGG , antisense CGGTTAACAGGC; Klp67A , sense CTCATCCGGGTC , antisense ACATTCTGTTTC; Klp10A sense ATTGCTGTCCATC , antisense CGATCCTTGTC . The mock dsRNA used as control was obtained from an EGFP vector ( Clontech ) sense sequence: AGCTGTTCACCG , antisense sequence TCACGAACTCCA . Preparations of S2 mitotic cells were carried out according to [22] . For Ndc80 and Klp67A immunostaining , cells were fixed for 10 min in 4% paraformaldheyde , incubated with PBS + 0 . 05% SDS for 30 min and then with 3% BSA in PBS for 30 min . In all the other indirect IF experiments , cells were fixed for 7 min in 3 . 7% formaldheyde and immunostaining was performed as described in [22] using the following antibodies , all diluted in PBS + 10% goat serum: anti-α tubulin monoclonal DM1A ( 1:100; Sigma ) ; rabbit anti-Spd2 ( 1:3500; [62] ) ; chicken anti-Cid ( 1:10000; [63] ) ; rabbit anti-Int6 ( 1:100 ) ; rabbit anti-Ndc80 ( 1:100; a gift of M . Goldberg , Cornell University ) ; rabbit anti-Klp67A ( 1:50 [27]; rabbit anti-cyclin B ( 1:100; [64] ) ; rabbit anti-GFP ( 1:100; Torres Pines Biolabs Inc ) . These primary antibodies were detected by incubation for 1 h with FITC-conjugated anti-mouse ( 1:10 , Jackson Laboratories ) , Cy3-conjugated anti-rabbit ( 1:300 , Life Technologies ) , Cy3-conjugated anti-chicken IgGs ( 1:100 , Jackson Laboratories ) , Rhodamine Red-conjugated anti-mouse ( 1:20 , Jackson Laboratories ) or FITC-conjugated anti-rabbit ( 1:50 Jackson Laboratories ) . Slides were mounted in Vectashield with DAPI ( Vector ) to stain DNA and reduce fluorescence fading . All images were captured using a CoolSnap HQ CCD camera ( Photometrics; Tucson , AZ ) connected to a Zeiss Axioplan fluorescence microscope equipped with an HBO 100 W mercury lamp . To quantify the spindle-associated Klp67-GFP , we stained preparations with both anti-GFP and anti-tubulin antibodies , which were detected by FITC-conjugated anti-rabbit and Rhodamine Red-conjugated anti-mouse , respectively . We measured the GFP and tubulin fluorescence and subtracted the background signal from each measure using the ImageJ software . We then calculated the ratio between the GFP and tubulin fluorescence . The inter- and intra-kinetochore distances were measured using the calipers tool of ImageJ ( NIH ) . For each metaphase , the inter-KD was calculated by measuring the distance between pairs of Cid signals associated with sister chromatids . The intra-KDs were calculated as [ΔNdc80-ΔCid/2] , where ΔNdc80 and ΔCid are the distances between the centers of paired Ndc80 and Cid signals; for each experimental condition we analyzed at least 200 sister kinetochores/centromeres . To assess the spindle length , metaphases were immunostained for tubulin and the centrosome marker Spd-2; we then measured the distance between the centrosomes associated with the opposite spindle poles using the calipers tools of ImageJ . We used two different methods for measuring the extension of the Cid signals; with both methods we considered only Cid signals that were distinct from other signals . In cells where the chromosomes were tightly aligned in metaphase ( mock-treated controls , MG132 treated cells , Int6 RNAi cells , and cells overexpressing Klp67A-GFP ) we directly measured the ratio between the major and minor axis of each Cid signal . We considered as major axis the one parallel or almost parallel to an ideal line orthogonal to the spindle axis ( the ideal line that connects the spindle poles ) bisecting the metaphase plate . Using this criterion , the axes ratio of some signal was lower than 1 . To measure the extension of the signals in cells where the chromosomes were not well aligned in a metaphase plate ( colchicine treated cells , mad2 RNAi cells , mad2 Int6 double RNAi cells , Ndc80 RNAi cells , Ndc80 Int6 double RNAi cells , Klp67A RNAi cells and Klp67A Int6 double RNAi cells ) we used the fit-ellipse function of the ImageJ software , which provided the length of the major and minor axis of each fluorescent signal . In vivo analysis of mitosis was performed on ( i ) mock-treated and Int6 RNAi cells expressing mCherry-tubulin and histone-GFP ( H2B-GFP ) ( a gift from Gohta Goshima ) , and ( ii ) cells expressing mCherry-tubulin and overexpressing Klp67A-GFP ( see below ) . Cells were plated on concanavalin-A ( Con-A ) -treated MatTek dishes . For filming the entire mitotic process , images were taken at 1 , 2 or 4 min intervals . To precisely measure the speed of chromatid movement during anaphase A , images were captured at 30–60 s intervals . 8 fluorescence optical sections were captured at 1 μm Z steps using a calibrated Prior Proscan stepping motor , with an EM-CCD camera ( Cascade II , Photometrics ) connected to a spinning-disk confocal head ( CarvII , Beckton Dickinson ) mounted on an inverted microscope ( Eclipse TE2000S , Nikon ) . Images were acquired using Metamorph software package ( Universal Imaging ) . Movies were made with the Metamorph software; each fluorescence image shown is a maximum-intensity projection of all sections . To calculate the velocity of chromatid-to-pole motion we divided the distance attained by the separating chromosomes sets at the end of anaphase A by the time elapsed from anaphase initiation; each measure was then divided by 2 . The graph reported in Fig 1G has been obtained by averaging the velocities of 12 control and 16 Int6 RNAi cells . All experiments were performed using S2 cells expressing either GFP-tubulin or mCherry-tubulin . FRAP was measured in a rectangular region of interest ( ROI ) positioned across a half-spindle as described in [23] , or within spot adjacent to a kinetochore as described in [65] . Raw intensities were background-corrected , normalized , and fitted in Matlab using the EasyFrap script [66] . Poleward flux rates in metaphase spindles were measured and analyzed according to [23] . A suspension of cells in culture medium was centrifuged at 1000 rpm in 50 ml falcon tubes for 5 min . After removal of the supernatant , the cell pellet was immediately pre-fixed in 2 . 5% glutaraldehyde dissolved in the culture medium; the pellet was then gently resuspended and left in the pre-fixation solution for 15 min and gently shaken . The pre-fixed sample was next transferred to 1 . 5 ml Eppendorf tubes and centrifuged at 1000 rpm for 5 min . After removal of the supernatant , the pellet was fixed in a fresh 2 . 5% glutaraldehyde in 0 . 1 M sodium cacodylate buffer ( pH 7 . 4 ) for 1 h at room temperature . Cells were then washed three times for 5 min each in 0 . 1 M sodium cacodylate buffer and post-fixed for 1 h in 1% water solution of osmium tetroxide containing few crystals of potassium ferricyanide ( K3[Fe ( CN ) 6] ) . After washing with three rounds of milliQ water , samples were incubated overnight at 4°C in 1% aqueous solution of uranyl acetate . On the next day , the cells were washed once with milliQ water and then dehydrated in ethanol series ( 30% , 50% , 70% , 96% for 10 min , and 100% for 20 min ) and acetone ( twice , for 20 min ) , and embedded in Agar 100 Resin ( Agar Scientific , Essex , UK ) . Complete polymerization of samples was conducted by keeping them for three days in the oven at 60°C . Semi-thin sections were obtained with Reichert-Jung ultracut microtome , stained with methylene blue and analyzed with a Zeiss Axioscop 40 light microscope . Ultra-thin sections were made using Leica Ultracut ultra-microtome and stained with Reynolds lead citrate . Sections were examined with JEOL JEM-100SX transmission electron microscope at 60kV . To obtain antibody against Drosophila Int6 , the Int6 sequence encoding aa 1–252 was cloned into pET200 vector ( Invitrogen ) , and the recombinant protein was purified by electro-elution . Immunization was carried out by Agro-Bio ( La Ferté St Aubin , France ) according to standard protocols . The antibodies were affinity purified as described in [67] . To perform co-transfections , cells were suspended in Schneider’s insect medium supplemented with 10% FBS at a concentration of 1×106 cells/ml and plated , 1 ml/well , in a six-well culture dish . Each culture was inoculated with 1 μg of plasmid supplemented with Effectene transfection reagent ( QIAGEN ) according to the manufacturer's instructions . For selection of stably transfected cultures , cells were diluted from 1:5 to 1:10 into the appropriate selective medium 72 h after transfection . For generating stable cell lines we used the following plasmids: H2B-GFP and mCherry- tubulin ( both gifts from Gohta Goshima , Nagoya University , Japan ) , pMT-Klp67A-GFP ( a gift from Ronald D . Vale , UCSF , CA ) , pCoHygro and pCoBlast ( both from Invitrogen ) . pMt-Klp67A-GFP expression was induced with 10 μM CuSO4 for 12 h ( for biochemical experiments ) or 100 μM CuSO4 for 48 h ( for IF experiments ) . For transient expression we used p-AWG ( from DGRC , Indiana University , Bloomington ) and a FLAG-tagged ubiquitin expressing plasmid generated in our laboratory ( the structure of this plasmid can be schematized as follows: pJZ4-Kpn1-6His-Xpress-FLAG-ubiquitin-Xba-pJZ4 ) . For immunoblotting of Drosophila proteins , S2 cells were washed in cold PBS and homogenized in lysis buffer ( 50 mM Hepes KOH pH 7 . 6 , 1 mM MgCl2 , 1mM EGTA , 1% Triton X-100 , 45 mM NaF , 45 mM β-glycerophosphate , 0 . 2 mM Na3VO4 ) in the presence of a cocktail of protease inhibitors ( Roche ) . Cell extracts were pelleted at 15 , 000 g in an Eppendorf centrifuge for 15 min at 4°C and the supernatants were analyzed by Western blotting according to [68] , using the following antibodies , all diluted in TBS-T ( TBS with 0 . 1% Tween 20 ) : rabbit anti-Klp67A ( 1:500; [27] ) ; rabbit anti-Int6 ( 1:1000 ) ; rabbit anti-Giotto ( 1:5000; [69] ) ; anti-α tubulin monoclonal DM1A ( 1:1000 , Sigma ) ; rabbit anti-GFP ( 1:2500; Torrey Pines Biolabs Inc ) ; rabbit anti-cyclin B ( 1:1000; [64] ) ; mouse anti-ubiquitin ( 1:1000; Covance ) ; anti-Flag-HRP-conjugated ( 1:3000 , Invitrogen ) . These primary antibodies ( except the anti-Flag-HRP-conjugated ) were detected using HRP conjugated anti-mouse and anti-rabbit IgGs and the ECL detection kit ( all from GE Healthcare ) . Band intensities were quantified by densitometric analysis with Image Lab software ( Bio-Rad ) . For the in vivo ubiquitination assay , S2 cell were transfected with p6His-Xpress-FLAG-ub , pAWG and pMt-Klp67A-GFP and after 72 h treated with int6 dsRNA as described above . Expression of pMt-Klp67A-GFP was induced by an overnight treatment with 10 μM CuSO4 . Cells were harvested and lysed with lysis buffer ( 50 mM Tris [pH 7 . 5] , 120 mM NaCl , and 0 . 5% NP40 ) containing 1% ( w/v ) sodium dodecyl sulfate ( SDS ) that was preheated to 100°C [70] . Before binding to the anti-Flag beads , NaCl and SDS concentration in the binding buffer were adjusted to 500 mM and 0 . 1% , respectively . After pull-down , the beads were washed with lysis buffer containing 0 . 1% SDS and were used for immunoblot analysis . Blots were imaged with the ChemiDoc MP imager ( Bio-Rad ) ; band intensities were quantified using Image Lab software ( Bio-Rad ) . | INT6 is an evolutionarily conserved gene originally identified as the insertion site of the mouse mammary tumor virus that causes tumors in mice . INT6 is downregulated in many human cancers , suggesting that it acts as tumor suppressor gene . The INT6 protein is involved in several biological processes , including translation and ubiquitin-mediated protein degradation . We performed RNA interference ( RNAi ) against the Drosophila homologue of INT6 ( Int6 ) and analyzed the effects of Int6 depletion on mitotic cell division . We found that loss of Int6 results in short spindles , delayed progression though metaphase and abnormally shaped centromeres/kinetochores . We also found that Int6-depleted cells fail to degrade the kinesin Klp67A . This protein , known to attenuate polymerization of microtubule ( MTs ) plus ends , accumulated at the kinetochores in Int6-depleted cells . We propose that this condition affects MT growth at the kinetochore , which in turn results in centromere/kinetochore deformation and delays satisfaction of the mitotic checkpoint . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"rna",
"interference",
"classical",
"mechanics",
"anaphase",
"chromosome",
"structure",
"and",
"function",
"metaphase",
"centromeres",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"animals",
"animal",
"models",
"drosophila",
"melanog... | 2017 | The Drosophila orthologue of the INT6 onco-protein regulates mitotic microtubule growth and kinetochore structure |
There are limited data describing the functional characteristics of HIV-1 specific antibodies in breast milk ( BM ) and their role in breastfeeding transmission . The ability of BM antibodies to bind HIV-1 envelope , neutralize heterologous and autologous viruses and direct antibody-dependent cell cytotoxicity ( ADCC ) were analyzed in BM and plasma obtained soon after delivery from 10 non-transmitting and 9 transmitting women with high systemic viral loads and plasma neutralizing antibodies ( NAbs ) . Because subtype A is the dominant subtype in this cohort , a subtype A envelope variant that was sensitive to plasma NAbs was used to assess the different antibody activities . We found that NAbs against the subtype A heterologous virus and/or the woman's autologous viruses were rare in IgG and IgA purified from breast milk supernatant ( BMS ) – only 4 of 19 women had any detectable NAb activity against either virus . Detected NAbs were of low potency ( median IC50 value of 10 versus 647 for the corresponding plasma ) and were not associated with infant infection ( p = 0 . 58 ) . The low NAb activity in BMS versus plasma was reflected in binding antibody levels: HIV-1 envelope specific IgG titers were 2 . 2 log10 lower ( compared to 0 . 59 log10 lower for IgA ) in BMS versus plasma . In contrast , antibodies capable of ADCC were common and could be detected in the BMS from all 19 women . BMS envelope-specific IgG titers were associated with both detection of IgG NAbs ( p = 0 . 0001 ) and BMS ADCC activity ( p = 0 . 014 ) . Importantly , BMS ADCC capacity was inversely associated with infant infection risk ( p = 0 . 039 ) . Our findings indicate that BMS has low levels of envelope specific IgG and IgA with limited neutralizing activity . However , this small study of women with high plasma viral loads suggests that breastmilk ADCC activity is a correlate of transmission that may impact infant infection risk .
Breast milk ( BM ) can be a vehicle for transmission of various pathogens , but the risk of infant infection is balanced by the potential clinical benefit of BM , which provides significant passive immunity and protection against many infectious agents [1]–[4] . In the case of HIV-1 , exposure to virus through breastfeeding accounts for almost half of the 30–40% of vertical transmissions that occur in untreated , breastfed infants of HIV-1 positive women [5]–[7] . Replacement feeding , avoidance of breastfeeding and reduced BM exposure by early weaning can significantly reduce BM transmission , however , these interventions have been associated with significant increase in infant morbidity and mortality [8]–[13] . Additionally , HIV-1 infected as well as exposed uninfected infants who do not breast feed have been shown to exhibit stunted growth [14] , [15] . These observations highlight the challenges facing HIV-1 infected women in sub- Saharan Africa where prolonged breastfeeding could lead to HIV-1 transmission but no breast feeding could increase the risk of morbidity and mortality resulting in a diluted benefit of HIV-1 free survival [16]–[18] . Consequently , greater understanding of BM protective factors in HIV-1 infection may open promising new ways to make breastfeeding safe for infants born toHIV-1 infected women . Approximately 15–20% of infants born to all HIV-1+ mothers in chronic infection acquireHIV-1 through BM [6] , [7] , [19] , [20] . This relatively low infection rate despite continued exposure suggests that either BM infectivity is low or that antiviral factors in BM may play a role in modulating transmission and/or acquisition of HIV-1 via the oral mucosa . Indeed , antiviral innate immune factors present in BM such as alpha defensins , bile salt-stimulated lipase , lactoferrin , and mucins have all been associated with modulating the risk of BM transmission [21]–[23] . BM is also composed of both innate and activated adaptive immune cells , presumably derived from other mucosal sites such as the gut associated lymphoid tissue . Indeed , HIV-1 specific CD8 T cells and B cells have been reported in BM [24]–[26] , but to date there have been no published studies that have explored the association between the functional immune responses in BM and risk ofHIV-1 transmission through breastfeeding . Vertical transmission , including BM transmission , is characterized by a transmission bottleneck [27]–[39] . In mother- to-child transmission ( MTCT ) , it has been suggested that this bottleneck is in part a result of selection pressure from Nabs because the viruses that are transmitted tend to be relatively insensitive to neutralization by maternal autologous antibodies ( Abs ) , even in mothers who harbor viruses with a range of neutralization sensitivities[32] , [39] . Consistent with the hypothesis that adaptive immunity plays a role in MTCT , several studies comparing levels of maternal plasma neutralizing antibody ( NAb ) titers reported that transmitting ( T ) mothers have lower levels of NAb in plasma compared to non-transmitting ( NT ) mothers [27] , [32] , [36]–[42] suggesting that maternal NAb may contribute to protection of the infant . However , the results of these studies are not consistent , particularly with respect to a role for NAb in protection by different routes of transmission [43]–[45] . Moreover , a recent study of passive Absin 100 HIV-1 exposed infants did not find evidence for a protective effect of broadly NAb on infant infection [46] . Until recently , most studies of BM HIV-1 Abs focused primarily on determining the association between the levels or presence of binding Abs to envelope ( env ) proteins and transmission . Several studies that have focused on BM IgG and IgA have showed no association between levels of these antibodies and transmission [47] , [48] . Notably , infant infection status in these early studies was determined by serology and/or clinical manifestation of AIDS , a situation that could result in misclassification of infant infection status . A more recent study that determined infant infection by DNA PCR showed increased levels of BM IgA in T compared to NT women suggesting that , rather than providing protection , BM HIV-1 env specific soluable IgA , is associated with increased risk of transmission [49] . However , all these studies used subtype B env proteins , in some cases from lab adapted viruses to detect HIV-1 binding Abs despite being conducted in sub-Saharan Africa where such variants are not typical of transmitted strains of HIV-1 [49] , [50] . Taken together , the results from BM binding studies have not provided clear evidence of a role of BM Abs in vertical transmission . BM Abs could provide benefit by directly neutralizing the virus within the milk or by non-neutralizing mechanisms such as antibody dependent cellular cytotoxicity ( ADCC ) that target infected cells . This could result in reduced levels of infectious cell-free virus and BM infected cells , which are both correlates of BM transmission [51]–[54] . The potential of Abs in BMto neutralize HIV-1 and/or mediate ADCC has only very recently been examined , and in this study of ARV-exposed , subtype C-infected women in Malawi , NAbs were detected in about half of the BM samples while ADCC activity was present in all BM samples obtained at 1 month after delivery [55] . There have been no studies to-date looking at BMS samples obtained from untreated T and NT women , particularly in colostrum and early milk , which is relevant given that virus levels are highest in colostrum [51]and the majority of BM transmissions occur early in life [6] , [20] . There has also been no study looking at how these BM Abs function in relation to MTCT . We evaluated neutralizing , binding and ADCC activity in BMS or BMS-derived IgG and IgA and matched plasma from antiretroviral ( ARV ) naïve T and NT mothers with high plasma viral loads and systemic NAbs . Our data shows that BM Nabs are rare and their levels are significantly lower than in plasma . However , we report a high frequency of ADCC activity in BMS that was significantly higher in NT women compared to T women . These data suggest that BMADCC mediating Abs but not Nabs may play a role in modulating HIV-1 transmission .
Women enrolled in a randomized clinical trial comparing breastfeeding to formula feeding in Nairobi Kenya provided BMS samples used in this study [6] . Subjects received coded identification numbers at the clinic and therefore BMS samples were anonymous to laboratory personnel . The ethical review committees of the University of Nairobi , the University of Washington and the Fred Hutchinson Cancer Research Center approved this study and the Kenyan ministry of health gave permission for the original study to be conducted . The methods for enrollment , counseling and follow up have been described elsewhere [6] , [51] . Briefly , HIV-1 positive women were enrolled at 32 weeks gestation and blood samples were taken for viral load and CD4 count testing . Maternal blood , breast milk samples , and infant blood samples were collected within the first week post-delivery , at 6 weeks , 14 weeks , 6 months and quarterly thereafter until 2 years . Infant HIV-1 status was determined using DNA PCR [56] . Breast milk samples were centrifuged to remove the lipid layer and the supernatant was stored at −70°C before being shipped either on dry ice or in liquid nitrogen to Seattle , Washington for long term storage at −70°C until use . Plasma and BM viral loads were determined using the Gen-Probe HIV-1 RNA assay ( Gen-Probe , La Jolla , Calf ) [51] , [57] . Breastmilk samples used in this study were chosen as the first available breastmilk sample after delivery for each woman and the reported breastmilk viral loads are contemporaneous . BMS IgG was purified using NAb Protein G spin columns ( Pierce , Biotech , Rockford , IL ) , with minimal changes to the manufactures instructions . Briefly , 250 ul of heat-inactivated BMS was added to 250 ul of binding buffer and the mixture was added to a protein G column followed by incubation at room temperature ( RT ) with end over end mixing for 30 min . Thereafter , the column was centrifuged to obtain the IgG flow through ( IgG stepFT ) which was saved for subsequent IgA purification . The column with bound Ab was washed 3 times with 400 ul of binding buffer . Bound Ab was eluted with 1 ml of elution buffer ( pH 2 . 8 ) and the eluate was neutralized by adding 100 ul of 1 M Tris . HCl ( pH 8 . 5 ) . Thus , the final purified IgG Ab was diluted 4-fold relative to the original BMS . The final eluted IgG and IgA was retained at a 1∶4 dilution of the original BMS and this was used undiluted in further neutralization assays . Coomassie blue staining ( Simply Blue , Invitrogen ) and ELISAs using Human IgG ELISA kit ( E-80G ) and human IgA ELISA kit ( E-80A ) ( Immunology Consultants laboratory , Newberg , OR ) were used to confirm the purity of Ab fractions . BMS IgA was purified from the IgG step FT using the method outlined by Hirbod et . al with some modifications [58] . Spin columns ( Thermo ) were packed with 400 ul of immobilized jacalin ( Pierce biotech , Rockford , IL ) and washed 3 times with 400 ul of PBS to equilibrate . The column was then loaded with 500 ul of the IgG step FT and incubated on an end over end roller for 2 hours at RT . After incubation , the column was centrifuged and a final flow through ( FT- fraction lacking IgG and IgA ) was collected and stored for analysis . The column was washed 3 times with PBS followed by a 3-hour incubation with 500 ul of 1 M Melibiose to elute bound IgA . The column was further washed with another 500 ul of elution buffer to maximize recovery and bring the final dilution of purified IgA fraction to 1∶4 relative to the original BMS , similar to the IgG fraction . As before , coomassie staining and ELISA were used to confirm the purity of Ab fractions . The subtype A HIV-1 envQ461 . d1 was cloned directly from peripheral blood mononuclear cells ( PBMCs ) of a recently infected Kenyan woman as described previously [59] . Autologous PBMC and BM cell derived clones have either been previously described or were obtained using the same protocol [39] , in some cases with modification of primers to allow amplification of the HIV-1 variant in that particular sample ( primers are available upon request ) . Plasmid DNA encoding the env of interest and a plasmid encoding an env-deficient HIV-1 subtype A proviral DNA , Q23Δenv [60] , were co-transfected into 293T cells at a 1∶2 molar ratio to generate pseudotyped viral particles as described [39] , [61] . Virus was harvested 48 hrs post-transfection and the infectivity was determined by single round infection of TZM-bl cells as described [39] . Pseudoviruses were also generated using Q23Δenv and simian immunodeficiency virus clone 8 ( SIV ) [62]oramphotropic murine leukimia virus ( MuLV ) envelope clones [63] . Neutralization was assessed by determining infection of a reporter cell line , TZM-bl , as previously described [39] . Briefly , 500 infectious particles were incubated with 2-fold serial dilutions of heat inactivated plasma or BMS , purified BMSIgG or IgA fraction , FT fraction or media only in a total volume of 50 ul at 37°C for 1 hour . TZM-bl cells in 100 ul of growth medium containing 30 ug/ml of diethylaminoethyl-dextran were then added . After 48 hours , neutralization was determined by measuring β-galactosidase activity present in the TZM-bl cell lysate . For each virus/Ab combination , at least two independent experiments were performed . Each experiment was performed intriplicate for plasma and BMS or duplicate for purified BMSAb fractions . Median inhibitory concentrations ( IC50s ) were defined as the reciprocal dilution of plasma , BMS or purified antibody that resulted in 50% inhibition , calculated by interpolation of the linear portion of the neutralization curve on the log2 scale as previously described [39] , [61] . Plasma and BMS samples were tested at 1∶100 and 1∶20 dilution respectively , while purified BMSAb fractions were tested at 1∶8 dilution ( a 2-fold dilution of the recovered purified fractions that were diluted 4 fold during processing ) . For the purposes of analysis , in cases in which the IC50s were less than the lowest dilutions tested , the midpoint value between the lowest dilution and zero was assigned . IC50s from replicate experiments were averaged by the geometric mean . Here IC50s indicate the geometric mean IC50 estimates [64] . Human IgG ELISA kit ( E-80G ) and human IgA ELISA kits ( E-80A ) ( Immunology Consultants laboratory , Newberg , OR ) were used to determine the levels of total IgG and IgA in un-purified BMS and plasma samples according to the manufacturer's instructions . HIV-1env specific ELISAs were performed using the protocol outlined by Sather et . al with minimal modifications [65] . Briefly , Immulon 2HB ELISA plates were coated with 25 ng/well of a HIV-1 subtype A Q461 . d1 soluble trimeric gp140 protein purified as described in [66] in 0 . 1 M NaHCO3 , pH 9 . 4 overnight at room temperature . Plates were blocked in phosphate buffered saline ( PBS ) , supplemented with 10% dry milk and 0 . 3% Tween-20 for 1 hr at 37°C . Unpurified BMS and plasma samples were diluted in 10% dry milk , 0 . 03% Tween in PBS . For detection of HIV-1env specific IgG and IgA , BMS samples were diluted at 1∶100 and were titrated 2-fold up to a maximum dilution of 12 , 800 . In cases where an end point titer could not be determined at this dilution , samples were diluted further up to a final dilution of 104 , 200 . For HIV-1 env specific plasma IgG , samples were diluted at 1∶100 , 000 followed by a 2-fold titration up to a maximum dilution of 12 , 800 , 000 while for IgA samples were initially diluted 1∶200 followed by a 2-fold dilution up to 25 , 600 . Samples were loaded in duplicate wells and incubated for 1 hr at 37°C . Plates were washed in a plate washer and bound IgG Ab was detected at 37°C for 1 hr with goat anti-human IgG- horseradish peroxidase ( HRP ) ( Bio-Rad , Hercules , CA ) diluted 1∶3000 while IgA was detected by goat anti human IgA HRP ( Invivogen , San Diego , CA ) diluted 1∶4000 . Plates were developed with 50 ul of 1-Step Ultra TMB-ELISA solution ( Pierce Biotech , Rockford . IL ) and stopped with 50 ul 1 N H2SO4 . Absorption at 450 nm was read on an EL808 Ultra Microplate Reader ( Bio-TEK Instruments . inc ) . In this study , end point titer ( EPT ) was defined as the BMS or plasma reciprocal dilution at which the average OD value was greater than or equal to two times the average OD value of background . The ability of BMS and their matched plasma to mediate ADCC activity was determined as described by Gomez-Roman et . al with a few modifications [67] . Briefly , CEM . NKr cells , a natural killer resistant cell line ( AIDS Research and Reference Reagent Program , NIAID , NIH ) were double stained with a membrane dye , PKH-26 ( Sigma , St . Louis , MO , USA ) and a viability dye , carboxyfluorescein diacetate , succinimidyl ester ( CFSE ) ( Molecular Probes , Eugene , OR , USA ) as recommended by the manufactures . After staining , 1×105 cells were coated for 1 hr at RT with 1 . 5 ug HIV subtype Agp120 protein obtained from an infant in the Nairobicohortat 6 weeks post-infection ( BL035 ) [39] . Coated cells were then washed once and resuspended in 1 ml of RPMI with 10%FBS . Five thousand coated or uncoated CEM . NKr cells were added to the appropriate duplicate wells containing 100 ul of 1∶100 or 1∶1000 heat inactivated BMS or plasma respectively . Similar experiments were performed using media only or HIV IgG ( NIH AIDS Research , Germantown , MD , USA ) as negative and positive controls , respectively . The antibody-target cell mixture was incubated at RT for 10 min to allow the antibody to interact with the antigen on the surface of target cells . Following incubation , 50 ul of effector cells ( HIV negative donor PBMCs ) were added to the mixture at an effector to target cell ( E/T ) ratio of 50∶1 and incubated for 4 hours at 37°C . For all 19 BMS and plasma samples , PBMCs from the same donor were used in parallel assays . Cells were then washed and fixed in 150 ul of 1% paraformaldehyde-PBS and stored at 4°C overnight . Fixed cells were analyzed within 24 hours of the ADCC assay using a BD LSRII instrument ( Becton Dickinson , San Jose , CA , USA ) . Flow cytometry data was analyzed using Flojo version 9 . 4 . 6 ( Tree Star Inc , Ashland , OR , USA ) . ADCC percent killing was defined as the percentage of membrane labeled cells ( PKH-26+ ) that had lost their viability dye ( CFSE− ) after subtracting two times the level of killing in the media only wells ( background ) , as described in ( 67 ) . Odds ratios ( OR ) for assessment of associations between detection of HIV-1 specific and non-specific activity in BMS and transmission were estimated by Fisher's Exact Test . IC50sfor HIV positive and HIV negative controls were compared by one-sided t-test on the log2 scale . All comparisons of Ab total concentrations and HIV-1 env specific titers were based on paired t-tests on the log10 scale , noting that differences on the log scale were approximately normally distributed , and corresponding multivariate adjustments were by linear regression . HIV-1 specific titers among those with detected virus neutralization by BMS IgG and IgA were each compared to titers among those with undetected neutralization using Welch's t-test on the log10 scale . All correlations were measured by Pearson's product moment correlation coefficient ( PPMCC ) , denoted r , with p-values based on the Student's t approximation for the distribution of the corresponding standardized test statistic . The relationship between maternal clinical correlates and BMS Ab neutralization , HIV-env specific binding titers and ADCC activity were each individually assessed by Welch's t-test with corresponding adjusted estimates by linear regression . Statistical analysis was performed using R 2 . 13 ISBN 3-900051-07-0 and STATA version 11 edition , ( College Station , TX ) .
The goal of this study was to determine the presence and functional capacity of BM HIV-specific antibodies and to determine if they impact MTCT . Therefore , we selected women who had high plasma viral loads ( greater than the cohort median of 4 . 6 log10 ) and thus were at increased risk of transmission . Among these women , we identified those who exhibited potent plasma NAb responses ( Majiwa and Overbaugh , unpublished data ) to maximize the chances of detecting BM NAbs . From this subset of women , we selected those that breast-fed for greater than 3 months to capture cases of BM HIV exposure to the infant . Women whose infants were HIV-1 positive before 6 weeks of life were excluded to ensure that transmission was as a result of BM and not late in-utero , or intra-partum exposure . An additional criteria was that women had available BMS samples collected at less than14 weeks after delivery because this early period is the window within which the majority of BM transmissions occur [6] and protein concentrations are highest [68] , [69] . Nineteen women with a median CD4 count of 360 cells/uL met these criteria . The median plasma and BM viral loads were5 . 22 and 2 . 44 log10 respectively , an ∼2-log difference that was also observed in the larger cohort [51] . Nine of these women transmitted HIV-1 to their infants via BM at various time-points postpartum ( Table 1 ) . The ability of heat inactivated BMS to neutralize virus bearing a highly sensitive env variant isolated from a Kenyan woman soon after her infection was determined . This heterologous HIV-1 subtype A env variant , Q461 . d1 , was chosen because >90% of plasma from individuals in the region showed detectable neutralization of this virus at a 1∶100 plasma dilution [70] . The results with plasma from 4 representative women are shown in Figure 1A . All4 plasma samples neutralized Q461 . d1 with IC50 values of ∼500 or greater . Importantly , 50% inhibitory activity was not achieved when testing plasma samples against SIV suggesting that the neutralization response was specific to HIV-1 . Overall , virtually all19 plasmas displayed potent HIV-1 specific neutralization , with IC50s ranging from 185 to 3144 ( Table 1 ) . We could not detect HIV-1 neutralization in any of the BMS at a similar starting dilution as plasma ( 1∶100 data not shown ) . At a very low starting dilution ( 1∶4 ) there was substantial non-specific inhibition of SIV and MuLV and preliminary assays suggested potential cytotoxic effect of more concentrated BMS , as reported previously [71] . BMS was therefore tested at a starting dilution of 1∶20 , hence 5× more concentrated compared to plasma . Results from BMS of 4 representative women against Q461 . d1 and SIV are shown in Figure 1B . While a low level of inhibition of HIV-1 was observed with some BMS such as MJ776 and MP199 , there was little difference in the magnitude of BMS neutralization of Q461 . d1 and SIV in all 4 cases . Among all 19 women , 9 BMSs - 6 from T and 3 from NT women - showed HIV-1 inhibition with IC50 values ranging from 21–85; there was no detectable inhibition by BMS from 3 T and 7 NT women . BMS from the majority of women also inhibited SIV and MuLV pseudoviruses , with IC50 values ranging from 20–95 ( Table 1 ) . A paired comparison of BMS HIV-1 IC50s with the geometric mean of IC50s for corresponding negative control viruses ( SIV and MuLV ) showed that HIV-1 IC50s were not statistically greater than those of the negative controls ( p = 0 . 44 ) . This observation suggested that the majority of inhibition we observed with BMS was likely not due to HIV-1 specific Abs . The presence of a non-specific inhibitor of HIV-1 in BMS could nonetheless be relevant to transmission risk . We thus examined the association between detection of non-specific activity and transmission and found that this relationship was not statistically significant ( OR = 4 . 77; 95% CI: 0 . 51 , 71 . 53; p = 0 . 17 ) . To determine what portion of the non-specific inhibition observed with unfractionated BMS was due to Abs versus other factors , we separately purified IgG and IgA Abs from BMS for use in the neutralization assays . Bands of the expected sizes for IgG and IgA were observed in the respective purified fractions by coomassie staining and cross contamination between Ab isotype fractions by total Ig ELISA was below detection ( data not shown ) . Purified Ab fractions were tested at a starting dilution of 1∶8 , which translated to a dilution 2 . 5 times higher than the most concentrated BMS we tested ( 1∶20 dilution ) . Using the purified BMS IgG fractions , neutralization of greater than 50%was detected in only 2 ( subjects MJ776 and MP199 ) of 19 purified BMS IgG tested , with IC50s of 9 . 4 and 9 . 9 respectively . ( These two examples are shown in figure 2A and a summary of the 19 in Table S1 ) . Of these women MJ776 transmitted HIV-1 to the infant while MP199 did not . In contrast , there was no detection of neutralization by purified BMS IgA fractions tested ( Results from 4 representative women are shown in figure 2B and a summary of the 19 in Table S1 ) . Importantly , purified BMS IgG and IgA fractions did not inhibit viruses pseudotyped with SIV env including the two BMS IgG fractions from subjects MJ776 and MP199 , which had detectable neutralization of virus pseudotyped with Q461 . d1env ( Figure 2A , B , and Table S1 ) . The FT fraction , which contained undetectable levels of BM IgG and IgA both by ELISA and coomassie staining , ( data not shown ) retained the non-specific activity displayed by BMS ( Table S1 ) . To ensure that we were not missing NAb responses by using a heterologous virus , we examined the ability of BMSAb fractions to neutralize autologous virus in a subset of the 19 women . BMS IgG and IgA Ab fractions and FT from a total of 8 women were each tested against 2 pseudoviruses bearing autologous env variants from blood [39] . Of the 8 women , 2 women both NTs , showed low potency neutralization of the blood-derived autologous virus to one of the two viruses tested . MM471 displayed low neutralization potency with anIC50 of 15against one of her autologous viruses when using IgG but not the IgA fraction ( representative experiment is shown in figure 3A ) . In contrast , MA411 displayed low neutralization potency with an IC50 of 9 against one of the autologous virus with IgA but not with IgG fractions ( a representative experiment is shown in figure 3B ) . BMS IgG and IgA fractions from the remaining six women , all Ts did not neutralize their respective autologous viruses above 50% . Autologous viruses for MJ776 and MP199 were not available for testing The ability of plasma and BMS purified Ab to neutralize variants obtained from BM was also determined for two subjects MF535 ( T ) and ML055 ( NT ) . Autologous plasma from MF535 and ML055 diluted at 1∶100 neutralized the respective BM viruses withIC50s of 152 and 718 , respectively . In contrast , there was no detectable neutralization by BMAb fractions against these autologous BM viruses ( data not shown ) . To determine if low NAbs in BMS reflected lower total BM Ab levels , we measured the levels of total and HIV-1envspecific IgG and IgA Abs in BMS and compared them to plasma ( Figure 4 ) . The levels of total BMS IgG were 0 . 88 log10 lower than BMS IgA ( p<0 . 0001 ) ( Figure 4A , black symbols ) . This is in contrast to plasma , where the IgG levels were found to be 1 . 02 log10 higher than IgA ( p<0 . 0001 ) ( Figure 4A , grey symbols ) . There was a pronounced difference between the magnitude of total IgG in BMS and plasma with BMS total IgG being2 . 25 log10 lower than plasma IgG ( p<0 . 0001 ) . In contrast , the total IgA levels in plasma were only slightly higher than in BMS , with a modest 0 . 39log10 difference between BMS and plasma ( p = 0 . 004 ) . We found statistically significant correlation between total BMS IgG and plasma IgG ( r = 0 . 67; p = 0 . 0034 ) while the levels of BMS total IgA correlated with total plasma IgA ( r = 0 . 78; p = 0 . 0003 ) . There was no significant correlation between BMS total IgG and BMS total IgA ( r = 0 . 39; p = 0 . 10 ) ( Table S2 ) . Next , we determined HIV-1 env specific IgG and IgA titers in unfractionated BMS and plasma against soluble gp140 protein derived from the subtype A variant , Q461 . d1 , that was used for the neutralization studies ( Figure 4B ) . HIV-1 env specific IgG titers were obtained in 100% of BMS and plasma samples . In contrast , HIV-1 env specific IgA titers were obtained in 50% of BMS and 90% of plasma samples; the rest were below the cut off value for EPT as defined in this study . BMS HIV-1 env specific IgG titers were 1 . 96 log10 higher compared to env specific IgA ( p<0 . 0001 ) ( Figure 4B , black symbols ) . Similarly , HIV-1 env specific IgG titers in plasma were higher by 3 . 63 log10 when compared to the env specific IgA titers ( p<0 . 0001 ) ( Figure 4B , grey symbols ) . Overall , similar to what we found for total IgG levels , BMS HIV-1 env specific responses were 2 . 22 log10 lower compared to that in plasma ( p<0 . 0001 ) ( Figure 4B ) . For HIV-1 env specific IgA , the log10 difference between BMS and plasma was 0 . 59 ( p = 0 . 0004 ) ( Figure 4B ) . BMS HIV-1 env-specific IgG titers were correlated with plasma HIV-1 env specific IgG titers ( r = 0 . 81; p<0 . 0001 ) and BMS total IgG ( r = 0 . 76; p = 0 . 0003 ) . There was no statistically significant correlation between BMS HIV-1 env specific IgG titers and BMS HIV-1 env specific IgA ( Table S2 ) . Similar to BMS HIV-1 env specific IgG titers and BMS total IgG , BMS HIV-1 env specific IgA titers and BMS total IgA levels were also positively correlated ( r = 0 . 69; p = 0 . 015 ) ( Table S2 . ) We examined the relationship between the levels of HIV-1 env specific titers in BMS and detection of neutralizing activity . The three women with IgG neutralizing activity had a log10 IgG titer of 4 . 41 as compared to a mean of 3 . 83 among non-IgG-neutralizers ( p = 0 . 0001 ) . The one woman with IgA NAbs also had the highest IgA env specific titer , which was1 . 10log10 greater than the group median . ( Figure . S1 ) . We determined the capacity of BMS binding antibodies and their matched plasma to mediate ADCC . The appropriate BMS and plasma dilution for the ADCC assay was determined by testing serial 10-fold dilutions of 4 representative BMS and plasma in the ADCC assay . The dilution that permitted detection of HIV-specific ADCC activity above background levels , but did not yield inhibition of ADCC activity that can occur with more concentrated samples [72] was chosen for testing ( 1∶100 for BMS and 1∶1000 for plasma ) . Using a single dilution also allowed us to test all 19 BMS and plasma samples with effector cells obtained from a single PBMC donor , which is critical for avoiding bias due to differences in effector cell activity observed from donor to donor . Overall , ADCC activity was detected in all BMS and plasma samples tested ( Figures 5 A and B ) . BMS ADCC mediated killing ranged from 1–27% ( median , 15% ) while that of plasma ranged from 16–36% ( median , 24% ) . BMS ADCC activity was correlated with gp140 env specific IgG titers ( r = 0 . 56 , p = 0 . 014 ) ( Figure 6 ) . A log10 increase in gp140 titers was associated with an absolute increase of 9 . 3 in % ADCC mediated killing by BMS ( 95% CI: 2 . 18 , 16 . 41; p = 0 . 013 ) . The relationship between maternal clinical correlates and BMS Ab neutralization , HIV-env specific binding titers and ADCC activity were each individually assessed . There was no statistically significant association between antibody titers and any of the clinical parameters examined . ( Table S3 ) . There was no statistically significant association between detection of NAbs and infant infection ( OR = 0 . 31; 95% CI: 0 . 0050 , 4 . 94; p = 0 . 58 ) . We observed a trend for statistical significance between infant infection and reduced BMSgp140 HIV-1 env specific IgG titers but not plasma titers ( estimated mean log10 difference 0 . 35 95% CI: −0 . 07 , 0 . 77; p = 0 . 098 ) in a univariate analysis ( Figure 7A ) . This association was in similar direction after controlling for plasma viral load ( p = 0 . 038 ) . Importantly , NT women were more likely to have higher BM ADCC activity compared to T women ( estimated mean % killing difference 6 . 89; 95% CI: 0 . 41 , 13 . 37; p = 0 . 039 ) ( Figure 7B ) . This relationship remained significant in a multivariate analysis controlling for plasma viral load ( p = 0 . 011 ) and both plasma and BM viral load ( P = 0 . 012 ) . There was no association between BM RNA viral load and BM ADCC activity ( p = 0 . 520 ) in these 19 women . There was also no significant difference between plasma ADCC in T and NT women ( Figure 7B ) .
The potential of HIV-1 specific Absin BM to inhibit HIV-1 or impact transmission risk has not been well defined . Despite the fact that the levels of both IgG and IgA were low in BM compared to plasma , we observed a trend for inverse correlation between the levels of HIV-1 specific IgG and risk of infant infection in the 19 women examined here . The effect of these antibodies did not appear to be through neutralization , as only 4 of 19 women had any detectable neutralizing IgG or IgA Abs and there was no correlation between detection of NAb and risk of infant infection . Rather , the important functional activity of these antibodies was linked to ADCC activity , as there was a statistically significant inverse correlation between the levels of ADCC activity and risk of infant infection . These data suggests that antibodies capable of mediating ADCC may be one factor that impacts the risk of BM HIV-1 transmission . We found that BM HIV-1 env-specific IgG titers were significantly higher than those of IgA but significantly lower when compared to IgG from matched plasma samples . A reduced IgA response at mucosal sites in HIV-1 infection is contrary to what is observed with mucosal responses to other pathogens but consistent with previous reports of a low HIV-1 specific binding IgA response in favor of IgG at various mucosal sites [73]–[76] . In general , low mucosal BM IgA might reflect an ability of HIV-1 to impair local immune responses as a means of evading the humoral immune system at the mucosal site . However , the observation that BM HIV-1 env specific IgG titers were correlated with total plasma IgG levels suggests that some of the BM IgG may originate from systemic circulation , a process that could help fight infection at the mucosal site . Despite low HIV-specific antibody levels in BMS compared to plasma , antibodies capable of ADCC were detected in all BMS samples . We found that the capacity to mediate ADCC was associated with the levels of HIV-1 env specific IgG titers , which is in agreement with data from previous studies [55] , [77]–[79] . This is perhaps not surprising given that envelope binding is a required step for ADCC activity measured in the assay used here . Using purified BMS antibodies from a subset of these women , we further confirmed that ADCC activity in BM was exclusively mediated by IgG ( data not shown ) . Thus , IgG mediated ADCC can be detected in unfractionated breastmilk , which includes IgA and other factors , as well as with purified antibody . ADCC titers have previously been shown to be generally higher compared to NAbs titers in the same individual possibly due to the specificity required to overcome the constraints posed by env protein in a bid to escape neutralization and also the fact that virus neutralization requires that all of the functional trimers be occupied by at least one antibody [80] , [81] . Thus it may be possible to elicit high levels of antibodies capable of ADCC using an HIV-specific immunogen even in cases where neutralizing responses are limited . BMS ADCC activity was significantly greater in NT compared to T women , suggesting a possible role in impacting infant infection . The mechanism by which BM ADCC might reduce transmission remains to be determined . ADCC would be expected to lead to effective clearance of infected cells . Given that the levels of HIV-infected cells in BM are correlated with transmission risk [52] , it is plausible that HIV-specific ADCC responses within BM may act through reducing cell-associated viral transmission . Other studies have implicated antibodies capable of ADCC in providing protection from infection and/or controlling an established infection . Several studies have shown that de novo ADCC responses to HIV and SIV infection are correlated with better viral control in chronic infection and/or clinical outcome . [77] , [78] , [82]–[85] . Vaccine-induced ADCC responses have also been correlated with reduced viral loads following SIV challenge [78] , [79] , [86]–[88] , supporting a potential role of Fc-mediated antibody responses in blunting a new infection in SIV-infected macaques . A study by Forthal et al . also provided evidence that antibody-dependent cell-mediated virus inhibition , which is a measure of ADCC in combination with other antiviral activities , was correlated with infection rate in the Vax004 vaccine trial , although ADCC alone was not directly examined in this study [89] . In addition , studies of passive immunization using HIV monoclonal antibodies in macaques suggest that FcγR binding is required for optimal protective efficacy [90] . These findings support a potential role for antibodies that act through ADCC in providing protection from infection in the non-human primate model . The current study is the first that reports an association between HIV-specific ADCC activity and risk of HIV infection in humans . This is the first study to examine BMS HIV-1 specific IgG and NAbs in relation to transmission risk using a relevant HIV-1 env representing recently transmitted virus from the dominant subtype in the population . This may explain our ability to detect a trend in association between binding antibodies and transmission , which was not seen in prior studies using other env proteins less representative of viruses in the study population to measure binding [47] , [48] . We used the same highly neutralization sensitive ( tier 1B ) subtype A HIV-1 env representing the dominant subtype in the population under study to optimize our chances of detecting NAbs in BMS . Importantly , plasma from all subjects had a potent NAb response against this virus , indicating that all subjects had generated NAbs capable of specifically recognizing this test virus . Only 4 BMS had Abs that could neutralize >50% of either heterologous or autologous blood-derived viruses and the presence of HIV-1 specific NAbs was not associated with infant infection . The neutralizing activity was observed in women with higher levels of total IgG Abs in BMS . Therefore , it is possible that generally low IgG and IgA titers in BM might explain the limited neutralization capacity displayed by BM Abs . The results of our study , showing low levels of HIV-1 env specific NAbs in BMS , are consistent with another recent study of BM HIV-1 NAbs [55] . In this study of a NVP-treated , clade C infected cohort , the levels of NAbs and HIV-1 env specific IgG were low in BM collected at 4 weeks post-delivery compared to plasma . We observed similarly low NAb levels in the breastmilk of ARV naïve women in a cohort that was enrolled prior to the availability of ARVs for prevention of MTCT [6] . Thus , collectively these studies indicate that the level of HIV-1 specific NAb are low in both early and mature milk , in both treated and untreated women and this is true no matter the infecting HIV-1 subtype . We detected non-specific inhibition of HIV-1 and unrelated viruses ( SIV , MLV ) with several unfractionated BMS samples . This observation is perhaps not surprising because innate factors in BM such as defensins , lipids and lactofferin have documented activity against many viruses including enveloped retroviruses [4] . The ability of unfractionated BMS to inhibit HIV-1 in the in vitro TZM-bl assay used here did not correlate with risk of infant infection . There are several limitations to our study , most notably the fact that we focused on a select group of women with high viral load and systemic NAbs in order to optimize our chances of detecting NAbs and to examine antibody levels in relation to transmission risk . Thus it is unknown if these findings are applicable to women with low viral loads or low systemic NAbs levels . Interestingly , a correlation between ADCC activity and viral control in SIV- infected macaques was only observed when animals with low viral load were excluded [86] . These authors suggested that a threshold of antigen may be needed to elicit robust ADCC . Certainly , larger studies using relevant env antigens to examine HIV-1 specific BM antibody responses in other populations will be needed to verify these findings and determine if the findings apply to women with lower viral levels and/or systemic NAb responses . In addition , while we focused on breastmilk antibodies in relation to post-partum transmission , there could be some misclassification of time of infection in this study . Specifically , the cases of transmission examined here were all cases of relatively early post-partum transmission and we cannot exclude that some were the result of intrapartum transmission , where BM antibody levels would be less relevant . Finally , while we did not see an association between BM viral RNA levels in this small study , but this does not rule out a relationship between ADCC and the cellular viral reservoir . Larger studies that include cell-associated virus levels and ADCC activity will be needed to clarify this issue . In conclusion , we found that the capacity of BM to neutralize heterologous and autologous viruses obtained from blood and BM is limited . This observation can be explained in part by the low titers of Abs in BM compared to plasma in general , particularly IgG . It is unclear if such low NAb levels could play a role in protection , but no association was observed in this small study . However , the association between HIV-1 env specific IgG titers and ADCC activity with infant infection suggest that BM Ab could be playing some role in modulating infection through non-neutralizing mechanisms . To the best of our knowledge , this is the first study to report a positive association between BM transmission and ADCC capacity in BM . If these results are verified in a larger study of MTCT , then it would suggest that immunogens tailored at enhancing BM Abs capable of ADCC might be of potential benefit , particularly to HIV-1 infected women with high viral loads , who are at the greatest risk of transmission . | In the absence of intervention , only about one third of infants born to HIV-1 infected mothers who are continuously exposed to maternal breast milk over prolonged periods get infected . This observation raises the possibility that immune factors in infected women play a role in limiting HIV-1 transmission . Identifying factors associated with reduced HIV-1 transmission risk will improve our understanding on the potential correlates of protection that should be the focus of generating effective immunogens and vaccination protocols . Here we assessed the functional role of breast milk antibodies in a group of women with high plasma viral loads and systemic NAbs and determined that overall , breast milk contains low levels of neutralizing antibodies when compared to plasma . In contrast , we observed a robust non-neutralizing activity in breast milk that was associated with infant infection status . Our study adds to the growing evidence of a potential role of non-neutralizing antibodies in limiting HIV-1 transmission and calls for more attention to this arm of the HIV-1 response . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"immunology",
"biology",
"microbiology"
] | 2012 | HIV-Specific Antibodies Capable of ADCC Are Common in Breastmilk and Are Associated with Reduced Risk of Transmission in Women with High Viral Loads |
Plants have evolved a considerable number of intrinsic tolerance strategies to acclimate to ambient temperature increase . However , their molecular mechanisms remain largely obscure . Here we report a DEAD-box RNA helicase , TOGR1 ( Thermotolerant Growth Required1 ) , prerequisite for rice growth themotolerance . Regulated by both temperature and the circadian clock , its expression is tightly coupled to daily temperature fluctuations and its helicase activities directly promoted by temperature increase . Located in the nucleolus and associated with the small subunit ( SSU ) pre-rRNA processome , TOGR1 maintains a normal rRNA homeostasis at high temperature . Natural variation in its transcript level is positively correlated with plant height and its overexpression significantly improves rice growth under hot conditions . Our findings reveal a novel molecular mechanism of RNA helicase as a key chaperone for rRNA homeostasis required for rice thermotolerant growth and provide a potential strategy to breed heat-tolerant crops by modulating the expression of TOGR1 and its orthologs .
Temperature rising caused by global warming has imposed significant negative effects on crop yields and most likely the damage level will keep rising in future [1 , 2] . On the other hand , the sessile lifestyle of plants necessitates specific adaptations to geographical variations of environmental temperatures in their living regions as well as to extensive temperature fluctuations caused by the day-night cycle , weather variations and seasonal changes [3] . To survive in these often stressful conditions , they have evolved a number of intrinsic tolerance strategies to adapt to high temperatures . Heat shock protein protection , membrane lipid unsaturation and reactive oxygen species scavenging have been all reported to confer thermotolerance to plants [4–6] . It was also reported that alternative histone H2A . Z coordinates a high temperature transcriptome [7] and a transcription regulator ELF3 controls plant thermoresponsive growth [8] . Ribosomal RNA homeostasis is crucial to normal growth and development . It is well known that RNA homeostasis is affected by cold stress , and protected by RNA helicases [9] , which mediate RNA conformational changes by hydrolyzing ATP and unwinding short RNA duplexes adjacent to their binding sites in a nonprocessive way [10 , 11] . They are widely involved in diverse cellular processes such as transcription , RNA splicing , RNA transport , degradation and translation , and have been well documented to be involved in cold stress responses in both bacteria and yeast [9 , 12] . So far , at least 7 yeast RNA helicases have been shown to associate with the SSU in the nucleolus [13 , 14] , which is essential for pre-rRNA processing [15] . One of those SSU associated RNA helicases , Rrp3 has been found to support an effective pre-rRNA processing required for cell proliferation [14] . Higher plants possess a larger and more diverse family of RNA helicases than other organisms [16] . For example , at least 73 RNA helicases are encoded by the rice genome [17] . This large number of RNA helicases suggests a predominant role of them in modulating cellular response to a diverse range of abiotic stresses encountered by plants [18 , 19] . Among them , one DEAD-box RNA helicase LOS4 has been shown to modulate chilling resistance by facilitating mRNA export from the nucleus to the cytoplasm [20 , 21] , and another member of this family AtRH25 also confers cold stress tolerance through an unknown mechanism [18] , while several other members appear to function in salinity stress tolerance [22–24] . However , little is known about the potential role of RNA helicases at high temperatures . Rice is a staple food crop initially domesticated in subtropical regions [25] , where it must cope with environmental challenges caused by high temperatures , and subsequent human selection has extended its cultivation to temperate zones with enhanced chilling tolerance [26] . It has been recently reported that a proteasome α2 subunit gene [27] and a receptor-like kinase ERECTA [28] both contribute to rice thermotolerance . Another recently identified gene COLD1 was reported to confer chilling tolerance in japonica rice by regulating G-protein signaling pathway [26] . However , the underlying molecular mechanisms of this temperature adaptation still remain mostly elusive . Here we report on TOGR1 , a DEAD-box RNA helicase protecting rice growth at high temperatures as an intrinsic pre-rRNA chaperone . Consistent with this role , we also found that it is regulated by both temperature and the circadian clock . Furthermore , TOGR1 is recruited to the SSU in the nucleolus to facilitate an effective pre-rRNA processing required for normal cell division and thus plant growth at high temperature . Transcript level of this gene is positively correlated with plant height in different cultivated varieties and wild rice species , and its overexpression significantly improves rice growth at high temperatures . Our results uncover an essential molecular mechanism of RNA helicase as a key chaperone for rRNA biogenesis required for plant thermotolerant growth , and a possibility to produce heat-tolerant crops by regulating the expression of TOGR1 and its orthologs .
To understand the intrinsic molecular mechanisms of rice acclimating to at high temperatures , we took advantage of a wide range of climate variations in China’s rice growing regions and carried out a genetic screen for temperature-sensitive mutants from a collection of rice varieties . Among them , a spontaneous recessive thermosensitive dwarf mutant ( thermotolerant growth required1-1 , togr1-1 ) was isolated from an indica variety , Zhongxian 3037 . When planted in paddy fields at three locations representing three divergent temperature regimes ( S1 Fig ) , togr1-1 exhibited a dramatic growth phenotypic variation ( Fig 1A; S2A–S2E Fig ) . Grown in Yangzhou’s hot summer-autumn with a highest daily maximum temperature near 38°C , 23 days exceeding 34°C , which is considered to be a heat-stress threshold temperature for rice [29] , and 71 days exceeding 30°C , togr1-1 was extremely dwarf with narrow leaf blades and did not set any seed . Similarly , under Beijing’s summer-autumn conditions ( daily maximum temperatures up to 36°C , 15 days exceeding 34°C and 72 days exceeding 30°C ) , togr1-1 was dwarf with narrow leaf blades before September , then recovered to semi-dwarf when temperature was cooling down , and finally produced small panicles with a few seeds . By contrast , in the cool winter-spring of Linshui ( daily maximum temperatures during vegetative growth stage generally below 28°C and never reaching 30°C ) , togr1-1 only showed a slight overall growth difference from wild-type . To further confirm the temperature dependence of the phenotypic variations of togr1-1 , we tested its effect on growth of the mutant seedlings under controlled conditions . After grown in a climate chamber at different temperatures for three weeks , upground plant height , maximum root length and crown root number were determined for WT and togr1-1 seedlings ( Fig 1B and 1C; S2F–S2H Fig ) . The mutant seedlings showed significantly reduced plant height , root length and crown root number at 30°C , 32 . 5°C and 35°C , whereas they grew similarly to wild-type at 25°C except for a slightly reduced plant height . Furthermore , the relative values of all of the three examined variables of the mutant seedlings compared to the wild-types showed significant strong negative correlations with temperature ( Fig 1C ) , supporting that the retarded growth of togr1-1 is high temperature dependent . To examine the morphological and cellular basis of the togr1-1 dwarfism , the plant architectures of togr1-1 grown in Beijing’s hot summer were examined at tillering stage . Its leaf blades were narrowed mainly due to a decrease in vascular bundle and lateral vein numbers , and all internodes were shortened ( S3A and S3B Fig ) . Furthermore , cross-sections of leaf sheath , leaf blades , internodes and root maturation zones revealed that togr1-1 and wild-type did not show a significant difference in cell elongation but a reduction in cell numbers ( S3C–S3E Fig ) , indicating that TOGR1 is required for normal cell division at high temperatures . To isolate the mutated gene that controls the thermosensitive phenotype of togr1-1 , an F2 population for positional cloning was generated from a cross between togr1-1 and a japonica variety Zhonghua-11 , and was then grown in the paddy fields under Beijing’s hot summer conditions . We mapped it to a 28 . 5-kb region of chromosome 3 ( S4A Fig ) , and found a single nucleotide G to T substitution at position 140 of the first exon of LOC_Os03g46610 ( named as TOGR1 ) , leading to a Gly to Val substitution at position 47 of amino acid sequence ( Fig 2A ) . When togr1-1 was transformed by a construct containing an entire ORF of TOGR1 and its putative promoter sequence , the transgenic plants exhibited normal growth under Beijing’s hot summer conditions ( S4B and S4C Fig ) . Furthermore , three other allelic recessive mutants of togr1 ( togr1-2 to -4 ) screened out from an EMS ( ethyl methanesulfonate ) induced mutation library using TILLING ( targeting induced local lesions in genomes ) method also showed a high temperature dependent growth repression ( S5 and S6 Figs ) . This demonstrated that the TOGR1 mutation is responsible for the thermosensitive phenotype . To characterize the expression of TOGR1 , GUS reporter fused to a putative promoter of TOGR1 was used to transform plants and revealed that it is widely expressed in roots , shoots , leaves , culms , spikelets and anthers during different developing stages ( S7 Fig ) , consistent with the microarray data from RiceXPro [30] , indicating a general requirement of TOGR1 by all organs . To examine if TOGR1’s expression is regulated by temperature , two-week-old Zhongxian3037 seedlings grown under 12-h-light-25°C/12-h-dark-20°C condition were transferred to 35°C at ZT4 and kept for seven hours . Transcription level of TOGR1 after 0–7 h high temperature treatment was analyzed by qRT-PCR ( Fig 2C ) . In comparison with its expression in the seedlings kept at 25°C , the highest level of transcription enhancement ( two-fold ) was detected after two hours high temperature treatment . Transcript level of TOGR1 were then analyzed at 20 , 25 , 30 , 35 and 40°C ( Fig 2D ) after two hours temperature treatment starting from ZT4 , and exhibited a strong positive correlation with temperature , indicating that its expression is high temperature inducible . In Fig 2C , TOGR1 reached peak expression after five hours treatment ( ZT9 ) at both temperatures and the induction of high temperature at that time was not as significant as that after two hours treatment , indicating control of other mechanisms as well . To examine if the expression of TOGR1 was controlled by the circadian clock , we kept temperature at constant 30°C and entrained two-month-old Zhongxian 3037 plants under 12-h-light/12-h-dark ( L/D ) condition , and subsequently detected its transcript level under both L/D and continuous light ( L/L ) conditions by qRT-PCR . The temporal expression level of TOGR1 ( Fig 2B ) waved in a cosine pattern with a period that lasted approximately 24 h under both L/D and L/L conditions , indicating a circadian regulation of its expression . In addition , its peak expression appeared at zeitgeber time ( ZT9 ) corresponding to afternoon of a subjective day , the hottest time of daily temperature fluctuation , indicating the circadian clock being used to anticipate high temperature in the afternoon and a coupled expression of TOGR1 with a daily temperature alteration . The amino acid sequence of TOGR1 contains nine motifs ( Fig 2A; S8 Fig ) which are widely conserved for ATP-dependent DEAD-box RNA helicases [31 , 32] and is highly similar to Rrp3 in S . cerevisiae [33] and DDX47 in H . sapiens [34] ( S8 and S9 Figs ) , suggesting that TOGR1 is functionally similar to those pre-rRNA processing helicases . To verify the RNA helicase activity of TOGR1 in vitro and examine any potential functional difference between TOGR1 and togr1-1 , we expressed and purified both TOGR1 and togr1-1 using E . coli expression system ( S10 Fig ) , and examined their activities on a randomly synthesized 3'-overhang RNA duplex ( Fig 2E ) at 25°C , 30°C , 35°C and 40°C , respectively . For both TOGR1 and togr1-1 ( Fig 2F ) , after incubation with ATP for 30 min , the single strand RNA ( ssRNA ) corresponding to the product of unwinding the double strand RNA substrate ( dsRNA ) was accumulated and no apparent ssRNA products were detected for the experimental controls including TOGR1 and togr1-1 without ATP , and the Prescission protease control with ATP . This detected capability of unwinding RNA duplexes into ssRNA products in the presence of ATP indicates that both TOGR1 and togr1-1 have ATP-dependent RNA helicase activities in vitro . It is noteworthy that the unwinding activity of TOGR1 was markedly enhanced following a temperature increase from 25°C to 40°C ( Fig 2F ) , indicating that temperature directly controls the helicase activities of TOGR1 . Intriguingly , compared with TOGR1 , neither the helicase activities nor the temperature sensitivity of togr1-1 was diminished , suggesting that this activity is not directly related to its defect ( see below ) . Taken together , these results showed that TOGR1 encodes a temperature-dependent DEAD-box RNA helicase . The implication of yeast Rrp3 in pre-rRNA processing [14] prompted us to investigate the role of TOGR1 in the rRNA biogenesis pathway . we compared the pattern of pre-rRNA intermediate accumulation in togr1-1 with that in wild-type as well as togr1-1 and the complemented plants using seedlings first grown at 25°C for two weeks and then subjected to just one day of high temperature at 38°C . As controls , seedlings grown at continuous 25°C were also investigated . Total RNA isolated from these plants was analyzed by northern hybridizations using oligonucleotide probes S1-S6 ( Fig 3A; S1 Table ) . The most remarkable difference between wild-type and togr1-1 was a 17S precursor P-A3 detected by probe S1 and S3 , which consists of partial 5' ETS , mature 17S rRNA and partial ITS1 ( Fig 3A ) . Under both temperature conditions , this intermediate was present at higher levels in togr1-1 than in wild-type and the complemented plants , with the difference at 38°C more prominent . When togr1-1 at 38°C and 25°C were further compared , a remarkable accumulation of P-A3 was revealed for the higher temperature . By contrast , it was detected at a relatively low level in wild-type at 38°C in comparison with that at 25°C . Similar patterns were also detected for 35S** , the highest molecular weight intermediate identified by probes S1 , S3 , S4 and S5 and known as a precursor for 17S , 5 . 8S and 25S rRNA [35] , and 27SA plus 27SB identified by probes S4 and S5 and known as precursors of 5 . 8S and 25S rRNA [35] , although not as prominent as P-A3 . Another apparent difference was that the intermediate 5'ETS-C2 detected by probes S1 , S3 and S4 was only present in togr1-1 at 38°C , but undetectable for wild-type and the complemented plants at both temperatures and for togr1-1 at 25°C . This intermediate presumably spans regions 5' ETS , 17S rRNA , ITS1 , 5 . 8S rRNA and partial ITS2 . Taken together , these results indicated that in wild-type plants acclimating to heat , the P-A3 , 35S** and 27SA&B intermediates are normally reduced as a consequence of an elevated pre-rRNA processing rate upon a shift to high ambient temperatures and by contrast these intermediates plus 5'ETS-C2 were accumulated in togr1-1 under the same conditions , indicating a significant delayed pre-rRNA processing in the togr1-1 thermosensitive mutant . To define the sequence of the P-A3 intermediate , its 5' and 3' ends were further analyzed by circular RT-PCR using total RNA isolated from wild-type and togr1-1 seedlings exposed to 38°C . In agreement with the northern blot analysis , it was revealed that the P-A3 intermediate was released by cleavages at two sites corresponding to the P and A3 sites in Arabidopsis ( Fig 3B; S11 Fig ) . The sequence of the A3 site ( AAGGAAC ) is conserved with that in Arabidopsis , providing a support to a previous hypothesis that this site is conserved within plants [36] . In addition , a significant higher level of the products was also obtained from togr1-1 than wild-type as the same amount of RNA was used ( Fig 3B ) . Taken together , these results showed that TOGR1 is required to maintain a normal pre-rRNA processing pathway at high temperatures and the mutation of togr1-1 caused a delayed pre-rRNA processing . Regarding that no apparent reduction of the helicase activities was detected for togr1-1 , we investigated other potential functional differences between TOGR1 and togr1-1 using budding yeast . It is known that the genetic depletion of Rrp3 in yeast results in a repressed proliferation [14] . We expressed C-terminal-HA-tagged TOGR1 and togr1-1 ( S12 Fig ) under the control of a galactose-inducible promoter in an Rrp3 conditionally depleted yeast strain and found that TOGR1-HA but not togr1-1-HA partially rescued the repressed cell proliferation at 38°C ( Fig 4A and 4B ) , whereas no complementation effects were detected for both TOGR1-HA and togr1-1-HA at 30°C ( Fig 4C and 4D ) , suggesting that TOGR1 but not togr1-1 functions similarly as Rrp3 at high temperatures and further supports the notion that TOGR1 elevates its activities following the temperature increase . It was also revealed that proliferation of the Rrp3 strain is sensitive to high temperature ( Fig 4B and 4D ) , similar to the togr1-1 rice mutant . Considering that Rrp3 is recruited to SSU by its central component , the U3 snoRNA [13 , 14] , we next investigated the association of TOGR1-HA and togr1-1-HA with the yeast U3 snoRNA using those transgenic yeast strains . Anti-HA antibodies were used to immunoprecipitate the HA tagged protein from cell lysates . RNA isolated from pellets was analyzed by northern blot with an RNA probe specific to the U3 snoRNA . As controls , immunoprecipitations were also performed on transgenic cells with no induction of TOGR1-1 expression . As expected , TOGR1-HA prominently coimmunoprecipitated the yeast U3 snoRNA , while togr1-1-HA exhibited no apparent association ( Fig 4E ) , showing that TOGR1 is an RNA helicase being tethered to SSU and the togr1-1 has a defective ability to be recruited to the SSU processome . Association of TOGR1 with the U3 snoRNA was further examined in rice . As expected , anti-HA antibodies immunoprecipitated the U3 snoRNA ( Fig 4F ) in the TOGR1-HA ( S12 Fig ) plants , while no signal was detected in WT and OsPRR1-HA controls , indicating that TOGR1 is associated with the U3 snoRNA and subsequently with SSU in plants . To characterize the subcellular localization of TOGR1 and togr1-1 , green fluorescent protein ( GFP ) was fused to the C-terminals of TOGR1 and togr1-1 , and transiently expressed in protoplasts prepared from rice seedlings under the control of a cauliflower mosaic virus 35S promoter . With Hoechst dye and RFP-tagged Arabidopsis protein HDT1 [37] indicating the nuclear and nucleolar localization , respectively , it was revealed that TOGR1-GFP accumulated in the nucleus and predominantly in the nucleolus ( Fig 5A and 5D ) . By contrast , togr1-1-GFP is only distributed in discrete spots surrounding the nucleus as it was expressed alone in protoplasts ( Fig 5B ) , indicating that togr1-1 has lost the ability being transported into the nucleus . Intriguingly , when togr1-1-GFP was coexpressed with HDT1-RFP , it was co-localized with the marker in the nucleolus ( S13A Fig ) , suggesting that a possible interaction between the two proteins caused an artificial location of togr1-1 . Further experiments showed that locations of togr1-2-GFP , togr1-3-GFP and togr1-4-GFP were all similar to TOGR1-GFP ( S13B–S13G Fig ) , indicating that these mutant proteins have other malfunctions rather than abnormal localization . Further experiments revealed that neither location of TOGR-GFP nor that of togr1-1-GFP is obviously affected by temperature ( S14 Fig ) . It is well established that the primary function of nucleolus is ribosome biogenesis , in which rRNAs are transcribed , processed and assembled with ribosomal proteins to form immature ribosomes [38] . At the early stage of pre-rRNA processing , the U3 snoRNA and its associated proteins are assembled to form the SSU processome onto the growing 35S pre-rRNA chain co-transcriptionally [13] , corresponding to the “terminal balls” at the 5' ends of the nascent pre-rRNA transcripts observed in Miller spreads under electron microscopes [38–40] . This is crucial for early pre-rRNA cleavage [41] . Hence , in agreement with the association of TOGR1 with the SSU and the defect of togr1-1 in this , the subcellular localizations of TOGR1 and togr1-1 explain the disordered rRNA maturation of the mutant , and thus further support the idea that TOGR1 is involved in pre-rRNA processing . The capability of TOGR1 unwinding randomly synthesized RNA duplex indicates its non-sequence specificity , akin to many other RNA helicases [31 , 42] . This is in contrary with their specialized functions in various biological processes [31 , 43] . Thus , RNA helicases are generally tethered with their target RNAs or RNP complexes to execute a specific function [10 , 43] . The mutation point in togr1-1 is a glycine in the N-terminal flanking sequence . This glycine is widely conserved in the homologues of TOGR1 in all analyzed species except yeast Rrp3 ( S8 Fig ) , suggesting its importance in addition to the well defined nine motifs . The togr1-1 protein has lost neither the helicase activities nor the sensitivity to temperature , but rather has lost the competence in being transported into the nucleus and tethered to SSU in the nucleolus , underscoring the critical role of the glycine allowing TOGR1 to be distributed to its functional site and the importance of tethering an RNA helicase to its target . To examine effect of elevated TOGR1 expression on plant thermotolerance , we overexpressed TOGR1 under control of its native promoter in a rice variety Yandao 8 ( S15A–S15C Fig ) . Two-week-old seedlings grown under 12-h-25°C/12-h-20°C condition were kept at the same condition or subjected to 45°C for 52 hours and subsequently recovered under 12-h-25°C/12-h-20°C for 7 days ( Fig 6A and 6B ) . The transgenic plants showed significantly enhanced thermotolerance than WT after a heat stress treatment , whereas no remarkable difference between them was detected under a cool condition . They were also grown in a paddy field under Beijing’s hot summer-autumn conditions ( S15F Fig ) . In comparison with the WT , the TOGR1-overexpressing plants exhibited a remarkable enhancement in plant height , 1000-grain weight and number of grains per panicle ( Fig 6C–6F ) , whereas panicle length and number of panicles per plant were not significantly affected ( S15D and S15E Fig ) . An exception in the number of grains per panicle was that TOGR1-ox3 had no significant increase like the other two transgenic lines . A possible reason for this is that insertion of T-DNA during transformation disrupted the function of a gene that has positive effect on seed setting and this disruption was then compensated by overexpression of TOGR1 . Consistently , qRT-PCR analysis of leaf blades from a collection of 38 rice varieties of O . sativa , O . granulata and O . rufipogon grown under Beijing’s hot summer-autumn conditions revealed that plant height has a moderate positive correlation ( r = 0 . 489; p < 0 . 01 ) with the expression level of TOGR1 ( Fig 6G ) . These results strongly support the idea that an enhancement of TOGR1 expression improves plant growth under hot conditions . In order to investigate the impact of rRNA biogenesis disorder on transcriptome , we performed RNA sequencing in three-week-old rice seedlings grown in chambers at 25°C and 30°C , respectively . It was shown that when temperature increases , both the wild-type and togr1-1 dramatically adjust the expression levels of more than 2 , 000 genes ( S16 Fig ) . Gene ontology ( GO ) enrichment analysis of the differentially expressed genes ( DEGs ) in the wild-type revealed several significantly enriched biological process categories required for growth including polysaccharide metabolic process ( GO:0005976 ) , carbohydrate metabolic process ( GO:0005975 ) , glucan metabolic process ( GO:0044042 ) , carbon fixation ( GO:0015977 ) and organic substance metabolic process ( GO:0071704 ) , in addition to response to stimulus ( GO:0050896; Fig 7A; S3 Table ) . By contrast , significantly enriched biological processes in togr1-1 contained response to stimulus , response to oxidative stress ( GO:0006979 ) and response to stress ( GO:0006950 ) , whereas the coordination of primary metabolic processes and carbon fixation were largely impaired or undetectable ( Fig 7A; S3 Table ) . Under a moderate high temperature ( 30°C ) that is known to promote growth of rice rather than cause heat stress [29] , togr1-1 appears to primarily divert resources to cope with environmental stimulus rather than to support growth , which normally occurs when plants are exposed to heat-stress conditions [44–46] . It was reported that ascorbate peroxidase 2 ( APX2 ) and galactinol synthase 2 ( GolS2 ) both protect plants from heat stress by scavenging of ROS ( reactive oxygen species ) and producing of osmoprotectants respectively [47–49] . To study if TOGR1’s function is related to these two genes , we next examined transcript levels of homologues of APX2 ( LOC_Os07g49400 ) and GolS2 ( LOC_Os03g20120 and LOC_Os07g48830 ) in WT rice and togr1-1seedlings ( S16B Fig ) . After one-day treatment of 45°C heat stress for seedlings previously grown at 25°C , all of the three genes showed significantly increased expressions in togr1-1 , whereas those in the WT either not significant or the increased levels much lower . This reflects a much stronger up-regulation of the three genes in response to heat stress in togr1-1 than that in the WT , and thus indicates a compensation to the internal disorders caused by dysfunction of TOGR1 , but does not support the hypothesis that TOGR1 functions in the same pathway with APX2 and GolS2 . Taken together , these results suggested that TOGR1 is required for a primary metabolism adaptation to at high temperatures and its mutation results in a system disorder impeding this adaptive mechanism required for normal cell division and thus plant growth at high temperatures .
As a ribosome factory , the nucleolus has been reported to be sensitive to various stresses including heat [50 , 51] . It has been proposed that during the early stage of pre-rRNA processing , the nascent pre-rRNA transcript might wrap around the SSU processome which mediates early stage cleavages [52] . Thus an effective processing requires proper RNA foldings and RNA-protein interactions regarding that RNA conformation is vulnerable to temperature changes [53] . In fact , the nucleolus contains numerous non-ribosomal proteins associated with the SSU processome to facilitate RNA folding and interaction of many components of it [54] . A number of them have been identified as RNA helicases [13 , 14] . Members of this protein family are known to protect organisms from various abiotic stresses including cold stress tolerance [9] , but little is known about their roles in at high temperature tolerance . Our study revealed that , an SSU-associated nucleolar RNA helicase , TOGR1 , appears to significantly affect early pre-rRNA cleavage ( Fig 3 ) and is essential for the thermotolerance mechanism of rice growth . This is mediated by priming pre-rRNA processing in response to ambient temperature increase as an RNA chaperone ( Fig 7B ) . At cool temperatures , likely with the help of other RNA helicases and the role of TOGR1 almost dispensable , the pre-rRNAs are in a functional conformation that can be readily processed into mature rRNAs to support plant growth . However , at high temperatures , the pre-rRNAs are misfolded and the interactions between pre-rRNAs and its processing proteins are disrupted , which would result in an ineffective processing of pre-rRNAs if without the assistance of TOGR1 . Being recruited to the SSU in the nucleolus and with its expression and activity enhanced by temperature increases , TOGR1 appears to aid pre-rRNAs to form native conformation and proper RNA-protein interactions as an RNA chaperone . This action thus ensures the production of normal amount of rRNAs required for a proper coordination of primary metabolisms and normal plant growth at high temperatures . By contrast , the dysfunction of togr1-1 in its inability of being transported to the nucleolus and thus recruited to SSU makes it unavailable when rRNA biogenesis needs a protection from high temperature leading to pre-rRNA processing and primary metabolism disorders and thus retarded plant growth . Up to date , this is the only plant RNA helicase that has been found to modulate the growth thermotolerance , revealing a novel molecular mechanism for plants to acclimate to hot weathers . Ribosomal RNA biogenesis is crucial for eukaryotic growth control [55 , 56] , and is tightly coupled to environmental changes [57] . Our results demonstrate that rice enhances the rate of rRNA biogenesis at high temperatures . However , temperature increase also leads to a disruption of RNA homeostasis which further causes the delayed pre-rRNA processing , similar to the well-established cold stress effect [9 , 18 , 19 , 21] . Thus , RNA helicases are needed to support an effective processing of pre-rRNA at high temperatures . The retarded rRNA biogenesis and plant growth defects exhibited by the togr1-1 mutant at high temperatures indicate an indispensable role of TOGR1 as an RNA chaperone in aiding pre-rRNA processing and normal plant growth as ambient temperature increasing . Plants sense ambient temperature change via a set of primary thermosensors including membrane localized Ca2+ channels [6] and H2A . Z containing nucleosomes [7] . All provide signals for plants to make internal adjustment to adapt to temperature fluctuation . To protect pre-rRNA processing from high temperature , TOGR1 appears to be up-regulated by temperature increasing signal and , in addition , its expression is regulated by the circadian clock and thus is synchronized with a daily temperature fluctuation which can be anticipated by the circadian clock . Our findings provide a remarkable example of high temperature induced RNA helicases , revealing a new role for a member of this protein family and , in particular , paving a way for studying the roles of RNA helicase and rRNA biogenesis in plant thermotolerance . TOGR1 appears to serve as a critical element of a “buffering” system against unfavourable temperature conditions for plants . That means this protein makes plants adaptive to a more broad range of temperatures by conferring thermotolerance . Nevertheless , further details about temperature sensing pathway of this system remain to be elucidated . Our study has also established a molecular connection between rRNA biogenesis and cell division . It has been argued that ribosome biogenesis is a key component of the signaling network controlling cell growth and division and play roles in specific aspects of development and physiology rather than a general growth control [55 , 56 , 58] . TOGR1 may participate in a coordination program of primary metabolisms via its effect on rRNA biogenesis . Coordinated primary metabolisms and sufficient rRNA availability facilitated by TOGR1 probably provide both signals and material basis for proper cell division at high temperatures . This is consistent with a previous report that mutation of 3 ribosome biogenesis genes in Arabidopsis caused the similar effect [58] . However , the pathways that ribosome biogenesis influences cell division remains to be explored in more detail . In conclusion , our results demonstrate that TOGR1 functions as a thermosensitive RNA chaperone in the nucleolus to mediate normal plant growth at high temperatures , providing a novel insight into the endogenous mechanisms protecting plants from hot weathers and a possibility to mitigate their adverse impacts on plant productivity .
The togr1-1 mutant was isolated from a rice ( Oryza sativa L . ssp . indica ) variety Zhongxian 3037 . Plants were grown in the paddy fields under natural conditions or in growth chambers ( 12L:12D cycle with a light intensity of 200 μmol quanta m-2 s-1 and 80% humidity , unless otherwise specified ) . For circadian analysis , two-month-old Zhongxian 3037 plants were first entrained under 12L:12D conditions for 10 days and then transferred to continuous light conditions , and temperature was kept at 30°C . Details of positional cloning , transgenic plants and plasmid constructions were described in S1 Text . Allelic mutants were obtained from http://croptilling . org/ and were screened out using a previously described TILLING method [59] . Primer pairs TOGR1T1F/ TOGR1T1R and TOGR1T2F/ TOGR1T2R ( S1 Text ) were used for amplification . Cross-sections of internodes , leaf blades and roots were stained by both safranin and fast green , and leaf sheaths were stained by safranin ( S1 Text ) . A Construct containing TOGR1pro::GUS was transformed into rice , and the transgenic plants were analyzed with a GUS staining assay as described previously [60] . The protein sequences homologous to TOGR1 were found by BLASTp ( http://www . ncbi . nlm . nih . gov/BLAST/ ) using the entire amino acid sequence of TOGR1 as a query . The obtained amino acid sequences were aligned using Clustal W software [61] ( http://www . ebi . ac . uk/clustalw/ ) and shaded using GenDoc . A rooted phylogenetic tree was constructed using the neighbor-joining method of the MEGA 5 . 2 software [62] with the following parameters: Poisson model , pairwise deletion , and bootstrap ( 1000 replicates ) . The fusion constructs ( 35S::TOGR1:GFP and 35S::togr1-1:GFP ) and control ( 35S::GFP ) were transiently expressed in protoplasts prepared from etiolated rice seedlings as described [63] . 35S::HDT1:RFP was used as a nucleolar marker . Protoplasts were stained by Hoechst 33258 ( Sigma-Aldrich ) before visualization and photography; see S1 Text . The 3′-overhang RNA duplex used for helicase assay were arbitrarily designed to minimize secondary structures [64] and labeled at the 5′ end of the indicated strand with ROX fluorophore ( Fig 2E; Takara Bio ) . The ROX labeled upper strand was used as a marker and the same strand without labeling was used as a trap for the displaced lower strand of the RNA duplex . Helicase assay was performed on purified proteins using a method modified from previously described [64] . Detailed procedures of protein expression and purification and helicase assay are provided in S1 Text . For complementation analysis , YPH499 ( MATa ura3-52 lys2-80 ade2-101 trp1-Δ63 his3-Δ200 leu2-Δ1 ) in which the chromosomally encoded Rrp3 was placed under the control of the DOX-repressible tetO7 promoter was used as parent strain [14] . Strain carrying empty pYES2 vector or pYES2 plasmid containing GAL1::TOGR1:HA or GAL1::togr1-1:HA was spotted on induction medium containing both DOX and D-galactose in a three 50-fold serial dilutions at OD600 0 . 5 , 0 . 01 and 0 . 002 . Growth curves were monitored in liquid induction medium , and strain carrying empty pYES2 grown in liquid synthetic minimal medium was used as control ( S1 Text ) . For rice , UBi-1::TOGR1:HA plants in togr1-1 background were grown for 60 days in growth chamber . Wild-type Zhongxian 3037 and UBi-1::OsPRR1:HA plants in Zhonghua-11 background were used as negative controls . Leaf blades were harvest and plant whole-cell extracts were prepared as previously described [65] . For yeast , tetO7::Rrp3 YPH499 strain harbouring GAL1::TOGR1:HA or GAL1::togr1-1:HA were cultured in liquid induction medium containing both DOX and D-galactose at 30°C . The GAL1::TOGR1:HA strain and the parent strain grown in liquid synthetic minimal medium were used as negative controls . At exponential stage , yeast whole-cell extracts were prepared according to a published method [66] using glass beads ( Sigma-Aldrich ) . Immunoprecipitation was performed as previously described [67] . Total RNA was then extracted for northern analysis . Detailed procedures are provided in S1 Text . For northern blotting , RNA probes antisense to the rice or yeast U3 snoRNA labeled with [α-32P] UTP and oligonucleotides labeled with [γ-32P]ATP at the 5′ ends were used to detect the U3 snoRNAs and pre-rRNA processing intermediates , respectively . The 5′ and 3′ ends of rRNA precursors were determined by circular RT-PCR as described previously [68] . For qRT-PCR , three biological and three technical repeats were performed in the experiments . Primers and detailed procedures are provided in S1 Text . RNA sequencing of Zhongxian 3037 at 25°C and 30°C and togr1-1 at 25°C and 30°C was conducted by BGI ( Shenzhen , China ) . DEGs ( differentially expressed genes ) with difference equal or higher than 2-fold between the two temperature levels were then mapped to GO terms in the database ( http://www . geneontology . org/ ) . Ultra-geometric test was used to find significantly enriched GO terms in DEGs comparing to the genome background , taking p≤0 . 05 as a threshold; see S1 Text . Protein samples extracted from plant leaves or yeast cells were separated on SDS-PAGE gels , and then directly visualized or analyzed by Western blot ( S1 Text ) . RNA-seq data have been deposited in the Gene Expression Omnibus ( GEO ) under accession number GSE42096 . | Global warming is increasingly posing negative impacts on crop productivity . In this study , we report a nucleolar-located RNA helicase TOGR1 for thermotolerant growth in rice . TOGR1 maintains pre-rRNA homeostasis under high temperature by securing a proper pre-rRNA structure via elevating its helicase activity . Its expression is high temperature inducible with an afternoon peak expression , consistent with a high temperature anticipation of the circadian clock . Transcriptome analysis revealed that TOGR1 is essential in coordinating primary metabolisms to support thermotolerant growth . Importantly , an enhanced expression of TOGR1 significantly increased biomass of rice . Our findings reveal a novel role of a RNA helicase in thermotolerance and provide a potential strategy to breed heat-tolerant rice cultivars and possibly other heat-tolerant crops . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"small",
"nucleolar",
"rnas",
"plant",
"anatomy",
"panicles",
"enzymes",
"gene",
"regulation",
"enzymology",
"cereal",
"crops",
"plant",
"science",
"rice",
"model",
"organisms",
"rna",
"helicases",
"crops",
"seedlings",
"cell",
"nucleus",
"inflorescences",
"plants",
... | 2016 | Nucleolar DEAD-Box RNA Helicase TOGR1 Regulates Thermotolerant Growth as a Pre-rRNA Chaperone in Rice |
CD4+ T cells orchestrate the adaptive immune response in vertebrates . While both experimental and modeling work has been conducted to understand the molecular genetic mechanisms involved in CD4+ T cell responses and fate attainment , the dynamic role of intrinsic ( produced by CD4+ T lymphocytes ) versus extrinsic ( produced by other cells ) components remains unclear , and the mechanistic and dynamic understanding of the plastic responses of these cells remains incomplete . In this work , we studied a regulatory network for the core transcription factors involved in CD4+ T cell-fate attainment . We first show that this core is not sufficient to recover common CD4+ T phenotypes . We thus postulate a minimal Boolean regulatory network model derived from a larger and more comprehensive network that is based on experimental data . The minimal network integrates transcriptional regulation , signaling pathways and the micro-environment . This network model recovers reported configurations of most of the characterized cell types ( Th0 , Th1 , Th2 , Th17 , Tfh , Th9 , iTreg , and Foxp3-independent T regulatory cells ) . This transcriptional-signaling regulatory network is robust and recovers mutant configurations that have been reported experimentally . Additionally , this model recovers many of the plasticity patterns documented for different T CD4+ cell types , as summarized in a cell-fate map . We tested the effects of various micro-environments and transient perturbations on such transitions among CD4+ T cell types . Interestingly , most cell-fate transitions were induced by transient activations , with the opposite behavior associated with transient inhibitions . Finally , we used a novel methodology was used to establish that T-bet , TGF-β and suppressors of cytokine signaling proteins are keys to recovering observed CD4+ T cell plastic responses . In conclusion , the observed CD4+ T cell-types and transition patterns emerge from the feedback between the intrinsic or intracellular regulatory core and the micro-environment . We discuss the broader use of this approach for other plastic systems and possible therapeutic interventions .
The immune system protects organisms against external agents that may cause various types of diseases . As the immune system mounts specialized responses to diverse pathogens , it relies on plastic responses to changing immunological challenges . At the same time , the immune system must maintain homeostasis and avoid auto-immune responses . Therefore , the immune system relies on resilience mechanisms that enable it to return to basal conditions once pathogens or immunogenic factors are no longer present [1–3] . CD4+ T cells , also known as T helper ( Th ) cells , are key in the response to infectious agents and in the plasticity of the immune system . Naive CD4+ T cells ( Th0 ) are activated when they recognize an antigen in a secondary lymphoid organ . Depending on the cytokine milieu and other signals in their micro-environment , CD4+ T cells attain different cell fates [2 , 4–7] . Nonetheless , we still do not have a complete understanding of the dynamic mechanisms underlying CD4+ T cell differentiation and plasticity [5] . Each CD4+ T cell type is associated with specific cytokines , receptors , transcription factors and functions ( Fig 1 ) . Th1 cells express T-bet , secrete interferon-γ ( IFN-γ ) and are associated with cellular immunity [8] . Th2 cells express GATA3 , secrete interleukin ( IL ) -4 and are associated with immunity to parasites [8] . Th17 cells express RORα and RORγt , secrete IL-17 and IL-21 , and are associated with neutrophil activation [9–10] . Follicular helper CD4+ T cells ( Tfh ) express Bcl6 and CXCR5 , secrete IL-21 and are associated with B cell maturation in germinal centers [11 , 12] . Th9 cells secrete IL-9 and exert anticancer activity [13 , 14] . Induced regulatory T cells express Foxp3 , secrete TGF-β and/or IL-10 , and are associated with immune tolerance [15 , 16] . There is also considerable overlap among the expression profiles of different CD4+ T cells . For example , IL-9 and IL-10 can be secreted by Th1 , Th2 , Th17 , iTreg cells and a variety of other immune cells [17–19] . T regulatory cells can also express IL-17 [20] . CD4+ T cells are highly plastic , switching from one type to another in response to environmental challenges ( Fig 1 ) [1 , 21–23] . Th17 cells can transform into Th1 cells [24–25] , and iTregs differentiate into Th17 in the presence of IL-6 [26] . Th2 cells can become IL-9 producing cells but may not easily become Th1 cells [27] . iTreg and Tfh cells can independently develop into other CD4+ T cell types , and they can be derived from Th1 , Th2 or Th17 cells [28–30] . The differentiation and plasticity of CD4+ T cells depends on the interactions among the cytokines produced by other immune cells , epithelial cells , adipocytes , or by the CD4+ T cells themselves; the transduction of those signals and the regulation of this signaling by suppressors of cytokine signalling ( SOCS ) proteins; the set of transcription factors expressed inside the cells; epigenetic regulation; certain metabolites; and also microRNAs [4 , 6 , 31–33] . Given the complexity of CD4+ T cell transitions and the difficulty of classifying a particular expression pattern as a subset or a lineage , we will refer to the different stable expression patterns of CD4+ T cells as “cell-types” . A mechanistic , integrative and system-level understanding of CD4+ T cell differentiation and plasticity requires dynamic regulatory network models that consider the concerted action of many components . These models can be used to prove whether the known biological interactions are necessary and sufficient to recover attractors that correspond to experimentally observed configurations in different CD4+ T cell types . Additionally , such models may be used to address whether the considered components and interactions also restrict and explain the observed patterns of transition among cell types . Finally , this type of model can be used to test the role of different network components in cell differentiation and plasticity . In such regulatory network models , the nodes correspond to the regulatory components of the network such as genes , proteins or signals , while the links correspond to the interactions among components . The state of each node is determined by the expression level of its regulators , and the logical functions describe the dynamic evolution of the node states . The attractors , the states to which such regulatory networks converge , can be interpreted as the profiles characterizing different cell types ( see reviews in: [34–36] ) . Previous studies have used regulatory network models to study CD4+ T cell differentiation and plasticity [37–40] . These models captured the dynamic and non-linear regulation of CD4+ T cells and recovered the attractors corresponding to the Th0 , Th1 , Th2 , iTreg and Th17 cell types . They have also been useful for preliminary studies of CD4+ T cell plasticity in the presence of different cytokines in the micro-environment [38] and fir studies of the effect of a specific molecule ( PPARγ ) in the Th17/iTreg switch [40] . However , as new T CD4+ cell types such as Tfh , regulatory Foxp3-independent , Th9 , and Th22 cells are described , it is necessary to develop an updated regulatory network that is able to recover the configurations that characterize such novel cell subsets . Additionally , to date no minimal model that incorporates the necessary and sufficient set of interactions to also recover the reported patterns of transitions among Th cells has been reported . Here , we specifically address whether CD4+ cell types and their transition patterns emerge as a result of the feedback between a minimal regulatory core of intra-cellular transcription factors and cytokines produced by the CD4+ T cell together with cytokines produced by other cells present in the micro-environment . Our results confirm that a regulatory network model that only considers the interactions among the master transcription factors is not sufficient to recover configurations that characterize the different CD4+ T cell types . Therefore , we then integrated a minimal network of master transcriptional factors with cytokine signaling pathway , including the cytokines produced by the cell and those present in the micro-environment , to integrate a network with the necessary and sufficient set of components to recover documented CD4+ T cell differentiation and plasticity patterns . The observed configurations of CD4+ T cells ( Th0 , Th1 , Th2 , Th17 , iTreg , Tfh , Th9 and Foxp3-independent T regulatory cells ) emerge from the feedback and cooperative dynamics among the multiple levels of regulation considered in the minimal model . In addition , this system is able to recover the plastic transition patterns and stability behavior that have been described for the different cell types in response to transitory perturbations and different micro-environments . Interestingly , our model predicts that transitions from particular cell types to others are caused by transient activations , while transient inhibitions usually cause cells to remain in their original cell types . Additionally , we show that T-bet , TGF-β and SOCS proteins are keys to recovering observed CD4+ T cell plastic responses . Finally , we discuss the relevance of our models for a system-level understanding of mammalian immunological responses and eventual biomedical interventions .
Boolean networks are capable of integrating qualitative interactions ( molecular , physical , chemical , etc . ) into a coherent picture and are useful ways to explore the minimal set of restrictions that are necessary and sufficient to produce emergent biological patterns and behaviors [41–43] . The regulatory interactions considered in the present model are grounded on experimental data . In the proposed regulatory network , the nodes represent the regulatory components of the network and the links the interactions among them ( S1 Table and S1 Fig ) . Given the complexity of the network , we simplified the model by removing intermediate components along a network path ( S1 File ) following a method proposed in [44] and checked the consistency of the reduced network using GINsim [45] . The predicted cell phenotypes arising from the steady states of the network are consistent with the available experimental data [2 , 4–7] . The model assumes that all interactions are synchronous , that all cytokine receptors are present , and that the TCR and its cofactors are activated ( being unable to model unactivated and anergic CD4+ T cells ) . The model ignores weak interactions , low levels of expression , and epigenetic regulation ( S1 File ) . To address whether a minimal transcriptional regulatory core could recover the observed configurations that characterize the main CD4+ T cell types that have been described up to now , we extracted from the general network under study a minimal regulatory module consisting only of transcriptional regulators ( Fig 2A , S2 Table , BioModels Database: MODEL1411170000 ) . Our aim was to test whether this minimal module contained a sufficient set of interactions to predict the observed configurations for the transcription factors included in the model that characterize different CD4+ T cell types . The nodes of the transcriptional regulatory network ( TRN ) correspond to the five “master” transcription factors associated with CD4+ T cell types: T-bet for Th1 , GATA3 for Th2 , RORγt for Th17 , Foxp3 for iTreg , and Bcl6 for Tfh . The dynamic analysis of this TRN recovered attractors corresponding to different CD4+ T cell types ( Fig 2B ) : Th0 , Th1 , Th2 , iTreg and the hybrid states T-bet+Foxp3+ [46] and GATA3+Foxp3+ [47] . However , this TRN did not converge to configurations that characterize the Th17 and Tfh cell types , implying that the expression of RORγt and Bcl6 is not sufficient to maintain such cell types . This result may be caused by the lack of feed-forward loops in the TRN . RORγt has no positive interactions with any of the transcription factors considered in the TRN and lacks a feedback loop mediated by transcription factors [48] . The mode of self-regulation of Bcl6 remains unclear , as it has been reported to either activate or inhibit its own expression in B cells [49–50] . The above result reveals which T CD4+ cell types rely only on the postulated TRN and which require extrinsic signals . To formally test this hypothesis , we extended the TRN network by introducing key components of signaling pathways and their regulators , as well as cytokines that have been shown to be fundamental in CD4+ T cell type attainment . This T CD4+ cell transcriptional-signaling regulatory network ( TSRN ) was then simplified ( S1 File , S1 Fig ) to obtain a minimal network . To reduce the number of nodes in the network , we assumed that the TCR signal was present and that the cytokine receptors were present in sufficient amounts to transduce a signal . This network lacks many important inflammatory cytokines ( such as IL-1 , TNFα ) , because while these cytokines are crucial for the immune response , they are dispensable for CD4+ T cell differentiation . The model analyzed in this paper also lacks extrinsic cytokines produced by other immune system cells and other cell types such as IL-12 and IL-18 . The network also lacks some transcription factors and cytokines associated with newly reported Th types such as IL-22 , as detailed experimental information linking them to the network model under analysis is not yet available . The nodes of the simplified TSRN represent ( Fig 3A , S3 Table , BioModels Database: MODEL1411170001 ) transcription factors , signaling pathways and extrinsic cytokines . The nodes corresponding to cytokine pathways are active if the signal is transduced; this means that if the cytokine is present , it forms a complex with the receptor that can activate a messenger molecule ( for example a STAT protein ) , which is then translocated to the nucleus . Cytokines can be produced by both CD4+ T cells ( intrinsic ) and by other cells of the immune system and the organism ( extrinsic ) . To resolve this ambiguity we added nodes representing the extrinsic cytokines produced by other cells and tissues of the immune system ( IL_e ) . This extended TSRN includes 18 nodes: the transcription factors ( Tbet , GATA3 , RORγt , Foxp3 , Bcl6 ) , the effector cytokines and their signaling pathways ( IFN-γ , IL-2 , IL-4 , IL-21 , IL-9 ) , the regulatory cytokines ( TGF-β and IL-10 ) and the extrinsic cytokines ( IFN-γe , IL-2e , IL-4e , IL-21e , TGF-βe and IL-10e ) . While IL-10 , IL-6 and IL-21 all signal using STAT3 , IL-6 and IL-21 cause inflammation , while IL-10 suppresses inflammation . To analyze this network , we assume that IL-10 signaling was mediated by a different pathway than IL-6/IL-21 , even though they share STAT3 as a messenger molecule . The production of these external cytokines is independent of regulation inside the CD4+ T cell , but their signaling can be blocked ( for example by SOCS proteins [51] ) . The resulting network includes two levels of regulation , the regulation in the nucleus by mutually inhibiting transcription factors and the regulation among the receptors and their signal transduction pathways mediated by SOCS proteins . The dynamic analysis of the TSRN yields stable configurations that correspond to: Th0 , Th1 , Th2 , Th17 , iTreg , Tfh , T regulatory Foxp3-independent cells , Th1R , Th2R and GATA3+IL4- cells ( Fig 3B ) . As this biological patterns can be obtained in the presence of various extrinsic cytokines , we labeled each attractor according to the active transcription factors and intrinsic cytokines . Resting CD4+ T cells ( labeled Th0 ) were defined as expressing no transcription factors or regulatory cytokines . Th1 was defined as Tbet and IFN-γ active [8] , Th2 as GATA3 and IL-4 active [8] and GATA3+ ( a Th2-like cell type ) as GATA3+IL4-[38] . Th17 was defined based on RORγt and STAT3 signaling mediated by IL-6 or IL-21 , all of which require the presence of TGF-βe [9–10] . iTreg expressed Foxp3 and TGF-β , IL-10 or both , all of which require the presence of IL-2e [16] . Interestingly , the TSRN model also predicts a novel set of steady states that had not been predicted by previous models but that correspond to reported biological cell types ( Fig 3B ) ; for example , Tfh cells with Bcl6 and STAT3 signaling mediated by IL-21 [12]; Th9 cells with IL-9 , requiring the presence of TGF-β and extrinsic IL-4 [27]; T regulatory cells , as Foxp3-independent CD4+ T cells ( TrFoxp3- ) with TGF-β , IL-10 or both , but not Foxp3 [52]; Th1 regulatory cells ( Th1R ) expressing a regulatory cytokine and T-bet [46]; and Th2 regulatory cells ( Th2rR ) expressing a regulatory cytokine and GATA3 [47] . The model does not consider the Th22 cell type [53] because IL-22 was not included in the network due to the lack of experimental data on this molecule . To validate the model with experimental data , we simulated loss and gain of function alterations for some nodes . In general , the results agree with the available experimental data , except in the case of the IL-2 knock-out . IL-2- causes the loss of iTreg cells as these cells require continuous IL-2 signaling [54 , 55] , but this differs from the actual IL-2 KO mutants , which lose most CD4+ T cell types because IL-2 is also critical for the activation and survival of CD4+ T cells . This model also allows us to predict the behavior of the Tr Foxp3- , Th1R and Th2R cell types in response to various knock-out and over-expression simulations for several transcription factors or signaling pathways where no experimental data are available . We performed a functional robustness analysis in which the logical functions of the network were altered ( S2 Fig ) to verify the construction of the functions and the structural properties of the model and to avoid over-fitting . Altering one of the functions of the network resulted in 1 . 389% of the initial states attaining a different final attractor than the original final state , and only 0 . 219% of the initial states arrived at an attractor that was not in the original set of attractors of the non-altered network . To further verify that the results of the Boolean network are not an artifact due to the discrete nature of the model and to further assess the robustness of the attractors to variations in the node values , we approximated the discrete step-like functions of the Boolean model with continuous interaction functions [44] ( S2 File ) . The continuous model recovers the same attractors as the Boolean regulatory network . Furthermore , these attractors are stable in response to small perturbations in the value of the nodes as predicted by the robustness analyses of the Boolean version of the model . Cytokines can be produced by the cell ( intrinsic ) or by other cells of the immune system ( extrinsic ) . These extrinsic cytokines constitute the micro-environment for CD4+ T cell differentiation . The role of polarizing micro-environments in CD4+ T cell differentiation was assessed using the TSRN model . In this network , the values of the extrinsic cytokines were fixed at a given expression level and the network response was analyzed again ( Fig 4 ) . Th0 , Th1 , Th2 and Tfh can be maintained in the absence of extrinsic cytokines or in the presence of effector cytokines such as IFN-γ , IL-2 , IL-4 or IL-21 . Th17 , iTreg and Th9 cells require extrinsic TGF-β , IL-2 and IL-4 , respectively , to maintain their homeostatic states [13 , 56] . TrFoxp3- states can be maintained in most polarizing micro-environments [57 , 58] . The recovered behaviors agree with the experimental data and also with previous models [38] . The importance of the extrinsic cytokines present in the micro-environment can be further analyzed when the system is studied under polarizing conditions . The presence of extrinsic signals for a given cell type increases the number of initial states that differentiate into that cell type , while the absence of extrinsic signals may lead to the loss of a cell type , as is the case with Th17 , iTreg and Th9 cells ( Fig 4 ) . The presence of the regulatory cytokines IL-10 and TGF-β inhibits most effector CD4+ T cells , except for Th17 . This finding may explain the presence of Th17 cells in regulatory micro-environments [59] and provides important insight concerning the relationship between Th17 and iTreg . Thus , this type of modeling framework and analysis may prove useful for finding therapeutic approaches to chronic inflammation . The polarization of the micro-environment towards a particular cell type increases the size of the basin of attraction and its resistance to transient perturbations . Basin size and attractor stability are not identical ( S3 Fig ) . In this way , the environmental signals promote specific cell types and increase their stability , which likely affects the population dynamics of CD4+ T cells . Nonetheless , different CD4+ T cell types coexist during immune responses . Even if the signals in the micro-environment promote a specific cell type , attractors corresponding to other cell types can still appear in this micro-environment , but their basin sizes and stability tend to be smaller . The ability of the immune system to dynamically respond to environmental challenges depends on its plastic responses . CD4+ T cells are phenotypically plastic , and once differentiated , their expression patterns can be altered depending on internal and external cues . This cell plasticity seems to be important for the overall plasticity of immune system responses [1] . To analyze CD4+ T cell plasticity , we transiently perturbed the attractors of the system . For each attractor we altered the value of one of its nodes and then evaluated the system until an attractor was reached . If the original attractor was reached , we considered the corresponding cell type as stable towards that perturbation . If a new cell-type was reached , we considered that the transition from one cell type to another corresponded to phenotypic plasticity . This analysis was repeated for every node and every attractor . This methodology allowed us determine all the transitions between cell types , the specific perturbation that caused the transition , and the path from one cell-type to another . These transient perturbations in the values of the nodes are equivalent to developmental noise or temporal changes in the micro-environment of the cell . The result is a cell-fate map where the nodes represent CD4+ T cell types recovered by the TSRNand the connections represent the possible transitions between pairs of differentiated cell types ( Fig 5 , S3 File ) . The model recovers the reported transitions corresponding to the polarization of naïve CD4+ T cells into canonical CD4+ T cell types , as well as various events of trans-differentiation between canonical CD4+ T cell types . Most of the predicted transitions are to or from Th0 or towards TrFoxp3- . It is important to clarify that the TCR complex was not included in the minimal model . Thus , in our model , the Th0 attractor represents resting CD4+ T cells . There are few direct transitions among the Th1 , Th2 , and Th17 cell types . The few direct transitions found towards iTreg and Tfh can only be achieved in polarizing micro-environments . It is also possible to transition from one of the main cell types to another one through the Th0 , TrFoxp3- , Th1R , Th2R or GATA3+IL4- attractors . This ability raises multiple questions about the signals necessary for plasticity in vivo . It is possible that in order to transition from one cell type to another , some signals have to be maintained for a certain period of time , or that more than one perturbation is necessary . Further studies are required to determine which conditions are necessary and sufficient for CD4+ T cell type transitions to further understand CD4+ T cell plasticity . Therefore , in the context of this study , we define plasticity as the potential of a given differentiated cell to attain other fates in response to alterations in the expression patterns of their intrinsic components and/or of the extrinsic micro-environment . Of the total of 121 possible transitions between cell types arising from those alterations , the TSRN network yielded 66 cell-type transitions . Thus , the topology or set of regulatory interactions proposed in this network generates restrictions in terms of cell types but also in terms of the patterns of cell-fate transitions . CD4+ T cells are typically under the influence of particular micro-environments , with specific cytokines affecting the dynamics of these cells . Depending on the combination of cytokines , some cell types are lost , and transitions among the remaining cell types are also restricted . To simulate polarizing micro-environments , we fixed the value of the cytokines associated with pro-Th1 ( IFGγe ) , pro-Th2 pro- ( IL-4e , IL-2e ) , pro-Th17 ( IL-21e , TGF-βe ) , pro-iTreg ( TGF-βe , IL-2e ) , pro-Tr ( IL-10 , ) pro-Tfh ( IL-21e ) and pro-Th9 ( IL-4 and TGF-βe ) . In general , the polarizing micro-environment increases the size of the attraction basin , the stability and the transition into the attractor . The biological nature of the polarizing signal affects the nature of the resulting transition . In response to regulatory signals ( IL-10e , TGF-βe ) , the majority of the transitions are towards TrFoxp3- , while inflammatory signals lead to more transition signals towards Th1 and Th2 . All of these results represent interesting predictions that could be tested experimentally . While all the elements of the TSRN have previously been shown to be necessary for the differentiation of CD4+ T cells , we wished to address their relative importance in cell plasticity responses . To evaluate this question , we perturbed each node of all the attractors and measured how many times the perturbed state changed to a new attractor ( Fig 7 ) and to which new cell type the system converged ( S4 Fig ) . This process is equivalent to the temporal activation or inactivation of a transcription factor or an element of the signaling pathway in response to noise . Alterations of T-bet and TGF-β usually caused the perturbed state to change from one attractor to another , while RORγt and IL-9 had the least effect on cell-fate transitions . In general , the system is more sensitive to perturbations in the master transcriptional regulators than to alterations of the cytokines . In contrast to previously published T CD4+ network models that only included SOCS1 [37–40] , several SOCS-type proteins were considered in the TSRN presented and analyzed here . SOCS proteins are important for the differentiation and plasticity of CD4+ T cells . SOCS1 is commonly silenced in inflammatory diseases , and over-expression of SOCS3 correlates with allergies [31 , 51] . To explore the role of SOCS proteins and the impact of alterations in these proteins on CD4+ T type transitions , we generated a network lacking the inhibitions mediated by these proteins ( Fig 8 ) . This altered system recovers the original attractors including Th0 , Th1 , Th2 , Th17 , iTreg , Tfh , TrFoxp3- , and Th9 , but it also predicts novel attractors expressing RORγt+IL-10+ ( Th17R ) and GATA3+IL-10+IL-9+ ( Th2RIL9+ ) , thus confirming the importance of SOCS proteins for attaining the Th17 and Th9 cell types . The importance of IL-10 for CD4+ T cell plasticity dramatically increased in the altered network , while the importance of the rest of the molecular elements decreased . This result suggests that SOCS proteins play an important role in stabilizing effector cell types and regulating the Th0 and TrFoxp3- cell types . SOCS proteins inhibit signal transduction; IL-10 in particular acts through these proteins to regulate CD4+ T cells . This regulation is important to buffer the effect of extrinsic cytokines in the TSRN network model . When SOCS proteins are absent , the network is more sensitive to changes in extrinsic cytokines and IL-10 . Further analyses of the effects of SOCS proteins on CD4+ T cells and the possibility of updates to the model based on experimental work should enable the evaluation of more subtle alterations in and combinations of SOCS proteins .
This model provides a mechanistic description of the way in which CD4+ T cell types and plasticity emerge from the interactions among the intrinsic and extrinsic components of the immune response . The study formally shows that , as expected , the interactions among master transcription factors considered in the TSRN are not sufficient to recover the configurations characteristic of CD4+ T cell types , nor the reported transition patterns . Furthermore , these results clearly demonstrate the necessity to include the feedback from signaling pathways in response to cytokines to recover most of the range of CD4+ T cell types ( Th0 , Th1 , Th2 , Th17 , Tfh , Th9 , iTreg and T regulatory Foxp3 independent cells ) and their transition pathways . As noted above , CD4+ T cell differentiation does not arise solely from the regulatory action of the core of the reported "master" transcription factors ( TF ) : T-bet , GATA3 , Foxp3 , RORγt and Bcl6 . This may be due to the lack of feedforward loops mediated by the transcription factors RORγt [48] and Bcl6 [49 , 50] . These results show that the transcriptional regulatory core of CD4+ T cell differentiation is necessary , but not sufficient for CD4+ T cell differentiation . The emergence of the different CD4+ T cell types and their transition patterns , requires the feedback from cytokine signaling pathways and external cues . This model provides a formal test for the emergence of different CD4+ T cell types from feedback or cooperative dynamics among master transcriptional factors , signaling pathway , cytokines produced by the cell and those present in the micro-environment . The proposed model recovers the observed configurations for the following CD4+ T cell types: Th0 , Th1 , Th2 , Th17 , Tfh , Th9 , iTreg and T regulatory Foxp3 independent cells [2 , 4–7] . The model also yields the cell types Tfh , Th9 and T regulatory Foxp3 independent cells that had not been previously incorporated into such models [37–40] . CD4+ T cell types depend on signals from other cells for their differentiation and maintenance . The cytokines in the micro-environment restrict which cell types and transitions can be attained . A cytokine micro-environment that promotes a particular cell type increases its attraction basin size , stability and increases the number of transitions towards the promoted cell type . Nonetheless , different CD4+ T cell types can coexist in micro-environments that do not promote all the present cell types . For example , the presence of pro-regulatory cytokines IL-10 and TGF-β inhibits most effector cells , except for Th17 . This finding may explain the presence of Th17 cells in regulatory micro-environments [59] and provides important insights concerning the relationship between Th17 and iTreg cells and the paradoxical role of TGF-β in inflammation [61] . Thus , the type of modeling framework and analyses presented here may prove to be useful for efforts to find therapeutic approaches to address chronic inflammation . The model was also used to analyze the plasticity of CD4+ T cells by systematically testing how transient perturbations affect the transition patterns among cell types under various micro-environments . Previous studies focused on cell plasticity in response to different micro-environments [38] or on the role of specific molecules [40] , rather than studying these phenomena as consequences of the global properties of the system . For example , the TSRN faithfully captures the polarization of resting CD4+ T cells into Th1 , Th2 , Th17 , iTreg and Foxp3-independent T cells , but the predicted cell-fate maps lack direct transitions from iTreg to Th17 and Th17 to Th1 [23–25] . The TSRN model may lack components , interactions or epigenetic mechanisms of regulation that are important to enabling such direct transitions [33] . An additional possibility is that signals must be combined during particular lengths of time to enable some transitions . Further theoretical and experimental research is required to understand the mechanisms underlying CD4+ T cell plasticity . However , the qualitative model proposed here can serve as a framework to incorporate additional details involved in CD4+ T plasticity . Our model shows that the activation of specific CD4+ T cell transcriptional-signaling regulatory network nodes generally induce cell type plasticity while inhibitions induce stability . The observed response patterns may be caused by the feedback loops and mutual inhibitions molecular network . These findings are coherent with the fact that the immune system generates a specific immunological response to particular challenges , maintains this response while the challenge remains present , and finally downregulates the immune response once the challenge has passed , thus maintaining homeostasis [3 , 61] . Our model suggests that T-bet , TGF-β and SOCS proteins are key network components to recover the observed CD4+ T cell plasticity . Although T-bet is a key transcription factor for Th1 , it also inhibits other transcription factors regulating the differentiation into different cell types [4] . TGF-β is a critical regulator of the immune response but also plays a key role during chronic inflammatory responses [61] . SOCS proteins regulate the phosphorylation of STAT proteins , playing a key role in modulating the signal transduction among different cell types [31 , 51] . Determining the key elements enabling cell-type plasticity has possible therapeutic implications , as these findings can help to identify therapeutic targets for modulating the immune response while predicting and avoiding secondary effects[3 , 62] . Given the complexity of CD4+ T cell expression patterns and transitions , it remains unclear whether cytokine expression profiles correspond to lineages or subsets [1–3 , 22] . The term lineage implies the stability of the cellular phenotype and that the cell has committed to an expression pattern and will maintain it in a fairly robust manner , regardless of environmental alterations . On the other hand , the term subset implies that the cell has a specified expression pattern but that extrinsic signals are required to maintain that pattern [1 , 22] . Cell types Th1 , Th2 , Tfh and TrFoxp3- can be considered lineages , as they exhibit commitment under different cytokine milieus , even if the extrinsic signals change , although environmental alterations can still affect their stability . However , Th17 , iTreg and Th9 cells , which require TGF-βe , IL-2e or IL-4e respectively , are potentially subsets . Th17 and iTreg cells also have small basins of attraction , low stability , and require extrinsic signals , exhibiting a lack of commitment . Th9 has a larger basin of attraction than Th17 or iTreg , but is less stable and susceptible to environmental alterations . Based on our analyses , we propose that the degree of dependence on extrinsic signals and the stability in response to changes in the micro-environment can provide clearer and more objective criteria to distinguish between CD4+ T cell subsets and lineages . CD4+ T cell differentiation and plasticity arises from the feedback among multiple levels of regulation: transcriptional regulation , signaling pathways and the micro-environment . Studying the molecular network as a dynamic system allows us to understand how the interactions among the components , the topology of the network , and the dynamic functions of the nodes give rise to the biological behavior . However , further theoretical and experimental research is required to understand CD4+ T cells . As our understanding of these cells improves , it will be possible to incorporate more detailed molecular information , such as the effect of relative expression levels and the characteristic time courses of expression in the system . This will , in turn , allow us to recover novel cell types and their relationship with other CD4+ T cell types and other cells of the immune system . The present model can now be extended to incorporate multiple cells and their population dynamics [39] , relationships with other cells of the immune system , and the formation of specialized niches that result from the dynamic interaction with the micro-environment . This approach will allow us to differentiate between CD4+ T cell subsets and lineages , to understand the developmental dynamics between the different cell types , and to propose approaches to immune system reprogramming that can be used in the clinic .
CD4+ T cell differentiation results from interactions among cytokines , signaling pathways and transcription factors . These interactions were modeled using Boolean networks that enabled us to integrate the qualitative nature of complex regulatory systems . A Boolean network is composed of nodes that represent the system´s molecular components ( i . e . , cytokines , signaling pathways or transcription factors ) . In a Boolean network , each node represents a component ( gene , protein , phenomenological signal ) that can be associated with a discrete variable denoting its current functional level of activity . If the node is functional its value is 1 , and if it is not functional , then its value is 0 ( see S1 File ) . Some nodes required special considerations concerning their activation states in the Boolean model . For example , in the case of GATA3 , which is continuously expressed during T-cell-lineage development and is necessary for lineage commitment and maintenance , GATA3low is set to 0 . As GATA3 is upregulated in Th2 differentiation [63] , we set GATA3high to 1 . Another example concerns STAT proteins , which are activated when the protein is phosphorylated , forming a dimer that translocates to the nucleus , where it activates its target genes . In this case , the value for STAT protein activation was only set to 1 when all the required conditions were met . The value of a node xi at a time t depends on the value of the input nodes ( including itself ) , referred to as its regulators . This value can be expressed with a logical function that describes the behavior of the node through time: xi ( t ) =ϕι ( τ , ξ1 , ξ2 , ξ2 , … , ξι , … , ξν ) . Weak interactions that are not necessary or sufficient , but only modulate a target factor , were not included in the input regulators of the truth tables ( S1 File ) . Such is the case for Foxp3 , which positively modulates the expression of IL-2Rα , which can be activated and functional in the absence of Foxp3 [64] . An input is a node that affects the values of the network but is independent of the network . The state of the network S can be represented by a vector that specifies the value of each node . The state of the network can be represented by a vector S composed of the values of all the nodes of the system . The state of the network corresponds to the expression patterns of a cell . To facilitate the analysis of the network and determine which components were necessary and sufficient to recover observed profiles and their patterns of transition , we reduced the extended regulatory network consisting of 85 nodes to one with 18 nodes , including 5 transcription factors , 7 signaling pathways and 6 extrinsic cytokines . To simplify the network , we assumed that the signal produced by the TCR and its co-factors was constitutive and ignored weak interactions as well as input and output nodes . Considering that the expression level of node xi at time t is represented by xi ( t ) , the attractors ( steady states ) that represent different phenotypes are determined by: xi ( t+1 ) = xi ( t ) . In that case , the mapping becomes a set of coupled Boolean algebraic equations . The explicit expressions of the attractors are then obtained by performing the algebraic operations according to the axioms of Boolean algebra [44] . Self-regulated nodes were not removed . If a node was removed , then the logical rules of its targets were modified , maintaining the regulatory logic and indirect regulation . To verify that we did not remove a necessary node , we recovered the attractors of the network and ensured that the configurations corresponding to the Th0 , Th1 , Th2 , Th17 and iTreg states could still be attained ( see the details of the reduction methods used in S1 File ) . The reduction was verified using the GINsim[45] software . GINsim uses decision diagrams to iteratively remove regulatory components and updates the components to maintain the indirect effects . This method preserves the dynamic properties of the original model . The simplification with GINsim returned a similar network to the one that we obtained with the Boolean logic reduction method proposed by Villareal et al . ( [44];S1 File ) . After inferring and simplifying the network , we studied its dynamic behavior . A regulatory network is a dynamic system . The state of a network will change over time depending on the logical functions associated with each node . When the values of a state vector S at t+1 are the same as those at time t , the system has attained an attractor: S* ( t ) = S ( t + n ) , n ≥ 1 . An attractor is interpreted as a stable expression phenotype of a cell , representing a cell type . All the states that lead to a solution S* constitute the basin of attraction of such an attractor . We determined the attractors and basins of attraction of the network using the R library BoolNet . Attractors were classified depending on the expression of both the master transcription factors and the main cytokine . Th0 was defined as expressing no transcription factors or regulatory cytokines . Th1 was defined as Tbet and IFN-γ active [8] , Th2 as GATA3 and IL-4 active [8] and GATA3+ ( a Th2-like cell type ) as GATA3+IL4-[38] . Th17 was identified by RORγt and STAT3 signaling mediated by IL-6 or IL-21 , all of which require the presence of TGF-βe [9–10] . The iTreg type was defined by Foxp3 and TGF-β , IL-10 or both , all of which require the presence of IL-2e [16] . Tfh cells were defined by Bcl6 and STAT3 signaling mediated by IL-21 [12] . Th9 cells express IL-9 , requiring the presence of TGF-β and extrinsic IL-4 [27] . T regulatory Foxp3-independent CD4+ T cells ( TrFoxp3- ) featured TGF-β , IL-10 or both , without expressing Foxp3 [52] . Th1 regulatory cells ( Th1R ) express a regulatory cytokine and T-bet [46] . Th2 regulatory cells ( Th2rR ) express a regulatory cytokine and GATA3 [47] . The attractors of the network correspond to cell types . A multi-stable system can have multiple attractors and switch between them in response to alterations in the state of the system [65] . To study the plasticity and robustness of the system we transiently perturbed the attractors of the network and then evaluated the functions until we arrived at an attractor . This methodology enabled us to obtain all the transitions between cell types , the specific perturbations that caused those transitions , and the path from one cell-type to another . We define an attractor as stable when the system remains in the same attractor in the presence of perturbations . The stability of each attractor in response to changes in the micro-environment and signaling pathways was analyzed by characterizing the evolution of the network in response to pulses of activation or inhibition of specific nodes . To quantify the stability of the attractors of the network , we perturbed the state vector of the solutions for one time step . Then , we counted how many of the perturbed state vectors stayed in the same attractor to quantify its stability . A system is plastic when it can transition from one state to another in response to alterations of the system . More specifically , the network was said to be plastic when a transition occurred from a given attractor to another in response to a transient perturbation in the value of one of its nodes . | CD4+ T cells orchestrate adaptive immune responses in vertebrates . These cells differentiate into several types depending on environmental signals and immunological challenges . Once these cells are committed to a particular fate , they can switch to different cell types , thus exhibiting plasticity that enables the immune system to dynamically adapt to novel challenges . We integrated available experimental data into a large network that was formally reduced to a minimal regulatory module with a sufficient set of components and interactions to recover most CD4+ T cell types and reported plasticity patterns in response to various micro-environments and transient perturbations . We formally demonstrate that transcriptional regulatory interactions are not sufficient to recover CD4+ T cell types and thus propose a minimal network that induces most observed phenotypes . This model is robust and was validated with mutant CD4+ T phenotypes . The model was also used to identify key components for cell differentiation and plasticity under varying immunogenic conditions . The model presented here may be a useful framework to study other plastic systems and guide therapeutic approaches to immune system modulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | A Minimal Regulatory Network of Extrinsic and Intrinsic Factors Recovers Observed Patterns of CD4+ T Cell Differentiation and Plasticity |
Invasion by infectious pathogens can elicit a range of cytokine responses from host cells . These cytokines provide the initial host defense mechanism . In this report , we demonstrate that TNF-α , a pro-inflammatory cytokine , can be induced by hepatitis C virus ( HCV ) in its host cells in a biphasic manner . The initial induction of TNF-α by HCV was prompt and could be blocked by the antibody directed against the HCV E2 envelope protein and by chemicals that inhibit endocytosis , indicating the specificity of endocytic uptake of HCV in this induction . Further studies indicated that the induction of TNF-α was dependent on toll-like receptors 7 and 8 ( TLR7/8 ) but not on other intracellular pattern recognition receptors . Consistently , siRNA-mediated gene silencing of the downstream effectors in the TLR7/8 signaling pathway including MyD88 , IRAK1 , TRAF6 , TAK1 and p65 NF-κB suppressed the expression of TNF-α . The role of p65 NF-κB in the induction of TNF-α via transcriptional up-regulation was further confirmed by the chromatin immunoprecipitation assay . TNF-α induced by HCV could activate its own receptor TNFR1 on hepatocytes to suppress HCV replication . This suppressive effect of TNF-α on HCV was due to its role in supporting interferon signaling , as the suppression of its expression led to the loss of IFNAR2 and impaired interferon signaling and the induction of interferon-stimulated genes . In conclusion , our results indicate that hepatocytes can sense HCV infection via TLR7/8 to induce the expression of TNF-α , which inhibits HCV replication via an autocrine mechanism to support interferon signaling .
Hepatitis C virus ( HCV ) is an enveloped virus with a single-stranded RNA genome of 9 . 6-Kb [1] . After binding to its receptors on hepatocytes , HCV is internalized by receptor-mediated endocytosis , and its genomic RNA is subsequently released into the cytosol to direct the synthesis of viral proteins using the internal ribosome entry site ( IRES ) located near its 5’-end . This leads to the production of a polyprotein with a length of approximately 3000 amino acids . The HCV polyprotein is proteolytically cleaved by host and viral proteases to give rise to individual viral proteins including the core protein , E1 and E2 envelope proteins , the p7 viroporin , and nonstructural proteins NS2 , NS3 , NS4A , NS4B , NS5A , and NS5B [2] . Pattern recognition receptors ( PRRs ) including toll-like receptors ( TLRs ) and RIG-I-like receptors are important components of the innate immune response . Upon the activation by the pathogen-associated molecular patterns ( PAMPs ) , these PRRs induce the expression of various cytokines via the downstream signaling pathways . Some TLRs are located on the cellular surface and sense extracellular PAMPs and some TLRs are located in the endosomes to detect internalized pathogens [3] . The TLR signaling is mediated by the TIR domain-containing cytosolic adaptors MyD88 , TIRAP/Mal and TRIF . The initial association of MyD88 to the receptor leads to the sequential recruitment and activation of IRAK4 and IRAK1 . The activated IRAK1 then binds to TRAF6 , after which the complex dissociates from the receptor for further signaling events including the activation of TAK1 . TAK1 can activate NF-κB and AP1 to stimulate the production of pro-inflammatory cytokines . IRAK1 and TRAF6 can also activate IRF7 to induce the expression of type I interferons ( IFNs ) [4 , 5] . Tumor necrosis factor-α ( TNF-α ) is a pro-inflammatory cytokine produced in response to infectious pathogens . The soluble TNF-α is produced as a result of cleavage from its precursor transmembrane TNF-α by the TNF-α-converting enzyme ( TACE ) . The secreted TNFα binds to its receptors , namely TNFR1 and TNFR2 , to exert its biological effects [6] . Multiple studies indicate that the blood level of TNF-α is increased in HCV patients and its level is positively correlated with HCV pathogenesis and the severity of liver diseases [7–9] . The major source of TNF-α in response to HCV infection is unclear and thought to be immune cells such as T lymphocytes and macrophages [10 , 11] . In this report , we provide evidence to demonstrate that hepatocytes can also produce TNF-α in response to HCV infection . This TNF-α induction is prompt and mediated by TLR7 and TLR8 . Furthermore , we also demonstrate that TNF-α , through an autocrine mechanism , prevents the depletion of IFNAR2 by HCV and is required to support interferon signaling in HCV-infected cells .
To determine whether HCV infection can directly induce the expression of TNF-α in its host cells , we infected Huh7 hepatoma cells with a cell culture-adapted HCV JFH1 variant using a multiplicity of infection ( MOI ) of 0 . 25 and collected the incubation media at different time points after infection for quantification of TNF-α using ELISA . The soluble TNF-α was initially detectable at 48 hours post-infection and its level further increased at 72 hours ( Fig 1A ) . When the quantitative RT-PCR ( qRT-PCR ) was used to analyze the expression of TNF-α RNA in cells , a similar induction profile was observed ( Fig 1A ) , although a ~10-fold induction of TNF-α was also observed at 24 hours ( see below ) . The induction of TNF-α in Huh7 cells could be detected by immunoblot at 24 hours post-infection when the MOI used was 1 or higher ( Fig 1B ) . To ascertain that the TNF-α induction by HCV was not specific to Huh7 cells , we also infected primary human hepatocytes ( PHH ) with HCV . As shown in Fig 1C , HCV infection of PHH also induced the expression of TNF-α after 24 hours when a semi-quantitative RT-PCR was used for the analysis . To determine how early after infection HCV could induce the expression of TNF-α , we analyzed the induction of TNF-α by HCV within the first 24 hours of infection using an MOI of 2 . As shown in Fig 2A , HCV could induce the expression of TNF-α as early as one hour post-infection , when the semi-quantitative RT-PCR was used for the analysis . This induction was increased at 2 hours , reduced at 4 and 8 hours and increased again at 24 hours ( Fig 2A ) . We also used qRT-PCR to analyze the effect of MOI on the induction of TNF-α and found that the induction of TNF-α by HCV was slight at 2 hours but significant ( ~10-fold ) at 24 hours post-infection when the MOI used was 0 . 25 ( Fig 2B ) . However , when the MOI of 1 was used , the fold induction of TNF-α at 2 hours and 24 hours post-infection was similar at about 15 , indicating a dose-effect of HCV on the induction of TNF-α at the early time points of infection . To rule out the possibility that the early induction of TNF-α was due to nonspecific factors in the HCV inoculum , we treated the HCV inoculum ( MOI = 1 ) with an anti-E2 antibody , which neutralized the infectivity of HCV [12 , 13] . As shown in Fig 2C , this neutralization antibody reduced the TNF-α RNA in Huh7 cells to almost the basal level , confirming the specificity of TNF-α induction by HCV . In addition , the early induction of TNF-α was inhibited , if cells were treated with actinomycin-D , an inhibitor of RNA synthesis , prior to HCV infection ( Fig 2D ) , indicating a transcriptional up-regulation of TNF-α . The effect of actinomycin-D on TNF-α was unlikely due to its effect on HCV entry , as this treatment slightly increased the HCV RNA level in cells ( S1 Fig ) . Considering that the induction of TNF-α occurred almost immediately after HCV infection , it did not appear likely that the translation or the replication of HCV genome RNA was involved . To test this possibility , we used UV-irradiation to inactivate HCV prior to infection . This UV inactivation did not inhibit the induction of TNF-α at two hours post infection ( Fig 2E ) , indicating that the integrity of the HCV genome was not essential for the induction of TNF-α at this early time point . However , the UV-inactivation of HCV reduced the second-phase induction of TNF-α at 24 hours , indicating that the HCV gene expression and/or replication was required for the efficient induction of TNF-α at this later time point . Besides TNF-α , the induction of other cytokines including IL-6 and IL-1β was also observed in both Huh7 cells and PHH at 2 hours post-infection ( S2 Fig ) . The finding that the induction of TNF-α was detected almost immediately after infection without the apparent need of an intact HCV genome suggested an early event of HCV infection in the induction of TNF-α , possibly during the viral entry . After binding to its co-receptors , HCV enters the cell via the clathrin-mediated endocytosis [1 , 14] . To test whether this endocytic uptake is required for HCV to induce TNF-α , we treated Huh7 cells with Dynasore , a cell-permeable inhibitor of dynamin GTPase , which mediates the scission of clathrin-coated vesicles from plasma membranes . As shown in Fig 3A , the inhibition of dynamin with Dynasore significantly inhibited the induction of TNF-α by HCV . This result suggested an important role of endocytic uptake of HCV in the induction of TNF-α . After the dissociation of clathrin , endocytic vesicles fuse with endosomes . The acidic content of endosomes then triggers the fusion of HCV envelope with endosomal membranes for the release of the viral genome into the cytosol . To examine the possible importance of endosomal acidification in the induction of TNF-α , we treated cells with chloroquine , an inhibitor of endosomal acidification . As shown in Fig 3B , the treatment of Huh7 cells with chloroquine abolished the TNF-α induction by HCV . Altogether , these results suggested that both the scission of endocytic vesicles from plasma membranes and the endosomal acidification were important for the induction of TNF-α by HCV . RIG-I and MDA5 are two cytosolic PRRs that recognize double-stranded RNAs ( dsRNAs ) , and the former has been shown to recognize the 3’-end UC-rich sequence of the HCV RNA [15] . To test the possible roles of these two PRRs in the induction of TNF-α by HCV , we performed the siRNA knockdown experiment to suppress the expression of these two proteins prior to HCV infection . As shown in S3A Fig , the suppression of RIG-I and MDA5 expression in Huh7 cells had only a marginal effect , if any , on the induction of TNF-α . The lack of effect of RIG-I on the induction of TNF-α by HCV was also confirmed by the infection of Huh7 . 5 cells with HCV . Huh7 . 5 cells were derived from Huh7 cells and expressed a defective RIG-I [16] . As shown in S3B Fig , HCV could induce TNF-α in Huh7 . 5 cells , confirming that RIG-I was not essential for the expression of TNF-α . Due to the lack of significant effect of RIG-I and MDA5 on the induction of TNF-α by HCV , we turned our attention to TLR3 , TLR7 , TLR8 and TLR9 , which are PRRs that reside in endosomes . Among them , TLR3 is activated by dsRNA; TLR7 and TLR8 , which share a high degree of structural similarity and are functionally active in Huh7 cells ( S4 Fig ) , are activated by single-stranded RNA ( ssRNA ) ; and TLR9 is activated by the unmethylated CpG motif of DNA [3] . To test the possible roles of TLR7 and TLR8 in the induction of TNF-α by HCV , we also performed the siRNA-knockdown experiments . HCV infection induced the expression of TNF-α at two hours post-infection and the simultaneous knockdown of both TLR7 and TLR8 ( TLR7/8 ) impaired this induction ( Fig 4A ) . It is not likely that the knockdown of TLR7/8 impaired HCV entry , as their simultaneous knockdown increased HCV RNA levels in cells at 24 hours post-infection ( S5A Fig ) . The single knockdown of TLR7 or TLR8 had only a marginal effect on the induction of TNF-α ( S5B Fig ) . This lack of significant effect of single knockdown of TLR7 or TLR8 on TNF-α was likely due to the compensatory increase of the expression of the other when the expression of either one of these two TLRs was suppressed ( S5C Fig ) . In contrast to TLR7/8 , the knockdown of TLR3 and TLR9 as well as TLR4 , which recognizes lipopolysaccharides , had no apparent effect on the induction of TNF-α by HCV ( S5D Fig ) . Note that HCV infection increased TLR7 and TLR8 RNA levels ( Fig 4A and S5B Fig ) and appeared to also slightly increase the TLR3 and TLR9 levels ( S5D Fig ) . The reason for this is unclear , but the induction of TLR7 and TLR8 was apparently mediated by NF-κB , as the knockdown of p65 , a subunit of NF-κB , largely abolished their induction by HCV ( S6 Fig ) . If HCV indeed induced the expression of TNF-α via TLR7 and TLR8 , then the suppression of expression of their downstream adaptor molecules should also inhibit the induction of TNF-α by HCV . As shown in Fig 4B , the suppression of expression of MyD88 , IRAK1 , TRAF6 or TAK1 , which mediates TLR7/8 signaling , all led to the reduction of TNF-α induction by HCV . The knockdown efficiency of these adaptor molecules was shown in S7 Fig . TAK1 activates the transcription factor NF-κB by phosphorylating IκB kinase β ( IKKβ ) . Its role in the induction of TNF-α by HCV was further confirmed by the observation that Celastrol , an inhibitor of the TAK1 kinase , impaired the induction of TNF-α by HCV ( Fig 4C ) . These adaptor molecules are not known to affect HCV entry and thus it is unlikely that their knockdown led to the reduction of TNF-α via the inhibition of HCV entry . Indeed , as shown in S8 Fig , the knockdown of TRAF6 did not decrease but rather increased the HCV RNA level . Taken together , our results demonstrated that HCV induced the expression of TNF-α via TLR7/8 and its downstream signaling molecules including MyD88 , IRAK1 , TRAF6 , and TAK1 . TAK1 activates IKKβ , which phosphorylates and destabilizes the NF-κB inhibitor IκB to result in the nuclear translocation of NF-κB . To determine whether NF-κB was indeed activated in the early time point of HCV infection , we performed the subcellular fractionation experiment . As shown in Fig 5A , HCV infection indeed induced the nuclear translocation of p65 , a subunit of NF-κB , at 2 hours post-infection . To further test the role of NF-κB in the activation of the TNF-α gene , we performed the chromatin immunoprecipitation ( ChIP ) assay to determine the binding activity of p65 NF-κB to the TNF-α promoter , which contains the NF-κB binding site [17] . As shown in Fig 5B , an enhanced binding of p65 NF-κB to the TNF-α promoter was observed in HCV-infected cells at 2 hours post-infection . The binding of p65 NF-κB to the TNF-α promoter was also detected at 24 hours , albeit to a lesser degree . To further verify the role of p65 in TNF-α induction , we performed the p65 NF-κB knockdown experiment . As shown in Fig 5C , p65 NF-κB knockdown reduced the ability of HCV to induce the expression of TNF-α at both 2 hours and 24 hours post-infection . Consistently , Bay-11-7085 , a chemical that inhibits the phosphorylation of IκBα and the activation of NF-κB , reduced the TNF-α expression ( Fig 5D ) . Taken together , our results clearly demonstrated a role of NF-κB in the induction of TNF-α by HCV . Note that previous studies indicated that HCV could induce oxidative stress , which could activate NF-κB [18 , 19] . However , it does not appear likely that oxidative stress was involved in the induction of TNF-α during the early time points of HCV infection , as the treatment of cells with the antioxidant N-acetylcysteine ( NAC ) had little effect on the induction of TNF-α by HCV at 2 hours and 48 hours post-infection ( S9 Fig ) . To determine whether TNF-α induced by HCV could directly affect HCV replication , we knocked down the expression of TNF-α using its specific siRNA prior to HCV infection . As shown in Fig 6A , the inhibition of TNF-α expression increased the levels of intracellular HCV RNA as well as the level of the HCV core protein , comparing with cells treated with the control siRNA . It also increased the HCV yield , as evidenced by the significant increase of HCV-positive cells when the progeny virus in the incubation media was harvested and used to infect naive cells ( Fig 6B ) . To confirm that TNF-α could indeed suppress HCV replication , we also treated Huh7 cells with recombinant TNF-α , which also reduced the HCV RNA level ( S10 Fig ) . Soluble TNF-α exerts its effect through its receptors TNFR1 or TNFR2 . TNFR1 is expressed in most cell types and believed to be responsible for most of the biological effects of TNF-α , while the expression of TNFR2 is primarily limited to endothelial cells and cells of hematopoietic lineages [20] . To understand how TNF-α exerted its inhibitory effect on HCV , we analyzed the expression of TNFR1 in HCV infected cells . As shown in Fig 6C , HCV caused the loss of TNFR1 at 24 hours post-infection . This loss of TNFR1 was a post-transcriptional event , as the TNFR1 RNA level was not affected by HCV ( S11 Fig ) . As TNFR1 , upon binding to TNF-α , is internalized by receptor-mediated endocytosis and degraded in lysosomes [21 , 22] , this result suggested an autocrine activation of TNFR1 . To test this possibility , we first treated naive Huh7 cells with TNF-α . This treatment indeed caused the loss of TNFR1 , which could be restored if cells were treated with Bafilomycin-A1 ( BafA1 ) ( Fig 6D ) , which inhibits the vacuolar ATPase activity and endocytic protein degradation in lysosomes [23] . We then treated HCV-infected cells at 24 hours post-infection with BafA1 . As shown in the same figure , BafA1 also restored the TNFR1 protein level in HCV-infected cells . These results were consistent with the degradation of TNFR1 in lysosomes . Similar to the TNF-α depletion , silencing TNFR1 with its siRNA also increased HCV RNA and core protein levels ( Fig 6E ) . These results strongly supported the argument that TNF-α suppressed HCV replication via an autocrine mechanism . TNF-α unlikely suppressed HCV replication via the induction of apoptosis , as the knockdown of its expression using the siRNA enhanced , rather than suppressed , apoptosis of HCV-infected cells ( S13 Fig ) . HCV infection can induce a modest level of type I interferons , which stimulate the expression of interferon stimulated genes ( ISGs ) [24–26] . To understand how TNF-α suppressed HCV replication , we analyzed the possible effect of TNF-α on IFN signaling in HCV-infected cells using a firefly luciferase reporter linked to the interferon-stimulated response element ( ISRE ) . As shown in Fig 7A , in agreement with the previous reports [24–26] , HCV infection slightly increased the ISRE activity and this increase was reduced to below the background level when TNF-α was depleted with the siRNA . The treatment of HCV-infected cells with IFN-α significantly increased the ISRE activity . Similarly , this increase was impaired , if TNF-α was depleted . The effect of TNF-α on ISRE was confirmed by analyzing the expression of OAS1 , ISG56 and MxA , three IFN-stimulated genes ( ISGs ) . As shown in Fig 7B , HCV could also increase the expression of these ISGs in Huh7 cells . This result was consistent with the previous reports that HCV could induce a low level of interferon response [24–26] . However , this induction was largely abolished , if the expression of TNF-α was inhibited with the siRNA . These results indicated that TNF-α , produced by the cells in response to HCV infection , was required to support interferon signaling and the expression of ISGs in HCV-infected cells . This is likely how TNF-α suppressed HCV replication . Type I IFNs bind to the IFN-α receptor ( IFNAR ) , which is composed of two subunits IFNAR1 and IFNAR2 . This binding activates Janus kinase 1 ( JAK1 ) and tyrosine kinase 2 ( TYK2 ) , which then phosphorylate and activate STAT1 and STAT2 to result in the activation of ISRE in the promoters of ISGs . To understand why TNF-α was required to support the expression of IFNs and ISGs , we analyzed the effect of TNF-α on the activation of STAT1 and STAT2 . As shown in S14B Fig , HCV did not apparently affect the phosphorylation of STAT1 and STAT2 at 24 hours post-infection , but it slightly increased the STAT1 phosphorylation at 48 hours post-infection , in agreement with its modest effect on ISRE . The phosphorylation of both STAT1 and STAT2 was clearly visible after the treatment with IFN-α . Interestingly , when TNF-α was depleted with its siRNA , the phosphorylation of STAT1 induced by IFN-α was significantly inhibited in HCV-infected cells ( Fig 7C ) . TNF-α could also enhance the phosphorylation of STAT1 and the expression of ISGs induced by IFN-α in naive Huh7 cells ( S14C Fig ) . To further investigate why the activation of STAT1 by IFN-α was inhibited when TNF-α was depleted , we analyzed the expression levels of IFNAR1 and IFNAR2 in HCV-infected cells . As shown in Fig 7D , the depletion of either TNF-α or TNFR1 led to the loss of IFNAR2 , but not IFNAR1 , in HCV-infected cells . This loss of IFNAR2 was a post-transcriptional event , as the level of IFNAR2 mRNA was not apparently affected by the depletion of TNF-α or TNFR1 ( S14D Fig ) , and most likely mediated by proteasomes , as its loss could be inhibited by the proteasome inhibitor MG132 but not by Bafilomycin A1 ( S15 Fig ) . Taken together , the results shown in Fig 7 indicated that TNF-α induced by HCV was required to maintain the IFNAR2 expression level to support IFN signaling . In the absence of TNF-α , IFNAR2 was depleted by HCV and IFN signaling was impaired .
HCV patients have an elevated serum level of TNF-α , and this level is positively correlated with the severity of liver diseases [7–9] . The source of TNF-α is unclear , but it is generally assumed that it is produced by immune cells such as macrophages [27] . In this report , we demonstrated that TNF-α could also be induced in HCV-infected cells . Although the amount of TNF-α produced by HCV-infected hepatocytes might be lower than that produced by professional immune cells such as macrophages [28] , it was sufficient to trigger an inhibitory response on HCV replication . Our finding is consistent with a previous report , which described an increased level of TNF-α in the hepatocytes of HCV patients [9] . The induction of TNF-α by HCV was specific , as it could be blocked by the antibody that neutralized the infectivity of HCV ( Fig 2C ) . This induction was biphasic , with the first phase of induction peaked at 2 hours post-infection ( Fig 2A ) . The induction of TNF-α in the first phase was dependent on TLR7/8 ( Fig 4A ) and required no HCV gene expression or replication ( Fig 2E ) . As TLR7 and TLR8 are activated by ssRNA , HCV either has to release the viral genomic RNA into endosomes during endocytosis to activate TLR7/8 or the HCV genomic RNA released into the cytosol after uncoating must be delivered immediately back into the endosomes . We favor the first scenario , as , if HCV RNA is released first into the cytosol , then it will likely also activate RIG-I and/or MDA5 , which are cytosolic PRRs . However , we found that these two PRRs did not play a significant role in the induction of TNF-α ( S3A Fig ) . If HCV indeed activates TLR7/8 during endocytosis , then the HCV virion must be disintegrated during this process for the genomic RNA to be released . This may be triggered by the acidic content of endosomes/lysosomes , which may destabilize HCV virion to release the viral RNA . The activation of the TLR7/8 signaling pathway by HCV led to the activation of NF-κB ( Fig 5A ) . This pathway was required for the induction of TNF-α by HCV in the first phase . The induction of TNF-α in the second phase also required NF-κB , as the depletion of p65 NF-κB also suppressed the second-phase induction of TNF-α by HCV ( Fig 5C ) . It does not appear likely that this second-phase induction of TNF-α was due to the second-round of infection by progeny virus particles , as this second-phase induction of TNF-α was long-lasting ( Fig 1A ) . A number of factors had been shown in the past to activate NF-κB in HCV-infected cells . These factors include TLR3 and protein kinase R ( PKR ) , which could both be activated by the double-stranded HCV RNA replicative intermediates [29 , 30] . These factors are likely the reasons why HCV was also able to induce TNF-α in the later phase of infection . TNF-α induced by HCV suppressed HCV replication ( Fig 6A ) . Our results indicated that this was likely due to its role in interferon signaling and the induction of ISGs ( Fig 7A and 7B ) . We found that both TNF-α and TNFR1 participated in IFN signaling by maintaining the stability of IFNAR2 , as in the absence of either one of them , IFNAR2 was lost in HCV-infected cells ( Fig 7D ) , apparently due to degradation by proteasomes ( S15 Fig ) . How HCV induced the degradation of IFNAR2 and how TNF-α antagonized this effect of HCV are interesting questions that remain to be determined . It is noteworthy that the degradation of IFNAR1 and IFNAR2 had previously been shown to be regulated by different mechanisms [31] , and thus the selective degradation of IFNAR2 by HCV without affecting IFNAR1 was not unexpected . Nevertheless , our results reveal an interesting interplay between the virus and the host cell , with the virus attempting to blunt the IFN response by depleting IFNAR2 and the host cell overcoming this blunting effect of HCV by using TNF-α to restore the expression of IFNAR2 . Although our results indicated that TNF-α could support IFN signaling to suppress HCV replication in cell cultures , the role of TNF-α in HCV replication and pathogenesis in vivo may be more complicated . This is due in part to our observation that TNF-α suppressed apoptosis of HCV-infected cells ( S13 Fig ) , which would favor HCV persistence , in part to the pro-inflammatory activities of this cytokine , and in part to a recent report that TNF-α could depolarize liver cells to enhance HCV entry [28] . Thus , it is tempting to speculate that TNF-α induced in the first phase may enhance HCV entry whereas it induced in the second phase may suppress HCV replication . This may explain why in the clinical trial of a limited number of HCV patients , the TNF-α inhibitor Etanercept was found to improve the therapeutic effect of IFN-α and ribavirin on HCV rather than to suppress it [32] .
Huh7 cells were maintained in Dulbecco’s modified eagle medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) . Huh7 . 5 cells were maintained in DMEM supplemented with 10% FBS and 1% nonessential amino acids . Primary human hepatocytes ( PHHs ) were obtained from the Cell Culture Core Facility of the USC Research Center for Liver Diseases . They were maintained in DMEM medium supplemented with what10% FBS . The JFH1 ( HCV genotype 2a ) variant , which produced a high level of infectious virus particles [33] , was propagated in Huh7 . 5 and used in our infection studies . Huh7 cells were infected with HCV with an MOI of 0 . 25 , and the incubation medium was replaced with the fresh medium after 3 hours of infection . The incubation medium was collected at the time points indicated and TNF-α was assayed using the human TNF-α Instant ELISA kit ( eBioscience ) following the instructions of the manufacturer . Cells were lysed in M-PER Mammalian Protein Extraction Reagent ( Thermo scientific ) containing the protease inhibitor cocktail , 1mM PMSF , 1mM sodium orthovanadate , and 1mM sodium fluoride for 10 minutes on ice followed by a brief sonication . Cell lysates were cleared by centrifugation at 14 , 000 x g for 2 minutes . The supernatant was collected , boiled in Laemmli buffer for 5 minutes , and used for immunoblot analysis or stored at −80°C for future use . The rabbit anti-HCV core antibody was prepared in our laboratory [34] . TNFR1 , p65 NF-κB , PARP , Caspase-8 , and IRAK1 antibodies were from Cell signaling , and actin and IFNAR2 antibodies were from Sigma . IFNAR1 antibody was from Abcam , and GAPDH , TLR7 , TLR8 , and TRAF6 antibodies were from Santa Cruz . Horseradish peroxidase ( HRP ) -conjugated goat anti-rabbit and rabbit anti-mouse secondary antibodies were also purchased from Abcam . siRNAs targeting TNF-α , TNFR1 , TLR3 , TLR4 , TLR7 , p65 NF-κB , MyD88 , IRAK1 and TAK1 were from Sigma . siRNAs targeting TLR7 , TLR8 , MDA5 and RIG-I were from Qiagen , and the siRNA targeting TRAF6 was from Santa Cruz . The transfection of siRNA into Huh7 cells was performed using Lipofectamine RNAiMax ( Invitrogen ) . Briefly , Huh7 cells seeded in 100 mm dishes were transfected with 600 pmole siRNA for 6 hours . The transfected cells were further incubated in fresh media for 48 hours prior to infection with HCV . For generation of stable cell line , lentiviral particles were first produced in 293T cells through the coexpression of pLKO . TRC plasmid with shRNA insertion that targets TLR7 or TLR8 and packaging vectors . Lentiviral particles were harvested 60 hours post transfection and filtered . Huh7 cells were then infected with these lentiviral particles and selected with puromycin ( 2 μg/mL ) . Protein lysates were extracted 10 days post selection and immunoblotted with antibodies against TLR7 and TLR8 . Total cellular RNA was isolated using TRIzol ( Invitrogen ) and reverse-transcribed to cDNA using SuperScript II First-Strand Synthesis System for RT-PCR ( Invitrogen ) . The cDNA was then subjected to PCR amplification with the primer sets listed in the table in the supplemental information ( S1 Table ) . For the quantification of HCV RNA , total cellular RNA was subjected to qRT-PCR using the TaqMan EZ RT-PCR Kit ( Applied Biosystems , Foster City , CA ) following the manufacturer’s instructions . HCV JFH1 primers 5′-TCTGCGGAACCGGTGAGTA-3′ ( forward ) and 5′-TCAGGCAGTACCACAAGGC-3′ ( reverse ) and the probe 5′-CACTCTATGCCCGGC CATTTGG-3′ were used for the qRT-PCR . The control GAPDH primer set with the probe was purchased from Applied Biosystems . For detection of other gene expressions , 100 ng total RNA was analyzed using the Power SYBR Green RNA-to-CT 1-Step Kit ( Applied Biosystems , Foster City , CA ) . The primers used are shown in S1 Table . Huh7 cells with or without HCV infection were rinsed once with phosphate-buffered saline ( PBS ) and then trypsinized . Cells were washed once more with PBS before the subcellular fractionation using NE-PER Nuclear and Cytoplasmic Extraction Reagents ( Thermo Scientific ) following the manufacturer’s instructions . Binding of p65 NF-κB to the TNF-α promoter was analyzed using the Abcam ChIP kit following the manufacturer’s instructions . Briefly , Huh7 cells were infected with HCV for the indicated time period . After which , 1 x 106 cells were fixed with 4% formaldehyde for 10 minutes at the room temperature . Cells were then lysed and the chromatin was sheared by sonication for 10 minutes . The chromatin was then immunoprecipitated using the anti-p65 antibody or a control IgG . The immunoprecipitate DNA samples were analyzed for the TNF-α promoter by PCR . | Hepatitis C virus ( HCV ) patients have increased levels of circulating tumor necrosis factor-α ( TNF-α ) . In this report , we demonstrate that HCV can directly induce the expression of TNF-α in hepatocytes in a biphasic manner via NF-κB . The induction of TNF-α by HCV in the first phase is prompt , requires no HCV gene expression and is dependent on TLR7 and TLR8 and their downstream effectors . TNF-α induced by HCV supports interferon signaling via an autocrine mechanism and suppresses HCV replication , as abolishing the expression of TNF-α or its receptor TNFR1 results in the loss of IFNAR2 , a subunit of the type I interferon receptor , and an increase of HCV replication . Our studies thus reveal an interesting interplay between HCV and hepatocytes , with the virus attempting to blunt the IFN response by depleting IFNAR2 and the host cell overcoming this blunting effect of HCV by using TNF-α to restore the expression of IFNAR2 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | TNF-α Induced by Hepatitis C Virus via TLR7 and TLR8 in Hepatocytes Supports Interferon Signaling via an Autocrine Mechanism |
The wide range of time scales involved in neural excitability and synaptic transmission might lead to ongoing change in the temporal structure of responses to recurring stimulus presentations on a trial-to-trial basis . This is probably the most severe biophysical constraint on putative time-based primitives of stimulus representation in neuronal networks . Here we show that in spontaneously developing large-scale random networks of cortical neurons in vitro the order in which neurons are recruited following each stimulus is a naturally emerging representation primitive that is invariant to significant temporal changes in spike times . With a relatively small number of randomly sampled neurons , the information about stimulus position is fully retrievable from the recruitment order . The effective connectivity that makes order-based representation invariant to time warping is characterized by the existence of stations through which activity is required to pass in order to propagate further into the network . This study uncovers a simple invariant in a noisy biological network in vitro; its applicability under in vivo constraints remains to be seen .
Sensory categorization is mediated , at least in part , by brain processes that extract information from the precise points in time at which neurons emit their first few spikes in response to the presentation of a sensory object [1]–[7] . In that context , a particularly attractive candidate representation primitive makes use of the order of neuronal recruitment , computed from the latencies to first spikes . The idea of representation by recruitment order is physiologically and computationally appealing because of its simplicity , rapidity , robustness and ease of implementation [7]–[10] . Spike order-based representation was shown to be applicable in-vivo , under conditions where the input has a fixed temporal order , either because of temporally structured stimulus features ( e . g . , [11] ) or due to unique structure of peripheral receptor tuning curves [8] . Order based representation can also result from an underlying feed-forward network structure ( e . g . , [12] ) . But what if these constraints are relaxed ? Is recruitment order applicable for representing stimuli that are not temporally ordered , in complex large-scale recurrent neural networks ? If applicable , how does it handle trial-to-trial variations in spike times of individual neurons ? How sensitive is it to the temporal resolution of ordering and the number of sampled neurons ? How much of the network's classification capacity is conserved when absolute times of spikes evoked in response to a given stimulus are compacted to vectors of recruitment orders ? The answers to these questions impact on the general applicability of recruitment order as an ensemble neural representation scheme . Here we approach the above questions by examining the capacity of recruitment order to classify between multiple stimuli delivered to a large-scale recurrent network of cortical neurons that develops spontaneously in-vitro; this is an experimental model that matches the generic biophysical nature of the subject matter , and provides exquisite control of relevant variables . Key functional properties of in-vivo networks are conserved in this preparation [13] , including cell types and their electrophysiological characteristics , synaptic and cellular level plasticity , developmental timeline and sensitivities to pharmacological agents . Sensory objects are defined in terms of identities of stimulating electrodes and the evoked neuronal activities are monitored through substrate-embedded array of spatially distributed extra-cellular recording electrodes . We show that recruitment order reliably classifies input sources on a trial-to-trial basis and is invariant to significant temporal changes in absolute spike times of individual neurons . Classification accuracy monotonously increases with the number of sampled neurons , and steeply sensitive to the temporal resolution of spike ordering . The infrastructural origin of rank order representation is interpreted in terms of effective network topology .
For spike-timing based representation to be applicable , latencies between spikes should be consistent in repeated presentations of a given sensory object , and distinctive between different objects . Both requirements strongly depend upon trial-to-trial variability of spike times . While directly stimulated individual cortical neurons can respond very reliably in terms of spike latencies [14] , [15] , evidence for trial-to-trial variability in spike times and spike counts of network-embedded neurons in-vivo abound [16]–[21] . As shown below , a large-scale recurrent network of cortical neurons that develops spontaneously in-vitro presents a similar dichotomy: Latencies to first spikes in “receptive sheath neurons”—i . e . , individual neurons that are directly activated by external stimuli—can be very reliable; in contrast , trial-to-trial variability of latencies and counts is extensive when “downstream” spikes are considered—i . e . , spikes that are generated by propagation of the activity from the receptive sheath neurons deeper into the network . Note that the recurrent nature of the networks implies that a given neuron may , and in most cases does , serve in both groups . We find that at stimulation frequencies of 3 Sec−1 or below , which is the range used in this study , the single spike evoked by each short ( 0 . 4 mSec ) stimulus that is directly applied to a receptive sheath neuron , occurs within less then 6–7 milliseconds following the stimulus . In concurrence with previously reported measurements from cortical neurons in-vitro [22] , [23] , the latency from the stimulus to the response in that range of stimulation frequencies is very reliable ( Figure S1 ) . In and by itself , reliable latency to first spike of directly activated neurons can support representation by recruitment order that is generated by stimulus dynamics ( different receptive neurons activated at different times [8] , [24] ) . But this is not what we are after here; the present study aims at the next level of processing , beyond the receptive sheath . Specifically , we ask how applicable recruitment order representation is downstream to the point of stimulus entry into a network , where spike time reliability is degraded by the dynamics of synapses , intricacies of propagation along axo-dendritic trees and the complexity of recurrent connectivity . Figure 1 shows latencies to first spikes measured downstream to the point of stimulus entry into a network . First spike latencies from 35 spike-sorted units [25] are shown , evoked in response to stimuli invading the network from two different well-defined loci ( “sources” ) , S1 and S2 . To assure that we only look at downstream neurons ( rather than receptive sheath neurons that are directly activated by the electrical stimulation ) , the first 10 mSec following each stimulus were removed from the data . Figure 1 ( Top two panels ) indicates that in spite of the relatively low rate of stimulation ( compare with Figure S1 ) , the latencies to first downstream spikes are severely warped over a range of 100 mSec and more , they wax and wane in a seemingly random , yet constrained manner . Note that the magnitude of time warping for a given neuron depends on the stimulation source . The bottom panel of Figure 1 shows that in many cases the correlation between latencies of different neurons is stimulus source specific . The above time warping is also observed at the population level , provided that activity is not averaged across trials . The average network response , expressed in terms of a population post-stimulus-time-histogram ( pPSTH ) , defined as the average number of spikes recorded throughout the network in a time window of 500 mSec following each stimulus , registered in 1 mSec time bins , shows a characteristic threshold-governed time amplitude trajectory that lasts 0 . 1–0 . 2 Seconds [26] , comparable to numerous observations in-vivo [1] , [27]–[29] . To differentiate from averaged population response , we denote the response of the network to a single stimulation event network spike , and define it in terms of the total number of neuronal action potentials counted over the entire population as a function of post-stimulus time [26] . Figure 2A and 2B show the pPSTH and the underlying variance between network spikes in response to a series of stimuli that were delivered to the network from a single stimulation site at a frequency of 0 . 3 Sec−1 . Trial-to-trial variations appear in the time-delay between the stimulus and the peak of the network spike , as well as in the overall shape of network spikes . Note that the range of temporal variations extends over several tens of milliseconds within which the network response warps , shifts to the right ( longer delays ) and back to the left ( shorter delays ) in a graceful manner or , sometimes , in what seems like a sudden switch between response modes . To appreciate the multiplicity and range of time scales involved , we have sorted the data shown in Figure 2B based on the time-delays from the stimulus to the peak of each network spike ( Figure 2C ) . Note the multiplicity of scales that are involved in the latency from stimulus to the peak and the width and the activity within network spikes , extending from below . 01 Sec−1 up to the ∼40 Sec−1 gamma range . The range and overall nature of population time warping does not depend on the stimulation source . This is demonstrated in Figure 3 that shows time warping of network spikes in response to two series of stimuli , delivered from two different stimulation sites , S1 and S2 . The above non-monotonic changes in absolute time delays between stimulus and neuronal responses at the levels of individual neurons and neuronal populations , set constrains on the capacity of temporal measures to reliably classify input sources on a trial-to-trial basis . In what follows we show how representation by recruitment order , computed from the latencies to first spikes , handles time-warped neuronal responses . For a propagation path ( and hence for a recruitment order ) to be invariant to neuronal response time warping in a large scale recurrent network , one of the following two options must be fulfilled: ( i ) Dynamics of membrane variables and synaptic efficacies are scaled and homogeneously distributed throughout the network; the idea is that under such conditions , paths of less resistance to propagation of activity remain stable . This option , however , is difficult to conceive biophysically and incompatible with previously reported results ( e . g . , [30] ) . ( ii ) Propagation paths , and hence order of recruitment , are constrained by chains of neuronal stations through which activity is required to pass in order to propagate further into the network , regardless of the status of membrane and synaptic dynamics; such stations are natural consequences of physical or effective connectivity that are inherent to the concept of synfire-chain [12] , [31] or certain forms of broadly-distributed network connectivity [26] , [32] ) . To identify chains of neuronal stations in large-scale neuronal networks under time-warping conditions , we have analyzed the recruitment order relationship between all recorded neuronal pairs: Given n neurons , there are n ( n−1 ) different ( i , j ) neuronal pairs ( i = 1 , 2 , 3 , … n; j = 1 , 2 , 3 , … n ) . The first spike times of a pair ( i , j ) , in response to each stimulation event , may appear in one of two orders , i→j and j→i ( we disregard the possibility of complete synchronous occurrence , for the sake of simplicity ) . We measure the probability of j to precede i for all possible pairs . If a given neuron ( or a group of neurons ) is an ideal station through which activity is required to pass in order to propagate further into the network , we expect that it always be preceded by activity of a given subset of neurons , and always be followed by the complement set of neurons . In other words , each genuine station in a chain would act like a bottleneck , separating all other neurons ( or groups of neurons ) into two groups based on the temporal relation between their first spikes and its own: a group of preceding neurons and a group of following neurons . Figure 4A demonstrates three pair-order matrices , generated from responses to three different stimulation sources of one network: The matrices ( each for one of the three stimulation sources ) depict the probability of each neuron to precede every other neuron . Neurons are presented , in each of the matrices , sorted by their average rank . All three matrices clearly show that recruitment is ordered ( see Figure S2 for more examples from different networks ) . The left and middle matrices of Figure 4A demonstrate cases in which our electrodes picked a small number of clear clusters of neurons; the right matrix demonstrates a fairly homogeneous arrangement along the diagonal and no clear clusters of neurons . The variety of matrix forms shown in Figure 4A ( and Figure S2 ) probably reflects the effect of sparse spatial sampling of a common underlying structure that enforces ordered recruitment: a chain of neural stations . Consider the simple three-stations arrangement , X→Y→Z . Let us assume that X , Y and Z are clusters of highly interconnected neurons , and that there is some overlap between these clusters ( i . e . , some neurons in cluster X are also part of cluster Y and so on ) . Cluster X activates cluster Y , which , in turn , activates cluster Z . In that respect , Y is a bottleneck station between X and Z . If cluster Y is outside the electrodes sampling area , the pair-order probability matrix is expected to show sharp separation between clusters , X and Z; the left and middle panels show this type of behavior , where white circles indicate the existence of bottlenecks ( the equivalents of Y ) that reside outside the sampling area . On the other hand , when all ( or most ) of the clusters are within the sampled area , we expect to see a more homogeneous diagonal arrangement ( i . e . a chain of bottleneck stations ) like the one exemplified in the right-hand panel of Figure 4A . Figure 4A also tells us that the rank order of different neurons is stimulus site specific: The small black arrows to the right of the middle panel depict a cluster of neurons that tend to respond close to each other in terms of their recruitment order; each arrow indicates one of these neurons . The dispersion of these arrows in the other ( right and left ) panels indicates that the rank of any given neuron is stimulus site specific . Thus , neurons appear at different ranks in responses to the three different stimulation sites , providing the infrastructure for recruitment order classification of input sources . Figure 4B shows that the three paths shown in Figure 4A do not result from a spatial wave-like propagation of activity across the recording area; rather , propagation paths appear to randomly connect between the recording electrodes , suggesting that the underlying structures are embedded in a non-trivial manner in space . In what follows we show that based on the two key features demonstrated in Figure 4A , ( ordered recruitment , which is stimulus specific ) , rapid and reliable classification of inputs is readily obtained by use of unsupervised as well as supervised algorithms . Throughout this study , we have imposed two constraints on which spikes are considered for analyses; both constraints are meant to avoid trivialization of the order-based classification task: ( i ) we omitted spikes that were evoked up to 10 mSec following each stimulus ( i . e . latency 0 means 10 mSec following a stimulus ) , thus avoiding classification by first spike latencies emitted from receptive sheath neurons . ( ii ) We only considered first spike latencies in neurons that responded to more than 90% of stimuli from all sources , thus making sure that classification is not based on neuronal identities ( in which case the task is trivialized by relying on a neuron , or group of neurons , that has high response probabilities to stimuli delivered from one of the sources , but not from other sources ) . Figure 5 demonstrates the process of extracting recruitment order from a network response to a stimulus . Note that recruitment order is a reduced form of absolute latencies to first spikes , a fact that will become crucial in a later section of this report , where we address the question of how much of the network capacity to classify input sources is lost in this reduction . To test the capability to classify inputs by recruitment order , we start by applying an unsupervised classification algorithm , that is – classification without the need to learn from labeled examples . To that end an order metric was applied , such that the distance between different recruitment orders can be measured: A single character symbolizes each neuron , and words are obtained , each of which represents the first spike order of neuronal recruitment in response to a given stimulus . For example , the word cgbdhefa stands for the order in which 8 neurons ( a–h ) were recruited in response to a given stimulus . The word cagbdhief stands for a response to another stimulus ( from the same or different input source ) , but this time 9 neurons ( a–i ) were recruited to respond . The Levenshtein Edit Distance string metric was used for measuring the distance between any two strings , expressed in terms of the minimum number of editing operations ( insertion , deletion , or substitution ) needed to transform one string into the other . Figure 6 demonstrates classification between two input sources based on the Edit Distance metric . A network was stimulated from two sources ( S1 and S2 ) intermittently , at four different frequency regimens . The top panel of Figure 6 shows the resulting Edit Distance matrix; responses are ordered according to their stimulation source , revealing clusters of similarity that clearly match the two sources of input ( depicted by white lines ) . Note that the four different frequency ( f ) regimens yielded very different population spike counts ( low panel of Figure 6 ) . Yet , the representation by recruitment order remains invariant to these changes . In nine different networks that were challenged with a two-source classification task , cluster analyses ( standard hierarchical algorithm , forced to identify two clusters ) using Edit Distance metric yielded average classification accuracies ranging from 0 . 6 to 0 . 98 ( median = 0 . 72; SD = 0 . 13 ) . The arbitrariness of our choice of Edit Distance metric is acknowledged; to avoid possible bias in the interpretation of our results , other order-based metrics were applied ( e . g . Spearman correlation and Euclidian metrics ) , yielding qualitatively similar results ( data not shown ) . The electrical activity of neuronal networks is expressed in terms of neuronal identities and their absolute spiking times; recruitment order is a dramatically reduced form of that data . How much of the classification capacity is conserved when absolute first spike times , which are evoked in response to a given stimulus , are compacted to lists of recruitment orders ? To answer that question , an estimate for classification capacity should be obtained for both absolute spike times and recruitment order representations . The dimensionality and statistical properties of our experimental data renders simple and direct estimation of classification capacity from response distributions impossible . To circumvent this limitation an alternative approach is adopted , where a lower bound on the classification capacity is estimated by the performance of a general purpose supervised classifier , trained to recognize the sources of stimulation . A Support Vector Machine ( SVM ) with an adaptive Gaussian kernel ( see Materials and Methods for details ) was trained to classify the different input sources based on labeled examples of the data , and the classification capacity was estimated by its performance on a test data set . In nine different networks that were challenged with a two-source classification task , the median accuracy ( test sets only ) obtained by use of absolute time to first spikes at 1 mSec resolution was . 99 ( SD = 0 . 03 ) ; the median accuracy obtained by use of recruitment rank order using 1 mSec resolution is . 98 ( SD = 0 . 12 ) . In six different networks that were challenged with a five-source classification task ( data shown in the context of the subsequent section ) , the median accuracy ( test sets only ) obtained by use of absolute time to first spikes at 1 mSec resolution was . 91 ( SD = 0 . 03 ) ; the median accuracy obtained by use of recruitment rank order using 1 mSec resolution is . 94 ( SD = 0 . 04 ) . ( Note that the slightly better performance of the recruitment order based classifier , compared to absolute time based classifier , reflects the lower dimensionality of the first relative to the latter , and the entailed effect on the sample size required for learning . ) Thus , in a two-source and five-source classification tasks , reduction from spike latencies to recruitment order representation is practically lossless . The classification capacity of these networks in a six input sources task yields similar results ( data shown in the context of the subsequent section ) . Up to this point , the presented classification results were based on all the electrodes that responded beyond the 90% criterion mentioned above . Theoretically , n neurons provide a space of n ! possible representations; thus , for instance , six sources can , in principle , be classified based on the recruitment order of only three neurons . How sensitive is classification by recruitment order to the number of sampled neurons ? Figure 7 shows the rank order representation accuracy as a function of the number of sampled electrodes for two different networks ( two source discrimination task ) . Here , each point in a continuous line depicts the mean accuracy of 200 random combinations of n electrodes ( abscissa ) . Note the monotonic increase in accuracy as a function of the number of sampled electrodes; similar results were obtained in six experiments of five sources classification task , shown in Figure S3 , left panel . Moreover , even though the classification accuracy based on spike timing is significantly higher than that of rank based representation at very small numbers of sampled electrodes , this difference disappears when the number of electrodes increased . In other words , as the number of sampled neurons increases , the information carried by spike times becomes redundant to the information carried by recruitment order . The apparent difference between the performances of the two classifiers in low electrode numbers implies that absolute spike times carry information about the stimulus source that is not captured by recruitment order . It is important to note , however , that the information carried by the exact latencies does not mean that this information is available to decoders that are based on precise times ( e . g . coincidence detectors ) . In fact , our data implies the contrary: The time-warping exemplified in the first part of the results suggests that absolute times to first spikes are not precise , at least in this preparation , and some non-trivial transformation of spike times is required before a satisfactory classification may be achieved . The relative simplicity of the recruitment order over spike times coding is apparent in the success of classification using unsupervised methods ( Figure 6 ) ; this method was unsuccessful when applied to spike times ( various distance metrics were tried; data not shown ) . Indeed , it seems likely that a tradeoff exists between the number of sampled neurons and the complexity of the neural code . Figure 8 demonstrates the impact of temporal resolution on the accuracy of classification in a case of two sources classification task ( left panel ) and six sources classification task ( right panel ) . Data was binned at temporal windows ranging from 1–100 mSec , and latency and rank vectors were prepared as explained above ( Figure 5 ) . Figure 8 shows that , in general , the classification capacity is very high as long as the temporal resolution is in the range of 10 mSec or less , decreasing significantly at lower resolutions; similar results were obtained in six experiments of five sources classification task , shown in Figure S3 , right panel . Importantly , the reduction of spike time data ( Red ) to recruitment order ( Blue ) does not degrade the classification accuracy throughout the range of analyzed resolutions . This result is evident in all of our experiments ( data not shown ) . In the nine different networks that were challenged with a two-source classification task , a 50% drop in accuracy was observed at a median temporal resolution of 43 mSec ( ±33 SD; range 15–71 mSec ) . The issue of temporal resolution required for accurate representation by recruitment order is rather subtle: our data indicates that while timing of individual spikes is lost in the transition to recruitment order , the temporal precision required for comparing the relative spike timing between different neurons is crucial for an accurate classification .
The extensive [33] and often conflicting array of hypotheses concerning neuronal representation of objects ( sensory , motor or “internally generated” ideas or concepts ) , reflects indeterminacy of data to theory . Regardless of what the nature of neural representation turns out to be , it should conform to elementary physiological constraints . Perhaps the most severe constraint in that respect is the multiplicity and wide range of timescales that are characteristic of neuronal excitability and synaptic communication: At each and every level of observation , physiologists report an ever-increasing range of reaction time scales that are involved in the generation of action potentials and their transformation to post synaptic signals [34]–[38] . It is hardly surprising , therefore , that the temporal structure of neuronal responses to repeated presentations of stimuli becomes inherently warped on a trial-by-trial base , reflecting the long tail of sensitivities to previous activation histories mediated by activity-dependent reactions underlying both exciting and restoring forces . Here we show that the effective structure that emerges spontaneously in large random networks of cortical neurons leads to representation by recruitment order that is invariant to response time warping . We bring evidence for the existence of neuronal stations through which activity is required to pass in order to propagate further into the network . We find it convenient to think about these sequences of neuronal stations in terms of chain-like effective structures [12]; thus , even in the face of activity-dependent changes in synaptic efficacies or membrane excitability , activity has nowhere else to go but through ordered stations , reassuring that the rank remains stable . Network architecture , in that sense , serves to protect the representation by order from the effects of the dynamics driven by activity dependence of reaction rates . Taken together with the observation that the ranking is stimulus site specific , the basic conditions for application of recruitment order representation are satisfied . ( Video S1 demonstrates a functional implementation of the above results: a Braitenberg vehicle that classifies objects in its visual field based on neuronal activity in a large-scale biological neural network . See caption of Video S1 for methodological comments ) . While ordered patterns of activations are observed at various spatiotemporal scales in-vivo and in-vitro in several neuronal preparations ( e . g . , [8] , [11] , [26] , [31] , [39]–[42] ) , the general applicability of representation by recruitment order at the random ensemble level is demonstrated , for the first time , in the present study . We provide a direct measure for the efficacy of rank order representation in actual classification tasks under well-controlled experimental conditions in large-scale recurrent networks of cortical neurons . Recruitment order is highly sensitive to the spatial features of stimuli and accurately classifies them on a trial-to-trial basis . The accuracy of spatial representation by the order of neuronal recruitment monotonically increases with the number of sampled neurons , and decreases with ordering time resolution . Furthermore , we show that the process of data compression , from absolute first spike latencies to recruitment order , is lossless from the point of view of stimulus classification accuracy . These results , even when taken together with the simplicity , rapidity , robustness and ease of physiological implementation of representation by recruitment order do not necessarily imply that it is superior to other representation primitives ( e . g . rate based or precise time delays ) ; more likely the balance between these features and task constraints govern the usefulness of a representation primitive in a given context . Several ways were proposed to realize a biologically plausible mechanism to decipher recruitment order . These include , for instance , a simple feed-forward network with shunting inhibition [7] , a decoder that is based on spike timing dependent plasticity that re-distributes synaptic weights at the single neuron level [43] , as well as a tempotron-based decoder that relies on adaptive integration time mechanisms [44] . Another issue that relates to the physiological plausibility of a decoder for recruitment order , has to do with the temporal reference point to which the code is time locked: In our experimental design , the classification is based on prior knowledge of the time at which a stimulus is delivered , but under real life situation such information is not available for the decoder . Recent studies , however , show that the temporal reference issue , which comes up whenever a time-based representation scheme is proposed , can be handled either by temporally referencing to population activity onset [45] , or by relying on time-based synaptic plasticity processes that tune the distribution of synaptic weights such that a neuron becomes sensitized to early spikes in a pattern [43] . Finally , from a more general perspective , it is interesting to contemplate on the possible functional relations between activity dependence of molecular transition rates underlying neuronal excitability ( e . g . , [34] ) , the space of possible combinations of precise latencies to first spikes and its degeneration to the form of recruitment order: While activity-dependence of neuronal reactions is a valuable driving force for exploration in a variety of adaptation and learning processes ( where representations are modified ) , it must be balanced by mechanisms that allow for stabilization and hence exploitation of existing representations . Representation by recruitment order provides a particularly attractive solution to this tradeoff: neurons dynamically change their absolute spike times relative to a reference signal ( e . g . stimulation time ) , thus exploring the space of possible associations driven by machineries of spike-timing dependent plasticity ( e . g . , [46] , [47] ) . Since many combinations of latencies to first spikes may realize any given representation by recruitment order , existing representations are invariant to the exploration process , as long as the latter does not degrade the order of neuronal recruitment . Effectively , a separation is formed between the level of absolute time delays , where exploration for new representations occurs , and the level of recruitment order where representations are stable enough to adaptively interact with the environment . The analogy to the separation between mutations at the genomic level , and selection at the proteomic ( phenotypic ) level immediately comes to mind .
Cortical neurons were obtained from newborn rats ( Sprague-Dawley ) within 24 hours after birth using mechanical and enzymatic procedures described in earlier studies [13] , [22] , [26] , [30] , [48]–[50] . The neurons were plated directly onto substrate-integrated multi-electrode arrays and allowed to develop functional and structural mature networks over a time period of 2–3 weeks . The number of neurons in a typical network is in the order of 300 , 000 , over an area of ∼300 mm2; various estimates of connectivity suggest that each neuron receives ∼1000 synapses , with ∼10% of these synapses being inhibitory ( see [13] for a comprehensive review of the preparation ) . The preparations were bathed in MEM supplemented with heat-inactivated horse serum ( 5% ) , glutamine ( 0 . 5 mM ) , glucose ( 20 mM ) , and gentamycin ( 10 µg/ml ) , and maintained in an atmosphere of 37°C , 5% CO2 and 95% air in an incubator as well as during the recording phases . Multi electrode arrays ( MEAs ) of 60 Ti/Au/TiN electrodes , 30 µm in diameter , and spaced 200 µm or 500 µm from each other ( Multi Channel Systems , MCS , Reutlingen , Germany ) were used . The insulation layer ( silicon nitride ) was pre-treated with poly-d-lysine . Long experiments lasting over 3 hours were conducted using a slow perfusion system with perfusion rates of ∼100 µL/hour . A commercial 60-channel amplifier ( MEA-1060-BC , MCS , Reutlingen , Germany ) with frequency limits of 1–5000 Hz and a gain of ×1024 was used . The MEA-1060-BC was connected to MCPPlus variable gain filter amplifiers ( Alpha-Omega , Nazareth , Israel ) for further amplification . Rectangular 200 µSec biphasic 10–50 µA current stimulation through randomly chosen pairs of adjacent MEA electrodes was performed using a dedicated stimulus generator ( MCS , Reutlingen , Germany ) coupled to a blanking circuit that disconnects the amplifiers during each input pulse . Data was digitized using two parallel 5200a/526 A/D boards ( Microstar Laboratories , WA , USA ) . Each channel was sampled at a frequency of 16–24 ksample/second and prepared for analysis using either the AlphaMap interface ( Alpha Omega , Nazareth , Israel ) or a dedicated Matlab ( MathWorks , Natwick , MA , USA ) interface developed by two of the authors ( D . E . and C . Z . ) . Thresholds ( ×8 RMS units; typically in the range of 10–20 µVolt ) were defined separately for each of the recording channels prior to the beginning of the experiment . All the activity recorded in the 60 electrodes up to 500 mSec following each stimulus were collected and stored for analyses . Where indicated , spike sorting procedures were applied , using the AlphaSort PCA package ( Alpha-Omega , Nazareth , Israel ) . Previous studies [13] , [22] , [26] , [30] , [48]–[50] show that the rate of spontaneous activity in these networks is , at least , one order of magnitude smaller compared to the activity evoked by stimulation , both when considered at the level of individual neurons as well as at the level of population responses; the interference of spontaneous activity with our analyses and interpretation of the results is minute . Mature networks were chosen for experimentation based on their ability to reliably respond to more than one source of low frequency ( 0 . 05 Sec−1 ) stimulation . Reliability of response is defined as a reverberating network activity ( that is time locked to a stimulus ) , observed in over 50% of stimulus presentations . The classification results presented here are based on a data set from 15 networks that were exposed to two ( n = 5 ) , three ( n = 2 ) , five ( n = 6 ) and six ( n = 2 ) different stimulation sources . In the two-source classification tasks the stimuli were delivered at several frequencies as explained in the results section . In all cases , the order at which stimuli were delivered through different stimulation sources was shuffled throughout the experiment . Wolfram's Mathematica 5 . 2 environment was used for calculation of the Levenshtein string metric and for cluster analysis . Neurons that fired within the same time bin were ranked according to their alphabet . SVM classification analysis was performed using MCSVM_1 . 0 ( http://www . cis . upenn . edu/~crammer ) , a C code package for multi-class SVM [51] using Gaussian Radial Based Function kernel . Kernel parameter and confidence intervals were set by a 5 fold cross-validation procedure . For SVM analyses , neurons that fired within the same time bin were credited an equal rank . | The idea that sensory objects are represented by the order in which neurons are recruited in response to stimulus presentation was put forward over a decade ago , largely based on computational biology considerations . While intensively analyzed in simulation studies , the general biological applicability of this highly compacted and efficient representation scheme , as an ensemble neural code , was never established . In recent years , algorithmic and experimental technologies advanced to a stage that allows for facing the challenge; here we took advantage of this progress . We let a large-scale random network of cortical neurons develop on top of a microfabricated , multielectrode array that enables electronic interrogation of the network , stimulating through various points in space , and simultaneously recorded the resulting activities from a large number of neurons . We applied state-of-the-art classification algorithms and asked how well the rank order representation scheme handles categorization tasks . We show that recruitment order is generally applicable as an ensemble code; it emerges spontaneously in a large “structureless” network of neurons as a functional code that is invariant to significant temporal variance in spike times and spike rates and flawlessly classifies inputs on a trial-to-trial basis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neuroscience/theoretical",
"neuroscience",
"neuroscience/sensory",
"systems"
] | 2008 | Order-Based Representation in Random Networks of Cortical Neurons |
Numerical models for simulating outbreaks of infectious diseases are powerful tools for informing surveillance and control strategy decisions . However , large-scale spatially explicit models can be limited by the amount of computational resources they require , which poses a problem when multiple scenarios need to be explored to provide policy recommendations . We introduce an easily implemented method that can reduce computation time in a standard Susceptible-Exposed-Infectious-Removed ( SEIR ) model without introducing any further approximations or truncations . It is based on a hierarchical infection process that operates on entire groups of spatially related nodes ( cells in a grid ) in order to efficiently filter out large volumes of susceptible nodes that would otherwise have required expensive calculations . After the filtering of the cells , only a subset of the nodes that were originally at risk are then evaluated for actual infection . The increase in efficiency is sensitive to the exact configuration of the grid , and we describe a simple method to find an estimate of the optimal configuration of a given landscape as well as a method to partition the landscape into a grid configuration . To investigate its efficiency , we compare the introduced methods to other algorithms and evaluate computation time , focusing on simulated outbreaks of foot-and-mouth disease ( FMD ) on the farm population of the USA , the UK and Sweden , as well as on three randomly generated populations with varying degree of clustering . The introduced method provided up to 500 times faster calculations than pairwise computation , and consistently performed as well or better than other available methods . This enables large scale , spatially explicit simulations such as for the entire continental USA without sacrificing realism or predictive power .
Models of infectious diseases are powerful tools for studying outbreak dynamics . Mass action mixing models assume equal probability of infection among all individuals in a population and have provided important theoretical insights for epidemiology . However , the importance of deviations from this assumption is now largely recognized [1] , and researchers are increasingly implementing stochastic simulation models that incorporate various levels of realism [2] . The effect of spatial heterogeneity can have a pronounced effect on outbreak dynamics [3] and is of particular concern when models are used to inform policy . Spatially explicit models can be used to identify geographical hotspots targeted for surveillance [4] or to compare control scenarios that are themselves spatially explicit ( e . g . ring vaccination strategies or regional movement restrictions ) . Here we focus on livestock disease models , but emphasize that the proposed methods are very broadly applicable . Livestock models typically consider infections at the farm level [5] , and since the farms have fixed spatial locations , spatially explicit models are appropriate . Distance dependent transmission is commonly modeled with a spatial kernel that describes how transmission risk varies with distance [5–7] . Large livestock disease outbreaks can have severe societal and economic implications , necessitating models that can simulate outbreaks at the national or even the continental scale [7–9] . With a large number of farms , computation time may be a limiting factor , particularly when stochastic simulation models are used to quantify uncertainty when comparing across multiple scenarios . In the absence of efficient algorithms to improve computational time , pairwise calculations must be implemented , whereby the distances between every pair of infectious and susceptible farms need to be calculated and the spatial , transmission kernel must be evaluated for those distances . When the spatial kernel is narrow compared to the spatial distribution of farms , there will be many farms that lie within the tail of the kernel where the risk of infection is low . Given this relatively low risk , it may be tempting to truncate the kernel at such large distances in order to dramatically reduce the computation time of the simulation . However , it is difficult to identify a spatial scale where such truncation does not influence the results and finding such a scale by trial and error may become a time consuming process . Rare long distance transmission , described by a fat tail of the spatial kernel , can have a pronounced effect on outbreak dynamics , sparking new infections in virgin areas of susceptible farms [5] . In addition , with two-dimensional space ( and assuming homogeneous farm locations ) , the number of potential transmission events also grows linearly with distance . Thus , even if the likelihood of infection of individual farms is low at large distances , there are many farms that can be infected and the probability of a transmission event occurring is not necessarily small . We therefore argue that truncation should be avoided due to its inaccuracy and that effort should be made to create algorithms to model transmission that circumvent the excessive number of kernel evaluations required for pairwise evaluation while maintaining accuracy . Brand et al . [10] introduced one such method , denoted the fast spectral rate recalculation ( FSR ) . This approach utilizes fast Fourier transformation ( FFT ) of the spatial kernel , and with only slight approximation of the transmission rates , it can speed up simulations by two order of magnitudes . Keeling and Rohani [11] introduced another technique , hereafter denoted the conditional entry algorithm , where the point pattern landscape of farm locations is overlaid with a grid , and the simulation of infections is split up into two steps: Our study introduces a novel method , which we will refer to as the conditional subsample algorithm . It builds on the approach of Keeling and Rohani [11] , and utilizes a similar gridding approach but with a different algorithm . Importantly , the conditional entry and conditional subsample methods do not approximate the epidemiological process ( beyond the temporal discretization , which are typically implemented also in pairwise simulation ) that is simulated; they merely speed up the computation by reducing the number of calculations , and hence preserve the accuracy of the simulation . Implementing the conditional entry or conditional subsample method requires the specification and construction of a grid structure ( Fig 1 ) , which raises a central question: how many farms should each grid cell contain in order to facilitate fast computation ? At both very large and very small cell sizes , both the conditional entry and conditional subsample algorithms require ( at least ) as many kernel evaluations as the pairwise algorithm . Thus , some intermediate grid size should optimize the speed of both algorithms . This study has three aims . Firstly , we introduce a novel transmission algorithm for grid-based calculations: the conditional subsample algorithm . Secondly , we propose an optimal grid size estimation method for determining a grid size configuration for the conditional subsample and conditional entry algorithms that ensures fast computation , and evaluate its performance . Thirdly , we investigate how this algorithm compares to simulations based on both the pairwise algorithm , the conditional entry algorithm presented by Keeling and Rohani [11] , and the FSR algorithm introduced by Brand et al . [10] . To address these aims , we simulate outbreaks on computer-generated farm locations with different levels of spatial clustering , as well as empirical farm populations from the USA , the UK and Sweden . These spatially heterogeneous farm distributions offers more realistic challenges to the proposed methods than would homogeneous distributions .
We use a spatially explicit kernel model based on an approach developed for the FMD outbreak in the UK in 2001 [5] . This is a versatile and common modeling approach that has also been used to model FMD in other countries such as the Netherlands [12] and Japan [6]; as well as other diseases , such as avian influenza in the USA [13] and bluetongue virus in the UK [14] . Our formulation is a standard farm level Susceptible-Exposed-Infectious-Removed ( SEIR ) model with discrete daily time steps and a daily rate of infection , Rij , between infectious farm i and susceptible farm j . This rate depends on the kernel K , which is a function of the Euclidean distance between i and j ( dij ) , as well as the transmissibility of i ( Ti ) and the susceptibility of j ( Sj ) . The daily infection rate is given by Rij=TiSjK ( dij ) . ( 1 ) The probability of a susceptible farm j becoming infected by farm i within a given span of days is obtained by discretizing Eq ( 1 ) pij=1−exp ( −SjTiK ( dij ) δt ) ( 2 ) where δt is the time step . By setting δt to one , pij becomes the daily probability of transmission from i to j . Once a farm was infected , an incubation period of four days was assumed after which the farm’s entire animal population was considered infectious for a period of five days and at the end of the infectious period the farm was considered removed from the population . Following Tildesley et al . [15] , transmissibility and susceptibility of farms were modeled as a nonlinear function of the number of cattle and sheep present on the premises as Ti=ϕcattlezcattle , iτcattle+ϕsheepzsheep , iτsheep ( 3 ) Si=ψcattlezcattle , iσcattle+ψsheepzsheep , iσsheep , ( 4 ) where all parameters are species-specific constants , and z is the number of animals of the respective species on the individual farms . For simplicity , we initially used the set of parameter values for τ , σ , Φ and Ψ fitted for Cumbria , England in [15] for all simulations ( see Table 2 for parameter values ) . These parameters resulted in outbreaks large enough to enable exploration of the efficiency of different grid configurations for the UK , but yielded small outbreak sizes when simulating infection across the other landscapes ( S1 Table ) . We were also interested in investigating the performance of the algorithms for large outbreaks , and therefore multiplied the ψcattle estimated for UK ( Cumbria ) by 200 when simulating outbreaks in all other landscapes , which resulted in substantially larger epidemics . The animal populations for these other landscapes contained only cattle , thus obviating the sheep terms in Eqs ( 3 ) and ( 4 ) . The local spread kernel , K , models the change in infection risk with distance between i and j , and includes all possible routes of infection except the shipment of animals between farms . For the purpose of comparing CE and CS and different optimized gridding schemes , we used the functional form K ( dij ) =α1+ ( dijβ ) γ , ( 5 ) parameterized as in [7] , ( Table 2 ) and referred to as the Buhnerkempe kernel . In this section , we consider three transmission algorithms: the pairwise ( PW ) , the conditional entry ( CE ) and the conditional subsample ( CS ) algorithms . The CE algorithm has previously been published by Keeling and Rohani [11] , whereas the CS algorithm is a novel approach . Example code for these three algorithms is provided as part of a C++ infection simulation model in the supplementary materials ( S1 Appendix and S2 Appendix ) . The transmission algorithms were compared using simulation run times . However , since this is a measure that is both system dependent and also possibly influenced by other processes running simultaneously on the system we also determined the total number of calls made to the distance kernel function ( Eq . 5 ) during a simulation . We considered this a measure of computational complexity and a proxy for relative computational time . Making a call to the kernel function is usually a relatively costly operation in itself and in a naïve implementation it will constitute most of the activity during a simulation as it takes place every time an infection probability is evaluated . For small populations of farms , it may be feasible to store the evaluated kernel values for each pair of farms in order to reduce the number of calculations , but for models on the individual farm level on national scales this approach easily becomes limited by memory availability . For comparison , representing the distances between all unique pairs as 64 bit double precision floating point numbers in a population of n nodes and making use of the fact that such a matrix is symmetrical to only store the upper or lower triangle requires approximately 0 . 373 GiB for n = 104; 37 . 3 GiB for n = 105 and 3725 . 3 GiB for n = 106 ( 2691 . 5 GiB for n = 850000 , which is roughly the size of the US farm population used in this study ) . Although supercomputer systems generally have memory per node in the ranges of 32 to 256 GiB , the availability of nodes in the higher range is usually limited and going above the limit of what a single node can handle necessitates a code that can handle shared memory between nodes , making a relatively simple problem significantly more complex . Also , even if it would be possible to run a simulation requiring a large amount of memory , lowering the memory requirement may allow multiple such simulations to run in parallel on the same system . A large number of independent replicates are generally needed , making numerical models for disease simulation prime examples of problems subject to embarrassingly simple parallelization . For the general case , this means that parallelization can be performed over the different replicates with no need for specialized code or extra overhead due to thread or process management , and that there is little to gain from more complex forms of parallelization . Throughout the explanation of the algorithms we will refer to the farms as nodes . A cell containing an infectious node is referred to as a , while a cell with susceptible nodes is referred to as b ( Fig 1C ) . The set of all cells that contain susceptible nodes is denoted B . An infectious node is referred to as i , a susceptible node as j . Sets of infectious and susceptible nodes are referred to as I and J respectively . These sets of nodes are sometimes subscripted with a cell to indicate a set of nodes within a cell . We use the notation of cardinality of the sets to refer to the number of elements in that set ( e . g . |Jb| , the number of susceptible nodes of cell b ) . The cell size ( number of nodes contained in each cell ) will have a large impact on the efficacy of both the CE and CS transmission algorithms . Conceptually , it is easy to see that either too small or too large cells will render the methods inefficient . At one extreme end , a single cell covering the entire landscape would equate the method to the pairwise algorithm . On the other hand , if each cell is so small that it only contains a single farm , the kernel similarly needs to be evaluated for every farm and iteration , plus an additional evaluation when a cell is entered . Clearly , some intermediate cell size is optimal . In theory , the best cell configuration is one where the difference between the upper bound of the probability of infection ( υib or ωab ) and the probability of infection itself is as small as possible because this will lead to fewer ‘false positives’ ( times where the outcome of step 1 is true for the CE algorithm , or a larger than necessary binomial sample for the CS algorithm ) . The amount of overestimation of υib and ωab can be reduced by using small cells with few nodes in each since that will minimize the difference between dab ( the cell-to-cell distance ) and dij ( the actual distance between the infectious node i and the susceptible nodes j ) . Furthermore , having a cell in which the distribution of the nodes’ susceptibility , Sj , is as homogenous as possible also contributes to a smaller amount of overestimation . At the same time , the more small cells there are , the closer the behavior of the transmission algorithm will be to that of the pairwise algorithm in that there will be a larger amount of operations required just to evaluate part one of respective algorithm . Finding a grid configuration with an optimal balance between these factors for a landscape with a spatially heterogeneous node distribution is not a trivial problem . Also , any perfect solution will be dependent on where in the landscape the infectious node is , further complicating the task . We propose a straightforward method to quickly find an approximation of the optimal average grid cell size , θ^* , for a given landscape . The method relies on a set of simplifying assumptions , where all farms are assumed to have equal susceptibility and transmissibility , and distributed uniformly in a quadratic landscape . Also , we only estimate θ^* for the initial phase of the outbreak , where only one node is considered infectious and all other nodes are susceptible . Note , however , that we challenge these assumptions below when we evaluate the performance of the gridding methods . The method to identify an optimal grid size is: This procedure was performed for 100 different grid configurations , each with total number of cells κ2 where κ ∈ {1 , … , 100} . For κ = 1 , a single cell overlays the landscape , and 100 was chosen as an upper bound that we determined to yield a large enough span of grid configurations to be sufficiently certain that the optima for the landscapes used in this study were found . The sum of expected number of kernel calls required for each evaluated cell was calculated as Ntot=∑a∈C ( ∑b∈Cb≠a ( Ni , b+1 ) +|Ja| ) , ( 14 ) where C is the set of all cells for the given configuration and Ni , b is the expected number of calls to the kernel function required to simulate infection spreading from one infectious node i in a to the nodes in b during one time step . For both the CE and CS transmission algorithms , there will always be one kernel call made for each cell that is checked in order to calculate the overestimated infection probability υib or ωab , as well as |Ja|calls associated with the internal pairwise checks of the susceptible nodes within a . Note that since the distribution of animals is uniform over the nodes and the nodes are uniform within the landscape , the transmissibility parameters from Eq ( 7 ) and Eq ( 10 ) , will have the same value ( Ti=T^a ) and the number of infectious nodes in a is one , so υib = ωab . After the initial overestimated probability υib or ωab has been calculated , the expected number of kernel calls for a given pair of infectious node i ( or cell a , it is the same in this context since there is only one infectious node in a ) and grid cell b is Ni , b=υib|Jb| . ( 15 ) This is simply the expected value of the binomial distribution given |Jb| draws and probability υib or ωab . Eq ( 15 ) is intuitive for the CS algorithm because each node in the binomial sample , drawn from the population of susceptible nodes within a grid cell during the execution of the algorithm , will be evaluated exactly once . This adds one kernel call for each node in the sample . The proof is somewhat less apparent for the CE algorithm , requiring a more comprehensive derivation . This is provided in the supplementary material ( S4 Appendix ) . Due to ωab increasing with the number of infectious nodes within a cell it was expected that during actual simulations with the CS algorithm the probability ωab would be close to or equal to one more often than υib would with the CE algorithm . This leads to the CS algorithm degenerating into the pairwise algorithm ( when ωab = 1 , nb = |Jb| ) more often than the CE algorithm which will still only occasionally enter the susceptible cell b , even when a cell contains a lot of infectious nodes . In order to make a fair comparison between the methods we therefore also calculated θ^* using the maximum number of animals on any farm in the landscape as well as the 75th percentile of the number of animals on the farms for transmission parameters rather than the median , and evaluated both using the CS and CE algorithms . With these higher transmission parameters , the cells become somewhat smaller which offsets this effect for the CS algorithm . The results showed mostly minute differences in run times for simulations depending on summary statistic used , but for the CS method using maximum number of animals was clearly optimal for the UK ( 3 . 2 and 2 . 9 times faster than 75th percentile and median , respectively ) . Based on this we chose to use maximum for all simulations with the CS algorithm and median for all simulations with the CE method . Two different methods were used to spatially divide the nodes of the landscapes into grids , each grid consisting of a set of square cells , C . The first such grid construction method , denoted regular grid construction , overlays the landscape with a set of uniformly sized square cells of predetermined spatial size ( Fig 1A ) , generated by selecting a number κ , describing the square root of the total number of cells ( or the number of cells along one dimension ) desired , and simply constructing κ2 square cells with side = l/κ , where l is the longest side of the rectangle bounding the landscape . Secondly , we used an adaptive grid construction approach with similarities to the quad-tree data structure common in computer science , where a spatially heterogeneous grid is constructed ( Fig 1B ) , with the aim of having an equal number of nodes per cell . Starting with one large cell covering the entire landscape , we recursively divided cells ( parents ) into four smaller equal-sized cells ( children ) , which in turn were divided into even smaller cells and so on . The process was continued for each cell as long as further subdivision satisfied the following condition: ( log ( |La| ) −log ( λ ) ) 2≥∑b∈B ( ( log ( |Lb| ) −log ( λ ) ) 2 ) |B| . ( 16 ) Here , La and Lb indicates the set of nodes within parent ( a ) and child ( b ) cells , respectively , B is the set of up to four child cells of a that contain nodes ( depending on the spatial distribution of nodes inside the parent cell , some child cells can end up without nodes ) and λ is the threshold number of nodes per cell . As such , the subdivision minimizes the squared difference on between log-number of nodes per cell and the specified log-λ . At the end , all cells that did not contain any nodes were removed from the final grid for both the adaptive and the regular grid construction methods . We applied the grid construction methods to farm landscapes that deviated from the underlying assumptions of the approximation in section Estimation of optimal grid cell size ( uniform node distribution inside a square with equal susceptibility and transmissibility ) , using grids according to the optimized θ^* as well as both finer and coarser grid configurations . We specified λi=ηiθ^* , i∈[−6 , −5 , … , 6] . ( 17 ) The constant η was set to 1 . 75 in order to give a large enough span of different grid configurations across all landscapes and the range of i was chosen as [6 , 6] to yield a manageable number of grid configurations with a sufficient level of detail as well as equal amount smaller and larger cell sizes in addition to θ^* itself . In order to make the regular grids somewhat comparable in size to the grids constructed with the adaptive method , the sets of values for κ were set to the integer values for which the average number of nodes per cell ( n/κ2 ) was closest to the sets of values of λ . For the landscape with uniform spatial distribution of nodes , the adaptive gridding approach generally resulted in configurations with cell sizes that deviated substantially from λ . Therefore , for this landscape , we used only the regular gridding method . The evaluation of the algorithms and the gridding schemes were performed on the empirical cattle and sheep farm population of the UK and the cattle farm populations of the 48 contiguous states of the USA ( i . e . excluding Alaska and Hawaii ) and Sweden , each with 177 855 , 832 514 and 24 275 farms respectively; as well as on three generated landscapes with differing degrees of clustering . The generated landscapes were created with the method of [16] and each had one quarter of the number of farms of the USA data ( 208 129 ) and an area that was one tenth of that of the contiguous USA . This scaling of the random landscapes was found to provide consistently large outbreak sizes through high enough farm density , while keeping simulation runtimes with the pairwise method at a manageable level . The number of animals on the farms for the generated landscapes were sampled randomly from the USA farm size distribution . See S2 Table for clustering statistics of the landscapes . The UK and Sweden keep detailed information about farms in central databases , including position and herd sizes . These data were made available for the study under confidentiality agreements and could be used when simulating outbreaks . There are however no equivalent data bases for the USA . Instead we used simulated demography data generated by the Farm Location and Agricultural Production Simulator ( FLAPS ) . FLAPS simulates spatially-explicit farm locations and farm sizes , based on county level demography information from the National Agricultural Statistics Service ( NASS ) , in combination with environmental features ( e . g . topography and climate ) , and anthropogenic factors ( e . g . roads and urban markets ) [17] . While not a perfect representation of the geographical distribution and demography of the USA farm population , it offers the most realistic depiction available . As such , it is well suited for investigating the performance of the presented disease simulation algorithms . The three randomized landscapes as well as the FLAPS realization for the USA farm population used in this study are available as supporting information ( S1–S4 Datasets ) . The usefulness of κ^ and θ^* as indicators of the actual optimal grid size for the original landscape was tested by comparing them to a number of other configurations generated using the two different grid construction methods ( regular and adaptive ) through simulations . Each simulation was seeded with one random farm as the initial case and was run until the outbreak either died out or until 10 , 000 cumulative farms had become infected . The run time and total number of kernel operations depends heavily on the number of nodes infected over the course of the simulated outbreaks , making comparisons between replicates with different outbreak sizes difficult . To circumvent this problem , the number of kernel operations of the simulations were recorded at the end of the time step where 10 , 100 , 1 , 000 and 10 , 000 nodes became infected ( we refer to these as outbreak stages ) . To improve the speed of all simulations regardless of transmission algorithm the kernel was evaluated for every integer distance between 1 and the maximum possible distance within the given landscape ( in meters ) and the distances calculated during the simulations were rounded to nearest integer . This does not , however , mean that the number of kernel calls as a relative measure of complexity changes as it still indicates the number of operations made where the kernel is involved . Also , even though this can be seen as an approximation , we argue that the error introduced by this approach will be smaller than the error in a set of landscape coordinates at the one-meter level . The simulations of the epidemic model using the grid construction methods and transmission algorithms , as well as the algorithm for finding θ^* , were all implemented in C++ ( S1 Appendix , S2 Appendix ) . All simulations were run at a supercomputer cluster consisting of 2 . 2GHz 8-core Intel Xeon E5-2660 processors . The CE and CS algorithms were compared to the fast spectral rate recalculation method ( FSR ) presented by Brand et al . [10] . Simulations with the FSR method and the other algorithms was performed using the US farm population with parameters as described in the [10] We compared the two methods using the distance kernel from the evaluation of the FSR method on the US population in [10] ( referred to as Brand kernel; notation converted to match that already in this paper ) Kγ ( d ) =αNγβ ( β2+d2 ) −γ/2 . ( 18 ) As closely as possible we tried to replicate the simulations in the original study and tested the methods with three different sets of parameters corresponding to three different shapes of the kernel ( Table 3 ) , Nγ being a normalization constant dependent on the choice of γ . In the comparison , the same node second closest to the center of Franklin County , Texas was seeded every replicate . The reason for choosing the node second closest to the center over the node closest to the center as in the original study , was that it had 19 animals in our data set as opposed to the central node which had only 1 , increasing chances of outbreaks taking off . Estimation of optimum grid size ( θ^* ) for the CS and CE method was performed as described under section Estimation of optimal grid cell size using the relevant kernel ( Table 3 ) . For these simulations , results were recorded at the time step when 10000 cumulative infected nodes was reached as well as at the point where the epidemic died out . A comparison to the FSR method was also attempted for the kernel and set of parameters used for the other analyses presented earlier in this paper ( Eq ( 5 ) and Table 3 ) . However , the very local nature of this kernel caused problems with the FSR method as the kernel shape necessitates a very fine resolution for the grid on which the image of infection is calculated in order to accurately capture the behavior of the kernel . For such fine resolutions , the efficiency of the algorithm drastically diminishes and made simulations until the end of the epidemic too long to be feasible . Thus , for this kernel , only the results from reaching 10000 cumulative infected farms was recorded before the simulations were terminated . Please note that the kernels used in the comparison with FSR are parameterized for a distance unit of kilometers rather than meters as for the other analyses presented .
For all landscapes , outbreak stages and grid construction methods using conditional subsample and conditional entry algorithms , even the worst grid configuration tested provided an improvement over the pairwise method , and the simulations with optimal suggested grid configurations ( θ^* ) yielded improvements in the range of 2 . 9–500 . 0 times faster ( Figs 2 and 3 , S1 Fig , S2 Fig ) . In almost all of those simulations the CS transmission algorithm consistently performed as well as or better than the CE method ( Fig 4 ) , and for that reason we present the results for the CS algorithm here and the results for the CE algorithm can be found in the supplementary material together with the result of the optimal grid estimation method for the six landscapes ( S3 Fig ) . The estimated optimal grid size , θ^* , coincided with θsim* for half of the combinations of landscapes and outbreak stages , meaning that the optimum grid estimation method worked well . However , for the situations where it was not the optimal grid size , it was generally very close to the optimal size , the most notable exception being for the uniformly random landscape at outbreak stage 10000 for which simulations using θ^* were 18 . 4% slower on average than for the optimal grid size θsim* ( Fig 2 , Table 4 ) . For the different outbreak size thresholds and landscapes using the conditional subsample transmission algorithm with the adaptive grid construction method , the positive values indicate that θ^* performed best in the simulations . For these cases the numbers represent the relative increase in average run time when comparing θ^* to the second-to-optimal grid cell size . Negative values indicate that θ^* did not perform best in the simulations and show the relative decrease in average run time when comparing θ^* to the value of θ that gave the best performance . No difference was evident in the number of time steps required in order to reach the different stages regardless of transmission algorithm and grid configuration , supporting our claim that the methods treat the dynamics of the outbreak equally ( S4 Fig , S5 Fig ) . Incidence curves further supporting the identical behavior for the CE and CS method to the pairwise simulations using the estimated optimal grid size θ^* can be found as supplementary information ( S6 Fig ) . Out of the 500 replicates for each of the 13 grid configurations , not all reached all outbreak stages . Information on the number of replicates that reached the different stages can be found in the supplementary material ( S1 Table ) . The regular grid construction method ( Fig 1A ) is only suitable for creating a grid configuration for a landscape with uniform spatial node distribution . If the node distribution is heterogeneous ( as is expected for most real landscapes ) , the estimated optimal grid configuration will not be close to the actual optimum . This is evident from S7 Fig , S8 Fig where the results of simulating outbreaks on the landscapes with heterogeneous node distribution using the regular gridding method are shown . The comparison between the transmission algorithms presented in this paper with the FSR method presented in [10] showed that the CS algorithm always performed best on outbreaks with up to 10000 cumulative infected nodes ( Fig 5 ) . For simulations that ran the entire course of outbreaks the FSR method performed better for Brand kernels with shape parameter a = 3 and a = 4 , while FSR and the CS algorithm was on par for shape a = 5 ( Fig 6 ) .
Stochastic disease simulation models are powerful tools for contingency planning and can be used to evaluate the efficiency of different control options [2 , 18 , 19] . However , a large number of replicates may have to be simulated to capture a representative range of possible outcomes; an issue that is amplified when different seeding conditions are included . Commonly , multiple control actions need to be considered [20 , 21] , further inflating the number of simulations . Thus , computation time quickly becomes a limiting factor . We have introduced a novel algorithm for this purpose , denoted the conditional subsample method , and demonstrated that it yield substantial reduction of computational complexity . When compared to two other available optimization methods , it outperformed the conditional entry algorithm in almost all situations , and in most cases also the FSR algorithm . We have in this study focused on spatially explicit simulation models , using as an application livestock disease models where farms are considered the infective unit . The introduced algorithm as well as the CE method [11] requires gridding of the spatial landscape ( e . g . Fig 1 ) , making the potential for speed-up sensitive to the grid configuration . For this purpose , we have introduced a method for estimation of the optimal number of nodes per cell . The estimation is based on the simplifying assumption that nodes are distributed randomly in a unit square . As expected , the method performs well in identifying an optimal grid size when simulating outbreaks under such ideal conditions ( Fig 3A ) . However , this is far too crude an assumption for most instances , as node locations are usually not randomly distributed [16] . To see how well our optimal grid size estimation method holds when this assumption is violated , we also simulated outbreaks in two computer generated landscapes with different levels of clustering , as well as three empirical landscapes of farm locations: Sweden , the UK and the USA . The analysis showed that the simulations based on the approximated optimal grid size did not perform well on these landscapes using a regular grid configuration ( Fig 1A , S7 Fig , S7 Fig ) . However , when using adaptive grid sizes , targeting equal numbers of farms per cell , the simulations based on the estimated optimum performed very well compared to the simulations with other grid configurations ( Fig 3 , S1 Fig , S2 Fig ) . This is encouraging , suggesting that our optimal grid size estimation method can be used to decide on the grid structure for a wide range of spatial configurations . Importantly , the simulations based on the estimated optimal grid size ( θ^* ) were substantially faster than the pairwise simulations , both during early and late stages of the outbreak ( Fig 2 , Fig 3 , Fig 4 , S1 Fig , S2 Fig ) . The observed speed-up when using the estimated optimal grid size ranged from a factor of 2 . 9 for the random high clustering landscape ( 161 nodes/cell , 10 infected nodes ) to a factor of 500 . 0 for the USA ( 333 nodes/cell , 10000 infected nodes; Fig 2 ) with the CS algorithm . The estimation of θ^* was based on simplified representations of the landscapes . Specifically , for in order to achieve the best estimation the transmissibility and susceptibility of the nodes were based on the median or maximum farm size across the un-simplified landscape for the CE and CS algorithms , respectively . The reason for this difference is that in the CE algorithm the transmissibility used for evaluating the entry into a cell or not is based on a single infectious node which corresponds closer to the median than the maximum farm size . In the CS algorithm , however , the transmissibility used is that of the most transmissible farm in cell a for which maximum farm size is a better proxy than the median . To further investigate the performance of the gridding algorithms and make comparisons to the FSR algorithm , we ran simulations with different spatial kernels , differing primarily in tail fatness [10] . We focused on the USA demography because this is the system the FSR method was introduced for . It is also the largest farm population and therefore where computational gain is the most important . Fig 5 shows that for all considered kernel shapes , the CS algorithm ( using a grid configuration determined by the method described in section Estimation of optimal grid cell size ) again outperformed the CE and pairwise methods . When comparing runtimes with the FSR method , the CS algorithm was consistently faster in terms of simulations to the first 10000 infected farms , whereas the results differed between kernel types in terms of computation time for entire outbreaks . One likely reason for this is that the FSR method works on susceptible farms and as they become fewer due to being infected and removed , the complexity of the algorithm decreases . The algorithms presented in this paper on the other hand has a complexity that grows with the number of infected farms and so , will be faster during the initial stages of the outbreak and lose relative speed as the epidemic grows . As such , there are instances where the FSR would be quicker , particularly if the simulated outbreaks are extremely large . The FSR method introduces a slight approximation unless a suitable grid size for the image of infection can be found . No apparent difference in terms of precision is revealed in Fig 5 for the Brand kernels; all simulation methods provide similar estimates of number of time steps to reach either 10000 infected farms ( Fig 5 ) or end of outbreak ( Fig 6 ) . Thus , the issue of approximation is of less concern in terms of choosing algorithm for kernels such as these . However , for the Buhnerkempe kernel which is particularly local , the grid requires a very fine resolution something that increases the computational burden significantly . The effect of increasing the grid resolution can be seen in the lower right panel of Fig 5 where the run time of FSR is markedly higher than for the other kernels . The grid resolution used for that simulation was high , but still too coarse to avoid introducing a slight error ( upper right panel ) and remedying that with an even finer grid would make the method even less competitive . It should be noted that the CS and CE algorithms are exact only beyond the discretizing into daily time steps . The Gillespie algorithm [22] can be used to model the continuous processes exactly , but to the knowledge of these authors , no optimization for this algorithm exists for the system considered here , and Brand et al . [10] showed for simulation of FMD outbreaks in the USA that the Gillespie algorithm is computationally very expensive . Beyond computation time , the discretization can be justified by the fact many process are cyclical at a finer resolution than the time scale of discretization ( i . e . daily discretization and diurnal cycles ) , or that data is typically available at discrete time scales [10] . To explore the efficiency of the set of methods described , we have applied them to outbreaks of FMD . This is a valuable approach , because we demonstrate the applicability of the method for relevant situations . Yet , a potential objection emerges–do the results hold outside of the context we have explored ? It is infeasible to consider all possible applications of the methodology; stochastic simulations of spatially explicit kernel models are used for a wide range of diseases and questions ( e . g . [6 , 12–14] ) . However , the study design along with the results provide some useful insight . We simulated outbreaks in a variable assortment of spatial landscapes with varying levels of clustering and farm densities , and found that our optimization method was efficient and constantly identified a grid configuration that sped up computations substantially , ranging from 100 to 500 times faster than the pairwise estimation for outbreaks of substantial size ( 10000 infected nodes ) . Next , we used the same optimization method to identify the optimal grid configuration for different kernels and found that in this novel context , the CS method improved computation time by a factor of 600 to 800 , depending on kernel . Here it also holds an edge over the FSR method when considering infections up to 10000 farms and vary in terms of computation time for extreme outbreaks ( Fig 6 ) . These simulations suggest that the CS method is a good candidate to consider for stochastic simulations of disease outbreak . Based on Fig 6 , the FSR algorithm may be faster for some applications , particularly when the simulations are expected to result in very large outbreaks . However , we are encouraged that in the instances where the CS algorithm performs slower than FSR , the difference is not vast ( approximately a factor of two; Fig 6 ) . Also , the FSR method was found to have some issues with very local kernels such as the one used in this work , so for such cases the CS or even the CE method would be more suitable . We argue that beyond the fast computation time , one strength of the CS algorithm lies in its simplicity . Because it does not approximate the epidemic model , it requires no tweaking to avoid loss of precision . As shown in the supplement ( S1 Appendix ) , the algorithm requires little additional code compared to the pairwise computation . In addition , the code structure is easy to combine with other processes . For instance , we have in our own ongoing work found it straightforward to combine a local transmission process , implemented with the CS algorithm , with a model for animal movements that leads to transmission over large distances [23] ( see supplement S1 Appendix for example code ) . Thus , we conclude that the transmission algorithms considered in this study are suitable for epidemic modeling of diseases where local area spread , modeled with a spatial kernel , is an important factor . By using the CS gridding approach , together with the introduced optimal grid cell size estimation , the computation can be sped up by several orders of magnitude . This allows for exploration of more scenarios , facilitating the use of disease simulation models for policy recommendations . | Numerical models for simulating the outbreak of infectious disease are powerful tools that can be used to inform policy decisions by simulating outbreaks and control actions . However , they rely on considerable computational power to explore all outcomes and scenarios of interest . Focusing on model types commonly used for livestock diseases , we here introduce novel algorithms for efficient computation , alongside techniques to optimize them based on simplifying assumptions . Through simulations of FMD outbreak in the US , the UK and Sweden , as well as in computer generated landscapes , we test how these methods perform under realistic conditions . We find that our optimization techniques works well , and when the introduced algorithms are implemented with these optimizations , computation time can be reduced by more than two orders of magnitude compared to pairwise calculations . We propose that the considered algorithms—which are straight forward to implement—will be useful for simulation of a wide range of diseases , and will promote the use of simulation models for policy recommendation . | [
"Abstract",
"Introduction",
"Method",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"ruminants",
"applied",
"mathematics",
"tropical",
"diseases",
"vertebrates",
"parasitic",
"diseases",
"animals",
"mammals",
"simulation",
"and",
"modeling",
"algorithms",
"kernel",
"functions",
"farms",
"mathematics",
"neglected",... | 2018 | Need for speed: An optimized gridding approach for spatially explicit disease simulations |
The signal transduction protein SmTK4 from Schistosoma mansoni belongs to the family of Syk kinases . In vertebrates , Syk kinases are known to play specialized roles in signaling pathways in cells of the hematopoietic system . Although Syk kinases were identified in some invertebrates , their role in this group of animals has not yet been elucidated . Since SmTK4 is the first Syk kinase from a parasitic helminth , shown to be predominantly expressed in the testes and ovary of adult worms , we investigated its function . To unravel signaling cascades in which SmTK4 is involved , yeast two-/three-hybrid library screenings were performed with either the tandem SH2-domain , or with the linker region including the tyrosine kinase domain of SmTK4 . Besides the Src kinase SmTK3 we identified a new Src kinase ( SmTK6 ) acting upstream of SmTK4 and a MAPK-activating protein , as well as mapmodulin acting downstream . Their identities and colocalization studies pointed to a role of SmTK4 in a signaling cascade regulating the proliferation and/or differentiation of cells in the gonads of schistosomes . To confirm this decisive role we performed biochemical and molecular approaches to knock down SmTK4 combined with a novel protocol for confocal laser scanning microscopy for morphological analyses . Using the Syk kinase-specific inhibitor Piceatannol or by RNAi treatment of adult schistosomes in vitro , corresponding phenotypes were detected in the testes and ovary . In the Xenopus oocyte system it was finally confirmed that Piceatannol suppressed the activity of the catalytic kinase domain of SmTK4 . Our findings demonstrate a pivotal role of SmTK4 in gametogenesis , a new function for Syk kinases in eukaryotes .
Helminth parasites of the genus Schistosoma are the causative agents of schistosomiasis , one of the most prevalent parasitic diseases for humans and animals worldwide [1] , [2] . More than 200 million people suffer from the pathological consequences of this disease , which originate from the massive egg production of schistosomes . The eggs cause inflammatory reactions in the gut , bladder , spleen and liver leading to granuloma formation [1] , [3] . Praziquantel is the only drug applicable to all schistosome species and is commonly used to treat patients , but treatment does not prevent reinfection . In the light of the absence of a vaccine and the probability of emerging resistance , a search for alternative treatments is a commonly accepted need for further research [4] , [5] . In this respect great international efforts are ongoing to analyze the genome of this blood fluke , its transcriptome , proteome , and glycome [6]–[10] . Besides their medical importance , schistosomes exhibit a nearly unique biological phenomenon–the pairing-dependent induction and maintenance of the sexual maturation of the female . During a constant pairing contact , the male activates signal transduction pathways in the female leading to the proliferation and differentiation of cells in the reproductive organs , such as , the ovary and vitellarium [11]–[14] . This is a prerequisite for the female to produce about 300 eggs each day [15] . One half reaches the outside of the definitive host to deliver miracidia continuing the life cycle . The remaining eggs are deposited in the host tissue causing pathogenesis . An egg from a mature female consists of one fertilized oocyte , originating in the ovary , and 30–40 surrounding vitelline cells produced in the vitellarium . Since growth and differentiation of vitelline cells and oocytes are probably controlled by signal transduction pathways , efforts have been made to identify and characterize the participating molecules . In the last decade , several genes encoding for signaling molecules from S . mansoni have been identified , some of which were found to be specifically or predominantly expressed in reproductive organs [reviewed in 16 , 17] . In contrast to the vitellarium , however , less is known about signaling molecules in the ovary . Among the molecules shown to be predominantly expressed in this organ is SmTK4 , a member of the Syk ( spleen tyrosine kinase ) tyrosine-kinase family [18] . Syk kinases are characterized by a tandem Src-homology 2 ( SH2 ) domain and a catalytic tyrosine kinase ( TK ) domain . Genome-project data have indicated that Syk kinase genes are absent in Caenorhabditis elegans , and in Drosophila melanogaster only the related kinase Shark ( SH2 domain ankyrin repeat kinase; [19] ) is present , which had suggested a recent evolutionary origin of kinases from the Syk family . However , Syk kinases were found in Hydra vulgaris as well as in sponge [20] , and with SmTK4 also in the parasitic helminth S . mansoni . In mammals , Syk kinases are expressed in hematopoietic cells playing well-characterized roles in inflammatory processes operating as downstream signaling molecules of immunoreceptors [21] . In the last years , evidence has accumulated for functions of Syk kinases in different signal transduction pathways also in non-hematopoietic cells [22] . Syk kinases regulate proliferation , differentiation , morphogenesis , and survival of epithelial [23] , [24] , endothelial [25] , and neuronal cells [26] . In the hematopoietic system , Syk kinases interact with immune and antigen receptors lacking intrinsic catalytic activity [27] . The tandem-like structure of the SH2 domains confers higher binding specificity of Syk kinases to phosphorylated tyrosine residues of upstream interaction partners compared to individual SH2 domains [28] . Following receptor activation , each SH2 domain interacts with one immunoreceptor tyrosine-based activation motif ( ITAM ) in the intracellular part of the receptor leading to a conformational change in Syk accompanied by an increase in its enzymatic activity [29] . In SmTK4 the conserved sequence within the SH2 domains responsible for this binding is absent , suggesting that this Syk kinase interacts with molecules without ITAMs . Binding of upstream partners stimulates autophosphorylation of Syk on tyrosines within the activation loop , which influences kinase activity or creates docking sites for SH2-containing proteins [30] . The phosphorylation of Syk can be enhanced by interacting Src ( Rous sarcoma virus kinase ) tyrosine kinases [27] . In addition , a variety of other signaling and adaptor molecules have been reported to associate with Syk kinases , but the relevance of these interactions have not been elucidated yet [27] . With respect to the very specialized function of Syk kinases in the hematopoietic system of mammals , the existence of a schistosome homolog was unexpected . SmTK4 was found to be transcribed in the larval stages as well as adults , independently from the pairing-status . Localization studies had demonstrated its predominant expression in the testes of the male and ovary of the female , but not in the vitellarium [18] . This narrow expression profile contrasts with other identified cellular kinases in adult schistosomes , such as the Src kinases SmTK3 and SmTK5 , whose expressions were demonstrated in all reproductive organs and other tissues [31] , [32] . Recently , an interaction of the Src kinase SmTK3 with SmDia ( Diaphanous-related formin ) and SmRho1 ( Ras-homologous GPTase ) was shown by yeast two-hybrid ( Y2H ) analyses and localization studies , indicating a role of SmTK3 in cytoskeleton organization processes in the gonads of adult S . mansoni [33] . To elucidate the potential function of SmTK4 in adult schistosomes , two different strategies were followed in this study . First , by isolating signaling molecules acting up- and downstream of SmTK4 we expected to discover interacting proteins , whose identity could provide evidence for functionally conserved signaling pathways . Using a recently established S . mansoni Y2H cDNA-library of females and males [33] and different constructs of SmTK4 as probes , several interacting proteins were identified , such as SmTK3 and a novel schistosome Src kinase ( SmTK6 ) , a MAPK ( mitogen-activated protein kinase ) -activating protein , and mapmodulin . Second , inhibitor and RNA interference ( RNAi ) experiments were performed to functionally knock-down SmTK4 activity . With the Syk kinase-specific inhibitor Piceatannol [34] , [35] or SmTK4-specific dsRNAs for RNAi , significant morphological changes in the testes and ovary of treated schistosomes were observed using carmine red-staining and confocal laser scanning microscopy ( CLSM ) . In Xenopus oocytes it was finally shown that Piceatannol is able to suppress the catalytic activity of the TK domain of SmTK4 . Taken together , our results in S . mansoni substantiate a pivotal role for the Syk kinase SmTK4 in gametogenesis , a function which has not yet been shown for a Syk kinase in other eukaryotes .
To identify upstream binding partners of the schistosome Syk kinase SmTK4 , an adult stage Y2H cDNA-library [33] was screened with the bait construct encoding the tandem SH2-domain of SmTK4 as a fusion protein with the Gal4-BD ( GAL4 DNA-binding domain ) and the TK domain of SmTK3 ( SmTK4-SH2SH2 + SmTK3-TK pBridge; Figure 1 ) . The schistosome TK domain was co-expressed in a kind of yeast three-hybrid ( Y3H ) approach to ensure the phosphorylation of tyrosine residues of potential interaction partners , since yeast does not possess specific , endogenous tyrosine-kinase activity [36] . Phosphorylation , however , is decisive for Syk-kinase interactions , because only phosphorylated tyrosine residues of binding partners acting upstream in a signaling hierarchy are favored binding sites for the pocket-like structure of the tandem SH2-domain of Syk kinases [28] , [37] . This principle has been successfully applied before to investigate the interaction between SH2 domain-containing proteins and tyrosine-containing substrates [38] , [39] . Without additional protein-protein interacting domains such as SH2 or SH3 , TK domains function promiscuously [40] . Therefore , we expected that tyrosine residues of yeast and library proteins were phosphorylated as a consequence of the expression of an individual TK domain . Indeed , Western-blot experiments with anti-phosphotyrosine antibodies have shown that tyrosine phosphorylation of yeast proteins was enhanced when the SmTK3 TK-domain was expressed ( Phillip , unpublished ) . The expression of both baits ( SmTK4-SH2SH2 and SmTK3-TK ) was confirmed at the transcriptional level by RT-PCR analyses using total RNA extracts from transformed yeast cells ( results not shown ) . Screening of the Y2H cDNA-library resulted in the identification of 77 initial prey clones , which underwent growth selection and β-galactosidase ( β-Gal ) filter assays . After isolation of the prey plasmids , sequence analyses of 14 remaining clones were performed by BlastX to unravel their identity ( Table 1 ) . Seven clones represented partial sequences from a schistosome homolog of a nonsense mRNA reducing factor ( NORF1 ) from Mus musculus ( accession number AAK08652 ) . Three clones encoded proteins with homology to a dipeptidyl peptidase III from Mus musculus ( accession number NP_598564 ) . Two clones showed similarity to the schistosome Src/Fyn kinase SmTK5 ( [32]; accession number AAF64151 ) and encoded a novel schistosome cellular tyrosine kinase ( CTK ) , named SmTK6 ( accession number FN397679 ) . Finally , two clones were identical to the Src kinase SmTK3 from S . mansoni , identified and characterized in a previous study ( [31]; accession number CAE51198 ) . To confirm their interactions and to determine their relative binding strength in a comparative approach , yeast cells ( strain AH109 ) were transformed with individual prey plasmids together with the original bait construct SmTK4-SH2SH2 + SmTK3-TK pBridge . Additionally , yeast cells were transformed with the prey plasmids and the bait construct SmTK4-SH2SH2 pBridge to analyze the dependence of the observed interactions on the additional tyrosine phosphorylation . Finally , the prey plasmids were used for transformation together with bait constructs that contained only one SH2 domain of SmTK4 ( SmTK4-SH2 ( 1 ) or SmTK4-SH2 ( 2 ) pBridge ) to test if the interactions depended on the presence of the tandem SH2-domain structure . After re-transformation of the prey plasmids with the bait constructs containing the tandem SH2-domain of SmTK4 , with or without the additional expression of the SmTK3 TK-domain , all yeast clones survived growth- ( Trp−/Leu−/Ade−/His− ) and color selection ( β-Gal filter assays ) thus confirming the observed interactions . However , when transformation was done with the bait plasmids containing only single SH2 domains of SmTK4 , the yeast clones containing the prey plasmids encoding the Src kinases SmTK6 or SmTK3 were no longer able to grow . This indicated that Src/Syk-kinase interaction depended on the presence of the tandem SH2-domain structure . The other clones survived the selection procedure , which is a hint for unspecific interactions ( Table S1 ) . To quantify the relative strengths of interaction β-Gal liquid assays were performed . To this end yeast cells were used , which were transformed with representative prey plasmids encoding for each potential binding partner and the bait plasmid SmTK4-SH2SH2 pBridge , or SmTK4-SH2SH2 + SmTK3-TK pBridge . These experiments again confirmed the observed interactions with SmTK4 . Furthermore , the results demonstrated the strongest interaction between the SmTK4 tandem SH2-domain and the novel Src kinase SmTK6 . This interaction was increased by a factor of five when the SmTK3 TK-domain was present indicating the significant influence of additional phosphorylation ( Figure 2 ) . The interaction between the SmTK4 tandem SH2-domain and the Src kinase SmTK3 was considerably weaker , but also enhanced by additional phosphorylation , although to a minor degree . The strength of interactions between the homologs of the nonsense mRNA reducing factor ( NORF1 ) and the dipeptidyl peptidase III were also weak and not , or only slightly , influenced by the state of phosphorylation ( Figure 2 ) . Since the novel Src kinase SmTK6 was the strongest upstream interaction partner of SmTK4 found in this screening , we carried out a preliminary characterization . The inserts of both SmTK6 prey clones were equal in size and about 1360 bp long . Comparisons by multiple alignment analysis revealed homology to CTKs and indicated that part of the 5′-region of the cDNA was missing in the isolated prey plasmids . Using schistosome sequencing data ( www . sanger . ac . uk ) , the whole coding sequence was identified in silico . To verify its existence in our S . mansoni strain , we performed RT-PCR analyses using the primer pair TK6-fl-5′ ( 5′-CTCATTATGGGAATTTGTTTGTG-3′ containing the ATG codon ) and TK6-fl-3′ ( 5′-AATTATCTAAATATTGAGCTTCTG-3′ containing the TAA stop codon ) , and total RNA from mixed-sex worms . Amplification products of the expected size were obtained and cloned . Sequence analysis showed that the complete cDNA sequence of SmTK6 is 1698 bp long encoding a protein of 566 amino acids ( accession number FN397679 ) . In silico analyses indicated the presence of one SH2 , one SH3 , and a TK domain that characterize CTKs of the Src family . Preliminary phylogenetic analyses showed that SmTK6 has also some similarity to the class of Abl kinases indicating that this kinase is a kind of Src/Abl intermediate ( Beckmann , unpublished ) . To additionally confirm the binding capacity of SmTK6 and SmTK4 , co-immunoprecipitation ( co-IP ) experiments were performed . To this end the tandem SH2-domain of SmTK4 was cloned as a FLAG-tagged construct into one multiple cloning site of the pESC-His yeast expression vector . In the second multiple cloning site of the same vector a nearly complete version of SmTK6 was cloned as a cMyc-tagged fusion . Following yeast transformation and subsequent growth selection , protein expression was induced by galactose . The expression of the recombinant proteins was proven by Western-blot analyses ( results not shown ) . Following co-IP , the presence of the FLAG- and cMyc-tagged schistosome proteins was finally confirmed by Western-blot analyses ( Figure 3 ) . Bands of the predicted sizes of 31 kDa ( SmTK4-SH2SH2 ) and 55 kDa ( SmTK6 ) were detected only in lysates precipitated by each of the antibodies , not in a control precipitated without antibody ( Figure 3B ) . This result confirmed that the tandem SH2-domain of SmTK4 is able to bind to SmTK6 independent of additional yeast Gal4-AD/BD ( GAL4 activating domain/DNA-binding domain ) fusion protein partners . By in situ hybridization , finally , SmTK6 was localized in the parenchyma of both genders , in the testes of the male and the ovary of the female ( Figure 4 , A–C ) . From the staining pattern obtained we cannot exclude that SmTK6 is also transcribed in the vitellarium . This testes- and ovary-preferential transcription pattern corresponded to that of SmTK4 [18] , additionally supporting the conclusion that these kinases interact . Downstream binding partners of Syk kinases are known to interact with the linker region or with the TK domain of a Syk kinase [41] . Furthermore , binding can be influenced by tyrosine residues within the TK domain , which are phosphorylated in trans by , e . g . , Src kinases [27] . For these reasons we used SmTK4-linker+TK + SmTK3-TK pBridge as the bait construct for library screening to identify binding partners acting downstream of SmTK4 ( Figure 1 ) . This bait construct expressed the complete linker region together with the TK domain of SmTK4 fused to the Gal4-BD ( multiple cloning site I , MCS I ) as well as the TK domain of SmTK3 ( multiple cloning site II , MCS II ) to ensure tyrosine phosphorylations of bait proteins ( Y3H ) . The expression of the bait constructs SmTK4-linker+TK ( + SmTK3-TK ) pBridge , SmTK4-linker pBridge , and SmTK4-TK pBridge was confirmed by RT-PCR analyses using total RNA extracts of the transformed yeast cells and appropriate primers ( results not shown ) . The mating with the S . mansoni Y2H library resulted , after subsequent growth selection and β-Gal filter assays , in 19 clones as potential candidates for interaction . The appropriate prey plasmids were isolated and their inserts sequenced . By BlastX analyses the identities of these potential binding partners were unraveled due to their homology to proteins in the NCBI-database ( Table 2 ) . Three clones showed homology to small heat-shock proteins ( HSPs ) from Caenorhabditis elegans ( accession number NP_001024376 ) . One clone encoded a related but not identical protein , which was also homologs to small HSPs ( accession number NP_001024376 ) , as well as to the major egg antigen Smp40 from S . mansoni ( accession number P12812 ) . Two prey clones encoded a protein with homology to caspase 3 from Homo sapiens ( accession number AAP36827 ) . The inserts of five further clones encoded homologs of the MAPK-activating proteins PM20/PM21 from different organisms , such as Homo sapiens ( accession number NP_001108072 ) . One clone represented a leucine-rich protein from S . mansoni ( accession number Q86QS6 ) , which exhibited homology to mapmodulin from Drosophila melanogaster ( accession number NP_001097361 ) . Seven clones showed the same insert sequence encoding a protein with no significant homology to any known proteins from other organisms . β-Gal liquid assays were performed to quantify the relative binding strengths of SmTK4 and its identified downstream interaction partners . To this end yeast cells were transformed with representative prey plasmids of the groups A–D together with the original bait construct SmTK4-linker+TK + SmTK3-TK pBridge to confirm the interactions and , additionally , with SmTK4-linker+TK pBridge to analyze the dependency of interactions on the supplementary tyrosine phosphorylation . Furthermore , to determine whether the linker region and/or TK domain of SmTK4 were responsible for binding , yeast cells were transformed with individual prey plasmids and further bait constructs containing only the linker region ( SmTK4-linker pBridge ) or only the TK domain of SmTK4 ( SmTK4-TK pBridge ) . The performed assays again confirmed the observed interactions with SmTK4 and , additionally , demonstrated significant differences in the relative binding strengths . The interactions between the small HSP and caspase 3 homologs with the different fragments of SmTK4 were very weak . With respect to the different SmTK4 fragments , the small HSP showed the strongest binding to the isolated linker region of SmTK4 , whereas the caspase 3 homolog showed no significant differences in the relative binding strengths ( Figure 5 ) . However , the interactions of the MAPK-activating protein ( PM20/PM21 ) and the mapmodulin with SmTK4 were stronger and differed between the SmTK4 fragments ( Figure 5 ) . In the case of the MAPK-activating protein , the interaction with the combined linker region and TK domain of SmTK4 was slightly enhanced by additional phosphorylation . However , with the isolated linker region , the interaction was nearly seven times stronger . In contrast , the interaction with the TK domain was very weak ( Figure 5 ) . These results showed that the MAPK-activating protein bound presumably to the linker region of SmTK4 , and that this binding was negatively influenced by the presence of the TK domain . Furthermore , binding to the linker region was strong even without additional phosphorylation . The interaction between mapmodulin and the combined linker and TK domain of SmTK4 was also slightly enhanced by the additional phosphorylation , as in the case with the MAPK-activating protein . Regarding the isolated linker region or TK domain , mapmodulin bound strongly to both with a slight bias towards the TK domain ( Figure 5 ) . Again , interaction was strong even without additional phosphorylation . As in the case with the MAPK-activating protein , the interactions of mapmodulin with the linker region or the TK domain were negatively influenced by the presence of both domains , indicating a potential inhibitory intramolecular conformation of these domains when expressed as a fusion protein . For the two strongest downstream binding partners of SmTK4 , MAPK-activating protein and mapmodulin , colocalization with SmTK4 in the ovary of the female and testes of the male was shown by in situ hybridizations ( Figure 4 , D–I ) supporting the conclusion for interaction . Corresponding to the SmTK4 transcriptional profile [18] , both binding partners were found to be transcribed also in parenchyma of both genders , but not in the vitellarium of the female . Recently , we introduced an experimental approach to investigate the role of schistosome CTKs by specific inhibitors . Using the Src kinase-specific inhibitor Herbimycin A for treatment of schistosome couples in vitro , we demonstrated that mitogenic activity and egg production were significantly reduced in females [42] . This was correlated with the reduced stability of the Src kinase SmTK3 from S . mansoni . Here we have combined the inhibitor approach with a novel way of phenotypic analysis at the morphological level . Based on the procedure established by Machado-Silva et al . [43] and Neves et al . [44] , we investigated inhibitor effects by treatment of schistosome couples in vitro with subsequent fixation and carmine red-staining . By CLSM finally , we looked for morphological effects with a focus on the testes of the male and the ovary of the female , because SmTK4 expression has been detected mainly in these organs [18] . To investigate whether the inhibitors influence egg production of paired females , the numbers of eggs were counted daily during the treatment period . Towards this end schistosome couples were treated in vitro with the Syk kinase-specific inhibitor Piceatannol ( 3 , 4 , 3′ , 5′-tetrahydroxy-trans-stilbene ) . This phenolic stilbenoid inhibits the activity of Syk kinases in cell culture with an IC50 value of 10 µM for the human Syk kinase [45] . For tissues it is used at higher concentrations of 100–200 µM [34] . For S . mansoni couples maintained in vitro , we used Piceatannol at concentrations of 35 µM , 70 µM , and 100 µM over a period of six days to investigate dosage- and time-dependent effects . The medium was changed daily along with the inhibitor , and the viability of the worms as well as their pairing stability was examined . During this time period , no alterations in behavior , mortality rates , or worm pairing compared with DMSO ( dimethyl sulfoxide ) -treated controls could be observed ( results not shown ) . Each day , an aliquot of treated worm couples was fixed in AFA , stained with carmine red and analyzed by CLSM . The control males and females , treated for the same time with DMSO , showed no morphological changes in the testes or ovaries ( Figure 6 , A–B ) compared to completely untreated schistosomes ( [43] , [44] , results not shown ) . The testes of untreated or DMSO-treated adult schistosome males are composed of several testicular lobes containing numerous spermatogonia and spermatocytes in different stages of maturation . Maturation of spermatocytes ( spermatogenesis ) begins in the dorsal part of the lobes with big round spermatogonia and ends in the ventral part with smaller elongated mature sperms ( spermatozoa , Figure 6A ) . In the ventral part of the testicular lobes and in the vas deferens elongated mature sperms can be detected as well as in the anterior sperm vesicle , which is full of sperms ( Figure 6A , arrows and asterisk ) . During treatment with 70 µM Piceatannol , however , this morphology changed considerably . After two days the size of the lobes was already slightly reduced , and the number of spermatocytes diminished ( results not shown ) . After six days these effects were even more dramatic . The testicular lobes were shrunk accompanied by a significant decrease in the number of spermatocytes per lobe ( Figure 6C ) . In the ventral part of the testicular lobes and in the anterior sperm vesicle , no elongated mature sperms were detected in most of the males ( Figure 6C , asterisk ) . Instead of mature sperms , the sperm vesicle contained , in several cases , undifferentiated round spermatocytes . In addition to the effects within the testes , the inhibitor also caused morphological changes in the ovary . In untreated or DMSO-treated mature females the ovary is composed of small oogonia and immature oocytes in the anterior part and larger primary oocytes in the posterior part ( [46] , Figure 6B ) . After treatment with 70 µM Piceatannol for six days , in most of the females the number of large primary oocytes was clearly increased compared to the small , immature oocytes . Furthermore , the high number of large oocytes was distributed all over the ovary and no longer concentrated to the posterior part ( Figure 6D ) . Using the lower concentration of 35 µM Piceatannol for worm treatment led to similar morphological changes in the testes and ovary , although with a delay of 1–2 days . A higher concentration ( 100 µM ) led to the same phenotypes , but in a shorter time period indicating a time- and dosage-depending effect of this inhibitor on schistosomes ( results not shown ) . Although at a comparatively low level , SmTK4 is also transcribed in subtegumental and parenchyma tissues in adults [18] . However , in these tissues we did not see an effect , which may be due to their low mitogenic activity compared to the gonads . To analyze the influence of Piceatannol on the egg production of paired females , the number of eggs was determined for treated couples maintained in vitro . 70 µM Piceatannol reduced the number of eggs per couple within 7 days to 51% compared to the DMSO control ( Figure 7 ) . Since obvious morphological changes after Piceatannol treatment were observed , the question arose , whether these effects could be ascribed to the inhibitor's effect on SmTK4 . To answer this question , evidential and experimental data were assembled pinpointing SmTK4 as the target of Piceatannol . First , Southern-blot analysis had shown that SmTK4 is a single copy gene [18] , and searches through the schistosome genome data ( [6] , assembly version 3 . 1; www . sanger . uk ) provided no evidence for the existence of other kinases of the Syk class . Therefore , we assume that SmTK4 is the only Syk kinase present in S . mansoni . Since at the concentrations used Piceatannol is specific for Syk kinases [34] , [35] , we concluded that SmTK4 is the only molecule targeted in S . mansoni . Furthermore , localization studies had indicated the testes and ovary as the only reproductive organs transcribing SmTK4 , which is not transcribed in the vitellarium . In this organ , no morphological changes following inhibitor treatment were observed ( result not shown ) . To specifically attribute the effect of Piceatannol to SmTK4 , we post-transcriptionally inhibited SmTK4 using dsRNAs ( RNAi ) . To this end , SmTK4-specific dsRNAs were generated spanning both SH2 domains , including the interdomain A ( linker region between the SH2 domains , Figure 1; 813 bp long ) . To exclude nonspecific side-effects on other tyrosine kinases , SmTK3-specific dsRNAs were also generated and applied as control . Based on the protocol from Ndegwa et al . [47] , ten S . mansoni couples were electroporated with a single square-wave impulse and 25 µg of SmTK4- or SmTK3-specific dsRNAs , respectively . As an additional control , worm couples were electroporated under the same conditions , but without dsRNA . To investigate the RNAi effects at the molecular level , we analyzed SmTK4 transcript levels in the three independent groups of electroporated worms by semi-quantitative RT-PCR ( Figure S1 ) . As endogenous standard , the housekeeping gene SmPDI ( protein disulfide isomerase; [48] ) was used . Five days after electroporation , total RNA was extracted from one half of the worm couples and the SmTK4 transcript level determined by RT-PCR analysis . The products were analyzed by agarose gel electrophoresis , and the intensities of the amplification products were densitometrically quantified using RT-PCR results of SmPDI for normalization . In three independent experiments , the SmTK4-mRNA levels were significantly reduced , although at different levels . In worms electroporated with SmTK4-dsRNA , transcript levels decreased to 10–32% compared to control worms , which had been electroporated without dsRNA . Worm couples treated with SmTK3-specific dsRNAs showed no alterations in the SmTK4-transcript level , indicating the specific silencing effect of the SmTK4-dsRNAs ( Figure S1 ) . Finally , by carmine red-staining and CLSM the morphology of those dsRNA-treated worms was investigated , which revealed the strongest silencing effect ( transcript reduction to 10% ) according to RT-PCR analysis . Control worms , electroporated without dsRNA , showed no alterations in the morphology of the testes or ovary ( Figure 6E–F ) . Additional control worms electroporated with SmTK3-specific dsRNA also showed no morphological alterations in these tissues ( results not shown ) . However , many of the SmTK4-dsRNA electroporated worm couples exhibited phenotypes that were qualitatively comparable to the phenotypes observed after Piceatannol treatment ( Figure 6G–H ) . In males , the size of the testicular lobes and the number of spermatocytes were reduced , and nearly no mature elongated sperms were detected in the ventral part of the lobes or within the sperm vesicle , but some immature spermatogonia ( Figure 6G , asterisk ) . In the ovary of the female , the number of mature oocytes was increased , but not to the same extent as observed after Piceatannol treatment ( Figure 6H ) . To finally test , whether SmTK4 exerts signal transduction activities which can be suppressed by Piceatannol , we made use of the Xenopus oocyte system [49] . Previous studies had shown that signal transduction proteins of schistosomes can be efficiently expressed in Xenopus oocytes . Moreover , parasite kinase activities were already studied in this heterologous system owing to their capacity to induce resumption of meiosis or germinal vesicle breakdown ( GVBD ) [50] , [51] . To analyze the GVBD-inducing capacity of SmTK4 and inhibitor effects on this kinase , two constructs were cloned as templates for cRNA synthesis , a full-length construct and a shortened version of SmTK4 containing only the catalytic TK domain . Following cRNA injection , GVBD was studied by the appearance of a white spot at the center of the animal pole of the oocyte . The results showed that in the absence of progesterone , a steroid inducer of GVBD [49] used for positive control , the TK domain of SmTK4 was able to induce 100% GVBD . This confirmed its catalytic activity ( Table S2 ) . In non-injected oocytes , no GVBD was observed under the same conditions as well as in oocytes transfected with the full-length variant of SmTK4 . This was probably due to a close conformation of the TK domain within the complete protein . When oocytes containing the catalytic TK domain of SmTK4 were incubated with increasing concentrations of Piceatannol , however , GVBD was negatively influenced in a concentration-dependent manner and completely suppressed at a concentration of 5 µM already . The inhibitor had no effect on non-injected oocytes , and did not inhibit progesterone-dependent maturation even when it was used at the concentration of 100 µM ( Table S2 ) . These results confirmed the specific action of Piceatannol on SmTK4 .
SmTK4 was the first Syk kinase identified in a parasitic helminth and shown to be predominantly expressed in the testes and ovary of adults [18] . To elucidate signaling cascades in which SmTK4 participates , screenings of a S . mansoni adult stage Y2H library were performed . Using the tandem SH2-domain as bait , upstream binding partners of SmTK4 were identified such as homologs of a nonsense mRNA reducing factor , a dipeptidyl peptidase III , SmTK3 , and SmTK6 . Although most clones represented homologs of a nonsense mRNA-reducing protein factor ( group A , [52] ) , binding tests with individual SH2 domains failed , indicating nonspecific interaction . Furthermore , interactions between Syk kinases and such factors are not known . Both arguments also apply to dipeptidyl peptidase III ( group B , [53] ) , which represents another nonspecific interaction partner . The other two upstream interaction partners belong to the Src family of CTKs ( groups C , D ) . One represented the already characterized SmTK3 [31] , whilst the other was a novel Src kinase named SmTK6 . In contrast to group A and B clones , the binding of both Src kinases occurred only with the tandem SH2-domain of SmTK4 , and was enhanced by additional phosphorylation in yeast , in the case of SmTK6 by a factor of five . This confirmed the specificity of the detected Syk-Src interactions . SmTK6 turned out to be the strongest upstream interaction partner , and in situ hybridizations demonstrated its transcription within the testes of the male and the ovary of the female . This corresponds to the transcription profiles of SmTK4 , but also SmTK3 , which among other tissues is also expressed in the testes and ovary [31] . In addition to the interaction analyses , the colocalization of SmTK6 , SmTK3 and SmTK4 allows the conclusion that these kinases cooperate in the gonads . Src-Syk interactions are already known from other systems . In mammalian immune cells , Src kinases recruit downstream-acting Syk kinases to the plasma membrane to activate them by phosphorylation [54] , [55] . Thus , high-affinity binding sites are created for molecules acting downstream of Syk [27] that bind to phosphotyrosine residues within the linker region , or the TK domain of Syk [27] , [41] , [56] . From these data we hypothesize that SmTK3 may recruit SmTK6 to the plasma membrane , where the latter one becomes phosphorylated . This may be a prerequisite for binding of SmTK4 by its SH2 domain . By co-IP we found further evidence for SmTK4-SmTK6 binding , which supports the sketched scenario . Future studies will aim to analyze these interactions in more detail . As downstream interaction partners of SmTK4 homologs of caspase 3 , small HSP , a PM20/PM21 type MAPK-activating protein , mapmodulin , and a protein with non-significant similarity were found . The caspase 3 homolog ( group B , [57] ) showed the weakest relative binding strength with all SmTK4 subfragments . Although for ZAP-70 , the second member of the Syk-kinase family in mammals , an influence on caspase pathways has been described [58] , this influence is indirect [59] , and no direct interaction between Syk kinases and caspases has been described yet . In light of these facts , a nonspecific binding may have occurred . The relative binding strength of the small HSP homolog ( group A ) was slightly stronger as for caspase 3 . Besides other functions [60] , HSPs fulfill chaperone functions , and a protective role of HSP90 for ZAP-70 has been demonstrated [61] . Therefore , we assume that the schistosome HSP homolog may fulfill a chaperone-like function for SmTK4 . Mapmodulin ( group D ) , a microtubule-associated protein [62] , [63] , showed high relative binding strengths to both the linker region and the TK domain indicating that it is able to bind to both fragments of SmTK4 . This suggests that SmTK4 may influence the reorganization of the cytoskeleton in spermatocytes or oocytes , which is supported by the in situ-hybridization data showing mapmodulin transcripts in the testes and ovary . Indeed , Syk kinases are known to phosphorylate substrates involved in cytoskeleton organization , or components of the cytoskeleton such as microtubules directly [23] , [64] , [65] . Finally , the MAPK-activating protein of the PM20/PM21 type ( group C ) showed the strongest relative binding activity to SmTK4 . Since this group of proteins lacks defined structural or functional domains , the molecular basis for interaction as well as the function of the schistosome MAPK-activating protein remains uncertain [66] , [67] . Results of our study provide first evidence for its binding to the linker region of SmTK4 . MAPK signaling is acknowledged to be initiated by upstream molecules , such as growth factors and RTKs , which are known to be components of signaling cascades controlling proliferation , differentiation , and survival of cells . A critical role of Syk for MAPK activation has been postulated [68] , although no direct interactions between Syk kinases and a MAPK-activating protein were shown yet . In situ-hybridization experiments colocalized transcripts of the schistosome MAPK-activating protein in the testes and ovary providing further evidence for interaction . This allows the speculation that SmTK4 may trigger a MAPK signaling-pathway in the gonads probably influencing cell proliferation . We tried to find evidence for MAPK activation using Western-blot analyses with antibodies directed against phosphorylated or non-phosphorylated forms of human MAP kinases and protein homogenates from mixed sex schistosomes . In another independent study , these antibodies showed a high specificity also for Echinococcus MAP kinases [69] . Although a highly conserved MAPK homolog exists in the genome of S . mansoni ( [6] , accession number XP_002575049 ) , its structure may be different since the antibodies were not able to detect a band of the expected size ( results not shown ) . However , the capacity of the TK domain of SmTK4 to induce GVBD in Xenopus oocytes provided at least indirect evidence for the potential of the schistosome Syk kinase to activate a MAP kinase cascade since GVBD is only induced when this signaling pathway has been activated [70] . The differentiation of germ cells in vertebrates and invertebrates depends on the re-organization of the actin and tubulin cytoskeletons [71]–[73] . The participation of Syk kinases in these processes , either directly by phosphorylation of cytoskeleton components [64] , or indirectly by the activation of further signaling molecules has already been described [74] , [75] . The identity of the isolated up- and downstream interaction partners of SmTK4 allowed first speculations for a role of SmTK4 in signaling pathways , regulating the proliferation and/or differentiation of germinal cells by activation of a MAPK pathway and/or by influencing cytoskeleton rearrangements . To investigate a cytoskeletal role , the Syk kinase-specific inhibitor Piceatannol was used for functional inhibition of the protein and second , SmTK4-specific dsRNAs were applied for post-transcriptional gene silencing . Since Southern-blot [18] and in silico analyses confirmed that SmTK4 is the only Syk kinase in S . mansoni , and since at the concentrations used Piceatannol only inhibits Syk kinases [34] , [45] , the morphological effects observed by CLSM could be specifically attributed to an inhibition of SmTK4 . This conclusion was supported by GVBD assays in Xenopus oocytes confirming that Piceatannol is able to specifically inhibit the catalytic activity of the TK domain of SmTK4 in a concentration-dependent manner . Inhibitor and dsRNA treatment seemed to disrupt sperm development in schistosome males at an early stage , since the number of mature sperms as well as of spermatocytes was reduced within the testicular lobes . It seems likely that the inhibition influenced the proliferation of the spermatogonial cells in the dorsal part of the testes . The proliferation of these cells is necessary for the initiation of spermatogenesis and the continuous production of mature sperms [76] . Thus , a dysfunction of spermatogenesis at this early stage would lead to the absence of mature sperms . This coincides with studies from other organisms ( mouse , human ) showing that the proliferation of spermatogonial cells is activated by CTKs . Accordingly , Src-specific inhibitors or RNAi led to dysfunctions in spermatogenesis [76] . For CTKs of the Src family , an involvement in spermatogenesis is well-known , and this has also been hypothesized for the Lyn- and Fyn-family kinases [76] , [77] . Indeed , the CLSM analysis of S . mansoni couples treated with the Src kinase-specific inhibitor Herbimycin A also showed significantly reduced numbers of sperms in the ventral part of the testicular lobes and within the sperm vesicle [78] . It was shown before that Herbimycin A affects the biochemical activity of SmTK3 [31] , but we cannot exclude that it also affects SmTK6 activity . In contrast to the Src-family kinases , however , the involvement of a Syk kinase in spermatogenesis has not been confirmed previously in eukaryotes . Harayama et al . [79] first reported the existence of a Syk kinase in spermatozoa of mammals . Here Syk is phosphorylated and activated by a cAMP-activated PKA indicating the involvement of a signaling pathway , which differs from that in hematopoietic cells . It was speculated that the PKA-Syk pathway is associated with the fertilization capacity of sperms [79] , but functional data have not been provided yet . In untreated schistosome females the anterior part of the ovary contains the immature developing oogonia which have stem-cell character , whereas the posterior part contains mature primary oocytes with enlarged cytoplasms . As in other trematodes , primary oocytes leave the ovary to become fertilized within the oviduct before entering meiosis [46] , [80] . Upon Piceatannol treatment , an increased number of primary oocytes and fewer immature oocytes were observed in females . One explanation for this structural change is an inhibition of processes occurring early in oocyte development such as oogonial divisions , leading to a reduced supply of primary oocytes . This is similar to the phenotype observed in the testes and supported by the result of the egg-count reduction test , which showed a significant reduction of egg production within one week in paired females treated with Piceatannol . We expected a complete fading of egg production with elongated time of treatment . However , a long-term effect of Piceatannol on egg production could not be studied because worms died nearly 10 days after treatment in culture before egg production had terminated . Furthermore , a hatching assay showed no morphological differences between developing eggs of inhibitor-treated worms or controls , indicating that SmTK4 may not have an essential function for egg maturation or developing miracidia ( results not shown ) . RNAi approaches performed to post-transcriptionally silence SmTK4 resulted in similar phenotypes in the testes and ovary of adult S . mansoni . This indicates that electroporation combined with soaking was suitable and efficient to induce RNA silencing in the gonads . In previous studies , RNAi effects were observed for genes expressed in the tegument or in the gastrodermis [47] , [81] , tissues more easily accessible to dsRNA . At this time-point , the RNAi protocol for adult schistosomes is not fully reproducible in our hands , as indicated by variable rates of SmTK4 knock-down in the three independent experiments , and by the absence of a phenotype in the SmTK3-dsRNA control . Since the SmTK4-dsRNAs reduced the transcript levels incompletely , phenotypes had been obtained that were qualitatively but not quantitatively comparable to the Piceatannol-induced phenotypes . The efficiency of this inhibitor to suppress SmTK4 function was higher resulting in more dramatic phenotypes in the gonads of adult schistosomes . In oocytes of marine nemertean worms ( Cerebratulus sp . ) , inhibitor studies also showed evidence for a role of Syk kinases , as well as of Src-like CTKs during oocyte maturation [82] . Using the nonspecific tyrosine kinase inhibitor genestein ( 50–100 µM ) , or the Syk-kinase specific inhibitor Piceatannol ( 50–100 µM ) a decrease in the maturation of oocytes was observed , indicated by reduced GVBD occurring in early meiosis , and also by reduced MAPK activity . With Src-kinase inhibitors ( 100 µM PP2 , 20 µM SU6656 ) no reduction in GVBD was observed , but in that of MAPK activity . Based on their data Stricker & Smythe [82] propose that in oocytes of nemertean worms the signal is transduced from RTKs to Syk kinases , followed by Src kinases and ending in a MAPK-signaling pathway that leads to the activation of MPF ( maturation-promoting factor ) and GVBD . This model is basically , but not completely supported by the findings of our study . Besides the indirect evidence for the MAP kinase cascade-inducing capacity of SmTK4 in Xenopus oocytes , Y2/3H analyses identified its interactions with the Src kinases SmTK6 and SmTK3 as upstream partners as well as with a MAPK-activating protein as one of two candidate downstream partner . Since Src kinases are activated by RTKs , it seems feasible to us that in S . mansoni oocytes the signal is transduced from a RTK to a Src kinase leading to the activation of the Syk kinase SmTK4 , which in turn activates a MAPK cascade . Its essential function for gonad development and the long-term killing effect of worms by Piceatannol in culture suggest that SmTK4 may be a candidate target for blocking transmission and disease progression . Although CTKs such as Syk kinases are conserved in the animal kingdom , there may be possibilities to target specific homologs . A comparison of human Syk ( accession number AAH11399 ) and SmTK4 ( accession number CAD13249 ) revealed differences at the amino acid level between 17% ( within the catalytic domain ) and 63% ( whole protein ) . This and the remarkable elongated linker-domain region of SmTK4 may provide a basis for selective drug design .
All experiments involving hamsters within this study have been performed in accordance with the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes ( ETS No 123; revised Appendix A ) and have been approved by the Regional Council ( Regierungspraesidium ) Giessen ( V54-19 c 20/15 c GI 18/10 ) . A Y2H cDNA-library based on RNA of adult male and female S . mansoni was constructed [33] . To this end , cDNAs were cloned into the prey vector pGADT7-Rec ( leucine nutritional marker LEU2 , Clontech ) in frame with the GAL4 activation domain ( Gal4-AD ) . For subsequent library screening , mating was performed according to the user manual ( Yeast protocols handbook , Clontech ) . Two yeast strains were used for the mating; the library-containing strain AH109 ( Mat a; reporter genes ADE2 , HIS3 , and LacZ ) and the bait-containing strain Y187 ( Mat α; reporter genes HIS3 , LacZ ) . As bait vector for library screening , the plasmid pBridge ( tryptophan nutritional marker TRP1 , Clontech ) was used , which contains two multiple cloning sites , MCS I and MCS II . MCS I allows the cloning of protein-coding gene sequences as fusion constructs with the GAL4 DNA binding domain ( Gal4-BD ) . MCS II permits the cloning of a second gene sequence for the expression of an additional protein and , therefore , allows the establishment of a Y3H system . Screening for upstream interaction partners of SmTK4 was performed with a bait vector containing the tandem SH2-domain of SmTK4 cloned into the MCS I in frame with the GAL4 DNA-binding domain ( Gal4-BD ) . The encoding sequence was amplified by PCR using the SmTK4-specific primer pair SmTK4-SH2SH2-5′ ( 5′-GGATCCGTGGAGCTATTCCAC-3′; containing a BamHI restriction site ) and SmTK4-SH2SH2-3′ ( 5′-CTGCAGTGATATACCACCGGA-3′; containing a PstI restriction site ) , and a full-length cDNA clone of SmTK4 as template . Amplification products of the expected size were cloned via BamHI/PstI into the MCS I of the vector pBridge . After cloning , the resulting construct SmTK4-SH2SH2 pBridge was sequenced to confirm the correct open reading frame of the Gal4-BD/SmTK4-SH2SH2 fusion . To perform a Y3H screening for upstream binding partners of SmTK4 a second bait vector was constructed , containing SmTK4-SH2SH2 in the MCS I and , additionally , the coding sequence for the TK domain of the schistosome Src kinase SmTK3 in the MCS II . To this end , the sequence for the SmTK3 TK-domain was amplified by PCR using the SmTK3-specific primers SmTK3-TK-5′ ( 5′-GCGGCCGCATCATCCAGAACCTGTGGG-3′; containing a BglII restriction site ) and SmTK3-TK-3′ ( 5′-AGATCTGCTGGTTGCTCATCTTCAC-3′; containing a NotI restriction site ) , and a full-length cDNA clone of SmTK3 as template . The PCR product was cloned via BglII/NotI into the MCS II of SmTK4-SH2SH2 pBridge resulting in the construct SmTK4-SH2SH2 + SmTK3-TK pBridge . The success of this cloning approach was confirmed by sequencing . For control studies to test binding specificities , modified versions of this construct were additionally cloned by deletion of either the N-terminal or the C-terminal SH2-domain of SmTK4 ( SmTK4-SH2 ( 1 ) + SmTK3-TK pBridge , SmTK4-SH2 ( 2 ) + SmTK3-TK pBridge ) . To screen for downstream interaction partners of SmTK4 three different bait vectors were constructed , containing either the linker-region of SmTK4 , the linker-region and the TK domain , or only the TK domain of SmTK4 in the MCS I . The coding sequences of these regions of SmTK4 were amplified by PCRs with the primer pairs TK4-bait1-5′ ( 5′-GGATCCTACAGAAACCAATACCAGTATC-3′ ) + TK4-bait1-3′ ( 5′-CTGCAGGGTATGCAAATATTTGTTTGT-3′ ) , TK4-bait1-5′ + TK4-bait2-3′ ( 5′-CTGCAGTTCAACAAGAAATTCGATG-3′ ) , or TK4-bait2-5′ ( 5′-GGATCCAAATTTATGATGAATTACCACC-3′ ) + TK4-bait2-3′ , respectively , and a full-length cDNA clone of SmTK4 as template . The 5′-primers contained a BamHI restriction site and the 3′-primers a PstI restriction site , respectively . Amplification products of the expected sizes were cloned via BamHI/PstI in the MCS I of pBridge in frame with the Gal4-BD , resulting in the constructs SmTK4-linker+TK pBridge , SmTK4-linker pBridge , and SmTK4-TK pBridge . For Y3H analysis , an additional vector was constructed , containing the SmTK4 linker region and TK domain in the MCS I , and the SmTK3 TK-domain in the MCS II ( SmTK4-linker+TK + SmTK3-TK pBridge ) . The integrity of the open reading frames with the Gal4-BD was confirmed by sequencing . The screening for upstream or downstream interaction partners of SmTK4 was performed with either the bait construct SmTK4-SH2SH2 + SmTK3-TK pBridge or SmTK4-linker+TK + SmTK3-TK pBridge , respectively . Yeast cells ( strain Y187 ) were individually transformed with both plasmids by the lithium acetate method ( Yeast protocols handbook , Clontech ) . For the screening of the S . mansoni library , bait-expressing Y187 cells were mated with library-containing AH109 cells . Mating efficiencies of 75% or 28% were obtained , respectively , which exceeded the required minimum of 2% in both cases ( Clontech ) . The first selection of diploid yeast cells containing interacting proteins was carried out on synthetic dropout medium lacking the amino acids tryptophan , leucine , and histidine ( Trp−/Leu−/His− ) . To enhance the selection pressure on clones with interacting proteins , grown colonies were plated onto synthetic dropout medium lacking the amino acids tryptophan , leucine , histidine , and adenine ( Trp−/Leu−/His−/Ade− ) . For further selection , β-Gal colony filter assays were performed using X-Gal as substrate according to the manufacturer's instructions ( Yeast protocols handbook , Clontech ) . From positively tested yeast clones , plasmid DNA was isolated using cell disruption by vortexing with glass beads ( Sigma ) followed by plasmid preparation ( PeqLab ) . Isolated plasmid DNA was transformed into heat shock-competent Escherichia coli cells ( DH5α ) , and the bacteria selected on LB-plates containing ampicillin ( 100 µg/µl ) . To differentiate bacterial colonies containing the pBridge bait-plasmid from those containing a pGADT7 prey-plasmid , colony PCRs with pGADT7-specific primers were performed . Prey plasmids from PCR-positive bacterial clones were isolated and sequenced commercially ( AGOWA , Berlin ) . For further binding analyses , the yeast strain AH109 was transformed with appropriate prey plasmids together with different bait plasmids . To confirm protein-protein interactions , the selection procedures were repeated . For quantification of relative interaction strengths , β-Gal liquid assays with ONPG ( o-nitrophenol-galactopyranoside , SIGMA ) as substrate were performed according to the Yeast protocols handbook from Clontech . For isolation of yeast total RNA , a 5 ml overnight culture of the appropriate yeast clone was centrifuged to harvest cells . The pellet was washed two times with PBS and afterwards frozen in liquid nitrogen . Cells were disrupted by three freeze/thaw cycles ( liquid nitrogen , 37°C water bath ) , 1 ml TriFast ( PeqLab ) was then added to the cell lysate , and total RNA was extracted according to the manufacturer's instructions . Total protein extracts from yeast cells were obtained using the urea/SDS method as described in the Yeast protocols handbook ( Clontech ) . For the inhibition of endogenous proteases and phosphatases , the buffer for cell disruption was supplemented with PMSF ( phenylmethylsulfonylfluoride , SIGMA ) , protease inhibitor cocktail ( Complete Mini , Roche ) , NaF ( sodium fluoride , 50 mM ) , and Na3VO4 ( sodium orthovanadate , 2 mM ) . Protein extracts were analyzed by standard SDS-PAGE and Western blotting with an anti-phosphotyrosine-specific antibody ( Santa Cruz Biotechnology ) . To confirm the transcription of the different bait-constructs inside the yeasts , RT-PCRs were performed with total RNA from bait-containing yeast clones as template . RT-PCRs were carried out stepwise in two separate reactions . First , 90 ng of total RNA was converted into cDNA using the SensiScript reverse transcriptase ( Qiagen ) and oligo-d ( T ) , or bait sequence-specific primers . One fourth of the RT reaction volume was used for PCR amplification using appropriate bait sequence-specific primers . Amplification products were analyzed by agarose gel electrophoresis . The pESC-His yeast expression system ( Novagen ) was used for immunoprecipitation experiments . The pESC-His vector contains two multiple cloning sites ( MCS I , contains a FLAG-tag sequence; MCS II contains a cMyc-tag sequence ) under the control of galactose-inducible GAL10 or GAL1 promoters . The tandem SH2-domain of SmTK4 was cloned into MCS I ( FLAG-tag at the C-terminus ) , and the nearly complete version of SmTK6 ( except 250 bp of the N-terminus ) was cloned into MCS II ( cMyc-tag at the N-terminus ) . For cloning , the tandem SH2-domain of SmTK4 was amplified by PCR using gene-specific primers containing appropriate restriction sites ( TK4-SH2SH2-pESC-His-5′ ( NotI ) : 5′-GCGGCCGCAATGGGAGCTATTCCACCG-3′; TK4-SH2SH2-pESC-His-3′ ( ClaI ) : 5′-ATCGATGATATACCACCGGAACCTGA-3′ ) . Before cloning into MCS II , the SmTK6 sequence was also PCR-amplified using gene-specific primers with restriction sites ( TK6-Voll-pESC-His-5′ ( XhoI ) : 5′-CTCGAGAATGTTGTGACTGATGTGCAT-3′; TK6-Voll-pESC-His-3′ ( SacII ) : 5′-CCGCGGTTATCTAAATATTGAGCTTCTGTGTGC-3′ ) . The integrity of the cloned constructs was confirmed by sequencing ( AGOWA , Germany ) . Following transformation , yeast cells ( strain YPH501 ) were grown for 5 days at 30°C on selection media ( synthetic drop-out medium [SD] , His− , + glucose ) to ensure the presence of the vector . Selected clones were grown over night in SD ( His− , + glucose ) liquid medium . The cells were centrifuged the next day and resuspendend in SG medium ( His− , + galactose ) and incubated for 5 h to induce transgene expression . As control , transformed yeast cells were alternatively resuspendend in SD ( His− , + glucose ) . Protein was isolated from 10 ml culture volume for electrophoresis ( 10 µg protein each; 10% SDS-PAGE ) . After blotting on nitrocellulose ( Schleicher & Schüll ) the induction of the expression of the recombinant schistosome proteins was confirmed by Western-blot analysis using an anti-FLAG-tag ( 2 . 5 µg/µl; Novagen ) or an anti-cMyc-tag antibodies ( 1 µg/µl; Novagen ) . A goat anti-rabbit-HRP ( horse raddish peroxidase ) secondary antibody was used for detection ( 1∶70 . 000; Novagen ) . Individual bands of the expected sizes of 31 kDa ( SmTK4-SH2SH2 ) or 55 kDa ( SmTK6 ) were observed in proteins obtained from the galactose-induced yeast culture but not in proteins obtained from the glucose-induced culture ( results not shown ) . For co-IP 70 µl of the protein lysate of the galactose-induced culture were pre-incubated with 30 µl protein A sepharose ( Sigma ) to remove proteins binding unspecifically . The remaining lysate was splitted in two parts , and each half incubated ( over night at 4°C ) with 4 µg anti-FLAG-tag antibody or 1 µg anti-cMyc-tag antibody , respectively . Antibody complexes were precipitated by protein A sepharose ( 2 h at 4°C ) . Following washing steps , the protein complexes were eluated from the protein A sepharose , and a Western-blot analysis was done with the recovered yeast protein as described above . After nitrocellulose transfer , the proteins were visualized by INDIA ink staining ( Pelikan , Germany; 1% acetic acid , 0 . 04% Tween 20 , 0 . 1% Fount India ) . To monitor the transcription of SmTK4 in control and treated S . mansoni , total RNA was extracted using TriFast ( PeqLab ) following the manufacturer's instructions . Residual DNA remaining in the RNA preparations was removed by DNase digestion using RNase-free DNaseI ( Fermentas ) . cDNA was synthesized using 1 µg total RNA , 1 µl oligo-d ( T ) primer ( dT24VN; 20 µM ) , 1 µl nonamer primer ( dN9; 20 µM ) and Superscript II reverse transcriptase ( Invitrogen ) . Subsequent PCRs were performed with 1/10 of the cDNA as template , FIREPol Taq polymerase ( Solis BioDyne ) , and the following primer combination: SmTK4-Sub3-5′ ( 5′-ATGACGTAAAAGATTCACGTG-3′ ) and SmTK4-Sub3-3′ ( 5′-TGCATGTTCTTCACTACAATC-3′ ) , which flank the region used as target for the dsRNAs . For normalization , the transcription of the housekeeping gene SmPDI [48] was monitored using the same cDNAs as template , but using the following primer combination: SmPDI-5′ ( 5′-GGGATTTATCAAGGATACGGACTC-3′ ) and SmPDI-3′ ( 5′-CACCAAGGAGCATACAGTTTGAC-3′ ) . All PCRs were performed in a final volume of 25 µl . PCR products were separated on 1 . 5% agarose gels stained with ethidium bromide . The relative intensities of the amplification products were determined densitometrically using the program ImageJ ( version 1 . 4 . 1; http://rsbweb . nih . gov/ij/index . html ) . For relative quantification of the SmTK4 products , the SmPDI products were used as endogenous standard . In situ hybridizations were done as described elsewhere in detail [31] , [33] . In short , adult worm pairs were fixed in Bouin's solution ( picric acid/acetic acid/formaldehyde; 15/1/5 ) before embedding in paraplast ( Histowax , Reichert-Jung ) . Sections of 5 µm were generated and incubated in xylol to remove the paraplast . Following re-hydration , proteins were removed by proteinase K treatment ( final concentration 1 µg/ml ) , and the sections were dehydrated . For hybridization , in vitro transcripts were labeled with digoxigenin following the manufacturers' instructions ( Roche ) . Labeled sense and antisense transcripts of SmTK6 ( unique site , length 354 bp ) , the MAPK activating protein PM20/21 ( length 452 bp ) , or mapmodulin ( length 420 bp ) , were size-controlled by gel electrophoresis . To prove their quality , transcript blots were made to confirm digoxigenin incorporation by alkaline phosphatase-conjugated anti-digoxigenin antibodies , naphthol-AS-phosphate , and Fast Red TR ( Sigma ) . All in situ hybridization were performed for 16 h at 42°C . Sections were stringently washed up to 0 . 5×SSC , and detection was achieved as described for transcript blots . A Liberian isolate of S . mansoni was maintained in Biomphalaria glabrata as intermediate host and in Syrian hamsters ( Mesocricetus auratus ) as definitive host [83] . Adult worms were obtained by hepatoportal perfusion at 42–49 days post-infection . After perfusion , adult schistosomes were washed three times with M199 medium before being cultured in vitro in M199 ( Gibco; including glucose , sodium bicarbonate , 4- ( 2-hydroxyethyl ) -1-piperazineethane sulfonic acid ) supplemented with an antibiotic/antimycotic mixture ( 1 . 25% , Sigma ) and FCS ( 10% , Gibco ) at 37°C and 5% CO2 [42] . For each experiment , 10–30 pairs of S . mansoni were kept in 60 mm diameter culture dishes in 3 ml culture medium . The medium was changed every 24 hours . If needed , schistosome pairs were carefully separated with fine tweezers . Piceatannol ( 3 , 4 , 3′ , 5′-Tetrahydroxy-trans-stilbene; Alexis Biochemicals ) was dissolved in dimethyl sulfoxide ( DMSO ) ( 5 µg/µl ) . For each inhibitor treatment experiment 20 adult couples of S . mansoni were maintained in 10 ml culture medium [42] , supplemented with various concentrations of Piceatannol ( 35 µM , 70 µM , 100 µM ) . Medium and inhibitor were refreshed every 24 hours during the treatment periods in vitro . For morphological analysis , adult worms were fixed for at least 24 hours in AFA ( alcohol 95% , formalin 3% , and glacial acetic acid 2% ) , stained for 30 minutes with 2 . 5% hydrochloric carmine ( Certistain® , Merck ) , and destained in acidic 70% ethanol . After dehydration for 5 minutes in 70% , 90% , and 100% ethanol , respectively , worms were preserved as whole-mounts in Canada balsam ( Merck ) on glass slides [43] , [44] . CLSM images were made on a Leica TSC SP2 microscope using a 488 nm He/Ne laser and a 470 nm long-pass filter in reflection mode . As basis for double-stranded RNA ( dsRNA ) synthesis , a 813 bp fragment of the SmTK4-coding DNA was amplified by PCR using the gene-specific primers SmTK4-5′ ( 5′-ATGCCTGGAGCTATTCCA-3′ ) and SmTK4-3′ ( 5′-TGATATACCACCGGA-3′ ) and a full-length cDNA clone of SmTK4 as template . The amplification product of expected size was cloned into the pDrive cloning vector ( Qiagen ) . The resulting construct , containing T7 and SP6 RNA polymerase promoters flanking the SmTK4 sequence was used to generate single-stranded RNA by in vitro transcription with T7 and SP6 RNA polymerases ( MEGAscript RNA transcription kit , Ambion ) . The single-stranded RNAs were purified by LiCl-precipitation , resuspended in dH2O , and quantified by spectrophotometry . Equal amounts of the single-stranded RNAs were mixed in annealing buffer ( 500 mM potassium acetate , 150 mM HEPES-KOH , pH 7 . 4 , 19 mM magnesium acetate , sterile dH2O ) and incubated at 68°C for 15 min . Annealing and integrity of the dsRNAs were confirmed by agarose gel electrophoresis . The dsRNA was delivered to adult worms according to the electroporation protocol of Correnti et al . [84] and Ndegwa et al . [47] . Briefly , electroporations were performed in 4 mm cuvettes with 10 couples in 50 µl electroporation buffer ( Ambion ) containing 25 µg dsRNA . A square-wave protocol was applied with a single 20 ms impulse at 125 V and at room temperature ( Gene Pulser XCell™ , Biorad ) . After electroporation , the worms were transferred to complete M199 medium and incubated for 5 days; 48 hours after electroporation the medium was refreshed . Capped messenger RNA ( cRNA ) encoding the full-length cDNA of SmTK4 or a shortened variant containing the catalytic TK domain of SmTK4 was synthesized in vitro using the T7 mMessage mMachine Kit ( Ambion , USA ) . Following synthesis , cRNA was injected into stage VI oocytes according to previously published protocols [50] , [85] . To investigate Piceatannol effects , transfected oocytes were incubated with increasing concentrations ( 1 , 2 , 5 , 10 , 20 , 50 , or 100 µM ) of Piceatannol . Non-injected control oocytes were incubated with the same inhibitor concentrations . As positive control progesterone was used , a steroid known to induce germinal vesicle break down ( GVBD ) in Xenopus oocytes [49] . This oocyte-specific physiological activity was detected by the appearance of a white spot at the center of the animal pole . The following public domain tools were used for sequence analyses: NCBI-BLAST ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) , the Wellcome Trust Sanger Institute S . mansoni OmniBlast server ( http://www . sanger . ac . uk/cgi-bin/blast/submitblast/s_mansoni/omni ) , and GeneDB ( http://www . genedb . org ) . For BLAST analyses to identify Syk kinases in the schistosome genome data set [6] we used the kinase domain sequence as template as the most conserved part of tyrosine kinases . | Parasitic blood flukes of the genus Schistosoma cause schistosomiasis , one of the most important infectious diseases for humans and animals worldwide . Besides their medical importance , schistosomes possess unique biological features . Among these is the sexual maturation of the female , which requires a constant pairing contact with the male . Pairing induces mitogenic activity and differentiation processes in the female that lead to gonad development . This is a prerequisite for egg production , which is closely connected with the pathological consequences of the disease since the eggs are trapped in different host organs , inducing inflammatory processes . Although these correlations are long known , the molecular basis of differentiation processes in female gonads are poorly understood . In the context of identification of signal transduction proteins controlling female reproductive development we identified SmTK4 of S . mansoni , the first Syk-family kinase of a parasite . By using biochemical and molecular approaches in combination with in vitro culture and a novel microscopical technique , we demonstrate in this study the pivotal role of the signaling protein SmTK4 in spermatogenesis and oogenesis of S . mansoni . This is a new attribute for Syk kinases of eukaryotes promoting SmTK4 as a candidate target for blocking transmission and disease progression . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology/germ",
"cells",
"genetics",
"and",
"genomics/gene",
"discovery",
"cell",
"biology/cell",
"signaling",
"developmental",
"biology/cell",
"differentiation",
"infectious",
"diseases/helminth",
"infections",
"genetics",
"and",
"genomics/gene",
"function",
... | 2010 | The Syk Kinase SmTK4 of Schistosoma mansoni Is Involved in the Regulation of Spermatogenesis and Oogenesis |
Frameworks such as BioNetGen , Kappa and Simmune use “reaction rules” to specify biochemical interactions compactly , where each rule specifies a mechanism such as binding or phosphorylation and its structural requirements . Current rule-based models of signaling pathways have tens to hundreds of rules , and these numbers are expected to increase as more molecule types and pathways are added . Visual representations are critical for conveying rule-based models , but current approaches to show rules and interactions between rules scale poorly with model size . Also , inferring design motifs that emerge from biochemical interactions is an open problem , so current approaches to visualize model architecture rely on manual interpretation of the model . Here , we present three new visualization tools that constitute an automated visualization framework for rule-based models: ( i ) a compact rule visualization that efficiently displays each rule , ( ii ) the atom-rule graph that conveys regulatory interactions in the model as a bipartite network , and ( iii ) a tunable compression pipeline that incorporates expert knowledge and produces compact diagrams of model architecture when applied to the atom-rule graph . The compressed graphs convey network motifs and architectural features useful for understanding both small and large rule-based models , as we show by application to specific examples . Our tools also produce more readable diagrams than current approaches , as we show by comparing visualizations of 27 published models using standard graph metrics . We provide an implementation in the open source and freely available BioNetGen framework , but the underlying methods are general and can be applied to rule-based models from the Kappa and Simmune frameworks also . We expect that these tools will promote communication and analysis of rule-based models and their eventual integration into comprehensive whole-cell models .
Rule-based frameworks such as BioNetGen [1–3] , Kappa [4–6] and Simmune [7 , 8] have been used to build detailed kinetic models of signaling pathways ( e . g . , FcεRI [9–11] , TCR [12] , EGFR [13 , 14] , and p53 [15] ) . A rule-based model is composed of multiple “reaction rules” , where each rule specifies a reaction mechanism and its structural requirements , e . g . , a phosphorylation rule would specify the set of binding interactions that bring the kinase into contact with substrate and the specific site on the substrate that is phosphorylated . Current models range in size from tens to hundreds of reaction rules , but these numbers are expected to increase as rule-based models are collectively organized in databases of kinetic interactions [10 , 12 , 14 , 16] and eventually integrated into whole cell models [17] . Large models , whether rule-based or otherwise , are difficult to understand or communicate without good visualization methods . Currently , the size of rule-based model that can be simulated far exceeds the size of model for which useful visualizations can be constructed automatically . In particular , we do not have visualizations that can present the regulatory interactions embedded in a model as a network diagram of signal flows . Also , other than using manual approaches , we do not have an effective approach to build compact pathway diagrams to communicate the model . Solving the automated diagramming problem is necessary to make the leap from opaque machine-readable model descriptions that can only be understood through manual annotation to transparent models that can be understood and explored by the wider community . Why is it challenging to visualize rule-based models ? Tools that formally visualize the model tend to focus on a single type of information , such as what molecular structures are being modeled ( contact map [6] ) , what rules have been defined on those structures ( Simmune [8] , Virtual Cell [18 , 19] , BioUML [20] ) , and how various rules interact with each other ( rule influence diagram [21] , Kappa story [22] ) . To communicate the architecture of the model at a global level , these different types of information have to be integrated into a single diagram , but current approaches such as the Extended Contact Map ( ECM ) [23] , the Systems Biology Graphical Notation: Entity Relationship Diagram ( SBGN:ER ) [24] and the Molecular Interaction Map ( MIM ) [25] rely on human interpretation , which decouples the diagram from the executable model . Methods to automate generation of diagrams include the Simmune Network Viewer [26] , which uses an interactive approach to visualization , and the rxncon regulatory graph [27] , which has a simplified representation of rule-based models that is more amenable for visualization than standard rules . In Fig 1 , we apply a contact map , a conventional rule visualization approach , a rule influence diagram and an extended contact map to a previously published model of immunoreceptor signaling [9] , and below , we discuss the issues raised by each type of information displayed in those diagrams . We also present more detailed comparisons to the remaining tools in Discussion . The contact map ( Fig 1A ) conveys the structural composition of a model ( e . g . in [28–30] and others ) by showing what types of molecular structures are available to compose reaction rules [6] . This includes structured objects called molecules , components , states and bonds , which we explain in more detail in the Methods section . Conventional rule visualizations ( Fig 1B ) show reaction rules as reactant to product transformations . The reactant side includes not just the structures that are to be modified in the rule , but also the structural requirements that need to be matched for the rule to be triggered . To determine the action of a rule , the reader has to compare reactants to products , which can be challenging for complex rules that have a number of structural requirements ( e . g . , rules in the center column of Fig 1B ) . Nevertheless , this is the standard approach to show rules ( e . g . , in [9 , 13 , 28] and others ) , whether using manually drawn diagrams such as Fig 1B or automated diagrams generated by various software ( Simmune [8] , Virtual Cell [18 , 19] , BioUML [20] ) . The rule influence diagram ( Fig 1C ) represents each rule with a single node and each computed interaction between rules as a directed edge [21 , 31] . Each rule interacts with other rules through shared structures , e . g . , a binding rule that produces a kinase-bound configuration regulates a phosphorylation rule that requires the same configuration . However , it is difficult to understand regulatory interactions from just the rule influence diagram because it does not show structures interacting with rules . Also , even moderate-sized models produce unreadably dense diagrams such as Fig 1C , and the computation of influences is quadratic in the number of rules , which is limiting for large models . Both BioNetGen and Kappa frameworks can generate rule influence diagrams , with the Kappa version allowing for different levels of precision [31] . The extended contact map ( Fig 1D ) is an expert-curated diagram that highlights functional roles of various structures and mechanisms as well as emergent regulatory architectures such as feedbacks and cascades [23] . It uses standard diagramming conventions to convey function ( e . g . , round arrowhead to indicate phosphorylation ) , annotation to relate diagram to model ( e . g . , edge label 2 pointing to rule number 2 ) , and secondary documentation to convey biological significance ( e . g . , an attached model guide that indexes and describes each rule ) . Each of these components is constructed manually , which is also true for related methods such as SBGN:ER [24] and MIM [25] ( see Discussion ) . Several recent models make use of the ECM ( [10 , 12 , 14 , 32] and others ) . In this work , we introduce three new methods that together constitute a new visualization framework for rule-based models . First , we introduce a novel compact rule visualization , which is more concise than conventional representations of rules and does not require visual comparison to convey the action of the rule . Second , we develop the atom-rule ( AR ) graph for showing regulatory interactions that can be efficiently derived from rules without pairwise comparisons . The bipartite AR graph displays a global view of how rules interact through the structures present in a model . Finally , because the raw AR graph is too dense for many applications , we present an AR graph compression pipeline that integrates expert knowledge and generates more readable diagrams . These methods are compatible with rules from the three widely-used frameworks of BioNetGen [1–3] , Kappa [4–6] and Simmune [7 , 8] and also with the proposed interchange format SBML-multi [33] . We have provided an implementation in BioNetGen 2 . 2 [3] , which is already available to users and to frameworks that incorporate BioNetGen , such as PySB [34] and Virtual Cell [18 , 19] . The remainder of the paper is organized as follows . In Methods , we briefly describe the new visualization methods and apply them to simple examples . In Results , we apply the methods to larger and more complex models , including a test set of 27 rule-based models from the literature . We use standard measures of graph readability to show that our methods produce more readable diagrams than current alternatives . In Discussion , we present additional comparisons with existing tools and discuss the potential benefits of the new tools for analysis of rule-based models .
In a rule-based model , molecules are structured objects composed of components . Fig 2A shows the BioNetGen language ( BNGL ) specification of molecules Enz and Sub representing enzyme and substrate respectively , along with corresponding visualizations . Enz has component sub and Sub has components enz , p1 and p2 . By convention , a component with a binding function is named after the molecule that it binds . So , sub on enzyme and enz on substrate represent binding sites for substrate and enzyme respectively . Components p1 and p2 represent phosphorylation sites . A component may have one or more modifications available to it , called internal states . For example , components p1 and p2 may be in the unphosphorylated state Y or phosphorylated state pY . Bonds can occur between pairs of components . Here , component sub on an Enz molecule can bind component enz on a Sub molecule to form an enzyme-substrate complex . Patterns , which are constructed from molecules , components , internal states and bonds , specify the reactants and products of a reaction rule . In Fig 2B , we show the BNGL specification of a simple enzyme-substrate system . Each rule requires a rate constant , with reversible rules , such as R1 , requiring rate constants for both forward and reverse directions . In Fig 2C , we visualize the rules using a conventional approach . Each reaction rule explicitly encodes model assumptions about a reaction mechanism . Structural features specified on the reactant side and modified on the product side constitute the reaction center . In rule R1 and its reverse , the sub-enz bond is formed in the forward direction and removed in the reverse direction , which indicates that R1 models reversible enzyme-substrate binding . In rule R2 , the unphosphorylated state of p1 is transformed to the phosphorylated state , which indicates that R2 models phosphorylation of component p1 . Analogously , rule R3 models phosphorylation of component p2 . Features that remain the same on both sides of a rule constitute reaction context , which describes the local conditions necessary for the mechanism to occur . In rules R2 and R3 , the sub-enz bond is present on both sides of the rule , which indicates that the respective phosphorylation mechanisms require the enzyme-substrate binding interaction . Features omitted on both sides of the rule are assumed not to affect the reaction mechanism . Components p1 and p2 are omitted in rule R1 and its reverse , which specifies that binding and unbinding mechanisms are independent of p1 and p2 . Similarly , rules R2 and R3 specify that phosphorylation at p1 is independent of p2 and vice versa . The site graph is a nested graph used to represent patterns [22] , such as the reactants and products in Fig 2C . In this work , we use site graph to refer to the visualization scheme where nodes representing molecules , components and internal states are nested hierarchically and bonds are shown as edges between components . In conventional rule visualization , as shown in Fig 2C , each reactant and product pattern is drawn separately as a site graph . To distinguish reaction center and reaction context , e . g . , to identify that rule rule R2 transforms the internal state of p1 and requires the sub-enz bond , the viewer has to visually compare the graphs from each side of the rule . This imposes a high mental load for complex rules , especially when a large amount of context obscures a much smaller reaction center . In this work , we introduce compact rule visualization ( Fig 2D ) , which does not require visual graph comparison and avoids drawing reaction context twice . We describe its derivation in S1 Appendix . Briefly , we identify and merge structures common to both sides of the rule , then use special nodes called graph operation nodes to represent the modifications performed . The directions of edges on the graph operation node indicate whether a structure is consumed or produced by that operation . In Fig 2D , each rule is shown with the respective operation node , namely AddBond ( R1 ) , DeleteBond ( _reverse_R1 ) , and ChangeState ( R2 , R3 ) respectively . BioNetGen also supports creating and deleting molecules ( AddMol , DeleteMol ) and multiple operations per rule ( S1 Fig ) . To interpret compact rule visualization , the viewer looks for graph operation nodes , which are visually distinguishable from molecule , component and internal state nodes . The structures adjacent to the graph operation nodes constitute the reaction center , whereas the remaining structures constitute reaction context . In this work , we introduce atoms and atom-rule graphs , which enable visualizing the regulatory architecture represented by a set of reaction rules . Atoms are elementary structural features found in patterns . In Fig 3A , using BioNetGen syntax as well as site graph visuals , we show instances of various types of atoms present in the product pattern of rule R2 . They include: The Atom-Rule ( AR ) graph indicates the relationship of a rule with various atoms , which can be reactant , product and/or context . We describe its derivation in detail in S1 Appendix . Briefly , a reactant or product edge is drawn if an instance of the atom is present in the reaction center , on the left or right side of the rule respectively . A context edge is drawn if an instance is present in the reaction context . In Fig 3D , we show AR graphs derived from the rules in Fig 2C , with atomic node labels in BioNetGen syntax . For convenience , the molecule atoms are omitted if there are no molecules added or deleted in the rule . To interpret the AR graph , one views each atom as a class of actionable sites present in the model . For example , Sub ( p1~Y ) represents the class of unphosphorylated states on p1 components that can potentially be acted upon by phosphorylation mechanisms . Then , one interprets each edge as an interaction between a mechanism and a class of sites . A reactant or product edge respectively indicates that a mechanism has a consumption or production effect on that particular class of sites . A context edge indicates that the mechanism requires that particular class of sites as a local condition . For example , from the AR graph of rule R2 in Fig 3B , we infer that R2 consumes unphosphorylated p1 , produces phosphorylated p1 , and requires that p1 be unbound and that enzyme be bound to substrate . The model AR graph , as in Fig 3C , is a bipartite graph between rules and atoms that is constructed by merging AR graphs of individual rules . Paths on the model AR graph that alternate between rules and atoms represent signal flows . A particular set of rules will always produce the same AR graph , which is a complete representation of signal flow in that rule set between atoms and rules . To build compact pathway diagrams that convey function , we provide a pipeline for reducing the complexity of the model AR graph ( Fig 4A ) while preserving relevant regulatory features . Briefly , it involves: The output of this pipeline is the compressed model AR graph . To decide which atoms and rules to remove ( Step 1 ) as well as which atoms belong together as groups ( Step 2 ) , we take a semi-automated approach . An automated heuristic grounded in commonly encountered biological scenarios makes a first pass through the full AR graph and outputs a template file containing the choices made by the heuristic . To account for nuances of individual systems , the user can edit this template to make alternate choices and import it back into the visualization tool ( see tutorial in S2 Appendix for a demonstration ) . Following this , an automated procedure examines each rule on the graph , the edges incident on the rule and the atom groups adjacent to the rule , then groups rules that share the same edge signature ( Step 3 ) . Currently , we support two types of edge signature: strict , which examines all three edge types , and permissive , which examines only reactant and product edges . Finally , an automated procedure replaces each group of nodes with a single representative node ( Step 4 ) . Edges incident on individual nodes are merged onto the representative node . A particular set of pipeline inputs ( edge signature , template ) will generate the same compressed AR graph , but these inputs can be tuned to produce different compressed AR graphs . Each step in the pipeline has a specific interpretation . Atoms and rules that are removed represent structures and mechanisms with low functional priority , which are typically free binding sites , unphosphorylated states , unbinding rules and dephosphorylation rules . Atom groups represent functional categories of biological structures , e . g . , the set of phosphorylation sites on a receptor . Rule groups represent categories of similarly acting mechanisms , e . g . , phosphorylation mechanisms active at a particular group of sites . Merging groups is equivalent to reducing the resolution of the graph from individual sites and processes to broad categories of those elements . Permissive grouping also introduces a weaker semantic for the context edge on the compressed graph: a merged group node with a context edge implies that at least one of its members prior to merging had the same context edge . An implementation of the methods described here is freely available as part of the open source BioNetGen distribution at http://bionetgen . org . A typical procedure involves calling a “visualize ( ) ” method from the BioNetGen model file with arguments for user input as well as a template file with edits , if applicable . The default template file can also be automatically generated as a text file . The typical visualization output is a file in GML format ( graph modeling language ) [41 , 42] encoding nodes , node labels , edges , edge directions and style attributes of nodes and edges such as color and shape . To lay out the graph , i . e . , assign specific coordinates to nodes , we recommend using a third party application such as the yEd graph editor ( http://yworks . com/yed ) , which was also used for the graphs shown in this paper . The tutorial in S2 Appendix provides a detailed walkthrough of the visualization tools using the model from Suderman and Deeds [40] as an example . We compiled a list of 27 rule-based models from the literature , which we list in S1 Table and attach in S1 Dataset . The models had 2239 rules in total , with the number of rules per model ranging from 6 to 625 . We applied to these models a suite of nine visualization tools: contact map , conventional rule visualization , compact rule visualization , Simmune Network Viewer , rule influence diagram and atom-rule graphs at various steps in the complexity reduction pipeline: full model AR graph , AR graph with background removed , AR graph compressed using a strict edge signature , and AR graph compressed using a permissive edge signature . The compression pipeline was applied automatically by making default choices for prioritizing and grouping nodes . On the output graphs , we computed number of nodes ( n ) and number of edges per node ( e/n ) , counting hierarchical relationships between nodes also as edges . We present these statistics in the Results section . Pseudocode for the algorithms underlying the tools as well as a detailed accounting of computational costs is available in S1 Appendix . Briefly , for compact rule visualization , the rate-limiting step is building a correspondence map between left and right sides of the rule . Given a maximum finite rule size , the cost can be considered as O ( 1 ) per rule . Examining the rule with the correspondence map to synthesize the AR graph is also O ( 1 ) per rule . Merging AR graphs of individual rules , grouping rules and merging groups are all O ( n ) , where n is the number of rules .
Visualizing individual rules promotes understanding the structural and kinetic assumptions encoded in a model . Unlike conventional rule diagrams , which require visually comparing reactant and product sides of a rule , compact rule visualization explicitly indicates which modification is performed on which set of structures . Specifically , it allows us to distinguish reaction center , the site of action of a rule , from reaction context , the structural requirements that need to be matched for the rule to fire . In Fig 5A , we show compact rule visualizations of four reaction rules from the immunoreceptor signaling model of Fig 1 . Rules R3 and R6 have AddBond operations and represent two distinct binding modes of Lyn kinase to the β domain of FcεRI receptor . In R3 , the U domain of Lyn binds the unphosphorylated β domain ( constitutive binding ) , whereas in R6 , the SH2 domain of Lyn binds the phosphorylated β domain ( activated binding ) . Rules R4 and R7 have ChangeState operations and represent phosphorylation of the β domain in receptor dimers , with the active kinase being Lyn recruited through constitutive and activated modes respectively . To understand a model , it is important to know how rules interact with each other and whether they form common motifs such as feedback or feedforward loops . For example , the rules in Fig 5A constitute a positive feedback loop: phosphorylation of β domain ( R4 , R7 ) activates Lyn binding ( R6 ) , which in turn promotes β phosphorylation ( R7 ) , but this is not obvious from conventional and compact rule visualizations . Current methods identify regulatory interactions between pairs of rules through graph comparison [21] , simulation [6 , 22] , or manual interpretation [23] . In contrast , the atom-rule graph , which is a bipartite graph showing regulatory interactions between rules and elementary structural features called atoms ( see Methods ) , is constructed efficiently by examining each rule’s reaction center and reaction context . In Fig 5B , we show an AR graph constructed from rules R3 , R4 , R6 and R7 , and the feedback loop is visible as a path on this graph . The model AR graph for the full immunoreceptor model ( Fig 6A ) is a complete representation of signal flow in the model , encompassing all 24 rules . The compression pipeline ( described in Methods ) extracts the essential features of signal flow from the model AR graph and displays them as a compact pathway diagram . The steps of the pipeline , which we apply to the model AR graph in Fig 6A , include: In Step 1 , we remove unphosphorylated states , dissociation rules and dephosphorylation rules from Fig 6A , producing the graph in Fig 6B . In Step 2 , we group bonds that link the same molecules ( Lig|Rec , Lyn|Rec , Rec|Syk ) and phosphorylation sites on molecules ( Rec_pY , Syk_pY ) , producing the atom groups shown in S2A Fig . In Step 3 , grouping rules that share similar reaction centers and contexts produces the rule groups shown in S2B Fig , whereas dropping the context similarity requirement produces the more inclusive rule groups shown in S2C Fig . In Step 4 , merging groups shown in S2B and S2C Fig produces the compressed AR graphs in Fig 6C and 6D respectively . Unlike the full AR graph , the compressed graphs are compact and easier to understand . It is also easier to trace specific signal flows on the compressed graphs , such as the feedback between Lyn-receptor binding and receptor phosphorylation ( edges marked x in Fig 6A–6D ) . Under default settings , the whole pipeline is automated , but the resolution of the compressed graphs and the quality of the output diagram can be tuned by providing user input , which includes customizing the heuristics for Steps 1 & 2 and choosing the grouping strategy for Step 3 . The strict grouping used in Fig 6C resolves three variants of Syk phosphorylation under various contexts ( nodes 1–3 ) and constitutive and phospho-activated Lyn|Rec binding modes ( nodes 4–5 ) , whereas the permissive grouping in Fig 6D merges variants of the same process and represents them with a single node ( nodes 6 , 7 ) . A specific set of pipeline inputs reproducibly generates the same compressed graph from the model and serves as diagram documentation . To test the scaling of our approach to the growing set of large rule-based models [10 , 12 , 14 , 16] , we applied the AR graph compression pipeline to two extensive models of receptor signaling: the FcεRI rule library constructed by Chylek et al . [10] ( 17 molecule types , 178 rules ) , and the ErbB signaling model constructed by Creamer et al . [14] ( 19 molecule types , 625 rules ) . The compressed graphs for these libraries are shown in Figs 7 and 8 respectively . Unlike the manually constructed Extended Contact Maps ( ECMs ) [23] that were published with these models , the graphs we show are pathway diagrams that were generated directly from the model specification . In S2 Appendix , we provide a tutorial on generating similar diagrams using the yeast pheromone signaling model of Suderman and Deeds [40] ( 26 molecule types , 272 rules ) as an example . The modeler can customize pipeline inputs to capture specific biochemical features in the model as well as strike a balance between compression and resolution on the output graph . For example , the default heuristic assumes that co-occurring phosphorylation sites can be grouped together , but for the FcεRI model , we wanted to distinguish between co-occurring phosphorylation sites with opposing functions , specifically those on Src family kinases Lyn and Fyn ( SFKs ) . So , during atom grouping , we grouped functionally similar sites across molecules , e . g . , the group SFK_Act_p contains activation-related phosphorylation sites on both Lyn and Fyn . As a result , the output graph ( Fig 7 ) resolves the regulatory interactions of a generic SFK rather than Lyn and Fyn individually . Similarly , for the much larger ErBb model , creating functional groups such as ligands , receptors , and receptor dimers caused a dramatic reduction in complexity , with the output graph ( Fig 8 ) showing signaling interactions of a generic ErbB receptor . Alternatively , grouping Lyn sites separately from Fyn or EGFR and ErbB2 receptors separately from ErbB3 and ErbB4 will produce graphs larger than those shown in Figs 7 & 8 , with regulatory interactions resolved in more detail . The compressed AR graph offers a convenient venue for analysis and exploration of a rule-based model . For example , on the FcεRI and ErbB graphs , we were able to identify well-known pathways such as MAPK ( transparent overlays in Figs 7 & 8 ) using a combination of node clustering and visual inspection . Also , on the FcεRI graph , we were able to trace network motifs encoded in the model ( Fig 9 ) by Chylek et al . [10] . Without the compressed AR graph , the same analyses would have required examining hundreds of complex rules in various combinations , which would have required significant effort . Thus , the compressed AR graph offers a useful proxy for the rule-based model that is more amenable to analysis . To assess the readability of various visualization tools , we examined the joint distribution of graph size n and edge density e/n for each visualization when applied to 27 published rule-based models ( see Methods ) , where n and e refer to number of nodes and edges respectively . In S3 Fig , we report these distributions for 9 visualization methods , and in Fig 10 , we show their geometric means . The choice of metrics follows from Ghoniem et al . [43] , who determined that user performance on visual graph analysis tasks decays with increasing graph size and edge density . Ghoniem et al . used much denser graphs than the ones in our test set , so we replaced their edge density metric √ ( e/n2 ) with e/n , which has a higher coefficient of variation for the graphs in our test set ( 2 . 54 vs 1 . 03 ) , and therefore higher discriminatory power . The results in Fig 10 confirm our qualitative observations on the readability of current visualizations and the improvements present in our new ones . Contact maps are generally compact with sparse edges as they only show structural composition and do not show individual mechanisms or signal flow . Rule visualizations , both conventional and compact , produce large graphs with sparse edges as they show the patterns encoded in each rule . However , compact rule visualizations are smaller than conventional ones as they make use of graph operation nodes . Diagrams showing interactions of rules are typically dense , such as rule influence diagrams and full AR graphs . However , full AR graphs have much lower edge density than rule influence diagrams as they use atoms to mediate interactions between rules . When the compression pipeline is applied , AR graphs’ size and edge density can be reduced to approach that of contact maps . This makes compressed AR graphs as readable as contact maps , while conveying substantially more information about the signaling architecture . The Simmune Network Viewer , which is intermediate between rule visualizations and full AR graphs , is discussed in detail below .
Edward R . Tufte , a pioneer of modern data visualization and analytic design , argues that “universal cognitive tasks” underlie how humans perceive information and motivates that “cognitive tasks should be turned into design principles” [44] . In the biochemical literature , diagrams and text employ a number of such cognitive tasks , and our automated methods recapitulate some of these . For example , one often describes a biochemical process using an action verb such as “binds” or “phosphorylates” . Graph operation nodes in compact rule visualization ( Fig 2D ) play a similar role in conveying the action of a rule . Similarly , one uses “site” to denote a molecular part that behaves distinctly or is targeted by a specific process . Atoms used in the atom-rule graph ( Fig 3A ) have a similar interpretation as types of actionable sites . Literature descriptions and diagrams also selectively emphasize active states over ground states and signal-activated processes over processes that attenuate the signal or occur in the background , which allows the reader to filter redundant information . Removing low priority nodes on the AR graph follows a similar principle ( Fig 4B ) . Text descriptions routinely categorize molecules and sites using principles such as homology and functional similarity [45–48] , and use broad terms to summarize information about specific molecules and sites . Grouping atoms and rules using the described heuristics ( Fig 4C–4D ) and compressing the AR graph ( Fig 4E ) recapitulates this approach . Whenever compression is applied to data , there exists a many-to-one relationship between the uncompressed and compressed representations . In the context of visualization , a rule-based model will generate the same conventional and compact rule visualizations and vice versa , but different models can generate the same contact map , rule influence diagram and AR graph . Therefore , one should use each tool at the resolution for which it is designed to be used . Compact rule visualization should be used to show the mechanism underlying each rule . The AR graph is less useful for this purpose , as it approximates each rule as a bipartite graph . Instead , it should be used to infer interactions between rules through formal or informal approaches . When applying the compression pipeline to the AR graph , one should verify that the choice of inputs is biologically reasonable . If this is the case , then the compressed AR graph is useful for both communicating the model to others as well as graph analysis . In addition to the approaches discussed in Introduction ( Fig 1A–1D ) and Methods ( Fig 2C ) , we show examples of other currently available tools ( Fig 11 ) and how they compare with compact rule visualizations and atom-rule graphs . The SBGN Process Description ( Fig 11A ) [24] is a visualization standard for reacting entities . It has the same limitation as conventional rule visualization , namely the need for visual graph comparison . The Kappa story ( Fig 11B ) [22] shows the causal order in which rules can be applied to generate specific outputs , and these are derived by analysis of model simulation trajectories . It is complementary to the statically derived AR graph for showing interactions between rules , but it does not show the structures that mediate these interactions nor does it provide a mechanism for grouping rules . Integrating Kappa stories with AR graphs is an interesting area for future work . The Simmune Network Viewer ( Fig 11C ) [26] compresses the representation of rules differently from the AR graph: it merges patterns that have the same molecules and bonds , but differ in internal states . Like the AR graph , it shows both structures and rules , and it produces diagrams with much lower density ( ‘sim’ in Fig 10 ) , but it obscures causal dependencies on internal states ( S4 Fig ) . The SBGN Entity Relationship diagram ( Fig 11D ) [24] and the Molecular Interaction Map ( Fig 11E ) [25] , like the Extended Contact Map [23] , are diagrams of model architecture that rely on manual analysis . The rxncon regulatory graph ( Fig 11F ) visualizes the rxncon model format [27] , which uses atoms ( called elemental states in rxncon ) to specify contextual influences on processes . This approach , which is also followed in Process Interaction Model[49] , is less expressive than the graph transformation approach used in BioNetGen , Kappa and Simmune ( S5 Fig ) . The AR graph we have developed generalizes the regulatory graph visualization so it can be derived from arbitrary types of rules found in BioNetGen , Kappa and Simmune models . The AR graph offers many advantages over existing methods , but there are a number of ways in which it could be improved or generalized . There are alternate ways to show the content of the AR graph , for example , as a two-dimensional matrix [43] . The compression algorithms can be extended to identify more complex relationships , for example , treating the consumption of an active state as an ‘inhibits’ relationship , grouping enzyme-binding and catalytic processes together as a Michaelis-Menten mechanism , etc . In the immediate future , we plan to add support for other features present in the BioNetGen model specification , such as compartmental states , transport rules and dependencies encoded in rate laws [50 , 51] . Additionally , the AR graph opens up rule-based models to a wide variety of analysis and visualization tools , as it transforms a complex rule-based model into a simple bipartite graph . For example , simulation fluxes can be conveniently visualized on a bipartite graph by mapping numeric values to node size or edge thickness [52] . Also , as rxncon developers have shown , one can perform stochastic Boolean simulations on a bipartite graph [53] . Model reduction approaches developed for rule-based models have previously used information on interactions between structures and rules [54] that can now be obtained directly from the AR graph . The AR graph also serves as a rich source of information that could be mined using formal approaches . Potential areas where new methods can be developed include identifying model subsystems ( as in Figs 7 and 8 ) by graph partitioning [55] , identifying network motifs ( as in Fig 9 ) by cycle detection [56] , dynamically grouping atoms and rules using graph structure discovery [57 , 58] , etc . Thus , adoption of the AR graph could pave the way for novel applications of graph analysis , data mining and machine learning to rule-based models . A natural future direction for signaling models is to explore the effects of complex input stimuli and crosstalk between pathways [59 , 60] on a comprehensive scale . This would require integrating rules from multiple sources , such as databases constructed in tandem by different groups ( e . g . [10 , 12 , 14 , 34] ) . The recently published whole cell model of Mycoplasma genitalium [17] makes effective use of databases to organize and visualize kinetic information [61–63] and provides proof-of-concept of a database-oriented approach . Currently , models of signaling from various receptors have as many as hundreds of rules [10 , 12 , 14] and this number is expected to increase by an order of magnitude to cover more molecule types , receptors and signal pathways . We expect that AR graphs will play a role in the construction , navigation and visualization of the rule-based databases of the future , similar to approaches deployed on other biological data ( VisANT [64] , ChiBE [65] ) . The AR graph will also be useful for frameworks that implement rule-based data structures ( SBML-Multi [33] , BioPax Level 3 [66] ) or integrate rules with higher-order model composition ( Virtual Cell [18 , 19] , PySB [34] ) . Thus , in addition to the immediate benefit of visualizing and understanding large models , the AR graph is expected to be useful in developing the comprehensive cell models of the future . | Signaling in living cells is mediated through a complex network of chemical interactions . Current predictive models of signal pathways have hundreds of reaction rules that specify chemical interactions , and a comprehensive model of a stem cell or cancer cell would be expected to have many more . Visualizations of rules and their interactions are needed to navigate , organize , communicate and analyze large signaling models . In this work , we have developed: ( i ) a novel visualization for individual rules that compactly conveys what each rule does , ( ii ) a comprehensive visualization of a set of rules as a network of regulatory interactions called an atom-rule ( AR ) graph , and ( iii ) a set of procedures for compressing the AR graph into a pathway diagram that highlights underlying signaling motifs such as feedback and feed-forward loops . We show that these visualizations are compact and informative across models of widely varying sizes . The methods developed here not only improve the understandability of current models , but also establish principles for organizing the much larger models of the future . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"phosphorylation",
"linguistics",
"enzymes",
"enzymology",
"social",
"sciences",
"signaling",
"networks",
"reactants",
"network",
"analysis",
"computer",
"and",
"information",
"sciences",
"syntax",
"network",
"motifs",
"proteins",
"chemistry",
"grammar",
"physics",
"bioch... | 2017 | Automated visualization of rule-based models |
Human T-Lymphotropic Virus Type 1 ( HTLV-1 ) infection causes lethal adult T-cell leukemia ( ATL ) and severely debilitating HTLV-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) in up to 5% of infected adults . HTLV-1 is endemic in parts of Africa and the highest prevalence in West Africa ( 5% ) has been reported in Caio , a rural area in the North-West of Guinea-Bissau . It is not known which HTLV-1 variants are present in this community . Sequence data can provide insights in the molecular epidemiology and help to understand the origin and spread of HTLV-1 . To gain insight into the molecular diversity of HTLV-1 in West Africa . HTLV-1 infected individuals were identified in community surveys between 1990–2007 . The complete Long Terminal Repeat ( LTR ) and p24 coding region of HTLV-1 was sequenced from infected subjects . Socio-demographic data were obtained from community census and from interviews performed by fieldworkers . Phylogenetic analyses were performed to characterize the relationship between the Caio HTLV-1 and HTLV-1 from other parts of the world . LTR and p24 sequences were obtained from 72 individuals ( 36 LTR , 24 p24 only and 12 both ) . Consistent with the low evolutionary change of HTLV-1 , many of the sequences from unrelated individuals showed 100% nucleotide identity . Most ( 45 of 46 ) of the LTR sequences clustered with the Cosmopolitan HTLV-1 subtype 1a , subgroup D ( 1aD ) . LTR and p24 sequences from two subjects were divergent and formed a significant cluster with HTLV-1 subtype 1g , and with the most divergent African Simian T-cell Lymphotropic Virus , Tan90 . The Cosmopolitan HTLV-1 1aD predominates in this rural West African community . However , HTLV-1 subtype 1g is also present . This subtype has not been described before in West Africa and may be more widespread than previously thought . These data are in line with the hypothesis that multiple monkey-to-man zoonotic events are contributing to HTLV-1 diversity .
The retrovirus Human T-Lymphotropic Virus type 1 ( HTLV-1 ) was first discovered in 1979 [1] . Approximately 5% of HTLV-1 carriers develop Adult T-cell Leukemia ( ATL ) or HTLV-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) . ATL ( and other HTLV-1 associated diseases have been described in West African individuals [2] , [3] , [4] . It is not known how many people are infected worldwide [5] , but the prevalence is high in Japan , Africa , the Caribbean and South America ( reviewed in [6] ) . Transmission of HTLV-1 is vertical ( mainly through prolonged breastfeeding ) , sexual and via contaminated blood products ( reviewed in [6] ) . HTLV-1 infection is believed to have originated in multiple zoonotic events from monkeys . Cross-species transmission of HTLV-1 from monkey to men is probably ongoing and still actively contributing to the endemic [7] , [8] . HTLV-1 is an ancient retrovirus with a very low evolutionary rate ( especially compared to HIV ) [9] . The Simian T-cell Lymphotropic Virus and HTLV are together referred to as Primate T-cell Lymphotropic Virus ( PTLV ) . Currently , 7 different HTLV-1 subtypes , based primarily on the nucleotide sequence of the LTR region , have been described; the Cosmopolitan subtype ( 1a ) that has spread worldwide and is further divided into subgroups ( A–E ) , 4 African subtypes ( 1b , 1d , 1e , 1f ) and a Melanesian/Australian subtype ( 1c ) . In 2005 a new African subtype g was described by Wolfe et al . based on sequence data from two infected subjects [7] . The origins of HTLV-1 are believed to be either in Asia or Africa and current data do not provide clear support for one or the other continent . The most divergent STLV-1 strain thus far identified is Asian ( MarB43 ) suggesting that MarB43 separated earlier from an ancestral virus than the PTLV-1s found in Africa [10] and supporting a hypothesis that the origin of PTLV-1 lies in Asia [10] . On the other hand , only HTLV-1 subtype a has been found in Asia , while all known HTLV-1 subtypes are present in Africa , providing support for an African origin of HTLV-1 [11] . Additional samples and phylogenetic data are needed to answer this question [12] . After an initial dispersion in Asia and Africa , the Cosmopolitan subtype 1a spread worldwide and is thought to have been introduced into the Americas during the post-Columbian slave trade [13] . In people of African descent in South America , mainly Cosmopolitan subtype 1a subgroup A is found [13] . In Africa , subgroup A has been found primarily in Southern Africa while in West Africa ( the origin of a large proportion of the slaves that were taken to South America ) only subgroups C and D have been identified so far [14] . This presents a puzzle - subgroup A is common in South Americans of African decent yet uncommon ( at least based on the currently available sequence data ) in the African regions from which a large proportion of slaves were brought to South America . Additional HTLV-1 sequence data from West Africa are needed to resolve this enigma . The study area of the current analysis is a rural community in Guinea-Bissau and has the highest reported HTLV-1 prevalence in West Africa ( 5% in 2007 ) [15] , but no sequence data exist from this area and the subtypes of HTLV-1 circulating in the area are not yet identified . Also , the contribution of the various HTLV transmission routes in this community is not well documented . To examine the molecular epidemiology and to define the routes of transmission of HTLV-1 in the Caio community in Guinea Bissau , we characterized the HTLV-1 LTR and p24 region from a collection of DNA samples isolated from children and adults from this cohort . We describe highly conserved examples of HTLV-1 subtype d , most closely related to isolates reported from Bissau and two examples of subtype g , which is the first time this subtype has been reported in West Africa .
Informed consent was obtained from all the study participants . From 2003 onwards , written informed consent was obtained as required by the Gambian Ethics Committee . In studies prior to the 2003 study , written informed consent was not required by the Gambia Government/MRC MRC Laboratories Joint Ethics Committee nor by the Ministry of Health of Guinea-Bissau and thus verbal consent was obtained . The documentation of the informed consent was done in a two-step process . After information on the study was given by the field worker , and after having responded to any questions that the prospective participant might have , the field worker asked whether the participant was willing to participate in the study . The field worker noted down the answer by ticking one of two boxes ( Accepting to participate in study or not ) ; a separate question was asked about providing a blood sample and the answer to this question was noted separately . . All studies , ( both prior to and after 2003 ) , were approved by the Gambia Government/MRC MRC Laboratories Joint Ethics Committee and by the Ministry of Health of Guinea-Bissau . The study area , Caio , is a settlement of 10 small villages with 10 , 000 inhabitants in North-Western Guinea-Bissau and has been described in detail [16] . A rolling census has been performed in the area since 1988 as part of a research project on HIV-2 [17] . In 1990 , 1997 and 2007 , population surveys were conducted to study the prevalence , incidence and risk factors of HIV-1 , HIV-2 , and HTLV-1 infection and these surveys included approximately 75% of the adult population [18] . A cohort was established to study the natural history of HIV-2; included were all HIV-2 infected adults and a similar number of controls , matched for age , sex , and village . This cohort was established in 1991 , and subjects were examined in 1996 , 2003 , and 2006 . In addition , in 2004 a cross-sectional study was performed to examine mother-to-child transmission of HTLV-1 [19] . HTLV-1 prevalence has been stable at around 5% in Caio throughout this period [15] . In the studies conducted in 2003 , 2006 and 2007 , DNA was extracted from blood samples ( whole blood and peripheral blood mononuclear cells ) , and stored for molecular analyses . Because these samples have been used in different studies , not each HTLV-1 infected individual had a stored sample; some samples had been depleted during other studies or could not be traced or storage had failed ( e . g . freezer breakdowns ) . The available samples were used for HTLV-1 sequencing for the current study . Samples for LTR sequencing were selected in 2010 from stored DNA samples . All complete LTR sequences obtained from mother-child pairs from the vertical transmission study in 2004 were also used in the current analysis ( accession numbers JN655856 to JN655872 ) . [19] . Samples for p24 sequencing were randomly selected in 2009 from stored DNA samples for an as yet unpublished immunological study . In the 1997 , 2003 and 2004 studies , plasma samples were tested for the presence of antibodies to HTLV-1 and HTLV-2 using a Murex HTLV-1+2 ELISA ( Abbott Murex Diagnostics , Dartford , UK ) . Reactive samples were retested by the same assay . HTLV-1 infection status was further determined by PCR using primers targeted to the tax/rex gene [20] . Each PCR reaction also included 10 copies of phage lambda DNA . This was co-amplified with specific primers in the same reaction to control for non-specific inhibition . Because the tax/rex primers amplified both HTLV-1 and HTLV-2 sequences , the amplicons were digested with the restriction enzyme Sau 3A ( which cuts only HTLV-1 ) to distinguish products from the two viruses . In this study all PCR-positive samples were confirmed to be HTLV-1 by Sau 3A digestion . The HTLV-1 testing in the 2007 study ( 2 ELISAs and PCR confirmation ) and the HIV testing of previous studies have been described in detail [15] . Amplification of the HTLV-1 proviral DNA was performed with nested PCR . PCR primers were optimized for a melting temperature ( Tm ) between 57 and 60°C and are listed in Table 1 . The complete LTR region was obtained by amplifying two overlapping fragments similar to the strategy used by Salemi et al . [21] . For each patient sample , a 438 bp 5′ LTR-gag ( primers MO195-198 ) and a 475 bp tax-3′LTR fragment ( primers MO199-202 ) were obtained . The p24 coding region was amplified on an 840 bp fragment using the nested primers MO076-079 . PCR products were purified using Qiaquick Gel Extraction ( Qiagen ) and sequenced using the inner PCR primers from both directions by Macrogen ( www . macrogen . com ) . Complete LTR and p24 sequences were assembled and ambiguities were resolved using the sequence alignment editor BioEdit . [22] All new sequences generated here have been deposited in Genbank with the following accession numbers ( JQ583778 to JQ583845 ) ( Table S1 ) . Multiple sequence alignments of the LTR and the related sequences in the GenBank/EMBL database were performed with the Clustal [23] and Dambe softwares [24] . A minimal editing of the alignment was performed manually with GeneDoc [25] . Neighbor-joining ( NJ ) and maximum-likelihood ( ML ) phylogenetic analyses were performed with PAUP* , version 4 . 0b10 [26] . The Tamura-Nei evolutionary model ( taking into account a different substitution rate for transversions , purine and pyrimidine transitions and allowing an intersite substitution rate heterogeneity modelled with a gamma distribution ) was chosen as best model ( alpha parameter of 0 . 6 ) using Modeltest 3 . 06 [27] . The NJ tree was constructed with an optimized nucleotide substitution rate matrix and gamma shape parameter using empirical base frequencies . The reliability of the NJ trees was evaluated by analyzing 1 , 000 bootstrap replicates and bootstrap values of >60% were considered significant . For the ML tree reconstruction a heuristic search with the subtree-pruning-regrafting branch swapping algorithm was performed using the NJ tree as starting tree including its optimised parameters . A likelihood ratio test was used to calculate the statistical support for the branches ( expressed in p-values ) . Trees were drawn with TreeView 1 . 4 [28] .
HTLV-1 LTR and/or p24 sequences were obtained from 72 individuals ( Table 2 ) . In order to obtain LTR sequences from persons who had been infected relatively recently ( between 1997 and 2007 ) or earlier ( prior to 1990 or 1997 ) , samples were selected from individuals diagnosed with HTLV-1 in 2007 . Other samples were randomly selected from the 2003 study . Of the 72 individuals yielding sequence , 14 ( 19% ) were diagnosed with HTLV-1 in 1990 , 17 ( 24% ) in 1997 , 5 ( 7% ) in 2003 and 28 ( 40% ) in 2007 ( Table S1 ) . Twenty-five ( 35% ) individuals were co-infected with HIV; 1 was HIV-1 , 19 were HIV-2 and 5 were HIV-1/HIV-2 dual co-infected . Eight ( 11% ) sequences came from children , who were all diagnosed with HTLV-1 in 2004 and who were all HIV negative . Information on relationships between individuals from whom sequences were available was obtained from census data and interviews; there was one married couple ( Caio5846 & Caio5884 ) , 2 siblings ( Caio4757 & Caio4758 ) and 6 mother-child pairs ( Caio4634 & Caio4635; Caio4647 & Caio4650; Caio4658 & Caio4659; Caio4743 & Caio4745; Caio4671& Caio4702; Caio4799 & Caio4801 ) . Within each of these 8 family pairs , the viruses showed 100% identical sequences . First , LTR sequences were analyzed to identify the HTLV-1 subtypes present in the village ( Table S1 ) . Identical HTLV-1 LTR sequences were found in known relatives but were also obtained from 10 unrelated individuals with no known epidemiological links ( not related or known sexual partners ) ( Figure 1 ) . Stringent control measures were in place to prevent cross-contamination during sample handling and PCR , and there was a consistent absence of product in all negative controls ( run with each set of PCR reactions ) . In addition there was no clustering of identical sequences by day of sample preparation . The nonidentical sequences exhibited 1 to 3% divergence for the LTR . A phylogenetic tree was constructed with all nonidentical LTR sequences ( Figure 1 - the identical sequences are listed in the figure legend ) . With a single exception the Caio sequences ( marked in red ) clustered with HTLV-1 subgroup D reference strains isolated from pregnant women from Bissau ( capital of Guinea-Bissau ) and other West and North African individuals . One virus LTR sequence , Caio4064 , was very divergent ( 6% ) from the major group of Caio subtype 1a isolates . Caio4064 clustered significantly with an STLV-1 sequence obtained from a wild African green monkey in the Central African Republic [29] ( Tan90 ) and with HTLV-1 subtype 1g isolated from a Cameroonian monkey hunter ( 2656ND ) [7] . The Caio4064 sample was obtained from a 65-year old woman diagnosed with HTLV-1 ( proviral load: 22 copies per 105 cells ) and HIV-2 ( viral load: undetectable ) infection in 1990 . She had not travelled outside of Guinea-Bissau and had not been bitten by a monkey or hunted monkeys ( information obtained in the 1997 survey ) . Her husband was HTLV-1 negative in the most recent survey ( 2007 ) . Further evidence for the presence of subtype 1g was obtained by examining a set of p24 sequences obtained from 36 individuals ( Table S1 , Figure 2 - the identical sequences are listed in the figure legend ) . Although both LTR and p24 sequences were not available for all subjects , there was sufficient overlap between the two sets of sequences to make some important conclusions . Sixteen individuals with no epidemiological links to each other ( not related and not sexual partners ) had identical p24 sequences . A ML-tree was constructed from the non-identical sequences; 32 of the Caio p24 sequences formed a single cluster . Consistent with the LTR phylogeny , Caio4064 p24 clustered significantly with the Tan90 p24 and an additional Caio sample , Caio65552 . Caio65552 was obtained from a 62 year old man who reported having hunted monkeys in the Caio area , but had no recollection of being bitten by a monkey . The subject was diagnosed with HTLV-1 ( proviral load: 2199 per 105 cells ) and HIV-2 ( viral load: undetectable ) in 1997 . He was not married and had no epidemiological link with the patient providing Caio4064 . No samples were available from the mothers of subjects providing Caio4064 or Caio65552 to examine vertical transmission .
This is the first characterization of the sequence diversity of HTLV-1 within a single , endemic community in West Africa with sequences obtained from 72 adults and children . An important feature of this study is that individuals were screened for HTLV-1 in population surveys and were not selected from a clinical setting , thus avoiding selection bias towards diseased individuals . Most of the sequences identified belonged to HTLV-1 Cosmopolitan subtype 1a subgroup D . However , two subjects were found to be infected with a divergent strain of HTLV-1 which clustered significantly with subtype 1g . This is the first time that this novel subtype has been isolated from individuals in West Africa and suggests this subtype has spread into the human population outside of Cameroon , either by human-to-human or interspecies transmission . Remarkable genetic stability in 1aD sequences was observed and in particular , 10 identical LTR sequences and 16 identical p24 sequences were found in people with no identified epidemiological links . These infections differed in time of diagnosis , route of infection ( vertically and horizontally ) and the length of the infection ( a few years vs . >14 years ) . This observation is consistent with the very low evolutionary rate of HTLV-1 ( estimated at 7 . 06×10−7−1 . 38×10−5 substitutions per site per year ) [9] , which is probably explained by clonal expansion of the virus ( rather than viral replication ) [30] . Whether this genetic stability is also influenced by host factors and whether specific viral variants are associated with proviral load and/or disease progression are important questions that remain to be elucidated . With such a limited degree of viral variation observed across a community cohort , combined with the low rate of progression to any HTLV-1 associated disease it is difficult to identify such an association . In Caio , 7 . 1% of HTLV-1 infected individuals were diagnosed with HTLV-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) [31] , but no data regarding Adult T-cell Leukemia are available . HTLV-1 infection in a large cohort study from Caio was found to be associated with increased mortality , especially among younger people and an increased proviral load was associated with increased mortality [32] . HTLV-1aD was the predominant variant in our study population , consistent with other studies from Bissau and Senegal [33] , [34] . Recently , this variant was also described in Europe in patients from West Africa who had emigrated to Portugal [35] . HTLV-1 phylogeny has been examined in a historical context to trace the origin of the virus and movement of people . While slaves who were brought to South America came from West Africa including Guinea-Bissau , surprisingly subgroup D has not been found among African descendants in South America . Subgroup A and C are prevalent among African descendants , for example in Brazil and French Guyana . Subgroup A could have been introduced into Brazil through slaves that came from Southern Africa [14] , [36] . Subgroup C has been found in the Noir Marron , a distinct population of African origin , who settled in French Guyana and the majority of whom originated from the Bight of Benin ( present-day Togo , Benin and a part of Nigeria ) and Gold Coast ( a part of Ivory Coast and Ghana ) [37] . The fact that subgroup D has not been described in these populations may be due to subgroup C being older and/or a later introduction of subgroup D into West Africa ( so that the prevalence was low or absent during the slave trade ) . It could also be that the number of slaves , originally from the countries where subgroup D was prevalent , was ( relatively ) low . To answer this question , additional analysis of both host and viral genetic markers is required . More sampling in West Africa and in groups of African descent in South America could shed light on this issue . The finding of two instances of HTLV-1 subtype 1g , Caio4064 and Caio65552 , from the Caio community is of great interest for several reasons . It indicates that this subtype is present beyond Cameroon and can also be found in a ‘normal’ community member , i . e . not only in people with close contact to monkeys such as hunters . That subtype 1g isolates cluster significantly with Tan90 , the most divergent African STLV-1 strain , gives support to an African origin of HTLV-1 with the oldest known STLV-1 as the most recent common ancestor of Caio4064/Caio65552 . Although the Asian MarB43 is the most divergent STLV-1 , it does not cluster with any known HTLV-1 strains and thus may be only distantly related to STLV subgroups that moved into humans . While most African STLV and HTLV strains cluster together , mainly by geographic location , Asian STLV and HTLV strains only cluster by species , suggesting cross-species transmission has frequently occurred in Africa and not in Asia [11] . No simian strain has been described yet which clusters with HTLV-1a , the most widespread subtype . Whether this is due to relatively infrequent sampling of non-human primates or due to other factors , remains to be resolved [38] . The presence of HTLV-1 subtype 1g infection could be the result of human to human contact ( Caio residents do travel extensively [39] ) . Alternatively , a zoonotic transmission may also explain the presence of this subtype . Hunting of monkeys and butchering for human consumption is commonly practiced [40] . In Caio , captured monkeys are often kept tethered in household compounds for future consumption and HTLV-1 infection has been found to be associated with monkey bites among older persons in the capital Bissau [41] . These social factors , combined with the low evolutionary rate of HTLV-1 [9] are consistent with subtype g virus appearing in Caio due to a transmission from a local monkey . Indeed , the 62 year old Caio man yielding Caio65552 reported hunting monkeys in the area; however , the woman from whom the Caio4064 isolate was obtained , reported that she did not hunt monkeys and was not bitten by one suggesting that her infection came about by another route . Iatrogenic spread should also be considered , as both individuals are older and co-infected with HIV-2 [42] , [43] . HTLV-1 infection is considered to be largely an infection of older women [44] . However , 6 young men ( 16 years of age ) were found to have become infected with HTLV-1 between 1997–2007 in the Caio community [15] , which led us to consider transmission during traditional local , non-sterile , male circumcision ceremonies . These take place every ten years in a sacred part of the Caio forest . This initiation ritual is exclusively for men and lasts three months . Three men included in the current analysis were known to have participated in the same circumcision ceremony . However , the LTR sequences from the viruses found in these 3 men ( yielding Caio5187 & Caio5801 & Caio7580 ) were not identical and no more closely related to each other than to sequences from the wider Caio population . ( i . e . they did not cluster separately from the larger subgroup D cluster ( Figure 1 ) ) . Both of these features would have been expected if the virus in these three subjects had been transmitted by infected blood in such a ceremony [45] , [46] . Thus , although a traditional circumcision ceremony has the potential to spread blood borne viruses , the phylogenetic data clearly do not support the idea that this ceremony was the source of HTLV-1 infection of these young men . Epidemiological surveillance and further sequencing are recommended to identify ( novel ) HTLV-1 subtypes [47] , [48] and to investigate the spread of subtype 1g . It would be of particular interest to investigate to what extent new introductions of HTLV subtypes ( and potentially other retroviruses ) [7] , [49] , [50] occur and whether non-sexual , non-vertical transmission takes place in this community . Further testing of indeterminate HTLV-1/2 ELISA samples found in this community [15] could indicate for example whether HTLV-2 is also present in Caio [51] . This may lead to a better understanding of the spread of other zoonotic events , of which the HIV pandemic is a major example [40] , and can have important public health implications . | Human T-Lymphotropic Virus type 1 ( HTLV-1 ) affects millions of people worldwide . It is very similar to Simian T-Lymphotropic Virus , a virus that circulates in monkeys . HTLV-1 causes a lethal form of leukemia ( Adult T-cell Leukemia ) and a debilitating neurological syndrome ( HTLV-associated myelopathy/tropical spastic paraparesis ) in approximately 5% of infected people . Based on sequence variation , HTLV-1 can be divided into 7 subtypes ( 1a–1g ) with the Cosmopolitan subtype 1a further subdivided into subgroups ( A–E ) . We examined HTLV-1 diversity in a rural area in Guinea-Bissau , a country in West Africa with a high HTLV-1 prevalence ( 5% ) . We found that most viruses belong to the Cosmopolitan subtype 1a , subgroup D , but 2 viruses belonged to subtype 1g . This subtype had thus far only been found in monkey hunters in Cameroon , who were probably recently infected by monkeys . Our findings indicate that this subtype has spread beyond Central Africa . An important , unresolved question is whether persons with this subtype were infected by monkeys or through human-to-human transmission . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"zoonoses",
"global",
"health",
"neglected",
"tropical",
"diseases",
"viral",
"diseases"
] | 2012 | Molecular Epidemiology of Endemic Human T-Lymphotropic Virus Type 1 in a Rural Community in Guinea-Bissau |
Hepatitis C virus ( HCV ) is a major cause of chronic liver disease affecting around 130 million people worldwide . While great progress has been made to define the principle steps of the viral life cycle , detailed knowledge how HCV interacts with its host cells is still limited . To overcome this limitation we conducted a comprehensive whole-virus RNA interference-based screen and identified 40 host dependency and 16 host restriction factors involved in HCV entry/replication or assembly/release . Of these factors , heterogeneous nuclear ribonucleoprotein K ( HNRNPK ) was found to suppress HCV particle production without affecting viral RNA replication . This suppression of virus production was specific to HCV , independent from assembly competence and genotype , and not found with the related Dengue virus . By using a knock-down rescue approach we identified the domains within HNRNPK required for suppression of HCV particle production . Importantly , HNRNPK was found to interact specifically with HCV RNA and this interaction was impaired by mutations that also reduced the ability to suppress HCV particle production . Finally , we found that in HCV-infected cells , subcellular distribution of HNRNPK was altered; the protein was recruited to sites in close proximity of lipid droplets and colocalized with core protein as well as HCV plus-strand RNA , which was not the case with HNRNPK variants unable to suppress HCV virion formation . These results suggest that HNRNPK might determine efficiency of HCV particle production by limiting the availability of viral RNA for incorporation into virions . This study adds a new function to HNRNPK that acts as central hub in the replication cycle of multiple other viruses .
Hepatitis C virus ( HCV ) is a major cause of liver disease affecting ∼130 million people worldwide [1] . Chronic HCV infection can cause steatosis , fibrosis , cirrhosis and hepatocellular carcinoma ( HCC ) and is a main indication for liver transplantation [2] . HCV is an enveloped virus that belongs to the Hepacivirus genus within the Flaviviridae family . The positive sense single-strand RNA genome encodes for a polyprotein that is cleaved by cellular and viral proteases into 10 viral proteins: three structural proteins ( core , envelope proteins E1 and E2 ) , the p7 polypeptide and six non-structural proteins ( NS2 , NS3 , NS4A , NS4B , NS5A , NS5B ) . The structural proteins are main constituents of the virus particle whereas most of the non-structural proteins are required for RNA replication . Assembly of virus particles is tightly linked to cytosolic lipid droplets ( LDs ) , where core accumulates , and components of the low density lipoprotein ( LDL ) pathway , most notably apolipoprotein E ( reviewed in [3] ) . As for all viruses , the HCV life cycle strongly depends on host cell factors promoting or restricting its replication . Thus , a tight interplay between viral and cellular proteins can be assumed to regulate virus replication and survival of the host cell . Numerous host cell factors involved in the HCV life cycle have been reported so far ( reviewed in [4] ) , often based on RNA interference screens with genome-wide or more selective siRNA libraries [5]–[17] . However , the overlap between identified cellular factors is marginal , likely resulting from the use of different siRNA libraries and experimental conditions , but also from several technical limitations that are inherent to high-throughput siRNA screens [18] . Importantly , in most cases the specific roles of these factors for the HCV life cycle have not been clarified . Heterogeneous nuclear ribonucleoprotein K ( HNRNPK ) is a polycytidine-binding protein originally identified as a component of the heterogeneous nuclear ribonucleoprotein complex [19] . HNRNPK is able to interact with RNA , DNA and multiple proteins and is involved in various cellular processes including chromatin remodeling , regulation of transcription , splicing and RNA translation [20] and was shown to assemble on DNA either as a transcriptional activator or as repressor [21] , [22] . In addition , mRNA stability [23] as well as alternative splicing [24] can be affected by binding of HNRNPK to mRNA structures . The presence of multiple phosphorylation sites in HNRNPK suggests that this protein is also involved in several cellular signaling pathways [25] . Interestingly , HNRNPK has been reported to be involved in the life cycle of different viruses by either direct interaction with viral proteins [26] , [27] or by affecting signal transduction and gene expression in a more indirect manner [28] , [29] . However , the effect of HNRNPK as a pro- or anti-viral factor strongly differs between the various virus systems . In the present study we conducted a siRNA-based screen and identified several cellular host factors affecting the HCV life cycle . This included HNRNPK that was found to selectively suppress production of infectious HCV particles . Our results suggest that HNRNPK might regulate the availability of viral RNA for the formation of infectious HCV particles .
In search for host cell factors involved in the HCV life cycle , a two-step high-throughput siRNA screen was conducted ( Fig . 1A ) . We used the extended druggable genome siRNA library covering a total of 9 , 102 human genes with known or predicted functions and suitable as potential drug targets . The first part of the screen covered HCV entry and replication , whereas the second part covered production of infectious extracellular virus particles ( assembly and release ) . For each gene , three different siRNAs were tested individually by using solid-phase reverse transfection . Huh7 . 5 FLuc cells , stably expressing the Firefly luciferase , were infected with the Renilla luciferase reporter virus JcR2a [9] that was derived from the highly assembly competent HCV chimera Jc1 [30] . Seventy-two hours after infection , cells were harvested to measure virus replication by Renilla luciferase assay , thus determining the impact of knock-down on HCV entry and replication . Firefly luciferase activity was quantified to exclude false-positive hits caused by knock-down-mediated alteration of cell growth and viability . To monitor the impact of knock-down on production of extracellular virus , culture supernatants of siRNA-transfected cells were used to inoculate naïve Huh7 . 5 FLuc cells and 72 h later , HCV replication was determined by Renilla luciferase assay; Firefly luciferase activity was also determined to exclude cell growth effects caused by the inoculum . The primary siRNA screen was performed in three replicates and identified in total 78 host dependency factors ( HDFs ) and 29 host restriction factors ( HRFs ) ( S1A Fig . ; S1A Table ) . Gene Ontology ( GO ) ( Gene Ontology Consortium , 2010 ) of candidate host factors revealed the biological processes implicated in the HCV life cycle: transport , transcription and transcription regulation ( S1B Fig . and S1C , D Table ) . Given the technical limitations that are inherent to such high-content siRNA-based screens , we attempted to increase comprehensiveness and reliability of our screen by conducting a meta-analysis that was based on the following data sets ( Fig . 1B and S1B Table ) : ( i ) our own and published HCV siRNA screens with genome-wide or selected siRNA libraries [5]–[10] , [12]–[16] , [31]–[37]; ( ii ) a genome-wide high-throughput yeast-two hybrid protein interaction study [38]; ( iii ) a comparative analysis of the proteome of crude HCV replication complexes ( CRCs ) conducted by us and others [39]; ( iv ) a comparative transcriptome analysis between low- and highly permissive Huh-7 cells [40] , [41] , because the degree of permissiveness might be due to expression levels of dependency and restriction factors; ( v ) a comparison of a mouse and a human hepatocytic cell line ( Hep56 . 1D and HuH6 , respectively ) with or without a subgenomic HCV replicon [42] , [43] , assuming that differences in expression levels of dependency and restriction factors might be more accentuated in cells that are much less permissive for HCV as compared to Huh-7 . In this way , a total of 204 genes ( 108 from our primary screen and 96 deduced from the meta-analysis ) were tested in an extended validation screen for their impact on the HCV life cycle . For the validation screen , we used 4 different siRNAs per gene purchased from a different supplier and the same set-up applied for the primary screen ( Fig . 1A ) . Hits were defined by a z-score of ≤−2 for HDFs or ≥+2 for HRFs , achieved with at least two different siRNAs per gene in 4 independent repetitions . Fig . 1C summarizes mean z-scores of all individual siRNAs , combining the four replicates , sorted from the lowest HDFs to the highest HRFs . In total , we were able to validate 40 HDFs and 16 HRFs affecting the HCV life cycle ( S1A Table ) . This corresponds to 21% of the hits identified in the primary screen and 28% of the candidates selected by our meta-analysis . To reveal cellular processes critically involved in the HCV life cycle , we performed an integrative computational analysis ( S2 Fig . ; S1C , D , F Table ) . Consistent with earlier reports , we identified enrichments of , amongst others , intracellular protein transport pathways such as the COP-I system [6] , factors involved in the epidermal growth factor receptor signaling pathway [44] , signal recognition particle receptor-dependent transport and signal peptide processing ( reviewed in [45] ) or the low-density-lipoprotein-pathway , consistent with the tight link of HCV assembly with intracellular lipid synthesis and storage systems [46] . One of the most striking phenotypes was obtained upon knock-down of heterogeneous nuclear ribonucleoprotein K ( HNRNPK ) , which had an opposing effect on both analyzed steps of the HCV life cycle . While knock-down of HNRNPK expression reduced virus entry/replication , albeit to a very moderate extent , virus assembly/release was profoundly enhanced in case of knock-down with 5 of the 7 selected siRNAs ( S1A Table ) . Moreover , HNRNPK was not only found in our meta-analysis , but also reported in earlier studies as HCV-modulating host cell factor [5] , [17] , [38] , arguing that HNRNPK might play an important role for the viral life cycle ( S1B Table ) . In fact , HNRNPK was reported to interact with HCV core [47] , NS3 [38] and the internal ribosome entry site ( IRES ) residing in the 5′ non-translated region of the viral genome [48] . Moreover , HNRNPK has also been reported to be involved in the life cycle of many other viruses such as Chikungunya virus [49] , Influenza A virus [50] or Sindbis virus [26] , interacting with viral proteins or nucleic acid structures ( Fig . 2A ) . Thus , given the consistency and robustness of results we obtained for HNRNPK and HCV , the opposing phenotype on entry/replication versus assembly/release and the unknown mechanism by which this host cell factor affects the HCV life cycle , we selected HNRNPK for further characterization . In the first set of experiments we determined whether HNRNPK expression would be affected by HCV infection . However , neither at the level of mRNA , nor with respect to HNRNPK abundance we observed alterations in HCV-infected cells as compared to control cells ( S3A and B Fig . , respectively ) . With the aim to separate the effect of HNRNPK on HCV assembly/release from its contribution to viral RNA replication , we co-electroporated HNRNPK-specific siRNAs , together with the JcR2a RNA genome , into Huh7-Lunet cells . These cells are highly permissive for viral replication as well as assembly and release , but support HCV entry only poorly due to low amounts of CD81 [51] . Thus , an apparent increase of viral replication due to virus spread could be excluded with these cells . Western Blot revealed efficient knock-down by each of the four siRNAs ( Fig . 2B ) . In contrast , non-targeting control siRNA ( siContr . ) as well as siRNAs targeting either the luciferase sequence within the genome of JcR2a or the assembly factor apolipoprotein E ( ApoE ) did not affect HNRNPK abundance . None of these siRNAs exerted cytotoxicity ( Fig . 2B ) . As shown in Fig . 2C , depletion of HNRNPK had no significant effect on viral replication whereas the positive control targeting the Renilla luciferase sequence in the viral genome ( siRLuc ) completely abolished replication . Importantly , amounts of infectious extracellular virus produced by HNRNPK-depleted cells revealed a ∼4-fold increase as compared to siContr . -transfected cells , corroborating the suppression of HCV production by HNRNPK ( Fig . 2D ) . Note that the conditions used here were optimized for HNRNPK- , but not ApoE-specific knock-down for which an shorter incubation period leads to higher silencing efficiency [52] . Therefore , reductions of virus titers achieved in the latter case were rather moderate . Although HNRNPK knock-down exerted a predominant effect on HCV particle production , in the primary screen it appeared as a dependency factor and reduced HCV entry/replication . Since an effect on RNA replication had been excluded , we determined the possible role of HNRNPK for virus entry by using HCV pseudoparticles ( HCVpp ) ( Fig . 2E ) . Therefore , we silenced HNRNPK expression for 48 h by using a mix of the 4 different siRNAs and infected these cells with a YFP gene-transducing HCVpp using Con1- ( gt1b ) derived envelope glycoproteins . After 72 h the number of infected cells was determined by flow cytometry . Consistent with the primary screen data we found that HNRNPK knock-down caused a slight , but statistically significant reduction of the number of infected cells as compared to cells transfected with the non-targeting control siRNA ( siContr . ) . Thus , HNRNPK appears to play a role in the early steps of the HCV replication cycle . However , given the more striking effect of HNRNPK knock-down on the production of infectious HCV particles , we focused our further analysis on this aspect . To corroborate our results with respect to the role of HNRNPK for HCV assembly/release , we conducted an immunofluorescence analysis of JcR2a-infected cells ( Fig . 3A ) . HNRNPK depletion did not affect HCV replication ( as determined by abundance of the Renilla luciferase protein encoded in the reporter virus genome ) at the single cell level ( Fig . 3A , left panel ) . In contrast , amount of infectious virus contained in the supernatant of HNRNPK-silenced cells was strongly increased , as reflected by a higher number of infected cells ( Fig . 3A , right panel ) . To exclude an effect of HNRNPK knock-down on Renilla luciferase and to determine whether HNRNPK enhanced virus production or particle infectivity , we co-transfected Huh7 . 5 cells with siRNA and a reporter-free Jc1 genome and determined intra- and extracellular core protein amounts along with infectious virus titers ( Fig . 3B ) . While amounts of core protein and infectious virus particles in transfected cells were unaltered by HNRNPK depletion , amounts of core and infectious virus particles in culture supernatants were clearly elevated ( Fig . 3B ) . However , total amount of core did not change , because intracellular core level was ∼10-fold higher and therefore , changes in amounts of secreted core protein caused only minor changes in total core protein abundance . Although there was a trend towards slightly increased specific infectivity of virus particles released from HNRNPK knock-down cells , the difference to control cells was not statistically significant suggesting that HNRNPK primarily contributes to virus production ( Fig . 3C ) . The fact that extracellular infectivity titers were elevated without concomitant reduction of intracellular infectivity titers argued against a pure virus release phenotype of HNRNPK knock-down . Moreover , intra- and extracelluar amounts of ApoE were not altered in HNRNPK knock-down cells ( Fig . 3D ) , excluding a general effect of HNRNPK depletion on the secretory pathway . Finally , we found that HNRNPK depletion elevated virus production independent from the analyzed genotype . By using chimeric genomes encoding for gt1a ( H77 ) or gt1b ( Con1 ) structural proteins we observed an increase of HCV assembly/release that was well comparable to the one detected with the gt2a chimera JcR2a ( S4A – C Fig . ) . Moreover , HNRNPK-dependent suppression of virus production was independent from the assembly competence of the used virus genome; elevated virus production upon HNRNPK knock-down was also observed with the original JFH-1 isolate that supports assembly only poorly as compared to the highly assembly-competent variant Jc1 [30] ( S4D Fig . ) . Given the close relationship between HCV and Dengue virus ( DENV ) that both belong to the family Flaviviridae and the role of HNRNPK in the life cycle of several other viruses ( Fig . 2A ) , we determined whether HNRNPK also affects production of infectious DENV particles . To this end , we utilized a DENV Renilla luciferase reporter virus that was tested in parallel to the HCV reporter virus JcR2a . In agreement with results described above , we detected profound enhancement of HCV assembly/release with no overt effect on RNA replication ( Fig . 3E , left panel ) . In contrast , HNRNPK knock-down had no impact on production of infectious DENV ( Fig . 3E , right panel ) . In summary , these results suggest that HNRNPK enhances production of infectious HCV particles , rather than particle release , and this effect appears to be specific for HCV . To exclude undesired off-target effects caused by the used siRNAs and to set up an assay allowing mapping studies of HNRNPK domains of relevance for HCV assembly , we tried to overexpress HNRNPK by lentiviral transduction . In the course of these experiments we noted that abundance of endogenous HNRNPK dropped upon expression of ectopic HNRNPK . As shown in Fig . 4A , cells expressing ectopically either wild type or HA-tagged HNRNPK did not contain higher amounts of total HNRNPK than control cells only expressing endogenous HNRNPK , arguing that expression level of this protein is tightly regulated . Due to this tight regulation of HNRNPK expression , we were not able to explore whether overexpression would result in a stronger suppression of HCV virus production . To overcome this limitation , we aimed to restore HNRNPK-mediated suppression of HCV particle production in knock-down cells by ectopic expression of a siRNA-resistant variant . Huh7 . 5 cell pools stably expressing either wild type or HA-tagged HNRNPK ( Huh7 . 5-wt-HNRNPK and Huh7 . 5-HA-HNRNPK , respectively ) , in addition to residual amounts of endogenous HNRNPK , were established . The ectopically expressed HNRNPK genes lacked the authentic 3′ NTR and thus , were resistant to the siRNA targeting the 3′ end of the endogenous HNRNPK mRNA ( si-3′NTR ) , but sensitive to siRNAs #1 - #4 targeting sequences in the coding region ( Fig . 4B ) . To demonstrate comparable knock-down efficiency of all 5 siRNAs , naïve Huh7 . 5 cells were transfected with each siRNA individually and infected with JcR2a ( Fig . 4C ) . Virus amounts contained in culture supernatants were determined by infection of naïve Huh7 . 5 cells , Renilla luciferase activity was quantified 72 h later and normalized to cell viability and the non-targeting control siRNA ( siContr . ) . A siRNA with a non-functional HNRNPK recognition site served as additional control ( siHNRNPKcontr . ) . All HNRNPK-specific siRNAs reduced expression of the target gene , concomitant with an increase of amounts of infectious extracellular HCV ( Fig . 4C ) . Importantly , HNRNPK silencing with siRNAs #1 - #4 depleted both endogenous and ectopic HNRNPK , whereas the siRNA targeting the 3′ NTR depleted only endogenous HNRNPK ( Fig . 4D , E ) . Although the absolute values of assembly/release enhancement caused by HNRNPK knock-down differed between individual experiments and cell pools , the relative capacity of ectopically expressed HNRNPK to suppress HCV particle production ( termed R3 value ) was well comparable between cells expressing different siRNA-resistant genes ( green arrow in Fig . 4D , E ) . These results unequivocally confirmed specificity of the knock-down phenotype and they showed that HA-tagged HNRNPK is fully functional ( Fig . 4E ) . Taking advantage of this knock-down/rescue assay , we next mapped HNRNPK domains that are of relevance for HCV particle production . HNRNPK is a multifunctional protein composed of different domains [20] ( Fig . 5A ) including a predicted nuclear localization signal ( NLS ) and a nuclear shuttling domain ( KNS ) , KH1 and KH2 domains capable of binding single strand RNA , a KH3 domain binding to DNA , a KI region responsible for protein-protein interaction and an interaction region with c-terminal kinase ( cKBR ) [20] . For mapping studies we generated a series of HNRNPK mutants lacking each of these domains individually ( Fig . 5B ) . These variants were tested for their capacity to restore suppression of HCV particle production in cells with knock-down of endogenous HNRNPK . As shown in Fig . 5C , HNRNPK mutants lacking the NLS , the cKBR domain or the KNS were only moderately affected in their capability to suppress HCV particle production . For these mutants , R3 values ranged between 89% and 75% ( Fig . 5C and D; for complete data set see S5 Fig . ) , showing that the deleted domains are not essential for HNRNPK-mediated suppression of HCV particle production . While deletion of the DNA binding domain ( KH3 ) resulted in an intermediate phenotype ( R3 = 59%; Fig . 5C and S5E Fig . ) , removal of either one of the two RNA binding domains ( KH1 and KH2 ) or the protein binding domain ( KI ) rendered the protein virtually inactive ( Fig . 5C , E and S5D , F Fig . ) . These results argue for an essential role of the RNA-binding KH-domains and the protein-protein interaction domain in limiting HCV particle production , whereas the other domains of HNRNPK appear to be dispensable . HNRNPK has been reported to be involved in the replication of multiple viruses and in some cases , regulatory functions of HNRNPK are based on its interaction with viral proteins ( Fig . 2A ) . With respect to HCV , an interaction between HNRNPK and full length or truncated core as well as NS3 has been described [38] , [47] . However , these interaction studies have been conducted with artificial expression systems , but the relevance in more authentic infection-based systems has not been addressed . Therefore , we investigated HNRNPK interaction with core and NS3 using either wild type or HA-tagged HNRNPK-expressing cells containing a subgenomic replicon or a full length HCV genome ( S6 Fig . ) . HA-specific immunoprecipitation assays confirmed that HA-HNRNPK coprecipitated with core and NS3 , although efficiency appeared rather low ( ∼1% of the respective intracellular HCV protein ) . Nevertheless , specificity of the pull-down was confirmed by lack of HNRNPK coprecipitation with NS5A or GAPDH and the absence of core and NS3 in immunoprecipitations with non-tagged HNRNPK ( S6 Fig . ) . Moreover , we found that mutants unable to suppress HCV assembly were also impaired in interaction with core protein as well as NS3 whereas a HNRNPK mutant capable to suppress virus particle production still interacted with these two viral proteins with efficiency comparable to the HA-tagged wild type ( S7 Fig . ) . To determine whether HNRNPK also interacts with HCV RNA , we conducted an analogous experiment , but immunocomplexes were analyzed for captured viral RNA using RT-qPCR , whereas efficiency of pull-down was determined by Western blot using HNRNPK- or HA-specific antibodies ( Fig . 6 ) . HNRNPK deletion mutants ΔKH1 , ΔKH2 and ΔKHI were included , because of their inability to suppress HCV particle production . Cells expressing the non-tagged ( wild type ) HNRNPK served as technical control to determine background binding of HCV RNA to HA-specific beads . Finally , we included cells transfected with a fully functional DENV genome to determine whether observed phenotypes are specific to HCV . All HNRNPK constructs were well expressed and HA-tagged proteins were efficiently captured with high specificity ( Fig . 6A ) . Analysis of immunocomplexes with an HCV- or DENV-specific RT-qPCR revealed specific co-capture of the subgenomic and the genomic HCV RNA with HA-HNRNPK whereas no such interaction was detected in case of DENV ( Fig . 6B ) . Pull-down efficiency ranged between 10% and 25% of detected intracellular HCV RNA . Moreover , HNRNPK deletion mutants unable to restrict HCV particle production did not interact with viral RNA . Importantly , loss of interaction with HCV RNA was also found with mutant ΔKI lacking the protein-protein interaction domain and unable to suppress HCV particle production ( Fig . 6 and 5 , respectively ) . Taken together , the correlation between suppression of HCV production by HNRNPK and its interaction with viral RNA argues for a link between these two processes . The observation that HNRNPK mutants unable to interact with core and NS3 only weakly coprecipitated HCV RNA argues for an indirect interaction . Inhibition of virus production by HNRNPK was specific to HCV and not found with the related DENV for which no interaction of HNRNPK with viral RNA could be found ( Fig . 6 ) . HCV assembly occurs in close proximity of LDs [53] where viral proteins , most notably core and NS5A accumulate [54] . Given the role of HNRNPK in HCV particle production and the interaction of this host cell factor with viral RNA we first determined whether subcellular localization of HNRNPK would be affected by HCV infection . To this end we purified LDs from HCV-infected cells by floatation gradient . Although we observed a strong enrichment of core protein in case of LDs isolated from HCV-infected cells , as well as copurification of the core-interacting protein DDX3 that served as positive control [55] , HNRNPK was not visible in LD fractions ( S8 Fig . ) . While this result suggested that HNRNPK is not recruited to the surface of LDs , we wondered whether subcellular distribution of HNRNPK might be affected by HCV infection . To address this question we conducted immunofluorescence studies comparing HCV-infected cells with control cells . Since abundance of HNRNPK in the cytoplasm is very low , we took advantage of the HA-tagged HNRNPK mutant lacking the predicted NLS . This mutant suppressed HCV particle production as efficiently as wild type HNRNPK and thus , was fully functional with respect to the studied phenotype ( Fig . 5C , D ) . Although removal of the NLS increased abundance of HNRNPK in the cytoplasm , a substantial proportion of this protein still resided in the nucleus ( Fig . 7A ) . Nevertheless , we found that in cells containing HCV , as determined by fluorescence in situ hybridization ( FISH ) of viral RNA ( see below ) , subcellular localization of cytoplasmic HNRNPK was altered and a fraction of HA-HNRNPKΔNLS was relocalized to ring-like structures that were not observed in the absence of HCV ( Fig . 7A ) . While this ring-like pattern is indicative of LDs , in the light of our subcellular fractionation results ( S8 Fig . ) we assumed that this staining pattern likely corresponds to ER membranes surrounding LDs . Indeed , by determining the colocalization of LDs with the ER marker protein disulphide isomerase ( PDI ) , HA-HNRNPKΔNLS and core were observed at LDs tightly surrounded by PDI-positive ER ( Fig . 7B ) . Since HNRNPK interacts with HCV RNA as determined by pull-down experiments , we determined subcellular localization of viral RNA by using fluorescence in situ hybridization ( FISH ) and colocalization of HCV RNA with HNRNPK as well as core . As shown in Fig . 7C , positive and negative strand HCV RNA as well as HNRNPK could be detected with high sensitivity . Specificity of HCV RNA detection of either polarity was corroborated by the absence of signal in cells that had been transfected with a replication-incompetent viral genome and analysis three days after transfection ( S9 Fig . ) . By using this method we observed colocalization of core , HA-HNRNPKΔNLS and viral RNA of positive and negative polarity in ring-like structures , corresponding most likely to sites in close proximity of LDs where HCV assembly takes place ( Fig . 7C and S10 Fig . , respectively ) . Moreover , this colocalization was not found in case of HA-HNRNPK variants ΔKH1 , ΔKH2 and ΔKI unable to suppress HCV assembly ( S11 Fig . ) . Although cytoplasmic abundance of these variants was lower as compared to HA-HNRNPKΔNLS these results suggest that HNRNPK proteins capable to suppress HCV assembly are relocalized to putative HCV assembly sites . This localization is consistent with a role of HNRNPK in regulating virus particle production . In summary , we conclude that HNRNPK is a host cell factor determining efficiency of HCV particle production .
Several high-content screens have reported putative HDFs and HRFs promoting or restricting the HCV life cycle . Given the poor overlap of these screens , in the present study we combined a high-content RNA interference-based screen with an extensive meta-analysis , finally leading to the identification of 56 host cell factors affecting either early ( entry and replication ) or late steps ( assembly and release ) of the HCV life cycle . Bioinformatic analysis revealed significant enrichment of hits in distinct host cell pathways . These include the COPI system required for retrograde vesicular transport as well as LD homeostasis [56] , or the SRP-dependent protein targeting machinery required for polyprotein processing ( reviewed in reference [45] ) . Enrichments were also found for the epidermal growth factor receptor signaling pathway that plays a role for HCV entry ( reviewed in reference [57] ) , and for LDL as well as plasma lipoprotein particle clearance pathways that are important for HCV assembly and release ( reviewed in [3] ) . This over-representation of distinct cellular processes in the interaction network emphasizes their importance for the HCV life cycle . We focused our analysis on HNRNPK that we identified as HRF limiting production of infectious HCV particles . In extension of earlier reports [48] , [58] , we demonstrate HNRNPK interaction with HCV RNA in the context of fully functional genomes . This interaction might be due to direct binding of HNRNPK to viral RNA or more indirectly , via interaction with the RNA binding proteins core or NS3 . In support of the latter assumption we found that HNRNPK mutants impaired in interaction with HCV RNA also were impaired in interaction with these two viral proteins . In any case , the finding that HNRNPK mutants unable to pull down HCV RNA also have lost their capability to suppress virus production provides strong evidence for a direct mechanistic link . The relocalization of HNRNPK in HCV-infected cells to the vicinity of LDs , together with core and viral RNA , is in line with this proposed role in the HCV life cycle . One plausible model explaining the underlying mechanism is that HNRNPK , via ( direct or indirect ) binding to HCV RNA , might regulate RNA availability for packaging into virions . In this model , HNRNPK would bind to viral genomes to feed them into a new cycle of RNA translation/replication , thus restricting viral RNA from incorporation into virus particles . Consequently , silencing of HNRNPK expression would increase the amount of viral RNA genomes available for packaging into nucleocapsids . In return , less genomes should enter a new cycle of RNA translation/replication . Indeed , by using cell lines containing a stable genotype 1 HCV replicon , previous studies reported a reduction of viral RNA replication upon knock-down of HNRNPK [59] , [60] , which is in line with this model . Although we also identified HNRNPK as a HDF in our entry/replication screen , in subsequent validation experiments , the impact of knock-down on HCV RNA replication was not statistically significant . Instead , we found that HNRNPK also plays a role in HCV entry . Moreover , HNRNPK has not been detected in affinity purified replication complexes [61] ( D . Paul and R . B . , unpublished ) , supporting the assumption that HNRNPK is not directly involved in HCV RNA replication . At first glance , this conclusion appears to contradict our model . However , we note that only a minor fraction of intracellular HCV RNA was coprecipitated with HNRNPK , suggesting that only a small proportion might be involved in virus assembly , which is consistent with the overall low assembly efficiency of HCV . Thus , releasing a HNRNPK-mediated block of assembly would have only a minor effect on the large pool of HCV RNA used for translation or replication and therefore , little effect on viral protein synthesis and RNA amplification , but well measurable effects on virus production . Knock-down of HNRNPK primarily increased amounts of infectious HCV particles in culture supernatants as deduced from elevated levels of core protein and infectivity . The fact that intracellular amounts of core protein and infectivity were not affected argues that HNRNPK is not directly involved in virus release . Instead , we assume that HNRNPK determines assembly efficiency of infectious HCV particles . In that case , one might wonder why enhanced assembly did not lead to elevated amounts of infectious intracellular HCV particles . It has been reported that only a subfraction of intracellular particles are released whereas the rest is targeted for degradation [62] . Moreover , the same study showed that infectious HCV particles are rapidly released . Thus , accelerated assembly does not necessarily result in elevated amounts of intracellular infectious virus particles or core protein , consistent with our finding . Accelerated , but otherwise normal , assembly would also explain why neither specific infectivity of HCV particles , nor their biophysical properties as determined by rate zonal centrifugation were affected by HNRNPK silencing ( M . P . and R . B . unpublished ) . Apart from HCV , HNRNPK plays an important role in the life cycle of many other viruses where it acts either as dependency or restriction factor . For instance , HNRNPK was reported to stimulate transcription of the hepatitis B virus surface gene by binding to the respective enhancer in the viral genome [29] . This stimulation is abolished by APOBEC3 ( apolipoprotein B mRNA editing enzyme catalytic polypeptide 3 ) that binds to and sequesters HNRNPK . In case of the human immunodeficiency virus , HNRNPK interacts with the accessory protein nef and activates signaling pathways that ultimately enhance transcription of the provirus [28] . Moreover , HNRNPK was reported to bind to a distinct stem-loop structure of the HIV-1 RNA genome , negatively affecting splicing [63] . In case of Sindbis virus , HNRNPK interacts with the subgenomic RNA [26] , whereas in Enterovirus 71 , it binds to a stem-loop structure in the 5′UTR [64] . Similar to what we observed for HCV , in these viruses the KI region as well as the RNA and DNA binding domains of HNRNPK are considered to be crucial for interaction with the viral RNAs . However , in all of these cases HNRNPK affects ( directly of indirectly ) viral replication whereas in the present study we describe a novel role of HNRNPK , i . e . regulating virus assembly . Although the detailed mechanistic aspects remain to be clarified , these examples illustrate that HNRNPK is usurped by evolutionary very distinct virus classes for various steps in their life cycles and presumably via different mechanisms . Our finding that HNRNPK is utilized by HCV to regulate specifically virus production adds a new facet to the complex regulation of virus - host cell interactions .
For infection experiments the highly permissive cell lines Huh7 . 5 [65] , Huh7-Lunet/CD81 [51] , Huh7hp [40] or Huh7 . 5 FLuc were used , all derived from naïve Huh-7 cells [66] . Huh7 . 5 FLuc are derived from Huh7 . 5 cells by lentiviral transduction of the gene encoding for the Firefly luciferase ( FLuc ) by using a transduction approach described elsewhere [51] and culturing in the presence of 900 µg/ml G-418 . Huh7 . 5 cells stably expressing wild type HA-tagged or untagged HNRNPK ( Huh7 . 5 wt-HNRNPK , Huh7 . 5 HA-HNRNPK ) or the mutant forms of HNRNPK ( Huh7 . 5 HNRNPKΔNLS , Huh7 . 5 HNRNPKΔKNS , Huh7 . 5 HNRNPKΔctK , Huh7 . 5 HNRNPKΔKI , Huh7 . 5 HNRNPKΔKH1 , Huh7 . 5 HNRNPKΔKH2 and Huh7 . 5 HNRNPKΔKH3 ) were generated by lentiviral transduction of the gene encoding for the HNRNPK variants by using the same transduction approach and culturing in the presence of 2 µg/ml puromycin . Cells were grown in Dulbecco's modified minimal essential medium ( DMEM; Life Technologies , Frankfurt , Germany ) supplemented with 2 mM L-glutamine , non-essential amino acids , 100 U/ml penicillin , 100 µg/ml streptomycin , and 10% fetal calf serum ( complete DMEM ) . The Renilla luciferase reporter virus ( JcR2a; derived from plasmid pFKI389Core-3′-Jc1 ) encodes a Renilla luciferase that is fused N-terminally with the 16 N-terminal amino acid residues of the core protein and C-terminally with the foot-and-mouth disease virus ( FMDV ) 2A peptide [9] . Jc1 [30] and JcR2a viruses were produced by transient transfection of Huh7 . 5 cells with in vitro transcribed RNA as described previously [67] . The DENV genome used in this study contains a Renilla luciferase reporter gene and has been described elsewhere [68] . In vitro transcripts were generated by using 10 µg plasmid DNA that had been linearized by 1h-digestion with MluI . DNA was extracted with phenol and chloroform and , after precipitation with isopropanol , dissolved in RNase-free water . In vitro transcription reaction mixtures ( total volume 100 µl ) contained 80 mM HEPES ( pH 7 . 5 ) , 12 mM MgCl2 , 2 mM spermidine , 40 mM dithiothreitol ( DTT ) , 3 . 125 mM of each nucleoside triphosphate , 1 U/µl RNasin ( Promega , Mannheim , Germany ) , 0 . 1 µg/µl of plasmid DNA , and 0 . 6 U/µl T7 RNA polymerase ( Promega , Mannheim , Germany ) . After 2 h incubation at 37°C , 0 . 3 U/µl reaction mixture of T7 RNA polymerase was added and the reaction mixture was incubated for 2 h at 37°C . Transcription was terminated by adding 1 . 2 U of RNase-free DNase ( Promega , Mannheim , Germany ) per µg plasmid DNA and 30 min incubation at 37°C . RNA was extracted with acidic phenol and chloroform , precipitated with isopropanol at room temperature and dissolved in RNase-free water . For virus production single-cell suspensions of Huh7 . 5 cells were prepared by trypsinization , washing with phosphate-buffered saline ( PBS ) and resuspension at a concentration of 1 . 5×107 cells/ml in Cytomix [67] supplemented with 2 mM ATP and 5 mM glutathione . Ten µg in vitro transcripts were mixed with 400 µl cell suspension and transfected by electroporation using a GenePulser system ( Bio-Rad , Hercules , CA ) and cuvettes with a gap width of 0 . 4 cm ( Bio-Rad ) at 975 µF and 270 V . Cells were immediately diluted in complete DMEM and seeded . Virus-containing supernatants were collected , titrated and used for infection experiments . The siRNA library used for the primary siRNA screen ( Ambion Silencer Extended druggable genome library V3 ) contains a total of 27 , 306 siRNAs targeting a subset of 9 , 102 human genes ( listed in S2a and b Table ) , with three independent siRNAs per gene . A custom-made siRNA library from Dharmacon was used for the validation screen , with four independent siRNAs per gene ( listed in S2c and d Table ) . The protocols for the high-throughput siRNA screening approach as well as statistical and bioinformatics analyses are described in Supplemental Experimental Procedures . Cells were lysed in luciferase lysis buffer ( 1% ( v/v ) Triton X-100 , 10% ( v/v ) glycerol , 25 mM glycylglycine ( pH 7 . 8 ) , 15 mM MgSO4 , 4 mM EGTA and 1 mM dithiothreitol ) . For dual luciferase measurement of Firefly and Renilla luciferase in 384- or 96-well microplates , cells were washed once with PBS , lysed directly on the plate in 20 µl ( 384-well plates ) or 30 µl ( 96-well plates ) luciferase lysis buffer per well and frozen at −80°C . Shortly before measurement lysates were allowed to thaw at RT for 30 to 60 min . Luciferase assay buffer ( 25 mM glycylglycin ( pH 7 . 8 ) , 15 mM K2PO4 , ( pH 7 . 8 ) , 15 mM MgSO4 , 4 mM EGTA , 1 mM DTT and 2 mM ATP ) , supplemented with 70 µM D-luciferin ( P . J . K . , Kleinblittersdorf , Germany ) , was added to each well using a Multidrop 384 dispenser ( Thermo-Fisher , Martinsried , Germany ) , and plates were incubated for 5 min at RT in the dark . Firefly luciferase activity was measured for 0 . 1 sec in a Mithras LB940 multimode microplate reader ( Berthold Technologies , Bad Wildbad , Germany ) . After addition of luciferase assay buffer , supplemented with 7 . 14 µM coelenterazine ( P . J . K . ) , Renilla luciferase activity was measured for 0 . 5 sec using a 475 nm filter in the same multiwell reader . Replication assays were performed with Huh7-Lunet cells , infection assays with Huh7 . 5 or Huh7 . 5Fluc cells . Single-cell suspensions were prepared by trypsinization , washed with phosphate-buffered saline ( PBS ) and resuspended at a concentration of 107 Huh7-Lunet cells or 1 . 5×107 Huh7 . 5 cells per ml in Cytomix [69] supplemented with 2 mM ATP and 5 mM glutathione . For knock-down experiments , 100 µl cell suspension was mixed with 2 . 5 µM siRNA and transfected by electroporation using a GenePulser system ( Bio-Rad , Hercules , CA ) and a 0 . 2 cm gap cuvette ( Bio-Rad , Hercules , CA ) at 500 µF and 166 V . Cells were immediately diluted in complete DMEM and seeded as required for the given assay . SiRNAs are listed in Supplemental Experimental Procedures . The used siRNA ( Eurofins MWG Opern , Ebersberg , Germany ) had the following sequences ( only sequences of the sense strands are given ) : siHNRNPK#1 5′-GCAAGAAUAUUAAGGCUCU-3′; siHNRNPK#2 5′-GGUCGUGGCUCAUAUGGUG-3′; siHNRNPK#3 5′-UGACAGAGUUGUUCUUAUU-3′; siHNRNPK#4 5′-UAAACGCCCUGCAGAAGAU-3′; siHNRNPK3′NTR 5′-CGUUAUUGUUGGUGGUUUA-3′; siHNRNPKcontr . 5′-GAAAGUUUUUCUAAGACUA-3′; siRluc 5′-GUAGCGCGGUGUAUUAUAC-3′; siApoE 5′-CUAGUUUAAUAAAGAUUCA-3′; siContr . 5′-UGGUUUACAUGUCGACUAA-3′ . Cells in a confluent 24-well cell culture plate were washed once with PBS and harvested in 100 µl per well of 2 x protein sample buffer ( 200 mM Tris , pH 8 . 8 , 5 mM EDTA , 0 . 1% Bromophenolblue , 10% sucrose , 3% SDS , 2% β-mercaptoethanol ) followed by an incubation at 37°C for 30 min with 50–75 U benzonase ( Merck , Darmstadt , Germany ) and heated for 5 min at 95°C . Proteins were separated by SDS-polyacrylamide gel electrophoresis and electrotransferred onto polyvinylidene membranes ( Perkin Elmer; Rodgau , MA ) , followed by blocking with PBS-0 . 5%Tween 20 containing 5% dried milk for 2 h prior to 1h-incubation with specific antibodies , each diluted 1∶1 , 000 in PBS containing 1% dry milk . Membranes were washed 3 times with PBS supplemented with 0 . 5% Tween 20 and incubated for 1 h with either horseradish-peroxidase-conjugated rabbit- or mouse-specific secondary antibodies ( Sigma-Aldrich , Steinheim , Germany ) diluted 1∶10 , 000 . Membranes were developed by using the Western Lightning Plus-ECL reagent ( Perkin Elmer; Rodgau , MA ) and bands were visualized on Amersham Hyperfilm ECL ( GE Healthcare Life Sciences , Uppsala , Sweden ) . For production of the lentiviral vectors , 2 . 4×106 293T cells were seeded per 6 cm-diameter dish in a volume of 4 ml DMEM complete one day prior to transfection by using the JetPEI transfection kit ( Polyplus Transfection , NY , USA ) as recommended by the manufacturer . Production of the lentiviral vectors has been described elsewhere [70] . In brief , 5 µg of the respective pWPI plasmid , 5 µg of the packaging plasmid ( pCMV-R8 . 74 ) and a VSV envelope glycoprotein expression plasmid ( pMD . G ) were transfected into 293T cells . After 24 h transfection medium was replaced by 4 ml fresh DMEM complete and after an additional 24h-incubation , lentivirus particles-containing supernatant was harvested . Supernatant was filtered through a 0 . 45 µm filter and 1 . 5 ml of the filtrate was used to infect Huh7 . 5 target cells that had been seeded at a density of 2×105 cells/well of a 6-well plate 24 h before inoculation . Transduction of target cells with the lentiviral particles was performed in total three times to achieve high number of integrates and thus high expression levels . Transduced cells were subjected to selection by using medium containing the appropriate drug . Mouse monoclonal antibody recognizing NS3 of the JFH-1 isolate ( NS3-2E3 ) was generated in co-operation with H . Tang , Florida State University , USA . The mouse monoclonal antibody 9E10 recognizing NS5A domain III of the HCV isolates Con1 and JFH-1 was a kind gift of C . M . Rice ( Rockefeller University , New York , USA ) . The mouse monoclonal antibody recognizing HCV core protein ( C7/50 ) was kindly provided by D . Moradpour ( University of Lausanne , Switzerland ) . The rabbit polyclonal antibody recognizing NS2 ( NS2-1519 ) has been described earlier [71] . Mouse monoclonal antibody against Renilla Luciferase was obtained from Chemicon Millipore ( USA ) . Mouse monoclonal antibodies against Lamin and GAPDH were obtained from Santa Cruz Biotechnology ( Heidelberg , Germany ) . The polyclonal rabbit antibody recognizing HNRNPK was purchased from Acris Antibodies ( Herford , Germany ) . Primary antibodies against the HA-tag ( mouse , H3663 ) , HA-specific agarose beads ( A2095 ) , rabbit polyclonal antibody reacting with protein disulfide isomerase ( PDI ) as well as secondary horse radish peroxidase-conjugated antibodies were purchased from Sigma-Aldrich . Chicken polycloclonal antibody against the HA-tag was obtained from Abcam ( Cambridge , UK ) For immunoprecipitation 4×106 Huh7 . 5 cells were electroporated and seeded into a 10 cm-diameter dish . After 48 h , cells were washed with ice-cold PBS and lysed for 1 h on ice in ice-cold lysis buffer ( 20 mM Tris-HCl , pH 7 . 5 , 0 . 5% Nonidet P-40 , 1 mM sodium deoxycholate , 10 mM NaF , 2 mM EDTA , protease inhibitor cocktail ( Roche , Mannheim , Germany ) , 0 . 5 mM DTT , 2 . 5 U/ml RNasin ) , supplemented with either 150 mM NaCl for co-immunoprecipitation of RNA or 100 mM NaCl for co-immunoprecipitation of viral proteins . Lysates were cleared by centrifugation at 13 , 000 x g for 30 min at 4°C . Non-specifically binding proteins contained in the supernatants were removed by 1 h-incubation with prot-A agarose at 4°C , followed by centrifugation at 6 , 000×g for 5 min . Supernatants were used for immunoprecipitation ( 3 h at 4°C ) by using mouse monoclonal anti-HA-coated agarose beads ( Sigma-Aldrich ) followed by washing 4 times with ice-cold lysis buffer . HA-peptide ( Sigma-Aldrich ) was used to elute bound antigen . One half of the immunocomplex was used for RNA isolation by using the Nucleo Spin RNAII Kit ( Macherey-Nagel , Düren , Germany ) as recommended by the manufacturer . The other half of the lysates was dissolved in 60 µl 4 x protein sample buffer ( 400 mM Tris-HCl pH 8 . 8 , 10 mM EDTA , 0 . 2% bromophenolblue , 20% sucrose , 3% SDS and 2% β-mercaptoethanol ) . Immuno-captured proteins were separated by electrophoresis into 10% polyacrylamide gels and dried gels were subjected to autoradiography using BioMax MS films ( Kodak , Rochester , MN ) . For quantification , gels were either analyzed by phosphoimaging or different exposures of the films were scanned and subjected to densitometry by using the QuantityOne software ( Bio-Rad , Munich , Germany ) . RT-qPCR reactions were carried out using the two-step real-time RT-PCR approach . In a first step RNA was reverse transcribed into cDNA using the Multiscribe reverse transcriptase ( Applied Biosystems ) according to the manufacturer's protocol . Synthesized cDNAs were directly used for real-time PCR or stored at −80°C until further use . The final volume of the real-time PCR reaction mix was 15 µl and contained the following components: 7 . 5 µl 2x Green DYE master mix ( P . J . K . , Kleinbittersdorf , Germany ) , 1 . 5 µl primer mix ( 5 µM each ) , 3 µl ddH2O and 3 µl cDNA . Reactions were performed on an ABI PRISM 7000 Sequence Detection System using the following settings: 95°C: 10 min → 40x [95°C: 30 sec → 58°C: 60 sec → 72°C: 60 sec] . For each primer set reactions were carried out in triplicates using HNRNPK-specific primers ( forward: TTCAGTCCCAGACAGCAGTG; reverse: TCCACAGCATCAGATTCGAG ) . The ΔΔCT method [72] was used to calculate the relative expression levels . GAPDH forward: GAAGGTGAAGGTCGGAGTC , reverse: GAAGATGGTGATGGGATTTC . Cells were washed once with PBS and fixed with 500 µl of 4% paraformaldehyde for 20 min at RT followed by three times washing with PBS . Fixed cells were permeabilized by 5 min incubation in 500 µl of 0 . 5% Triton X-100 in PBS and washed 3 times with PBS . Immunostaining was performed by using the rabbit polyclonal HNRNPK antibody ( 1∶200 dilution ) , mouse monoclonal Renilla luciferase antibody ( 1∶1 , 000 dilution ) , mouse monoclonal core antibody ( 1∶200 dilution ) or rabbit polyclonal PDI antibody ( 1∶200 dilution ) . Antibodies were diluted in PBS supplemented with 3% bovine serum albumin ( BSA ) . After 60 min incubation and three times washing with PBS , cells were incubated with secondary antibodies conjugated to Alexa-Fluor 546 , 488 or 405 at a dilution of 1∶1 , 000 in PBS containing 3% BSA for 60 min in the dark . Nuclei were stained with DAPI ( Sigma-Aldrich ) . LDs were stained with HCS LipidTOX Deep Red neutral lipid stain ( Molecular Probes ) . Coverslips were mounted on glass slides with Fluoromount G ( Southern Biotechnology Associates , Birmingham , USA ) and samples examined using a Leica SP2 confocal laser scanning microscope ( Leica , Wetzlar , Germany ) or Perkin Elmer spinning disk confocal microscope . Images were edited and merged by using the ImageJ software package ( Rasband , W . S . , ImageJ , U . S . National Institutes of Health , Bethesda , Maryland , USA; http://rsb . info . nih . gov/ij/ ) . For Fig . 7B , images were deconvolved with the AutoQuantX3 using blind deconvolusion . Reconstructed 3D images were created using the Imaris 7 . 7 . 2 software . RNA in situ hybridization and indirect immunofluorescence . Cells were washed once with PBS and fixed with 500 µl of 4% paraformaldehyde for 20 min at RT , followed by three times washing with PBS . Fixed cells were permeabilized by 1 h incubation in 500 µl 70% ethanol and washed 3 times with PBS . Positive and negative-strand HCV RNA was detected by FISH using the QuantiGene ViewRNA ISH Cell Assay ( Affymetrix ) as recommended by the manufacturer . Immunostaining was performed by using a chicken polyclonal antibody recognizing the HA-tag ( 1∶200 dilution ) and a mouse monoclonal antibody recognizing HCV core protein ( C7/50 ) ( 1∶200 dilution ) after blocking the cells with 20% goat serum in PBS/0 . 01% Triton X-100 . Antibodies were diluted in PBS , supplemented with 10% goat serum . After 60 min incubation cells were washed three times with PBS and incubated with secondary antibodies conjugated to Alexa-Fluor 546 or 488 at a dilution of 1∶1 , 000 in PBS containing 10% goat serum for 60 min in the dark . Nuclei were stained with DAPI ( Sigma-Aldrich ) . Coverslips were mounted on glass slides with Fluoromount G ( Southern Biotechnology Associates , Birmingham , USA ) and samples examined using a Leica SP2 confocal laser scanning microscope ( Leica , Wetzlar , Germany ) . Images were edited and merged by using the ImageJ software package ( Rasband , W . S . , ImageJ , U . S . National Institutes of Health , Bethesda , Maryland , USA; http://rsb . info . nih . gov/ij/ ) . The cell proliferation reagent WST-1 ( Roche ) was used as recommended by the manufacturer . LDs were enriched as described earlier [73] . In brief , Huh7 . 5 cells were electroporated with 10 µg full length in vitro transcripts of the JFH-1 isolate and treated with 200 µM oleic acid ( Sigma-Aldrich ) for 16 h prior to harvesting . Cells were harvested 72 h post electroporation by scraping into PBS containing 5 mM EDTA . Cells were pelleted by centrifugation , resuspended in 1 . 3 ml HLM buffer ( 20 mM Tris , pH 7 . 4 , 1 mM EDTA , protease inhibitor cocktail ) and incubated on ice for 10 min . Cells were disrupted by douncing and nuclei were removed by centrifugation at 1 , 000×g for 10 min at 4°C . The post nuclear fraction was mixed with 0 . 5 volumes 60% ( wt/vol ) sucrose dissolved in HLM buffer and overlaid with 5 ml of HLM-5% ( wt/vol ) sucrose , followed by 5 ml of HLM buffer . The gradient was centrifuged at 13 , 000×g for 30 min at 4°C and then the centrifuge was stopped without brake . Four hundred µl of the LD fraction was collected from the top of the gradient and proteins were precipitated by using an acidified aceton/methanol mixture . Pellets were resuspended in SDS-containing 2x protein sample buffer and analyzed by Western-blot using a 10% polyacrylamide gel . Huh7 . 5 cells were co-electroporated with 2 . 5 µM of siRNA and 5 µg of Jc1 RNA and seeded into 6-well plates . To determine the amounts of extracellular infectivity , supernatant was harvested 72 h after electroporation , filtered through a 0 . 45 µm-pore-size filter and stored at 4°C . To quantify amounts of intracellular infectivity , cells were rinsed three times with PBS and scraped into 0 . 5 ml PBS . Cells were pelleted by centrifugation for 5 min at 700×g , resuspended in 0 . 5 ml of complete DMEM and subjected to three freeze-thaw cycles . Cell debris was removed by centrifugation for 10 min at 20 , 000×g . Virus titers were determined by limiting-dilution assay using Huh7 . 5 target cells and staining of the NS3 protein with the 2E3 antibody as described elsewhere [70] . To determine amounts of core protein , cells and supernatant were subjected to three freeze-thaw cycles and diluted to a final concentration of 0 , 5% Triton X-100/PBS/protease inhibitor ( Roche ) . Lysates were cleared by centrifugation at 20 , 000×g for 10 min at 4°C . HCV core protein was quantified using a commercial chemiluminescent microparticle immunoassay ( CMIA ) ( 6L47 , ARCHITECT HCV Ag Reagent Kit , Abbott Diagnostics , Abbott Park , USA ) according to the instructions of the manufacturer . To determine ApoE amounts , cells and supernatant were subjected to three freeze-thaw cycles . ApoE protein was quantified using a commercial Human Apo E ELISA Kit ( Cell Biolabs , USA ) according to the instructions of the manufacturer . HIV-based pseudotypes containing HCV envelope glycoproteins of the Con1 isolate were generated by transfection of 293T cells . Briefly , 6×107 cells were seeded into 10 cm-diameter dishes in a volume of 8 ml DMEM complete ( Life Technologies ) one day prior to transfection by using the JetPEI transfection kit ( Polyplus Transfection , NY , USA ) as recommended by the manufacturer . Production of the lentiviral vectors has been described elsewhere [74] . In brief , 5 µg of the respective HCV envelope protein expression construct ( pcDNAΔC-E1E2 ) , the HIV Gag-Polymerase construct ( pHIT60 ) and the VENUS-transducing lentiviral vector were transfected into 293T cells . After 24 h transfection medium was replaced by 8 ml fresh DMEM complete and after an additional 24 h and 48h-incubation , HCVpp-containing supernatant was harvested . Supernatant was filtered through a 0 . 45 µm filter and overlaid with 4 ml 20% ( wt/vol ) sucrose , followed by centrifugation at 26 , 000×g for 2 h at 4°C . Pellets were suspended in DMEM complete and used for infection of Huh7 . 5 cells that had been seeded into 6-well plates for 72 h . Cells were detached by adding trypsin , fixed with 500 µl of 2% paraformaldehyde for 1 h at RT followed by washing with PBS . Samples were analyzed by flow cytometry and data were processed by using the FlowJo Analysis Software package ( Tree Star , USA ) . | As obligate intracellular parasites with limited gene coding capacity viruses exploit host cell machineries for the sake of efficient replication and spread . Thus , identification of these cellular machineries and factors is necessary to understand how a given virus achieves efficient replication and eventually causes host cell damage . Hepatitis C virus ( HCV ) is an RNA virus replicating in the cytoplasm of hepatocytes . While viral proteins have been studied in great detail , our knowledge about how host cell factors are used by HCV for efficient replication and spread is still scarce . In the present study we conducted a comprehensive RNA-interference-based screen and identified 40 genes that promote the HCV lifecycle and 16 genes that suppress it . Follow-up studies revealed that one of these genes , the heterogeneous nuclear ribonucleoprotein K ( HNRNPK ) , selectively suppresses production of infectious HCV particles . We mapped the domains of HNRNPK required for this suppression and demonstrate that this protein selectively binds to the HCV RNA genome . Based on the correlation between suppression of virus production , HCV RNA binding and recruitment to lipid droplets , we propose that HNRNPK might limit the amount of viral RNA genomes available for incorporation into virus particles . This study provides novel insights into the complexity of reactions that are involved in the formation of HCV virions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"cell",
"biology",
"gastroenterology",
"and",
"hepatology",
"biology",
"and",
"life",
"sciences",
"molecular",
"biology"
] | 2015 | Identification of HNRNPK as Regulator of Hepatitis C Virus Particle Production |
FSD-1 , a designed small ultrafast folder with a ββα fold , has been actively studied in the last few years as a model system for studying protein folding mechanisms and for testing of the accuracy of computational models . The suitability of this protein to describe the folding of naturally occurring α/β proteins has recently been challenged based on the observation that the melting transition is very broad , with ill-resolved baselines . Using molecular dynamics simulations with the AMBER protein force field ( ff96 ) coupled with the implicit solvent model ( IGB = 5 ) , we shed new light into the nature of this transition and resolve the experimental controversies . We show that the melting transition corresponds to the melting of the protein as a whole , and not solely to the helix-coil transition . The breadth of the folding transition arises from the spread in the melting temperatures ( from ∼325 K to ∼302 K ) of the individual transitions: formation of the hydrophobic core , β-hairpin and tertiary fold , with the helix formed earlier . Our simulations initiated from an extended chain accurately predict the native structure , provide a reasonable estimate of the transition barrier height , and explicitly demonstrate the existence of multiple pathways and multiple transition states for folding . Our exhaustive sampling enables us to assess the quality of the Amber ff96/igb5 combination and reveals that while this force field can predict the correct native fold , it nonetheless overstabilizes the α-helix portion of the protein ( Tm = ∼387K ) as well as the denatured structures .
Small “ultrafast” folders ( proteins that fold on the order of microseconds ) , both naturally occurring and designed , have received considerable attention in the last few years . These proteins have the singular advantage of being computationally tractable , thus bridging the gap between experimental and in silico studies . They permit not only a testing of the accuracy of computational force fields , but also ( in the event that the force field proves to be adequate ) an assessment of protein folding theories . Folding mechanisms and the possibility of multiple folding pathways , both of which can be difficult to determine from standard bulk measurements , can be resolved through an analysis of molecular dynamics simulations . FSD-1 is a 28 residue designed ultrafast folder with a ββα ( hairpin/helix ) fold [1] . The protein has a well-defined hydrophobic core , and unlike the more commonly studied ββα BBA5 protein ( which has a D-proline at the β-turn position ) , only contains naturally occurring residues . The folding time of FSD-1 has not been reported , but the folding kinetics of a close analog , FSD-1ss ( involving substitution of two non-natural aromatic residues at positions 6 and 26 ) have been monitored using laser-induced temperature-jump spectroscopy ( Table 1 ) [2] . This modified protein displayed two folding phases ( τ1∼150 ns and τ2∼4 . 5 µs ) at 322 K , placing FSD-1ss at the top range of known ultrafast folders . Although the N-terminal region FSD-1ss ( residues 16–23 ) adopts a loose U-shape rather than the tight β-hairpin seen in FSD-1 , the overall tertiary structure of FSD-1ss is similar to FSD-1 with Cα Root Mean Square Deviation ( Cα-RMSD ) of only 2 . 2 Å . One can therefore expect that FSD-1 folds on similar microseconds timescales as FSD-1ss . The thermal unfolding of FSD-1 , as determined from Circular Dichroism ( CD ) [1] and Differential Scanning Calorimetry ( DSC ) [3] , is reversible , but weakly cooperative , with a relatively low melting temperature ( Tm = 315 K ) . Mayo and co-workers , who designed this protein , attributed the observed transition to the melting of the entire protein . Feng et al . [3] have however recently challenged this interpretation and have proposed that the broad transition , which lacks clearly defined folding or unfolding baselines , in fact reflects only the melting of the α-helical segment ( residues 14–26 ) of the protein . Should this interpretation be correct , then one would have to reconsider FSD-1 as a model system for studying the folding of α/β proteins . Prior simulations of the FSD-1 protein have met with mixed ( and sometimes conflicting ) results , and have not provided a clear picture of the nature of the melting transition . For instance , a replica exchange molecular dynamics ( REMD ) simulation of FSD-1 in explicit solvent , using the Amber protein force field ( ff03 ) , TIP3P water and the NVT ensemble , predicted a melting temperature of 411 . 59 K [4] , ∼100 K higher than the experimental value of 315 K . Simulations performed using a different force field , water model and simulation protocol ( OPLS-AA/L 2001 force field , TIP4P water model and NPT ensembles with REMD simulation ) lead to a melting temperature that is 84K higher than what is observed experimentally [3] . These unsatisfactory results may be the result of inadequate force fields , or/and due to insufficient sampling . In order to overcome possible sampling issues related to the use of explicit solvents , a number of groups have turned to coarse-grained protein models [5] or to implicit water models . Pak and co-workers , using the CHARMM 19 force field in conjunction with a GB solvation model [6] witnessed the folding of FSD-1 to a structure 2 . 56 Å Cα-RMSD from the NMR structure at 440 K in 15 ns , i . e . at a temperature well above the experimental Tm and with a folding time off by orders of magnitude . The authors had better success using replica exchange molecular dynamics ( REMD ) simulations and a newly modified version ( param99MOD5 ) of AMBER 99 with GBSA implicit solvent [7] , obtaining a computational melting temperature of ∼309 K . However , the predictive power of these simulations is uncertain given that FSD-1 was used as a training peptide in the optimization of the force field . Finally , using a newly optimized force field in combination with the recently developed implicit solvent ( IGB = 5 ) [8] , Lei and co-workers [9] were able to fold the double mutant of FSD-1 ( FSDEY ) [10] , into its native state with high population ( >64 . 2% ) and high fidelity ( 1 . 29 Å ) . However , the computationally generated heat capacities failed to produce a melting transition at 315K , and the melting of the helix was observed at 360K which is higher than that of stable helical protein [11] . In the present paper , we investigate the folding of FSD-1 using the Amber ff96 protein force field combined with the implicit water solvent IGB = 5 . This combination has been shown by Dill and co-workers to have a good balance of α/β propensity in the case of small peptides [12] . We have recently been able to use this combination to investigate the conformations adopted by natively disordered amyloid peptides ( e . g . prion fragment [13] and amylin [14] ) which can sample a variety of conformations including α , β and α/β . Recent successes of this force field/implicit solvent combination include the successful folding of the 39-residue NTL9 protein [15] by the Pande group and the unfolding of the 64-residue protein L [16] . The majority of studies involving the ff96/igb5 model have focused on assessing the model's reliability in generating accurate structural properties of proteins . Here , we present a thorough thermodynamic and kinetic analysis that enables , through a direct comparison with experimental data , an in-depth evaluation of the strength and weaknesses of this force field/implicit water model combination . Our simulations offer the first comprehensive interpretation of the broad transition at 315 K and resolve the issue of whether it corresponds to the melting of the protein or to an only helix-coil transition . Furthermore , our extensive simulations enable a thorough exploration of the energy landscape for folding , a structural characterization of transition state ensemble , and explicitly demonstrate the existence of multiple folding routes . Our simulations indicate that while ff96/igb5 can be reliably used to identify the folded state of the protein , as well as predict the thermodynamic order of formation of the structural elements in folding , there remain areas in which the protein force field/implicit water model needs to be improved . In particular , we find that ff96/igb5 overstabilizes both the denatured state and the helical portion of the protein .
To directly probe folding routes , we performed 20 CMD simulations ( 1 µs each ) at 300K starting from an extended conformation . The last snapshot of each trajectory ( Text S1 ) offers a quick structural assessment: a successful α/β fold was formed in 2 trajectories ( A–B ) , an unpacked α/β fold was observed in 1 trajectory ( C ) , a partial α-helix was formed in 6 trajectories ( D–I ) , a partial β-sheet was formed in 7 trajectories ( J–P ) and a compact coil-turn fold was formed in 4 trajectories ( Q–T ) . A representative structure for each structural family is shown in Fig . 5 . Altogether , 10% of the trajectories folded to the correct α/β fold , yielding an estimated folding time of 10±7 µs ( see Methods section ) . This time needs to be corrected taking into account the fact that our simulations used a friction coefficient 1/60 of the one in water . The corrected folding time is at least 600 µs , far larger than our estimate of 2 . 2 µs from the barrier height from REMD ( or the 4 . 5 µs folding time of the FSD1-ss mutant ) . The implication is that the CMD simulations lead to a much rugged folding landscape than “reality” and that the ff96/igb5 combination may overstablize unfolded structures ( e . g . C–Q in Fig . 5 ) . The Cα-RMSD of each trajectory ( Text S1 ) was calculated ( defined as Cα-RMSD<3 . 0 Å for at least consecutive 30 ns ) for each trajectory . The two successful folding trajectories are shown in Fig . 6 and Fig . 7 . By applying the clustering analysis described in the Methods section to each trajectory , we identified structural families whose population is greater than 1% of the total snapshots . The representative structures ( the centroid of each structural family ) , with their respective time of occurrence and abundance , are shown in Fig . 6 and Fig . 7 . Strikingly , two different folding routes are observed in these two trajectories .
FSD-1 , a designed small ββα ultrafast folder , has received considerable experimental and computational attention as a model system for studying protein folding mechanisms and for testing computational models . However , the validity of using FSD-1 as a prototypical folder has recently been challenged by Feng et al [3] . They argue that the broad melting transition at 315 K observed by CD and DSC is mainly due to the melting of the helical portion of the protein , rather than to the melting of the entire protein . They point out that an overall protein transition should exhibit a better-defined baseline and a higher melting temperature ( for comparison , HP35 folds at 342 K ) and the β-hairpin part appears to be flexible and lack of stability based on their REMD simulations using OPLS-AA/L 2001 force field , TIP4P water model and NPT ensemble . Their simulations and others [4] , thus far , have not been able to address this controversy and provide an explanation for the broad melting transition of FSD-1 , with computational folding transitions all lying at much higher temperatures than the experimental ones . Reliable folding of mixed α/β fold proteins like FSD-1 is notoriously difficult , mostly because most force fields are either heliophilic [20] , [21] or β-centric [12] . In the present paper , we investigate the folding of FSD-1 using the Amber ff96 force field combined with the implicit solvent IGB = 5 , which appears to offer a reasonable balance of helical and beta propensities . Our REMD simulations show two broad peaks in the excess heat capacity at Tm = 321±6 K and Tm = 387±4K . These peaks correspond to the two structural transitions identified from the computational thermal-denaturing curves: 1 ) formation of the hydrophobic core ( Tm = 323±6 K ) , β-hairpin ( Tm = 307±3 K ) and tertiary fold ( Tm = 305±2 K ) and 2 ) formation of the α-helix ( Tm = 373±10K ) . By comparing the simulations results with the DSC and CD experiments , we find that the first transition qualitatively agrees with the experiments , whereas the second transition is an artifact of the simulations resulting from the ff96/igb5 induced overstabilization of the α-helix . This suggests that the improved igb5 solvation model does not fully resolve the secondary balance problem inherent to the heliophilic Amber ff96 force field and that further refinement of the protein force field is necessary . While the force field clearly overstabilizes the helix , the fact that the helix is more stable than the other secondary structural elements is confirmed by the CD spectra . A re-examination of the experimental CD spectra of FSD-1 at 353K and 277 K presented in Fig . 2 of ref 3 shows that the spectrum above the experimental Tm = 315 K possesses helical features rather than coil-only features and the CD spectrum below the experimental Tm = 315 K possesses both β-hairpin-rich and helix-rich features rather than helix-only features . Indeed , the spectrum at 353 K lacks random-coil features ( i . e . very low ellipticity above 210 nm and negative bands near 195 nm ) ; instead it shows helical negative bands at 203 nm and 222 nm , indicating the presence of helical structure . In addition , the CD spectrum of FSD-1 at 277 K not only shows two helix bands at 207 and 220 nm but also contains a sheet band at 218 nm , intimating that the protein shows signs of a folded α/β protein at this temperature . Furthermore , the CD spectrum of FSD-1 is significantly different form the CD spectra of the helix-only protein [11] , suggesting it contains both α and β secondary structure . Put together , our prediction is as follows: although the α-helix is the most stable structural component ( i . e . the helix-to-coil transition is dominant in the denaturation process ) , the broad denaturation transition seen experimentally also has a significant contribution arising from the spread in melting temperatures of the hydrophobic core , the β-hairpin . In other words , the breadth does not arise solely from the helix-coil transition as had been proposed by Feng and coworkers [3] . Our simulations explain the experimentally observed lack of well-defined baselines as due to the marginal stability of this protein , coupled with the spread in melting temperatures of the individual secondary and tertiary components . We suggest that mutations that would enhance the stability of the β-hairpin could improve the overall stability and the cooperativity of this designed protein , making it an even better model for natural α/β proteins . The folding free energy landscape at temperatures near the folding temperature ( Tm = 321±6 K ) shows two well-defined folded and unfolded basins . Transition state structures identified from the structures at the top of the barrier satisfying the Pfold analysis revealed common features: full formation of the C-terminal helix and partial formation of the hydrophobic core and the N-terminal β-hairpin . The free energy barrier height enables us to calculate the folding time of FSD-1 . Following Kramer's theory , the folding time is estimated to be τfolding = 2 . 2±0 . 7 µs , which is of the same order of magnitude as that of FSD-1ss [2] , which shows two folding phases ( τ1∼150 ns and τ2∼4 . 5 µs ) . Should FSD-1 present the same two phases , we could assign the first fast phase to the formation of hydrophobic core , and assign the concurrent formation of N-terminal β-hairpin and the tertiary fold to the slower phase ( ∼4 . 5 µs ) . Furthermore , our CMD trajectories provide detailed atomic information of possible folding routes starting from a straight chain . Although the general folding features gleaned from the REMD thermodynamic analysis ( e . g . early formation of the C-terminal helix and thereafter folding N-terminal hairpin and the tertiary fold ) are seen in the two successful folding trajectories , a significant heterogeneity is present across the two folding trajectories: 1 ) the C-terminal helix can initiate from different sites ( residues 15–20 in Trajectory 1 vs . residues 19–22 in Trajectory 2; 2 ) different structures were visited along the folding routes; 3 ) the transition state structures found in each trajectory , validated by Pfold analysis , differ , supporting the notion of different transition structures for multiple routes [22]; 4 ) significant non-native topologies and secondary structures are also sampled along the two folding routes . The presence of these structures underlines the importance of considering non-native interactions , in addition to the natively favored interactions used in Go-like models [23] . Non-native conformations play a role in modulating folding times and mechanisms . Put together , our all-atom CMD data provide direct evidences to a funneled energy landscape [18] with multiple folding pathways and a diverse transition state ensemble . Nonetheless , the low number of successful folding runs ( only 2 out of 20 trajectories ) and the resulting lengthy folding time ( ∼600 µs vs . 2 . 2 µs ) indicates that the CMD simulations generate a much more rugged folding landscape than reality . A better modeling by the implicit protein force field for the unfolded state might resolve the problem . It is interesting to compare the folding mechanism of FSD-1 to a similar protein ( BBA5 ) with ββα fold . In the case of BBA5 , Pande and coworkers showed that this protein follows a diffusion-collision model [17] , [24]: docking of the preformed α-helix and β-hairpin . Thus , for BBA5 , the formation of the β-hairpin precedes the formation of the tertiary fold , whereas in the case of FSD-1 , we have shown that the β-hairpin and the tertiary fold form concurrently . Clearly , the D-proline present at the turn region of the β-hairpin of BBA facilitates the formation of this structural element , but , as a result , it may reduce the overall folding cooperativity of the protein . It is possible that the introduction of non-natural amino-acids makes this protein less “funnel-like” . Nonetheless , BBA5 and FSD-1 fold in microsecond despite these differences in folding mechanism , consistent with contact order theory that suggests that native topology is a major factor in determining the folding time [25] .
An initial energy minimization was performed on an extended chain conformation and the minimized structure was used as the input for both replica exchange molecular dynamics ( REMD ) and conventional molecular dynamics simulations ( CMD ) simulations . In REMD simulation [28] , [29] , [30] , [31] , 16 replicas were set up with initial temperatures exponentially spaced from 271 to 465 K for solution phase calculations ( i . e 271 . 0 280 . 0 289 . 3 300 . 0 311 . 2 322 . 7 334 . 7 347 . 2 360 . 1 373 . 4 387 . 3 401 . 7 416 . 6 432 . 1 448 . 2 465 . 0 , see reference [32] for the algorithm used to optimize them ) . 20 CMD simulations were conducted at 300 K . Initial velocities for each trajectory were generated according to the Maxwell-Boltzmann distribution for its target temperature . The first 1 . 0 ns of REMD simulation was performed to equilibrate the system at its target temperatures . After equilibrium , exchanges between neighboring replicas were attempted every 1000 MD steps ( 2 . 0 ps ) and the exchange rate among replicas was ∼20% . SHAKE [33] was applied to constrain all bonds connecting hydrogen atoms and thus a time step of 2 . 0 fs . In order to reduce computation time , non-bonded forces were calculated using a two-stage RESPA ( reference system propagator algorithm approach ) [34] where the forces within a 12 Å radius were updated every step and those beyond 12 Å were updated every two steps . The Langevin dynamics was used to control the temperature 300K using a collision frequency of 1 . 0 ps−1 . The lower collision frequency than a typical value ( ∼60 ps−1 ) for water solvent was used for a better conformational sampling . The center of mass translation and rotation were removed every 250 MD steps ( 0 . 5 ps ) . Each trajectory was run for 1 . 25 µs and 1 . 0 µs respectively in REMD and CMD simulations , giving a cumulative simulation time of 20 . 0 µs . The trajectories were saved at 2 . 0 ps intervals for further analysis . For analysis of secondary structure , the STRIDE program of Frishman and Argos [35] is used . For analysis of tertiary structural families , the snapshots are clustered by the GROMACS protocol [36] , in which the structure similarity score is based on pair wise Cα-RMSD of 2 . 0 Å . This is done in order to reduce a large number of the sampled structures into a few structural families . The structure with the largest number of neighboring structures within the cutoff , was selected as the representative structure of the structural family/cluster . The formation of hydrophobic core , secondary structures and tertiary structure are important events in the protein folding process . To monitor the formation of the hydrophobic core of FSD-1 , we calculate the radius of gyration for the hydrophobic core formed by the residues Ala5 , Ile7 , Phe12 , Leu18 , Phe21 , Ile22 , and Phe25 . To characterize the secondary and tertiary structure formation , the Cα-RMSDs of the N-terminal hairpin ( residues 3–13 ) , the C-terminal α-helix ( residues 14–26 ) and the whole protein ( residues 3–26 ) , are calculated against the NMR structure ( pdb id: 1FSD ) . The terminal residues are flexible and thus are not included the calculation . The fractions of the folded state at various temperatures are directly calculated from the REMD trajectories based on each of the four order parameters . Here , the folded state is defined by setting a cutoff of Cα-RMSD or the radius of gyration ( Rg ) for each order parameter . The cutoff is the value that separates the two populations in the distribution of each order parameter at 323 K ( Text S1 ) . The values are listed in Table 2 . The melting transition temperature is obtained at the midpoint where = 0 . 5 . Assuming a two-state thermodynamic model , the thermodynamic parameters in the direction of folding to unfolding can be obtained from the following van't Hoff analysis [37]: ( 1 ) ( 2 ) ( 3 ) where , and ( assumed to be constant across temperature ) are the changes in the van't Hoff enthalpy , entropy at and heat capacity at constant pressure . Also , ( i . e . peak temperature values ) and the denaturation enthalpy , ( 4 ) can be obtained from the excess heat capacity profile , which is calculated by subtracting the heat capacity of the native state ( i . e . the lowest temperature ) from the absolute heat capacity profile [38]: ( 5 ) The absolute heat capacity ( ) is estimated from the potential energy distribution at each temperature from the REMD simulation by ( 6 ) where E is the potential energy , R is the gas constant , and T is the temperature . Estimation of the folding time: The folding time can be obtained from Kramers' theory of unimolecular reaction rates in solution [39] , [40] , [41]: ( 7 ) where is the height of the free energy barrier , R is the gas constant and T is the absolute temperature , A is the pre-exponential factor . As to the 40-residue protein BBL , A is about 0 . 8 µs [42] estimated by the measured relaxation time [43] of the Förster resonance energy transfer ( FRET ) efficiency for the acid-denatured state of the BBL with donor and acceptor fluorophores attached to the N and C termini at 305 K . Using the linear length scaling suggested by the homopolymer collapse theory [44] , the pre-exponent factor A for a 28 residue FSD-1 is about 0 . 56 µs ( 0 . 80 µs * 28/40 ) . Alternatively , the folding time based on two-state folding model can be estimated from a large number of CMD simulations ( ) of a short duration ( ) [45]: ( 8 ) where is the number of the folded trajectories . | The protein folding process , in which a linear chain of amino acids reaches its biologically active three-dimensional shape , is fundamental to life . Small “ultrafast” folders , proteins that fold in microseconds , have received considerable attention , because these proteins serve as model systems for the folding of larger proteins , and thus permit a testing of the accuracy of computational models as well as an assessment of protein folding theories . FSD-1 , a designed small ultrafast folder with a ββα fold , has been actively studied in the last few years as a model system for mixed α/β fold proteins . The suitability of this protein to describe the folding of naturally occurring proteins has however recently been challenged based on the observation that the melting transition is very broad , with ill-resolved baselines . Prior simulations have not been successful in providing an interpretation of this broad melting transition . In the present study , our extensive molecular dynamics simulations using the AMBER protein force field ( ff96 ) coupled with the implicit solvent model ( IGB = 5 ) shed new light on the nature of the folding transition of this protein , as well as reveal the strengths and weaknesses of the force field in predicting the thermodynamics and kinetics of folding . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biophysics/protein",
"folding",
"biophysics/experimental",
"biophysical",
"methods",
"biophysics/theory",
"and",
"simulation"
] | 2010 | On the Origins of the Weak Folding Cooperativity of a Designed ββα Ultrafast Protein FSD-1 |
Clonorchiasis is a neglected tropical disease caused by Chinese liver fluke , Clonorchis sinensis infection . C . sinensis is a biological carcinogen causing cholangiocarcinoma in humans . In the mammalian host , C . sinensis newly excysted juveniles ( CsNEJs ) migrate from the duodenum into the bile duct . Bile drives the chemotactic behavior of CsNEJs . Little is known about which components of bile induce the chemotaxis . We designed a chemotaxis assay panel and measured the chemotactic behavior of CsNEJs in response to bile or bile acids . The CsNEJs migrated toward 0 . 1–1% bile but away from 5–10% bile . The CsNEJs showed strong chemoattraction to cholic acid ≥25 mM , but chemorepulsion to lithocholic acid ≥0 . 25 mM . To the CsNEJs , mixture of cholic acid and lithocholic acid was chemoattractive at a ratio greater than 25:1 but chemorepulsive at one smaller than that . Regarding migration in the mammalian hosts , high concentration of lithocholic acid in the gallbladder bile may repel CsNEJs from entering it . However , bile in the hepatic bile duct has a chemoattractive strength of cholic acid but a trace amount of lithocholic acid . Collectively , our results explain why the CsNEJs migrate principally to the hepatic bile ducts , bypassing the gallbladder .
Many parasites seek out and invade hosts using host-emitted chemical cues , exhibiting a trait known as chemotaxis . Recent studies have encompassed a range of examples , including the miracidia of Schistosoma species swim along a chemical gradient toward the snail host [1 , 2] . The larvae of Echinostoma species , both miracidia and cercariae , locate their snail hosts by sensing chemotactic cues [3] . Invasion by S . mansoni schistosomula is induced by chemo-orientation toward d-glucose and l-arginine in the host serum [4] . When Diplostomum spathaceum cercariae invade fish , they recognize monosaccharides , glycoproteins , and fatty acids [5] . Within the hosts , these kinds of chemotaxis also are crucial for successful parasitism and survival although the parasites can migrate to atypical location [6] . Bile and bile acids provide pivotal cues to gastrointestinal parasites . Glycine-conjugated cholic acid stimulates excystation of Fasciola hepatica metacercariae [7] . Whole bovine bile and dehydrocholic acid stimulate the locomotor cycle of F . hepatica juveniles [8] . Oviposition of S . mansoni adults is increased by bile components , especially tauroursodeoxycholic acid [9] . Survival rate of newly excysted Clonorchis sinensis juveniles ( CsNEJs ) increases in low concentration of bile [10] . Bile acids and conjugated bile salts , except lithocholic acid , enhance activity of CsNEJs [10] . Bile acids , which comprise most of the organic compounds in bile , include cholic acid ( 34% ) , chenodeoxycholic acid ( 39% ) , and deoxycholic acid ( 26% ) as major components and lithocholic acid ( <0 . 5% ) as a trace constituent in both gallbladder and hepatic bile duct [11 , 12] . In mammals , the main function of bile acids is to facilitate the formation of micelles for fat absorption . Several bile acids , such as whole bovine bile , dehydrocholic acid and tauroursodeoxycholic acid , are known to influence physiological or kinetic activities of flukes [8–10] . On the basis of these studies , we suspected that the chemotactic behavior of CsNEJs toward bile could be associated with bile acids . As a bile-dwelling parasite , C . sinensis is the most prevalent liver fluke in East Asian countries , infecting more than 200 million people [13] . The World Health Organization has recognized C . sinensis as a biological carcinogen for cholangiocarcinoma in human [14] . The mammalian hosts become infected by eating freshwater fish containing C . sinensis metacercariae . The ingested metacercariae excyst in the duodenum , and CsNEJs promptly migrate to the intrahepatic bile duct . Bile is assumed to be a chemoattractant to CsNEJs , since they migrate only to the bile duct of a rabbit whose gallbladder is stimulated to release bile [6] . However , it is not known which component of bile drives the migration of CsNEJs , and why they prefer to move to the liver rather than to enter the proximal and bile-rich gallbladder . In order to investigate chemotactic migration of C . sinensis , we designed and fabricated a custom-made chemotaxis assay trough similar to the tubular route from the ampulla of Vater to the biliary passages in the mammalian host . With the chemotaxis assay panel , we investigated which bile components induce this peculiar chemotactic behavior of CsNEJs .
Naturally infected Pseudorasbora parva , the second intermediate host of C . sinensis , was purchased at Hunhe fish market in Shenyang , Liaoning Province , People’s Republic of China . The fish was ground and digested in artificial gastric juice ( 0 . 5% pepsin [MP Biochemicals Co . , Solon , OH , USA] , pH 2 . 0 ) for 2 h at 37°C [15] . The solid matter was removed from digested content by filtration through a sieve of 212-μm mesh diameter . The C . sinensis metacercariae were collected using sieves of 106- and 53-μm mesh diameter and washed thoroughly several times with 0 . 85% saline . The C . sinensis metacercariae were gathered under a dissecting microscope and stored in phosphate-buffered saline ( PBS ) at 4°C until use . The metacercariae were excysted in 0 . 005% trypsin ( Difco , Detroit , MI , USA ) and used as CsNEJs for experiments . A custom-made chemotaxis assay panel with 8 troughs was crafted . Eight half-round troughs with dimensions of 100 mm long , 10 mm wide and 5 mm deep , were carved in a polycarbonate block ( Fig 1A and 1B ) . Each trough was graduated , with 0 at the center , +10 to +50 mm on the left side , and −10 to −50 mm on the right side . CsNEJs were placed at the center of trough in the custom-made chemotaxis assay panel , and then serially diluted bile solutions were dropped at one end ( Fig 1A and 1C ) . All experiments were performed at our lab in Chung-Ang University College of Medicine . A walk-in incubator ( Model No . J-RHC; JISICO CO . , LTD , Seoul , Korea; http://www . jisico . co . kr ) was built-in to maintain constant temperature and humidity with dimensions , 250 cm wide , 296 cm long and 210 cm high ( S1A and S1B Fig ) . For all experiments , the walk-in incubator was equilibrated at temperature 37°C and at humidity 80% for 1 h prior to the experiments . In all chemotaxis assays , 1× Locke’s solution was used as a base solution [10] . Each trough was filled with 1 ml of 1× Locke’s solution and approximately 20 CsNEJs were placed at the center 0 point using a micropipette . After allowing 5 min for adaptation , CsNEJs were stimulated by dropping bile or bile acid solutions ( Sigma-Aldrich , St . Louis , MO , USA ) at the +50 mm point . Behavior and migration distance of CsNEJs were observed under a dissecting microscope at each time point . All experiments were performed inside a walk-in incubator maintained at 37°C and 80% humidity as described above . To minimize temperature fluctuation , the chemotaxis panel was covered with acrylic lid except when the chemicals were applied or CsNEJs were observed . Bile or bile acid solutions were freshly made immediately before use . For bile chemotaxis assay , 10 μl of 0 . 01–10% bile solution was dropped at the +50 mm point , and then migration distance of CsNEJs was recorded at a given time interval . As a control , 10 μl of 1× Locke’s solution was used . Four kinds of bile acids , i . e . , cholic acid , deoxycholic acid , chenodeoxycholic acid and lithocholic acid , were used in the evaluation . Four microliters of bile acid solution were dropped at the +50 mm point in the trough , and then migration distance of CsNEJs was recorded . As bile acids were dissolved in dimethyl sulfoxide ( DMSO , Sigma-Aldrich , St . Louis , MO , USA ) , 4 μl DMSO was dropped at the +50 mm point as a control . The CsNEJs that died during the assay were excluded . To allow a longer migration distance , the scale in the chemotaxis assay trough was rearranged: the starting 0 point was moved to the right end of the trough and the graduation was marked +10–+80 mm toward the left end . At the beginning , CsNEJs were placed at 0 point in the trough and 4 μl of 50 mM cholic acid was dropped at the +30 mm . Every 10 min , the cholic acid solution was sequentially dropped at the + 50 mm and +70 mm point in the trough . Mean chemotactic distance ( mm ) was calculated by summing the migration distances of all CsNEJs and dividing it by the number of CsNEJs . All experiments were performed in triplicate with different batches of CsNEJs . Each value was presented as a mean ± standard error of mean . The significance of difference was statistically analyzed by Student’s t-test with p-value less than 0 . 05 considered significant .
Chemotactic migration distance of CsNEJs was normalized to the control group ( Fig 2A ) . The CsNEJs responded immediately and migrated concentration-dependently toward 0 . 1–1% bile , but away from 5 and 10% bile during 6 h of observation . When exposed to 10% bile solution , some CsNEJs shrank , moved very slowly , or stopped moving , but other CsNEJs immediately moved away from the bile solution . Chemotactic response of CsNEJs to individual bile acids was investigated . CsNEJs migrated toward cholic acid of concentrations 25 mM or higher during 6 h ( Fig 2B ) . In fact , the majority of the CsNEJs movement was observed during the first 1 h . Closer observation revealed that the CsNEJs migrated very quickly toward cholic acid within as short as 10 min , slowed for 20 min , and then moved only minimally from that point ( Fig 2C ) . The half-maximum effective concentration ( EC50 ) was estimated as 17 mM ( Fig 2D ) . The chemoattractive effect of cholic acid was saturated at concentrations above 50 mM . Cholic acid of 25 mM concentration was attractive to CsNEJs at 25–40°C . A higher temperature of 34–40°C induced CsNEJs to migrate a longer distance toward cholic acid than a lower temperature of 25–31°C did ( S2 Fig ) . Deoxycholic acid and chenodeoxycholic acid , major bile components , were less attractive to CsNEJs . Toward the two bile acids , CsNEJs showed insignificant chemotactic behaviors ( S3A and S3B Fig ) . In the chemotactic assay trough , CsNEJs migrated quickly toward cholic acid for a short time and then lingered at a point . It was suspected that CsNEJs could not sense a concentration gradient of cholic acid because it became dissipated and equilibrated as time passed . To test whether CsNEJs remained sensitive to cholic acid when they stopped moving , they were re-attracted by adding 50 mM cholic acid to the trough . The CsNEJs responded immediately and moved to the added cholic acid , stopped moving within 5 min , and remained at that point . The CsNEJs were found to move at different speeds and were divided arbitrarily into two groups: fast and slow movers . The fast movers reacted swiftly to cholic acid , moved quickly , and migrated farther than the slow movers . Nevertheless , in both groups , the resting CsNEJs still retained sensitivity to cholic acid , and resumed migration to repetitive attractions of cholic acid ( Fig 3 ) . In contrast to the chemoattractive cholic acid , lithocholic acid acted as a chemorepellent to CsNEJs . Toward 0 . 13 mM or lower concentrations of lithocholic acid , CsNEJs wriggled near the starting line , but moved away from lithocholic acid at a concentration of 0 . 25 mM or higher ( Fig 4A and 4C ) . At a threshold concentration of 0 . 25 mM lithocholic acid , CsNEJs moved minimally compared to controls for 1 . 5 h , then slowly migrated 4 . 0 mm after a total of 3 h . The EC50 was estimated as 0 . 38 mM ( Fig 4B ) . Lithocholic acid at 1 . 25 mM stimulated CsNEJs to migrate 2 . 8 mm in 1 h . As the lithocholic acid concentration increased , the migrating speed increased in a concentration-dependent manner with a maximum reaching 8 . 4 mm at 5 mM . However , as the concentration of lithocholic acid and the duration increased , increasing number of flukes began to shrink and eventually died . When 5 mM lithocholic acid was applied , all CsNEJs died in 1 h ( Fig 4A ) . CsNEJs encounter chemoattractive cholic acid and chemorepellent lithocholic acid concurrently in the mammalian host . The chemotactic response of CsNEJs to a mixture of cholic acid and lithocholic acid was tested . To a mixed solution with cholic acid fixed at 50 mM , the migration distance decreased as lithocholic acid concentration increased to 1 mM . When lithocholic acid concentration was 2 mM or higher , CsNEJs turned around and moved in the opposite direction from the mixed solution ( Fig 5A ) . Conversely , when lithocholic acid was fixed at 1 mM concentration and cholic acid concentration was successively increased from 3 . 13 mM , CsNEJs moved away from the mixed solutions of a cholic acid concentration lower than 25 mM , but reversed their migration toward the mixed solution of cholic acid concentration equal to or exceeding 25 mM ( Fig 5B ) . The mixed solution of cholic acid and lithocholic acid was chemoattractive to CsNEJs at a ratio greater than 25:1 .
Taken together , we found that CsNEJs had chemotaxis to bile , principally to cholic acid . In the mammalian host , upon excystation in the upper duodenum , CsNEJs sense the cholic acid and migrate chemotactically on the duodenal mucosal surface and enter the common bile duct . When encountered en route the concentrated gallbladder bile with strong chemorepellent lithocholic acid , the CsNEJs may swerve from it and find their way to the intrahepatic bile duct . | We previously reported that Clonorchis sinensis newly excysted juveniles ( CsNEJs ) were chemotactically attracted to bile . However , there is still a paucity of information regarding which components and what concentration of bile induce the chemotactic behavior . Here , we show , among various bile components tested , two have opposing chemotactic influences on the CsNEJs; cholic acid was characterized as a chemoattractant and lithocholic acid as a chemorepellent . Chemorepulsive migration was dependent on the concentration of lithocholic acid . Notably , the ratio ( 25:1 ) of cholic acid and lithocholic acid plays a critical role in defining chemotactic preferences of CsNEJs . We suspect that this bile acid ratio directs the parasites in the mammalian host , i . e . the high concentration of lithocholic acid in the gallbladder bile may repel CsNEJs from entering it . Bile in the hepatic bile duct has a chemoattractive level of cholic acid but a trace amount of lithocholic acid . These findings may explain why the CsNEJs preferentially migrate to the common and hepatic bile ducts rather than the gallbladder . Deeper understanding on the parasitism of the liver fluke is likely to have major implications for the studies on other parasites . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"biliary",
"system",
"invertebrates",
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"liver",
"body",
"fluids",
"helminths",
"gallbladder",
"bile",
"social",
"sciences",
"animals",
"trematodes",
"clonorchis",
"sinensis",
"chemotaxis",
"assay",
"animal",
... | 2018 | Bile acids drive chemotaxis of Clonorchis sinensis juveniles to the bile duct |
Drosophila melanogaster Piwi functions within the germline stem cells ( GSCs ) and the somatic niche to regulate GSC self-renewal and differentiation . How Piwi influences GSCs is largely unknown . We uncovered a genetic interaction between Piwi and c-Fos in the somatic niche that influences GSCs . c-Fos is a proto-oncogene that influences many cell and developmental processes . In wild-type ovarian cells , c-Fos is post-transcriptionally repressed by Piwi , which destabilized the c-Fos mRNA by promoting the processing of its 3′ untranslated region ( UTR ) into Piwi-interacting RNAs ( piRNAs ) . The c-Fos 3′ UTR was sufficient to trigger Piwi-dependent destabilization of a GFP reporter . Piwi represses c-Fos in the somatic niche to regulate GSC maintenance and differentiation and in the somatic follicle cells to affect somatic cell disorganization , tissue dysmorphogenesis , oocyte maturation arrest , and infertility .
Two major stem cell types are present in the Drosophila ovary: germline stem cells ( GSCs ) and somatic stem cells . Somatic stem cells differentiate into somatic follicle cells that provide structural support of the egg chamber . GSCs differentiate into germ cells , which become nurse cells or oocytes ( Fig 1A ) . The somatic niche ( or the GSC microenvironment ) promotes GSC maintenance via signaling factors such as dpp/BMP [1 , 2] . Piwi in GSCs and in the somatic niche promotes GSC maintenance and differentiation [3–7] . piwi mutant flies have no or markedly underdeveloped ovaries . Phenotypic studies of piwi mutant mosaic clones suggest that Piwi also affects oogenesis [3 , 4 , 8] . A genome-wide screen identified genetic interactors of Piwi [9] , and follow-up studies revealed that Piwi interacts with Corto and Polycomb Group proteins to regulate GSCs [10 , 11] . However , the molecular mechanism by which Piwi regulates GSC maintenance and differentiation is not well understood . Piwi associates with Piwi-interacting RNAs ( piRNAs ) , which are small ( 26–32 nt ) RNAs that preferentially contain uridine as the first residue [12] and possess a 2-O-methylation site at the 3′ end [13] . The biogenesis of piRNAs and their repression of transposon activities to safeguard germline genome integrity have been extensively studied [14–16] . Primary piRNAs are generated from long , single-stranded precursor RNAs and undergo amplification through the ping-pong pathway to generate secondary piRNAs [17 , 18] . Although the molecular factors and mechanisms that control the amplification of secondary piRNAs are well characterized , primary piRNA biogenesis mechanism is less understood . We followed-up on a previous genetic screening analysis [9] , and found that the proto-oncogene c-Fos is involved in Piwi-mediated regulation of GSCs . As part of the activator protein-1 complex , c-Fos regulates genes that control cell proliferation , differentiation , and survival [19–21] . We found that Piwi-mediated repression of c-Fos in the somatic niche regulates GSC maintenance and differentiation . Further , we reveal that the c-Fos mRNA serves as a piRNA precursor that is negatively regulated by Piwi , and this destabilization promotes somatic cell organization during ovarian tissue morphogenesis and animal fertility .
A genetic screen previously identified genomic regions whose heterozygous deficiency partially suppressed ovariole developmental defects in piwi mutant flies [9] . To follow up on this screen , we analyzed 31 fly lines with well-characterized genetic mutations located within these genomic regions . Each mutation was analyzed in flies with trans-heterozygous piwi mutant alleles 1 and 2 [3 , 22] at day 4 post eclosion to ensure approximate developmental equivalency . Further , wild-type and piwi mutant ovaries had similar germ cell to somatic cell ratios , as revealed by a comparison of Vasa ( germ cells ) and Tj ( somatic cells ) mRNA and protein levels ( S1A and S1B Fig ) . Small ( approximately less than 200 μm ) ovaries with few ovarioles , similar to piwi mutant in Fig 1Bii , were categorized as having ovary defects . Large ( approximately larger than 200 μm ) ovaries with greater than 10 ovarioles , similar to piwi; c-Fos/+ mutant in Fig 1Biii , were categorized as having partially suppressed ovariole defects . We identified 2 independent P-element insertions in c-Fos , EY01644 and EY08232 alleles , that partially suppressed the ovariole defects in piwi[1/2] mutants , homozygous piwi[1/1] , and heterozygous piwi[2/06839] mutants ( Figs 1B and 1C and S1C ) . It was previously shown that transgenic expression of c-Fos cDNA rescued the homozygous lethal phenotype of the EY01644 mutant allele [23] . We found that the homozygous lethality of the EY08232 allele is also rescued by transgenic expression of c-Fos cDNA ( S1D Fig ) . Thus , genetic reduction of c-Fos partially suppresses the ovariole defects in piwi mutant flies . Each Drosophila ovary is composed of 18–22 ovarioles , which are spatially organized to house germline development and maturation [24] . The germarium at the tip of each ovariole contains GSCs and somatic stem cells [25] . In wild-type flies , GSCs are defined by their apical position at the germarium , the cytoplasmic expression of Vasa , the localization of Hts to the spectrosome , ( a GSC-specific form of the fusome [24 , 26]; Fig 1C ) , and the ability to differentiate into germ cells in egg chambers . We used this functional definition to quantify the number of GSCs in wild type , piwi single and piwi;c-Fos/+ double mutant fly lines . We found that wild type flies contained 1–3 GSCs/germarium . Both of the piwi; c-Fos/+ double mutant lines displayed higher numbers of GSCs/germarium than the piwi mutant ( Fig 1D and 1E ) . These findings indicate that c-Fos mutations partially suppress GSC loss in piwi mutant ovaries . Next , we investigated the cell type ( s ) that underlie the suppressive effect of c-Fos mutations in the ovaries of piwi mutant flies . We used the Gal4/UAS system to drive cell type-specific expression of small hairpin RNAs ( shRNAs ) targeting piwi ( 22235 , 33724 ) and/or c-Fos ( II , Val10 , and III; S1E and S1F Fig ) [27–29] . Nos:Gal4 , Tj:Gal4 and C587:Gal4 drivers were used to induce shRNA expression in ovarian germ , somatic , and escort cells , respectively . We found that flies with reduced piwi expression in germ cells lacked ovarioles , as expected; however , this phenotype was not suppressed by loss of c-Fos ( S1G Fig ) . In contrast , flies with reduced piwi expression in ovarian somatic or escort cells displayed ovariole defects and reduction of c-Fos in the respective cells did partially suppress these ovariole defects ( Fig 1F and S1H Fig ) . These results indicate that Piwi interacts with c-Fos in the somatic niche to regulate germline development . To investigate whether c-Fos function in GSCs or germ cells affects fertility , we used Nos:Gal4 driving c-Fos shRNAs to deplete c-Fos specifically in the germ cells . We found that these flies laid significantly fewer eggs than control animals ( S2A Fig ) . This finding suggests that c-Fos in the GSCs and germ cells is required for normal fertility . We examined various cellular processes that are important for germ cells , such as meiotic double-stranded DNA break repair , ring canal structure , oocyte axis patterning , and detected no significant difference between flies depleted of c-Fos and wild-type controls ( S2B–S2I Fig ) . These results suggest that c-Fos is required for normal speed of germ cell maturation and egg production . Thus , c-Fos in the GSCs and germ cells is required for female fecundity . Germaria of the piwi mutants contained more spectrosomes than those of the wild type ( S3A Fig ) , consistent with previous findings [5 , 6] . Unexpectedly , the average numbers of germ cells containing spectrosomes per germarium were significantly higher in the piwi; c-Fos/+ double mutants than in the piwi mutant ( S3A and S3B Fig ) . This is likely caused by differentiation defects , which would result in little to no egg chamber formation . To determine whether this was the case , we quantified the number of egg chambers per ovariole . We found that the percentage of ovarioles with 3 or more egg chambers and the average number of egg chambers per ovariole were significantly higher in the piwi;c-Fos/+ double mutants than the piwi mutant ( S3C and S3D Fig ) . These findings indicate that germ cell differentiation is partially rescued by the c-Fos/+ mutations in the piwi mutant ovaries . The combination of increased undifferentiated germ cells with partially rescued egg chamber formation in the piwi;c-Fos/+ double mutants likely reflect that piwi;c-Fos/+ double mutants have two populations of germ cells: a population that continually proliferates but cannot differentiate , and a different population that continually proliferate and differentiate . Altogether these results suggest that c-Fos functionally interacts with Piwi to affect the maintenance and differentiation of GSCs . c-Fos is known to affect dpp signaling during embryogenesis [30] and in follicle cells of late-stage egg chambers [31] . Therefore , we examined whether c-Fos affects GSCs via dpp signaling . dpp/BMP signaling in the somatic niche induces Mad phosphorylation , which in turn represses the differentiation factor bag-of marbles ( bam ) to promote GSC maintenance [2 , 32 , 33] . We used immunofluorescence to quantify the number of phosphorylated Mad ( pMad ) -positive germ cells ( S4A Fig ) . The identification of GSCs in piwi mutants is complicated by the presence of some germ cells containing spectrosomes but not pMad ( S4A Fig ) . However , we found no significant difference in the number of pMad-positive germ cells between piwi and piwi;c-Fos/+ double mutant ovarian tissues ( S4B Fig ) . These findings were confirmed in flies with somatic cell-specific RNAi knockdown of piwi and/or c-Fos ( S4C Fig ) . Further , we found that mutations of Jun kinase , which phosphorylates c-Fos as part of JUNK signaling , did not suppress the ovariole defects in piwi mutants ( S4D Fig ) . These findings suggest that c-Fos does not affect Piwi-mediated regulation of GSC function via dpp/BMP or JNK signaling pathways . The finding that c-Fos reduction can partially suppress piwi mutant phenotypes suggests that Piwi represses c-Fos in ovaries . We collected c-Fos and Piwi expression data across 26 Drosophila tissue and cell types from FlyAtlas [34] . We noticed that the expression of c-Fos and Piwi were anti-correlative specifically in the ovary , where c-Fos was expressed at a significantly low level and Piwi at a significantly high level ( Fig 2A; p = 1 . 485 × 10−7 by the Grubbs test ) . We analyzed c-Fos expression by RT-qPCR using 2 different primer sets and found that the c-Fos mRNA level in piwi[1/2] mutant ovarian cells was significantly higher than that in wild-type ovaries ( Fig 2B ) . In contrast , c-Fos mRNA levels did not differ between piwi mutant and wild-type larval cells ( Fig 2B ) . To determine if a developmental stage difference between wild-type and piwi mutant ovaries might underlie altered c-Fos expression , we analyzed ovo mutant ovaries , which have similar defects to piwi mutant ovaries [35 , 36] . Although the ovo mutant ovaries displayed increased RPL40 and c-Fos mRNA levels ( S5A Fig ) , the increase in c-Fos mRNA observed in piwi mutant ovaries was higher and more specific ( RPL40 did not increase in piwi mutant ) . To determine if Piwi affects transcription of the c-Fos locus , we examined 3 replicate data sets of RNA polymerase II ChIP-seq in wild-type and piwi mutant ovarian cells [11] . We found that the piwi mutations did not affect RNA polymerase II binding to the c-Fos promoter , which was confirmed by ChIP-qPCR ( S5B Fig ) . These data suggest that Piwi does not affect transcription of c-Fos . Thus , data from FlyAtlas , RT-qPCR , and chromatin immunoprecipitation ( IP ) -qPCR suggest that c-Fos repression is Piwi-dependent , post-transcriptional , and specific to the ovary . We also evaluated c-Fos protein levels by immunofluorescence ( IF ) and Western blotting ( WB ) . The specificity of the antibodies used for IF and WB was confirmed by RNAi-mediated c-Fos depletion ( S1E and S1F Fig ) . c-Fos protein levels were higher in GSCs and the adjacent cystoblasts than in somatic and other germ cells , as shown by IF and confocal microscopy ( using the same imaging parameters for the wild type and the piwi mutant; Fig 2C ) . Piwi was present in all nuclei of the ovary , and c-Fos localized to both the cytoplasm and the nucleus ( Fig 2C and 2D ) . Further , c-Fos protein levels were high in all cells in the piwi mutant ovarian cells ( Fig 2E ) . Quantitation of the IF signals indicated that c-Fos was significantly increased , whereas Piwi was significantly decreased , in piwi mutant cells ( Fig 2F ) . Indeed , we observed a 2-fold increase in c-Fos protein levels in piwi mutant compared to the wild type ovaries ( Fig 2G ) . Together , these data suggest that Piwi represses c-Fos expression in in the ovarian somatic cells . We next examined whether piRNAs are involved in Piwi-mediated repression of c-Fos . We aligned published piRNA sequences p and found that 429 piRNA sequences uniquely mapped to the entire c-Fos locus . Of these 429 piRNA sequences , 135 mapped to the 3′ UTR ( Fig 3A ) . A binomial test to determine the significance of unique piRNA enrichment at the c-Fos 3′ UTR yielded a p-value of 2 . 2 × 10−16 , indicating significantly higher enrichment of piRNAs at the 3′ UTR than the rest of the locus . Sequences of the entire c-Fos 3′ UTR or the piRNAs do not exhibit homology to retrotransposon sequences . The published studies [12 , 37–39] and our study have all used size selection of approximately 26-30nt for sequencing small non-coding RNAs . This size selection excludes the presence of RNAs outside this size range; thus we cannot rule out a potential scenario of other non-coding RNAs originated from or targeting the c-Fos 3′ UTR . Nevertheless , we found that 126/135 piRNAs were in the sense orientation , consistent with primary piRNAs [12 , 17 , 40] . These data suggest that the c-Fos 3′ UTR is a primary piRNA precursor . Next , we validated the putative piRNAs unique to the c-Fos 3′ UTR . Because of the relative low abundance of these piRNAs ( S6A Fig ) , we used a stem-loop RT primer to amplify these piRNAs for RT-qPCR [41] . We designed TaqMan probes that are specific to these RT-PCR products and do not recognize the longer piRNA precursors ( S6B Fig ) . TaqMan assays were designed to detect 3 predicted piRNAs unique to the c-Fos 3′ UTR , termed piRNA1-3 ( S6C Fig ) . We detected expression of these piRNAs in fly ovaries , which was reduced upon depletion of c-Fos by RNAi ( S6D Fig ) , confirming specificity of the TaqMan assays . These data suggest that piRNAs are specifically generated from the c-Fos 3’UTR in fly ovaries . To determine if c-Fos piRNAs associate with Piwi , we performed immunoprecipitation ( IP ) experiments . We IP’d Piwi and IgG from Drosophila ovarian cell extracts and detected Piwi but not Aub ( the closest homolog of Piwi ) in Piwi IPs , confirming the specificity of the Piwi IP ( Fig 3B ) . We radioactively end-labeled RNAs that co-precipitated with Piwi and IgG , and observed an enrichment of small RNAs in the Piwi IP ( S6E Fig ) . We performed TaqMan RT-qPCR and found that c-Fos piRNAs 1–3 were enriched in the Piwi IP ( Fig 3C ) . In comparison , 2S rRNA enrichment in the Piwi IP is significantly lower ( Fig 3C ) . We also analyzed small RNAs purified from the ovaries of piwi mutants and wild-type flies and found significantly lower levels of piRNAs 1–3 in the three piwi mutant lines than in the wild type ( Fig 3D ) . The association of c-Fos piRNAs with Piwi and the requirement of Piwi for their biogenesis/stability provide additional biological support for these computationally identified piRNAs . Previous studies utilized an in vitro cell line , ovarian somatic cells ( OSC ) , to examine the effect of Piwi and piRNA biogenesis factors on OSC transcriptomes [42 , 43] . c-Fos FPKM levels from Sienski et al . and Ohtani et al . in OSCs are ( i ) significantly higher than that in wild type ovarian cells , ( ii ) unaffected by depletion of biogenesis factors Piwi , Armi , or Mael by siRNAs , and ( iii ) of similar c-Fos level in the piwi mutant ovarian cells from our data ( S7A Fig ) . Yet , piRNAs from c-Fos are detected in the OSCs . One explanation is that OSCs and the ovarian cells differ in genes involved in germline development: down-regulated genes in the ovarian cells are enriched in cell adhesion , motion , and morphogenesis , while upregulated genes are enriched in reproductive processes , game production , eggshell formation , oogenesis , and cytoplasm organization ( S7B and S7C Fig ) . Our findings by FlyAtlas gene expression profiling , RT-qPCR , IF , and WB support the conclusion that Piwi represses c-Fos in the ovarian cells . The molecular differences observed between Drosophila OSCs and ovarian cells ( S7B and S7C Fig ) suggest that Piwi requires a yet-identified mechanism/factor ( s ) present in the ovarian cells but absent in OSCs to mediate developmentally important gene regulation . To determine if the c-Fos 3’UTR is sufficient to repress gene expression in fly ovaries , we generated transgenic Drosophila expressing GFP reporters of the c-Fos 3ʹ UTR either by random site integration ( GFP-c-Fos-UTR-1 , -2 , -3 on the UASp vector ) or PhiC31-mediated integration into the 89E11 site ( GFP-ss-cFos-UTR in the plasmid pWALIUM-10 vector , http://www . flyrnai . org ) . We also obtained control GFP reporters of the K10 3’UTR ( GFP-K10UTR in the UASp vector ) or the Ftz intron ( GFP-Ftz-intron in the pWALIUM-10 vector; integration into the 89E11 site by PhiC31 ) . We used the Gal4/UAS system to drive expression of the GFP reporters in somatic ( by Tj:Gal4 ) cells . We found that the GFP-K10UTR and the GFP-Ftz-intron transgenes were more highly expressed than the GFP-c-Fos-UTR transgene in somatic cells , as determined by IF and WB ( Fig 4A–4C ) . In contrast , the GFP-c-Fos-UTR transgene was not repressed in larval cells ( Fig 4D ) . Reducing piwi expression in somatic cells by RNAi increased the expression of GFP-c-Fos-UTR in somatic cells by approximately 2-fold ( Fig 4E , S8A and S8B Fig ) ; piwi reduction by RNAi phenocopies piwi[1/2] mutant [5 , 6 , 44] . These results suggest that the c-Fos 3’UTR is sufficient to reduce gene expression in ovarian somatic cells , and that Piwi is required for this repression . Next , we examined whether Piwi protein interacts with the c-Fos transcripts . We found that the coding region and 3’UTR of c-Fos mRNA , but not rp49 mRNA , were enriched in a Piwi IP from wild-type ovarian cells compared with to IgG IP ( Fig 5A ) . rp49 is a ribosomal subunit and expressed in the same cell types as c-Fos . This enrichment was also observed by crosslinking followed by IP ( S8C Fig ) . To determine whether the c-Fos 3ʹ UTR is sufficient to recruit Piwi , we evaluated the enrichment of the GFP-Ftz-intron or GFP-c-Fos-UTR reporter mRNAs ( both driven by Tj:Gal4 ) in Piwi IPs from ovarian cells . We found that GFP-c-Fos-3’UTR mRNA and endogenous c-Fos mRNA , but not GFP-Ftz-intron mRNA or rp49 mRNA ( lacking the c-Fos 3’UTR ) , were enriched in Piwi IPs from ovarian cells ( Fig 5B–5D ) . Thus , the c-Fos 3’UTR is sufficient to recruit Piwi . If Piwi and the c-Fos 3’ UTR repress gene expression through the generation of primary piRNAs , then the GFP-c-Fos-UTR transgene would be predicted to increase the biogenesis of these specific piRNAs ( Fig 6A ) . We quantified the levels of c-Fos piRNAs 1–3 and found that they increase by 2- to 20-fold in GFP-c-Fos-3’UTR lines compared to control lines not expressing the transgene ( Fig 6B ) . We then purified small RNAs from ovarian cells of Tj:Gal4 , Tj:Gal4;GFP-K10UTR , or Tj:Gal4;GFP-c-Fos-3’UTR and performed RNA-seq . Computational filtering ( see Materials and Methods ) to identify piRNA sequences aligned to the c-Fos 3’UTR that were all in the sense orientation , had a median size of 26 nt , and contained the molecular signature of the first base being uridine in more than 70% of the piRNAs ( Fig 6C and S8D Fig ) . The c-Fos-specific piRNAs increased by approximately 9 fold ( 37 versus 4 fragments per kilobase of transcript per million mapped reads in controls ) and unique sense piRNA sequences increased by approximately 4–5 fold in in GFP-c-Fos-3’UTR ovarian tissue ( Fig 6C ) . We did not detect antisense piRNAs unique to the c-Fos-3’UTR , increase in miRNAs , or piRNAs aligned to the GFP coding sequence . Thus , the repression of c-Fos and GFP-c-Fos-3’UTR coincides with increased primary piRNA generation by Piwi . Given the developmental importance of c-Fos [19–21] and our finding that Piwi mediates c-Fos repression , we examined the functional consequences of ectopic c-Fos expression in the Drosophila ovary . To overexpress c-Fos , we replaced the 3’UTR in the c-Fos transgene with the K10 3’UTR to disrupt Piwi-mediated repression ( S9A Fig ) . We also generated control animals with UASp:c-Fos containing its own 3′ UTR and a c-Fos knockdown line ( c-Fos-K10 3’UTR; c-Fos shRNA-Val10 ) . We found that overexpression of the c-Fos-K10 3’UTR transgene in somatic stem cells and somatic follicle cells abolished egg production ( Fig 7A and S9A Fig ) . Further , egg production was restored in flies expressing c-Fos shRNA-Val10 to reduce c-Fos-K10 3’UTR in somatic stem cells and somatic follicle cells ( S9B and S9C Fig ) . This finding showed that repression of c-Fos in ovarian somatic cells , mediated by its 3′ UTR , is required for female fecundity . Overexpression of somatic c-Fos by Tj:Gal4 driving c-Fos-K10UTR resulted in enlarged ovarian tissues , longer ovarioles , and more egg chambers per ovariole ( Figs 7B and 7C and S9D ) . Vasa IF analysis showed that Vasa expression was persistent in all egg chambers with somatic c-Fos overexpression , but low in the mid- and late-stage egg chambers of control ( Tj:Gal4 ) ovaries ( Fig 7B ) . We observed defective egg chamber morphology in ovarioles overexpressing somatic c-Fos ( S9E Fig ) and rampant necrosis in late-stage germ cells ( S9F Fig ) . These findings suggest that c-Fos overexpression in ovarian somatic cells results in the arrest of oocyte maturation , retention of egg chambers in the ovarioles , necrosis of late-stage germ cells , and failure of egg production by the animal . Further analyses of germaria and egg chambers by Tj and Vasa IF staining revealed various cellular defects due to the overexpression of somatic c-Fos ( Fig 7D ) , such as abnormal cell organization in 57%–65% of germaria ( Fig 7E ) , excessive and disorganized somatic cells in 10% –32% of germaria ( Fig 7Di and 7F ) , and cyst accumulation ( >3 cysts ) in 44% of germaria ( Fig 7Dii and 7G ) . In 25%–33% of the egg chambers , c-Fos overexpression in somatic stem cells and somatic follicle cells led to disorganization , accumulation into multiple cell layers , and invasion into the germ cell compartment ( Fig 7Diii and 7H ) . These findings indicate that c-Fos regulation is required for somatic stem cell and somatic cell organization for ovarian tissue morphogenesis . Quantitation of the S phase ( IF of PCNA ) and mitosis ( IF of phosphorylated serine 10 in histone H3 ) revealed no differences between control cells and cells overexpressing c-Fos ( S9G Fig ) . Therefore , overexpression of c-Fos does not increase cell proliferation in ovarian somatic stem cells and somatic cells . Our findings suggest that an important function of Piwi in the Drosophila ovary is to repress c-Fos in the somatic niche and somatic ovarian cells , and that animals with piwi loss of function and c-Fos overexpression share similar phenotypes , including somatic cell disorganization . Although we observed GSC loss or differentiation defects in piwi mutant flies but not c-Fos-overexpressing flies , this difference is likely due to the presence of Piwi in c-Fos-overexpressing ovaries . To examine the potential molecular similarities between animals with piwi loss and c-Fos overexpression , we compared the transcriptomes of the piwi[1/2] mutant and c-Fos overexpressing ovaries . In the gene expression profiling by RNA-seq , we found the mean FPKM values of c-Fos to be 12 . 6 in w[1118] , 34 . 2 in piwi[1/2] , and 70 . 1 in c-Fos-K10UTR ( overexpressing transgenic c-Fos ) ovarian cells ( summarized in S7A ) . Remarkably , more than 65% of differentially expressed genes ( compared to the wild type ) were the same in the piwi mutant and c-Fos overexpressing ovaries ( Fig 8A–8C ) . Genes upregulated by c-Fos overexpression or piwi loss were enriched in the functional categories of actin cytoskeleton organization , morphogenesis , development , and cell motility categories , whereas downregulated genes were enriched in microtubule cytoskeleton organization , cell cycle , cell division , mitosis , DNA replication , and chromosome organization ( Fig 8D ) . We thus propose that Piwi regulates these processes by repressing c-Fos in the ovarian somatic cells ( Fig 8E ) to promote cell organization and tissue morphogenesis .
Piwi functions in both the somatic niche and in GSCs to maintain GSCs , but its underlying mechanisms are not well understood [4 , 45] . We found that an important function of piwi in the Drosophila ovary development is to repress c-Fos . Piwi-mediated repression of c-Fos in somatic stem cells and somatic follicle cells was required for somatic cell organization and ovarian tissue morphogenesis . Our data suggest that the 3′ UTR of the c-Fos mRNA recruits Piwi , which regulates the activities of as yet identified nucleases to generate primary piRNAs from the c-Fos 3’UTR , leading to destabilization and post-transcriptional repression of c-Fos ( Fig 9 ) . Unclear aspects of the proposed model ( Fig 9 ) include the mechanism by which the c-Fos 3’UTR recruits Piwi protein , identity of the nucleases involved in generating the primary piRNAs , and the extent by which the mRNA degradation machinery is involved . In GSCs , c-Fos expression was comparatively high and important for fertility , suggesting that a Piwi-independent mechanism regulates c-Fos . Relatively little is known about how Piwi protein targets non-transposon mRNAs . Two recent genomic studies uncovered that piRNAs and the mouse Piwi protein MIWI cause instability of a subset of mRNAs in the mouse testes [46 , 47] . Moreover , MIWI-mediated targeting of mRNAs and long noncoding RNAs depends on retrotransposon sequences and occurs in the cytoplasm [46] . Our study showed that the piRNAs need not be of retrotranspon origin ( none of the piRNAs from the c-Fos 3’UTR are homologous to retrotransposon sequences ) and that this gene regulation functionally impacts germ cell development and animal fertility , thereby contributing to the understanding of gene regulation by Piwi and piRNAs . Open questions include how Piwi and piRNAs target individual mRNAs , and whether a direct mechanism links Piwi-piRNAs and the mRNA degradation machinery to mediate gene repression . The regulation of c-Fos and potentially other genes by Piwi-dependent processing into piRNAs in the ovary supports the concept that modest gene regulation is important during developmental events . Piwi and piRNAs repress the expression of c-Fos ( a proto-oncogene with a pervasive role in development and disease ) by an average of 2-fold . This modest repression is similar to that seen in the dosage compensation of sex chromosomes [48] or gene modulation by miRNAs [49] . However , the deregulation of these molecular processes can have severe , and often lethal , consequences on the developing organism . It is reasonable to propose that modest gene regulation by various molecular processes offers flexible modes of gene expression and potentially accommodates the many dynamic cellular events occurring during development . Unexpectedly , c-Fos overexpression in the somatic niche did not significantly affect GSCs . This milder phenotype is likely a consequence of the nonoverlapping functions of Piwi and c-Fos in the somatic niche , and the regulation of additional molecular events by Piwi ( e . g . , dpp/BMP signaling ) besides inhibiting c-Fos to affect GSC functions . Therefore , c-Fos appears to be a part of an extensive Piwi-centric network that safeguards GSC functions . Our study uncovers a novel mechanism involving Piwi and c-Fos that regulates somatic cell organization for tissue morphogenesis of the Drosophila ovary . piwi reduction in the inner sheath cells or escort cells of ovaries is known to trigger somatic cell disorganization in the ovarioles [6] . This phenotype had not been studied in detail , likely because the tissue dysmorphogenesis phenotypes are masked by GSC loss and differentiation defects that occur in animals with mutations of piwi or factors in primary piRNA biogenesis [3 , 4 , 50] . Further , phenotypic analyses of piwi mutant mosaic clones in late-stage egg chambers revealed no observable defects [3 , 4] . This finding indicates that piwi inactivation does not affect somatic follicle cells . However , the aforementioned somatic clonal analysis was carried out in differentiated follicle cells and not somatic stem cells , because piwi inactivation leads to loss of somatic stem cells . Our study circumvented this technical hurdle to uncover a function for Piwi in oogenesis . Another intriguing finding was that c-Fos overexpression in somatic stem cells and somatic follicle cells was sufficient to result in persistent Vasa expression and arrest in egg chamber maturation ( Fig 5B ) . This finding suggests that either c-Fos repression in the somatic cells is required for normal germ cell maturation or that c-Fos overexpression disrupts a yet-unidentified soma-to-germ cell signaling event that is required for normal germ cell maturation . Thus , the somatic cell organization mediated by c-Fos is likely not only important for tissue morphogenesis but also critical for ensuring germ cell maturation . Non-transposon gene regulation by Piwi and piRNAs is not well-understood , possibly because only a few of these gene targets have been characterized , which are Tj [51] , Nanos in embryonic axis determination [52] , and Masc in sex determination of the silkworm [53] . Although piRNAs are generated from many genic transcripts in Drosophila ovaries , this often does not lead to repression of the genic transcripts . Our study is only the beginning of a more comprehensive effort to uncover non-transposon gene regulatory functions of Piwi and piRNAs to affect germ cell development . Future discoveries of other piRNA precursors repressed by Piwi and piRNAs in the germ cells and detailed mechanistic studies would be necessary to determine whether this c-Fos regulatory mechanism by Piwi and piRNAs is a broader post-transcriptional gene regulatory process .
Adult Drosophila flies at day 4 post-eclosion were used for all genetic assays to ensure approximately developmental equivalency . Germ cell to somatic cell ratios were similar between wild type and piwi[1/2] mutants , as shown by similar Vasa and Tj levels ( S1A and S1B Fig ) . Wild type is w[1118] . Most strains are from BDSC: piwi[1]/CyO ( #43319 ) , piwi[2]/CyO ( #43637 ) , piwi[06843]/CyO ( deletion of scar and piwi; #12225 ) [5] , Df ( 2L ) BSC145 ( deletion of chr2:32C1 that includes the piwi locus; #9505 ) , c-Fos[EY01644]/TM3 ( #15077 ) , c-Fos [EY08232]/TM3 ( #16882 ) , Tj:Gal4 ( #50105 ) Nos:Gal4 ( #25751 ) , UAS:GFP RNAi ( #35786 ) , UAS:shc-Fos ( pVALIUM10; #27722 ) , and UAS:shPiwi ( pVALIUM20; #33724 & #34866 ) . UAS:shPiwi-22235 is from VDRC . UAS:c-Fos RNAi ( II ) is from D . Bohmann [27] . UAS:c-Fos ( II ) and ( III ) contain the K10 UTR and are from P . Emery [54] . Antibody names , IF dilutions ( unless otherwise stated ) , and sources are as follows: Mouse 1B1 anti-Hts , 1:50 , DSHB; rabbit anti-pMad , 1:300 , Abcam ab52903; rabbit anti-c-Fos [55] , 1:50 , S . Subhabrata; guinea pig anti-Piwi , 1:200 for IF and 6ug/IP , Peng lab; mouse anti-Piwi [56] , 1:100 for WB , M . Siomi; mouse anti-Aub [56] , 1:200 , M . Siomi; rabbit anti-c-Fos , 1:1000 for WB , Abnova PAB8948; rabbit anti-GFP , 1:500 and 1:2000 for WB , Life Technologies A011122; chicken anti-GFP , 1:250 and 1:2000 for WB , Life Technologies A10262; guinea pig anti-Tj , 1:250 , Peng lab; rabbit anti-Vasa , 1:200 , Santa Cruz sc-30210; Alexa dye–conjugated donkey secondary antibodies , 1:500 , Jackson ImmunoResearch . PBS is 137mM NaCl , 2 . 7mM KCl , 10mM Na2HPO4 , 1 . 8mM KH2PO4 pH 7 . 4 . HEPM is 25 mM HEPES pH 7 . 9 , 10 mM EGTA , 60 mM PIPES , 2 mM MgCl2 . Buffer D is 300mM KCl , 20mM HEPES pH 7 . 9 , 0 . 2mM EDTA , 0 . 1% TritonX-100 , 25% glycerol , 1x protease inhibitors ( Roche , 11873580001 ) , 1mM DTT . RNA was purified with the GeneJET RNA purification kit ( Thermo Scientific , K0732 ) . The cDNA was generated from 200 ng of RNA by using the High-Capacity cDNA RT kit with random hexamer or oligo dT ( Applied Biosystems , 4374966 ) . To distinguish sense from antisense transcripts , gene-specific reverse transcription was performed using Superscript IV Reverse Transcriptase ( Thermo Scientific , 18090050 ) . qPCR in the iQ SYBR Green Supermix ( Bio-Rad , 170–8880 ) was analyzed on a Bio-Rad CFX96 system . RT reactions were performed in triplicate for quantitation . rp49 and rpl40 are used for normalization because they are ribosomal subunits with ubiquitous expression . S1 Table lists the primer sequences . Dissected ovaries were homogenized , extracted in buffer A with 0 . 1% Triton-X-100 for 4 minutes on ice , and centrifuged to obtain the supernatant as the cytoplasmic fraction . The nuclear pellet was washed once in buffer A and then extracted in buffer D to obtain the nuclear fraction . Equal amounts of protein extracts ( in buffer D ) were separated by SDS-PAGE and transferred onto a nitrocellulose membrane ( 162–0115 , Bio-Rad ) . Membranes were blocked by 2% bovine serum albumin ( BSA ) in HEPM , incubated in primary antibodies ( diluted in 1% BSA , HEPM 0 . 1% Triton X-100 ) overnight at 4°C , washed in PBS 0 . 1% Triton X-100 , incubated in IRdye-conjugated secondary antibodies , and imaged on an Odyssey Fc system ( LI-COR ) . Signals were quantitated with the Image Studio software ( LI-COR ) . The Student’s t test was used for statistical analyses . Drosophila ovaries were fixed in 3% paraformaldehyde in PBS , permeabilized in PBS with 0 . 3% Triton X-100 overnight at 4°C , blocked with 2% normal donkey serum in HEPM , probed with primary antibodies in HEPM and 0 . 05% Triton X-100 overnight at 4°C , washed with PBST , probed with secondary antibodies , washed with PBST , stained with DAPI , washed with PBST , and mounted in ProLong Gold Antifade Mountant ( Life Technologies , P36930 ) . The egg chamber and spectrosome were quantitated on a Zeiss Axio Imager . M2 . GSCs were quantitated on a Nikon C2 . Images were acquired with a Zeiss LSM780 , a Leica TCS SP5 , or a Zeiss LightSheet Z . 1 . For IF quantitation , images were acquired with the same parameters and analyzed by the Zen Black software ( Zeiss ) to obtain signal density , which is the background-subtracted average intensity per μm2 . The Student’s t test was used for statistical analyses . Ovarian cytoplasmic fractions were extracted and discarded . Nuclear pellets were fixed , washed , and sonicated in the lysis buffer by using the Bioruptor Pico ( Diagenode ) . Equal amounts of chromatin were added to Dynabeads ( Life Technologies ) prebound with 4ug of IgG or Piwi antibodies . After overnight incubation , beads were washed and immunoprecipitates were eluted . Purified DNAs from the input and eluates were analyzed by qPCR . S2 Table lists the primer sequences . Published piRNA sequences [12 , 37–39] were downloaded from Gene Expression Omnibus ( GEO; GSM154618 , GSM154620 , GSM154621 , GSM154622 , GSE9138 , GSE13081 , and GSE26507 ) to generate 2 . 2 million unique and 4 . 2 million multi-aligned libraries . For piRNAs from ovarian somatic cell lines , data [57] were downloaded from GEO ( GSM1119289 ) and mapped to the BDGP R5/dm3 reference genome by using GSNAP [58] . 6 . 7 million reads were uniquely mapped . For in-house piRNA analysis , small RNAs were isolated by the mirVana miRNA isolation kit ( Life Technologies , AM1561 ) and separated by PAGE gel . 50 ng of the gel-extracted RNAs ( 20–30 nt ) was used to construct libraries with the TruSeq small RNA prep kit ( Illumina , RS-200-0012 ) . Libraries were sequenced on a Hi-Seq 2500 ( Illumina ) . The adaptor-trimmed sequencing reads were aligned to the BDGP R5/dm3 reference genome by using GSNAP [58] and filtered by size and non-coding RNA type ( tRNA , snoRNA , snRNA , rRNA , pre-miRNA , and miRNA ) . For miRNA quantification , reads were aligned to the miRBase hairpin precursors . The binomial test was used to compare piRNA enrichment at the c-Fos 3′ UTR against the rest of the c-Fos locus . The one-sided Student’s t test was performed to compare unique piRNAs . Seq data from this study were deposited in GEO by the identifier GSE69722 . For analysis of small RNA data from Handler et al . , data were downloaded from GEO ( GSM1119289 ) and mapped to the BDGP R5/dm3 reference genome using GSNAP [58] . 6 . 7 million reads were uniquely mapped . Among them , 4751 were mapped to c-Fos ( 3R:25591717–25619835 ) , 3660 of which mapped to the 3′ UTR ( 3R:25618747–25619835 ) . Drosophila ovaries were dissected and homogenized , and small RNAs were isolated by the mirVana miRNA isolation kit ( Life Technologies , AM1561 ) . smRNAs ( 5 ng each ) were used in RT reactions with the TaqMan MicroRNA Reverse-Transcription Kit ( Life Technologies , 4366596 ) , and RT reactions were analyzed by the TaqMan 2S rRNA assay ( Life Technologies , 4427975 ) and the custom smRNA assays 1–4 ( Life Technologies , assays ID CS1RULS , CS20SR0 , CS39QX8 , and CSS07ER ) by using TaqMan Universal MM II ( Life Technologies , 4440043 ) . To understand the unique interaction of piwi and c-Fos in ovary germline , we compared gene expression profiles between ovary germline and OSCs . PolyA-selected RNAseq data of OSC from Ohtani et al ( 2013 ) were downloaded from GEO ( GSE47006 ) and mapped to the BDGP R5/dm3 reference genome using STAR [59] , which was used in mapping of in-house generated RNAseq data . Gene expression values were estimated with Cufflinks [60] , and compared with Cufflinks-generated expression of in-house data . Unsupervised hierarchical clustering analysis was done using all the genes that expressed ( FPKM>1 ) in at least one samples . Differentially expressed genes were selected with p-value of less than 10−5 and fold change of greater than 4 . Gene Ontology analysis was done using DAVID [61] . The biological difference between OSC and germline can be confounded by difference in data generation , however , gene ontology analysis of differentially expressed genes points to developmentally meaningful processes , thus indicating that the effect of biological differences is much stronger than the batch effect . The 3′ UTR was cloned into the pPGW vector ( 1077 , Drosophila Genomics Resource Center ) by Gateway ( Life Technologies ) . Site-specific integration transgenic constructs were assembled by Gibson assembly and recombined into the pWALIUM10roe vector ( TRiP , Harvard Medical School ) by Gateway . Cesium chloride-prepped DNAs were sent to BestGene , Inc . , for injection into w[1118] or Bloomington stock 9744 ( integration site at 89E11 ) embryos to generate transgenic lines . | The Drosophila melanogaster ovary is consisted of germ cells differentiated from GSCs and ovarian somatic cells that provide structural support to the organ . Piwi is a ribonucleoprotein required for GSC maintenance and differentiation in the Drosophila ovary . Piwi does so by influencing GSC-autonomous mechanisms or the somatic niche signaling . The somatic niche is composed of cells adjacent to the GSC that signal to promote GSC functions . Piwi and piRNAs are also known to repress transposons . We found that Piwi genetically interacts with c-Fos in the somatic niche and represses c-Fos expression in the ovarian somatic cells . Piwi destabilizes the c-Fos mRNA by mediating the generation of piRNAs from its 3′ UTR . Piwi represses c-Fos in the somatic niche to promote GSC functions and in the somatic ovarian cells to influence cell organization , tissue morphogenesis , and egg production . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"3'",
"utr",
"messenger",
"rna",
"animals",
"cell",
"differentiation",
"germ",
"cells",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms"... | 2016 | c-Fos Repression by Piwi Regulates Drosophila Ovarian Germline Formation and Tissue Morphogenesis |
Dengue is the commonest arboviral disease of humans . An early and accurate diagnosis of dengue can support clinical management , surveillance and disease control and is central to achieving the World Health Organisation target of a 50% reduction in dengue case mortality by 2020 . 5729 children with fever of <72hrs duration were enrolled into this multicenter prospective study in southern Vietnam between 2010-2012 . A composite of gold standard diagnostic tests identified 1692 dengue cases . Using statistical methods , a novel Early Dengue Classifier ( EDC ) was developed that used patient age , white blood cell count and platelet count to discriminate dengue cases from non-dengue cases . The EDC had a sensitivity of 74 . 8% ( 95%CI: 73 . 0-76 . 8% ) and specificity of 76 . 3% ( 95%CI: 75 . 2-77 . 6% ) for the diagnosis of dengue . As an adjunctive test alongside NS1 rapid testing , sensitivity of the composite test was 91 . 6% ( 95%CI: 90 . 4-92 . 9% ) . We demonstrate that the early diagnosis of dengue can be enhanced beyond the current standard of care using a simple evidence-based algorithm . The results should support patient management and clinical trials of specific therapies .
Dengue is an acute , systemic viral infection and a public health problem in the tropical world [1] . The etiological agents of dengue are any of the four dengue viruses ( DENV-1-4 ) . In endemic countries it is common for all four DENV serotypes to co-circulate . Late-stage trials of a dengue vaccine with intermediate efficacy have recently been reported , offering hope of a public health intervention [2 , 3] . The World Health Organisation ( WHO ) has a stated goal of reducing global dengue mortality by 50% by 2020 [1] . Improvements in case diagnosis and management will be central to achieving this aim . Significant loss of intravascular plasma volume leading to hypovolemic shock ( dengue shock syndrome ( DSS ) ) , usually between the 4th-6th day of illness , is the commonest life-threatening complication of dengue [1 , 4] . It’s widely held that the case-incidence of DSS can be reduced via careful monitoring and the judicious use of parenteral fluids to maintain an adequate intravascular volume [1] . Ideally , this case management approach is enabled because the attending physician had made an early diagnosis and thus alerted clinicians , nurses and family caregivers to the signs and symptoms suggestive of clinical worsening . Additional benefits of an early diagnosis include support to community level public health interventions and improvements in the sensitivity of case surveillance systems and disease burden estimates . Furthermore , it is likely that the therapeutic window of opportunity for a dengue antiviral drug lies in the first 48–72 hours of illness [5] . Thus , programmatic use of therapeutic interventions in the future will likely go hand in hand with strategies for early diagnosis . Yet there are numerous challenges for busy primary care clinicians in making a diagnosis of dengue in the first few days of illness . Rapid lateral flow tests , based on the detection of the viral NS1 antigen , are available in some settings and can provide a confirmatory diagnosis [6–8] . The diagnostic performance of the WHO dengue case definition , which relies on non-specific signs and symptoms that overlap with other infectious diseases , is unknown in the first few days of illness [1] . Potts et al concluded that more prospective studies were needed to construct a valid and generalizable algorithm to guide the differential diagnosis of dengue in endemic countries [9] . To this end , several prospective studies have described the creation of classifiers for the diagnosis of dengue [10–12] . However none of these studies have exclusively focused on pediatric fever cases presenting to primary care facilities with short illness histories , a very common scenario in dengue endemic settings . Against this backdrop , the purpose of the current study was to prospectively derive a dengue diagnostic algorithm from routinely collected clinical and laboratory findings in pediatric patients with <72 hours of illness history and compare this approach against the diagnostic performance of a leading NS1 rapid test ( BioRad NS1 Ag STRIP ) in the same patients . The results provide pragmatic methods to enhance the early diagnosis of dengue in primary care settings .
The study protocol was approved by the Hospital for Tropical Diseases scientific and ethical committee and the Oxford University Tropical Research Ethical Committee ( OXTREC 35–10 ) . The accompanying parent/guardian of each child provided written informed consent . Recruitment occurred in the public sector outpatient departments of Children’s Hospital No . 1 ( HCMC ) , Children’s Hospital No . 2 ( HCMC ) , The Hospital for Tropical Diseases ( HCMC ) , Tien Giang Provincial Hospital , Dong Nai Children’s Hospital , Binh Duong Provincial Hospital and Long An Provincial Hospital . These outpatient departments function as primary care providers to their local communities . A patient presenting to one of the study sites was eligible for enrolment if they met the following inclusion criteria—a ) fever at presentation ( or history of fever ) and less than 72 hours of symptom history , b ) in the attending physicians opinion dengue was a possible diagnosis , c ) 1–15 years of age inclusive , d ) accompanying family member or guardian had a mobile phone and e ) written informed consent for the child to participate was provided by the parent/guardian . Patients were excluded if- a ) the attending physician believed they were unlikely to be able to attend follow-up or b ) the attending physician believed another ( non-dengue ) diagnosis was more likely . Patient enrolment occurred consecutively during normal clinical hours on weekdays without restriction . All patients were enrolled into the study before the attending physician received the results of any routine laboratory tests . At the time of enrolment , information on the patient’s age , sex , illness history , presenting signs and symptoms were recorded in a case report form . The definitions used to support standardized data capture are shown in S1 Table . Blood samples were drawn for routine hematology , biochemistry and NS1 rapid test . All NS1 rapid tests ( NS1 Ag STRIP , BioRad ) were performed on the same day of patient enrolment by one of two trained laboratory technicians at the Hospital for Tropical Diseases . Routine hematology results , but not biochemistry or NS1 rapid test results , were made available to the attending physician , who decided on the management approach , i . e . hospitalization or ambulatory follow-up . A 2nd blood sample for the purposes of serology was collected around the time of defervescence from all patients that were hospitalized anytime during their acute illness . If the patient was managed solely on an ambulatory basis for the duration of their illness , then a 2nd early convalescence blood sample for the purposes of serology was collected only from a randomly selected 10% of this patient population . The randomization code to select ambulatory cases for follow-up was generated by software . All clinical and laboratory data were stored in an ICH-GCP compliant , clinical data management platform called “CLIRES” . Demographic and clinical data were double entered . Electronic data files containing hematological results were uploaded directly to CLIRES . Independent study monitoring was performed by the Clinical Trials Unit of the Oxford University Clinical Research Unit which examined adherence to the trial procedures , data collection and recording and compliance with ICH-GCP . The gold standard diagnostic result was a composite derived from three tests; RT-PCR , IgM serology and NS1 detection by ELISA . First , all enrolment plasma samples were tested with a validated , quantitative RT-PCR assay to detect DENV RNA [13] . Next , any enrolment plasma samples that were negative in the RT-PCR assay were tested using the Platelia Dengue NS1 Ag ELISA assay ( BioRad ) and scored according to the manufacturer's instructions . Samples with equivalent results were repeated and if still equivocal they were scored as negative . Next , IgM ELISA serology ( Panbio , Brisbane , Australia ) was performed according to the manufacturer's instructions for patients who had paired plasma samples ( enrolment and early convalescence ) and who were negative in both the DENV RT-PCR assay and Platelia Dengue NS1 ELISA . Any patient who was—a ) DENV RT-PCR positive , b ) NS1 ELISA positive , or c ) had DENV IgM seroconversion in paired plasma samples , was classified as a laboratory-confirmed dengue case . IgM seroconversion was defined as a change in the MAC ELISA test result from negative to positive in paired plasma samples with the 2nd sample collected 6 or more days after illness onset and >2 days after the 1st sample . Any patient who was DENV RT-PCR negative , NS1 ELISA negative and did not IgM seroconvert in paired plasma samples was classified as “not dengue” . Any patient who was DENV RT-PCR negative and NS1 ELISA negative at the time of enrolment , but did not have paired samples available for serology , was classified as a “presumptive not-dengue” case . For analysis , data from “not dengue” and “presumptive not-dengue” cases were pooled . Plasma samples were enriched for proteins with molecular weight >100kDa using Amicon filtration units ( Millipore ) . Briefly , 200μl of plasma was concentrated to ~30μl and then tested in the Platelia NS1 ELISA . All concentrated samples were tested in parallel with an aliquot of the original plasma samples and the filtrate ( containing proteins with molecular weight <100kDa ) . Logistic regression was used for the development of the diagnostic algorithm . A detailed assessment of the model assumptions of linearity and additivity was performed ( S1 Text ) . All pre-defined candidate predictors listed in S1 Table and significant interaction terms were included in the full model . The model was then simplified using step-wise backwards selection using Akaike’s Information Criterion ( AIC ) and stability selection [14] . Alternative statistical models such as classification and regression trees ( CART ) and random forests ( RF ) were also investigated in order to find an optimal diagnostic algorithm [15 , 16] . The performance of the model was assessed with respect to discrimination ( receiver operating characteristic curves ( ROCs ) and area under the ROC curve ( AUC ) ) , calibration ( calibration plots and calibration intercepts and slopes ) , and standard accuracy criteria of binary diagnostic tests ( sensitivity , specificity , negative and positive predictive values ) . We selected the cut-off point to classify a patient as dengue positive at a predicted risk of dengue of ≥33 . 3% , corresponding to assuming that the “cost” of missing a true dengue patient is twice as large as the cost of a false-positive [17] . To avoid over-optimistic estimates of model accuracy and performance due to model derivation and evaluation on the same dataset , all accuracy measures were corrected for optimism by validation . Validation was performed for the whole model development process including variable selection . Two validation schemes were employed to mimic external validation: The final logistic model was also presented as a nomogram for direct clinical use . All statistical analyses were performed using the statistical software R v3 . 1 . 1 ( R foundation for statistical computing , Vienna , Austria ) and its companion packages c060 version 0 . 2–3 ( for stability selection ) , randomForest version 4 . 6–7 ( for random forest ) and rpart version 4 . 1–8 ( for CART ) .
5729 children with fever of less than 72 hours were enrolled at one of the seven clinical study sites in southern Vietnam between October 2010 and December 2012 . A summary of the patient screening , enrolment and diagnostic outcomes is shown in S1 Fig A total of 5707 patients were included in the analyses . 1692 ( 29 . 6% ) participants had laboratory-confirmed dengue . The baseline characteristics of the dengue and non-dengue cases are shown in Table 1 . Notably , dengue cases were older than non-dengue cases . All four DENV serotypes were detected; DENV-1 was the commonest serotype , followed by DENV-4 , -2 and -3 . Enrolment plasma samples ( n = 5707 ) were tested for the presence of NS1 by NS1 Ag Strip test in a blinded , real-time fashion . Against the composite gold-standard reference diagnostic result , the NS1 Ag Strip test had a sensitivity of 70 . 4% ( 95%CI: 68 . 2–72 . 6% ) , specificity of 99 . 2% ( 95%CI: 98 . 9–99 . 5% ) , positive predictive value ( PPV ) of 97 . 4% ( 95%CI: 96 . 3–98 . 2% ) , and negative predictive value ( NPV ) of 88 . 9% ( 95%CI: 87 . 9–89 . 8% ) for the diagnosis of dengue ( Table 2 ) . There was a striking difference in diagnostic performance by serotype , with NS1 detection being less sensitive in DENV-2 infections irrespective of the serological response ( primary vs secondary ) ( S2 Table ) . The detection of NS1 was strongly associated with the concentration of DENV RNA in the same plasma sample; the odds of NS1 detection increased by 1 . 8 ( 95%CI: 1 . 6–1 . 9 ) for each 10-fold higher DENV RNA concentration ( Table 2 ) . These data define the strengths and weaknesses of NS1 rapid testing; it is highly specific but is compromised by suboptimal sensitivity , especially for DENV-2 cases . Volume enrichment of the plasma molecular weight fraction containing multimeric NS1 ( >100 , 000kDa ) was performed on plasma samples from 21 viremic dengue cases enrolled in this study . However , despite 5-10-fold concentration of plasma , this processing failed to materially improve the diagnostic yield , with only 1 of 11 samples changing their status from negative ( original sample ) to positive ( concentrated sample ) in the Platelia NS1 ELISA ( S3 Table ) . Multivariate logistic regression analyses of clinical , demographic and laboratory data from 5707 patients were performed to generate a practical dengue diagnostic classifier that could replace or augment NS1-based diagnosis in the first 72 hours of illness . The most parsimonious model , derived from stability selection , used the patient’s age , white cell count and platelet count at the time of enrolment to classify dengue from non-dengue cases ( Table 3 ) . Alternative approaches to feature selection yielded models with only slightly higher performance but relied on many more ( more than ten ) variables ( S4 Table ) . The most parsimonious model , herein called the Early Dengue Classifier ( EDC ) , had a sensitivity of 74 . 8% ( 95%CI: 73 . 0–76 . 8% ) , specificity of 76 . 3% ( 95%CI: 75 . 2–77 . 6% ) , positive predictive value of 57 . 1% ( 95%CI: 56 . 2–59 . 0% ) , and negative predictive value of 87 . 8% ( 95%CI: 86 . 8–88 . 5% ) for correctly classifying dengue cases in the entire dataset at the pre-defined cut-off of 33 . 3% . Of note , this pre-defined cut-off reflecting clinical priorities was very close to the cut-off corresponding to the point on the ROC curve closest to the upper left corner ( perfect model ) , which was 34 . 2% ( Fig 1A ) . The area under the ROC curve ( AUC ) was 0 . 829 ( Fig 1B ) and the predicted risk of dengue agreed well with the observed risk ( Fig 1C ) . The EDC had sensitivity of 72 . 9% ( 95% CI: 69 . 6–76 . 6% ) for DENV1 , 74 . 7% ( 95%CI: 71 . 0–79 . 7% ) for DENV2 , 68 . 4% ( 95%CI: 59 . 2–74 . 5% ) for DENV3 and 78 . 2% ( 95%CI: 75 . 5–83 . 3% ) for DENV4 infection . The overall performance characteristics of the EDC under temporal , leave-one-site-out validation or seasonality ( rainy versus dry season ) , are summarized in S5 Table . These results suggest that , in settings where NS1 rapid tests are not routinely available , the EDC could assist primary care physicians in dengue diagnosis . In settings where NS1 rapid tests are routinely used , the EDC can be combined with the NS1 rapid test as a composite test ( classified as positive when either NS1 rapid test or EDC are positive , and classified as negative when both NS1 rapid test and EDC are negative ) . This composite test had sensitivity of 91 . 6% ( 95%CI: 90 . 4–92 . 9% ) while the specificity was 75 . 7% ( 95%CI: 74 . 5–77 . 0% ) . Corresponding positive and negative predictive values were 61 . 7% ( 95%CI: 60 . 6–63 . 1% ) and 95 . 5% ( 95%CI: 94 . 9–96 . 1% ) . If a higher specificity was desired , a higher cut-off value of the EDC could be used for the combined test instead , e . g . a cut-off of 50% would lead to a sensitivity of 86 . 0% ( 95%CI: 84 . 5–87 . 6% ) and specificity of 89 . 6% ( 95%CI: 88 . 7–90 . 5% ) . These results imply that the EDC is useful in settings with and without NS1 rapid testing . Fig 2 presents a nomogram of the EDC . The nomogram assigns points to all risk factors and translates the total point score to a predicted risk for dengue . For example , a 9-year-old patient with platelet count 100x103/mm3 , and white blood cell count 5x103/mm3 has a total points score of 15+32+84 = 131 , and the corresponding risk of dengue is about 70% . The predicted risk of dengue is larger than 33 . 3% so the patient would be classified as dengue positive . Alternatively , the EDC could be implemented as a smartphone app . The exact formula for the estimated risk of dengue ( p ) is given by the following logistic equation: logit ( p ) = 1 . 236 + 0 . 139*age ( in years ) – 0 . 254*white blood cell ( in 103/mm3 ) – 0 . 006 *platelet ( in 103/mm3 ) .
The early and accurate diagnosis of dengue on the grounds of clinical signs and symptoms alone is problematic [9] . Physicians need better tools to assist in early diagnosis if the WHO ambition of a 50% reduction in global dengue mortality is to be achieved by 2020 . This study characterized the performance of three diagnostic approaches; the NS1 rapid test , a stand-alone diagnostic classifier and the combination of NS1 rapid test and diagnostic classifier together . Our results highlight the utility of NS1 rapid tests for an early specific diagnosis , yet also remind that 2nd generation tests are needed with improved sensitivity . The diagnostic classifier described here could help guide diagnosis in endemic settings , or be used as an adjunct to help exclude dengue in patients returning a negative NS1 rapid test result . There is a body of literature describing the performance of NS1 rapid tests for the diagnosis of dengue [6–8 , 19–22] . This current study extends that literature in several ways . First , by virtue of the large sample size we demonstrate with high precision the differential sensitivity of the NS1 Ag STRIP for different DENV serotypes . This test was sensitive ( between 75–85% ) for DENV-1 , -3 and -4 infections , but poorly sensitive in DENV-2 infections ( 46 . 4% ) . Lower sensitivity was partially attributable to the great majority of DENV-2 infections being associated with secondary serological responses , although we note sensitivity was also low in primary DENV-2 infections . This suggests that there are particular virological ( e . g . lower viral burdens in vivo ) or intrinsic aspects of the NS1 test , that limit DENV-2 NS1 detection . [23–26] . Second , we make the novel observation that 5–10 fold enrichment of proteins with molecular weight >100kDa in plasma specimens ( the NS1 hexamer has predicted molecular weight of 310kDa [27] ) did not lead to improved NS1 detection rates . These data suggest that dengue patients who return negative NS1 rapid test results in the first 3 days of illness have free plasma NS1 concentrations substantially below the limit of sensitivity of existing assays and that 2nd generation tests might need to be at least an order of magnitude more sensitive . Nonetheless , better NS1 rapid diagnostic tests are needed if they are going to be widely adopted by clinical services in primary care settings . In malaria , HRP2 rapid diagnostic tests for Plasmodium falciparum infection are an example of how improvements to assay performance can lead to recognition as a diagnostic standard of care [28] . Finally , although serum NS1 concentrations have been proposed to have prognostic value in a small study , this is yet to be independently validated and is likely to be difficult given that blood NS1 concentrations vary widely according to the infecting DENV serotype , serological response and day of illness [8 , 24 , 29 , 30] . Previous studies have described clinical and/or routine laboratory findings that distinguish patients with dengue from those with other febrile illnesses [12 , 31–37] . What is striking in the literature is that only three prospective studies have considered dengue diagnostic algorithms exclusively in children and of these the largest contained 1227 patients , of who 614 had dengue [11 , 38 , 39] . More generally , most diagnostic studies failed to report positive and negative predictive values for their diagnostic algorithms , thus making it difficult to assess their utility in routine practice . Against this backdrop , a strength of the current study is the large sample size , the presence of all four DENV serotypes , robust statistical validation techniques and transparent performance characteristics . The clinical signs and symptoms that make up the WHO case definition for dengue were not used in the final , parsimonious diagnostic EDC classifier . Instead , we found that only three variables—patient age , white blood cell count and platelet count , provided similar discriminatory information as alternative models that relied upon a much larger set of clinical data . The purpose of this study was to explore whether it was possible to develop any kind of simple , evidence-based algorithm for early diagnosis—the results demonstrate this feasibility , albeit the performance characteristics of the end-result algorithm are not so outstanding that they will result in widespread adoption or change the practice of experienced clinicians . We concur with Potts et al in the belief that diagnostic rules for dengue are not a replacement for good clinical acumen and management [9] . Nonetheless , the EDC described here offers an evidence-based guide that can likely improve the prevailing diagnostic accuracy of most Vietnamese physicians working in primary care who do not possess extensive experience in dengue diagnosis and management . In particular , in settings where NS1 rapid tests are not routinely available or affordable , or where DENV-2 is the most prevalent virus in circulation , the EDC could help guide clinicians in making their differential diagnosis . An early diagnosis of dengue can assist in patient triage and management by directing clinical/caregiver attention to clinical warning signs and/or the appearance of capillary permeability , for which supportive oral and/or parenteral fluid therapy is recommended in order to prevent circulatory compromise . Additionally , in the first days of illness many dengue cases are infectious to Aedes aegypti mosquitoes and hence an early diagnosis could support measures to prevent further transmission , e . g . by use of topical repellents and local mosquito control [40] . Our study has several design features and limitations that might preclude its wider generalizability . The EDC relies on routine hematology findings that are commonly accessible in primary care settings in Vietnam but might not be available everywhere . By design , our study focused on patients with <72 hours of illness and hence our results might not be applicable to patients who present to medical care at later time-points . By using the age of the patient as a component of the EDC , it’s likely that the EDC would not perform well in settings where the burden of dengue falls on age-groups different from that in southern Vietnam . Nonetheless , this study has delivered the largest population-based and quantitative framework to guide early diagnosis of pediatric dengue . Further prospective validation in Vietnam and other endemic countries with similar epidemiology will be needed to establish the clinical utility of the EDC . | Dengue is a very common acute infectious disease in the tropical world . Health care professionals are able to better care for dengue patients if they can make an early diagnosis and make a plan for case management . This current study investigated fever in 5729 children in Vietnam with 3 days or less of fever and identified 1692 dengue cases using advanced , gold standard methods . We systematically collected a range of medical and laboratory findings on each patient when they entered the study and used statistical tools to determine if these medical and laboratory findings could enable early diagnosis , independent of sophisticated , gold-standard laboratory tests . Our results , called the Early Dengue Classifier , had performance characteristics suggesting it could improve the diagnostic proficiency of health care professionals . However the performance of the Early Dengue Classifier is not perfect and likely will not change the practice of experienced doctors in dengue endemic settings . Our study highlights the need for 2nd generation , easy-to-use rapid diagnostic tests that can accurately diagnose dengue in the first few days of fever . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Sensitivity and Specificity of a Novel Classifier for the Early Diagnosis of Dengue |
N6-methyladenosine ( m6A ) is the most abundant methylation , existing in >25% of human mRNAs . Exciting recent discoveries indicate the close involvement of m6A in regulating many different aspects of mRNA metabolism and diseases like cancer . However , our current knowledge about how m6A levels are controlled and whether and how regulation of m6A levels of a specific gene can play a role in cancer and other diseases is mostly elusive . We propose in this paper a computational scheme for predicting m6A-regulated genes and m6A-associated disease , which includes Deep-m6A , the first model for detecting condition-specific m6A sites from MeRIP-Seq data with a single base resolution using deep learning and Hot-m6A , a new network-based pipeline that prioritizes functional significant m6A genes and its associated diseases using the Protein-Protein Interaction ( PPI ) and gene-disease heterogeneous networks . We applied Deep-m6A and this pipeline to 75 MeRIP-seq human samples , which produced a compact set of 709 functionally significant m6A-regulated genes and nine functionally enriched subnetworks . The functional enrichment analysis of these genes and networks reveal that m6A targets key genes of many critical biological processes including transcription , cell organization and transport , and cell proliferation and cancer-related pathways such as Wnt pathway . The m6A-associated disease analysis prioritized five significantly associated diseases including leukemia and renal cell carcinoma . These results demonstrate the power of our proposed computational scheme and provide new leads for understanding m6A regulatory functions and its roles in diseases .
N6-methyl-adenosine ( m6A ) methylation is a paradigm-shifting research filled with exciting discoveries . Recent research has shown that m6A exists in > 25% of mRNAs in mammalian cells [1 , 2] and forms an important regulatory circuitry that controls many aspects of RNA metabolism [3–10] . Evidence of m6A’s involvement in cancer and other diseases [11–21] and its role in regulating viral life cycle [22–25] are also accumulating . However , our current knowledge about how m6A levels are regulated and whether and how regulation of m6A levels of a specific gene can play a role in cancer and other diseases is largely elusive . The purpose of this study is to conduct a comprehensive prediction of m6A mediated functions and associated diseases through global analysis of m6A regulated genes using 75 human methylated RNA immunoprecipitation sequencing ( MeRIP-seq ) [1 , 2] samples curated by MeT-DB2 [26] . To this end , prediction of context-specific m6A sites is an essential first step . Several informatics tools have been developed to predict condition independent m6A sites from RNA sequences [27–35] or condition-specific m6A peaks in MeRIP-seq data [36 , 37] . Chen et al . [35] proposed the first sequence-based model iRNA-Methyl to predict m6A site using features extracted from RNA sequences . Subsequently , Zhou et al . developed SRAMP [29] to improve the performance of predicting single-base m6A sites using three kinds of features extracted from pri- and mature RNA sequences . Since then , a line of sequence-based algorithms [38–40] , all based on different handcrafted features extracted from RNA sequences have been developed . As an alternative to handcrafted RNA sequence features , Wei et al . [33] proposed DeepM6APred to extract features automatically using a deep belief network ( DBN ) . However , because RNA sequences are independent of any study conditions , none of these sequence-based models can predict condition-specific m6A sites . Because m6A has been shown to play different regulatory roles in different cell conditions and disease types , its methylation status is highly dynamic in nature , being either methylated or demethylated depending on the biological contexts . Also , m6A status can be experimentally manipulated through silencing or overexpressing key m6A related proteins , a commonly used approach for studying m6A functions . Therefore , predicting condition-specific single-base m6A site is an essential task in m6A research . Currently , algorithms including exomePeak [41] and MeTPeak [42] are proposed to predict context-specific m6A peaks from MeRIP-seq data . However , MeRIP-Seq has a limited resolution of ~100bp and its large biological and technical variations often result in high false positive rates in the predicted peaks . No existing algorithm can predict condition-specific m6A site at a single-base resolution . In addition to a lack of single-base , context-specific site prediction algorithms , computational prediction of m6A functions has not been adequately addressed . In our previous work [43] , we developed m6A-Driver , a network-based approach to identify m6A driven genes with significant functions under a specific context . However , m6A-Driver has several limitations . First , m6A-Driver to identify m6A driven genes in two different conditions , e . g . , gene knock-down vs . normal . Therefore , m6A-Driver cannot be used for the intended global analysis that includes samples from multiple conditions . Second , because of the small sample size , m6A-Driver tends to identify a large number of significant genes , which makes it difficult to prioritize these genes . In this work , we propose to develop a new approach to address these limitations of the site or peak detection algorithms and m6A-Driver so that a global prediction of m6A mediated functions and associated diseases across samples from multiple conditions can be reliably performed . To improve the resolution and accuracy of condition-specific m6A site prediction from MeRIP-seq , we propose a novel Convolutional Neural Network ( CNN ) [44] based method named Deep-m6A to detect m6A sites from MeRIP-Seq peak regions with a single base resolution . This model integrates mRNA sequence information with MeRIP-seq data and trained on single-base m6A sites identified by the miCLIP [45] , the state-of-the-art high throughput technology for single-base detection of m6A sites . Deep-m6A makes it possible to identify single-base m6A sites from the 75 human MeRIP-seq samples with higher accuracy and resolution , providing us a chance to investigate the relationship of m6A methylation with gene expression and diseases in a global manner . To this end , we propose Hot-m6A , a new network-based pipeline . We hypothesize that m6A regulates biological processes and pathways by regulating a set of functionally interacted genes through regulating their mRNA stability . Therefore , the expression of a gene regulated by an m6A site is likely to be correlated with the m6A level across multiple conditions and this regulation also influences its neighboring genes in the PPI network . The stronger an m6A regulates a gene , the more significant the m6A-expression correlation and the higher the degree of influence on its neighbors . Our goal is to identify m6A regulated genes , whose expressions are correlated with their m6A methylations across 75 samples and which are closely interacting with other m6A genes in a PPI network . To achieve this goal , we adopt the HotNet2 [46] algorithm . The basic idea of Hot-m6A is to diffuse “heat” , i . e . , correlation of gene expression and m6A level in this study , of genes in a network and select the significant genes , where the “heats” they diffused to each other are high . Therefore , the m6A regulated genes identified by HotNet2 will in general have relatively high expression-methylation correlation and closely interacted with each other in the PPI network . However , HotNet2 still can also identify m6A regulated genes with lower “heat” but closely interacted with “hot” genes . After m6A regulated genes are identified , our pipeline prioritizes the m6A associated diseases by applying the random walk with restart on heterogeneous network ( RWRH ) method [47] on a gene-disease heterogeneous network . We applied Deep-m6A and Hot-m6A to the 75 MeRIP-seq samples , which produced a compact set of 709 functional significant m6A-regulated genes and nine functional enriched subnetworks . The functional enrichment analysis of these genes and networks reveal that m6A targets key genes of many critical biological processes including transcription , cell organization and transport , cell proliferation and cancer-related pathways such as the Wnt pathway . The m6A-associated disease analysis prioritized five significantly associated diseases including leukemia and renal cell carcinoma .
The flowchart of our pipeline is illustrated in Fig 1 . We intend to perform a global analysis of existing human m6A site in different samples to uncover m6A regulated functions and associated disease . To achieve this , we first set out to determine the single-base m6A sites . To this end , we developed a novel CNN model , called Deep-m6A , which takes both sequence feature and MeRIP-Seq IP reads count as an input to predict context specific single-base m6A sites from MeRIP-Seq data . We then applied Deep-m6A to predict single-base m6A sites for all 75 human MeRIP-Seq samples extracted from MeT-DB V2 . 0 [26] . Then , we extracted the m6A sites appeared in at least 12 samples according to a test based on Fisher’s z transformation [48] and calculated the revised Fisher’s z-transformation [49] of the Pearson correlation of methylation degree and gene expression level across all their appeared samples ( see Methods and material for details ) . Genes that contain at least one m6A site that occurred in more than 12 samples and whose methylation degree and gene expression level are significantly correlated were defined as candidate m6A regulated genes . We selected the largest absolute z-transformed correlation from all m6A sites in a gene to denote the methylation-expression correlation of the gene . To further study the m6A regulated functions and select functional m6A-regulated genes , we developed Hot-m6A and applied it to 4 different PPI networks , taking the absolute z-transformed correlations for all candidate m6A-regulated genes as the heat vector , to identify significant PPI subnetworks that are regulated by m6A more than expected by chance . The 4 PPI networks includes BioGRID [50] , HINT+HI2012 [51 , 52] , MultiNet [53] , and iRefIndex [54] . The last three networks are used in the HotNet2 paper . The edges that are identified by all the four networks were extracted to form significant consensus subnetworks , which were then extended to include edges identified in 3 , 2 , and 1network , respectively ( See Methods for the detail ) . All the genes in the extended subnetworks are defined as significant m6A-regulated genes and genes reported in the significant consensus networks are defined as consensus m6A-regulated genes . Finally , we took all these significant m6A-regulated genes as gene seeds and their correlated diseases according the OMIM as disease seeds and applied RWRH on a heterogeneous gene-disease network . We also built four heterogeneous gene-disease networks corresponding to the four PPI networks . For each heterogeneous gene-disease network , the top 10 ranked diseases were selected as candidate m6A-associated disease . A random network test was applied to calculate an empirical p-value for each candidate m6A-associated disease . The candidate disease with a p-value < 0 . 05 was selected as significant candidate m6A-associated disease . The significant candidate m6A-associated disease that identified by all the 4 networks were finally identified as m6A-associated disease . We first evaluated the performance of Deep-m6A using the training data from the HEK293 cell lines and compared its performance with Deep-m6A-S . As described in Methods , Deep-m6A-S and Deep-m6A use the same CNN structure but different input features . Deep-m6A-S takes 101-nt long sequences centered at a DRACH motif as input , whereas Deep-m6A takes both sequence and the corresponding IP reads count as input . As detailed in the “Dataset” section , this training data include 4 , 742 positive samples that are CITS miCLIP m6A sites in the MeRIP-seq peak regions and also centered at DRACH motif , and 33 , 718 negative samples that are also centered at a DRACH motif in the MeRIP-seq peak regions but are > 50-nt away from the positive samples and not CIMS miCLIP sites or single-base m6A sites reported in other experiments [55] . Before model comparison , we investigated the difference between MeRIP-seq IP reads coverage in the sequence regions of the positive and negative samples . As shown in Fig 2 ( A ) , the average reads count of positive samples are higher and more centered at the DRACH m6A motif than those of the negative samples . The CNN architecture of Deep-m6A and Deep-m6A-S includes one convolutional layer followed by a max-pooling layer , a fully connected layer and an output layer . We used 10-fold cross validation to evaluate the performance of Deep-m6A and Deep-m6A-S and took one of the 10 CVs to optimize the hyperparameters . To reduce the impact of significant imbalance between the positive and negative samples , we split the negative samples into 7 subsets , each of which has the equal size to positive samples and for each CV , we trained 7 models for each balanced pair of positive/negative samples and used the average predicted score of the 7 models as the final predicted probability of a test DRACH motif to be a single base m6A methylation site . The CNN architecture was determined by a grid search method [56] , where the searched parameters are the kernel size: 4x3 , 4x4 and 4x5; # of kernels: 16 , 32 and 48; the max-pooling size: 1x3 , 1x4 and 1x5; # of the nodes of the fully connected layer: 8 , 12 , 16 and 32; the fully connected layer dropout rate: 0 . 2 , 0 . 25 , 0 . 3 , and 0 . 5; and the softmax output layer dropout rate: 0 . 2 , 0 . 25 , 0 . 3 and 0 . 5 . Table 1 showed the optimized hyperparameters by grid search . The categorical cross entropy loss function and Adadelta [57] optimizer were adopted in the training . Fig 2 ( B ) and 2 ( C ) shows the receiver operating characteristic ( ROC ) curves and the precision-recall ( PR ) curves of the CV test . We have trained seven CNN models to balance the scale of the positive and negative training samples and the predicted probability of a site is an average predicted probably of these seven models . We can see Deep-m6A achieved a higher area under the ROC curve ( AUC = 0 . 898 ) and area under the PR curve ( PRAUC = 0 . 554 ) than Deep-m6A-S model ( AUC = 0 . 884 , PRAUC = 0 . 489 ) . The AUC and PRAUC of Deep-m6A are 1 . 4% and 5 . 6% higher than Deep-m6A-S , respectively . There is especially a higher improvement in precision . This suggests that including IP reads help lower the false positive rate , especially in the higher ranked predictions . Having a higher precision is particularly important for subsequent biological validation and functional study because in practice attention is most likely given to a limited top ranked predictions . To further validate the performance of Deep-m6A , we applied it along with Deep-m6A-S on the independent MOLM13 dataset and compared their performance with SRAMP [29] . As described in the “Dataset” section , this dataset includes 726 positive and 6 , 577 negative samples . Deep-m6A and Deep-m6A-S were trained on HEK293 dataset as described in the previous section . For SRAMP , we downloaded the SRAMP tool from their webserver ( http://www . cuilab . cn/sramp/ ) and applied it to the 101-nt RNA ( without intron ) and DNA ( with intron ) sequences of each sample . The RNA sequences are used by the SRAMP mature RNA model and the DNA sequences are used by SRAMP full DNA model . When feeding these sequences into the SRAMP tools , they output probabilities of all DRACH motifs contained in the sequence . We only picked the results for the centered motif of each sample as the prediction probabilities for SRAMP . Fig 2 ( D ) and 2 ( E ) shows the ROC curve and PR curve results of these models . First , among the three sequence-based models , Deep-m6A-S reports about 1% improvement in AUC and 2% in PRAUC over the two SRAMP models . This suggests that the CNN model can capture additional discriminate sequence features than SRAMP . Second , Deep-m6A outperforms all three sequence-based prediction models in both AUC and PRAUC . The improvement in precision is even more pronounced ( ~6% over Deep-m6A-S ) . This once again speaks the benefit of including IP reads in the prediction . The relatively large number of human MeRIP-Seq data of different cells and tissues under different conditions curated by MeT-DB2 gives us a chance to investigate the relationship of m6A methylation with gene expression and diseases in a global manner . As the first step of this global investigation , we applied Deep-m6A to the 75 human MeRIP-seq samples to identify candidate m6A-regulated genes . A positive site was predicted when the prediction probability calculated by Deep-m6A is greater than 0 . 907; this threshold is chosen because at this threshold the 10-fold CV test in training can achieve a precision of 0 . 7 . This resulted in 23 , 456 single-base m6A sites in all 75 samples . Fig 3 ( A ) shows the frequency of a site being predicted in the 75 samples . As shown , ~23% of these sites are sample specific , i . e . they only appear in one sample . In contrast , ~30% sites appear in more than 12 data samples . To conduct a global analysis across all the 75 samples , we extracted sites that appear in more than 12 samples . The distributions of these sites on mRNA as well as all predicted m6A sites were tally using Guitar R package [58] and shown in Fig 3 ( B ) . As illustrated , for all the predicted sites , the distribution tends to be enriched in 3’ UTR and 5’ UTR , whereas the sites appeared in more than 12 samples are more enriched around stop codon and in 3’ UTR . Because 3’UTR contain binding sites of miRNA and RNA binding proteins such as HuR that are known to post-transcriptionally regulate gene expression , this distribution may indicate that the sites that appeared in > = 12 samples may potentially be involved in regulating gene expression . Next , we selected the genes which harbor m6A sites that appear in > = 12 samples as candidate m6A-regulated genes . In total , 3 , 670 m6A-regulated genes that contain totally 7 , 090 m6A sites were extracted . To study the relationship of their expression and m6A methylation , we calculated the Pearson’s correlation of methylation degree and gene expression level for all the candidate m6A-regulated genes and then converted it to a revised Fisher’s z score as described in “Methods and material” section . We then set out to identify functional significant m6A-regulated genes , which are defined as candidate genes whose expression levels are influenced by the m6A methylation more than expected by chance and which are closely interacting with each other in PPI networks . We applied Hot-m6A ( See Methods and material for detailed algorithms ) to the candidate m6A-regulated genes , taking their absolute revised Fisher’s z score as the heat vector and performing heat diffusion in the PPI network to identify functional interacted genes with relative high correlation . 49 consensus m6A-regulated genes were identified across all four PPI networks using Hot-m6A and 709 m6A-regulated genes were identified in the extended network ( S1 Fig; see Methods and material for detail ) . Most of the 709 m6A-regulated genes ( 681 or ~96% ) are in the biggest subnetwork . We then compared the distributions of the absolute revised correlation z scores of the 49 consensus m6A-regulated genes , 709 m6A-regulated genes , and remaining non-significant candidate genes ( Fig 3 ( C ) ) . We can see that most of the m6A-regulated genes have larger correlation than other nonsignificant candidate genes and the consensus m6A-regulated genes tend to the largest correlations among the three groups . This result speaks for the ability of our pipeline to identify likely m6A regulated genes . Notice there are some m6A-regulated genes with low correlations; they are identified by Hot-m6A because they interacted tightly with “hot” genes with high correlation in the PPI network . To further detect the function module from these 709 m6A-regulated genes , we removed edges that are identified only in one PPI network , and obtained 22 subnetworks including 113 genes in total . We further isolated the 9 largest functional significant subnetworks that have at least 3 genes ( S2 Fig ) . We further examined the functional enrichment for all the 709 significant genes using DAVID [59] ( Fig 3 ( D ) ) . Among the twelve enriched GO BPs ( PBenjamini < 0 . 05 ) , seven are directly related to transcription including transcription from RNA polymerase III promoter ( 10 genes , PBenjamini = 4 . 78×10−3 ) , positive regulation of DNA-templated transcription ( 44 genes , PBenjamini = 5 . 79×10−3 ) , transcription , DNA-templated ( 116 genes , PBenjamini = 1 . 67×10−2 ) , tRNA transcription from RNA polymerase III promoter ( 5 genes , PBenjamini = 1 . 62×10−2 ) , 5S class rRNA transcription from RNA polymerase III type 1 promoter ( 5 genes , PBenjamini = 1 . 62×10−2 ) , negative regulation of transcription from RNA polymerase II promoter ( 52 genes , PBenjamini = 2 . 92×10−2 ) and negative regulation of transcription , DNA-templated ( 39 genes , PBenjamini = 4 . 82×10−2 ) . Indeed , the m6A is shown to involve in every stage of RNA metabolism including transcription . It is shown to regulate mRNA stability and speculated to also regulate mRNA splicing [60] . Also , we also see that m6A genes could regulate cell organization and transport as they are enriched in vesicle-mediated transport ( 25 genes , PBenjamini = 3 . 83×10−5 ) , cell−cell adhesion ( 32 genes , PBenjamini = 3 . 61×10−4 ) and extracellular matrix organization ( 22 genes , PBenjamini = 2 . 04×10−2 ) are also enriched . We also noticed that that the Wnt signaling pathway is enriched ( 10 genes , PBenjamini = 1 . 26×10−2 ) . Wnt is an important pathway widely involved in cancer and cell development [61–63] . The potential involvement of m6A in cancer is reinforced by the enriched proteoglycans in cancer KEEG pathway ( 22 genes , PBenjamini = 2 . 43×10−2 ) . This is not surprising as there is increasing evidence demonstrating its regulatory roles in different cancer [20 , 64–70] . Another enriched KEGG pathways are Protein processing in the endoplasmic reticulum ( 20 genes , PBenjamini = 2 . 37×10−2 ) and Lysosome pathway ( 18 genes , PBenjamini = 6 . 31×10−3 ) , both of which are protein processing pathways . These predictions are corroborated by our knowledge of the m6A’s role in regulating translational efficiency [71 , 72] . We next performed GO and KEGG pathway enrichment analysis to the nine largest significant subnetworks ( S3 and S4 Figs ) . These enriched BP and pathways present reasonably clear and consistent interpretations of these subnetworks . Subnetwork A is mostly involved in protein processing ( 6 genes enriched in Protein processing in endoplasmic reticulum , 6 genes enriched in ER-associated ubiquitin-dependent protein catabolic process , 3 genes enriched in retrograde protein transport , ER to cytosol and 4 gene enriched in protein stabilization ) and potentially regulate mRNA stability via SMG9 , UPF1 , SMG8 and SMG1 genes enriched in nuclear-transcribed mRNA catabolic process , nonsense-mediated decay BP term . Subnetwork B is predicted to mostly involve in Notch signaling pathway via NOTCH2 , MAML1 and MAML3 genes . Subnetwork C is closely related to cell motility , proliferation and survival through enrichment of m6A-regulated genes including VEGFB , PDGFB , VEGFA , COL5A1 , NRP1 , SLIT2 and SPARC in pathways like Focal adhesion ( PBenjamini = 1 . 75×10−2 ) , positive regulation of endothelial cell proliferation ( PBenjamini = 1 . 56×10−3 ) , cell migration involved in sprouting angiogenesis ( PBenjamini = 1 . 26×10−3 ) and positive regulation of endothelial cell migration ( PBenjamini = 1 . 26×10−3 ) . It is also involved in neuronal development related pathways including Axon guidance ( NRP1 , PLXNA1 , SEMA3F and SLIT2 genes with PBenjamini = 1 . 30×10−3 ) , branchiomotor neuron axon guidance ( NRP1 , PLXNA1 and SEMA3F genes with PBenjamini = 8 . 54×10−3 ) , semaphorin-plexin signaling pathway involved in axon guidance ( NRP1 , PLXNA1 and SEMA3F genes with PBenjamini = 1 . 19×10−3 ) and axon extension involved in axon guidance ( NRP1 , SEMA3F and SLIT2 genes with PBenjamini = 1 . 19×10−3 ) . The function of subnetwork D is mostly related to transcription including regulation of transcription from RNA polymerase II promoter ( MED26 , MED18 , MED9 , MED11 and MED21 genes with PBenjamini = 4 . 16×10−4 ) , transcription , DNA-templated ( MED29 , MED26 , POLR2I , MED9 , MED11 and MED21 genes with PBenjamini = 4 . 92×10−3 ) and transcription initiation from RNA polymerase II promoter ( MED26 , POLR2I and MED13 genes with PBenjamini = 1 . 48×10−2 ) . Subnetwork E’s function is defined through mostly the enrichment Wnt signaling pathway ( 5 genes enriched including FZD8 , DKK1 , LRP6 , FZD5 , LRP5 with PBenjamini = 6 . 75×10−6 ) . For subnetwork F , 3 m6A-regulated genes are STK11 , WDR6 and STRADA and they are involved in cell cycle and cell proliferation related pathways including cell cycle arrest , mTOR signaling pathway and AMPK signaling pathway with PBenjamini < 0 . 05 . Subnetwork G serves a role in antigen processing ( 3 genes , TAP2 , HLA-E , TAPBP enriched in Antigen processing and presentation , antigen processing and presentation of peptide antigen via MHC class I and antigen processing and presentation of endogenous peptide antigen via MHC class I with PBenjamini < 0 . 05 ) . Subnetwork H includes ITGAV , ITGB5 and CYR61 as m6A regulated genes , which are involved in cell adhesion and subnetwork I are involved in MAPK signaling pathway via gene MAP3K4 , GADD45B and MAP2K7 . Taken together , these results suggest that m6A-regulated genes are enriched in significant biological process and pathways that can influence RNA transcription , cell motility , cell proliferation , cell survival and cell death , and therefore are likely involved in caner and cancer related pathways . Moreover , m6A regulated genes tend to be in the upstream of these enriched pathways ( S5 Fig ) . We next investigated the association of m6A with diseases . We first searched the OMIM database for diseases related to m6A-regulated genes , where we found 308 phenotypes associated with 177 m6A-regulated genes according to OMIM phenotype-disease relationship record ( S1 File ) . Most of the selected phenotypes are associated with only one m6A-regulated gene . To prioritize these diseases to identify significant m6A-associated diseases , we mapped the 709 m6A-regulated genes to each of the four PPI networks and the 308 phenotypes to the disease network and constructed four gene-disease networks . We then applied RWRH to each of the four networks ( See “Methods and material” for the detailed algorithm ) , taking the regulated genes and their correlated diseases as seed nodes . Then , we selected the top 10 ranked diseases as candidate m6A-associated diseases and calculated an empirical p-value to assess if a disease association is selected by chance . Five significant m6A-associated diseases with p-value < 0 . 05 and reported by all the four networks were finally identified as the m6A-associated diseases ( Table 2 ) . As shown in Table 2 , acute myeloid leukemia is prioritized as the most significant m6A associated disease , where m6A-regulated genes NSD1 , PICALM , and ABL2 are OMIM-annotated disease genes , suggesting that they may be potential m6A-associated biomarkers in leukemia . Several lines of evidence have shown the direct involvement of m6A in regulating leukemia [68–70] . The second significant disease is type 2 diabetes mellitus . Yang et al . has established this association and reported that glucose is involved in the dynamic regulation of m6A in patients with type 2 diabetes [73] . Our prediction identified the m6A genes HNF1B , HNF1A , WFS1 , and IRS2 as the potential disease genes , which could provide a new clue to study the role of m6A in type 2 diabetes . Also , m6A is predicted to have a role in renal cell carcinoma . This is also corroborated by Xiao et al . [74] , which reported that the m6A methyltransferase METTL3 acted as a tumor suppressor in renal cell carcinoma and publications in [75 , 76] , which identified YTHDF2 , an m6A binding protein , to be involved in renal cell carcinoma . Taken together , we found existing evidence to support 3 out of 5 predicted association diseases . These results demonstrate the power of the proposed network-based analysis and the RWRH algorithm in identifying m6A associated diseases .
The accumulation of a large number of MeRIP-Seq samples from different cells and tissues under different conditions gives us a chance to analyze m6A-regulated genes and m6A-associated functions in a global manner . However , existing informatics tools for predicting m6A sites from MeRIP-seq are hampered by high false positive rates and low resolutions . To address this issue , we developed Deep-m6A , the first CNN model that predicts single-base m6A sites in MeRIP-Seq peak regions by integrating mRNA sequence features with MeRIP-Seq IP reads . Test results from 10-fold CV on training HEK293 data and an independent OMLM13 dataset showed that Deep-m6A outperformed sequence-based algorithms including Deep-m6A-S and SRAMP in both precision and sensitivity . Although the miCLIP technology has been proposed to profile transcriptome-wide m6A at a single-base resolution , its adoption is still very limited because of its more complex protocol . Therefore , MeRIP-seq will continue to serve as the go-to high throughput technology for global m6A profiling in the near future . Give the ability of Deep-m6A to provide MeRIP-seq a single-base detection resolution , we expect Deep-m6A to be an important tool in m6A research . Currently , Deep-m6A is trained to detect sites that reside on DRACH motifs . Even though this requirement of containing motifs helps reduce the false positive predictions , it also sacrifices the prediction sensitivity and will inevitably miss the positive sites that do not have any motifs . Therefore , further improvement of Deep-m6A in the future to be able to detect sites without motifs will provide additional value for Deep-m6A . In our scheme , to further reduce the false positive rate and prioritize functional significant m6A-regulated genes , we examined the correlation between expression and m6A methylation of m6A-regulated genes and assessed their function significance using Hot-m6A implemented on the PPI network . We applied Deep-m6A and Hot-m6A to 75 human MeRIP-seq data , which resulted in a compact collection of 709 m6A-regualted genes and several interacting subnetworks of m6A-regualted genes . Functional analysis revealed that these genes are mainly involved in transcription and Wnt pathway ( Fig 2 ( D ) ) . Wnt pathway is one of the key cascades regulating cell migration and cell development , and is also tightly associated with cancer such as glioblastoma ( GBM ) . Even though direct involvement of m6A in Wnt pathway has not been reported , YTHDF2 , an m6A reader , has been shown to suppress cancer cell migration by inhibiting EMT in an m6A dependent manner [77] . Also , ALKBH5 , an m6A demethylase , is reported to promote GBM tumorigenesis by stabilizing nascent FOXM1 transcripts through mediating its m6A levels [78] and FOXM1 has also been show to control Wnt target gene expression in GBM . It is highly likely that there are alternative pathways such as Wnt , by which m6A regulates cell migration and tumorigenesis . Functional enrichment of subnetworks also presents a highly consistent interpretations of their functions ( S3 and S4 Figs ) . This suggests that the m6A-regulated genes in the subnetworks are likely to involve in the same processes and pathways and thus share similar functions . In terms of their enriched processes and functions , we observed again important pathways such as focal adhesion , mTOR signaling pathway and AMPK signaling pathway , which are also known to regulate cell cycle , and cell migration , and are critical in cancer . Finally , the network-based disease association analysis on the m6A-regulated genes reported 5 significant associated diseases , where 3 of them have been corroborated by the existing publications . Our predictions could provide clues for the mechanisms , by which m6A regulating these diseases . While there is no published evidence to support the other two associated diseases , our prediction points to potentially new disease associations of m6A . Last but not the least , our proposed methods have several issues that need to be further addressed in the future . First , because of the lack of miCLIP data , the model has only been trained on data from HEK293 cell line; this may not be enough to capture the features of all different kinds of true positive m6A sites . Second , the scale of phenotype-gene relationships in the OMIM database are relative small , which might not be able to capture all potential significant disease-gene correlations . A potential solution is to integrate other disease-gene annotation information like DisGeNET [79] to make the network more complete . Finally , we only analyzed the common m6A sites across many samples and their influence on gene expression . However , some of these context specific m6A sites appear only in a unique sample and they could be important for understanding m6A functions under this specific condition . Developing algorithms to detect these unique genes would be another future work .
The positive single-base m6A sites for our training data for Deep-m6A and Deep-m6A-S were obtained from [45] , which developed m6A individual-nucleotide-resolution cross-linking and immunoprecipitation ( miCLIP ) technology . This paper includes two alternative technologies including cross-linking–induced mutation sites ( CIMSs ) and cross-linking–induced truncation sites ( CITSs ) for human embryonic kidney ( HEK293 ) cells using total cellular RNA . The CIMS miCLIP uses Abcam as antibody and C→T transitions as feature , whereas the CITS miCLIP uses SySy as antibody and truncations as feature . We chose CITS miCLIP generated single base m6A sites as true positive m6A sites because the corresponding MeRIP-Seq data of the HEK293 cell line that we used for training were also generated with SySy as antibody [77] . Another reason is that 74% ( 4847/6543 ) of CITS sites can be mapped to peaks generated by MeRIP-Seq data , whereas this ratio is only 55% ( 5202/9536 ) for CIMS sites . The MeRIP-Seq peaks were detected using the exomePeak R package [41] from 2 replicates of MeRIP-Seq samples [77] . A consistent peak that appears in both replicates and also contains at least one CITS miCLIP site was determined as a single-base m6A-containing MeRIP-Seq peak . The mRNA sequences of these peaks ( without introns ) were subsequently extracted and any sites in the sequences that contain DRACH motifs were defined as candidate m6A sites . A candidate m6A site was then extended to 101 nt centered at the “A” of the DRACH motif to capture the sequence and reads count features around the motif . The candidate sites that also are CITS miCLIP m6A sites were determined as the positive samples ( 4 , 742 in total ) and other candidate sites that are at least 50-nt away from a positive “A” and are not any CIMS miCLIP sites or single base m6A sites reported in other experiments [55] were defined as the negative samples ( 33 , 718 in total ) . The independent miCLIP test data were obtained from [69] , which has 3 miCLIP replicates for the leukemia cell line MOLM13 , each of which contains 8 , 113 , 2 , 886 and 2 , 050 miCLIP sites , respectively . There were in total 11 , 746 miCLIP m6A sites after combining all the three replicates , whereas only 136 sites were common across all the three replicates . Here , we considered miCLIP m6A sites that appeared in at least two replicates as true positive m6A sites ( 1 , 147 in total ) . The corresponding MeRIP-Seq test data were obtained from [80] , which included 2 MeRIP-Seq replicates for the MOLM13 cell line . One of the 2 replicates has very low sequencing depth , so we took the combined peaks of these two replicates reported by exomePeak as MeRIP-seq peaks . Similar as the training data , the MeRIP-Seq peaks that contain at least one miCLIP m6A sites were determined as single-base m6A-containing MeRIP-Seq peaks and the sites with DRACH motifs in these regions were defined as candidate single-base m6A sites . The candidate m6A sites that are also miCLIP m6A sites were determined as positive samples ( 726 in total ) and other candidate sites that are >50-nt away from any positive sites and are not miCLIP sites from any of the three miCLIP replicates were determined as negative samples ( 6 , 577 in total ) . The human MeRIP-Seq data were downloaded from MeT-DB V2 . 0 [26] , which include 75 human samples from different human cell lines and tissues under different conditions , including 9 cell lines ( A549 , Dendritic cells , embryonic stem cell , Endoderm , HEK293T , HeLa , HepG2 , neural progenitor cells , OKMS inducible fibroblasts and U2OS ) and brain tissue samples under different conditions . The reference PPI networks were built based on BioGRID ( release 3 . 4 . 128 ) [50] , HINT+HI2012 [51 , 52] , MultiNet [53] , and iRefIndex [54] . After removing the isolated proteins and self-interaction proteins , we established a PPI network with a total of 16 , 062 proteins and 152 , 676 interactions . The last three PPI networks were downloaded from http://compbio . cs . brown . edu/pancancer/hotnet2/ . The HINT+HI2012 network contains 9 , 858 genes and 40 , 704 edges; the iRefIndex network contains 12 , 128 genes and 91 , 808 edges; and the Multinet network contains 14 , 398 genes and 109 , 569 edges . Diseases ontology terms were collected from the Disease ontology [81] . The “doSim” function of R package “DOSE” [82] was used to calculate the semantic similarity between two DO terms , which are used as the edge weight of the disease network . The disease gene relationship information was extracted from the OMIM database [83] . The OMIM ID was mapped to DO ID so that we can use OMIM gene–phenotype relationship to connect the PPI network and DO disease network to construct a gene-disease network . Four gene-disease heterogeneous networks were built for each of the PPI networks . We developed 2 CNN models for single-base m6A prediction , one called Deep-m6A and the other was Deep-m6A-S . These 2 models use the same CNN structure and hyperparameters but with different inputs . For Deep-m6A-S , the input is the OneHot encoded 101nt RNA sequences centered at the “A” of a DRACH motif . OneHot encoding translates the A , U , C , G characters into a binary vector of ( 1 , 0 , 0 , 0 ) , ( 0 , 1 , 0 , 0 ) , ( 0 , 0 , 1 , 0 ) and ( 0 , 0 , 0 , 1 ) , respectively . Therefore , the input of Deep-m6A-S becomes a 4 x 101 matrix , Ms . On the contrary , Deep-m6A takes both RNA sequences and the features of MeRIP-Seq IP reads count at each nucleotide of the RNA sequence . However , the input for Deep-m6A is similar to OneHot encoded Ms for Deep-m6A-S but with 1s replaced by the IP reads count feature at that nucleotide . The IP reads count features were calculated by the same approach as exomePeak , i . e . , RCnorm=ln ( RC/RCtotal*108 ) ( 1 ) where RCnorm is a 101-dimension vector of normalized reads count , RC is another 101-dimension vector of raw reads count , and RCtotal is the total number of reads for the MeRIP-Seq IP sample . Because there are two replicates in each of the HEK293 and MOLM13 datasets , we calculated the average RCnorm of the two replicates as input IP reads count . The Deep-m6A input is then calculated as Msr=MsDRCnorm ( 2 ) where DRCnorm is a 101x101 diagonal matrix with diagonal entries being those of RCnorm . Deep-m6A and Deep-m6A-S use CNN [44] to capture the non-linear features of input sequences and IP reads count . The adopted CNN architecture consists of a convolutional , a max-pooling and two fully connected layers ( Fig 4 ) . The convolutional layer outputs the pointwise product between the input matrix ( Msr for Deep-m6A and Ms for Deep-m6A-S ) and filters , which is followed by a rectified linear ( ReLU ) activation . Then , a max-pooling layer , which selects the maximum value over a window , is applied to reduce the dimensionality , which is followed by a dropout operation to reduce the complexity of the model . Finally , a fully connected dense layer is added followed by another ReLU and dropout to combine all the features learned by each filter . The output of the dense layer is passed on to the softmax function to generate the probability of the input sample to be an m6A site . Because the positive and negative sample sizes are imbalanced , we split the negative samples into seven subsets of equal size and trained 7 CNN models , each on a set that paired positive samples with a subset of negative samples . Seven models are trained as a result . For any prediction , the averaged prediction probability of the seven models is taken as the final predicted probability for an input sample . We implemented the CNN models using the keras R package . We refer Deep-m6A and Deep-m6A-S as the combination of the seven models . Input data files and Deep-m6A source code and the models are made publicly available on GitHub ( https://github . com/NWPU-903PR/Deepm6A . git ) . The inputs of the Deep-m6A function are a MeRIP-Seq IP sample bam file and a bed file that annotates the peaks identified by exomePeak R package from MeRIP-Seq data . The output is an excel file , where each row contains information about a predicted single-base m6A site extracted from an exomePeak peak region and each column denotes the chromatin , the chromatin start , the chromatin end , Entrez gene ID , the predicted probability , the strand and the motif at this location , respectively . We applied Deep-m6A to predict single-base m6A sites in 75 human MeRIP-Seq samples from MeT-DB2 . First , we applied exomePeak to predict m6A peaks in each MeRIP-Seq sample and then searched for DRACH motifs in the peak regions . “A”s in these motifs were treated as candidate single-base m6A sites and the 101nt RNA sequences centered at these “A”s and the corresponding IP reads counts were extracted to construct the input matrix Msr . For each candidate site , we separately applied Deep-m6A trained on the training data to calculate the probability of it to be a m6A site . We predict a positive site when the output probability is greater than 0 . 907; this threshold is chosen because at this threshold the 10-fold CV test in training can achieve a precision of 0 . 7 . In this way , we detected in total 23 , 456 single-base m6A sites . We define m6A-regulated genes as genes whose expression level is influenced by its m6A methylation level more than expected by chance . We propose a new network based algorithm , Hot-m6A , to identify m6A-regulated genes . Hot-m6A first determines that genes whose expression levels are significantly influenced by m6A by assessing the correlation between the methylation level and the corresponding expression . If the expression levels of a gene change together with its m6A levels across more samples , then there will be a higher chance that the gene expression is influenced by m6A . To obtain a statistically meaningful correlation , we need to have the same m6A sites appearing in many samples . In this study , we estimated the sample size needed to detect a correlation of 0 . 8 with a significance level of α = 0 . 05 . Using a two-sided test based on the Fisher’s z transformation , at α = 0 . 05 with power 90% , the required sample size is approximately 12 ( i . e . , n = 12 ) [48] . Thus , we select single-base m6A sites that appear in at least 12 samples and calculated the Pearson’s correlations of the methylation degree and gene expression level across all occurred samples . The methylation degree was calculated as: Methlevel=ln ( mean ( eRCnorm ) /GFPKM ) ( 3 ) where RCnorm is defined in ( 1 ) , GFPKM is the FPKM of gene harbored the corresponding m6A site , calculated from the MeRIP-seq input sample by Cufflinks [84] . The expression level is denoted by the ln scale gene FPKM . The correlation coefficient is then transformed into a z score using a revised Fisher’s z-transformation ( which is known as Hotelling's ( 1953 ) second-order transformations ) [49] that considers the influence of sample size n , which is expressed as z=12ln ( 1+r1−r ) ( 4 ) z*=z−3z+r4n ( 5 ) where r is the correlation , n is the sample size , and z is Fisher’s transformation . We used z* of each gene as its correlation for the subsequent analysis in the pipeline . Genes containing this kind of m6A sites are treated as candidate m6A-regulated genes . For a candidate gene that has multiple m6A sites , the z* score with the largest absolute value among all its harbored m6A sites is selected as the correlation degree for that gene and the absolute z* score are used to denote the degree of regulation . To further select m6A regulated genes more than expected by chance , Hot-m6A adopts the HotNet2 algorithm [85] , which was originally developed to identify significant mutation genes for cancer . HotNet2 takes the gene mutation frequency or mutation score as input heat vector and applies a heat diffusion method to identify subnetworks of a genome-scale interaction network that is mutated more than expected by chance . The advantage of HotNet2 is that it not only detects significant mutated genes with high mutation frequency but also can identify significant mutation genes with relatively low mutation frequency but interact closely with other significant genes . In our case , we also want to identify significant m6A-regulated genes that not only have high m6A-expression correlations but also cooperate with each other functionally in a functional network . To this end , Hot-m6A takes the absolute m6A-expression correlation coefficients z* scores of candidate m6A regulated genes as the input heat vector and uses the four PPI networks as the reference functional network , which includes BioGRID , which has been shown to contain the significant interactions between m6A methylated genes [43] , and the other three PPI networks that were used in the HotNet2 paper . Hot-m6A has two parameters β and δ; β is the fraction of the heat that a node in the network retains for itself at each time step and δ is the threshold , which determines whether there is an edge between 2 nodes in the final subnetwork . β can influence the amount of heat that a gene shares with its neighbors and is determined by the topology of the PPI network . For the three networks used in the HotNet2 paper , we selected β as 0 . 4 for HINT+HI2012 , 0 . 45 for iRefIndex , and 0 . 5 for Multinet , which are reported values in [46] . For the BioGRID network , we performed the same analysis as the HotNet2 paper did to determine the value of β . For each different β from {0 . 05 , 0 . 1 , 0 . 15 , … , 0 . 95} , we analyzed the inflection point at which the heat kept in the direct interacting neighbors of a gene drops and selected β as the one with the largest inflection point . As shown in S6 Fig , β is selected as 0 . 5 for the BioGRID PPI network . On the other hand , δ influences the scale of the subnetwork that we generated . To automatically determine it , for each PPI network , we firstly generated 100 random PPI networks [86] . We used the same “heat” vector at each run and select δ as the minimum one where all strongly connected subnetwork components identified by Hot-m6A have a size less than and equal to a threshold Lmax . For each Lmax = 5 , 10 , 15 , 20 , we reported the median of the 100 δmins for the 100 random networks . For each run , we selected the smallest δ with the most significant ( P < 0 . 05 ) subnetwork sizes k as described in HotNet2 [46] . The P-value , which denotes the significance of subnetwork size k is computed for the statistic Xk , the number of subnetworks of size ≥ k reported by Hot-m6A . To compute an empirical distribution of Xk for computing the P-value , we permuted the heat scores , i . e . , the z transformation of correlation , among the genes in the original PPI network for 1000 times and then applied HotNet2 to the network using the permutated heat scores . In the end , δ was selected as 0 . 00769 for BioGRID , 0 . 0148 for HINT+HI2012 , 0 . 00998 for iRefIndex , and 0 . 00777 for Multinet . We applied Hot-m6A to each of the four PPI networks and pooled all the genes and edges reported in at least one network to form a candidate m6A regulated gene network , G . We then assigned the number of the PPI networks in which an edge exists as the weight of this edge . Next , we initialized the consensus subnetworks C as the connected components of G with edges of the weight = 4 , i . e . , the edges that are reported in all 4 PPI networks . Then we extended each consensus subnetwork S ∈ C by adding edges of weight = 3 , and afterward further extended them by adding edges of weight = 2 and then 1 . Finally , we defined the genes in the final extended networks as the m6A regulated genes and the genes in the initial consensus subnetworks as the consensus m6A regulated genes . To further detect the functional modules of m6A regulated genes , we removed edges with weight = 1 in the m6A regulated subnetworks to generate functional significant subnetworks . To further reveal the association between m6A and disease , we adopted the RWRH ( random walk with restart on heterogeneous network ) [47] algorithm to prioritize the candidate m6A regulated diseases . RWRH is a well-known heterogeneous network-based algorithm to infer the gene-phenotype relationship . The heterogeneous network contains three parts: gene-gene interaction network , phenotype ( disease ) network , and gene-phenotype relationship . Similarly , we also built four heterogeneous networks based on each of the four PPI networks . The disease network was generated from the semantic similarities between two DO terms of Disease Ontology database and gene-phenotype relationships were extracted from OMIM . Specifically , the OMIM phenotype ID was mapped to DO disease ID to match the disease network with the gene-phenotype relationship . RWRH performs a random walk with restart from certain gene and disease seed nodes in the heterogeneous network . After several steps of random walk , the probability of each node , that the seed genes and seed diseases will walk to , become steady and is returned as vectors pg∞ and pd∞ . pgi∞ denotes the probability that the seed genes and seed diseases will walk to gene i and pdj∞ is the probability that the seed genes and seed diseases will walk to disease j . In another world , RWRH can prioritize how closely a gene or a disease is correlated with the seed genes and seed diseases . To study the correlation between m6A and diseases , we first mapped all the 709 significant m6A-regulated genes obtained from HotNet2 to OMIM gene-phenotype relationship , which resulted in 308 phenotypes regulated by 177 m6A genes . Most of these phenotypes contain only one disease gene ( The details of the m6A gene-disease relationships are included in S1 File ) . Next , to obtain candidate m6A-associated diseases , we mapped these 308 m6A gene-correlated OMIM phenotypes to DO ID , which resulted in 90 m6A gene-associated DO diseases . For each of the four heterogeneous networks , these 90 diseases were taken as disease seeds for RWRH , whereas all the 709 significant m6A-regulated genes were mapped to the PPI network to serve as gene seeds for RWRH . Finally , we applied RWRH using these genes and disease as seed nodes to prioritize all the DO diseases in the heterogeneous networks . Fig 5 illustrates the heterogeneous network in this study . As shown , there is a chance that d3 may be ranked as significant m6A-associated diseases even though there is no m6A-regulated gene directly connected with it . The reason is that the genes ( i . e . , g3 , g7 ) that closely interacted with m6A regulated genes ( i . e . , g5 ) are connected to d3 and m6A regulated genes ( i . e . , g5 ) connected with disease ( i . e . , d2 , d4 ) also closely interacted with d3 . RWRH can efficiently capture this significant network topology and prioritized d3 as significant , which showed the power of this study to prioritize the potential m6A-associated disease with no prior knowledge that correlated with m6A-regulated genes . The top 10 ranked diseases based on the RWRH output probability pd∞ were selected as candidate m6A-associated diseases . To ensure these top diseases were really influenced by m6A-regulated genes rather than by chance , we implemented a random test to calculate an empirical p-value to assess if the disease is randomly selected . For this test , 100 random PPI networks that only keep the degree distribution of original network were generated using the method in [86] . Then , for each random PPI network , we connected it with the disease network using the disease-gene relationship and deleted the relationship between m6A-regulated genes and their corresponding diseases to generate a corresponding random heterogeneous network whose gene interaction relationship is random and contains no prior knowledge of the relationship between m6A regulated genes and any disease . After that , we applied RWRH in each of the 100 random heterogeneous networks . Finally , for each candidate m6A-associated disease dj ( j = 1 , 2 , ⋯ , 10 ) , the empirical p-value was calculated as: p=#{π ( dj ) }100 , where π ( dj ) is a random heterogeneous network in which dj is found as the top 10 candidate disease after RWRH . The empirical p-value is calculated as the probability that a candidate m6A-associated disease is selected by random . The candidate m6A-associated diseases that with p < 0 . 05 were selected as significant candidate m6A-associated diseases . The significant candidate m6A-associated diseases that reported by all the four heterogeneous networks are determined as significant m6A-associated diseases . | The goal of this work is to identify functional significant m6A-regulated genes and m6A-associated diseases from analyzing an extensive collection of MeRIP-seq data . To achieve this , we first developed Deep-m6A , a CNN model for single-base m6A prediction . To our knowledge , this is the first condition-specific single-base m6A site prediction model that combines mRNA sequence feature and MeRIP-Seq data . The 10-fold cross-validation and test on an independent dataset show that Deep-m6A outperformed two sequence-based models . We applied Deep-m6A followed by network-based analysis using HotNet2 and RWRH to 75 human MeRIP-Seq samples from various cells and tissue under different conditions to globally detect m6A-regulated genes and further predict m6A mediated functions and associated diseases . This is also to our knowledge the first attempt to predict m6A functions and associated diseases using only computational methods in a global manner on a large number of human MeRIP-Seq samples . The predicted functions and diseases show considerable consistent with those reported in the literature , which demonstrated the power of our proposed pipeline to predict potential m6A mediated functions and associated diseases . | [
"Abstract",
"Introduction",
"Result",
"Discussion",
"Methods",
"and",
"materials"
] | [
"genetic",
"networks",
"rna",
"sequences",
"gene",
"regulation",
"protein",
"interaction",
"networks",
"methylation",
"network",
"analysis",
"sequence",
"motif",
"analysis",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"computer",
"and",
"informatio... | 2019 | Global analysis of N6-methyladenosine functions and its disease association using deep learning and network-based methods |
The liver plays a key role in removing harmful chemicals from the body and is therefore often the first tissue to suffer potentially adverse consequences . To protect public health it is necessary to quantitatively estimate the risk of long-term low dose exposure to environmental pollutants . Animal testing is the primary tool for extrapolating human risk but it is fraught with uncertainty , necessitating novel alternative approaches . Our goal is to integrate in vitro liver experiments with agent-based cellular models to simulate a spatially extended hepatic lobule . Here we describe a graphical model of the sinusoidal network that efficiently simulates portal to centrilobular mass transfer in the hepatic lobule . We analyzed the effects of vascular topology and metabolism on the cell-level distribution following oral exposure to chemicals . The spatial distribution of metabolically inactive chemicals was similar across different vascular networks and a baseline well-mixed compartment . When chemicals were rapidly metabolized , concentration heterogeneity of the parent compound increased across the vascular network . As a result , our spatially extended lobule generated greater variability in dose-dependent cellular responses , in this case apoptosis , than were observed in the classical well-mixed liver or in a parallel tubes model . The mass-balanced graphical approach to modeling the hepatic lobule is computationally efficient for simulating long-term exposure , modular for incorporating complex cellular interactions , and flexible for dealing with evolving tissues .
As the number of man-made environmental chemicals continues to grow , there is an urgent need to develop effective tools to test their potential risk to humans . The number of environmental chemicals that are produced in substantial quantities now numbers approximately 10 , 000 [1] . In order to determine the potential risk to humans of exposure to these compounds , it is critical to establish a dose-response curve – the functional dependence of toxic endpoints , e . g . hepatic lesions , on exposure to that compound . Traditional long-term animal testing to determine dose-response is time consuming , expensive , and requires the sacrifice of thousands of animals without clear relevance to humans . Recognizing this need for new approaches to toxicity testing [2] , [3] , [4] , the U . S . Environmental Protection Agency is conducting ongoing efforts to collect in vitro data [5] to make inferences about in vivo toxicity in both test animals [6] and humans [7] . Without appropriate context , in vitro testing is insufficient for predicting effects in vivo . Context can be established through informatics , i . e . correlating in vitro data with known in vivo phenotypes , or modeling efforts in which abstract rules are hypothesized to determine in vivo outcomes as a function of variables , some of which may be determined in vitro . Whereas empirical models describe the available data and are therefore best limited to interpolation , physiologic models attempt to describe the underlying biology in sufficient detail to emulate the true dynamics . Physiologic models generate new hypotheses which can subsequently be tested to refine the model . Both informatics and modeling approaches create frameworks without which there could be little meaningful interpretation of in vitro data . Our goal is to establish an in silico model for dose-response that can be calibrated using in vitro characterizations of chemical effects . The liver is often the site of initial exposure to hazardous compounds and their metabolites due to first-pass metabolism of blood from the gastro-intestinal tract via the hepatic vein . In mammals the hierarchical structure of the liver terminates in 105 to 106 functional units called lobules [8] first identified by Kiernan [9] . Each hepatic lobule receives blood from up to six portal triads , each typically consisting of a hepatic arteriole and a portal venule in addition to a bile ductule [10] . Blood flows through intervening spaces between the cells , i . e . sinusoids [11] , and drains into the central vein . Hepatocytes are arranged in plates one to two cells thick , organized radially around the central vein . A two-dimensional slice of a hepatic lobule is shown in Figure 1 . Compounds within the blood are exchanged with the hepatocytes sequentially as blood passes through the sinusoids . The action of the enzymes within the hepatocytes on compounds produces metabolites that may be more or less harmful than the parent compound . Although mechanisms of chronic chemical-induced injury are not completely understood , it is believed to involve multiscale molecular and cellular interactions that culminate in tissue damage . Tissue dosimetry is traditionally estimated using physiologically-based pharmacokinetic ( PBPK ) models . A PBPK model consists of a system of ordinary differential equations ( ODEs ) for the concentration of a compound ( or compounds ) in different tissues . Typically some key tissues are treated as separate compartments for which a tissue-specific concentration is calculated , while other tissues are modeled using aggregate compartments . More complicated dynamics within a tissue , such as diffusion or membrane transport , are often modeled with additional sub-compartments but each sub-compartment is well-mixed . The equations are parameterized by subject- or species-specific physiologic parameters such as cardiac output and tissue volumes as well as compound-specific parameters such as diffusion/transport rates and tissue-specific plasma to tissue “partition coefficients” corresponding to the assumption of a rapidly-established equilibrium between concentration of compound stored in the tissue and the concentration of compound in the plasma flowing through the tissue . PBPK models relate the concentration of compounds inhaled or ingested from the environment to internal tissue doses [12] , [13] , [14] . In addition to the well-mixed approach , the parallel tubes model of liver function has often been used to calculate in vivo hepatic metabolism based upon in vitro measures such as intrinsic hepatic clearance [15] , [16] . Typically used at steady-state , the parallel tubes model assumes that each lobule is a tube connecting a portal triad and central vein , along which concentration varies spatially . Though in vitro studies typically average over the response of a many hepatocytes within a well , hepatocyte function is known to vary significantly in vivo [17] , e . g . , hepatocytes near the central vein may express very different levels of enzymes than those nearer to the portal triad . For this reason the lobule is divided into zones of approximately similar hepatocyte function depending on location within the lobule . The heterogeneity between these zones is thought to arise from gradients in oxygen availability , exposure to nutrients from the portal venules , and hormone concentration [18] . Modeling the differences between regions of the lobule should provide key insights into the differences between phenomena observed in homogenous in vitro conditions and heterogeneous in vivo reality . The first multi-compartment geometric model of the liver was developed by Andersen et al . [19] . In that model there were no cells , but the concentrations of compounds in different zones of lobules were modeled continuously and could therefore be coupled to a PBPK model . Liu et al . [20] have followed a similar sub-compartment coupled to PBPK approach for modeling zonal heterogeneity due to transporters and enzymes . Recent approaches to simulating the response of the liver include that of Ohno et al . [21] who coupled independent realizations of a model for cellular dynamics into a linear array to allow some instances of the model to be close to the source of nutrients and foreign compounds while others were further removed . Höhme et al . [22] have developed a discrete model of the hepatic lobule that considers the biochemical forces between hepatocytes to simulate recovery following acute chemical toxicity . Ierapetritou et al . [18] recently conducted a thorough review of liver tissue simulation approaches in which they summarize the previously mentioned approaches as well as higher dimensional models including fluid dynamics approaches based upon approximations of the Navier-Stokes partial differential equations . Such approaches are data- and computationally-intensive , especially given the convoluted and dynamic cellular boundary of the sinusoidal spaces . Hunt et al . [23] have taken a unique agent-based approach with individual hepatocytes represented by agents wherein metabolism can occur . The environment of the agents is determined using a hybrid graph and grid approach in which compounds are represented by objects moving through the lobule . Cell-oriented agent-based modeling ( ABM ) of tissues offers a number of unique advantages [24] , [25] . First , since cells are the functional units of tissues , the ABM has more physiologic relevance than a continuum model . Second , the agent responses can be calibrated and verified through comparison with actual cells in vitro ( or ex vivo ) . Third , spatial outcomes from the ABM can be translated to histopathologic effects such as acute lesions and tumor formation . While the agent-based strategy is suitable for modeling tissue responses , the approaches to the liver taken so far have not provided a framework for estimating tissue dosimetry . Though the spatial distribution of a compound has previously been modeled , past approaches have represented compounds as agents that are difficult to link to traditional exposure modeling . Due to the spatial heterogeneity of the hepatic lobule , both molecular and cellular , it is important to model the microanatomic distribution of chemicals and to relate this to continuous variation in chemical concentration resulting from changes in human environmental exposure . We have implemented a microdosimetry model that relates whole-body chemical exposures to cell-scale concentrations . Our objective was to develop the framework for simulating the microanatomic distribution of various environmental chemicals in a canonical lobule for extended periods of time ranging from hours to months . This required an approach that is quantitative , efficient in computational resources , and sufficiently flexible to account for anatomic changes ( due to chemical insult or other factors ) [26] . First , we approximated the microanatomic architecture of the hepatic vasculature and parenchyma assuming a discrete topology by a graphical model . This allowed us to systematically explore the consequences of morphologic changes on the concentration distribution . Second , we transformed the sinusoidal elements of the vascular network into a system of microscopic well-mixed compartments through which material flow was assumed to be one-dimensional . Third , we connected the virtual lobule to a PBPK model to relate individual exposure to microdosimetry . For a range of physiologically relevant morphologic parameters we evaluated the microdosimetry in response to xenobiotic exposure levels and varying physico-chemical attributes .
The two dimensional morphologic characteristics of the mammalian hepatic lobule were represented as a discrete connectivity graph , in which the edges captured spatial proximity . The two main anatomic entities considered are hepatocytes , the parenchymal cells responsible for the metabolism of chemicals , and vasculature , i . e . sinusoids through which chemicals flow to the hepatocytes . These are represented by different node types including: hepatocytes , sinusoidal primitives , arterial and venous sources , and the central vein , while edges represent connectivity and spatial proximity between the nodes . Mass transfer through the sinusoidal network occurs through edges: The edges connecting vascular nodes transfer material through the sinusoids , whereas edges between the vascular and cellular nodes exchange material between the sinusoids and parenchyma . A simplified geometry of the lobule was defined using the following morphologic parameters: the number portal triads ( defining the vascular inputs ) , the branching factor of the sinusoids , the number of sinusoids entering the central vein , and the sizes of sinusoids , hepatocytes , and the lobule . The graphical model of the lobule was constructed algorithmically using these parameters and visualized spatially ( Figure 2 ) . The “virtual lobules” generated in this manner presented a complex sinusoidal architecture representing a substantial challenge for estimating the distribution of xenobiotics and nutrients . The graphical model of the lobule was generated iteratively ( the algorithm is described in the methods section ) . The sinusoidal network was constructed starting with the central vein and extending radially outwards to the portal region . Beginning with a node representing the central vein , sinusoid primitives ( nodes ) were sequentially appended to form the initial vasculature . Small random variations in the placement and branching of sinusoidal primitives were used to reconstruct the histologic appearance of a hepatic lobule . Second , the hepatic arterioles and portal venules , were placed at the perimeter of the lobule and connected sinusoidal network . Third , the parenchymal cells were placed contiguously with the sinusoidal network . Because we chose to connect the portal venule and arterioles to the central vein in two dimensions the spatial layout was not completely space-filling . The approach described above is flexible , allowing the generation of diverse lobular topologies through which flows can be simulated . Five basic morphologies were examined , as depicted in Figure 3 , in which the number of portal triads ( more accurately dyads since bile was neglected ) , the probability of sinusoid branching Pbr , and the presence of random noise were all varied . No random noise or branching and one portal dyad produced a lobule with a single tube ( panel a in Figure 3 ) that in the limit of many sinusoidal segments approaches a parallel tubes model . With multiple portal dyads a classical lobule structure [27] that allows both direct flow from the portal triads to the central vein and mixing flow between portal triads is produced ( panel b in Figure 3 ) . A 10% chance of sinusoid branching ( panels c and d ) produced nearly space-filling lobule graph while a 5% chance of sinusoid branching ( panel e ) did not . Miller et al . ( 1979 ) observed that the branching of sinusoids is greater near the portal triad than near the central vein [28] . Human lobules have been observed to typically have between three and six portal triads per lobule [8] , [10] , [27] , though many “triads” actually consist of dyads missing either an arteriole , bile duct , or most commonly a portal venule [10] . Given these observations , we believe that the geometries that include multiple portal triads and random branching of the sinusoids ( panels c , d , and e ) appear qualitatively more physiologic . Blood circulation through the graphical model of the vasculature was simulated as a network flow ( Figure 4 ) . Because the sinusoidal diameter is much smaller than hepatocytes [29] , there are a large number of sinusoid primitives in each virtual lobule . To efficiently solve for the flow , the sinusoid primitives were aggregated into the following components: “straight” or linear sequences and “branch” sections where straights meet and mix . As shown in Figure 5 , graph aggregation results in a smaller graph that preserves the spatial distribution of the sinusoids . Each aggregated node was assumed to be well-mixed , that is , each constituent sinusoid primitive i has the same concentration Cμi ( see Table 1 for a list of all symbols used in this document ) . Mass-balanced flow through the aggregate graph was determined by solving for the flow across each edge of the sinusoid graph G ( V , E ) due to the sources at both the arterial and venous elements of each portal triad . In general , solving for network flow from node i to node j across edge Eij requires |E| different weights Qμij ( i . e . , flow rates ) . Mass-balance provides only |V| constraints – one at each node – so additional constraints were needed . We made use of the hemodynamical equivalent of Ohm's law [30] , [31]:where Pi is the pressure at node i and the resistance was assumed to have the same value R for all edges . We note that Rij could be determined using schemes such as the cross-sectional area of each branch . Hemodynamics provides |E| additional constraints , but introduces |V| additional unknown pressures Pi . Together with mass balance we have |E|+|V| constraints for |E|+|V| unknowns . This system of equations can be represented with a matrix and , given source flows and outlet pressure , can be solved by diagonalization . Since we are not currently interested in sinusoidal pressure , R and the outlet pressure were taken as one . This assumption does not effect the quantitative values of Qμij since they depend only on the relatively values of Pi . As can be seen in Figure 4 , randomly generating sinusoids can lead to dead-end sinusoids for which no flow is predicted . These sinusoids are removed from the lobule and additional hepatocytes are added where possible . To evaluate the appropriateness of these assumptions and the suitability of the approach to arbitrary graphical structures , we return to Figure 3 , where predictions are made for a rat liver lobule and compared to measurements made by Komatsu et al . ( 1990 ) for the radial dependence of flow of erythrocytes in the sinusoids with distance from the central vein . In vivo microscopy was used by Komatsu et al . to observe the exposed livers of ten Sprague-Dawley rats and flow was measured in three zones – near the central vein , near the portal venule , and intermediate [31] . Flow was observed to increase with distance from the portal venule , presumably as blood from the portal arteriole and other portal triads mixed in . As can be seen on the left-hand side of Figure 3 , only geometries where random branching is present ( panels c , d , and e ) , produce profiles with increasing flow as the central vein is approached . Given the indeterminacy in where flow was measured relatively to the central vein , it is hard to determining the precise radial profile of the flow . All geometries produce mean flow within a factor of two of the measured values , supporting the approximate appropriateness of this graphical approach to hemodynamics in the hepatic sinusoids . A list of simulation parameters used is given in Table 2 . The final step needed to determine the concentration Cμi for each sinusoid i is to find the concentration of compound ( s ) in the blood flowing into the liver . Our approach requires the rate and chemical concentration ( s ) for blood flow from the gut and the hepatic arteries . We used a simple PBPK model ( Figure 6 ) with oral and inhalation routes of exposure ( PBPK model parameters are listed in Table 3 ) . Microdosimetry for each lobule was determined from the pharmacokinetic exposure model by assuming an arteriole flow equal to Qμart = Qliv/Rliv∶lob/NPT and a venule flow Qμven = Qgut/Rliv∶lob/NPT where Rliv∶lob is the ratio of liver to lobule volume and NPT is the number of portal triads per central vein . Concentrations within the lobule are determined by solvingwhere Mi is the summed clearance of all hepatocytes adjacent to aggregate sinusoid i , and I is the identity matrix . Note that at steady state the flow can be determined from just the geometry G ( V , E ) and the metabolism M , in which case and there is no need to solve for dynamic concentration changes , since new concentrations can be calculated analytically . For a completely physiologic , three-dimensional lobule Rliv∶lob would be equal to the number of lobules in the liver – approximately 106 [32] . We determined Rliv∶lob , the ratio of the total volume of the liver to the total volume of the sinusoidal spaces and hepatocytes in the simulated lobule , to be approximately 108 , which is roughly 100 times greater than the physiologic value . We expect a greater value for two reasons: First , many components of the lobule other than the sinusoidal spaces and hepatocytes , such as endothelial and stellate cells , extracellular space , and bile ducts , contribute to the volume of the lobule . Including these additional components , and therefore increasing the volume of the simulated lobule , will reduce Rliv∶lob . Second , each simulated lobule is assumed to have a thickness equal to a sinusoidal diameter ( 23 . 5 µm [29] ) . As is illustrated in Figure 7 , many ( quasi- ) two-dimensional lobules are needed to fill the same volume ( and thus preserve mass balance ) as single three-dimensional lobule . The difference between Rliv∶lob and the actual number of lobules indicates that 100 simulated lobules are currently needed to fill the space of a single physiologic lobule . Using the lobule geometries ( given in Figure 3 ) ensembles of ten lobules were used for simulating blood flow . For each geometry the flow was simulated for an oral exposure of 10 µMol total ( equivalent to 0 . 03 mg per kg body weight for a 200 molecular weight compound and a 70 kg subject ) with an intrinsic hepatic clearance due to metabolism of 10 µL/min/million hepatocytes . We compared the average concentration throughout the lobule , as predicted by our approach , with the prediction Cliv for a PBPK model with a well-mixed liver compartment with equivalent metabolic clearance ( i . e . the product of the clearance per hepatocyte , the total number of hepatocytes in a lobule , and the effective number of lobules Rliv∶lob ) . It is important to note that the overall pharmacokinetics depends on the lobule layout because the effective number of lobules Rliv∶lob is determined by volume alone and therefore the total clearance of the liver depends on the number of hepatocytes relative to the volume of the lobule . Though the overall clearance varied with geometry , the impact of different geometries on the average concentration in the lobule was small . As shown in Figure 8 , for the assumed metabolism rate the mean predicted concentration did not vary greatly from what would be predicted for a more traditional well-mixed compartment . To compare results between geometries the concentrations were scaled by Cliv predicted for the appropriate CL . We find that in all cases the predicted average concentration slightly exceeds the well-mixed PBPK prediction , but that otherwise the pharmacokinetics are very similar . Plotted on the right-hand side of Figure 3 is the radial-dependence of concentration on position relative to the central vein at tmax – the time at which the lobule reaches maximum average concentration , . In all cases the mean concentration decreased slightly from the portal triads to the central vein – the predicted concentration was similar to the parallel tubes model . Thus , the mean predictions were similar to typical approaches for predicting liver concentrations . Geometry had a much greater impact on the variability in predicted concentrations Figure 3 . For all the lobules with random branching great variability was observed at the edges of the lobule , maximally distant from the central vein . Some regions receive slightly higher concentrations while other , stagnant regions received almost none . This supports the idea of considering sinusoidal topology for estimating changes in the local environment of a hepatocyte in addition to radial location between the central vein and the portal triad ( i . e . zone I , II , or III ) . Since there were not large differences between the predictions for the three lobules with random branching , we arbitrarily chose to simulate lobules with six portal triads and 10% chance of branching ( geometry c in Figure 3 ) for the remained of the studies in this paper . A larger ensemble of fifty lobules was generated for these studies . To test whether a continuum approximation ( ODEs ) was appropriate for modeling mass transfer in the sinusoidal graph we estimated the number of molecules at a hepatocyte . If the number of molecules at higher concentrations is not large enough a stochastic approach [33] would be preferable . As shown in Figure 9 , the upper bound on the number molecules at a total dose of 10 µM is was nearly a million molecules per hepatocyte , as calculated by multiplying the concentration in the sinusoid adjacent to each hepatocyte and dividing by the number of hepatocytes accessing that sinusoid . Though a small fraction of hepatocytes are exposed to almost no molecules , a continuum approach appears appropriate . The maximum concentration in the tissue following a dose is a commonly used measure of tissue exposure in pharmacokinetics . For the simulated lobule a local Cμi , max can be calculated for each hepatocyte as a result of different sinusoids receiving different concentrations . Figure 10 shows the distribution of Cμi , max experienced by all the hepatocytes in an ensemble of fifty lobules with intrinsic hepatic metabolic clearance of 10 µL/min/million hepatocytes . The values have been normalized to the Cmax predicted for a well-mixed liver with the same overall metabolic clearance ( indicated but the solid line ) . The peak for the distribution is in excess of the well-mixed prediction , while the breadth is quite wide , indicating that at this rate of metabolism some hepatocytes receive exposures nearly 40% greater than would be predicted for a well-mixed liver while others receive almost no exposure . Ito and Houston [15] summarize a range of intrinsic metabolism rates including values as low as 1 . 4 µL/min/million hepatocytes ( caffeine ) and as large as 1800 µL/min/million hepatocytes ( propranolol ) . This wide variability in metabolism rate has consequences for the variability predicted across the lobule . As shown in Figure 11 , the variability in exposure received by different hepatocytes grows from a few percent to nearly 800% for a metabolism rate of 1000 µL/min/million hepatocytes . For rapid metabolism those hepatocytes first exposed to blood from the portal triad receive eight times the exposure that would be predicted for a well-mixed liver , while downstream hepatocytes receive almost no exposure to the parent compound . Heterogeneity within the lobule is dynamic [34]; a low metabolism rate may be due to limited distribution of metabolizing enzymes , while a high rate of metabolism may lead to induction of enzymes , perhaps heterogeneously . Both of the distributions in Figure 9 and Figure 10 are broad , indicating that the average response of the ensemble is not necessarily characteristic of the response of any one simulated lobule . Given that these and other variability have been observed , any model of hepatic effect that depends upon local concentrations , particularly threshold models , may have a different response for a spatially-extended simulation than with a well-mixed simulation . The relevance of this heterogeneity will depend on the parameter regime – for low metabolism and little variability , the well-mixed approximation is likely to be sufficient . If large variability is present , e . g . for rapidly metabolized compounds , it may be crucial to determine which hepatocytes receive large exposures . This is especially useful for modeling spatial effects such as the development of lesions in one region , but not another . We conducted a preliminary analysis of the cellular effects due to microdosimetry using a simple agent-based model for hepatocytes . Each agent was defined by a fixed , identical xenobiotic metabolism rate , and functional states that were updated at each time step via state transition rules . A simple approach was used to encode probabilistic state transition rules conditioned on inputs from the agent environment . Future cellular models will be able to take better advantage of the freedom to proliferate and move provided by this approach since flow for a new arrangement can be determined rapidly by updating the sinusoid and contact graphs . Here we considered normal hepatocytes and cell death following exposure to threshold cytotoxic concentration . The ABM was integrated with the sinusoidal flow model with each being updated alternately . We simulated twelve minutes of the flow followed by eight iterations of the ABM – intended to be sufficiently small time periods for each model to respond realistically to changes in the other . Experimental verification will be needed to determine the appropriate time scales . Given the current cellular model and the predicted increase in variability with metabolism rate shown in Figure 11 , two types of comparisons were made: a spatially-extended hepatic lobule with an approximate “parallel tube” model ( given by the lobule geometry in Figure 3a ) and variability due to rapid metabolism for low ( 1 µL/min/million hepatocytes ) and high ( 1000 µL/min/million hepatocytes ) rates of metabolism . An arbitrary threshold of chemical concentration has been assumed , above which cell stress and apoptosis become much more common . Since different metabolic clearances and lobule geometries lead to different pharmacokinetics the simulations were normalized by varying the threshold for enhanced apoptosis – the threshold was set to 110% of the maximum average lobule concentration predicted for each configuration . For a well-mixed lobule , a threshold in excess of maximum lobule concentration should have no effect . Instead , as shown in Figure 12 we observed that spatial heterogeneity in toxicant concentration across the lobule enhanced cell injury before the chemical was cleared . This effect was not observed in the approximate parallel tubes model . Enhanced cell death was not observed at low xenobiotic metabolism rates in the spatially-extended lobule . Though there is some baseline apoptosis at the lower metabolism rate , there is roughly five times greater apoptosis for higher metabolism , i . e . greater variability in exposure . This suggests that lobular geometry is not solely responsible for the cell behavior and hepatocyte metabolism is required for the variability in the cellular response . Variation in cellular responses is frequently observed [17] and is thus physiologically relevant . While additional work is required to evaluate the responses in our model , these findings suggest the value of spatially extended tissue level models of microcirculation and cellular dynamics .
We have described a microdosimetry model to relate environmental exposures to cellular exposures . This is only a step toward developing virtual tissues that can predict the in vivo consequences of chemical exposure based upon in vitro information . The liver lobule is known to be spatially heterogeneous [18] , [34] . Zonal differences between central and peripheral hepatocytes include oxygen availability , hormone concentration , expression of metabolizing enzymes , ( e . g . , CYP 3A4 ) , gluconeogenesis , and glycolysis [18] . One clear conclusion of this modeling work is that morphology of the liver alone is insufficient to explain the observed zonation in hepatocyte function or even gradients in concentration across the lobule . We observed variations that are driven by the action of hepatocytes , i . e . metabolism , and not by geometry alone . A model for a spatially-extended hepatic lobule sets the stage for investigating emergent behavior in models of hepatocyte function . If the action of hepatocytes creates spatial variation across the lobule then any cellular dynamic response that depends on chemical or nutrient concentration may in turn be altered , which could be a prelude to zonal patterns of biological functions . More extreme effects , such as central lobular necrosis , may be due to the transformation of the compound via metabolism into a more potent compound or zone-dependent variation in sensitivity of the hepatocytes . In contrast to the well-stirred model of the liver , the simulated lobule provides a means of accessing a variety of inter- and intracellular dynamics . Though the results we obtain are in some respects similar to previous models , we gain the additional capability of allowing hepatocyte-specific dosimetry as well as the potential to alter lobule geometry , e . g . lesions or necrosis , in response to chemical injury . Since numerical approaches often allow even large systems of ODEs to be solved much more rapidly than analogous systems of PDEs [35] and since numerous algorithms exist for analysis of graphs [36] , we believe this approach is tractable for simulating sub-chronic and chronic xenobiotic exposure scenarios while preserving mass-balance . Because we use a flexible graphical model of tissues , the remaining micro-anatomic structures ( other cell types , extracellular matrix , bile ducts , etc . ) can be included incrementally without significant changes in our approach . In contrast to computationally intensive , spatially continuous approaches such as fluid dynamics , this graph-theoretic approach has hopefully sacrificed little physiologic detail but gained a great deal in terms of computational efficiency . Calculating hemodynamical flow on a graph allows rapid determination of flow given minimal boundary conditions , which will be especially useful for recalculating flow as morphology changes ( e . g . lesion formation ) or as individual sinusoids are temporarily blocked ( e . g . Kupffer cells ) . A faster dosimetry model allows the focus to center on cellular phenotypes , which are the key to modeling disease pathogenesis . A computationally-tractable approach allows for simulating the long run times associated with sub- and chronic toxicity studies as well as simulating large populations . We evaluated our approach to hepatic blood flow in three ways . First , we qualitatively tuned the appearance of the lobule to match actual physiology . Second , we compared the predicted pharmacokinetics for our spatially-extended lobule with traditional approaches , finding regimes in which our approach reduced to the well-mixed liver and the parallel tubes model . Third , we quantitatively compared the flow predicted for a rat with observations made in vivo of actual flow . Though all three lines of evaluations supported our approach , they also all pointed toward further refinements that may be necessary for simulating dose-response . This work addresses the dose portion of the dose-response curve , allowing assessment of how changes in exposure impact the hepatic lobule . The greater body of work remains with modeling response . Sufficiently complex models for hepatocellular dynamics , and eventually models for additional cell types , especially the Kupffer cells responsible for inflammatory responses , must be developed before we arrive at a useful model for homeostatic liver function . It remains to be seen whether three-dimensionality or even a departure from the classical lobule paradigm to simulate multiple lobules will be needed . To establish the safety of a compound one ideally finds the dose-response curve for various toxicity endpoints , so that an acceptable level of exposure can be determined . Currently the gold standard of toxicology is animal testing , but the need and desire for in vitro testing is growing . An in silico model for predicting dose-response would , at a minimum , provide a screen for prioritizing compounds that requiring further testing and perhaps may ultimately be able to predict in vivo consequences for the large number of compounds for which there is little or no toxicity data . The multiscale approach describe here is intended to be fast and verifiable , and would allow the determination of whether an observed in vitro response is relevant in vivo . The limitations in developing a homeostatic model of liver function are not computational , but biological . Additional data is needed , especially information on the statistical distribution of lobule morphology and the determination of cell state in response to local inputs . This model provides a framework for making use of two types of readily available data – histopathology slides and in vitro measures of cell function . In all likelihood direct comparison to liver toxicology data will be met initially with more failures than successes , but where we initially fail we will learn . Histopathology images have long been used to obtain information on microanatomic regions , vasculature , individual cells , cell types , and cell phenotypes from two- and three-dimensional images . Though traditionally time-intensive , advances in automated extraction of information from histopathology images are making it possible to analyze these images at a single cell resolution [37] , [38] , [39] . Additionally it is possible to extract information about the functional state of cells using cytomorphologic features or molecular markers [40] . Though cell-scale assay technology is still developing , it will be essential for fully calibrating and evaluating models such as this in order to provide simulated in vivo context for the results of in vitro assays . True variability in the response of a given hepatocyte is either a product of independent microdosimetry and cell variability , or is a function of the two , depending on the degree of correlation . To determine the significance of a chemical perturbation it is not enough to understand the cellular dynamics , but also the context in which those dynamics exist – i . e . , microdosimetry .
We have implemented a microdosimetry model for relating whole-body chemical exposures to cell-scale concentrations . The model is written in the freely available statistical language R , version 2 . 8 . 1 [41] . Given morphologic parameters Nt , the number of portal triads; Ns , the number of sinusoids per source/sink; Pbranch , the probability of a sinusoid branching; and Dmax , the size of the lobule , and calculating θCV is the angle to the central vein , given current position: Recursive Sinusoid Placement Algorithm ( SPA ) : The aggregation process is performed using the following algorithm: | Virtual tissues are emerging as a powerful tool for computational biology . By encoding known biology into a simulation of tissue function , gaps in knowledge can be identified . As a simulation of tissue function , in silico experiments can be performed inexpensively and rapidly . There are over 6000 chemicals produced in large quantities that may be present in our environment , many of which have not been thoroughly examined for human toxicity . Traditional toxicity testing is expensive , lengthy , and relies heavily upon the use of animals . For this reason in vitro toxicity testing techniques are being developed . However , techniques are needed to relate in vitro results to in vivo conditions . The liver is often the first tissue to show signs of toxicity and therefore a predictive liver toxicity simulator would be a powerful tool to reduce the financial and animal cost of toxicity testing . As a first step , we have developed a model for relating environmental exposure to cell-level concentrations; a model for virtual tissue microdosimetry . We identify regimes in which this approach is equivalent to previous techniques , as well as regimes where large cell-to-cell variability exists . This variability should have consequences both for normal liver function and the onset of injury . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"oncology",
"computer",
"science/applications",
"computational",
"biology/synthetic",
"biology",
"gastroenterology",
"and",
"hepatology/hepatology",
"biophysics/theory",
"and",
"simulation",
"pathology/histopathology",
"pharmacology/drug",
"development",
"physiology/gastroenterology",
... | 2010 | Simulating Microdosimetry in a Virtual Hepatic Lobule |
A SARS-CoV lacking the full-length E gene ( SARS-CoV-∆E ) was attenuated and an effective vaccine . Here , we show that this mutant virus regained fitness after serial passages in cell culture or in vivo , resulting in the partial duplication of the membrane gene or in the insertion of a new sequence in gene 8a , respectively . The chimeric proteins generated in cell culture increased virus fitness in vitro but remained attenuated in mice . In contrast , during SARS-CoV-∆E passage in mice , the virus incorporated a mutated variant of 8a protein , resulting in reversion to a virulent phenotype . When the full-length E protein was deleted or its PDZ-binding motif ( PBM ) was mutated , the revertant viruses either incorporated a novel chimeric protein with a PBM or restored the sequence of the PBM on the E protein , respectively . Similarly , after passage in mice , SARS-CoV-∆E protein 8a mutated , to now encode a PBM , and also regained virulence . These data indicated that the virus requires a PBM on a transmembrane protein to compensate for removal of this motif from the E protein . To increase the genetic stability of the vaccine candidate , we introduced small attenuating deletions in E gene that did not affect the endogenous PBM , preventing the incorporation of novel chimeric proteins in the virus genome . In addition , to increase vaccine biosafety , we introduced additional attenuating mutations into the nsp1 protein . Deletions in the carboxy-terminal region of nsp1 protein led to higher host interferon responses and virus attenuation . Recombinant viruses including attenuating mutations in E and nsp1 genes maintained their attenuation after passage in vitro and in vivo . Further , these viruses fully protected mice against challenge with the lethal parental virus , and are therefore safe and stable vaccine candidates for protection against SARS-CoV .
Coronaviruses ( CoVs ) are pathogens responsible for a wide range of existing and emerging diseases in humans and other animals [1] . A novel coronavirus causing the severe acute respiratory syndrome ( SARS-CoV ) was identified in Southeast China in 2002 . SARS-CoV rapidly spread worldwide to more than 30 countries within six months , infecting 8000 people and leading to death in approximately 10% of the cases [2 , 3] . While SARS-CoV has not reappeared in humans , CoVs including those similar to SARS-CoV , are widely disseminated in bats circulating all over the world , making future SARS-CoV outbreaks possible [4–7] . Furthermore , in September 2012 , a novel coronavirus infecting humans , the Middle East respiratory syndrome coronavirus ( MERS-CoV ) , was identified in two patients with severe respiratory disease in Saudi Arabia [8 , 9] , again indicating that emergence of other highly pathogenic CoVs is likely . Thus , development of efficacious and safe vaccines and anti-virus therapies for these pathogens is essential . SARS-CoV is an enveloped virus with a positive sense RNA genome of 29 . 7 kb that belongs to the Coronavirinae subfamily , genus β [2] . The virion envelope contains embedded three structural proteins , spike ( S ) , envelope ( E ) , and membrane ( M ) and several group specific proteins: 3a , 3b , 6 , 7a , and 7b [10–12] . The S protein , which mediates virus entry into host cells , the 3a protein and the M proteins , induce neutralizing antibodies , with those specific for S protein being most protective [13–16] . The SARS-CoV S and N proteins trigger T cell responses [17] , which are also important for protection and enhance the kinetics of virus clearance [18 , 19] . SARS-CoV E protein is a small integral membrane protein of 76 amino acids that contains a short hydrophilic amino-terminus followed by a hydrophobic transmembrane domain and a hydrophilic carboxy-terminus [20] . E protein oligomerizes to form an ion-conductive pore in membranes [21–23] , and contains a PDZ-binding motif ( PBM ) formed by its last four carboxy-terminal amino acids [24 , 25] . PDZ domains are protein-protein recognition sequences , consisting of 80 to 90 amino acids that bind to peptide sequences ( PBMs ) [26–28] . These protein-protein interactions modulate cellular pathways important for viral replication , dissemination in the host and pathogenesis [29] . We previously demonstrated that a SARS-CoV lacking the E gene ( SARS-CoV-∆E ) was attenuated in different animal models [30–34] , indicating that SARS-CoV E protein is a virulence factor . SARS-CoV lacking the E protein fully protected both young and elderly BALB/c mice against challenge with virulent mouse-adapted SARS-CoV [32]; therefore , rSARS-CoV-∆E is a promising vaccine candidate . Live attenuated vaccines are considered highly effective because of their ability to replicate within host cells , resulting in high levels of antigenic stimulation , and robust long-term immunological memory [35 , 36] . However , a major safety concern with live attenuated vaccines is the possibility of reversion to a pathogenic form . CoVs are prone to RNA recombination and mutation in tissue culture and during animal infection [37] , so it is crucial that rSARS-CoV-∆E be thoroughly studied after serial passage . In this study , we show that passage of viruses lacking all or part of the E protein in Vero E6 cells and mouse DBT-mACE2 cells [38] led to the incorporation of compensatory insertions . Interestingly , passage of SARS-CoVs lacking the E protein PBM led to regeneration of viral proteins containing PBMs , either by incorporation of novel chimeric proteins or by the insertion of new PBMs into existing proteins , such as the 8a protein or the E protein if E was only partially deleted . Strikingly , these modifications were not observed after passage of SARS-CoVs with mutated E protein that retained the PBM . In fact , we have shown that if instead of deleting the full-length E protein , only small deletions of 8–12 amino acids were introduced into the carboxy-terminus of E protein with retention of the PBM , the SARS-CoVs generated were attenuated and genetically stable both in cell culture or in mice . While this partial E protein deletion resulted in virus stability , we also augmented vaccine safety by introducing mutations into the SARS-CoV nsp1 protein . The nsp1 protein of CoVs suppresses host gene expression by inducing host mRNA degradation and inhibiting protein translation [39–44] , and is an IFN antagonist [45–47] . Nsp1 deletions resulted in attenuated murine coronaviruses that fully protected against the challenge with parental virus [48 , 49] . In this manuscript we show that small deletions within SARS-CoV nsp1 protein resulted in virus attenuation , associated with reduction of inflammation and higher levels of IFN-β and interferon-stimulated genes ( ISGs ) . Vaccination with mutated nsp1 variants protected against challenge with the virulent mouse-adapted SARS-CoV ( rSARS-CoV ) virus . To generate safer vaccine candidates , viruses incorporating deletions in both the nsp1 and E proteins were constructed . These double mutants were protective against virulent virus challenge , and were genetically stable .
To determine the stability of SARS-CoV-∆E , or of virus containing deletions of the E protein and several group specific genes including 6 , 7a , 7b , 8a , 8b and 9b ( SARS-CoV-∆[E , 6-9b] ) , we infected Vero E6 and DBT-mACE2 cells with rSARS-CoV , rSARS-CoV-∆E or rSARS-CoV-∆[E , 6-9b] . Supernatants were serially passaged 16 times and the distal third of the genome , from the S gene to the 3´ end ( around 8 kb ) , was sequenced using specific primers ( S1 Table ) . In all cases , an insertion consisting of a partially duplicated M gene fused to the SARS-CoV leader RNA sequence , a 5´ sequence common to coronavirus mRNAs [50–52] was detected upstream of the native M protein ( Fig 1A ) . In contrast , no chimeric proteins were detected after serial passage of the parental virus . All MCH genes encoded the amino terminus and the three transmembrane domains of M and also different PDZ-binding motifs at the carboxy-terminus of the protein ( Fig 1B ) . Genomic evolution occurred rapidly , as the chimeric genes were already detected within 5 passages in both cells lines , Vero E6 and DBT-mACE2 . The expression of viral sgmRNAs corresponding to the chimeric genes was characterized by RT-PCR ( S1A Fig ) . We used RNA harvested from infected cells after serial passage , plaque purification and amplification along with specific primers ( S2 Table ) . PCR products corresponding to specific MCH sgmRNAs were identified in MCH-Vero and MCH-DBT-infected cells ( S1B Fig ) . Expression of the chimeric proteins encoded by these sgmRNAs was confirmed using an antibody specific for all M and MCH proteins and a second one that recognized the MCH-DBT protein . Vero E6 cells were mock infected or infected with different recombinant viruses ( rSARS-CoV , rSARS-CoV-∆E , rSARS-CoV-∆E-MCH-Vero and rSARS-CoV-∆E-MCH-DBT ) at a moi of 0 . 3 . The expression of native M and MCH was confirmed at 24 hpi by Western blot analysis ( Fig 1C and 1D ) . These results indicated that , after serial passages of SARS-CoVs lacking the E protein in cell culture , a similar type of chimeric membrane protein was generated in three independent experimental settings . To test whether the presence of MCH genes conferred a replication advantage to SARS-CoV-∆E in vitro , the growth kinetics of SARS-CoV-∆E-MCH-Vero ( MCH-Vero ) and SARS-CoV-∆E-MCH-DBT ( MCH-DBT ) were analyzed . Vero E6 and DBT-mACE2 cells were infected with the recombinant viruses ( moi of 0 . 001 ) and viral titers were determined at the indicated hpi ( Fig 2 ) . MCH-Vero and MCH-DBT viruses showed lower titers at 24 hpi in Vero E6 cells compared to rSARS-CoV but both virus titers of the two chimeric viruses and rSARS-CoV were similar at 72 hpi . In contrast , a 100-fold decrease in viral growth was observed in rSARS-CoV-∆E-infected cells ( Fig 2 ) . Interestingly , chimeric proteins seemed to be specific for each cell type , as a ∆E virus containing a chimeric protein generated in Vero E6 cells ( MCH-Vero ) grew better in this cell line than in DBT-mACE cells , and the MCH-DBT virus generated in DBT-mACE2 cells specifically enhanced its growth in this cell line . This result indicated that the MCH protein provided a growth advantage for the virus , which partly compensated for the lack of the E protein . To determine the effect of the MCH protein in pathogenesis , BALB/c mice were intranasally infected with the recombinant viruses using 100 , 000 pfu , and weight loss and survival were monitored for 10 days ( Fig 3A ) . Mice infected with the parental virus ( wt ) showed signs of clinical disease at 2 days post infection ( dpi ) , reflected by ruffled fur , shaking , loss of mobility and weight loss , resulting in the death of all mice at 6 dpi ( Fig 3A ) . In contrast , mock-infected mice or mice infected with the ∆E virus , independently of whether the chimeric proteins ( ∆E , MCH-Vero and MCH-DBT ) were present or absent , did not lose weight and all of them survived without symptoms of disease ( Fig 3A ) . To analyze the effect of the MCH protein on virus growth in vivo , BALB/c mice were intranasally inoculated with the recombinant viruses and euthanized at 2 and 4 dpi . Virus titers in the lungs were determined ( Fig 3B ) . Viruses lacking the E protein , in the presence or absence of MCH protein , grew to lower titers in lungs at 2 and 4 dpi , as compared with those observed in mice infected with rSARS-CoV . Notably , rSARS-CoV-∆E replicated in the lung to higher levels than those containing the chimeric proteins ( MCH-Vero and MCH-DBT ) at both days p . i . Chimeric proteins only increased fitness in a cell type-dependent manner , i . e . , viruses with the chimeric protein only grew better in the cell system in which this chimeric protein was generated . The virus containing a chimeric protein generated in Vero E6 cells grew better in this cell line , and the virus containing a chimeric protein generated in DBT-mACE2 cells showed an increased growth in these cells ( Fig 2 ) . Lungs of mice infected with rSARS-CoV were highly edematous and showed profuse hemorrhagic areas at 2 and especially at 4 dpi ( S2A Fig ) , leading to a significant increase in lung weight at 4 dpi ( S2B Fig ) . Lung sections from mock-infected mice or mice infected with rSARS-CoV-∆E , MCH-Vero and MCH-DBT ( Fig 3C ) showed minimal damage at 2 and 4 dpi . In contrast , analysis of the lungs of mice infected with rSARS-CoV revealed extensive inflammatory cell infiltration and edema in alveolar and bronchiolar airways ( Fig 3C ) . The chimeric protein MCH was generated after rSARS-CoV-∆E passage in cell culture , but not when full-length E protein was present , suggesting that it compensated for functions originally performed by E protein . To identify such E protein functional domains , a set of recombinant SARS-CoVs ( Fig 4A ) , with mutations or deletions in different regions of E protein [23 , 24 , 53] , was passaged 16 times in Vero E6 cells ( Fig 4 ) . In Mut 1 , several amino acid substitutions were introduced at the E protein amino-terminal region . rSARS-CoV deletion mutants ∆2 , ∆3 , ∆4 , ∆5 and ∆6 included sequential or partially overlapping small deletions of 6 to 12 amino acids in the carboxy-terminus of E protein . Interestingly , the last 6 amino acids within the Δ6 virus ( YSRVKN; Fig 4 ) revealed an alternative PBM at the carboxy-terminal domain of the protein [53] . In recombinant ∆PBM , the last 9 amino acids of E protein were deleted , truncating the carboxy-terminus and eliminating the E protein PBM . In mutPBM , the PBM was abolished by mutating the last 4 amino acids to glycine , maintaining the full-length E protein . In contrast , in altPBM , 4 amino acids within E protein carboxy-terminal region were mutated to alanine , maintaining an active PBM domain [24] . In SARS-CoV N15A and V25F mutants , ion channel activity of E protein was abolished by one point mutation in the transmembrane domain ( Fig 4A ) . Vero E6 cells were infected with each of the recombinant viruses at a moi of 0 . 5 and supernatants were serially passaged for 16 times , and the presence of MCH gene was determined by sequence analysis . The results indicated that the MCH was not generated when the E protein contained a PBM sequence ( rSARS-CoV , Mut 1 , ∆2 , ∆3 , ∆4 , ∆5 , ∆6 , altPBM , N15A and V25F ) ( Fig 4B ) . When the PBM was absent ( ∆PBM and mutPBM ) , a new PBM containing the original sequence was added to the carboxy-terminal end of mutated E protein in all cases , reinforcing the importance of the PBM domain during infection . A virus incorporating the chimeric MCH protein was only generated when E protein was completely deleted ( ∆E ) , i . e . , when the restoration of a PBM on the E protein was not possible . To further analyze whether SARS-CoV requires a transmembrane protein containing a PBM , we generated two recombinant rSARS-CoV-∆E that contained artificial chimeric proteins ( Fig 4C ) . In SARS-CoV-∆E-MCH-EPBM ( MCH-EPBM ) , a chimeric protein containing the first transmembrane domain of the M protein fused to the last nine amino acids of E protein , encompassing the PBM , was introduced . In SARS-CoV-∆E-3aCH-3aPBM ( 3aCH-3aPBM ) , the chimeric protein was formed by the first transmembrane domain of 3a protein and a PBM composed by the last nine amino acids of 3a protein ( Fig 4C ) . Both viruses were passaged 16 times in Vero E6 cells and compensatory mutations were not detected after sequencing . All these data indicated that the virus requires a transmembrane protein displaying a PBM and that novel proteins with a PBM compensate for the loss of the E protein PBM . MCH-Vero and MCH-DBT exhibited an attenuated phenotype ( see above ) . To analyze the genetic stability of recombinant SARS-CoV-∆E in BALB/c mice , virus was passaged every 48 hours by intranasal inoculation . A partial duplication of 45 nucleotides was found within 8a gene ( Fig 5 ) , leading to the incorporation of a fragment of 15 amino acids at the carboxy-terminus of 8a protein , and generating a novel 8a protein ( 8a-dup ) with an internal PBM ( CTVV ) ( Fig 5A and 5B ) . To determine the virulence of this novel virus ( SARS-CoV-∆E-8a-dup ) , we infected a new cohort of BALB/c mice and assessed survival and weight loss , and measured virus titers . Mice infected with SARS-CoV-∆E did not lose weight and all survived . In contrast , SARS-CoV-∆E-8a-dup grew to titers similar to those of rSARS-CoV and developed profound weight loss , developing signs of illness and death by 7 dpi ( Fig 6A and 6B ) . Histological examination of lungs from SARS-CoV-∆E-infected mice showed absence of lung damage at both dpi . In contrast , lungs of mice infected with rSARS-CoV or SARS-CoV-∆E-8a-dup showed substantial perivascular , peribronchial and interstitial cellular infiltration and edema at 2 and 4 dpi ( Fig 6C ) . To further confirm the relevance of the partial duplication within 8a gene in the induction of virulence , a recombinant SARS-CoV-∆E with an 8a-dup gene was generated ( rSARS-CoV-∆E-8a-dup ) . Virulence during infection with rSARS-CoV-∆E-8a-dup was evaluated as described above ( Fig 6D ) . The engineered virus ( rSARS-CoV-∆E-8a-dup ) was as virulent as the SARS-CoV-∆E-8a-dup generated after SARS-CoV-∆E passage in vivo ( Fig 6D ) . These results indicated that SARS-CoV-∆E regained virulence after serial passage in mice . SARS-CoV infection is associated with p38 mitogen-activated protein kinase ( MAPK ) activation and elevated levels of pro-inflammatory cytokines [24 , 54–56] . To begin to determine the basis of increased virulence exhibited by SARS-CoV-∆E-8a-dup , p38 MAPK activation and pro-inflammatory cytokine expression were analyzed during infection ( Fig 7 ) . p38 MAPK activation was analyzed by Western blot at 24 hpi using a phospho-p38 MAPK ( p-p38 ) specific antibody . Antibodies recognizing the total endogenous p38 MAPK and actin were used as controls . A significant increase in p38 MAPK activation , assessed at 24 hpi using a phospho-p38 MAPK ( p-p38 ) -specific antibody and Western blot analysis , was observed in SARS-CoV-∆E-8a-dup-infected compared to SARS-CoV-∆E or mock-infected cells ( Fig 7A and 7B ) . To test whether pro-inflammatory cytokine expression was induced during SARS-CoV-∆E-8a-dup infection , we analyzed the expression of several genes previously associated with SARS-CoV pathology [24 , 57] including: chemokine ( C-X-C motif ) ligand 10 ( CXCL10 ) , chemokine ( C-C motif ) ligand 2 ( CCL2 ) and interleukin 6 ( IL6 ) ( S3 Table ) . 18S ribosomal RNA was used to normalize the data , as previously described [58 , 59] . BALB/c mice were mock-infected or infected with 100 , 000 pfu of recombinant SARS-CoV , and lungs were collected at 2 dpi . A significant increase in the expression of all pro-inflammatory cytokines in the lungs was observed during infection with virulent viruses ( rSARS-CoV and SARS-CoV-∆E-8a-dup ) ( Fig 7C ) . In contrast , infection with the recombinant virus lacking E protein at passage 0 ( SARS-CoV-∆E ) did not induce the expression of pro-inflammatory cytokines . Activation of p38 MAPK and pro-inflammatory cytokines expression during infection with rSARS-CoV-∆E-8a-dup were evaluated as described above . The engineered virus ( rSARS-CoV-∆E-8a-dup ) induced similar p38 MAPK activation ( Fig 7D and 7E ) and overexpression of proinflammatory cytokines ( Fig 7F ) as compared with the SARS-CoV-∆E-8a-dup generated after SARS-CoV-∆E passage in vivo . These data indicated that reversion of SARS-CoV-∆E-8a-dup to a virulent phenotype was associated with an exacerbated immune response similar to that triggered during infection with the rSARS-CoV . SARS-CoV-ΔE reverted to a virulent phenotype after serial passages in mice . To increase the genetic stability of the vaccine candidate , we introduced small attenuating deletions in the E gene , instead of deleting of the full-length E protein [53] as described above . In addition , to increase vaccine biosafety , we introduced additional attenuating mutations within the SARS-CoV nsp1 gene . To identify domains within the SARS-CoV nsp1 protein that could contribute to virulence , we compared the sequence to that of the MHV nsp1 ( Fig 8A ) , with the goal of identifying conserved regions that could be functionally important . Based on this information , four mutant viruses ( rSARS-CoV-nsp1* ) were generated by introducing deletions of 8 to 11 amino acids into the nsp1 protein ( rSARS-CoV-nsp1-∆A , -∆B , -∆C and -∆D Fig 8A ) . All the deletion mutants grew to similar titers as rSARS-CoV in Vero E6 cells ( Fig 8B ) . However the ∆A and ∆B mutants grew to lower titers in DBT-mACE2 cells . To analyze the effects of these deletions in vivo , mice were intranasally infected with mutants rSARS-CoV-nsp1-∆A , -∆B , -∆C and -∆D , and daily monitored for 10 days . SARS-CoV-nsp1-∆C and -∆D infected mice transiently lost a small amount of weight and all mice survived . In contrast , mice infected with SARS-CoV lost weight , and all died by day 5 ( Fig 9A ) . Mice infected with rSARS-CoV-nsp1-∆A or rSARS-CoV-nsp1-∆B lost 20 and 15% of their initial weight by day 3 , with survival reduced to 60 and 80% , respectively ( Fig 9A ) . These data indicated that deletion of regions C and D within nsp1 protein led to attenuated mutants , whereas deletion of regions A and B were only partially attenuating . With the exception of rSARS-CoV-nsp1-∆A at day 2 p . i . , virus titers were reduced compared to rSARS-CoV-infected mice ( Fig 9B ) . There was not a strict correlation between virus titers and virulence , possibly because nsp1 is involved in the countering IFN production after infection . No significant changes on gross inspection of lungs or in their weight were observed when the lungs of mock-infected and SARS-CoV-nsp1-∆C and -∆D-infected mice were compared ( Figs 9C and S3 ) . In contrast , lungs from mice infected with the rSARS-CoV and , to a much lower extent SARS-CoV-nsp1-∆A and -∆B-infected mice lungs , showed evidence of hemorrhage ( S3 Fig ) . In addition , rSARS-CoV-infected mice showed lung weight increase , not observed with the lungs of SARS-CoV-nsp1*-infected mice . Compared to mock-infected mice , lungs from rSARS-CoV-infected mice showed clear inflammatory infiltrates and alveolar and bronchiolar edema ( Fig 9C ) . In contrast , mice infected with the viruses rSARS-CoV-nsp1* showed no ( SARS-CoV-nsp1-∆C and -∆D ) , or minimal ( SARS-CoV-nsp1-∆A and -∆B ) lung damage . These data correlated well with the virulence observed for the SARS-CoV-nsp1* mutants , showing that the most attenuated viruses were those that induced less lung pathology in vivo . Since nsp1 has anti-interferon activity , we next measured expression of IFN and IFN-stimulated genes after infection with rSARS-CoV , rSARS-CoV-nsp1-∆C and -∆D . We focused on the ∆C and ∆D viruses , because these viruses were fully attenuated and had an efficient growth in vivo . SARS-CoV-nsp1-∆C and -∆D induced higher levels of IFN-β and ISGs ( IRF1 , DDX58 , and STAT1 ) , compared to mock-infected and rSARS-CoV-infected cells ( Fig 10A and 10B ) . This effect was specific , as the expression of control 18S rRNA was the same in virus-infected cells or mock-infected cells ( Fig 10B ) . These results indicated that deletion of regions C and D of nsp1 restored IFN responses , leading to virus attenuation . To analyze whether SARS-CoV-nsp1-∆C and -∆D induced protective immune responses , mice were intranasally vaccinated with SARS-CoV-nsp1-∆C and -∆D , and challenged 21 days later with rSARS-CoV . After challenge , mock-vaccinated mice rapidly lost weight , and all mice died by day 6 ( Fig 11A and 11B ) . In contrast , mice immunized with SARS-CoV-nsp1-∆C and -∆D viruses , did not significantly lose weight , and 100% survived the challenge ( Fig 11A and 11B ) . In order to develop a safe vaccine candidate , mutant viruses with deletions in both nsp1 and E genes were engineered . A rSARS-CoV deleted in the nsp1 D domain and the E protein ( SARS-CoV-nsp1ΔD-ΔE ) , and a second mutant virus with deletions of the nsp1 D domain coupled with a small deletion ( E∆3 ) ( Fig 4A ) in the E protein ( SARS-CoV-nsp1ΔD-EΔ3 ) were generated . EΔ3 deletion mutant was selected for further studies because this deletion led to a virus that grew to titers similar or higher than the SARS-CoV-∆E , in cell culture or in mice , respectively ( Fig 12A and 12B ) . More importantly , the E∆3 virus was genetically stable after 10 passages in cell culture [53] or in vivo , maintaining its attenuated phenotype ( Fig 12C ) , in contrast to the ∆E virus ( Figs 1 and 5 ) . This deletion was combined with another one in SARS-CoV nsp1 protein ( nsp1ΔD ) , which was fully attenuating . The resulting virus grew to relatively high titers in vivo ( Figs 9 and 13A ) . Viruses were rescued in Vero E6 cells , cloned and sequenced to confirm the presence of the desired mutations . To analyze the stability of the viruses in tissue culture cells , SARS-CoV-nsp1ΔD-ΔE and SARS-CoV-nsp1ΔD-EΔ3 were passaged 10 times in Vero E6 cells , followed by sequencing of the nsp1 and E genes . The deletions introduced in both nsp1 and E genes were conserved , suggesting that these deletions were genetically stable in vitro . SARS-CoV-nsp1ΔD-ΔE and SARS-CoV-nsp1ΔD-EΔ3 at passage 1 reached peak titers at 72 hpi ( 5·105 pfu/ml and 5·104 pfu/ml in Vero E6 and DBT-mACE2 cells , respectively ) ( Fig 13B ) . Decreased virus growth was likely due to deletions in the E gene , as previously described [53] ( Fig 8B ) . After 10 passages , viruses showed a slight increase in titer ( Fig 13B ) , suggesting incorporation of additional mutations but not in E or nsp1 . To analyze the pathogenicity of SARS-CoV-nsp1ΔD-ΔE and SARS-CoV-nsp1ΔD-EΔ3 mutants at passages 1 and 10 ( p1 and p10C , respectively ) , BALB/c mice were intranasally inoculated with recombinant viruses . All mice infected with these viruses maintained their weight and survived ( Fig 14A ) . In contrast , all mice infected with rSARS-CoV died ( Fig 14A ) . These results indicated that viruses including deletions in both nsp1 and E proteins were attenuated in vivo . SARS-CoV-nsp1ΔD-EΔ3 , especially the p10C virus , grew more efficiently than SARS-CoV-nsp1ΔD-ΔE ( Fig 14B ) . Nevertheless , no obvious gross lesions or changes in weight were observed in the lungs of mice infected with any of these doubly mutant rSARS-CoV ( Figs 14C and S4 ) . In contrast , mice infected with rSARS-CoV showed lung injury and a marked increase in the weight of lungs , as described above ( S4 Fig ) . Histological examination of lungs from SARS-CoV-nsp1ΔD-ΔE and SARS-CoV-nsp1ΔD-EΔ3 ( p1 and p10C viruses ) -infected mice showed only minimal evidence of damage or leukocyte infiltration at days 2 and 4 post-infection ( Fig 14C ) while rSARS-CoV-infected mice showed extensive cellular infiltration and edema . The rSARS-CoV-nsp1∆D-E∆3 was selected for further study because this virus showed higher titers in vivo as compared with the rSARS-CoV-nsp1∆D-∆E virus ( Fig 14B ) , therefore it could promote a higher immunization . In addition , ∆E mutation led to unstable viruses that incorporated new chimeric proteins in cell culture , or a novel 8a protein in vivo , causing reversion to a virulent phenotype ( Fig 6 ) . In contrast , viruses containing the E∆3 mutation remained stable after 10 passages and maintained their attenuated phenotype ( Fig 12 ) . To analyze the stability of rSARS-CoV with mutations in nsp1 and E in mice , SARS-CoV-nsp1ΔD-EΔ3 was passaged 10 times ( p10M ) , followed by sequencing from the S gene to the 3´ end of the genome . Only two changes were observed in the viral sequence , one in the E gene ( A26250T N→I ) , and a second one in the M gene ( A26450G Q→R ) . The deletions introduced in the nsp1 and E genes were conserved , suggesting that this virus was essentially genetically stable in vivo . To analyze whether the E and M mutations could be compensatory mutations , virus titers at p1 and p10 were compared in Vero E6 and DBT-mACE2 cells . Titers of SARS-CoV-nsp1ΔD-EΔ3 ( p10M ) at different times post-infection were the same as those observed for rSARS-CoV ( Fig 15A ) , indicating that the mutations increased virus replication . To evaluate whether the compensatory mutations restored the pathogenicity of the virus , BALB/c mice were intranasally inoculated with rSARS-CoV and SARS-CoV-nsp1ΔD-EΔ3 ( p10 ) , and were daily monitored for 10 days . Mice infected with rSARS-CoV started to lose weight by day 2 , and died by day 6 ( Fig 15B ) . In contrast , although mice infected with the SARS-CoV-nsp1ΔD-EΔ3-p10M mutant initially lost 10% of their weight , at day 5 the mice started to regain weight , fully recovered from the disease , and 100% survived ( Fig 15B ) . SARS-CoV-nsp1ΔD-EΔ3-p10M grew to similar titers as rSARS-CoV ( Fig 15C ) . Nevertheless , unlike the parental virus , lungs of SARS-CoV-nsp1ΔD-EΔ3-p10M-infected mice presented no significant increase of inflammatory cytokines . In addition , no obvious lung lesions , weight increases , nor substantial inflammatory cell infiltration , as determined by macroscopic and histological examination , were observed ( S5 Fig ) . To further support the stability and attenuation of the double mutant virus , additional passages ( up to 20 ) were conducted in mice . Evaluation of the passaged virus virulence showed that rSARS-CoV-nsp1∆D-E∆3 remained attenuated ( Fig 15D ) . These results indicated that despite the mutations that the virus incorporated after their passage in mice , the virus maintained the in vivo attenuated phenotype . To determine whether SARS-CoV-nsp1-ΔD-EΔ3 confers protection against challenge with rSARS-CoV , BALB/c mice were either immunized with SARS-CoV-nsp1ΔD-EΔ3-p1 , -p10C and -p10M or non-immunized , as a control . At 21 days postimmunization , mice were challenged with rSARS-CoV administered by the same route . Non-immunized mice lost weight and all died on day 6 after the challenge ( Fig 16A and 16B ) . In contrast , vaccination with the attenuated mutant viruses completely protected mice against challenge with rSARS-CoV ( Fig 16A and 16B ) , indicating that the double mutant virus is a promising vaccine candidate .
We have previously shown that deletion of SARS-CoV E gene leads to an attenuated virus that is a promising vaccine candidate [30–34] . However , since safety and stability are main concerns of live attenuated vaccine candidates , we focused on rSARS-CoV-∆E stability in vitro and in vivo and on the generation of a safe vaccine candidate by identifying the mechanisms of reversion to virulence . Unexpectedly , serial passage of rSARS-CoV-∆E in cell culture resulted in the generation of chimeric proteins composed of a partial duplication of the membrane gene fused to a part of the leader sequence . Our results are in agreement with a recombinant MHV lacking the E protein ( rMHV-∆E ) that was viable but its replication was drastically impaired [60] . rMHV-∆E replicated to 10 , 000-fold lower titers than the parental virus and remarkably , evolved similarly to SARS-CoV-∆E after serial passage in tissue culture cells [61] . Despite that the generation of a chimeric protein was previously observed in MHV when E gene was deleted [61] , the presence of a PBM motif within the inserted chimeric sequence and its main role in providing genetic stability to SARS-CoV-ΔE during passage described in this manuscript , were not previously noticed . Alteration of coronavirus genome was not unexpected due to the high frequency of RNA recombination and mutation described for these viruses , both in cell culture and in animals [37 , 62–64] . All chimeric SARS-CoV M proteins generated in cell culture were expressed , enhancing virus growth in cell culture compared to rSARS-CoV-∆E . Similarly , rMHV-∆E that contained chimeric proteins also showed significant increases in viral yields [61 , 65] . In contrast , mice infected with rSARS-CoVs containing chimeric proteins generated in cell culture showed a decrease in viral titers in the lungs of infected mice even when compared with rSARS-CoV-∆E virus . Similarly , extensive passage in cell culture of other CoVs , including porcine epidemic diarrhea virus ( PEDV ) and transmissible gastroenteritis virus ( TGEV ) , led to less pathogenic strains compared to wild-type viruses , possibly due to the emergence of deletion mutants that lost sequence domain not needed for their growth in cell culture , but that influenced their tropism and in vivo replication [66–68] . Chimeric proteins containing a PBM inserted in the viral genome after passage increased viral fitness in cell culture in a tissue specific manner , i . e . , the chimeric protein inserted into viral genome during passage in monkey cells promoted virus growth in cells from this species , whereas the one inserted during passage in murine cells specifically increased virus fitness in cells from mice . Furthermore , these chimeric proteins did not enhance viral growth nor virulence in vivo . These data indicated that the insertion of chimeric proteins specifically adapted the virus for an optimum growth in cell culture but did not enhance in vivo growth nor virulence . In this context , it is also important to note that the activity of PBMs is dependent on their specific sequence , and also on the sequence context in which they are inserted [27 , 28 , 69 , 70] . Serial passage of rSARS-CoV-∆E in mice introduced a partial duplication of 45 nucleotides in the 8a protein , resulting in its reversion to a virulent phenotype . This phenotype was associated to its ability to activate p38 MAPK and to the induction of inflammatory cytokine expression and increased lung damage , as previously described [24 , 71] . 8a protein is a short transmembrane protein composed of 39 amino acids that forms cation-selective ion channels [72] . SARS-CoV variants with deletions in 8a ORF , have been transmitted and maintained in humans in the late phases of SARS-CoV epidemic [73 , 74] . Interestingly , an 8a protein mutant generated during virus passage in vivo contained a new potential PBM ( CTTV ) localized in the internal region of the carboxy-terminal domain of the protein ( Fig 5 ) . Despite the CTVV sequence was already present in the original 8a protein , it most likely does not represent a functional PBM , as it is located within the transmembrane domain of the 8a protein . Active PBMs are in general located in exposed regions of the proteins , usually the end of the carboxy-terminus or , exceptionally , in internal positions within the carboxy-terminal domain , allowing their interaction with PDZ domains , such as it has been observed in the NS5 proteins of tick-borne encephalitis virus ( TBEV ) and Dengue virus [75 , 76] . However , PBMs forming part of a transmembrane domain are not accessible to PDZ-containing proteins and have not been described [29] . Therefore , the new CTVV sequence placed in an exposed environment may constitute a novel and active PBM . The PBM insertion within 8a protein after passaging in vivo could be due to the fact that the ORF8 is one of the regions where most variations were observed between human and animal isolates of SARS-CoV [4] . In fact , a complete genome sequence of SARS-like coronaviruses in bats isolates showed the presence of a PBM within the ORF8 [5] . As mentioned above , species of bats are a natural host of coronaviruses closely related to those responsible for the SARS outbreak . PDZ domains are among the modules most frequently involved in protein-protein interactions found in all metazoans [69] . In the human genome , there are more than 900 PDZ domains in at least 400 different proteins [77] . Many pathogenic viruses produce PDZ ligands that disrupt host protein complexes for their own benefit , such as hepatitis B virus , influenza virus , rabies virus and human immunodeficiency virus , influencing their replication , dissemination in the host , transmission and virulence [29] . Generation of new proteins containing a PBM after passaging in cell culture and in mice may affect their interaction with a wide range of cellular PDZ-containing proteins , affecting diverse biological functions with high relevance in pathogenesis . E protein PBM participates in two different and independent issues , virus stability and virulence . Our data suggest that when the PBM was present in a proper environment at the end of the E protein , either a native or a mutant protein , viruses remained stable . An independent observation is that the presence of a PBM within E protein confers pathogenicity to the virus [24] . This virulence is prevented either by PBM removal [24] or by the introduction of small deletions within the carboxy-terminus of E protein [53] , which by themselves may cause attenuation or , alternatively , by indirectly affecting the PBM . Our results highlighted the critical requirement of viral proteins containing a PBM in the generation of CoVs with virulent phenotypes , and opened up new approaches for the rational design of genetically stable vaccines . Maintaining the attenuated phenotype of the vaccine candidate after passage in vitro was crucial to avoid the reversion to a virulent phenotype during the design and production of a genetically stable vaccine candidate . To this end , the identification of the relevance of the presence of a functional PBM motif at the carboxy-terminus of a transmembrane protein of the virus has been instrumental in the development of a stable SARS-CoV vaccine candidate . To minimize the risk of regain of virulence after passage , we engineered viruses with small deletions in E gene , instead of deletion of the entire E gene . By preserving the PBM , we observed no evidence for the development of chimeric proteins and thus no gain in virulence . As additional measures to ensure safety of this live attenuated vaccine candidate , we incorporated attenuating mutations into nsp1 , in the context of the rSARS-CoV-ΔE or EΔ3 . Nsp1 was chosen as a second attenuation target because this gene is located at a distant site ( >20 kb ) from that of the E gene in the viral genome , making it very unlikely that a single recombination event with a circulating wt coronavirus could result in the restoration of a virulent phenotype . To analyze the role of SARS-CoV nsp1 in the pathogenesis of the virus , recombinant viruses encoding four different small deletions were generated . Deletion of amino acids 121–129 and 154–165 , in the carboxy terminal region of nsp1 led to virus attenuation , indicating that nsp1 enhanced virus pathogenicity , as was previously shown for MHV [45 , 48 , 49] . Interestingly , these attenuated mutants grew in mice to lower titers than rSARS-CoV , probably by inducing higher IFN responses , indicating that these regions of nsp1 are critical for IFN antagonism . The induction of a higher innate immune response by the nsp1 deletion is most probably responsible for the decrease in SARS-CoV-nsp1* virus titers observed in mice and , to a lesser extent , in DBT-mACE–2 cells . In fact , a rSARS-CoV lacking the nsp1 protein grew poorly in IFN competent cells , but replicated as efficiently as the wt virus in IFN deficient cells [46] , consistent with our findings . Similarly , titers of MHV deleted in nsp1 are restored almost to wild type levels in type I IFN receptor-deficient mice [48] . Immunization with singly deleted rSARS-CoV protected mice against challenge with rSARS-CoV , as it was previously shown with MHV nsp1-deletion mutants [48 , 49] . SARS-CoV-nsp1ΔD-EΔ3 , which contained deletions in nsp1 and E protein , maintained its attenuated phenotype after passage in Vero E6 cells and in mice . In addition , immunization with this double mutant fully protected mice from challenge with the parental virulent virus , indicating that it is a promising vaccine candidate in terms of both stability and efficacy . Both humoral and cellular responses are relevant to protect from SARS [18 , 19 , 78 , 79] . The viruses generated in this work express all viral proteins , except for small regions deleted in the E and nsp1 proteins , therefore have the potential of inducing both antibody and T cell responses , making this type of live vaccine more attractive than subunit or non replicating virus vaccines . Understanding of the molecular mechanisms by which an attenuated SARS-CoV reverted to a virulent phenotype could also be applied to the development of other relevant CoVs vaccines , such as MERS-CoV .
Animal experimental protocols were approved by the Ethical Committee of The Center for Animal Health Research ( CISA-INIA ) ( permit numbers: 2011–009 and 2011–09 ) in strict accordance with Spanish National Royal Decree ( RD 1201/2005 ) and international EU guidelines 2010/63/UE about protection of animals used for experimentation and other scientific purposes and Spanish national law 32/2007 about animal welfare in their exploitation , transport and sacrifice and also in accordance with the Royal Decree ( RD 1201/2005 ) . Infected mice were housed in a ventilated rack ( Allentown , NJ ) . The mouse-adapted ( MA15 ) [71] parental virus ( wt ) , and recombinant viruses were rescued from infectious cDNA clones generated in a bacterial artificial chromosome ( BAC ) in our laboratory [32 , 33 , 53 , 80] . Vero E6 and BHK cells were kindly provided by E . Snijder ( University of Leiden , The Netherlands ) and H . Laude ( Unité de Virologie et Immunologie Molecularies , INRA , France ) , respectively . The mouse delayed brain tumor ( DBT ) cells expressing the murine receptor ( ACE2 ) for SARS-CoV ( DBT-mACE2 ) were generated in our laboratory [38] . In all cases , cells were grown in Dulbecco's modified Eagle's medium ( DMEM , GIBCO ) supplemented with 25 mM HEPES , 2 mM L-glutamine ( SIGMA ) , 1% non essential amino acids ( SIGMA ) and 10% fetal bovine serum ( FBS , Biowhittaker ) . Virus titrations were performed in Vero E6 cells as previously described [33] . 8 week-old specific-pathogen-free BALB/c Ola Hsd mice females were purchased from Harlan Laboratories . BALB/c mice were maintained for 8 additional weeks in the animal care facility at the National Center of Biotechnology ( Madrid ) . For infection experiments , mice were anesthetized with isoflurane and intranasally inoculated at the age of 16 weeks with 100 , 000 plaque forming units ( pfu ) of the indicated viruses . All work with infected animals was performed in a BSL3 laboratory ( CISA , INIA ) . Mutant viruses ( SARS-CoV-nsp1* ) with small deletions covering different regions of nsp1 protein ( SARS-CoV-nsp1-∆A , -∆B , -∆C and -∆D ) , were constructed using an infectious cDNA clone . cDNA encoding the genome of SARS-CoV-MA15 strain was assembled in a bacterial artificial chromosome ( BAC ) ( plasmid pBAC-SARS-CoV-MA15 ) [32 , 57 , 80] . DNA fragments containing nucleotides 8142 to 9211 , comprising the nsp1 gene of the SARS-CoV genome were generated by overlap extension PCR using as template the plasmid pBAC-SARS-CoV-MA15 and the primers indicated in S4 Table . The final PCR products were digested with the enzymes AvrII and BstBI and cloned into the intermediate plasmid pBAC-SfoI-MluI-SARS-CoV that contains the first 7452 nucleotides of the SARS-CoV infectious cDNA clone [80] , to generate plasmids pBAC-SfoI-MluI SARS-CoV-nsp1* ( pBAC-∆A , -∆B , -∆C , and -∆D ) [80] . The plasmids pBAC-SfoI-MluI-SARS-CoV-nsp1* were digested with the restriction enzymes SfoI and MluI and the fragments were inserted into the plasmid pBAC-SARS-CoV-MA15 , digested with the same restriction enzymes , to generate pBAC-SARS-CoV-MA15-nsp1* plasmids . Mutant viruses SARS-CoV-nsp1ΔD-ΔE and SARS-CoV-nsp1ΔD-EΔ3 were generated using the plasmids pBAC-SARS-CoV-MA15-ΔE and -EΔ3 [33 , 53] . The plasmids were digested with the enzymes BamHI and RsrII and the digested fragments were exchanged with the fragment of plasmid pBAC-SARS-CoV-MA15-nsp1∆D , to generate pBAC-SARS-CoV-MA15-nsp1ΔD-ΔE and pBAC-SARS-CoV-MA15-nsp1ΔD-EΔ3 plasmids . Two fragments representing the nucleotides containing the chimeric proteins MCH-EPBM and 3aCH-3aPBM were chemically synthesized ( BioBasic Inc ) to generate SARS-CoV mutants . The final PCR products and synthesis fragments were digested with enzymes BamHI and MfeI and cloned into the intermediate plasmid psl1190+BamHI/SacII-SARS-CoV to generate the plasmids psl1190-∆E-MCH-EPBM and psl1190-∆E-3aCH-3aPBM . The plasmid psl1190+BamHI/SacII SARS-CoV contains a fragment corresponding to nucleotides 26045 to 30091 of the SARS-CoV infectious cDNA clone engineered into plasmid psl1190 ( Pharmacia ) [80] . These constructs were cloned in the infectious pBAC-SARS-CoV-MA15-∆E with the enzymes BamHI and SacII . Mutant virus rSARS-CoV-∆E-8a-dup with small duplication of 8a protein was constructed using an infectious cDNA clone . cDNA encoding the genome of SARS-CoV-MA15-∆E strain was assembled in a bacterial artificial chromosome ( BAC ) ( plasmid pBAC-SARS-CoV-MA15-∆E ) [32 , 57 , 80] . DNA fragments containing nucleotides 27779 to 27898 , comprising the 8a gene of the SARS-CoV genome were generated by overlap extension PCR using as template the plasmid pBAC-SARS-CoV-MA15 and the primers indicated in S4 Table . The final PCR products were digested with the enzymes XcmI and NheI and cloned into the intermediate plasmid pBAC-BamHI-NheI-SARS-CoV that contains the nucleotides ( nt ) 26044 to 28753 nucleotides of the SARS-CoV infectious cDNA clone [80] , to generate plasmid pBAC-BamHI-NheI-SARS-CoV-∆E-8a-dup [80] . The plasmid pBAC-BamHI-NheI-SARS-CoV-∆E-8a-dup was digested with the restriction enzymes BamHI and NheI and the fragments were inserted into the plasmid pBAC-SARS-CoV-MA15 , digested with the same restriction enzymes , to generate pBAC-SARS-CoV-MA15-∆E-8a-dup plasmid . The viruses were rescued in BHK and Vero E6 cells as previously described [33] . Viruses were cloned by three rounds of plaque purification . Subconfluent monolayers ( 90% confluency ) of Vero E6 and DBT-mACE2 on 12 . 5 cm2 flasks were infected at a multiplicity of infection ( moi ) of 0 . 001 with the indicated viruses . Culture supernatants were collected at 0 , 4 , 24 , 48 and 72 hpi and virus titers were determined as previously described [33] . BALB/c mice were anesthetized with isoflurane and intranasally inoculated with 100 , 000 plaque forming units ( pfu ) of virus in 50 μL of DMEM . Weight loss and mortality were evaluated daily . For protection experiments mice were immunized intranasally with 6000 pfu of the attenuated viruses , and then challenged with an intranasal inoculation of 100 , 000 pfu of SARS-CoV at 21 days post-immunization . Mice were monitored daily for weight loss and mortality . To determine SARS-CoV titers , lungs were homogenized in PBS containing 100 UI/ml penicillin , 0 . 1 mg/ml streptomycin , 50 μg/ml gentamicin , and 0 . 5 μg/ml amphotericin B ( Fungizone ) , using a gentleMACS dissociator ( Miltenyi Biotec ) and virus titrations were performed in Vero E6 cells as described above . Viral titers were expressed as pfu/g tissue . Mice were sacrificed at 2 and 4 dpi . Lungs were removed , fixed in 10% zinc formalin for 24 h at 4°C and paraffin embedded . Histological examination was performed using hematoxylin and eosin staining of sections . BALB/c mice were anesthetized with isoflurane and intranasally inoculated with 100 , 000 pfu of the indicated recombinant viruses in 50 μL of DMEM . Two days after inoculation , mice were euthanized , and their lungs were removed and homogenized as previously described . The lung homogenate was clarified by low-speed centrifugation at 3 , 000 rpm for 12 min , and 100 μL of the supernatant was administered intranasally to naive mice . Intranasal inoculation of BALB/c mice with clarified supernatants of lung homogenates collected 2 dpi was repeated 10 times . Cell lysates were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) , transferred to a nitrocellulose membrane by wet immunotransfer and processed for Western blotting . The blots were probed with monoclonal antibodies for p38 MAPK ( dilution 1:500; Cell Signaling ) , phospho-p38 MAPK ( dilution 1:500; Cell Signaling ) and actin ( dilution 1:10 , 000; Abcam ) or polyclonal antibodies specific for M ( dilution 1:1000; Biogenes ) and MCH-DBT ( dilution 1:1000; Biogenes ) proteins . Both polyclonal antibodies recognizing the parental SARS-CoV M protein or the MCH-DBT protein were generated by Biogenes ( Germany ) as previously described [81] using synthetic peptides corresponding to the residues RTRSMWSFNPETNILLNVPLRGTIVTRPLM and PLMNLSLVL , respectively . Bound antibodies were detected with horseradish peroxidase-conjugated goat anti-rabbit ( dilution 1:30 , 000; Cappel ) and the Immobilon Western chemiluminescent substrate ( Millipore ) . DBT-mACE2 cells were infected with SARS-CoV , SARS-CoV-nsp1-ΔC and -ΔD at a moi of 0 . 125 . Total RNAs from DBT-mACE2 infected cells were extracted at 48 hpi using the Qiagen RNeasy kit according to the manufacturer’s instructions . Quantitative reverse transcription-polymerase chain reaction ( qRT-PCR ) reactions were performed at 37°C for 2 h using the High Capacity cDNA transcription kit ( Applied Biosystems ) and 100 ng of total RNA and random hexamer oligonucleotides . Cellular gene expression was analyzed using TaqMan gene expression assays ( Applied Biosystems ) specific for Mus musculus genes ( S3 Table ) . Data were acquired with an ABI PRISM 7000 sequence detection system ( Applied Biosystems ) and analyzed with ABI PRISM 7000 SDS version 1 . 0 software . Gene expression in mock-infected cells and SARS-CoV , SARS-CoV-nsp1-ΔC and -ΔD-infected cells was compared . Quantification was achieved using the 2-ΔΔCt method , which analyzes relative changes in gene expression in qPCR experiments ( Livak and Scmittgen , 2001 ) . The results of three independent experiments were analyzed . All experiments and data analysis were MIQE compliant [82] . Lung sections from infected animals were collected at 2 dpi and homogenized using gentleMACS Dissociator ( Miltenyibiotec ) . Then , total RNA was extracted using the RNeasy purification kit ( Qiagen ) . Reactions were performed at 37°C for 2 h using a High Capacity cDNA transcription kit ( Applied Biosystems ) with 100 ng of total RNA and random hexamer oligonucleotides . Cellular gene expression was analyzed using TaqMan gene expression assays ( Applied Biosystems ) specific for mouse genes ( S4 Table ) . Data representing the average of three independent experiments were acquired and analyzed as previously described [57] . All experiments and data analysis were MIQE compliant [82] . The computer modeling of 8a protein structure was performed with the raptorX server http://raptorx . uchicago . edu [83] . The predicted structures were visualized using Pymol ( http://www . pymol . org/ ) . Student´s t test was used to analyze differences in mean values between groups . All results are expressed as means ± standard errors of the means . P values of <0 . 05 were considered statistically significant . The UniProt ( http://www . uniprot . org/ ) accession numbers for genes and proteins discussed in this paper are: SARS-CoV E protein , P59637; SARS-CoV 8a protein , Q19QW2; SARS 3a protein , P59632; SARS M protein , P59596; mouse IFN-β , P01575; mouse IRF1 , P15314; mouse DDX58 , Q6Q899; mouse STAT1 , P42225; human p38 MAPK , Q16539; human ACE2 , Q9BYF1; mouse ACE2 , Q8R0I0; mouse CXCL10 , P17515; mouse CCL2 , P10148; mouse IL6 , P08505; mouse 18S , O35130; human actin , P60709 . | Zoonotic coronaviruses , including SARS-CoV , Middle East respiratory syndrome ( MERS-CoV ) , porcine epidemic diarrhea virus ( PEDV ) and swine delta coronavirus ( SDCoV ) have recently emerged causing high morbidity and mortality in human or piglets . No fully protective therapy is still available for these CoVs . Therefore , the development of efficient vaccines is a high priority . Live attenuated vaccines are considered most effective compared to other types of vaccines , as they induce a long-lived , balanced immune response . However , safety is the main concern of this type of vaccines because attenuated viruses can eventually revert to a virulent phenotype . Therefore , an essential feature of any live attenuated vaccine candidate is its stability . In addition , introduction of several safety guards is advisable to increase vaccine safety . In this manuscript , we analyzed the mechanisms by which an attenuated SARS-CoV reverted to a virulent phenotype and describe the introduction of attenuating deletions that maintained virus stability . The virus , engineered with two safety guards , provided full protection against challenge with a lethal SARS-CoV . Understanding the molecular mechanisms leading to pathogenicity and the in vivo evaluation of vaccine genetic stability contributed to a rational design of a promising SARS-CoV vaccine . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Identification of the Mechanisms Causing Reversion to Virulence in an Attenuated SARS-CoV for the Design of a Genetically Stable Vaccine |
Dengue is an emerging health problem in several coastlines along the Red Sea . The objective of the present work is to elucidate spatial and temporal patterns of dengue transmission in Port Sudan . A longitudinal study with three cross-sectional surveys was carried out in upper , middle and lower class neighborhoods , from November 2008 to October 2009 . Monthly , entomological surveys were followed by serological surveys in dengue vector-positive houses . Meteorological records were obtained from two weather stations in the city during the same time . Overall , 2825 houses were inspected . Aedes aegypti represented 65% ( 35 , 714/54 , 944 ) and 68% ( 2526/3715 ) of the collected larvae and pupae , respectively . Out of 4640 drinking water containers , 2297 were positive for Ae . aegypti . Clay-pots “Zeirr” followed by plastic barrels were key productive containers for pupae of dengue vector , 63% ( n = 3959 ) and 26% ( n = 1651 ) , respectively . A total of 791 blood samples were tested using PanBio Capture/Indirect IgM ELISA . Overall , the sero-prevalence rate of dengue ranged between 3%–8% ( 41/791 ) , compared to an incidence of 29–40 new cases per 10 , 000 ( 193/54886 ) in the same examined population . Lower and middle class neighborhoods had higher entomological indices compared with upper class ones ( p<0 . 001 ) . Although , dengue incidence rate was significantly lower in the middle and lower class neighborhoods ( F = 73 . 97 , d . f . = 2 , p<0 . 001 ) , no difference in IgM prevalence was shown . The city is subject to two transmission peaks in the winter ( i . e . November–January ) , and summer ( i . e . June–August ) . The serological peaks of dengue are preceded by entomological peaks that occur before the onset of winter ( November ) and summer ( March ) respectively . Dengue incidence is heterogeneously distributed across the neighborhoods of Port Sudan and exhibits a bi-cyclic intra-annual pattern . Hence , it should be feasible to carry out timely vector control measures to prevent or reduce dengue transmission .
Dengue fever is the most important mosquito-borne viral disease in the world . The incidence of dengue has increased over ten-fold over the last three decades with an estimated 50 million cases in over 100 countries [1] , [2] . The global emergence of dengue is linked to increasing travel of viremic people as well as dispersal of its main vectors ( i . e . Aedes aegypti and Ae . albopictus ) into new locations [3]–[5] . Climate variability and unplanned urbanization may contribute to dengue epidemics [6]–[8] . The Red Sea is a semi-enclosed Mediterranean sea surrounded by the African and Eurasian continents . It has a surface area of 438 , 000 km2 and water volume of 233 , 000 km3 that is linked to the Indian Ocean by a very shallow sill . Although dengue is reported from several surrounding countries such as Saudi Arabia , Yemen , Djibouti , Somalia and Sudan [9]–[16] , little is known about dengue epidemiology along coastlines of the Red Sea . Port Sudan city is Sudan's main seaport on the Red Sea . The main dengue vector Ae . aegypti has been reported in the area since the 1930s [17] . The dengue virus serotypes DEN1 and DEN2 were first detected in the 1980s in Port Sudan , while DEN-3 was recently identified in an outbreak [14] , [16] . During the last two decades , the city has been subject to a number of dengue outbreaks ( Ministry of health , personnel communications ) . In the present study , entomological and serological surveys in upper , middle and lower class neighborhoods of Port Sudan were coupled with meteorological parameters . The main objective of the present study was to elucidate spatial and temporal patterns of dengue transmission in Port Sudan city .
Ethical approval for the study was granted by the Ethical Review Committee of the Ministry of Health , Sudan ( 2008 ) . The objectives and procedures of the study were explained to the local health authorities , medical assistants and householders at each study site . Informed consent was obtained from all participants in accordance with the ethical standards of the Sudan committee . This was a descriptive stratified longitudinal study in upper , middle and lower class neighborhoods . The first survey was launched in October 2008 and the last one finished in October 2009 . Monthly , two types of surveys were carried out: entomological pupal/demographic surveysand household serological surveys . Port Sudan ( 19 58 N , 37 21 E ) , is about 300 , 000 km2 with an estimated population of 450 , 000 people . Port Sudan has a humid Mediterranean climate and a service-based economy linked to shipping operations and trade . In addition , there are no suburbs or surrounding villages so food is transported to the city from other regions of Sudan and neighboring countries . Port Sudan is administratively divided into three sectors ( Eastern , Middle and Southern ) which are further divided into 39 neighborhoods . Nine residential neighborhoods were selected because they are a good representation of the city by class . We grouped these neighborhoods using indicators of living conditions such as method of water supply , on-site sanitation , and building material of houses into three strata: upper ( Abuhasheish , Downtown and Elthora ) , middle ( Elmatar , Salalab East , and Dar Elnaeem ) , and lower class neighborhoods ( Dar Elsalam , Elgadisia and Elwihda ) . Approximate locations of the study sites are depicted in the map of the city ( Figure 1 ) . Average monthly records from November 2008 to October 2009 were obtained from the meteorological authority . Meteorological parameters ( minimum and maximum temperature , relative humidity , precipitation and , evaporation rates , wind speed and direction ) were recorded on daily basis in two stations in Port Sudan . A random stratified sampling strategy was followed . All dengue cases during the study period were reviewed retrospectively . These cases were reported through the health information system vertically from health dispensaries and the main hospitals of Port Sudan up to the central level . Then , only cases whose home address was from the study neighborhoods were selected for the study . All clinical criteria and laboratory data from each case were checked further by an epidemiologist to confirm its accordance with the dengue case definition and management protocol of the Ministry of Health and WHO guidelines [1] . Monthly , entomological indices were calculated for each study site . These include both Stegomyia indices: House Index ( HI ) = percentage of houses or premises positive for Aedes aquatic stages , Container Index ( CI ) = percentage of water containers positive for Aedes aquatic stages , Breteau Index ( BI ) = number of positive containers per 100 houses in a specific location; Pupal indices ( Pupal/Person ( P/P ) = total number of collected pupae/total number of inhabitants in the inspected households Pupal/children ( P/C ) = total number of collected pupae/total number of children under five years in the inspected households . All the data analysis was performed using version 2 . 3 of OpenEpi software for Windows [22] . Comparison between two groups was done using a Chi square test . ANOVA was utilized to compare between the study's strata . Pearson correlation was performed to associate entomological , serological and meteorological data .
A total of 2825 households were accessible and inspected in Port Sudan city . Average family size was larger in the upper class neighborhood ( 6 . 1 ) compared to the middle and lower strata ( 5 . 8 and 5 . 6 , respectively ) . This was associated with greater consumption rate of drinking water in the high stratum ( 13 liters/person ) compared to the other two strata ( 9 and 7 liters/person , respectively ) . Although donkey-drawn water tankers were the main method of water supply in Port Sudan ( 75% ) , this was only true in 51% of households in the upper class neighborhood due to the presence of public water pipes ( 33% ) and motorized tankers ( 16% ) in these areas ( Table 1 ) . While the mean minimum temperature ( 19 . 3°C ) was recorded on March 2009 , the mean maximum one ( 43 . 7°C ) was on July 08 . The minimum relative humidity was recorded on June 09 ( 30% ) , compared to the maximum ( 67% ) on November 08 . In addition , a short rainy season occurred in two months December 2008 ( 1 . 2 mm ) and January 2009 ( 3 mm ) , with very little rainfall in November , February and July ( <0 . 00001 mm ) . The highest wind speed occurred in January 2009 ( 1 . 3 knots ) while the lowest one was in October 2009 ( 6 knots ) . The highest evaporation rate was on July 2009 ( 15 . 7 mm ) while the lowest one was on November 2008 ( 6 . 4 mm ) . A total of 2297 out of 4640 water containers ( 49 . 4% ) were found positive for Ae . aegypti . However , over 70% of the positive containers were covered with lids , and 98% of these were located indoors . Clay-pots ( in Arabic “Zeirr” ) followed by plastic barrels represented the key breeding containers for pupae of Aedes mosquito , containing 63% ( number of pupae = 3959 ) and 26% ( number of pupae = 1651 ) of pupae , respectively . Other containers ( representing <10% ) included: pools of excess tap water , underground tanks , pans , pools filtered water from clay-pots , wells and plastic Jerry cans . The highest dispersion of pupae among container types was shown in May ( i . e . six types found positive for Aedes ) , compared to the lowest in October ( i . e . three types ) ( Figure 2 ) . A total of 54 , 944 larvae and 3715 pupae of mosquitoes were collected during the entomological surveys ( Table 2 ) . Morphological identification showed that Ae . aegypti constituted 65% and 68% , of the collected larvae and pupae , respectively . Fewer pupae and larvae of Aedes aegypti were collected in upper class neighborhoods ( 45% and 59% ) compared to the middle ( 66% and 75% ) , and lower class ( 65% and 72% ) ones , respectively . Other collected mosquitoes were identified as: Culex quinquefasciatus Say ( 18 . 5% ) , Culex sitiens Wiedemann ( 12% ) and Ae . caspius Pallas ( 0 . 1% ) , and unidentified Culex spp . belonging to group IV ( 11% ) . The dengue vector was co-breeding with one of the above species in few numbers of mixed containers ( 1 . 68% ) . A total of 791 individual blood specimens was tested using PanBio Capture/Indirect IgM ELISA ( Table 3 ) . Participants ranged in age from 3 months to 80 years old; females represented 64% ( 506/791 ) of the study population . Dengue IgM high-titer serums were shown in 5 . 2% ( 41/791 ) specimens . There is a significant difference in IgM positive results between age groups :<5 , 6–17 , 18–39 , 40–60 and >60 years ( χ2 = 5 . 05 , d . f . = 5 , p = 0 . 03 ) . Among these age groups , higher dengue attack rates were shown in adults aged 18–39 and above 40 years , 2 . 14% ( 17/791 ) and 1 . 77% ( 14/791 ) , respectively . Although females constituted 3 . 16% ( 25/791 ) of the positive results , the gender difference was not significant ( χ2 = 0 . 168 , d . f . = 1 , p = 0 . 4 ) . A total of 193 dengue cases were reported from the selected neighborhoods . Among these reported cases , 69% were males ( 133/193 ) compared to 21% females ( 60/193 ) . This gender difference in attack rate was significant ( λ2 = 5 . 1 , d . f . = 1 , p = 0 . 02 ) . Another significant difference on attack rate was shown between the age groups ( λ2 = 19 . 6 , d . f . = 4 , p = 0 . 0005 ) . Accordingly , while 42% ( 83/193 ) of the reported cases ranged between ( 18–39 ) years old , 40% ( 77/193 ) were ranged ( 6–17 ) years old .
This study shows dengue transmission in Port Sudan is autochthonous and related to storage of drinking water . Drinking water in Port Sudan is mainly sourced from Khor Arbaat , i . e . a seasonal stream or Wadi . There is a reservoir in Khor Arbaat about 20 km north of Port Sudan [23] . Drinking water is either pumped through pipelines or transported via motorized tankers to the city . Also , there are two desalinization facilities in the city . However , all these water sources are insufficient and supply only one third of the needed drinking water [24] . The main containers for indoor breeding of dengue vector in Port Sudan were clay pots and barrels . Owing to shortages of drinking water , the residents of Port Sudan usually preserve drinking water in these containers in close proximity to their houses . A common factor in the emergence of dengue in urban settings in developing countries is a lack of basic services for economically marginalized and growing populations [6] , [25] . Both IgM seroprevalence ( ranged between 3%–8% among the healthy residents ) and incidence rate ( 35 new clinical cases per 10 , 000 individuals ) reveals that dengue is a considerable burden on the population of Port Sudan . Similarly , a recent study in the city found a 7% IgM seroprevalence rate among pregnant women [15] . No virus serotype ( s ) has been determined in the current work . Co-circulation of DEN-1 and DEN-2 was confirmed in Port Sudan in 1984 [14] . In this hospital study , about 72% of the symptomatic patients were males and their ages averaged 28 years , ranging between 12–70 years . Introduction of DEN-3 in Port Sudan was confirmed during the 2004/2005 outbreak [16] . In Central Brazil , co-circulation of the three serotypes was shown [26] . Females were more affected and about 85% of the infected individuals were adults ranging between 20–70 years . Thus , individuals of working age ( 18–60 years old ) appeared to be more vulnerable to dengue transmission than other age groups , either because this group is subject to a secondary infection of dengue or more susceptible after a few years of transmission . Further work is needed to quantify the economic burden of dengue on the community of Port Sudan . Dengue has an uneven spatial distribution in Port Sudan . Although lower and middle class neighborhoods have low consumption rates of drinking water compared to upper class neighborhoods , such neighborhoods have higher entomological density indices than the latter . This may be due to the large number of small containers such as clay pots utilized in the lower and middle class neighborhoods . This is in line with our finding that the incidence rate of clinical dengue is low in the middle and lower class neighborhoods . Perhaps the low and middle class neighborhoods were affected first and have higher rates of herd immunity . However , this was not supported here by a significant difference on IgM prevalence between the three study strata . Therefore , it may be linked with health seeking behavior in these economically distinct areas . The temporal pattern of dengue in Port Sudan showed a bi-cyclic trend . Hence , the city was likely subject to two transmission peaks: the first short peak in the winter ( i . e . extending for 2 months in November and December ) , and a second long peak in the summer ( i . e . 3 months from June to August ) . These two peaks were preceded by peaks of mosquito densities in December and June . The current work confirms historical records of Ae . aegypti as the principle vector of dengue in Port Sudan [17] . No reports on invasion of Ae . albopictus in the city from sea ports of South East Asia was traced . However , the disappearance of the dengue vector ( Ae . aegypti ) in September demands further research to define whether there is a true disappearance and if so how and from where the vector is reintroduced . The coastlines of the Red Sea are subject to two monsoons: a northeasterly winter monsoon ( October–April ) and southwesterly summer one ( May–September ) [27] . The maximum temperature recorded in July and August preceded the observed crash of the Aedes population in August and September . However , further research is needed to determine if there is a relationship between dengue outbreaks and climate variability in the Red sea region . In conclusion , there are spatiotemporal patterns of dengue transmission in Port Sudan . Hence it should be feasible to carry out timely vector control measures to prevent or reduce dengue transmission in this coastline area . Coastlines of the Red Sea face similar situations of insufficient drinking water and the zone is prone to dengue epidemics . Climatic variability and increased shipping traffic along the Red Sea ports ( trade with China , South East Asia and Latin America ) during the last decade may be key drivers for dengue outbreaks . Further research is needed to study the impact of climatic and socioeconomic changes on emergence of dengue in the Red Sea region . | Dengue is a tropical infectious disease that is of emerging global importance . As a dengue vaccine is still a distant prospect , descriptive epidemiological studies are a vital tool for developing a surveillance system capable of preventing dengue outbreaks . In the current work , the investigators describe epidemiology of dengue in Port Sudan on the Red Sea . There , the disease has distinctive spatial and temporal patterns . Households use water storage containers to supplement Port Sudan’s poor public water supply . These containers provide breeding sites for dengue mosquitoes . Differences in water consumption patterns across neighborhoods result in differences in disease incidence . Our results suggest that Lower class neighborhoods may suffer dengue more than upper class ones . In addition , two transmission peaks are observed during the winter and summer seasons . These spatial and temporal patterns of dengue may describe dengue epidemiology along similarly affected coastlines of the Red Sea . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"survey",
"methods",
"entomology",
"epidemiology",
"infectious",
"disease",
"epidemiology",
"dengue",
"fever",
"neglected",
"tropical",
"diseases",
"biology",
"zoology"
] | 2012 | Spatial and Temporal Patterns of Dengue Transmission along a Red Sea Coastline: A Longitudinal Entomological and Serological Survey in Port Sudan City |
Suppressor of cytokine signalling 3 ( SOCS3 ) negatively regulates STAT3 activation in response to several cytokines such as those in the gp130-containing IL-6 receptor family . Thus , SOCS3 may play a major role in immune responses to pathogens . In the present study , the role of SOCS3 in M . tuberculosis infection was examined . All Socs3fl/fl LysM cre , Socs3fl/fl lck cre ( with SOCS3-deficient myeloid and lymphoid cells , respectively ) and gp130F/F mice , with a mutation in gp130 that impedes binding to SOCS3 , showed increased susceptibility to infection with M . tuberculosis . SOCS3 binding to gp130 in myeloid cells conveyed resistance to M . tuberculosis infection via the regulation of IL-6/STAT3 signalling . SOCS3 was redundant for mycobacterial control by macrophages in vitro . Instead , SOCS3 expression in infected macrophages and DCs prevented the IL-6-mediated inhibition of TNF and IL-12 secretion and contributed to a timely CD4+ cell-dependent IFN-γ expression in vivo . In T cells , SOCS3 expression was essential for a gp130-independent control of infection with M . tuberculosis , but was neither required for the control of infection with attenuated M . bovis BCG nor for M . tuberculosis in BCG-vaccinated mice . Socs3fl/fl lck cre mice showed an increased frequency of γδ+ T cells in different organs and an enhanced secretion of IL-17 by γδ+ T cells in response to infection . Socs3fl/fl lck cre γδ+ T cells impaired the control of infection with M . tuberculosis . Thus , SOCS3 expression in either lymphoid or myeloid cells is essential for resistance against M . tuberculosis via discrete mechanisms .
Tuberculosis ( TB ) , an infectious disease caused by Mycobacterium tuberculosis , remains a leading public health problem worldwide . The global incidence of TB is rising with 8 . 8 million new cases and 2 million deaths each year [1] . However , while immune responses to TB clearly show their importance in host defence , it is clear that there are still gaps in our knowledge of the host factors determining the outcome of infection . Host responses of mycobacterial infections are primarily Th1 immune responses involving cellular effector mechanisms such as macrophage activation . IFN-γ is known to be an important mediator of mycobacterial control during clinical and experimental infections [2] . IL-12 is crucial for optimal differentiation and maintenance of IFN-γ-secreting antigen-specific Th1 cells [3] , [4] , and in controlling mycobacterial infections in mice and man [5] , [6] . The “suppressor of cytokine signalling” ( SOCS ) proteins are a family of eight members that inhibit STAT activation by different receptors . SOCS proteins bind either the Janus-activated kinases ( JAKs ) directly inhibiting their kinase activity , or the intracellular domain of cytokine receptors thereby targeting the receptor complex for ubiquitination and subsequent proteasome-mediated degradation [7] . SOCS3 inhibits STAT3-mediated signalling by binding to the IL-6 receptor family subunit gp130 , G-CSF , leptin and the IL-12 receptor [8] . Since SOCS3-deficient mice die during embryogenesis [9] , [10] , the role of SOCS3 in vivo has been studied using conditional knockdown mice . Conditional knockdown of SOCS3 in macrophages protects mice from LPS shock by reducing the secretion of IL-12 and TNF due to the enhanced anti-inflammatory effect of STAT3 [11] . However , mice with SOCS3-deficient macrophages and neutrophils succumb to toxoplasmosis , probably due to reduced IL-12 and IFN-γ responses [12] . Furthermore , SOCS3 can also inhibit STAT1 activation thereby preventing IFN-γ-like responses in cells stimulated with IL-6 [13] , [14] . SOCS3 also may have several roles in T cell function . SOCS3 expression in T cells can both obstruct the differentiation of inflammatory IL-17-producing Th17 cells [15] , [16] and inhibit the secretion of anti-inflammatory IL-10 and TGF-β by T cells [17] and mice with SOCS3-deficient T cells are more susceptible to infection with Leishmania major [17] . On the other hand , SOCS3 has also been shown to impair T-cell memory development , T cell-mediated IFN-γ secretion and LCMV virus clearance in mice [18] . In the present study , the role of SOCS3 in the outcome of infection with M . tuberculosis was investigated . We report that the expression of SOCS3 , in either myeloid or T cells , is independently required for the control of M . tuberculosis infection in mice . SOCS3 expression in myeloid cells allows a proper IL-12 secretion by hampering an IL-6-mediated inhibition of IL-12 expression . SOCS3 expression in T cells reduces the frequency of γδ+ T cells in different organs and the secretion of IL-17 by + T cells in response to infection in a gp130-independent manner .
First , the role of Socs3 expression in myeloid cells in the control of infection with M . tuberculosis was examined by using Socs3fl/fl LysM cre mice [19] . Lungs and spleens from Socs3fl/fl LysM cre mice showed significantly higher M . tuberculosis levels than Socs3fl/fl littermates at 16 and 28 days of infection ( Figure 1A , B ) . A larger area of the lung parenchyma of Socs3fl/fl LysM cre mice was occupied by granulomas as compared to controls 4 weeks after infection ( Figure 1C , D ) . Furthermore , M . tuberculosis-infected Socs3fl/fl LysM cre mice also showed a higher cumulative mortality ( Figure 1E ) . Socs3fl/fl LysM cre mice infected with the attenuated M . bovis BCG displayed higher bacterial levels in the lungs and spleen ( but not the liver ) , although the differences in BCG levels with infected Socs3fl/fl littermates were not as striking as those observed after infection with M . tuberculosis ( Figure 1F ) . Since the LysM promoter is active in neutrophils and SOCS3 has been shown to be a negative regulator of granulopoiesis [20] , [21] , we studied whether the increased susceptibility to M . tuberculosis of Socs3fl/fl LysM cre mice was associated to increased numbers of neutrophils at the site of infection . Comparable numbers of Gr1+/F4/80- neutrophils and similar mRNA levels of the neutrophil enzyme myeloperoxidase were detected in lungs from M . tuberculosis-infected Socs3fl/fl LysM cre and control mice ( Figure 1G , H ) . Next , we studied the expression and role of SOCS3 in mycobacteria-infected macrophages . Bone marrow-derived macrophages ( BMM ) from wild type ( WT ) mice showed increased accumulation of Socs3 mRNA after infection with either M . tuberculosis or BCG ( Figure 2A–C , S1A , B ) . Recognition by innate immune receptors was required for SOCS3 expression , since Socs3 mRNA levels after infection were reduced in the Toll-like receptor adaptor molecule MyD88−/− BMM and BMM incubated with a NF-κB inhibitor but not in Irf3−/− BMM ( Figure 2A , B and S1A , B ) . IRF3 has been shown to detrimentally affect M . tuberculosis infection [22] . As expected , Socs3 mRNA levels were reduced in M . tuberculosis-infected Socs3fl/fl LysM cre BMM when compared to controls ( Figure 2C ) , and in vitro infection of BMM with M . tuberculosis stimulated STAT3 phosphorylation that was prolonged in Socs3fl/fl LysM cre BMM ( Figure 2D ) . Whether a defect of macrophages to control intracellular mycobacterial growth could account for the enhanced susceptibility of Socs3fl/fl LysM cre mice to mycobacteria was then studied . Socs3fl/fl LysM cre BMM , pulmonary and peritoneal macrophages showed diminished intracellular levels of M . tuberculosis ( Figure 2E–G and data not shown ) . The IFN-γ-mediated control of mycobacteria by macrophages is essential for the intracellular control of M . tuberculosis . Incubation of BMM with IFN-γ decreased the number of intracellular M . tuberculosis . Similar bacterial levels were measured in Socs3fl/fl LysM cre and control BMM after incubation with IFN-γ ( Figure 2H ) . Macrophages have been shown to kill mycobacteria through the generation of nitric oxide ( NO ) by the IFN-γ-regulated inducible NO synthase ( iNOS ) [23] . M . tuberculosis-infected Socs3fl/fl LysM cre BMM contained higher iNos mRNA and nitrite levels than Socs3fl/fl BMM ( Figure 2I , J ) . Similarly , infection of Socs3fl/fl LysM cre BMM with BCG or stimulation with Pam3CSK4 , an agonist for TLR2 , a receptor that plays a prominent role in the initiation of responses against M . tuberculosis [24] , led to higher NO and iNos mRNA levels compared to controls ( Figure S1C , D ) . Cells derived from gp130F/F mice , harbouring a gp130 Y757F mutation to ablate SOCS3 binding to gp130 , show an exaggerated gp130-mediated STAT3 signalling as a consequence of an impaired negative feedback loop by SOCS3 to down-modulate gp130/STAT3 signalling [25] . Similar to Socs3fl/fl LysM cre BMM , gp130F/F BMM showed increased iNos mRNA and nitrite levels after infection with either M . tuberculosis or BCG , or stimulation with Pam3CSK4 ( Figure 2K and S1E , F ) . Thus , the increased iNOS response of SOCS3-deficient macrophages was dependent on signalling via gp130 . Similarly , the mRNA expression levels of the IFN-γ-induced chemokine CXCL10 was also increased in either M . tuberculosis- or BCG-infected Socs3fl/fl LysM cre BMM ( Figure 2L and S1G ) . Altogether , these data demonstrated that the higher susceptibility to M . tuberculosis of Socs3fl/fl LysM cre mice was not associated with a defect of BMM or pulmonary macrophages in controlling intracellular bacterial growth in vitro . After infection with M . tuberculosis , Il-6 mRNA levels were strikingly increased in lungs from Socs3fl/fl LysM cre mice when compared to littermates ( Figure 3A ) . Therefore , we evaluated whether IL-6 secretion was also elevated in mycobacteria-infected Socs3fl/fl LysM cre BMM . Unexpectedly , we found diminished Il-6 mRNA and protein levels in M . tuberculosis- or BCG-infected or in Pam3CSK4-stimulated Socs3fl/fl LysM cre BMM , peritoneal and pulmonary macrophages as compared to controls ( Figure 3B , C and S2A–C ) . Similarly , gp130F/F BMM expressed lower IL-6 protein and mRNA levels after infection with either M . tuberculosis or BCG , or stimulation with Pam3CSK4 ( Figure 3D and S2D ) . In conclusion , although IL-6 levels are increased in lungs of infected Socs3fl/fl LysM cre mice , SOCS3 does not impair IL-6 secretion by mycobacteria-infected BMM . The LPS-induced production of TNF and IL-12 is reduced in SOCS3-deficient macrophages if IL-6 is added [11] . Consistent with this observation , the Tnf mRNA or protein accumulation was reduced in Socs3fl/fl LysM cre BMM incubated with either M . tuberculosis , BCG- or Pam3CSK4 as compared to controls ( Figure 3E and S2E ) . Next , we studied whether IL-6 signalling accounted for SOCS3/gp130-mediated regulation of TNF levels . Tnf mRNA and protein levels were reduced in M . tuberculosis- or BCG-infected gp130F/F BMM and were partially restored in gp130F/F Il-6−/− BMM ( Figure 3G and S2F , G ) , indicating that SOCS3 allows proper infection-induced TNF secretion in macrophages by hampering gp130/IL-6 receptor-mediated signalling . Moreover , the co-incubation with recombinant IL-6 ( rIL-6 ) further diminished TNF levels in M . tuberculosis-infected Socs3fl/fl LysM cre but not in control BMM ( Figure 3E ) . Similar to the results observed for TNF , the Il-12 p40 mRNA and protein levels were reduced in Socs3fl/fl LysM cre BMM after stimulation with Pam3CSK4 , BCG or M . tuberculosis when compared to controls ( Figure 4A–C ) . IL-12 levels were further decreased in cultures of infected mutant macrophages incubated with rIL-6 ( Figure 4B , C ) . Accordingly , IL-12 p40 mRNA and protein levels were reduced in M . tuberculosis- or BCG-infected gp130F/F BMM , and such defect was restored in infected gp130F/F Il-6−/−BMM ( Figure 4D and S3A ) . Since IL-12 secretion by dendritic cells ( DCs ) is required for Th1 differentiation , we investigated whether DCs from Socs3fl/fl LysM cre mice also showed an impaired secretion of IL-12 . Socs3 mRNA levels were reduced in Socs3fl/fl LysM cre bone marrow-derived dendritic cells ( BMDC ) indicating the expression of the LysM cre recombinase ( Figure 4E ) . The Il-12 p40 mRNA and IL-12 protein expression by Socs3fl/fl LysM cre BMDC after infection with M . tuberculosis was reduced as compared to controls ( Figure 4F , G ) . As shown for BMM , incubation with exogenous IL-6 further diminished IL-12 secretion by M . tuberculosis-infected Socs3fl/fl LysM cre BMDC ( Figure 4G ) . In order to exclude that the diminished IL-12 secretion in BMDC cultures was due to the response of contaminant macrophages in the culture , the expression of Il-12 p40 mRNA was tested in CD11c+ sorted cells . Both , CD11c+ and CD11c- cells showed diminished IL-12 p40 mRNA accumulation after infection with M . tuberculosis compared with controls ( Figure 4H ) . Moreover , Socs3fl/fl LysM cre splenic DCs displayed a diminished secretion of IL-12 after M . tuberculosis ( Figure 4I ) . Thus , SOCS3 expression promotes IL-12 secretion in M . tuberculosis-stimulated DCs . Since IL-12 is required for IFN-γ secretion by NK cells , we tested the effect of SOCS3 expression by M . tuberculosis-infected DCs in the regulation of IFN-γ secretion by NK cells . Co-incubation with M . tuberculosis-infected splenic CD11c+ DCs induced IFN-γ secretion by NK cells . IFN-γ expression by NK cells was reduced when these cells were incubated with Socs3fl/fl LysM cre DCs ( Figure S3B , C ) . Lower Il-12 p40 mRNA levels were also found in lungs of M . tuberculosis-infected Socs3fl/fl LysM cre mice compared to controls ( Figure 4J ) and similarly , lower Ifn-γ mRNA accumulation in lungs from Socs3fl/fl LysM cre mice was detected 2 . 5 weeks after M . tuberculosis infection ( Figure 4K ) . In order to examine whether an effect of SOCS3-deficient myeloid cells on the cytokine production by CD4+ T cells accounted for the elevated numbers of bacteria in Socs3fl/fl LysM cre mice , we depleted CD4+ cells during M . tuberculosis infection by administration of anti-CD4 neutralizing antibodies ( Figure S4A ) . CD4+ cell depletion decreased the Ifn-γ mRNA accumulation in lungs from Socs3fl/fl mice . The Ifn-γ mRNA levels in infected Socs3fl/fl LysM cre mice were similar to those measured in CD4+ cell-depleted mice ( Figure 4K ) . Moreover , lungs from Socs3fl/fl LysM cre and control mice depleted of CD4+ cells showed similar bacterial levels ( Figure 4L ) . In contrast to the decreased Ifn-γ mRNA expression , the frequency of CD44+ and CD62L+ CD4+ activated T cells in the lungs of M . tuberculosis-infected Socs3fl/fl LysM cre and Socs3fl/fl mice was similar , and higher than in uninfected animals ( Figure S4B , C ) . When we compared cytokine levels in Socs3fl/fl LysM cre and Socs3fl/fl littermates at later time points after M . tuberculosis infection . Higher Ifn-γ and iNos , but similar Tnf-mRNA levels were measured in lungs from Socs3fl/fl LysM cre compared to control mice 8 weeks after infection ( Figure S4D–F ) . Regulatory FoxP3+ T-cells have been shown to expand in mice with SOCS3-deficient DCs [26] . However , comparable levels of FoxP3+ CD4+ T cells were found in lungs and pulmonary lymph nodes of infected mutant and control mice ( Figure S4G ) , suggesting that the susceptibility of Socs3fl/fl LysM cre mice to M . tuberculosis is not due to higher frequencies of regulatory T cells . Altogether , the enhanced susceptibility of Socs3fl/fl LysM cre mice to M . tuberculosis could be associated to a reduced IL-12 secretion resulting in a delayed CD4+ -cell dependent IFN-γ-expression . Next , the role of SOCS3 expression by T cells in the control of infection with M . tuberculosis was studied . Lungs and spleens from Socs3fl/fl lck cre mice showed higher numbers of M . tuberculosis bacteria 4 weeks after aerosol infection ( Figure 5A and data not shown ) , with 500-fold higher bacterial levels in lungs compared to Socs3fl/fl littermate controls . In contrast , no differences in bacterial load were registered 2 weeks after infection ( Figure 5A ) . Infected Socs3fl/fl lck cre mice had a median survival of 38 days after infection while controls survived more than 200 days ( Figure 5B ) . Four weeks after infection , Socs3fl/fl lck cre mice displayed an increased severity of pulmonary pathology ( Figure 5C–E ) with granulomas containing large necrotic areas ( Figure 5F ) and elevated levels of neutrophil myeloperoxidase transcripts ( Figure 5G ) . The frequency of Foxp3+ CD4+ regulatory T cells in pulmonary lymph nodes was similar ( Figure 5H ) . Whether the susceptibility of Socs3fl/fl lck cre mice was associated to an altered frequency of T cell populations was then evaluated . While the percentage of CD4+ and CD8+ T cells in lungs and spleens from Socs3fl/fl lck cre and Socs3fl/fl mice before or after M . tuberculosis infection was similar ( Figure S5A–D ) , the percentage of γδ+ T cells in the thymus , spleen and lungs of uninfected Socs3fl/fl lck cre mice was strikingly elevated and remained high after M . tuberculosis infection when compared to Socs3fl/fl controls ( Figure 6A–C , E ) . T cells accumulated in the lungs after M . tuberculosis infection and higher numbers of γδ+ T cells were observed in lungs from infected Socs3fl/fl lck cre compared to Socs3fl/fl mice ( Figure 6D ) . The outcome of infection with M . tuberculosis of Rag1−/− mice reconstituted with control or Socs3fl/fl lck cre T cells was then compared . Lungs and spleens from Rag1−/− mice transferred with Socs3fl/fl total T cells ( CD90+ ) showed lower bacterial levels than non-transferred mice while Socs3fl/fl lck cre T cells failed to transfer protection ( Figure 6F ) . Moreover , the transfer of a 1∶1 mixture of Socs3fl/fl lck cre and control T cells conferred no protection to Rag1−/− mice ( Figure 6F ) , suggesting that SOCS3-deficient T cells can hamper M . tuberculosis control by wild type T cells . Rag1−/− mice transferred with CD4+ Socs3fl/fl lck cre cells unlike those transferred with total T cells from the same mice , showed reduced M . tuberculosis levels compared to non-transferred controls indicating that in Socs3fl/fl lck cre mice CD3+CD4− T cells hamper the protective ability of CD4+ cells ( Figure 6G ) . Therefore , we examined whether γδ+ T cells could account for the suppressive activity of CD3+CD4− cells . Indeed , Rag1−/− mice transferred with γδ+ cell-depleted CD90+ Socs3fl/fl lck cre T cells showed lower bacterial levels than those transferred with total Socs3fl/fl lck cre T cells ( Figure 6 H ) . Since SOCS3 expression in T cells has been shown to impair IL-17 production [27] , we speculated that a differential release of IL-17 could be related with the susceptibility of Socs3fl/fl lck cre mice to M . tuberculosis . Lungs from M . tuberculosis-infected Socs3fl/fl lck cre mice showed higher levels of Il-17 mRNA than controls ( Figure 7A ) . The levels of IL-17 in supernatants from lung cells of Socs3fl/fl lck cre mice stimulated or not with mycobacterial Purified protein derivate ( PPD ) were higher than controls 2 . 5 weeks after M . tuberculosis infection , when no differences in bacterial load in lungs were detected ( Figure 7B ) . Thus , the impaired control of M . tuberculosis-infection in Socs3fl/fl lck cre mice was associated with increased IL-17 levels . γδ+ T cells have been shown to dominate IL-17 secretion during infection with M . tuberculosis [28] . In line with this observation , the incubation of naïve spleen T cells with mycobacteria-infected BMDCs or their supernatants resulted in the secretion of IL-17 ( Figure 7C ) . Furthermore , IL-17 was secreted by γδ+ and total ( CD90+ ) but not by CD4+ T cells after incubation supernatants from mycobacteria-infected BMDCs . The levels of IL-17 secreted by γδ+ T cells were higher than those by similar numbers of total T cells . The IL-17 content in supernatants from γδ+ and total Socs3fl/fl lck cre T cells was higher compared to Socs3fl/fl controls ( Figure 7D ) . IL-17-secreting cells were enumerated by intracellular cytokine staining in PMA/ionomycin-stimulated lung cell suspensions from Socs3fl/fl lck cre and Socs3fl/fl mice 16 days after M . tuberculosis infection . The majority of IL-17-secreting lung T cells in infected mice were γδ+ rather than CD4+ or CD8+ cells ( Figure 7E , G ) . Furthermore , M . tuberculosis infection stimulated the IL-17 secretion capability of γδ+ T cells ( Figure 7F ) . However , the frequency of IL-17-secreting cells among γδ+ T cells from infected Socs3fl/fl lck cre and Socs3fl/fl was similar , suggesting that the lack of SOCS3 does not alter the differentiation of γδ+ T cells into IL-17-secreting cells . IL-17 was measured in supernatants from lung cell suspensions from CD4+ or CD90+ Socs3fl/fl lck cre T cell-transferred Rag1−/− mice 4 weeks after M . tuberculosis infection . While IL-17 levels were strikingly higher in supernatants from mice transferred with total T cells compared with non-transferred controls , the IL-17 concentration in cultures from mice inoculated with CD4+ cells was not increased ( Figure 7H ) . Thus , the inhibition of CD4+ cell-mediated protection in Rag1−/− mice transferred with CD90+ cells ( Figure 6G ) was associated to an increased IL-17 secretion . Lung cells from M . tuberculosis-infected Socs3fl/fl lck cre mice contained higher levels of Ifn-γ mRNA ( Figure 8A ) and secreted higher levels of IFN-γ when stimulated with PPD than Socs3fl/fl controls ( Figure 8B ) . Thus , the susceptibility of Socs3fl/fl lck cre mice was not associated to an impaired secretion of IFN-γ . The accumulation of Il-6 mRNA was increased in lungs of Socs3fl/fl lck cre M . tuberculosis-infected mice at 4 but not at 2 . 5 weeks after infection compared to controls ( Figure S6A ) . The IL-6 levels in supernatants from Socs3fl/fl lck cre and Socs3fl/fl lung cells obtained 2 . 5 weeks after infection , stimulated or not with PPD , were similar ( Figure S6B ) . SOCS3 has been shown to regulate IL-10 secretion by T cells [17] . However , Socs3fl/fl lck cre and Socs3fl/fl lung cells from M . tuberculosis-infected mice , stimulated or not with PPD , secreted similar levels of IL-10 ( Figure S6C ) . In contrast to results from M . tuberculosis-infected mice , Il-6 and Il-17a mRNA levels in lungs from BCG-infected Socs3fl/fl lck cre and Socs3fl/fl mice were similar ( Figure 8C and S6D ) . Spleen cells from BCG-immunized Socs3fl/fl lck cre and control mice showed comparable IL-17 , IFN-γ or IL-6 secretion in response to PPD stimulation when compared with controls ( Figure 8D , E and S6E ) . The organs from Socs3fl/fl lck cre and Socs3fl/fl mice infected with M . bovis BCG contained similar bacterial levels ( Figure 8F ) . Moreover , bacterial levels in Socs3fl/fl lck cre and Socs3fl/fl mice challenged with M . tuberculosis after BCG immunization were comparable ( Figure 8G ) . Next , we studied whether a differing stimulation of SOCS3 expression could explain the divergent susceptibility of Socs3fl/fl lck cre mice to M . tuberculosis and BCG infection . Lungs from mice infected with either M . tuberculosis or BCG showed higher Socs3 mRNA levels compared to uninfected mice . Socs3 transcript levels were higher in M . tuberculosis- than in BCG-infected animals ( Figure 8H , I ) . However , Socs3 mRNA levels in pulmonary T cells before or after infection with BCG or M . tuberculosis were similar ( Figure 8J ) . Thus , different expression levels of SOCS3 in T cells do not explain the distinct susceptibility of Socs3fl/fl lck cre mice to M . tuberculosis and BCG infection . To further characterize the function of SOCS3 in myeloid and T cells in the control of infection with M . tuberculosis , gp130F/F knock-in mice were used . gp130F/F mice displayed a dramatically enhanced susceptibility to M . tuberculosis as measured by their increased bacterial load , severity of pathology in lungs ( Figure 9A , B ) , and increased cumulative mortality ( gp130F/F mice died before 48 days after infection whereas all WT controls survived for more than 100 days ) . Since IL-6 mediated , at least in part , the inhibition of TNF and IL-12 secretion by SOCS3-deficient BMM and BMDC , the role of IL-6 in the susceptibility of gp130F/F mice to infection with M . tuberculosis was studied . We found that gp130F/FIl-6−/− as well as gp130F/FStat3+/− mutant mice displayed lower levels of M . tuberculosis bacteria in lungs and diminished severity of pulmonary pathology when compared to gp130F/F mice , indicating that the increased susceptibility of gp130F/F mice is in part mediated by IL-6 and STAT3 activation ( Figure 9A , B ) . Lungs from gp130F/F mice displayed higher Il-6 and lower Il-12 p40 mRNA accumulation than infected control mice , while Il-12 p40 mRNA accumulation in lungs from gp130F/FIl-6−/− was comparable to WT mice ( Figure 9C , D ) . Surprisingly , lungs and spleens from M . tuberculosis-infected Rag1−/− mice transferred prior to infection with either gp130F/F or WT T cells contained similar bacterial levels , indicating that T cells play , if any , a redundant role in the gp130-mediated control of M . tuberculosis . Bacterial levels in mice transferred with WT or gp130F/F cells were lower than those of non-transferred controls ( Figure 9E ) . Moreover , frequencies of γδ+ T cells in organs from WT or gp130F/F were similar ( data not shown ) . Thus , the susceptibility of Socs3fl/fl lck cre mice to infection with M . tuberculosis is mediated by receptors other than gp130 . Since T cells did not account for the increased susceptibility of gp130F/F mice and gp130F/F mice were significantly more susceptible to M . tuberculosis than Socs3fl/fl LysM cre mice , we studied the relative contributions of hematopoietic and non-hematopoietic cell lineages to the susceptibility of gp130F/F mice to infection with M . tuberculosis . Reciprocal bone marrow ( BM ) radiation chimeras between WT and gp130F/F mice were generated by inoculation of BM cells into irradiated recipients . WT mice reconstituted with gp130F/F BM contained higher titers of M . tuberculosis in the lungs than those reconstituted with WT BM ( sham chimeric mice ) , although differences in bacterial levels were notably lower than those observed in the gp130F/F mice ( Figure 9F ) . Although gp130F/F recipients showed significant mortality after irradiation , the few survivors inoculated with WT BM showed very high bacterial levels , similar to those from non-irradiated gp130F/F mice ( data not shown ) . Thus , these data suggest that non-lymphoid , hematopoietic cells only partially account for the susceptibility of gp130F/F mice to M . tuberculosis and suggest a relevant role for non-hematopoietic cells in the high sensitivity of these mice to infection .
In the present study , we demonstrated that SOCS3 expression in lymphoid and myeloid cell populations is essential for the resistance against M . tuberculosis in mice via distinct mechanisms . M . tuberculosis and BCG infections were potent stimuli for Socs3 expression in vivo and in myeloid cell populations in vitro . In line with previous studies , we found that SOCS3 expression in macrophages was mediated by MyD88 and NF-κB [29] . By using Socs3fl/fl LysM cre mice , our data suggest that myeloid SOCS3 expression contributes to a timely CD4+ cell-dependent IFN-γ-secretion rather than to improved innate effector immune mechanisms by macrophages . The observation that CD4+ cell-depleted Socs3fl/fl LysM cre and control mice had similar bacterial loads supports this hypothesis . Moreover , SOCS3-deficient and control BMM as well as pulmonary macrophages showed comparable intracellular mycobacterial growth , and IFN-γ diminished bacterial numbers with equal efficiency in Socs3fl/fl LysM cre and WT macrophages , as also recently demonstrated for T . gondii [12] . However , whether a defective IFN-γ secretion either by antigen-specific CD4+ T cells , or alternatively by NKT or CD8+ T cells underlies the susceptibility of Socs3fl/fl LysM cre mice to infection with M . tuberculosis remains to be determined . SOCS3-deficient BMM showed increased STAT3 activation and diminished secretion of TNF and IL-12 after infection with either M . tuberculosis or BCG . This confirms previous data showing reduced TNF and IL-12 release by Socs3fl/fl LysM cre and gp130F/F macrophages in response to LPS , when co-incubated with IL-6 [11] , [30] . Moreover , constitutively active STAT3 has been found to inhibit Il-12 p40 mRNA accumulation in LPS-stimulated BMDC [31] . Even though the LysM promoter is primarily active in neutrophils and macrophages , LysM promoter activity in DCs has previously been shown [26] . Socs3fl/fl LysM cre BMDC showed limited IL-12 production in response to mycobacterial stimulation and Il-12 p40 levels were also reduced in the lungs of infected Socs3fl/fl LysM cre and gp130F/F mice . Importantly , we found that IFN-γ levels were diminished in lungs of Socs3fl/fl LysM cre at 16 days but not at later time points after infection with M . tuberculosis . This delay in the establishment of immune protective responses might underlie the increased susceptibility of Socs3fl/fl LysM cre mice . In support of this notion , the resistance of different mouse strains to M . tuberculosis is associated with the timing of IFN-γ responses [32] . Although we observed a reduced secretion of IFN-γ by NK-cells during in vitro co-culture with M . tuberculosis-infected Socs3fl/fl LysM cre and control splenic DCs , the NK cell involvement in the enhanced susceptibility of Socs3fl/fl LysM cre mice is unlikely since NK cells were not required for controlling mycobacterial infections [33] , [34] . TNF is of major importance in the control of M . tuberculosis [35] . Although TNF secretion by infected SOCS3-deficient macrophages is reduced , TNF expression in the lungs of M . tuberculosis-infected Socs3fl/fl LysM cre mice was not diminished , suggesting that a role for TNF in the susceptibility to infection of Socs3fl/fl LysM cre mice is unlikely . The reduction of TNF and IL-12 levels observed in gp130F/F BMM was reversed when using gp130F/F Il-6−/− cells , and addition of rIL-6 further diminished the release of IL-12 and TNF by either mycobacteria-infected , or Pam3CSK4-stimulated Socs3fl/fl LysM cre BMM , suggesting that in macrophages , SOCS3 allows proper TNF and IL-12 secretion by hampering an IL-6-mediated inhibition of the secretion of these cytokines . Lungs from M . tuberculosis-infected Socs3fl/fl LysM cre and lck cre mice contained higher Il-6 mRNA levels than controls . However , neither macrophages nor T cells are likely to account for the elevated IL-6 levels in M . tuberculosis-infected SOCS3-deficient mice . Epithelial cells , fibroblasts and adipocytes have all been shown to secrete IL-6 in response to inflammatory stimuli [36] , [37] . Whether non-hematopoietic cells are major IL-6 producers during M . tuberculosis infection remains to be investigated . Socs3fl/fl lck cre mice showed a dramatically enhanced susceptibility to M . tuberculosis infection . However , SOCS3 expression in T cells was not required for the development of protective immune responses against M . tuberculosis in BCG-vaccinated mice . Thus , the requirement of SOCS3 in the control of mycobacterial infection depends on the mycobacterial species and on the immune status of the host . Our results show a hitherto unknown role for SOCS3 controlling the frequency of γδ+ T cells in different organs before and during M . tuberculosis infection while frequencies of CD4+ or CD8+ T cells were not regulated by SOCS3 . Moreover , γδ+ T cells impaired the transfer of protection by Socs3fl/fl lck cre CD4+T cells , suggesting that SOCS3 inhibits a non-redundant detrimental role of γδ+ T cells in the outcome of infection with M . tuberculosis . The detrimental activity of SOCS3-deficient γδ+ T cells contrasts with previous reports that have shown a minor role of WT γδ+ T cells in resistance to M . tuberculosis [38] , [39] . SOCS3 can impair the secretion of IL-17 [16] , [40] . The increased IL-17 mRNA and protein levels in Socs3fl/fl lck cre mice in M . tuberculosis- , but not in BCG-infected mice suggested that IL-17 levels might be causally associated to the increased susceptibility to M . tuberculosis infection . When γδ+ T cells but not CD4+ Socs3fl/fl lck cre T cells were adoptively transferred in Rag1−/− mice , an increased IL-17 secretion by lung cells and impaired transfer of protection against M . tuberculosis was observed . The function of IL-17 during primary mycobacterial infections is controversial since only after high dose intratracheal infection mice deficient in IL-17 were reported to be unable to control M . tuberculosis infection [41]–[43] . On the other hand , IL-17 has been implicated to increase bacterial dissemination , recruitment of neutrophils and morbidity during infection with M . tuberculosis [44]–[46] . Socs3fl/fl lck cre mice showed elevated levels of neutrophil-derived molecules and necrotic granulomas during M . tuberculosis infection . Our results confirmed a previous report indicating that γδ+T cells dominate IL-17 production during M . tuberculosis infection [28] . SOCS3 also hampered the secretion of IL-17 by γδ+ T cells when incubated with infected DCs or their supernatants . However , SOCS3 did not impair the development of γδ+ T cells that are capable of secreting IL-17 . Thus , the increased IL-17 levels in Socs3fl/fl lck cre mice are probably the consequence of both the increased numbers of γδ+ T cells and their unrestricted secretion of IL-17 in response to cytokines released by mycobacteria-stimulated DCs . The increased susceptibility to M . tuberculosis of Socs3fl/fl lck cre mice was not associated to an impaired IFN-γ secretion by antigen-specific T cells , suggesting that SOCS3 is not required for IFN-γ secretion by T cells and that γδ+ T cells do not modulate IFN-γ secretion by αβ+ T cells . Since SOCS3 regulates signalling via various receptors , we investigated whether signals mediated via the gp130 receptor account for the susceptibility to M . tuberculosis of SOCS3 conditional knockdown animals . We found that gp130F/F mice are highly susceptible to infection with M . tuberculosis . The susceptibility of gp130F/F is mediated by both IL-6-dependent as well as IL-6-independent signalling events , since gp130F/FIl-6−/− mice showed lower bacterial load than gp130F/F mice but higher bacterial levels than controls . The cytokine responses of gp130F/F and Socs3fl/fl LysM cre macrophages to mycobacterial infections were similar suggesting that the protective role of SOCS3 in myeloid cells is dependent on gp130 . IL-1β and IL-23 have been shown to stimulate IL-17 production by γδ+ T cells [47] , [48] , but SOCS3 impairs IL-23 signalling [27] . Thus , IL-1 β and IL-23 might mediate the elevated IL-17 secretion in Socs3fl/fl lck cre T cells . Moreover , IL-23 signalling [49] as well as the increased frequency of γδ+ T cells in Socs3fl/fl lck cre mice are both independent of gp130 signalling ( data not shown ) . Accordingly , T cells from highly susceptible gp130F/F mice transferred resistance to M . tuberculosis as previously shown for T . gondii infection [50] . Since gp130F/F were more susceptible to M . tuberculosis than Socs3fl/fl LysM cre mice , and lethally irradiated WT that were reconstituted with gp130F/F BM were more resistant to infection than gp130F/F or WT BM→ gp130F/F mice , we also suggest that gp130-dependent SOCS3-signalling in non-hematopoietic cells contributes to the control of infection with M . tuberculosis . Collectively , our data indicate that the expression of SOCS3 either in myeloid or in T cells is essential for control of M . tuberculosis infection ( Figure 10 ) . SOCS3 mediates protection through inhibition of IL-6/gp-130 signalling in myeloid cells , while gp130-independent , SOCS3-mediated mechanisms in T cells contribute to the control of M . tuberculosis .
The animals were housed and handled at the Dept . of Microbiology , Tumor and Cell Biology and the Astrid Fagreus Laboratory , Karolinska Institute , Stockholm , according to directives and guidelines of the Swedish Board of Agriculture , the Swedish Animal Protection Agency , and the Karolinska Institute ( djurskyddslagen 1988:534; djurskyddsförordningen 1988:539; djurskyddsmyndigheten DFS 2004:4 ) . The study was performed under approval of the Stockholm North Ethical Committee on Animal Experiments permit number N302/10 and N487/11 . Animals were housed under specific pathogen-free conditions . Socs3fl/fl mice containing loxP-flanked Socs3 alleles have been described before [51] . For a T cell-specific deletion Socs3fl/fl mice were bred with transgenic lck cre mice [52] and for a myeloid-specific deletion with transgenic LysM cre mice [19] . Offsprings were genotyped as described [51] and Socs3fl/fl littermates negative for cre expression were used as controls for all experiments . Gp130F/F mice possess a homozygous substitution of tyrosine ( Y ) 757 to phenylalanine ( F ) within the common IL-6 family receptor gp130 abrogating the SOCS3 binding site . Gp130F/F mice and their corresponding compound mutant homozygous null for IL-6 ( gp130F/F Il-6−/− ) or heterozygous for STAT3 ( gp130F/F Stat3+/− ) have been described previously [53] , [54] , all on a mixed C57Bl/6×129/Sv background which were used as controls . Rag1−/− mice were generated by homologous recombination in embryonic stem cells [55] and crossed to C57Bl/6 background . BCG Montreal and M . tuberculosis Harlingen and H37Rv were grown in Middlebrook 7H9 ( Difco , Detroit , MI ) supplemented with albumin , dextrose , catalase and , for BCG cultures , 50 µg/ml hygromycin ( Sigma , St . Louis , MO ) . BMM and BMDC were infected at the indicated multiplicity of infection ( MOI ) and after 4 hours cells were washed twice with PBS to remove extracellular bacteria . Mice were infected i . v . with 1×106 BCG or 250 M . tuberculosis Harlingen strain by aerosol using a nose-only exposure unit ( In-tox Products , Uppsala , Sweden ) [56] . A 15-ml suspension of 1×106 M . tuberculosis per ml was loaded into a nebulizer , and animals inhaled the bacteria aerosol for 20 min . Bacteria were quantified on Middlebrook 7H11 agar containing 10% enrichment of oleic acid , albumin , dextrose , catalase , 5 µg of amphotericin B per ml and 8 µg/ml polymyxin B grown for 3 weeks at 37°C . Briefly , single-cell suspensions from spleens were selected for CD4+ or CD90+ T cells with magnetic beads ( Miltenyi Biotech , Cologne , Germany ) as specified by the manufacturer . When indicated , CD90+ cells were depleted of γδ+ cells by FACS sorting . 1–3×106 T cells were inoculated i . v . into Rag1−/− mice . Two weeks after transfer , mice were infected with M . tuberculosis Harlingen . Mice were injected i . p . three consecutive days with 0 . 5 mg/mouse of Sepharose G affinity-purified anti-CD4 ( GK1 . 5 ) antibody one week before infection . CD4-specific depletion was controlled in blood using flow analysis . Two weeks after the first injection , additional 0 . 5 mg/mouse anti-CD4 antibody was injected to maintain CD4+ cells depleted . Mice were immunized with 5×106 heat-killed BCG ( 60 min at 80°C ) s . c . and boosted after 2 weeks with 2 . 5×106 heat-killed BCG . 4 weeks after the first injection , splenocytes from immunized and non-immunized mice were re-stimulated with 15 µg/ml PPD ( Statens Seruminstitut , Copenhagen , Denmark ) and supernatants collected after 72 h . Mice were immunized with 1×106 BCG i . v . and kept for 10 weeks before aerosol infection with M . tuberculosis together with non-immunized controls . Mice were sacrificed 4 weeks after M . tuberculosis strain Harlingen infection and bacterial loads were determined in the lungs . Recipient C57Bl/6×129/Sv and gp130F/F were irradiated 2× with 550 rad and received 5×106 BM cell from either C57Bl/6×129/Sv or gp130F/F mice . Mice were kept for 3 weeks on antibiotics ( Tribrissen in drinking water ) and were infected with M . tuberculosis 8 weeks after transfer . Bone marrow was extracted from tibia and femurs of mice and resuspended in Dulbecco's modified Eagle's medium ( DMEM ) containing glucose and supplemented with 2 mM L-glutamine , 10% FCS , 10 mM Hepes , 100 µg/ml streptomycin , 100 U/ml penicillin ( all from Sigma ) , and 30% L929 cell-conditioned medium ( as a source of macrophage-colony stimulating factor ) . Bone marrow cells were passed through a 70 µm cell strainer , plated and incubated for 6 days at 37°C , 5% CO2 . Bone marrow-derived macrophage ( BMM ) cultures were then washed vigorously to remove non-adherent cells , trypsinized , counted and cultured for one day at 37°C in 24 , 12 or 6 well plates . We have previously shown by immunofluorescence staining that these BMM are F4/80+ , CD14+ and Mac-3+ [57] . Mouse bone marrow-derived dendritic cells ( BMDC ) were differentiated as previously described [58] . Briefly , bone marrow was extracted from tibia and femurs and cell suspensions cultured in RPMI-1640 medium containing 10% FCS , 100 U/ml penicillin , 100 µg/ml streptomycin and 2 ng/ml GM-CSF ( Peprotech , Rocky Hill , NJ ) . Fresh medium and cytokine were replaced after 3 days . After six days of culture , loosely adherent cells were harvested and seeded in concentrations for infection . In some cases , harvested cells were further selected for CD11c expression with magnetic beads ( Miltenyi Biotech ) before seeding . Pulmonary macrophages were isolated as previously described [59] . Briefly , lungs from Socs3fl/fl and Socs3fl/fl LysM cre mice were dissected , digested with 1 . 8 U/ml dispase for 60 min at RT , followed by digestion with DNase ( both form Sigma ) for 30 min at 37°C . After red blood cell lysis , hematopoietic CD45+ lung cells were positively enriched using magnetic beads ( Miltenyi Biotech ) , and pulmonary macrophages selected by plastic-adherence . Forty-eight hours after seeding to culture plates , CD45+ adherent cells were washed four times with RPMI media . Splenic DCs were isolated as previously described [60] . Briefly , splenocyte suspensions were positively selected using anti-CD11c-coupled magnetic beads ( Miltenyi Biotech ) . This protocol lead to a purity >95% and an approximate yield of 0 . 5–1×106 DCs per spleen . 0 . 5×106 BMDC were seeded in 500 µl medium and infected with BCG ( MOI2 ) . After 24 h , supernatants were transferred on either CD90+ , CD4+ or γδ+ T cells . All cells had been separated for CD90 with magnetic beads ( Miltenyi Biotech ) followed by flow cytometry-based sorting ( for CD4+: CD3+ and CD4+ , for γδ+: CD3+ , CD4− , CD8− , γδ+ TCR ) . 72 h after transfer , supernatants were harvested and IFN-γ and IL-17 concentrations were determined by ELISA . Transcripts were quantified by real time PCR as previously described [56] . Hprt was used as a control gene to calculate the ΔCt values for individual samples . The relative amount of cytokine/Hprt transcripts was calculated using the 2− ( ΔΔCt ) method . These values were then used to calculate the relative expression of cytokine mRNA in uninfected and infected cells and tissues . Concentrations of cytokines in supernatants of stimulated cells were determined either by using cytometric bead array ( CBA ) mouse Th1/Th2/Th17 cytokine kit ( BD Biosciences , San Jose , CA ) or by enzyme-linked immunosorbent assays ( ELISA ) for IL-6 , IFN-γ , TNF ( BD Biosciences ) , IL-12 and IL-10 ( eBioscience , San Diego , CA ) and IL-17 ( R&D systems , Minneapolis , MN ) following the manufacturers' recommendations . Lungs were perfused with PBS through the heart before removal from mice . Following digestion with Collagenase D and DNase I , erythrocytes were lysed and single-cell suspensions prepared by filtering lung tissue through 70-µm nylon cell strainers . Single spleen cell suspensions were obtained by mechanical disruption , lysis of erythrocytes and straining over a 70-µm nylon mesh . Lung cells and splenocytes were stained for CD3 , CD4 , CD8 and γδ TCR ( all eBioscience ) or F4/80 and Gr1 ( BD Biosciences ) and fixed before acquisition . For determination of IL-17-producing cells , lung cells were incubated with 50 ng/ml phorbol myristate acetate ( PMA ) and 2 µg/ml ionomycin ( Sigma ) in presence of brefeldin A ( 5 µg/ml ) for 6 hours , stained with cell population-specific antibodies , fixed , permeabilized using leukocyte permeabilization reagent IntraPre ( Immunotech , Marseille , France ) and stained with anti-IL-17a ( eBioscience ) . CD11c+ splenic DCs were infected with M . tuberculosis H37Rv MOI5 for 4 hours , washed and cultured overnight with DX5+ NK cells separated from spleens with magnetic beads ( Miltenyi Biotech ) . The next day , cells were treated with brefeldin A ( 5 µg/ml ) for 4 hours , followed by a FACS stain for DX5 ( PE , BD Biosciences ) and intracellular IFN-γ ( eBioscience ) . Data were acquired in CyAn ADP flow cytometer ( Beckman Coulter ) and analyzed using FlowJo software ( Tree star Inc . , Ashland , OR ) . M . tuberculosis-infected and uninfected BMM were lysed and separated on 10% separating/5% stacking SDS-polyacrylamide gels . Samples were then transferred onto nitrocellulose membranes ( BioRad , Hercules , CA ) by electroblotting at 100 V , 250 mA for 80 min . Immunostaining was performed using polyclonal rabbit anti-phosphorylated ( Tyr701 ) STAT3 , total STAT3 ( Cell signaling technology , Beverly , MA ) or anti-actin ( Sigma ) . Membranes were then washed and incubated with horse-radish peroxidase-conjugated polyclonal goat anti-rabbit immunoglobulin ( DAKO ) and developed using ECL-Plus ( Amersham Biosciences , Buckinghamshire , UK ) and photographed using a Fuji intelligent dark box II digital camera . Formalin fixed left lungs of mice experimentally inoculated with M . tuberculosis were blocked on paraffin . From each lung sample 4 sections were obtained , one longitudinal along the long axis of the lobe and 3 across/transversal of the remaining piece of lung . The blocks were processed and sections were stained with haematoxylin-eosin . All sections were interpreted by the same pathologist ( D . G-W . ) and scored semi-quantitatively , blinded to the variables of the experiment . The following features were scored: | Tuberculosis is a severe disease caused by infection with the intracellular bacteria Mycobacterium tuberculosis . The protein “suppressor of cytokine signalling 3” ( SOCS3 ) inhibits the responses of cells to several cytokines and growth factors that signal via the STAT3 transcription factor . Since STAT3 is a major controller of immune and inflammatory responses , we studied the role of SOCS3 in the control of infection with M . tuberculosis . Mice deficient in the expression of SOCS3 either in myeloid or lymphoid cells were extremely susceptible to infection with M . tuberculosis as measured by elevated bacterial levels , worsened pathology and reduced survival . In myeloid cells , SOCS3 hindered a detrimental role of IL-6 . In absence of SOCS3 , IL-6 hampered the release of IL-12 by antigen-presenting cells . In T cells , SOCS3-mediated protection was independent of IL-6 signals , and of adequate IFN-γ secretion by antigen-specific T cells . Instead , SOCS3 inhibited the in vivo accumulation of , and the IL-17 secretion by γδ+ T cells . γδ+ T cells accounted in part for the susceptibility to M . tuberculosis infection of mice with SOCS3-deficient T cells . Thus , SOCS3 controls diverse immune mechanisms of myeloid and lymphoid cells that are required for containment of M . tuberculosis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] | [
"immunity",
"immunity",
"to",
"infections",
"immune",
"defense",
"immunology",
"biology",
"immunomodulation"
] | 2013 | Critical and Independent Role for SOCS3 in Either Myeloid or T Cells in Resistance to Mycobacterium tuberculosis |
Array-based comparative genomic hybridization ( Array-CGH ) is an important technology in molecular biology for the detection of DNA copy number polymorphisms between closely related genomes . Hidden Markov Models ( HMMs ) are popular tools for the analysis of Array-CGH data , but current methods are only based on first-order HMMs having constrained abilities to model spatial dependencies between measurements of closely adjacent chromosomal regions . Here , we develop parsimonious higher-order HMMs enabling the interpolation between a mixture model ignoring spatial dependencies and a higher-order HMM exhaustively modeling spatial dependencies . We apply parsimonious higher-order HMMs to the analysis of Array-CGH data of the accessions C24 and Col-0 of the model plant Arabidopsis thaliana . We compare these models against first-order HMMs and other existing methods using a reference of known deletions and sequence deviations . We find that parsimonious higher-order HMMs clearly improve the identification of these polymorphisms . Moreover , we perform a functional analysis of identified polymorphisms revealing novel details of genomic differences between C24 and Col-0 . Additional model evaluations are done on widely considered Array-CGH data of human cell lines indicating that parsimonious HMMs are also well-suited for the analysis of non-plant specific data . All these results indicate that parsimonious higher-order HMMs are useful for Array-CGH analyses . An implementation of parsimonious higher-order HMMs is available as part of the open source Java library Jstacs ( www . jstacs . de/index . php/PHHMM ) .
In recent years , the method of array-based comparative genomic hybridization ( Array-CGH ) [1]–[5] has been widely applied for the detection of DNA copy number polymorphisms between closely related genomes . Most Array-CGH studies have their focus in cancer research for the genome-wide identification of deletions and amplifications of genomic regions in tumor compared to healthy tissue [6]–[10] . With the availability of the genome sequence of the accession Columbia ( Col-0 ) of the model plant Arabidopsis thaliana [11] , studies comparing the genomes of different accessions have been performed using the Array-CGH approach to analyze evolutionary processes and phenotypic features at a molecular level [12]–[17] . All these studies require efficient bioinformatics methods for the precise identification of copy number polymorphisms from Array-CGH data . Over the last years , a large number of different methods for the identification of copy number polymorphisms from Array-CGH data have been developed including approaches based on Gaussian mixture models [18] , circular binary segmentation [19]–[21] , genetic local search algorithms [22] , [23] , dynamic programming [24]–[26] , hierarchical clustering [27] , sparse Bayesian learning [28] , variational methods [29] , [30] , smoothing techniques [31]–[34] , regression models [35] , [36] , or wavelets [37] , [38] . In-depth contributions to the comparison of different methods have been made by two studies [39] , [40] . Selected well-performing methods have been made publicly available by webservers [41]–[44] . Despite these different methods , the identification of copy number polymorphisms by methods based on Hidden Markov Models ( HMMs ) is very popular [45]–[61] providing a natural way for modeling genomic spatial dependencies present in Array-CGH data . Most of these HMM -based methods use three up to six states with specific Gaussian emission densities for the modeling of Array-CGH measurements . Greater differences exist in learning principles used for adapting models to data . The Baum-Welch algorithm [62]–[65] has been used in [47] , [48] , [56] , [59] , [61] for estimating the parameters of the HMM by maximizing the likelihood without integrating prior knowledge on the distribution of Array-CGH measurements . Due to specific model extensions , numerical estimations of the likelihood have been considered in [50] , [51] . Bayesian approaches using Markov Chain Monte Carlo simulations have been developed in [52]–[55] , [58] , a numerical Bayesian estimation has been applied in [57] , and a Bayesian Baum-Welch algorithm has been utilized in [60] . All these Bayesian approaches enable the integration of prior knowledge on the distribution of Array-CGH measurements for improving the identification of copy number polymorphisms . A characteristic of all these HMMs is that they are based on the mathematical theory of standard first-order HMMs [65] , [66] . This leads to a common limitation that all these HMMs can only model dependencies between Array-CGH measurements of two directly adjacent chromosomal regions . Yet , no attention has been paid to higher-order HMMs enabling the modeling of dependencies between a chromosomal region and its most recent predecessors that are clearly present in Array-CGH data ( e . g . Figure 1 ) . In contrast to the broad usage of first-order HMMs in applied sciences [66]–[68] , published applications of higher-order HMMs are relatively rare , but they have been demonstrated to be powerful extensions of first-order HMMs for several applications including speech recognition [69]–[76] , image segmentation [77]–[79] , robotic [80] , handwriting recognition [81] , or DNA and protein sequence analysis [82]–[85] . Extensions of the mathematical theory of first-order HMMs to higher-order HMMs are comprehensively described in [86]–[89] . The improved modeling of spatio-temporal dependencies by higher-order HMMs is realized by a more complex state-transition process defined on the basis of a higher-order Markov model reviewed in [90] . A limitation of this improved modeling is the exponential increase of transition parameters with increasing model order requiring growing amounts of data and computational resources for model training and evaluation . This has generally limited the usage of large model orders . Consequently , most existing studies have only focused on second-order HMMs [69]–[73] , [78] , [80] , [82] , [84] . To enable the usage of improved modeling characteristics of greater model orders by simultaneously overcoming the exponential increase of transition parameters , a fast incremental training has been developed in the domain of speech recognition [87] , [91] . This heuristic algorithm iteratively increases the model order by only including transition parameters that are required for the representation of the training data . That has led to higher-order HMMs with reduced model complexities [87] , [91] , [92] and to mixed-order HMMs [93]–[95] reaching improved results in speech recognition in comparison to first-order HMMs and standard higher-order HMMs . In addition , a variable-length HMM has been developed to improve the modeling of motion capture data [96] , [97] . The state-transition process of this model is defined by a variable memory Markov chain for which the transition parameters are determined by a minimum entropy criterion integrated into an extended Baum-Welch training . However , since implementations of both approaches for reducing the number of transition parameters are not publicly available and since algorithmic extensions would be necessary to enable the integration of prior knowledge , these models cannot directly be utilized for the analysis of Array-CGH data . Here , we develop the novel model class of parsimonious higher-order HMMs enabling the interpolation between a mixture model ignoring spatial dependencies and a higher-order HMM exhaustively modeling spatial dependencies between measurements of closely adjacent chromosomal regions . This interpolation is realized by incorporating a dynamic programming approach [98] , [99] into a specifically developed Bayesian Baum-Welch training algorithm enabling the integration of prior knowledge and a data-dependent reduction of transition parameters . Based on that interpolation , a parsimonious higher-order HMM can effectively model spatial dependencies between measurements of closely adjacent chromosomal regions . In an in-depth case study with the model plant Arabidopsis thaliana , we apply parsimonious higher-order HMMs to compare the genomes of the accessions C24 and Col-0 based on a publicly available Array-CGH data set . This enables the identification of DNA polymorphisms ( deletions or sequence deviations , amplifications ) in C24 with respect to the reference genome of Col-0 [11] . We evaluate and compare parsimonious higher-order HMMs against standard first-order HMMs and other existing methods by making use of deletions or sequence deviations identified in an independent array-based resequencing experiment of C24 [100] , [101] . Moreover , we perform a functional analysis of identified genomic differences revealing novel details of differences between C24 and Col-0 , and we also consider widely used human cell lines [102] for additional model comparisons .
In this section , the Arabidopsis Array-CGH data set is introduced and candidate regions of deletions or sequence deviations for model evaluation identified in resequencing data are considered . This section provides the basics of parsimonious higher-order HMMs . In the following , these models are introduced , a prior distribution for integrating prior knowledge into the training is specified , a model-specific Bayesian Baum-Welch training algorithm is developed , and details to the parameter initialization are given . Finally , a link to related work is given .
The modeling of the partial autocorrelation function [108] of the Arabidopsis Array-CGH profiles by higher-order HMMs was initially studied to determine a range of model orders for an in-depth analysis by parsimonious HMMs . The partial autocorrelation function quantifies linear dependencies between measurements of chromosomal regions in close chromosomal proximity for an increasing distance of regions . As shown in Figure 1 , such dependencies are clearly present in the Arabidopsis Array-CGH profiles motivating the application of HMMs of different model orders for modeling of these dependencies . Initially , HMMs of order zero up to five were trained on the Arabidopsis Array-CGH profiles using the Bayesian Baum-Welch algorithm . Next , each HMM was used to sample artificial profiles with log-ratios . These profiles were used to compute the mean partial autocorrelation function modeled by each HMM . As expected from theory , the HMM of order zero ( mixture model ) does not model dependencies between log-ratios in any chromosomal distance . The first-order HMM shows a clear improvement in comparison to the mixture model , but especially HMMs of order three up to five reached the best , nearly identical approximation of the partial autocorrelation function of Array-CGH profiles . A better modeling of the partial autocorrelation function by higher-order HMMs is expected from theory because of their more complex state-transition processes enabling an improved modeling of spatial dependencies compared to HMMs with a smaller model order . Still , none of these HMMs was able to perfectly approximate the partial autocorrelation structure of the Array-CGH profiles . But , despite of that , this study helped to determine a range of model orders for further analyses . The results of this study are summarized in Figure S3 in Text S1 . Based on this initial study with higher-order HMMs , parsimonious HMMs of order one up to five are subsequently investigated in detailed studies to analyze their abilities to identify DNA polymorphisms between C24 and Col-0 . An Array-CGH data set by [103] comparing the genomes of the accessions C24 and Col-0 of A . thaliana is used to identify polymorphic regions between both genomes by parsimonious higher-order HMMs . These models are evaluated based on deletions or sequence deviations determined in [101] for the genome of C24 in comparison to the reference genome of Col-0 using publicly available array-based resequencing data [100] . The mapping of these polymorphic regions to corresponding chromosomal regions in the Array-CGH data set shows an obvious coupling with potential deletions or sequence deviations present in the Array-CGH data set ( Figure 2b ) . These potential deletions or sequence deviations are used as reference for model comparisons . Parsimonious higher-order HMMs of different model complexities were adapted to the Array-CGH data using the developed Bayesian Baum-Welch training . For each model , all chromosomal regions in the Array-CGH data set were ranked in decreasing order of their state-posterior probabilities of state ‘’ modeling deletions or sequence deviations . Using the knowledge about potential deletions or sequence deviations in the Array-CGH data set , the identification of these polymorphic regions was quantified for each model in terms of the true-positive-rate ( TPR ) at 1% false-positive-rate ( FPR ) . The mean TPRs obtained for twenty different initializations of each model at 1% FPR are shown in Figure 4a ( see Figure S4a in Text S1 for standard deviations of TPRs and see Figure S5a in Text S1 for FPRs at fixed TPR ) . The application of parsimonious higher-order HMMs has clearly improved the identification of deletions or sequence deviations in comparison to the standard first-order HMM . Moreover , parsimonious higher-order HMMs with much smaller model complexities than corresponding higher-order HMMs can also reach a clearly improved accuracy for identifying polymorphic regions in comparison to corresponding higher-order models . The best parsimonious higher-order HMMs have model complexities in the range of 3 up to 9 leaves . This range of model complexities includes parsimonious HMMs of order two up to five that nearly reach the same performance for identifying deletions or sequence deviations . State-context trees underlying well-performing parsimonious HMMs of order three up to five are clearly reduced leading to model complexities comparable with that of parsimonious second-order HMMs . Thus , not all higher-order dependencies are required for reaching a good performance at the stringent level of 1% FPR . Similar results are shown in Figure S6a in Text S1 using a less restrictive mapping of the independently determined deletions or sequence deviations from [101] to the Array-CGH data set for model comparisons . Parsimonious higher-order HMMs have initially been compared against the standard first-order HMM and higher-order HMMs at a stringent FPR of 1% . Next , these models are compared at a less stringent FPR of 2 . 5% . That leads to an identification of deletions or sequence deviations comparable with those obtained by applying the state-posterior decoding algorithm [65] , [89] that computes for each chromosomal region in the Array-CGH data set the most likely state under the given model . The results are shown in Figure 4b ( see Figure S4b in Text S1 for standard deviations of TPRs and see Figure S5b in Text S1 for FPRs at fixed TPR ) . Generally , parsimonious higher-order HMMs reach a higher accuracy for the identification of deletions or sequence deviations than the standard first-order HMM . The best parsimonious higher-order HMMs also reach an accuracy that is comparable or slightly better than that of corresponding higher-order HMMs . This accuracy is obtained at much lower model complexities than for higher-order HMMs . That can become particularly useful for avoiding overfitting in small data . In comparison to the results at 1% FPR , the complexity of the best models is more shifted into the range of 9 to 27 leaves at 2 . 5% FPR ( Figure 4 and Figure S4 in Text S1 ) . This indicates that the identification of polymorphic regions is more complicated . Because at a higher FPR , the Array-CGH measurements of additionally identified polymorphic regions are more similar to that of non-polymorphic regions . These difficulties tend to be managed best by parsimonious higher-order HMMs . The best models in Figure 4b are among the fourth-order parsimonious higher-order HMMs . A tree structure of one of the best models is shown in Figure 5 . The underlying parsimonious fourth-order HMM has still some specific fourth-order transition probabilities for the states ‘’ ( deletion or sequence deviation ) and ‘’ ( non-polymorphic ) , whereas those of state ‘’ ( amplification ) are completely reduced to second-order transition probabilities . This unbalanced reduction of transition parameters tends to be coupled with the asymmetry of the Array-CGH measurement distribution in Figure 2a . Most of the chromosomal regions in the Array-CGH data set are non-polymorphic , a small proportion tends to be deleted or affected by sequence deviations , whereas only a very small proportion of regions tends to be amplified . The tree structure indicates that these tendencies are transferred to the number of transition parameters per state . This parsimonious fourth-order HMM is considered in all further studies with the Arabidopsis Array-CGH data set because of its good performance at the level of 2 . 5% FPR comparable with the results obtained by applying the state-posterior decoding algorithm enabling an in-depth analyses of genomic differences between C24 and Col-0 . Generally , similar tendencies like shown in Figure 4b are also present in Figure S6b in Text S1 considering a less restrictive mapping of the independently determined deletions or sequence deviations from [101] to the Arabidopsis Array-CGH data set for model comparisons . Here , the well-performing parsimonious fourth-order HMM is compared against other existing methods on the Arabidopsis data set . Then , another widely considered human cell lines data set by [102] is used for additional model comparisons . Subsequent to this , the focus is on comparative genomics of the accessions C24 and Col-0 of A . thaliana . The genome annotation of the reference genome of Col-0 provides the opportunity to investigate what is functionally behind chromosomal regions where the genomes of C24 and Col-0 differ . The parsimonious fourth-order HMM with underlying parsimonious tree structure in Figure 5 was applied to identify polymorphic regions in the Arabidopsis Array-CGH data set . The state-posterior decoding algorithm [65] , [89] was used to classify each chromosomal region in the Array-CGH data set either as a deletion or sequence deviation , as unchanged , or as an amplification in C24 with respect to the reference genome of Col-0 . This algorithm assigns the most likely state of the three-state architecture of the HMM ( Figure S2 in Text S1 ) to each chromosomal region measured in the Array-CGH data set . The identification of deletions or sequence deviations by state-posterior decoding is comparable to that shown in Figure 4b . In total , about 4 . 7% ( 17 , 306 of 364 , 339 ) of all chromosomal regions of the reference genome of Col-0 were identified as being affected by deletions or sequence deviations in the genome of C24 , and about 0 . 2% ( 855 of 364 , 339 ) of all chromosomal regions were identified as amplified in C24 ( Table S1 ) . This asymmetry in predictions is expected from the distribution of measurements in the Array-CGH data set ( Figure 2 ) reflecting the design of the tiling array that only represents chromosomal regions present in the reference genome of Col-0 [11] . Of the 17 , 306 chromosomal regions identified as being affected by deletions or sequence deviations , 2 , 647 are singletons consisting of only one tile and 76 . 5% of these singletons are containing a micro-deletion or sequence deviation in C24 compared to Col-0 that is covering at least 40% of the underlying tile . In all , genomic regions affected by deletions or sequence deviations represent about 5 . 59 Mb of the Col-0 reference genome . This is in good accordance with the findings in [100] , [101] . Subsequently , all identified genomic differences are analyzed in detail . The development of parsimonious higher-order HMMs for the analysis of Array-CGH data has been motivated by the observation of strong spatial dependencies between measurements in close chromosomal proximity . A parsimonious higher-order HMM represents an interpolation between a mixture model ignoring spatial dependencies and a higher-order HMM exhaustively modeling spatial dependencies . To enable this interpolation , the mathematical theory of widely used first-order HMMs has been extended . A central point is the extension of the Bayesian Baum-Welch training by incorporating a dynamic programming approach [98] , [99] enabling a data-dependent modeling of spatial dependencies . In a detailed study based on Array-CGH data for comparing the genomes of the Arabidopsis thaliana accessions C24 and Col-0 , parsimonious higher-order HMMs clearly improved the identification of deletions or sequence deviations in comparison to typically used first-order HMMs and other existing methods . Especially , parsimonious HMMs of order three up to five with clearly reduced model complexities in comparison to corresponding higher-order HMMs reached the best results . In-depth functional analyses of identified DNA polymorphisms revealed that most of these genomic differences between C24 and Col-0 are caused by transposons . Genic regions as well as 5′ and 3′ untranslated regions are less affected , but still genes with functions in ATP-binding , cellular signaling , or cell pathogen defense have been found to be specifically affected by deletions or sequence deviations in C24 in comparison to the reference genome of Col-0 . These findings are in accordance with other studies [17] , [100] and might indicate specific environmental adaptations of both accessions . Additionally , a superfamily classification of transposons has revealed that specific retrotransposon and DNA transposon superfamilies tend to be more involved than others in driving the evolution of C24 and Col-0 . Additional model evaluations performed on widely considered human cell lines showed that parsimonious HMMs are also well-suited for the analysis of non-plant-specific Array-CGH data sets . All these results indicate that parsimonious higher-order HMMs are useful tools for the analysis of Array-CGH data . Potential future applications could include other domains in which standard first-order HMMs are frequently used . This might include the HMM -based analysis of ChIP-chip data [117]–[120] or the analysis of next-generation sequencing data [121]–[125] . | Array-based comparative genomics is a standard approach for the identification of DNA copy number polymorphisms between closely related genomes . The huge amounts of data produced by these experiments require efficient and accurate bioinformatics tools for the identification of copy number polymorphisms . Hidden Markov Models ( HMMs ) are frequently used for analyzing such data sets , but current models are based on first-order HMMs only having limited capabilities to model spatial dependencies between measurements of closely adjacent chromosomal regions . We develop parsimonious higher-order HMMs enabling the interpolation between a mixture model ignoring spatial dependencies and a higher-order HMM exhaustively modeling these dependencies to overcome this limitation . In an in-depth case study with Arabidopsis thaliana , we find that parsimonious higher-order HMMs clearly improve the identification of copy number polymorphisms in comparison to standard first-order HMMs and other frequently used methods . Functional analysis of identified polymorphisms revealed details of genomic differences between the accessions C24 and Col-0 of Arabidopsis thaliana . An additional study on human cell lines further indicates that parsimonious HMMs are well-suited for the analysis of Array-CGH data . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results/Discussion"
] | [
"algorithms",
"computer",
"science",
"model",
"organisms",
"plant",
"and",
"algal",
"models",
"plant",
"biology",
"biology",
"computational",
"biology"
] | 2012 | Parsimonious Higher-Order Hidden Markov Models for Improved Array-CGH Analysis with Applications to Arabidopsis thaliana |
Morphological variation in natural populations is a genomic test bed for studying the interface between molecular evolution and population genetics , but some of the most interesting questions involve non-model organisms that lack well annotated reference genomes . Many felid species exhibit polymorphism for melanism but the relative roles played by genetic drift , natural selection , and interspecies hybridization remain uncertain . We identify mutations of Agouti signaling protein ( ASIP ) or the Melanocortin 1 receptor ( MC1R ) as independent causes of melanism in three closely related South American species: the pampas cat ( Leopardus colocolo ) , the kodkod ( Leopardus guigna ) , and Geoffroy’s cat ( Leopardus geoffroyi ) . To assess population level variation in the regions surrounding the causative mutations we apply genomic resources from the domestic cat to carry out clone-based capture and targeted resequencing of 299 kb and 251 kb segments that contain ASIP and MC1R , respectively , from 54 individuals ( 13–21 per species ) , achieving enrichment of ~500–2500-fold and ~150x coverage . Our analysis points to unique evolutionary histories for each of the three species , with a strong selective sweep in the pampas cat , a distinctive but short melanism-specific haplotype in the Geoffroy’s cat , and reduced nucleotide diversity for both ancestral and melanism-bearing chromosomes in the kodkod . These results reveal an important role for natural selection in a trait of longstanding interest to ecologists , geneticists , and the lay community , and provide a platform for comparative studies of morphological variation in other natural populations .
Color variation in natural populations is a useful entry point to investigate the evolution of mammalian phenotypic diversity . Pigmentary differences can be readily observed and quantified; much is known about the underlying biochemistry and cell biology , and there are many examples of apparent convergent evolution . Fundamental questions such as the relative roles played by regulatory vs . protein-coding variation , the phylogenetic origin of similar phenotypes shared among different lineages , and the potential impact of pleiotropic mutations can all be explored from a molecular genetic perspective [1–3] . Mammalian color diversity is particularly apparent in the Felidae , with a wide range of patterns and base colors represented among 37 species with a common ancestor ~11 million years ago [4 , 5] . In addition to color patterns such as rosettes , stripes , or spots , two characteristic pigmentary phenotypes appear in multiple species [6 , 7] . The ticked phenotype , a brushed appearance that hides dark tabby markings , is characteristic of 4 wild species ( the lion , puma , caracal , and jaguarundi ) , and melanism , a black coat with residual or “ghost” dark tabby markings in some individuals , has been described in 13 wild species ( see [8] for references ) . Felid melanism is especially interesting because it is polymorphic within each species , but its possible adaptive significance has received little attention to date [9] . In rodents and in birds , an adaptive role of melanism is well-established from ecological and field studies [1 , 2 , 10 , 11] , but the evolutionary forces that underlie melanism in larger mammals are less clear , with examples of introgression in the case of North American black wolves [12] , and a founder effect in the case of Malaysian black leopards [13] . A recent study by Allen et al . [9] found that the presence of melanism across felid species is correlated with habitat and/or behavioral diversity , and suggested a common underlying mechanism of disruptive selection . A particularly interesting group of animals from this perspective is an endemic lineage of Neotropical wild cats that includes multiple species exhibiting melanism as a naturally occurring phenotypic variant . This lineage comprises eight species of small cats belonging to the genus Leopardus , which diverged from each other after the colonization of South America by a common ancestor ~3 million years ago ( MYA ) [4 , 5] . In three of these species , the pampas cat ( L . colocolo ) , the kodkod ( L . guigna , referred to locally as the güiña or hüiña in Chile or Argentina , respectively ) , and Geoffroy’s cat ( L . geoffroyi ) , melanistic individuals comprise 20% or more of the population in some areas [14–17] . Inter-species hybridization within Leopardus has been reported [18 , 19]; thus , high frequencies of melanism within these species could reflect introgression , an ancient trans-specific polymorphism , or independent evolution after divergence from a common ancestor . To distinguish among these possibilities and to better understand how similar phenotypes evolve in closely related felids , we used massively parallel targeted resequencing to delineate the molecular cause and population genetic history of melanism mutations in each of the three species . The closest reference genome to Leopardus spp . is that of the domestic cat , with a last common ancestor ~6 MYA . As an alternative to oligonucleotide capture hybridization , which is constrained to protein-coding regions of closely related species [20] , we adapted an approach in which large insert clones ( from bacterial artificial chromosome or fosmid vectors ) are used as a template to generate biotinylated RNA hybridization probes for solution hybridization capture of homologous segments in the three species . This approach , termed CATCH-Seq ( for “Clone Adapted Template Capture Hybridization” ) was developed to assess dense human genetic variation in regions that are otherwise difficult to assess by other methods of genotyping [21] , e . g . the MHC , but , as shown here , can be extended to facilitate resequencing of targeted genomic regions in any organism for which clone-based reagents from a related species are available . We identify different molecular causes of melanism in the pampas cat , the kodkod , and the Geoffroy’s cat . Haplotype-based analyses point to distinctive population histories in all three species , with one—the pampas cat—exhibiting strong evidence of selection for melanism . Our results have implications for pigmentary biology and genetics , and yield new insight into the evolution of wild felids .
The pampas and Geoffroy’s cats are widely distributed through South America , while the kodkod is geographically more restricted ( Fig . 1 ) . All three species , which diverged from a common ancestor ~2 . 5 MYA , are relatively small , inhabit savanna and grassland areas as well as temperate rainforests ( particularly the kodkod ) , and prey on small mammals , birds , and/or amphibians and reptiles . To mitigate against confounding effects of population structure , we sampled individuals from well-defined geographic regions ( Fig . 1 ) in Emas National Park ( Brazil ) , the southern region of Rio Grande do Sul ( Brazil ) , and Chiloé Island ( Chile ) , where the frequencies of melanism are 25% , 20% , and 29% for the pampas cat , Geoffroy’s cat and kodkod , respectively ( Table 1 , Table S1 in S1 Data ) . CATCH-Seq is best suited to explore sequence variation once specific regions of interest are identified; therefore we initiated molecular analyses of melanism by conventional capillary-based sequencing of candidate genes . In mammals , melanism represents a shift in the balance between red-yellow pheomelanin and brown-black eumelanin , controlled by the activity of a melanocyte-specific G protein-coupled receptor , the Melanocortin 1 receptor ( MC1R ) [22–25] . The most common causes of melanism mutations are gain-of-function alterations in MC1R , or loss-of-function alterations in ASIP , which encodes Agouti signaling protein , a paracrine signaling molecule that inhibits MC1R signaling [26] . Consistent with the nature of the mutations , melanism caused by MC1R mutations is dominantly inherited , while melanism caused by ASIP mutations is recessively inherited; however , the inheritance pattern of melanism in pampas , Geoffroy’s cat , and the kodkod is not known . ( Additional , less frequent causes of melanism in some mammals include mutations of beta-defensin 103 [an alternative ligand for MC1R] [27] , Attractin [an accessory receptor for ASIP] [28] , or Mahogunin [an E3 ubiquitin ligase that acts upstream of the MC1R] [29] ) . Using the domestic cat genome as a starting point for the design of oligonucleotide PCR primers , we determined the protein-coding sequence of ASIP and MC1R in 18 pampas cats ( 10 melanistic ) , 23 Geoffroy’s cats ( 7 melanistic ) , and 16 kodkods ( 5 melanistic ) . Animals were selected from specific geographic areas as described above , and collection was independent of coat color phenotype ( Table S1 in S1 Data ) . In each species , we identified missense alterations whose molecular nature , pattern of association , and apparent mode of inheritance made a compelling case for them being the cause of melanism ( Table 1 , Table S4 in S1 Data , Fig . 1 ) . In both the pampas cat and the kodkod , mutations in ASIP were identified that are predicted to cause a loss-of-function , and homozygosity for these mutations was completely associated with melanism . In the pampas cat , an inferred Arg to Cys substitution in the C-terminal region of ASIP ( p . R120C ) lies in the critical RFF loop ( e . g . a triplet motif made of one arginine and two phenylalanine residues ) , required for binding to the MC1R [30] , and also creates an odd number of cysteine residues , predicted to interfere with folding of the disulfide-rich domain . Similarly , in the kodkod , an inferred Cys to Tyr substitution in ASIP ( p . C126Y ) affects the key disulfide bond that stabilizes the RFF loop [30] . All melanistic pampas cat individuals were homozygous for ASIPR120C ( Chi-square = 18 , p<0 . 005 ) and all melanistic kodkod indivduals were homozygous for ASIPC126Y ( Chi-square = 16 , p<0 . 005 ) . No other ASIP coding sequence variants were identified in the pampas cat or in the kodkod ( Table S5 in S1 Data ) ; four intraspecific variants in MC1R were present in the pampas cat , but none of them exhibited an association with melanism ( Table S4 in S1 Data ) . In Geoffroy’s cat , we identified an MC1R mutation as the likely cause of melanism . All melanistic individuals were heterozygous for four variants that predicted three nonsynonymous substitutions , p . C125R , p . T177I , and p . G194S ( Table S6 in S1 Data ) . All individuals were either homozygous for the ancestral alleles ( 125Cys , 177Thr , 194Gly ) or heterozygous for the derivative alleles ( 125Arg , 177Ile , 194Ser ) ; we observed complete association of heterozygosity for the derivative alleles with melanism ( Chi-square = 16 , p<0 . 005 ) . In the laboratory mouse , a Cys to Arg substitution at the site homologous to felid residue 125 causes constitutive activation of the receptor [31] , and the exact same change is thought to be responsible for melanism in the Alaska silver fox [32]; therefore , we consider C125R to be the likely causative alteration in the Geoffroy’s cat . Transmission of melanism has not been described in these species , but the amino acid substitutions and patterns of association predict that melanism in the pampas cat and kodkod is recessively inherited , and that melanism in the Geoffroy’s cat is dominantly inherited . Frequencies of the melanism alleles are 0 . 71 , 0 . 5 , and 0 . 15 for the pampas cat , the kodkod , and Geoffroy’s cat , respectively , and are consistent with the field-based assessments of melanism frequency in the populations from which each sample was drawn ( Table 1 ) . Like other wild felids that occur at low densities , small sample sizes constrain the power of studies based on estimates of genotype frequency . With that caveat , a relatively high prevalence of melanism due to independent mutations in these three species suggests that the phenotype is not deleterious [33] . Originally developed as a cost-effective way for targeted resequencing of human regions that are highly polymorphic , clone-based target capture is also uniquely suited for studying non-model organisms for which large insert genomic libraries from a closely related organism are available . To survey genetic variation within and around ASIP and MC1R in Leopardus spp . , we identified a series of fosmids from the domestic cat reference genome that contain and surround each locus ( Fig . 2 ) . For ASIP , we used 8 fosmids that contain 299 kb of DNA within an 842 kb interval; for MC1R , we used 5 fosmids that contain 251 kb of DNA within a 409 kb interval ( Fig . 2 , Table S7 in S1 Data ) . Illumina paired-end libraries were constructed from sheared genomic DNA of 57 animals ( Table S1 in S1 Data ) . From 54 libraries that passed quality control , material selected by CATCH-Seq using biotinylated probes prepared from the 13 fosmids was sequenced in a multiplexed format; a total of 1 , 053 , 331 , 306 paired-end reads ( 50 bp ) were collected . Summary alignment and mapping statistics are presented in Table 2 . The current domestic cat assembly ( Felis_catus-6 . 2 ) spans 2 . 43 Gb and represents ~14x Q20 coverage , but has not yet been fully annotated . Approximately 70% of our sequence reads aligned to the cat genome; of these , ~58% aligned to one of the ASIP or MC1R target fosmids considering the entire sequence of the fosmids , and ~7% aligned to the ASIP or MC1R target considering the repeat-masked sequence of the fosmids , consistent with an enrichment value of ~500–~2500-fold ( Table 2 , Table S9 in S1 Data ) . The mean coverage of each fosmid was 150x , with 71%–99% ( mean , 92% ) of non-repetitive sequence covered at > 10x ( Table S10 in S1 Data ) . Nucleotide divergence between the domestic cat and Leopardus spp . was ~1–2% , making the domestic cat unsuitable as a reference for variant-calling; therefore , we used SAMtools to develop a consensus sequence for each species separately ( combining melanistic and non-melanistic individuals within each species ) . The species-specific consensus sequence was then used to establish high-confidence variant calls for each individual according to best practices for the Genome Analysis Toolkit , and likely haplotype structure was inferred using Beagle . We identified 149 , 383 , and 474 SNPs in the kodkod , pampas cat , and Geoffroy’s cat , respectively , which yield estimates of percent nucleotide diversities of 0 . 0180 ( kodkod ) , 0 . 0451 ( pampas cat ) , and 0 . 0572 ( Geoffroy’s cat ) ( Table S11 in S1 Data ) . The different nucleotide diversities are not artifacts of ascertainment , since alignment statistics and coverage are similar for the three species ( Supplemental Tables S9 , S10 ) . ( Coverage of fosmids in the Geoffroy’s cat was not as complete , ~15x , as in the kodkod ( ~35x ) or pampas cat ( ~20x ) [Table S10 in S1 Data] , but more SNPs were identified in Geoffroy’s cat than in the other two species ) . Instead , the differences in nucleotide diversity likely reflect species-specific demographic histories as described below . To help validate our pipeline , and to further explore the pattern of variant distribution between and within species , we carried out a phylogenetic analysis for the ASIP and MC1R regions using a Bayesian Markov Chain Monte Carlo ( MCMC ) approach [34] , in which individual chromosomes for each locus were clustered and the clade credibility for the major nodes was examined . Although this approach does not account for recombination and therefore is not informative with regard to gene genealogy within species , the tree topologies should approximate evolutionary relationships among species , and also provide some assessment of quality control with regard to potential population structure as well as the accuracy of species identification . As depicted in Fig . 3 , the posterior clade probability for the nodes representing the divergence of the three species is very high for both ASIP and MC1R loci , with basal relationships that recapitulate the known topology in which the kodkod and Geoffroy’s cat are more closely related to each other than to the pampas cat . For the pampas cat and Geoffroy’s cat , chromosomes bearing melanism mutations ( ASIP for pampas cat , MC1R for Geoffroy’s cat ) do not cluster together; for the kodkod , there is some clustering of melanistic ASIP chromosomes that may reflect recent population history , as described below . A fine-scale view of haplotype patterns and nucleotide diversity allows additional insight into the evolutionary history underlying felid melanism . Positive selection is expected to give rise to reduced nucleotide diversity at the melanism locus ( ASIP in the pampas cat and kodkod , MC1R in Geoffroy’s cat ) , but not at the “control” locus ( MC1R in the pampas cat and kodkod , ASIP in Geoffroy’s cat ) . Fig . 4 depicts the pattern of variation for individual chromosomes at the ASIP ( pampas cat and kodkod ) and MC1R ( Geoffroy’s cat ) loci , grouped according to whether the chromosome carries an ancestral or derivative allele , and whether it was observed in a melanistic ( m/m or +/M ) or a non-melanistic ( +/+ or +/m ) animal . At the level of individual haplotypes , multiple variants in the pampas cat define a large haplotype block ( Fig . 4A ) that extends throughout the AsipD fosmid and that contains the ASIP melanism mutation . Haplotype diversity is also reduced for melanism-bearing chromosomes in the pampas cat: within the AsipD fosmid , there are 3 haplotypes among 26 melanism-bearing chromosomes compared to 8 haplotypes among 10 ancestral chromosomes . In the Geoffroy’s cat , four variants within the Mc1rC fosmid delineate a ~1kb haplotype block almost perfectly associated with melanism ( Fig . 4B ) ; beyond this central core , there is no obvious distinction in haplotype structure or haplotype diversity between chromosomes that carry melanism mutations and those that do not . Haplotype patterns in the kodkod reveal striking differences between melanism-bearing and ancestral chromosomes , but relatively little within-group difference; in the AsipD fosmid , there are 3 haplotypes among 10 melanism-bearing chromosomes , and 5 haplotypes among 14 ancestral chromosomes ( Fig . 4C ) . We next compared nucleotide diversity between melanism-bearing and ancestral chromosomes ( Fig . 5A , B , C ) , using fosmid origin ( Fig . 2 ) as a window for proximity to the causative variant . In the pampas cat , melanism-bearing chromosomes exhibit a clear signature of selection ( Fig . 5A ) , with reduction of nucleotide diversity that extends across a ~500 kb interval surrounding the causative variant ( ASIP fosmids AsipB , AsipC , AsipD , AsipE , AsipF , AsipG ) , but levels of nucleotide diversity similar to ancestral chromosomes in flanking fosmids ( ASIP fosmids AsipA , AsipH ) . In the Geoffroy’s cat ( Fig . 5B ) , melanism-bearing chromosomes also exhibit a window of reduced nucleotide diversity in a region ~50 kb upstream of the causative variant ( MC1R fosmid Mc1rB ) , although the extent of the difference is not as striking as in the pampas cat . In the kodkod , nucleotide diversity is similarly low for both melanism-bearing and ancestral chromosomes in the AsipD fosmid; in two adjacent fosmids ( AsipE , AsipF ) , levels are modestly higher for melanism-bearing compared to ancestral chromosomes ( Fig . 5C ) . We also calculated nucleotide diversity for melanistic compared to non-melanistic individuals ( Fig . 5D , E , F ) ; although less sensitive for evaluating differences in derivative vs . ancestral chromosomes ( since non-melanistic pampas cats and kodkods are a mixture of +/m and +/+ ) , this approach allows a direct comparison of nucleotide diversity at a locus unlinked to the causative variant ( MC1R for the pampas cat and kodod; ASIP for the Geoffroy’s cat ) . There are no systematic differences between melanistic and non-melanistic individuals for loci that are unlinked to the causative variant , whereas differences observed for the pampas cat and Geoffroy’s cat in a by-chromosome analysis of the causative locus ( Fig . 5A , B ) persist in a by-individual analysis of the causative locus ( Fig . 5D , E ) . A signature of selection in the pampas cat is also readily apparent from graphs of extended haplotype homozygosity ( EHH ) and haplotype bifurcation diagrams ( Fig . 6A ) ; EHH for melanism-bearing chromosomes in the pampas cat extends for hundreds of kilobases with relatively few bifurcations . By contrast , the Geoffroy’s cat and the kodkod exhibit similar EHH for ancestral and derivative chromosomes ( Fig . 6B , C ) . Haplotype diversity , nucleotide diversity , and EHH in the kodkod are unusual: nucleotide diversity graphs for melanism-bearing and ancestral chromosomes intersect and cross within the AsipD fosmid ( Fig . 5C ) , and the EHH graphs in the kodkod are similar for ancestral and melanism-bearing chromosomes ( Fig . 5C ) . Although this pattern does not lend itself to a simple interpretation , genetic drift is likely to be a contributing factor given the overall reduction of nucleotide diversity; 0 . 018% in the kodkod compared to 0 . 045% and 0 . 057% in the pampas and Geoffroy’s cats , respectively .
The application of short-read DNA sequencing technology to longstanding questions in mammalian evolution offers significant opportunities as well as challenges . Morphological variation within a species , or between closely related species , is often hypothesized to serve an adaptive role , a question that can be informed by knowing the molecular basis and evolutionary history of the underlying events . This effort can be challenging , however , in settings where a reference genome does not yet exist . As shown here , our application of a clone-based targeted capture-resequencing strategy reveals distinct evolutionary histories for melanism in three closely related Leopardus species . The results have implications for understanding the biological and genetic basis of melanism in other mammals , inform ongoing efforts in conservation biology , and provide a molecular example that can be applied more widely in population and evolutionary genomics . Melanism is one of the most common color morphs in domestic and natural populations of birds and mammals , and in most species it is caused by a loss-of-function mutation in ASIP or a gain-of-function mutation in MC1R [3 , 10 , 26] . The same now appears to be true in felids; including the three Leopardus species presented here , ASIP or MC1R mutations have now been found to cause melanism in 8 felid species [8 , 35] , a trend that seems likely to continue for the remaining 5 felid species ( the jungle cat , the marbled cat , the bobcat , the tigrina , and the serval ) , in which melanism is recognized but not characterized from a molecular genetic perspective . Loss-of-function mutations in two additional and related pigment type-switching components , Attractin and Mahogunin , cause melanism in the laboratory mouse or rat , but these mutations also cause brain abnormalities and would probably be subject to negative selection in natural populations [36] . In North American wolves , an unusual gain-of-function alteration in beta-defensin 103 causes dominantly inherited melanism and was acquired by hybridization with domestic dogs; however , beta-defensin variants that affect pigmentation have not been recognized outside of dogs , wolves , or coyotes [12 , 27] . Interspecies hybridization is not uncommon among Leopardus species [19] , but we observed no examples of variants that were shared between species . Our results show that at least three independent melanism mutations have occurred during recent evolution of this genus , and have remained polymorphic within each species . Molecular genetic evidence for an adaptive role is strongest for the pampas cat , in which the causative ASIP variant lies on a single major haplotype that extends for hundreds of kilobases as part of a selective sweep . However , absence of long haplotype patterns that clearly distinguish ancestral and melanism-bearing chromosomes in the Geoffroy’s cat and kodkod does not exclude the possibility that melanism is adaptive in these species . The Geoffroy’s cat in particular exhibits patterns of variation that deviate from neutral expectation , with all melanism-bearing chromosomes carrying three non-synonymous mutations in MC1R-coding sequence , and reduced levels of nucleotide diversity in a region ~50 kb upstream of MC1R . Potential selection for melanism in the Geoffroy’s cat may have occurred earlier than in the pampas cat; alternatively , or in addition , a potentially higher recombination rate at MC1R than at ASIP may have diminished the size of an extended haplotype in the Geoffroy’s cat . Studies of additional Geoffroy’s cat populations may help reveal whether the unusual haplotype pattern associated with melanism reflects a specialized population history in Southern Brazil or a more general aspect of the derivative chromosome . In the kodkod , haplotype structure is similar between melanism-bearing and ancestral chromosomes , but the patterns of nucleotide diversity are unusual . Furthermore , within the fosmid that carries the causative variant ( AsipD ) , multiple SNPs sort the melanism-bearing and ancestral chromosomes into two different lineages . A recent population bottleneck associated with colonization of Chiloé Island could account for the latter observation , but does not easily explain why ancestral chromosomes exhibit a local reduction of nucleotide diversity anchored by the ASIP gene . One intriguing possibility is frequency-dependent balancing selection , such that both melanistic and non-melanistic phenotypes have adaptive value , but only at low frequencies . Larger sample sizes and evaluation of multiple loci would provide additional insight on this hypothesis . The kodkod is also striking for its very low nucleotide diversity ( 0 . 018% ) , among the lowest described among all mammals [37] , which is consistent with assessments from ecological studies indicating that the restricted geographic range of this species coupled with deforestation of Central-southern Chile has reduced its effective population size [14–17 , 38] . Although it can be challenging to distinguish if loss of genetic diversity is a cause or consequence in any natural population , it is directly related to population reduction . Melanism is a quintessential example of natural selection in many animals; mechanisms that underlie adaptation can be difficult to discern in larger mammals , although recent work from Allen et al . [9] provides evidence for disruptive selection . In felids , crypsis , presumably as an aid to predation , is thought to drive variation in color patterns , but variation in base color is often postulated to facilitate temperature regulation or response to UV radiation . The latter two mechanisms are oft-cited explanations for Gloger’s rule ( correlation between prevalence of darker coloration and equatorial proximity ) , but are unlikely to explain persistent polymorphism of melanism within Leopardus spp . Our results point to an important role for natural selection in the evolution of melanism for at least one of the analyzed species , but reveal very different genomic signals that likely reflect unique demographic histories , influenced by different selective processes in addition to drift and recombination . Application of similar approaches to other species will extend our knowledge of how natural selection has shaped color variation , and offer new avenues to investigate the evolutionary history of phenotypic diversity .
We studied a total of 57 individuals ( 18 pampas cat , 16 kodkod , and 23 Geoffroy’s cat ) whose origin and phenotype are listed in Table S1 in S1 Data . Biological material was either blood ( collected from wild-caught individuals in the context of field ecology studies ) or tissue ( from road-killed animals encountered during routine wildlife surveys ) ; in all cases , sampling was performed following appropriate national regulations for handling animals and biological materials . DNA was prepared by extraction with phenol/chloroform , and assessed for concentration and quality by fluorometry and agarose gel electrophoresis , respectively . When this work was initiated , we made use of an annotated but incomplete ( 1 . 9x ) genome assembly of the domestic cat , felCat3 [39] . During the course of the work , two additional assemblies became available , and all coordinates used in the manuscript now refer to felCat5/Felis_catus-6 . 2 , http://genome . wustl . edu/genomes/detail/felis-catus/ . After PCR amplification , protein-coding exons of ASIP [8] and MC1R ( Table S2 in S1 Data ) were analyzed by automated capillary sequencing . We first examined 8 individuals of each species ( 4 of each phenotype ) , then extended those results to all available samples ( Supplemental Tables S1 , S4 , S5 , S6 ) . Sequencing electropherograms were verified and corrected with Sequencher 4 . 2 ( GeneCodes Corporation ) , and every potential variant was carefully inspected for confirmation . Homologous nucleotide and amino acid sequences of each gene from additional mammalian species were obtained from GenBank ( Table S3 in S1 Data ) and aligned using ClustalW . DNA libraries for Illumina sequencing were generated using standard protocols . Briefly , 700ng to 2 . 5ug of input DNA ( depending on the level of DNA degradation assessed by agarose gel electrophoresis before shearing ) was sheared to a size range of 100–500 bp , and custom inline barcodes were added during adapter ligation for paired-end sequencing . For target selection , we utilized a set of fosmid libraries that had been previously end-sequenced and mapped to a 3x draft assembly , V17e/felCat4 [40]; identity of all fosmids was verified by PCR of each end . A detailed protocol for probe preparation , hybridization , and capture is described elsewhere ( Day et al . , manuscript submitted ) . Briefly , fosmids were pooled according to their inferred individual mass , a total of 1 . 5ug of input was sheared to 100–500 bp , and DNA fragments were used to prepare biotinylated RNA with a Megascript T7 kit ( Ambion ) . After DNAse treatment and removal of unincorporated nucleotides , hybridizations were carried out in a volume of 26ul that contained 300 ng of probe and 500ng of library DNA ( 125ng each of 4 inline barcoded libraries ) . Selected DNA was recovered with magnetic beads and amplified by PCR ( 20 cycles using standard Illumina primers ) . Library 4-plex pools were sequenced as 12-plex library sets on a single Illumina HiSeq 2000 lane . We used ASIP and MC1R regions from chromosomes A3 and E2 , respectively , as reference sequences , extracted from the Felis_catus-6 . 2 assembly . The genomic reference was masked for transposable elements and low complexity regions with RepeatMasker [41] . To minimize genotyping errors related to inter-species differences in allele structure and distribution , we first mapped all raw sequence reads against the domestic cat reference using the Burrows-Wheeler Aligner ( BWA ) with default parameters [42] , then used SAMtools [43] to compute consensus sequences for these preliminary alignments , and to de novo assemble targeted regions in each of the three species separately . Sequence reads were then remapped to these de novo species-specific consensus sequences and subsequently used to for variant calling according to GATK best practices guidelines [44] . Briefly , we started by applying alignment quality control procedures available in GATK to detect sequence intervals with low quality mappings ( i . e . possibly related to the presence of sequence variants , such as small insertions or deletions , in subsets of the analyzed samples ) . In all such cases , a thorough local realignment of the reads was performed to minimize the number of mismatching bases . The GATK Unified Genotyper ( UG ) tool was applied on realigned reads to infer the genotype structure simultaneously across all samples for each species . The raw genotype calls were subjected to a filtering procedure by imposing thresholds on a set of quality criteria , including minor allele frequency MAF>10% , Phred-scaled mapping quality MQ>40 , UG quality by depth QD>2 and UG HaplotypeScore>13 . The filtered calls were further restricted to a small subset of high quality calls , to satisfy an average variant density threshold of ~1 per 1kb of target sequence . Finally , the BEAGLE genetic analysis software package [45] was used to check genotype consistency across all samples of each species and infer the haplotype phase of selected variants . Curated genotype and haplotype data was then used for downstream analyses , including nucleotide diversity and polymorphism statistics with DnaSP 5 . 10 [46] , EHH [47] , and generation of median-joining haplotype networks ( Network 4 . 5 . 0 . 0; http://www . fluxus-engineering . com ) . Haplotype bifurcation plots in Fig . 5 were generated with Sweep 1 . 1 ( http://www . broadinstitute . org/mpg/sweep/index . html ) and rehh 1 . 0 [48] . DNA sequences reported in this manuscript are publicly available from DRYAD ( doi:10 . 5061/dryad . pq482 ) . | Color polymorphism in closely related animal species provides an opportunity to study how the balance between natural selection and genetic drift shapes the evolution of appearance and form . The cat family , Felidae , is especially interesting; 13 of 37 extant species exhibit polymorphism for melanism , but evidence for any adaptive role is lacking , in part because the potential benefits of melanism to felid predators are not clear , and in part because the tools for genomic analysis of natural populations are limited . We identify the mutations responsible for melanism in three closely related South American wild felids , the pampas cat , the kodkod , and Geoffroy’s cat , then adapt a new approach for targeted genome sequencing to characterize molecular variation in the region surrounding each melanism mutation . We find that each mutation has developed independently , with strong evidence for natural selection in the black pampas cat , and reduced genetic variation in the entire population of kodkods . Our results demonstrate that some “black cats” are black not by chance , but by selection for a mutation that provides increased fitness . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Recurrent Evolution of Melanism in South American Felids |
Taenia solium , a pork-borne parasitic zoonosis , is the cause of taeniasis and cysticercosis in humans . In Vietnam , poor sanitation , the practice of outdoor defecation and consumption of raw/undercooked pork have been associated with infection/exposure to T . solium in both humans and pigs . The broad-scale geographic distribution of the prevalence of T . solium varies throughout the country with infection restricted to isolated foci in the north and a more sporadic geographic distribution in the Central Highlands and the south . While cross-sectional studies have allowed the broad-scale geographic distribution of T . solium to be described , details of the geographic distribution of T . solium at finer spatial scales have not been described in detail . This study provides a descriptive spatial analysis of T . solium exposure in humans and pigs and T . solium taeniasis in humans within individual households in village communities of Dak Lak in the Central Highlands of Vietnam . We used Ripley’s K-function to describe spatial dependence in T . solium exposure positive and negative human and pig households and T . solium taeniasis exposure positive and negative households in villages within the districts of Buon Don , Krong Nang and M’Drak of Dak Lak province in the Central Highlands of Vietnam . The prevalence of exposure to T . solium in pigs in Dak Lak province was 9 ( 95% CI 5 to 17 ) cases per 1000 pigs at risk . The prevalence of exposure to the parasite in humans was somewhat higher at 5 ( 95% CI 3 to 8 ) cases per 100 individuals at risk . Spatial aggregations of T . solium exposure-positive pig and human households occurred in some , but not all of the villages in the three study districts . Human exposure-positive households were found to be aggregated within a distance of 200 to 300 m in villages in Krong Nang district compared with distances of up to 1500 m for pig exposure-positive households in villages in M’Drak district . Although this study demonstrated the aggregation of households in which either T . solium exposure- or taeniasis-positive individuals were present , we were unable to identify an association between the two due to the very low number of T . solium taeniasis-positive households . Spatial aggregations of T . solium exposure-positive pig and human households occurred in some , but not all of the villages in the three study districts . We were unable to definitively identify reasons for these findings but speculate that they were due to a combination of demographic , anthropological and micro-environmental factors . To more definitively identify characteristics that increase cysticercosis risk we propose that cross-sectional studies similar in design to that described in this paper should be applied in other provinces of Vietnam .
Taenia solium is a pork-borne zoonosis of major public health and economic importance . The parasite causes cysticercosis/neurocysticercosis in humans and pigs in many low-income communities in Latin America , Africa and Asia [1] . Poor sanitation , allowing pigs to free roam , the practice of outdoor defecation and consumption of raw/undercooked pork are risk factors for T . solium infection . Cysticercosis infection in humans and pigs occurs due to the accidental ingestion of T . solium eggs shed through the feces of humans with T . solium taeniasis ( tapeworm carriers ) . Taeniasis occurs when humans consume raw/undercooked pork with T . solium cysticerci . T . solium infection results in not only an economic burden in low-income communities due to loss of productivity in affected individuals and the cost of treatment , but also losses arising from the condemnation of pig carcasses destined for human consumption . It was estimated that approximately US $185 million and 2 . 1 million disability-adjusted life years ( DALYs ) were lost in north India due to human cysticercosis in 2011 [2] . It was estimated that the number of DALYs for human cysticercosis in Mozambique in 2007 [3] was 6 . 0 per thousand person-years , 0 . 2 in 2011 in Mexico [4] and 0 . 7 in 2012 in Tanzania [5] . The pork industry in four provinces of Laos estimated losses of between US $55 , 000 to 96 , 000 arising from 20% of carcasses identified as cysticerci-positive over a 21 month period [6] . In Latin America , the number of neurocysticercosis infections has been estimated to be between 11 and 29 million with 1 . 3 million individuals suffering from neurocysticercosis-related epilepsy [7] . Globally , cysticercosis was estimated to be the cause of over 28 , 000 deaths in 2010 [8] . Although risk factors for T . solium taeniasis and cysticercosis include outdoor defecation , the consumption of raw/undercooked pork and allowing pigs to free roam , transmission patterns and the prevalence of the parasite can vary considerably and may prove inconsistent within and/or between regions and communities [9] thus impeding efforts to achieve parasite eradication . Because of the negative impact of T . solium on human health and the economy [10] , controlling the disease is a priority . Understanding the spatial distribution of T . solium is one important step towards development of effective control strategies . Studies in communities of Latin America and Africa , where T . solium infection is hyperendemic , identified clustering of T . solium cysticercosis at both the household and community level [11–13] and a strong geographical association between T . solium carriers and cysticercosis in humans and pigs [14 , 15] . Spatial analyses were used to define an appropriate radius for a control area in a community where T . solium was hyperendemic in Peru . Targeting interventions within this control area was effective in reducing the number of T . solium carriers and the sero-incidence of porcine cysticercosis [16] . T . solium is endemic in Vietnam . A systematic review of cross-sectional studies carried out in Vietnam between 1999 to 2011 showed that the prevalence ranged from 0 to 130 T . solium positive individuals per 1000 individuals at risk [17] . The distribution of T . solium in Vietnam is characterized by hotspots or foci of infection in communities in the northern provinces , including Phu Tho and Bac Ninh [18 , 19] . In the Central Highlands and the south of the country the distribution of T . solium is sporadic [20] . While previous studies [15 , 16 , 17] have provided useful information at the regional and provincial level , we are aware of no investigations that have investigated the distribution of T . solium at finer spatial scales . With this background , the aim of this study was to describe the spatial distribution of households in which one or more individual humans or pigs were T . solium exposure positive and households in which one or more individual ( humans ) were T . solium taeniasis positive . Quantitative estimates of the prevalence and geographic distribution of T . solium exposure and T . solium taeniasis positive households will provide evidence to allow public health authorities to decide between treatment programs applied at the whole community level as opposed to treatment programs applied at either the individual household or small area level .
This study was reviewed and approved by the Behavioral and Social Sciences Human Ethics Sub-committee , the University of Melbourne ( reference number 1443512 ) and the Animal Ethics and Scientific Committee , Tay Nguyen University ( reference number 50 . KCNTY ) . This study was conducted under the supervision of the local Center for Public Health and the local Center for Animal Health , Dak Lak , Vietnam . This research on pigs was based on the International Guiding Principle for Biomedical Research Involving Animals issued by the Council for the International Organization of Medical Sciences . The cross-sectional study was carried out between May and October 2015 in Dak Lak province in the Central Highlands of Vietnam . Dak Lak is comprised of 15 districts with approximately 70% of the total population of 1 . 8 million people living in rural areas [21] . Within the province , three districts namely Buon Don , Krong Nang and M’Drak were chosen as the study sites based on their diverse geographic characteristics ( Fig 1 ) . The characteristics of these districts have been described in detail elsewhere [22] . A sampling frame listing the name of all villages in Dak Lak province was obtained from Sub-Department of Animal Health office within the Ministry of Agriculture and Rural Development . Villages eligible for sampling comprised those with more than 1000 pigs , as recorded by the Sub-Department of Animal Health . All eligible villages within each district were assigned a number and two numbers chosen at random to select villages from each district for inclusion in the study . A list of householder names within each selected village was obtained from each village head person , and householder names were assigned a numeric code . A sheet of paper listing numeric household codes for each village were cut into pieces and placed face-down on a table . The village head person was asked to select 50 households at random for human sampling and between 100 and 140 households for pig sampling . All households selected for human sampling and pig sampling were visited several days before the proposed sampling date to obtain consent from householders to take part in the study . Householders eligible for inclusion in the study were individuals who were healthy , not pregnant and over seven years of age . Householders requested to take part in the study signed a consent form . Those that were under 18 years of age were required to provide written consent as well as written consent from either their parents or legal guardians . At the time of consent each study participant was given a labeled stool container , with instructions that the container would be collected on the date of sampling , several days later . Sampling of households ( Fig 2 ) was carried out in two stages . Humans from each of the 50 consenting study households were sampled between May and October 2015; pigs from each of the 100 to 150 consenting study households were sampled between June and October 2015 . At the time of each household visit , a questionnaire was administered to each of the study participants soliciting details regarding demography , sanitation and hygiene status , food culture and religion , practice of pig management and the longitude and latitude coordinates of the main doorway of entry of the dwelling used for sleeping . On the date of sampling , consenting householders were visited by staff from the Sub-Department of Health and 5 mL of venous blood collected into plain clotting tubes from consenting study participants . Stool samples were fixed in 5% potassium dichromate ( w/v ) for molecular analysis . Pigs that were pregnant , ill or aged less than 2 months of age were excluded from sampling . Approximately , 10 mL of blood was obtained from the cranial vena cava of each pig into plain blood collection tubes . Blood samples were allowed to clot at ambient temperature prior to centrifugation at 3200 × g for 5 minutes to collect serum . Serum was dispensed into 1 . 5 mL aliquots and stored at -20°C until analysis . The longitude and latitude of the main doorway of entry of sampled households was recorded using a handheld global positioning system ( GPS ) device ( Garmin GPSMAP64 , Taiwan ) . Human stool samples were tested to determine T . solium tapeworm carriers using a real-time PCR ( T3qPCR ) described by Ng-Nguyen et al . ( 2017 ) [22] . The assay has been reported to have a diagnostic sensitivity of 94% and a diagnostic specificity of 98% . Human serum samples were tested for the presence of antibody against T . solium cysticerci using a lentil-lectin purified glycoprotein-enzyme-linked immunoelectrotransfer blot ( LLGP-EITB ) . This assay has a diagnostic sensitivity of 98% and a diagnostic specificity of 100% . The LLGP-EITB was performed using the methodology described by Tsang et al . ( 1989 ) [23] . Pig serum samples were tested for the presence antibody against T . solium cysticerci using rT24H antigen in the enzyme-linked immunoelectrotransfer blot ( EITB ) format . The rT24H-EITB assay showed no cross-reaction to T . hydatigena and had a diagnostic sensitivity and specificity of 100% when tested on 29 cysticercosis-negative USA pig sera , 12 necropsy-positive T . solium-positive Peruvian pig sera and four T . hydatigena necropsy-positive Vietnamese pig sera [24] . The performance of the rT24H-EITB assay was carried out using the methodology described by Noh et al . 2014 [25] . Positive samples resulting from the rT24H-EITB assay were confirmed using the LLGP-EITB assay . Details collected during each of the household visits and laboratory test results were stored in a relational database ( Microsoft Access 2007 , Microsoft Corporation , Redmond , USA ) . Longitude and latitude coordinates of household locations ( recorded in degrees , minutes and seconds ) were converted to decimal degrees and re-projected to the Universal Transverse Mercator Zone 48N projection using the World Geodetic System 1984 datum . Analyses were carried out to describe the spatial characteristics of: ( 1 ) T . solium exposure-positive and T . solium exposure-negative households for humans; ( 2 ) T . solium exposure-positive and T . solium exposure-negative households for pigs; and ( 3 ) human or pig T . solium or T . solium taeniasis-positive and negative households . Our classification of T . solium exposure-positive households for humans and pigs was based on the LLGP-EITB and rT24H-EITB assays , respectively . Classification of Taenia-positive households was based on parallel interpretation of the test results of the LLGP-EITB , rT24H-EITB and T3qPCR assays ( positive for either T . solium exposure or T . solium taeniasis ) . Ripley’s K-function [26] provides a summary measure of spatial dependence among point locations as a function of their Euclidean distance . The K-function is defined as the expected number of points that are located within a distance h of an arbitrarily selected point location , divided by the overall density of points [27] . Where there is spatial dependence in a point pattern , point events are likely to be surrounded by other point events and , for small vales of distance h , K ( h ) will be relatively large . Conversely , if point events are regularly spaced , each point is likely to be surrounded by empty space and , for small values of distance h , K ( h ) will be small . To facilitate inference , we developed separate K-function plots for T . solium exposure-positive and T . solium exposure-negative households . For each value of h we then calculated the K-function difference as D ( h ) = K ( h ) positive – K ( h ) negative . If exposure-positive households were spatially aggregated , over and above that of the exposure-negative households , then D ( h ) will appear graphically as peaks ( or troughs ) as a function of distance . Three sets of K-function analyses were carried out for: ( 1 ) human T . solium exposure-positive households and human T . solium exposure-negative households; ( 2 ) pig T . solium exposure-positive households and pig T . solium exposure-negative households; and ( 3 ) T . solium taeniasis exposure positive and T . solium taeniasis negative households . Monte Carlo simulation was used to construct critical envelopes for each K-function difference plot . Here , we randomly assigned the observed number of positive households across the population of study household locations and re-computed D ( h ) each time . The critical envelopes are based on 1000 Monte Carlo simulations of the data . Departures of the observed value of D ( h ) above the limits of the upper and lower critical envelopes provided an indication of spatial aggregation of exposure-positive households beyond that which would be expected by chance , and at what spatial scale .
The total numbers of households visited for collecting samples from humans and pigs , respectively , were 190 and 408 in Buon Don , Krong Nang and M’Drak districts . Within the 190 households , a total of 342 individuals consented to participate in the study . The number of pigs sampled from the 408 households was 1281 . Four of the 190 households ( 2 . 1% , 95% CI 0 . 6 to 5 . 6 ) housed T . solium tapeworm carriers and the percentages of households housing individuals and pigs that were T . solium exposure positive was 8 . 9% ( 17/190 , 95% CI 5 . 5 to 14 ) and 2 . 7% ( 11/408 , 95% CI 1 . 4 to 4 . 9 ) , respectively . Amongst the 11 T . solium exposure-positive households for pigs , there was one household that had more than one pig antibody-positive to T . solium cysticerci; all other exposure-positive households had a single pig that was seropositive . All T . solium exposure-positive households for humans had a single individual that was antibody-positive . Of the 561 households that were visited for either human or pig sampling , 31 had either humans or pigs that were either human T . solium exposure-positive , pig T . solium exposure-positive or T . solium taeniasis positive ( Table 1 ) . There were 29 households with single infections; one household with exposure-positive pigs and one household had an individual infected with T . solium taeniasis and an individual that was T . solium exposure-positive . Human T . solium exposure- and taeniasis-positive households were present in all three districts . There was no pig T . solium exposure-positive households in Buon Don . In three study districts , the prevalence of having a household latrine was relatively low ranging from 13% to 47% . This meant that outdoor defecation was a practice reported by between 15% and 74% of the study population . Our data showed that allowing pigs to free roam was common practice in Buon Bon , Krong Nang , and M’Drak . The percentage of pigs that consumed human feces was high in Buon Don and M’Drak ( Table 2 ) . Of 11 exposure-positive households for pigs , there were nine households in M’Drak . Pigs kept in seven of the nine households were allowed to roam freely within the village . No exposed pigs were detected in Buon Don ( Fig 3C and 3D ) and two exposed pigs were detected in Krong Nang ( Fig 4C and 4D ) . In M’Drak , the four exposed pigs that were detected were within a distance of 250 m of each other ( Fig 5C ) and pairs of exposure-positive pig households were less than 100 m apart ( Fig 5D ) . The K-function difference plot for T . solium exposure in pigs in M’Drak shows K ( h ) positive in excess of K ( h ) negative up to a distance of 1500 m ( Fig 6F ) . There were small numbers of human T . solium exposure-positive households in close proximity in Krong Nang ( Fig 4B ) . The K-function difference plot for exposure to T . solium in humans in Krong Nang supported this observation , where K ( h ) positive was in excess of K ( h ) negative up to a distance of 200 to 300 m ( Fig 6C ) . The spatial distribution of human exposure-positive households in Buon Don ( Fig 3A and 3B ) and M’Drak ( Fig 5A and 5B ) were more regularly distributed; there was no evidence of significant differences between K ( h ) positive and K ( h ) negative up to a distance of 3000 m ( Fig 6A and 6E ) . When we considered households that were either human T . solium exposure , taeniasis or pig T . solium exposure as a single group , the K-function difference plot showed all T . solium exposure-positive and taeniasis-positive households were aggregated up to a distance of 1000 m in M’Drak ( Fig 6g ) . Similar associations were evident in Buon Bon and Krong Nang but the observed K-function difference plot did not exceed the simulation envelope limits at any distance ( Fig 6B and 6D ) . On inspection , however , we observed a group of taeniasis- and exposure-positive households in close proximity to each other in the village of Cu Mta in M’Drak district ( Fig 7 ) .
This study describes the fine scale spatial distribution of T . solium exposure in pigs and humans in Vietnam for the first time . The geographic distribution of T . solium exposure- and taeniasis-positive households varied markedly across the districts of Buon Don , Krong Nang and M’Drak of Dak Lak province . A prominent feature of this data is that the prevalence of T . solium exposure in both species and T . solium taeniasis was relatively low ( in humans 9 exposure- and 2 taeniasis-positive households per 100 households at risk; in pigs 3 exposure-positive households per 100 households at risk ) making it difficult to definitively identify characteristics of the spatial distribution of positive households that are likely to exist across all districts of Dak Lak , and indeed all districts of Vietnam . Spatial aggregations of T . solium exposure-positive households for humans occurred in some ( the village of Ea Wer in Buon Don district ( Fig 3B ) , Dlieya in Krong Nang district ( Fig 4B ) and Krong Jing and Cu Mta in M’Drak ( Fig 5A and 5B ) ) , but not all , of the villages in the three study districts . Our K-function difference plots showed that T . solium exposure-positive households for humans showed the same pattern of spatial dependence as T . solium exposure-negative households in Buon Bon and M’Drak ( Fig 6A and 6E ) . In Krong Nang , compared with human exposure-negative households , human exposure-positive households were aggregated up to a distance of 200 to 300 m ( Fig 6C ) . We speculate that if the prevalence of exposure was higher in Buon Don and M’Drak and sufficient resources were available to allow larger sample sizes in each of the two districts to be collected , a similar pattern of spatial dependence would be evident . Although spatial aggregation of T . solium exposure-positive households for humans in Krong Nang was beyond that expected by chance ( and entirely due to a collection of five positive households in the village of Dlieya ) , its overall magnitude was relatively small ( Fig 6C ) . Dlieya is small village comprised of less than 200 households in a remote area of Krong Nang . In Dlieya the number of individuals per household was considerably larger than that of the other villages in the study ( median 5; minimum 2 to maximum 11 ) and a notable feature was that it was common for several generations of a family to live together in close proximity , and a highly prevalent custom was that food was shared with neighbours and relatives on a daily basis , often associated with community ceremonies ( e . g . weddings and anniversary of deaths ) . In the district of Krong Nang , houses are typically surrounded by a large garden comprised of coffee or pepper trees . We hypothesize that in this district , where a high proportion of the study population were known to defaecate outdoors , individuals were more likely to defecate in their own garden ( as opposed to communal areas ) which means that it was more likely for T . solium eggs to be present in close spatial proximity to a given household where exposure-positive individuals were present . We speculate that the anthropological and fine-scale environmental characteristics of Dlieya were sufficient to allow spatial clustering of human exposure infection to be detected even in the presence of a modest sampling effort ( n = 30 households ) . Spatial aggregations of exposure to T . solium in pigs occurred but this was infrequent . In three study districts , there was a single aggregation of pig exposure-positive households in M’Drak ( Fig 5C ) . Our K-function difference plot for M’Drak ( Fig 6F ) showed pig exposure-positive households were clustered within a distance of 1500 m . Presumably , this was due to the larger range over which free-roaming pigs forage . Copado et al . , 2004 [26] reported that free-roaming pigs travel daily within a distance ranging from 1000 to 3000 m . In a 12 hour period , pigs traveled a distance of up to 4000 m and spent , on average , 47% of their time outside of their homestead [28] . Our findings are supported by those of Ngowi et al . ( 2010 ) [12] who conducted a cross-sectional study of cysticercosis in 784 pig-owning households in northern Tanzania . In the study of Ngowi et al . ( 2010 ) it was shown that porcine cysticercosis was clustered within the distance of 600 m and 10 km . Morales et al . ( 2008 ) [29] conducted a cross-sectional study of 562 pigs in the state of Morelos in Mexico in 2003 . In this study the prevalence of porcine cysticercosis was relatively high ( 13%; 95% CI 11 to 17 ) and while free-roaming pigs had a greater risk of being cysticercosis-positive , no geographical clustering of positivity was found . Spatial clustering of T . solium exposure in pigs in M’Drak could have been associated with the age of the resident pig population , the absence of pigsties and the regular habit of coprophagy amongst pigs . In total , there were 11 T . solium pig exposure-positive households in Dak Lak . Nine of these 11 positive households were in M’Drak , aggregated in groups of two to four households ( Fig 5C and 5D ) . Of the nine T . solium pig exposure-positive households in M’Drak , in seven households pigs were allowed to roam freely within the village , increasing the chance of exposure to T . solium eggs . The terrain in M’Drak is generally flat . The quality of the soil is poor supporting predominantly natural grasslands . For these reasons , the practice of allowing pigs to free roam is more common compared with the two other districts . M’Drak had a high proportion of pigs that were not confined ( 17% [211 of 1281] , 95% CI 15 to 19 ) . Of the total number of pigs sampled in this study , 511 ( 40% ) were from M’Drak . Of the 511 M’Drak pigs that were sampled , it was reported that 193 ( 38% ) consumed human feces and 211 ( 41% ) regularly scavenged for food ( Table 2 ) . When we considered households that were human and/or pig T . solium exposure-positive or taeniasis positive as a single group , our K-function difference plot showed these positive households were aggregated up to a distance of 1000 m in M’Drak ( Fig 6G ) , but not in Buon Bon ( Fig 6B ) and Krong Nang ( Fig 6D ) . O’Nea at al . ( 2012 ) and Pray et al . ( 2017 ) showed that human and/or porcine cysticercosis cases were strongly associated with the presence of tapeworm carriers [14 , 15] . Individuals and pigs living in close proximity to tapeworm carriers are more likely to be infected with T . solium cysticercosis [11 , 15 , 30] . Given this unique data set , with contemporary sampling of humans and pigs , it was of interest to us to determine if there was a spatial dependence between T . solium exposure- and T . solium taeniasis-positive households . Although households that had either T . solium exposure- or taeniasis-positive cases were spatially aggregated , we were unable to identify an association between the two because of the extremely low number ( n = 4 ) of taeniasis-positive households across the three study districts . On inspection we observed a group of taeniasis- and exposure-positive households in close proximity to each other in the village of Cu Mta in M’Drak district ( Fig 7 ) . Madinga et al . ( 2017 ) [26] and Morales et al . ( 2008 ) [29] indicated that there was no spatial correlation of T . solium exposure in pigs and T . solium taeniasis . Spatial aggregations of human and pig exposure-positive households occurred in some , but not all , of the villages in the three study districts . We can only speculate about the reasons for this pattern , as discussed above . Since cysticercosis occurs throughout Vietnam [17] it is likely that foci of infection are present in other areas . The relatively low prevalence of exposure to T . solium indicates that massive deworming programs in the communities of Dak Lak province are , for the most part , unnecessary . Instead , we recommend that if a human is identified as T . solium positive then either: ( a ) individuals resident in the immediate area should be tested to rule out the presence of an exposure or infection cluster; or ( b ) anthelmintic treatment is offered to individuals resident within a 2000 m radius of the identified case . With respect to the second approach , privacy issues would need to be handled appropriately , particularly in small communities . A limitation of this study was that for logistic reasons sampling of humans and pigs were carried out independently resulting in a lack of overlap of the locations of households where humans and pigs were sampled ( see , for example , Ea Nuol in the district of Buon Don , Fig 3A and 3C ) . While this limited our ability to identify an association ( if any ) between human and pig T . solium exposure-positive households , assessment of the spatial dependence of exposure status by species ( human , pigs ) was possible . Although households that had either T . solium exposure- or taeniasis-positive cases were spatially aggregated , we were unable to quantify their spatial association due to the extremely low number of T . solium taeniasis-positive households . | Taenia solium is a pork-bone zoonotic parasite . Humans acquire taeniasis from consumption of raw/undercooked pork contaminated with T . solium cysticerci . Pigs and humans acquire cysticercosis following consumption of food contaminated with eggs shed from the feces of humans with T . solium taeniasis . In Vietnam , the geographic distribution of T . solium varies throughout the country with hotspots or foci of infection in communities in the North and a more sporadic distribution in the Central Highlands and the South . While information on the distribution at the regional and provincial level is available , there is no available information on the spatial distribution of T . solium at fine spatial scales and factors influencing its distribution at fine spatial scales have not been described in detail . In this cross-sectional study , we collected information on the geographic coordinates of study households and utilized spatial analytical techniques to quantify both the fine scale spatial pattern of exposure to T . solium as well as the tendency for T . solium exposure-positive households to be located close to other T . solium exposure-positive households ( spatial autocorrelation ) in three districts in Dak Lak province . We found that in some of the study villages T . solium exposure-positive households were more likely to be surrounded by other T . solium exposure-positive households . Human exposure-positive households were found to be aggregated within a distance of 200 to 300 m in villages in Krong Nang district; whilst spatial aggregation of pig exposure-positive households was found up to distances of 1500 m in villages in M’Drak district . Although households that had either T . solium exposure- or taeniasis-positive cases were aggregated , we were unable to quantify their spatial association due to the extremely low number of T . solium taeniasis-positive households . This study shows that in the Central Highlands of Vietnam , T . solium exposure tend to cluster within foci . This information can be used to inform community intervention programs to lower its incidence in both humans and pigs . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"cross-sectional",
"studies",
"population",
"dynamics",
"tropical",
"diseases",
"geographical",
"locations",
"vertebrates",
"diet",
"parasitic",
"diseases",
"animals",
"mammals",
"animal",
"products",
"research",
"design",
"nutriti... | 2018 | Spatial distribution of Taenia solium exposure in humans and pigs in the Central Highlands of Vietnam |
The intestinal epithelium serves critical physiologic functions that are shared among all vertebrates . However , it is unknown how the transcriptional regulatory mechanisms underlying these functions have changed over the course of vertebrate evolution . We generated genome-wide mRNA and accessible chromatin data from adult intestinal epithelial cells ( IECs ) in zebrafish , stickleback , mouse , and human species to determine if conserved IEC functions are achieved through common transcriptional regulation . We found evidence for substantial common regulation and conservation of gene expression regionally along the length of the intestine from fish to mammals and identified a core set of genes comprising a vertebrate IEC signature . We also identified transcriptional start sites and other putative regulatory regions that are differentially accessible in IECs in all 4 species . Although these sites rarely showed sequence conservation from fish to mammals , surprisingly , they drove highly conserved IEC expression in a zebrafish reporter assay . Common putative transcription factor binding sites ( TFBS ) found at these sites in multiple species indicate that sequence conservation alone is insufficient to identify much of the functionally conserved IEC regulatory information . Among the rare , highly sequence-conserved , IEC-specific regulatory regions , we discovered an ancient enhancer upstream from her6/HES1 that is active in a distinct population of Notch-positive cells in the intestinal epithelium . Together , these results show how combining accessible chromatin and mRNA datasets with TFBS prediction and in vivo reporter assays can reveal tissue-specific regulatory information conserved across 420 million years of vertebrate evolution . We define an IEC transcriptional regulatory network that is shared between fish and mammals and establish an experimental platform for studying how evolutionarily distilled regulatory information commonly controls IEC development and physiology .
Epithelial cells lining the intestinal tract serve important and evolutionarily conserved functions in animal physiology . The intestinal epithelium is the primary site for absorption and metabolism of diverse dietary nutrients and xenobiotics , relays metabolic and immunological signals to the rest of the body , and provides a critical barrier to microorganisms that reside within the intestinal lumen [1] . Dysfunction in the development and physiology of intestinal epithelial cells ( IECs ) has been implicated in a growing number of human diseases , such as inflammatory bowel diseases [1] , colorectal cancer [2] , food allergy [3] , obesity [4 , 5] , malnutrition [6] , and infectious diarrheas [7] . These insights have fueled considerable interest in the molecular and cellular mechanisms underlying IEC biology . Due to the common evolutionary origins of the animal intestine , animal models are invaluable tools in understanding the intestinal epithelium , including its normal development and dysfunction . The appearance of a “through gut” with a distinct mouth , anus , and intermediate regions was an early step in bilaterian animal evolution [1] . It is thought that many of the anatomic and physiologic features of the intestine are conserved between bilaterian lineages , with mammals ( members of Sarcopterygii ) and bony fishes ( members of Actinopterygii ) last sharing a common ancestor approximately 420 million years ago [8] . Although lineages within these vertebrate taxa have evolved specific adaptations in their intestinal anatomy and physiology , fundamental aspects appear to be conserved [9] . For example , the intestinal epithelium in mammals and fishes comprises functionally similar IEC subtypes , including absorptive enterocytes and secretory cells such as goblet cells and enteroendocrine cells . These differentiated cells are rapidly renewed through the action of IEC stem or progenitor cells residing at the base of villi or rugae [10 , 11] . Another prominent conserved feature of the vertebrate intestine is anatomic and physiologic specialization along the anteroposterior axis . In mammals , the gut is generally composed of a small intestine , which includes a duodenum in which chemical digestion occurs , a jejunum in which the majority of nutrients are absorbed , an ileum that specifically absorbs bile salts and vitamin B12 , and a colon or large intestine in which absorption of water and salts occurs . Though the intestinal tract of zebrafish and other fishes display anteroposterior regional specialization , the evolutionary relationship with mammalian intestinal regions has remained unclear . The zebrafish intestine was originally described to consist of 3 histologically defined segments: ( 1 ) anterior or rostral intestine , also known as the intestinal bulb or segment I; ( 2 ) the middle intestine or segment II; and ( 3 ) the posterior or caudal intestine or segment III [12–14] . Though this 3-segment nomenclature has been used to describe the zebrafish intestine from larval to adult stages , the extent to which intestinal segmental programs are maintained across zebrafish life stages remains unresolved . Transcriptomic characterization has shown that the anterior intestine of the adult zebrafish generally expresses genes with similar function to the mammalian small intestine , while the posterior zebrafish intestine corresponds to the mammalian large intestine [15] . However , it is generally unknown which gene sets are expressed in similar anteroposterior patterns across multiple species . Recent studies have begun to identify transcription factors ( TFs ) and regulatory regions that contribute to the identity and function of IECs in individual species [16–19] . For example , CDX2 acts as a master sequence-specific TF that regulates intestinal patterning and epithelial identity in mice and zebrafish [18 , 20 , 21] and controls chromatin access to regulatory regions for other TFs that specify IEC identity in mammals , such as the small intestinal TFs HNF4A and GATA4 [18 , 22–24] . IEC subtype specification is similarly controlled through transcriptional regulatory mechanisms . These include Wnt signaling [25] and a Notch signaling cascade that uses downstream TFs such as RBPJ , ATOH1 , and HES1 , which direct IEC specification into secretory or absorptive lineages [12 , 17] . Still , it remains uncertain which portions of the regulatory framework defining IEC function are conserved , hindering the utility of model organisms to help dissect relevant signaling mechanisms , transcriptional programs , and disease states . Genome-wide accessible chromatin assays can identify cell type and condition-specific cis-regulatory regions . These nucleosome-depleted regulatory regions contain transcription factor binding sites ( TFBS ) that provide a critical insight into the underlying transcriptional networks that define tissue identity . However , recent studies have found accessible regulatory regions largely similar when comparing IEC stem cells and their downstream subtype progenitor intermediates , in spite of differences in gene expression [17] . Similarly , gene expression changes induced in IECs upon colonization with a microbiota were not associated with overt alterations in the accessible chromatin landscape [16] . Together , these findings suggest that aspects of IEC cell plasticity , differentiation , and environmental response are not driven by gross changes in the accessible chromatin landscape , making key regulatory regions difficult to identify . Differential expression or binding of lineage-specific or environmentally responsive TFs [16 , 26] or chromatin modifiers like the histone deacetylase HDAC3 , which plays important roles in how IECs respond to microbes [27] , may partially explain the lack of gross chromatin accessibility changes in certain IEC populations [16 , 17] . However , it remains unclear which regulatory mechanisms serve similar roles in IEC function in different species or if conserved accessible chromatin regions across species can identify important regulatory mechanisms that have not been easily identified within a single species [28] . In this study , we tested the hypothesis that conserved IEC functions are achieved using conserved transcriptional regulatory mechanisms . We profiled the transcriptome and accessible chromatin landscape of IECs from 4 evolutionarily distant vertebrates: zebrafish , stickleback , mouse , and human . We found substantial overlap at a transcriptional level including a common group of IEC signature genes , evidence for common regulation of IEC subtype specification , and unexpected similarity between gene expression along the length of the intestine from fish to mammals . These transcriptional similarities were not easily explained by neighboring conserved regions that were commonly accessible in IECs . However , using accessible chromatin regions and TFBS prediction we were able to recover common IEC-related regulatory information genome wide and at several important representative loci , despite a scarcity of sequence conservation around genes commonly expressed in IECs for over 420 million years .
In order to understand the extent of gene expression similarity across vertebrate intestinal epithelia , we compared newly generated gene expression data from IECs isolated from adult human colon , adult zebrafish intestine , adult stickleback intestine , and from data we previously generated from adult mouse colon and ileum [16] ( Fig 1A ) . We find a strong correlation in gene expression between fish and mammalian IECs throughout the dynamic range of the transcriptome ( Fig 1B and 1C , S1A Fig ) . Using principal component analysis ( PCA ) and hierarchical clustering , we find that the expression of orthologous genes is more similar amongst IECs across species than mouse IECs are to other mouse tissues ( Fig 1B–1E , S1 and S2 Figs ) . Furthermore , unrelated RNA sequencing ( RNA-seq ) data from whole mouse intestine cluster with data from all vertebrate IECs ( Fig 1D and 1E ) [29] . These results reveal that gene expression levels in IECs are similar across these 4 vertebrate species and suggest that many aspects of IEC physiology have been conserved since the common ancestor of mammals and fish . In our PCA of mRNA expression from IECs and mouse tissues , we found that principal component 1 ( PC1 ) separated mammalian and fish IECs from all other tissues . We identified 470 genes whose expression levels highly correlate with PC1 and exhibit high expression in IECs relative to other tissues , though their expression and function is not necessarily exclusive to IECs ( Fig 1D and 1E , S1 Fig , S1 Table , Materials and methods ) . These IEC signature genes are representative of physiologic functions and cell types in the intestinal epithelium and include genes involved in lipid , carbohydrate , and protein metabolism ( Fig 1F and 1G , S2 Table ) . IEC signature genes , including retinol binding protein 2 ( RBP2 , with rbp2a assigned as the zebrafish ortholog by Ensembl ) , fatty acid binding protein 6 ( FABP6 ) , and cadherin 17 ( CDH17 ) , are amongst the most highly expressed genes in IECs from several species ( Fig 1H , S1D Fig ) . In addition , we identified genes indicative of different IEC subtypes within the intestinal epithelium including RBP2 [30] and FABP6 ( enterocyte ) [31] , peptide YY ( PYY; enteroendocrine ) [32] , and polypeptide N-acetylgalactosaminyltransferase 6 ( GALNT6; goblet cells ) [33] , consistent with our RNA-seq data representing heterogeneous populations of IECs ( S1D Fig ) . Ribosomal protein genes and translation components were also highly represented within this signature , consistent with the intestinal epithelium being one of the most highly-proliferative tissues [33] . Amongst the IEC signature genes , we found TFs known to be involved in development and function in the intestine , including the epithelial-specific E26 transformation-specific ( ETS ) TF ELF3 [34] , HNF4A [18] , HNF4G [35] , FXR [36] , GATA5 [37] , and OSR2 [38] , suggesting a conserved basal IEC transcriptional network . Several IEC signature TFs have known associations with human Inflammatory Bowel Diseases ( IBD ) , including SMAD7 [39] , CEBPG [40] , STAT3 [40–42] , XBP1 [43] , HNF4A [44] , ELF3 [41] , IRF1 [45] , and NFκB components IKBKB , IKBKG , and NFKBIZ [46] . Furthermore , the most common human disorders associated with IEC signature genes were obesity-related traits , IBD , and Type 2 Diabetes ( S1 Table ) . Together these data highlight the utility of mouse , zebrafish , and stickleback in modeling human intestinal development and disease and suggest a basal similarity in the transcriptional mechanisms underlying intestinal epithelial homeostasis in vertebrates . The results described above revealed conserved IEC signatures by comparing RNA levels from intestinal IECs to other tissues . However , we speculated that signatures of intestinal identity might be further resolved by comparing gene expression along the intestine’s anteroposterior axis in zebrafish and mice . Using previously published datasets we identified gene orthologs that showed similar expression patterns along the adult zebrafish intestine ( divided into 7 sections of equal length ) [15] and along the adult mouse intestine ( divided into duodenum , jejunum , ileum , and colon ) ( Fig 2 , S3 Fig , Materials and methods ) [47] . Of 493 genes that showed high expression in the anterior of the zebrafish intestine , we found over 70 genes sharing similar high expression in the anterior intestine of mouse [15] , including IEC signature genes like Rbp2 , Aldob , and Ehhadh ( Fig 2A and 2B , S3 Fig ) . Many of these genes ( e . g . , Fabp2 , Acsl5 , Agpat2 , Slc27a4 , and Dgat2 ) are critical in lipid metabolism and uptake , which is consistent with lipid absorption and metabolism taking place mostly in the small intestine in mouse and anterior portion of the intestine in zebrafish [48] . However , ordering genes by mouse duodenum expression level reveals that some of these genes have surprisingly high similarities in cross-species expression patterns along the length of the intestine ( Fig 2A and 2B , S3C Fig ) . For example , adenosine deaminase ( Ada ) , which is most highly expressed in duodenum in mouse [49] , is most highly expressed in the zebrafish sections 1–2 ( Fig 2B , S3C Fig ) . Similarly , Fabp2 and Enpep , which are expressed most highly in the jejunum and ileum in mouse , are most highly expressed in sections 3–5 in zebrafish intestine ( Fig 2C and S3C Fig ) . This suggests an unappreciated similarity between IEC gene expression along the small intestine in mammals ( mouse ) and teleosts ( zebrafish ) and the potential for further examples of subregionalization in the zebrafish intestine that have not been previously described [13–15] . In support of this conserved regional specification , we also found a small group of genes that are expressed highly only in the terminal portion of the zebrafish anterior intestine ( section 5 [15] ) , which occupies a location similar to the mammalian ileum ( Fig 2C and S3D Fig ) . The mammalian ileum is involved in the uptake of bile salts following their use in emulsification of lipids in the anterior small intestine [50] . Two IEC signature genes involved in bile handling , Fabp6 and Slc10a2 , show high expression in this narrow region of zebrafish intestine and mouse ileum ( Fig 2C and S3D Fig ) . In addition , the proteases Lgmn , Scpep1 , and 3 cathepsins are in this cluster and show similar high expression largely in the mouse ileum , suggesting a regionally conserved utilization of lysosomal-cathepsin—mediated degradation ( Fig 2C and S3D Fig ) [51] . These observations suggest that the cellular differentiation and physiological programs deployed in this region of the zebrafish intestine are specialized for bile salt recovery , with strong homology to the mammalian ileum . We also found numerous genes expressed more highly in the posterior end of the mouse and zebrafish intestine , suggesting similar physiologic functions in zebrafish distal intestine and mouse colon ( Fig 2D , S3E and S3F Fig ) [15] . Collectively , these observations suggest that the teleost intestine has the capacity to articulate complex gene differentiation patterns along the length of the intestine that are functionally and spatially analogous to segments in the evolutionarily distant mammalian intestine . To test the hypothesis that conserved regions of accessible chromatin underlie the transcriptional similarities we found in IECs , we profiled accessible chromatin in the same cell preparations that we used to generate our RNA-seq data using Formaldehyde-Assisted Isolation of Regulatory Elements sequencing ( FAIRE-seq ) from stickleback , zebrafish , and human colon IECs . We also used recently published data from our group that profiled mouse ileum and colon IECs using DNase I hypersensitive sites sequencing ( DNase-seq ) [16] . Combining IEC accessible chromatin maps with species-transferable regulatory landmarks such as the transcription start site ( TSS ) and conserved nonexonic elements ( CNEs ) allowed us to profile and define common regulatory information utilized in all species . Accessible chromatin peaks were frequently enriched at orthologous TSS in IECs , including at IEC signature genes such as ELF3 ( Fig 3A , S3 Table and S2 Fig ) . This accessible chromatin signal is consistent with typical genome-wide distributions , though the relationship with accessibility may not be strictly driven by regulatory regions or transcription that is specific to IECs ( Fig 3B and 3C ) . However , the related IEC PC1 correlation values , transcription levels , and presence of IEC signature genes were both higher on average at the TSS with higher accessible chromatin levels , suggesting that the magnitude or presence of accessible chromatin may be conserved at related regulatory regions in IECs in distantly related species ( Fig 3B and 3C , S4A and S4B Fig ) . We compared accessible TSS call overlaps at the TSS for all IEC samples and several cell lines and tissues from mouse and human Encyclopedia of DNA Elements ( ENCODE ) /Roadmap [52–54] to determine if accessible chromatin status could identify IEC-specific regions . We clustered this information to identify common patterns of accessible chromatin status in all these samples ( Fig 3D ) . While the majority of TSS regions appeared to be constitutively accessible at most genes in most species and tissues represented , a group of IEC signature genes had TSSs that were accessible frequently in IECs but less often in other tissues ( Fig 3E ) . Importantly , genes within this group were also almost always accessible at the TSS in several independent datasets of intestinal tissue from mouse and human ( Fig 3E ) . This group included key genes involved in IEC biology such as HNF4A , HNF4G , RBP2 , A1CF , CFTR , and CDH17 . Further , we found 3 genes , FABP6 , SLC10A2 , and TMIGD1 , that showed high similarity of mRNA levels along the length of the intestine at the mouse ileum and zebrafish section 5 ( Fig 2C ) and showed limited accessibility in nonintestinal tissues with accessibility in some of the intestine-related datasets ( Fig 3E ) . To determine if the regions that show chromatin accessibility largely in IECs also showed conservation at a sequence level , we used multiple metrics that measure sequence conservation from teleost to mammals . This included zebrafish conserved nonexonic elements ( zCNEs ) , a stringent comparison of noncoding regions of 14 species , including human and mouse [55] . We also used UCSC liftOver and zebrafish-to-human and zebrafish-to-mouse multiple alignment format ( MAF ) blocks to identify potentially DNA-conserved regions . Though metrics at the RNA and chromatin levels suggest substantial similarities between teleost and mammalian intestine and these TSS regulatory regions often show sequence constraint at some level , less than 15% of IEC signature genes had detectable conservation from zebrafish to mouse or human in TSS regions . This hampered our ability to infer which regulatory regions and putative TFBS are actually conserved across these species ( Fig 3E ) [56] . Additionally , most of these conserved regions were identified by MAF blocks and appeared to have only small regions of highly degenerate sequence conservation from teleosts to mammals that were located immediately upstream of transcribed regions and suggestive of minimal functional conservation . We speculated that common regulatory information could still be shared in these TSS regions because short , modular TFBS could escape sequence conservation metrics [56–58] . We looked for enrichment of TFBS using a library of 303 position weight matrices ( PWMs ) ( primarily derived from human ChIP-seq datasets ) within 2 sets of accessible chromatin peaks: ( 1 ) those between 10 kb upstream of the TSS to 10 kb downstream of the transcription termination site ( TTS ) [59] ( Fig 3F ) and ( 2 ) those at the TSS of IEC signature genes ( S4C and S4D Fig ) . Common enrichment of PWMs in accessible chromatin regions at IEC signature genes of different species identified several TF motifs known to regulate IEC expression in mammals including HNF1 , HNF4A , GATA4 , KLF5 , and the often similar ETS factors , including ELF3 and SPDEF ( Fig 3F , S4C and S4D Fig ) [60–63] , that may represent a TF network that control core conserved aspects of IECs in animals . Conservation metrics rely on relatively long stretches of DNA sequence , so we tried to identify sequence properties that varied between species and might interrupt the maintenance of easily detected conserved sequence . By comparing the percentage of DNA bases that are either adenosine or thymine ( AT% ) at TSSs ordered by zebrafish FAIRE signal , we found that gross sequence characteristics are substantially different at the TSS of orthologs between species ( Fig 3G and 3H ) . AT% decreases at the TSS from approximately 51% in zebrafish to 46% in stickleback to 35% in mouse and human ( Fig 3H ) . These differences are maintained on average in the area surrounding the TSS ( Fig 3G and 3H ) . This general phenomenon likely has a substantial influence on the maintenance of sequence at these regulatory regions [64 , 65] and may represent a particular challenge for the identification of conserved regions and regulators . However , the substantial difference in AT% observed at TSS is absent on average at CNEs , which , based on the method that they are identified , are inherently similar in sequence ( Fig 3I ) . This suggests that general sequence utilization differences seen at TSSs in different species on average are important but may not influence all regulatory regions . The apparent maintenance of TFBS enrichment suggested that our accessible chromatin and RNA-seq data were identifying functionally conserved regions , although they frequently lacked conserved DNA sequence . To test this , we cloned regions upstream of the rbp2a and fabp6 TSS that showed no substantial sequence conservation from fish to mammals , but appeared accessible largely in IECs , and tested them using a functional in vivo zebrafish reporter assay ( Materials and methods ) . Because both RBP2 and FABP6 are IEC signature genes that showed strong regional conservation of expression in zebrafish and mouse ( Fig 2 ) , this allowed us to simultaneously test if the conservation of regulatory information was interpreted by zebrafish to specify intestinal regionality in addition to IEC expression generally . We generated a transgenic reporter construct with the 1 . 3 kb region upstream of the TSS of zebrafish retinol binding protein gene , rbp2a , cloned upstream of the mouse cFos minimal promoter and green fluorescent protein ( GFP ) [Tg ( rbp2a:GFP ) ] ( Figs 4A and 3E , S5A , S6A Figs and S4 Table ) . Tg ( rbp2a:GFP ) was capable of driving high expression in the anterior portion of the intestinal epithelium in larvae ( Fig 4A ) , which was consistent with the known expression patterns of rbp2a [15 , 66 , 67] , suggesting that sufficient regulatory information to drive regional expression in IECs is contained within this fragment . This expression pattern is distinct from a general control reporter construct in which no additional DNA was cloned upstream of the mouse cFos minimal promoter driving GFP , and no consistent IEC expression is found ( S6B–S6I Fig ) [68] . Using the same reporter assay , we then tested the region immediately upstream of Rbp2 from mouse Tg ( Mmu . Rbp2:GFP ) and RBP2 from human Tg ( Hsa . RBP2:GFP ) and found that both were capable of driving GFP in IECs in the anterior portion of the larval zebrafish intestine ( Fig 4A and S5A Fig ) . We tested if we could identify common , small TFBS that might have escaped detection by conservation metrics in these regions but explain the conserved patterns of IEC expression . TF motif searching within the RBP2 fragments for zebrafish , mouse , and human identified shared strong matches to HNF4A and GATA factors within a few hundred bases of the TSS ( Fig 4B and 4C ) . To test how often HNF4A and GATA motifs occurred generally , we queried a similarly sized area 150 bp upstream of the TSS of all genes . Only . 23% ( 79 genes ) , . 25% ( 51 genes ) , and . 25% ( 47 genes ) had both HNF4A and GATA4 motifs in zebrafish , mouse , and human , respectively , and no other 1-to-1-to-1 orthologs contained both HNF4A and GATA4 motifs at this location in all 3 species . This suggests that the presence of HNF4A and GATA sites in the RBP2 TSS is conserved among vertebrate genomes and is unlikely to have occurred by chance . Indeed , HNF4A and GATA4 are likely a common vertebrate regulatory cassette as studies in multiple species have identified their importance in small intestinal IEC biology ( S3G and S3H Fig ) [26 , 69] . Collectively , these results establish that accessible chromatin maps can help discern conserved motif information and how in vivo reporter assays can be used to test for potential conserved tissue-specific regulatory activity . We next sought to interrogate the regulatory potential of regions upstream of FABP6 , an IEC signature gene expressed primarily in the ileum in mouse and human and whose TSS was accessible in all IECs tested except mouse and human colon ( Figs 3E and 4D ) . The region 258 bp upstream of the zebrafish fabp6 TSS [Tg ( fabp6:GFP ) ] drove a very specific GFP expression domain exclusively in IECs in the middle of the larval zebrafish intestine , consistent with the endogenous pattern of fabp6 mRNA expression ( Figs 2C and 4D , S5B and S5G–S5J Fig ) [15 , 47 , 70] . Unlike rbp2a , fabp6 had a small region upstream of the fabp6 TSS that was conserved to mouse and human . However , this appeared to only correspond to a TATA-box ( Fig 4G , S5B and S5C Fig ) , and cloning of this small region from mouse to test in our zebrafish reporter assay did not drive expression in IECs . However , when we included the entire accessible chromatin region from mouse Tg ( Mmu . Fabp6:GFP ) ( a 503 bp region upstream of mouse Fabp6 TSS ) , which included the minimal conserved region , we found this larger sequence sufficient to drive an IEC expression pattern that was positionally identical to the corresponding region from zebrafish ( Fig 4D ) . This suggests the regulatory information necessary to drive IEC expression in the putative zebrafish ileum is within the additional region defined by accessible chromatin from mouse and not solely detected by conservation . In order to define the relationship between this fabp6 domain in the context of the canonical 3 segments of the zebrafish intestine [14] , we compared the larval expression pattern of Tg ( fabp6:GFP ) with zebrafish segment 2 marker TgBAC ( lamp2-RFP ) [71] and intestinal segment 1 marker Tg ( -4 . 5fabp2:DsRed ) [72 , 73] . Strikingly , we found Tg ( fabp6:GFP ) did not overlap with either , suggesting the intestinal region marked by Tg ( fabp6:GFP ) is a novel distinct segment of the zebrafish larval intestine ( Fig 4E–4F ) . Similar regions upstream from human FABP6 were negative for driving expression in zebrafish IECs , despite the putative presence of shared IEC-related TFBS like CDX2 , RBPJ , and RXR ( Fig 4G , S5B Fig and S4 Table ) . The combined evidence of conserved positional expression of Fabp6 and other ileal genes ( Fig 2C and S3D Fig ) together with the maintenance of region-specific cis-regulatory information at zebrafish and mouse Fabp6 orthologs ( Fig 4D–4G ) indicate that an intestinal segment functionally and regionally homologous to mammalian ileum is maintained in zebrafish larvae and likely specified by similar regulators . To determine if the transcriptional patterns and domains we detect in the larval zebrafish intestine also occur in adult stages , we assayed expression patterns in the adult ( 3+ months ) zebrafish intestine using the same stable transgenic lines we queried in larvae . Tg ( rbp2a:GFP ) showed a similar expression pattern restricted to the anterior intestine as larval zebrafish; however , consistent with adult expression data ( Fig 2B ) , high GFP expression did not extend to the most anterior IECs ( Fig 5A–5B ) . This suggests that additional transcriptional domains or functional differences exist in the most anterior zebrafish intestine . We also found that Tg ( fabp6:GFP ) had a very similar pattern between larval and adult stages with a relatively small and discreet region of IEC GFP expression immediately after the second bend and between the segment 2 marker TgBAC ( lamp2-RFP ) ( Fig 5F–5G ) and the segment 1 marker Tg ( -4 . 5fabp2:DsRed ) expression domains ( Fig 5H–5I ) . Collectively , this suggests that regional transcriptional programs in the zebrafish intestine are maintained between larval and adult stages . Further , the extent of functional homology between the zebrafish and mammalian intestine may be greater than previously appreciated ( Fig 5J–5K ) . We propose a working model with at least 5 transcriptional/functional domains in zebrafish , although additional studies are needed to comprehensively resolve these domains , their interplay , boundaries , regulators , as well as the full nature and limitations of the homology between teleost and mammalian regional IEC programs ( Fig 5K ) . In addition to TSS regions , we also specifically queried a published dataset of 54 , 533 zCNEs , of which 11 , 792 are also conserved to mouse and human , for accessibility in IECs [55] . Ordering zCNEs by the FAIRE-seq signal from zebrafish identified that the neighboring genes of the most accessible CNEs were also highly expressed in other-species IEC samples ( Fig 6A ) . There was also a surprising overlap in the magnitude of accessibility at these conserved sites between species ( Fig 6A , 6B and 6F ) . However , 48 of the 77 CNEs that were accessible in all zebrafish , mouse ileum and colon , and human colon IEC datasets were also accessible in at least 82% ( 14/17 ) of additional nonintestinal ENCODE and Human Roadmap datasets ( Fig 6B–6D ) [52–54] . This suggests that these regions are not specifically responsible for IEC expression . However , these frequently accessible conserved sites could represent a particular pan-vertebrate primitive transcriptional networks as zCNEs are commonly found at TSSs and near developmental and TF genes ( Fig 6B–6E ) [55] . Some of these conserved regions that are accessible in most tissues were near genes known to have roles in IEC biology , like Egr1 [74] , Nr1d1 [75] , and Jun [75] , and we did not want to exclude that these regions could still be important in IEC expression . Cloning constitutively accessible regions from nr1d1 and jun were negative for IEC expression . When we cloned the egr1-neighboring CNE regions from zebrafish ( zCNE_11264 ) , mouse ( mzCNE_11264 ) , and human ( hzCNE_11264 ) and tested them separately using our reporter assay , unsurprisingly , multiple tissues showed GFP expression . However , we observed a distinct differential pattern across the intestine , with GFP expressing most highly in IECs in the mid intestine in zebrafish reporter lines representing all 3 species ( Fig 6G–6J ) . TF motif searching identified multiple ETS and CA/T-rich-G ( CArG ) /MCM1 , AGAMOUS , DEFICIENS , and SRF ( MADS ) box sites in all 3 species in CNE_11264 , which are often immediately adjacent ( Fig 6K ) , consistent with a serum response element that has been characterized at human EGR1 [76–78] . We noticed the size of the conserved region and spacing and number of CArG and ETS binding sites of zCNE_11264 ( 458 bp ) and mzCNE_11264 ( 352 bp ) were greater than hzCNE_11264 ( 120 bp ) ( Fig 6G–6L ) . Searching for local CArG and ETS binding sites identified a neighboring cluster of additional CArG and ETS binding sites , approximately 200 bp outside of the conserved hzCNE_11264 region ( Fig 6K ) . hzCNE_11264 was capable of driving similar IEC expression without these additional putative redundant conserved binding sites ( Fig 6G , 6J and 6K ) . However , this highlights the imperfect nature of defining conserved regions across distantly related species and suggests that accessible chromatin maps combined with searching for common neighboring motif language seeds may identify additional sequence conservation information ( S7J–S7L Fig ) . This same combination of ETS and CArG TFBS was found immediately upstream of Egr1 in the stickleback genome , although zCNEs are not specifically annotated in stickleback ( Fig 6K ) . Identifying these discrete nonoverlapping ETS and CArG binding sites at CNE_11264 in multiple species suggests that multiple ETS and CArG sites have functional relevance for this regulatory region ( Fig 6K ) . However , this also demonstrates the complexity of identifying conservation that often relies on flexible , redundant regulatory logic ( Fig 6K and 6L ) . Similarly , the diversity of detected CArG boxes also show how degenerate TF sequences can deviate while potentially maintaining a similar functional output ( Fig 6M ) [79] . In an attempt to find zCNEs and upstream regulators that act primarily in IECs , we identified a group of 15 zCNEs that were not accessible in at least 9 out of 17 non-IEC tissues but were always accessible in IECs in zebrafish , mouse , and human ( Fig 6E ) . Motif analysis of these IEC-accessible CNEs revealed common putative IEC-related TFBS in zebrafish , mouse , and humans such as GATA , HNF4A , HOX , CDX2 , HNF6 , HNF1 , and TEAD ( S7A–S7I and S8 Figs ) . One of these zCNEs neighbors , the gene hairy and enhancer of split-1/hairy-related 6 ( HES1/her6 ) ( zCNE_44665 ) , showed remarkable accessible chromatin specificity within intestinal datasets ( Fig 6E; S9A and S9B Fig ) . Hes1 is a transcriptional repressor known to play diverse roles in many tissues including embryogenesis and neural and T-cell development [80–82] . Importantly , it also plays critical roles in the differentiation of IEC subtypes from intestinal stem cell progenitors and in mouse is expressed exclusively in the intestinal crypt [17 , 83] . While crypts are not present in the zebrafish intestine , and IEC progenitor or stem cells in fish have only been recently characterized [11] , hes1/her6 has been found to be expressed in a distinct subset of IECs in an analogous compartment at the base of zebrafish intestinal folds [84] . It is not fully known what genomic regions regulate Hes1 IEC expression and if these regions control aspects of Hes1’s transcriptional response to microbes [85] or in intestinal cancer [86] . We were curious if our IEC chromatin data did discriminate important conserved regulatory regions that drove expression in IECs ( Fig 7A ) . In our reporter assay zCNE_44665 , a region approximately 3 , 600 bp upstream of the hes1/ her6 TSS in zebrafish , revealed strong IEC expression and expression in other tissues including liver ( Fig 7B–7D , S10 Fig ) . Whole-mount and cross-sections of 7 dpf fish identified that high GFP expression was within a subset of IECs , often at the base of slight invaginations of this cell layer ( Fig 7B–7E , S10A Fig ) . These invaginations are ultimately analogous to the base of epithelial folds ( rugae ) seen in older fish and the intestinal crypt in mammals , although , at 7 dpf , substantial folds are not articulated ( Fig 7D and 7E ) [12] . This Tg ( zCNE_44665:GFP ) population did not overlap with cells that were positive with the enterocyte marker Tg ( -4 . 5fabp2:DsRed ) ( Fig 7F and S10G Fig ) [73] , the enteroendocrine marker Tg ( neurod1:TagRFP ) ( Fig 7G and S10H Fig ) [87] , or cells with characteristic goblet cell morphology . This suggests the hes1-neighboring conserved region zCNE_44665 drives GFP in an undercharacterized population of IECs that is distinct from these known differentiated IEC subtypes in zebrafish . We scanned the sequence of the orthologous CNEs from zebrafish , mouse , and human and found TFBS for hepatocyte nuclear factor 1 ( HNF1 ) , hypoxia inducible factor 2 alpha ( HIF2A ) , and recombination signal binding protein for immunoglobulin kappa J region ( RBPJ ) in all 3 CNEs ( Fig 7L; S9A–S9C Fig ) . RBPJ is known to regulate HES1 expression in the presence of Notch signaling [88] , resulting in the alteration of the proportion of secretory and absorptive IEC lineages [17] . Therefore , we tested if this Tg ( zCNE_44665:GFP ) population was positive for Notch signaling . Crosses of Tg ( zCNE_44665:GFP ) with Tg ( EPV . Tp1-Ocu . Hbb2:hmgb1-mCherry ) , which uses a viral-derived promoter with Notch-responsive RBPJ binding sites [89] , revealed substantial overlap between mCherry+ and GFP+ cells ( Fig 7H and 7I , S10 Fig , S1 Movie ) . The mCherry+ IECs were always GFP+ , and only 64 . 6% of GFP+ cells were mCherry+ , which could be due in part to the relatively slow maturation of mCherry protein . This overlap suggests Notch signaling is important in regulating this element upstream of HES1/her6 ( Fig 7H and 7I , S10 Fig ) . A cross-section of dissected intestine from 8-week-old fish revealed a more sophisticated expression pattern in relationship to the now articulated intestinal folds ( Fig 7I ) . GFP+ cells included cells at the base and typically in the bottom half of the intestinal folds and included cells positive for Notch signaling . Interestingly , these cells frequently had more apical nuclei than most IECs , whose nuclei typically are located basally within the epithelium ( Fig 7I and S10E Fig ) . To determine if a similar regulatory capacity was conserved to mammals , we then tested the hzCNE_44665 from human ( approximately 10 kb from the TSS of HES1 ) . This human region also drove intestinal expression that was similarly limited to a subset of IECs that often overlapped with Notch-positive cells ( 30 . 5% GFP+/mCherry+ , 33 . 6% GFP+ only , 35 . 7% mCherry+ only ) ( Fig 7J and 7K ) . To determine if common putative TFBSs found in CNE_44665 were regulating expression in IECs , we generated a zebrafish line with the HNF1 binding site abolished through mutation in the zCNE_44665 GFP reporter construct Tg ( zCNE_44665 ΔHNF1:GFP ) ( Fig 7L and 7M ) . We still identified GFP expression in a subset of cells that was coincident with Notch signaling in IECs from stable fish lines containing this construct ( Fig 7N ) . However , when we abolished the RBPJ binding site , we found that high GFP expression was completely lost in IECs , including the subset of cells that overlapped with Notch-positive cells ( Fig 7O ) . This suggests a conserved RBPJ binding site is necessary for expression in IECs and may contribute to common HES1 regulation in fish and mammals in progenitor IEC populations ( Fig 7O ) . Collectively , our chromatin data analysis is capable of distinguishing IEC regulatory regions and putative causative TFBS from complex regulatory landscapes that may regulate a broadly expressed gene’s transcription in IECs .
Conserved regulatory elements that are stable over millions of years are likely to have analogous functions in their respective genomes and could coordinate conserved tissue-specific transcriptional patterns . However , identifying and functionally annotating these regions at individual loci or across a single genome is problematic , tedious , and does not accurately predict or robustly identify conserved function . As demonstrated above , our strategy of combining tissue-specific transcription and accessible chromatin datasets with conservation and TFBS prediction inferred from DNA sequences from multiple species identified putative conserved regulatory and functional information in IECs that could not have been identified by any one data set alone . Tissues and cell types like IECs are defined by complex patterns of gene expression . Despite specializations in each animal species , IECs serve core inherent functions such as absorption and metabolism of dietary nutrients and xenobiotics and as a barrier to microbes residing within the intestinal lumen . As evidence for this common conserved function , we identified 470 orthologous genes expressed highly in IECs with relative tissue-specificity across 420 million years of vertebrate evolution . Functional conservation was maintained across a broad range of IEC biology , including genes involved in IEC subtypes , lipid transport and metabolism , and a response to microbes and inflammation . We highlight commonly expressed TFs because they may underlie the expression of conserved networks that are associated with IEC function , identity , and regionalization . YBX1 , HNF4A , ELF3 , XBP1 , ID3 , HMGB2 , IRF1 , STAT3 , GATA5 , and OSR2 , amongst other TFs , appear to be more highly expressed in IECs than other tissues . Importantly , many of these highly expressed TFs also show enrichment for their cognate TFBS in accessible chromatin surrounding IEC signature genes in multiple species including the ETS factors ELF3 and ELF4 , HNF4A , GATA5 , and STAT3 . Our understanding of intestinal evolution has been hindered by a lack of information about the degree to which anteroposterior segments in extant vertebrate species are ancestral or derived traits . We found patterns of conserved expression along the length of the zebrafish and mouse intestine , suggesting that conserved discrete transcriptional regulatory programs may specify homologous duodenal , jejunal , ileal , and colonic segments along the zebrafish intestine [15] . We found striking evidence for transcriptional regulation underlying this conserved similarity , as the genomic region immediately upstream of the highly expressed IEC signature gene FABP6 , in both zebrafish and mouse , was capable of driving GFP expression coincident between the zebrafish intestinal segment 1 and segment 2 domains [12–15] . We infer that this discrete segment specified by the expression of FABP6 and other markers functions as the conserved homologous zebrafish ileum and that the zebrafish intestine is more completely defined as at least 5 distinct segments with further evidence that transcriptional domains similar to the duodenum , jejunum , and colon exist ( Figs 2 and 5 ) . Collectively , these results indicate that the transcriptional underpinnings of the well-characterized segmental program present in the mammalian intestine are ancestral to the last common ancestor with bony fishes and that the utility of the zebrafish as a model for human intestinal biology is even greater than previously appreciated . Our initial analysis to utilize the presumed regulatory information at the TSS of IEC signature genes identified a number of regions with largely IEC-specific accessible chromatin status . However , we were unable to identify a substantial number of highly sequence-conserved regions at these TSSs . To circumvent the apparent lack of conserved regulatory information despite clear transcriptional similarities , we applied strategies through which regulatory information could be inferred without directly using traditional sequence conservation metrics . Searching for significantly enriched TF motifs found in accessible chromatin regions surrounding genes expressed in IECs in multiple species allowed us to identify common presumptive TF motifs that are used in the regulation of these genes , including HNF1 , HNF4A , GATA , and ETS factors [60 , 62 , 63] . In addition , we looked for common predicted TFBS in TSS regions that appeared to have accessible chromatin in IECs but no strong sequence conservation . Accordingly , despite the lack of sequence conservation , the RBP2 ( a ) promoter regions from zebrafish , mouse , and human are capable of driving highly similar expression in the IECs of zebrafish . This conserved expression is presumably largely due to common HNF4A and GATA motifs in zebrafish , mouse , and human that escape detection by commonly used sequence conservation metrics . Recently , a microbially responsive element in the zebrafish angptl4 gene was shown to contain an element with HNF4A and GATA motifs that were involved in driving expression that is essentially identical to the expression pattern from our rbp2a fragment [68] . Fabp2 also shows the same regional expression as rbp2a and angptl4 in the intestine and has binding sites for HNF4A and GATA that are shared to mammals [72] . Furthermore , the TFs HNF4A and GATA , and FXR , that have putative binding sites in the regulatory regions from Rbp2 and Fabp6 , respectively , show intestinal expression patterns that appear to explain much of the regional IEC expression of these genes ( S3G and S3H Fig ) . This suggests these common combinations [18] are conserved in regulating genes in the IECS of anterior intestines from teleosts to mammals . Intriguingly , HNF4A and FXR were also recently shown to mediate IEC responses to microbiota [26 , 90] , indicating complex relationships between tissue-specific and microbially-responsive transcriptional programs in the intestinal epithelium . We were able to identify a small number of highly conserved noncoding elements with apparent conserved IEC-specific chromatin accessibility , representing excellent candidates to understand conserved regulatory mechanisms that drive IEC expression ( Fig 6 and S7 Fig ) . Our strategy highlights the utility of these data in identifying tissue-discriminating regulatory regions at genes that lack clear tissue-specific transcription and may be selectively regulated by discrete regulatory regions in different tissues . Similar strategies using available mammalian accessible chromatin datasets could annotate the remaining CNEs with putative functions . We focused on a CNE upstream of HES1 because of its exceptional accessible chromatin specificity in the intestine ( Fig 6E , S9A and S9B Fig ) and the known importance of HES1 in IEC biology [91] . We identified a necessary RBPJ binding site within the hes1 CNE that drove expression in IECs coincident with Notch signaling . Notch signaling likely plays a complex function at the hes1 locus , as amongst the 6 hes1/her6 neighboring zCNEs , 4 contain conserved RBPJ binding sites in their zebrafish , mouse , and human CNE counterpart ( zCNE_44665 , 44657 , 44661 , 44663 ) , 1 CNE ( zCNE_44658 ) contains an RBPJ site in zebrafish and human , and only 1 additional CNE ( zCNE_44660 ) contains no RBPJ sites [92 , 93] . The only HES1/her6-neighboring zCNE containing a predicted binding site for HNF1 was zCNE 44665; however , loss of the HNF1 binding site was not sufficient to ablate expression in Notch-positive IECs , suggesting that other mechanisms confer the specific accessibility and expression in IECs . Due to the existence of other RBPJ sites at this locus , it seems unlikely that Notch signaling strictly underlies the IEC specificity , although tissue-specific expression of Notch ligands may also contribute to IEC expression [94] . Hes1 is expressed exclusively in the crypt of the mammalian intestine , including in stem cells and transit-amplifying progenitors coincident with Notch signaling [83 , 94 , 95] . Similarly , we see the zebrafish and human HES1 CNE_44665 is capable of driving expression in IECs at the base of and bottom half of intestinal folds using our zebrafish GFP reporter assay , also coincident with Notch signaling [84] . It will be interesting to determine if Hes1 and the CNE_44665-marked cells play an analogous progenitor role in fish as well as mammals . Though we focused on the sequenced conserved region that also showed accessible chromatin specificity , a larger region outside of this conserved region showed intestinal accessible chromatin specificity in human and mouse ( S9A and S9B Fig ) . Interestingly , a GFI1B binding site detected in the zebrafish zCNE , but absent in the mouse and human zCNE , was detected within the adjacent region that was accessible specifically in mouse and human intestinal datasets . GFI1B is a repressor TF that helps specify the IEC-subtype tuft cell [96] . This larger regulatory region may facilitate conserved regulation of HES1 expression and restrict Hes1 expression in tuft cells , even though this GFI1B binding site is not found in a region that is conserved in sequence from zebrafish to human ( S9A–S9C Fig ) [96] . These results provide an important context for further exploring how conserved DNA regulatory regions and multiple TFs function cooperatively to regulate expression of the HES1 gene in IECs and other distinct tissues . While we framed this study around the idea that particular regulatory regions and transcripts could have specific functionality in IEC cell types , an interesting premise is that important regulatory regions and transcripts may be exclusive to multiple distinct organs or cell types , and these circuits could be conserved across species . Of course , even IECs are heterogeneous , so a complete understanding of IEC-specific programs will require higher resolution maps of IECs and multiple additional tissues [33] . Our examples of Fabp6 and Rbp2 do indeed seem to show high GFP expression that is largely limited to IECs ( Fig 2 ) . However , HNF4A and GATA also function in the liver [97] where rbp2a is expressed [66]; therefore , the definition of specificity and understanding of insulation of transcriptional regulation in different cell types requires further study . Though we believe our approach successfully identified genes and regulatory regions that are important in IECs , we did notice that the TSS at many IEC signature genes also showed accessibility in kidney and liver tissue ( Fig 3E ) , suggesting an important overlapping utilization of the regulatory regions and gene functions in these other tissues . We do not want to minimize the importance of the concept that many regulatory regions are used in multiple cell types and should perhaps holistically be thought to function , be selected upon , and be conserved in an interorgan/cell-type regulatory network . The strategies and methods we used here were able to detect diverse types of conserved transcriptional and regulatory information in fish and mammalian IECs; however , important limitations apply . Specifically , with our strategy , some pathways utilized broadly in many tissues , genes expressed in a small number of cells , or lowly expressed genes may be difficult to characterize for their conservation across species in the intestine and other tissues . For example , though the Wnt pathway functions in a broad range of tissues [98] , it has been established as functioning in a similar manner in zebrafish and mammalian intestine . The Wnt coactivator TCF4 is tied to microbially regulated epithelial cell proliferation in zebrafish [99] . Remarkably , deletion of the Wnt pathway inhibitor , Apc , results in intestinal tumors in both zebrafish [100] and mammals [101] . However , only a small number of Wnt-related genes , including FZD5 , were identified here as IEC signature genes , although individual ligands or pathway components may signal the use of broader pathways in IECs . As a result , the lack of highlighting a conserved gene or pathway cannot be considered as the lack of conservation , generally . Furthermore , the ability to find conserved function in a particular cell type , gene family , or pathway may depend highly on the proposed definition of conservation and be complicated by each genome’s unique history . Teleosts underwent a genome duplication approximately 340 million years ago [102] . Approximately 20% of these duplicated gene pairs are maintained in the extant zebrafish and other teleost species , although the same duplicated genes are not always maintained in each genome [102–104] . Retained duplicated genes can undergo , amongst other fates , neo- and subfunctionalization specific to a lineage , spurring adaptation and potentially driving speciation [105] . This duplication can complicate comparative genomics and the parsing of function across vertebrate species with limited transcriptional profiling , but it also provides a rich platform for understanding gene evolution , function , and regulatory regions [106] . To create a focused strategy using genes that likely had maintained function and therefore detectable transcription and regulatory signal across species , our RNA analysis was largely limited to 1-to-1 orthologs across the 4 species we assayed , although our CNE analysis was essentially independent of orthology . We suspect that the diversity of types of regulatory information aren’t unique to genes with specific types of orthology across species , although novel regulation of gene function may more frequently arise during sub- and neofunctionalization of duplicated paralogs [105] . In addition , certain gene groups such as TFs are less likely to be lost following duplication in the zebrafish genome [102] , suggesting that additional analysis may be required to uncover and integrate the functions and regulatory information that are contained within groups of genes that do not show 1-to-1 orthology . We found instances at Rbp2 , Fabp6 , Egr1 , and Hes1 in which traditional sequence conservation metrics were not sufficient to fully identify common putative regulatory information from fish to mammals even when it is located approximately the same distance relative to the gene body in all species [57 , 107] . In all instances , accessible chromatin data provided additional context to identify conserved regulatory information . Importantly , because TFBS can be modular , conserved transcriptional regulation can occur with almost no sequence conservation signal being detectable across species . This property is largely due to the inherent nature of TFs because plasticity in the number , arrangement , and affinity of TFBS can result in nearly identical transcriptional responsiveness and output even in the absence of long stretches of sequence conservation . To more accurately identify conserved regulatory information , metrics are needed that incorporate short degenerate TF motifs or identify highly conserved short motifs between closely related species that are then identified as a conserved block in distantly related groups of similar species in syntenic regions . This information could be further anchored by accessible chromatin [108 , 109] . This may only partially circumvent the somewhat inherent statistical problem of short sequences arising by chance . The interrogation of regulatory DNA sequences using assays such as GFP reporters can determine how complex orthologous regulatory sequences are interpreted in vivo . However , these assays are not necessarily sufficient to determine functional conservation across species . For example , orthologous TFs may show an altered TF binding motif preference in each species , although the regulation by this common factor is itself otherwise conserved . Alternatively , functional TFBS from one species may be invalidated by other neighboring sequences in cross-species assays , although the same TF is successfully regulating the same gene in each species . This issue is highlighted by the difference in AT% usage at the TSS of the 4 species used in this study that presumably is partially distinct from traditional TF binding content ( Fig 3G ) . This property may partially be driven by more general sequence characteristics that require even less stringency in conserving specific long blocks of DNA like CpG methylation [64] , nucleosome positioning , and overall dynamics of nucleosome accessibility [110] , binding competition dynamics [111] , or TSS selection [65] that may vary from organism to organism . Detecting sequences and principles that account for these confounding factors will need to be accomplished and compensated for before a complete understanding of the similarity and differences between analogous organs and transcriptional programs can be realized . Tissue-specific transspecies high-throughput enhancer activity assays [112] may be required to sufficiently sample and test transcription rules and conservation for specificity across species . These will need to be further combined with a similar deep understanding of posttranscriptional control and epigenetic mechanisms to develop a more complete picture of signals encoded in DNA and if they are commonly or divergently utilized . Our experimental strategy utilized IECs from healthy adult animals representing 4 vertebrate lineages to reveal conserved mechanisms underlying tissue- and region-specific IEC transcriptional regulation . These results provide an important frame of reference for future efforts to uncover similar mechanisms in IEC subtypes or in the context of other developmental stages , disease states , or environmental exposures . Many of the genes and upstream TFs broadly implicated here as conserved features of vertebrate IECs have already been implicated in human diseases ( S1 Table ) and in the intestinal response to microbiota [26 , 90] , prompting further studies into the mechanistic relationships between transcriptional regulatory networks governing IEC identity , environmental sensitivity , and disease pathogenesis . We believe that using the strategy of simultaneously leveraging genome-wide data sets from multiple species can identify key ancient aspects of biology more quickly than studying any species alone .
Zebrafish studies were approved by the Institutional Animal Care and Use Committees of Duke University ( protocol A165-13-06 ) and University of North Carolina at Chapel Hill ( protocol 12–058 . 0 ) . Stickleback studies were approved by the Institutional Animal Care and Use Committee of Stanford University ( protocol 13834 ) . Stickleback were collected under California Scientific Collecting Permit #3260 . Studies involving human tissues were performed under University of North Carolina at Chapel Hill IRB approval numbers 10–0355 and 14–2445 . Mouse tissue dissection and IEC extraction protocols and all initially processed data were previously described ( GSE57919 ) [16] . Putative regulatory elements were amplified from genomic DNA with primers containing FseI and AscI restriction site overhangs and then cloned , maintaining orientation relative to their native TSS , into the p5E-FSE-ASC entry plasmid ( 381 ) ( http://tol2kit . genetics . utah . edu/; Tol2kit v1 . 2 ) [125] . Putative clones were confirmed by PCR and sequencing to ensure an exact match to genomic sequence from the corresponding species . Four-way LR reactions were performed using the LR Clonase II Plus Kit ( 12538120 , Invitrogen ) combining p5E-FSE-ASC modified with a putative regulatory element , pME cFos ( S4 Table ) , p3E EGFP ( 366 ) , and pDestTol2pA2 ( 394 ) using the provided protocol . This generated a single plasmid recombined with a putative regulatory region upstream of a minimal mouse cFos promoter-driving eGFP and flanked by Tol2 transposon insertion sites , which was confirmed to contain the putative regulatory region by PCR . Site-directed mutagenesis constructs were generated by randomly changing key bases over 10 bp in TFBS , detected by Homer within the zebrafish hes1/her6-neighboring zCNE_44665 . Homer could no longer detect the targeted TFBS using the mutated genomic sequence in the context of the otherwise original zCNE_44665 sequence . For each site-directed mutation , complementary primers containing the 10 bp mutated region flanked by wildtype 20 bp sequences on either side were used in circular PCR with the original zCNE_44665 containing 381 plasmid as template followed by DpnI treatment to digest methylated plasmid . Site-directed mutagenesis was confirmed by sequencing . | The epithelium lining the intestine is an ancient animal tissue that serves as a primary site of nutrient absorption and interaction with microbiota . Its formation and function require complex patterns of gene transcription that vary along the intestine and in specialized intestinal epithelial cell ( IEC ) subtypes . However , it is unknown how the underlying transcriptional regulatory mechanisms have changed over the course of vertebrate evolution . Here , we used genome-wide profiling of mRNA levels and chromatin accessibility to identify conserved IEC genes and regulatory regions in 4 vertebrate species ( zebrafish , stickleback , mouse , and human ) separated from a common ancestor by 420 million years . We identified substantial similarities in genes expressed along the vertebrate intestine . These data disclosed putative conserved transcription factor binding sites ( TFBS ) enriched in accessible chromatin near IEC genes and in regulatory sites with accessibility restricted to IECs . Fluorescent reporter assays in transparent zebrafish showed that these regions , which frequently lacked sequence conservation , were still capable of driving conserved expression patterns . We also found a highly conserved region near mammalian and fish hes1 sufficient to drive expression in a specific population of IECs with active Notch signaling . These results establish a platform to define the conserved transcriptional networks underlying vertebrate IEC physiology . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"gene",
"regulation",
"vertebrates",
"animals",
"animal",
"models",
"osteichthyes",
"model",
"organisms",
"experimental",
"organism",
"systems",
"sequence",
"motif",
"analysis",
"epigenetics",
"mammalian",
"genomics",
"chromatin",
... | 2017 | Genomic dissection of conserved transcriptional regulation in intestinal epithelial cells |
Extracellular signaling is a mechanism that higher eukaryotes have evolved to facilitate organismal homeostasis . Recent years have seen an emerging interest in the role of secreted microvesicles , termed extracellular vesicles ( EV ) or exosomes in this signaling network . EV contents can be modified by the cell in response to stimuli , allowing them to relay information to neighboring cells , influencing their physiology . Here we show that the tumor virus Kaposi’s Sarcoma-associated herpesvirus ( KSHV ) hijacks this signaling pathway to induce cell proliferation , migration , and transcriptome reprogramming in cells not infected with the virus . KSHV-EV activates the canonical MEK/ERK pathway , while not alerting innate immune regulators , allowing the virus to exert these changes without cellular pathogen recognition . Collectively , we propose that KSHV establishes a niche favorable for viral spread and cell transformation through cell-derived vesicles , all while avoiding detection .
Extracellular communication is pivotal to maintain organismal homeostasis and a disease state . One of the mechanisms by which cells communicate to their surroundings is through extracellular vesicles ( EV ) . Cells package a number of biological molecules into EV such as proteins , nucleic acids , lipids , and metabolites . EV are released into the microenvironment as well as into the circulation via the blood and lymphatic vessels . All vessels are lined with endothelial cells ( EC ) , which are permanently exposed to EV , and uptake the EV-loaded cargo . This may induce changes in differentiation , metabolism , migration , and gene expression ( reviewed in [1] , for further examples see [2] ) . Ample experimental evidence has connected EV to cancer metastasis , immune signaling and the response to invading pathogens , though the molecular details of EV biology are far from established and tend to differ dramatically among experimental systems [1] . As in many aspects of modern biology it is difficult to come up with experimental approaches that are both tractable and physiologically relevant . Kaposi Sarcoma ( KS ) and Kaposi-sarcoma-associated herpesvirus ( KSHV ) lymphomas , specifically primary effusion lymphoma ( PEL ) , represent one such system to study EV biology . KS is one of the most angiogenic cancers in humans , and was one of the first identifiable markers for AIDS ( reviewed in [3] ) . KSHV is the etiological agent of KS and PEL , which induce a unique tumor microenvironment that remodels tumor and lymphatic vasculature [4–6] . Of note , the transdifferentiation induced by the virus is as dramatic in uninfected neighboring cells as in virus-infected cells . We had shown that KSHV-infected cells release EV containing all viral micro RNAs ( miRNA ) , but not the virus itself , and that the viral miRNAs are present at high concentrations in KSHV-lymphoma derived EV ( KSHV-EV ) in culture and in patients [7] . This establishes the KSHV-EV:EC interaction as physiologically relevant , experimentally robust , and as we show here , highly tractable . EV membranes are enriched for phosphatidylserine ( PS ) , which is recognized by Annexin-V and plays a role in EV adsorption . Tetraspanins , such as CD9 , CD81 , and CD63 , are found on the majority of EV and have been used to affinity-purify EV [8–12] . Alix and Flotillins-1 and 2 are additional molecules that define EV ( see http://www . exocarta . org/ ) . Multiple EV purification methods have been devised [13] . These purification schemes yield comparable preparation of EV though the resultant fractions can be quite heterogeneous and need to be carefully characterized in each experiment . Exosomes are a subtype of EV that is defined by their intracellular biogenesis . Exosomes originate from the inward budding of the late endosome into the multivesicular body and traffic from there to the plasma membrane where they are released . When studies use primary patient material and/or cell culture supernatant , the origin of the EV cannot unequivocally be attributed to the multivesicular body; and the term EV rather than exosome is used . To date , no EV-specific receptors have been defined; evidence for tissue-specific uptake is limited to specific scenarios such as neuronal or immunological synapses [14] . Unlike viruses , EV are believed to be able to enter all cell types . Viruses that modulate EV maturation and content include Human and Simian immunodeficiency virus ( HIV and SIV ) , vaccinia virus , hepatitis C virus , and herpesviruses [1 , 7 , 12] . This led Gould et al . to propose the trojan horse hypothesis [15] , whereby viruses use EV to modulate the cell physiology of neighboring cells in order to further infection . In this study , we sought to characterize how EV taken from the HIV-associated PEL , either cell culture or primary patient fluid , influence EC behavior . We discovered that affinity purified EV mediated cell migration , proliferation , and secretion of human interleukin-6 ( IL-6 ) through the extracellular signal related kinases ( ERK1/2 ) pathway . This was accomplished without tripping of innate sensors such as interferon regulatory factor 3 ( IRF3 ) , stimulator of interferon genes ( STING ) [16–21] , or nuclear factor kappa B ( NF-κB ) . It allows KSHV to modulate its immediate environment without alerting innate immune signaling pathways . Chronic exposure to PEL-derived EV , mimicking the pleural environment , resulted in a reprogramming of the recipient cells’ mRNA profile , without activating innate immune signaling cascades and/or interferon stimulatory genes . Collectively , these results provide a new paradigm for EV function in modifying the phenotype of recipient cells and highlight a previously unknown method of virus-induced cellular reprogramming without alerting viral sensors .
EV research is still a developing field . Functional studies are dependent on consistent and pure EV preparations . Hence , a carefully controlled purification strategy was established and validated . EV from a KSHV-negative B-cell lymphoma ( BJAB ) and a KSHV-positive primary effusion lymphoma ( PEL ) cell line ( BCBL1 ) were compared to each other and to EV from primary KSHV-tumor effusions and normal plasma . To allow for large scale EV purification , an initial concentration step using polyethylene glycol ( PEG ) was used [13] . No discernable size or concentration differences were observed between EV isolated from BJAB and BCBL1 cell supernatant . The mean and mode sizes were <100 nm ( Fig 1A and 1B ) , the particle concentrations were similar ( Fig 1C ) , and acetylcholine esterase ( AchE ) activity was comparable across all preparations ( Fig 1D ) . The biophysical properties of EV from PEL were similar to those of the Epstein-Barr virus ( EBV ) -negative B cell lymphoma BJAB , and the EBV-positive B cell lymphoma cell line Namalwa ( S1 Fig ) . Following established standards [22] the EV markers CD63 , CD81 , CD9 , and Flotillin-2 were evaluated by Western blot ( Fig 1E ) . All proteins were present at comparable levels in EV and none were detected in EV-depleted media . The KSHV latency-associated nuclear antigen ( LANA ) was present in BCBL1 lysate , but not BJAB lysate and not in EV . GAPDH was present in whole cell lysates , but not EV . MS/MS confirmed the presence of additional , prototypical EV markers , but no viral proteins . There was no difference in EV marker protein content as defined by exocarta . org among EV from BJAB vs BCBL1 ( S2 Fig ) . KSHV-encoded miRNAs are incorporated at high concentrations into EV and provide a high sensitivity marker to trace BCBL1 derived EV [7] . The miRK-12-5 was present in BCBL1 cells and BCBL1-derived EV , but not BJAB cells or BJAB-derived EV ( Fig 1F ) . Negative staining electron microscopy ( EM ) of EV showed small , rounded vesicles of ~40–70 nm diameter ( Fig 1G ) . After PEG enrichment , EV were further purified by either ultracentrifugation or column purification . Both procedures yielded comparable results ( S3 Fig and S4 Fig ) . In sum , this experimental approach resulted in clean and concentrated EV preparations . To link the cell line studies to primary patient material , EV from healthy donor ( HD ) plasma and primary PEL fluid were isolated . The EV exhibited the same characteristic size distribution across different donors ( S5 Fig ) . Their concentrations were at times higher than from culture supernatant . AchE activities , likewise , were comparable . These experiments demonstrate that EV isolated by this method from virus-negative and virus-positive cell supernatants were biophysically and biochemically similar to each other and to normal plasma or primary PEL fluid . To exclude the presence of virion particles in the EV preparation , EV were subsequently affinity-purified using antibody conjugated beads . Recapitulating our prior results [7] , this step retained EV-associated proteins ( S6A Fig ) and miRNAs ( S6B Fig ) while removing KSHV particles , as measured by DNA content ( S6C Fig ) . NTA analysis of the affinity-purified EV from BJAB , BCBL1 PEL , HD , and primary PEL yielded consistent characteristics across ( S6D–S6F Fig ) , and EM detected the presence of cup-like vesicles typically of EV , but no evidence of virion particles ( S6G and S6H Fig ) . Through this method , we recovered all but one of the KSHV-encoded miRNAs in the CD63+ fraction of EV from BCBL1 cells . Neither the viral transcript LANA nor the cellular encoded GAPDH mRNA were present , demonstrating enrichment for the viral miRNAs , with primary PEL CD63+ EV serving as our positive control ( Fig 2A ) . To show that the samples contained similar fractions of CD63+ EV , the EV were incubated with Dil , a membrane-intercalating dye , ExoGreen , an internal esterase-substrate , or both , bound to antibody-coated beads and subjected to flow cytometry ( S7 Fig ) . Results were consistent between CD63 ( Fig 2B ) , CD9 ( Fig 2C and S8 Fig ) , and CD81 beads ( Fig 2D and S9 Fig ) . This demonstrated that PEL derived EV carry at least three tetraspanin markers ( CD63 , CD81 , CD9 ) , which are suitable for positive affinity purification and that our three-step approach ( PEG > column > affinity-bead ) yielded a highly purified , virus and DNA-free , EV fraction . The principal target of transformation by KSHV are endothelial cells ( EC , reviewed in [3] ) . EC are also exposed to the highest concentration ( ~ 1011/mL ) of EV via lymphatic and blood vessels and thus can be considered a bona-fide target of systemically circulating EV [23] . For these reasons , hTERT-immortalized human umbilical vein endothelial cells ( hTERT-HUVEC ) were used to investigate EV uptake . To establish that the purified EV were endocytic , 1010 CD63+ EV were labeled with Dil ( 1 , 1’-dioctadecyl-3 , 3 , 3’3’-tetramethylindocarbocyanine perchlorate , red ) and ExoGreen ( green ) and added to 105 hTERT-HUVECs ( MOI = 105 ) and after 12 hours , fixed and analyzed by fluorescence microscopy . Uptake was equivalent for EV from all sources: BJAB , BCBL1 , HD , and primary effusion fluid ( Fig 3 ) . The red and green labels co-localized in the cytoplasm . Not every incidence of red and green EV labels co-localized after cells were exposed for 12 hours . This was likely due to differences in the recycling nature of the biomolecules stained ( lipids vs . proteins ) . As the time course was extended ( S10 Fig and S11 Fig ) the lipid dye Dil redistributed throughout the cell , in contrast to the ExoGreen protein dye which remained in punctate structures . EV uptake was blocked by Annexin-V , but not heparin sulfate at a concentration , which reliably blocks KSHV entry [24] ( Fig 4 ) . This establishes that , after extensive purification , the EV retained endocytic activity , which was dependent on phosphatidylserine ( PS ) but not heparin . Henceforth , this material is referred to as KSHV-EV . To test the hypothesis that KSHV-EV act as chemoattractants and paracrine growth factors for EC , cell migration was monitored continuously using the xCelligence system . The xCelligence system allowed us to evaluate a number of physiological phenotypes that are associated with EV reprogramming . First , KSHV-EV served as a chemoattractant for EC migration , whereas EV from human donor blood ( HD ) did not . This chemoattractant property was phenocopied by EV isolated from primary PEL fluid ( Primary PEL EV ) ( Fig 5A and 5B ) . Second , we asked if synergy existed between KSHV-EV and KS-relevant cytokines ( VEGF , IL-6 , PDGF-β , or SDF1-α ) . The hTERT-HUVECs were exposed to EV and plated into the top chamber and monitored for migration into the bottom chamber , which contained the cytokine in serum-free media . KSHV-EV significantly enhanced cell migration in response VEGF and IL-6 , but not PDGF-β or SDF1-α ( S12 Fig ) . Third , a scratch assay was performed . The hTERT-HUVECs were grown in the presence of EV , the confluent monolayer disrupted , and closure monitored over time . KSHV-EV and primary PEL EV enhanced migration , EV from HD did not ( Fig 5C ) . The positive control , VEGF , alone or added to EV from HD yielded the same degree of scratch closures as induced by KSHV-EV ( Fig 5D and 5E ) . Fourth , to test the hypothesis that EV induced cytokines , which could amplify or mediate the migration phenotype , supernatants were analyzed for IL-6 , IL-10 , which are implicated in KS biology , as well as the immune response cytokines IL-18 , IL-1β , and interferon alpha ( IFN-α ) . Only IL-6 was significantly induced in response to KSHV-EV and primary PEL EV ( Fig 5F and S13 Fig ) . These experiments establish that KSHV-EV , but not EV circulating in healthy patients contributes to the pathophysiology of KS by modulating EC function and inducing human IL-6 . To test whether KSHV-EV elicited an innate immune response in EC , a number of known innate immune signaling pathways were examined . On the one hand , such a phenotype would be expected as PEL and KS represent an inflammatory microenvironment ( reviewed in [3] ) ; on the other hand , pro- as well as anti-inflammatory phenotypes have been reported for infected-cell derived EV from different viruses ( reviewed [1] ) . The innate immune response involves membrane-bound receptors , such as toll-like receptors ( TLRs ) , as well as cytoplasmic RIG-like receptors , such as RIG-I and MDR-5 . Both pathways ultimately converge onto interferon regulatory factor 3 ( IRF3 ) and nuclear factor kappa B ( NF-κB ) . IRF3 and NF-κB are normally sequestered in the cytoplasm . Upon stimulation , they translocate to the nucleus . KSHV-EV , or primary PEL EV did not induce nuclear translocation of IRF3 ( Fig 6A and 6B ) and did not induce IRF3 phosphorylation ( Fig 6C ) . This was in contrast to infection with West Nile Virus or stimulation with Polyinosinic:polycytidylic acid ( PolyI:C ) . KSHV-EV did not inhibit the phosphorylation of IRF3 in response to PolyI:C ( Fig 6D ) , suggesting the KSHV EV did not actively inhibit signaling . KSHV-EV , or primary PEL EV did not induce nuclear translocation of NF-κB either ( S14 Fig ) . Likewise , targeted transcriptional profiling of ninety NF-κB-regulated genes showed no response to KSHV-EV ( S15 Fig ) . The IRF3 and NF-κB transcription factors represent the endpoint of a multitude of RNA sensing pathways . As neither was activated , it is unlikely that any of the upstream receptors ( TLR , RLR ) were activated to the degree that authentic viral infection would . KSHV is a DNA virus , and its reactivation from latency is curbed by cGAS/STING [25] . To test for the induction of cGAS-STING signaling by KSHV-EV , we monitored the induction of interferon-beta ( IFN-β ) . KSHV-EV did not change IFN-β transcript levels , and KSHV-EV did not modulate the cGAS/STING response to Interferon Stimulatory DNA ( ISD ) or Poly I:C ( Fig 7A ) . As a second , independent measure of cGAS/STING activation we measured TANK binding kinase ( TBK ) . TANK is phosphorylated upon recognition of cytosolic nucleic acids in a MAVS dependent manner , and physically interacts with STING . KSHV-EV did not affect phospho-TBK levels alone or in conjunction with the positive inducers ( Fig 7B and 7C ) . In sum , neither TLR , RIG-I , nor cGAS/STING pathways were activated by KSHV-EV; at least under these conditions KSHV-EV did not inhibit the activation of these pathways by physiological triggers either . To identify molecular pathways that could explain KSHV-EV induced endothelial cell migration , we explored ERK1/2 signaling . ERK1/2 has been implicated in IL-6 signaling as well as EC migration [26 , 27] . Treatment of hTERT-HUVEC with KSHV-EV and primary PEL EV , but not HD EV induced ERK1/2 phosphorylation ( p-ERK1/2 ) ( Fig 8A ) . The EV themselves did not contain p-ERK1/2 ( Fig 8B ) . To exclude the possibility that IL-6 induced ERK1/2 phosphorylation as part of a secondary feedback loop , the experiment was repeated in the presence of antagonistic anti-IL-6 receptor antibodies . Despite blocking IL-6 signaling , ERK1/2 became phosphorylated upon KSHV-EV exposure ( Fig 8C ) . Pre-incubation with Annexin-V , which blocks EV adsorption , significantly reduced p-ERK1/2 levels ( Fig 8C ) . This result is consistent with the notion that ERK activation was a direct result of EV exposure . ERK1/2 is phosphorylated by MEK , which can be targeted pharmacologically . To test the hypothesis that MEK kinase activity was required for p-ERK1/2 in response to KSHV-EV and primary PEL EV , we used AZD6244 . Pre-treatment of cells with AZD6244 blocked ERK1/2 activation by primary PEL EV relative to DMSO control ( Fig 8D ) . AZD6244 also reduced primary PEL EV- and KSHV-EV-dependent cell migration ( Fig 8E and 8F ) . To ensure that AZD6244 did not exert off-target effects , we repeated the cell migration assay with a different MEK inhibitor , PD184352 . Treatment with PD184352 antagonized the enhanced cell migration phenotype of hTERT-HUVECs treated with KSHV-EV ( S16 Fig ) . These experiments demonstrate that KSHV-EV activate the Ras/Raf/MEK/ERK pathway , which leads to EC activation , proliferation , and migration . In PEL and KS patients , KSHV-EV are continually released into the microenvironment and systemically into the blood and lymphatic circulation . Next to hemangioma , KS is the most angiogenic cancer in humans ( reviewed in [3] ) . PEL grow as effusions , bathing the cavity walls in KSHV-EV . Thus , a physiological relevant experimental design would expose EC repeatedly to KSHV-EV . Such a design measures long-term cellular reprogramming ( Fig 9A ) . Here , hTERT-HUVECs were exposed to KSHV-EV or BJAB-derived EV over a period of 12 days . Every 24 hours the media was replenished with fresh KSHV-EV or control-tumor EV , and cellular programming was analyzed by RNAseq . The time course was divided into an acute phase ( day 2 and day 4 ) , an intermediate phase ( day 6 and day 8 ) , and a chronic phase ( day 10 and 12 ) . KSHV-EV induced synchronized , progressive , and directional transcription profile changes compared to control ( Fig 9B , see S17 Fig for a principal component analysis of hierarchical clustering ) . To identify continuous transcript changes over time , a likelihood ratio test was used . This identified 67 transcripts that were significantly upregulated and 84 that were significantly downregulated genes over the course of treatment ( Fig 9C , and S2 Table ) . Heatmap representation of these altered genes show distinct contrasts between the treatment groups ( Fig 9D–left ) , which were maintained over the 12 days of KSHV-EV exposure . To test the hypothesis that KSHV-EV mediate any or all of the transcriptionally changes hitherto ascribed to KSHV miRNA expression within the infected cell , we analyzed transcriptional changes in genes that were previously identified in KSHV-infected EC [6 , 28 , 29] . Changes in these particular sets of mRNAs signal BEC to LEC transcription upon direct infection of KSHV . These mRNAs remained largely unchanged ( S18 Fig ) . The notable exception was MAF1 mRNA , which was previously shown to be a direct target of multiple KSHV miRNAs with eleven predicted target sites in the 3’UTR [6] , which was robustly down-regulated . As a control we analyzed a predefined set of interferon stimulatory genes ( ISGs ) [30] ( Fig 9D , right ) . These showed induction during the acute phase for a few genes; however , these did not differ between control and KSHV-EV . This induction was not maintained over time , consistent with the idea that KSHV-EV reprogram EC towards a proliferative , activated phenotype , which is different from phenotypic changes induced by inflammation . Time course analysis allowed for the identification of patterns of transcriptional changes for individual genes . The minimal set of EV-exposure biomarkers was comprised of genes with the highest and most consistent mRNA changes over time ( Fig 10A and 10B ) . Among them were CD9 and JUNB , which we chose to validate at the level of protein expression ( Fig 10C ) . CD9 protein levels were greatly reduced in KSHV-EV treated cells , particularly in the intermediate and chronic time points . As an internal control , we monitored the protein levels of a separate EV protein Tsg101 , which remained constant in both treatment groups over time . JUNB levels were very low in the hTERT-HUVEC cells to begin with and further reduced at the intermediate time frame in the KSHV-EV treatment group . Thus , the transcriptional changes in response to KSHV-EV translated into protein level changes and physiological changes . The advantage of genome-wide transcriptional profiling lies in the identification of signaling networks , rather than individual genes . Hence , the significantly altered genes were clustered into gene ontology ( GO ) pathways [31] . The preeminent pathways identified herein related to extracellular matrix modulation , cell adhesion , growth and migration ( Fig 11 ) . Next , we explored phenotypic changes that would be consistent with the transcriptional pathways that dominated GO analysis . These are summarized in S3 Table . Experiments showing increased cell growth and migration ( and thus decreased adhesion ) in response to KSHV-EV , but not control-EV were already noted above . CD9 is a tetraspanin , which is involved in cell adhesion , motility and junctional integrity . It is also involved in EV biogenesis . To test whether the KSHV-EV-induced downregulation of CD9 at the mRNA and protein level led to a reduction in EV secretion in the recipient cells , we measured EV in the supernatant of hTERT-HUVEC cells at 24 hours after treatment with KSHV-EV . This experiment was possible because earlier studies had shown that exogenously added EV , analogous to liposomal transfection , are taken up within a few hours after addition to media [12 , 32] . There were no changes in the biophysical characteristics of the hTERT-HUVEC-derived EV , but the total amount was reduced by comparison to control ( Fig 12 and S18 Fig ) . To test the hypothesis that KSHV-EV induce some , but not all phenotypes as KSHV infection , morphological differences in the recipient hTERT-HUVEC were evaluated . HTERT-HUVEC were exposed to KSHV-EV , mock , or BJAB-derived , control EV for four consecutive days . Chronically infected HUVECs served as a control . KSHV-EV treatment did not alter Tubulin ( Fig 13A ) or Actin ( Fig 13B ) organization . KSHV LANA was present in infected , but not EV-treated cells ( Fig 13C ) . By contrast , the proliferation marker Ki-67 was dramatically induced by KSHV-EV , but not control EV . Ki-67 positivity was similar to KSHV infected cells ( Fig 13D and 13E ) . H&E stain revealed a greater cell density in KSHV-EV compared to mock or BJAB-EV treated cells ( S20 Fig ) . KSHV-infected hTERT cells exhibited greatly increased cell size , as previously described and consistent with mTOR/S6K activation [33] . Overall these results mirror the dramatic dysregulation of infected as well as uninfected EC in KS lesions , where the normal vasculature and extracellular environment is essentially destroyed and slit-like empty spaces develop . This analysis demonstrated that KSHV-EV inducing a long-lasting reprogramming of EC , which results in transcription signatures and pathway alterations consistent with the phenotypic changes observed in KS lesions .
KS is an incredibly angiogenic cancer , second only to hemangioma [3] . It is driven by KSHV-infected EC and defined by a unique molecular mechanism that manifests itself in aberrant EC behavior . Many studies have focused on the cell autonomous roles of KSHV [6 , 28 , 29 , 33–38] . In addition , studies by Mesri and others ( reviewed in [39] ) have established that the KS phenotype depends to a large degree on paracrine signaling mechanisms to reprogram neighboring uninfected EC . This report establishes that KSHV-EV mediate some of the paracrine phenotypes of KS ( summarized in S3 Table ) . EV mediate a large variety of phenotypes in the immediate microenvironment as well as at distant sites . EV have established roles in cell differentiation , angiogenesis , cell migration as well as metastasis [40–44] . We had shown earlier that KSHV miRNAs are present in systemically circulating EV in KS patients , PEL fluid as well as in transgenic mice , which carry the KSHV miRNAs , but are not competent to make virions [7] . PEL are of post-GC lineage lymphoma , approaching almost plasmablastic stage . They grow i . p . ( in body cavities ) , unlike Burkitt lymphoma , which are also post-GC lymphoma , but pre-plasmablastic and grow as a solid mass in lymph nodes , not as an effusion . This may explain the prominence that EV have in the biology of PEL and KSHV vis-a-vis other tumors . Crucial to the study of EV is a well-validated purification pipeline [22 , 45] . In the context of virus infections , it is important to exclude viral particles , which tend to co-purify with EV in ultracentrifugation , crowding-agent , and size exclusion chromatography approaches . Hence , we added affinity purification using antibody-coated beads directed against CD63 or other tetraspanins as the final purification step . This step depletes virions to below the limit of detection [7 , 12] , as it positively retains EV on a column rather than collecting a precipitate . It also reduces the complexity of the EV populations [10 , 46] to only those EV that are of narrow size , tetraspanin-positive , and inhibited by Annexin-V . Adding an RNAse and DNAse step as well as size exclusion chromatography eliminated contaminating free RNA and DNA and selected against vesicles that are released non-specifically by dying cells . This has not always been done and may explain reports of EV preparations that induce heavy DNA and RNA-dependent immune stimulation in recipient cells . We believe the final product of our purification pipeline represents biologically-relevant KSHV-EV at or slightly below physiological concentration . It has been a matter of debate as to whether EV induce or suppress the innate immune response . This phenotype depends largely on the specific virus and target cell . Professional immune cells , such as dendritic cells and macrophages are known to receive and transmit pro-inflammatory signals through EV [47–49] . RNA viruses induce a dramatic innate response and large amounts of secondary messengers , such as IFN-β or cGAMP [48] . DNA viruses , such as herpesviruses may also transmit pro-inflammatory signals through EV , which can be sensed by professional antigen presenting cells [47 , 50] . While we cannot exclude that lytically replicating cells or professional antigen presenting cells infected with KSHV behave differently , these experiments demonstrate that EC , the primary target of KSHV infection , do not become activated by EV from KSHV-infected lymphoma cells or by EV from primary PEL fluid . They may become activated by cytokines or small soluble molecules , though a non-activated phenotype of uninfected cells would also be consistent with the biology of KSHV as most primary KSHV infections are clinically asymptomatic and not associated with mononucleosis-like symptoms or autoimmunity as seen with EBV or human cytomegalovirus infection . These results support a model whereby cellular proteins , cellular and viral miRNAs that are carried in KSHV-EV modulate long-term reprogramming of EC in the immediate microenvironment of the tumor and systemically in the human host . Whereas prior studies focused on the immediate effects of a single bolus of EV , these experiments were designed to mimic continuous KSHV-EV exposure as seen in KS and PEL patients or patients with a high latent virus burden . The concentration of EV in normal blood is ~1010–1011/mL [23] , which is ~ 6 orders of magnitude higher than the median concentration of KSHV in the blood of symptomatic , untreated AIDS-KS patients [51] . We added 1010/mL to 106 cells ( MOI = 10 , 000 ) . Based on our prior work and Fig 3 , we assume that all EV are endocytic-competent and all KSHV-EV carry the KSHV miRNAs . EV adsorption plateaus within hours of exposure [32] . Using a MOI bolus of 4000 vs . 1000 is unlikely to result in a qualitatively different response after 4 days . The novelty of this experimental design is mimicking chronic exposure , as is the case in the KS microenvironment or for any endothelial cell lining the blood vessels . Whereas PEL and KS , in the context of uncontrolled progression to AIDS , are rapidly fatal , de novo KSHV infection per se is not . HIV-negative endemic , pediatric and classic KS have rapid as well as smoldering clinical progression [52] . KSHV-associated multicentric Castleman’s disease has a waxing and waning presentation , closely associated with IL-6 levels [53] . IL-6 , IL-10 , VEGF , PDGF and other inflammatory cytokines are elevated in PEL , KS and MCD , and agents , such as pomalidomide , rapamycin and tocilizumab which modulate their levels , modulate disease [54–56] . KSHV-EV consistently induced human IL-6 in uninfected cells . These experiments show that in addition to cytokines EV also transmit pro-growth signals and can reprogram EC . KSHV-EV induced a much more long-lasting phenotype than acute phase cytokines , which mimics differentiation and trans-differentiation . Whereas it was difficult to identify a single master regulator of this trans-differentiation phenotype , network analysis showed significant changes of transcriptional modules that regulate extracellular matrix remodeling , translation and exosome biogenesis . By comparison , IFN-β and NF-κB transcriptional networks were unaffected . KSHV-EV signaled through MAPK/ERK , which is consistent with MAPK/ERK’s role in modulating EC motility and vascular behavior ( reviewed in [57] ) . Our observations are consistent with a recent study by Yogev et al . [58] , who showed that KSHV-EV ( derived from infected EC ) induce metabolic remodeling of nearby uninfected cells . This represents perhaps the initiating step of trans-differentiation . Afterwards , continued KSHV-EV exposure resulted in continued reprogramming as has been described for KSHV-infected and KSHV-miRNA transfected EC [4–6 , 34] and these experiments verified that most KSHV-miRNAs are present in KSHV-EV . Reprogramming here is used to defined an altered state of gene transcription and cell lineage , such as published by Hansen et al . [6] , upon transfection of the KSHV miRNAs into EC , or upon infection of EC with KSHV [4 , 28 , 29 , 59–61] . At this point we do not know if this reprogramming will persist after EV exposure has subsided and if not , how quickly the cells return to their normal state . Clinically , KS lesions and KS-associated edema regress as KSHV is cleared by immune restoration upon cART or lowering of immunosuppressive drugs in the context of transplant KS . Hence , we speculate that that the KSHV-EV induced phenotype likewise is transient . This would be in contrast to permanent lineage reprogramming , which is most commonly associated with epigenetic changes to the cellular DNA . Evidence for KSHV-infection induced chromatin remodeling has been published [62–65] . If in addition to transcriptional reprogramming , the viral miRNAs ( or other molecules ) that are contained in KSHV-EV also alter chromatin accessibility stably and irreversibly is a fascinating hypothesis and the subject of future studies . The gradual and long-lasting reprogramming of transcriptional networks is consistent with the mechanism of action for miRNAs , which have their most physiological impact in development rather than acute signaling . EBV and KSHV express miRNAs and in infected cells these miRNAs account for as much as 50% of the miRNA pool . KSHV and EBV incorporate the viral miRNAs into EV [7 , 66–68] . Both viruses substantially modulate the protein composition of EV [69] . In addition , EBV incorporates the LMP-1 oncogene into EV [70 , 71] , whereas no KSHV proteins were detected in EV thus far . This phenotype is consistent with the idea that EV are pivotal for establishing local tissue homeostasis and provides a molecular mechanism for it . Further studies are needed , but for the first time there now exists a highly reproducible , physiologically relevant experimental design to study long-term EV-EC interactions . In conclusion , our findings point toward a novel means of cellular reprogramming by viruses . They pinpoint novel , actionable pathways for intervention and biomarker development . KSHV is able to infect EC , but the larger importance of EV stems from the fact that these vesicles can carry viral components to distant locations and transfer them into cells that the virus cannot enter . This may explain some of the phenotypes that viruses , including HIV , have on uninfected cells and it may explain why clinical sequalae persist long after the virus has been cleared or entered molecular latency .
All cells were grown at 37°C in 5% CO2 . hTERT-immortalized HUVECs were cultured in endothelial growth medium ( EGM-2 media; Lonza ) supplemented with the EGM-2 Bulletkit ( Lonza ) and 10% EV-depleted fetal bovine serum ( FBS ) . BCBL1 ( from the laboratory of Dr . D . Gamen ) cells were cultured in RPMI 1640 ( Gibco ) supplemented with 10% Tetracycline-free , EV-depleted FBS ( Clontech ) , 100 U/mL penicillin G ( Gibco ) , 100 μg/mL streptomycin sulfate ( Gibco ) and 2 mM L-glutamine ( Gibco ) . BJAB ( from the laboratory of Dr . D . Gamen ) cells were cultured in RPMI 1640 supplemented with 10% EV-depleted FBS , 100 U/mL penicillin , 100 μg/mL streptomycin sulfate , and 2 mM L-glutamine . Namalwa ( EBV-positive , obtained from the ATCC #CRL-1432 ) were grown in the same conditions as BCBL1 cells . Total EV were isolated using approximately 400 mL of cell culture supernatant . Cells were pelleted at 4˚C at 800x g for 10 minutes . Supernatant was then passed through a 0 . 22 μm Nalgene Rapid Flow Filter ( Thermo Fisher ) . Filtered supernatants were aliquoted into individual 50 mL conical tubes ( Corning ) . EV were precipitated with 40 mg/mL PEG-8000 and incubation at 4˚C for >8 hours . Precipitates were then spun down at 4˚C at 1 , 200x g for 60 minutes . Pellets were resuspended in 500 μL of ice-cold 1X PBS ( Gibco ) . Removal of non-associated molecules were done by ( i ) ultracentrifugation or ( ii ) column chromatography . ( i ) For ultracentrifugation , the volume was increased to ~4 mL with 1X PBS and centrifuged at 4˚C at 120 , 000x g for 60 minutes using a Beckmann SW32 rotor . The pellet was then resuspended in 4 mL of fresh 1X PBS and centrifuged again . A total number of three washes was done . The final pellet was resuspended in 100 μL of fresh , ice-cold 1X PBS . ( ii ) For column chromatography , GE Sephadex G-200 was equilibrated with ice-cold 1X PBS for a total of 4 compacted bead volumes ( 4 mL ) . The resuspended EV were added to the equilibrated column and allowed to flow through the column by gravity . EV were collected in the first 1 mL of fresh , cold 1X PBS . EV were also isolated from isolated from 50 mL plasma from health donors or 10 mL PEL . Briefly , blood was processed and erythrocytes , leukocytes , platelets , and plasma were separated using Ficoll reagent ( GE 17-1440-02 ) as above . Samples were enriched for CD63 , CD9 , and CD81 positive EV using magnetic beads ( ThermoFisher 1060D , 10620D , and 10622D , respectively ) . Briefly , the total EV isolated as above were added to 80 μL of equilibrated , antibody-coated magnetic beads . Non-specific IgG-coated beads were used as a control . EV were bound to beads overnight at 4˚C and beads were washed 3X with 1X PBS . EV were eluted in 100 μL of elution buffer ( Invitrogen ) or 0 . 2 M Glycine pH = 2 . 0 for further analysis . Cell were authenticated by targeted amplification of STR typing loci using Ion Torrent Precision ID GlobalFiler NGS STR Panel and compared against the STR database of the German Collection of Microorganisms and Cell Cultures GmbH . Total EV were labeled with 1 μM Dil ( 1 , 1’-dioctadecyl-3 , 3 , 3’3’-tetramethylindocarbocyanine perchlorate; Sigma ) , 0 . 5X ExoGreen ( SystemBio ) , followed by G25 column filtration and incubated with 40 μL CD63 , CD9 , or CD81 beads , washed 3x with PBS and analyzed using the BD Accuri C6 Plus flow cytometer ( BD Biosciences ) . FITC and PE settings were used to detect ExoGreen and Dil with an excitation laser of 488 nm and emission filters of 533 nm and 585 nm , respectively . Unlabeled EV were used to set background fluorescence . Results were analyzed using FloJo 2 . 0 . hTERT-HUVECs were grown on a cover slip in 6 well plates in a total volume of 3 mL and treated with labeled 109 EV/mL for the indicated time period at 37°C . Cells were then rinsed with PBS and fixed in 4% paraformaldehyde for 10 minutes at RT , washed 3 times with PBS and the cells permeabilized using 0 . 5% Triton X-100 in PBS for 10 minutes and washed 3x with PBS . For indirect immunofluorescence , coverslips were blocked in a solution of 10% goat serum ( Vector Labs ) in PBS with 0 . 2% Triton X-100 and incubated with primary antibodies: anti-IRF3 antibody ( Cell Signaling , #4962 , 1:100 dilution ) anti-P-NF-κB p65 ( S536 ) ( clone 93H1 , Cell Signaling , 1:100 dilution ) . Coverslips were washed three times with PBS-0 . 2% Triton with 2% BSA and incubated with FITC-conjugated anti-rabbit secondary antibody ( #FI-1000 , Vector Labs Inc . 1:500 dilution ) . Cells were washed three times with PBS-0 . 2% Triton X-100 with 2% BSA and stained with 0 . 2 μg/mL DAPI ( Sigma ) prior to mounting in VectaShield ( Vector Labs ) . Cells were imaged on a Leica DM4000B microscope with a Q-Imaging Retiga-2000RV camera and HCX-PL-APO 506187 lens at 63x magnification . De-convoluted images ( Simple PCI 6 software Metamorph v 7 . 8 . 12 . 0 , 10 iterations RB , GB or RGB ) were then opened in Imaris V 9 . 2 . 0 . and background subtraction of all channels was done using recommended settings of 400 um filter width . Localizations of EV-delivered Dil and ExoGreen were done using the “Add Spots” command using spots of different sizes depending on fluorescence intensity . Regions of spot calling were standardized to linear detection ranges using absolute intensity . For nuclei staining , the “Add New Surfaces” command was used . As a positive control for IRF3 activation , hTERT-HUVECs were infected with West Nile Virus NY99 ( WNV ) at a MOI of 5 ( observed after 36 hours ) or Poly I:C at 5 μg/mL ( observed after 12 hours ) . Pellets ( 1010 EV or 106 cells ) were lysed in 100 μL NP-40 lysis buffer and run on an 8% SDS-PAGE gel , transferred to a nitrocellulose membrane ( Hybond ) and blocked in 5% dry milk in TBS overnight at 4°C . Antibodies are listed in S1 Table . For detection of tetraspanins , non-reducing conditions were used . To visualize total protein by silver stain , we used the Pierce Silver Stain Kit ( ThermoFisher ) after which bands were excised and analyzed by mass spectrometry at the UT Southwester core ( https://www . utsouthwestern . edu/research/core-facilities/proteomics-core . html ) . To block EV entry , EV were incubated with recombinant Annexin-V ( 2 μg/mL , BD Biosciences ) for 30 minutes prior to addition to hTERT-HUVECs . To block virus entry EV were incubated with 50 μg/mL heparin ( Lonza ) . To block autocrine IL-6 feedback , 10 ng/mL IL-6 receptor ( sIL-6R , Peprotech #200-06R ) was added to cells 24 hours before EV addition . To inhibit MEK and ERK1/2 , AZD6244 ( SelleckChem ) and PD184352 ( SelleckChem ) , respectively were used to treat cells 24 hours prior to addition EV addition at indicated concentrations . ( i ) Scratch assays were performed as previously described [7] . Briefly , hTERT-HUVECs were grown in a 24-well plate ( Corning ) prior to treatment with EV . The wound was initiated using a standard 200 μL pipette tip and the cells were then washed and replaced with fresh media containing one of the following as a chemo-attractant: 10% FBS , 10 ng/mL VEGF ( Peprotech ) , 1 U/mL hIL-6 ( Peprotech ) , 10 ng/mL PDGF-β ( Peprotech ) or 10 ng/mL SDF-1α ( CXCL12 ) ( R&D Systems ) . The culture was monitored over time . Images were obtained using a Leica DMIL microscope with a HI Plan 10x/0 . 25 PHI objective and QImaging camera ( Cooled color , RTV 10 bit ) paired with QCapture imaging software 3 . 0 . Images are shown at 100x magnification and were analyzed using ImageJ software to calculate the percent wound closure at a given time point . ( ii ) Cell proliferation and migration was analyzed using the xCelligence RTCA DP instrument as previously described [7] . Briefly , hTERT-HUVECs were treated with EV for 24 hours and proliferation measured by conductance . For migration , both sides of the xCelligence CIM Plate 16 ( Acea Biosciences ) plate membrane were coated with 20 μg/mL fibronectin prior to assembly and media containing FBS or a specified cytokine was placed in the lower chamber as the chemo-attractant . Cells were plated at 15 , 000 cells per well of the upper chamber . Reads were taken every 2 minutes for a period of 12 hours . The cell index reflects the degree of cellular migration towards the specified chemo-attractant . Levels of IL-6 , IL-8 , IL-10 , IL-18 and IL-1β were determined by ELISA according to the manufacturer’s protocol ( eBioscience , #88–7066 ( IL-6 ) , #88–8086 ( IL-8 ) , #88–7106 ( IL-10 ) , #BMS267INST ( IL-18 ) and #88–7010 ( IL-1β ) . The average of at least three wells is reported for each biological replicate . Total RNA was isolated using TRI reagent ( Molecular Research Center ) as previously described ( https://www . med . unc . edu/vironomics/services/protocols/ ) , treated with Turbo DNA-free kit ( Ambion , Life Technologies ) and 100 ng of DNA-free RNA , as determined by Nanodrop , was used as input for High Capacity cDNA synthesis kit ( Applied Biosystems , Life Technologies ) . Custom NF-κB and endothelial lineage real-time qPCR arrays were used previously published [72] . EV were adsorbed on a glow-charged carbon coated 400-mesh copper grids for 2 minutes and then stained with 2% ( weight/volume ) uranyl acetate in water . Transmission electron microscopy ( TEM ) images were taken using a Philips CM12 electron microscope at 80 kilovolts . Images were captured on a Gatan Orius camera ( 2000 x 2000 pixels ) using the Digital Micrograph software ( Gatan , Pleasanton , CA ) . Images were then cropped in Adobe Photoshop . ( i ) For continuous , variable pairwise T-tests were performed to determine statistical significance for n ≥ 3 biological replicates . ( ii ) For comparison of mass spectrometry data , a hypergeometric test was used . ( iii ) For analysis of RNAseq data we used a custom pipeline . The decision to use our particular analysis is discussed at length at https://support . bioconductor . org/p/62684/ . In brief , STAR-Aligned BAM files representing table of counts for each samples were processed using DESeq and other Bioconductor packages: https://bioconductor . org/packages/devel/bioc/vignettes/GenomicAlignments/inst/doc/summarizeOverlaps . pdf . | The role of extracellular vesicles ( EV ) has received considerable attention in recent years . The contents of these cell-derived vesicles have been shown to be modulated upon challenge by a virus or neoplastic transformation , and can influence the behavior of recipient cells . Here we demonstrate that purified EV from an AIDS-associated , virus-driven lymphoma induces unique cellular signaling , motility , and gene expression reprogramming in recipient endothelial cells . This was accomplished without activation of innate immune activation , even after prolonged exposure to these EV . Collectively our results point toward a model in which tumor-derived EV condition neighboring cell physiology , while avoiding detection by immune regulators . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"flow",
"cytometry",
"phosphorylation",
"cell",
"motility",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"natural",
"antisense",
"transcripts",
"gene",
"regulati... | 2019 | Extracellular vesicles from Kaposi Sarcoma-associated herpesvirus lymphoma induce long-term endothelial cell reprogramming |
The goal of training is to produce learning for a range of activities that are typically more general than the training task itself . Despite a century of research , predicting the scope of learning from the content of training has proven extremely difficult , with the same task producing narrowly focused learning strategies in some cases and broadly scoped learning strategies in others . Here we test the hypothesis that human subjects will prefer a decision strategy that maximizes performance and reduces uncertainty given the demands of the training task and that the strategy chosen will then predict the extent to which learning is transferable . To test this hypothesis , we trained subjects on a moving dot extrapolation task that makes distinct predictions for two types of learning strategy: a narrow model-free strategy that learns an input-output mapping for training stimuli , and a general model-based strategy that utilizes humans' default predictive model for a class of trajectories . When the number of distinct training trajectories is low , we predict better performance for the mapping strategy , but as the number increases , a predictive model is increasingly favored . Consonant with predictions , subject extrapolations for test trajectories were consistent with using a mapping strategy when trained on a small number of training trajectories and a predictive model when trained on a larger number . The general framework developed here can thus be useful both in interpreting previous patterns of task-specific versus task-general learning , as well as in building future training paradigms with certain desired outcomes .
One of the core problems in learning is determining the range of tasks and circumstances that a training paradigm will impact . Training can produce both learning that generalizes to new tasks and circumstances and learning that is restricted to the exact training conditions . The difficulty is that there are many paths to good performance in a given task – from more demanding routes in which extensive knowledge is acquired , to special purpose shortcuts that allow good performance with restricted knowledge . Without knowing which of the myriad possible approaches the subject will take during learning , there is no way to predict the generality of the eventual learning . Even mundane tasks like learning when it is safe to cross the street have both narrow and general solutions . A general , and more difficult path , requires learning internal models for car and subject motion that can be used to look-ahead and predict future locations . To avoid being hit , you need to estimate the distance , speed , and direction of nearby cars and then to mentally simulate the cars' path through time . A similar simulation must also be run forward for your own progression across the street . The two simulations must be merged to determine if you are likely to be occupying the same physical space as a car at the same point in time ( which we do not advise ) . While in general this works for any configuration of cars and pedestrians , such general strategies usually come with performance costs . Indeed , the sheer number of pedestrians struck by cars suggests that this type of look ahead is indeed both cognitively demanding and subject to error [1] . Even when not hit by a car , prediction errors can make us hurry a bit more than expected . Such errors are simply an intrinsic part of every task that requires look-ahead . Each time a prediction is made about a future state , some uncertainty is necessarily associated with that prediction ( Figure 1 ) . The further ahead you are asked to predict , the larger the accompanying uncertainty term becomes and the more frequent and larger will be the prediction errors you make [2] . One route to learning is improving this look-ahead process . This may be accomplished by enhancing the ability to estimate the initial state of the cars ( distance/velocity ) or by honing the internal model that is used to predict the progression of the cars ( e . g . , to gain more accurate knowledge of possible lane changes ) . Critically , any solution in this family will result in a reasonably general improvement in the probability of successfully crossing the street . Honing the internal model will result in benefits that do not depend on the exact set of cars , positions , or paths . However , there is an alternative route to successful street crossing that does not require using ( or improving ) an internal model . Prediction can be eliminated by learning to map cues in the environment directly onto responses . The strategy of finding a mapping between perceptual information and actions that bypasses prediction is termed a policy in computer science . To illustrate , a simple policy for street crossing is to wait until the walk sign turns green . Policies greatly decrease both the cognitive overhead ( thus freeing resources for other endeavors ) and the error rate associated with predicting car trajectories . However , efficiency comes at a cost of inflexibility - policies are usually narrow in scope . The benefits of using a policy often completely disappear when various contextual changes are made to the environment ( the most obvious being if no crosswalk sign is present , if the crosswalk sign is malfunctioning or unintelligible , etc . ) . We emphasize that response policies are not the same as response biases . Response biases correspond to systematic errors in performance , and any number of strategies may give rise to this behavior . Response policies are specific action selection strategies that rely upon learned associations between stimuli and responses as described above . As there are clearly situations where humans do employ predictive models [3] , but also situations where human behavior is consistent with inflexible mappings characteristic of using response policies , is it possible to predict when one strategy will be favored over the other ? To address this question , we began with a simple hypothesis: namely that humans will employ the strategy that maximizes overall performance . If the demands of the task make response policies complex and difficult to learn , we predict that subjects will utilize predictive methods . Conversely , if the task demands make learning policies reasonably simple , we predict that subjects will eschew the computational costs of prediction in favor of exploiting simple mappings . To test our hypothesis , we chose to utilize a task in which human subjects have been shown repeatedly to rely upon a predictive model: visual extrapolation through occlusion . Thus , for the purpose of this study , we have narrowed the focus to identifying the conditions under which subjects learn to adopt a simpler policy-based strategy that abandons predictive extrapolation in favor of a direct mapping between particular stimulus inputs and specific extrapolation endpoints . The extrapolation task we employed was selected because it meets several key criteria . First , we needed a task dimension , in this case training set variability , that could be manipulated to either strongly favor a predictive model or a policy-based strategy . We show that extensive repetition of a small set of stimulus-response pairs should strongly favor learning mappings , while fewer repetitions of a large set of stimulus-response pairs should favor continued use of a predictive model . The second criterion is that the task allows subjects to freely use either type of strategy when selecting responses . Finally , and perhaps most critically , the relative efficacy of both strategies needs to be independently estimated . To this end , we created a motion version of a well-studied extrapolation paradigm ( e . g . , [4]–[5] ) . In the context of this task , we assess the use of two different subject-adopted strategies: one that is model-based and one that relies upon categorization-based stimulus-response mapping . In two experiments , we provide clear evidence that the strategy chosen by subjects , and thus the resulting generality/specificity , can be predicted by computing the best strategy given the training set they experience . The framework we develop here allows us to successfully make an explicit a priori determination as to which strategy would be most effective under what conditions and with clear predictions regarding the behavioral outcomes that would be observed in two experimental groups trained on two different sized sets of stimulus-response pairs as described below . The framework also provides a satisfying and parsimonious explanation for the question of “why” subjects select a given learning route – they simply choose the route that maximizes performance on the given task .
In Experiment 1 , subjects watched a dot travel along a circular trajectory before disappearing behind a stationary occluder ( see Figure 2A ) . After the dot's disappearance , subjects were asked to choose which of 20 “bins” ( each with a span of 9° ) residing at the opposite edge of the occluder the dot would reemerge within . The bin size was chosen to approximate subjects' inherent reliability in the extrapolation of curved contours using a similar stimulus [4]–[5] . The basic design consisted of a pre-test , extrapolation training with different set sizes , followed by a post-test to determine the impact of training ( Figure 2B , see Methods for a full description ) . During both pre-test and post-test , circular trajectories were drawn randomly from the full 2D space of curvature and orientation and no feedback was given so as to discourage learning . Subjects also completed a trajectory generation task in which they were shown the same moving dot stimulus , but after the dot reached the occluder the subjects were asked to use the mouse to draw the remainder of the trajectory . The ability to perform this type of data generation task is strongly indicative of having a predictive model . During training , subjects were provided feedback on their performance . These trials comprised a subset of fixed trajectories , which were repeated ( in pseduorandom order ) for a total of 200 exposures . We picked two training set sizes , “small” and “large” , with the former being predicted to favor mappings and the latter predictive models . We generated predictions by simulating performance by both a forward predictive model ( a Kalman filter ) that provides excellent fits to pre-test data and an agent that learns direct mappings between the visible trajectory statistics and response bins ( a model-free Q-learner; see Figure 3A , Methods and Text S1 and Text S2 of the Supporting Information for details ) . The simulations allowed us to make clear and quantitative a priori predictions about the effect of training set size on the performance of a predictive model strategy versus a policy-based strategy . This analysis suggested that a training set of 4 trajectory-response pairs would favor learning a response policy , while a training set of 20 trajectory-response pairs would favor the predictive model strategy , with the total number of training trials equated across groups . These values were used in the experiment . Subjects were randomly assigned to either a group that was trained on only 4 of the set of possible trajectories with a corresponding 4-bin response space ( 4Traj ) , or a second group that was trained on 20 different trajectories corresponding to the full 20-bin response space ( 20Traj; see Figure 3B ) . Both groups completed 5 training blocks of 80 trials each ( see Methods for more details ) . Therefore , within these blocks , either four trajectories were presented 20 times per block ( in the 4Traj group ) or 20 trajectories were each presented four times each ( in the 20Traj group ) . The four trajectories presented to the 4Traj group were a subset of the twenty trajectories presented to the 20Ttraj group . To assess the effects of training , both groups underwent a post-test identical to the pre-test ( both the choice and generation tasks ) . Recall that the post-test comprised a set of no-feedback trials using trajectories that differed from the training sets . Our prediction was that the 4Traj group would transition away from using a predictive model and toward using a simple mapping strategy . This should manifest in a number of ways . First , during training we expected to see substantial gains in accuracy in the 4Traj group ( far exceeding what would be possible using a predictive model ) . In fact , given the perceptual separation between the 4Traj trajectories , we expected performance during training to approach ceiling levels . However , at post-test , we expected that the use of the same ( now inappropriate ) mapping strategy would result in markedly poorer performance than was seen at pre-test . Again , this trade-off reflects the very heart of learning a response policy . Bypassing the need for a predictive model allows for extremely accurate responses to be made on training stimuli , but provides for no flexibility to deal with stimuli not in the training set . Conversely , we predicted that the 20Traj–trained subjects would not change their strategy from pre-test to post-test . In the 20Traj group , we predicted only modest improvements in performance at best as the number of training trials is theoretically unlikely to dramatically improve predictive performance via either changes in the state estimate or the internal trajectory model . There were no significant differences in pre-test behavior between the two groups in choice task accuracy ( both groups ∼25% correct ( chance performance = 5% ) , t ( 15 ) = 0 . 72 , p = . 49; see Figure 4A ) . On average , the 20Traj group had only a slight advantage with a mean absolute distance of 1 . 22 bins ( +/−0 . 0753 SEM ) from true over the 4Traj group's mean absolute distance of 1 . 42 bins ( +/−0 . 0876; see Figure 4B ) . We also considered whether subjects were biased to under- or overshoot the true trajectories' endpoints by taking the signed mean distance where negative values correspond to a tendency to undershoot the endpoint ( as if underestimating curvature ) . There was no systematic bias across subjects in either direction ( M20Traj = 0 . 0067+/−0 . 11 bins; M4Traj = −0 . 16 +/−0 . 12 ) Additionally , Kolmogorov-Smirnov tests failed to reject the null hypothesis that the two group distributions did not differ from their respective ideal choice bin distributions ( D = 0 . 0294 , p = 0 . 1243 ( 4Traj ) & D = 0 . 0281 , p = 0 . 334 ( 20Traj ) see Figure 4B ) . Finally , we considered subjects' bin choice confusability . An ideal subject accurately estimates the curvature and orientation of the visible portion of each trajectory and extrapolates to the correct bin . However , due to sensory noise , actual human subjects often misestimate the visible trajectories and extrapolate to the wrong bin . Sensory noise also causes different trajectories to be perceptually less discriminable and thus more confusable . We measure subjects' trajectory confusability by the range or number of bins around the true bin that the subject deems acceptable – the larger the range , the greater the confusability . At pre-test , there is no difference in confusability across the two groups ( t ( 15 ) = 1 . 6201 , p = 0 . 126; Figure 4C ) . Taken together , the pre-test results demonstrate that neither group had a pre-training bias toward disproportionately choosing specific bins . The pre-test results also verify that subjects are employing predictive models to arrive at their choices . In the absence of continuous visible information ( i . e . , the impact of occlusion ) and the requirement that a response be made later in time and space , subjects may either rely on an internal belief about where the dot is going ( i . e . , a model-based prediction ) when choosing a bin , or they must pick one at random - our data clearly rule out the latter strategy . How subjects learn these predictive models is beyond the scope of this paper and will be addressed in future work . The model simulations predicted that by the end of training , 4Traj subjects learning trajectory-to-bin mappings would achieve near ceiling performance . This is a significant advantage over what could be expected by relying on a predictive strategy ( approximately 1 . 5 times better , see Figure 3A ) . Indeed , as predicted , the 4Traj group choice accuracy improved significantly to reach 87 . 3% on average by the final training block ( +/−2% SEM; F ( 1 , 9 ) = 738 . 04 , p<0 . 001 ) . The accuracy of the 20Traj group exhibited modest , but nevertheless significant improvements during training , with accuracy climbing to 35 . 2% on average by the final training block ( +/−3% SEM; F ( 1 , 6 ) = 10 . 85 , p = 0 . 0165 ) . Thus , both groups were able to use the feedback during training to improve their performance . The critical test of our primary hypothesis , however , was the strategy each training group adopted during the post-test session . In direct contrast to pre-test , clear differences in choice behavior were evident between the two groups . As depicted in Figure 4A , the accuracy of the 20Traj group remained stable as compared to pre-test performance ( F ( 1 , 6 ) = 0 . 17 , p = 0 . 69 ) , a result consistent with continued use of the same type of response strategy . We note that although the 20Traj group accuracy remained stable , the choice distributions became even better aligned with the ideal ( ‘true’ ) distributions at post-test relative to pre-test as revealed by a smaller computed distance using the same Kolmogorov-Smirnov test ( D = 0 . 0246 , p = 0 . 5045 ) . Thus , although we did not expect a strategy shift or a drastic improvement in performance , the 20Traj training did provide modest benefit . By contrast , the accuracy of the 4Traj group showed a decline in accuracy from pre-test to post-test . At the group level , this decline was not significant ( F ( 1 , 9 ) = 2 . 59 , p = 0 . 1423 ) ; however , as will be discussed in detail next , seven out of ten subjects continued to rely almost exclusively on the four trained bins at post-test while the remaining three subjects returned to using the full set of bins . The former group showed a significant decline in choice accuracy relative to their pre-test performance ( F ( 1 , 6 ) = 7 . 21 , p = 0 . 036 ) whereas the latter group showed no change in accuracy , like their 20Traj counterparts ( F ( 1 , 3 ) = 0 . 81 , p = 0 . 463; Figure 4A ) . The decrease in accuracy for these subjects is accompanied by an increase in mean absolute distance between their bin choices and the true bins ( see Figure 4B ) . The decline in performance seen in the majority of 4Traj subjects is consistent with the predicted impact of learning to utilize an inflexible policy to make decisions . Moreover , the mapping strategy should rely on the use of the same bin whenever the input is deemed the same as one of the training inputs . As is evident in Figures 4B & 4c , there was a large increase in the probability of 4Traj subjects selecting one of the trained bins associated with a significant increase in confusability ( F ( 1 , 6 ) = 30 . 91 , p = 0 . 0014 ) for those subjects . That is , subjects were willing to accept the four trained bins as appropriate responses for a much broader range of trajectories than they were at pre-test . This was a pattern that emerged early in training – being evident even in the first no-feedback transfer block – and remained throughout in the majority ( 7/10 ) of 4Traj subjects as described ( Figure S1 ) . We note that the key difference between the transfer and post-test trials is that the post-test trials were randomly sampled from the full space of possible trajectories , as opposed to the fixed set of transfer block trajectories , which contained both trained and untrained trajectories . Further analysis of the transfer block data revealed that the 4Traj-trained subject bin choices were significantly more accurate for trained trajectories versus untrained trajectories ( t ( 18 ) = −4 . 5736 , p<0 . 001 ) . This is due largely to the fact that the majority of these subjects continue to use the four trained bins for all trajectories during these blocks ( see Figure S1 ) . The 20Traj subjects also exhibit a small but non-significant advantage for trained trajectories ( t ( 18 ) = −1 . 8681 , p>0 . 05 ) . Generally , the 20Traj group's use of a prediction-based strategy provides them with a performance advantage in comparison with their 4Traj-trained counterparts who rely on a categorization-based strategy for untrained trajectories ( see Figure S2 ) . Finally , subject data on the generation task , in which subjects drew the extended trajectories of the transfer trajectory set , show a qualitative change in behavior after 4Traj training . The plots in Figure 5 depict the sets of trajectories drawn by three 4Traj subjects at pre-test and post-test ( the first two columns; see Figure S3 for the remaining subjects' trajectories ) . For the purpose of comparison , the bins from the bin choice task have been included along the curved edge of the occluder with very small tickmarks , and the black dots correspond to the four trained bins . The true trajectory endpoint distribution is depicted by the transfer set ‘TSF’ in Figure 2B , which we note does not span the full edge of the occluder . Of particular note is that 4Traj-trained subjects who used only 4 bins in the choice task during transfer and post-test showed a similar trend in the drawing data . That the drawn trajectories show no resemblance to a circular trajectory demonstrates that they were not using their previous predictive model , but instead were aligning their drawn trajectories with one of the four trained endpoint locations . Given that the drawing trajectory set included the 4Traj training trajectories , we analyzed the data by removing these trajectories and looking at performance on the remaining trials . When we compare the distributions of the subjects' drawn endpoints with those of the true trajectory for these diagnostic trials , we find that no 20Traj subjects deviate from true at post-test ( Kolmogorov-Smirnov tests , p>0 . 05; see the drawn trajectories for the three sample 20Traj subjects depicted in the final two columns of Figure 5 and the remaining individual 20Traj subjects in Figure S3 ) . By contrast , half of the 4Traj subjects deviate significantly from true at post-test ( similar to the 7/10 subject count who continue to use the four trained bins during the post-test of the choice task ) . For validation purposes , none of the subjects in either group drew trajectories with endpoints that deviate from true for the trajectories that do terminate at the locations where the four trained bins occurred in the choice task . The post-test choices of the 4Traj group suggest that confusable inputs ( i . e . , similar trajectories ) were mapped to the trained responses , resulting in a four-response classification . The rationale for this idea is that the pre-test data show clear evidence of the confusability of trajectories ( recall Figure 4 ) . The set of trajectories that map to a bin is well-modeled by a Gaussian distribution over trajectory orientation and curvature . Learning a mapping with input confusability is equivalent to learning to divide the 2D space of orientation and curvature into categories ( responses ) . To assess the extent to which this strategy truly corresponded with subject behavior we simulated the performance of two types of decision makers . One simulated decision maker had an accurate understanding of the 20 trajectory parameter categories corresponding to the 20 response bins ( i . e . , could properly divide the 2D space of orientation and curvature into the 20 response bins ) , while the second divided the same space into 4 categories ( see Figure 6 and Text S3 of the Supporting Information ) , comprising the sets of new trajectories that are confusable with the four trained trajectories . The subjects' bin choice distributions replotted in the 2D parameter space ( Figure 7 and Figure S4 ) demonstrates that the distribution of choices at pre-test ( recall Figure 4 ) is consistent with 20-category based responses for both groups , though with some blurring of the categories . This is to be expected with non-perfect estimation and extrapolation . To quantify our qualitative observation , we then assessed the extent to which each of the two models correctly predicted individual subject choice behavior at both pre- and post-test . The Bayes factors of the four-category classifier model performance relative to the full model for each individual subject can be seen in Figure 8 . At pre-test all subjects were better fit by the 20-category than the 4-category model as expected . However , at post-test , a majority of the 4Traj subjects were better fit by the 4-category model ( Figure 7 and Figure S4 ) . Finally , in order to demonstrate that the observed behavior was the result of a strategy shift in the service of maximizing performance , we established that the persistence of the 4-category response strategy into the transfer blocks was completely dependent on a lack of feedback ( thus , the subject had no error signal that would indicate the strategy was not just as efficient as during the training ) . Indeed , if subjects were provided feedback on their performance with the post-test trajectories ( which end in all 20 bins ) , it should be immediately apparent that the four category mapping strategy resulted in far larger errors than would be the case for a predictive model . Four additional subjects underwent the same 4Traj training as described above . The only difference was that feedback was provided for the second half of the post-test . As can be seen in Figure 9 and Figure S5 , the data from the pre-test and the first half of the post-test closely mirror the previous results with the subjects at pre-test responding in a manner consistent with the 20-category model and at post-test with the 4-category model ( 3/4 subjects ) . However , once feedback was provided , an immediate switch back to the 20-category model was observed . It is worth noting that this final behavior effectively rules out one additional alternative hypothesis – namely that the behavior of the 4Traj subjects could be the result of combining the results of a predictive model with a prior learned over bins . Because the subjects experienced correct responses at only four of the bins , the posterior over bins would thus also share that property . Since data accumulated into a prior converges to a delta function in the limit of infinite data assuming the identifiability of the sufficient statistics on x , the impact of additional data on the prior monotonically decreases as the amount of data goes to infinity . Thus , undoing this type of learned prior could only be accomplished via a significant amount of data , far less than appears to be required for them to completely alter their choice strategy after feedback is provided in the post-test . Instead , the behavior does indeed appear to be most consistent with a shift toward a pure categorization approach with no predictive model being utilized . The results of Experiment 1 suggest that human subjects rely on contextual experience in order to determine the simplest strategy that maximizes task performance . In the context of trajectory extrapolation , all subjects initially relied upon a model-based extrapolation strategy that extends the visible trajectory through the occlusion region via a trajectory dynamics model , in this case , one that is consistent with a parameter space partitioned into 20 categories . However , a simpler classification strategy emerges with repetitive training on a limited set of trajectories that indicates to the subject that only a few specific responses are needed . This strategy remains the preferred one until new information is provided that the context has changed , upon which subjects rapidly switch strategies to improve performance and more effectively meet task demands . However , it is not clear what kind of mapping the 4Traj strategy classification corresponds to . One possibility is that the mapping is between trajectories and response bins - subjects may learn that bins 1 , 7 , 13 , &17 are the only appropriate responses in this experiment . Alternatively , subjects may learn a more general mapping between trajectories and regions of space , for example , the regions of space indicated by the predictive model for the trained stimuli . In other words , feedback might serve to crystalize or reinforce particular model extrapolations , rather than particular responses . We tested these possibilities in a second experiment by having subjects make their bin choices at multiple distances from the point of occlusion ( i . e . , along the curved edge of multiple sized half-disk occluders ) after training . If classifying subjects map inputs to response bins , this manipulation should have no effect on the persistent use of the identical four bins at post-test . If subjects map inputs to particular model extrapolations , a new set of four should emerge with each new occluder radius that is sensible for that prediction distance and set of trajectories experienced during training . Predictions for each of the two mapping hypotheses are shown in Figure 9a , with response mapping represented by the blue dashed lines and extrapolation mapping represented by the green dashed lines . A third possibility is that subjects only learn a mapping for the specific occluder used in training—that is , only for the mid-sized . In this case , we expect choice behavior would revert to pre-test . For subjects showing the response mapping strategy ( 7 out of 10 ) we found that they used the same response bins for all three occluder conditions ( Figure 10A and Figure S6 ) , strongly supporting the idea that subjects are learning a response mapping . Consistent with the results of Experiment 1 and the behavioral predictions , the 20Traj subjects consistently performed in accordance with the full model throughout the experiment ( Figure 10B and Figure S7 ) . We conclude from these results that 4Traj subjects who adopt a classification response strategy at post-test use this strategy to bypass extrapolation altogether . If they were encoding a four-bin response strategy at a decision-level that runs the trajectory forward through occlusion ( i . e . , via extrapolation ) , a different set of four bins would be expected with each occluder condition because the trained trajectories reemerge in different bins for the different occluders . Instead , the categorizing subjects chose among the same four trained bins for all three occluder radii . This strategy saves computational cost as it eliminates using a forward predictive model .
In two experiments , we have provided clear evidence that the strategy learning route chosen by subjects , and thus the resulting generality/specificity of that strategy , can be predicted by computing the best strategy given the training set they experience . The framework that we utilized in this endeavor stands in contrast to previous approaches that have relied on somewhat loose ( and often seemingly post-hoc ) descriptions of task conditions that result in flexible and/or inflexible learning ( e . g . , whether the task is “easy” or “hard” [6] ) . Here we were able to make an explicit a priori determination as to which strategy would be most effective under what conditions and thus were able to make clear predictions regarding the behavioral outcomes that would be observed in our 4Traj and 20Traj groups . The framework also provides a satisfying and parsimonious explanation for the question of why subjects select a given learning route – they simply choose the route that maximizes performance on the given task . In the context of a motion extrapolation task , we assessed the use of two different subject-adopted strategies: one that is model-based and one based on categorization-based stimulus-response mapping . When the training paradigm did not provide a reliable route to bypassing model-based extrapolation , subjects continued to use extrapolation and thus suffered no decrements in performance at post-test . Conversely , when training provided a viable opportunity to forego costly and error prone model-based predictions in favor of a simpler categorization-based mapping , subjects quickly transitioned to the less demanding approach . While this shift in tactic provided substantial benefits on the trained task itself ( indeed , in our experiments 4Traj subjects increased their accuracy by approximately 60% ) , the fact that the categorization scheme was only truly appropriate for the training task conditions meant that there was a significant cost to performance when the conditions changed ( i . e . , at post-test ) . In essence , the mapping strategy was both highly efficient and highly inflexible - it allowed the subjects to perform the single trained task very well , but to do little else when the context changed . Subjects' adoption of the categorization-based strategy revealed two additional insights about action selection . First , adoption of simpler , training-specific strategies did not occur at the expense of more general abilities . That is , when subjects were provided with feedback that the categorization strategy is not suitable for the task at post-test , they rapidly returned to their original model-based strategy . Therefore , task training adds to the toolbox , rather than replaces old ones , allowing individuals to choose the best tool for the immediate task . Second , use of the simpler categorization strategy may not actually be for the purposes of successful extrapolation more generally per se , but instead , for the purpose of minimizing costs associated with action planning . Not only do subjects continue to use the categorization scheme at post-test , they continue to use the same scheme at post-test for multiple extrapolation distances . That is , they use the strategy to bypass extrapolation altogether . This is the least computationally demanding strategy supported by these subjects' experience , as it never requires a forward predictive model . Thus , we additionally conclude that people exploit any shortcuts that may be revealed through training . It is worth noting though that there is room for improvement in these a priori predictions . In particular , approximately 25% of the 4Traj subjects did not conform . Rather than continuing to utilize the categorization strategy during the post-test , they immediately switched back to using their predictive model . We could find no particular trend in the data to allow us to predict which subjects would be part of this 25% . Those that did switch back may have had other knowledge or made alternative inferences regarding the task context that we could not directly measure with the current paradigm . According to our overarching framework , however , we would predict that the cognitive cost associated with running the predictive model forward should be measurably less in these subjects . This is a hypothesis that remains to be tested in future work . Furthermore , subjects in the 20Traj condition learned very little during the course of training . However , we expected the number of trials in 20Traj training to be insufficient to learn a 20-category strategy . This result is reasonable given that 20Traj subjects had only 20 trials training per trajectory , rather than the 100 trials per trajectory experienced by the 4Traj subjects . Moreover , given trajectory confusability , the error feedback in training contained little information that could impose changes on the model . Thus , our 4Traj and 20Traj training conditions were designed to create a scenario in which subjects could learn a simpler categorization model or stick with their original predictive model . Future work will be devoted to establishing conditions under which subjects will be expected to learn a new predictive model , or , more likely , improve an existing one , in order to benefit task performance . Such work will contribute further insight into the effects of training on action strategy learning and selection . The distinction between flexible and inflexible learning has been made in many distinct areas of the learning literature [7]–[9] , but without knowing the set of alternative strategies and the cost/benefit ratios of those various strategies in advance , it is difficult to predict when each will arise . Here , we show that by starting with an inherently predictive task , training conditions can be established that directly promote inflexible learning . Overall , the general outcome observed in 4Traj subject performance is consistent with the finding of inflexible ( ‘specific’ ) learning prevalent in the perceptual learning literature . That the majority of the 4Traj subjects were unable to adapt to the new trajectories at posttest despite achieving near ceiling performance during training , is analogous to typical reports of this field: substantial improvements in performance on the trained task , but no ability to perform new tasks ( i . e . , no transfer of learning ) ( see [10] for a review ) . For instance , subjects trained to discriminate whether a field of moving dots was moving just barely clockwise or counterclockwise from straight up will show a substantial improvement in accuracy through training . However , when they are subsequently asked to perform the same discrimination task around a new angle ( e . g . , straight left ) , no benefits of the training are observed [11]–[13] . Similarly inflexible learning has been observed for low-level visual features such as orientation , contrast , texture , and retinal location [14]–[18] . Not surprisingly , given the results reported here , the training conditions most commonly utilized throughout this literature are consistent with those that we have shown promote the use of a task specific strategy – namely large numbers of trials and very small stimulus sets . This same basic finding has been shown in other domains as well , such as motor and cognitive training . Similarly , the fact that only the subjects trained with a more variable stimulus set ( i . e . , the 20Traj subjects ) retained the ability to generalize is also consistent with the examples in the literature of training paradigms that produce flexible ( ‘general’ ) learning [19]–[23] . In particular , these training experiments have commonly used highly variable stimulus sets , which prevent learning of specific strategies that are reliable . Although our task does differ from these and other classic learning tasks in a number of potentially key ways , the parallels between our findings and those of other learning domains suggests important links can be made . We believe that our findings and framework offer insight into general principles of learning , beyond any specific learning context . For example , the notion of ‘overfitting’ ( see [10] for a review ) in perceptual learning can be operationalized as policy learning , as described above . Nevertheless , it will be left to future work to determine whether the framework we put forth here can directly account for the results in more ‘classic’ learning tasks . Outside of the domain of perceptual learning , policy learning , or learning a direct mapping between a condition and a response , is implicated as a major hypothesis for habit-based learning . A similar prediction of habit-based dominance was put forward by [24] . In their work , the reliability of performance determines whether a model-based strategy is adopted versus a simpler model-free strategy . Other domains also describe the phenomenon of automaticity . For instance , when subjects are asked to simulate how they would move the steering wheel of a car to make a lane change to the left , most fail somewhat dramatically [25] , moving the wheel to the left and then straightening ( which would very quickly see them go off the left side of the road ) , rather than moving the wheel to the left and then back to the right . In other words , removing the input breaks the trained or automatic habitual response . In essence , these subjects can initiate a lane-change but lack a method of predicting what the next motor action should be and thus they cannot run this seemingly simple process forward for the two seconds that it would take to change lanes . Instead , subjects appear to rely entirely on a set of automatically executed mappings between cues in the environment and actions they should take ( e . g . , if the car is going off toward the left side of the road , turn the steering wheel back right ) . Similar automaticity is observed in other domains , such as in catching fly balls in baseball . This is a clear situation where one could , in principle , perform an extensive look ahead when the ball is struck . This would allow the fielder to run directly to the most likely landing location ( which would be constantly updated as more and more of the trajectory was observed ) . However , in practice , human fielders appear to execute a series of automatic responses based on the flight of the ball ( e . g . , to run in such a way as to cancel the vertical acceleration of the image of the ball on the retina; [26]–[28] ) . Our results also bear similarity to findings in the statistical learning literature . For example , Chalk et al [29] provide evidence that subjects develop biases to report perceived motion direction consistent with the motion distribution of the training stimuli . Our subjects similarly exhibit a change in their responses whereby the choices are consistent with the distribution of responses used during training . However , unlike previous authors who conclude that the statistical learning biases perception of the stimuli , we conclude that learning biases the responses made . This is supported by the fact that subjects immediately switch strategies when they are explicitly signaled that the task context has changed; that is , we have no evidence that our subjects perceive the stimulus differently , nor evidence to suggest that our subjects are using fundamentally different perceptual information in the categorization and the extrapolation cases . In sum , our primary goal is to contribute new understanding about the mechanisms underlying training-based changes in task performance . Towards that end , we establish a new approach towards predicting and training the learning of different action strategies . We also provide results that span accounts from several literatures . Specifically , the categorization-based responses of our 4Traj subjects link these literatures , as the responses embody the characteristics of inflexible ( non-generalizable ) learning of model-free policies that map trajectories directly to specific responses , with little sensitivity to changes in context ( i . e . , significant failures once removed from the trained scenario ) . Finally , we provide clear evidence that learning is the direct result of the training paradigm , establishing perceptual learning as one potential future avenue of research . The importance of the goal of training must be reflected in the design of any training paradigm . If the goal is to produce a large amount of learning on only a specific task , with no need for generalization , then the best route is large numbers of training trials on only that task . However , if the goal is to produce learning that generalizes beyond the specifics of the training task , it is essential to ensure that the parameters of the task provide no feasible option to transition to a computationally simpler mapping strategy .
All research was carried out with approval by the University of Minnesota IRB , protocol number: 0201M16281 and Assurance of Compliance number: FWA00000312 . The goal of the model simulations was to assess the relative efficacy of a mapping strategy versus a predictive model strategy ( given a human-like ability to estimate trajectory parameters such as curvature and orientation ) as a function of the number of trajectory-response pairs that needed to be learned . More specifically , we wished to identify conditions that strongly favored one strategy over the other , which could then be used in an experiment with human subjects . We believe this provides a significant advance in the study of conditions that result in flexible versus rigid learning . To evaluate the effectiveness of utilizing a predictive model on the task , we used a linear state space model , often termed a Kalman filter , a standard algorithm used in modeling trajectory estimation [30]–[33] . At each time step , the Kalman filter maintains an internal representation of the dot's state – its position , velocity , and acceleration . During the visible portion of the trajectory , the Kalman filter repeatedly runs through a cycle of prediction , observation , and correction . It applies its internal model to the dot's state to make a prediction about the dot's state at the next time step . Then it observes the actual position at the next time step ( corrupted slightly by noise ) , and corrects its estimate based on this new observation for each time step where the dot is observable . When the dot hits the occluder , the Kalman filter continues its predictions through the duration of occlusion without observation/correction steps ( see Text S2 of the Supporting Information for further details ) . The total number of trials was set to 2000 and simulated for 4 , 8 , 12 , 16 , and 20 trajectories . To evaluate the effectiveness of learning mappings , we simulated Q-learning [34]–[36] , a standard iterative learning approach used in the reinforcement learning literature to learn a response mapping ( see Text S1 of the Supporting Information for details ) . Briefly , the agent learns an action-value function that gives the expected value , in terms of performance of each response in a given state – the trajectory . Actions-values are learned via experience – for each trajectory , the agent makes a response and tabulates its performance feedback for this trajectory-response pair . The agent's “knowledge” is thus represented by a look-up table , Q , in which each row corresponds to a specific trajectory , and each column corresponds to one of the possible choice bins . We performed five separate Q-learning simulations . We simulated performance after learning under the same conditions as above ( 2000 trials each for training sets 4 , 8 , 12 , 16 , & 20 trajectories ) . Because there are more repetitions of individual trajectories for small set sizes , mapping performance will degrade with training set size . Relative model performance was quantified as the log ratio of percent correct choices . Figure 2C shows that the mapping strategy outperformed the predictive model strategy for training sets less than 10 , after which the performance reversed . We chose to test subjects on set sizes of 4 and 20 as these conditions strongly favored one of the strategies . | Predicting what humans will learn from a training task , in particular , whether learning will generalize beyond the specifics of the given experience , is of both significant practical and theoretical interest . However , a principled understanding of the relationship between training conditions and learning generalization remains elusive . In this paper , we develop a computational framework for predicting which of two basic decision-making strategies will be utilized by human subjects - 1 ) simple stimulus-response mappings or 2 ) predictive models . Through simulation , we show that the nature of the training experience determines which of these categories leads to better in-task performance; repetitive training on a small set of examples favors simple stimulus-response mappings , whereas training on a large set of examples favors predictive strategies . We then show that humans trained under these various conditions do indeed utilize the predicted strategy . Finally , we show that the strategies that are utilized during training predict generalization of learning . Those who learn simple mappings fail to generalize their new skills , in contrast to those who use default predictive strategies . The framework developed here is useful both in interpreting previous patterns of learning , as well as in building training paradigms with given desired outcomes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"experimental",
"psychology",
"learning",
"psychology",
"social",
"and",
"behavioral",
"sciences",
"cognitive",
"psychology",
"psychophysics",
"sensory",
"perception"
] | 2014 | Task-Specific Response Strategy Selection on the Basis of Recent Training Experience |
Multiple Acyl-CoA Dehydrogenase Deficiency ( MADD ) is a severe mitochondrial disorder featuring multi-organ dysfunction . Mutations in either the ETFA , ETFB , and ETFDH genes can cause MADD but very little is known about disease specific mechanisms due to a paucity of animal models . We report a novel zebrafish mutant dark xavier ( dxavu463 ) that has an inactivating mutation in the etfa gene . dxavu463 recapitulates numerous pathological and biochemical features seen in patients with MADD including brain , liver , and kidney disease . Similar to children with MADD , homozygote mutant dxavu463 zebrafish have a spectrum of phenotypes ranging from moderate to severe . Interestingly , excessive maternal feeding significantly exacerbated the phenotype . Homozygous mutant dxavu463 zebrafish have swollen and hyperplastic neural progenitor cells , hepatocytes and kidney tubule cells as well as elevations in triacylglycerol , cerebroside sulfate and cholesterol levels . Their mitochondria were also greatly enlarged , lacked normal cristae , and were dysfunctional . We also found increased signaling of the mechanistic target of rapamycin complex 1 ( mTORC1 ) with enlarged cell size and proliferation . Treatment with rapamycin partially reversed these abnormalities . Our results indicate that etfa gene function is remarkably conserved in zebrafish as compared to humans with highly similar pathological , biochemical abnormalities to those reported in children with MADD . Altered mTORC1 signaling and maternal nutritional status may play critical roles in MADD disease progression and suggest novel treatment approaches that may ameliorate disease severity .
Multiple acyl-CoA dehydrogenase deficiency ( MADD ) , also known as glutaric aciduria type II ( GA-II , OMIM #231680 ) , is a rare autosomal recessive inherited metabolic disorder first described in 1976 [1] . The precise incidence and prevalence are unknown but are likely underreported given the variability in clinical presentation . MADD is caused by mutations in electron transfer flavoprotein genes A ( ETFA ) , B ( ETFB ) or the ETF dehydrogenase ( ETFDH ) [2] . The ETFA and ETFB gene products , ETFα and ETFβ respectively , form an ETF heterodimer located in the mitochondria matrix [3] . This complex receives electrons from at least nine distinct dehydrogenases that are involved in fatty acid β-oxidation , amino acid and choline metabolism [4] , [5] , [6] , [7] . Patients with MADD are classified by disease severity with type 1 having severe neonatal-onset with congenital anomalies , rapid deterioration and death [8] . Type 2 patients with MADD do not have congenital anomalies but still have a severe course with death usually during the few years of life [9] . Finally , type 3 patients have later onset and an overall milder course . However they still have hypoglycemia , metabolic acidosis , cardiomyopathy , hepatomegaly , kidney defects and neurological manifestations such as encephalopathy and leukodystrophy [10] , [11] . Current treatments are mainly aimed at relieving symptoms though anecdotal reports of improvement after administration of riboflavin or Coenzyme Q have been reported [11] . While all types of MADD can be caused by ETFA , ETFB or ETFDH mutations , it is not understood why there is such variability in disease severity . Several reports indicate a marked buildup of fatty acids , amino acid or toxic compounds in multiple organs in patients with MADD . However , comprehensive cellular and molecular analyses have not been possible as there are no animal models available that recapitulates the spectrum of abnormalities seen in patients with MADD . The first animal model of MADD was created by inactivating the zebrafish etfdh gene [12] . This mutant zebrafish was named xavier ( xav ) with conserved metabolic abnormalities also observed in MADD patients including increased levels of acylcarnitines and glutaric acid . However xav mutant zebrafish did not recapitulate morphological defects observed in MADD patients . This may be due to early lethality seen in this model prior to later stages of organogenesis . Using forward genetic screening for mutants with abnormal livers , we identified a mutant zebrafish called dark xavier ( dxavu463 , termed hereafter as dxa ) due to its phenotype of a dark fatty liver and hepatomegaly . Dxa mutant zebrafish have a nonsense mutation in the etfa gene resulting in widespread abnormalities broadly similar to those observed in MADD patients . We found large increases of acylcarnitines and glutaric acid in dxa mutants associated with multiple abnormalities of various organs including brain , liver , kidneys and heart . Marked accumulation of neutral lipid drops including cerebroside sulfate and free cholesterol in multiple organs was also observed . Analyses by mass spectrometry [13] found a large increase in triacylglycerides in dxa mutants but also a significant decrease of phosphatidylserine species which was also observed in human tissue derived from a patient with MADD [14] . The multiple defects seen in dxa mutant zebrafish closely recapitulate many core abnormalities observed in human patients with MADD . Interestingly , dxavu463 mutant developed hyperplasia with increased cell size in multiple organs including brain , liver and kidney suggesting activation of mTORC1 signaling . Excessive maternal feeding also exacerbated the phenotype in dxa mutants . We confirmed that mTORC1 signaling is highly elevated in dxa with increased phosphorylation of S6 and 4E-BP1 . Treatment of dxa zebrafish with rapamycin alleviated a subset of signaling and cellular proliferation abnormalities suggesting that targeting mTORC1 signaling could be a rational therapeutic approach for patients with MADD .
We identified dxa mutants during a forward genetic screen using ENU mutagenized zebrafish . Homozygous dxa mutants had a large and dark colored liver at 7 days post fertilization ( dpf ) ( Figure 1A ) . However , when more closely examined , dxa mutants had a broad spectrum of defects during development and post developmental stages ( Figure 1A ) . About 20% of mutants had severe congenital defects ( type I ) that included a small head and cardiac edema , these larvae died by 5–6 dpf . Approximately 18% of mutants were type II with moderate defects including an abnormal head and dark liver , intestine and brain . These died by 7–8 dpf . The remainder of the mutants ( approximately 62% ) classified as type III had mild defects that were morphologically close to wild type zebrafish except for a darker appearing liver , intestine and brain ( n = 218 ) . Type III mutants lived for 10 dpf in the unfed state whereas control siblings live for 10–12 dpf . Overall , type I , II and III mutants accounted for approximately 25% of the total zebrafish in each cross suggesting the dxa phenotype was due to a defect in an autosomal recessive gene . This was later confirmed ( see below ) as a mutation in the etfa gene known to be involved in mitochondrial function . Given the potential for metabolic influences on mitochondrial disease , we studied whether maternal overfeeding prior to egg laying could influence the phenotype . One week of extra feeding caused a dramatic shift in severity with 57% of dxa mutant zebrafish now classified as type I , 32% type II and 11% type III ( n = 151 ) ( Figure 1B ) . This result suggests that the maternal nutritional state dramatically affects the severity of dxa zebrafish and may also explain similar phenotypic variability reported in patients with MADD . To identify the mutant gene in dxa zebrafish , we performed conventional linkage mapping and were able to map the likely gene to approximately 0 . 18 cM from the galk2 gene , located on zebrafish chromosome 25 ( 1/547 recombination , data not shown ) . Whole genome sequencing of dxavu463 and control zebrafish ultimately identified a mutation in the etfa gene approximately 360 kb from galk2 . This G to T mutation introduces a premature stop codon ( G290X ) in etfa ( Figure 1C ) . Zebrafish etfa has 80% homology to the human ETFA gene suggesting a highly conserved function ( sequence alignment not shown ) . Whole mount in situ hybridization of etfa mRNA shows maternal and ubiquitous expression during early development with subsequent high expression levels maintained in the midbrain and blood vessels at 30 hours post fertilization ( hpf ) as well as liver and pectoral fins at 2 dpf ( Figure S1A ) . Immunofluorescent staining with an anti-Etfa antibody also revealed high expression of Etfa protein in neural progenitor cells located adjacent to the ventricles of the brain of wild type larvae ( Figure 1D , head , inset ) . We also saw strong expression in neuromast hair cells as well as kidney , liver and skeletal muscle of the pectoral fin of wild type at 9 dpf ( Figure 1D , trunk , inset , n = 9/9 ) . However , negligible Etfa protein was detected in dxa zebrafish ( Figure 1D , bottom panel , n = 9/9 . There is some residual signal in the dxa zebrafish that we interpret as non-specific binding of the secondary antibody to the outer pial membranes of the brain and outer eye ( Figure 1D ) . Immunoblot analyses also detected a very minimal amount of Etfa protein in the dxa mutant ( Figure S1C ) . This also supports a loss of function mutation due to non-sense mediated decay of etfa mRNA given the location of the premature stop codon in exon 10 and the in situ expression data ( Figure 1D lower panel , Figure S1B . 10/10 ) . This expression pattern of etfa further supports an important role in high energy demanding cell types such as neural progenitors within the brain , hepatocytes and kidney tubule cells . While genetic testing of affected patients is ideal , MADD can be strongly suspected in symptomatic children who exhibit increased serum acylcarnitines and glutaric aciduria [7] . Using tandem mass spectroscopy to determine acylcarnitine levels , we found significantly higher level of multiple long- , medium- and short-chain acyl-CoA species and isovalerylcarnitine in the dxa mutant larvae compared to control siblings ( Figure S2A ) . This suggests that dysregulation of mitochondrial β-oxidation is highly similar in dxa to that observed in patients with MADD . Further analysis of organic acids using gas chromatography-mass spectrometry found approximately 6 . 5 µg of glutaric acid per dxa larvae ( Figure S2C ) , but no detectable amount seen in control siblings ( Figure S2B ) . This pattern is highly reminiscent of that seen in patients with MADD , also known as glutaric aciduria Type II ( GA-II ) . Hepatic steatosis is a central sign in MADD , likely resulting from defective fatty acid β-oxidation that may be exacerbated during episodes of hypoglycemia . Dxa mutants exhibit progressive accumulation of dark colored granules in multiple organs including brain , liver and intestine after 6 dpf ( see Figure 1A ) . Oil Red O ( ORO ) staining in whole mounts of type II dxa mutant larvae and coronal sections showed massive accumulations of neutral lipid in the brain , liver and intestine as well as blood vessels at 8 dpf ( Figure 2A , B ) . Interestingly , toluidine blue staining of thick sections used for Transmission Electron Microscopy ( TEM ) revealed many heterogeneous sized vacuoles in the liver with brown colored drops in the cytosol of hepatocytes ( Figure 2C ) . As glycosphingolipid can include glucose or galactose and sulfate groups ( cerebroside sulfate ) that can be stained with toluidine blue , the brown drops within hepatocytes may contain cerebroside sulfate . Intensive Periodic Acid Schiff ( PAS ) staining was also seen in dxa liver ( Figure 2C , middle ) suggesting that these lipid drops are comprised of cerebroside sulfate as liver glycogen is normally undetectable in unfed 8 dpf zebrafish . Additional support that the drops do not contain glycogen is supported by TEM analyses where we did not observe any glycogen containing granules at 6 or 8 dpf ( data not shown ) . We also found high levels of free cholesterol in the cytosol of dxa hepatocytes using filipin staining ( Figure 2C , right ) . However , we did not see lipid and free cholesterol accumulation in dxa liver at 6 dpf although the mutants already exhibit hepatomegaly and enlarged hepatocytes ( Figure S3A–D ) Enlarged cell size was a consistent phenotype at later stages ( Figure S3E , F ) . Dxa hepatocytes were approximately three times larger those seen in control siblings ( Figure S3G ) . These results suggest that intrinsic abnormalities of hepatocytes led to both lipid and cholesterol accumulation in dxa mutant zebrafish . We then analyzed cellular ultrastructure using TEM to investigate possible organelle defects . The internal mitochondrial cristae density was markedly decreased in type II dxa hepatocytes at 6 dpf although total mitochondrial size was not changed ( Figure 2D ) . Strikingly , we found extremely large mitochondria with minimal cristae in type II mutants just 2 days later at 8 dpf ( Figure 2E ) . From these TEM results , we conclude that the “vacuoles” we saw in the mutant liver are actually grossly swollen mitochondria ( Figure 2C , E ) . This suggests that Etfa is required for mitochondrial maintenance as well as energy metabolism . We assessed mitochondrial function in dxa mutants by measuring oxygen consumption over time . We found significantly decreased oxygen flux in dxa mutant zebrafish compared to sibling controls ( Figure S4 ) . This strongly supports an impairment of mitochondria function in etfa mutant cells . It has been reported that many patients with severe neonatal onset MADD have polycystic kidney disease though Bohn et al . and Harkin et al . reported that these kidneys were pathologically distinct from typical polycystic kidney disease [15] , [16] . We did note high Etfa expression within pronephric tubules of wild type kidneys ( Figure 1D , trunk ) . Histological analysis of toluidine blue stained sections of dxa zebrafish kidney showed clear abnormalities possibly resulting from hypertrophy of the pronephric tubular epithelium . We also found a large number of prominent vacuoles in both type II and III dxa kidney epithelium compared to wild type ( Figure 3A ) . TEM analysis of type II dxa mutants showed similar to hepatocytes , they are very large mitochondria with minimal cristae ( Figure 3C , 2E ) . This finding implies that the “cystic kidney” pathology ascribed to patients with MADD may be due to massively swollen mitochondria in kidney tubule cells . We further found lipid and free cholesterol accumulation in the cytosol of dxa mutant kidney cells ( Figure 3B , D ) . However , we did not see significant increases of lipid and free cholesterol in dxa mutants that have only mild defects at earlier stages , though hypertrophic kidney tubule cells with swollen mitochondria are already present . ORO staining also revealed extensive lipid drops in the brain ( Figure 2A ) . We then analyzed brain sections to more precisely determine the location and cell types that contain lipid . We found large lipid accumulation in the ventricular zone ( VZ ) of the brain where the neural progenitor cells are found ( Figure 4A ) . Within the same region of dxa mutant brain , we also found cerebroside sulfate containing lipid drops at 9 dpf ( Figure 4B ) . Interestingly , type II mutants with more severe defects have VZ cells with very large nuclei compared to other neurons , but these large cells had pale intracellular staining ( Figure 4B ) . TEM revealed that these cells do not have a discernible subcellular structure other than swollen nucleus and mitochondria ( Figure 4C , D ) . These findings are suggestive of ongoing necrosis although we could not identify ruptured cell membranes . Of note we did not see enlarged nuclei at 6 dpf but swollen mitochondria were still observed in the type II mutant brain ( data not shown ) . These results suggest progressive and rapid brain damage after 6 dpf in dxa zebrafish . To analyze whether those abnormal cells are neural progenitors , we performed immunostaining for Sox2 in both type II and III dxa larvae at 8 dpf . We found increased numbers of Sox2 positive cells in both type II and III mutants ( n = 9/9 ) . Statistical analysis of type II mutants showed a 75% increase of Sox2 positive cells in the dorsal part of the VZ at 8 dpf ( Figure S5 ) . Finally , we found that brain lipid-binding protein ( BLBP ) positive neural progenitor cells were increased and distorted in their morphology though had decreased processes within the white matter of dxa mutants ( Figure 4H , asterisk and yellow magnified inset , n = 6/6 ) . Altogether , these results suggest that neural progenitor cells in the dxa mutant are hyperplastic and hyperproliferative . They may also be unable to properly generate neurons given the abnormal appearing grey matter observed in dxa mutant brain ( Figure 4 ) . Given the overt increases in lipids seen by Oil Red O staining , we performed lipid profiling with mass spectrophotometry ( MS ) to identify differences of lipid molecular species between control and dxa mutant larvae at 8 dpf . We found moderately decreased monoacylglycerol ( MAG ) and diacylglycerol ( DAG ) in the dxa mutant though only the MAG decrease was statistically significant . However a large increase of triacylglycerol ( TAG ) was observed in the dxavu463 mutants ( Figure S6 ) . Using MS analyses of glycerophospholipids , we also found significant decreases in phosphatidylserine ( PS ) species ( Figure S6 ) . By contrast , the two most abundant phospholipid species phosphatidylcholine ( PC ) and phosphatidylethanolamine ( PE ) did not show any statistically significant differences between control and dxa mutant zebrafish . These results provide a rationale for future lipid modifying therapies in patients with MADD and will help focus future experiments on lipid abnormalities seen in etfa mutant zebrafish . High Etfa expression was found within neuromast cells that are zebrafish sensory organs ( Figure 1D ) . Interestingly , type I and II dxa mutants had a decreased response to touch stimulation ( data not shown ) that was correlated with the severity of defects and increased age . Neuromast cells had short or absent kinocilia in type II dxa mutant and using ORO staining we found the dark granules seen in dxa neuromasts are comprised of lipids ( Figure 5A , seen in 10/10 mutant larvae examined ) . Given these widespread lipid abnormalities in sensory structures , we also looked for alterations in sensory nerves tracts . Using acetylated tubulin , we noted decreased staining with a disorganized appearing , “kinked” axonal track in type II dxa mutants ( Figure 5B , n = 8/9 ) . We then examined expression of myelin basic protein ( MBP ) , the most abundant myelin associated protein in the brain and spinal cord . We found decreased MBP staining in dxa type II mutants ( Figure 5C , n = 6/6 ) but did not see significant changes in type III mutants ( data not shown , n = 5/6 ) . Using TEM to examine myelination in the spinal cord , we again found abnormally increased size and morphology of mitochondria in the Mauthner axon track ( sensory pathway mediating escape responses ) as well as decreased myelination ( Figure 5D ) . These hypomyelination findings are reminiscent of the leukodystrophy reported in patients with MADD [17] . The markedly enlarged cells in dxa mutant brain , liver and kidney suggests that mTORC1 could be involved as signaling through this kinase is a key controller of cell size [18] . We previously showed activation of this pathway in zebrafish causes increased cell size that can be reversed with rapamycin , a potent mTORC1 inhibitor [19] . We then evaluated the phosphorylation status of the mTORC1 downstream effectors S6 ribosomal protein and 4E-BP1 by immunofluorescence . We found markedly elevated phospho-S6 and phospho-4E-BP1 in dxa mutants especially in neural progenitor cells ( n = 4/6 ( phospho-S6 ) , 6/6 ( phospho-4E-BP1 ) ) and pial cells ( n = 6/6 ( phospho-S6 ) , 6/6 ( phospho-4E-BP1 ) ) on the midbrain and central canal of the hindbrain ( Figure 6A , B ) . Increased level of phospho-4E-BP1 was seen more broadly than phospho-S6 in dxa mutants notably within midline cells and the central canal of the hindbrain as well as the intestine ( Figure 6B ) . Immunoblots of dxa mutants also revealed increased mTORC1 signaling compared to control larvae at 6 dpf ( Figure S7A–C ) . We also found increased phospho-S6 and phospho-4E-BP1 in the kidney and liver at 8 dpf ( Figure 6B ) . Even at earlier time points before the pathology was overt , we still found increased mTORC1 signaling in the liver ( Figure S7D ) . Given these findings we hypothesized that mTORC1 inhibition could potentially rescue dxa mutants . However , rapamycin treatment from 3–9 dpf was not able to suppress the dxa mutant phenotype ( data not shown ) . To address whether activated mTORC1 observed in dxa mutants is rapamycin sensitive , we treated with rapamycin daily from 5 dpf to 8 dpf at a concentration of 300 nM . Rapamycin freely crosses the blood brain barrier in zebrafish and typically inhibits mTORC1 downstream completely [19] . In contrast to results we have obtained with other zebrafish models of human disease , we found that levels of phospho-S6 and phospho-4E-BP1 were not fully suppressed in the brains of treated larvae ( Figure 6C ) . In the liver rapamycin did in fact suppress phospho-S6 levels but phospho-4E-BP1 actually appeared to be increased ( Figure 6D ) [20] . This intriguing finding suggests a novel regulation of mTORC1 downstream effectors in dxa mutant zebrafish . We previously found that tsc2 mutant zebrafish with prominent mTORC1 activation had both increased cell size and increased proliferation [19] . To assess proliferation in dxa mutants , we analyzed the proportion of kidney and liver cells expressing proliferating cell nuclear antigen ( PCNA ) at 8 dpf . Very few PCNA positive cells were detected in control siblings but a highly increased proportion of cells express PCNA in type III dxa mutant at 8 dpf ( Figure 7A ) . We quantified these differences in the liver and found 1/246 PCNA positive cells in control larvae versus 20/245 in type III dxa mutant zebrafish , this difference was statistically significant , p<0 . 006 ( Figure 7B ) . We then treated with rapamycin from 5 to 8 dpf to verify if suppression of mTORC1 signaling could rescue aspects of the dxa phenotype . While mutants treated with rapamycin still developed a fatty liver , there was a clear decrease in cellular proliferation ( Figure 7B ) . This result suggests that mTORC1 may be considered as a potential therapeutic target for some of the pathological features seen in patients with MADD .
MADD is a complex genetic disease with multi-organ involvement and widespread biochemical abnormalities . These features likely reflect the impairment of multiple acyl-CoA dehydrogenases with each enzyme normally handling different substrates . Additional MADD complexity may be due to distinct mutations in ETFA , ETFB or ETFDH . While patients with ETFDH mutations predominate in the literature , this is possibly due to a bias of genetic testing for relatively milder forms of MADD that are compatible with longer survival . Patients with ETFA mutations in contrast may have a more severe course and rapidly succumb to this disease prior to an accurate clinical , biochemical and genetic assessment . MADD is now screened in newborns in many countries and the true prevalence of all genotypes should eventually emerge from prospective analysis of confirmed positive cases . Comprehensive analysis of MADD features and pathological mechanisms including genotype/phenotype relationships has been severely hampered by the lack of genetic animal models that recapitulates key features of MADD . In this study we analyzed a novel zebrafish model with a loss of function mutation of the etfa gene . Remarkably , dxa zebrafish recapitulates many key MADD features including biochemical abnormalities , a phenotypic spectrum from severe ( type I and II ) to moderate ( type III ) and multi-organ defects of the brain , liver and kidney . The shared phenotype of hepatic steatosis and dysmorphic kidneys seen in patients with MADD and dxa mutant zebrafish are likely due to defects of fatty acid β-oxidation as well as disruptions of amino acid and choline metabolism . C4 ( butyryl ) and C5 ( isovaleryl ) acylcarnitines and glutaric acid were highly elevated , this confirms the remarkable conservation of zebrafish and human mitochondrial function . However we did note species-specific differences . For example , patients with MADD have elevations of C14:1 but we also observed increases of fully saturated C16 and C18 in zebrafish . This suggests that the substrate availability for very long chain acyl-CoA dehydrogenase ( VLCAD ) in the zebrafish diet differs from the human fatty acid pool and that zebrafish primarily oxidize saturated fatty acids . In contrast to MADD/GAII , GA Type I is caused by mutation in glutaryl-CoA dehydrogenase ( GCDH ) , however this is one of dehydrogenases coupled to the ETF complex . Gcdh knockout mice did not have any obvious brain defects , but on a high lysine diet , these mice had neuronal loss , defective myelination and swollen mitochondria [21] , [22] . Though abnormalities of neural progenitor cells were not reported , their results suggest that accumulation of glutaric acid may be sufficient to cause defective myelination and mitochondrial abnormalities although the clinical differences between GA-I and GA-II support distinct pathological mechanisms for each disease . Strikingly , we found the severity may be caused by the nutritional state of the parents as extra feedings prior to egg fertilization produced a much higher proportion of type I and II mutants in each cross . Ongoing studies in our laboratory will investigate whether this mechanism is due to additional metabolic “stress” or from alterations of key maternal proteins , lipids or mRNA in the yolk . However , our findings indicate that a better understanding of nutrition and overfeeding may positively impact fetuses with MADD and could reduce severe congenital anomalies . The low frequency of this disease and the lack of prenatal diagnosis in families without a previously diagnosed proband makes this scenario unlikely but given the trend towards precise genetic diagnoses for all aspects of medicine , maternal diet potentially exacerbating the MADD phenotype may be a crucial finding . mTORC1 signaling is a key mediator of cell size control and differentiation . Using other zebrafish models of human disease , rapamycin treatment reversed abnormalities of cell size and mTORC1 signaling in the brain [19] . In marked contrast , brain abnormalities and other aspects of mTORC1 signaling in dxa zebrafish appeared to be rapamycin resistant . Phospho-S6 was entirely inhibited in the liver of dxa zebrafish but levels of phospho-4E-BP1 were actually elevated by rapamycin . It was previously shown that rapamycin inhibits phosphorylation of S6 and 4E-BP1 differentially [20] . This group reported that S6K and S6 phosphorylation were readily abolished throughout the duration of rapamycin treatment but phosphorylation of 4E-BP1 can recover despite initial inhibition and repeated application of rapamycin . We do not understand the mechanism leading to rapamycin resistant mTORC1 signaling in the brain and liver but speculate it may be related to increased amino acids in dxa zebrafish that could activate Rag proteins [23] . Increased leucine for example is sufficient to cause Rag GTPase dependent translocation of mTORC1 to lysosomes [24] . Isovaleric Co-A dehydrogenase requires the ETF complex and loss of function mutations in the ISOVALERYL-CoA DEHYDROGENASE ( IVD ) gene are known to cause accumulation of isovaleric acid , a metabolite of leucine [25] , [26] . Leucine accumulation in dxa may then be activating mTORC1 . We found markedly increased leucine levels in dxa larvae supporting this potential mechanism ( Figure S8 ) . We also found markedly increased p62/sequestosome 1 in dxa mutant zebrafish ( data not shown ) that was recently shown to be essential to activate mTORC1 [27] . These findings suggest that restricting intake of leucine and other branched amino acids may be important in MADD to suppress symptoms due to mTORC1 activation . However , increased aerobic glycolysis was observed in etfdh mutant zebrafish , this may compensate for a failure of mitochondrial beta oxidation [12] . Increased glycolysis may provide key intermediates for cell proliferation [28] and elevated mTORC1 signaling could further increase glycolysis by modulating transcription of genes required for this metabolic process [29] . This may represent a compensatory mechanisms and inhibition of mTORC1 with rapamycin could exacerbate the MADD phenotype or precipitate a metabolic crisis . We have seen no evidence for this in our animal model but caution against the use of mTORC1 inhibitors in patients with MADD outside of well regulated clinical trials . The myelination defects in etfa mutant zebrafish are notable given the severe neurologic deficits including encephalopathy that is usually seen in patients with MADD . Lysosomal disorders such as metachromatic leukodystrophy ( MLD ) [30] , Krabbe disease [31] and Gaucher disease [32] all have accumulation of cerebroside sulfate that appears to cause myelination defects in nerve system as well as hepatomegaly . We also see accumulation of cerebroside sulfate in radial glia and hepatocytes in dxa mutant larvae . It is possible that inhibition of autophagy by mTORC1 activation might contribute to symptoms in MADD . The markedly increased p62 levels in dxa mutants supports such a mechanism . In conclusion , we report the first animal model of MADD due to mutations of the etfa gene . Dxa mutant zebrafish larvae have an array of biochemical and pathological features that strongly indicates this is a relevant model for MADD . Dxa zebrafish can be effectively employed to generate and test further hypotheses about disease specific mechanisms . In addition , dxa mutant zebrafish will be invaluable for future in vivo chemical screens to identify therapeutic compounds that may ameliorate disease aspects of MADD and potentially other mitochondrial disorders .
Zebrafish strains used in this study included AB* and dxavu463 . Embryos were obtained from natural matings and raised at 28 . 5°C in egg water ( 0 . 3 g of sea salts/L ) . For overfeeding experiments , we gave an extra meal of TetraMin Tropical Flakes daily for one week prior to fertilization of eggs . The normal diet is twice a day meal of brine shrimp and Tropical Flakes Monday through Friday and once day on Saturday and Sunday of each week . Short term extra feeding does not cause any obvious phenotypes . 5 pairs of heterozygous siblings were used for this experiment . We fed normally one week and each pair of each was mated . Then we gave the same zebrafish extra food for the subsequent week and mated again . This cycle was repeated three times . Antisense digoxigenin-labeled RNA probe for etfa was produced using a DIG-RNA labeling kit ( Ambion ) . Embryos were fixed in 4% paraformaldehyde overnight , and dehydrated in 100% methanol at −20°C . Whole mount hybridization was performed using standard protocols [33] . BCIP/NBT ( Vector laboratories ) mixture was used as a chromogenic substrate . In situ images were acquired using a Zeiss Axioscope and Nikon Coolpix 4500 digital camera . To avoid staining variation , 3 control and 3 dxa mutant larvae were processed together in the same slide glass . Slides were processed in a Sequenza Slide Rack . Embryos were fixed in 4% paraformaldehyde from overnight to two days at 4°C . Fixed embryos were embedded in 1 . 2% agarose/5% sucrose and saturated in 30% sucrose at 4°C for 1 to 2 days . Tissue blocks were frozen in 2-methyl butane . 10 µm sections were collected on microscope slides using a Leica cryostat . Sections were kept in −80°C before use . Sections were rehydrated in 1× PBS and blocked in 5% sheep serum in PBS for 2 hours , they were then incubated with primary antibodies to Etfa ( Genetex , #GTX124324 , dilution 1∶300 ) , Sox2 ( abcam , #97959 , dilution 1∶500 ) , PCNA ( Sigma , #P8825 , dilution 1∶3000 ) , BLBP ( abcam , #ab32423 , dilution 1∶500 ) , phospho-S6 ribosomal protein ( Cell Signaling #2215 Ser235/236 , dilution 1∶300 ) , and phospho-4E-BP1 ( Cell Signaling #2855 Thr37/46 , dilution 1∶300 ) overnight at 4°C , washed 10 minutes×3 times with 1× PBS and then incubated for 2 hours with Alexa Fluor conjugated goat anti-rabbit secondary antibodies . Sections were then washed with 1× PBS for 30 minutes and mounted in Vectashield with DAPI ( Vector laboratories ) . Antigen retrieval for PCNA staining was performed for 30 minutes of boiling in 10 mM sodium citrate before blocking . Images were acquired using Zeiss Axiovert 200M microscope with Zeiss AxioCam MRm and Hamamathu digital camera . Digital images were processed using Adobe Photoshop CS5 and Adobe illustrator CS5 . All images received only minor modifications with control and mutant sections always processed in parallel . Fixed samples were rinsed with PBS-DT ( 1× PBS , 0 . 5% Triton X-100 , 2% DMSO ) and both control and mutant were incubated with blocking solution ( PBS-DT , 5% goat serum ) for 2 hours at room temperature in a single tube . The antibody against acetylated-tubulin ( Sigma , #T7451 , dilution 1∶500 ) was used overnight at 4°C . Larvae were rinsed with PBS-DT 3 times ( 10 minutes each ) . Secondary was a goat Cy3-anti-mouse for overnight at 4°C . Specimens were rinsed with 700 µL of PBS-Tween for 10 minutes and repeated 5 times . Zebrafish were fixed in 4% PFA and mounted in glycerol before being imaged . For whole mount staining at larvae stage , larvae were fixed in 4% PFA overnight . Control and dxa mutant larvae were rinsed three times ( 5 minutes each ) with 1× PBS/0 . 5% Tween-20 ( PBS-Tween ) . After removing PBS-Tween , larvae were stained with mixture of 300 µL of 0 . 5% ORO in 100% isopropyl alcohol and 200 µL of distilled water for 15 minutes . Larvae were then rinsed with 1× PBS-Tween for three times . Larvae were rinsed twice in 60% isopropyl alcohol for 5 minutes each . They were briefly rinsed in PBS-Tween and fixed in 4% PFA for 10 minutes . Larvae were mounted in glycerol prior to imaging . For high resolution ORO staining on transversely sectioned larvae , 10 µm sections were dried at room temperature for 5 minutes . 150 µL of working ORO solution was added to slides and stained for 30 seconds . They were then washed with tap water and mounted using Vectashield with DAPI . For free cholesterol staining on transversely sectioned larvae , slides were soaked with 1× PBS for 5 minutes , then Filipin complex diluted 1∶500 ( Sigma , F-976 ) was added directly to slides and stained for 1 minute in the dark . Slides were washed with PBS and mounted with 75% glycerol . Images were taken using the DAPI channel of a fluorescent microscope . Frozen sections were used for PAS staining . The PAS stain was conducted in the Translational Pathology Core laboratory at Vanderbilt University using a DAKO Artisan Link Staining System . Glycerophospholipids from zebrafish larvae were extracted using a modified Bligh and Dyer procedure [34] . Forty of 8 dpf larvae of each genotype , either mutant or sibling control were homogenized in 800 µl of ice-cold 0 . 1 N HCl∶CH3OH ( 1∶1 ) using a tight-fit glass homogenizer ( Kimble/Kontes Glass Co , Vineland , NJ ) for about 1 minute on ice . The suspension was then transferred to cold 1 . 5 mL Eppendorf tubes and vortexed with 400 µl of cold CHCl3 for 1 min . Centrifugation ( 5 minutes at 4°C , 18 , 000× g ) to separate the two phases . Lower organic layer was collected , an odd carbon internal standard was added and solvent evaporated . The resulting lipid film was dissolved in 100 µl of isopropanol∶hexane∶100 mM NH4COOH ( aqueous ) 58∶40∶2 ( mobile phase A ) . Quantification of glycerophospholipids was achieved by the use of an LC-MS technique employing synthetic odd-carbon diacyl and lysophospholipid standards . Typically , 200 ng of each odd-carbon standard was added per sample . Glycerophospholipids were analyzed on an Applied Biosystems/MDS SCIEX 4000 Q TRAP hybrid triple quadrupole/linear ion trap mass spectrometer ( Applied Biosystems ) and a Shimadzu high pressure liquid chromatography system with a Phenomenex Luna Silica column ( 5-µm particle size ) using a gradient elution as previously described [35] , [36] . Individual species were identified based on their chromatographic and mass spectral characteristics . This analysis allows identification of the two fatty acid moieties but does not determine their position on the glycerol backbone ( sn-1 versus sn-2 ) . Neutral lipids from zebrafish ( forty of 8 dpf larvae/sample ) were extracted by homogenization in the presence of internal standards ( 500 ng 14∶0 monoacylglycerol and 24∶0 diacylglycerol and 1 µg 42∶0 triacylglycerol ) in 2 ml 1× PBS and extracting with 2 mL ethyl acetate∶trimethylpentane ( 25∶75 ) . A dried lipid film was dissolved in 1 mL hexan∶sopropanol ( 4∶1 ) and passed through a bed of Silica gel 60 Å to remove remaining polar phospholipids . Solvent from the collected fractions was evaporated and lipid film was redissolved in 90 µl 9∶1 CH3OH∶CHCl3 , containing 10 µl of 100 mM CH3COONa for MS analysis essentially as previously described [36] , [37] . Samples were analyzed in triplicates and p-values determined using Student's t-test . Forty 9 dpf control and dxa mutant larvae were lysed using pellet pestles ( Sigma , #Z359947 ) and passed through a 25 gauge syringe in 150 µL of PBS . For acylcarnitine analysis , the total lysate was placed into a 96 well plate containing stable isotope labeled internal standards ( Cambridge Isotope Laboratories , Andover , MA ) and acylcarnitine analysis performed according to the published methods [38] . Briefly , the lysate was dried under nitrogen , reconstituted with fifty µL of acetonitrile and one µL was injected into a Xevo-TQS tandem mass spectrometer ( Waters Corp . Waltham , MA ) . Acylcarnitines were quantified against an isotope–labeled internal standard of the nearest chain-length using the parent ions of the carnitine-specific fragment of m/z 85 . For organic acid analysis , the total lysate was made up to a final volume of 2 . 5 mL using deionized water , acidified to pH 2 . 0 and the acid fraction extracted three times into equal volumes of ethyl acetate . The pooled organic phases were dried down under a stream of nitrogen at room temperature and trimethylsilyl derivatives analyzed by gas chromatography-mass spectrometry using an Agilent 7890A gas chromatograph fitted with a 5975C Mass Selective Detector ( Agilent Technologies , Santa Clara , CA ) using a method initially developed for urine and vitreous humor analysis [39] . The acylcarnitine and organic acid assays are validated and in routine clinical use and were also previously used for analyses of etfdh mutant zebrafish [12] . In brief , samples were fixed in 2 . 5% gluteraldehyde for 1 hour then transferred to 4°C overnight . Samples were washed 3 times in 0 . 1 M cacodylate buffer then incubated for 1 hour in 1% osmium tetraoxide and washed with cacodylate buffer . Samples were dehydrated through a graded series of ethanol , then incubated in ethanol and propylene oxide ( PO ) . Samples were infiltrated with 25% Epon 812 resin and 75% PO for 35 minutes , then 50% Epon 812 resin and 50% PO for 1 hour then exchanged with new 50% Epon 812 resin and 50% PO and incubated overnight . Samples were exchanged with 75%: 25% ( resin: PO ) , then pure epoxy resin for 3–4 hours , then overnight . Finally , the resin was exchanged with epoxy resin for 3 hours , embedded in epoxy resin and polymerized at 60°C for 48 hours . Sectioning and Imaging: 500 nm to 1 µm thick sections were collected using a Leica Ultracut microtome . Thick sections were stained with 1% toluidine blue and . 70–80 nm ultra-thin sections were cut from this block and collected on 300-mesh copper grids and stained with 2% uranyl acetate ( aqueous ) for 16 minutes and then with lead citrate for 12 minutes . Samples were imaged on the Philips/FEI Tecnai T12 electron microscope at various magnifications . Average basal oxygen flux was quantified by high resolution respirometry using the Oroboros O2k Oxygraph ( Oroboros Instruments , Innsbruck , Austria ) . Ten larvae ( 8 dpf ) per chamber were maintained in Instant Ocean at 28°C and initially equilibrated to room air . Measurements of oxygen concentration were recorded every 2 seconds with no stirring . When the measured O2 concentration stabilized , chambers were stirred at 100 rpm for as short a time as possible to permit recording of a new stable O2 concentration , which was reflective of the true O2 concentration in solution . Stirrers were then turned off . This process was repeated until no further decrement in O2 concentration was measured and the fish were no longer motile . Average oxygen flux was calculated from the change in O2 concentration over time from the beginning of the experiment to the end . Data are from 4 measurements made with 10 larvae of each genotype . Forty control siblings and homozygote dxa mutant larvae from 3 to 5 clutches were homogenized in 100–750 µL of 0 . 1 M TCA , containing 10 mM sodium acetate , 100 µM EDTA , 5 ng/ml isoproterenol as an internal standard and 10 . 5% methanol at pH 3 . 8 . Samples were centrifuged at 10 , 000× g for 20 minutes . Supernatant was removed and stored at −80 degrees . Samples of the supernatant were then analyzed for biogenic monoamines and/or amino acids . Leucine was quantified with a Waters AccQ-Tag system with a Waters 474 Scanning Fluorescence Detector . Ten µL samples of the supernatant are diluted with 70 µL of borate buffer to which 20 µL aliquots of 6-Aminoquinol-N-Hydroxysuccinimidyl Carbamate and 10 µL 250 pmol/µL α-aminobutyric acid ( as internal standard ) are added to form fluorescent derivatives . 10 µL of sample was then injected into the HPLC system , and separation of the amino acids accomplished by means of a Waters amino acid column and supplied buffers using a specific gradient profile . Error bars in Figure S3 and S6 represent standard error of the mean ( SEM ) , error bars in Figure 7 and S4 represent standard deviations ( SD ) . Student's t-test was used to determine statistical significance . All animal experiments were done with the approval of the Vanderbilt University IACUC . | Mitochondrial disorders have multiple genetic causes and are usually associated with severe , multi-organ disease . We report a novel zebrafish model of mitochondrial disease by inactivating the etfa gene . Loss of this gene in humans causes multiple acyl-Co dehydrogenase deficiency ( MADD ) that manifests with brain , liver , heart , and kidney disease . While presentations are variable , many children with MADD have a severe form of the disease that rapidly leads to death . We report that etfa gene function is highly conserved in zebrafish as compared to humans . In addition we uncovered potential disease mechanisms that were previously unknown . These include the impact of maternal nutrition on disease severity in their offspring as well as the role mTOR kinase signaling . Inhibition of this kinase with the drug rapamycin partially reversed some of the symptoms suggesting this may be a new approach to treat mitochondrial disorders . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"metabolic",
"disorders",
"biology",
"medicine"
] | 2013 | Multi-organ Abnormalities and mTORC1 Activation in Zebrafish Model of Multiple Acyl-CoA Dehydrogenase Deficiency |
In mammals , the synaptonemal complex is a structure required to complete crossover recombination . Although suggested by cytological work , in vivo links between the structural proteins of the synaptonemal complex and the proteins of the recombination process have not previously been made . The central element of the synaptonemal complex is traversed by DNA at sites of recombination and presents a logical place to look for interactions between these components . There are four known central element proteins , three of which have previously been mutated . Here , we complete the set by creating a null mutation in the Syce1 gene in mouse . The resulting disruption of synapsis in these animals has allowed us to demonstrate a biochemical interaction between the structural protein SYCE2 and the repair protein RAD51 . In normal meiosis , this interaction may be responsible for promoting homologous synapsis from sites of recombination .
Meiosis is a specialised process in which the replicated diploid genome undergoes two rounds of cell division without an intervening DNA replication . Production of haploid gametes from the diploid germ line is a complex process requiring the accurate separation of the two parental genomes to avoid the aneuploidy which would result from errors . Meiotic recombination imposes the additional requirement that the two genomes be precisely aligned for exchange of genetic information . In organisms from budding yeast to humans a key component of the meiotic cellular machinery used to enforce this is the synaptonemal complex ( SC ) . This is a widely occurring , proteinaceous structure which physically links the pairs of sister chromatids ( for review see [1] ) and is visualised in the electron microscope as a zipper like structure with two lateral elements ( LE ) and the central element ( CE ) in between . Lateral elements are derived from axial elements ( AE ) that connect sister chromatids after premeiotic DNA replication . To date , numerous protein components of the SC have been defined in a variety of organisms ( reviewed in [1] ) . They can be classified as components either of the LE/AE or of the CE . In mammals AE proteins include cohesins and coiled coil domain proteins such as SYCP3 and SYCP2 [2]–[4] . The CE contains the recently described proteins SYCE1 , SYCE2 and TEX12 [5] , [6] . SYCP1 is a key protein , which links AEs to the CE through its central coiled coil domain and by having C and N terminal globular domains anchored in AE and CE respectively [7]–[9] . In many organisms the formation of the SC is dependent on double strand breaks ( DSBs ) which can be processed to crossover or , more frequently , non crossover pathways . The SC may play a role in regulating the non random distribution of crossovers known as interference . However the requirement for and intact SC is sexually dimorphic in mice and it is not required for interference in female meiosis [10] . In male mice the fully assembled SC is required to complete crossover recombination and genetic exchange . Mutations in axial element components Sycp2 and Sycp3 result in failure of SC formation and infertility in the male . Milder meiotic defects in female meiosis result in increased aneuploidy and reduced litter sizes [11]–[13] . To date mutagenesis of known components of the CE in mouse suggest that an intact CE is required in both sexes . In Sycp1 null mice synapsis is completely abolished and although the MSH4 foci indicative of intermediate stages of recombination are present neither sex forms the MLH1 foci , which are the cytological markers of crossover , and both sexes are infertile [14] . Syce2 null mice , in which the axial elements align but do not synapse , also do not form MLH1 foci in either sex although again proteins indicative of earlier stages of the recombination process such as RAD51 and MSH4 are present [15] . TEX 12 , a central element protein which interacts with SYCE2 , has recently been shown to have a similar null phenotype with the absence of crossover recombination in both sexes [16] . Since these proteins are mutually dependent for localisation to and formation of the CE this similarity is not surprising . Based on known interactions between SYCP1 , SYCE1 , SYCE2 and TEX12 ( Figure S1 ) we have suggested that the assembly of the SC is a multi-step process which is blocked at different stages by the absence of SYCE1 and 2 and probably TEX12 [15] . In the presence of SYCE2 and the absence of SYCE1 the prediction is that points of synapsis , as observed in the Syce2−/− animals , do not occur . Here we report the phenotype of such mutant animals . Importantly this phenotype has suggested interactions between these structural components of the SC and the recombination machinery .
We disrupted the mouse Syce1 gene by gene targeting in AB2 . 2 ES cells . The targeting vector was designed to replace exons 2–11 of the Syce1 gene with the LacZ- Neor selection cassette ( Figure S2A ) . Correct targeting was confirmed by Southern Blot analysis ( Figure S2B ) . Correctly targeted ES cells were injected into C56BL/6 blastocysts and produced two germline transmitting chimeras . Offspring produced by mating these chimeras to C56BL/6 females were genotyped by PCR ( Figure S2C ) and Syce1+/tm1HGU animals intercrossed . Animals were produced from these matings with all genotypes in Mendelian ratios . To confirm the absence of the SYCE1 protein in the Syce1tm1HGU /tm1HGU ( Syce1−/− ) animals we used Western blotting . A polyclonal antibody raised against C-term of SYCE1 detects a protein band of the expected size ( 45 KDa ) in wild-type testis extracts but not in the Syce1−/− , confirming the specificity of antibodies as well as indicating that the Syce1 disruption described here results in a null mutation ( Figure S2D ) . The lack of detectable proteins demonstrates the absence of splicing between the Neor gene and remaining Syce1 exons which might produce truncated proteins . Syce1−/− mice are infertile . Mating of both sexes with wild-type animals failed to yield any offspring although Syce1−/− males produced copulatory plugs suggesting normal sexual behaviour . Syce1 mutant ovaries were minute and testes size was only 20–30% of wild-type littermates , which is similar to other meiotic mutants [12] , [14]–[16] . We observed no phenotypes in other tissues of these animals . Histological analysis of adult Syce1−/− gonads revealed an almost complete lack of follicles in ovaries ( Figure 1A ) , suggesting a disruption during meiosis followed by apoptosis , and lack of postmeiotic cells in the testis ( Figure 1B ) . Primary spermatocytes were the most common germ cell type indicating a spermatogenesis arrest at prophase I . Elevated levels of apoptosis were detectable in some tubules by TUNEL staining ( Figure 1B , insets ) suggesting that arrested cells are eliminated by this mechanism . The high number of positive cells in a fraction of tubules indicates that most of the cells undergo apoptosis at the same epithelial stage , which was determined to be stage IV ( data not shown ) . Syce1−/− females show a meiotic prophase phenotype similar to males indicating that SYCE1 plays the same role in both male and female meiosis . The lack of mature gametes is consistent with the expected role of SYCE1 protein in meiosis and demonstrates that Syce1 is an essential gene for both male and female fertility . To investigate the cause of the meiotic defect in more detail we prepared surface spread chromosomes from Syce1−/− spermatocytes . Normally during meiotic prophase I homologous chromosomes are closely juxtaposed and are then physically connected by the SC along the entire length of chromosome axes . Immunostaining for SYCP3 , SYCP2 and STAG3 proteins revealed that AEs are formed normally in the absence of SYCE1 ( Figure 2 and S3 ) and that homologous chromosomes align in close juxtaposition . The sex chromosomes are an exception to this; as in Sycp1 , Tex12 and Syce2 null mutants the pseudoautosomal regions do not pair and a sex body is not formed ( Figure 2D , arrows ) . Wild-type spermatocytes at pachynema are characterised by the presence of ribbon-like structures seen by staining for SYCP1 . These represent fully formed SCs linking homologous chromosomes ( Figure 2A ) . In Syce1−/− cells , although AEs are formed and aligned SCs do not assemble between them as indicated by the absence of continuous SYCP1 staining ( Figure 2B , D ) . Interestingly a weak discontinuous SYCP1 signal was observed associated with AE whether they are closely aligned or not ( Figure 2B , D ) . We used immunostaining for SYCE2 and TEX12 , two other markers of synapsis that in the wild-type co-localise with SYCP1 ( Figure 2E ) to further investigate synaptic failure . Although SYCE2 and TEX12 foci co-localise as expected , immunostaining for SYCE2 or TEX12 does not resemble that of the wild-type animals . Instead they were found in intermittent foci between closely aligned AEs ( Figure 2F ) . This is consistent with the observations that their localisation to the SC is co-dependent and their known interactions ( Figure S1 ) [6] , [15] , [16] . Unlike in wild-type spermatocytes , in Syce1−/− spermatocytes SYCE2 does not always follow SYCP1 signal either locally within a pair of homologs or globally in one nucleus ( Figure 2D , B respectively ) . A subset of cells shows accumulation of SYCP1 on both AEs without accompanying SYCE2 , suggesting that the SYCP1 C-terminal region can bind to AEs in the absence of SYCE1 . Additionally in Syce1/Syce2 double knockout SYCP1 still binds to aligned AEs suggesting that it is the presence of SYCE1 that restricts SYCP1 binding to synapsed axes when all components are present ( not shown ) . Syce1−/− oocytes display very similar defects in chromosome synapsis to males ( Figure 2G–H ) . AE are fully formed and homologous chromosomes align , however tripartite synaptonemal complex is not formed along the length of chromosomes . In some cases AEs are in very close apposition along their length with spacing similar to that of the normal SC with SYCE2 and SYCP1 co-localised between them . In order to determine whether these sites of co-localisation of CE proteins represent SC formation we have performed electron microscopy on testis sections from Syce1−/− animals . Extensive analysis of the mutant material revealed presence of parallel AEs but failed to find any signs of the CE ( Figure 3 ) . This is in contrast to the Syce2 or Tex12 nulls , where CE-like structures were observed [15] , [16] . Based on the observations from all three mutants we propose that the SYCE1 protein is required not only to stabilise SYCP1 dimers within central element but also to stack the transverse filaments into layers to form CE and determine the thickness of the SC . Meiotic recombination is initiated by SPO11-mediated double strand breaks ( DSB ) [17] . The generation and the repair of these breaks are required for chromosomal synapsis in most organisms including mammals [18]–[21] . The appearance of these breaks is accompanied by the phosphorylation of histone H2AX on large domains of chromatin around the break . As meiosis proceeds to the pachytene stage γH2AX is removed from synapsed chromosomes and is restricted to the largely asynapsed sex chromosomes in the XY body [22]–[24] ( Figure 4A ) . Syce1−/− spermatocytes showed extensive γH2AX staining in early cells that persisted to the most advanced spermatocyte stages ( Figure 4B ) ( in these animals the sex body does not form ) . Oocytes show the same pattern of staining ( Figure 4J ) . This suggests that DSB are generated in the Syce1−/− mutants but are not efficiently repaired . To assess the state of DSB repair in mutant spermatocytes and oocytes we analysed the distribution of proteins involved in different steps of meiotic repair and recombination [25] , [26] . First the strand exchange proteins RAD51 and DMC1 are recruited to the sites of DSB and form early recombination nodules ( EN ) . RAD51/DMC1 mediate the homology search and the single end invasion of the homologous chromosome [27] . Cytologically , RAD51 and DMC1 manifest as numerous foci along chromosome cores , typically several hundred occur in a mouse meiotic nucleus [28] . During normal meiosis numbers of RAD51/DMC1 foci peak at leptonema and disappear by mid-pachynema except along asynapsed cores of sex chromosomes in males ( Figure 4C and K ) . RAD51 foci are highly abundant in both Syce1−/− spermatocytes and oocytes and are localised to both aligned and unaligned chromosome cores ( Figure 4D and L ) . Fifteen percent of cells lack RAD51 foci entirely . The MutS homologs MSH4 and MSH5 have been proposed to function in stabilization or resolution of recombination intermediates and possibly also during synapsis in earlier stages of prophase I [29]–[31] . In normal meiosis MSH4 foci appear concurrently with synapsis at early zygotene , peaking at late zygotene and starting to decrease at early pachytene ( Figure 4E and M ) . In Syce1−/− spermatocytes and oocytes MSH4 foci appear without synapsis and are found only between aligned chromosome cores ( Figure 4F and N ) . This indicates that MSH4/MSH5 mediated DNA-DNA interactions between homologous chromosomes can occur in the absence of SYCE1 . Spermatocytes of mice lacking other proteins such as SYCP1 and SYCE2 which are required for synapsis also have MSH4 foci . After MutS homologs MSH4/MSH5 associate with DNA a complex of MutL homologs MLH1/MLH3 is recruited to sites now termed late recombination nodules ( RN ) . Together they are implicated in the processing of DSB through the double Holliday junction ( dHJ ) recombination intermediates that result in crossover . Mlh1 was shown to be essential for crossover formation in mammals and yeast [32]–[34] . In wild-type meiosis MLH1 appears at late prophase in pachytene and is present in a few sites that correspond in number and distribution to the number of crossover events estimated genetically [35] ( Figure 4G and O ) . We stained Syce1−/− spermatocytes and oocytes with an anti-MLH1 antibody and failed to observe any MLH1 foci ( Figure 4H and P ) . This indicates that despite MSH4 associated recombination intermediates MLH1 can not be recruited to resolve them into crossover in the absence of SYCE1 and synapsis or that cell death occurs before that stage . Taken together , analysis of the progress of meiotic recombination suggests that SYCE1 is dispensable for the initiation of recombination but is essential for stable homologue interactions mediated by the SC and crossover formation . Recombination and synapsis are co-dependent and physically linked in yeast where synapsis is initiated at sites of recombination destined to be crossovers [36] , [37] . To our knowledge no such link has been described in the mouse . In Syce1−/− spermatocytes we noticed that the pattern of SYCE2/TEX12 foci between closely juxtaposed AEs resembles that of RAD51 . To confirm our observations we immunostained Syce1−/− spermatocytes with anti-SYCE2 and -RAD51 antibodies . A subset of cells ( 42% , n = 435 ) with high number of RAD51 foci ( approximately two hundred per nucleus ) did not have any SYCE2 staining ( Figure S4 ) However , cells with approximately half the number of RAD51 foci , located between aligned AE , showed co-localised staining for SYCE2 ( 43% n = 435 ) ( Figure 5B ) . SYCE2 was almost always accompanied by a RAD51 signal in these cells ( Figure 5B , lower panel in offset ) . To test if this co-localisation reflects a biochemical interaction between SYCE2 and RAD51 we used immunoprecipitation ( IP ) from wild-type and Syce1−/− testicular extracts . We have immunoprecipitated proteins using both anti-SYCE2 antibody and preimmune serum as a control , and checked for interacting proteins by probing western blot with anti-RAD51 antibodies . We were able to detect RAD51 as a band of approximately 37 KDa in the input as well as weakly in the wild-type and Syce1−/− IP samples but not in the control ( Figure 6A ) . As a further control we have used Syce2−/− testis extract for IP with anti-SYCE2 antibodies and failed to detect RAD51 ( Figure 6B ) . To check if this interaction is specific and not due to the precipitation of the whole SC we tested SYCE2 IP samples with antiSYCP3 antibodies and did not detect SYCP3 in the immunoprecipitated sample ( Figure 6C ) . Although we detect SYCE2 and RAD51 in the same complex we can not and do not conclude that this interaction is direct . Our attempts to demonstrate that using an in vitro assay have been inconclusive due to insolubility of proteins when co-overexpressed or to RAD51-GST interactions in pull down reactions . We proceeded to check if SYCE2 also co-localises with MSH4 which appears when chromosomes synapse and which succeeds RAD51 in the recombination nodules . Co-immunostaining of Syce1−/− spermatocytes for SYCE2 and MSH4 revealed that these two proteins only partially co-localise . ( Figure 5D , and inset ) . There are different classes of cells: one which has only SYCE2 signals and no MSH4 ( 7 . 5% , n = 189 , not shown ) , another which stains for both ( 36% , n = 189 ) ( Figure 5D ) and the remaining largest group shows only MSH4 foci ( 50% , n = 189 ) ( Figure S4 ) . This would suggest that as RAD51 is displaced by MSH4 , SYCE2 is no longer associated with chromosomes in the Syce1−/− animals . Altogether , this data suggests that central clement protein SYCE2 interacts , directly or indirectly , with the recombination protein RAD51 . Is synapsis dependent on the RAD51/SYCE2 interaction ? Spo11 null mice are unable to generate meiotic DSB and as a result RAD51 is absent from the nucleus . Despite this , various degrees of synapsis , mostly nonhomologous , were observed in the Spo11 null , on the basis of SYCP1 staining [20] , [21] . We have stained Spo11−/− spermatocytes for SYCE1 and SYCE2 to check if these proteins are components of this DSB independent synapsis . Our results show that both SYCE1 and SYCE2 co-localise with SYCP1 on the SC in the Spo11 mutants indicating that apparently normal synapsis can form in the absence of RAD51 and DSB ( Figure S5 ) , but in this case between random chromosomes .
Successful completion of meiosis in mouse depends on the assembly of the SC . Recent work using targeted mutagenesis to make null mutations in three ( Sycp1 , Syce2 and Tex12 ) of the four known protein components of the CE has shown that the CE is a critical component of this structure [14]–[16] . Here we complete the set by mutating the remaining known component SYCE1 . As predicted from the known multiple interactions of the proteins ( Figure S1 ) Syce1−/− animals have a phenotype which is very similar to that of the other three null mutations . DNA repair is incomplete , the SC and the sex body are absent , homologous alignments at variable distances of the AEs occur , early ( RAD51 ) and intermediate ( MSH4 ) markers of recombination are present but there is a complete absence of MLH1 marking crossovers . In the testis cells are eliminated by apoptosis and both sexes are infertile . Complete assembly of the SC is co-dependent on the presence of all four proteins ( SYCP1 , SYCE1 , SYCE2 and TEX12 ) and perhaps on others as yet undiscovered . However the mice null for different CE components are likely blocked in different states of SC assembly and provide tools to dissect this essential process . There are distinct features of the Syce1−/− phenotype . In the absence of SYCE1 transverse filament protein SYCP1 binds to AEs when they are closely aligned and presumably forms N-termini associations [9] . This may reflect the protein's ability to form polycomplexes with dimensions corresponding to SCs [38] . However SYCP1 is also associated with AEs that are further apart confirming the proposal in our model that SYCP1 N-terminal associations alone are insufficient to promote SC assembly and require SYCE1 for stability in physiological conditions . The extensive association of SYCP1 with AEs in the Syce1−/− animals suggests that SYCE1 could play a role in restricting SYCP1 binding in wild-type synapsis . These associations with unpaired AEs are absent in the Syce2−/− and Tex12−/− males where SYCE1 is present [15] , [16] . The Syce1−/− phenotype further supports the idea that SYCE2 and TEX12 act in concert . From published data we know that their localisation to the SC is co-dependent [15] , [16] and in the absence of SYCE1 ( this paper ) both SYCE2 and TEX12 co-localise as foci between aligned AEs , therefore their recruitment to chromosome axes is SYCE1 independent . Previously , in our model for synaptonemal complex assembly we suggested that SYCE1 stabilises N-terminal interactions of SYCP1 in the CE and that SYCE2/TEX12 is required for the elongation of the SC . The Syce1−/− phenotype is consistent with this model . Given the presence of three out of four CE components and interactions between SYCP1 and SYCE2 we expected some form of CE to be present in Syce1−/− spermatocytes as found in Syce2−/− and Tex12−/− spermatocytes . Our extensive analysis of testis sections at the EM level failed to detect a CE . Our model for CE assembly was two dimensional , reflecting observations in the light microscope and in EM sections but the SC has a thickness which we had not taken into account and of which SYCE1 may be a component [39] . In a revised model although the three CE proteins ( SYCP1 , SYCE2 and TEX12 ) co-localise they do not produce a visible CE in the microscope due to the absence of multiple layers of proteins dependent on SYCE1 . We propose that SYCE1 stabilises the N-termini associations of SYCP1 ( width ) and regulates formation of transverse filament stacking ( thickness ) in addition to being required for SC extension through its interactions with SYCE2 and SYCP1 . Studies of the SC functions in various organisms revealed that the SC is essential for normal progression of meiotic recombination and formation of crossovers in yeast , plants and mammals [14] , [40] , [41] . It has been also shown that proper assembly of the SC between homologous chromosomes depends on recombination . In the absence of the SPO11 induced DSBs that initiate recombination , levels of SC formation are highly reduced or form between nonhomologous chromosomes [20] , [21] . Additionally , the correct processing of DSBs at the early stages of recombination is essential for synapsis to occur [29] , [31] , [42] , [43] . Impaired recombination in mouse mutants lacking the CE points to the possibility that interactions between the structural components of the CE and the recombination machinery occur and are essential for crossover . Prior to synapsis the recombinase RAD51 is recruited to the DSBs and disappears as chromosomes synapse . In mutants that lack the SC RAD51 persists longer and is associated with the AEs . It is not possible to study the function of RAD51 in meiosis due to embryonic lethality of the Rad51 mutation [44] . However , the phenotypes of recently reported mutations in the Tex15 and Tex11 ( Zip4H ) genes show that both recruitment as well as timely disappearance of RAD51 are crucial for synapsis and meiotic recombination . In the Tex15 mutant RAD51 foci are highly reduced in number whereas in the Tex11 ( Zip4H ) mutant the number of these foci increases , probably as a result of delayed processing of DSB . Both mutants show synapsis defects . In Tex11 null some chromosomes do not synapse at all and in Tex15−/− spermatocytes synapsis is completely abolished . As a result the number of MLH1 foci present in spermatocytes is reduced or eliminated , respectively [45]–[47] . In wild-type meiosis several different types of structures containing recombination proteins have been described based on immuno-histochemsitry . In leptotene RAD51/DMC1 foci have been termed early nodules ( EN ) , later they begin to contain RPA in addition to RAD51/DMC1 and when synapsis is complete RAD51 is absent in RPA containing transition nodules ( TN ) . The MLH1 containing recombination nodules ( RN ) appear last [26] . Based on our observation that SYCE2 and RAD51 co-localise in a subset of the Syce1−/− spermatocytes and that interactions between these proteins can be detected in testis extracts we propose that this interaction promotes synaptonemal complex assembly/extension . From a yeast two hybrid assay and in vitro pull down experiments it was previously suggested that SYCP1 interacts with RAD51 but not with DMC1 [48] . SYCP1 was also shown to recruit SYCE1 and SYCE2 to the SC as these proteins are not chromosomally localised in Sycp1−/− spermatocytes [5] , [6] and hence must be involved in the RAD51/SYCE2 interaction . Although all four CE proteins are needed for complete synapsis , structures suggestive of sites of initiation of synapsis can be seen at both light and electron microscope resolution in the absence of SYCE2 or TEX12 but not in the absence of SYCE1 . In the SYCE1 null animals we observe co-localisation of SYCE2 and RAD51 which we suggest occurs in normal mouse meiosis but is obscured by the subsequent rapid assembly of the SC . This concentration of SYCE2 may function to promote SC extension . We can not exclude that TEX12 , a SYCE2 binding partner , plays a specific role in its interaction with RAD51 . Interestingly , it was shown that in DSB deficient mutants , when breaks are introduced artificially , the number of RAD51 foci representing induced DSB correlate with the extent of synapsis [49] . This also points out the link between RAD51 and synapsis . However , it seems that RAD51 is not required in Spo11 mutants for initiation and partial assembly of the SC [20] , [21] but in these animals the SC is not formed between homologous chromosomes . Perhaps the presence of RAD51 at the sites of DSB favours the extension of homologous SC assembly over that of non homologous SC in a competitive and ( in terms of aneuploidy ) potentially disastrous situation . Feedback from SC assembly must be required for the maturation of a small set of TN into the RN marking sites of recombination . The combination of cytology and enzymology has pointed to the ability of cellular structures to recruit and perhaps modify the function of repair enzymes for use in meiosis [50] . Our results here suggest that this process may also operate in the reverse direction with repair proteins playing a role in the assembly of structures essential for meiosis and fertility .
To inactivate the Syce1 gene , we designed a targeting vector to replace exons 2–11 by selection cassette . This construct was based on a modified pBluescript vector containing DTA cassette , En2SA-IRES-LacZ-pA and floxed tk-NEO gene . A 5 . 2 kb ApaI fragment containing part of intron 1 of the mouse Syce1 gene was cloned between DTA and LacZ-Neo cassettes and a 2 . 2 kb SacI fragment containing exons 12–13 of the Syce1 gene was cloned downstream of Neo cassette . The linearised Syce1 targeting construct was electroporated to AB2 . 2 ES cells . After selection with G418 ES cell clones were screened by PCR ( FP: CAACCTCCCTCACCACCTTA , RP: TTGCTGAAGTTGTGCCAGAC ) . Potential positive clones were expanded and DNA was extracted for Southern blot analysis . DNA was digested with EcoRI and hybridised with external probe ( See Figure S2 ) . Cells from one of the correctly targeted ES clones were injected into C57/B6 blastocysts to obtain chimeras . Chimeric males were mated to C57/B6 females and progeny was genotyped using primers ( FP:CCAGAAGCCTGAACATCTGACA , RP:TACCATCCTCCATGAGCTGTCT , Neo:AGGACATAGCGTTGGCTACCC ) . To produce Syce1ko mice we intercrossed heterozygous offspring . Tissues for histological examinations were dissected and fixed in Bouin's fixative . Subsequently , tissues were embedded in paraffin and 6 µm sections were cut . Mounted sections were deparaffinised , rehydrated , and stained with hematoxylin and eosin . Apoptosis was assayed using DeadEnd Fluorometric TUNEL System ( Promega ) according to the manufacturer's protocol Spread chromosomes from males and females were prepared and stained as previously described [5] , Images were captured using a system comprising a charge-coupled device camera ( Orca-AG; Hamamatsu ) , a fluorescence microscope ( Axioplan II; Carl Zeiss MicroImaging , Inc . ) with Plan-neofluar objectives ( 100× NA 1 . 3 ) , a 100-W Hg source ( Carl Zeiss MicroImaging , Inc . ) , and quadruple band-pass filter set ( model 86000; Chroma Technology Corp . ) , with the single excitation and emission filters installed in motorised filter wheels ( Prior Scientific Instruments ) . Image capture was performed using in-house scripts written for IPLab Spectrum ( Scanalytics ) . Images were processed using Adobe Photoshop . Electron microscopy was performed using ultra thin sections of testis tissue fixed in 2 . 5% glutaraldehyde and 1% OsO4 as described previously [51] . The primary antibodies used were: rabbit anti-SYCE1; rabbit anti-SYCE2 [5]; guinea pig anti-SYCE1; guinea pig anti-SYCE2; guinea pig anti-TEX12 [6]; rabbit anti-SYCP1 ( Abcam ) ; mouse anti-SYCP3 [52]; rabbit anti-SYCP3 ( Abcam ) ; rabbit anti-STAG3 [53]; rabbit anti-SYCP2 [54]; rabbit anti-γH2AX ( Upstate Biotechnology ) ; mouse anti-Rad51 ( Upstate Biotechnology ) ; mouse anti-MLH1 ( BD Biosciences ) ; rabbit anti-Msh4 ( Abcam ) . Secondary antibodies used were Alexa Dyes ( AlexaFluor-488 , 594 and 647 ) conjugates ( Molecular Probes ) . Protein extraction , immunoprecipitation and detection were carried out as previously described [5] | Production of sperm and eggs , also known as gametes , requires a reduction in the number of copies of the genome , from the two found in most cells of the body to the single copy found in gametes . This is a complex process , made even more complex because it is coupled with recombination , a process that is an important contributor to genetic diversity . Mammals and many other organisms achieve reduction and recombination through a process called meiosis , which is recognisable by the presence of a distinctive structure—the synaptonemal complex—that links the chromosomes together and is essential for meiosis to complete . We have made mice that lack SYCE1 , a protein component of the synaptonemal complex . In these animals , meiosis is blocked at a particular stage , and this has allowed us to detect co-localisation and interactions—likely indirect—between enzymes involved in recombination and structural proteins involved in meiosis . This provides a starting point to understand in biochemical detail the protein links between structure and function in meiosis . Mutations or variants in the genes encoding such proteins are likely contributors to variations in fertility and to abnormalities in chromosome number . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology/germ",
"cells",
"cell",
"biology",
"genetics",
"and",
"genomics/chromosome",
"biology"
] | 2009 | Mutation of the Mouse Syce1 Gene Disrupts Synapsis and Suggests a Link between Synaptonemal Complex Structural Components and DNA Repair |
Ebola virus emerged in West Africa in December 2013 . The high population mobility and poor public health infrastructure in this region led to the development of the largest Ebola virus disease ( EVD ) outbreak to date . On September 26 , 2014 , China dispatched a Mobile Biosafety Level-3 Laboratory ( MBSL-3 Lab ) and a well-trained diagnostic team to Sierra Leone to assist in EVD diagnosis using quantitative real-time PCR , which allowed the diagnosis of suspected EVD cases in less than 4 hours from the time of sample receiving . This laboratory was composed of three container vehicles equipped with advanced ventilation system , communication system , electricity and gas supply system . We strictly applied multiple safety precautions to reduce exposure risks . Personnel , materials , water and air flow management were the key elements of the biosafety measures in the MBSL-3 Lab . Air samples were regularly collected from the MBSL-3 Lab , but no evidence of Ebola virus infectious aerosols was detected . Potentially contaminated objects were also tested by collecting swabs . On one occasion , a pipette tested positive for EVD . A total of 1 , 635 suspected EVD cases ( 824 positive [50 . 4%] ) were tested from September 28 to November 11 , 2014 , and no member of the diagnostic team was infected with Ebola virus or other pathogens , including Lassa fever . The specimens tested included blood ( 69 . 2% ) and oral swabs ( 30 . 8% ) with positivity rates of 54 . 2% and 41 . 9% , respectively . The China mobile laboratory was thus instrumental in the EVD outbreak response by providing timely and reliable diagnostics . The MBSL-3 Lab significantly contributed to establishing a suitable laboratory response capacity during the emergence of EVD in Sierra Leone .
Ebola virus belongs to the Filoviridae family of enveloped viruses and contains a non-segmented negative-strand RNA genome [1 , 2] . Infection in humans can cause Ebola hemorrhagic fever , with exceptionally high case-fatality rates of more than 50% [3 , 4] . The incubation period of Ebola virus disease ( EVD ) is 2 to 21 days [5] . The clinical signs and symptoms are extremely similar to those of the Marburg virus and include fever , body aches , vomiting , diarrhea , rash and , in some cases , both internal and external bleeding [5] . Patients usually die of multiple-organ failure or hypovolemic shock . No licensed therapeutic or prophylactic treatments are currently available . The largest outbreak of EVD has been ongoing in West Africa since December 2013 . As of April 15 , 2015 , 25 , 826 cases ( 10 , 704 deaths [41 . 4%] ) had been reported by the World Health Organization ( WHO ) [6] . Although direct contact is the main route of transmission [7–10] , EVD is still easily contagious , and healthcare workers have constituted a considerable proportion of all cases . In particular , by April 11 , 2015 , 864 healthcare workers ( 503 deaths [58 . 2%] ) had been infected [6] . Ebola virus is classified as a biosafety level-4 agent . Clinical specimen inactivation should be performed in a biosafety level-3 laboratory , and subsequent to this step , routine testing can be performed in a biosafety level-2 laboratory . However , at the time of the outbreak , West Africa had few high-level biosafety facilities , so scientists had to work under difficult and dangerous conditions associated with potential exposure risks [11] . It would take a fairly long time , a large staff and many resources to construct a new fixed biosafety facility , thus delaying prevention and control of the epidemic . Therefore , a mobile unit [12 , 13] with both biosafety and flexibility was urgently needed to manage epidemics and emergent public health incidents such as the EVD outbreak . In September 2014 , China responded to the appeal made by the United Nations and WHO and offered assistance to the government of Sierra Leone . A truck-based mobile biosafety level-3 laboratory ( MBSL-3 Lab ) and a well-trained diagnostic team were then dispatched and deployed to the Sierra Leone-China Friendship Hospital , in one of the hardest-hit areas , near Freetown , to assist in EVD diagnosis . The team members and aid supplies arrived on September 17 , 2014 . It took approximately one week to rebuild part of the hospital into multiple functional regions to meet the specimen testing requirements , including a specimen-receiving region , a supply-storage region , a waste-incineration region , a nucleic-acid-detection region , and a staff-rest area , among others . The MBSL-3 Lab was transported by an airlift jet aircraft ( Antonov An-124 Ruslan , Russia ) from Beijing Capital International Airport on September 24 , 2014 , at 03:00 ( Beijing time ) to Freetown International Airport on September 25 , 2014 , at 14:00 ( Freetown time ) , with a flight duration of 43 h . It took another three and a half hours to drive the MBSL-3 Lab to the Sierra Leone-China Friendship Hospital . With strict training and standard operating procedures ( SOPs ) , clinical specimen testing began within 60 h after the arrival of the MBSL-3 Lab , enabling the diagnosis of suspected EVD cases in less than 4 hours from the time of sample receiving . In total , 1 , 635 suspected EVD cases ( 824 positive [50 . 4%] ) were tested from September 28 to November 11 , 2014 , and none of the staff members was infected with Ebola virus or other pathogens . Here , we provide a brief overview of the MBSL-3 Lab and the biosafety precautions applied to manage the EVD outbreak .
This Ebola outbreak response was a humanitarian aid mission . The SOPs used were approved by the WHO and the Sierra Leone Ministry of Health and Sanitation ( MoHS ) . The diagnostic results were released immediately after the specimen analyses were completed . Specimens were delivered to our worksite daily from two sources: the emergency operations center jointly established by the Sierra Leone MoHS and the China medical aid team who accompanied us and was also deployed to the Sierra Leone-China Friendship Hospital . When picking up the specimens , the staff wore one layer of personal protective equipment ( PPE ) , including a protective suit ( Lakeland INC or DuPont , USA ) , an N95 mask ( 3M , USA ) , an anti-impact goggle ( 3M , USA ) , two pairs of latex gloves with the inner pair taped to the protective suit and a pair of dedicated shoes and waterproof shoe covers ( S1 Fig ) . The surface of the specimen bucket and the packing bag were disinfected by spraying with 0 . 25% chlorine-containing disinfectants . The staff extracted RNA in the BSL-3 Lab wearing two layers of PPE . The inner PPE included a protective suit , an N95 mask , a pair of inner gloves and a pair of dedicated shoes and waterproof shoe covers ( S1 Fig ) . The external PPE included a HEPA filter-equipped powered air purifying respirator ( 3M , USA ) , a disposable sterilized surgical gown , a pair of external gloves and waterproof shoe covers ( S1 Fig ) . The specimen bucket was opened within the biosafety cabinet . As Buffer AVL in the QIAamp Viral RNA Mini Kit ( Qiagen , Germantown , MD , USA ) was insufficient to inactivate samples [14] , a combination of physical and chemical inactivation was performed to enhance the inactivation efficiency . The specimens were first inactivated by incubation in a water bath at 62°C for 1h before opening the tube cap to pipette the samples and were then further inactivated by the addition of Buffer AVL to the samples . RNA was extracted using the QIAamp Viral RNA Mini Kit ( Qiagen , Germantown , MD , USA ) according to the manufacturer’s protocol . All waste was first chemically inactivated ( with 0 . 25% chlorine-containing disinfectant ) , then sterilized using a double-leaf autoclave and finally incinerated . Quantitative real-time PCR ( Q-RT-PCR ) assays were performed using a set of published primers and probes [15] , targeting regions of the glycoprotein gene ( F: 5’-TGGGCTGAAAAYTGCTACAATC-3’; R: 5’-CTTTGTGMACATASCGGCAC-3’; Probe: FAM-5′-CTACCAGCAGCGCCAGACGG-3′-TAMRA ) . RNA was amplified using the One Step PrimeScript RT-PCR Kit ( TaKaRa , Japan ) , and 40-cycle Q-RT-PCR assays were run on the LightCycle 96 System ( Roche , Switzerland ) . Melt curve analysis was performed to confirm the identity of the amplification products . The specimens were considered positive if there was an apparent logarithmic phase in the amplification curve , with melting point confirmed amplification products and the Ct value≤36 ( Ct value<26 , intense positive; 26≤Ct value≤ 36 , weak positive ) . In contrast , the specimens were considered negative if there was no apparent logarithmic phase , with the Ct value undetermined , and they were considered suspect when 36<Ct value≤40 . The MBSL-3 Lab was equipped with a -20°C freezer and a -80°C freezer , and there was another -80°C freezer outside the MBSL-3 Lab . As a result , we could store a total of 1500–2000 specimens . For short-term storage , namely , within 1 day , we stored the specimens at -20°C . For long-term storage , we stored the specimens at -80°C . The specimens were well packed and surface disinfected with 0 . 25% chlorine-containing disinfectant before storage . The Sierra Leone-China Friendship Hospital was guarded by the military guard of Sierra Leone , and the freezers were well locked . Every patient was assigned a unique Outbreak Case ID by the emergency operations center jointly established by the MoHS . Each time a sample was collected , the patient was asked to complete a “VIRAL HEMORRHAGIC FEVER CASE INVESTIGATION FORM” . The sample tube and the investigation form were marked with the Outbreak Case ID and patient name and were then delivered to us . Therefore , the Outbreak Case ID provided a unique number for tracking the patient , the specimen and the test result . The information in our testing report included the Outbreak Case ID , the Ct value yielded by Q-RT-PCR and the confirmed result ( Yes/No/Suspect ) . According to an agreement with the MoHS , we usually did not contact hospitals directly . Instead , we submitted the testing report to the WHO and the MoHS , which was in charge of delivering the results to hospitals . In particular , the China medical aid team who came with us and was also deployed to the Sierra Leone-China Friendship Hospital could get testing results from us directly .
The China MBSL-3 Lab arrived in Sierra Leone on September 25 , 2014 , and specimen tests were carried out within 60 h of its arrival . The worksite layout was shown in Fig 1 . After receiving specimens , scientists sent them to the MBSL-3 Lab , where RNA was extracted . One room in the hospital was rebuilt and used for subsequent Q-RT-PCR analysis . The MBSL-3 Lab was powered by alternate use of 200kW diesel generators . Lab and household trash was incinerated away from the lab or structures in a pit . There were surveillance cameras all around the worksite and inside every experimental room , and scientists could watch real-time surveillance video and communicate with the experimenters in the laboratory . An overview of the composition of the China mobile laboratory diagnostic team and the team members’ tasks was shown in Table 1 . One scientist was in charge of contacting the MoHS to coordinate issues such as sending specimens and releasing analysis results . In addition , eight scientists engaged in virus detection . Technical support personnel were in charge of the operation of the MBSL-3 Lab , including overseeing the water and electricity supply , maintenance and repair of equipment , sterilization and incineration of lab trash as well as watching and recording the daily experimental process . Two medical doctors monitored the health conditions of every staff member . The MBSL-3 Lab was composed of three container vehicles . The container encompassing the BSL-3 laboratory was called the main container ( L×W×H: 9125×2438×2896mm ) ; the second container , of the same size , was used for personnel cleaning and technology support and was called the auxiliary container; and the third container was the command container ( L×W×H: 6300×2460×2100mm ) . As shown in Fig 2 , the main and auxiliary containers were connected by an airtight soft connection and together formed a complete BSL-3 Lab . From the entrance to the inside , in order , there was the outside locker room ( 0-5Pa ) , the inside locker room ( Buffer room-2 , -10Pa ) , the semi-contaminated channel ( -20±5Pa ) , the air lock room ( Buffer room-1 , -45±5Pa ) and the BSL-3 laboratory ( -70±10Pa ) . The doors were interlocking . The checklist for the different workplaces and instruments in the MBSL-3 Lab was listed in S1 Table . The MBSL-3 Lab provided triple protection for humans , specimens and the environment . The main performance of the MBSL-3 Lab was detailed as follows . “Four Flows” management were the key elements of biosafety measures in the MBSL-3 Lab ( Fig 4 ) . To assess the aerosol exposure risk when working in or around the MBSL-3 Lab , air samples were collected from the BSL-3 lab , locker rooms , water treatment room , equipment room , exhaust outlet and command container and were concentrated for EVD detection every 15 days ( S1 Fig ) . Fortunately , all results were negative . We also collected swabs from the surfaces of potentially contaminated objects to determine whether there was an existing exposure risk ( S2 Table ) . On one occasion , the pipette used to pipette samples from the blood-collection tubes tested positive for EVD , with a Ct value of 27 . 75 . The diagnostic algorithm for laboratory testing and the rationale for positive/negative/suspect test results were presented in Fig 5 . We repeated the testing of the suspect and negative cases and strongly recommended collecting specimens again if collection was performed <3 days post onset of symptoms . We found no evidence of RNA contamination during the entire operation . We added positive and negative controls to every experiment , and all controls produced the expected results . Overall , 1 , 635 suspected EVD specimens were tested from September 28 to November 11 , 2014 , primarily blood/serum samples ( 69 . 2% ) and oral swabs ( 30 . 8% ) . The sample sources and test results were presented in Table 2 . In total , 824 cases ( 50 . 4% ) were identified as positive , and the positive rate of the swab samples ( 41 . 9% ) was slightly lower than that of the blood samples ( 54 . 2% ) .
The number of various paroxysmal public health events has been growing , and most have occurred in poverty-stricken areas . However , the resources for medical treatment , outbreak management and laboratory research are concentrated in developed regions , and substantial expenditure would be required to build new medical systems in these areas . Because epidemic situations are always urgent , scientists thus work under inadequate conditions and face exposure risks . Therefore , rapid , safe and flexible outbreak response capacity is urgently needed [17] . A mobile laboratory unit can easily be promptly deployed when needed and can provide a safe working environment , which will be a vital part of the outbreak response to emerging public health events or bioterrorism acts and will make great contributions to lessening and controlling epidemics . Several mobile units have previously been used in natural disaster scenarios [18 , 19] , in health surveys [20 , 21] , during the outbreak of severe infectious diseases [22–24] and in military campaigns [25] . Our MBSL-3 Lab meets the requirements of on-site collection , isolation , cultivation and detection of emergent infectious pathogens . This laboratory also protects humans as well as the environment and specimens , and it was designed to be functional in a field setting , even without logistical support . The major challenges in a remote location may be power supply and water supply , but there are ways to overcome them . There was an 80kVA ( ≈70kW ) diesel generating set in the auxiliary container of the MBSL-3 Lab . Full fuel in the oil box can power the MBSL-3 Lab in continuous operation for 24h . We can bring as much fuel with us as possible using oil tanks , and wherever the MBSL-3 Lab can arrive , a refueling truck could also arrive . The MBSL-3 Lab is also equipped with a water storage tank and a water softener , and water can be re-supplied with water from a well or clear stream . If the experimenters could do not take a shower in the MBSL-3 Lab , the water requirement is not large , approximately 200L per day . In addition , the MBSL-3 Lab is equipped with a leveling system , but it still needs a 20m×8m level ground . This was the first time that we executed a mission in Africa . In total , 1 , 635 specimens were tested from September 28 to November 11 , 2014 , accounting for more than one quarter of the nation’s specimen volume during the same period . In all , 824 ( 50 . 4% ) specimens were EVD-positive , representing 33 . 3% of the total number of confirmed cases reported in Sierra Leone during the same period . The maximum number of specimens that we could reasonably process in one day is approximately 120–150 . We developed strict SOPs , adopted comprehensive protective measures and used comprehensive medical and logistical support systems to ensure safe and orderly performance of the virus diagnosis task . In particular , the “Four Flows” biosafety protocol was strictly followed . We monitored the exposure risk during clinical specimen testing . Air samples were collected from every workspace , and the test results were all negative , indicating that the working environment was relatively safe . The surfaces of potentially contaminated objects were also swabbed . On one occasion , the pipette used to pipette samples from blood-collection tubes tested positive . Given that a portion of the specimens contained only a small sample volume , the pipette had to be placed deep into the tubes and was easily contaminated by touching the inner wall . Therefore , it was suggested that the barrel of the pipette should be disinfected with disinfectant-containing gauze after pipetting each sample to avoid personnel infection and cross-contamination of samples . The test results played an important role in the disposal of symptomatic individuals and might , in a sense , determine their fates . For positive cases , the patients would be properly isolated and treated without visiting family members , and traditional religious funerals for the dead were forbidden . For negative cases , the patients would be separated from the positive cases and kept in an observation ward for follow-up testing or discharge to relieve the limited wards . Hence , the accuracy of the test results was crucial . False-positive results might lead to the individual being infected by positive patients , whereas false-negative results might lead to the spread of EVD to families and even the community . Our diagnostic algorithm suggested a suspect conclusion when 36<Ct value≤40 and strongly recommended resampling and considering clinical information and epidemiological links . Q-RT-PCR is now a preferred method for pathogen diagnosis due to its rapid and sensitive features [26] , but it is prone to contamination and may result in false-positive results . Therefore , we conducted every experiment in the biosafety cabinet . The cabinet and PCR room were exposed periodically to ultraviolet radiation to eliminate nucleic acid contamination . Additionally , PCR tubes were never opened . Every control included in the PCR assays produced the expected result , indicating high experimental accuracy . Moreover , the MoHS was in charge of retrospective look at the disease progresses of the patients , and to date , we have not received any feedback regarding a false diagnostic case from the MoHS . We have shown that the positive rate of oral swabs was lower than that of blood samples . The technique and efficiency of swabbing might be one of the most important factors . Swab samples should be obtained by vigorous sampling to acquire sufficient biologic material for testing [27] . A quality-control PCR target ( housekeeping gene target ) , such as Beta 2 Microglobulin ( B2M ) , should be added for sample integrity assessment in the future . Our MBSL-3 Lab continuously worked for six months and managed 4 , 867 specimens for EVD diagnostics . During that time , the China CDC established a fixed BSL-3 Lab near the Sierra Leone-China Friendship Hospital for long-term surveillance and to serve as the public health system for future outbreaks and epidemics . Currently , the EVD epidemic situation is under effective control , and our MBSL-3 Lab has been proven to be an important force for disease control and emergency disposal . | A Mobile Biosafety Level-3 Laboratory ( MBSL-3 Lab ) and a well-trained diagnostic team were dispatched to Sierra Leone to assist in Ebola virus disease ( EVD ) diagnosis when the largest outbreak of EVD to date emerged in West Africa in 2014 . This setup allowed for the diagnosis of suspected EVD cases in less than 4 hours from the time of sample receiving . The laboratory was composed of three container vehicles and was equipped with advanced ventilation system , communication system , electricity and gas supply system . Multiple safety precautions were strictly applied to reduce exposure risks . A total of 1 , 635 suspected EVD cases were evaluated from September 28 to November 11 , 2014 , and none of the staff members was infected with Ebola virus or other pathogens . The China mobile laboratory was thus instrumental in the EVD outbreak response by providing timely and accurate diagnostics . Therefore , the MBSL-3 Lab played a significant role in establishing a suitable laboratory response capacity during the emergence of EVD in Sierra Leone . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"laboratory",
"equipment",
"engineering",
"and",
"technology",
"pathogens",
"rna",
"extraction",
"microbiology",
"disinfection",
"health",
"care",
"viruses",
"preventive",
"medicine",
... | 2017 | Rapid deployment of a mobile biosafety level-3 laboratory in Sierra Leone during the 2014 Ebola virus epidemic |
Paramutations represent heritable epigenetic alterations that cause departures from Mendelian inheritance . While the mechanism responsible is largely unknown , recent results in both mouse and maize suggest paramutations are correlated with RNA molecules capable of affecting changes in gene expression patterns . In maize , multiple required to maintain repression ( rmr ) loci stabilize these paramutant states . Here we show rmr1 encodes a novel Snf2 protein that affects both small RNA accumulation and cytosine methylation of a proximal transposon fragment at the Pl1-Rhoades allele . However , these cytosine methylation differences do not define the various epigenetic states associated with paramutations . Pedigree analyses also show RMR1 does not mediate the allelic interactions that typically establish paramutations . Strikingly , our mutant analyses show that Pl1-Rhoades RNA transcript levels are altered independently of transcription rates , implicating a post-transcriptional level of RMR1 action . These results suggest the RNA component of maize paramutation maintains small heterochromatic-like domains that can affect , via the activity of a Snf2 protein , the stability of nascent transcripts from adjacent genes by way of a cotranscriptional repression process . These findings highlight a mechanism by which alleles of endogenous loci can acquire novel expression patterns that are meiotically transmissible .
The term “paramutation” describes a genetic behavior in which the regulatory state of specific alleles is heritably altered through interactions with their homologous partners in trans [1 , 2] . This behavior presents an exception to the Mendelian principle that alleles segregate from a heterozygous state unchanged [3] . Paramutations have been best characterized at loci encoding transcriptional regulators of pigment biosynthesis in maize , but similar behaviors have been described in other plant and animal systems , most recently in mice [4 , 5] . While the broader roles of paramutation in genome-wide regulation and evolution remain to be seen , the Pl1-Rhoades allele of the maize purple plant1 ( pl1 ) locus presents a tractable system to study the paramutation process . The pl1 locus encodes a Myb-like protein that acts as a transcriptional activator of genes required for anthocyanin pigment production [6] . Inheritance patterns illustrate that the Pl1-Rhoades allele can exist in quantitatively distinct regulatory states , reflected by differences in plant color . When individuals with a highly expressed reference state of Pl1-Rhoades , termed Pl-Rh , are crossed with plants having a repressed state , referred to as Pl′ , only progeny with weak pigmentation are produced [7 , 8] . Pl-Rh states invariably change to Pl′ in Pl-Rh/Pl′ heterozygotes [7]; this is a typical hallmark of paramutation . Relative to Pl-Rh , the Pl′ state displays reductions in both Pl1-Rhoades RNA levels ( ∼10-fold ) and transcription rate ( ∼3-fold ) that are associated with a reduction in plant pigment [8] . This repressed Pl′ state is meiotically stable when maintained in a Pl1-Rhoades homozygote , with no reversion to Pl-Rh seen to date . Pl′ can , however , revert to Pl-Rh when heterozygous with some pl1 alleles other than Pl1-Rhoades , when maintained in a hemizygous condition , or in the presence of specific recessive mutations [9–12] . Genetic screens for ethyl methanesulfonate ( EMS ) –induced recessive mutations identify at least ten loci , including required to maintain repression1 ( rmr1 ) , rmr2 , rmr6 , and mediator of paramutation1 ( mop1 ) , whose normal functions maintain the repressed Pl′ state ( [10 , 11 , 13]; J . B . H . , unpublished data ) . These rmr mutations specifically affect the expression of Pl1-Rhoades and not other pl1 alleles [10 , 11] , indicating that the Pl1-Rhoades allele is a direct and specific target of paramutation-based epigenetic changes . mop1 was recently identified [14 , 15] as encoding the putative ortholog of the Arabidopsis protein RDR2 , a presumed RNA-dependent RNA polymerase involved in siRNA-based maintenance of de novo cytosine methylation [16] . Recessive mutations defining rmr1 , rmr2 , and rmr6 destabilize the repressed Pl′ state , resulting in darkly pigmented plant tissues , an increase in pl1 RNA levels , and meiotic transmission of Pl-Rh revertant states [10 , 11] . To date , the molecular identity of these rmr factors remains unknown . In this report we identify rmr1 as encoding a novel Snf2 protein that represents a founding member of a subgroup of factors similar to proteins involved in plant small RNA metabolism . Our analyses show that RMR1 affects both pl1 RNA transcript stability as well as small interfering RNA ( siRNA ) accumulation and DNA methylation patterns at Pl1-Rhoades . These results support a model in which maintenance of paramutant states is dependent on a repression mechanism similar to the recently proposed cotranscriptional gene silencing mechanism in fission yeast [17 , 18] . To our knowledge , RMR1 is the first protein identified that maintains trans-generationally repressed states established by paramutation .
The rmr1 locus is defined by four recessive mutations ( Protocol S1 ) characterized by a darkly pigmented plant phenotype that results from loss of Pl′ repression . Previous RNase protection experiments showed a 26-fold increase in pl1 RNA in floret tissue between rmr1–1 mutant plants and heterozygous siblings [10] . However , these experiments did not address if changes in pl1 transcript abundance correlated with changes in actual transcription at the pl1 locus . In vitro transcription assays using nuclei isolated from husk leaf tissue revealed there was no statistically significant change in relative transcription rates of the Pl1-Rhoades allele between rmr1–1 mutants and heterozygous siblings ( Figure S1 ) . However , transcription rates of anthocyaninless1 ( a1 ) , a direct target of the PL1 transcriptional activator [7 , 19] , were ∼4-fold greater in rmr1–1 mutants ( Figure S1 ) , reflecting significantly increased PL1 activity . Transcription rates from colored plant1 ( b1 ) —a locus encoding a basic helix-loop-helix factor genetically required for a1 transcription— remained unchanged . These results were recapitulated in comparisons between nuclei isolated from rmr1–3 mutants and heterozygous siblings in which in vitro transcription assays revealed no significant change in transcription rate of Pl1-Rhoades ( Figures 1A and S1; n = 4 , two-tailed two-sample t-test , t = 0 . 8 , p = 0 . 5 ) while RNase protection experiments showed a 5 . 7-fold increase in pl1 RNA for rmr1–3 mutants ( Figure 1B and 1C; n = 2 , two-tailed two-sample t-test , t = 10 . 8 , p < 0 . 01 ) using RNA isolated from the same tissues of the same individuals . Similar comparisons from identical tissues but in a different genetic background again showed that transcription rates at pl1 remained unchanged while pl1 RNA levels increased 7 . 52-fold in rmr1–3 mutants compared to heterozygous siblings ( n = 1; see Protocol S1 ) . These RNA expression results sharply contrast those of previous reports using identical in vitro transcription assays that detected significant differences in Pl1-Rhoades transcription rates between Pl′ and Pl-Rh states and between rmr6 mutants and non-mutants [8 , 11] . This indicates our in vitro results represent an accurate assessment of transcription rates and not a limitation of the assay to detect rate differences at the pl1 locus . Combined , these results imply an increase of pl1 RNA abundance disproportionate to insignificant changes in transcription rate in rmr1 mutants , the most direct interpretation being that RMR1 functions at a post-transcriptional level to stabilize Pl1-Rhoades RNA . To better understand Rmr1 function and the paramutation mechanism , we used a map-based approach to identify the rmr1 gene . Using a polymorphic F2 population we looked for genetic linkage between the mutant phenotype and previously mapped chromosome markers [20] . The dark-color phenotype of rmr1–1 homozygotes showed invariant cosegregation with the mutant parent polymorphism of SSLP markers bnlg1174a ( 680 chromosomes tested; <0 . 15 cM ) and npi252 ( 60 chromosomes tested; <1 . 7 cM ) , indicating rmr1 was tightly linked to those markers in bin 6 . 05 on Chromosome 6 . We used the high degree of synteny between this region and rice Chromosome 5 to identify candidate rmr1 orthologs ( Figure 2A and 2B ) . Within the syntenic rice region we identified a gene model , Os05g32610 ( http://rice . tigr . org/ ) , predicted to encode a Snf2 protein . The Snf2 protein family is composed of members similar to Saccharomyces cerevisiae Snf2p with a bipartite helicase domain containing Pfam SNF2_N and Helicase_C profiles , and includes many proteins involved in ATP-dependent chromatin remodeling [21 , 22] . While there was no public maize expressed sequence tag for this candidate , we used BLAST searches to identify genomic survey sequence similar to Os05g32610 . Oligonucleotide primers were designed from these sequences and used to generate PCR amplicons spanning the maize Os05g32610 ortholog , which were sequenced from individuals homozygous for Rmr1 progenitor alleles and mutant derivatives ( see Materials and Methods and Dataset S1 ) . The maize sequence generated from each of the homozygous mutants revealed single unique transition-type base pair changes consistent with EMS mutagenesis relative to the progenitor ( Figure 2C ) . The amino acid change associated with the rmr1–1 allele is predicted to prevent proper folding of the helicase domain [23] , while the non-conservative amino acid substitutions associated with the rmr1–2 and rmr1–4 alleles occur at highly conserved residues in the SNF2_N profile ( Figure 2D ) . The rmr1–3 allele is associated with a nonsense mutation predicted to truncate the peptide before the conserved helicase domain . CAPS markers were designed to the potential rmr1–1 and rmr1–3 lesions and used to show that the base pair polymorphisms at each of the probable lesions invariably cosegregate with the mutant phenotype ( see Materials and Methods ) . These results support these polymorphisms as bona fide molecular lesions in the rmr1 gene . Based upon molecular genetic mapping data , DNA sequencing results , and the relevance of the fact that Snf2 proteins affect chromatin environments , we conclude the rmr1 locus encodes a protein containing a Snf2 helicase domain . Os05g32610 gene models and our cDNA sequencing analysis ( see Materials and Methods ) indicate rmr1 encodes a 1 , 435-amino-acid protein . In addition to having the conserved Snf2 helicase domain , the protein has a large N-terminal region with no significant identity to any known or predicted proteins . Phylogenetic comparison with other known Snf2 proteins in maize , rice , Arabidopsis , and budding yeast shows RMR1 is a member of a Rad54-like subfamily defined by DRD1 ( Figure 3 ) . Arabidopsis DRD1 is a putative chromatin remodeling factor affecting RNA-directed DNA methylation ( RdDM ) patterns [24–26] . In the emerging RdDM pathway model , DNA sequences are targeted for de novo cytosine methylation by complementary siRNA molecules generated from “aberrant” RNA transcripts . The putative MOP1 ortholog in Arabidopsis , RDR2 , is required in this pathway to presumably generate double-stranded RNA from these transcripts and provide a substrate for siRNA biogenesis through activity of a Dicer-like enzyme [27] . DRD1 is thought to be a downstream effector protein that facilitates de novo methylation of targeted DNA sequences , possibly by modulating chromatin architecture to provide access to de novo methyltransferases [24–26 , 28] . The DRD1 subfamily also includes the recently identified CLSY1 protein implicated in the systemic spreading of siRNA-mediated silencing in Arabidopsis [29] . Multiple sequence alignments ( Figure S2 ) indicate RMR1 is not the structural ortholog of either DRD1 or CLSY1 . The DRD1 subfamily can be divided into three distinct monophyletic groups , with RMR1 , DRD1 , and CLSY1 defining different groups ( Figure 3 ) . The presumed maize ortholog of DRD1 is likely one of two proteins in the DRD1 subgroup , Chromatin remodeling complex subunit R 127 ( CHR127 ) ( http://chromdb . org/ ) , a partial protein predicted from maize expressed sequence tag sequences , or CHR156 , a full-length protein predicted from maize genomic sequence ( see Materials and Methods ) . RMR1 is more similar to Arabidopsis proteins predicted from At1g05490 and At3g24340 . RNA interference knockdowns of these putative Arabidopsis orthologs are known to have little to no effect in response to DNA damage [30] . Taking into account the phylogenetic analysis of the predicted coding sequence , it is possible RMR1 function may be similar to , but distinct from , that of DRD1 and CLSY1 . The three proteins may fulfill a similar role in RdDM , but perhaps function under different conditions or in distinct genomic contexts . Alternatively , they could perform different roles within an RdDM pathway , or function in separate epigenetic mechanisms altogether . Given the results of our pl1 RNA expression analyses , it is possible that RMR1 represents a Snf2 protein that links chromatin organization to RNA transcript stability . In the described Arabidopsis RdDM pathway , DRD1 maintains cytosine methylation at nonsymmetrical CNN sequences represented by siRNAs [24–26] . Many endogenous genomic targets of DRD1 appear to be repetitive elements [31] . At Pl1-Rhoades there is a 402-bp terminal fragment of a CACTA-like type II DNA transposon , similar to doppia , 129 bp upstream of the translational start site [8 , 32 , 33] . Assuming analogous functional roles of RMR1 and DRD1 we compared DNA methylation patterns at this upstream repetitive element in rmr1 mutants and non-mutant siblings . Previous restriction-enzyme-based comparisons of DNA methylation status between Pl-Rh and Pl′ states found no differences , although few 5′ proximal sites were evaluated [8] . Using Southern blot hybridization analysis following digestion of genomic DNA with methylation-sensitive restriction enzymes , we found that the doppia fragment is hypomethylated at specific sites in plants homozygous for the rmr1–1 mutation compared to heterozygous wild-type siblings ( Figures 4A , 4B , and S3 ) . Consistent with findings in Arabidopsis RdDM mutants [16 , 34–36] , the sites hypomethylated in rmr1 mutants were of the CNN context . A relative hypomethylation pattern in 5′ sequences is also present in plants homozygous for mutations at either rmr6 or mop1 ( Figures S4 and S5 ) . In rmr6 mutants the extent of hypomethylation was greater than that of either rmr1 or mop1 mutants and encompassed CG methylation sites as well as non-CG targets , suggesting Rmr6 has a broader effect in cytosine methylation maintenance . The presence of these methylation differences in multiple mutant backgrounds indicates that this hypomethylation pattern reflects the chromatin status at doppia in plants where maintenance of repressed paramutant states is compromised . Consistent with the Arabidopsis RdDM model , small RNAs ( ∼26 nt ) with sequence similarity to the doppia element are detected in wild-type Pl′ plants in both sense and antisense orientations ( Figures 4D and S6 ) . These small RNAs are undetectable in rmr1 mutants , unlike in wild-type siblings . This result contrasts those in Arabidopsis showing that DRD1 deficiencies do not affect the abundance of endogenous siRNAs representing repetitive elements [31] . However , it has been reported that the abundance of endogenous siRNA and trans-acting siRNA populations are highly reduced in CLSY1 mutants [29] . To test if the doppia fragment hypomethylation was indicative of genome-wide changes we assayed the cytosine methylation status at centromeres and 45S repeat sequences . Cytosine methylation patterns were unaffected in either of these regions in rmr1 mutants as compared to non-mutant siblings ( Figure S7 ) . Additionally , we examined the methylation status of doppia-like loci genome-wide ( Figure 4E ) and found no obvious differences between rmr1 mutants and non-mutant siblings . These results indicate that while RMR1 acts on the doppia sequence upstream of Pl1-Rhoades , doppia elements appear unaffected throughout the genome . This specificity of RMR1 function may be due to its intimate and exclusive involvement with alleles that undergo paramutation , or may be indicative of differential regulation of repetitive elements depending on their genomic and epigenetic context . If RMR1 is involved in maintaining cytosine methylation patterns characteristic of repressed paramutant states then a prediction would be that the methylation differences seen between mutants and non-mutants would reflect the Pl′ and Pl-Rh regulatory states . Surprisingly , there are no methylation differences at the doppia fragment between Pl-Rh and Pl′ states ( Figures 4C and S8 ) . These results suggest that while the upstream doppia element of Pl1-Rhoades is a target of multiple factors involved in maintaining the epigenetic repression associated with paramutation , the actual process of paramutation does not result in similar changes of DNA methylation at this element . Based on a reverse transcriptase PCR ( RT-PCR ) expression profile ( Figure S9 ) rmr1 appears to be expressed in all rapidly dividing somatic tissues , consistent with a role in maintaining paramutant states throughout development . However , since the methylation patterns maintained by RMR1 appear unrelated to the paramutant state of Pl1-Rhoades , we questioned whether RMR1 is directly required for paramutation to occur . This process results in the invariable establishment of the Pl′ state in Pl′/Pl-Rh plants , as evidenced by the observation that only Pl′/Pl′ progeny are found when Pl′/Pl-Rh plants are crossed to Pl-Rh/Pl-Rh testers [7 , 8] . If RMR1 were directly involved in this process we would expect that an rmr1 deficiency might interfere with the Pl′ establishment event . To test this , we tracked the behavior of individual Pl1-Rhoades alleles in test crosses to assess the ability of the Pl′ state to facilitate paramutations in Pl′/Pl-Rh; rmr1–1/rmr1–2 plants . The Pl1-Rhoades allele in a Pl-Rh state was genetically linked ( ∼1 . 5 cM ) to a T6–9 translocation breakpoint ( T6–9 ) . The T6–9 interchange can act as a dominant semi-sterility marker , allowing us to trace specific Pl1-Rhoades alleles through genetic crosses [11] . rmr1 mutants heterozygous for the T6–9 interchange ( T6–9 Pl-Rh/Pl′ ) were crossed to a Pl-Rh/Pl-Rh tester ( Figure 5; Table S1 ) . If establishment of the Pl′ state was prevented in rmr1 mutants , we would expect all progeny receiving the interchange to display a Pl-Rh/Pl-Rh phenotype ( dark anther pigmentation ) . We observed that over half the progeny inheriting the interchange displayed a Pl′/Pl′-like phenotype ( light anther pigmentation ) , indicating that paramutation was established in the rmr1 mutant parent . It should also be noted that Pl-Rh/Pl-Rh plants , and those of an intermediate phenotype of partial pigmentation [7] , were present in both progeny inheriting the interchange and those inheriting a normal chromosome . These results are consistent with previous work showing Pl′ can revert to a Pl-Rh state in rmr1 mutants [10] . Corresponding analysis of the establishment of paramutant states at the b1 locus generated similar results ( Table S2 ) . The repressed B′ state of the B1-Intense allele [37] was established in B′/B-I rmr1 mutants greater than 95% of the time . While it is possible that rmr1 defects affect establishment efficiency , it will be difficult to differentiate any such effects from its clear role in maintenance [11] . These results point to an interesting duality in RMR1 function in which the wild-type protein is necessary for meiotic heritability of repressed epigenetic states , but is not required to establish these states . This duality is markedly different from results generated in the analysis of DRD1 , which was shown to be necessary for the maintenance , establishment , and removal of repressive epigenetic marks [24 , 25] .
RMR1 is the first protein identified whose function acts to maintain trans-generationally repressed states associated with paramutation , a genetic behavior that affects meiotically heritable epigenetic variation through allelic interactions at endogenous loci . The identification of RMR1 as a Snf2 protein highlights an emerging role of these proteins in establishing and maintaining epigenetic marks . In Arabidopsis the Snf2 proteins DRD1 and DDM1 [38 , 39] are known to maintain cytosine methylation patterns . Lsh1 , the mammalian protein most closely related to DDM1 , is also required for normal DNA methylation patterns [40–42] . There are some 42 Snf2 proteins in Arabidopsis and at least as many in maize ( http://chromdb . org/ ) . This diversity likely represents great functional specialization amongst these proteins . We have placed RMR1 in an RdDM pathway based on its helicase domain similarity to DRD1 and the recent identification of MOP1 as an RDR2 ortholog [14 , 15] . Consistent with this proposed pathway , the rmr1 mRNA expression profile ( Figure S9 ) closely matches that of mop1 [15] . Additionally , both RMR1 and MOP1 are necessary to maintain cytosine methylation patterns at silenced transgenes [43] , the Pl1-Rhoades doppia sequences , and certain Mutator transposable elements ( [15 , 44]; J . B . H . and D . Lisch , unpublished data ) . DRD1 is also known to target repetitive elements found in euchromatic contexts through an RdDM pathway [31] . However , the role RMR1 plays to maintain the repressed paramutant states at Pl1-Rhoades appears different than the function of DRD1 in the Arabidopsis RdDM pathway , as RMR1 has , in addition to its requirement for CNN methylation at doppia , a role in the normal accumulation of small RNAs with similarity to that element . It is unclear how RMR1 mediates the post-transcriptional regulation of pl1 transcripts as suggested by the in vitro transcription and RNase protection assays reported here . It is possible that pl1 transcripts resulting from Pl1-Rhoades in the Pl′ state are less stable than those produced from the Pl-Rh state because of differences in the chromatin environment of Pl1-Rhoades . However , there do not appear to be any Pl′-specific small RNAs produced from the pl1 coding region [12] . In S . pombe it has been shown that the chromatin environment of a locus can affect RNA transcript levels without altering RNA polymerase II occupancy of that locus , leading to the proposal of a cotranscriptional gene silencing mechanism whereby nascent transcripts initiating in a heterochromatic environment are degraded by complexes targeted via heterochromatic small RNAs [17 , 18] . Chromatin differences in the upstream region of Pl1-Rhoades may favor recruitment of alternative RNA-processing factors or RNA polymerases , which in turn influence the stability of pl1 transcripts . In plants , localization of the large subunit 1a of RNA polymerase IV to loci targeted for RdDM appears necessary for the biogenesis of siRNAs from these loci [28] . When Pl′ repression is disrupted in rmr1 mutants , this alternate genesis or processing of the pl1 transcript may also be lost . Alternatively , our results may highlight a novel role for RMR1-like Snf2 proteins in directly interacting with nascent RNA transcripts via a helicase domain , or in recruiting factors that directly destabilize these transcripts . Importantly , our analysis of rmr1 mutants calls into question the relationship between RMR1 function and the mechanism of paramutation at Pl1-Rhoades . The mutational screens identifying rmr1 , rmr6 , and mop1 were designed to discover genetic components necessary to maintain the repressed state of Pl′ , not necessarily factors needed to establish this repressed state [10 , 13] . Therefore , it is possible that loci thus far identified may be indirectly related to the paramutation mechanism . Our results are consistent with a model wherein RMR1 functions in an RdDM pathway , along with an RDR2-like enzyme , MOP1 , to maintain a persistent heterochromatic-like chromatin structure at the repetitive element found directly upstream of the pl1 coding region . While it is not clear where RMR1 acts in this pathway it presumably acts coordinately with the maize orthologs of known RdDM components identified in Arabidopsis , namely DCL3 [16 , 45] , the DRM methyltransferases [36] , AGO4 [46 , 47] , the RNA polymerase IV subunits , and the maize DRD1 ortholog ( Figure 6A ) . In this model , doppia transcripts , perhaps because of the repetitive nature of the doppia genomic elements and/or the numerous internal subterminal repeats that are present in these elements [32 , 48] , are the source of aberrant RNA that is processed via MOP1 and a DCL3 enzyme into siRNAs . This small RNA production is carried out in a manner that is dependent on RMR1 activity , possibly via direct interaction with a small RNA processing complex or by making the DNA accessible to factors necessary for siRNA precursor generation such as polymerase IVa . These siRNAs , through the activity of AGO4 , DRM enzymes , and polymerase IVb , then establish a heterochromatic state at the Pl1-Rhoades doppia-like element that is present in both Pl-Rh and Pl′ states . The methylation effects seen in rmr1 mutants might indicate that this heterochromatization machinery depends on the activity of RMR1 to feed back on the doppia element , or loss of RMR1 may short circuit this pathway and thus affect methylation activity indirectly . An RMR1 defect then affects stability of paramutant states at pl1 because of the chromatin context of the Pl1-Rhoades allele , and not through direct disruption of components required for paramutations to occur . This is in line with a report that MOP1-dependent small RNAs produced at the b1 locus are insufficient to mediate paramutation [49] . The relationship between RMR1 action , the chromatin organization of Pl1-Rhoades , and the repressed Pl′ state is not clearly understood at this time . It is possible that derepression of the upstream repetitive element makes the region more accessible to general transcription factors whose actions could destabilize repressive Pl′ chromatin states that are independent of those maintained at doppia ( Figure 6A ) . Indeed , RNA polymerase processivity can lead to changes in the chromatin environment through histone modifications or histone replacement [50 , 51] . Alternatively , Pl′ chromatin states may represent a spreading of the heterochromatic domain at doppia into a euchromatic region defined by the Pl1-Rhoades gene space ( Figure 6B ) . In fission yeast , heterochromatic domains nucleated by small RNAs have the ability to spread in cis through successive H3 K9 methylation [52] . In this situation , loss of RMR1 function would alleviate Pl′ repression by disrupting maintenance of this expanded heterochromatic domain . In either of these situations RMR1 affects Pl1-Rhoades paramutations by virtue of its role in maintaining heterochromatic states at a proximal repetitive element . McClintock was the first to describe derivative alleles in which transposons acted to control the expression patterns of attendant genes [53] . It is now clear that epigenetic modulations of the transposons themselves—what McClintock referred to as “changes in state”—can alter the regulatory properties of individual genes both somatically [54] and trans-generationally [55 , 56] . Our results indicate that even transient changes in state of the Pl1-Rhoades doppia fragment can have trans-generational effects on pl1 gene expression patterns . These experimental examples , in the context of McClintock's thesis [53] , point to a dynamic source of regulatory , and potentially adaptive , variation adjunct to the DNA itself . Precisely how this epi-variation relates to existing genome structure and function , as well as its evolutionary potential , remains a largely unexplored area of investigation . Currently , well-characterized examples of paramutation are limited to loci where expression states have a clear phenotypic read-out , such as pigment synthesis . cis-Elements required to facilitate paramutation have been functionally identified at specific alleles of b1 and colored1 ( r1 ) [57–59] . To date , there is no evidence that the chromatin status of these cis-elements is affected by mutations at trans-acting loci required for maintenance of repressed paramutant states . It appears that paramutations represent a type of emergent system wherein genomic context and maintenance of chromatin states interact to facilitate meiotically heritable epigenetic variation . In this view , it is possible that cis- and trans-elements necessary for maintenance of such variation might not interact in a direct and predictable manner . What remains to be seen is the extent to which this type of system acts throughout the genome . Genome-wide screens for paramutation-like behavior , in which expression states are affected by allele history , remain technologically and conceptually challenging . Recent work by Kasschau et al . [60] suggests that in Arabidopsis , few endogenous genes are regulated by proximal presumed RdDM targets . However , it is tempting to speculate that examples of paramutation represent an exception to this trend , representing a mechanism by which populations can quickly , and heritably , change their transcriptome profile and regulation .
Plants were scored as carrying Pl-Rh or Pl′ states through visual inspection of anther pigmentation and assignment of an anther color score as previously described [7] . Pl′/Pl′ ( anther color score 1 to 4 ) anthers show little to no pigmentation while Pl-Rh/Pl-Rh ( anther color score 7 ) anthers are dark red to purple . Mutants were scored in the same way , with rmr and mop mutants showing a Pl-Rh/Pl-Rh-like phenotype , except in the case of the F2 rmr1 mapping populations , in which mutants were chosen on the basis of a dark seedling leaf phenotype [10] . Elite inbred lines ( B73 , A619 , and A632 ) were provided by the North Central Regional Plant Introduction Station ( http://www . ars . usda . gov/main/site_main . htm ? modecode=36-25-12–00 ) . Color-converted versions of A619 and A632 inbred lines were created by introgressing the Pl1-Rhoades allele into each [11] . The rmr1–1 , rmr1–2 , mop1–1 , and rmr6–1 alleles have been previously described [8 , 10 , 13] . The rmr1–3 allele was derived from identical materials used to isolate rmr1–1 and rmr1–2; rmr1–4 was derived from EMS-treated pollen from an A619 color-converted line applied to a color-converted A632 line [11] ( see Protocol S1 and Table S3 for complementation tests ) . The T6–9 translocation line carrying the Pl1-Rhoades allele used in Pl′ establishment tests has been described previously [11] . In vitro transcription assays ( rmr1–1 and rmr1–3; Figures 1 and S1 ) and RNase protection assays ( rmr1–3 only; Figure 1 ) were carried out as described [8] with husk nuclei and RNA isolated from single ears of the same genetic stocks used to measure pl1 RNA differences in rmr1–1 anthers [10] . The b1 and pl1 genotypes of these plants are as follows: B1-Intense ( B-I ) /B-I; Pl1-Rhoades ( Pl′ ) Rmr1/Pl′ rmr1–1 and B-I/B-I; Pl′ rmr1–1/Pl′ rmr1–1 , or B-I/B-I; Pl′ Rmr1/Pl′ rmr1–3 and B-I/B-I; Pl′ rmr1–3/Pl′ rmr1–3 . Identical procedures were applied to single ears from plants homozygous for Pl′ and either homozygous or heterozygous for rmr1–3 following a single backcross into the KYS inbred line [12] . Additional details regarding stock syntheses are available upon request . A F2 mapping population was created from inbred ( S9 ) rmr1–1/rmr1–1 , Pl′/Pl′ , and color-converted A632 inbred ( Pl′/Pl′ , >93% A632 ) parents . DNA was isolated using the DNeasy 96 plant kit ( Qiagen , http://www1 . qiagen . com/ ) from F2 mutant seedlings , mapping parents , and F1 hybrid leaf tissue . These DNA samples were screened with SSLP markers developed from the Maize Mapping Project ( http://www . maizemap . org/; US National Science Foundation award number 9872655; primer sequences and protocol available at http://maizegdb . org/ ) . Initial marker choice was restricted to Chromosomes 6 and 9 because of linkage of rmr1 to a T6–9 breakpoint . In addition to the rmr1–1 mapping population , a second F2 mapping population created with inbred ( S7 ) rmr1–3/rmr1–3 , Pl′/Pl′ , and color-converted A632 parents showed similar cosegregation with marker bnlg1174a ( 178 chromosomes tested; <0 . 56 cM ) . CAPS [61] markers were designed to test cosegregation of the rmr1–1- and rmr1–3-associated lesions with the rmr1 mutant phenotype ( see Protocol S1 for details ) . No recombinant chromosomes ( 876 chromosomes tested for rmr1–1 , 268 chromosomes tested for rmr1–3 ) were found using either marker . A BLAST search using the rice Os05g32610 ORF as a query identified maize GSS and sorghum expressed sequence tag sequences that were used to generate a contig representing the putative maize gene ( see Protocol S1 for sequence identifiers ) . Oligonucleotide primers ( Sigma-Genosys , http://www . sigmaaldrich . com/Brands/Sigma_Genosys . html ) were designed from these sequences and used in PCR amplification of genomic DNA from three separate individuals homozygous for each rmr1 mutant allele as well as functional reference alleles Rmr1-B73 , Rmr1-A632 , and Rmr1-A619 . PCR amplicons were purified using QIAquick gel extraction kit ( Qiagen ) and dideoxy sequenced ( UC Berkeley DNA Sequencing Facility , http://mcb . berkeley . edu/barker/dnaseq/ ) . To verify the intron/exon structure of rmr1 , cDNA was generated from rmr1–1 mutants as well as non-mutant B73 plants as described [15] , and rmr1 was amplified via RT-PCR . The resulting products , which were the predicted size for spliced rmr1 transcript , were sequenced to validate the intron/exon structure shown in Figure 2C . See Protocol S1 and Table S4 for all oligonucleotide primer sequences used . Sequencing reads from genomic and cDNA were aligned and edited with Sequencher ( Gene Codes , http://www . genecodes . com/ ) to create a contig representing rmr1 . The N-terminal prediction is based on alignment of RMR1 with the protein model for Os05g32610 . A search of the Pfam database ( http://www . sanger . ac . uk/Software/Pfam/ ) with the predicted RMR1 protein sequence was used to identify the conserved SNF2_N and Helicase_C protein profiles of the Snf2 helicase domain . MUSCLE [62] was used to generate an alignment between RMR1 and proteins from Arabidopsis , rice , maize ( CHR127 and CHR156 ) , and budding yeast over the helicase domain ( Figure S2 ) . Sequences for CHR127 and CHR156 were retrieved from ChromDB ( http://www . chromdb . org/ ) . Additional sequence information for CHR156 was identified from BAC CH201-3L17 ( GenBank accession AC194602 ) , and gene model prediction was performed using FGENESH+ ( Softberry , http://www . softberry . com/ ) with RMR1 as similar protein support . A distance tree was created and bootstrap values were calculated using PAUP* 4 . 0 from the above alignment ( Sinauer Associates , http://www . sinauer . com/ ) . Genomic DNA was isolated as described [13] from the terminal flag leaves of adult plants segregating for rmr1 , rmr6 , and mop1 mutants and heterozygous siblings as well as Pl′ and Pl-Rh plants as assayed by anther pigmentation [7 , 8 , 10 , 13] . Restriction digest and subsequent Southern blots were carried out as previously described [13] , using the restriction enzymes listed in Figure 4 ( New England Biolabs , http://www . neb . com/ ) . The probes specific to pl1 are shown in Figure 4; the 45S and centromere probes are as described [13] . Small RNAs were prepared from 10-mm immature ear tissue and used to generate small RNA northern blots as previously described [63] . In Figure 4D the small RNAs were run with a 27-bp DNA oligonucleotide containing doppia sequence that hybridized with the riboprobe used to identify the small RNAs . The riboprobe was synthesized as described [63] from a plasmid containing the region denoted probe B in Figure 4A linearized at an AseI site so as to contain only doppia sequence . Establishment of the Pl′ state in rmr1 mutants was assayed essentially as described previously [11] . When the T6–9 interchange pair is heterozygous with structurally normal chromosomes , the plants display ∼50% pollen sterility due to meiotic-segregation-induced aneuploidy in the resulting gametes . Pollen sterility was assayed in the field using a pocket microscope . rmr1 mutants were crossed to Pl-Rh/Pl-Rh A619 or A632 inbreds ( Table S1 ) , and the resultant progeny were scored with respect to Pl1-Rhoades expression state . | Genetics is founded on the principle that heritable changes in genes are caused by mutations and that the regulatory state of gene pairs ( alleles ) is passed on to progeny unchanged . An exception to this rule , paramutations—which reflect the outcome of interactions between alleles—produce changes in gene control that are stably inherited without altering the DNA sequence . It is currently thought that these allelic interactions cause structural alterations to the chromatin surrounding the gene . Recent work in both maize and mice suggests that RNA molecules may be responsible for paramutations . Several genes are required to maintain the repressed paramutant state of a maize purple plant1 ( pl1 ) allele , and here we report that one of these genes encodes a protein ( RMR1 ) with similarity to a protein previously implicated in facilitating genomic DNA modifications via small RNA molecules . Genetic and molecular experiments support a similar role for RMR1 acting at a repeated sequence found adjacent to this pl1 gene . Although loss of these DNA modifications leads to heritable changes in gene regulation , the data indicate these changes do not represent the heritable feature responsible for paramutation . These findings highlight an unusual but dynamic role for repeated genomic features and small RNA molecules in affecting heritable genetic changes independent of the DNA template . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"plant",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
] | 2007 | A Novel Snf2 Protein Maintains trans-Generational Regulatory States Established by Paramutation in Maize |
Atherosclerosis is the main cause of coronary heart disease and stroke , the two major causes of death in developed society . There is emerging evidence of excess risk of cardiovascular disease at low radiation doses in various occupationally exposed groups receiving small daily radiation doses . Assuming that they are causal , the mechanisms for effects of chronic fractionated radiation exposures on cardiovascular disease are unclear . We outline a spatial reaction-diffusion model for atherosclerosis and perform stability analysis , based wherever possible on human data . We show that a predicted consequence of multiple small radiation doses is to cause mean chemo-attractant ( MCP-1 ) concentration to increase linearly with cumulative dose . The main driver for the increase in MCP-1 is monocyte death , and consequent reduction in MCP-1 degradation . The radiation-induced risks predicted by the model are quantitatively consistent with those observed in a number of occupationally-exposed groups . The changes in equilibrium MCP-1 concentrations with low density lipoprotein cholesterol concentration are also consistent with experimental and epidemiologic data . This proposed mechanism would be experimentally testable . If true , it also has substantive implications for radiological protection , which at present does not take cardiovascular disease into account . The Japanese A-bomb survivor data implies that cardiovascular disease and cancer mortality contribute similarly to radiogenic risk . The major uncertainty in assessing the low-dose risk of cardiovascular disease is the shape of the dose response relationship , which is unclear in the Japanese data . The analysis of the present paper suggests that linear extrapolation would be appropriate for this endpoint .
Atherosclerosis is the main cause of coronary heart disease and stroke , the two major causes of death in developed society [1] . Though previously initiation of atherosclerosis was attributed mainly to lipid accumulation within the arterial walls , it is now widely accepted that inflammation plays a vital role in the initiation and progression of the disease [2]–[5] . For some time cardiovascular effects of high dose radiotherapy ( RT ) have been known [6] , [7] . A variety of effects are observed , presumed to result from inactivation of large numbers of cells and associated functional impairment of the affected tissue . Among such effects are direct damage to the structures of the heart – including marked diffuse fibrotic damage , especially of the pericardium and myocardium , pericardial adhesions , microvascular damage and stenosis of the valves − and to the coronary arteries; these sorts of damage occur both in patients receiving RT and in experimental animals [6] . There is emerging evidence of excess risk of cardiovascular disease at much lower radiation doses and occurring a long time after radiation exposure in the Japanese atomic bomb survivor Life Span Study ( LSS ) cohort [8] , [9] and in various occupationally-exposed groups [10]–[14] although not in all ( e . g . , [15] ) . Assuming that they are causal , the likely mechanisms for such effects of low dose and/or chronic radiation exposures on cardiovascular disease are not clear [16] , [17] . It is of interest that elevated levels of the pro-inflammatory cytokines IL-6 , CRP , TNF-α and INF-γ , but also increased levels of the ( generally ) anti-inflammatory cytokine IL-10 , have been observed in the Japanese atomic bomb survivors [18] , [19] . There was also dose-related elevation in erythrocyte sedimentation rate and in levels of IgG , IgA and total immunoglobulins in this cohort , all markers of systemic inflammation [19] . In this paper we outline a mathematical formulation of a model of cardiovascular disease that is largely based on the inflammatory hypothesis articulated by Ross [2] , [3] . The motivation behind the mathematical modelling is to encompass various factors contributing to the inflammatory process and subsequently to atherosclerotic formation . As atherosclerosis is not only a multifactorial , but also a multi-step disease , we concentrate on modelling chronic inflammation , primarily at early stages in the disease , but outlining a treatment for the later stages that lead to plaque rupture . The model is to some extent based on a model of McKay et al . [20] , although there are significant departures from and elaborations of this model . In particular , features are borrowed from the generally rather simpler models of Cobbold et al . [21] and Ibragimov et al . [22] . Stability analysis of a simplified version of the model will be performed . We shall be particularly concerned with mechanisms for effects of cholesterol and fractionated low dose radiation exposure in this inflammation model , and outline a case for radiation-induced monocyte cell death as a candidate pathway .
In this section we shall consider a spatial atherosclerosis model based on a simplification of the biology outlined in Text S1 section A . 1 . The model is entirely concerned with processes in the intima ( the tissue immediately adjacent to the endothelial cell lining of the arteries ) , , where the disease process is thought to be initiated , with boundary conditions determined in part by species concentrations in the lumen , . Specifically , the model is concerned with atherosclerosis in the large arteries , for example the coronary , carotid and other cerebral arteries , lesions in which account for the largest part of cardiovascular morbidity and mortality [4] . In Text S1 section A . 2 we outline a rather fuller version of this spatial model , incorporating more of the biological detail of section A . 1 of Text S1 . The main point of the section is the stability analysis that we perform in the final part , this being the reason for the simplifications . The processes are a combination of stage 2 and stage 3 processes outlined in section A . 1 of Text S1 . The set of reaction-diffusion equations is as follows: ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) ( 6 ) ( 7 ) ( 8 ) ( 9 ) where are the undamaged and damaged EC concentrations , is the chemo-attractant ( monocyte chemo-attractant protein 1 ( MCP-1 ) ) concentration , is the proliferation factor ( macrophage colony-stimulating factor ( M-CSF ) ) , is the monocyte concentration , is the macrophage concentration , is the bound lipid concentration and is the necrotic core concentration . is the LDL concentration with oxidation state ( the number of vitamin E molecules unoxidized - 1 ) ( so is the LDL with all vitamin E oxidized , although itself unoxidised and is the fully oxidised LDL concentration ) . are the chemotactic factors ( assumed constant ) associated with monocytes , macrophages and T-lymphocytes , respectively; the mechanism for chemotaxis ( as given by the terms involving these coefficients ) is similar to that of Keller and Segel [23] , [24] . are the rates of diffusion of the associated species . MCP-1 ( also known as CCL2 ) is known to recruit monocytes , T-lymphocytes and dendritic cells to sites of tissue injury . Table S1 gives further details of candidate molecules for some of the model variables . While many of the components of these equations are standard ( further details are given in Text S1 section A ) , a few deserve further explanation . In equation ( 3 ) we assume that chemo-attractant is degraded ( via the term ) at a rate proportional to the concentration of macrophages , T-cells and monocytes; McKay et al . [20] do not assume such degradation . We assume this because chemo-attractant molecules are assumed to adhere to cell-surface markers on these cell species ( the mechanism by which they are assumed to attract ) ; a similar assumption was made by Ibragimov et al . [22] . In equation ( 7 ) we assume that the bound lipid concentration , , is increased ( internalised within macrophages ) at a rate determined by the concentration of macrophages , , and the concentration of fully oxidized LDL , ( the term in equation ( 7 ) ) , but that this bound lipid is released when the macrophages die ( the term in equation ( 7 ) ) , a function of macrophage concentration and bound lipid concentration , as given in McKay et al . [20] . As discussed in Text S1 section A ( equations ( A . 23 ) – ( A . 24 ) and preceding ) , we shall assume that and . The macrophage flux , as per equation ( 6 ) , is given by . Therefore the bound lipid flux ( all carried by macrophage diffusion and chemotaxis ) is . This leads to a chemotaxis term similar to that assumed by McKay et al . [20] . We obtain a diffusion term that is entirely due to macrophage diffusion , , in contrast with McKay et al . who assume a standard diffusion term in the lipid , ; we fail to see how bound lipid can diffuse apart from macrophages - by definition it is bound within macrophages . [ is the divergence ( div ) operator , is the gradient ( grad ) operator and is the Laplacean operator . ] Further details are given in Text S1 section A . Let be the outward unit normal on the boundary ( ) , where is the boundary between intima and lumen , and is the boundary between intima and media . Then we have: ( 10 ) ( 11 ) ( 12 ) ( 13 ) ( 14 ) ( 15 ) ( 16 ) [ are the monocyte and T-lymphocyte concentrations on the immediate lumenal side of the EC layer . ] These boundary conditions are similar to those assumed by Ibragimov et al . [22] . Note that in ( 12 ) and ( 15 ) we convert from the monocyte and T-lymphocyte flux ( respectively ) to the rate of change of concentration per unit distance ( respectively ) via the inverse of the respective diffusion constants . We assume the following parametric form of the boundary monocyte and T-lymphocyte flux in ( 12 ) and ( 15 ) : ( 17 ) ( 18 ) The fundamentally linear form of these is inspired by data in Takaku et al . [25] and Klouche et al . [26] . The threshold levels of chemo-attractant , , below which these fluxes are zero , is inspired by similar assumptions made by Ibragimov et al . [22]; as we discuss below , non-zero threshold levels are needed for there to be a stable solution .
We assume that the system is in spatial and temporal equilibrium at some time , and is subject to some perturbation after that point . Let be the equilibrium values of the various quantities , and let be the differences from these equilibrium values after perturbation – so that , for example , , and similarly for the other species . Therefore , for to be the equilibrium values we have by ( 1 ) – ( 9 ) : ( 19 ) ( 20 ) ( 21 ) ( 22 ) ( 23 ) ( 24 ) ( 25 ) ( 26 ) ( 27 ) In order that the system be in equilibrium , the boundary monocyte and T-lymphocyte flux must be zero , so we must have that ; for the remainder of this section we therefore assume this . Assuming the coefficients are non-trivial ( , , , ) these simplify to ( 19 ) and ( 20 ) and: ( 28 ) ( 29 ) Putting together ( 19 ) , ( 20 ) and ( 29 ) we see that: ( 30 ) This implies a non-linear relationship between LDL and the equilibrium chemo-attractant level . However , as is clear from Figures 1–2 , only for very high levels of LDL , multiples in excess of 50 of the baseline levels ( see Table S3 ) , are there appreciable departures from linearity . We shall often assume that . In all that follows we assume a limiting process , so that , . If we perform the obvious linearisations in equations ( 1 ) – ( 9 ) and ignore all second and higher order terms in we obtain: ( 31 ) ( 32 ) ( 33 ) ( 34 ) ( 35 ) ( 36 ) ( 37 ) ( 38 ) ( 39 ) The boundary conditions ( 10 ) – ( 16 ) , together with ( 17 ) , ( 18 ) translate to: ( 40 ) Green's first identity ( for general scalar functions and domain ) states that: ( 41 ) Integrating ( 33 ) – ( 39 ) over the intima , , we have by ( 41 ) that: ( 42 ) ( 43 ) ( 44 ) ( 45 ) ( 46 ) ( 47 ) ( 48 ) In Text S1 section B we outline solutions to ( 31 ) – ( 32 ) , ( 42 ) – ( 48 ) in various cases . As shown there , if and then: ( 49 ) From Table S2 it is clear that these conditions are likely to be always satisfied: and . If then using the results of Text S1 section B ( ( B . 11 ) – ( B . 16 ) ) : ( 50 ) ( 51 ) ( 52 ) ( 53 ) ( 54 ) ( 55 ) for some , so that in general do not decay to 0 as , although by ( B . 18 ) does ( at least in ) . Note in particular that unless and , or possibly ( more likely ) that , then there will be time trends in the averaged quantities for , and , so that in particular stability cannot be re-attained . In Figures 3–7 we plot the variation of the spatially-averaged chemo-attractant , , derived assuming a non-zero equilibrium concentration of monocytes ( Table S3 ) and using ( B . 1b’ ) . The perturbation is assumed to take place via killing of monocytes in the intima , which in this case could be produced by ionizing radiation , and also via damage to endothelial cells , produced in the same way . We do not assume instantaneous changes in any of the other species , i . e . , . The reason for this is that by ( 28 ) ( assuming as we do that ) , so that radiation would not have any species to act on in equilibrium . For the parameters used here ( given in Tables S2 and S3 ) , the overwhelming contribution is via monocyte killing: by 80 seconds the contribution from this term is 4 . 5×10−17 M ml−1 compared with a contribution of −1 . 6×10−18 M ml−1 via damage to endothelial cells . As can be seen , the change in chemo-attractant concentration occurs ( for monocytes and in aggregate ) relatively quickly , over a timescale of minutes , although the endothelial cell killing component varies more slowly , over a timescale of hours; after 24 hours this and all other averaged quantities are virtually constant . In contrast to the above general case , when only the monocyte population ( of all the species ) is perturbed , for a sufficiently long time after exposure ( days or more ) we have by ( 50 ) , ( 52 ) that and , and by ( 51 ) , ( 53 ) , ( 54 ) the averaged change in all other species tends to zero . In this case we see that: ( 56 ) approximated to first order , so that by ( 30 ) , at least in average , equilibrium can be re-established at these new values of and . We conjecture that in fact equilibrium is re-established for all quantities in this case . It is easy to see that if there were to be further small perturbations in , at intervals of days or more , the resulting changes in the spatially-averaged quantities would be approximately additive in the corresponding increments , as shown in Figure 7 . Moreover , from ( 50 ) the excess chemo-attractant ( MCP-1 ) in relation to the monocyte perturbation is . Therefore , so long as the individual monocyte perturbations are small and temporally separated ( by a day or more ) , the increment in chemo-attractant will not depend on anything other than the cumulative absorbed dose , as indicated in Figure 7 . In Figure 8 we plot the percent proportion of the population whose cumulative chemo-attractant ( MCP-1 ) concentration exceeds the threshold ; as we discuss below , this threshold is the critical point for system stability , exceedance of which makes development of cardiovascular disease much more likely . [The probability is derived assuming that the population distribution MCP-1 is Gaussian with mean and standard deviation ( SD ) determined by the adult female data of Cannon et al . [27]; the mean is augmented by the radiation-induced increment , given by ( B . 1b’ ) . ] For a range of threshold values between 0 . 25 and 1 . 00 times the SD in excess of the mean , we have baseline risks of exceeding the threshold ( i . e . , cardiovascular disease ) of 16–40% . Most developed countries have cumulative cardiovascular disease mortality in the range 20–40% and the world mean is 30% [28] , so that this range of values of the MCP-1 threshold , , is plausible . For this range of threshold values , Figure 8 demonstrates that risks vary remarkably linearly with dose over the dose interval 0–4 Gy . As for Figure 7 , the risk will not depend on anything other than the cumulative absorbed dose , as long as this is given in small daily increments .
We have outlined a model for early stage atherosclerotic lesion formation , and performed a stability analysis for a simplified version of the model . While some components of the system ( in particular the T-lymphocyte concentration , ) are stable , in the sense that after perturbations of the system the species concentrations return to their equilibrium value , various other species , in particular the proliferation factor concentration , , the chemo-attractant concentration , , the monocyte concentration , , and the necrotic core , , are generally not stable . In particular , the mean level of chemo-attractant increases continuously and rapidly after instantaneous perturbation by a radiation dose , over a timescale of minutes . However , as we note below , because of cellular repair processes , which are not taken into account in our model , there are reasons for assuming that perturbation by radiation would not be instantaneous , so that this process might be extended over at least a period of hours after exposure . The main driver for the increase in chemo-attractant is the death of monocytes and the consequent reduction in monocyte-induced degradation in chemo-attractant concentration , the term in ( 3 ) . It is well known that radiation can cause cell death [29] , and the degree of cell killing and damage that we assume is consistent with radio-biological expectation [30] , [31] . Although the change in chemo-attractant ( MCP-1 ) concentration that we assume after 10 mGy is relatively modest , 4 . 5×10−17 M ml−1 , a fractionated dose of 1 Gy would result in 4 . 5×10−15 M ml−1 , comparable with the normal concentration of MCP-1 in adult plasma , 7 . 9×10−15 M ml−1 [27] . The fact that the range of excess relative risks predicted by our model , 0 . 49–0 . 93 Gy−1 , is consistent with those in a number of occupational studies ( Table 1 ) adds to the plausibility of this mechanism . We have also shown that the model predicts that equilibrium level of chemo-attractant ( MCP-1 ) increases more or less directly with levels of LDL , and in particular oxidized LDL , with slight non-linearity at very high levels of MCP-1 . This is in accordance with experimental [32] , [33] and epidemiological observations [34] . Specifically , there is experimental evidence that addition of minimally-oxidised LDL results in a ≈22-fold increase in levels of MCP-1 in ECs in an in vitro co-culture system [32] . In a group of baboons fed a high cholesterol , high fat diet , oxLDL in serum increased by about 19 . 6% ( 95% −28 . 9 , 68 . 1 ) after 7 weeks , resulting in an increase in serum MCP-1 at that point of 66 . 7% ( 95% 54 . 2 , 79 . 1 ) [33] . Both of these are consistent with the linear relationship ( without constant term ) predicted by our model ( 30 ) . If radiation dose were to be given in a fractionated manner , with doses separated by a period of hours or more , the model predicts that chemo-attractant ( e . g . , MCP-1 ) would increase linearly with cumulative accumulated dose , with a corresponding decrease in the intimal monocyte concentration , as shown in Figure 7 . This would carry on until the chemo-attractant concentration at the boundary , , exceeds one or other of the thresholds , beyond which point an equilibrium solution is no longer possible . At these points , there would be increased trans-intimal flux of monocytes and T-lymphocytes from the lumen , which would result ( via ( 48 ) ) in a continuous increase in necrotic lesion size , and therefore risk of atherosclerosis . The doses used here are moderate ( 10 mGy/day ) , such as might occur in occupational exposure settings , and would account for the observed radiation-associated excess risk that has been seen in various groups of nuclear workers [10]–[14] . The model implies that at least until the chemo-attractant threshold is exceeded the system is stable , assuming that the conjecture we make after ( 56 ) is valid . If the chemo-attractant threshold is exceeded as a result of the perturbation term , , resulting in monocyte or T-lymphocyte flux across the EC layer , then extra terms need adding to the right hand side of ( 44 ) and ( 47 ) , ( 57 ) and ( 58 ) respectively . Apparently paradoxically , if , then we must have for the terms inside the integrals to contribute non-trivially , and so these terms will be negative and therefore tend to reduce the averaged levels of monocytes and T-lymphocytes in the system . By ( 42 ) this will tend to increase the chemo-attractant concentration still further . In other words , once this chemo-attractant threshold is crossed the system tends ( on average ) to become yet more unstable . There are of course other agents that damage monocytes or ECs that would cause the chemo-attractant level to increase , so that although for an individual this threshold might never be passed , in a large population there would be a continuous ( and approximately linear ) increase in cardiovascular risk with dose as shown in Figure 8 . The same phenomenon would also occur at higher doses ( e . g . , at radiotherapeutic levels of dose ) , at a correspondingly higher level , although the relative magnitude of the perturbations would make the neglect of all but first order perturbations that we assumed in deriving ( 31 ) – ( 39 ) possibly invalid; there is abundant evidence of radiation-induced disease in groups exposed to certain forms of RT [6] , [7] . Critical to our model , and indeed the understanding of atherosclerosis , is whether there really are such thresholds in chemo-attractant levels for the trans-intimal monocyte and T-lymphocyte flux . We assume the presence of such thresholds for the purposes of our stability analysis , as we have to if there is to be a stable solution , but it is possible nevertheless that these thresholds are zero , in which case , assuming the model is correct , the atherosclerotic process must be inherently unstable . As indicated above , if this is model is correct and is to be consistent with the observed cumulative cardiovascular disease mortality in developed populations [28] , then the chemo-attractant ( MCP-1 ) threshold must lie in the range [mean+0 . 25×population SD , mean+1 . 00× population SD] ( the mean and population SD being as in Cannon et al . [27] ) , in other words [1 . 0 , 1 . 7]×10−14 M ml−1 . We implicitly assume that atherosclerosis is mainly responsible for the observed excess risk of cardiovascular morbidity or mortality following fractionated low-dose irradiation of the heart and major arteries . This assumption is supported by experimental data in ApoE−/− mice [35] . However , some human symptoms are due to ( myocardial ) ischaemia which could be caused by either macrovascular ( atherosclerotic ) or microvascular damage . At higher ( radiotherapy ) doses , both human and animal data suggest that both types of lesion occur [6] . Although the generally high prevalence of atherosclerosis in humans suggests that this is the more probable cause of ischaemia following low-dose radiation , it is possible that microvascular disease also plays a role . It should be noted that we have been addressing mechanisms for induction of atherosclerosis following fractionated low-dose radiation to the large arteries ( coronary , carotid etc ) . There is a large literature on fibrotic , pericardial , myocardial and other morbidity sequelae of high-dose irradiation of the heart and large arteries , both for humans and animals [6] . The pro-inflammatory mechanisms for these are reasonably well understood , and quite different from those hypothesized here [36] . That the true mechanisms for low-dose effects are likely to be very different is also suggested by the pronounced fractionation effect seen for high-dose exposure in relation to heart failure in rats [37] , [38] , in contrast to the somewhat lower risks observed in the Japanese atomic bomb survivors compared with occupationally exposed groups ( Table 1 ) . Indirect mechanisms for the action of radiation could also be postulated . At high doses it is clear that inflammatory markers are up-regulated in vitro and in vivo , although at lower doses if anything the evidence points to down-regulation of inflammation [16] . In terms of the model this could be mediated by an increase in radical flux , which could , via lipid peroxidation , lead to EC damage . This in turn would lead to an increase in the chemo-attractant signal . Radiation is known to cause long-term variation in certain T-cell subpopulations ( CD4+ ) in the Japanese atomic bomb survivors [39] , and this mechanism could also be readily incorporated in the model . Long-term radiation-associated changes in cholesterol concentration have been observed in the Japanese atomic bomb survivors [40] , presumably a result of some change in liver metabolism; these too could be easily incorporated in the model . It is of interest in this respect that there is a highly statistically significant trend with internal ( plutonium α-particle ) dose to the liver for ischaemic heart disease and cerebrovascular disease in the latest analysis of the Mayak worker data [14] . Set against that , there is little evidence of excess risk of circulatory disease risk , specifically cardiac disease in groups exposed to the diagnostic contrast medium Thorotrast , which delivered a substantial α-particle liver dose [41] , [42] . An important consideration in estimating dose to the intima , and which may have a bearing on interpretation of certain epidemiological studies , is the role of oxygen diffusion . This has been modelled by Richardson [43]–[45] , who has highlighted the pronounced variations with oxygen concentration across the intima , which also varies with age as a result of modifications in arterial geometry [44] . It is well known that with decreasing oxygenation the effective dose reduces [45] , and this implies that biologically effective dose per unit exposure reduces by 8–12% from age 0 . 5 to 70 years , whether for high linear energy transfer ( LET ) ( 222Rn , 218Po , 214Po ) or for low LET radiation [45] . This needs to be addressed in the dosimetry of any study; assuming that , as we argue above , intimal dose is of the most relevance to cardiovascular risk , not doing so would imply a modest negative bias in modifications of the radiation response by age at exposure . Whilst the inflammatory process is recognized as an integral part of the atherosclerotic process [5] it does not explain the observation that the proliferation of vascular smooth muscle cells ( VSMC ) during atherosclerotic plaque development appears to be monoclonal [46] . Clonality suggests that plaque VSMCs must have undergone multiple rounds of division , and telomere loss studies argue that this is between 7–13 cumulative population doublings [47] . However , clonality itself is not synonymous with transformation of a single cell , and subsequent studies have shown that large patches in the normal vessel media are monoclonal [48] , [49] . Thus , clonality is more likely to be explained by the presence of developmental clones in the normal vessel wall , rather than a mutation . Finally , in contrast to tumours , plaque VSMCs show poor proliferation , enhanced apoptosis , and early senescence [50] . These features would not confer a proliferative or survival advantage to plaque VSMCs . Furthermore , plaque VSMC proliferation is now seen to be beneficial in atherosclerosis [51] , so that the pathological consequences of a mutation promoting VSMC proliferation are unclear . The limitations of the modelling performed here should be acknowledged . Even in the fuller model considered in Text S1 section A there is much biology not included – simplifications have been made for analytical simplicity . Although not strictly a defect in the model , we assume in our motivating example that a certain ( dose-dependent ) fraction of the monocytes are killed instantaneously by radiation exposure . The magnitude of this fraction is based on data from a human bone-marrow colony-forming assay ( for cells under hypoxic conditions ) of Gordon [30] ( Table S3 ) , performed 9 days after irradiation . It is known that cells take a variable length of time to die after irradiation , as a result of the repair and mis-repair processes they are thought to be subject to [52] . As such , a possibly more realistic scenario would have assumed this total cell damage exponentially distributed over time rather than occurring instantaneously . However , it is unlikely that the variable delay in expression of monocyte mortality , which is likely to be 99 . 7% complete within three hours of irradiation [52] , will make much difference to the predictions of our model , concerned as it largely is with the consequences of fractionated radiation doses separated by days or more . It would not be too difficult to modify the equations ( 5 ) , ( 6 ) and ( 8 ) to incorporate the simple repair-misrepair model outlined in Brenner et al . [52] , although for the purposes of the present paper we regard this as an unnecessary elaboration . That said , the simpler model proposed here we trust captures what is known about the main features of interaction of oxidized LDL and various other molecular species ( MCP-1 , G-CSF , bound lipid ) with the various cellular species ( monocytes , macrophages , T-lymphocytes ) that are known to be of significance for induction of atherosclerosis . The mathematics underlying these reaction and diffusion processes is reasonably standard . What is interesting and novel about the present paper is that using only experimentally derived parameters ( taken wherever possible from human data ) ( Tables S2 , S3 ) we have reproduced what is observed in other experimental and epidemiologic data ( Figures 7–8 , Table 1 ) . This proposed mechanism would in principle be experimentally testable . This would best be done in vitro , looking for changes in MCP-1 levels , or other potential chemo-attractants , in a co-culture system similar to that developed by Takaku et al . [25] . This could be explored under a range of radiation exposure conditions ( both localized and fractionated ) and subsequent effects on , for example , adhesion properties could also be examined . In vivo experiments would be more complex ( and expensive ) , but could also be performed , for example , using the ApoE−/− knockout mouse model employed by Stewart et al . [35] , [53] . Even human data could be envisaged . In particular , if arterial tissue could be sampled from patients who have , a short time previously , received low-dose radiotherapy or high-dose diagnostic procedures ( e . g . , computerized tomography ) , together with suitable ( age-matched ) controls , one could determine whether intimal concentration of MCP-1 was significantly increased and the manner in which concentration changed with dose . If the proposed mechanism were true , it also has substantive implications for radiological protection , which at present does not take cardiovascular disease into account [54] . Analysis of the Japanese atomic bomb survivor data implies that non-cancer disease mortality , in particular cardiovascular mortality , contributes almost equally as cancer mortality to the radiogenic excess risk [8] . The major uncertainty in assessing the low-dose risk of cardiovascular disease is the shape of the dose response relationship , which is very unclear in the Japanese data [8] , [55] . The analysis of the present paper suggests that linear extrapolation would be generally appropriate for this endpoint . | Atherosclerosis is the main cause of coronary heart disease and stroke , the two major causes of death in developed society . There is emerging evidence of excess risk of cardiovascular disease in various occupationally exposed groups , exposed to fractionated radiation doses with small doses/fraction . The mechanisms for such effects of fractionated low-dose radiation exposures on cardiovascular disease are unclear . We outline a spatial reaction-diffusion model for early stage atherosclerotic lesion formation and perform a stability analysis , based on experimentally derived parameters . We show that following multiple small radiation doses the chemo-attractant ( MCP-1 ) concentration increases proportionally to cumulative dose; this is driven by radiation-induced monocyte death . This will result in risk of atherosclerosis increasing approximately linearly with cumulative dose . This proposed mechanism would be testable . If true , it also has substantive implications for radiological protection , which at present does not take cardiovascular disease into account . The major uncertainty in assessing low-dose risk of cardiovascular disease is the shape of the dose response relationship , which is unclear in high dose data . Our analysis suggests that linear extrapolation would be appropriate . | [
"Abstract",
"Introduction",
"Models",
"Results",
"Discussion"
] | [
"public",
"health",
"and",
"epidemiology/occupational",
"and",
"industrial",
"medicine",
"cardiovascular",
"disorders/vascular",
"biology",
"mathematics",
"public",
"health",
"and",
"epidemiology/epidemiology",
"cardiovascular",
"disorders/coronary",
"artery",
"disease",
"cardi... | 2009 | A Model of Cardiovascular Disease Giving a Plausible Mechanism for the Effect of Fractionated Low-Dose Ionizing Radiation Exposure |
Type I chaperonins are large , double-ring complexes present in bacteria ( GroEL ) , mitochondria ( Hsp60 ) , and chloroplasts ( Cpn60 ) , which are involved in mediating the folding of newly synthesized , translocated , or stress-denatured proteins . In Escherichia coli , GroEL comprises 14 identical subunits and has been exquisitely optimized to fold its broad range of substrates . However , multiple Cpn60 subunits with different expression profiles have evolved in chloroplasts . Here , we show that , in Arabidopsis thaliana , the minor subunit Cpn60β4 forms a heterooligomeric Cpn60 complex with Cpn60α1 and Cpn60β1–β3 and is specifically required for the folding of NdhH , a subunit of the chloroplast NADH dehydrogenase-like complex ( NDH ) . Other Cpn60β subunits cannot complement the function of Cpn60β4 . Furthermore , the unique C-terminus of Cpn60β4 is required for the full activity of the unique Cpn60 complex containing Cpn60β4 for folding of NdhH . Our findings suggest that this unusual kind of subunit enables the Cpn60 complex to assist the folding of some particular substrates , whereas other dominant Cpn60 subunits maintain a housekeeping chaperonin function by facilitating the folding of other obligate substrates .
Chaperonins are large double-ring assemblies that assist in the efficient folding of substrate proteins ( reviewed in [1]–[3] ) . Two types of chaperonins have been identified: type I in bacteria ( GroEL ) , mitochondria ( Hsp60 ) , and chloroplasts ( Cpn60 ) , and type II in archaea ( thermosome ) and eukaryotic ( TRiC/CCT ) cytosol ( reviewed in [4] ) . Whereas a type I chaperonin ring is composed of seven subunits , a type II chaperonin ring consists of eight or nine subunits that are not identical but are homologous . Type I chaperonin requires co-chaperonin GroES/Hsp10 for substrate encapsulation , whereas type II chaperonin is independent of GroES/Hsp10 factors . Both types of chaperonins utilize ATP as energy to drive a series of structural rearrangements that allow them to capture , encapsulate , and release the substrate proteins ( reviewed in [4] ) . The GroEL/GroES complex from Escherichia coli ( E . coli ) represents the type I chaperonins and its structure and function have been studied extensively ( reviewed in [4] , [5] ) . GroEL consists of 14 identical subunits of ∼57 kDa and these subunits form two heptameric rings stacked back-to-back with a central cavity in each ring [6] . Each subunit contains three domains . An equatorial domain comprises the ATP/ADP binding site and an apical domain contains the hydrophobic surface toward the ring cavity for polypeptide binding . The intermediate domain links the equatorial and apical domains [6] , [7] . The co-chaperonin GroES is a homoheptameric single-ring composed of 10 kDa subunits [8] . GroES can rapidly bind the substrate-captured GroEL ring ( cis ring ) in the presence of ATP; hence , the GroEL/GroES complex provides an encapsulated cavity for protein folding [9] , [10] . Due to structural rearrangements in the apical and intermediate domains , the cis cavity becomes enlarged and the physical features of the cavity wall change . This process lasts about 10–15 s and is accompanied by the hydrolysis of seven ATP molecules . After hydrolysis , ATP and other non-native peptides bind to the GroEL in the trans ring , triggering dissociation of the GroES from the opposite ring . The folded protein is then released from the chaperonin complex ( reviewed in [1]–[5] ) . Proteome-wide analysis of chaperonin-dependent protein folding has shown that GroEL interacts with about 250 different proteins and these substrates are categorized into three classes [11] . The class I substrates are independent of GroEL/GroES , whereas class II substrates are partially dependent on GroEL/GroES , and they can utilize other chaperone systems , such as DnaK , for folding . A total of 84 proteins are grouped into class III and they are potential obligate substrates of GroEL/GroES in vivo [11] . More recently , Fujiwara et al . [12] employed a more direct approach by testing the solubility of class III substrates in GroE-depleted cells and found that only ∼60% ( 49 out of 84 ) of the class III proteins are absolutely dependent on GroEL/GroES for folding . Furthermore , an additional eight proteins that were not identified as class III proteins were also found to be GroEL/GroES obligate substrates and the authors defined these 57 proteins as class IV obligate substrates [12] . The majority of the class IV proteins are involved in metabolic reactions . Bioinformatic analysis has shown that nearly half of the class IV proteins contain TIM-barrel folds . In addition to these TIM-barrel folds , FAD/NAD ( P ) -binding domains , PLP-dependent transferase-like folds , and thiolase folds are also highly enriched in the group [12] . These data suggest that GroEL/GroES has been optimized to facilitate the folding of a variety of substrates during evolution . The basic features of the mechanisms for GroEL/GroES-mediated folding of the nonnative substrates have been demonstrated by a great number of functional and structural studies . However , these studies have focused primarily on the model chaperonin system that is composed of uniform subunits , such as GroEL from E . coli . In contrast to E . coli , nearly 30% of bacterial genomes contain two or more chaperonin genes [13] . Furthermore , almost all mitochondria and chloroplasts studied in higher plants possess multiple chaperonin subunits [14] . There have been few reports focusing on the role played by multiple chaperonin genes . In Sinorhizobium meliloti , one of the five GroEL paralogs , GroEL1 was shown to be required for NodD protein folding [15] . However , overexpression of another GroEL protein can suppress the defect of the groEL1 mutant [15] . There are similar reports in other bacteria; Bradyrhizobium japonicum possesses at least five highly conserved groESL operons . Although nitrogen fixation activity was reduced to approximately 5% of the wild-type ( WT ) level in the double mutant defective in groEL3 and groEL4 , overexpression of two of the other groESL operons partially suppressed this phenotype [16] . Of the three chaperonin genes of Rhizobium leguminosarum , only Cpn60 . 1 is essential for growth . Overexpression of the Cpn60 . 3 gene in the cpn60 . 1 mutant sustains bacterial growth , but the complemented strain is sensitive to high temperature , suggesting that Cpn60 . 3 does not facilitate the folding of particular proteins supporting the growth of bacteria at high temperature [17] . By contrast , the specificity of GroEL1 function in Mycobacterium smegmatis seems to be absolute . This chaperonin is not essential for growth but is required for mycolic acid biosynthesis during mature biofilm formation [18] . GroEL1 may be specifically involved in the folding of two proteins , KasA and SmEG4308 , which are required for mycolic acid synthesis , or in converting KasA between two isoforms [18] . These lines of evidence suggest that the functions of the multiple chaperonin subunits are specialized , although they have different degrees of specificity . Chloroplast type I chaperonin complex ( Cpn60 ) is similar in structure to GroEL and also consists of two stacked heptameric rings [19] , [20] . In contrast to GroEL , which is composed of identical subunits , Cpn60 comprises two different subunit types , Cpn60α and Cpn60β [21]–[23] , and they are only approximately 50% identical to each other [14] . In vitro reconstitution studies of the chloroplast Cpn60 complex suggested a stoichiometry of α7β7 in the Cpn60 complex [24] , which is in accordance with the observation that roughly equal amounts of α and β subunits are present in chaperonin oligomers purified from spinach chloroplasts [25] . However , it is still unclear how these subunits are organized within a complex . Furthermore , the Arabidopsis thaliana genome contains two genes encoding Cpn60α subunits and four genes encoding Cpn60β subunits [14] , and they have different expression profiles [26] , [27] . Cpn60α1 ( At2g28000 ) , Cpn60β1 ( At1g55490 ) , and Cpn60β2 ( At3g13470 ) are the dominant Cpn60 subunits , whereas Cpn60α2 ( At5g18820 ) , Cpn60β3 ( At5g56500 ) , and Cpn60β4 ( At1g26230 ) are present at very low levels . Disruption of the Cpn60α1 gene results in general defects in plastid function , leading to embryonic lethality [28] , which highlights the critical role of Cpn60α1 in maintaining plastid function . The cpn60β1β2 double mutant also shows the lethal phenotype [29] , suggesting that Cpn60α1 and Cpn60β1–β2 form a heterooligomer that provides the housekeeping chaperonin function in chloroplasts by assisting the folding of a wide range of proteins . Multiple subunits also occur in type II chaperonin CCT and certain subunits are responsible for the binding of specific substrates , such as actin and tubulin [30] . Recently , it has been suggested that different subunits of CCT play unique roles in determining substrate specificity [31] . However , few reports have focused on the function of the multiple subunits in the chloroplast chaperonin system . In particular , whereas the amino acid sequences of the Cpn60β1–β3 subunits share 90%–95% identity , Cpn60β4 is only 60% identical to each of the other three Cpn60β subunits [14] . So far , there has been no explanation why plants evolved this unusual kind of Cpn60β subunit . In this study , we showed that Cpn60β4 is strictly and specifically required for the folding of the NdhH protein , a subunit of the chloroplast NADH dehydrogenase-like complex ( NDH ) .
NDH is a multi-subunit complex embedded in the thylakoid membrane and is involved in chlororespiration and photosystem I ( PSI ) cyclic electron transport ( Figure 1A ) [32] . The activity of NDH can be monitored as a post-illumination rise in chlorophyll fluorescence ( Figure 1B ) due to reduction of the plastoquinone ( PQ ) pool by the NDH complex in the dark [33] . Based on this phenomenon , we isolated dozens of Arabidopsis mutants specifically defective in NDH activity , which we referred to as chlororespiratory reduction ( crr ) mutants . Characterization of these mutants led to the identification of several NDH subunits and a large body of proteins involved in the expression of subunit genes and the assembly or stabilization of the NDH complex ( reviewed in [34] , [35] ) . Here , we identify two mutants , cpn60α1 and crr27 , neither of which showed the increase in fluorescence after actinic light ( AL ) illumination , indicating impaired NDH activity ( Figure 1B ) . In contrast to crr mutants , the cpn60α1 mutant exhibited retarded growth and pale green-leaf phenotypes ( Figure 1C ) . Map-based cloning identified a single amino acid substitution ( D335A ) at a conserved position in cpn60α1 ( Figure 1D ) . Although the total levels of Cpn60α and Cpn60β were increased in cpn60α1 , possibly due to complementation effects , the levels of many other chloroplast proteins , including NDH subunits , were reduced to various extents ( Figure S1 ) , supporting the idea that Cpn60 has a diverse set of substrates . The reduction in NDH ( 25%–50% ) in cpn60α1 at least partly explains the failure to detect NDH activity by chlorophyll fluorescence . The crr27 mutants were isolated by screening Ds transposon-tagged lines using PAM ( pulse amplitude modulation ) fluorometry [36] , [37] . Unlike cpn60α1 , crr27 mutants did not exhibit any visible phenotype besides impaired NDH activity ( Figure 1B and 1C ) . The Cpn60β4 gene was knocked out by Ds or T-DNA insertions in three crr27 alleles , and reverse transcription ( RT ) -PCR analysis did not detect any Cpn60β4 transcripts ( Figure 1E and 1F ) . Full NDH activity was rescued by the introduction of the WT Cpn60β4 gene into crr27-1 ( Figure 1B ) . Two chlorophyll fluorescence parameters , ETR ( electron transport rate ) and NPQ ( nonphotochemical quenching ) , indicate subtle defects in photosynthesis and are therefore often used to characterize mutants with defective photosynthetic apparatus . ETR was only slightly reduced and NPQ was not affected in crr27-1 ( Figure S2 ) , which is consistent with the phenotypes of other crr mutants with specific defects in NDH activity . NDH interacts with at least two copies of PSI to form the NDH-PSI supercomplex ( Figure 2A ) [35] , which can be separated by blue native ( BN ) -PAGE [38] , [39] . To study the role of Cpn60β4 in biogenesis of the NDH complex , thylakoid protein complexes from WT and crr27 mutants were separated by BN-PAGE . No difference was found in the major complex bands between WT and crr27 ( Figures 2B and S3A ) . However , band I , corresponding to the NDH-PSI supercomplex detected in WT , was replaced by band II , corresponding to the subsupercomplex in crr27 , as in the NdhL-defective ndhl ( crr23 ) mutant ( Figure 2B ) [38] . Based on extensive genetic and biochemical characterizations , we divided the NDH complex into four categories: membrane , lumen , and A and B subcomplexes ( Figure 2A ) [39] . Previous mass analysis revealed that only subcomplex A , which is composed of four plastid-encoded subunits ( NdhH–NdhK ) and four nucleus-encoded subunits ( NdhL–NdhO ) , was absent in band II ( Figure 2A ) [39] . Immunoblot analysis confirmed that the levels of subcomplex A subunits NdhH and NdhL were dramatically decreased in crr27 , whereas subunits of the other NDH subcomplexes , NDH18 , NDF1 , and FKBP16-2 , were only slightly reduced ( Figure 2C ) . Consistent with the invisible growth phenotype of crr27 , identical levels of D1 ( PSII complex ) , PsaA ( PSI complex ) , and cytochrome ( Cyt ) f ( Cyt b6f complex ) were detected in crr27 and WT ( Figure 2C ) . In addition , no significant difference in stromal protein levels was detected between WT and crr27-1 mutants by clear native ( CN ) -PAGE and subsequent two-dimensional ( 2D ) /SDS-PAGE ( Figure S3B ) . From these results , we conclude that the accumulation of NDH subcomplex A was specifically impaired in the absence of Cpn60β4 . Previous transcriptomic analysis [26] indicated that the expression level of the Cpn60β4 gene is lower than that of other Cpn60β genes . Furthermore , the Cpn60β4 subunit could not be visualized with Coomassie Brilliant Blue ( CBB ) staining in the 2D CN/SDS-PAGE gel , whereas other Cpn60β subunits and Cpn60α1 were detected [27] , suggesting that the stoichiometry of Cpn60β4 is extremely low compared to the other Cpn60β subunits . To confirm the accumulation of Cpn60β4 in WT , we separated total stromal protein complexes isolated from WT plants by CN-PAGE . The protein band corresponding to the position of the Cpn60 complex was excised from the gel ( Figure S4A ) and analyzed by liquid chromatography-tandem mass spectrometry ( LC-MS/MS ) analysis using the linear ion-trap triple quadrupole ( LTQ ) -Orbitrap XL-HTC-PAL system , which provides high mass accuracy , high resolution , and high sensitivity . The values of Mowse score , Protein match , and emPAI ( exponentially modified Protein Abundance Index ) are commonly used to estimate relative protein levels . LC-MS/MS analyses detected the Cpn60β4 protein , but its level was significantly lower than those of the other three Cpn60β proteins ( Figure S4B; Table S1 ) . Consistent with the apparent mutant phenotype ( Figures 1 and 2 ) , we confirmed the accumulation of Cpn60β4 in WT ( Figure S4C ) . In contrast to cpn60α1 , the crr27 mutation did not affect total Cpn60α or Cpn60β levels ( Figures 3A and S1A ) . To study the role of Cpn60β4 , a chimeric gene encoding an HA ( influenza hemagglutinin protein epitope ) tag fused to the C-terminus of Cpn60β4 was introduced into crr27-1 . This transformation fully restored NDH activity ( Figure 1B ) , indicating that the HA-tag did not affect the function of Cpn60β4 . We also overexpressed HA-tagged Cpn60β1 and Cpn60β4 in crr27-1 under control of the cauliflower mosaic virus ( CaMV ) 35S promoter . NDH activity and NDH subcomplex A level were rescued only in the 35S::Cpn60β4-HA lines ( Figures 1B and 2C ) , indicating that Cpn60β1 cannot complement the function of Cpn60β4 , even under the control of the same promoter . Cpn60β4 localized to the chloroplast stroma ( Figure 3A ) and co-migrated with other Cpn60β and Cpn60α subunits in CN-PAGE ( Figures 3B and S4 ) , suggesting that Cpn60β4 is an intrinsic subunit of the Cpn60 complex . To examine this possibility , HA-tagged Cpn60β4 was enriched from the stromal fraction isolated from crr27-1 plants complemented with Cpn60β4-HA using the µMACS HA isolation kit ( Miltenyi Biotec ) under previously established conditions [11] . Because the additional HA tag in the C-terminus does not affect the function of Cpn60β4 ( Figure 1B ) , their interacting proteins might also be co-purified . Total isolated proteins were separated by SDS-PAGE ( Figure S5A ) and further analyzed by LC-MS/MS analysis . Both Cpn60α1 and Cpn60β1–β4 subunits were detected in the purified sample and MS analysis showed that they were the most abundant proteins ( Tables 1 and S2 ) , implying that Cpn60β4 forms a specific heterooligomeric Cpn60 complex with Cpn60α1 and other Cpn60β subunits . The molecular masses of Cpn60β4-HA , Cpn60β1–β3 , and Cpn60α1 are 64 . 4 , 58 . 2 , and 57 . 1 kDa , respectively , which enables us to distinguish them in SDS-PAGE . The purified proteins were subjected to 7 . 5% SDS-PAGE and three major bands with molecular masses of approximately 60 kDa were visualized with CBB staining ( Figure 3C ) . Based on the mobility , three bands should correspond to Cpn60β4-HA , Cpn60β1–β3 , and Cpn60α1 ( Figure 3C ) . Quantitative estimation of these signals showed that the level of Cpn60α1 is about 50% of all of the chaperonin subunits , which is consistent with the proposal that the Cpn60 complex is composed in a α7β7 stoichiometry [24] , [25] . However , the level of Cpn60β4 in the chaperonin complex was lower ( ∼15% ) ( Figure 3C ) , implying that approximately two molecules of Cpn60β4 are included in double rings , as well as five molecules of other Cpn60β subunits . Based on the stoichiometry of the Cpn60β4 subunit in total Cpn60 subunits [26] , [27] , the majority of the Cpn60 complex is unlikely to contain Cpn60β4 , and we estimated its stoichiometry in the specific complex including Cpn60β4 . Given that Cpn60β4 is an intrinsic subunit of the chaperonin complex ( Figure 3 ) and that the NDH subcomplex A was missing in the crr27 mutants ( Figure 2 ) , it is very likely that the specific chaperonin complex containing Cpn60β4 is required for the folding of at least one subunit of the NDH subcomplex A . If this is the case , the interacting NDH subunit would be copurified with Cpn60β4-HA . To determine differences in substrate specificity , protein complexes containing Cpn60β1 were isolated using the 35S::Cpn60β1-HA lines . Neither Cpn60 nor NDH subunits were detected in untransformed WT plants , which were used as a negative control ( Figure S5A; Table S2 ) , excluding the possibility of non-specific binding to the magnetic beads . Cpn60α1 and Cpn60β1–β3 were co-purified with both Cpn60β1 and Cpn60β4 ( Tables 1 and S2 ) . These results confirmed that Cpn60α1 and Cpn60β1–β4 form a heterooligomeric complex in vivo . The ratio between protein emPAI scores can be used to estimate the relative amounts of protein in the different samples [40] . The emPAI ratios of Cpn60α1 and Cpn60β1–β3 detected in Cpn60β4- and Cpn60β1-purified samples were 3 . 9 and 0 . 47–1 . 26 ( Cpn60β4/Cpn60β1 ) , respectively , suggesting that comparable amounts of Cpn60 complexes were used for MS analysis . This estimation was confirmed by the similar intensity of the Cpn60 subunit bands detected by SDS-PAGE ( Figure S5A ) . Interestingly , 24 peptides of an NDH subunit , NdhH , were found in the Cpn60β4-purified fraction . No other NDH subunits or NDH biogenesis factors were found in any sample other than the NdhJ detected in the Cpn60β1-purified extraction ( Table 1 ) . Although NdhH was also co-purified with Cpn60β1 , only two NdhH peptides were detected ( Table 1 and Figure S5B ) and the emPAI ratio of NdhH from Cpn60β4- and Cpn60β1-purified samples was 43 . 4 . Aside from the nonspecific proteins detected in WT as well as chaperonin subunits detected in Cpn60β4-IP sample , NdhH was the most abundant protein found in the Cpn60β4-purified sample ( Table S2 ) . These results indicate that the Cpn60 complex containing Cpn60β4 can specifically recognize unfolded NdhH . NdhH was also detected in the Cpn60β1-purified sample ( Table 1 ) , suggesting that Cpn60 complexes containing Cpn60β1 also can interact with NdhH . However , when Cpn60β1-HA was introduced into crr27-1 , the transformation did not rescue NDH activity ( Figures 1 and 2 ) , suggesting that Cpn60 complexes lacking Cpn60β4 cannot produce native NdhH even though they can bind to it . This idea is consistent with the fact that crr27 accumulates the band II subsupercomplex ( Figure 2 ) . Although the functional NDH complex is localized to the thylakoids , three assembly intermediate complexes including NdhH are present in the chloroplast stroma ( Figure S6 ) [41] . Nuclear-encoded factors CRR6 and CRR7 may be required for integration of these intermediates into thylakoids to form the functional NDH complex . In crr27-1 , the level of stroma-localized NdhH was significantly reduced ( Figure S6A ) . Furthermore , 2D CN/SDS-PAGE and immunoblot studies showed that the accumulation of the 500 kDa and 400 kDa intermediate complexes was impaired in crr27-1 ( Figure S6B ) , implying that only NdhH folded by the Cpn60 complex including Cpn60β4 can be efficiently incorporated into these two assembly intermediates and further into thylakoids . These results also suggest that the folding of NdhH via the Cpn60 complex containing Cpn60β4 occurs at the initial step of NDH subcomplex A biogenesis . Based on a transcriptome database of Arabidopsis , ATTED-II [42] , we found that the Cpn60β4 gene , but not other Cpn60 genes , is co-expressed with genes encoding NDH subunits and NDH biogenesis factors ( Table S3 ) . This pattern is consistent with our findings that Cpn60β4 is specifically required for the folding of NdhH . However , the question remains as to why the other Cpn60β proteins cannot complement the function of Cpn60β4 . The mycobacterial GroEL1 has a histidine-rich C-terminus that appears to be critical for its specific function in association with proteins required for bacterial biofilm formation [18] . Cpn60β4 also contains a C-terminal extension that is not conserved in other Cpn60β proteins ( Figures 4A and S7 ) . Although the C-terminus of Cpn60β4 is not conserved in plants , the region is rich in positively charged residues ( Figure 4A ) . To study whether the C-terminus is important for Cpn60β4 function , HA-tagged Cpn60β4 lacking the C-terminus or exchanged by the short C-terminal tail of Cpn60β1 was expressed in crr27-1 ( Figure 4B ) . 2D CN/SDS-PAGE immunoblot analysis showed that the mutant versions of Cpn60β4 can be incorporated into the Cpn60 complex ( Figure 4C ) , excluding the possibility that the C-terminus of Cpn60β4 is required for the stabilization or formation of the chaperonin complex . Although the levels of mutant Cpn60β4 were approximately twice those of the WT version of Cpn60β4 in the stroma , NdhH levels in thylakoids were reduced by approximately one half in the Cpn60β4-HA lines ( Figure 4B ) , resulting in the reduction of NDH activity ( Figure 4D ) . These results indicate that the folding efficiency of NdhH was reduced in the absence of the Cpn60β4-specific C-terminus . In the absence of the C-terminal tail , NdhH is still partially assembled ( Figure 5B ) . We also transformed crr27 with Cpn60β1 fused with the Cpn60β4 C-terminal tail , but NdhH folding activity was not complemented ( unpublished data ) . These results suggest that other features of Cpn60β4 are required for its specific function . Protein sequence alignment revealed that the ATP/ADP and Mg2+ binding sites are highly conserved between Cpn60β4 and the other three Cpn60β subunits ( Figure S7 ) , which is consistent with the fact that Cpn60β4 is an intrinsic subunit of the Cpn60 complex ( Figure 3 ) . However , the proposed substrate-binding residues are less conserved ( Figure S7 ) , which may explain why Cpn60β4 has a high affinity specifically for NdhH . Protein sequence alignment also showed that up to 31 amino acid residues are highly conserved among the putative Cpn60β4 orthologs , but their properties are different from the corresponding residues in other Cpn60β subunits ( Figures S7 and S8 ) . Three-dimensional ( 3-D ) structure analyses mapped these residues to the apical , intermediate , and equatorial domains of the Cpn60β4 subunit ( Figure S8 ) . Of the 31 conserved residues in Cpn60β4 , several charged amino acids are located inside the cavity ( Figure S8 ) . The negatively charged GroEL cavity wall is required for rapid folding of some substrates [43] . In E . coli , each GroEL subunit has 27 negatively and 21 positively charged amino acids exposed to the central cavity in the cis-conformation , resulting in a net charge of −6 [44] . By analogy with E . coli GroEL , Cpn60α1 and Cpn60β1 have net charges of −4 and −6 , respectively . However , Cpn60β4 in Arabidopsis has a more positive charge of 0 , and this trend is found for Cpn60β4 in other plants ( charges ranging from −2 to +2 ) ( Figure 4E ) . To investigate the significance of the positively charged cavity wall in the folding of NdhH , the multiple charged residues , which are highly conserved in Cpn60β4 but not in Cpn60β1–β3 subunits , were converted to the corresponding amino acids of AtCpn60β1 and the mutant genes were introduced into crr27-1 plants . The sites correspond to 5 out of 31 amino acid residues indicated in Figure S8 . Although the mutant Cpn60β4 has a net charge of −4 ( Figure 4E ) , NdhH level and NDH activity were fully rescued in the transformed plants ( Figure 4D and 4F ) . We also transformed a version of Cpn60β4 with the amino acid alterations on the wall of the central cavity and the deletion of the C-terminus into crr27-1 . In these lines , NdhH levels in thylakoids were reduced to ∼50% of the Cpn60β4-HA lines ( Figure 4G ) , similar to the results in crr27-1 transformed by Cpn60β4 lacking its C-terminus ( Figure 4B ) . These results suggest that the positive charge of the cavity wall is not crucial for the folding of NdhH . The residues specifically conserved in the putative Cpn60β4 orthologs are dispersed throughout the molecule except for five positively charged sites facing the cavity wall ( Figure S8 ) , and it is not feasible to determine the sites responsible for the specific function by the site-directed mutagenesis . Proper folding of NdhH may require both the drastic alteration in the sequence as well as the C-terminal extension . Both chloroplast Cpn60 and NDH complexes are thought to have originated from their cyanobacterial ancestors , GroEL2 [45] and NDH-1 [32] . NdhH is highly conserved in Arabidopsis , Physcomitrella patens ( moss ) and cyanobacteria ( Figure S9 ) . NdhH is a 45 . 5 kDa protein with α+β domains ( Figure S9 ) [46] , and theoretically it can be fully encapsulated within the chaperonin cage . In contrast to NdhH , the structure of Cpn60 is not conserved among organisms . To clarify the evolution and ancestry of Cpn60β subunits in plants , we compared the amino acid sequences of members in several fully sequenced genomes ( Figure 5A ) . In addition to the distinct clades of Cpn60α and Cpn60β , Cpn60β proteins were further divided into two clades: putative AtCpn60β4 orthologs and other Cpn60β genes ( major Cpn60β ) . The orthologs of AtCpn60β1–β3 were found in the closely related Arabidopsis lyrata , but two major Cpn60β proteins of poplar ( Populus trichocarpa ) were related only to AtCpn60β3 . In contrast , the major Cpn60β subunits of monocots form a different subclade , and maize ( Zea mays ) and rice ( Oryza sativa ) each contain two major Cpn60β subunits ( Figure 5A ) . These facts suggest that gene duplication of major Cpn60β subunit genes took place independently both in monocots and eudicots . In contrast , a single copy of the putative Cpn60β4 ortholog was detected in angiosperms ( Figure 5A ) . A total of three Cpn60β subunits were found in Physcomitrella patens and they are related to major Cpn60β subunits in angiosperms ( Figure 5A ) . Notably , no ortholog of Cpn60β4 was found in P . patens . The phylogenetic tree indicates that the origin of Cpn60β4 can be traced to the origin of land plants and that Cpn60β4 was lost in the descendent lineage of bryophytes . Due to the low bootstrap support of the evolutionary relationships between angiosperm and bryophyte major Cpn60β subunits ( 72/0 . 90 as shown in Figure 5 ) , it is also likely that the Cpn60β4 orthologs were produced by a gene duplication event that took place only in a common ancestor of angiosperms and underwent a rapid rate of evolution to obtain the novel function . In any case , P . patens should use a different mechanism to assist the folding of NdhH , as it also contains the chloroplast NDH complex . To study whether the Cpn60β subunits in bryophytes can facilitate the folding of NdhH , we introduced the Cpn60β gene isolated from liverwort ( Marchantia polymorpha ) into crr27-1 ( Figure 5B ) . We identified only one gene copy encoding Cpn60β in M . polymorpha , possibly due to incomplete genome information . Immunoblots detected a trace amount of NdhH in the thylakoids of three transgenic lines , although the levels of MpCpn60β were comparable to those of AtCpn60β4 in crr27-1 transformed by AtCpn60β4-HA ( Figure 5B ) , suggesting that MpCpn60β partially rescues the phenotype of crr27-1 and that Cpn60β in bryophytes retains its ability in assisting the folding of NdhH . As MpCpn60β forms the heterologous Cpn60 complex with AtCpn60α , we could not compare the efficiency of the NdhH folding with the complex including AtCpn60β .
Although multiple chaperonin genes are present in a significant proportion of bacteria and eukaryotes , the function and biological significance of this kind of divergent evolution have yet to be revealed [13] . Generally , it is thought that the major subunits fulfill the housekeeping chaperonin function . The minor chaperonin subunits may increase the general chaperoning ability by elevating the chaperonin abundance in response to different environmental conditions . In this study , we demonstrated that the highly divergent chloroplast chaperonin subunit Cpn60β4 is specifically and strictly required for the folding of the NDH subunit NdhH . Although Cpn60β4 is highly divergent from the major Cpn60β subunits ( Figure 5 ) , it is an intrinsic chaperonin subunit and forms a heterooligomeric Cpn60 complex with Cpn60α1 and other Cpn60β subunits ( Figure 3 and Table 1 ) , suggesting the involvement of this specific Cpn60 complex in assisting the folding of some proteins . In line with this idea and the crr27 phenotype , the NDH subunit NdhH was copurified with the heterooligomeric Cpn60 complex including Cpn60β4 ( Table 1 ) . Although the Cpn60 complex lacking Cpn60β4 also interacts with NdhH with less affinity ( Tables 1 and S2 ) , it cannot produce native NdhH for further NDH complex assembly ( Figures 1 and 2 ) , implying that Cpn60β4 is required for both high-affinity binding and folding of NdhH . In contrast with the observation in R . meliloti , B . japonicum , and R . leguminosarum [15]–[17] , in which the function of the unusual chaperonins can be partially replaced by other chaperonins , Cpn60β4 is absolutely required for the folding of NdhH . These data support the proposal that multiple chaperonin subunits have evolved to assist the folding of specific proteins , although the functional specialization is not absolute in some cases . What is the structural basis for the functional specialization of Cpn60β4 ? We clarified that the unusual C-terminus of Cpn60β4 is required for the full activity of the Cpn60 complex containing Cpn60β4 for folding NdhH . The aforementioned GroEL1 protein in M . smegmatis also has an unusual histidine-rich C-terminus , which was shown to be essential for the specific function of GroEL1 [18] . These observations suggest that modification of the C-terminus is necessary to facilitate the folding of specific targets . This idea is consistent with the fact that the many bacterial genomes encode an additional chaperonin with an unusual C-terminus [13] . The C-terminus of Cpn60β4 is not required for the formation of the specific chaperonin complex containing Cpn60β4 ( Figure 4C ) . Thus , it should have some special functions in other steps during the folding of NdhH . The chaperonin complex containing Cpn60β4 and its substrate NdhH can be purified via the HA tag fused with the C-terminus of Cpn60β4 ( Table 1 and Figure 3C ) . As the cis ring is capped by a co-chaperone , it is likely that the C-terminal tail of Cpn60β4 extends from the trans ring so that the chaperonin complex containing Cpn60β4 can be trapped by microbeads coupled with HA antibody . Alternatively , the chaperonin complex was purified by the C-terminal tail extruding from the cis ring , which was not capped but already associated with the substrate NdhH . The protruded C-terminal tail of Cpn60β4 might promote the high affinity binding with NdhH , ensuring that the nonnative NdhH can be efficiently captured by the Cpn60 complex containing Cpn60β4 . In addition , enclosure of the nonnative NdhH protein inside the cavity by a co-chaperonin will lead to the encapsulation of the C-terminus . Consequently , the C-terminus might also contribute to the specific function of Cpn60β4 in assisting the folding of NdhH in the cavity . It has been reported that changing the length of the C-terminus of GroEL can affect the folding speed of some substrates [43] , [47] . Farr et al . provided further evidence that the elongated C-terminus perturbed the ATPase activity , and the disturbance of the rate of ATP hydrolysis resulted in the modification of the folding rate of some substrates [48] . The physical properties of the C-terminus [49] such as the length [43] , [47] , hydrophilicity [50] , and hydrophobicity [43] have been proposed to be critical for the substrate folding in the cavity . We also found that the C-terminus of the Cpn60β4 is rich in positively charged residues ( Figure 4A ) . Thus , it is likely that these unusual C-termini provide specific environments in the chaperonin cavity and/or modify the chaperonin ATPase rate of the chaperonin complex . The deletion of up to 27 residues of the C-terminal tail of GroEL does not affect the growth of E . coli [51] , [52] , suggesting that the C-terminal motif does not play an essential role in assisting the folding of substrates . It is also true that Cpn60β4 lacking its unusual C-terminal tail still can partially assist the folding of NdhH ( Figure 4B ) . Our results showed that the charged residues exposed on the cavity wall do not play a critical role for the specific function ( Figure 4F ) . In addition , we discovered many residues that are potentially important for the specific function of Cpn60β4 ( Figures S7 and S8 ) . Among them , several residues are mapped to the intermediate domain or near the ATP/ADP binding site ( Figure S8 ) . The E . coli GroEL with specific mutations in this region can improve the folding activity for green fluorescent protein ( GFP ) , most likely through the adjustment of the ATPase activity [53] . Recently , the apical domain of GroEL1 from Mycobacterium tuberculosis was shown to be sufficient for binding the specific substrate KasA [54] . Notably , some residues are specifically conserved in the apical domain of the putative Cpn60β4 orthologs ( Figure S8 ) . In addition , other conserved residues in AtCpn60β4 orthologs may be required for the formation of the Cpn60 complex including Cpn60β4 with certain stoichiometry or to provide certain physical features of the cavity wall . With the exception of the C-terminal extension , we could not specify the residues that are required for Cpn60β4 function . It is likely that some residues that are specifically conserved in the putative Cpn60β4 orthologs cumulatively contribute to their specific function , and this drastic evolution may have been necessary to assist the folding of NdhH . However , it is puzzling that MpCpn60β , which is related to the major AtCpn60β subunits , partly complemented the function of Cpn60β4 and that bryophytes do not contain the Cpn60β4 orthlogs ( Figure 5 ) . Specific mutations of GroEL can improve the folding activity of a specific protein [53] . However , mutated GroEL has a reduced ability to fold a variety of natural substrates , suggesting a conflict between the specific ability of GroEL to fold particular substrates and its general ability in assisting the folding of a wide range of substrates [53] . By combining all of the features acquired during the evolution of plants , Cpn60β4 allows the Cpn60 complex to assist the folding of the specific substrate , NdhH . However , other Cpn60 subunits , especially the major Cpn60β proteins , may have become optimized to support the efficient folding of other obligate substrates . Through this kind of divergent evolution , the chaperonin system can resolve the apparent conflict between specialization and generalization of its function .
Arabidopsis thaliana ( ecotypes Col-0 and Nössen ) plants were grown in a growth chamber ( 50 µmol photons m−2 s−1 , 16 h photoperiod , 23°C ) for 3 to 4 wk . cpn60α1 was mutagenized by ethyl methanesulfonate [55] . crr27-3 ( ecotypes Nössen ) was isolated from a collection of Ds transposon insertion lines [36] . crr27-1 ( SALK_136518 , Col-0 ) and crr27-2 ( SALK_064887 , Col-0 ) were obtained from the ABRC Stock Center . For complementation experiments , vectors were transferred into Agrobacterium tumefaciens C58C by electroporation , and the bacteria were used to transform crr27-1 by floral dipping . Chlorophyll fluorescence was measured using a MINI-PAM ( pulse amplitude modulation ) portable chlorophyll fluorometer ( Walz , Effeltrich , Germany ) . The transient increase in chlorophyll fluorescence after turning off AL was monitored as previously described [33] . Leaves were exposed to AL ( 50 µmol photons m−2 s−1 ) for 5 min . AL was turned off and the subsequent transient rise in fluorescence ascribed to NDH activity was monitored . Fluorescence levels were standardized to the maximum fluorescence levels of closed PSII ( Fm ) by applying saturating-light pulses ( SP ) . To investigate the light intensity dependence of two chlorophyll fluorescence parameters , ETR and NPQ , measuring light ( 650 nm , 0 . 1 µmol photons m−2 s−1 ) was used to excite the minimum fluorescence at open PSII centers in the dark-adapted state ( Fo ) . A saturating pulse of white light ( 800 ms , 8 , 000 µmol photons m−2 s−1 ) was applied to determine the maximum fluorescence at closed PSII centers in the dark ( Fm ) or during light illumination ( Fm’ ) . The steady-state fluorescence level ( Fs ) was recorded during AL illumination ( 15–1 , 000 µmol photons m−2 s−1 ) . These photosynthetic parameters were recoded 2 min after the change of AL intensity . ΦPSII was calculated as ( Fm′ – Fs ) /Fm′ . ETR and NPQ were calculated as ΦPSII×photon flux density ( µmol photons m−2 s−1 ) and ( Fm – Fm′ ) /Fm′ , respectively . Freshly isolated chloroplasts were osmotically ruptured in buffer containing 20 mM HEPES–KOH ( pH 7 . 6 ) , 5 mM MgCl2 , and 2 . 5 mM EDTA . Thylakoid membranes were separated from the stromal fraction by centrifugation ( 17 , 000 g for 10 min at 4°C ) . CN-PAGE , BN-PAGE , and 2-D CN/SDS-PAGE were performed according to previous reports [27] , [38] . For immunoblot analysis , thylakoid and stromal proteins were loaded by equal chlorophyll and protein content , respectively . The stromal protein contents were determined with a Bio-Rad Protein Assay Kit ( cat . No . 500-0006 ) . Immunoblot signals were detected with an ECL plus Western Blotting Detection Kit ( GE Healthcare ) and visualized with a Luminescent image analyzer ( LAS ) -3000 ( Fuji Film ) . Immunoblots were quantified by Imagemaster software ( Amersham Pharmacia Biotech ) . Total RNA was isolated from Arabidopsis leaves with an RNeasy Plant Mini Kit ( Qiagen ) and treated with DNase I ( Invitrogen ) . Total RNA ( 5 µg ) was reverse transcribed using the SuperScript III First-Strand Synthesis System ( Invitrogen ) in a total volume of 20 µl . The cDNA was used in 35 cycles of PCR with the following primers: 5′-TGGCTCTGTCACCAAGAAGCTTCAG-3′ and 5′-GCTTTCTGGGTGAATCCGTTGGTAA-3′ . RT-PCR products were separated on agarose gels and detected by ethidium bromide staining . Cpn60-substrate complexes were isolated from crr27-1 plants expressing HA-tagged Cpn60β subunits with the µMACS HA isolation kit under the conditions previously established [11] , with a minor modification . Freshly isolated chloroplasts were osmotically ruptured in lysis buffer ( 50 mM Tris-HCl pH 8 . 0 , 0 . 01% Tween 20 , 10 mM MgCl2 , 20 mM glucose , 20 U/ml hexokinase ) plus protease inhibitors ( Complete mini , Roche ) . Within 10 s after lysis , ADP was added to a final concentration of 10 mM . Thylakoid membranes were pelleted by centrifugation at 20 , 000 g for 10 min at 4°C and the supernatants were transferred to new tubes . NaCl was added to the supernatants to a final concentration of 150 mM and then mixed with 50 µl anti-HA MicroBeads ( Miltenyi Biotec . ) . After incubation for 2 h at 4°C , the beads were transferred to columns placed in a magnetic field . Columns were rinsed four times with 200 µl washing buffer I ( 50 mM Tris-HCl pH 8 . 0 , 1% Triton , 0 . 5% Sodium deoxycholate , 150 mM NaCl , 5 mM ADP ) and twice with 200 µl washing buffer II ( 50 mM Tris-HCl pH 8 . 0 , 1% Triton , 150 mM NaCl , 5 mM ADP ) . After final washing with 200 µl washing buffer III ( 25 mM Tris-HCl pH 7 . 5 , 5 mM ADP ) , the chaperonin-substrate complex was eluted with elution buffer ( 50 mM Tris-HCl pH 6 . 8 , 50 mM DTT , 1% SDS , 1 mM EDTA , 0 . 005% bromophenol blue , 10% glycerol ) . Total protein was separated on 12 . 5% SDS-PAGE gels ( Perfect NT Gel , DRC ) and stained with CBB . SDS-PAGE lanes were cut into four slices and analyzed by LC-MS/MS analyses . Peptide Preparation and LC-MS/MS analyses were performed as previously described [39] . The excised bands were treated twice with 25 mM ammonium bicarbonate in 30% ( v/v ) acetonitrile for 10 min and 100% ( v/v ) acetonitrile for 15 min , and then dried in a vacuum concentrator . The dried gel pieces were treated with 0 . 01 mg/ml trypsin ( sequence grade; Promega ) /50 mM ammonium bicarbonate and incubated at 37°C for 16 h . The digested peptides were recovered twice with 20 µl 5% ( v/v ) formic acid/50% ( v/v ) acetonitrile . The extracted peptides were combined and then dried in a vacuum concentrator . LC-MS/MS analyses were performed on an LTQ-Orbitrap XL-HTC-PAL system . MS/MS spectra were compared by the MASCOT server ( v . 2 . 2 ) against TAIR8 ( The Arabidopsis Information Resource ) with the following search parameters: set-off threshold at 0 . 05 in the ion-score cut-off; peptide tolerance , 10 ppm; MS/MS tolerance , ±0 . 8 Da; peptide charge , 2+ or 3+; trypsin as enzyme allowing up to one missed cleavage; carboxymethylation on cysteines as a fixed modification; and oxidation on methionine as a variable modification . Chaperonin protein sequences were first aligned using the CLUSTALX 1 . 83 program [56] . The protein alignment was further refined manually and 534 conserved sites were used for phylogenetic analysis . Phylogenetic trees were constructed by using both the maximum likelihood ( ML ) and Bayesian methods to ensure the robustness of our analysis . ML trees were constructed by using PHYML version 2 . 4 [57] with WAG model selected via MODELTEST 3 . 06 [58] , and support for each branch was assessed using bootstrap analyses with 100 bootstrap replicates . Bayesian trees were constructed using MrBayes software [59] with the WAG model . Four chains of Markov chain Monte Carlo were run , sampling one tree every 1 , 000 generations for 1 , 000 , 000 generations , starting with a random tree . The first 50 , 000 generations were excluded as burn-in to ensure that the chains reached stationary . The posterior probability was used to estimate nodal robustness . The structure model of the Cpn60β4 protein was obtained by homology modeling using the SWISS-MODEL server ( http://swissmodel . expasy . org/ ) [60] , and the crystal structure of E . coli GroEL ( PDB 1AON , Chain A ) was used as a modeling template . The 3-D predicted structure was visualized using the PyMol software . Sequence data from this article can be found in the Arabidopsis Genome Initiative or GenBank/EMBL databases under the following accession numbers: At ( Arabidopsis thaliana ) Cpn60α1 ( At2g28000 ) , AtCpn60α2 ( At5g18820 ) , AtCpn60β1 ( At1g55490 ) , AtCpn60β2 ( At3g13470 ) , AtCpn60β3 ( At5g56500 ) , AtCpn60β4 ( At1g26230 ) , AtNdhH ( BAA84443 ) , Al ( Arabidopsis lyrata ) Cpn60α1 ( 481708 ) , AlCpn60α2 ( 488768 ) , AlCpn60β1 ( 474606 ) , AlCpn60β3 ( 495739 ) , AlCpn60β4 ( 890123 ) , Sy ( Synechocystis SP . PCC 6803 ) GroEL-1 ( NP_440731 ) , SyGroEL-2 ( NP_442170 ) , SyNdhH ( CAA43057 ) , GroEL ( Escherichia coli ) ( NP_418567 ) , and Pp ( Physcomitrella patens ) NdhH ( BAC85094 ) . | Chaperonins assist the folding of some nascent and denatured proteins to their native , functional forms . Each chaperonin consists of a pair of protein complexes resembling two stacked toroids; folding occurs inside the toroid cavity . Chaperonins are ubiquitous in both bacteria and more complex nucleated cells , as well as in the intracellular organelles that have evolved from bacteria by endosymbiosis: mitochondria and , in plants , chloroplasts . They are indispensable for cellular function . Many different chaperonin subunits have evolved in various species of bacteria as well as in most mitochondria and chloroplasts . The physiological and functional relevance of these multiple chaperonin subunits is poorly understood , however . In this study , we have characterized the minor chaperonin subunit Cpn60β4 from Arabidopsis chloroplasts , which differs in structure from other chloroplast chaperonins . When the Cpn60β4 gene is defective , the plants fail to accumulate one protein complex in particular: the chloroplast NADH dehydrogenase-like complex ( NDH ) . We discovered that Cpn60β4 forms a complex with other Cpn60 α and β subunits and that this complex is essential for the folding of the NDH subunit NdhH . Cpn60β4 has a unique protein “tail” that is required for the efficient folding of NdhH . Our findings suggest that Cpn60β4 has evolved with distinctive structural features that facilitate the folding of one specific substrate and that this strategy is used by plants to satisfy their conflicting requirements for chaperonins with both specialized and general functions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology"
] | 2011 | A Chaperonin Subunit with Unique Structures Is Essential for Folding
of a Specific Substrate |
Epstein-Barr virus ( EBV ) is a human herpesvirus associated with B-cell and epithelial cell malignancies . EBV lytically infects normal differentiated oral epithelial cells , where it causes a tongue lesion known as oral hairy leukoplakia ( OHL ) in immunosuppressed patients . However , the cellular mechanism ( s ) that enable EBV to establish exclusively lytic infection in normal differentiated oral epithelial cells are not currently understood . Here we show that a cellular transcription factor known to promote epithelial cell differentiation , KLF4 , induces differentiation-dependent lytic EBV infection by binding to and activating the two EBV immediate-early gene ( BZLF1 and BRLF1 ) promoters . We demonstrate that latently EBV-infected , telomerase-immortalized normal oral keratinocyte ( NOKs ) cells undergo lytic viral reactivation confined to the more differentiated cell layers in organotypic raft culture . Furthermore , we show that endogenous KLF4 expression is required for efficient lytic viral reactivation in response to phorbol ester and sodium butyrate treatment in several different EBV-infected epithelial cell lines , and that the combination of KLF4 and another differentiation-dependent cellular transcription factor , BLIMP1 , is highly synergistic for inducing lytic EBV infection . We confirm that both KLF4 and BLIMP1 are expressed in differentiated , but not undifferentiated , epithelial cells in normal tongue tissue , and show that KLF4 and BLIMP1 are both expressed in a patient-derived OHL lesion . In contrast , KLF4 protein is not detectably expressed in B cells , where EBV normally enters latent infection , although KLF4 over-expression is sufficient to induce lytic EBV reactivation in Burkitt lymphoma cells . Thus , KLF4 , together with BLIMP1 , plays a critical role in mediating lytic EBV reactivation in epithelial cells .
Epstein-Barr Virus ( EBV ) is a human gamma-herpesvirus that causes the clinical syndrome infectious mononucleosis [1] , and contributes to several types of human malignancy . EBV , which primarily infects B cells and oropharyngeal epithelial cells , is associated with the development of both B cell and epithelial cell tumors in humans , including Burkitt lymphoma , Hodgkin Disease , nasopharyngeal carcinoma ( NPC ) and gastric carcinoma [2 , 3] . Like all herpesviruses , EBV undergoes both latent and lytic forms of infection in normal cells , and both types of infection are essential for the long-term success of the virus . However , EBV-infected tumors primarily contain cells with latent viral infection , since this type of infection allows expression of the major viral transforming proteins but does not cause virally-mediated cell killing [2 , 4] . In contrast to B cells , relatively little is known about the regulation of EBV infection in normal epithelial cells . The memory B cell compartment serves as the major reservoir for life-long latent EBV infection in humans [5] . EBV-infected B cells can be reactivated to the lytic form of infection , which is required for production of infectious viral particles , following strong B cell receptor ( BCR ) stimulation and/or plasma cell differentiation [4 , 6–8] . Normal ( untransformed ) oropharyngeal epithelial cells also support the lytic form of EBV infection [9–11] , but there is currently little evidence that these cells can undergo persistent latent infection . Long-term latent EBV persistence following infection of telomerase-immortalized nasopharygeal epithelial cells has been reported to require over-expression of the oncogene , cyclin D1 , as well as repression of the p16 tumor suppressor protein [12] . Thus , the ability of EBV to establish long-term latency in epithelial cells in vitro may require that the cells already be abnormal . Much of our current knowledge regarding EBV infection of untransformed epithelium in humans is derived from papers examining EBV gene/protein expression in oral hairy leukoplakia ( OHL ) lesions of immunosuppressed patients [10 , 13 , 14] . These studies have suggested that EBV infection in OHL lesions is limited to the more differentiated layers of the tongue epithelium , and is completely lytic . Consistent with these findings , a recent in vitro study examining EBV infection in normal stratified oral epithelial cells grown in organotypic raft culture found completely lytic EBV infection in the differentiated cell layers , but no evidence of latent or lytic infection in undifferentiated basal cells [15] . Whether epithelial cell differentiation promotes lytic EBV reactivation and/or whether EBV can even infect normal undifferentiated epithelial cells in humans is not totally clear . It remains possible that EBV ( similar to human papilloma virus [HPV] ) [16 , 17] maintains persistent latent infection in a small number of undifferentiated normal basal epithelial cells and converts to productive viral infection during epithelial cell differentiation . Studies done by several groups in vitro have suggested a correlation between epithelial cell differentiation of EBV-infected carcinoma lines and EBV lytic reactivation , but the methods used to induce differentiation in these studies have multiple different effects and thus may have activated lytic EBV gene expression through mechanism ( s ) distinct from differentiation per se [18–21] . Ideally , the effect of differentiation on EBV gene expression should be examined using air-interface organotypic raft cultures , as has previously been done in HPV-infected cell lines [22] . However , such studies have been limited by the availability of a long-term EBV infected epithelial cell line that retains the ability to undergo differentiation using this technique . The lytic viral cascade in latently EBV-infected cells is initiated by expression of the two viral immediate-early ( IE ) genes , BZLF1 ( Z ) and BRLF1 ( R ) , that encode the transcription factors Z and R respectively [23–30] . The Z and R proteins initially activate each other’s promoters and then cooperate to induce expression of the entire cadre of lytic genes required for lytic viral DNA replication and virion production [28 , 31–35] . While overexpression of Z alone is sufficient to induce EBV lytic reactivation in essentially all previously examined latently EBV-infected cell lines , we recently identified an EBV-infected cell line ( EBV-infected telomerase-immortalized normal oral keratinocytes , referred to here as NOKs-Akata cells ) in which only R , and not Z , overexpression can induce lytic viral reactivation [36] . These results suggest that EBV lytic reactivation in normal epithelial cells , in contrast to EBV-infected carcinoma cells and EBV-infected B cells , may be more dependent upon cellular transcription factors that activate the viral IE promoter ( Rp ) driving R transcription rather than the IE promoter ( Zp ) driving Z transcription . We recently showed that BLIMP1 , a cellular transcription factor whose expression is greatly enhanced by both B cell and epithelial cell differentiation [37 , 38] , can activate Rp in reporter gene assays and reactivate low-level lytic gene EBV expression in a subset of EBV-infected cell lines [39] . However , the cellular factors that activate Rp ( and potentially Zp ) during epithelial cell differentiation remain largely unexplored . In this paper , we have used organotypic raft cultures of NOKs-Akata cells to demonstrate that epithelial cell differentiation of latently infected cells induces lytic EBV reactivation . Furthermore , we show that a differentiation-dependent cellular transcription factor , KLF4 [40 , 41] , binds to and activates both the Zp and Rp EBV promoters , and collaborates with BLIMP1 to synergistically induce high-level lytic EBV reactivation in latently infected epithelial cells . We also find that KLF4 and BLIMP1 are only expressed within differentiated cells in normal tongue epithelium , and that epithelial cells in a patient-derived OHL lesion express both KLF4 and BLIMP1 . In contrast , we do not detect KLF4 protein in either EBV-infected or uninfected B cells . We propose that the increased expression of KLF4 and BLIMP1 that occurs during normal epithelial cell differentiation promotes lytic EBV infection . Conversely , the lack of both KLF4 and BLIMP1 expression in normal undifferentiated epithelial cells , undifferentiated NPC tumors , and B cells promotes viral latency .
The differentiation program of stratified epithelia can be recapitulated in vitro using the organotypic culture technique in which epithelial cells are plated on a dermal equivalent composed of human fibroblasts embedded in collagen , allowed to grow to confluence , then raised to the air/liquid interface ( thereby commonly referred to as 'raft' cultures ) and cultured for approximately two weeks’ time during which the epithelial cells proliferate , stratify , and daughter cells that lose contact with the dermal equivalent undergo terminal differentiation . Rafting is considered the gold standard method for recapitulating the normal differentiation program of stratified epithelial in vitro . To determine if EBV-infected NOKs cells ( NOKs-Akata ) retain the ability to be differentiated , uninfected and EBV-infected NOKs were grown in raft cultures . As shown in Fig 1A , the EBV-infected NOKs cells undergo stratification as does the uninfected parental cell population , with classical morphological signs of differentiation including the presence of terminally differentiated squames at the top surface , clearly visible in the EBV-infected NOKs . Closer examination of the EBV infected cells , however , showed evidence of a much less well organized basal layer of epithelial cells ( the layer of cells directly in contact with the underlying dermal equivalent ) , with signs of epithelial migration/invasion into the underlying dermal equivalent , suggestive of a partial disruption in the normal program of epithelial cell differentiation . Consistent with these morphological changes , the pattern of expression of cytokeratin 10 ( K10 ) and involucrin , cellular proteins whose expression is induced in suprabasal cells , was less uniform in the EBV-infected raft culture ( Fig 1B and Fig 1C ) . These observations validate recent findings made using other approaches for inducing epithelial cell differentiation that less well recapitulate the in vivo process: suspension of epithelial cells in methylcellulose or growth of epithelial cells in high concentration of calcium chloride [42] . We next examined whether signs of lytic EBV reactivation arise within raft cultures of two independently isolated populations of EBV-infected NOKs ( Fig 2A and 2B ) . First we stained for EBV-encoded small nuclear non-coding RNAs ( EBERs ) , which are highly expressed in latently EBV-infected cells , to confirm that the raft cultures retained EBV . EBERs were detected by in situ hybridization throughout the raft cultures of EBV-infected NOKs , while uninfected cells had no detectable EBERs ( S1 Fig ) . Next we stained for two markers of lytic reactivation , the EBV immediate-early BZLF1 gene encoding the Z protein and the early lytic BMRF1 gene encoding the viral DNA polymerase processivity factor . Cells positive for these markers were exclusively detected within the suprabasal compartment of the raft culture . Uninfected NOKs cells did not stain positively ( S1 Fig ) . Immunofluorescence co-staining using antibodies directed against K10 and Z showed that while expression of Z can be associated with expression of differentiation marker K10 ( Fig 2C , left panel ) , some Z-positive cells do not express K10 ( Fig 2C , right panel ) . To determine if epithelial cell differentiation is associated with EBV genome amplification , fluorescence in situ hybridization ( FISH ) assays were performed using a probe that detects EBV genome DNA . While small green foci representing latent EBV genomes were present in every cell , cells with highly positive signals for the EBV genome ( due to lytically-infected EBV ) were only detected in the more differentiated cell layers ( S2 Fig ) . Together , these results indicate that expression of lytic EBV genes and lytic viral DNA replication arises preferentially within the differentiating , suprabasal compartment , indicating that lytic reactivation arises when epithelial cells are triggered to undergo terminal differentiation , even though that differentiation is partially perturbed by EBV . The relative rarity of lytically infected cells in rafted NOKs-Akata cells in comparison to EBV-infected rafted primary oral epithelial cells [15] may reflect the decreased differentiation in NOKs-Akata cells . We next determined whether other treatments that have been reported to induce epithelial cell differentiation in vitro , such as phorbol ester ( TPA ) , and calcium chloride/fetal bovine serum ( FBS ) , can differentiate NOKs-Akata cells , and if they have any effect on lytic EBV reactivation . Treatment with either TPA or calcium chloride ( given simultaneously with 10% FBS in RPMI media ) promoted differentiation of NOKs-Akata cells , as indicated by increased expression of the differentiation-dependent cellular protein , involucrin ( Fig 3A ) . Both treatments also led to lytic EBV reactivation , as indicated by increased expression of the lytic viral proteins , Z and BMRF1 ( Fig 3A ) . To determine if TPA-mediated lytic reactivation of NOKs-Akata cells is at least partially differentiation-dependent , cells were treated with TPA in the presence or absence of a Rho-associated , coiled-coil-containing protein kinase ( ROCK ) inhibitor ( Y27632 ) that has been reported to inhibit epithelial cell differentiation [43] . The ROCK inhibitor decreased both the TPA-induced differentiation marker ( involucrin ) and TPA-mediated lytic EBV reactivation ( Fig 3B ) , suggesting that the TPA effect on EBV gene expression in NOKs-Akata cells is at least partially differentiation-dependent . Since lytic EBV reactivation in NOKs-Akata cells primarily depends upon activation of the viral Rp IE promoter [36] , we next examined whether this promoter can be activated by the Kruppel-like factor 4 ( KLF4 ) cellular transcription factor . KLF4 is selectively expressed in the upper spinous and granular layers of skin epithelial cells and is required for terminal epithelial cell differentiation in skin [40 , 41 , 44] . Although our lab previously showed that the cellular Sp1 transcription factor can bind to and activate the Rp [45] , the effect of KLF4 ( a member of the zinc-finger family of proteins that binds to Sp1-like sites ) on Rp activity has not been reported . Of note , KLF4 was recently reported to activate the immediate-early EBV Z promoter in reporter gene assays via a motif that also binds Sp1 [46] , although its ability to induce lytic reactivation in EBV-infected B cells or epithelial cells has not been examined . As shown in Fig 4A , co-transfection of Rp- and Zp-driven luciferase vectors with a KLF4 expression vector greatly increased the activity of both EBV IE promoters in EBV-negative NOKs cells . Given the ability of KLF4 to activate both of the EBV IE promoters , we next asked if KLF4 over-expression is sufficient to induce lytic viral reactivation in latently infected NOKs-Akata cells ( previously shown to be reactivated by R but not Z expression ) , or HONE-Akata cells ( an epithelial carcinoma cell line , super-infected with the Akata strain of EBV , that can be reactivated by either Z or R expression ) [36] . As shown in Fig 4B and 4C , overexpression of KLF4 was sufficient to activate expression of the EBV immediate-early proteins , Z and R , as well as the early lytic protein , BMRF1 , in both epithelial cell lines . These results indicate that KLF4 is sufficient to induce lytic EBV reactivation in an epithelial cell environment . Although expression of KLF4 is differentiation-dependent in normal epithelia , KLF4 is also commonly overexpressed in squamous cell carcinomas where it can act as an oncogene [47] . To determine whether endogenous KLF4 expression plays an important role in mediating lytic EBV reactivation in EBV-infected epithelial cell lines , KLF4 was knocked down using siRNA in two different EBV-positive epithelial cell lines , followed by treatment with TPA or sodium butyrate to induce lytic viral reactivation . As shown in Fig 5 , CNE-2-Akata cells have detectable expression of KLF4 , and knockdown of endogenous KLF4 inhibited the ability of both TPA and sodium butyrate treatment to induce lytic viral reactivation in CNE-2-Akata cells ( Fig 5A and 5B ) . KLF4 knockdown also decreased the ability of TPA to induce lytic reactivation in NOKs-Akata cells ( Fig 5C ) . Thus , constitutive KLF4 expression is required for efficient TPA-and sodium butyrate-mediated lytic EBV reactivation in EBV-infected epithelial cell lines . To determine the mechanism by which KLF4 activates the BRLF1 promoter , we compared the ability of KLF4 to activate a series of 5’ Rp deletion mutants in reporter gene assays . KLF4 activation of Rp was significantly decreased when promoter sequences between -551 and -486 were deleted ( Fig 6A ) . Site-directed mutagenesis was then performed to mutate two different consensus KLF4 motifs ( located between -500 and -496 , and between -452 and -448 ) ( Fig 6B ) alone , or in combination , in the Rp -551 luciferase construct . Mutation of either site alone partially decreased KLF4 activation of the promoter , and mutation of both sites simultaneously resulted in a more dramatic decrease ( Fig 6C ) . This result suggests that both KLF4 consensus binding sites contribute to KLF4 activation of the BRLF1 promoter . To examine whether KLF4 associates directly with the EBV Rp and/or Zp promoters in vivo , ChIP assays were performed in latently infected HONE-Akata cells transfected with either control vector or a KLF4 expression vector . These studies confirmed that KLF4 binds to both the Rp and Zp promoters of the endogenous viral genome in EBV-infected HONE cells ( Fig 6D ) , but not to the EBV EBNA promoter , Cp . ChIP for RNA polymerase II ( pol2 ) also revealed that KLF4 increased Rp and Zp occupancy with active pol2 ( Fig 6D ) , consistent with its ability to activate transcription from both viral IE promoters . Similar results were obtained in NOKs-Akata cells ( S3 Fig ) . The cellular transcription factor , BLIMP1 , activates both Rp [39] and Zp [48] in reporter gene assays , and , like KLF4 , is induced during epithelial cell differentiation . To determine if the combination of BLIMP1 and KLF4 activates the Rp and/or Zp promoters more strongly than either BLIMP1 or KLF4 alone , we compared the effects of each transcription factor alone , or in combination , on Zp and Rp activity in EBV-negative NOKs cells . Since the ability of KLF4 to activate some promoters is affected by promoter methylation [49 , 50] , we also examined whether DNA methylation of the Rp or Zp promoters alters the KLF4 and/or BLIMP1 effect . As shown in Fig 7A and 7B , the combination of KLF4 and BLIMP1 together produced highly synergistic activation of both the Rp and Zp promoters , regardless of promoter methylation status . Of note , the ability of KLF4 and BLIMP1 alone to activate the Rp -673 promoter construct was significantly reduced when the promoter was methylated , while activation mediated by the BLIMP1/KLF4 combination was less affected . Thus , the availability of both KLF4 and BLIMP1 may be particularly important for activating the Rp in situations where it is highly methylated ( as occurs in EBV-positive NPC [51] ) . Conversely , methylation of Rp and loss of either KLF4 and/or BLIMP1 expression might be one of the mechanisms by which EBV achieves latency in these tumor cells . To determine whether KLF4 also synergizes with BLIMP1 to induce lytic reactivation in latently infected EBV-positive epithelial cell lines , we transfected KLF4 and BLIMP1 expression vectors ( alone or in combination ) into HONE-Akata , NOKs-Akata or SNU . 719 cells ( an EBV-positive gastric carcinoma line ) . While BLIMP1 and KLF4 alone both induced detectable expression of the two IE proteins ( Z and R ) , as well as the early protein , BMRF1 , in each cell line , the combination of KLF4 and BLIMP1 together produced dramatically more lytic EBV protein expression than either KLF4 or BLIMP1 alone in each line ( Fig 8A–8C ) . KLF4 also synergized with BLIMP1 to induce the expression of late viral capsid protein , p18 , in EBV-infected HONE , NOKs and SNU . 719 cells ( S4 Fig ) . Likewise , the combination of KLF4 and BLIMP1 synergistically increased the release of infectious virion particles from CNE-2 Akata cells ( Fig 8D and 8E ) , and the amount of intracellular EBV DNA in HONE-Akata cells ( S4 Fig ) . Of note , neither KLF4 nor BLIMP1 induced expression of the other protein in epithelial cells . Since we observed constitutive KLF4 ( but not BLIMP1 ) expression in a number of EBV-infected epithelial cell lines ( Fig 5 ) , we also determined whether endogenous KLF4 expression is required for BLIMP1-mediated lytic EBV reactivation . The ability of transfected BLIMP1 to induce expression of EBV lytic proteins was compared in HONE-Akata cells where endogenous KLF4 expression was knocked out using CRISPR-Cas9 technology , versus HONE-Akata cells infected with a non-targeting control CRISPR-Cas9 vector . Disruption of the KLF4 gene significantly reduced the ability of BLIMP1 to induce lytic EBV gene expression ( Fig 8F ) . These results confirm that KLF4 and BLIMP1 strongly collaborate to induce lytic EBV reactivation , and suggest that differentiation-associated lytic reactivation in normal epithelial cells is likely to be at least partially mediated through KLF4 and BLIMP1 . We were unable to create a derivative of NOKs-Akata cell line in which KLF4 was stably knocked down or knocked out using either shRNA or CRISPR-Cas9 technology , respectively , suggesting that KLF4 serves as an essential survival factor for this cell line . Although previous studies have shown that expression of KLF4 is differentiation-dependent in normal skin [41 , 44] , to our knowledge the effect of differentiation on the expression of KLF4 in human oral mucosal epithelium has not been examined . To determine whether KLF4 and BLIMP1 expression is regulated by differentiation in normal human tongue tissue , we performed immunohistochemistry analysis using antibodies directed against these two cellular proteins . Expression of both KLF4 and BLIMP1 was highest in cells within the suprabasal compartment ( Fig 9A ) . Similar results were obtained in normal human tonsil epithelium ( S5 Fig ) . In NOKs-Akata cells grown in raft cultures , BLIMP1 expression was differentiation dependent . However , KLF4 was expressed in a portion of undifferentiated basal epithelial cells as well as in the more differentiated cells ( Fig 9B ) , consistent with our finding that KLF4 is required for the long-term survival of this cell line . To confirm that KLF4 and BLIMP1 are also both expressed in the highly lytic OHL lesions that can occur within differentiated tongue epithelium of immunosuppressed patients , we examined KLF4 and BLIMP1 expression in a patient-derived OHL lesion . We confirmed that cells with the morphology typical of OHL cells ( “ballooning” cells with a large amount of cytoplasm- see arrows ) not only expressed the lytic viral proteins , Z and BMRF1 , but also the cellular proteins , KLF4 and BLIMP1 ( Fig 10 ) . Collectively , these results suggest that the expression of KLF4 and BLIMP1 correlates with lytic EBV infection in the tongue . To examine further the effect of epithelial cell differentiation state on EBV gene expression in humans , and to explore the possibility that undifferentiated cells can support low-level latent EBV infection , epithelial cells were isolated using laser dissection from the undifferentiated ( basal ) , partially differentiated ( middle ) , or differentiated ( superficial ) layers of normal tonsil epithelium , or OHL lesions; only normal tonsils that had EBER+ B cells were examined ( S6 Fig ) . Total RNA was purified , DNase treated , reverse transcribed , and amplified with a series of primers ( Table 1 ) to detect both epithelial cell-specific , B cell-specific , and viral transcripts . As shown in Table 2 , the cellular ∆Np63 transcript ( expressed specifically in basal epithelial cells ) was detected in basal , but not superficial , OHL lesions , confirming our ability to separate these layers accurately . Furthermore , the B-cell-specific CD20 transcript was not detected ( using primers that can detect even one CD20-positive cell in 1000 cells ) , making it very unlikely that the isolated RNA was partially derived from infiltrating B cells . In the majority ( 3/4 ) of OHL tissue , lytic EBV transcripts ( including BZLF1 , BRLF1 , and BcLF1 ) could not be detected in the undifferentiated epithelial cells , but were easily detected in differentiated cells . Interestingly , however , EBER transcripts were detected in the basal epithelial cells in the majority ( 3/4 ) of OHL biopsies ( Table 2 ) . Furthermore , EBER transcripts were also detected in undifferentiated basal cells in normal tonsil tissue , although lytic EBV transcripts were not detected ( Table 3 ) . In addition , low level EBER staining was detected in undifferentiated epithelium of tonsils that had a high number of EBER-positive B cells ( S6 Fig ) . These results suggest that EBV , like HPV , may establish persistent latent infection in some basal epithelial cells , and confirm that lytic EBV reactivation is confined to more differentiated epithelial cell layers . Very little is known regarding KLF4 protein expression in B cells , in which EBV normally enters the latent form of infection . To further examine the potential role of KLF4 in regulating EBV gene expression in B cells , we performed immunoblot analysis to compare KLF4 levels in a variety of EBV-infected epithelial and B cell lines . Interestingly , we found that the cell line with the highest level of KLF4 expression ( AGS-Akata ) is an EBV-superinfected gastric carcinoma line that we previously reported supports unusually high level lytic EBV protein expression [52] ( Fig 11A ) . Consistent with a previous report showing that KLF4 expression is decreased in NPC tumor specimens [53 , 54] , we found that KLF4 expression in C666 . 1 cells ( the only authentic EBV+ NPC tumor cell line in this panel ) is relatively low compared to that in the EBV-superinfected epithelial cell lines ( NOKs and HONE ) . Furthermore , KLF4 was not detected in either EBV-positive Burkitt lines or EBV-transformed lymphoblastoid cells . Consistent with the lack of KLF4 protein expression in Burkitt cell lines , Human Protein Atlas studies also recently reported no detectable KLF4 protein expression in the lymphoid tissue of normal human tonsil or spleen [55] . To determine if KLF4 expression can be induced in Burkitt lymphoma cells treated with agents that activate the lytic form of EBV infection , Burkitt cells were treated with a variety of different lytic-inducing agents , including anti-IgG ( which induces B-cell receptor activation ) , 5-aza-2’-deoxycytidine ( a demethylating agent ) and TGF-β . These treatments did not restore KLF4 expression in Burkitt cells , although these agents induced lytic EBV reactivation as expected ( S7 Fig ) . Nevertheless , overexpression of KLF4 in both the Raji and Jijoye latently infected , EBV-positive , Burkitt lymphoma cell lines was sufficient to induce the expression of the EBV immediate-early proteins , Z and R , as well as the early lytic protein , BMRF1 ( Figs 11B and 11C and S8 ) . Thus , KLF4 can induce lytic EBV reactivation even in a B cell-environment and , hence , the lack of KLF4 expression in this cell type may promote viral latency .
EBV infection of oropharyngeal epithelial cells is associated with both malignant and non-malignant human diseases , but the viral and cellular factors that regulate EBV gene expression in epithelial cells are poorly understood . Here , we show in both in vitro and in vivo studies that the induction of differentiation of epithelial cells promotes the lytic form of EBV infection , while undifferentiated basal epithelial cells support latent EBV infection . In addition , we demonstrate that the differentiation-dependent cellular transcription factors , KLF4 and BLIMP1 , collaboratively link EBV lytic reactivation to epithelial cell differentiation . Furthermore , we find that the absence of KLF4 expression in B cells may help to promote viral latency in this cell type . Together , these studies suggest that KLF4 plays a key role in regulating EBV gene expression , particularly in epithelial cells . KLF4 , a zinc-finger protein that binds to Sp1-like sites , can act as either a positive or negative regulator of transcription when bound to promoters , and has highly divergent functions depending upon the cell type and context in which it is expressed [56–61] . KLF4 is perhaps best known for its ability to convert differentiated cells into IPS cells when delivered with three other cellular transcription factors , Oct4 , Sox2 and c-Myc [62] . In this context , KLF4 has been reported to act as a “pioneer” transcription factor that can bind to DNA sites within heterochromatin , and subsequently convert the chromatin to a more open form that can then be bound by other transcription factors such as c-myc [63] . However , KLF4 also inhibits gene expression in some instances , by binding to co-repressor proteins such as HDACs [64] . The role of KLF4 in tumor formation is also very complex , as KLF4 can function as either a tumor suppressor , or tumor promoter , depending upon the cell type and other factors . KLF4 is thought to function as a tumor suppressor in certain epithelial cell tumors , and its expression is decreased in both EBV-positive NPC and in gastric carcinoma [54 , 65–67] . Consistent with its role as a tumor suppressor , KLF4 inhibits cell cycle progression in some , but not all , cell types [59 , 68] . In the case of normal epithelium , KLF4 is expressed in differentiated , but not undifferentiated , cells , and the knockout of the KLF4 gene in mice results in abnormal epithelial cell differentiation [40] . Furthermore , KLF4 binds to the involucrin gene promoter , and mediates differentiation-dependent expression of this gene , in human epithelium [44] . The results here suggest that KLF4 likewise helps to promote differentiation-dependent expression of the two EBV IE genes in oral epithelium . However , KLF4 may also function as a tumor promoter in some types of epithelial cell carcinomas , including squamous cell carcinomas ( SCC ) [47] . Consistent with a survival role for KLF4 in certain epithelial cell tumors , KLF4 induces squamous epithelial cell dysplasia when expressed in basal keratinocytes in mice [69] , transforms rat kidney epithelial cells in vitro [70] , and is correlated with bad prognosis when over-expressed in human head and neck carcinomas [47 , 71] . The mechanism ( s ) by which KLF4 switches to a tumor activator in certain types of epithelial cell carcinoma are not totally understood , but may be related to the multiple different types of post-translational modifications that can occur on the KLF4 protein ( including phosphorylation , acetylation and sumoylation ) and alter its functions [72–77] . Similar to its potential role as an oncogene in certain SCCs , we found that KLF4 is required for long-term survival of NOKs cells and thus we were unable to knockdown KLF4 expression ( except transiently ) in this cell type . In B cells , KLF4 has been reported to halt cell cycle progression and act as a tumor suppressor [78–80] . Even though KLF4 transcripts can be detected in normal B cells , recent immunohistochemistry staining studies reported by the Human Protein Atlas consortium indicated that KLF4 protein is undetectable within the lymphoid regions of normal tonsils , spleen and lymph nodes [55] . We were likewise unable to detect KLF4 protein expression in any of the EBV-infected B cell types that we examined by immunoblot analysis ( Fig 11 ) , including Burkitt lymphoma cell lines with various forms of EBV latency and EBV-transformed lymphoblastoid cell lines , and did not detect KLF4 expression ( by IHC analysis ) in the lymphoid areas of normal human tonsil and spleen . The absence of KLF4 expression in EBV-infected B cells may provide a mechanism to promote viral latency . In any event , our finding that KLF4 expression is sufficient to induce lytic EBV gene expression in Burkitt lymphoma cell lines indicates that KLF4 can promote lytic EBV gene expression outside the context of epithelial cell differentiation . We also show here that KLF4 binds to the endogenous viral Rp IE promoter in EBV-infected HONE and NOKs cells , and that KLF4 binding sites in the promoter are required for efficient KLF4-mediated activation of the promoter in reporter gene assays . In addition , we show that KLF4 also binds to the Zp IE promoter in vivo . This result is consistent with the report of another group showing that KLF4 can activate Zp in a reporter gene assay through previously mapped Sp1 motifs [46] . Since Sp1 has been shown to bind to and activate the Zp and Rp promoters in vitro [45 , 46] , and KLF4 and Sp1 bind to the same motif , it has not been clear what role , if any , KLF4 plays in mediating viral reactivation in vivo . Our results here suggest that KLF4 is critical for promoting efficient lytic reactivation during normal epithelial cell differentiation , but is less likely to play an important role in B cells , since B cells do not appear to express KLF4 protein within normal lymphoid tissue . Nevertheless , it remains possible that KLF4 can be activated in EBV-infected B cells under certain circumstances and contribute to lytic reactivation in specific contexts . We also show here that the combination of the KLF4 and BLIMP1 transcription factors results in much more lytic EBV gene expression than either factor alone . This synergistic effect on EBV IE promoter activity may be most biologically relevant in differentiating epithelial cells such as normal tongue tissue , where we show that the two transcription factors are coordinately expressed . Since we found no evidence that BLIMP1 affects KLF4 expression , or vice versa , we postulate that the two transcription factors independently activate Zp and Rp transcription through different mechanisms , and that the combined effect of these two factors is synergistic . BLIMP1 has been previously shown to activate Zp in reporter gene assays [48] , and we recently reported that BLIMP1 can also activate Rp in both EBV-negative epithelial cells and B cells [39] . BLIMP1 activation of Rp is at least partially mediated through Rp sequences centered at -660 relative to the transcriptional start site [39] . Although BLIMP1 is complexed to Rp in vivo in ChIP assays , it does not bind to this site directly in vitro , based upon EMSA results [39] , and the exact mechanism ( s ) by which BLIMP1 activates either Rp or Zp has not been clearly defined . Since BLIMP1 binding to promoters usually results in inhibition of promoter activity [81–84] , it is possible that BLIMP1 down-regulates a cellular factor ( s ) which in turn inhibits Zp and Rp activity . As BLIMP1 expression is increased by differentiation of both epithelial and plasma cells , it likely plays a role in the lytic reactivation of EBV in both cell types . Our finding that endogenous KLF4 expression in epithelial cells is required for efficient BLIMP1-mediated lytic reactivation in that cell type raises the interesting prospect that another member of the Kruppel-like cellular transcription factor family serves a similar role in differentiation- dependent lytic reactivation in plasma cells . Finally , our studies examining the expression of latent ( EBER ) and lytic EBV transcripts in different layers of OHL lesions and normal tonsil epithelium confirm that lytic EBV transcripts are largely confined to differentiated epithelium in humans , although EBER transcripts can be observed in both the undifferentiated and differentiated cells ( consistent with the differentiation effects in NOKs-Akata cells ) . This raises the question of whether low-level latent EBV infection in basal epithelial cells ( too low to be reproducibly detected by EBER in situ hybridization staining ) actually occurs as a normal component of EBV infection . If so , this would help to explain how EBV infection can lead to the undifferentiated form of nasopharyrngeal carcinoma . The relative paucity of lytic versus latent EBV transcripts in normal tonsil tissues of immunocompetent individuals is perhaps somewhat surprising , given how much lytic protein expression is observed in the differentiated epithelial cells of OHL lesions . We speculate that the innate and/or adaptive immune responses in immunocompetent individuals very efficiently inhibits viral reactivation and/or eliminates cells with lytic infection , and that latently infected cells are more resistant to the host immune responses . In the future , it will be important to examine oral epithelium from patients who have recently recovered from infectious mononucleosis , since such patients secrete very high levels of infectious viral particles in their saliva for up to one year after infection , and thus latent and lytic EBV transcripts ( and proteins ) may be easier to detect in the normal oropharyngeal cells of these individuals .
The research using oral hairy leukoplakia lesions for IHC analysis in Fig 10 was approved by Institutional Review Board ( IRB ) of our institution ( University of Wisconsin Madison ) , as well as the IRB of our collaborating institution ( University of California , San Francisco ) . All adult subjects provided written informed consent ( and no children were in the study ) . The oral hairy leukoplakia studies shown in Table 2 , approved by the IRB of our collaborating institution ( Louisiana State University Health Sciences Center , Shreveport ) , used de-identified tissues obtained from the AIDS and Cancer Resource Specimen resource . The tonsil studies shown in Table 3 and S6 Fig , approved by the IRB of our collaborating institution ( Louisiana State University Health Sciences Center , Shreveport ) , used de-identified samples considered “left over” material by the pathologist that would have been discarded otherwise . The IHC studies on normal tongue and tonsil tissue shown in Figs 9 and S5 used de-identified tissues purchased commercially from Abcam and IHC World and were approved by the University of Wisconsin School of Medicine IRB . The NOKs cell line ( a gift from Karl Munger , Tufts University ) is a telomerase immortalized normal oral keratinocyte cell line that was derived as previously described [85] . NOKs cells were cultured in keratinocyte serum free medium ( KSFM ) ( Life technologies , Inc . ) supplemented with epidermal growth factor and bovine pituitary extract . The NOKs-Akata cell line was derived by co-culturing the NOKs cells with Burkitt lymphoma cells containing the Akata strain of EBV ( with an inserted G418 resistance selectable marker and a green fluorescent protein ( GFP ) gene ) [86] and then selecting with 50 ug/ml G418 , as previously described [36] . The Akata-GFP Burkitt lymphoma cell line was a gift from Kenzo Takada [received from Bill Sugden] ) . HONE-Akata cells ( a gift from Lawrence Young , University of Birmingham ) and CNE2-Akata cells ( a gift from K . W . Lo at the Chinese University of Hong Kong [received via Diane Hayward] ) are EBV-superinfected ( Akata strain ) HONE and CNE2 epithelial cell carcinoma cell lines that were originally thought to be derived from nasopharyngeal carcinomas but have been recently shown to be HPV infected and at least partially derived from HeLa cells [87] . Both of these cell lines were cultured in RPMI medium with 10% fetal bovine serum ( FBS ) , 1% penicillin-streptomycin ( pen-strep ) and 400 μg/mL G418 . C666-1 cells ( a gift from Dolly Huang ) , an EBV-infected nasopharyngeal carcinoma line [88] , and SNU . 719 cells , a gastric carcinoma line harboring EBV [89] , were cultured in RPMI with 10% FBS and 1% pen-strep . AGS gastric carcinoma cells ( obtained from ATCC ) were maintained in F-12 medium supplemented with 10% FBS and 1% pen-strep . AGS-Akata cell line is derived from AGS cells superinfected with Akata strain of EBV and selected for G418 resistance as previously described [90] . Raji ( obtained from ATCC ) , Mutu I ( a gift from Alan Rickinson ) , Mutu III , Jijoye ( ATCC ) , and Kem I ( a gift from Jeffrey Sample ) are EBV-positive Burkitt lymphoma cell lines and were cultured in RPMI supplemented with 10% FBS and 1% pen-strep . D4 LCL is an EBV-transformed ( B95 . 8 ) B cell lymphoblastoid cell line ( LCL ) . Plasmid DNAs were purified using QIAGEN Plasmid Maxi Kits as described by the manufacturer . pCDNA3 . 1-HA- KLF4 ( Addgene plasmid # 34593 , a gift from Michael Ruppert ) expresses an amino-terminal HA-tagged KLF4 [91] . pCDNA3-BLIMP1 expresses an amino-terminal FLAG-tagged BLIMP1 [92] . Plasmid pCpGL- Zp -668 is a luciferase reporter construct containing nucleotides -668 through +15 ( relative to the transcription initiation site ) of the EBV BZLF1 IE promoter ( Zp ) cloned between the SpeI and BglII restriction sites of pCpGL ( a gift from Michael Rehli [93] ) , a CpG-free vector driving expression of the luciferase gene . The pCpGL-Zp-40 construct contains nucleotides -40 through +15 ( relative to the transcription initiation site ) of the EBV BZLF1 IE promoter and serves as a negative control in some experiments . Plasmid pCpGL-Rp-1068 is a luciferase reporter construct containing nucleotides -1068 through +38 ( relative to the transcription initiation site ) of EBV BRLF1 IE promoter ( Rp ) cloned between the SpeI and BglII restriction sites of pCpGL . The 5’ Rp deletion mutants were constructed as described previously [39] . The names of the 5’ promoter deletions indicate the number of promoter nucleotides present in each construct upstream of transcription start site . Site-directed Rp mutants altering KLF4 consensus sites were constructed in the pCpGL-Rp-551 vector using the Strategene QuikChange Site-Directed Mutagenesis Kit , as per the manufacturer’s protocol , using the following primers: Mutant 1–5’-CTCTGGACATCCGCACGAATCAAATCACAATTTTTGGAGACCCGTC- 3’ and Mutant 2–5’- GCCCGGAGCAATGACTCTAGTTTGTCCTTGTGTGAGGTC-3’ . Reporter gene constructs were methylated in vitro using CpG methyltransferase M . SssI ( New England Biolabs ) as per the manufacturer’s protocol . Methylated and mock treated reporter gene constructs were cleaned by phenol chloroform extraction and salt precipitation , and complete methylation of these constructs was then confirmed by cutting the DNA with both HpaII ( which cannot digest methylated DNA ) and MspI ( which cuts irrespective of the methylation state ) . Transwell inserts ( 24 mm in diameter and 0 . 4 μm in pore size; Costar ) were coated with 1 ml of collagen ( 3 . 0 mg/ml; Wako Chemicals ) premix consisting of F-12 medium , 10% FBS and 1% Pen/Strep . Human foreskin fibroblasts ( 600 μl at 7 . 5 × 105 cells/ml ) were embedded into the remaining collagen mix and 2 . 5 ml was plated onto the collagen-coated Transwell inserts . The collagen-coated Transwell insert with embedded human fibroblasts was allowed to incubate for 4 days in a 5% CO2 incubator at 37°C in F-12 medium containing 10% FBS and 1% Pen/Strep . After 4 days , 150 μl of keratinocytes ( 1 . 4 x 106 cells/ml ) in keratinocyte plating medium ( F medium [1 . 88 mM Ca2+] ) containing 0 . 5% FBS , adenine ( 24 μg/ml ) , cholera toxin ( 8 . 4 ng/ml ) , hydrocortisone ( 2 . 4 μg/ml ) , and insulin ( 5 μg/ml ) were plated onto the collagen dermal equivalent . Four days after plating , the Transwell inserts were placed onto three 1-in2 cotton pads ( Bio-Rad ) in a six well tray ( BD Biosciences ) . The rafts were fed from below the Transwell insert with cornification medium ( keratinocyte plating medium containing 5% FBS and 10 μM C8:0 ) every other day . Eleven days after being lifted to the liquid-air interface , the rafts were fed for 8 h with cornification medium containing 10 μM bromodeoxyuridine ( BrdU ) . Subsequently , the rafts were embedded in 2% agar-1% formalin , fixed in 4% formalin at 4°C overnight , embedded in paraffin , and cut into 4-μm-thick cross sections . NOKs-Akata cells were treated for 48 hours with the following chemical reagents to induce lytic EBV reactivation: phorbol 12-myristate 13-acetate ( TPA; 20 ng/ml; Sigma ) , sodium butyrate ( 3 mM; Sigma ) and calcium chloride dehydrate ( 1 . 2 mM , Sigma ) . For ROCK inhibitor experiments , cells were treated with ROCK inhibitor Y27632 ( 10 μM , Enzo lifesciences ) at the same time as the TPA treatment . B cells were treated with the following chemical reagents to induce lytic EBV reactivation: anti-human IgG ( 10 μg/ml; Sigma ) , 5-aza-2’-deoxycytidine ( 2μM; Acros Organics ) , and TGFβ ( 5 ng/ml; Biolegend ) . Formalin-fixed , paraffin-embedded tissue sections were deparaffinized and then examined by IF as previously described [94] . Primary antibodies used were anti-Z ( BZ . 1 ) monoclonal antibody ( 1:200 , Santa Cruz Biotechnology SC-53904 ) , anti-K10 polyclonal antibody ( 1:1 , 000 , Covance PRB-159P ) , and anti-involucrin ( SY5 ) monoclonal antibody ( 1:1000 , Sigma , I9018 ) . Secondary antibodies used were Alexa 594 conjugated goat anti- rabbit ( red ) ( Invitrogen A-31571 ) and Alexa 488 conjugated goat anti-mouse ( green ) ( Invitrogen 21206 ) . Formalin-fixed , paraffin-embedded tissue sections were deparaffinized and then examined by IHC as previously described [95] . Antibodies used included anti-Z ( BZ . 1 ) monoclonal antibody ( 1:200 , Santa Cruz Biotechnology SC-53904 ) , anti-BMRF1 monoclonal antibody ( 1:200 , Vector Laboratories VP-E608 ) , anti-KLF4 polyclonal antibody ( 1:500 , Sigma-Aldrich HPA002926 ) and anti-BLIMP1 ( 1:1000 , Sigma-Aldrich HPA030033 ) . Human normal tongue tissue and tonsil tissue slides were commercially purchased ( Abcam and IHC World ) . EBER in situ hybridization studies were performed using the PNA ISH Detection Kit ( DakoCytomation ) according to the manufacturer’s protocol as previously described [95] . Formalin-fixed , paraffin-embedded tissue sections were deparaffinized and then examined by FISH . A digoxigenin ( DIG-11-dUTP , Roche ) -labeled probe was used to analyze viral DNA amplification . Nick translation was used to label EBV bacmid DNA ( B95 . 8 ) with digoxigenin to make the probe . Deparaffinized sections were incubated in pre-hybridization buffer ( 2XSCC , 0 . 5% IPECAL , pH 7 . 0 ) for 30 minutes at 37°C . Sections were dehydrated using a series of ice cold ethanols ( 70% , 80% , 95% ) for 2 minutes each . Sections were dried by placing them in an empty container at 50°C for 5 minutes . Sections were then placed in denaturation solution ( 28 mL formamide , 4 mL 20X SSC pH 5 . 3 , 8 mL water ) at 72°C for 2 minutes . The ethanol series was repeated again , and after drying the sections , denatured probe was added to the slides . The probe was hybridized to the raft sections overnight at 37°C in a humidified chamber . After washing for 30 minutes twice with 2XSSC and 50% formamide at 50°C and 30 minutes twice with 2XSSC at 50°C , signals were detected with a digoxigenin-specific antibody conjugated to fluorescein isothiocyanate ( Sigma , F3523 ) at 2% by volume in STM solution ( 4X SSC , 5% non-fat dried milk , 0 . 05% Tween-20 , 0 . 002% sodium azide ) for 30 minutes at 37°C . Nuclei were counterstained with DAPI . Plasmid DNA was transfected into epithelial cells with Lipofectamine 2000 transfection reagent ( Invitrogen ) according to the manufacturer’s protocol . In general , epithelial cells ( in a 12 well plate ) were transfected with either 100 ng pcDNA3-KLF4 , 100 ng pcDNA3-BLIMP1 , or 50 ng pcDNA1-KLF4 plus 50 ng pcDNA3-BLIMP1 , in addition to 400 ng pcDNA3 . 1 . KLF4 siRNA ( Origene SR306162 ) was transfected into epithelial cells with Lipofectamine RNAiMax ( Invitrogen ) , as per the manufacturer’s protocol . Raji and Jijoye cells were transfected using Amaxa cell line nucleofactor kit V , as per the manufacturer’s protocol . Cell lysates were harvested in Sumo lysis buffer including protease inhibitors ( Roche ) as described previously [96] . Protein concentration was determined using the Sumo protein assay ( Biorad ) , and proteins were separated in SDS-10% polyacrylamide gels and then transferred onto a nitro-cellulose membrane . Membranes were blocked in PBS containing 5% milk , and 0 . 1% Tween 20 solution . Membranes were then incubated in the following primary antibodies: anti-Z ( Santa Cruz , product # sc-53904 , 1:250 ) , anti-BMRF1 ( Millipore , product # MAB8186 , 1:3 , 000 ) , anti-R rabbit polyclonal antibody directed against the R peptide ( peptide sequence EDPDEETSQAVKALREMAD , 1:2 , 500 ) , anti-KLF4 ( Cell Signaling , product # 4038 , 1:1 , 000 ) , anti-BLIMP1 ( Cell Signaling , product # 9115 , 1:1 , 000 ) , anti-β-actin ( Sigma , product # A5441 , 1:5 , 000 ) , anti-tubulin ( Sigma , product # T5168 , 1:2 , 000 ) , and anti-involucrin ( Sigma , product # I9018 , 1:3000 ) . The secondary antibodies used were horseradish peroxidase ( HRP ) - labelled goat anti-mouse antibody ( Fisher Scientific , 1:5 , 000 ) and HRP- labeled anti-rabbit antibody ( Fisher scientific , 1:5 , 000 ) . Cells were washed with cold PBS and harvested 48 hours after transfection in 1X reporter lysis buffer ( Promega ) , subjected to one freeze-thaw cycle , and then the relative luciferase units were quantified using a BD Monolight 3010 luminometer ( BD Biosciences ) and luciferase assay reagent ( Promega ) . The fold change for each condition was calculated relative to the promoter activity in the presence of the control vector , pCDNA . For each condition , at least 3 independent experiments were performed in duplicates . 2x107 cells were cross-linked in 1% ( w/v ) formaldehyde ( Sigma ) for 5 min at room temperature and the cross-linking reaction was quenched by addition of glycine to a final concentration of 0 . 125M . Cells were washed twice with cold PBS and lysed in 1 ml of lysis buffer ( 50 mM Tris-HCl [pH 8 . 1] , 10 mM EDTA , 1% [w/v] SDS , 1 mM PMSF , 1 μg/ml leupeptin , 20 μg/ml aprotinin ) for 30 min on ice before extensive sonication with a Qsonica LLC Q700 sonicator . After extract clearing by centrifugation , supernatants were diluted 1:10 in dilution buffer ( 16 . 7 mM Tris-HCl [pH 8 . 1] , 1 . 2 mM EDTA , 167 mM NaCl , 1 . 1% [v/v] Triton X-100 , 0 . 01% [w/v] SDS , 1 mM PMSF , 1 μg/ml leupeptin , 20 μg/ml aprotinin ) . Aliquots of each input chromatin lysate were reserved for qPCR analysis . 1ml of diluted chromatin lysate was incubated with ChIP antibodies with rotation at 4°C overnight . Primary antibodies used were anti-KLF4 polyclonal antibody ( H-180 , Santa Cruz 20691 ) and anti-RNA polymerase II S5 phospho-specific antibody ( 4H8 , Abcam ab5408 ) . 15ul Protein A/G magnetic beads ( Thermo 88802 ) were added to each 1ml ChIP and incubated for 1 hour at 4°C with rotation . Next , magnetic beads were pelleted with magnetic separation rack and washed once with cold low salt wash buffer ( 20 mM Tris-HCl [pH8 . 1] , 2 mM EDTA , 150 mM NaCl , 1% [v/v] Triton X-100 , 0 . 1% [w/v] SDS ) , once with high salt wash buffer ( identical to low salt wash buffer , except 500 mM NaCl ) , once with LiCl wash buffer ( 10 mM Tris-HCl [pH8 . 1] , 1 mM EDTA , 0 . 25 M LiCl , 1% [v/v] NP40 , 1% Deoxycholic acid ) , and finally twice with TE buffer ( 10 mM Tris-HCl [pH8 . 1] , 1 mM EDTA ) . Samples were then resuspended in 150 μl of elution buffer ( 0 . 1 M NaHCO3 , 1% [w/v] SDS ) and rotated for 20 min at room temperature . Two rounds of elution of protein-DNA complexes were pooled . Reversal of cross-linking was accomplished by incubation of pooled eluates at 65°C for 4 hours after addition of NaCl to final concentration of 200mM and 100ug/ml Proteinase K . DNA was purified using the QIAquick PCR purification kit ( 28706; Qiagen ) and quantified using a BioRad CFX96 system with the iTaq universal SYBR Green supermix ( 1725121;Bio-Rad ) . Purified input chromatin lysate was used in real-time PCR reactions for standardization . Viral titration assays were performed in CNE2-Akata cells , as previously described [97] . CNE2-Akata cells were transfected with either control vector , KLF4 , BLIMP1 or KLF4 and BLIMP1 together ( for synergy studies ) . Supernatant was harvested , 96 hours after transfection , and was passed through a 0 . 8 um filter . 10 uL of this supernatant ( for each condition ) was used to infect Raji cells ( 2 x 105 cells/ condition ) , followed by the addition of phorbol-12-myristate-13-acetate ( TPA ) ( 20 ng/ml ) and sodium butyrate ( 3 mM final concentration ) , 24 hours after infection . Viral titer was determined by counting the number of GFP-positive Raji cells , 48 hours after infection . HONE-Akata cells were transfected with either control vector , KLF4 , BLIMP1or KLF4 and BLIMP1 together . Intracellular DNA was harvested from these cells , 96 hours after transfection , using the GenElute mammalian genomic DNA miniprep kit ( Sigma ) according to manufacturer’s protocol . Primers directed against the EBV BZLF1 promoter ( forward primer – 5’-TGCCTGTGGCTCATGCATAGTTTC-3’ and reverse primer – 5’-GCCATGCATATTTCAACTGGGCTG–3’ ) were used to quantify viral DNA . DNA was also amplified using primers directed against the beta-globin gene ( forward primer – 5’-GAGGCTCTGACCATAACCAAA-3’ and reverse primer- 5’-GACAAGGCTGCAAGCTATACTA-3’ ) , and the EBV quantification was normalized to the beta-globin result to correct for variations in DNA quality and quantity . All samples were assayed in duplicate . DNA was amplified using the iTaq universal SYBR Green supermix as suggested by the manufacturer ( catalogue # 1725121; Bio-Rad ) in a BioRad CFX96 machine . The PCR amplification protocol was initiated at 98°C for 2 minutes followed by 39 PCR cycles consisting of 5 seconds at 98°C followed by 60°C for 30 seconds . Mutagenesis of KLF4 was performed using the CRISPR-Cas9 technology , as previously described [98 , 99] . The following oligos were annealed and cloned into LentiCRISPRV . 2 plasmid ( Addgene plasmid # 52961 , gift from Feng Zheng ) [98]: Oligo 1: 5’-CACCGGGAGCCGGTGCGGCTTGCGG-3’ and Oligo 2–5’-AAACCCGCAAGCCGCACCGGCTCCC-3’ . 4 ug of either the control LentiCRISPRV . 2 plasmid or plasmid containing the KLF4 guide RNA , 0 . 6 ug VSV-G and 1 . 4 ug of ps-Pax2 ( a gift from Didier Trono [Addgene plasmid # 12260] ) were co-transfected into 293T cells in a 10 cm dish to package the lentivirus . Supernatant , containing the lentivirus , was harvested at 2 and 3 days post infection and was used to infect HONE-Akata cells . Infected HONE-Akata cells were selected using 1 ug/mL of puromycin . KLF4 mutagenesis ( knock out ) was confirmed using western blot analysis . Paraffin-embedded specimens were sectioned to a thickness of 4–5 μm , stained with hematoxylin and eosin and kept in desiccant until used for laser capture microdissection ( LCM ) . Institutional Review Board approval was obtained . The cells of interest were identified by their morphology and captured by a PixCell IIe LCM system ( Arcturus Engineering , Inc . , Mountain View , CA , USA ) . CapSure HS LCM caps ( Arcturus ) coated with infrared light absorbing ethylene vinyl acetate ( EVA ) were placed over the tissue . The laser spot size and power were adjusted to melt the EVA film and capture cells only in the area irradiated by the very low-power infrared targeting beam . The laser power was 55–60 mW , laser pulse duration was 1 . 5–1 . 8 msec , and laser spot size was 7 . 5 μm in diameter . Total RNA was extracted , purified , DNase treated , and reverse transcribed from the LCM captured cells using the Paradise Whole Transcript RT Reagent System ( Arcturus ) according to the manufacturer’s instructions . A human formalin-fixed universal reference RNA was reverse transcribed in parallel for use as a positive control . Real-time quantitative PCR ( QPCR ) was performed using an ABI Prism 7000 Sequence Detector with SYBR Green . The PCR reactions were set up in a 96-well optical plate in duplicate by adding the following reagents into each well: 2 . 5 μl of cDNA , 12 . 5 μl of SYBR Green PCR Master Mix ( Applied Biosystems , Foster City , CA , USA ) ; the final concentrations of primers were 0 . 3 μmol/L in a final volume of 25 μl . The PCR amplification protocol was initiated at 50°C for 2 minutes followed by 10 minutes at 95°C and 40 PCR cycles consisting of 15 seconds at 95°C followed by 60°C for 1 minute . To exclude the possibility of contamination with genomic DNA each reaction also contained a control PCR amplification of isolated RNA to which no reverse transcriptase had been added and a water only control . The sequences of the primers that were used are summarized in Table 1 . All samples were tested with the reference genes glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) or Cyclophilin A ( CycA ) for data normalization to correct for variations in RNA quality and quantity . All samples were assayed in duplicate or triplicate , and values were expressed as mean ± standard deviation . Control experiments with freshly isolated peripheral B cells admixed with cultured epithelial cells showed that the sensitivity of the assay was such that 1 B cell could be detected in a background of 1 , 000 epithelial cells with threshold cycle ( Ct ) values for GAPDH and CD20 of 26 . 40±0 . 16 and 38 . 78±0 . 18 respectively . The specificity of amplification of targets with high Ct values was confirmed by analysis of the temperature dissociation curves . | Lytic EBV infection of differentiated oral epithelial cells results in the release of infectious viral particles and is required for efficient transmission of EBV from host to host . Lytic infection also causes a tongue lesion known as oral hairy leukoplakia ( OHL ) . However , surprisingly little is known in regard to how EBV gene expression is regulated in epithelial cells . Using a stably EBV- infected , telomerase-immortalized normal oral keratinocyte cell line , we show here that undifferentiated basal epithelial cells support latent EBV infection , while differentiation of epithelial cells promotes lytic reactivation . Furthermore , we demonstrate that the KLF4 cellular transcription factor , which is required for normal epithelial cell differentiation and is expressed in differentiated , but not undifferentiated , normal epithelial cells , induces lytic EBV reactivation by activating transcription from the two EBV immediate-early gene promoters . We also show that the combination of KLF4 and another differentiation-dependent cellular transcription factor , BLIMP1 , synergistically activates lytic gene expression in epithelial cells . We confirm that KLF4 and BLIMP1 expression in normal tongue epithelium is confined to differentiated cells , and that KLF4 and BLIMP1 are expressed in a patient-derived OHL tongue lesion . These results suggest that differentiation-dependent expression of KLF4 and BLIMP1 in epithelial cells promotes lytic EBV infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Differentiation-Dependent KLF4 Expression Promotes Lytic Epstein-Barr Virus Infection in Epithelial Cells |
HTLV-1 orf-I is linked to immune evasion , viral replication and persistence . Examining the orf-I sequence of 160 HTLV-1-infected individuals; we found polymorphism of orf-I that alters the relative amounts of p12 and its cleavage product p8 . Three groups were identified on the basis of p12 and p8 expression: predominantly p12 , predominantly p8 and balanced expression of p12 and p8 . We found a significant association between balanced expression of p12 and p8 with high viral DNA loads , a correlate of disease development . To determine the individual roles of p12 and p8 in viral persistence , we constructed infectious molecular clones expressing p12 and p8 ( D26 ) , predominantly p12 ( G29S ) or predominantly p8 ( N26 ) . As we previously showed , cells expressing N26 had a higher level of virus transmission in vitro . However , when inoculated into Rhesus macaques , cells producing N26 virus caused only a partial seroconversion in 3 of 4 animals and only 1 of those animals was HTLV-1 DNA positive by PCR . None of the animals exposed to G29S virus seroconverted or had detectable viral DNA . In contrast , 3 of 4 animals exposed to D26 virus seroconverted and were HTLV-1 positive by PCR . In vitro studies in THP-1 cells suggested that expression of p8 was sufficient for productive infection of monocytes . Since orf-I plays a role in T-cell activation and recognition; we compared the CTL response elicited by CD4+ T-cells infected with the different HTLV-1 clones . Although supernatant p19 levels and viral DNA loads for all four infected lines were similar , a significant difference in Tax-specific HLA . A2-restricted killing was observed . Cells infected with Orf-I-knockout virus ( 12KO ) , G29S or N26 were killed by CTLs , whereas cells infected with D26 virus were resistant to CTL killing . These results indicate that efficient viral persistence and spread require the combined functions of p12 and p8 .
HTLV-1 causes Adult T-cell Leukemia/Lymphoma ( ATLL ) [1] , [2] or HTLV-1 Associated Myelopathy/Tropical Spastic Paraparesis ( HAM/TSP ) in approximately 2–3% of the 15–20 million individuals infected by the virus worldwide [3] , [4] . HTLV-1 persists in the host despite a vigorous cellular and antibody response , suggesting that the virus has developed effective mechanisms to counteract host immune surveillance [5] . The HTLV-1 open reading frame-I ( orf-I ) protein products p12 and p8 increase NFAT activity [6] , [7] , STAT-5 transcriptional activity and IL-2 production [8]–[10] in T-cells . In addition , they cause down-regulation of ICAM-1 and -2 , but not ICAM-3 surface expression , allowing escape of infected cells from NK cell killing [11] . The p12 protein precursor is processed by proteolytic cleavage that removes a non-canonical endoplasmic reticulum ( ER ) retention/retrieval signal at its amino-terminus to yield p8 ( Figure 1A ) [12] , [13] . The p8 protein traffics to the cell surface , is recruited to the immunological synapse following T-cell receptor ( TCR ) ligation , and down-regulates TCR proximal signaling [14] . In addition , p8 increases cell adhesion and virus transmission and is transferred to neighboring cells via cellular conduits [15] . Both p8 and p12 can form homo- or hetero-dimers through a highly conserved single cysteine ( position 39 ) or are palmitoylated and remain monomeric ( Figure 1A ) [16] , [17] . Orf-I knockout viruses are not infectious in non-human primates [18] , suggesting the importance of orf-I in human infection . Here , studying a cohort of 160 HTLV-1 infected individuals , using an experimental model of macaque infection and using in vitro relevant models of HTLV-1 infection , we demonstrate that natural mutations within orf-I can affect the relative amounts of p12 and p8 , which in turn , correlate with viral DNA levels in blood , the best predictor of risk for the development of HAM/TSP or ATLL [19]–[22] . In addition , we demonstrate that both proteins are essential for the in vitro resistance to cytotoxic T-lymphocyte ( CTL ) killing of HTLV-1 infected cells .
Analysis of orf-I was performed on 160 HTLV-1 infected individuals from various geographical areas ( Table 1 ) , 79 had HAM/TSP and 81 were carriers . Genomic DNA was isolated from patient PBMCs and used to quantify the viral DNA load and for studies on the orf-I gene . As expected , individuals with HAM/TSP had significantly ( p = 0 . 0001 ) higher PBMC viral DNA loads than carriers ( Figure 1B ) . We obtained DNA sequences for a total of 834 clones from these patients and compared them to our reference orf-I cDNA [13] and found 216 variants ( i . e . , one or more nucleotide changes compared to the consensus sequence ) . One hundred thirty of these variants ( 85% ) were unique . The most frequent non-synonymous mutations within the orf-I gene yielded G29S , P34L , S63P , R88K , and S91P amino acid changes . In line with orf-I being necessary for infection , none of the approximately 1600 orf-I sequences analyzed had a premature termination codon . We selected 17 non-synonymous mutations based on either their proximity to the cleavage sites or their frequency in humans and inserted them into the reference orf-I cDNA ( herein defined as p12WT ) and transfected the expression constructs into 293T-cells . The relative amount of p8 and p12 , evaluated by Western blot and densitometric scans was calculated as a percentage of total expression from the orf-I gene ( Figure 1C ) . A minimum of two up to 20 independent Western blot experiments were performed as indicated for each mutant in the legend of Figure 1C . We observed 3 distinct patterns of expression ( Figure 1C and Figure 1D ) . The first consisted of balanced expression of p8 and p12 , as for p12WT . All these mutations were downstream of both cleavage sites ( mutants S91P-61% of patients , R88K-10% of patients , S69G-12% of patients and R83C-6% of patients ) . The second class consisted of predominant expression of p12 ( mutants F3L-4% of patients , P45L-9% of patients , S23P-15% of patients , P34L/F61L-23% of patients , S63P-62 . 5% of patients , L66P-10% of patients and G29S-30% of patients ) ( Figure 1A ) . The viral DNA loads for patients with the G29S mutations ( Supplemental Table S3 ) follow the same trend as the overall patient pool in that patients with HAM/TSP had higher viral DNA loads . The third pattern was generated by a rare mutation in position 26 between the two cleavage sites from aspartic acid ( D ) ( present in p12 WT , see Figure 1A ) to either asparagine ( N ) ( mutated in 5% of patients ) or glutamic acid ( E ) ( mutated in 2% of patients ) , resulting in the predominant expression of p8 [12] . Analysis of the three orf-I expression patterns and viral DNA levels in blood revealed significantly higher viral levels in individuals whose cDNA expressed both p8 and p12 ( p = 0 . 05 ) , compared to those that predominantly expressed either p8 or p12 ( Figure 1E ) . No correlation with disease status was observed within this patient cohort . To directly assess the requirement of p8 and p12 on viral infectivity and persistence , we engineered the HTLV-1 molecular clone pAB [18] , that carries an orf-I identical to p12WT designated here as pAB-D26 ( Figure 2A ) . Glycine 29 was substituted with serine to generate pAB-G29S as this mutation impairs cleavage of p12 to p8 , resulting predominantly in p12 expression ( Figure 1D ) [12]–[18] . Because substitution of N or E at position 26 , results in predominant expression of p8 ( Figure 1D ) , we generated pAB-N26 . Importantly , the mutations introduced in the orf-I gene did not alter the amino acid sequences of the hbz or orf-II genes that overlap with orf-I . The isogenic clone pAB-p12KO , mutated at the orf-I initiation ATG to eliminate expression of both p8 and p12 ( Figure 2A ) was used as a control since it is infectious in vitro but not infectious in vivo [18] . The molecular clones were co-transfected with an HTLV-1-LTR-Lucifease construct into 293T-cells to demonstrate their equivalent ability to produce the Tax protein and activate the viral LTR ( Figure 2B ) . All viruses produced equivalent amounts of intracellular p24Gag ( Figure 2C ) and extracellular p19Gag ( Figure 2D ) . We generated stable 729 . 6 human B-cell lines producing the viral mutants as described [18] . These cell lines were clonal and expressed equivalent levels of intracellular p24Gag and extracellular p19Gag ( Figure 2E , lower panels ) . We observed differences in viral transmission when the cell lines were co-cultured with the reporter cell line , BHK1E6 [23] , which contains the β-galactosidase gene under the control of the HTLV-1-LTR promoter . The D26 , G29S , or 12KO viruses were transmitted equivalently , but the N26 virus was transmitted 10-fold more efficiently ( Figure 2E ) , consistent with the ability of p8 to increase virus transmission [15] . Whether it is p8 , p12 , or both that contribute to the requirement of orf-I for infection in vivo remains unclear [18] . To address this point , we inoculated intravenously the lethally γ-irradiated B-cell lines producing equivalent levels of p19Gag ( Supplemental Table S1 ) from the D26 virus into four macaques , the N26 virus into four macaques and the G29S virus into four macaques . One animal was inoculated with parental uninfected 729 . 6 cells as a control . Three ( P834 , P840 , P872 ) of the four animals exposed to the D26 virus became HTLV-1 positive by PCR with viral levels greater than 50 copies per one million cells for at least one time point throughout the study and fully seroconverted for viral antigens ( Figure 3 ) . In contrast , only one animal ( P845 ) exposed to the N26 virus was PCR positive for viral DNA and only three animals showed weak reactivity to HTLV-1 antigens . None of the four animals in this study or four animals from a previous study exposed to G29S virus became PCR positive or seroconverted ( Figure 3 ) . We verified that the virus in animal P845 , infected with N26 virus , retained the mutation at position 26 by cloning and sequencing orf-I from its PBMCs . These results suggest that expression of both p12 and p8 is required for efficient HTLV-1 infection and viral persistence . However , they also suggest that p8 may be sufficient for infection and at least partial seroconversion; particularly since none of the eight animals inoculated with virus predominantly expressing p12 seroconverted or had detectable viral DNA . HTLV-1 infects monocytes and dendritic cells [24]–[26] but the role of infected monocytes to HTLV-1 pathogenesis remains unclear . We have previously demonstrated that the abrogation of orf-I expression results in loss of HTLV-1 infectivity of primary monocyte-derived dendritic cells [18] and further that infection of the monocytic cell line THP-1 mirrored results of ex vivo , primary dendritic cells [25] . To define the relative contribution of p8 , p12 , or both to monocyte infection , we exposed the monocytic cell line THP-1 to equivalent amounts of virus as measured by p19Gag from unfiltered cell-free supernatants . Representative cultures are shown ( Figure 4 ) . Cultures exposed to D26 or N26 viruses had greater than 10 , 000 pg/ml of p19Gag in their supernatants at week 2 ( Figure 4A ) and virus production was maintained up to 16 weeks . In contrast , cultures infected with G29S or 12KO viruses had only background levels of p19Gag , as seen in control cultures ( mock-infected with 729 . 6 culture supernatant ) . Genomic DNA isolated from the exposed THP-1 cells at week 18 was tested by nested PCR for viral DNA . The level of HTLV-1 DNA detected by PCR was consistent with the level of p19Gag released into the supernatant and was highest in the cultures infected with the D26 and N26 viruses ( Figure 4B ) . Quantitative PCR showed that the D26 and N26 infected cultures contained 3–4 viral DNA copies per cell , while the G29S and 12KO infected cultures contained less than 1 copy per cell ( Figure 4C ) . Interestingly , despite the differences in viral production , all HTLV-1 infected THP-1 cultures displayed down-regulation of CD14 and up-regulation of the activation markers HLA-DR and CCR7 ( Figure 4D ) . These results suggest that p8 expression is necessary and sufficient for productive HTLV-1 infection in monocytes since p8 is expressed in both D26 and N26 , but not in G29S and 12KO . CTLs play an important role in limiting viral replication and spread by recognizing and lysing virally infected cells . The orf-I protein products interfere with the normal trafficking of the MHC-class-I molecule and are thought to reduce CTL recognition [27] , [28] . To dissect the impact of p12 and/or p8 on CTL responses in the context of the whole virus , we generated immortalized infected CD4+ T-cells lines from an HLA . A2 healthy donor that allowed the use of the human CTL clone from a HAM/TSP patient that recognize the HLA . A2 restricted Tax peptide [11]–[19] [29] . The CD4+ T-cells were cultured for over a year prior to analysis; viral production , viral DNA copy numbers , and the level of expression of the orf-I gene in the infected cultures is summarized in Supplementary Table S2 . In line with previous results [15] , T-cells producing N26 transmitted virus better than those producing D26 , G29S and p12KO ( Figure 5A ) . We reported previously that orf-I expression down-regulates the surface expression of major histocompatibility complex ( MHC ) -class-I in overexpression models [28] . Interestingly , in here , we observed that the surface expression of HLA . A2 was clearly down-regulated in CD4+T-cells that produce the G29S virus indicating that p12 expressed by the virus , down-regulates MHC-class-I in primary human CD4+ T-cells ( Figure 5B ) . Next , we studied whether infection of T-cells with the different viruses affected their susceptibility to CTL killing . The CD4+ T-cells infected with D26 , N26 , G29S and 12KO were loaded with equivalent amounts of the immunodominant Tax [11]–[19] peptides and co-cultured with the CTL clone at various effector-to-target ratios . We observed the highest CTL killing of the 12KO cells , suggesting that the absence of both p8 and p12 makes cells susceptible to CTL killing ( Figure 5C ) . We found reduced killing of cells infected with N26 that predominantly express p8 , as well as in G29S that predominantly express p12 . Strikingly , we observed nearly complete resistance to CTL killing at all effector-to-target ratios of cells infected with D26 that express a balanced level of p8 and p12 ( Figure 5C ) . The resistance to CTL killing of the D26 infected CD4+ T-cells was abrogated by siRNAs targeting the orf-I mRNA but not control siRNA ( Figure 5D and 5E ) . These results suggest that balanced expression of p12 and p8 is required to protect HTLV-1 infected cells from CTL killing .
The p12 precursor , encoded by orf-I , contains two proteolytic cleavage sites , the first site , between amino acids 9 and 10 and the second site , between amino acids 29 and 30 [12] . The p12 precursor is an ER associated protein and its cleavage removes a non-canonical ER retention/retrieval signal that generates p8 , a protein that localizes to the cell surface [12] . Both p8 and p12 interact with the β and γc chains of the interleukin-2 receptor ( IL-2R ) [10] , the heavy chain of the MHC-class-I [28] , calreticulin and calnexin [30] , and ICAM-1 and ICAM-2 [11] . The p8 protein traffics to lipid rafts , is recruited to the immunologic synapse following T-cell receptor ( TCR ) ligation where it down-regulates TCR proximal signaling [12] and co-localizes with lymphocyte function-associated antigen-1 ( LFA-1 ) , increasing its clustering [15] . The p8 protein also increases T-cell adhesion , the formation of cellular conduits , and HTLV-1 transmission [15] . A novel feature of p8 is its ability to be rapidly transferred from cell-to-cell through cellular conduits [15] . Here , we investigated the specific contribution of each isoform to viral infectivity of T-cells and monocytes in vitro and in viral persistence in vivo . We hypothesized , that genetic mutations , which surround the putative cleavage sites , affect the relative levels of p8 and p12 that may have consequences in HTLV-1 infection . By using reverse genetics on samples from HTLV-1 infected individuals , we have identified genetic polymorphisms that affect the efficiency of cleavage of the p12 precursor protein into p8 . The first and most frequent group of mutations results in an intermediate efficiency of cleavage that yields an equivalent amount of p8 and p12 . This phenotype is associated with productive infection of monocytes and a high viral DNA level in blood that is a correlate of disease development [20]–[22] , [31] . The second most frequent phenotype , yields predominantly p12 and affects the ability of the virus to productively infect monocytes . The third rarer phenotype , results in higher levels of p8 and a virus that retains its ability to productively infect monocytes . However , consistent with our studies in macaques and CTL sensitivity , both the second and third phenotypes are associated with low virus DNA levels in blood of naturally infected humans . We found that HLA-DR expression , as well as CD80 and CCR7 were up-regulated in monocyte cultures infected by HTLV-1 , even when low or undetectable viral proteins are expressed . The role of infected monocytes to HTLV-1 pathogenesis is unclear . However , HTLV-1 infection of monocytes has been demonstrated [25] , [26] , [32] , [33] and viral infection is associated with an increased frequency of more differentiated monocytes ( CD16bright ) that may spread the virus to tissues [34] . These results suggest that orf-I plays a role in viral persistence however , an early study by Furukawa and colleagues [35] found in one HAM/TSP patient a virus with a mutation at the start codon of orf-I and that this virus was transmitted in the individual's family . In contrast to our work , Furukawa et al . did not clone the orf-I products and assess its expression and stability [35] . In addition , the authors have not ruled out that the orf-I gene was expressed in those individuals through alternative splicing . Several groups have shown that cryptic splice sites and donor sites are present in retroviral sequences and that gene products can be produced through alternative splice acceptor/splice donor usage . Thus , although we have not demonstrated an absolute requirement for orf-I in HTLV-1 infection in humans , it is clearly required in non-human primates [18] . In addition , the results of this study and that of Furukawa et al . [35] , finding only 1 in 304 patients which do not retain orf-I expression ( 0 . 3% ) , suggests that orf-I expression is likely to provide an advantage in HTLV-1 persistence . Over-expression studies showed that p12 contributes to evasion from CTL by interacting with the MHC-class-I Heavy chain ( Hc ) in the ER and preventing its association with β2-microglobulin [27] , [28] , [36] . This interaction induces the MHC-class-I Hc retro-translocation into the cytosol for degradation by the proteasome , decreasing cell surface MHC-class-I . The p8 protein was recently shown in exogenous expression studies to be transferred to uninfected cells [15] . Therefore , we speculate that the contribution of p8 to CTL escape may be ascribed to the ability of this protein to be transferred to CD8+ T-cells , whereby it may down-regulate TCR signaling , resulting in the weakening of the strength of the immunological synapse [15] , and inhibition of CTL degranulation . Indeed , the p8 protein is recruited to lipid rafts within the immunological synapse upon engagement of TCR by CD3 ligation and causes T-cell anergy [12] . More recent studies demonstrated a reduction in the strength of the immunological synapse in the presence of p8 [15] . This is in line with the finding that not only the number of HTLV-1-specific CTLs is important , but also their functional avidity [37] and even if they are abundant [38]–[40] , they do not clear infection . Collectively our results suggest a model whereby a combination of effects of p8 and p12 on monocyte infectivity , viral transmission , and escape from CTL favors viral persistence ( Figure 6 ) . A virus expressing both p12 and p8 ( D26 ) infects monocytes , is efficiently transmitted to CD4+ T-cells , renders them less prone to CTL lysis and persists ( Figure 6A ) . In contrast , a virus ablated in p8 and p12 expression ( 12KO ) is poorly infectious in monocytes and CD4+T-cells in vitro , the infected cells are susceptible to CTL killing and infection is not sustained in vivo ( Figure 6B ) . Virus expressing mainly p8 ( N26 ) , has an intermediate phenotype; it maintains its infectiousness for monocytes and CD4+ T-cells , but because it only partially protects infected CD4+ T-cells from CTL does not cause a robust infection in vivo ( Figure 6C ) . Consistent with the concept of co-dependence of p8 and p12 functions for viral persistence in the host , a virus expressing mainly p12 ( G29S ) is poorly infectious in monocytes , the CD4+ infected T-cells are partially susceptible to CTL killing and the virus is not infectious in macaques ( Figure 6D and Figure 3 ) . It is likely that p12 and p8 also affect other steps in antigen processing and presentation of HTLV-1 peptides on MHC-class-I . The p12/p8 proteins interact with calnexin and calreticulin [30] which may affect the folding of MHC-class-I and its peptide loading [41] . Similarly , the interaction of p12 and p8 with the 16 kDa protein of the V-ATPase that occurs in the ER [42] may prevent the assembly of the mature form of the V-ATPase and acidification of the secretory pathway . Recent work shows that the p8 protein traffics to the cell surface via the secretory pathway [12] . Thus , p8's association with the V-ATPase could alter not only the secretory and the endocytic pathway , but also receptor recycling on the cell membrane . Thus , future work is necessary to assess whether p12 and p8 interaction with the V-ATPase has important functional implication in antigen processing and presentation . Our results from the macaque studies and CTL killing assays suggest that HTLV-1 virus expressing p12 only should be efficiently eliminated . However , we do find infected individuals , both carriers and HAM/TSP patients , harboring HTLV-1-p12 only virus . Several factors may influence the persistence of HTLV-1 p12 only virus . First , from earlier studies by Nicot et al [8] , p12 expression through its activation of STAT5 decreases the IL-2 requirement and thus confers a proliferative advantage to infected cells . Second , we see that expression of p12 alone does down-modulate MHC class 1 expression on infected T-cells . The down-modulation could be sufficient in vivo to allow escape from some CTL clones . Third , p12 protein has been shown to down-modulate ICAM-1 and ICAM-2 suggesting that infected cells would be less susceptible to NK cell killing [11] . Further , while we find that HTLV-1 G29S virus does not productively infect THP-1 cells , infection does occur and in preliminary studies we find that activation of infected cells stimulates infectious virus production . This would allow the virus to persist undetected in an infected individual and upon activation spread of the virus . Finally , from our studies on p13 and Tax [43] we find that there is significant interplay between viral proteins . Our studies have focused on orf-I mutations , but it is possible that changes in other viral genes can impact the role of orf-I in immune evasion . In conclusion , our data suggest that while infection of monocytes is important in HTLV-1 infection , viral persistence also necessitates a coordinated expression of p12 and p8 to avoid CTL recognition of infected cells . Thus , pharmacologically altering the efficiency of cleavage of the p12 precursor could have profound effects on viral persistence , by restoring the effectiveness of the host immune response to HTLV-1 and ultimately decreasing the risk of disease development through the reduction of the number of HTLV-1 infected cells .
This study was carried out in strict accordance with the recommendations described in the Guide for the Care and Use of Laboratory Animals of the National Institute of Health , the Office of Animal Welfare and the United States Department of Agriculture . All non-human primate work was approved by the NCI Division of Intramural Research Animal Care and Use Committees ( IACUC; protocol no . 458 ) . The animals were housed , feed , given environmental enrichment and handled in accordance with the standards of the Association for the Assessment and Accreditation of Laboratory Animal Care International . Appropriate steps were taken to minimize suffering in accordance with the Weatherall report ( “The use of non-human primates in research” ) . The animals were housed and experiments conducted at Advance Bioscience Laboratories in Rockville , MD in accordance with the standards of the American Association for Accreditation of Laboratory Animal Care . Non-human primates are housed in a rolling rack system and the cage configuration within the rooms allow for establishment of visual contact with other species members . Positive human interaction with the staff includes providing food treats , positive verbal and non-verbal communication , systematic husbandry and consistent staffing . A dietary enrichment and novel food program has been in place in the colony since 1987 . Each animal is provided with sensory and cognitive enrichment that include foraging and food-based enrichment strategies , toys , auditory and visual enrichment and hideaways . All procedures were carried out under anesthesia ( Telazol , Ketamine/Xylazine or Ketamine HCl ) by trained personnel under the supervision of veterinary staff and all efforts were made to ameliorate the welfare and to minimize animal suffering in accordance with the Weatherall report for the use of non-human primates recommendations . Early endpoint criteria , as specified by the IACUC approved score parameters , were used to determine when animals should be humanely euthanized . Blood samples from HTLV-1-infected patients and non-infected ( ND ) donors were obtained from the Centre Hospitalier Universitaire de Fort-de-France in Martinique and Institut Pasteur de Cayenne in French Guyana , the Bahia School of Medicine and Public Health and the National Institutes of Health Clinical Center . Patients suffering from HAM/TSP or HTLV-1 asymptomatic carriers were recruited according to World Health Organization ( WHO ) criteria . All subjects gave fully informed , written consent and all clinical investigations have been conducted according to the principles expressed in the Declaration of Helsinik . All samples were anonymized and research conformed to the guidelines of the ethics review board of the National Cancer Institute . The study comprised 160 HTLV-1 infected individuals from different geographical regions ( Caribbean , France , North America , Africa , and Brazil ) with different disease status ( Table 1 ) . The subjects for the analysis were participants in research studies conducted at the institutions of the authors . Informed consent was written and obtained from each subject in accordance with the Declaration of Helsinki . DNA extracted from PBMCs of HTLV-1 infected individuals was used to determine the viral DNA load . Real-time PCR analysis of HTLV-1 ( Tax ) was performed with 100 ng of cellular DNA as previously described [19] . HTLV-1 viral DNA levels were calculated by the following formula: ( copies of HTLV-1 ( pX ) / ( copies of beta-actin/2 ) ×100 cells . We are using the term viral DNA load since our assay does not distinguish between integrated and unintegrated viral DNA . The same DNA was used as templates for PCR reactions using Platinum High Fidelity PCR Supermix ( Invitrogen , Carlsbad , CA ) according to the manufacturer's protocol . In the reaction , 10 pmol/µl of each primer: 12-Fwd 5′-CACCTCGCCTTCCAACTG-3′ , p12-p30-Rev 5′-GGAGTATTTGCGCATGGCC-3′ were used for amplification of the p12-p30 ( 872 bp ) region at Tm = 55°C . For samples with no visible amplified PCR product 2 µl of the PCR reaction was used as a template for nested PCR with primers: p12-nested-F 5′-GTCTAGTATAGCCATCAACC-3′ and p30-mid-nested-Rev 5′- CTGGACAGGTGGCCAGTA-3′ . PCR products were purified by gel electrophoresis and QIAquick Gel Extraction Kit ( Qiagen , Valencia , CA ) and subsequently , cloned into pCR4 TA TOPO vector ( Invitrogen , Carlsbad , CA ) according to the manufacturer's protocol . QIAprep Spin Miniprep Kit ( Qiagen , Valencia , CA ) was used for plasmids isolation . Five to 20 clones per patient were isolated and sequenced . The 1530 orf-I sequences for the HTLV-1 infected individuals are available from Genbank under the accession numbers in Text S1 . The study on the immunophenotype of blood monocytes was performed on patient samples obtained through the NIH Clinical Center ( Table 1 ) . The pME18S p12deltaSL expression plasmid has been described previously [14] . This plasmid served as a backbone for generation of p12 mutants by means of PCR or by QuickChange Site-Directed Mutagenesis Kit ( Stratagene , La Jolla , CA ) using site-specific mutagenic oligonucleotides according to the manufacturer's instructions . The following oligonucleotides were used and the sequence of plasmid clones was analyzed to confirm the mutations . F3L-F: 5′-CCTAGCACTATGCTGCTTCGCCTTCTCAGCfCCCT-3′ F3L-R: 5′-AGGGGCTGAGAAGGCGAAGCAGCATAGTGCTAGG-3′ S23P-F: 5′-GCTCCTGCTCTTCCTGCTTCCTCCGGGCGACGTCAGCG-3′ S23P-R: 5′-CGCTGACGTCGCCCGGAGGAAGCAGGAAGAGCAGGAGC-3′ D26N-F: 5′-CCTGCTTTCTCCGGGCAACGTCAGCGGCCTTC-3′ ( for p12 subgroup A template – with S ( serine ) at the 23rd amino acid position in p12 ) D26N-R: 5′-GAAGGCCGCTGACGTTGCCCGGAGAAAGCAGG-3′ ( for p12 subgroup A template ) D26N-F: 5′ -CCTGCTTCCTCCGGGCAACGTCAGCGGCCTTC-3′ ( for p12 subtype B template - with P ( proline ) at the 23rd amino acid position in p12 ) D26N-R: 5′-GAAGGCCGCTGACGTTGCCCGGAGGAAGCAGG-3′ ( for p12 subtype B template ) D26E-F: 5′-CTGCTTTCTCCGGGCGAAGTCAGCGGCCTTCTTC-3′ D26E-R: 5′- GAAGAAGGCCGCTGACTTCGCCCGGAGAAAGCAG-3′ G29S-F: 5′-TGCTTTCTCCGGGCGACGTCAGCAGCCTTCTTCTC-3′ G29S-R: 5′-GCGGAGAAGAAGGCTGCTGACGTCGCC-3′ delta29-F: 5′-GTGGCTCGAGACCATGCTTCTTCTCCGCCCGCCTC-3′ delta29-R: 5′-TCGGTCTAGAAACAACAACAATTGCATTCATTTTATGTTTCAGGTTCA-3′ P34L-F: 5′-GGCCTTCTTCTCCGCCTGCCTCCTGCGCCGTGC-3′ P34L-R: 5′-GCACGGCGCAGGAGGCAGGCGGAGAAGAAGGCC-3′ P45L-F: 5′-GCCTTCTCCTCTTCCTTCTTTTTCAAATACTCAGC-3′ P45L-R: 5′-GCTGAGTATTTGAAAAAGAAGGAAGAGGAGAAGGC-3′ S63P-F: 5′-CTCCCGCTCTTTTTTCCGCTTCCTCTTCTCCTC-3′ S63P-R: 5′-GAGGAGAAGAGGAAGCGGAAAAAAGAGCGGGAG-3′ L66P-F: 5′-GCTCTTTTTTTCGCTTCCTCCTCTCCTCAGCCCGTCGCTGCCG-3′ L66P-R: 5′-CGGCAGCGACGGGCTGAGGAGAGGAGGAAGCGAAAAAAAGAGC-3′ S69G-F: 5′-GCTTCCTCTTCTCCTCGGCCCGTCGCTGCCGAT-3′ S69G-R: 5′-ATCGGCAGCGACGGGCCGAGGAGAAGAGGAAGC-3′ R88K-F: 5′-GGCTCTTTCTCCCCTGGAAGGCCCCGTCGCAGCCGGCCG-3′ R88K-R: 5′-CGGCCGGCTGCGACGGGGCCTTCCAGGGGAGAAAGAGCC-3′ S91P-F: 5′-CCCCTGGAGGGCCCCGCCGCAGCCGGCCGCGGC-3′ S91P-R: 5′-GCCGCGGCCGGCTGCGGCGGGGCCCTCCAGGGG-3′ . Expression of all mutants were assessed by western blot analysis using the anti-HA1 antibody clones12CA5 and 3F10-HRP ( Roche Applied Science , Indianapolis , IN ) . 293T- and BHK1E6 cells were grown in Dulbecco's modified Eagle medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) , 2 mM penicillin-streptomycin and 5 mM L-glutamine . The 729 . 6 B-cells were grown in RPMI 1640 supplemented with 10% FBS , 2 mM penicillin-streptomycin and 5 mM L-glutamine . The HTLV-1 molecular clones pAB-D26 ( WT ) , pAB-G29S ( p12 ) , and pAB-p12KO were previously described . To generate pAB-N26 ( p8 ) , mutation of GAC to AAC at amino acid 26 of orf-I ( a glutamic acid to asparagine substitution ) was introduced into the pBST ClaI/SalI cassette using QuickChange Site-Directed Mutagenesis Kit ( Stratagene , La Jolla , CA ) and then ligated to the pACH backbone . The mutant clones were verified by DNA sequencing of the ClaI/SalI fragment inserted in the provirus . To confirm that the clones were producing virus , they were transfected into 293T-cells using Effectene reagent ( Qiagen , Valencia , CA ) . Briefly , 10 µg of DNA of pAB-D26 ( WT ) , pAB-N26 , pAB-G29S , and pAB-p12KO was transfected into 10 cm dish of 293T-cells . After 48 hours , the cells were extracted for total protein with radioimmunoprecipitation assay ( RIPA ) buffer and analysis of intracellular HTLV-1-p24 ( Advanced BioScience Laboratories Inc . , Rockville , MD ) and tubulin ( Sigma-Aldrich , St . Louis , MO ) . Intracellular Tax expression was characterized by co-transfecting molecular clones and an HTLV-1-LTR-luciferase reporter into 293T-cells . The pRL-TKLuc plasmid was used as a transfection control . After 48 hours , cells were extracted with Passive Lysis Buffer ( Promega , Milwaukee , WI ) and protein samples analyzed with Dual-Glo reagent ( Promega , Milwaukee , WI ) for LTR activation . The culture supernatant from these transfections were spun down to remove any cell debris and analyzed by p19Gag ELISA ( ZeptoMetrix , Buffalo , NY ) for virus production . The siRNA nucleofection assays were performed using the Human T-cell Nucleofection Kit ( Lonza , Basel , Switzerland ) and program O-017 as described by the manufacturer . Briefly , CD4+ D26 producing T-cells ( 2×106 ) were incubated with 20 nM of either control siRNA or siRNA to orf-I ( 5′GCACUAUGCUGUUUCGCCUUCUCAG3′ ) ( Stealth RNA , Invitrogen , Carlsbad , CA ) . Forty-eight hours after nucleofection , cells were used in the cytotoxicity assay . Knockdown of Orf-I expression was monitored by transient transfection of Orf-I expression constructs and siRNA into 293T-cells using Lipofectamine 2000 ( Invitrogen , Carlsbad , CA ) as described by the manufacturer . Stable HTLV-1 producing 729 . 6 human lymphoblastoid B-cells were produced as described previously [18] . Briefly 729 . 6 cells ( 5×106 ) were electroporated with 5 µg of pAB-D26 , pAB-G29S , pAB-N26 or pAB-p12KO using AMAXA Nucleofector II , Nucleofection kit V at M-013 ( Lonza , Basel , Switzerland ) according to the manufacturer's guidelines . Infected cells were selected by culture in neomycin as previously described [44] . The supernatant p19Gag production was measured by ELISA assay ( ZeptoMetrix , Buffalo , NY ) . To analyze the clonality of the cell lines , 2–5×107 cells were grown up in the absence of G418 and genomic DNA extracted using a Genomic DNA Wizard Kit ( Promega , Milwaukee , WI ) following the manufacturer's guidelines . DNA ( 25 µg ) was digested with either High Fidelity EcoRI or XhoI ( New England Biolabs , Ipswich , MA ) , run on an 0 . 8% agarose gel overnight , denatured with Denaturing Solution ( Biosource , Madison , WI ) and neutralized with Neutralizing Buffer ( KD Medical , Columbia , MD ) , then blotted overnight to Immobilon NY+ membrane and crosslinked at 1 . 5×106 J with a UV crosslinker . Biotinylated probe was synthesized with a Phototope kit and pAB-D26 molecular clone ( New England Biolabs , Ipswich , MA ) . Membrane was prehybridized in Ultrahyb buffer ( Ambion Life Technologies , Grand Island , NY ) at 42°C for 2 hours , 10 pmol of probe added and hybridized overnight . Membrane was washed according to the Ultrahyb manufacturer guidelines and developed using the Phototope Star chemiluminescence kit ( New England Biolabs , Ipswich , MA ) . Using negative selection beads ( Invitrogen , Carlsbad , CA ) , CD4+ T-cells were isolated from un-infected peripheral blood mononuclear cells . Stable HTLV-1 producing CD4+ T-cell lines were established by co-cultivation of donor un-infected primary HLA . A2+/CD4+ T-cells with lethally γ-irradiated 729 . 6-HTLV-1 infected lines . T-cells were cultured in RPMI supplemented with 20% FBS and 100 U of interleukin-2 for one year . Virus production was monitored by p19Gag ELISA ( ZeptoMetrix , Buffalo , NY ) and viral genomic sequences verified by sequencing of the ClaI-SalI fragment as described above . Chronically infected THP-1 cells were produced as previously described [25] . Briefly , supernatant from 729 . 6 HTLV-1 producer cell lines were collected and ultra-centrifuged at 23000 rpm for two hours and thirty minutes at 4°C to concentrate the virus . Pellets were suspended in PBS and p19Gag measured by ELISA assay ( ZeptoMetrix , Buffalo , NY ) . Equivalent amounts of p19Gag were used for infection of THP-1 cells . Briefly , THP-1 cells were suspended in virus preparations and centrifuged at 3000 rpm for one hour at room temperature in the presence of 8 µg/ml polybrene ( Sigma , St . Louis , MO ) . Cultures were maintained in RPMI 1640 , 10% FBS , with 50 µM β-mercaptoethanol and p19Gag production monitored . BHK1E6 cells ( 1×105 ) containing a lacZ reporter gene downstream of the , LTR promoter [23] , were co-cultured for 48 hours with either control un-infected cells or HTLV-1 producers D26 , N26 , G29S and 12KO ( 1×106 ) . Monolayers were washed twice with PBS to remove medium and suspension cells and assayed using a β-galactosidase Staining Kit according to manufacturer's instructions ( Active Motif , Carlsbad , CA ) . The β-galactosidase expressing cells were counted by brightfield microscopy . The macaques used in this study were male colony-bred Indian Rhesus Macaques ( RMs ) obtained from Covance Research Products . The animals were housed , feed , given environmental enrichment and handled in accordance with the standards of the Association for the Assessment and Accreditation of Laboratory Animal Care International . Appropriate steps were taken to minimize suffering in accordance with the Weatherall report ( “The use of non-human primates in research” ) . The care and use of the animals were in compliance with all relevant institutional ( National Institutes of Health ) guidelines . All macaques were 2–3 years of age and seronegative for simian T-cell lymphotropic virus 1 and simian immunodeficiency virus at the initiation of the study . RMs were inoculated with lethally irradiated 729 . 6 producer cells . Supernatant p19 levels were measured prior to inoculation and cell numbers were adjusted to give equivalent amounts of p19Gag per animal ( Supplementary Table S1 ) . Four macaques each were used for the N26 and D26 viruses and eight each for the G29S virus . One macaque was infused with irradiated control parental 729 . 6 cells . All macaques received an equivalent dose of virus based on p19Gag expression levels per million cells , as previously described [18] . Mononuclear cells were separated from whole blood specimens by density gradient centrifugation ( Ficoll ) . PBMCs ( 5×106 ) from animals positive for proviruses were washed in PBS and DNA isolated using the Genomic DNA Wizard kit as described by the manufacturer ( Promega , Milwaukee , WI ) . Reactivity to specific viral antigens in the sera of infected animals was detected with the use of a commercial HTLV-1 western immunoblot assay ( GeneLabs Diagnostics , Redwood City , CA ) . Quantitative Real-time PCR analysis was performed as described previously [18] . Proviral loads were normalized to the macaque albumin gene and expressed as the number of HTLV-1 proviral DNA copies per 106 PBMCs . The limit of detection for the PCR assay is one copy in ten thousand cells . DNA sequencing of the orf-I genes was performed from ex vivo samples of macaque cellular DNA to check for reversions of point mutations to wild-type . After genomic DNA was isolated from 5×106 PBMCs , PCR was performed using primers p12-Fwd , 5′-CACCTCGCCTCCCAACTG-3′ and p30-Rev , 5′-GGAGTATTTGCG- CATGGCC-3′ which amplified the fragment ( 871 nucleotides ) spanning positions 6414 to 7285 of the HTLV-1 genome . The PCR amplicon was cloned into the pCR4TOPO vector ( Invitrogen , Carlsbad , CA ) using the manufacturer's protocol and 10 unique colonies sequenced . The level of orf-I mRNA was quantified using splice site-specific quantitative RT-PCR ( qRT-PCR ) as described [45] . After isolation of total cellular RNA , RT-PCR was performed using primers: p12-1B , 5′-GTCCGCCGTCTAG∧CACTATG-3′; p12 reverse , 5′-GGAGGAAGCAGGAAGAGC-3′; probe , MP-1 5′ ( FAM ) -TTCGCCTTCTCAGCCCCTTGTCT-3′ ( TAMRA ) . All samples were normalized to Gapdh mRNA copy number . Surface staining for CD4+ T-cells was performed for 30 minutes at room temperature with antibodies to CD4 and HLA . A2 from BD Biosciences ( San Jose , CA ) . All cells were fixed with 1% paraformaldehyde and at least 50 , 000 events acquired on an LSRII ( BD Bioscience , San Jose , CA ) . Data analysis was performed with FlowJo 9 . 4 software ( Tree Star Inc . , Ashland , OR ) . THP-1 staining was performed with antibodies to CD14 , CD83 , HLA-DR from Biolegend ( San Diego , CA ) and CD80/CD86 and CCR7 from BD Bioscience ( San Jose , CA ) . The cytolytic activity against target cells was assayed using previously characterized HTLV-1 Tax11-19 ( LLFGYPVYV ) -specific CD8+ CTL clone [29] . The CTL clone was maintained by weekly stimulation with peptide-pulsed ( 1 mg/ml ) irradiated PBMCs from an HLA-A201+ non-HTLV-1 infected individual . CTL culture medium was IMDM containing 10% human serum , 2 mM L-glutamine , 100 U/ml penicillin and 100 µg/ml streptomycin . Human recombinant interleukin-2 ( 50 U/ml ) was added on the next day of stimulation . Target cells were CD4+ T-cells infected with HTLV-1 D26 , N26 , G29S or 12KO and autologous Epstein-Barr virus-transformed B-cells as a positive control . The cytotoxicity assay was performed using DELFIA EuTDA Cytotoxicity assay ( Perkin Elmer ) . Target T-cells were loaded with bis ( acetoxymethyl ) 2 , 2′:6′ , 2″-terpyridine-6 , 6″-dicarboxylase ( BATDA ) and pulsed with or without 100 ng/ml of Tax peptide . Target cells ( 3×103 ) were incubated with CTL clones for 3 hours at 37°C in 96 well plates at indicated effector-to-target ratios . The supernatant ( 20 µl ) was incubated with 200 µl of Europium solution , and the fluorescence was measured in a fluorometer ( Wallac 1420 VICTOR3; Perkin Elmer ) . The percent specific lysis was calculated as ( experimental release−spontaneous release ) / ( maximum release−spontaneous release ) ×100 . The assay was performed in triplicate . | HTLV-1 persists despite a vigorous host immune response . We found that polymorphism of HTLV-1 orf-I alter the relative amounts of the p12 precursor and its cleavage product p8 , and is associated with differences in blood virus levels in humans , a correlate of disease risk . Reverse genetics in 160 HTLV-1 infected individuals demonstrated that equivalent levels of p8 and p12 are associated with high virus levels and , accordingly , genetically engineered HTLV-1s that express either predominantly p12 or p8 are poorly infectious in macaques . We found that expression of p8 is sufficient for productive infection of monocytes . Expression of either p12 alone or p8 alone is insufficient to protect infected cells from MHC-class-I restricted CTL killing . However , the balanced expression of both provides resistance of infected cells to CTL killing . Together , our findings provide the rationale to explore novel approaches to target the cleavage of the p12 protein , an essential step for viral infectivity and persistence . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases",
"biology",
"and",
"life",
"sciences",
"immunology",
"medicine",
"and",
"health",
"sciences"
] | 2014 | Co-dependence of HTLV-1 p12 and p8 Functions in Virus Persistence |
Olfactory receptors ( ORs ) , which are involved in odorant recognition , form the largest mammalian protein superfamily . The genomic content of OR genes is considerably reduced in humans , as reflected by the relatively small repertoire size and the high fraction ( ∼55% ) of human pseudogenes . Since several recent low-resolution surveys suggested that OR genomic loci are frequently affected by copy-number variants ( CNVs ) , we hypothesized that CNVs may play an important role in the evolution of the human olfactory repertoire . We used high-resolution oligonucleotide tiling microarrays to detect CNVs across 851 OR gene and pseudogene loci . Examining genomic DNA from 25 individuals with ancestry from three populations , we identified 93 OR gene loci and 151 pseudogene loci affected by CNVs , generating a mosaic of OR dosages across persons . Our data suggest that ∼50% of the CNVs involve more than one OR , with the largest CNV spanning 11 loci . In contrast to earlier reports , we observe that CNVs are more frequent among OR pseudogenes than among intact genes , presumably due to both selective constraints and CNV formation biases . Furthermore , our results show an enrichment of CNVs among ORs with a close human paralog or lacking a one-to-one ortholog in chimpanzee . Interestingly , among the latter we observed an enrichment in CNV losses over gains , a finding potentially related to the known diminution of the human OR repertoire . Quantitative PCR experiments performed for 122 sampled ORs agreed well with the microarray results and uncovered 23 additional CNVs . Importantly , these experiments allowed us to uncover nine common deletion alleles that affect 15 OR genes and five pseudogenes . Comparison to the chimpanzee reference genome revealed that all of the deletion alleles are human derived , therefore indicating a profound effect of human-specific deletions on the individual OR gene content . Furthermore , these deletion alleles may be used in future genetic association studies of olfactory inter-individual differences .
Olfaction , the sense of smell , is characterized by the remarkable ability to detect and discriminate millions of odorous compounds . It contributes significantly to our perception of the environment and to our quality of life [1] . At the molecular level , olfaction is mediated by a conserved signal transduction cascade , which is initiated by the binding of odorants to specific G-protein coupled receptors , known as olfactory receptors ( ORs ) [2] , [3] . The human OR repertoire is comprised of 851 genes and pseudogenes , organized in clusters on almost every chromosome [4] . During human evolution these gene clusters underwent dynamic processes of expansion , diversification and duplication as well as diminution and pseudogenization [5] , processes which may be still ongoing . Since members of the human species depend on their sense of smell to a lesser degree than other mammals , their OR repertoire has undergone an accelerated process of pseudogenization , resulting in functional inactivation of more than 50% of the ORs by frame-disrupting mutations . Furthermore , the lower degree of purifying selection in human olfaction also appears to result in enhanced inter-individual genome diversity , e . g . the prevalence of high-frequency inactivating single nucleotide polymorphisms ( SNPs ) , i . e . segregating pseudogenes [6] . Thus , the human OR repertoire serves as an interesting model for genome evolution and variability . Human olfactory perception exhibits considerable phenotypic variation , for which a genetic basis has been proposed [7]–[10] . Recently two studies associated common SNPs affecting OR genes OR7D4 and OR11H7P to human sensitivity and perception of the respective odorants androstenone and isovaleric acid [11] , [12] . These findings support a relationship between genotypic and phenotypic variability in human olfaction . However , recent results indicate that most of the variation in the human genome is not accounted for by SNPs . Genome structural variants , usually defined as kilobase ( kb ) to megabase ( Mb ) deletions , duplications , insertions , and inversions , emerge as responsible for the majority of variable base-pairs among individuals [13]–[15] and may be a main basis for phenotypic differences [16] . Structural variants affecting the copy-number of a ≥1 kb genomic region ( e . g . through deletion or duplication ) are widely referred to as Copy Number Variants ( CNVs; [17] ) . Several genome wide surveys have recently reported a prevalence of OR genes among CNVs [13] , [14] , [18] , [19] , suggesting an impact of CNVs on the individual OR gene content . Moreover , recent reviews have suggested that CNVs have an impact on the genome in longer evolutionary terms by facilitating the expansion or diminution of gene families [20] , [21] . However , these genome-wide surveys did not focus on ORs specifically , and were either carried out at low-resolution , suitable for genomic clusters of ORs rather than for single OR loci , or considered few ( i . e . , 8 ) individuals . A gene-level resolution survey of copy-number variability in the largest mammalian gene superfamily should thus serve as an excellent model for studying long-term effects of CNVs on the genome . Furthermore , such a survey may help identifying candidates for examining potential phenotypic consequences of CNVs on smell perception . Here , we report a high-resolution analysis of CNVs affecting the human OR repertoire . We use custom high-resolution oligonucleotide tiling microarrays to study OR gene copy-number variation in 25 individuals with ancestry in three human populations , and report that about a fourth of all OR genes are commonly affected by CNVs . In addition , we find that CNVs affecting ORs are more complex than previously appreciated . Furthermore , evolutionarily “young” ORs , as well as OR pseudogenes , show a strong tendency for copy-number variation . In addition , using confirmatory quantitative PCR ( qPCR ) , we report a total of 15 OR gene loci with an appreciable prevalence of homozygous deletion genotypes in humans . These deletions may provide an additional , CNV-based , genotypic basis for variations in human olfactory perception .
Analyzing the microarray results we identified OR loci with median normalized microarray log2-intensity ratios R that significantly deviated from expected measures , and were thus scored as CNVs ( see Methods ) . Altogether , we observed 1301 CNV events ( gains and losses ) affecting loci harboring intact OR genes and pseudogenes in the 25 individuals analyzed ( Figure 1A ) . Notably , due to the comparative nature of our analysis , this results in 24 , rather than 25 , sets of CNVs . Of the 851 OR loci interrogated , 244 ORs ( 28% , including 93 intact genes and 151 pseudogenes ) were observed to be affected by CNVs ( Tables 1 , S1 ) . Next , we examined to what extent CNVs affect the individual genomic OR content . On average , we observed CNV events in 22 OR genes and 32 OR pseudogenes per sample ( mean number of CNV events per sample; see Methods ) . The vast majority of these probably represent true positives , as we generally estimate a false-positive rate of less than 4% for CNV assignments based on a control ( self-vs . -self hybridization; see Methods ) . Interestingly , meta-analysis of lower resolution data performed by Nozawa et al . [19] indicated a mean difference of only 10 . 9 intact OR genes and 11 . 3 OR pseudogenes between individuals . Our considerably higher numbers are presumably explained by the fact that many CNVs detected in our study are below the resolution of the CNV-detection methods that were available to Nozawa et al . ( see e . g . discussions in recent studies using high-resolution mapping approaches [14] , [15] , [29] ) . A more detailed view of the microarray results was generated for two subsets of the OR repertoire . One ( Figure 1B , left panel ) encompasses the OR gene subset for which qPCR has been performed . The second , an even stronger , zoom view is afforded by the left panel of Figure 1C , where the most variable 25 loci , with 56 OR genes , are shown . Both views highlight the fact that CNVs create unique mosaics of OR dosages across individuals that might contribute to the functional individuality of the human nose . Furthermore , the qPCR results led us to detect 23 additional CNVs that were not detected by the microarrays and were added to our map . Fifteen of these intersect with CNVs reported in the Database of Genomic Variants ( DGV; http://projects . tcag . ca/variation/; see Table 2 ) , which corroborates our qPCR results . Such a false-negative rate was not unexpected for the array-based CNV-calls , since microarrays have been reported to miss some true-positive CNVs [13]–[14] . The high resolution of our approach also allowed us to more precisely predict the location of CNV breakpoints from microarrays . To this end we used the microarray-data based BreakPtr algorithm [23] to identify instances where the breakpoints of CNVs are located within an OR locus ( see Methods ) . Disrupted/split loci were detected for 88 ORs , suggesting a surprising complexity of CNV genotypes ( Table S3 ) . For eight of these loci our data suggest recurrent CNV formation events , since for different samples distinct breakpoint locations were predicted ( Table S3 ) . While most CNV breakpoints do not disrupt the ∼900 bp OR coding regions , the BreakPtr algorithm predicts disruption of coding regions of four specific OR loci ( OR4C16 , OR1M1 , OR6B2 , and OR4N2; see Table S3 ) . Although further experiments will be necessary to validate these predicted coding region disruptions , additional sequence analyses described below indicate that OR coding sequence may be directly involved in CNV formation , e . g . leading to novel OR fusion genes . We used three different methods for CNV validation , i . e . ( i ) comparing our CNV-calls to previously reported CNVs from DGV [30] , ( ii ) performing quantitative PCR ( qPCR ) , and ( iii ) performing conventional PCR . Initially , when comparing our results to DGV , we found that in total , 134 ( 55% ) copy-number variable OR loci intersected with previously described CNVs ( Table 1 ) . Our capacity to discover a large number of new variable loci , despite the fact that relatively few individuals were examined when compared to the multiple studies contributing to DGV , stresses the sensitivity and overall value of our high-resolution CNV map . Subsequently , we validated CNVs by performing qPCR on 122 OR loci ( 104 ORgenes and 18 pseudogenes ) , which were selected to represent both high and low variability loci ( Figure 2 , S2 ) and which included similar amounts of novel and already known CNVs . Sixty validation experiments were carried out with a panel of only 13 samples , while the other 62 experiments were performed in 23 samples . We initially compared individual microarray intensities ( normalized log2-intensity ratios ) to qPCR outcomes ( normalized Cp-values ) . In general , qPCR results revealed an acceptable correlation to the normalized microarray values ( Figure S3A , S3B ) , with the exception of a small group of 23 OR loci that displayed considerable qPCR inter-subject variability despite low-variability microarray values ( Figure S3A , S3B ) . We added these 23 cases to our list of copy-number variable OR loci ( Table 2; Figure S4 ) . Next , we examined the rate at which CNV calls were validated by qPCR ( qPCR results were therefore normalized relative to reference individual NA19154 ) : 87% of all CNVs tested yielded negatively correlating Cp-values , as expected , indicating successful validation , with similar success rates for gains and losses . In the majority of cases , rounded absolute qPCR-measures were equal to Cp = 1 , in line with an abundance of simple gains and losses ( i . e . heterozygous deletions and duplications ) . Taken together , most CNV calls can be validated by qPCR , demonstrating a very reasonable specificity of our platform . Finally , for two cases ( deletions I and II , Table 3 ) , we validated qPCR results that indicated homozygous deletions , i . e . such that consistently failed in specific DNA samples , by conventional PCR with an additional set of primers ( Figure S5 ) . For both deletion alleles , two positive and two negative samples were tested . Standard PCR confirmed our results in both cases , validating the presence of homozygous deletion I in NA12003 and NA12246 , and homozygous deletion II in NA19103 and NA19141 . CNVs affecting OR loci are strongly non-uniformly distributed , causing a clustering of CNVs into ‘hotspots’ , and an associated clustering of regions with high variance in normalized microarray intensities ( see Figures 1A , 2 ) . Accordingly , there is a strong correlation ( 0 . 8 , P-value = 10−18 ) between the variance ( based on measure R ) of a variable OR to the variance of adjacently located ORs ( Figure S6A ) . Some of the clustering can be explained with an enrichment of CNVs near telomeres and centromeres ( Figure S6B ) , consistent with observations from a previous report [13] . Genomic CNV formation biases are a plausible explanation for the clustering . Furthermore , CNV size ( e . g . CNVs affecting multiple ORs ) and biases in evolutionary selection are likely also responsible for the observed non-uniform distribution in our data ( see below ) . We analyzed apparent CNV ‘hotspots’ in more detail , and observed that the 244 variable OR loci were clustered into 149 extended genomic blocks displaying a common pattern of variability ( Figure S7 ) . To examine whether observed blocks of high copy-number variability actually behave as single ( large ) CNVs we performed an analysis addressing the correlation of cross-individual patterns within each of these blocks ( Figure S6C ) . Indeed , the analysis identified 31 copy-number blocks containing 105 OR loci with a Pearson correlation coefficient >0 . 8 in cross-individual normalized microarray measures R , indicating that the blocks likely correspond to a single , large CNV . The largest such block is located on chromosome 14 and contains 11 OR loci . On the other hand , this suggests that many of the remaining candidate hotspots may underlie several independent ( i . e . smaller ) CNVs occurring in close proximity . However , although large CNVs ( such that span more than one OR locus ) are relatively rare ( involving 43 out of OR 149 loci only ) , they contribute to our inter-individual variability to approximately the same extent as small CNVs . Moreover , the multi-locus CNVs are important from a functional aspect , as large CNVs that affect several ORs simultaneously are more likely to result in detectable phenotypes than small ones . Our high-resolution map of CNVs affecting OR genes further enabled us to address questions relating to the evolution of the OR repertoire . We first tested whether the evolutionary age of an OR gene is correlated with its propensity to be affected by a CNV . Interestingly , we found that ORs with a closely related paralog in the human genome , evaluated using the level of sequence identity as a measure , are significantly more likely to be affected by CNVs than ORs lacking a closely related paralog ( Figure 3 ) . In other words , evolutionarily “younger” ORs tend to be more frequently affected by CNVs than more “ancient” ORs . To confirm this trend using a different approach to classify OR genes into “young” and “ancient” we used one-to-one orthology relationships with the chimpanzee genome and categorized OR genes into “young” if they lacked a one-to-one ortholog and “ancient” in the case of unambiguous orthology: indeed , we found that OR genes that recently exhibited a duplication/loss event in the human , or the chimpanzee genome ( i . e . such genes lacking one-to-one orthologs between human and chimp ) , are significantly more likely to be affected by CNVs than OR genes with unambiguous one-to-one orthologs ( Pvalue<0 . 001 , Figure 4A ) . We note that these findings are compatible with a suggested general model of copy-number variation as an evolutionary basis of paralog birth [20] , [21] , whereby novel paralogs , manifested as CNVs , may later become fixed in the population . We further note that the enrichment of CNVs in “younger” OR genes may be due to two reasons: First , the higher number of CNVs observed in ”younger” ORs may be due to decreased selective pressures in “young” ORs compared to “ancient” ORs ( see below ) . Second , different CNV formation biases acting on genomic regions harboring “young” and “ancient” ORs may be responsible . We first evaluated the impact of CNV formation bias through testing whether pairs of tandemly oriented segmental duplications ( SDs; [31] ) – known mediators of CNV de novo formation through induction of non-allelic homologous recombination ( NAHR ) – are enriched among the ORs affected by CNVs . When performed at single gene resolution , this analysis yielded inconclusive results , which may be due to a confounding positional bias caused by the fact that neighboring OR loci are often located in close vicinity in the genome , forming genomic clusters ( see Text S1 ) . We thus examined the potential impact of NAHR also at the level of 135 genomic clusters of OR loci , which are listed in the HORDE database ( http://bioportal . weizmann . ac . il/HORDE/index . html ) , and observed a significant and robust enrichment ( with P-values<0 . 01 ) of SD-pairs among highly copy-number variable clusters compared to clusters for which no single CNV has been observed in our study ( see Text S1 for details and Figure S8 ) . Thus , our data indicate that NAHR has likely biased the distribution of CNVs within the OR repertoire . Next , we tested whether , and to what extent , evolutionary selection may influence CNVs affecting the OR repertoire . In this regard , we first revisited the recent report that OR pseudogenes are equally likely to be affected by CNVs as OR genes , a finding suggesting a neutral evolution of OR copy number variation [19] . In particular , we compared CNVs in OR genes and pseudogenes , and found that OR pseudogenes are significantly more likely to be affected by CNVs than OR genes ( Table 1 , Pvalue = 0 . 007 , χ2 statistic = 7 . 38 , DF = 1 ) . A consistent and significant signal was also observed when carrying out the analysis at the level of microarray probes , either only for the coding regions or for the adjacent non-coding regions ( see supplementary Text S1 ) . This may indicate that stronger evolutionary constraints act on OR genes than on pseudogenes . Additionally , we analyzed the frequency of CNVs among OR genes and pseudogenes separately for the “young” and “ancient” ORs ( Figure 4B ) . This analysis showed that the difference between “young” and “ancient” ORs is greater than the difference between OR genes and pseudogenes . Nonetheless , in both groups OR genes were less prone to CNV formation than pseudogenes , indicating a mixed contribution of formation bias and selection ( see below ) . Having established that “young” ORs are more prone to CNVs than the “ancient” ones , we further addressed which types of CNVs , i . e . gains or losses , are most common in the two OR groups . Interestingly , for both “ancient” and “young” ORs we observed significant imbalances between gains and losses ( Figure 5A , Pvalue = 0 . 006 , χ2 statistic = 10 . 3 , DF = 2 and Pvalue = 0 , χ2 statistic = 43 . 8 , DF = 2 , respectively ) . Notably , the two groups exhibited an opposite over-all trend , i . e . “ancient” ORs displayed significantly more gains than losses and “young” ORs showed significantly more losses than gains . As we arbitrarily picked a reference individual in our study , we also tested whether the observed trend was robust if neglecting the reference , by looking for ORs that exhibited only one type of CNV – gain or loss – in at least 50% of the samples ( Text S1 material ) . Although this analysis revealed that some of the reported events are likely attributable to rare alleles in the reference individual , the trend of opposite balances between gains and losses in the two groups remained significant ( Pvalue = 0 . 004 , χ2 statistic = 11 . 1 , DF = 2 ) . CNVs affecting OR loci may encompass deletions , duplications or more complex allelic structures . In this realm , homozygosity for deletion alleles is the CNV-related genotype that is most likely to cause a direct olfactory phenotypic effect . This is because , with some exceptions , OR gene disruption is likely to be recessive , i . e . only absence of the functional gene product will result in an observable phenotype . In particular , a dosage effect stemming from a different number of active OR copies is not expected due to clonal exclusion , whereby only one paralog is expressed in any given sensory neuron [32] , [33] . Thus , we particularly focused on the identification of frequent deletion alleles , which may lead to common differences in smell perception between individuals . While carrying out qPCR-based validation experiments , for several OR loci we immediately identified samples that consistently failed to amplify during qPCR , suggesting a potential homozygous deletion of the OR locus ( Table 3 ) . For example , two samples ( NA12003 and NA12246 ) consistently failed to amplify neighboring ORs from a specific locus on chromosome 11 ( Figure S5 ) . In particular , this variation was found to harbor a genomic region of 82 kb involving 4 OR genes and 2 OR pseudogenes . The distribution of median normalized microarray intensity log2-ratios ( measure R ) of OR loci in this region ( Figure 5 ) is consistent with a bi-allelic CNV and with dosage levels corresponding to 0 , 1 , or 2 copies . Furthermore , PCR with uniquely designed primers for ORs in that region yielded no product for NA12003 and NA12246 and bands of expected size in positive controls ( NA12004 and NA19141 ) , thus supporting both the microarray and the qPCR results ( Figure S5 ) . We note that a subset of common OR deletions may not be tractable with the approach presented here , as they were deleted from the individual ( s ) contributing to the human reference genome , which we used to design our high-resolution arrays . Thus , to identify additional potential deletions affecting OR genes , we further searched for discrepancies between the human genome reference assembly and other human genome sequencing projects . In particular , we searched for OR genes that reside in regions of disagreement between the reference genome and publicly available data from the initial Celera human genome assembly [34] , Craig Venter's genome [28] , James Watson's genome [35] , and fosmid clones which are not part of the current reference assembly . Altogether , we identified three OR genes , OR8U8 , OR8U9 and OR9G9 , which are not present in the current reference assembly . The first two present an interesting case , whereby two intact OR genes ( OR8U8 and OR8U9 ) appear to have recently fused through NAHR , which led to the formation of a chimeric OR gene ( OR8U1; Figure S9 ) . Moreover , sequence comparison between OR8U8 , OR8U9 , and OR8U1 identified a plausible NAHR recombination region , i . e . a 119 bp interval ( bp 472–590 in Figure S9 ) , which displays 100% DNA sequence identity between the three genes . qPCR with allele specific primers , designed at the interface of the recombination region , identified 4 individuals heterozygous for the deletion ( Table 3 ) . A similar case , where the recombination between OR51A2 and OR51A4 results in a new gene encoding an amino acid sequence identical to OR51A4 and upstream regions from the OR51A2 gene was reported previously [14] . Together , these observations indicate that ORs themselves , rather than their genomic environment , may frequently promote CNV formation through NAHR . In the third case , OR9G9 , although absent from the reference genome , was identified in all individuals tested by qPCR . However , its closest paralog – OR9G1 was homozygously deleted in 3 out of 23 individuals tested . Altogether , fosmids and the individual genomes of Craig Venter and James Watson supplied additional support for 3 deletion loci ( deletions I , II and VI; Table 3 ) identified by qPCR ( Table 3 ) . We further followed up 9 regions , encompassing 15 OR genes , for which deletion alleles were initially identified by qPCR . For 8 out of the 9 deletions we estimated deletion allele frequencies from the number of homozygously deleted individuals in our cohort at approximately 0 . 2 up to 0 . 6 ( Table 3 ) . Notably , deletion VII appears only in the Yoruban individuals , which may suggest population-specificity of this allele . Yet , this bias should be interpreted with caution , as the number of samples we employed from each population is insufficient for meaningful comparison between the populations . Examination of the chimpanzee reference genome ( see Methods ) shows that all deletions identified in this study are likely to represent human derived alleles ( Table 3 ) , barring presumably infrequent shared polymorphisms . A recent survey [15] on genomic structural variation in different individuals , which identified deletions with an initial resolution in breakpoint assignment of 20–40 kb ( and base-pair resolution for several cases in which breakpoint junctions were sequenced ) , reported variants that are consistent with all deletion alleles reported in our study . This validates our findings and confirms the frequent occurrence of these variants; thus , a considerable number of OR genes are frequently deleted in humans , with homozygous alleles occurring commonly in the population .
We have carried out a high-resolution analysis of CNVs affecting OR gene and pseudogene loci , and identified many OR-related CNVs for which copy-number variability has not been reported previously . In contrast to a previous low-resolution study [19] we observe that CNVs are enriched among evolutionary young ORs as well as pseudogenes . Our results differ from those published by Nozawa et al . , probably due to the significant increase in resolution ( nearly 2 orders of magnitude , from 50–100 kb to <1 kb ) , which enabled us to focus on single loci , rather than clusters , and thus to distinguish unaffected OR genes ( non-CNVs ) from adjacently located affected OR pseudogenes ( CNVs ) . Our analysis suggests that both formation bias and evolutionary constraints have likely shaped the distribution of CNVs in the human OR repertoire . In fact , both biases are difficult to distinguish . For instance , pseudogenes and other repeats are known to be enriched in vertebrate gene deserts [36] , which is presumably both due to mutational and selective biases . Also , we defined OR pseudogenes based on the absence of an intact open reading frame in the human reference genome . This may lead to misclassification of some of the intact genes , which may not be functional in reality , due to missense mutations [37] or mutations affecting non-coding regulatory elements . However , both of these confounding factors are presumably affecting only a minority of loci and thus are unlikely to influence our conclusions in relation to a depletion of CNVs affecting OR genes relative to OR pseudogenes . Furthermore , our data showed a bias for CNV-enriched OR loci to be located between tandemly oriented segmental duplications , which are known to induce NAHR [38] . Besides NAHR , other formation-processes such as non-homologous end-joining ( NHEJ; [39] ) are likely to play a role in the genesis of CNVs affecting OR loci . In the future , determining the relative contribution of such mutational processes will require the identification of the breakpoint junction sequences of numerous CNVs . In addition to regional biases caused by different mutational processes involved in CNV formation , large CNVs may sometimes span both genes and adjacently located unprocessed pseudogenes . Unprocessed pseudogenes often arise through tandem duplication of OR loci followed by inactivation of the newly generated , neutrally evolving , paralogous loci . This may lead to biases in the frequency at which pseudogenes vary in terms of copy-number , depending on selective constraints acting on adjacent functional loci . Finally , even in the absence of large CNVs such a bias may occur , as CNVs affecting pseudogenes in the proximity of genes may be detrimental due to long-range regulatory effects ( e . g . , through interfering with non-coding regulatory elements ) . Furthermore , the enrichment of CNVs among ORs located in close proximity to telomeres/centromeres may also be reflective of CNV formation biases or selective biases . In this regard , human subtelomeric regions are enriched for segmental duplications , and NHEJ and NAHR presumably operate efficiently in those regions [39] . At least for some cases , we were already able to present sequence-based evidence for the likely involvement of NAHR in CNV formation . In particular , we demonstrated that CNVs causing a fusion of tandemly oriented OR genes were presumably formed through NAHR ( Figure S9 and [14] ) . Such events exemplify a potential mechanism for accelerated functional diversification of ORs , where paralogs are originally created with new function or regulation pattern , rather than through the process of sequential duplication and diversion . Consequently it may be hypothesized that large OR subfamilies came to existence through frequent and/or large duplication events , implying that genes from large OR subfamilies will be prone to reside in CNV loci . However , our data showed no obvious correlation between OR subfamily size and averaged variances of R ( Figure S10 ) . This may partly be due to a confounding factor , namely the reduced sensitivity of microarrays in detecting CNVs within regions sharing very high sequence similarity ( see additional discussion bellow ) ; such regions are enriched in the largest OR subfamilies , and our microarrays may have failed to detect CNVs in these . As discussed above , our high-resolution data helped to clarify that CNVs do not randomly affect genes and pseudogenes , and that for OR genes purifying selection may operate on top of formation biases . In evolutionary terms , CNVs , which are variants en route to fixation , have good potential to influence the OR repertoire size . Here , we have presented evidence for an abundance of polymorphic gene loss events affecting the most copy-number variable group of ORs , i . e . a group classified here as evolutionarily “young” . This may point to one possible underlying mechanism for the well-documented diminution of the human OR repertoire as it is reflected in the considerably reduced human OR repertoire size ( i . e . 851 ORs ) compared with dog and chimpanzee ( ∼1000 ORs ) , and with rat and mouse ( ∼1400 ORs ) ( [5] and references therein ) . It should be noted , however , that although these ORs were herein classified as “young” for simplicity , they do not necessarily have to represent recent gene duplicates . In particular , due to the orthology assignments , ORs that underwent deletion or duplication in the chimpanzee genome are also classified as “young” in our study . In contrast , the more “ancient” ORs potentially provide a more stable backbone of the olfactory subgenome , which is less affected by CNVs and also appears to have an overall positive balance between gains and losses . This slight enrichment for gains may imply stronger evolutionary constraints acting on these ORs , as losses are thought to be more detrimental than gains , and genes under purifying selection are more biased away from deletions than from duplications [13] . The identification of 9 deletion alleles , encompassing 15 OR gene loci and present at appreciable frequency , is significant for studies of olfactory function . Previously , functional OR gene inactivation alleles , involving SNPs leading to in-frame stop codons or substitution of conserved amino acids in an otherwise unmodified OR locus sequence , have been reported [6] , [37] . Such alleles were subsequently linked to individual human responses to specific odorants , using both in-vitro and association study approaches [12] . However , large deletions have not so far been reported among the variants used for genetic association studies . The present identification of a number of unexpectedly frequent deletion alleles ( with deletion allele frequencies of up to 0 . 6 ) , some of which encompass several genes from the same OR subfamily , thus provides additional strong candidates for genetic association studies of human olfaction . To this end , we have recently initiated a CNV genotyping experiment using qPCR for the herein reported deletion I ( Table 3 ) against a Caucasian cohort [40] of 94 subjects , phenotyped for olfactory acuity towards eight odorants ( unpublished ) . The results were inconclusive , probably due to the low number of samples and odorants involved . Future association studies will require larger sample sizes and , ideally , a-priori in-vitro assessment of ligand specificity for the affected ORs . Our study also has certain technical limitations . First , while microarrays represent the most cost-effective method for studying CNVs at large scale and high resolution , cross-hybridization limits their specificity and sensitivity in repetitive genomic regions . Cross-hybridization results in averaging of the signal over several loci , and is thus more likely to lead to false negative than to false positive CNV calls when using a stringent cutoff for scoring the arrays . A second potentially confounding factor is the inter-individual sequence variability , i . e . SNPs and small indels , that may affect probe hybridization . Yet , different probes on our arrays are generally ≫1 bp apart from each other , there are dozens of probes for each locus , and the signal is analyzed over all probes mapping to an OR locus . Thus , inter-individual sequence variability in specific probes , typically at the level of 1 SNP per kilobase , is unlikely to considerably affect our CNV calls . Furthermore , our qPCR results indicate that the false-positive rate is relatively low in our microarray experiments , as opposed to a considerable false negative rate , which was expected due to the stringent cutoff applied for scoring the arrays . Third , the comparative nature of our analysis may introduce an overestimation of the frequencies of some CNVs , if the reference sample carries a rare allele . In such cases , the rest of the samples are expected to show only one type of change – gain or loss , across a majority of the samples . Importantly , this would not change the CNV status of the locus ( i . e . whether the OR is considered to be copy-number variable or not ) , and thus did not affect the main conclusions of this study . We nevertheless specifically addressed this issue by calling CNVs independent from the reference individual ( see Text S1 ) , an analysis that did not considerably affect our overall CNV counts and did not change the conclusions of our study . Fourth , a considerable fraction of CNVs may represent recurrent , rather than common variants emerging from single mutational events ( Table S3 ) . Distinguishing recurrent from common CNVs coherently will become a challenging task that will require breakpoint-resolution data , which is currently available only for few CNVs affecting ORs . Finally , a large portion of CNVs ( 62% , Table 1 ) reported in DGV to intersect with OR loci are not observed in our study . This is likely to be , in part , attributed to the relatively low number of samples we analyzed and to false-negative calls in our study , but also to the fact that for most CNVs listed in DGV the size-ranges have been overestimated ( in this regard , note the excellent recent survey published in [29] ) . Furthermore , a parallel survey of CNVs affecting functional OR loci was published while the present paper has been under review [41] . In this study , the authors report a statistical analysis of a subset of CNVs listed in DGV , as well as an experimental validation of CNVs recorded in DGV , which affect a set of 37 OR loci . In agreement with our data , they failed to validate 16 out of the 37 CNVs tested , despite using 50 samples of diverse ancestry . Altogether , these results support the size over-estimation of previous CNV surveys at low resolution ( [29] , [41] ) and stress the relevance of systematic follow-up studies focusing on CNV subsets . In conclusion , our results emphasize the importance of carrying out genome-wide CNV surveys at high resolution . This is especially important , if one aims to identify events relevant to association studies , which requires the delineation of CNV event nature ( i . e . deletion/duplication or complex , common or recurrent ) , exact CNV boundaries , and CNV population frequencies . Thus our study both provides insights into the evolution of the largest human gene family , and suggests specific targets for subsequent association studies .
To determine copy-number variation at OR loci genome-wide and at exon-level resolution , custom high-resolution oligonucleotide tiling microarrays were designed using Nimblegen/Roche technology . The microarrays contained 71 , 980 different oligonucleotide probes of length 45–85 bp , adjusted in C+G content to yield similar optimum hybridization temperatures [22] . Further , to save experimental costs and make the platform more efficient , multiplex ( 4-plex ) microarrays were used . We designed the probes to cover all 851 OR loci represented in the reference genome ( build35 ) , i . e . 388 OR gene loci and 463 OR pseudogene loci including the respective 5′-regions ( i . e . , 10–20 kb upstream of genes and pseudogenes ) and 3′-regions ( 2 kb downstream of genes/pseudogenes; depiction of probe locations for an example OR locus are presented in Figure S1 ) . Before assigning oligonucleotide probes to loci , we separated the frequently partially overlapping OR loci within genomic gene clusters to avoid assigning probes to multiple ORs loci . Overlapping loci were separated according to the orientation of the OR open reading frames they contained . If both ORs were oriented 5′ to 5′ or 3′ to 3′ , the region between them was divided equally between the 2 loci . However , if the orientation was 5′ to 3′ , the separation was such that 15% of the region ( not exceeding 2 kb ) was assigned to the 3′ of one of the OR loci , and 85% were assigned to the 5′ of the other OR locus . Genomic coordinates of OR genes and pseudogenes were obtained from the HORDE database ( http://bip . weizmann . ac . il/HORDE/ ) . A median genomic distance between interrogated oligonucleotides of 148 bp was used on the arrays , a density expected to allow the identification of breakpoints of CNVs at ∼500 bp resolution [23] . Our custom microarray design approach focused on optimum oligonucleotide probes and excluded highly repetitive regions . In particular , ( i ) oligonucleotides of length 85 bp were extracted from the human reference genome; each oligo was truncated according to its GC content [22] and subsequently aligned against the genome using BLASTN ( ncbi ) ; ( ii ) the array design was initiated by picking oligonucleotides with an offset of 148 bp , thereby considering oligos only if these did not reveal an additional hit to the reference genome at a sequence identity cutoff S of 90% , indicating probe redundancy . ( iii ) If less than the average number of probes per gene/pseudogene locus were picked in the previous step , additional oligos were added to the array , initially by changing the offset to 74 bp , and then by incrementally increasing S from 91% to 99% . The median number of probes per locus was 86; in some instances , i . e . in case of highly similar ( paralogous ) regions in the genome , the minimum number of independent probes was 20 for each OR locus ( as an exception , three small OR pseudogene loci were included in the set , having 14 , 15 , and 16 independent probes , respectively ) . In 3 cases , to reach a minimum of 20 independent probe measurements per OR locus and facilitate robust CNV detection , duplicate probes were included in the set ( as for all other probes used , these duplicates matched perfectly to the target locus only ) . Twenty-five genomic DNA samples ( DNA from cell lines obtained from Coriell; mostly HapMap samples ) were used in this study , covering three populations ( European , Asian , Nigeria/Yoruba ) , as follows: NA12003 , NA12004 , NA12005 , NA12006 , NA12246 , NA12248 , NA12865 , NA10851 , NA11997- Europeans ( CEPH ) ; NA15510- presumably European [14]; NA18611- Asian ( Chinese origin ) ; NA18945 , NA18946 , NA18972- Asians ( Japanese origin ) ; NA18504 , NA18508 , NA18856 , NA19103 , NA19141 , NA18501 , NA18502 , NA18505 , NA18506 , NA19128 , NA19154- Africans ( with origin from Nigeria; Yoruba ) . We interrogated genomic DNA from pairs of two individuals ( one labeled with Cy3 , one with Cy5 , measured in different channels ) by following the high-resolution comparative genome hybridization protocol previously described [22] . In our analysis pipeline , Cy3 and Cy5 labeled DNA was independently analyzed – that is , two distinct ‘intensity-measurements’ were retrieved from each panel of the novel Nimblegen multiplex platform ( eight distinct measurements per 4-plex microarray experiment; each panel carried 71 , 980 probes ) . Quantile normalization ( i . e . , the algorithm normalize . quantiles from www . bioconductor . org ) was used to normalize all channel measurements simultaneously , yielding overall identical intensity distributions for each channel across all experiments . Subsequently , for each oligonucleotide probe on the array , log2-ratios r were obtained by relating normalized intensities i for each probe to median normalized probe intensities calculated across three replicates c1 , c2 , and c3 of a designated control individual ( NA19154 ) : For NA12246 , NA15510 , NA19103 , and NA19141 the microarray experiments were performed in replicate , each replicate was normalized separately . Averaged values of the log2-ratios r obtained from the replicates were used for further analysis . CNVs affecting OR loci were called using the following approach: we calculated R as the median r of all probes that unambiguously map to an OR locus and scored loci as being affected by a CNV if abs ( R ) is greater or equal to a cutoff C = 0 . 18 . At this cutoff value , an independent additional technical replicate of NA19541 yielded only 2 predicted CNVs ( false positives ) affecting OR genes or pseudogenes; this suggests a false positive rate of <4% for our CNV calls , as the mean number of such CNVs in other individuals was 54 ( median 53 ) . Since C represents a conservative cutoff for R-measures that also led us to overlook some true CNVs ( see below ) , we decided to apply raw R-values , rather than copy-number calls , for measuring some of the correlations presented below ( e . g . such presented in Figures 3 , S3 , and S6 ) . Following the scoring of the array data according to cutoff C ( see above ) , microarray intensities were converted to discrete values of 1 , −1 and 0 , representing gains , losses and neutral values respectively . For each sample we counted the total number of CNV events ( gains or losses ) . Next we calculated the average of the total number of events per sample found in 24 individuals . Pearson correlations and respective P-values were calculated using Wessa , P . ( 2008 ) , Free Statistics Software , version 1 . 1 . 23-r1 , at http://www . wessa . net/ . Chi-square tests and contingency tables probabilities were calculated using http://www . physics . csbsju . edu/stats . Quantitative PCR ( qPCR ) was carried out using Absolute Blue SYBR Green Rox Mix ( ABgene ) on a Roche LightCycler 480 . Concentrations of the DNA samples were measured using a NanoDrop 1000 spectrophotometer . The samples were diluted to 1 ng/µl stock . Prior to each experiment , 8 µl of every sample was dispensed into the 96 well qPCR plate by Biomek 3000 robot , in duplicates , and dried over-night . The reactions were carried out in a 96-well plate in 10 µl volume , containing 5 µl of Blue-SYBR-Green mix , 0 . 5 µM of each primer and 8 ng of genomic DNA . The following thermocycling program was used: Enzyme activation , 95°C , 15 min; 40 cycles of denaturation at 95°C , 15 sec , and annealing & extension at 60 C , 1 min . After amplification the temperature was slowly raised and fluorescence was continuously monitored to produce melting curves of every product , so as to verify product specificity . Reactions that indicated more than one peak in the melting curves , were removed from further analysis . All qPCR results are summarized in Table S1 . The qPCR results were normalized using qPCR results of the regulator of calcineurin 1 ( RCAN1 ) gene ( previously Down Syndrome Critical Region 1 gene ( DSCR1 ) ) , presumed not to vary in copy number in normal individuals[24] , [25] . For comparisons between microarray and qPCR results we performed a further normalization by log scale subtraction of the value for the designated reference sample NA19154 . The 56 most variable OR loci were selected based on the variance of qPCR results , such that 25 most variable blocks of OR loci were represented . ORs that did not show gains or losses , but revealed a variance in qPCR results >0 . 15 , were considered as additional CNVs . Polymerase chain reaction ( PCR ) amplification was performed in 10 µl volume , using Qiagen HotStart Taq polymerase enzyme , and standard supplier protocol . The following thermocycling program was used: enzyme activation , 95°C , 15 min; followed by 45 cycles of denaturation at 95°C , 15 sec , annealing at 58°C , 30 sec , and extension at 72 C , 1 min . Products were analyzed by gel electrophoresis , in 2% agarose gel with 100 bp marker . Primers were designed with Primer3 ( http://fokker . wi . mit . edu/primer3/input . htm ) software using the following parameters: melting temperatures ( Tm ) between 56 and 60°C , GC-content between 30% and 70% and length from 18 to 28 bp . Uniqueness of primers and amplicons was checked using BLAT and in-silico-PCR against the hg18 reference assembly ( available at http://genome . ucsc . edu ) [26] . In the case of qPCR , all primers were designed to amplify 100–200 bp segments . In cases where automatic design failed to produce specific primers we designed them manually using sequence alignments . All primers were adjusted for the thermodynamical properties used in Primer3 , and checked for uniqueness as described above . CNV breakpoints affecting OR loci were identified using two complementary approaches . ( i ) Breakpoints of CNVs were predicted from the high-resolution microarray data using the BreakPtr algorithm [23] . Namely , although most CNV breakpoints likely occur outside OR gene/pseudogene loci , we tested whether some CNVs disrupt the loci targeted in this study . We therefore applied the default parameters of BreakPtr ( ‘core parameterization’ [23] ) to fine-map CNV breakpoints . We conservatively considered breakpoint-assignments within a locus as robust , if a minimum of 5 probes were present on either side of the breakpoint junction and supported the predicted dosage change [23] ( only such robust breakpoint assignments were reported by us ) . ( ii ) Deletion breakpoints were identified computationally through analyzing genomic clones deposited in Genbank that are not included in the current reference genome . Namely , fosmids were selected from Genbank by using the keywords “fosmid” and “complete sequence” . This library , which contained 1807 clones , was screened specifically for OR genes using the TBLASTN ( ncbi ) algorithm . Altogether , we found 30 fosmids that contained at least one full-length OR gene . These 30 fosmids were subsequently aligned to the human genome in 20 kb fragments using the BLAT algorithm [26] . Fragments that aligned to regions of lengths different from 20 kb were suspected to contain deletions or insertions , and were inspected manually . The clone AC208786 was found to contain the OR8U8 and OR8U9 genes , both of which are not present in the reference genome . AC193144 contains deletion VI , and AC210900 contains deletion I , listed in Table 3 . OR9G9 was found on chr11 of the Celera assembly , and also on clone AC212901 . The Celera assembly was mined using methodologies described in [27] . To identify deleted ORs from Craig Venter's genome [28] we analyzed table “HuRef_homozygous_indels_inversion_gff . txt” from the supplementary material of [28] . We identified OR loci with a one-to-one ortholog in the chimpanzee genome by comparing OR gene/pseudogene coding regions in a pair-wise and bidirectional fashion between human and chimpanzee using BLAST [26] . OR loci were assumed to have a one-to-one ortholog in chimpanzee if , and only if , a corresponding bi-directional ( i . e . reciprocal ) best hit was found in the chimpanzee genome . The median protein sequence identity of bi-directional best hits was 98% . Ancestral alleles for deletions were determined based on bidirectional best hits between the whole deletion region and the individual OR genes versus the Pan troglodyes reference assembly ( March 2006 , Build 2 . 1 ) . For that purpose , orthologous regions between human and chimpanzee were aligned using MEGABLAST ( ncbi ) and BLAT . For the OR8U1 locus we also used a sequence from Celera's latest human genome assembly ( R27c ) . Our searches usually revealed syntenic and highly identical orthologous regions ( coverage>98% and maximum identity>96% ) , except for OR2T10 ( coverage = 82% and maximum identity = 80% ) , for which the region coincides with a gap in the chimpanzee genome . | Copy-number variants ( CNVs ) are deletions and duplications of DNA segments , responsible for most of the genome variation in mammals . To help elucidate the impact of CNVs on evolution and function , we provide a high-resolution CNV map of the largest gene superfamily in humans , i . e . , the olfactory receptor ( OR ) gene superfamily . Our map reveals twice as many olfactory CNVs per person than previously reported , indicating considerable OR dosage variations in humans . In particular , our findings indicate that CNVs are specifically enriched among evolutionary “young” ORs , some of which originated following the human-chimpanzee split , implying that CNVs may play an important role in the gene-birth and gene-loss processes that continuously shape the human OR repertoire . Furthermore , we describe 15 OR gene loci showing frequent human-specific deletion alleles . Additionally , we present evidence for a recent non-allelic homologous recombination event involving a pair of OR genes , forming a novel fusion OR that may harbor novel odorant-binding properties . Such events may potentially relate to individual functional “holes” in the human smell-detection repertoire , and future studies will address the specific chemosensory impact of our genomic variation map . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics",
"computational",
"biology"
] | 2008 | High-Resolution Copy-Number Variation Map Reflects Human Olfactory Receptor Diversity and Evolution |
Buruli ulcer disease ( BUD ) , caused by Mycobacterium ( M . ) ulcerans , is the third most common mycobacterial disease after tuberculosis and leprosy . BUD causes necrotic skin lesions and is a significant problem for health care in the affected countries . As for other mycobacterial infections , T cell mediated immune responses are important for protection and recovery during treatment , but detailed studies investigating these immune responses in BUD patients are scarce . In this study , we aimed to characterise M . ulcerans-specific CD4+ T cell responses in BUD patients and to analyse specific cytokine-producing T cells in the context of disease severity and progression . For this case-control study , whole blood samples of BUD patients ( N = 36 , 1 . 5–17 years of age ) and healthy contacts ( N = 22 , 3–15 years of age ) were stimulated with antigen prepared from M . ulcerans and CD4+ T cells were analysed for the expression of TNFα , IFNγ and CD40L by flow cytometry . The proportions and profile of cytokine producing CD4+ T cells was compared between the two study groups and correlated with disease progression and severity . Proportions of cytokine double-positive IFNγ+TNFα+ , TNFα+CD40L+ , IFNγ+CD40L+ ( p = 0 . 014 , p = 0 . 010 , p = 0 . 002 , respectively ) and triple positive IFNγ+TNFα+CD40L+ ( p = 0 . 010 ) producing CD4+ T cell subsets were increased in BUD patients . In addition , TNFα+CD40L-IFNγ- CD4+ T cells differed between patients and controls ( p = 0 . 034 ) . TNFα+CD40L-IFNγ- CD4+ T cells were correlated with lesion size ( p = 0 . 010 ) and proportion were higher in ‘slow’ healers compared to ‘fast healers’ ( p = 0 . 030 ) . We were able to identify M . ulcerans-specific CD4+ T cell subsets with specific cytokine profiles . In particular a CD4+ T cell subset , producing TNFα but not IFNγ and CD40L , showed association with lesion size and healing progress . Further studies are required to investigate , if the identified CD4+ T cell subset has the potential to be used as biomarker for diagnosis , severity and/or progression of disease .
Buruli ulcer disease ( BUD ) , caused by Mycobacterium ulcerans , is a neglected tropical disease with reported cases in 33 subtropical and tropical countries [1] . The majority of cases have been recorded in 12 countries , mainly in Western and Central Africa with Ghana among those countries where BUD is a significant problem for public health [1] . Notably about half of all cases are diagnosed in children and adolescents with dramatic consequences for their health and social life . At early stages BUD is characterised by painless , subcutaneous papules , nodules , plaques or oedemas . Without treatment , most lesions enlarge progressively and ulcerate developing undermined skin borders [2 , 3] . The disease can eventually destroy tissues , affect bones leading to deformation and may cause permanent disabilities [4 , 5] . However , in some cases non-ulcerative lesions are stable and do not progressively enlarge or ulcerate . Treatment outcome has substantially improved with the introduction of a combination therapy with rifampicin ( given orally ) and streptomycin ( intramuscular injection ) [1 , 6 , 7] . However an effective vaccine against infection is currently not available . Studies on protection of Bacillus Calmette-Guérin ( BCG ) vaccination led to contradictory results [8–10] . In the absence of protective vaccines and with limited knowledge about the exact mode of transmission [11] , prevention of BUD is currently not feasible [12] . Therefore disease management relies on early case detection , reliable diagnosis and on early effective treatment . Classical methods of diagnosis such as isolation by M . ulcerans culture or smear microscopy of acid-fast bacilli take many weeks and/or lack sensitivity [13 , 14] . The current gold standard for diagnosis is based on PCR dependent detection of the M . ulcerans-specific insertion sequence IS2404 from lesion specimens . This test shows high sensitivity and specificity , but is not helpful in monitoring disease progression . Immune-based assays provide a reliable tool for the detection of other mycobacterial diseases . In tuberculosis , Interferon-gamma release assays ( IGRAs ) are important for the diagnosis of Mycobacterium tuberculosis infections [15] . IGRAs detect interferon-gamma ( IFNγ ) following in vitro stimulation with specific antigens . IFNγ is either quantified from plasma of cultured blood ( QuantiFERON ) or by detecting IFNγ producing T cells ( T-SPOT . TB ) . A comparable test based on the activation of M . ulcerans-specific T cells is currently not available . However , it is tempting to speculate , that such an immune based test could be beneficial for early diagnosis , prognosis of healing progress and monitoring response to antibiotic treatment in BUD patients . For most mycobacterial infections , including tuberculosis , acquired cellular immunity is important for protection , but cellular immune responses are not well defined in BUD . It is known that up to one-third of lesions can heal spontaneously [16–18] and formation of granulomas has been reportedly associated with an induction of a proinflammatory and a down-modulation of inhibitory immune responses likely affecting the number of bacilli in the lesions [19 , 20] . Therefore , it has been suggested that cellular immunity plays an important role in the immune response against M . ulcerans . So far only few studies exist characterising immune responses in the context of M . ulcerans infection [21–25] . For instance , a broad analysis of systemic serum chemokines and cytokines revealed suppression of proteins including macrophage inflammatory protein ( MIP ) -1β , monocyte chemoattractant protein ( MCP ) -1 and IL-8 indicating immune modulation during M . ulcerans infection [26] . Immune responses to M . tuberculosis infections strongly depend on T helper type 1 cytokines IFNγ and tumor necrosis factor alpha ( TNFα ) . Counteracting , regulatory immune responses based on the inhibitory cytokines Interleukin ( IL- ) 10 and TGF-β are of major importance in several infectious diseases including tuberculosis [27] . In lesions of BUD patients , the key cytokines IFNγ and TNFα and the inhibitory cytokines IL-10 and TGF-β are produced , but the relative expression of these cytokines varied with the stage of the disease [20] . Modulated concentrations of TNFα were reported in serum of BUD patients if compared to controls [28] . Higher levels IFNγ and IL-10 were detected in BUD patients compared to household contacts or non-endemic controls following stimulation with M . ulcerans sonicate , in a study focusing on IFNγ and IL-10 from supernatants of whole blood [29] . However , there are no studies investigating proportions of cytokine producing CD4+ T cells in BUD patients , which allow dissecting regulation of specific cytokine-producing CD4 T cell subsets in the context of BUD , including multiple cytokine producing CD4+ T cells or T cells producing a single cytokine ( based on what has been measured ) . The analysis of specific cytokine producing CD4+ T cells in active tuberculosis and latently infected tuberculosis patients revealed differences in expression of CD40L- T cells [30] . Expression of CD40L on CD4+ T cells has been associated with a TH1 signature in infections with M . tuberculosis [31 , 32] . Elsewhere , differences in IFNγ+TNFα+IL-2+ , TNFα-single positive CD4+ T cells or a general increase in multi-cytokine producing T cells expression were observed between tuberculosis patients and contacts [33 , 34] . Higher proportions CD40L+IL-2- CD4+ T cells were associated with cystic fibrosis patients infected with Mycobacterium abscessus [35] , highlighting the potential of characterizing the precise profile of cytokine produced by CD4+ T cells in mycobacterial infections . Therefore the aim of this study was to analyse the profile and frequencies of cytokine producing CD4+ T cells after stimulation with M . ulcerans antigen in BUD patients and healthy contacts and to analyse specific cytokine-secreting CD4+ T cell subsets in the context of disease severity and progression .
Ethical approval of the study was obtained from the ‘Committee on Human Research , Publication and Ethics ( CHRPE ) at the School of Medical Sciences , Kwame Nkrumah University of Science and Technology , Kumasi , Ghana ( Ref . : CHRPE/AP/275/14 and CHRPE/AP/301/15 ) . In addition , ethical approval was granted by the Ethical committee of the medical faculty of the Heinrich-Heine-University Dusseldorf ( Ref . : 3903 ) . The aims and procedures were explained to participants and/or their parents/guardians prior recruitment into the study . Only compliant patients were recruited and they were free to withdraw at any point during the study . Written consent was obtained from the parents or guardians . In some cases ( illiterate guardians/parents ) consent was confirmed by thumbprint , a procedure approved by the review board . Between September 2014 and February 2016 , patients with BUD were recruited at the Agogo Presbyterian Hospital in the Asante Akim North District , where there is high incidence of BUD in the middle forest belt of Ashanti region of Ghana [36] . A patient was recruited when the presenting lesion was consistent with the WHO clinical disease definition for BUD and diagnosis later confirmed by M . ulcerans IS2404 PCR . This study is part of a larger study investigating immune modulation in BUD patients with and without concomitant co-infections in children and adolescence . Up to 50% of all BUD cases in Africa are diagnosed in children below the age of 17 years [1 , 37] with age being a significant factor for the clinical presentation of the disease [38] . The present study is specifically focusing on children and adolescent . Participants had to be ≤ 17 years of age , to be recruited into the study . Fifty-one of 101 ( 50 . 5% ) patients were excluded for falling outside the age limit . Patients were also excluded if they i ) had a history of BUD , tuberculosis or leprosy , ii ) had a history of liver or kidney diseases , iii ) were HIV positive ( none excluded ) or tested positive for any helminth infection , iv ) had a recent or current antibiotic use . A flow chart indicating selection of the study group is shown in S1 Fig . A control group of age- and gender matched healthy contacts was recruited from siblings , relatives or household contacts of participants . All participants were asked about their presenting active BUD disease , family history of BUD exposure , their previous medical history , household demographics and previous treatments . Clinical BUD cases were confirmed by PCR for the IS2404 repeat sequence specific for M . ulcerans [39] . Lesions were classified by their form: nodule , oedema , plaque or ulcer . Lesion size was determined by measuring the widest diameter of a lesion and surface area lesion in cm2 ( to take into account differences in the shape of lesions ) . Lesions were categorized following WHO guidelines: Category I: lesion <5cm in the widest diameter , Category II: lesion <15 cm , category III: lesion >15 cm , multiple lesions , osteomyelitis . Stool samples were taken for diagnosis of soil-transmitted helminths ( STH ) and infection with S . mansoni using the Formol-Ether concentration method . Microfilariae were detected using Sedgewick chamber using 100 μl of whole blood [40] . Depending on the initial result , up to 1000 μl of blood was filtered on a nucleopore filter membrane ( 3 μm ) ( Whatmann ) . Filters were stained with Giemsa and analysed by microscopy [40] . Malaria ( Plasmodium falciparum ) status was determined using the rapid test CareStart Malaria HRP2 pf ( Access Bio , Inc . ) . Haematological parameters were assessed using the Sysmex XS-800i system ( Sysmex ) . Patients with BUD received a standard treatment regime with 15 mg/kg streptomycin and 10 mg/kg rifampicin daily for 8 weeks , as recommended by the WHO [7] . Patients presented every two weeks during antibiotic treatment and monthly subsequently for monitoring of healing progress until complete healing . Up to 10 mL of venous blood was collected into heparinised blood collection tubes ( BD Biosciences ) between 9am and 12noon , prior initiation of antibiotic treatment . Blood samples were transported to the laboratory based in Kumasi and immediately processed ( in less than 6 hours after blood has been taken [41 , 42] ) . Surface stains to identify T , B , NK and CD16+ myeloid cells was performed directly in 100 μl of whole blood , diluted with 100 μl RPMI 1640 supplemented with 100 U/mL penicillin and 100 μg/mL streptomycin . Following incubation with the fluorochrome-conjugated antibodies for 30 min , red blood cells were lysed using a red blood cell lysis buffer ( Roche ) and remaining leucocytes were fixed for 15 min using Cytofix Solution ( Biolegend ) . For detection of intracellular cytokines , 100 μl of whole blood was stimulated in 96 well round bottom plates with M . ulcerans antigen at a final concentration of 5 μg/mL or with Staphylococcal enterotoxin b ( SEB; Sigma ) at a final concentration of 15 μg/mL . M . ulcerans antigen sonicate was prepared from a M . ulcerans 1 isolate of African origin . The origin and preparation of the M . ulcerans antigen are described in detail elsewhere [29] . Stimulated blood was incubated for a total of 17 . 5 hr at 37°C , 5% CO2 and Brefeldin A ( Sigma ) was added after an initial incubation of 2 . 5 hr . Whole blood cultured in medium without any stimuli was used as unstimulated negative control . Following incubation , red blood cells were lysed and the remaining cells fixed and permeabilized using a permeabilization Wash Buffer ( Biolegend ) . Fluorochrome-conjugated antibodies ( CD4 , TNFα , IFNγ , CD40L ) were added and incubated for 30 min followed by two additional wash steps . Stained blood was acquired on a BD Accuri C6 Flow Cytometer ( BD Biosciences ) and analysed using FlowJo v10 ( TreeStar ) . For further analysis , M . ulcerans or SEB-specific responses were determined by subtracting unstimulated control values from SEB or M . ulcerans stimulated cells . If subtracted values were ≤ 0 , values were set to 0 . 001 for illustration purposes . Following anti-human antibodies were used for flow cytometric analysis: APC-conjugated CD16 ( clone 3G8 ) , PerCP-Cy5 . 5 . -conjugated CD3 ( HIT3a ) , FITC-conjugated CD56 ( HCD56 ) , Alexa488-conjugated CD4 ( RPTA-4 ) , APC-conjugated TNFα ( MAB11 ) , PerCP-Cy5 . 5 . -conjugated CD154/CD40L ( 24–31; all Biolegend ) , PE-conjugated IFNγ ( 2572311 , BD Bioscience ) and PE-conjugated CD20 ( 2H7 , eBioscience ) . All staining panels were evaluated using fluorescence minus one and unstained controls . Since analysed data were not normally distributed ( Shapiro-Wilk ) and did not meet assumptions for parametric tests , non-parametric tests were applied . For comparison of two groups ( e . g . BUD patients vs . contacts ) Mann-Whitney U test was used . Correlations were tested using Spearman’s rho analysis . Statistical analyses were performed using SPSS v23 ( IBM Corp . ) and were taken as significant if ≤ 0 . 05 . Graphical illustration was done using Graphpad v7 . 0 ( GraphPad Software , Inc . ) . Lesion types were separated into groups based on their lesion form: non-ulcerative forms ( nodule/oedema/plaque ) and ulcerative forms . Patients were also separated into two groups based on the time to healing following the start of antibiotic treatment , using a cut-off of 111 days ( based on the median healing time = 111 . 0 days , range 14–337 days ) or alternatively based on the time healing was first recognized ( cut-off based on the median of 56 days , range 14–231 days ) .
Blood samples were obtained from 36 BUD patients with a median age of 8 . 5 years ( range 1 . 5–17 years ) and from healthy contacts ( median age 7 . 0 years; range 3 . 0–15 years ) . Detailed characterisation of the study groups are provided in Table 1 . Haematological parameters were compared between BUD patients and contacts . There was a moderate , but significant increase in the frequency of basophils ( Table 1 ) . There was no difference between major lymphocyte subsets including T cells , B cells and NK cells ( S2A–S2C Fig ) , but proportions of CD16+ myeloid cells were increased in BUD patients ( S2D Fig ) . Additional haematological parameters did not differ between the two groups ( Table 1 ) . Clinical characteristics of the BUD group are shown in Table 2 . Lesion size varied between individual BUD patients ( 2 . 3–79 . 4 cm2 ) . Based on the widest diameter all lesions but one were classified as category I or II lesions ( for details see methods section ) . One lesion with 15 . 7 cm in the widest diameter was classified as category III lesion . The majority of patients presented with either nodules ( 50 . 0% ) or with ulceration ( 30 . 6% ) ( Table 2 ) . None of the patients included into the study presented with multiple lesions or osteomyelitis . To determine the specific cytokine producing profile of M . ulcerans specific CD4+ T cells , whole blood samples were stimulated overnight with M . ulcerans antigen and analysed for TNFα , IFNγ and CD40L producing T cells by flow cytometry . An example of the according gating and cytokine production is shown in S3 Fig . BUD patients had significant higher proportions of CD4+ T cells producing two ( IFNγ+TNFα+ , TNFα+CD40L+ , IFNγ+CD40L+ ) or all three ( IFNγ+TNFα+CD40L+ ) cytokines when compared to contacts ( Fig 1A , upper panels ) . Only a minor fraction of the BUD contacts showed detectable levels of these cytokine-producing cells in response to M . ulcerans antigen . In addition , frequencies of two subsets characterised as TNFα+CD40L- and CD40L+IFNγ- , were increased in BUD patients ( Fig 1B , upper panels ) . Further characterisation of these subsets revealed , that TNFα+CD40L- CD4+ T cells were almost exclusively IFNγ- ( median of IFNγ- = 100% ) and were therefore denoted as TNFα+CD40L-IFNγ- . In contrast , CD40L+IFNγ- contained both TNFα+ and TNFα- cells ( median of TNFα+ = 42 . 9% ) . Neither CD40L+IFNγ-TNFα- nor IFNγ+CD40L-TNFα- single positive CD4+ T cells differed between BUD patients and contacts ( p = 0 . 134 and p = 0 . 881 respectively ) . No significant differences between both groups were observed following stimulation with Staphylococcal enterotoxin b used as polyclonal control stimulation ( Fig 1A and 1B , lower panels ) . In summary , CD4+ T cells of BUD patients showed specific cytokine profile characterised by induction of multiple cytokine producing cells as well as an increase in the frequency of TNFα+CD40L-IFNγ- CD4+ T cells . BUD patients presented at different stages of disease characterized by different types of lesions and varying lesion sizes ( Table 2 ) . Cytokine producing CD4+ T cells were analysed in the context of these clinical presentations to determine their potential as biomarkers . Data were split into two groups comprising patients with non-ulcerative forms ( nodules , oedema , plaque ) or ulcer . Neither multiple cytokine producing CD4+ T cells producing TNFα+IFNγ+ , TNFα+CD40L+ , TNFα+IFNγ+CD40L+ differed between the two groups ( Fig 2A ) nor TNFα+CD40L-IFNγ- CD4+ T cells showed any differences ( Fig 2B ) . However , the size of lesions did not differ significantly between these two groups ( p = 0 . 508 ) . Therefore the lesion size was correlated with cytokine producing CD4+ T cells . There was no correlation between multiple cytokine producing T cells ( Fig 2C ) . However , TNFα+CD40L-IFNγ- CD4+ T cells were positively correlated with surface area of the lesion ( rho = 0 . 456; p = 0 . 010 ) ( Fig 2D ) , which was reflected by a significant correlation with the widest diameter of the lesion ( rho = 0 . 529; p = 0 . 004 ) ( Fig 2E ) . None of the additional CD4+ T cell subsets presented in Fig 1B ( TNFα+IFNγ- , IFNγ+TNFα- , IFNγ+CD40L- , CD40L+TNFα- ) was correlated with lesion size ( S4 Fig ) . Patients were treated with a combination of rifampicin and streptomycin for eight weeks and time until complete healing was monitored . Applying the healing time , patients could be divided into ‘fast’ and ‘slow’ healers ( Fig 3A ) . A cut-off of 111 . 0 days ( median healing time ) was applied to classify and distinguish these two groups . Notably , we detected significantly increased proportions of TNFα+CD40L-IFNγ- CD4+ T cells in ‘slow’ compared to ‘fast’ healers , whereas none of the other cytokine producing subsets were significantly different between the two groups ( Fig 3B and 3C ) . Of note , ‘fast’ and ‘slow’ healers did not differ significantly by age ( p = 0 . 193 ) or gender ( p = 0 . 079 ) or original lesion size ( Fig 3D ) . There was no significant correlation between healing rate within the first four weeks ( expressed as mm/week ) and TNFα+CD40L-IFNγ- CD4+ T cells ( Fig 3E ) . Some lesions do not immediately start to heal following initiation of antibiotic treatment , a phenomenon that has been described in BUD patients . For BUD patients included in this study healing started at a median duration of 56 days and this value was used to distinguish two groups . BUD patients with lesion starting to heal in less than 56 days had significant lower proportions of TNFα+CD40L-IFNγ- CD4+ T cells compared to patients with lesions starting after more than 56 days ( Fig 3F ) . Of note , is the fact that TNFα+CD40L-IFNγ- CD4+ T cells did not differ between BCG scar positives and negatives ( p = 0 . 815 ) .
In the present study , we show that the TNFα+CD40L-IFNγ- CD4+ T cell subset , induced by M . ulcerans antigen , provides a promising approach for establishing an immune-based assay for monitoring BUD . We evaluated the expression of T helper type 1 associated cytokines in combination with CD40L following stimulation with M . ulcerans sonicate . In particular multiple cytokine expressing CD4+ T cells ( e . g . IFNγ+TNFα+ , TNFα+CD40L+ , IFNγ+CD40L+ ) were induced in BUD patients compared to healthy contacts . In addition , proportion of TNFα+CD40L-IFNγ- were also higher in BUD patients making it a unique single cytokine producing subset . Both multiple cytokine producing CD4+ T cells as well as TNFα single positive T cells have been identified to discriminate between latent and active tuberculosis [33] . TNFα-single positive T cells were also the strongest predictor of an active tuberculosis in a larger study [43] . In BUD this unique TNFα-single positive T cell subset may reflect a strong inflammatory rather than a protective response . However , the functional role of TNFα+CD40L-IFNγ- CD4+ T cells in BUD needs to be investigated in more detail . Diagnostic tests based on immunological responses are routinely used for detecting an infection with M . tuberculosis . IGRA’s are based on the antigen-induced production of IFNγ , but comparable assays are not available for BUD . Our findings may provide an approach to develop such an immune-based assay . In BUD patients , suppression of immune responses can be recognized locally within lesions [20 , 21 , 44] . Whether a generalised systemic immune suppression can be detected is not yet clarified with contradictory findings in different studies [21 , 23 , 29 , 45] . In our study , none of the analysed cytokine-producing CD4+ T cell subsets were lower in BUD patients compared with contacts suggesting that patients are able to mount TH1 responses confirming earlier results [29] . If immune suppression can be detected systemically as found by others [21 , 46] , may depend on the stage of disease , time of sampling , stimulation and method of analysis . As there is currently no vaccine for BUD and the mode of transmission remains unclear disease management and optimization of treatment will remain highly important for the foreseeable future . Therefore we evaluated cytokine producing CD4+ T cells in the context of several clinical characteristics of BUD . First we compared BUD patients with different types of lesions ( ulcer versus non-ulcerative forms ) . None of the analysed subsets including TNFα+CD40L-IFNγ- CD4+ T cells differed between the two groups . Since the type of lesions was not necessarily associated with the size of lesion , we also correlated our identified cytokine producing CD4+ T cells subsets with surface area and with the widest diameter of lesions . Interestingly the only parameter showing a correlation was the TNFα+CD40L-IFNγ- , further supporting the role of this subset as biomarker . An effective treatment based on rifampicin and streptomycin has been established and is recommended by the World Health Organization [1 , 7] . The efficacy of the treatment regime is more than 90% [47 , 48] . Nevertheless the healing process varies widely . Healing may start directly upon initiation of antibiotic treatment or it may take several weeks until healing becomes obvious and time until complete healing is obtained varies considerably . In addition , lesions may increase in size during or following initiation of antibiotic treatment , a phenomenon , which is mainly attributed to recovery of immune responses and is referred to as ‘paradoxical reaction’ [49–51] . Such paradoxical reactions are reported in more than 20% of cases [49 , 52] . Given this , immunological markers , which help to predict healing and potentially ‘paradoxical reactions’ could be beneficial in terms of patient care and optimisation of antibiotic treatment . Since paradoxical reactions are thought to be based on changes in immune response [53] , identifying biomarkers is a promising approach . In the present study complete healing was observed between 14 and 337 days ( median 111 . 0 ) , which is in the range of observed in other studies [47 , 48] . Comparing ‘slow’ versus ‘fast’ healers revealed indeed that only TNFα+CD40L-IFNγ- expression differed between the two groups . The difference in time to healing could theoretically be attributed to the fact that larger lesions need longer time for complete healing . However , in our study , the size of lesions did not differ significantly between ‘slow’ and ‘fast’ healers . Therefore the fact that TNFα+CD40L-IFNγ- differed between the two groups could not solely be attributed to an association with the original lesion size . In addition the healing rate within the first four weeks after start of antibiotic treatment did not correlate with TNFα+CD40L-IFNγ- . The time until first signs of healing were also recorded and varied between 14 and 231 days ( median 56 days ) , consequently we analysed two groups based on this median . Indeed , the frequency of TNFα+CD40L-IFNγ- CD4+ T cells was higher in patients , which showed a delayed start of healing . Evaluating the healing progress using biomarkers may also be of importance in the evaluation of novel treatment regimens , such as full oral therapies , which are currently tested [54] . Of note , we did not have sufficient numbers of patients with a paradoxical reaction ( N = 3 ) to evaluate differences in cytokine producing CD4+ T cells as prognostic marker for this . Therefore it needs to be further analysed if TNFα+CD40L-IFNγ- T cells are useful in this context in a larger study with more patients included . The current study is focusing on children and adolescent . Age is an important factor affecting the clinical presentation of BUD [38] and in Africa more than 50% of all BUD cases are diagnosed in children . Therefore this age group would benefit most of improved biomarkers . However , in addition , TNFα+CD40L-IFNγ- producing CD4+ T cells as biomarker has to be tested in older BUD patients , which are more likely to develop severe forms of BUD [55] . In summary , we present here the first study analysing cytokine producing CD4+ T cells stimulated with M . ulcerans antigen focusing at T helper type 1 CD4+ T cells . Proportions of multiple cytokine as well as TNFα+CD40L-IFNγ- CD4+ T cells differed between BUD patients and healthy contacts and the later one was associated with lesion size and differed between ‘slow’ and ‘fast’ healers . Hence TNFα+CD40L-IFNγ- CD4+ T cells are a potential biomarker , which may allow the development of tests comparable to IGRAs . | Buruli ulcer disease ( BUD ) is a devastating skin disease characterised by nodules , plaque or oedema at early stages , which progress to a characteristic form of ulcer . Without treatment the disease can cause broader tissue destruction , affect bones and cause permanent disabilities . BUD is treated with a combination of two antibiotics over a period of eight weeks . Overall this treatment scheme is effective , but duration of the healing process varies strongly . Specific immune responses are of major importance for controlling other mycobacterial infections ( e . g . tuberculosis ) , but are also used for immune-based diagnostic tests . However , detailed knowledge of specific immune responses in BUD patients is missing . The aim of this study was to characterise the specific cytokine response of CD4+ T helper cells in BUD patients and to correlate specific T cell subsets to clinical disease progression . Using flow cytometry on Mycobacterium ulcerans-stimulated blood samples , we identified specific cytokine-secreting T cell subsets , induced in BUD patients . Proportions of a CD4+ T cell subset , producing a single cytokine were positively associated with lesion size and healing time . Results of this study provide deeper insights in the specific immune response in BUD patients and identify a potential indicator for early diagnosis and disease progression . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"antimicrobials",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"body",
"fluids",
"drugs",
"immunology",
"microbiology",
"p... | 2017 | Analysis of Mycobacterium ulcerans-specific T-cell cytokines for diagnosis of Buruli ulcer disease and as potential indicator for disease progression |
We report a genome-wide assessment of single nucleotide polymorphisms ( SNPs ) and copy number variants ( CNVs ) in schizophrenia . We investigated SNPs using 871 patients and 863 controls , following up the top hits in four independent cohorts comprising 1 , 460 patients and 12 , 995 controls , all of European origin . We found no genome-wide significant associations , nor could we provide support for any previously reported candidate gene or genome-wide associations . We went on to examine CNVs using a subset of 1 , 013 cases and 1 , 084 controls of European ancestry , and a further set of 60 cases and 64 controls of African ancestry . We found that eight cases and zero controls carried deletions greater than 2 Mb , of which two , at 8p22 and 16p13 . 11-p12 . 4 , are newly reported here . A further evaluation of 1 , 378 controls identified no deletions greater than 2 Mb , suggesting a high prior probability of disease involvement when such deletions are observed in cases . We also provide further evidence for some smaller , previously reported , schizophrenia-associated CNVs , such as those in NRXN1 and APBA2 . We could not provide strong support for the hypothesis that schizophrenia patients have a significantly greater “load” of large ( >100 kb ) , rare CNVs , nor could we find common CNVs that associate with schizophrenia . Finally , we did not provide support for the suggestion that schizophrenia-associated CNVs may preferentially disrupt genes in neurodevelopmental pathways . Collectively , these analyses provide the first integrated study of SNPs and CNVs in schizophrenia and support the emerging view that rare deleterious variants may be more important in schizophrenia predisposition than common polymorphisms . While our analyses do not suggest that implicated CNVs impinge on particular key pathways , we do support the contribution of specific genomic regions in schizophrenia , presumably due to recurrent mutation . On balance , these data suggest that very few schizophrenia patients share identical genomic causation , potentially complicating efforts to personalize treatment regimens .
Schizophrenia is a common neuropsychiatric disorder that is characterized by positive symptoms such as delusions , paranoia and hallucinations , negative symptoms including apathy , anhedonia , and social withdrawal , and extensive cognitive impairments that may have the greatest impact on overall function [1] , [2] . While current antipsychotic drug treatments control positive symptoms in most patients , negative symptoms and cognitive impairments are much less improved by these agents [3] . A possible way to improve the treatment of schizophrenia is to identify genetic risk factors that might elucidate the underlying pathophysiological bases as well as help to subclassify patients at a molecular level in a manner helpful to therapy . The etiology of schizophrenia as presently defined is not well understood . While there are clear environmental contributors to disease [4]–[10] , it is clear that genetic predisposition is the major determinant of who develops schizophrenia , with heritability estimates as high as 80% [11] , [12] , placing schizophrenia amongst the most heritable of the common diseases . Schizophrenia genetic research has traditionally focused on identifying linkage regions or on candidate genes and polymorphisms , such as the val158met polymorphism in the dopamine metabolizing gene COMT , or other types of variants such as VNTRs ( MAOA , DAT1 , SLC6A4 ) . Such studies have implicated dozens of genes and variants , but none is generally accepted as definitively associated with schizophrenia [13]–[15] . It is now possible to represent the majority of common genetic variation by genotyping a selected set of tagging SNPs [16] . Such hypothesis-free genome-wide association studies allow the discovery of new genes and pathways affecting complex traits such as schizophrenia with much greater power to detect small effects than linkage studies . To date , there have been five genome-wide association studies ( GWAS ) of schizophrenia . The first study used a small sample size of 178 cases and 144 controls self-identifying as Caucasian and recruited in the US , and reported the association of a SNP in the pseudoautosomal region of the y chromosome at p = 3 . 7×10−7[17] . The second used pooled DNA samples from 600 cases and 2 , 771 controls , all Ashkenazi Jews , and found no genome-wide significant association , although they reported a strong effect of a RELN SNP in females only [18] . The third used pooled DNA from 574 schizophrenia trios and 605 unaffected controls , all recruited in Bulgaria and again found no genome-wide significant association [19] . The next study of 738 cases and 733 controls ( each about 30% African-American , 56% European American and 14% Other ) found no evidence for the involvement of common SNPs in schizophrenia [14] . The most recent study included 479 cases compared to 2 , 937 WTCCC controls and replicated the top SNPs in two further datasets respectively comprising 1 , 664 cases and 3 , 541 controls and 6 , 666 cases and 7 , 897 controls [20] . Three of the loci remained associated after all analyses , one in ZNF804A and two in intergenic regions . While the genotyping arrays used in genome-wide association studies have very limited capacity to detect the effects of rare single site variants , large copy number variants can be readily identified using these arrays , even if they occur in only one or a few subjects . Recently , considerable attention has turned towards identifying rare copy number variants that show elevated frequencies in various human diseases using these platforms . In schizophrenia , four genome-wide screens for large CNVs have recently appeared . Two papers showed that large ( >100 kb ) , rare deletions and duplications that disrupted genes were significantly more common in schizophrenia cases than controls [21] , [22] , and that the disrupted genes in patients were disproportionately from neurodevelopmental pathways [21] . Another showed that de novo CNVs were eight times more frequent in sporadic cases of schizophrenia than they were in familial cases or unaffected controls [23] . While neither of these papers succeeded in identifying particular CNVs as definitive schizophrenia risk factors , the greater load of CNVs reported in cases implicate this type of genetic variant in schizophrenia . Consistent with this , Stefansson et al . [24] recently screened for de novo CNVs and focused on three recurrent CNVs in 4 , 718 patients and 41 , 201 controls ( including , for replication purposes only , all samples investigated in this study ) , located at 1q21 . 1 , 15q11 . 2 and 15q13 . 3 , with odds ratios of 14 . 8 , 2 . 7 and 11 . 5 . Two of these same loci were also reported as risk factors by the International Schizophrenia Consortium ( also including the Aberdeen samples used here ) [22] . These papers collectively suggest that the common disease-common variant hypothesis may be less relevant to schizophrenia than rare variants with highly penetrant effects [25] . However it should be noted that , to date , no WGA SNP study has been well powered to detect effects of common SNPs , since they have either been performed using pooled DNA , ethnically heterogeneous samples or small samples sizes , so it is not yet possible even to rule out reasonably large effects of common SNPs in schizophrenia . Additionally , despite these strong suggestions of a role for rare highly penetrant CNVs , there has been no test of whether any common CNVs also contribute to the risk of schizophrenia . Here we investigated the effects of common SNPs , and both common and rare CNVs , on schizophrenia risk using genome-wide SNP data from the Illumina HumanHap genotyping BeadChips .
We tested for SNP associations with schizophrenia with a logistic regression model using the PLINK software [26] and including sex and curated EIGENSTRAT axes as covariates . An additive genetic model was tested . A series of quality control procedures were undertaken before this analysis ( for details see Text S1 ) . The results were then annotated using the WGAViewer software [27] . No single polymorphism showed a genome-wide significant association in the discovery cohort . The top 100 associated SNPs are shown in Table 1 . The most strongly associated SNPs at this stage were in the ADAMTSL3 gene ( lowest p = 1 . 34×10−6 , Table 1 ) . Following these analyses , we genotyped the top 100 polymorphisms in a further independent Munich cohort of 298 schizophrenic patients and 713 healthy controls , all self-identifying as of German or central European ancestry . Using the Sequenom iPLEX system , we successfully genotyped 98 of the 100 SNPs and found that 8 of these 98 variants showed an association that was significant at the 0 . 05 level in the independent cohort ( rs2135551 , rs950169 , rs1911155 , rs4745431 , rs4745430 , rs4487082 , rs3748376 and rs11635597 ) . These included the most strongly associated three SNPs in the list: rs2135551 , rs950169 and rs1911155 in ADAMTSL3 ( in linkage disequilibrium with one another ) ( Table 1 ) . Since 3 of these 8 SNPs are in strong LD , this is approximately the number of significant associations we would expect by chance at p<0 . 05 , however in all 8 cases the direction of effect was the same as in the original cohort . The combined p value for the strongest associated SNP ( rs2135551 ) across the original and first replication studies is 1 . 3×10−7 . If we use a Bonferroni correction for all the SNPs considered in this study ( 312 , 565 SNPs that passed quality control and the minor allele restriction ) , the 0 . 05 experiment-wide cut-off is 0 . 05/312565 = 1 . 6×10−7 , which means that this association is suggestive , but falls short of the proposed threshold for genome-wide significance of <5×10−8 [28] . The most associated polymorphism ( rs2135551 ) is in the 3′UTR of exon 30 of ADAMTSL3 . To investigate a possible functional mechanism for this SNP , we tested for association with alternative splicing events in the associated region ( exons 28 , 29 , and 30 ) using brain tissue . We found that the associated SNPs rs950169 and rs2135551 showed a highly significant correlation with the use of an alternative splice acceptor site , resulting in a truncated PLAC ( protease and lacunin ) domain in the ADAMTSL3 protein ( p<0 . 0001 , Figure S2 ) . We then confirmed a causal relationship between rs950169 and the observed splicing pattern using a MINIGENE system ( Text S1 ) , and showed an association between rs950169 genotype and the splice form of ADAMTSL3 in brain tissue from both healthy controls and Alzheimer's disease patients ( Figure S3 ) . Finally , we showed that ADAMTSL3 is particularly strongly expressed in hippocampal pyramidal cells ( Figure S4 ) . To try to further confirm this association , we genotyped the ADAMTSL3 SNP rs2135551 ( and , by proxy , rs950169 ) using TaqMan assays in a further 394 cases and 524 controls from Italy . However , in this cohort the p value was 0 . 311 . We then investigated the SNP in 589 schizophrenia cases and 11 , 491 controls from Iceland and found that it did not associate with schizophrenia in these subjects either ( p = 0 . 12 ) ( Hreinn Stefánsson , personal communication ) . Finally , we failed to replicate this association in a third cohort of 179 cases and 267 controls of European-American ancestry genotyped using the Illumina-610 Quad genotyping chip ( and passing through the same quality control procedures used for the discovery cohorts ) . The p value for rs213551 was 0 . 19 and for rs950169 it was 0 . 22 . Since we had whole-genome data for the European-American cohort , we also checked the top 100 SNPs from the discovery cohort but none of the SNPs associated after being corrected for 100 tests ( lowest raw p value = 0 . 02 ) . In the combined Munich and Aberdeen discovery cohort , the MAF of rs2135551 was 0 . 23 in cases and 0 . 30 in controls . In the Italian cohort , the MAF was a little higher at 0 . 33 in cases and 0 . 36 in controls , in Iceland the MAF was 0 . 25 in cases and 0 . 27 in controls , and in the European-American sample it was 0 . 28 in cases and 0 . 25 in controls . These data indicate that despite its real functional effect , the top association with schizophrenia in the discovery and first replication cohorts is likely to be a false positive . To assess more formally the combined evidence of association for the ADAMTSL3 SNPS we extended the Bayesian framework developed by Wakefield [29] to consider the cumulative posterior odds for the hypothesis of true association with schizophrenia , as data from successive datasets are added ( see Methods ) . We found that the posterior odds for true association at rs2135551 tracked from 0 . 10 ( after GWAS ) to 0 . 68 ( after Replication 1 in Germany ) to 0 . 30 ( after Replication 2 in Italy ) to 0 . 15 ( after Replication 3 in deCODE ) to 0 . 02 ( after Replication 4 in the US cohort ) . Thus , under these assumptions , the odds for the association being true never rise above one , and finally reduce to the null hypothesis being 50 times more likely than the alternative hypothesis of true association . To investigate the degree of weight to put on the functional effect of rs950169 , using data from a genome-wide association study examining the effects of approximately 550 , 000 SNPs on expression of 1 . 41 million exons in human frontal cortex tissue ( Text S1 ) , we looked to see what percentage of exons show association with a nearby ( +/− 100 kb ) SNP at or below the level of significance that rs950169 associates with the expression of the ADAMTSL3 exon 3605495 ( https://www . affymetrix . com/analysis/netaffx/index . affx ) . We found that 84 , 840 , or 6% , of exons showed a SNP association of this magnitude . This illustrates the importance of weighing functional evidence against an appropriate null hypothesis , and we note that in many cases arriving at a quantitative evaluation of this kind can be very difficult . This fact , along with the failed replication in the Italian , Icelandic and US data , suggests that evidence of functional effect for SNPs implicated in GWAS studies should not be considered an appropriate substitute for confirmation in replication datasets . For analysis of copy number variation , we used the three cohorts with genome-wide SNP genotype data , namely Aberdeen , Munich , and an American cohort that has not yet been studied ( for copy number variation ) in any previous publications . All samples that passed SNP-QC procedures ( see Methods ) were entered into the CNV analysis , whereupon further QC was performed to determine if accurate CNV calling could be expected ( see Methods ) . In Aberdeen , 12 samples ( 10 cases and 2 controls ) failed CNV QC , in Munich 39 samples ( 9 cases and 31 controls ) and in the American 49 samples ( 29 cases and 14 controls ) failed CNV QC . These samples were excluded from further analysis , leaving a final dataset of 422 cases and 381 controls from Munich , 441 cases and 439 controls from Aberdeen and 150 cases and 264 controls from the US ( European origin ) , a total of 1 , 013 cases and 1 , 084 controls . We also examined both previously implicated regions , and regions newly implicated here in 60 African-American schizophrenia patients and 64 African American controls ( after excluding 8 and 1 respectively for CNV QC failure ) .
While our genome-scan identified no definitive associations between SNPs and schizophrenia risk , SNPs in the ADAMTSL3 gene were the most strongly associated . One of the ADAMTSL3 SNPs is clearly functional and influences the proportion of two alternative transcript species . Nevertheless , study of additional cohorts strongly suggested that ADAMTLS3 is not related to schizophrenia risk and that functional evidence should not be used to strengthen claims for modestly associated variants . We also failed to replicate any of the SNPs previously identified in either genome-wide association studies or candidate gene studies as schizophrenia risk factors . Notably , not only did none of these show genome-wide significance , none showed significant evidence even if we corrected only for the 782 SNPs from previously associated candidate genes . We have calculated here that by using genome-wide genotyping in the discovery cohorts and then typing the top 100 SNPs in the first replication cohort , we have 80% power to detect an allele with MAF≥0 . 1 at an odds ratio of 1 . 8 or larger . This is one of the largest whole genome association studies reported for schizophrenia and is similar in size or larger than studies that have successfully identified risk factors for common diseases [46] , [47] . While our sample size is not large enough to identify risk factors of small effect in a genome-wide context , we have introduced an approach for assessing whether our negative results are consistent with a model in which common SNPs explain most of the heritability of schizophrenia . While this analysis shows that we cannot rule out such a possibility , it would require a large number of SNPs and an implausible genetic model . There are a number of possible explanations for the lack of compelling genetic associations in schizophrenia GWAS to date . Firstly , it has been postulated that schizophrenia is not an homogenous condition , but is in fact a group of several different genetically heterogeneous syndromes that are classed together due to overlap of particular diagnostic symptoms [48] . This hypothesis may be tested in the near future with the publication of collaborative datasets combining several thousand cases and controls , which will presumably include reasonable sample sizes of each subtype . However , in order to distinguish between the presence of reasonable effect sizes in subgroups of the patients and very small effect sizes common to all , it will be necessary to examine more detailed information about the patients including symptoms , putative disease subtypes , and perhaps measures of cognition and other endophenotypes . Similarly , large datasets would also be necessary to investigate the possibility that common variants exert their effects only through interaction with other genetic variants , a plausible hypothesis that it has not yet been possible to investigate in any powerful way . Alternatively , genetic risk factors may interact so strongly with the environment that their marginal effects over all environments are too low to detect in a sample of this size . Further investigation of this will require much larger , longitudinal studies of patients and controls . The final possibility is that much of schizophrenia risk is due to rare , moderate-to-high penetrance variants whose population frequencies reside somewhere below the threshold of detection of genome-wide screens . Due to the complex patterns of schizophrenia heredity , and the relative lack of families with Mendelian schizophrenia syndromes , this cannot account fully for schizophrenia susceptibility . On balance , however , the data presented here are most consistent with this interpretation . First , we find no evidence of association for common SNPs , but clear evidence that a fraction of cases are due to very rare , very highly penetrant structural variants . One interpretation of this pattern is that selection for reliable cognitive function has been sufficiently strong to keep the genome free of common variants that predispose to schizophrenia , and that it is only rare deleterious variants that influence risk [49] . This model for schizophrenia genetics presents clear challenges to the hope that genetics will rapidly reveal new therapeutic opportunities or partition patients up into a small number of clinically manageable subgroups . More encouragingly , however , despite the rarity of these events , we nonetheless replicated several specific regions from previous studies that were not known to be recurrent schizophrenia-associated CNVs , including those affecting APBA2 and the surrounding region [40] . Additionally , we have implicated new regions , including a large deletion at 16p13 . 11-p12 . 4 that may be an important risk factor for other neuropsychiatric conditions . This region also intimates at the possibility that while patients may have different genetic contributors , some of the different events may point towards the same pathway , given that the 16p region includes a gene known to encode a binding partner of DISC1 , a gene with confirmed involvement in schizophrenia . These observations suggest that a full catalogue of rare determinants of schizophrenia could identify a number of specific genomic regions or events that unite a fraction of patients as having the same or similar underlying causes . These subgroups of patients can be further investigated to see if the genetic contributors can elucidate the molecular mechanisms underlying particular symptoms or drug-response phenotypes . Finally , even if most of the schizophrenia risk is due to rare relatively highly penetrant causes , it seems unlikely this would all be structural . It is likely that rare single site changes disrupting gene function must contribute as well , and will likely only be determined through full genome resequencing , which must be considered a goal for future schizophrenia genomic research .
All cases and controls gave informed consent . The study was approved by both local and multiregional academic ethical committees . The SNP discovery cohort consisted of two distinct sub-cohorts . The first cohort comprised 439 schizophrenia patients ( age 39 . 2±10 . 4 yr , range 19–70 ) and 418 healthy controls ( age 48 . 8±14 . 7 yr , range 22–75 ) , all self-identifying as of German or central European ancestry and collected in Munich . The second cohort comprised 461 schizophrenia patients and 459 controls , all self-identifying as of Scottish or north European ancestry , collected in Aberdeen , Scotland . Critically , patients for the two cohorts were selected using a consistent clinical protocol . To be enrolled as a case , participants must have had both a DSM-IV and an ICD-10 diagnosis of schizophrenia [50] . In the Munich and Aberdeen cohorts respectively , subtypes were observed in the following proportions: paranoid 77 . 6% and 86 . 2% , disorganized 15 . 6% and 7 . 5% , catatonic 2 . 2% and 2 . 1% and undifferentiated 4 . 6% and 4 . 2% . Detailed medical and psychiatric histories were collected , including a clinical interview using the Structured Clinical Interview for DSM-IV ( SCID ) , to evaluate lifetime Axis I and II diagnoses . Cohen's Kappa ( Cohen , 1960 [51] ) of 0 . 80 indicated good inter-rater reliability . Exclusion criteria included a history of head injury or neurological diseases . All case participants were outpatients or stable in-patients . Further details of the Munich cohort and protocol are available in Van den Oord et al . ( 2006 ) [52] . All cases and controls gave informed consent . The study was approved by both local and multiregional academic ethical committees . Healthy volunteers were randomly selected from the general population both for the Munich and Aberdeen cohorts ( ascertained by mail for Munich , and by general practitioners for Aberdeen ) . In the Aberdeen study volunteers were screened for absence of psychiatric disorders and only those with no major psychiatric episodes or major mental illness in a first degree relative were included in the study . In the Munich cohort several screenings were conducted before the volunteers were enrolled in the study in order to exclude subjects with central neurological diseases and psychotic disorders or subjects who had first-degree relatives with psychotic disorders . First , subjects who responded were screened by phone for the absence of neuropsychiatric disorders . Second , detailed medical and psychiatric histories were assessed for the volunteers and their first-degree relatives using systematic forms . Third , if no exclusion criteria were fulfilled , they were invited to a comprehensive interview including the SCID [52] to validate the absence of psychotic disorders . Finally , a neurological examination was conducted to exclude subjects with current CNS impairment . In the case that the volunteers were older than 60 years , the Mini Mental Status Test [53] was performed to exclude subjects with possible cognitive impairment . The enrolment procedure was similar for the Aberdeen controls , although a formal SCID was not undertaken . The first replication cohort comprised 298 schizophrenia patients ( age 37 . 3±11 . 8 yr , range 18–66 ) and 713 healthy controls ( age 45 . 5±16 . 1 yr , range 19–72 ) . The recruitment protocol is identical to that used for the Munich discovery sample . Schizophrenia subtypes were observed in the following proportions: paranoid 75 . 5% , disorganized 16 . 1% , catatonic 4 . 7% and undifferentiated 3 . 7% . The sample comprised a total of 918 subjects of whom 394 ( mean age±SD 43 . 5±12 . 8 years , range 19–80 ) had a DSM-IV-TR [50] diagnosis of schizophrenia and 524 ( mean age±SD 47 . 3±29 . 7 years , range 19–87 ) were healthy controls . Patients and controls were of Caucasian ancestry for at least two generations , lived in northern Italy , were unrelated to other participants , and fulfilled predefined group-specific inclusion and exclusion criteria . The different subtypes of schizophrenia were observed as follows: paranoid 61 . 9% , undifferentiated 17 . 0% , residual 10 . 4% , disorganized 9 . 4% , catatonic 1 . 3% . The patients were enrolled from those voluntarily admitted to the Brescia IRCCS Fatebenefratelli . The inclusion criteria were a DSM-IV-TR diagnosis of schizophrenia [54] and a level of understanding and attention judged sufficient to give true informed consent; a lifetime comorbidity with other DSM-IV-TR Axis I disorders was an exclusion criterion . All participants underwent detailed clinical interviews , implemented , when required , by DSM-IV-TR adjusted versions of the Structural Clinical Interview for DSM-IV Axis I Disorders . Moreover , to attribute the schizophrenia subtype , a checklist of the symptoms dominating the clinical picture at the screening visit and in the previous 4 weeks was used . All patients and controls enrolled in the study provided a written informed consent approved by local Ethical Committee ( CEIOC , Brescia , Italy ) . A concise , but unequivocal explanation about the aims of the study was included on the written consent form . The healthy unrelated participants were recruited through different sources ( randomly selected among university , consenting doctors , nurses , employees and attendants of Brescia IRCCS Fatebenefratelli and elderly association ) . All participants underwent a psychiatric interview to exclude Axis I disorders and Axis I diagnosis of first-degree relatives . Absence of relevant neurological diseases was mandatory for the inclusion in the study . The Mini Mental Status Test 29 was performed to subjects older than 65 years , to exclude possible cognitive impairment . For the third replication , we used information provided by Dr . Hreinn Stefánsson by personal communication , with reference to the deCODE samples used in previous publications , e . g . [24] . The US patients were part of an NIMH-funded Clinical Research Center at Case Western Reserve University and prospective clinical trials at Vanderbilt University . Information about recruitment and assessment has been previously reported [55] , [56] . The healthy controls were recruited as part of the Genetics of Memory/ Genetics of Epilepsy studied at Duke . All subjects were cognitively normal and free of neuropsychiatric disorders . The extra 1 , 547 controls were also part of the Genetics of Memory/ Genetics of Epilepsy studied at Duke and genotyped in the same facility using the Illumina Infinium HumanHap 550K , and subject to identical quality control procedures . They comprised healthy controls who performed normally in a series of cognitive tests ( age range = 18–85 , mean = 25 . 5 , median = 22 ) . The majority were of European origin but also included were approximately 10% each of African-American , East Asian ( mostly Chinese ) and South Asian ( mostly Indian ) as well as 5% Hispanic . The Munich cohort was genotyped using the Illumina HumanHap300 chip with a total of 317 , 503 SNPs and the Aberdeen cohort was genotyped using the Illumina HumanHap550 chip with a total of 555 , 352 SNPs . We carried out a series of quality control ( QC ) checks and tests of cryptic relatedness , ultimately excluding a total of 15 and 28 participants in Munich and Aberdeen respectively ( Text S1 ) . We also employed a “one percent rule” that discarded from analysis any SNP that had more than 1% of samples that could not be reliably scored , to reduce the scope for spurious association . After employing this rule the average success rate of genotyping was 98 . 4% and the concordance rate for duplicate genotyping was 99 . 997% . The US cohort was genotyped using the Human-610 Quad Beadchip at the Institute for Genome Sciences and Policy Genotyping Core , and the same quality control procedures were applied as those used for the discovery cohorts . Our core association analyses to identify schizophrenia risk factors focused on single-marker tests of the 312 , 565 QC-passed SNPs that were genotyped in both cohorts . To control for the possibility of spurious associations resulting from population stratification we used the EIGENSTRAT approach of Price et al [57] . This method derives the principal components of the correlations among gene variants and corrects for those correlations in the association tests . In principle , therefore the principal components in the analyses should reflect population ancestry . We have noticed however that some of the leading axes appear to depend on other sources of correlation , such as sets of variants near one another that show extended association . We have documented the potential for inversions to create this effect and it may be created by other causes of extended linkage disequilibrium as well ( Text S1 ) . For this reason we inspected the SNP ‘loadings’ for each of the leading axes to determine if they depended on many or relatively few SNPs , as would be expected if the given axis reflected population ancestry or a more localized linkage disequilibrium effect respectively . This analysis identified several axes clearly due to inversions and suggested that four axes should be retained for ancestry adjustment ( Text S1 ) . We therefore assessed significance using four principal components emerging from the EIGENSTRAT analyses as covariates in a logistic regression model which also incorporated sex as a covariate and combined samples from Munich and Aberdeen ( a division which clearly drove the first EIGENSTRAT axis ) . Following Wakefield [29] , we found the estimated log-odds for association , , under a multiplicative genetic model for rs2135551 , together with its estimated variance V , from standard logistic regression of each dataset . Given a prior odds of PO for the association being true , and a prior distribution of ∼N ( μ , W ) for θ under the hypothesis of true association , we found the posterior odds having observed new data at each stage as , and updated the posterior distribution of θ under the hypothesis of true association as . We then entered these posteriors as priors into the analysis of the next set of data . To start , we set PO = 1/100000 following the Wellcome Trust Case Control Consortium [58] ( i . e . , assuming a million independent regions of the genome and 10 detectible causal loci for schizophrenia ) , and following Wakefield , 2007 [29] we set μ = 0 and W = ( log ( 1 . 5 ) /1 . 96 ) 2 ( i . e . , assuming that 95% of all casual effects fall between 2/3 and 3/2 per allele under a multiplicative genetic model ) . Alternative transcripts were identified searching ExonHit Therapeutics SpliceArray portal ( http://portal . splicearray . com ) and blasting exon-intron boundary sequences against human cDNA libraries . For semi-quantitative evaluation of transcript ratio differences , primers flanking the common 5′ splice donor site in exon 29 ( forward primer: 5′-TTGGGCCCTCCTGTGATA-3′ , location shown in Figure S1A ) and the alternative 3′ splice acceptor site in exon 30 ( reverse primer: 5′-TGGCAGCACCTTTGTTTGTA-3′ , location shown in Figure S1A ) were used to simultaneously amplify all four transcript forms ( Figure S1A ) . The fragments were separated on a 3 . 5% NuSieve agarose gel and direct sequencing was used to confirm expected transcript forms . Taqman-based real time PCR was used to quantitatively determine ratios of alternative transcripts in human brain tissue . Assays were custom designed through Applied Biosystems by targeting unique exon-exon boundaries ( for primer and probe sequences see Text S1 ) . β-actin mRNA expression level was quantified using a commercially available Taqman assay ( Applied Biosystems ) . Fluorescence outputs were quantified in real time using a 7900HT Fast Real Time PCR System and the data were analyzed using SDS software v . 2 . 2 . 2 ( Applied Biosystems ) . One way analysis of variance was used to determine the correlation of alternative transcript abundance with the rs950169 and rs2135551 genotypes in human brain tissue . Statistical analyses were performed both separately in control and Alzheimer's disease prefrontal cortex samples , and as a combined subject analysis . A genomic DNA fragment of 4028 bp from the ADAMTSL3 gene that included exons 29 and 30 with flanking intron sequences was PCR-amplified from a reference genomic DNA using the following primers: gggaattcAAGGGCAGATACCCCAAAGT and taggatccCGCTTGCTCTTCCAACTACC . Subsequently , the PCR fragment was subcloned into pSPL3 ( GibcoBRL ) as a minigene . The minor allele of rs950169 was generated in the minigene by mutagenesis ( QuikChange Mutagenesis kit , Stratagene ) and the sequences were confirmed by DNA sequencing . The minigenes were transfected into HEK293 cells using Lipofectamine2000 ( Invitrogen ) . After the 48 h transfection , RNA was extracted using RNeasy kit ( Qiagen ) and converted into cDNA using High-Capacity cDNA Archive Kit ( Applied Biosystems ) . Alternative splicing of exon 29 and exon 30 was detected by Taqman assays and agarose gel . Following Chapman et al . [59] , we assumed that the test statistic from a case-control trend test of association follows a non-central chi-square distribution with 1 d . f . and non-centrality parameter η = ( n−1 ) r2H , where n is the sample size , r2 is the LD between the causal SNP and it's tag SNP on the GWAS genotyping panel , and H is the proportion of variation explained by the SNP if it were typed directly . In a case-control setting , , where Π is the proportion of cases in the total sample , p is the frequency of causal alleles in controls and in the general population ( assuming a rare disease ) , p′ is the frequency of causal alleles in cases ( where θ is the allelic relative risk or odds ratio assuming a rare disease ) , and is the causal allele frequency in the study as a whole . We simulated sets of 100 , 000 X1 values from a Normal distribution with mean = √η1 and variance = 1 , where η1 is the presumed non-centrality parameter from the GWAS study ( n = 1734 , Π = 0 . 506 ) , and an additional set of 100 , 000 X2 values from a Normal distribution with mean = √η2 and variance = 1 , where η2 is the presumed non-centrality parameter from the first replication study ( n = 1011 , Π = 0 . 295 ) . To score a “hit” , we required both that X1 exceeded the upper critical value for a two-tailed test at α = 0 . 0003 ( to mimic being passed to the 1st replication stage ) , and that both X1 and X2 had the same sign and had a joint P<1 . 6×10−7 when combined using Stouffer's weighted-Z method [60] ( to mimic achieving a Bonferroni-corrected genome-wide significance level after both stages ) . Power was defined as the number of hits divided by 10 , 000 . Solutions to θ based on fixed values of the other parameters were found by an iterative root-finding procedure ( function “uniroot” in the R statistical package , http://www . r-project . org/ ) . The Total Lambda-s expected based on k independent SNPs each with a given OR and MAF was found using equations in Camp et al . [61] . All subjects that passed SNP QC procedures were entered into the CNV analysis . This comprised 892 samples from Aberdeen ( 441 controls , 451 cases ) , 842 samples from Munich ( 412 controls , 430 cases ) and 443 samples from the US ( 267 controls , 176 cases ) . The CNV calls were generated using the PennCNV software ( version 2008jun26 version [62] ) using the Log R ratio ( LRR ) and B allele frequency ( BAF ) measures automatically computed from the signal intensity files by BeadStudio , and the standard hg18 “all” PennCNV hidden Markov model ( hmm ) and population frequency of B allele ( pfb ) files for the 317 and 550 BeadChips . For the samples genotyped on the 610-Quad BeadChips , we used the hh550_610 . hg18 pfb and gc model files separately provided by Dr . Kai Wang to ensure inclusion of all CNV-specific markers . Because many of the samples had below optimal genomic wave QC values , for Aberdeen and Munich we implemented the gc model wave adjustment procedure . We used the PennCNV checks to exclude samples that failed quality control . These included samples that had a LRR standard deviation >0 . 28 , BAF median>0 . 55 or <0 . 45 , BAF drift >0 . 002 or WF>0 . 04 or <−0 . 04 . For the US cohort we found an excess of CNVs in samples with LRR_SD values between 0 . 25 and 0 . 28 , so the LRR_SD cut-off was reduced to 0 . 25 for both cases and controls from the US . All samples that failed QC after the wave adjustment procedure were removed . Due to the complications of hemizygosity in males and X-chromosome inactivation in females , all analyses were restricted to autosomes . Additionally , to ensure that we were working with high-confidence CNVs , we excluded any CNV for which the difference of the log likelihood of the most likely copy number state and the less likely copy number state was less than 10 ( generated using the -conf function in PennCNV ) . Finally , some centromeric and telomeric regions are not well mapped , and this can potentially result in CNV-calling errors in these regions ( Dr . Kai Wang , personal communication ) . Also , genomic regions coding for immunoglobulin genes have previously been shown to be potential sites of false-positive PennCNV calls [62] . Our own research has shown that calls in both of these types of region differed significantly depending on the sample type used for DNA extraction ( significant difference p<10−10 for deletion and/or duplication frequencies between samples genotyped on DNA extracted from blood or saliva , data not shown ) . We therefore excluded any CNV that overlapped any of the following regions by 50% or more of its length: chr2: 87 . 0–92 . 0 , chr14: 18–23 . 6 Mb , chr14: 104 . 5–106 . 5 Mb , chr15: 17 . 0–21 . 0 , chr16: 31 . 8–36 . 0 Mb , chr22: 20 . 5–21 . 8 Mb ( immunoglobin regions ) ; chr1:0–4 Mb , 240–247 Mb; chr2: 87 . 0–92 . 0 Mb; chr4: 0–1 . 43 Mb , 48 . 75–49 Mb , 190 . 7–191 . 3 Mb; chr7:0–200 kb , 56 . 5–62 . 5 Mb; chr8: 39–45 Mb , 145–146 . 3 Mb; chr9: 44 . 5–70 . 1 Mb; 138–140 . 2 Mb; chr10: 38 . 5–42 Mb , 134–135 . 4 Mb; chr11: 0–1 . 8 Mb; chr14: 18–23 . 6 Mb , 104 . 5–106 . 5 Mb; chr15: 17 . 0–21 . 100–100 . 3 Mb; chr16: 0–2 . 1 Mb , 31 . 8–36 . 0 Mb , 86 . 6–88 . 9 Mb; chr17:0–1 Mb , 76 . 5–78 . 8 Mb; chr18: 14–16 Mb , 75 . 5–76 Mb; chr19: 0–2 . 1 Mb , 25 . 7–28 . 3 Mb , 61 . 5–62 . 5 Mb; chr20: 25 . 7–28 . 3 Mb , 61 . 5–62 . 5 Mb; chr21:9 . 7–14 . 3 Mb; chr22:14 . 4–14 . 7 Mb , 20 . 5–21 . 8 Mb ( centromeric and telomeric regions , some overlapping immunoglobin regions as above ) . We also removed CNVs that spanned centromeres by searching for those larger than 1 Mb with fewer than 50 SNPs and checking their genomic locations . Following Walsh et al . , we defined rare copy number variations as those with at least 100 kb in size , at least 20 SNPs and not previously described the Database of Genomic Variants ( DGV , http://projects . tcag . ca/variation/; dgv18v6 ) . Any previously described event that had at least a 60% overlap with a newly discovered event was considered ‘not rare’ and excluded from further evaluation ( for details see [21] ) . We then looked to see if there was an increase in particular types of rare CNVs between cases and controls using a 2-tailed Fisher's Exact test to compare number of cases versus number of controls with and without the event . In order to implement a genome-wide screen for the effect of common CNVs on schizophrenia predisposition , the number of deletions and duplications affecting each SNP was counted up and compared between cases and controls using Fisher's exact test . For each population , separate analyses were done for deletions , duplications and loci affected by both deletions and duplications . To enter the deletion analysis , a SNP had to be deleted in 3 or more samples and duplicated in fewer than 2 samples , for the duplication analysis a SNP had to be duplicated in 3 or more samples and deleted in fewer than 2 samples and the third analysis included all SNPs that are deleted in 2 or more samples and duplicated in 2 or more samples . Events that only occurred in one or two individuals were not analyzed . For the screen for schizophrenia-specific recurring events , we performed the same statistical test , and again stipulated that the event must occur in at least three individuals , but this time we did not filter out sites that were affected by duplications from the deletion analysis nor those affected by deletions from the duplication analysis , in order to maximize the search space for each test . Firstly , we screened the genes that are affected by the rare CNVs greater than 100 kb ( described above ) . To do this we mapped all genomic coordinates for SNPs used in defining CNVs into the most updated human genome variation build ( Ensembl variation build 50_36l , dbSNP build 129 ) . We then aligned the genomic coordinates of the rare CNVs with the most updated human genome build ( Ensembl core build 50_36l , human genome build 36 ) . Any protein-coding gene that was either broken by or fully included in a rare CNV was considered “affected” . We detected the following gene counts that were affected by deletions in cases and controls respectively in the different populations: Aberdeen: 407 , 210; Munich: 109 , 55; US: 70 , 159;and for duplications , Aberdeen: 294 , 180; Munich: 217 , 127; US: 34 , 17 . We then used Ingenuity Pathway Analysis ( IPA , see [21] ) to perform a pathway enrichment ( over-presenting ) analysis separately for the four groups of genes we detected . The statistical significance was evaluated using Fisher's exact test . To avoid false positives , we further stipulated that at least two genes in a pathway must be disrupted for that pathway to be considered enriched , in addition to the P values from Fisher's exact test . | Schizophrenia is a highly heritable disease . While the drugs commonly used to treat schizophrenia offer important relief from some symptoms , other symptoms are not well treated , and the drugs cause serious adverse effects in many individuals . This has fueled intense interest over the years in identifying genetic contributors to schizophrenia . In this paper , we first show that common genetic variants , the focus of most research until recently , do not seem to have a major impact on schizophrenia predisposition . We then provide further evidence that very rare , large DNA deletions and duplications contribute to or explain a minority of schizophrenia cases . Although the small number of events identified here do not restrict focus to a finite set of molecular pathways , we do show one event that deletes a gene known to interact with DISC1 , a gene known to cause psychiatric problems in one family . Such convergent findings have potential implications for the development of new therapies and patient subclassifications . We conclude that schizophrenia genetics research must turn sharply toward the identification of rare genetic contributors and that the most important tool in this effort will be complete whole-genome sequencing of patients whose clinical characteristics have been very thoroughly assessed . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/medical",
"genetics"
] | 2009 | A Genome-Wide Investigation of SNPs and CNVs in Schizophrenia |
Trachoma has been endemic in The Gambia for decades . National trachoma control activities have been in place since the mid-1980's , but with no mass antibiotic treatment campaign . We aimed to assess the prevalence of active trachoma and of actual ocular Chlamydia trachomatis infection as measured by polymerase chain reaction ( PCR ) in the two Gambian regions that had had the highest prevalence of trachoma in the last national survey in 1996 prior to planned national mass antibiotic treatment distribution in 2006 . Two stage random sampling survey in 61 randomly selected Enumeration Areas ( EAs ) in North Bank Region ( NBR ) and Lower River Region ( LRR ) . Fifty randomly selected children aged under 10 years were examined per EA for clinical signs of trachoma . In LRR , swabs were taken to test for ocular C . trachomatis infection . Unadjusted prevalences of active trachoma were calculated , as would be done in a trachoma control programme . The prevalence of trachomatous inflammation , follicular ( TF ) in the 2777 children aged 1–9 years was 12 . 3% ( 95% CI 8 . 8%–17 . 0% ) in LRR and 10 . 0% ( 95% CI 7 . 7%–13 . 0% ) in NBR , with significant variation within divisions ( p<0 . 01 ) , and a design effect of 3 . 474 . Infection with C . trachomatis was found in only 0 . 3% ( 3/940 ) of children in LRR . This study shows a large discrepancy between the prevalence of trachoma clinical signs and ocular C . trachomatis infection in two Gambian regions . Assessment of trachoma based on clinical signs alone may lead to unnecessary treatment , since the prevalence of active trachoma remains high but C . trachomatis infection has all but disappeared . Assuming that repeated infection is required for progression to blinding sequelae , blinding trachoma is on course for elimination by 2020 in The Gambia .
Trachoma is the leading infectious cause of blindness worldwide . [1] It is caused by repeated re-infection with the ocular serotypes ( A , B , Ba and C ) of the bacterium Chlamydia trachomatis , and is predominantly found in the poorest countries in the world . Active trachoma , characterised by the presence of subepithelial follicles ( trachomatous inflammation , follicular ( TF ) ) and/or inflammation ( trachomatous inflammation , intense ( TI ) ) , is usually found in children . After years of repeated re-infection scarring may occur , which can lead to distortion of the eyelid , causing the eyelashes to turn inwards ( trichiasis ( TT ) ) and scratch the cornea , resulting in corneal opacity and blindness . [2] The World Health Organization ( WHO ) strategy for Global Elimination of Blinding Trachoma by the year 2020 ( GET2020 ) is through employment of the SAFE strategy ( Surgery for trichiasis , Antibiotics for active trachoma , Facial cleanliness , and Environmental improvement ) . [3] The Gambian National Eye Care Programme ( NECP ) , established in 1986 , expanded its national intervention programme to cover the whole country by 1996 . NECP activities include the training of health workers in primary eye care , surgery for trichiasis cases , recognition and treatment of conjunctivitis , school screening , and face-washing promotion . [4] The NECP also has a network of nyateros ( Friends of the Eye ) who are non-health professionals identified by their own communities , and have been trained to promote good eye health practices in the community . The programme policy is to treat active trachoma cases and household contacts with tetracycline eye ointment . At the time the survey presented here was conducted ( January to March 2006 ) The Gambia was yet to receive a donation of the antibiotic azithromycin from the International Trachoma Initiative ( ITI ) for mass treatment . Evidence from two national surveys carried out in 1986 and 1996 suggested that the prevalence of trachoma fell in The Gambia , with age standardised prevalences of blinding trachomatous corneal opacities falling from 0 . 10% to 0 . 02% . In the same time span , the prevalence of active trachoma in 0–14 year-olds fell from 10 . 4% to 4 . 9% . [5] Disease prevalence data collected in surveys is of importance to any national control programmes seeking to eliminate blinding trachoma in accordance with the WHO definitions ( prevalence of TT less than 1 case per 1000 total population , and prevalence of TF in 1–9 year-old children less than 5% ) . [3] Data collected in surveys also allow control efforts to be directed to trachoma endemic areas . However , many studies have demonstrated that the prevalence of detected ocular C . trachomatis infection is lower than the prevalence of active trachoma , especially in mass treated and low prevalence settings . [6] , [7] , [8] , [9] The Gambia has not collected infection data in its national surveys but these data may lead to a better understanding of the disease's epidemiology with potential implications for the introduction of trachoma control interventions . Mass antibiotic treatment for trachoma is likely to be effective in treating communities with C . trachomatis infection , but of questionable value where no infection can be demonstrated . [10] , [11] We aimed to estimate the prevalence of active trachoma and ocular C . trachomatis infection in children aged less than 10 years , a decade after the last national survey , in Lower River Region ( LRR ) and North Bank Region ( NBR ) . These two regions had the highest prevalence of active trachoma in children aged 0–9 years in 1996 ( 11 . 5% and 7 . 7% , respectively ) . [5] This study reports the results of this cluster-randomised cross-sectional survey in The Gambia , conducted in 2006 .
Research was done in accordance with the declaration of Helsinki . Ethical approval was obtained from the London School of Hygiene and Tropical Medicine ( LSHTM ) , UK , Ethics Committee and The Gambia government/Medical Research Council ( MRC ) Joint Ethics Committee , The Gambia . Written ( thumbprint or signature ) informed consent was obtained from the guardians of all children . The estimated populations of NBR and LRR in 2003 were 172 , 835 and 72 , 167 , respectively . [12] The sample size was constructed for 80% power , with 95% confidence , to detect a region prevalence of TF in 0–9 year olds above 10% if the true prevalence was 12% , or below 10% if the true prevalence was 8% , allowing for the geographical clustering of TF cases by assuming a design effect of 4 . The design effect shows the effect of the study design on the estimate's variance , and increases with cluster sample size and within-cluster homogeneity . Since the distribution of trachoma is clustered , and a cluster sampling strategy was employed , a design effect of 4 adjusts the sample size to obtain the same estimate precision as if the disease were homogenously distributed and a simple random sample had been taken . [13] , [14] The survey methods have been described in detail elsewhere . [15] Briefly , a two-stage cluster random sampling strategy with probability of selection proportional to size was employed . Sixty-one enumeration areas ( EAs ) , geographical units of approximately the same population size , were chosen at random from the two regions ( 42 in NBR and 19 in LRR ) . EAs are classified in The Gambia's census as rural or urban . [12] A household head list was made for each selected EA and a random selection of households was made by dividing random numbers generated in Excel ( MS Excel v2000 ) with the reciprocal of the number of households in the EA . Households were selected sequentially from the top of this list , duplicates excluded , until 50 children aged under 10 years were included , based on the assumption that the average household size was 7 . 5 people , of whom about half would be aged 0–9 years . Reserve households were included in case 50 children could not be obtained with the first set number of selected households . The day before examination , an enumeration team censused the de facto population ( those who had slept in the household the night before ) of the selected households , recording name , alias names , age and sex . The enumeration team identified the children aged under 10 years and informed the households that the examination team would be coming the following day . The examination team examined the children for clinical signs of trachoma in the same order as they had been selected until a total of 50 children per EA was obtained . Two experienced trachoma clinical graders were used , who had successfully achieved a chance corrected agreement ( Cohen's kappa statistic [16] ) with the standard over the scoring of each sign ( TF , TI , TS , TT ) of 0 . 8 or greater in validation exercises with an experienced observer ( RLB ) . These exercises were conducted both in the field and using the WHO trachoma grading slides , and were further supplemented using an in-house slide and photograph collection . Both eyes were graded using a 2 . 5× magnifying loupe and torchlight . Grading was according to the WHO simplified grading system and the results of the worst eye were reported . [17] In LRR only , ocular Dacron swabs ( Hardwood Products Company , Gilford , ME , USA ) were taken from the everted tarsal conjunctiva of the child's right eye , using a highly standardised technique . [11] All individuals diagnosed with trichiasis were referred to the nearest health centre for surgery provided free of charge . Individuals with TF or TI were offered treatment with tetracycline eye ointment . Other ocular morbidities were managed according to NECP guidelines . The swabs were kept cool in the field , frozen within 10 hours , and processed by Amplicor Polymerase Chain Reaction ( PCR ) assay ( Roche Molecular Systems , Branchburg , NJ , USA ) at MRC Laboratories , The Gambia . A panel of samples was successfully completed by the laboratory technicians , who had been masked to the sample results . In order to demonstrate that adequate ocular specimens were taken in the field , all samples were also tested for the presence of human-specific hypervariable 1 ( HV1 ) D-loop region mitochondrial DNA ( mtDNA ) , using D-loop HV1 upper primer L15997 , 5′-CAC CAT TAG CAC CCA AAG CT-3′ and D-loop HV1 lower primer H16236 , 5′-CTT TGG AGT TGC AGT TGA TG-3′ ( Sigma-Genosys , Gillingham , UK ) . [18] The reaction mixture contained 2 µL of Amplicor extract , 1X Quantitect SYBR Green PCR mastermix ( Qiagen , Crawley , UK ) , each primer at 0 . 3 µM , and was made up to 10 µL with DEPC-treated sterile water . After denaturation at 95°C for 15 minutes , samples were subjected to 45 cycles of thermal cycling ( 15 seconds at 95°C , 30 seconds at 60°C and 90 seconds at 72°C ) on a Rotor-Gene RG3000 ( Qiagen , Crawley , UK ) . The samples were analysed by melt analysis ( 72–95°C ) , which consisted of a single hold/acquisition for 45 seconds at 72°C and then 46 hold/acquisitions at increments of +0 . 5°C . The HV1 D-loop region amplicon melted at a mean temperature of 81 . 2°C ( range 80 . 78–81 . 37 , mean standard deviation 0 . 41 ) . During optimisation , we used gel electrophoresis to confirm the presence of a discrete 278 bp amplicon . Results were double-entered by different entry clerks and verified in Microsoft Access ( MS Access v2000/2003XP ) . Data cleaning and analyses were performed in Stata ( v9 . 2 , STATA Corp . , College Station , TX , USA ) . Any discrepancies after verification and cleaning were checked against the original paper forms . The unadjusted prevalence of TF in 1–9 year-olds is presented as a percentage , as standard practice for a trachoma control programme , at EA , district , and regional levels . At regional level , the design effect and adjusted estimates of prevalence and corresponding 95% confidence bounds were obtained , accounting for the two-stage sampling framework and population size . At EA and district levels , exact binomial confidence intervals were computed around the prevalence , as the study was designed to have an accurate estimate of TF prevalence only at the region level . Associations between the prevalence of TF and districts were tested using the chi-squared ( χ2 ) statistic . The spatial distribution of TF prevalence by district and EA was presented graphically using ArcView 3 . 3 software ( Environmental Systems Research Institute , Inc . Redlands , CA , USA ) .
Overall , TF was found in 310 ( 10 . 4% ) children aged 0–9 years ( Table 1 ) . Children aged 3–5 years had the highest prevalence of TF ( 17 . 0% ) , followed by those aged 1–2 years ( 13 . 1% ) . Children under one year old and those aged 6–9 years had TF prevalences of 5 . 6% and 6 . 3% , respectively . Only three ( 0 . 1% ) children had TI ( two in NBR and 1 in LRR ) , and two of these also had TF . The left and right eye were concordant for TF ( both clinically normal or both with TF ) in 2933 ( 98 . 1% ) children . Considering the 1–9 year age group , the overall TF prevalence ( the WHO indicator ) was 10 . 7% ( 95% CI 8 . 7–13 . 1 ) . The prevalence of TF in LRR and NBR , respectively , was 12 . 3% and 10 . 0% ( Table 2 ) . The design effect for heterogeneity among EAs was 3 . 474 overall; 3 . 322 for LRR and 3 . 561 for NBR . The district prevalence varied significantly in LRR ( p = 0 . 006 ) from 5 . 7% to 18 . 2% , and in NBR ( p = 0 . 001 ) from 5 . 6% to 15 . 1% . There was a trend towards higher rates in the eastern districts compared to those in the west in LRR ( χ2 test for trend p<0 . 001 ) but not in NBR ( p = 0 . 287 ) ( Figure 1 ) . Prevalence also appeared to vary dramatically between EAs in the same district . For example , in Jarra West in LRR , prevalence ranged from 0% to 36 . 7% , and in Central Baddibu in NBR , prevalence ranged from 6 . 3% to 34 . 0% . Comparing the results with those from 1996 indicates little change overall: LRR increased slightly from 11 . 5% to 11 . 9% , and NBR increased from 7 . 7% to 9 . 7% . [5] . There were three urban EAs in LRR and nine in NBR . Overall , 53 of 552 ( 9 . 6% , 95% CI 7 . 3–12 . 4 ) urban children aged 1–9 years had TF , compared with 245 of 2225 rural children ( 11 . 0% , 95% CI 9 . 7–12 . 4 ) . There was no evidence of variation in TF prevalence between urban and rural EAs in either LRR ( p = 0 . 395 ) or NBR ( p = 0 . 854 ) . Only 3 of the 950 samples collected in LRR tested positive for ocular C . trachomatis by Amplicor . Sample was available for the testing of human-specific hypervariable D-loop region mtDNA in 942 of the 947 Amplicor-negative samples , of which 937 ( 99 . 5% ) tested positive . Thus , the prevalence of ocular C . trachomatis infection in samples positive for human mtDNA was 0 . 3% ( 3/940 ) . Three field air controls ( swabs waved in the air during fieldwork to control for any field contamination ) were also tested for both ocular C . trachomatis infection and human mtDNA , giving negative results . The three children who tested positive for ocular C . trachomatis infection by Amplicor were aged 9 , 8 and 5 years old . One of these was in Jarra West , in an EA with a TF prevalence of 8 . 0% in 0–9 year-olds , and two were from Kiang East , in an EA with a TF prevalence of 16 . 0% in 0–9 year-olds . Two Amplicor positive children were clinically normal according to the WHO simplified grading system definition , and the other had bilateral TF . All three Amplicor positives were confirmed positive by a real-time PCR targeting the ompA gene . [15]
The results of this survey , covering two regions of The Gambia with a population of approximately 245 , 000 people , suggest that the overall prevalence of active trachoma in children aged 0–9 years has not noticeably changed since the 1996 national survey . There is some evidence of a secular trend in trachoma prevalence in The Gambia prior to 1996: in one rural village active disease prevalence reduced from the 65 . 7% documented in 1959 to 2 . 4% in 1996 in the absence of direct control interventions [19] . However there is no evidence that such a trend continued in LRR and NBR between 1996 and 2006 despite trachoma control efforts being in place in these areas . In contrast to TF , the prevalence of C . trachomatis infection was extremely low in this survey , with only 3 of the 940 samples from LRR that were positive for human mtDNA also testing positive for C . trachomatis . No infection data were collected in the previous Gambian national surveys so no trends over time can be estimated . In addition , other countries have not collected data on ocular C . trachomatis infection when conducting population-based surveys of active trachoma , so between-country comparisons cannot be made . However , our results show a lower prevalence of infection compared with research studies conducted in The Gambia over the last two decades . [10] , [20] , [21] , [22] In 1991 , the prevalence of infection among all inhabitants in two Gambian villages was 17 . 2% . [20] In 1999 , the overall prevalence in eight Gambian villages was 35 . 9% . [21] Burton et al . reported an overall prevalence of 7 . 2% in 14 Gambian villages in 2003 , [10] and 19 . 8% in two villages in 2006 . [22] It is important to note that these projects were focused on villages with notable trachoma public health problems usually identified by finding a high prevalence of active trachoma on prior screening , with later laboratory demonstration of high rates of ocular C . trachomatis infection . In these villages , therefore , higher prevalence rates may have been recorded than would be estimated in the national Gambian population . A disparity between the prevalence of active trachoma and of ocular C . trachomatis infection has been reported by others , especially in communities where the prevalence of disease is declining , or in communities that have received mass treatment with antibiotics . [6] , [9] , [10] , [20] , [23] , [24] , [25] , [26] , [27] , [28] Solomon et al . observed a lag time of several years between infection elimination and the prevalence of TF falling below 5% in a mass-treated Tanzanian community . [7] , [26] In a low prevalence region of Nepal , 6 . 3% of children aged 1–10 years had clinical signs but no C . trachomatis was detected . [6] Our results suggest that trachoma may be in decline in The Gambia , with the clearance of clinical signs lagging behind that of infection . That TF cases remain clustered by EA with a design effect of 3 . 474 may suggest that previous transmission is being captured , or that undetected infections may have existed in close contacts . Despite diagnostic assays within the last decade becoming increasingly more sensitive , the prevalence of detectable C . trachomatis infection is now very low . Our observation that the vast majority of cases were TF , with only three cases of TI , supports this hypothesis , as TI cases are more likely to have detectable C . trachomatis infection and with a higher organism load . [11] , [21] , [29] , [30] In the one TI case tested here , C . trachomatis infection was not detected . Clinically , our impression was that many of the TF cases were rather mild , in the senses that cases with very large numbers of follicles were rarely present , and also that ‘minor signs’ of trachoma such as active pannus and limbal follicles[31] , were rare . Possibly therefore increasing the 10% TF threshold for intervention , increasing the number of central zone follicles required to diagnose TF , and including minor signs of trachoma in the assessment , might all improve the ability of clinical signs to predict communities or districts with significant ocular C . trachomatis infection . Much more data would however be needed to justify any such departure from the current simplified survey methods . This association between prevalence of disease , severity of disease , and load of C . trachomatis infection could provide another explanation for the observed disparity between prevalence of clinical signs and detected ocular C . trachomatis infection . The bacterial load may have been too low for detection by Amplicor PCR . However , the detection limit of Amplicor PCR has been placed at about 2–4 elementary bodies ( or 20–40 plasmid copies ) per 100 µl , [32] and thus even a few bacteria collected on the swab should have yielded a positive result . The use of a Dacron polyester-tipped swab may have lowered the PCR detection level as it absorbs approximately 200 µl of Amplicor lysis buffer and therefore a higher volume is required to elute the sample . The use of a flocked swab could help overcome this . [33] In terms of clinical grading , the two examiners were experienced and had fulfilled the criteria of the validation sessions . Swabs were taken according to a standardised protocol [11] by experienced ophthalmic nurses , and 99 . 5% ( 937/942 ) of the samples tested positive for human-specific mtDNA . Thus , we believe that the observed prevalence of active trachoma , as well as the low prevalence of ocular C . trachomatis infection , is genuine . The follicular conjunctivitis ( TF ) observed in this study could alternatively be due to organisms other than C . trachomatis . Potential causes include adenovirus , Herpes simplex virus , Epstein Barr virus , Molluscum contagiosum , Moraxella spp . and other species of Chlamydia , such as C . pneumoniae , which was first isolated from the eyes of children with trachoma . [6] , [34] , [35] , [36] , [37] , [38] Further laboratory investigations to test the C . trachomatis negative swabs from clinically active children for these other organisms may be warranted . Our results have two important implications . The first relates to policy decisions regarding control efforts , particularly now that The Gambia has for the first time received a donation of azithromycin for trachoma control . Currently , according to WHO criteria , districts and communities with a TF prevalence greater than or equal to 10% in children aged 1–9 years should receive mass treatment annually , in addition to the ‘F’ and ‘E’ components of the SAFE strategy , for at least three years , until the prevalence of TF falls below 10% . [39] According to these criteria , 3 districts and 10 EAs in LRR , and 3 districts and 17 EAs in NBR , should be treated . However , our sampling strategy was designed to have an accurate estimate of TF prevalence at the region level and not at the community or district level . The size of Gambian regions is similar to the average size of a district in many countries ( 100 , 000 to 150 , 000 people ) . On this basis , our results suggest that the whole of LRR and NBR should undergo mass treatment according to WHO criteria , as their respective TF prevalence in 1–9 year-olds is 12 . 3% and 10 . 0% . Conversely , the very low prevalence of C . trachomatis infection indicates that treatment with antibiotics is almost certainly not necessary in many of these communities , since transmission of C . trachomatis is now presumed to be rare . Treatment decisions based on clinical signs of trachoma may lead to unnecessary mass treatment of whole regions or districts , wasting the scarce resources available , although there may be benefits beyond trachoma control to the mass distribution of azithromycin , such as a decline on overall mortality as shown in Ethiopia . [40] A cheap , rapid , point-of-care test that can detect ocular C . trachomatis infection could allow policy makers to devise treatment strategies based on the prevalence of infection . At present , no such test is available despite previous encouraging results [41] . Alternatively , the use of laboratory tests , such as PCR , has been suggested as a means of detecting ocular C . trachomatis infection . Despite reduced costs through specimen pooling[42] , national programmes are unlikely to adopt these assays as they require expensive reagents , electricity-dependent equipment , and highly trained technicians . [43] In the absence of a test for infection for use by national programmes , our observation that TF clustered by EA supports the strategy of treating households containing someone with active trachoma in order to reach infected but clinically normal individuals . [10] The second implication concerns itself with the question of whether children with clinical signs of trachoma in the absence of C . trachomatis infection will develop blinding sequelae . The epidemiology of trachoma is not sufficiently understood to enable this question to be answered for certain , although the evidence that repeated infection is required for disease progression suggests that it is unlikely . [44] , [45] This study shows a large discrepancy between the prevalence of clinical signs and of C . trachomatis infection in two Gambian regions in the absence of a national mass treatment programme . The prevalence of TF in children aged less than 10 years in the two Gambian regions that were sampled remains largely unchanged from the prevalence 10 years ago , at around 10% . In contrast , the prevalence of C . trachomatis infection is low ( 0 . 3% ) . Our results indicate that The Gambia may not only be on course for certification of blinding trachoma elimination according to the WHO definition ( prevalence of TT less than 1 case per 1000 total population , and prevalence of TF in 1–9 year-old children less than 5% ) [3] , but also on course to eliminate the ocular strains of C . trachomatis prior to any mass azithromycin distribution . | Trachoma is the leading infectious cause of blindness worldwide , and is mainly found in tropical and poor countries . It is caused by infection of the eyes with the bacterium Chlamydia trachomatis . However , sometimes the clinical signs of disease can be present without infection being detected . Control efforts involve surgery , antibiotic treatment , face washing , and environmental improvement for better hygiene . Surveys of trachoma help countries to know whether and where they should implement control interventions . The Gambia is found in West Africa and has suffered from trachoma for decades . We conducted a survey of two Gambian regions to look at how much trachoma disease and C . trachomatis infection there is in the eyes . We found that although there was enough disease ( ≥10% ) to warrant antibiotic treatment for everyone in the regions , there was nearly no infection ( 0 . 3% ) . This means that using clinical signs alone to make treatment decisions in low prevalence settings like The Gambia can lead to the waste of scarce resources . Our results also suggest that since less than 1% of children are infected with C . trachomatis , The Gambia is on course to achieve the World Health Organization's aim of eliminating blinding trachoma by the year 2020 . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases",
"ophthalmology/eye",
"infections",
"public",
"health",
"and",
"epidemiology",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"public",
"health",
"and",
"epidemiology/epidemiology",
"public",
"health",
"and",
"epidemiology/infectious",
"d... | 2009 | Active Trachoma and Ocular Chlamydia trachomatis Infection in Two Gambian Regions: On Course for Elimination by 2020? |
The mucosal surface of the intestinal tract represents a major entry route for many microbes . Despite recent progress in the understanding of the IL-21/IL-21R signaling axis in the generation of germinal center B cells , the roles played by this signaling pathway in the context of enteric microbial infections is not well-understood . Here , we demonstrate that Il21r-/- mice are more susceptible to colonic microbial infection , and in the process discovered that the IL-21/IL-21R signaling axis surprisingly collaborates with the IFN-γ/IFN-γR signaling pathway to enhance the expression of interferon-stimulated genes ( ISGs ) required for protection , via amplifying activation of STAT1 in mucosal CD4+ T cells in a murine model of Citrobacter rodentium colitis . As expected , conditional deletion of STAT3 in CD4+ T cells indicated that STAT3 also contributed importantly to host defense against C . rodentium infection in the colon . However , the collaboration between IL-21 and IFN-γ to enhance the phosphorylation of STAT1 and upregulate ISGs was independent of STAT3 . Unveiling this previously unreported crosstalk between these two cytokine networks and their downstream genes induced will provide insight into the development of novel therapeutic targets for colonic infections , inflammatory bowel disease , and promotion of mucosal vaccine efficacy .
Several microbial pathogens elicit the type I ( e . g . IFN-α , IFN-β ) , type II ( IFN-γ ) or type III ( e . g . IFN-λ ) interferons , leading to the transcription of several hundred interferon-stimulated genes ( ISGs ) [1 , 2] . The activation of each of these interferon systems induces a distinct but partially overlapping set of signature ISGs which have been reported to contribute important roles to host defense against a wide variety of pathogens , including viruses , bacteria , and parasites , by directly targeting genes conferring resistance to infection [3 , 4] . The essential requirement of ISGs in host immunity has been shown by the fact that mice deficient in one or more ISGs are more susceptible to infections with multiple viral and bacterial pathogens [5–7] . The role of ISGs in host defense at mucosal surfaces of the gut is not fully understood . In an attempt to elucidate the mechanisms by which infection is controlled at mucosal surfaces of the small intestine , the cooperation between two other cytokine networks , interferon lambda ( IFN-λ ) ( not IFN-γ ) and IL-22 , was shown to be required for optimal control of viral replication in a mouse model of rotavirus infection [6] . However , the role of the IL-21/IL-21R signaling axis in colitis of any origin and in collaboration with IFN-γ is not known . Infection of mice with the murine enteric pathogen , Citrobacter rodentium , is considered a robust model to study host immune response to minimally-invasive gut pathogens at the mucosal surfaces of the large intestine [8–10] . This extracellular bacterial pathogen shares virulence features with the closely related human enteropathogenic Escherichia coli ( EPEC ) and enterohemorrhagic Escherichia coli ( EHEC ) , making it an ideal surrogate model to study the human disease [8 , 10] . Several lines of evidence suggest that the immune response to the human EPEC and EHEC relies on both innate [11–13] and adaptive [14–17] arms of the immune system . The C . rodentium colonization elicits a robust TH1 response , characterized by the predominant production of IFN-γ , IL-12 [14] and TNF-α [18] , as well as a TH17 response producing IL-17 [19] . The absolute requirement of TH1 responses in host defense against C . rodentium has been reinforced by the observations that CD4+ T cell-deficient mice ( CD4-/- ) ( but not CD8+ T cell-deficient mice ) IFN-γ knockout ( Ifng-/- ) mice and IL-12-deficient mice ( IL-12p40-/- ) showed greater susceptibility to enteric infection , increased systemic dissemination of the bacterium and enhanced pathology after C . rodentium infection [16 , 17 , 20] . IL-21 binding to its cognate receptor , IL-21R , results in the activation of Janus kinase 1 ( JAK1 ) and JAK3 and the subsequent phosphorylation of signal transducer and activator of transcription ( STAT ) proteins , mainly STAT3 but also STAT1 and STAT5 [21 , 22] . The activated phospho-STAT proteins dimerize and translocate into the nucleus , bind to the interferon ( IFN ) -γ-activated sequence ( GAS ) motif and initiate a transcription program that includes some IL-21 target genes [23] . Although the cascade of events occurring after the IL-21-activation of STAT3 is well-studied , it is still not fully understood how the activation of STAT1 via IL-21 influences downstream target genes [23] . However , it is interesting that IL-21 via STAT1 can augment expression of both Tbx21 and Ifng gene expression as well as expression of certain interferon-regulated genes Ifit1 and Ifit2 , that are also IL-21 targets , and that STAT3 activation diminishes these effects , either in mice or humans [24] . Despite the critical requirement of the IL-21/IL-21R signaling axis in the generation of germinal center B cells [25] , the roles played by this signaling pathway in the context of enteric microbial infection is not well-characterized . Based on this , we investigated the role of IL-21/IL-21R axis , in collaboration with the type II interferon , IFN-γ , in protection against enteric microbial infection with C . rodentium in the colon . We found that an intact IL-21/IL-21R axis was required for resistance against and clearance of enteric infection with this pathogen and that activated CD4+ T cells were the exclusive expressors of IL-21 following infection with C . rodentium in the distal colon . The CD4+ T cell-derived IL-21 curtailed enteric infection and also contributed to C . rodentium-induced inflammation and pathology at the mucosal surfaces of the colon . We also found that IL-21 acted in concert with IFN-γ to optimally activate STAT1 in CD4+ T cells and to promote subsequent optimal expression of ISGs in the distal colon . These events were independent of STAT3 . Our findings revealed a previously unknown effector function for the IL-21/IL-21R signaling axis in amplification of ISG expression and the modulation of host response to microbial infection at the mucosal surfaces of the gut . The understanding of the mechanisms by which the IL-21/IL-21R signaling axis regulates intestinal epithelial integrity and host immunity after infection with minimally-invasive gut pathogens ( e . g . Escherichia coli ) will provide insights into novel preventive and therapeutic targets for the control of human infections with enteric bacterial pathogens . Such collaboration between IL-21 and IFN-γ also provides mechanisms by which the IL-21/IL-21R signaling axis regulates inflammation in the colon and provides insights into novel preventive and therapeutic targets for inflammatory conditions in humans , including inflammatory bowel disease , as well as inflammation-induced cancers .
Using the murine intestinal pathogen , C . rodentium , we investigated the requirement of the IL-21/IL-21R signaling axis in protection against mucosal microbial infection in the gut . Our findings indicated that Il21r-/- mice had significantly higher ( 2 logs ) bacterial burden in the feces as compared with WT controls , both early and late in the infection , although the peak bacterial load was comparable ( Fig 1A and 1B ) . While the WT controls were able to control the infection with C . rodentium by day 21 p . i . , Il21r-/- mice had an impaired ability to clear the enteric infection with this pathogen ( Fig 1A and 1B ) . The differences in fecal bacterial burdens were noticeable as early as day 2 p . i . in Il21r-/-mice , suggesting an important roleplayed by this signaling pathway in early host protection events to this enteric pathogen . Although the WT and Il21r-/- mice were cohoused for 2 weeks to equilibrate microbiota , we also bred matched WT heterozygotes and homozygous Il21r-/- mice from the same set of homozygous Il21r-/- females by crossing with WT or Il21r-/- males , and then kept the litters nursing together until weaned , so that they would obtain the same microbiota from their mothers intrapartum and during nursing/foster nursing . Consistently , mice homozygous for the targeted mutation of IL-21R ( Il21r-/- ) showed significantly higher bacterial burdens and delayed clearance of C . rodentium infection as compared with their heterozygous littermate controls ( Il21-/+ ) that were bred together ( S1 Fig ) , confirming that the difference in susceptibility to C . rodentium infection was not due to distinct microbiomes . Furthermore , comparable infection kinetics were observed following infection with an OVA-expressing C . rodentium , indicating that overexpression of a plasmid carrying the chicken ovalbumin did not alter the infectivity , or the ability to colonize the mucosal surfaces , of OVA-Citrobacter as compared with WT-Citrobacter ( S2 Fig ) . Collectively , these findings were consistent with one previous report indicating a role played by the IL-21/IL-21R axis in protection against C . rodentium [26] . Despite effective bacterial replication and significantly higher bacterial burden in Il21r-/- mice , histology demonstrated only a moderate increase in inflammatory cell recruitment , predominantly in the mucosa , and mild hyperplasia accompanied with loss of goblet cells in the distal colons of Il21r-/- mice 9 days after infection with C . rodentium , the time of peak infection when bacterial loads are not significantly different between WT and Il21r-/- mice ( Fig 1A–1C ) . In WT controls increased numbers of inflammatory cells were observed in the mucosa extending into the submucosa of the distal colon . Epithelial cell erosion and ulcerations , and submucosa edema , were more prominent in WT mice than in their Il21r-/- counterparts at day 9 p . i . ( Fig 1C ) . Crypt hyperplasia and loss of goblet cells accompanied by significant submucosal edema with significantly higher perivascular inflammatory cells were more noticeable in the distal colon of WT mice 9 days after infection than in the Il21r-/- mice ( Fig 1D and 1E ) . However , the percentage and the absolute numbers of those cells were comparable in whole colons between the two genotypes at days 3 and 9 after C . rodentium infection ( Fig 1F and S3A–S3G Fig ) . Overall , despite having higher bacterial burden , the Il21r-/- mice appeared to have less inflammation , indicating that IL-21 response was a critical factor in the inflammatory response and at least part of that inflammatory response may be necessary for control of the bacterial load . Using an ex-vivo organ culture system , we examined the kinetics of IL-21 production by ELISA in the distal colon of WT mice following C . rodentium infection . As shown in Fig 2A , the peak of IL-21 production in the distal colon of WT mice infected with C . rodentium occurred 9 days p . i . , when the infection was at its peak . Likewise , similar kinetics were observed for other cytokines important for protection against C . rodentium infection , including IFN-γ , IL-17A and IL-22 in the distal colon of WT ( Fig 2A ) . To further investigate the extent to which immune or non-immune cell types in the distal colon contributed to the expression of IL-21 after C . rodentium infection , we sorted cells in the colon into hematopoietic ( CD45+EpCAM- ) or non-hematopoietic ( CD45-EpCAM+ ) cells and sorted for immune cell types . Using the Nanostring method , we observed that IL-21 was almost exclusively expressed by mucosal CD4+ T cells ( CD45+EpCAM-CD3+CD4+ ) in WT mice 9 days after C . rodentium infection , while other cells of hematopoietic origin , such as natural killer ( NK ) cells , dendritic cells ( DCs ) , neutrophils , and macrophages , did not express significant levels of IL-21 transcripts following infection ( Fig 2B ) . Furthermore , mucosal CD4+ T cells expressed higher levels of transcripts for IL-21R as compared with other immune and non-immune cells after infection with C . rodentium ( Fig 2C ) . However , there was some IL-21R expression by innate cells such as NK cells , neutrophils , macrophages and DCs that may contribute to the difference in early control of C . rodentium at day 2–3 pi . ( Fig 1A and 1B ) . Interestingly , the colonic intestinal epithelial cells ( IECs ) expressed neither detectable transcripts for IL-21 ( Fig 2B ) nor mRNA for IL-21R following infection ( Fig 2C ) . The latter observations were consistent with lack of IL-21R expression in a C57BL/6 colon carcinoma cell line ( MC-38 ) model of colonic IECs ( S4 Fig ) . We performed principal component analysis ( PCA ) of gene expression profiles in the whole distal colon of WT and Il21r-/- mice using Nanostring . Our analysis showed four distinct clusters between uninfected and infected WT and Il21r-/- mice 9 days after infection with C . rodentium ( Fig 3A ) . While the uninfected WT and Il21r-/- mice clustered closed to each other , the infected WT and Il21r-/- mice formed two distinct clusters that were separated far from each other , indicating differences in their gene expression profiles . Several lines of evidence suggest that colonization with C . rodentium elicits a robust , highly polarized TH1 response in the colon , as shown by increased expression of IFN-γ , tumor necrosis factor ( TNF ) -α and IL-12 [14] . The critical requirement of IFN-γ during C . rodentium infection has been highlighted by the observations that IFN-γ knockout ( Ifnγ-/- ) mice had an impaired ability to clear infection [14 , 17] . In this model , IFN-γ produced by antigen-experienced CD4+ T cells mediates the mucosal immune response to C . rodentium and its subsequent eradication [20] . Our findings demonstrated that type I- and type II-specific , but not type III-specific , ISGs were induced after C . rodentium infection in the distal colons of WT mice ( Fig 3B , 3C and 3D and S5 Fig ) . Considering the key requirement of IFN-γ and downstream ISGs in host defense against C . rodentium infection , we next investigated the contribution of IL-21/IL-21R signaling axis to the optimal expression of ISGs in the colon of Il21r-/- mice under homeostatic conditions as well as 9 days after enteric infection . Our findings demonstrated that most genes impaired in the whole distal colon of Il21r-/- mice infected with C . rodentium were known ISGs ( Fig 3C and 3D ) . In each case , the fold increase during infection compared to uninfected controls was substantially less in the Il21r-/- mice than in the WT controls ( Fig 3C ) , and in most cases the increase in Il21r-/- mice was not significant , and the increase in the WT was significantly greater than the increase in Il21r-/- mice . Interestingly , the expression of both the type I- and type II-specific ISGs were impaired in the whole colon of Il21r-/- mice 9 days after C . rodentium infection ( Fig 3D and 3E ) . To further explore the extent of impaired expression of ISGs in the absence of an intact IL-21/IL-21R axis specifically in CD4+ T cells , FACS-sorted CD4+ T cells ( the main producers of IL-21 and main expressors of IL-21R , Fig 2B and 2C ) were isolated from the distal colon lamina propria ( LP ) of Il21r-/- mice and WT controls 9 days after C . rodentium infection and analyzed by Nanostring . Consistently , our findings indicated that the majority of genes impaired in CD4+ T cells isolated from the LP of Il21r-/- mice were known ISGs ( Fig 3F and 3G ) , indicated by red bars ( Fig 3G ) . Gene ontology analysis of processes enriched in mucosal CD4+ T cells impaired from the distal colons of Il21r-/- mice identified genes with a wide range of functions ( Fig 3H ) . Likewise , the expression of type I- and type II-specific , but not type III-specific , ISGs by CD4+ T cells was impaired in Il21r-/- mice as compared with WT controls ( Fig 3I ) . In view of these results , we hypothesized that IL-21 and IFN-γ produced in response to C . rodentium infection may act in concert for the optimal expression of ISGs in the colon and are required for the control of C . rodentium infection in vivo . To experimentally test the hypothesis , we treated naïve splenocytes with recombinant murine IL-21 or IFN-γ alone or in combination for 24 hr and the expression of representative ISGs LAG3 and granzyme A by CD4+ T cells was measured by flow cytometry . When cells were treated with a combination of recombinant murine IL-21 or IFN-γ the expression of representative ISGs by CD4+ T cells isolated from WT mice was significantly upregulated compared to cells treated with either IL-21 or IFN-γ alone ( Fig 3J for LAG-3 and S6 Fig for granzyme A ) . Interestingly , almost all CD4+ T cells positive for IFN-γR were also positive for IL-21R as well ( S7 Fig ) , so the same cells could respond to both cytokines , not a sum of some cells that could respond to IFN-γ and some to IL-21 . To further address whether differences in the gene expression profiles between WT and Il21r-/- mice did not merely reflect the severity of inflammation , we investigated the expression of a representative ISG , LAG-3 , expressed by CD4+ T cells isolated from the colonic LP of naïve ( uninfected ) WT and Il21r-/- mice . Remarkably , Il21r-/- mice expressed significantly lower surface expression of LAG-3 ( p = 0 . 001; n = 7 animals ) even in the absence of colonic inflammation ( S8 Fig ) . These findings indicate that the differences in the gene expression profiles between WT and Il21r-/- mice are not likely a consequence of the inflammation severity . Collectively , these findings indicate previously unrecognized collaboration between IL-21 and the IFN-γ signaling pathway to optimally express ISGs and a requirement for an intact IL-21/IL-21R signaling axis for the optimal expression of ISGs by CD4+ T cells . The absolute requirement of CD4+ T cell responses during C . rodentium infection has been reinforced by the observations that CD4+ T cell-deficient mice ( CD4-/- ) , but not CD8+ T cell-deficient mice ( β2m-/- ) , showed greater susceptibility to C . rodentium-induced colitis and increased systemic dissemination of the bacterium to extra-intestinal sites [16] . Consistent with our findings that IFN-γ , but not IFN-α , was the dominant interferon expressed in the colon after infection with C . rodentium ( Fig 4A ) , significantly lower concentrations of IFN-α than IFN-γ were detected in the colon of both infected WT and Il21r-/- mice ( Fig 4B ) . Collectively , these findings suggested that IFN-γ is the predominant interferon produced in the colon in response to C . rodentium and that the lack of an intact IL-21/IL-21R signaling axis does not negatively affect the IFN-γ expression and production in response to infection in Il21r-/- mice . Considering that both type I- and type II-specific ISGs were impaired in infected Il21r-/- mice , we investigated which type of interferon was required for the control of infection . Type I interferons have been studied mostly in the context of viral infections [7] . The roles played by type I interferons in non-viral infections , including bacterial infections have recently been investigated [7 , 27 , 28] . We further experimentally determined roles played by type I interferons in host protection following C . rodentium infection . Consistent with these results , mice deficient in interferon-α/β receptor ( Ifnar-/- ) had bacterial burdens comparable to those of their WT controls and were able to efficiently clear infection after oral challenge ( Fig 4C ) . However , mice deficient in IFN-γ ( Ifng-/- ) showed a delayed clearance similar to that seen in Il21r-/- mice and exhibited significant weight loss early in the course of C . rodentium infection ( Fig 4D and 4E ) . These findings suggested that type I IFNs played minimal , if any , roles in protection against C . rodentium infection in mice and that the type II IFN ( i . e . IFN-γ ) contributed significantly to host protection . The earliest step at which IL-21 could influence production of ISGs is in the production of IFN-γ itself , so we addressed these levels in the Il21r-/- mice . It has been shown by other investigators that Il21r-/- mice express significantly higher levels of IFN-γ in the colon LP compared with WT counterparts during dextran sulfate sodium ( DSS ) -induced colitis [29] . Our findings demonstrated that IFN-γ was the dominant interferon , mainly expressed by CD4+ T cells . However , dendritic cells , to a much lesser extent , expressed IFN-β , but not IFN-α or IFN-λ or much IFN-γ ( Fig 4A ) . Consistent with these findings , by using an ex-vivo organ culture system we found that the lack of the IL-21/IL-21R signaling axis did not negatively affect the production of IFN-γ in the distal colon of Il21r-/- mice , as evidenced by significantly higher levels of IFN-γ production in the distal colon as compared with WT controls 9 days after C . rodentium infection ( Fig 4B ) . Although the expression of IFN-γ by gut-associated CD4+ T cells isolated from Il21r-/- mice was impaired ( ~3 fold ) as compared with WT controls , higher expression of the cytokine by NK cells ( ~3 fold ) could explain higher levels of IFN-γ observed in the whole distal colon of the Il21r-/- mice ( S9 Fig ) . Likewise , significantly higher levels of IL-17A were noted in the distal colon of Il21r-/- mice as compared with WT controls at that time ( Fig 4B ) . This contrasts with the poorer control of the infection , which should benefit from increased IFN-γ and IL-17 . No significant differences were observed in the intracellular expression of IFN-γ by CD4+ T cells isolated from the colonic LP of either naïve Il21r-/- mice or naïve WT controls ( Fig 4F ) . Intracellular staining for IFN-γ expression by ovalbumin-specific mucosal CD4+ T cells in WT mice after OVA-Citrobacter infection demonstrated that CD4+ T cells are a major source of Citrobacter-induced IFN-γ following infection in the intestine ( Fig 4G ) . Because IFN-γ production is higher , not lower , in the colons of Il21r-/- mice , the reduction in ISGs cannot be due to a decrease in IFN-γ itself or to the need for IL-21 signaling to optimally induce IFN-γ . Thus , the effect on ISGs must be downstream of levels of IFN-γ itself . We therefore asked whether the effect on ISGs was due to a decrease in IFN-γ receptor expression , the next step that might account for the lower levels of ISGs . The lack of the IL-21/IL-21R signaling axis did not impair the expression of IFN-γR1 and IFN-γR2 ( CD119 ) by CD4+ T cells isolated from the LP of Il21r-/- mice 9 days after C . rodentium infection as compared with WT controls ( Fig 4H–4K ) . Indeed , CD4+ T cells isolated from the LP of Il21r-/- mice expressed similar levels of IFN-γR1 and IFN-γR2 as WT controls ( Fig 4H–4K ) . Collectively , these findings demonstrated that the IL-21/IL-21R axis was not required for optimal expression and production of IFN-γR β-chain and IFN-γ and that the increased bacterial burden in Il21r-/- mice and reduced expression of ISGs were not due to impairment of IFN-γ or its receptor in these mice , but must be further downstream in the IFN-γ signaling pathway . If the IL-21/IL-21R axis is not necessary for expression of IFN-γ or its receptor , we asked whether it affected the next step in signal transduction downstream of the IFN-γR , STAT1 phosphorylation . It is known that IL-21 signals mainly via STAT3 but also via STAT1 and STAT5 [21 , 30 , 31] . We therefore hypothesized that IL-21 acts in concert with IFN-γ to facilitate the activation of STAT1 in CD4+ T cells and that might explain why Il21r-/- mice failed to optimally express multiple ISGs in the distal colon after C . rodentium infection . To address this hypothesis , we stimulated total splenocytes with either recombinant mIL-21 or mIFN-γ alone or in combination and analyzed the activation of STAT proteins by CD4+ T cells . Interestingly , a combination of mIL-21 and mIFN-γ ( 20 ng/ml of each cytokine ) led to significantly enhanced phosphorylation of STAT1 in CD4+ T cells , as compared with CD4+ T cells stimulated with either mIL-21 or mIFN-γ alone ( Fig 5A , top row , Fig 5B ) . As expected , the stimulation of CD4+ T cells isolated from Il21r-/- mice with a combination of mIL-21 and mIFN-γ did not enhance the activation of STAT1 , as compared with cells stimulated with mIFN-γ alone ( Fig 5A , bottom row , Fig 5B ) . However , no significant differences were observed between the activation of STAT1 in CD4+ T cells stimulated by mIFN-γ alone in the two genotypes , indicating that the IFN-γ axis is functional in the absence of IL-21/IL-21 R signaling in Il21r-/- mice . However , the treatment of splenocytes with a combination of IFN-γ with IL-17A or IL-22 did not result in enhanced activation of STAT1 in CD4+ T cells isolated from WT or Il21r-/- mice ( S10A–S10D Fig ) . These findings suggest that IL-21 , in collaboration with IFN-γ , enhances the expression of ISGs by CD4+ T cells via enhanced phosphorylation of STAT1 , but that IL-21/IL-21R signaling axis is not necessary for IFN-γ to induce the phosphorylation of STAT1 . In addition , we next investigated other possible complementary mechanisms besides a direct effect of IL-21 on STAT1 . It is known that the IL-21 binding to its cognate receptor , IL-21R , leads to the activation of STAT3 protein and subsequently activates a transcription program that includes some IL-21 target genes [31] . Remarkably , some of the IL-21 target genes ( i . e . Gzma , Gzmb , Il10 ) are known ISGs with non-redundant critical roles in host protection against a wide variety of microbial pathogens , including enteric infections [30] . Based on this , we analyzed the activation of STAT3 following stimulation with a mIL-21 or mIFN-γ alone or in combination . Although phosphorylation of STAT3 was induced in CD4+ T cells by the IL-21 stimulation alone , the combined application of IL-21 and IFN-γ did not further enhance STAT3 phosphorylation ( Fig 5C , Fig 5D , top row , and S11 Fig ) . As expected , CD4+ T cells isolated from Il21r-/- mice failed to induce the STAT3 activation upon treatment with IL-21 alone or in combination with IFN-γ ( Fig 5C and Fig 5D , bottom row , Fig 5E ) . To test whether STAT3 played any role in the collaboration observed between IL-21 and IFN-γ in ISG induction , we generated conditional knockout mice . The biological functions of IL-21 are mediated mainly via the activation of the STAT3 signaling axis downstream of the IL-21R in a wide variety of hematopoietic cells , although IL-21 is also known to exert some biological effects via the activation of STAT1 and STAT5 [23] . It has been shown that STAT3 is activated in intestinal epithelial cells following C . rodentium infection in vivo and that mice conditionally deficient in STAT3 in epithelial cells ( Stat3ΔIEC ) were highly susceptible to infection and developed severe colitis after infection with C . rodentium [32] . Our findings suggested that IL-21 was exclusively expressed by mucosal CD4+ T cells and that mucosal CD4+ T cells expressed higher levels of transcripts for IL-21R than other mucosal cells tested , although several immune cell types express this receptor ( Fig 2B and 2C ) . We bred CD4-conditional STAT3-/- mice by crossing CD4-Cre mice with STAT3flox/flox mice ( See Materials and Methods ) . The abrogation of the STAT3 signaling pathway in these mice was confirmed by the lack of STAT3 activation upon IL-6 treatment of CD4+ T cells isolated from CD4stat3-/- mice ( Fig 6A ) . To address the role played by the IL-21/IL-21R signaling axis in CD4+ T cells in host protection after C . rodentium infection , we assessed the bacterial burden , the infection kinetics and survival rates in conditional knockout mice with a CD4+ T cells-specific deletion of STAT3 activity ( Fig 6B–6D ) . The CD4stat3-/- conditional deficient mice showed increased bacterial burden at early time points ( days 3 and 7 ) , a higher peak bacterial load at day 9 p . i . , and impaired clearance at days 14 , 17 , 21 and 29 . Thus , they were even more impaired in their ability to handle C . rodentium infection than the Il21r-/- mice . That may be because STAT3 is critical not only for IL-21 signaling , but also for IL-17 , and induction of TH17 cells , which are another key mediator of C . rodentium clearance . Survival of these mice after C . rodentium infection was also significantly impaired ( Fig 6C ) . These finding are consistent with previous reports demonstrating that the STAT3 activation in TH17 and TH22 CD4+ T cells is important for protection against C . rodentium [33] . This confirms that a deficiency in these cytokines in CD4+ T cells alone is sufficient to seriously impair their ability to handle this bacterial colonic infection . At necropsy , the conditional deletion of STAT3 signaling in CD4+ T cells resulted in watery stool as well as hematomas along the lengths of the distal colons of CD4stat3-/- mice 9 days after C . rodentium infection ( Fig 6E ) . Histological examinations of the distal colons of mice 9 days after infection demonstrated significantly lower crypt hyperplasia scores , a hallmark of pathology during C . rodentium infection , and considerably shorter crypt lengths in of CD4stat3-/- mice ( Fig 6F–6H ) . Collectively , these findings indicated that C . rodentium infection induced moderate pathological changes in CD4stat3-/- mice as compared with littermate STAT3flox/flox control mice , despite significantly higher bacterial burdens in the distal colons of those mice . It is known that IL-21 exerts some of its downstream effects via the activation of STAT1 , STAT3 and STAT5 [21] , whereas the induction of IFN-γ-target genes exclusively requires the activation of STAT1 signaling pathway [34] . We already have shown that the activation of STAT3 in CD4+ T cells did not increase when stimulated with a combination of IL-21 and IFN-γ compared with cells treated with IL-21 alone ( Fig 5C and 5D ) . Hypothetically , the increased STAT1 phosphorylation when IL-21 was combined with IFN-γ could have been due to a direct effect of IL-21 on STAT1 , or to an indirect effect of some events downstream of IL-21’s principal signaling pathway through STAT3 . Thus , we sought to determine whether the optimal expression of ISGs requires intact STAT3 signaling in CD4+ T cells or is occurring in a STAT3-independent manner . To specifically target STAT3 in CD4+ T cells and to avoid the off-target effects of the global STAT3 deletion and its possible indirect effects of these on CD4+ T cells , we stimulated conditional Stat3-/- CD4+ T cells ( lacking STAT3 only in CD4+ T cells ) or control CD4+ T cells with IFN-γ or IL-21 or a combination of both and determined the expression of ISGs following the single or dual cytokine treatment . In particular , we sought to determine whether the optimal expression of ISGs by CD4+ T cells requires an intact STAT3 signaling or that the enhanced expression of ISGs in CD4+ T cells following treatment with both IFN-γ and IL-21 occurs in a STAT3-independent manner . Further analyses demonstrated that Stat3-/- CD4+ T cells upregulated an ISG profile in response to exogenous IFN-γ or IFN-γ+IL-21 in a fashion similar to CD4+ T cells from STAT3flox/flox controls ( Fig 7B , top and middle rows ) . Accordingly , Stat3-/- CD4+ T upregulated a gene signature profile very similar to the one induced in control animals in response to a combination of rmIFN-γ and IL-21 ( Fig 7B , bottom row , and Fig 7C ) . These findings show that the lack of a functional STAT3 in CD4+ T cells does not impair the expression of ISGs in Stat3-/- CD4+ T cells and that both Stat3-/- CD4+ T cells and cells from STAT3flox/flox littermates upregulated analogous gene profiles in response to IFN-γ and IL-21 . Furthermore , the lower panels comparing the cytokine combination with IFN-γ alone show that many ISGs are upregulated more by the combination than by IFN-γ alone in both the intact and STAT3-deficient CD4+ T cells . We conclude that the enhanced induction of most ISGs when IL-21 is added to IFN-γ does not depend on the former’s signaling through STAT3 . To more mechanistically delineate the roles played by IL-21 in enhanced expression of ISGs via the facilitated activation of STAT1 , we particularly asked whether the enhancement of STAT1 phosphorylation by combining IL-21 with IFN-γ was a direct effect of IL-21 on STAT1 or an indirect effect dependent on STAT3 . To address this question , we measured STAT1 activation in Stat3-/- CD4+ and Stat3+/+ CD4+ T cells in response to IFN-γ or IL-21 alone or in combination . Our findings indicated that Stat3-/- CD4+ T cells and control Stat3+/+ CD4+ T cells significantly and comparably phosphorylated STAT1 more in response to a combination of IFN-γ and IL-21 than to the single-cytokine-treated CD4+ T cells ( Fig 7D and 7E ) . These findings suggest that IL-21-induced enhancement of STAT1 activation was independent of STAT3 . Interestingly , the IL-21 treatment of CD4+ T cells isolated from mice deficient in IFN-γ/IFN-γR signaling ( Ifngr-/- ) induced the activation of STAT1 in these cells , indicating that the endogenous IFN-γ signaling pathway is not required for the STAT1 activation by IL-21 ( S12A Fig ) . Moreover , the expression of LAG-3 in CD4+ T cells isolated from Ifngr-/- mice was not induced by IFN-γ treatment alone , and was not higher with combined IL-21/IFN-γ than with IL-21 alone , indicating the expression of LAG-3 is not enhanced by IL-21 in the absence of an intact IFN-γ/IFN-γR signaling pathway ( S12B Fig ) .
In this study we discovered a previously unknown collaboration between IL-21 and IFN-γ inducing interferon-stimulated genes ( ISGs ) , in the course of studies on the role of the IL-21/IL-21R signaling pathway in resistance to and clearance of infection with a minimally-invasive murine intestinal pathogen C . rodentium . Deficiency in this pathway leads to attenuated inflammation in the colon following infection with this pathogen and impaired clearance of the pathogen . These effects appear to be partially dependent on ISGs , even though IL-21-induced STAT3 activation could play a role in IL-17- or IL-22-mediated protection against C . rodentium . Nanostring analysis identified that the majority of genes substantially impaired ( ≥ 2-fold ) in the whole distal colon of Il21r-/- mice as well as in CD4+ T cells isolated from those animals after infection with C . rodentium were ISGs . We showed that IFN-γ , but not IFN-α/β , mediated resistance to and clearance of C . rodentium . Importantly , we have discovered unexpectedly that IFN-γ collaboratively interacts with IL-21 for the optimal activation of STAT1 and subsequent induction of ISGs . Thus , the collaboration occurs at the level of STAT1 downstream of the IFN-γR , not in the expression of IFN-γ itself or its receptor . We further showed the absolute requirement of the STAT3 signaling pathway in CD4+ T cells for host defense against C . rodentium , by a separate mechanism because the enhanced activation of STAT1 and the subsequent induction of ISGs in CD4+ T cells by the combination of IL-21 and IFN-γ occurred in a STAT3-independent manner . STAT3 is known to be critical for IL-17 function which is important for host defense against extracellular pathogens [33] . The expression of ISGs is tightly regulated by the immune system to avoid excessive and persistent induction causing inflammation and tissue damage [35] . Excessive expression of ISGs has been linked to several inflammatory conditions [34] . Multiple microbial pathogens trigger type-specific interferons , resulting in the transcription of a wide range of downstream gene signatures with distinct or overlapping functions . Some of these genes have roles in the regulation of immunity and inflammation in different immune compartments , including the colon [36–38] . Several ISGs play important roles in host defense against viral , bacterial , and parasitic infections by directly targeting genes conferring protection against infection [39] . Mice deficient in Isg15 ( Isg15-/- ) are more susceptible to infections with several viral [40 , 41] and bacterial infections [42] . As such , studies in vivo showed that IFN-β produced by Legionella-infected macrophages promoted host defense via the upregulation of ISGs and this induction was required for host defense against L . pneumophila [43] . In addition to protective roles , ISGs can mediate inflammatory responses in the colon , predisposing the host to colitis-associated colon cancer [44] . Similarly , IL-21 was highly expressed in the colons of C57BL/6 mice with dextran sulfate sodium ( DSS ) -induced colitis and Il21r-/- mice manifested milder DSS-induced colitis as compared with their WT counterparts [29] . These findings suggest that the IL-21/IL-21R signaling axis could be part of a positive feedback loop that amplifies an inflammatory response in the gut . Our data support this interpretation , because the Il21r-/- mice had less inflammation and edema in the distal colons after C . rodentium infection despite having a higher bacterial burden ( Fig 1C ) . A network of cytokines signals through the STAT3 pathway with overlapping or opposing pro- or anti-inflammatory properties , including IL-6 , IL-10 , IL-17 , IL-21 and IL-22 . These cytokines activate the STAT3 signaling cascade by phosphorylation and subsequent STAT3 dimerization and translocation into the nucleus [45 , 46] . In a mouse model of intestinal inflammation ( DSS-induced colitis ) , for example , IL-22 was excessively secreted and the antibody blockade of IL-22 led to exacerbated inflammation in the colon , whereas the IL-22 overexpression resulted in attenuated inflammation [45] . These results suggest anti-inflammatory roles for IL-22 in the colon following intestinal inflammation . IL-6 is historically considered a pro-inflammatory cytokine and is known to promote inflammation in several models of inflammatory disease [46] . Conversely , the genetic loss of IL-6 or the antibody blockade of IL-6 results in attenuated colitis following intestinal inflammation [47] . While STAT3 plays an important role in protection against C . rodentium infection as we have seen , we found that it is not necessary for the collaboration between IL-21 and IFN-γ . Minimally-invasive enteric microbial pathogens adhere very closely to the intestinal epithelial surfaces and induce drastic physiological and cytoskeletal changes and structural reorganization in underlying epithelial cells [48] . Several defensive mechanisms have evolved to control enteric microbial infections at the intestinal epithelial barrier [49] . A cooperation between innate lymphoid cell ( ILC ) -derived IL-22 and IFN-λ was required for the optimal expression of STAT1 in IECs , leading to enhanced expression of ISGs by these cells and subsequent control of rotavirus infection in vivo [6] . It has been suggested that an evolutionary collaboration between two distinct but related cytokine signaling pathways facilitates the control of infection in the intestine . Our finding demonstrated that the receptor for IL-21 was more highly expressed by mucosal CD4+ T cells during intestinal infection than by other immune subsets in the colon LP ( i . e . dendritic cells , macrophages , neutrophils , NK cells ) and that mucosal CD4+ T cells expressed significant levels of transcripts for the IFN-γ receptor . Concurrent engagement of both receptors on CD4+ T cells was essential for optimal induction of ISGs in those cells . Our findings also indicated that colonic IECs did not express the receptor for IL-21 in response to intestinal microbial infection , whereas IFN-γ receptor was expressed by these cells . The lack of the expression of the IL-21 receptor by IECs excludes these cells as targets for the collaborative effects of IL-21 and subsequent modulation of ISG expression during C . rodentium infection as shown in other models of intestinal infection , in which IECs were the main targets of IL-22 and IFN-λ cooperation [6] . The indispensable role of CD4+ T cells in protection against intestinal infection with C . rodentium has been established in several studies [16 , 17] . Mice deficient in CD4+ T cells ( but not CD8+ T cells ) are extremely susceptible to intestinal infection with C . rodentium as well as to the systemic spread of the bacterium to extra-intestinal sites , including the mesenteric lymph nodes ( MLNs ) , spleen and the liver [15] . Intestinal infection with this pathogen elicits a TH1-biased immune response , characterized by the induction of IFN-γ and TNF-α ( 17 ) . CD4+ T cells play a central role in immune response to intestinal infection with C . rodentium via the production of cytokines required for host resistance to this pathogen , including IL-17 , IL-22 and IFN-γ , and they have been shown to be a main source of antigen-specific induction of IFN-γ during intestinal infection with this bacterium [17] . Our observations show dual roles for mucosal CD4+ T cells following infection with C . rodentium in the colon: First , IL-21 was exclusively expressed by mucosal CD4+ T cells of the colon LP and second , mucosal CD4+ T cells express significant levels of transcripts for IL-21R and IFN-γR and are targets for the collaborative activity of IFN-γ and IL-21 . These finding identify a previously unidentified collaboration between two distinct signaling pathways in the optimal expression of ISGs in CD4+ T cells , and protection against C . rodentium , and may further explain the absolute requirement of CD4+ T cells in the regulation of mucosal immune response to this pathogen in the colon . The type I IFNs have emerged as key players in host defense against both extracellular and intracellular bacterial pathogens in recent years [7 , 27] . We observed that both type I- and type-II specific ISGs were induced in the distal colon following infection with C . rodentium . Further analysis demonstrated that mice deficient in IFN-α/β signaling ( Ifnar-/- ) managed to clear infection with the bacterium in kinetics similar to WT controls . However , mice lacking functional IFN-γ ( Ifng-/- ) failed to clear the infection , indicating that type II-specific , but not type I-specific , ISGs were required for host defense in the colon . Other effects of IL-21 receptor deficiency on intestinal host defense have been reported very recently involving defective IgA responses to atypical commensal bacteria , such as segmented filamentous bacteria ( SFB ) and Helicobacter species , indirectly affecting C . rodentium-induced immunopathology [50] . However , in this situation , the effect was mostly on inflammation and the effect on bacterial burden was minor . This is clearly distinct from our model in which it is clear that IFN-γ plays a critical role and the role of IL-21 is mainly to amplify the IFN-γ signal in CD4+ T cells . Furthermore , we have cohoused mice for at least 2 weeks in all experiments to equalize intestinal microbiota , as mice , which are coprophagic , rapidly equilibrate their microbiomes when cohoused for 2 weeks [51–54] , and have replicated the same results with Il21r-/- mice vs . heterozygous controls that were bred from the same group of Il21r-/- mothers , to further ensure that the microbiota were equivalent ( S1 Fig ) . The collaborative link we established here between IFN-γ and IL-21 signaling pathways provides mechanistic elucidation to the enduring conundrum as to why the lack of an intact IL-21/IL-21R signaling axis renders host susceptible to a wide range of pathogens [55 , 56] . It also provides insight into why the lack of an intact IL-21/IL-21R pathway leads to attenuated inflammation in the colon following insults during infectious and non-infectious insults [29 , 44] . Considering that IECs , as the first line of defense against minimally-invasive intestinal pathogens including C . rodentium , do not express the receptor for IL-21 , our findings suggest the collaborative effect we describe between IFN-γ and IL-21/IL-21R signaling axes acts deep in the lamina propria of the colon as a second defensive layer against mucosal pathogens . Understanding the interactions of these cytokine networks and their signaling pathways should allow development of novel therapeutic targets for colonic infections , inflammatory bowel disease , and promotion of mucosal vaccine efficacy .
Six- to eight-week old sex- and age-matched female mice were used in all experiments . The mice were bred in-house or purchased from the Jackson Laboratory ( Bar Harbor , ME , USA ) . In order to exclude the effects of differing microbiome compositions between mouse genotypes on the experimental designs , the knockout and WT mice ( both on a C57BL/6 background ) were cohoused for 2 weeks at a 1:1 ratio before all experiments as described before [51–54] . In addition , to further exclude microbiome differences , we bred WT or Il21r-/- males to the same group of Il21r-/- females to produce matched pups that were either heterozygotes ( WT phenotype ) or homozygous Il21r-/- and that were then foster nursed together , so the microbiome obtained from their mothers and nursing mothers would be identical ( S1 Fig ) . C57BL/6 were purchased from Charles River Laboratories ( Wilmington , MA , USA ) . B6N . 129-Il21rtm1Kopf/J ( 019115 ) were purchased from the Jackson Laboratory . Ifnar-/- and Ifngr-/- mice on a C57BL/6 background were kindly provided by Dr . Howard Young ( NCI/NIH ) . We generated mice in which STAT3 protein is conditionally deleted in CD4+ T cells ( CD4stat3-/- ) by breeding CD4-Cre mice ( Tg ( Cd4-cre ) 1Cwi/BfluJ; 017336 ) to STAT3flox/flox ( B6 . 129S1-Stat3tm1Xyfu/J mice; 016923 ( both strains from the Jackson Laboratory ) as described earlier [57] . Mice were genotyped using DNA isolated from tail snips . STAT3flox/flox littermates were used as wild-type ( WT ) controls . All experiments were carried out in accordance with guidelines and protocols approved by the National Cancer Institute Animal Care and Use Committee in compliance with the National Institutes of Health Guidelines ( VB-014 ) . Spleens were aseptically removed from naïve WT , Il21r-/- , STAT3flox/flox or CD4stat3-/- mice and cells were mechanically disrupted through a 100 μm cell strainer ( BD Biosciences , San Jose , CA ) using the plunger of a 6 ml syringe . RBCs were lysed in ACK lysing buffer ( Lonza , Walkersville , MD ) , and the remaining cells were washed twice with ice-cold PBS . A total of 5×106 splenocytes were cultured in duplicate in 1 ml of RPMI 1640 ( Gibco ) supplemented with 10% FBS and 100 μg/ml penicillin/streptomycin ( all from Gibco ) . Cells were allowed to rest for 2 additional hours at 37°C and subsequently were stimulated in the presence of recombinant murine IFN-γ , IL-17A , IL-21 and IL-22 ( 20 ng/ml; Peprotech , Rocky Hill , New Jersey , USA ) alone or in combination ( IFN-γ+IL-17A; IFN-γ+IL-21; IFN-γ+IL-22; 20 ng/ml each ) . Cells receiving no cytokine treatments were used as controls . Cells were harvested at different intervals following the addition of cytokines and were stained with indicated antibodies . C . rodentium strain DBS100 ( ATCC 51459 ) was propagated in Luria-Bertani ( LB ) broth at 37°C , harvested by centrifugation , and resuspended in PBS at a concentration of 5×109 colony forming units ( CFU ) /mL . In some experiments , an ovalbumin-expressing C . rodentium ( OVA-Citrobacter ) under a kanamycin-resistance gene was used for infection . Mice infected with OVA-Citrobacter were given kanamycin ( 1g/L ) in drinking water ad libitum , starting 4 days before infection and during the entire course of infection to prevent loss of OVA expression . Mice were infected with 100 μl of the bacterial suspension containing 5×108 CFU of C . rodentium/mouse by oral gavage as described previously [58 , 59] . For bacterial quantification , fecal pellets ( 50–100 mg ) were weighed , homogenized in 2 mL of sterile PBS , serially diluted , and plated onto MacConkey agar as described before [58 , 59] . The detection limit of the culture method was 103 CFU/g feces . The colons were cleaned , rolled into a Swiss roll configuration , fixed in 10% buffered formalin overnight , followed by fixation in 70% ethanol and subsequently embedded in paraffin . Tissue sections ( 5-μm thick ) were stained with hematoxylin and eosin ( H&E ) and digitized with Aperio ScanScope ( Aperio , Vista CA ) and were analyzed using Aperio ImageScope software . The severity of colitis was assessed by an unbiased ( blinded ) observer using a scoring system developed previously [60] . An ex vivo organ culture system was used to determine the kinetics of cytokine production in the colon of mice after C . rodentium infection as described before [26] . The distal colons from infected mice at different time-points or uninfected controls were removed , washed briefly and opened longitudinally and were cultured in RPMI1640 culture medium ( Gibco ) supplemented with 10% FBS , and 100 μg/ml penicillin/streptomycin ( Gibco ) . Culture supernatants were collected at different intervals post-culture and were examined for cytokine concentrations by ELISA . Mouse ELISA kits for IFN-α , IFN-γ , IL-17A , IL-21 and IL-22 ( all from eBioscience ) were used . Results were expressed as picograms per g tissue ( pg/g ) . The colonic LPLs were isolated as described elsewhere [59] . Briefly , the distal colons were removed , cut longitudinally , and washed with ice-cold PBS . The colons then were cut into small pieces and incubated for 20 min in pre-digestion solution ( PBS , pH7 . 4 , containing 30 mM EDTA and 1 mM dithiothreitol ) at 37°C with agitation . After incubation , the samples were vortexed for 20 seconds and the supernatants were discarded . Subsequently , the tissues were washed twice with ice-cold PBS , minced , and digested in RPMI1640 ( Gibco ) , containing 420 μg/ml Liberase TL ( Roche , Indianapolis , IN , USA ) , and 0 . 1 mg/ml DNase ( Roche ) , for 45 min at 37°C . Digesting tissues were further mechanically dissociated by vortexing vigorously every 10 min . Digested tissues were vortexed for an additional 20 seconds and passed through a 70-μm cell strainer ( DB Falcon , San Jose , CA , USA ) . Isolated cells were washed twice with ice-cold PBS , counted , and stained with the indicated antibodies . Isolated LPLs were incubated with 1 μg/106 cells anti-mouse CD16/CD32 ( clone 93; Biolegend , San Diego , CA , USA ) in FACS buffer ( PBS supplemented with 3% FBS ) for 20 min on ice to block Fc receptor binding , followed by live/dead cell labeling using a LIVE/DEAD Fixable Aqua Dead Cell Stain Kit ( Life Technologies , Eugene , OR , USA ) for 20 min at 4°C in the dark per the manufacturer’s instructions . For surface marker staining , cells were stained in duplicate with a cocktail of the following conjugated antibodies in FACS buffer: anti-CD3-BV421 ( clone 17A2 ) , anti-CD4-Alexa Fluor 488 ( clone GK1 . 5 ) , anti-CD8-Alexa Fluor 700 ( clone 53–6 . 7 ) , anti-NK1 . 1-APC ( clone PK136 ) , anti-CD45 ( clone 30-F11 ) , anti-CD11b-APC ( clone M1/70 ) , anti-MHC class II ( I-A/I-E ) -Pacific Blue ( clone M5/114 . 15 . 2 ) , anti-F4/80-Brilliant Violet 421 ( clone BM8 ) , anti-LAG-3-PerCP/Cy5 . 5 ( clone C9B7W ) , anti-EpCAM-PE-Cy7 ( clone G8 . 8 ) , anti-IFN-γR-β-PE ( clone MOB-47; all from Biolegend ) , anti-CD11c-Texas Red ( clone MCD11C17; Invitrogen ) and LAG-3-PE ( clone C9B7W; eBioscience ) . For measuring the surface expression of IL-21R , LPLs were stained using a biotinylated anti-IL-21R Ab ( eBio4A9; eBioscience ) , followed by staining with PE-streptavidin ( Biolegend ) . The immune cell number in the colon was quantified by flow cytometry using CountBright absolute counting beads according to the manufacturer’s instructions ( Molecular Probes , Invitrogen ) . For intracellular staining , the LPLs isolated from the colon were stained without stimulation or were stimulated ex vivo for 8 h in the presence of PMA ( 50 ng/ml ) and ionomycin ( 350 ng/ml ) or chicken egg ovalbumin ( OVA ) ( 100 μg/ml; Sigma ) , adding brefeldin A at 10 μg/ml ( all from Sigma-Aldrich ) for the last 6 h . The cells then were fixed , permeabilized , and stained with anti-IFN-γ-PE ( clone XMG1 . 2; Biolegend ) . Data were acquired using an LSRII flow cytometer ( BD Biosciences ) and were analyzed using FlowJo software ( Tree Star , Inc . , San Carlos , CA , USA ) . We applied phospho-flow cytometry analysis to investigate the phosphorylation of STAT1 or STAT3 proteins following the treatment of naïve splenocytes with a combination of recombinant murine IFN-γ , IL-17A , IL-21 , IL-22 or IL-6 . Following the treatment of cells ( 5×106/well ) with IFN-γ , IL-17A , IL-21 or IL-22 ( 20 ng/ml each ) alone or in combination ( IFN-γ+IL-17A; IFN-γ+IL-21; IFN-γ+IL-22; 20 ng/ml each ) , the cells were fixed and permeabilized followed by staining with either anti-phospho-STAT1 ( pY701; clone 4a; BD Biosciences ) , anti-STAT3 ( pY705; clone 13A3-1; Biolegend ) or mouse IgG1κ isotype control ( clone P3 . 6 . 2 . 8 . 1; eBioscience ) . Data were acquired using an LSRII flow cytometer as described earlier . LPLs were isolated as described above from distal colons of naïve mice or from mice infected with C . rodentium 9 days p . i . The colonic LPL were FACS-sorted into CD4+ T cells ( EpCAM-CD45+CD3+CD4+ ) by using a FACSAria cell sorter ( Becton Dickinson ) . Splenic CD4+ T cells were purified ( ≥95% purity ) using negative selection with a mouse CD4+ T cell isolation kit ( Miltenyi Biotec , Auburn , CA ) and total RNA was isolated from CD4+ T cells using a Qiagen RNeasy Plus Micro Kit ( Qiagen , Hilden , Germany ) . In some experiments total RNA from whole distal colons was isolated using a Qiagen RNeasy Plus Mini Kit ( Qiagen , Hilden , Germany ) and used as the source of RNA for further analysis . Total RNA ( 100 ng ) was used as samples for probe-based NanoString system ( nCounter XT Code Set; Seattle , WA , USA ) . The raw data for each gene was compile and normalized against the spike-in positive ( 6 genes ) and negative ( 8 genes ) internal reference genes and were expressed as normalized counts by using the nSolver analysis software version 3 . 0 ( NanoString Technologies ) . The gene expression data were further normalized to the geometric mean of the expression of internal reference genes and presented as normalized counts/gene/biological sample . The status of impaired genes in the whole distal colon or CD4+ T cells isolated from the distal colon was tallied against the Interferome database ( http://interferome . org ) to establish whether the expression of a given gene is affected by type I , type II or type III interferons in the previously published databases [61] . The gene classification based on biological processes and gene ontology ( GO ) were performed using the Database for Annotation , Visualization and Integrated Discovery ( DAVID , version 6 . 8; http://david . abcc . ncifcrf . gov ) Bioinformatics Resources [62 , 63] . The normalized Nanostring values were used to generate heatmaps by using the web-based open software Morpheus ( http://software . broadinstitute . org/morpheus ) . Data were analyzed using GraphPad Prism , version 7 . 03 , software ( GraphPad , San Diego , CA , USA ) and expressed as the Mean ± SEM . For statistical analyses , a 2-tailed Mann-Whitney U test or a one-way ANOVA followed by Bonferroni post-hoc adjustment test for multiple comparison were employed . *p < 0 . 05 was considered statistically significant . | Diarrheal diseases still remain the second leading cause of mortality in children younger than 5 years old worldwide , leading to 1 . 3 million deaths per annum . The diarrheagenic Escherichia coli ( DEC ) pathotypes are considered NIAID Biodefense Category B agents . Human infections with enteropathogenic and enterohemorrhagic Escherichia coli ( EPEC and EHEC , respectively ) are associated with human disease . EPEC is a common cause of infantile diarrhea in the developing and underdeveloped world , and EHEC is considered an emerging zoonotic infection . These enteric pathogens cause a wide range of clinical symptoms , varying from mild diarrhea to more complicated clinical presentations , including hemolytic-uremic syndrome ( HUS ) and hemorrhagic colitis . Using a murine model of Citrobacter rodentium infection , we found the requirement of a functional IL-21/IL-21R signaling axis in the control of enteric microbial infections via augmenting activation of STAT1 in mucosal CD4+ T cells in a murine model of Citrobacter rodentium colitis . Understanding how the IL-21/IL-21R signaling pathway contributes to the host immunity in the colon will further provide insights into the development of novel preventive and therapeutic targets for human subjects with enteric microbial infections and other inflammatory conditions , including inflammatory bowel disease ( IBD ) and celiac disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"animal",
"models",
"developmental",
"biology",
"model",
"organ... | 2019 | Interleukin 21 collaborates with interferon-γ for the optimal expression of interferon-stimulated genes and enhances protection against enteric microbial infection |
The SUPPRESSOR OF rps4-RLD1 ( SRFR1 ) gene was identified based on enhanced AvrRps4-triggered resistance in the naturally susceptible Arabidopsis accession RLD . No other phenotypic effects were recorded , and the extent of SRFR1 involvement in regulating effector-triggered immunity was unknown . Here we show that mutations in SRFR1 in the accession Columbia-0 ( Col-0 ) lead to severe stunting and constitutive expression of the defense gene PR1 . These phenotypes were temperature-dependent . A cross between srfr1-1 ( RLD background ) and srfr1-4 ( Col-0 ) showed that stunting was caused by a recessive locus in Col-0 . Mapping and targeted crosses identified the Col-0-specific resistance gene SNC1 as the locus that causes stunting . SRFR1 was proposed to function as a transcriptional repressor , and SNC1 is indeed overexpressed in srfr1-4 . Interestingly , co-regulated genes in the SNC1 cluster are also upregulated in the srfr1-4 snc1-11 double mutant , indicating that the overexpression of SNC1 is not a secondary effect of constitutive defense activation . In addition , a Col-0 RPS4 mutant showed full susceptibility to bacteria expressing avrRps4 at 24°C but not at 22°C , while RLD susceptibility was not temperature-dependent . The rps4-2 snc1-11 double mutant showed increased , but not full , susceptibility at 22°C , indicating that additional cross-talk between resistance pathways may exist . Intriguingly , when transiently expressed in Nicotiana benthamiana , SRFR1 , RPS4 and SNC1 are in a common protein complex in a cytoplasmic microsomal compartment . Our results highlight SRFR1 as a convergence point in at least a subset of TIR-NBS-LRR protein-mediated immunity in Arabidopsis . Based on the cross-talk evident from our results , they also suggest that reports of constitutive resistance phenotypes in Col-0 need to consider the possible involvement of SNC1 .
Plants possess a highly effective immune system that responds to conserved non-self molecular patterns , or to specific pathogen-derived molecules deployed to alter host defenses [1]–[3] . The latter response , called effector-triggered immunity ( ETI ) , is largely mediated by resistance ( R ) proteins that directly or indirectly detect the presence of pathogen effectors [3] , [4] , although mechanistically overlap between ETI and the response to molecular patterns can be observed [5] , [6] . ETI can lead to programmed cell death termed the hypersensitive response ( HR ) [7] , [8] . In the case of resistance to some viral and hemi-biotrophic bacterial pathogens , it has been shown that the HR is not causally related to resistance [9]–[13] . Nevertheless , the plant immune response is deleterious to plant growth , normal development , and seed set even in the absence of HR , and therefore needs to be tightly controlled [14] . In order to explore the molecular mechanisms that negatively regulate ETI , we performed a suppressor screen for reactivated AvrRps4-triggered resistance in the naturally susceptible Arabidopsis ( Arabidopsis thaliana ) accession RLD [15] . This screen yielded two mutant alleles in SUPPRESSOR OF rps4-RLD1 ( SRFR1 ) . Mutations in srfr1 enhanced resistance of RLD specifically to Pseudomonas syringae pv . tomato strain DC3000 ( DC3000 ) expressing avrRps4 , while susceptibility to the virulent strain DC3000 was unchanged [15] . Apart from re-establishing a certain level of resistance to avrRps4 , no other marked phenotype was noted . RPS4 encodes an R protein of the Toll/Interleukin-1 receptor ( TIR ) - nucleotide binding site ( NBS ) - leucine-rich repeat ( LRR ) ( TNL ) class [16] , and was found to require the defense regulator EDS1 to trigger immunity [17] . This is in contrast to the coiled-coil ( CC ) -NBS-LRR ( CNL ) R proteins RPS2 , RPM1 and RPS5 , which require the defense gene NDR1 [17] . Combining mutations in SRFR1 and the CNL pathway genes RPM1 , RPS2 or NDR1 did not measurably alter the susceptibility to the cognate effector genes . The partial resistance to avrRps4 in srfr1 mutants required EDS1 [15] , [18] . In addition , mutations in RPS6 , another TNL gene that requires EDS1 [12] , led to susceptibility to DC3000 ( hopA1 ) that was diminished in srfr1-1 rps6-1 double mutants [19] . Taken together , these data indicated that SRFR1 function is closely associated with the EDS1 resistance pathway . Here we show that a mutation in SRFR1 in the accession Columbia-0 ( Col-0 ) , srfr1-4 , activates the Col-0 specific and EDS1-dependent R-like gene SNC1 , consistent with the genetic function of SRFR1 as a negative regulator of R gene-mediated resistance . Activation of constitutive defenses in srfr1-4 was temperature-dependent . In addition , RPS4 and SNC1 contributed redundantly to susceptibility to DC3000 ( avrRps4 ) in Col-0 at 22°C , whereas at 24°C RPS4 activity was the sole determinant of resistance . Interestingly , SRFR1 interacted with both RPS4 and SNC1 . Our data thus provide evidence for cross-talk between these TNL pathways that converge on SRFR1 , suggesting that SRFR1 may have a general function in regulating TNL protein signal output .
We previously had isolated the mutant alleles srfr1-1 and srfr1-2 from the Arabidopsis accession RLD [15] . Apart from enhanced resistance to DC3000 ( avrRps4 ) , they did not display marked phenotypes . To further investigate the function of SRFR1 , we aimed at isolating T-DNA tagged lines of SRFR1 in the accession Col-0 [20] , [21] . Out of four lines , one did not germinate ( SALK_106212 ) , and one was untagged ( SALK_095440 ) . We could verify a T-DNA insertion far upstream of the open reading frame in SALK_039199 , without causing an apparent phenotype . Interestingly , the fourth line , SAIL_412_E08 with a T-DNA insertion in the second intron of SRFR1 ( Figure 1A ) , showed pronounced stunting ( Figure 1B ) in one-fourth of plants ( 22 out of 97 plants; χ2 = 0 . 28 ) . Genotyping showed that the T-DNA insertion in SRFR1 segregated in the original seed stock , and that stunted plants were invariably homozygous for the T-DNA insertion . Reverse transcription ( RT ) PCR showed that no srfr1 mRNA was detected with primers on either side of the insertion ( Figure S1 ) . A low level of srfr1 mRNA could be detected with primers located 3′ of the T-DNA insertion , but this mRNA contained the T-DNA ( Figure S1 ) , indicating that srfr1-4 mRNA does not encode functional protein . Consistent with this , Li and co-workers recently showed that no SRFR1 protein can be detected in this knock-out line [22] . We named this line srfr1-4 . Subsequently , we back-crossed srfr1-4 to Col-0 . The stunted phenotype co-segregated with homozygosity of the srfr1-4 T-DNA tagged allele in F2 plants ( Table 1 ) . To prove that the phenotype originated from the srfr1-4 allele , we transformed healthy heterozygous srfr1-4 plants with pSHK102 containing a genomic clone of SRFR1 [18] , and by scoring for antibiotic resistance selected 5 single-locus homozygous transgenic SRFR1 T3 lines that contained at least one copy of the srfr1-4 T-DNA allele based on genotyping . Because the transgenic copy of SRFR1 prevented us from determining whether these T3 lines were homozygous or heterozygous for the srfr1-4 allele , we tested whether srfr1-4 segregated in the next generation by genotyping 15 progeny for each line . Three of the 5 lines were shown in this way to be homozygous for the srfr1-4 allele , and the transgenic copy of SRFR1 reversed the stunted phenotype in each case ( Figure 1B ) . We concluded that the stunted growth phenotype is caused by the T-DNA insertion in SRFR1 . The stunted srfr1-4 phenotype was in marked contrast to the normal phenotype of srfr1-1 and srfr1-2 plants . To determine whether the specific allele of SRFR1 or the Col-0 genetic background causes the severe phenotype of srfr1-4 , we first reexamined more closely F3 families of important break-point plants retained from the SRFR1 mapping populations . Plants in these F3 families were generated by crossing srfr1-1 or srfr1-2 ( RLD background ) to the SAIL RPS4 T-DNA knockout line rps4-1 ( Col-0 background ) [15] , [18] and were progeny of F2 plants selected for resistance to DC3000 ( avrRps4 ) . They were therefore homozygous for srfr1-1 or srfr1-2 , with varying degrees of Col-0 background . Two out of 4 srfr1-1 and 2 out of 6 srfr1-2 F3 families contained no individuals with abnormal growth phenotypes . However , the remaining F3 families gave rise to plants with phenotypes similar to srfr1-4 . The combined total number of stunted plants in these families was 20 out of 107 plants , consistent with the segregation of a single recessive gene in these populations ( χ2 = 2 . 43 , P>0 . 1 ) . We concluded that most likely the mutant alleles srfr1-1 and srfr1-2 also induce stunting in the Col-0 background and that Col-0 possesses a recessive genetic modifier that alters the srfr1 phenotype . We tested these predictions directly by out-crossing srfr1-4 to RLD and srfr1-1 . In the cross to srfr1-1 , 14 out of 46 plants were stunted , consistent with both srfr1-1 and srfr1-4 causing stunting and the segregation of a recessive gene ( χ2 = 0 . 45 , P>0 . 5 ) . In the cross to RLD , segregation of the stunted phenotype in the F2 generation was explained by two recessive genes , and genotyping showed that while all stunted plants were homozygous srfr1-4 , not all srfr1-4/srfr1-4 plants were automatically stunted ( Table 2 ) . In this cross , stunted F2 plants were also selected to determine a rough map position for the presumptive Col-0 modifier gene . This mapping placed the Col-0 modifier gene onto chromosome 4 ( Table 3 ) . Interestingly , in addition to the bottom of chromosome 4 where SRFR1 is located , individual break-point plants identified a map position towards the top of chromosome 4 between markers ciw6 and CH42 for the Col-0 modifier gene . The map position for the modifier gene contained the Col-0-specific TNL R gene homolog SNC1 , which was originally identified through a point mutation that autoactivates the SNC1 protein and constitutively induces PR genes even in the npr1 mutant line [23] . Additional work showed that wild-type SNC1 is easily autoactivated when expression of SNC1 is misregulated [24] . For example , mutations in BON1 , a member of the copine gene family encoding a plasma membrane-localized putative calcium-dependent phospholipid-binding protein [25] , [26] , lead to higher SNC1 expression levels , constitutive defense responses and reduced plant growth [27] . When the Col bon1-1 mutant was outcrossed to other Arabidopsis accessions , it was found that the wild-type SNC1 gene from Col-0 behaved as a recessive locus that causes stunting [27] . Our segregation data also indicated that the Col-0 modifier was recessive ( Table 2 ) . We therefore tested additional phenotypes displayed by bon1-1 plants , such as temperature dependence of constitutive defense activation and growth phenotypes . The stunted phenotype in srfr1-4 was severe at 22°C , but was intermediate at 24°C and absent at 28°C ( Figure 2A ) , reminiscent of the Arabidopsis bon1-1 mutant phenotype . In srfr1-1 and srfr1-2 plants , resistance to DC3000 ( avrRps4 ) was enhanced , but remained unchanged to virulent DC3000 , and plant growth was normal [15] , [18] . Interestingly , the srfr1-4 mutants were resistant not only to avirulent DC3000 ( avrRps4 ) , but also to virulent DC3000 and non-pathogenic DC3000 hrcC− ( Figure 2B ) . The srfr1-4 line showed approximately 50-fold lower DC3000 and DC3000 ( avrRps4 ) growth than wild type Col-0 , whereas the growth of DC3000 hrcC− in srfr1-4 was about 10-fold less than in Col-0 , suggesting that mutations in SRFR1 in Col-0 increased basal defenses at 24°C that were additive to AvrRps4-triggered immunity ( Figure 2B ) . Complemented srfr1-4 lines did not show either enhanced resistance phenotype ( Figure 2B ) . We could not test bacterial growth at 22°C because srfr1-4 plants were severely stunted at this temperature . However , consistent with an upregulation of salicylic acid ( SA ) -based defenses , PR1 and PR2 mRNA levels were upregulated and PDF1 . 2 levels down-regulated in srfr1-4 at 22°C ( Figure 2C ) . Characterization of the srfr1-4 phenotype and mapping therefore strongly suggested that the Col-0 modifier is SNC1 . To test this directly , we crossed srfr1-4 to snc1-11 , a T-DNA insertion allele in the first exon of SNC1 [27] . In the F2 population , the number of stunted plants was consistent with the segregation of two recessive loci ( srfr1-4 and wild-type SNC1 ) ( Table 4 ) . All of the stunted plants were homozygous for the srfr1-4 allele and the wild-type SNC1 allele . In contrast , all plants of normal stature that were homozygous for the srfr1-4 T-DNA allele possessed at least one copy of the snc1-11 T-DNA allele ( Table 4 ) . Therefore , the stunted phenotype of srfr1-4 plants requires two copies of SNC1 in Col-0 , analogous to the phenotype of bon1-1 plants [27] . We quantified the effect of mutations in SRFR1 on plant growth by measuring the shoot weight of srfr1 mutants in Col-0 and RLD ( Figure 3 ) . Shoot weights were close to normal in the original srfr1-1 and srfr1-2 plants compared to wild-type RLD . Mutations in srfr1 caused severe reductions in shoot weight in the Col-0 background that were completely reversed by introgressing snc1-11 . Interestingly , the shoot weight of srfr1 SNC1 plants was more strongly reduced than in bon1-1 plants ( Figure 3 ) , indicating that perhaps SRFR1 functions downstream of additional R genes apart from regulating SNC1 . Together with the negative regulation in AvrRps4- and HopA1-triggered immunity , these results show that SRFR1 is a negative regulator of plant immune responses of broader specificity than originally described . Previous studies had suggested that the readily autoactivatable SNC1 is limited to the Col-0 accession , but these studies had not included RLD [27] . We therefore sequenced the likely RLD ortholog of SNC1 in RLD to determine the molecular basis for the very different phenotypes of Col-0 and RLD srfr1 mutants . At the 5′-end , SNC1-specific primers consistently amplified a sequence with high overall similarity to SNC1-Col ( Figure 4A and 4B ) . SNC1-specific primers designed to amplify the complete SNC1 gene or the 3′-half of SNC1 failed to result in a unique RLD product . This reflected the very duplicated nature of the 3′-half of SNC1 in Col-0 . Whole sections of the gene are not only duplicated within SNC1 with 100% sequence identity , but are also found in linked family members [28] . We were not able to experimentally determine unequivocally which genomic PCR product from the 3′-end was physically linked to the 5′-end of SNC1-RLD . We therefore determined the SNC1 mRNA sequence from RLD using a combination of 3′-Rapid Amplification of cDNA Ends ( 3′-RACE ) and RT-PCR . As shown in Figure 4A , the open reading frame of SNC1-RLD predicted a protein of 619 amino acids , including a TIR and NBS domain but only a partial LRR domain . The predicted amino acid sequence identity between SNC1-Col and SNC1-RLD within the first three exons was 87% . However , our SNC1-RLD cDNA sequence was missing the fourth and fifth exons , leading to an in-frame stop codon at position 620 ( Figure 4B ) . Interestingly , in the SNC1-RLD cDNA the very 3′-end of the open-reading frame and the 3′-untranslated region showed high nucleotide sequence identity with the corresponding region in SNC1-Col . Because we only obtained cDNA sequence of SNC1-RLD at the 3′-end , we could not determine whether the 3′-end of the SNC1-RLD coding sequence is interrupted by introns . We also obtained RT-PCR products from Col-0 . These indicated that in contrast to the annotation of SNC1 in TAIR , we did not find evidence for the splicing of intron 5 , which does not contain in-frame stop codons ( Figure 4B ) . This alternative SNC1 transcript encoded a SNC1 protein of 1404 amino acids rather than the annotated 1301 amino acids . Taken together , sequencing of the RLD SNC1 ortholog provided evidence for polymorphisms at the 5′-end and major alterations in the 3′-half of the gene compared to Col-0 , consistent with the fact that RLD does not have a SNC1 ortholog that triggers stunted growth in the absence of SRFR1 . Activation of SNC1 , either by intragenic autoactivating mutations [23] or by mutations in negative regulators of SNC1 such as BON1 [27] , leads to constitutively enhanced resistance . Consistent with this and the constitutive expression of PR genes in srfr1-4 ( Figure 2C ) , we observed with in planta bacterial growth assays increased resistance of srfr1-4 to DC3000 ( avrRps4 ) and to virulent DC3000 ( Figure 2B ) . The latter shows that srfr1-4 plants possess elevated basal resistance that is independent of particular avirulence genes . To test if enhanced basal resistance in srfr1-4 , like stunted growth , is fully dependent on SNC1 , we performed in planta bacterial growth assays at varying temperatures . As noted before , we were not able to infiltrate srfr1-4 plants at 22°C because of the severe growth phenotype . At both 22°C and 24°C , the growth of DC3000 and DC3000 ( avrRps4 ) was reduced in srfr1-4 snc1-11 compared to growth in wild type Col-0 , even though the growth of DC3000 ( avrRps4 ) in srfr1-4 snc1-11 was slightly higher than that in srfr1-4 at 24°C ( Figure 5A and 5B ) . This remnant enhanced basal resistance in srfr1-4 snc1-11 plants may be related to the induced defense gene mRNA levels observed in RLD srfr1-1 and srfr1-2 plants , although the latter plants do not show enhanced basal resistance [18] , [29] . These results demonstrate that although the stunted phenotype of srfr1-4 at 22°C and 24°C is fully mediated by SNC1 , enhanced basal resistance at these temperatures in srfr1-4 is not entirely mediated by SNC1 . At 28°C , both basal and AvrRps4-triggered resistance were abolished in srfr1-4 and srfr1-4 snc1-11 plants ( Figure S2A ) . In addition , AvrRps4-triggered resistance was also abolished in wild-type Col-0 , confirming previous results [30] , and in snc1-11 plants ( Figure S2A ) . Consistent with normal growth and absence of resistance at 28°C , SNC1 and PR1 expression were not elevated in srfr1-4 or srfr1-4 snc1-11 plants ( Figure S2B ) . Previously , we showed that several defense-related genes were up-regulated in RLD srfr1 mutants , supporting our hypothesis that SRFR1 may function as a repressor in plant innate immunity by negatively regulating defense gene expression levels [29] . The growth and constitutive defense phenotypes of srfr1-4 at 22°C and 24°C prompted us to quantify defense-related gene mRNA levels in srfr1-4 at these temperatures using quantitative reverse transcription real-time PCR ( qPCR ) , and to determine whether all changes in expression in srfr1-4 can be attributed to SNC1 . As expected , SNC1 transcript levels were higher in srfr1-4 than in Col-0 at 22°C and 24°C , as were those of RPP4 and At4g16950 ( Figure 6A ) , two TNL genes in the SNC1 cluster that are co-regulated with SNC1 [31] . Interestingly , RPP4 and At4g16950 expression levels were higher also in the srfr1-4 snc1-11 double mutant ( Figure 6A ) , showing that higher mRNA levels of these genes is not an indirect effect of SNC1 activation . Similarly , we observed increased mRNA levels of the CNL R gene RPS2 , and to a lesser extent of RPM1 , in srfr1-4 and srfr1-4 snc1-11 plants at both 22°C ( Figure S3A ) and 24°C ( Figure S3B ) , indicating that upregulation of R genes by mutations in SRFR1 is not limited to TNL genes in Col-0 . In contrast to SNC1-RLD , upregulation of RPM1 and RPS2 was not observed in the RLD mutant srfr1-1 ( Figure S3C ) , possibly reflecting the presence of additional accession-specific SNC1-like genes in Col-0 [32] that may lead to enhanced expression of CNL genes . SA-dependent defense related gene mRNA levels were also higher in srfr1-4 than in wild-type at 22°C and 24°C ( Figure 6B ) . Unlike for TNL and CNL genes , these expression levels were reduced in srfr1-4 snc1-11 compared to srfr1-4 to varying degrees , although they were still higher than in wild-type ( Figure 6B ) . Interestingly , NPR1 and EDS1 mRNA levels in the double srfr1-4 snc1-11 mutant showed additive increases compared to the wild-type and single mutants at 22°C ( Figure 6B ) . In contrast , mRNA levels of PDF1 . 2 , a defensin gene whose expression is under negative regulation by the JA-responsive transcription factor JIN1 [33] , was strongly repressed at 22°C in srfr1-4 but induced in srfr1-4 snc1-11 plants compared to wild-type . PDF1 . 2 expression levels were not significantly different among the genotypes at 24°C ( Figure 6C ) . These results point towards complex modular control of defense gene expression that is influenced by a combination of SRFR1 , SNC1 and temperature to varying proportions . The Arabidopsis accession RLD carries a natural mutation in RPS4 and is fully susceptible to DC3000 ( avrRps4 ) [34] , [35] . In addition , introduction of RPS4 from Col-0 or Ler into RLD is sufficient to provide full resistance to DC3000 ( avrRps4 ) when compared to Col-0 and Ler [16] , [35] . We also observed susceptibility of rps4-1 , an RPS4 T-DNA allele in the Col-0 background , under our conditions that were used to map SRFR1 [15] . However , it was reported that rps4-2 , a second RPS4 T-DNA allele in the Col-0 background , was only slightly more susceptible to DC3000 ( avrRps4 ) [36] . Based on the accession-specific presence of SNC1 in Col-0 , the temperature-dependent srfr1-4 phenotype and the fact that SRFR1 was identified in a screen for enhanced DC3000 ( avrRps4 ) resistance in RLD , we speculated that the rps4-2 phenotype might be temperature-dependent . Indeed , when directly comparing plants grown in identical growth chambers at 22°C or 24°C , we observed a strong temperature dependence: rps4-2 plants were as resistant to DC3000 ( avrRps4 ) as Col-0 at 22°C , while at 24°C they were as susceptible as Col-0 treated with virulent DC3000 and as susceptible as RLD treated with either strain ( Figure 7 ) . Given the effect of temperature , we next tested whether SNC1 interferes with the susceptible phenotype at 22°C . Interestingly , rps4-2 snc1-11 double mutants displayed approximately 30-fold increased bacterial growth of DC3000 ( avrRps4 ) compared to Col-0 or rps4-2 at 22°C ( Figure 7A ) , suggesting that SNC1 in the absence of RPS4 contributes to AvrRps4-triggered immunity at 22°C in Col-0 . However , susceptibility of rps4-2 snc1-11 to DC3000 ( avrRps4 ) was not complete compared to Col-0 treated with virulent DC3000 or to RLD treated with either strain , indicating that additional factors interfere with rps4-caused susceptibility ( Figure 7A ) . No significant difference of DC3000 ( avrRps4 ) growth in rps4-2 and rps4-2 snc1-11 was observed at 24°C , reflecting full susceptibility of rps4-2 to DC3000 ( avrRps4 ) at this temperature ( Figure 7B ) . Recently , RRS1 was shown to be involved in DC3000 ( avrRps4 ) -mediated resistance [37] , [38] . However , we observed no temperature-dependent resistance to DC3000 ( avrRps4 ) in the Ws-0 mutants rps4-21 and rrs1-1 ( Figure S4 ) . As was observed before , mutations in either RPS4 or RRS1 had equal effects on DC3000 ( avrRps4 ) susceptibility , which was qualitatively different from the redundancy between SNC1 and RPS4 ( Figure 7 ) . Interestingly , as reported before [38] , we reproducibly observed approximately 10-fold higher growth of DC3000 compared to DC3000 ( avrRps4 ) in the single rps4-21 and rrs1-1 mutants and the double mutant , indicating that additional layers of resistance exist . The redundancy between RPS4 and SNC1 suggests that they function in parallel to provide resistance to DC3000 ( avrRps4 ) at 22°C . We speculated that this cross-talk between two R proteins might occur if both interact with proteins in a common complex . Perturbation of this complex by an effector could trigger one or the other R protein , and both need to be absent to observe susceptibility . Based on the results presented here , we reasoned that SRFR1 might be a common interaction partner of RPS4 and SNC1 . In the past , transient expression of SRFR1 in Nicotiana benthamiana led to variable protein expression levels and required a silencing inhibitor for detectable expression [18] . We therefore generated stable transgenic N . benthamiana plants expressing HA-SRFR1 encoded by a genomic clone driven by the native Arabidopsis SRFR1 promoter . We first determined the functionality of this genomic HA-SRFR1 construct in Arabidopsis by testing for complementation of the stunted srfr1-4 phenotype . Transgenic plants expressing HA-SRFR1 in the srfr1-4 background showed normal growth and development ( Figure S5A ) . Immunoblot analysis detected the expression of the transgene product in these transgenic plants ( Figure S5A ) . HA-SRFR1 in these plants localized to microsomal and nuclear fractions ( Figure S5B ) . This localization was consistent with the nuclear and punctate cytoplasmic localization of GFP-SRFR1 transiently expressed in N . benthamiana [18] . We observed improved and reproducible HA-SRFR1 expression in the stable transgenic N . benthamiana lines . As in Arabidopsis , HA-SRFR1 localized to the microsomal and nuclear fractions in N . benthamiana ( Figure 8A ) . A previous study showed that RPS4 was predominantly localized to microsomes [36] . Immunoblot assays of Myc-SNC1 transiently expressed in N . benthamiana suggested that SNC1 was mainly a soluble cytoplasmic protein , although a sizeable portion accumulated in the microsomal fractions ( Figure 8B ) . We also detected some SNC1 in the nuclear fraction ( Figure 8B ) . We tested for SRFR1 interaction with SNC1 and RPS4 by transiently expressing Myc-SNC1 , Myc-RPS4 or Myc-eGFP as a negative control in transgenic HA-SRFR1 N . benthamiana plants . Co-immunoprecipitation analysis on protein isolated 48 h after infiltration of Agrobacterium tumefaciens strains showed that SRFR1 interacted with both SNC1 and RPS4 in the microsomal fraction ( Figure 9 ) . No significant interaction between SRFR1 and SNC1 was detected in the soluble fraction , even though SNC1 was detected in this fraction . No interaction with eGFP was detected in either fraction ( Figure 9 ) . As an additional control , we probed SRFR1 co-immunoprecipitated samples for the presence of GAPDH and V-ATPase . Neither protein was co-immunoprecipitated with SRFR1 ( Figure S6A and S6B ) , indicating that the interactions of SRFR1 with SNC1 and RPS4 are specific .
Here we extend our analysis to the Col-0 specific TNL R-like gene SNC1 and show that mutations in SRFR1 activate SNC1 . SNC1 was originally identified based on an autoactivated allele that led to constitutive expression of PR1 [23] . Subsequently , it was shown that perturbation of wild-type SNC1 expression readily leads to autoactivation [24] , [27] , [40] . Our finding that SNC1 is activated in srfr1 mutants is reminiscent of the bon1/cpn1 phenotype [25]–[27] . How the absence of BON1 leads to SNC1 activation is not known . In particular , it is not known if sub-pools of BON1 and SNC1 reside in the same protein complex . Together , our data show that mutations in SRFR1 impact three resistance specificities , namely AvrRps4- , HopA1- and SNC1-triggered immunity . The impact of srfr1 mutations on SNC1 is novel , given that previously we observed effects of SRFR1 mutations only in the absence of the R genes RPS4 or RPS6 . SNC1 is therefore the first TNL gene for which a genetically direct negative regulation by SRFR1 could be shown . Whether this is also mechanistically direct remains to be determined . Consistent with the proposed function of SRFR1 as a transcriptional repressor , we found increased mRNA levels for SNC1 , RPP4 and At4g16950 in srfr1-4 plants . This altered expression level was not an indirect effect of SNC1 activation , since RPP4 and At4g18950 were also upregulated in the srfr1-4 snc1-11 double mutant . Because these members of the SNC1 locus were previously shown to be co-regulated with SNC1 [31] and because changes in SNC1 expression levels have been shown to cause autoactivation of SNC1 [24] , we propose that mutations in SRFR1 lead to misregulated expression of SNC1 , which in turn activates constitutive expression of an enhanced defense phenotype . The genetic connection of SNC1 and RPS4 via SRFR1 was measurable as cross-talk between these resistance pathways in disease assays under specific environmental conditions . Because it had been convincingly shown that the Col-0 rps4-2 mutant was not fully susceptible to DC3000 ( avrRps4 ) [36] , while we observed complete susceptibility , we tested whether environmental conditions had an influence on the Col-0 phenotypic response to DC3000 ( avrRps4 ) . Surprisingly , we found that a mere 2°C difference in temperature changed the phenotype of rps4-2 from almost completely resistant to DC3000 ( avrRps4 ) to fully susceptible . Other environmental factors that are likely to impact this response are humidity [26] , with drier conditions favoring resistance , and light intensity . Because cis or second-site mutants with activated SNC1 have a well-described conditional phenotype influenced by temperature and humidity , we tested whether the partial phenotype of rps4-2 is influenced by SNC1 . Indeed , we were able to measure a synergistic effect of mutations in RPS4 and SNC1 on susceptibility to DC3000 ( avrRps4 ) at 22°C . In addition , in the accessions RLD and Ws-0 that do not have SNC1 , mutations in RPS4 result in susceptibility to DC3000 ( avrRps4 ) that is not influenced by changes in temperature in the range investigated here . SNC1 was originally identified in a screen for mutants with constitutively activated defenses , and to date no cognate avirulence gene has been identified . Nevertheless , some suppressor mutants of the constitutive snc1-1 phenotype such as mos7 also impact effector-triggered immunity [41] . Our finding that SNC1 contributes to AvrRps4-triggered immunity further indicates that SNC1 can be considered a bona fide R gene . Conceptually , cross-talk between resistance pathways can occur if an effector protein has more than one target , or if R proteins guard a common target . The former seems to be the case for RPM1 and TAO1 , which additively contribute to full resistance to DC3000 ( avrB ) [42] . In contrast , AvrRpm1 induced measurable defenses in rpm1 plants that were dependent on RPS2 , presumably because both RPM1 and RPS2 guard RIN4 , a protein that is the target for both AvrRpm1 and AvrRpt2 [19] . As a first step to distinguish between these models , we tested whether SNC1 and RPS4 co-localize with a common protein . Given the regulatory function of SRFR1 on SNC1 and on AvrRps4-triggered resistance , we speculated that SRFR1 might be such a common protein . Interestingly , the microsomal pool of SRFR1 was found to be in a complex with SNC1 . Transiently expressed GFP-SRFR1 in N . benthamiana localized to the nucleus and cytoplasm [18] . The cytoplasmic localization was punctate . Here , further analysis of the cytoplasmic pool showed that most SRFR1 localized to the microsomal cytoplasmic fraction , and very little was soluble . Because the majority of SNC1 was in the soluble cytoplasmic pool , it was not possible to determine whether the microsomal pool of SNC1 diminishes in the absence of SRFR1 . In addition , the native N . benthamiana pool of SRFR1 may be sufficient to localize some proportion of SNC1 to microsomes . Most likely , SNC1 is in a higher-order complex with SRFR1 in a microsomal compartment of unknown identity . Interestingly , we found that RPS4 also interacted with SRFR1 in the same cell fraction . This suggests that perhaps additional R proteins localize to a common complex . The localization of SRFR1 and interactions with RPS4 and SNC1 are reminiscent of CRT1 [43] . However , the functions of CRT1 and SRFR1 likely differ , because mutations in CRT1 compromise , not enhance , effector-triggered immunity . Because mutations in SRFR1 lead to increased , not decreased resistance , we do not propose that SRFR1 is analogous to RIN4 as the guardee of RPS4 or SNC1 , since deletion of a guardee should prevent recognition of the specific effector that targets the guardee . The function of guardee for SNC1 may be fulfilled by BON1 [44] , although BON1 is localized to the plasma membrane [25] and to our knowledge it has not been determined whether BON1 interacts with SNC1 . Also , because no cognate effector is known for SNC1 and because deletion of BON1 leads to autoactivation of SNC1 , it is difficult to quantify the effects of BON1 mutations on disease resistance and susceptibility . Interestingly , we consistently observed a more severe growth phenotype of srfr1-4 plants compared to bon1-1 plants , yet the srfr1-4 growth phenotype is completely reversed by snc1-11 . Apart from negatively regulating the activation of SNC1 , SRFR1 most likely regulates additional R proteins . Because of positive feed-back , all these pathways may be turned on once SNC1 is activated . While in bon1-1 plants SRFR1 is still present to downregulate these other R proteins , this is not the case in srfr1-4 plants . Therefore , this observation is suggestive of a broad and central function of SRFR1 in downregulating R protein output . It is currently unknown where in the cell the recognition of AvrRps4 by RPS4 occurs . Several plant R proteins , including RPS4 , have been shown to function in the nucleus to trigger immunity [36] , [45] . Because the cytoplasmic pool of these R proteins predominates over the nuclear pool , it is difficult to establish whether R proteins translocate to the nucleus upon effector perception , or continuously cycle between the cytoplasmic and nuclear compartment . We also detected a low amount of SNC1 in the nucleus , whereas the autoactivated mutant snc1-1 protein appears to accumulate to higher levels in the nucleus [41] . It was also found that snc1-1 needed to be in the nucleus to cause a stunted phenotype [41] , and that temperature modulated the localization of snc1-1 [46] . Interestingly , a balanced partitioning of EDS1 between the cytoplasm and nucleus was recently shown to be required for full EDS1-mediated resistance [47] , indicating that immune regulatory proteins may have coordinated cytoplasmic and nuclear functions during the immune response . Here we found that SRFR1 interacts with RPS4 and SNC1 in the cytoplasm , and also that mutations in SRFR1 alter the expression of defense genes independent of a snc1 phenotype . Because of the low amount of RPS4 [36] , SRFR1 and SNC1 protein in the nucleus , so far we have not been able to ascertain whether they also interact in the nucleus . However , our results seem to suggest that at resting state , the majority of SRFR1 , RPS4 and SNC1 protein is extra-nuclear localized and forms a complex in the microsomal fraction . SRFR1 may therefore negatively regulate RPS4 and SNC1 translocation to the nucleus . We propose that a second point of regulation is in the nucleus , where SRFR1 may negatively regulate the transcriptional reprogramming upon pathogen perception . More detailed analyses before and during a defense response are required to substantiate these hypotheses . The genetics of enhanced resistance in RLD srfr1 mutants were originally interpreted to signify that an additional specific R gene is required for resistance [15] . In the mapping crosses rps4-1×srfr1-1 and rps4-1×srfr1-2 , resistant F2 plants were identified in the ratio 13 susceptible to 3 resistant , consistent with segregation of a recessive locus ( srfr1 ) and a dominant locus that was proposed to be a second specific R gene with weak recognition of AvrRps4 [15] . In light of the results presented here , we needed to reinterpret these results . Retesting our mapping population provided evidence for severely stunted plants at the expected ratio of one in 16 stunted plants . These would be double recessives ( srfr1 and wild-type SNC1 ) and would have been lost from our usual phenotypic analysis because of preferential retention of vigorously growing seedlings after planting for disease assays . Upon reinspection , the segregation of resistant plants in the two mapping populations was indeed statistically consistent with the segregation of a single recessive locus ( srfr1 ) in a population where 1/16th of the population ( genotype srfr1/srfr1 SNC1/SNC1 ) that would have been expected to be resistant was eliminated from consideration . In addition , in both mapping populations we had noticed an apparent suppression of recombination along chromosome 4 in retained plants [15] , which is consistent with the fact that both SRFR1 and SNC1 are located on chromosome 4 . At the same time , we show here that the original model for resistance in srfr1 mutants mediated by other R genes with weaker recognition of AvrRps4 is still valid because cross-talk between R genes exists in response to AvrRps4 . However , we now consider it unlikely that one single additional R gene is responsible for resistance in srfr1 mutants . In conclusion , our data contribute to evidence for extensive cross-talk between at least three TNL pathways that converge on SRFR1 , indicating that SRFR1 perhaps has a central function in regulating the output of additional TNL proteins . The present data also allow us to propose more directly that SRFR1 negatively regulates R proteins or R gene expression . While models for SRFR1 so far have focused on a nuclear-localized transcriptional repressor function [18] , the data here suggest that SRFR1 also has a function in the cytoplasm . Consistent with this , Li and co-workers recently showed that SRFR1 interacts with SGT1 in the cytoplasm [22] . Whether SRFR1 is merely an accessory protein in a cytoplasmic “resistasome” or has regulatory functions and migrates to the nucleus remains to be established . Nevertheless , our data highlight molecular architecture aspects of a subset of TNL-mediated resistance pathways that will allow further mechanistic insight into the function of TNL R proteins . The cross-talk evident from our results also means that any reports of constitutive resistance phenotypes in Col-0 need to consider the possible involvement of SNC1 .
The srfr1-4 line ( SAIL_412_E08 ) from the Syngenta Arabidopsis Insertion Library [21] was obtained from the Arabidopsis Biological Resource Center . The T-DNA insertion site in srfr1-4 in the second intron was determined by sequencing and was found to be upstream of the insertion site suggested by raw flanking sequence from the T-DNA Express website ( http://signal . salk . edu/cgi-bin/tdnaexpress ) . rps4-2 ( SALK_057697 ) was isolated from the Salk T-DNA knockout lines [20] . snc1-11 ( SALK_047058 ) and bon1-1 were a kind gift from Jian Hua ( Cornell University ) . Using snc1-11 as a recipient , srfr1-4 snc1-11 and rps4-2 snc1-11 double homozygous mutants were generated . The mutant lines rps4-21 , rrs1-1 and rps4-2 rrs1-1 in the Ws-0 background were kindly provided by Yoshihiro Narusaka ( Research Institute for Biological Sciences , Japan ) . The mapping populations generated by crossing srfr1-1 or srfr1-2 to rps4-1 ( SAIL_519_B09 ) were described previously [15] . Complemented srfr1-4 transgenic lines were generated by transforming srfr1-4 with pSHK102 , a genomic SRFR1 clone in vector pCAMBIA2300 [18] , using the floral dip method [48] . Single locus transgenic lines homozygous for the transgenic copy of wild-type SRFR1 were selected by scoring for kanamycin resistance , the selectable marker of pCAMBIA2300 ( the selectable marker for SAIL lines is BASTA ) . Among these homozygous lines , those with at least one copy of the srfr1-4 allele were selected by genotyping and propagated to the next generation . Lines homozygous for both the SRFR1 transgene and the srfr1-4 allele were identified as those where srfr1-4 did not segregate in the next generation . SNC1 was mapped by genotyping stunted plants in the F2 generation from the cross RLD×srfr1-4 using SSLP and CAPS markers [49] , [50] . Unless otherwise noted , Arabidopsis plants used in this study were grown in E-7/2 reach-in growth chambers ( Controlled Environments Ltd . , Winnipeg , Manitoba , Canada ) under an 8 h light/16 h dark cycle at 24°C and 22°C , with 70% relative humidity and a light intensity of 90–140 µmol photons m−2 s−1 . Virulent Pseudomonas syringae pv . tomato strain DC3000 containing the empty vector ( ev ) pVSP61 or DC3000 expressing avrRps4 from plasmid pVSP61 was grown as described previously [16] . To generate DC3000 hrcC− ( ev ) , pVSP61 was mobilized into the recipient DC3000 hrcC− mutant by triparental mating using the helper plasmid pRK2013 . In planta bacterial growth assays were performed by syringe infiltration . Leaves of 4-week old plants were infiltrated with bacterial suspensions of 5×104 cfu/mL . Leaf discs with a total area of 0 . 5 cm2 per sample were ground in 10 mM MgCl2 , and solutions were plated in serial dilutions on selective medium in triplicate at the indicated time points . Statistical comparison of bacterial growth was tested using a two-tailed Student's t-test . Quantitative reverse transcription PCR was performed as described previously [18] . Briefly , total RNA was extracted from the indicated plant lines using TRIZOL ( Invitrogen , Carlsbad , CA , USA ) . For RT-PCR experiments , cDNA was synthesized from 2 µg of total RNA using an oligo ( dT ) 15 primer and Moloney murine leukemia virus ( MMLV ) reverse transcriptase ( Promega , Madison , WI , USA ) following the manufacturer's protocol . Quantitative real-time reverse transcription PCR ( qPCR ) was performed with SYBR GREEN PCR Master Mix and an ABI 7500 system ( Applied Biosystems , Warrington , UK ) according to the manufacturer's instructions . The levels of transcripts were normalized using SAND gene ( At2g28390 ) mRNA levels as an internal standard . These experiments were performed at least twice with similar results . Semi-quantitative RT-PCR was performed from total RNA extracted from Col-0 and srfr1-4 . Equivalent amounts of cDNA from both samples were used to detect PR1 , PR2 and PDF1 . 2 . ACTIN2 ( At3g18780 ) was used as an internal control . Table S1 lists the oligonucleotide primer sequences used in qPCR and semi-quantitative RT-PCR . To determine the SNC1 cDNA sequence from RLD and Col-0 , the 3′-RACE procedure ( Invitrogen , Carlsbad , CA , USA ) and RT-PCR ( see above ) were performed as described previously [19] . PCR products were ligated into the pGEM-T Easy vector ( Promega ) for sequencing . See Table S1 for oligonucleotide primer sequences used in these experiments . All clones were verified by sequencing . To generate epitope-tagged SNC1 constructs , genomic SNC1 DNA including introns was amplified by PCR from Col-0 using SNC1 GATE primers listed in Table S1 . In vitro BP Clonase recombination reactions were carried out to insert the PCR product into the pDONR201 entry vector according to the manufacturer's instructions ( Invitrogen ) . LR reactions were performed to recombine the entry clones into GATEWAY-compatible destination vectors . Using BP and LR reactions , we constructed Myc-gSNC1 with six Myc tags under the control of the cauliflower mosaic virus 35S promoter . Similarly , Myc-gRPS4 was generated by amplifying the genomic fragment of RPS4 from the FLAG-gRPS4 construct [51] using the primers RPS4 FOR and RPS4 REV ( Table S1 ) . To construct the binary vector expressing genomic HA-tagged SRFR1 from its native promoter ( HA-gSRFR1 ) , independent PCR reactions were performed with the primer combinations HA-SRFR1 FOR/gSRFR1 XbaI REV and pCAMBIA PmeI FOR/HA-SRFR1 REV using the template pSHK102 [18] . The PCR products were mixed and used for overlap PCR with the pCAMBIA Pme I FOR/gSRFR1 XbaI REV primers . The 2 . 2 kb PCR product was digested with PmeI and XbaI and used for replacing the PmeI-XbaI fragment of pSHK102 . The resulting binary vector was electroporated into Agrobacterium tumefaciens strain C58C1 . Transgenic N . benthamiana plants expressing HA-gSRFR1 from the Arabidopsis native promoter were generated by stable Agrobacterium-mediated transformation as previously described [52] . Transgenic plants were selected on media containing 100 µg/ml kanamycin . Microsomal and soluble fractions were prepared according to published procedures [53] . Briefly , plant materials were ground in buffer H ( 50 mM HEPES , pH 7 . 5 , 250 mM sucrose , 15 mM EDTA , 5% glycerol , 0 . 5% polyvinylpyrrolidone ) containing 3 mM DTT and 1×protease cocktail inhibitors ( Sigma , St . Louis , MO ) . The extracts were filtered through two layers of miracloth pre-wetted with buffer H and centrifuged at 2000×g for 15 min at 4°C . The supernatant consisting of the cytoplasmic fraction was further subjected to ultracentrifugation at 100 , 000×g to separate the soluble and microsomal ( pellet ) fractions . The pellet was resuspended in buffer H . Nuclear extracts were prepared using the CelLytic™ PN Isolation/Extraction Kit ( Sigma ) following the manufacturer's instructions . Total protein concentrations of fractions were determined by Bradford assays with BSA as standard . Extracts were normalized to 1 µg/ml with buffer H . For co-immunoprecipitation assays , the nonionic detergent Igepal CA-630 ( Sigma ) was added to 0 . 2% and 1% final concentration to the soluble and microsomal fractions , respectively . The extracts were incubated overnight with 20 µl of anti-HA or anti-Myc agarose beads ( Sigma ) . The beads were washed three times with buffer H containing 0 . 2% Igepal CA-630 . The immunoprecipitates were analyzed by immunoblot assays with anti-Myc-HRP ( Santa Cruz Biotechnology ) or anti-HA-HRP ( Roche ) antibodies . The degree of enrichment in cellular fractionation was determined by immunoblot analyses with anti-GAPDH ( Genscript , Piscataway , NJ ) , anti-V-ATPase ( Agrisera , Vännäs , Sweden ) , anti-histone H3 ( Abcam , Cambridge , MA ) and anti-RNA pol I ( Agrisera ) antibodies . SNC1: At4g16890; SRFR1: At4g37460; RPS4: At5g45250; RPP4: At4g16860; NPR1: At1g64280; EDS1: At3g48090; PAD4: At3g52430; SID2: At1g74710; PR1: At2g14610; PR2: At3g57260; PDF1 . 2: At5g44420; SAND: At2g28390; ACTIN2: At3g18780 . | Plants , like humans , have an immune system to defend against disease . This immune system seeks out the presence of disease-causing microbes and other invaders by detecting non-plant molecules and proteins . Plants rely on this surveillance to activate an antimicrobial response of appropriate strength at the right time; as with humans , an overactive immune system can be harmful to plants . We study how plants achieve an appropriate balance , using genetics and the interaction between the reference plant Arabidopsis thaliana and the bacterial plant pathogen Pseudomonas syringae . So-called plant resistance proteins are important activators of immunity that directly or indirectly intercept foreign proteins deployed by pathogens . Resistance proteins are generally thought to be highly specific detectors that only respond to a single pathogen protein . However , while working with a negative regulator of plant immunity called SRFR1 , we discovered a surprising level of cross-talk between different resistance proteins that becomes evident only under certain environmental conditions such as low temperature . We also show that SRFR1 and these resistance proteins bind to each other , possibly explaining the observed cross-talk . Our work thus highlights linkages between resistance pathways and provides insight into the molecular architecture of the plant innate immune response . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"plant",
"biology/plant-biotic",
"interactions",
"genetics",
"and",
"genomics/gene",
"discovery",
"plant",
"biology/plant",
"genetics",
"and",
"gene",
"expression"
] | 2010 | The Arabidopsis Resistance-Like Gene SNC1 Is Activated by Mutations in SRFR1 and Contributes to Resistance to the Bacterial Effector AvrRps4 |
Rabies has been eliminated from domestic dog populations in Western Europe and North America , but continues to kill many thousands of people throughout Africa and Asia every year . A quantitative understanding of transmission dynamics in domestic dog populations provides critical information to assess whether global elimination of canine rabies is possible . We report extensive observations of individual rabid animals in Tanzania and generate a uniquely detailed analysis of transmission biology , which explains important epidemiological features , including the level of variation in epidemic trajectories . We found that the basic reproductive number for rabies , R0 , is very low in our study area in rural Africa ( ∼1 . 2 ) and throughout its historic global range ( <2 ) . This finding provides strong support for the feasibility of controlling endemic canine rabies by vaccination , even near wildlife areas with large wild carnivore populations . However , we show that rapid turnover of domestic dog populations has been a major obstacle to successful control in developing countries , thus regular pulse vaccinations will be required to maintain population-level immunity between campaigns . Nonetheless our analyses suggest that with sustained , international commitment , global elimination of rabies from domestic dog populations , the most dangerous vector to humans , is a realistic goal .
Rabies has been one of the most feared diseases throughout human history and has the highest human case-fatality proportion of any infectious disease [1 , 2] . Every year over 7 million people receive post-exposure prophylaxis , and an estimated 55 , 000 people die from rabies [3] ( more than yellow fever , dengue fever , or Japanese encephalitis [4] ) . Over 99% of these deaths occur in developing countries where rabies is endemic in domestic dog populations [5] . However , the impacts of canine rabies are often overlooked , largely because human rabies deaths are now extremely rare in Western Europe and North America , where mass vaccination successfully eliminated the disease from domestic dog populations [6] . Increasing incidence of canine rabies in Africa and Asia has prompted concerns that similar strategies may not be effective in these areas [7 , 8] . The critical question now is whether global elimination of domestic dog rabies is achievable . Keys to answering this question include: a quantitative understanding of the transmission dynamics of rabies in domestic dog populations , particularly the basic reproductive number , R0; a quantitative understanding of domestic dog demography; and information about the practicality and effectiveness of various vaccination strategies . While recent data support the feasibility and practicality of domestic dog vaccination strategies [9–11] , there are very little quantitative data on rabies transmission dynamics [12] and the underlying demographic processes . Transmission is the most important process underlying infectious disease dynamics [13] , but it is also the least understood . Rates of transmission are usually inferred from population patterns of disease incidence , but population-level analyses do not capture between-individual variation in transmission resulting from differences in behaviour , genetics , immune status , and environmental and stochastic factors , which play an important role in determining disease dynamics [14 , 15] . Contact tracing has been used to directly measure case-to-case transmission , and applications of the technique to emerging infections such as SARS have generated important insights into disease transmission and control in human populations [16 , 17] , but transmission processes for diseases circulating in animal populations are much harder to study . Rabies is an acute viral encephalitis that is spread through the saliva of infected hosts [2] . Clinical manifestations vary , but the neurological phase often includes increased aggression and the tendency to bite and thereby transmit infection; rapid progression to death is inevitable [4] . These distinctive signs make transmission of rabies easier to track than that of most other diseases and provide an unusual opportunity to explore epidemiological patterns at the scale of the individual . Here , we present data on rabies transmission in two districts of rural Tanzania , Serengeti and Ngorongoro ( Figure 1 ) . We were able to monitor the spread of infection using contact-tracing methods , which were feasible due to the discrete and memorable nature of transmission events . We recorded >3 , 000 potential transmission events between 2002 and 2006 and reconstructed case histories of over 1 , 000 suspect rabid animals that illustrate heterogeneity in several aspects of transmission , including the latency , movement patterns , and biting propensity of infected individuals . Although these districts border the Serengeti ecosystem , we have argued that domestic dogs are the sole maintenance population of rabies in this community: they make up over 90% of our observations of rabid animals , and the >70 isolates that have been sequenced ( from 13 host species ) are all consistent with the Africa 1b canid strain [18 , 19] . This is one of the most extensive datasets on individual transmission events assembled in an animal population; it has potential to shed light on critical , but often elusive , details of infectious disease transmission . We also analyze data from rabies outbreaks around the world , which provide a global and historical context for the Tanzania dataset .
Analyses of the contact-tracing data generated robust estimates of epidemiological parameters that have important implications for rabies control ( Table 1 , Figures 2 and 3 , and Figure S1 ) and provide insight into how infectious disease transmission scales from individual behaviour to population-level dynamics . We estimated R0 for rabies in Serengeti and Ngorongoro districts directly from infectious histories , from reconstructed epidemic trees based on the spatiotemporal proximity of cases , and from the exponential rate of increase in cases at the beginning of an epidemic . Biting behaviour of rabid dogs during the course of infectious periods was highly variable ( mean bites per rabid dog = 2 . 15 , 95% confidence interval ( CI ) from fitting a negative binomial distribution: 1 . 95–2 . 37; variance = 5 . 61 , CI: 4 . 63–6 . 92; shape parameter k = 1 . 33; CI: 1 . 23–1 . 42 ) ( Figure 3A ) . The probability that an unvaccinated dog developed rabies after being bitten by an infectious animal was high ( Prabies|bite = 0 . 49 , CI: 0 . 45–0 . 52 ) ( Table 1 ) if the bitten dog was not vaccinated or killed immediately after exposure . Multiplying the average number of dogs bitten per rabid dog by the probability of developing rabies following exposure gave an R0 estimate of 1 . 05 ( CI: 0 . 96–1 . 14 ) ( Figure 3A and Table 1 ) . These estimates should be regarded as lower bounds , because not all transmission events were observed ( this calculation excludes rabid dogs that were killed before biting other animals or that disappeared and likely corresponded to unknown or unobserved rabid dogs in other areas; see Materials and Methods ) . Detailed data on the timing and location of transmission events and infections allowed us to estimate the spatial infection kernel and generation interval ( distances and times between source cases and their resulting infections , respectively ) ( Figure 2 ) and probabilistically reconstruct transmission networks ( Videos S1 and S2 ) . Calculating the average number of secondary cases per rabid dog during the period of exponential epidemic growth ( before vaccinations were implemented ) from these reconstructions gave similar R0 estimates of 1 . 1 in Serengeti district and 1 . 3 in Ngorongoro ( CIs: 1 . 04–1 . 10 and 1 . 26–1 . 42 , respectively ) ( Table 1 ) . The more traditional approach of estimating R0 , by fitting a curve to incidence data over the same interval of exponential epidemic growth , also produced similar estimates of 1 . 2 in Serengeti and 1 . 1 in Ngorongoro ( CIs: 1 . 12–1 . 41 and 0 . 94–1 . 32 , respectively ) ( Table 1 and Figure 3B ) . This approach is robust to underreporting ( Text S1 and Figure S2 ) but should likewise be considered a lower bound , because some local control measures were instituted ( such as tying or killing ) . We also estimated R0 from the intrinsic growth rate of outbreaks of domestic dog rabies elsewhere in the world ( Table 2 ) and obtained values between 1 . 05 and 1 . 85 , which are consistent with our estimates from northwest Tanzania . For many diseases , R0 is expected to increase with host density [12 , 13 , 20 , 21] . Despite the domestic dog population density in Serengeti ( 9 . 38 dogs/km2 ) being considerably higher than the dog population density in Ngorongoro ( 1 . 36 dogs/ km2 , see Table 3 ) , we were unable to detect significant differences in our estimated values of R0 between the two districts . Nor did we find any conspicuous differences in R0 estimated from the outbreaks listed in Table 2 , which represent a wide range of population densities . There may , in fact , be no relationship between R0 and population density for canine rabies . On the other hand , a subtle relationship between dog density and transmission rates might be difficult to detect for a number of reasons . To investigate whether it would be possible to decipher systematic differences in R0 across the range of values that we estimated , we simulated outbreaks using our epidemiological parameter estimates , but varied R0 ( from R0 = 1 to R0 = 2 ) , whilst maintaining individual variance in biting behaviour ( same shape parameter k , see Text S2 ) . Although the mean estimates of R0 from fitting to these simulated trajectories were accurate , they were surrounded by wide confidence intervals ( Figures S2 and S5 ) , suggesting that if only a small number of epidemics were sampled , any underlying relationship might not be apparent . Several mass domestic dog vaccination campaigns were carried out in villages in the study districts during the 5-y period . We analysed the impacts of these interventions at the village level to capture the wide variation in achieved levels of vaccination coverage . We incorporated demographic processes ( Table 3 gives demographic parameter estimates ) and waning of vaccine-induced immunity ( see Materials and Methods ) , because these affect the level of herd immunity within the population at any one time . There were no rabies outbreaks ( defined as at least two cases not interrupted by an interval of more than one month ) in villages when vaccination coverage exceeded >70% . Small outbreaks occurred in villages with lower coverage and the largest ( and longest ) outbreaks only occurred in villages with <20% coverage . Observed outbreak sizes were within the range expected from the heterogeneity of biting behaviour and the coverage achieved by village-level vaccination campaigns ( Figure 4A ) . The effective reproduction number R , which describes transmission once an epidemic is underway , declined during the course of the observed epidemics ( Figure 3C ) . At the level of individuals , vaccination coverage reduced the number of secondary cases per rabid dog ( Figure 4B ) . More than 300 vaccinated dogs were identified by contact tracing as having been bitten by rabid animals . Only ten of these animals showed any signs indicative of rabies , although in the absence of vaccination approximately 50% ( Prabies|bite = 0 . 49 ) ( Table 1 ) of these would have been expected to succumb to the disease . Individual actions by dog owners such as tying or killing exposed or infectious animals also had an impact . By killing rabid dogs , villagers reduced the overall average infectious period by around 16% ( 3 . 7 d for rabid animals that died from the disease versus 3 . 1 d for all infected animals , including those that were killed ) . However , there were no consistent declines through time in the number of bites by rabid dogs ( Figure S3 ) . Thus we consider vaccination to have been the overwhelming factor in curtailing the outbreaks ( Figure 4A ) . From our estimates of R0 , we calculate the deterministic critical vaccination threshold for rabies elimination in rural Tanzania to be only 20% ( Pcrit = 1 – 1/R0 ) , and even in areas where R0 is higher , Pcrit rises to just 40% ( Table 2 ) . Our observations and simulations ( Figure 4 ) demonstrate that small outbreaks occur by chance even when coverage exceeds Pcrit and should be expected more frequently when there is individual variation in transmission ( Text S2 ) . Herd immunity declines rapidly in the interval between vaccination campaigns because of births and deaths in the domestic dog population ( Figure 4 , inset ) . To maintain herd immunity above Pcrit between campaigns , therefore , requires a larger proportion of the dog population , Ptarget , to be vaccinated ( Ptarget = e ( ν+d+r ) T Pcrit , where r is the rate of dog population growth , d is the death rate , 1/ν is the duration of vaccine-induced immunity , and T is the interval between campaigns ( see Materials and Methods ) ) . By incorporating demographic parameters ( Table 3 ) , we estimate that annual campaigns should therefore aim to vaccinate 60% of the dog population to avoid coverage falling below Pcrit .
The basic reproductive number , R0 , is the average number of secondary infections produced by an infected individual in an otherwise fully susceptible population [20] . R0 is the most important parameter in infectious disease epidemiology , and considerable effort has been devoted to its estimation and to understanding its implications for disease control [20 , 22–26] , although it is important to note that some factors not incorporated in R0 , e . g . , host births as well as deaths , may also have important control implications . Depending upon the quality and quantity of data , a number of approaches can be used to estimate R0 . Choosing the most appropriate method and assessing its accuracy can be difficult , given the associated assumptions and shortcomings [22] . Most methods do not account for variability in the pathogenesis and behaviour of infected animals; some methods make inferences from quantities that are confounded by ( often unmeasured ) responses to disease incidence ( e . g . , epidemic size or prevalence at equilibrium ) ; and different methods are variously biased due to measurement and process error . Although our attempts to estimate R0 are also imperfect , they do incorporate individual variation in behaviour and pathogenesis , explicitly address several common assumptions , and have been carefully checked for biases through extensive simulations . The overall consistency in the low values of R0 that we estimated ( ∼1 . 1 < R0 < 2 ) is therefore reassuring and provides optimism for the feasibility of canine rabies control by vaccination . If R0 increases with host density in this system , different threshold levels of vaccination coverage would be necessary to eliminate disease in different density populations [12 , 20] . However , our data on individual variation in biting behaviour also illustrate that it would be difficult to detect statistical differences in the range of R0 values that we estimated ( Figure S2 ) . Thus in practice , when only a small number of epidemics are observed , individual variation in transmission may mask any underlying variation in R0 driven by population density . So although we cannot decipher the relationship between population density and rabies transmission , the consistency of our individual- and population-level estimates from Tanzania and from a wide range of sites around the world allow us to estimate the threshold vaccination coverage necessary to eliminate the disease . Our estimates of R0 predict that only relatively low levels of vaccination coverage are required to eliminate rabies ( ∼20–45% ) , but there is considerable variation in empirically observed levels of coverage that have successfully controlled the disease; low levels of coverage ( 30–50% ) have been successful in some circumstances [27] , although higher levels have also failed [28] . Our analyses suggest that these inconsistencies are , in large part , a consequence of host demography . When vaccinations are carried out in pulses , births and deaths within the host population will continuously reduce the level of herd immunity attained during campaigns ( Figure 4 , inset ) . Turnover of domestic dogs in rural Tanzania is very high ( Table 3 ) ; therefore , annual campaigns should aim to vaccinate 60% of the dog population to maintain vaccination coverage above Pcrit for the duration of the interval between campaigns . When successive campaigns have achieved this , rabies incidence has declined dramatically despite high endemic levels in adjacent areas [29] . Domestic dog population turnover therefore appears to have had a marked influence on rabies dynamics that explains the variable success of vaccination efforts . The empirically derived consensus that 70% coverage is sufficient for long-term rabies elimination [30 , 31] was likely reached because it is effective as a target for annual campaigns in almost all demographic settings , including those with particularly high turnover such as those we describe from Tanzania . There are other potential explanations and caveats . The nutritional and health status of animals might affect the development of protective immunity in response to vaccination . However , more than 97% of dogs sampled from Serengeti district developed strong antibody titres ( >0 . 5 IU/ml ) in response to vaccination [32] , suggesting that these factors do not impair the efficacy of dog vaccination in rural Tanzania . In addition , numerous practicalities—such as occasional failures in the cold chain , improper vaccination of animals , mistaken registrations , etc . —will all reduce the level of population immunity below the estimated vaccination coverage . Furthermore , our observations and simulations confirm that small outbreaks may occur simply by chance even when coverage exceeds Pcrit [33] , and these are particularly likely when there is individual variation in transmission ( Figure 4 ) . Higher levels of coverage are therefore necessary to reduce the chance of outbreaks with greater certainty; especially where the risk from imported infections is highest ( Figure 4C ) . This could be a concern if canine rabies were to be eliminated from domestic dog populations but continued to circulate in sympatric wildlife; however , canine rabies was successfully eliminated in Western Europe and North America despite the presence of wildlife hosts capable of transmission . Thousands of people die every year from this horrific and preventable disease , because the control of canine rabies has been severely neglected in developing countries [2] . Inherent inter-annual periodicity of epidemics exacerbates the situation , with rabies only intermittently perceived as problematic [6] , as illustrated by the recent outbreak in China [34] . The problem of canine rabies has often been considered intractable in rural Africa , because of poor infrastructure , limited capacity , and the misperception that large populations of wild carnivores are responsible for disease persistence . Our analyses show that global control of canine rabies is entirely feasible and that successful elimination of canine rabies in many parts of the world has likely been achieved precisely because R0 is so low and institutional commitment to maintain high levels of vaccination coverage has been sustained [6] . Achieving vaccination coverage of 60% or more in dog populations in Africa is both logistically and economically feasible through annual vaccination campaigns [9–11 , 29] . The resultant reduction in costs of human post-exposure prophylaxis suggest that vaccination interventions targeted at domestic dog populations could translate into appreciable savings for the public health sector [3 , 8 , 29] . Furthermore , the inherently low R0 and the tractability of rabies contact-tracing indicates that once endemic rabies is controlled , elimination could be achieved through active case detection in remnant foci of infection ( much like the strategy used to eradicate smallpox [35] ) ; similar measures are proving effective in programmes to eliminate canine rabies in the Americas [36] . However , the most crucial step towards global elimination of canine rabies will be sustained commitment and coordinated efforts to maintain sufficient vaccination coverage in domestic dog populations .
We collected data from two districts in northwest Tanzania: Serengeti , inhabited by multi-ethnic , agro-pastoralist communities and high-density dog populations , and Ngorongoro , a multiple-use controlled wildlife area , inhabited by low-density pastoralist communities , predominantly Maasai , and lower-density dog populations ( Figure 1 ) . Attributes of the dog populations in these districts are presented in Table 3 . Wildlife populations also differ in the two districts , but domestic dogs are the focus of this study because they are the only maintenance population of rabies in the area [18] . Data on patients with animal-bite injuries from hospitals and dispensaries , case reports of rabid animals from livestock offices , and community-based surveillance activities were used as primary sources [18] . Visits were made to investigate incidents reported in 2002 to 2006 involving suspected rabid animals . Cases were mapped at the site of the incident ( wherever possible ) and villagers interviewed to evaluate the status of the biting animal , determine its case history , and identify its source of exposure and subsequent contacts ( if known ) . The same procedure was exhaustively followed for all associated exposures/cases . Interviews were conducted with veterinary officers , local community leaders , and livestock field officers in attendance , resulting in an active reporting network . Cases were diagnosed on epidemiological and clinical criteria , adapting the “six-step” method through retrospective interviews with witnesses [37] . Rabies was suspected if an animal displayed clinical signs [37] and either ( a ) disappeared or died within 10 days , or ( b ) was killed , but had a history of a bite by another animal or was of unknown origin . Additional clinical criteria for wild carnivores ( ∼10% of human exposures were caused by wild animals and ∼10% of inferred transmission events involved rabid wildlife ) included tameness , loss of fear of humans , diurnal activity ( for nocturnal species ) , and unprovoked biting of objects and animals without feeding . When multiple incidents involving suspected rabid wildlife were reported on the same/consecutive days within neighbouring homesteads , we assumed a single animal was involved . Brain samples were collected and tested for confirmation wherever possible , but despite efforts to obtain diagnostic samples , most cases reported here were suspected rather than confirmed . Inadequate sample preservation such as storage at room temperature and long intervals between sample collection and testing ( during which samples underwent repeated freeze-thaw cycles ) probably caused specimens to deteriorate . Composite samples of each brain necessary to achieve the highest test reliability were also rarely available . Nevertheless , a high percentage of samples from suspected cases of rabies were confirmed by laboratory diagnosis ( ∼75% ) suggesting that use of epidemiological and clinical criteria is justified and reliable [18] . Researchers are encouraged to contact the authors regarding data availability . Dog vaccination campaigns in Serengeti district in 2000 resulted in low and patchy vaccination coverage ( 35–40% estimated from post-vaccination household surveys ) . Annual campaigns conducted from 2003 onwards in a 10-km zone adjacent to the western border of Serengeti National Park achieved higher coverage levels of between 40 and 80% . In 2004 , the Tanzanian government conducted vaccinations in villages in Serengeti district beyond the 10-km zone reaching 55% coverage across the remainder of the district , but in subsequent years , campaigns were less systematic and conducted in fewer villages . Vaccination in Ngorongoro was restricted to small-scale localised campaigns in the district town centre until 2004 , whereupon widespread annual vaccinations were implemented with overall coverage exceeding 80% [9] . Data on the number of dogs vaccinated in each village and on each campaign date were collected from 2003 onwards . The incubation period and duration of infectiousness were estimated for rabies in domestic dogs from records of when individual dogs were bitten , developed clinical signs , and were killed or died . Gamma distributions were fitted to these data using maximum likelihood with interval censoring to account for cases where the relevant data were only approximately known ( Figure 2 and Table 1 ) . To estimate the probability distribution of the generation interval , G ( t ) , an incubation and an infectious period were drawn from their respective distributions , a “time-to-bite” deviate was drawn from a uniform distribution over the interval of the infectious period , and the two intervals were summed . There was a significant correlation between the length of the infectious and incubation periods , but significance was entirely due to a single data point; we therefore treated the distributions as independent . The spatial infection kernel K ( d ) was estimated by fitting a gamma distribution to the distances between known source cases and animals that they contacted . Many contacts occurred within the same , or neighbouring , homesteads . In these cases , the precise distance was not always recorded , but we assumed it was less than 100 m . We therefore replaced the probability of a contact within 100 m by the probability distribution averaged over the range 0–100 m . ( 1 ) Direct estimates from infectious histories . Using maximum likelihood , we fitted a negative binomial distribution to data on biting behaviour of rabid dogs ( Figure 3A ) . The probability of developing rabies following a bite ( Prabies|bite ) was estimated , excluding bitten animals that had previously been vaccinated , or that were either killed or vaccinated immediately after the bite , and binomial confidence intervals were calculated . R0 was estimated as the probability Prabies|bite multiplied by the average number of bites per rabid dog and confidence intervals were calculated using a resampling procedure . Dogs that were removed ( killed or tied up ) before causing secondary cases in other dogs ( even if they bit people ) were excluded from this calculation , as were suspect rabid dogs that either disappeared before biting other dogs or that were of unknown origin and were killed before being observed to bite other dogs ( Figure 3A ) . We pooled data from both districts for this estimate because insufficient complete case-histories of rabid dogs ( after excluding cases with interventions ) were traced to accurately estimate R0 for Ngorongoro ( 35 versus 477 in Serengeti ) . We also estimated R0 directly from the distribution of secondary cases per rabid dog . Dogs that were bitten by rabid animals but did not develop rabies because of interventions ( previous vaccination or being killed/vaccinated immediately after the bite ) were multiplied by Prabies|bite and added to observed secondary cases , giving an expected number of secondary cases per rabid dog in the absence of intervention and a similar estimate of R0 ( 1 . 14 , CI: 1 . 03–1 . 25 ) ( Figure S1 ) . ( 2 ) Epidemic tree reconstruction . We used an algorithm for probabilistically constructing epidemic trees based on the location of cases in space and time [38] . For each suspected case ( i ) , we chose a progenitor ( j ) at random with probability pij from all n cases preceding that case , where: G is the distribution of generation times , tij is the length of time ( in days ) between the occurrence of case i and its potential progenitor j ( G ( t ) = 0 for t < 0 ) , K is the spatial infection kernel , and dij is the distance ( in km ) between the locations of case i and its potential progenitor j ( using the average probability when distance <100 m , see above ) . Because the dates that some individuals were bitten or developed rabies were only approximately known , 1 , 000 bootstrapped datasets were generated with the dates drawn randomly from a uniform distribution over the window of uncertainty and a consensus tree of the most probable links was determined and used to generate secondary case distributions illustrated in Figure S1 . Because transmission of rabies from livestock is recorded extremely rarely , we did not allow livestock progenitors , which considerably improved the match between known and assigned links compared to an algorithm where all species could be assigned as progenitors . All detected cases in carnivores ( including domestic cat and wildlife cases ) were included in the tree reconstructions using the spatial infection kernel and generation interval parameters estimated for domestic dogs . The contribution of nondomestic dog carnivores to the overall epidemic was small , and estimates of within- and between-species transmission are described elsewhere [18] . When known links between primary and secondary cases were not retained in the trees , they were correctly assigned in more than 60% of cases in both districts , indicating that probabilistic reconstruction was effective . The average number of secondary cases putatively produced from each primary case was calculated from the bootstrapped trees . R0 was estimated as the average number of infections caused per rabid dog that was infectious during the period of exponential epidemic growth . Determining the period of exponential growth is somewhat subjective; for consistency between methods , we used the interval that gave the median R0 value for time series regression estimates ( see below ) . The choice of interval caused more variance in R0 estimates for this reconstruction technique than for other methods because it averages the heterogeneous behaviour of a small number of individual animals that spark an epidemic . Thus inclusion or exclusion of particularly infectious individuals has a large effect on R0 . ( 3 ) Inference from the epidemic curve . A single infection will cause future cases distributed according to the probability distribution of the generation interval . Therefore the number of cases arising in any given interval is the result of those cases that occurred at times in the past whose secondary cases occur in this interval and is determined by the probability distribution of the generation interval . This intuitive description is formalized by the Euler-Lotka equation , adapted for an infection process [25] and an expression for R0 can be obtained: We estimated the initial growth rate of the epidemic ( r ) by fitting an exponential curve to incidence data using a generalized linear model . We compared Akaike's Information Criterion values to determine the appropriate error structure ( Poisson or negative binomial ) . The choice of which part of the epidemic curve the model should be fit to was subjective , therefore the model was fit to all possible sections of the epidemic curve ( using a minimum of nine consecutive months ) and the median , the 2 . 5th and the 97 . 5th percentile of the R0 estimates are presented in Table 1 . Figure 3B ( inset ) shows that the estimate of R0 was robust to the interval chosen for fitting the curve . We used the same method to estimate R0 from data that we had compiled on outbreaks of canine rabies from elsewhere in the world . For these time series , we fitted exponential curves to the intervals between the first recorded case and the month ( or week ) with highest rabies incidence ( Table 2 ) and converted the estimated growth rates to estimates of R0 using the serial interval distribution data gathered by contact tracing in Tanzania . For partly vaccinated populations , we corrected our R0 estimates by dividing by the fraction of dogs which were vaccinated prior to the outbreak [12] . For all the outbreaks considered , including those in Tanzania , some localized and individual control measures may have been instituted ( such as tying up or killing infected animals ) , and therefore our R0 estimates should be regarded as lower bounds . However simulations also revealed that for very low values of R0 ( <1 . 2 ) , estimates from the epidemic trajectory can be slightly biased upwards ( Figure S2 ) . This is probably because at very low levels of R0 , most introductions do not initiate further cases and therefore a small number of individuals with higher than average biting behaviour are needed to trigger epidemics , thus biasing trajectories . The effective reproductive number R measures the average number of secondary cases per primary infection once an epidemic is underway . R changes through space and time depending upon the implementation of control measures , the depletion of susceptibles and the build-up of local correlations in the spatial distribution of infected and susceptible individuals . Numbers of secondary cases per rabid dog ( inferred from the epidemic tree reconstructions ) were calculated monthly and averaged across bootstrapped trees to give a time-varying estimate of R ( Figure 3C ) . Although R declined through time in both districts , there was no apparent temporal trend in the biting behaviour of rabid dogs ( Figure S3 ) , suggesting that domestic dog vaccination was the main factor responsible for reducing transmission . To calculate vaccination coverage and the decline in herd immunity due to population turnover and waning of vaccine-induced immunity , it was necessary to estimate the size of the domestic dog populations ( N ) and their rates of growth ( rdogs ) . We projected human population sizes in both districts using 2002 national census data [39 , 40] , and we calculated human:dog ratios from household questionnaires conducted in 1994 , 2003 , and 2008 in Serengeti district and in 1994 and 2004 in Ngorongoro district . We then estimated dog populations from the projected human population sizes and the human:dog ratios and calculated the rate of increase of the dog population in each district ( rdogs = log ( Nt/N0 ) /t ) ( Table 3 ) . An alternative estimate of the rate of domestic dog population growth was derived from demographic data collected using household questionnaires . The death rate of dogs ( d ) was calculated using a Cox proportional hazards model of survival from longitudinal data ( n = 802 ) . When pups ( dogs under 3 months of age ) were excluded from the model , neither age nor sex significantly affected survival . The per-capita birth rate ( b ) was assumed to be the product of the sex ratio ( ρ ) , the average litter size ( l ) , and frequency ( ϕ ) and pup survival ( s ) ( b = ρlϕs ) . These demographic parameters were estimated from cross-sectional data ( 309 litters ) and the rate of increase was calculated ( rdogs = b – d ) . Pup survival was estimated from a subset of puppies that remained in the household , because of the unknown fate of puppies that were given away or sold . We suspect that mortality of female puppies is greater than males . However , obtaining reliable data to accurately estimate pup survival is difficult , and the result of assuming equal mortality rates is an estimate of rdogs that is more conservative with respect to vaccination coverage ( i . e . , results in lower population-level immunity ) . This estimate of rdogs ( 0 . 088 dogs/y ) was similar to other estimates from the region [41 , 42] and close to those calculated directly from population sizes ( rSerengeti = 0 . 090 dogs/y , rNgorongoro = 0 . 102 dogs/y ) ( Table 3 ) . A comparison of the stable age distribution ( calculated from cross-sectional data assuming a roughly constant rate of population growth ) was consistent with age distributions predicted from the estimated demographic parameters . To evaluate whether the predicted level of vaccination coverage required to control rabies ( Pcrit = 1 – 1/R0 ) was sufficient in practice [20] , we plotted the size of village-level outbreaks ( an outbreak was defined as at least two cases not interrupted by an interval of more than one month ) against vaccination coverage in that village at the time of the case that initiated the outbreak . Vaccination coverage was modeled by susceptible reconstruction using demographic parameters described above ( we show the results from using the largest estimate of rdogs ( 0 . 10 dogs/y ) because this gives the most conservative predictions of the impacts of vaccination , but results are very similar using the lower rdogs estimates ) . We assumed coverage was approximately 20% in January 2002 and that the duration of vaccine-induced immunity ( 1/ν ) was approximately 3 y ( http://www . intervet . co . uk/Products_Public/Nobivac_Rabies/090_Product_Datasheet . asp ) . Numbers of vaccinated and susceptible animals within a village were adjusted according to the doses of vaccine used at village vaccination stations on each campaign date ( sufficient vaccine was provided such that all animals brought to the station could be vaccinated ) . A time series of cases in a village and the associated susceptible reconstruction are shown in the inset of Figure 4A . To predict the expected size of outbreaks given the observed variability in transmission , we simulated outbreaks in a starting population of 500 dogs ( similar to the domestic dog population size in an average village ) ; this choice had little effect on our results . We used our parameter estimates ( Table 1 ) to randomly assign secondary cases and corresponding generation intervals . Each realization was seeded by a single animal and the starting population was initialized with vaccination coverage generated from a binomial distribution . For comparison with the outbreak data we conditioned each realization upon >1 secondary case ( Figure 4A ) . Demographic parameters were incorporated , and 10 , 000 runs were completed for each starting condition . We also calculated the probability of an outbreak of a particular size or larger being seeded by one infectious case to evaluate the coverage needed to prevent outbreaks with different degrees of certainty ( Figure 4C and Figure S4 ) . If V and N denote numbers of vaccinated individuals and the total population size respectively , then vaccination coverage can be expressed as a proportion P = V/N . The number of vaccinated dogs declines following a campaign as individuals die and as vaccine-induced immunity wanes ( Vt = V0e– ( d+ν ) t , where d is the death rate and 1/ν is the duration of vaccine-induced immunity ) , whereas the total population grows at the rate of population increase ( Nt = N0ert ) . To prevent sustained endemic transmission , vaccination coverage must be maintained above Pcrit ( such that R is held below 1 ) . From our estimates of demographic parameters and R0 , we calculated the proportion of the population that needs to be vaccinated , Ptarget , to prevent vaccination coverage falling below Pcrit during the interval , T , between campaigns: Ptarget = e ( ν+d+r ) T Pcrit . This formulation for estimating the coverage needed to interrupt endemic transmission given turnover in the domestic dog population assumes that immunity from vaccination lasts an average of 1/ν time units and declines exponentially . In reality , vaccine-induced immunity is likely to be closer to a fixed duration , and thus fewer dogs would be expected to lose immunity during the 1-y interval between campaigns than under the exponential model . This indicates that our estimate of Ptarget may be slightly overestimated , although this is an important area for further investigation . | Canine rabies has been successfully eliminated from Western Europe and North America , but in the developing world , someone dies every ten minutes from this horrific disease , which is primarily spread by domestic dogs . A quantitative understanding of rabies transmission dynamics in domestic dog populations is crucial to determining whether global elimination can be achieved . The unique pathology of rabies allowed us to trace case-to-case transmission directly , during a rabies outbreak in northern Tanzania . From these unusual data , we generated a detailed analysis of rabies transmission biology and found evidence for surprisingly low levels of transmission . We also analysed outbreak data from around the world and found that the transmission of canine rabies has been inherently low throughout its global historic range , explaining the success of control efforts in developed countries . However , we show that when birth and death rates in domestic dog populations are high , such as in our study populations in Tanzania , it is more difficult to maintain population-level immunity in between vaccination campaigns . Nonetheless , we conclude that , although the level of vaccination coverage required is higher than would be predicted from naïve transmission models , global elimination of canine rabies can be achieved through appropriately designed , sustained domestic dog vaccination campaigns . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"ecology"
] | 2009 | Transmission Dynamics and Prospects for the Elimination of Canine Rabies |
The malaria agent Plasmodium falciparum is predicted to export a “secretome” of several hundred proteins to remodel the host erythrocyte . Prediction of protein export is based on the presence of an ER-type signal sequence and a downstream Host-Targeting ( HT ) motif ( which is similar to , but distinct from , the closely related Plasmodium Export Element [PEXEL] ) . Previous attempts to determine the entire secretome , using either the HT-motif or the PEXEL , have yielded large sets of proteins , which have not been comprehensively tested . We present here an expanded secretome that is optimized for both P . falciparum signal sequences and the HT-motif . From the most conservative of these three secretome predictions , we identify 11 proteins that are preserved across human- and rodent-infecting Plasmodium species . The conservation of these proteins likely indicates that they perform important functions in the interaction with and remodeling of the host erythrocyte important for all Plasmodium parasites . Using the piggyBac transposition system , we validate their export and find a positive prediction rate of ∼70% . Even for proteins identified by all secretomes , the positive prediction rate is not likely to exceed ∼75% . Attempted deletions of the genes encoding the conserved exported proteins were not successful , but additional functional analyses revealed the first conserved secretome function . This gave new insight into mechanisms for the assembly of the parasite-induced tubovesicular network needed for import of nutrients into the infected erythrocyte . Thus , genomic screens combined with functional assays provide unexpected and fundamental insights into host remodeling by this major human pathogen .
Plasmodium falciparum is the protozoan parasite responsible for the most deadly forms of malaria . The symptoms of malaria , which include fevers and chills and can include coma and death , are caused by infection of human erythrocytes by the parasite . After invasion of the erythrocyte , the parasite is contained within a membrane-bound compartment , the parasitophorous vacuole ( PV; Fig . 1A ) . Intracellular parasites induce major changes in several properties of the erythrocyte , including its deformability [1]–[3] , permeability of its plasma membrane [4] , [5] and its adhesiveness to the endothelium [6] . Underlying these changes are proteins produced by the parasite and exported past the PV membrane ( PVM ) into the host cell . Examples include ring-infected erythrocyte surface antigen ( RESA ) , which increases the heat-resistance of the erythrocyte [7] , [8] , and P . falciparum erythrocyte membrane protein 1 ( PfEMP1 ) , a cell surface adhesin that , together with another parasite protein , Knob-Associated Histidine-Rich Protein ( KAHRP ) , forms knobs on the surface of the erythrocyte that increase the adhesiveness of the infected erythrocyte [9]–[11] . Another important change in the erythrocyte is the appearance of a large membranous network , the tubovesicular network ( TVN; Fig . 1A ) [12] , which plays a role in nutrient import into the parasite . The formation of this import organelle is entirely dependent on the parasite , and previous studies have shown that development of the TVN is linked to nutrient import into infected erythrocytes [12] , [13] . For an exported protein to reach the cytosol or membrane of the erythrocyte , it needs to cross two membranes: the parasite plasma membrane and the PVM ( Fig . 1A ) . The first step requires a canonical ER-type signal sequence , while export of many ( but not all ) proteins from the PV into the erythrocyte depends on a sequence motif referred to as the Host Targeting ( HT ) -motif [14] [alternatively known as Plasmodium Export Element ( PEXEL ) ] [15] positioned downstream and proximal to the signal sequence . The HT-motif and PEXEL are identified by different algorithms and have slightly different specificities , but recognize the same core sequence ( RxLxE/Q/D ) [14] , [15] . Identification of this export motif allowed prediction of the set of proteins exported into the erythrocyte ( the HT-based Hiller secretome and PEXEL-based Marti secretome ) , and recently an expanded version of the PEXEL-based secretome ( Sargeant secretome ) , identified with the PEXEL-based prediction program ExportPred , was published . All three secretomes are unexpectedly large , containing over 250 proteins each . Excluding the large protein families of RIFINs ( 165 proteins ) and STEVORs ( 22 proteins , of which only one is synthesized in an individual parasite [16] ) , the Hiller secretome contains 113 proteins , the Marti secretome 158 , and the Sargeant secretome 267 [17] . The overlap of the Hiller and Marti secretomes is only 59 proteins ( 53% of Hiller set , 37% of Marti set ) , underscoring the differences in the prediction algorithms . A large majority of the identified proteins cannot be annotated , and unfortunately , export of nearly all ‘hypothetical’ proteins has not been experimentally verified . In just two cases have full-length fusions of a hypothetical protein with the Green Fluorescent Protein ( GFP ) been shown to be exported [14] , [18] . In six additional cases only the N-terminal region of the proteins was tested [14] , [17] . There was no subsequent verification with full-length proteins . Hence the contribution of additional sequences to the export signal was not established . Interestingly , all P . falciparum proteins known to be exported into the host cell are species-specific . Hence little is known about the functions of exported proteins shared by all Plasmodium species . It is these proteins that are very likely to be involved in the processes that allow the parasites to survive within the erythrocyte , and would make excellent targets for prophylaxis . Thus identification of the exported proteins and their function will give better insight into the parasite-erythrocyte interaction . We have therefore investigated several parameters of protein export in P . falciparum to refine the export prediction . We show that the HT motif-based algorithm PlasmoHT is limited by the identification of ER-type signal sequences in P . falciparum and that this represents an important difference with ExportPred . We furthermore identify the putatively exported proteins conserved in other Plasmodium species and use this as a high value candidate set to validate the export prediction . Importantly , we find that testing N-terminal regions alone can lead to critical oversight on location of the protein and full-length fusions should be evaluated in order to validate positive predictions . Finally , this study shows that despite the fact that essential genes cannot be knocked out in blood stage Plasmodium berghei , insight into biological processes can be attained by utilizing rapid transgene expression . Even so , limitations in the genetic system of P . falciparum have not allowed any large scale validation of export predictions , and the testing here improves significantly on previous attempts and enables functional analyses .
Secreted P . falciparum proteins can have recessed signal sequences and thus be difficult to identify with standard signal sequence identification programs such as Signal P [19] . In prior attempts to identify exported proteins Hiller et al . modified SignalP to examine the first 100 residues instead of the default 70 residues [14] , while Sargeant et al . , abandoned SignalP but instead required the presence of a stretch of 10–25 hydrophobic residues near the N-terminus , separated by a spacer region from the start methionine [17] . However , neither prediction was based on multiple validated sequences . We therefore set out to determine how well SignalP predicts the secretion of known secreted P . falciparum proteins in order to obtain a more accurate prediction of secretion . SignalP by default examines the N-terminal 70 amino acids of a protein and determines the most likely cleavage site ( represented by the C and Y scores ) , the maximal signal sequence residues ( MaxS ) , and the average signal sequence residues ( MeanS ) as measured from the N-terminus to the most likely cleavage site [19] . We plotted the MaxS and MeanS values of 82 proteins known to be transported through the secretory pathway ( with destinations as diverse as the apicoplast , parasite plasma membrane , rhoptry and erythrocyte cytosol and membrane , see Table S1 ) ( Fig . 1B ) . Seventy-six proteins had a score above the MaxS threshold for a signal sequence ( 0 . 82 ) , and in many cases the values were close to maximal . However , six ( 7 . 3% ) had a MaxS score below the threshold , while 21 ( 25 . 6% ) had a MeanS score below the threshold ( 0 . 47 ) ; four of these proteins ( 4 . 9% ) had both a MeanS and MaxS score below threshold . Figure 1B shows that a protein ( RESA ) with a MaxS score of 0 . 571 and MeanS score of 0 . 113 can be secreted . Apparent from Fig . 1B is that as the MeanS scores decrease , the MaxS values become more scattered . Since a lower MeanS score often is indicative of a recessed signal sequence , it is possible that the increased length of the N-terminus before the hydrophobic core of the signal sequence allows for lower MaxS scores . The large percentage of proteins with a low MeanS score is likely a reflection of the prevalence of recessed signal sequences in plasmodial proteins . Thus clearly , while many secreted P . falciparum proteins adhere to the same principals of secretion as higher eukaryotes , there may be a degree of flexibility in the signal that allows proteins with lower MaxS scores to be secreted , which in turn leads to under-prediction of secreted proteins by SignalP . To evaluate whether proteins predicted to be exported to the erythrocyte contained any distinguishing features in their signal sequences , we plotted the MaxS and MeanS scores of the proteins in the previously published Hiller and Sargeant secretomes [14] , [17] . Since Hiller et al . used the default 0 . 82 MaxS cut-off to identify signal sequences , all MaxS scores predictably were above 0 . 82 , but the MeanS scores were scattered within the same range of scores as seen in the group of known secreted proteins ( Fig . 1C ) . The Sargeant secretome is based on the SignalP-independent algorithm ExportPred . As shown in Fig . 1D , like the Hiller secretome , the proteins in the Sargeant secretome fall on the same general curve as the known secreted set , i . e . , as the MeanS decreases , the MaxS values become more scattered . However , 50 out of 267 proteins ( 18% ) had signal sequence scores below the lowest experimentally proven functional MaxS score of 0 . 571 for Plasmodium; the fitness of these values for secretion remains to be validated experimentally . Thus one major reason for the different predictions in the Sargeant secretome relative to the Hiller secretome lies in the extremely low SignalP scores the former allows . To obtain a maximal secretome whose signal sequence predictions were within the range of experimentally verified signals for Plasmodium , we re-evaluated the HT-based secretome based on a MaxS cutoff of 0 . 571 . After manual curation this identified an additional 22 putatively exported proteins . We also accommodated recent updates in gene calling in the P . falciparum genome to revise secretome predictions obtained by MEME/MAST , the programs used to identify the initial Hiller secretome . Furthermore we utilized a second algorithm , HMMER , to expand the MAST output , which yielded an additional 171 proteins . Combining these proteins with the Hiller secretome ( 251 proteins ) yielded an expanded secretome ( van Ooij secretome ) of 422 proteins ( Table S2 ) . The algorithmic predictions of this secretome were further curated for proteins with MaxS scores between 0 . 58 and 0 . 82 . No further curation was undertaken based on expression , functional or structural data . Removing the RIFIN and STEVOR family members left 224 proteins ( no PfEMP1 family members are present in this secretome since they contain an internal signal sequence and are thus not recognized by Signal P ) . The van Ooij secretome approaches the Sargeant secretome ( 267 proteins ) in size , but utilizes signal sequence parameters that closely mimic experimental predictions and an HT-motif that is closely linked to the Hiller secretome . The MaxS and MeanS values of the proteins in the van Ooij secretome ( omitting the RIFINs and STEVORs ) were plotted against each other and were found to follow a similar pattern to the Hiller secretome ( Fig . 1D ) . A comparison of the secretomes and the distribution of MaxS scores is listed in Table S3 . It has been noted previously that exported proteins are frequently encoded by genes consisting of two exons [15] , [17] . This effect is exaggerated somewhat by the conservation of this exon-intron structure in the rif and stevor genes , which make up 38 . 5 and 7 . 2% , respectively , of the expanded secretome . Even after removal of the rif and stevor genes , many proteins in the van Ooij secretome are encoded by genes that have this two-exon structure , comprising 69 . 4% percent of the secretome , and 65% in the Sargeant secretome , with one-exon structure the second-most prevalent in both sets . Distribution of exon structures is shown in Table S4 . Previous investigations of exported proteins have focused primarily on the proteins unique to P . falciparum or members of large antigenic families unique in other Plasmodium species . We were interested in identifying those proteins that are conserved among all Plasmodium species because they likely perform functions necessary for the interaction of every Plasmodium species with the host erythrocyte . Hence they define the core of the interactions of the parasite with the host cell but remain completely unknown . Our initial analysis of the secretome of rodent malaria parasites P . berghei , Plasmodium chabaudi , and Plasmodium yoelii ( Rodent Malaria Parasites ( RMP ) ; [14] ) and studies by Sargeant et al . [17] on the secretomes of several Plasmodium species describe smaller secretomes than those of P . falciparum . This could reflect a less complex interaction of those species with the host erythrocyte but is at least in part due to the less complete annotation of the genomes ( see Table S5 for examples of changes in annotation of RMP and Plasmodium vivax proteins that uncover signal sequence and HT-motifs ) . Therefore , conserved genes were detected by searching the RMP genomes for orthologues of P . falciparum secretome proteins . Since synteny breakpoints often contain species-specific genes [20] , we narrowed our search to proteins encoded by genes that had maintained their genomic localization , which additionally aids in identifying those orthologues in which parts of the protein have diverged and thus have a lower score in a BLAST analysis , but are nonetheless bona fide orthologues . On the basis of these criteria , 11 proteins were identified in the original Hiller secretome ( Table 1 ) , while the van Ooij secretome contains an additional 18 . All syntenic genes were also conserved in Plasmodium vivax , indicating the widespread conservation of the genes . A similar search by Sargeant et al . identified nine putatively exported P . falciparum proteins ( identified using the PEXEL ) that were conserved in P . vivax and P . yoelii [17] . The overlap of core set predictions by Sargeant et al . and those listed in Table 1 consists of only four proteins ( indicated by an asterisk ) . Two conserved proteins identified by Sargeant et al . but not listed in Table 1 were not recognized as having an HT-motif ( underscoring the difference between the predictive algorithms PlasmoHT and ExportPred ) , while two others did not have a clearly recognizable RMP orthologue in the PlasmoDB database . The remaining protein had a MaxS score of 0 . 72 and was identified when the MaxS threshold was set at 0 . 58 . Only two of the eleven syntenic genes , pf13_0090 and pfd0495c , have the classic two-exon structure . In pf14_0607 the signal sequence is encoded by a small exon , with the HT-motif encoded close to the 5′ end of the second exon , but the entire gene contains twelve additional exons . pfl1660c is encoded by five exons , with the signal sequence and the HT-motif , as well as the majority of the protein , encoded by the first exon . The other seven genes consist of a single exon . It is possible that the classic two-exon structure may reflect a mechanism by which P . falciparum has been able to convert non-exported proteins to exported proteins through the addition of a small exon encoding a signal sequence and HT-motif . None of the syntenic genes are located near the telomeres . This is not surprising as the telomeres contain many species-specific genes . These proteins provide a high value set to test the predictive value of each of these secretomes . But since we were interested in testing the prediction of the export motif independent of the SS , we restricted our subsequent analyses to conserved proteins identified in the Hiller secretome ( Table 1 ) , which by virtue of using a MaxS cut-off of 0 . 82 dramatically reduces false positive predictions for recruitment into the secretory pathway , likely to be more prevalent in the more expansive Sargeant and van Ooij secretomes . This assumption is justified by the data in Fig . 1B showing that the vast majority of known secreted P . falciparum proteins have MaxS values higher than 0 . 82 . Sargeant et al . showed that the N-terminal residues of one of the conserved proteins of the Hiller secretome listed in Table 1 , PF14_0607 , targeted a GFP-fusion to the lumen of the PV , but were not able to promote export to the erythrocyte [17] . This raised the possibility that the leader sequence did not faithfully reflect the export properties of the complete protein . We therefore tested fusions of the full-length gene and the first 89 codons of pf14_0607 to gfp . As shown in Fig . 2A , green fluorescence associated with the full-length fusion was indeed detected in bright punctate spots in the erythrocyte . However , the fusion of the first 89 codons ( containing the signal sequence and the HT-motif , but missing the transmembrane domains ) was not ( or extremely poorly ) exported to the erythrocyte ( Fig . 2B ) . Sargeant et al . postulated that the lack of export of the fusion of the N-terminal region of PF14_0607 was due to the presence of a phenylalanine in position 4 of the HT-motif . When this residue was changed to an alanine in the 89 codon-GFP chimera , the protein was indeed robustly exported ( Fig . 2D ) into the erythrocyte cytosol in an HT-dependent manner ( compare Fig . 2C to 2D ) . These data suggest that a phenylalanine residue at position 4 can indeed hinder export when acting in a short fusion protein that contains just the signal sequence and the HT motif , but in the full-length protein it can be sustained , likely due to structural constraints on the leader imposed by the rest of the protein . Moreover the fact that the mutated 89 codon-GFP fusion is delivered to the erythrocyte cytosol ( Fig . 2D ) , while the full length gene of interest localizes to punctate intraerythrocytic structures ( Fig . 2A ) , suggests that transmembrane and other regions of the protein influence its final destination in the host cell . These data are consistent with our prior studies that have systematically established that while the HT motif is essential for export to the erythrocyte , sequences upstream and downstream provide information [14] , [21]–[23] for export resides in overall domain structure that can be conserved across evolutionary distance . The data in Fig . 2 highlight a case where sequences significantly downstream of the HT-motif in the functional protein can likely influence overall structural information needed for HT-motif dependent export . Therefore all subsequent studies were performed with full-length ( or nearly full-length ) protein fusions ( see Table 1 ) . Genetic manipulation in Plasmodium , while possible , remains a slow process . Successful generation of stable transgenic parasites and initial expansion in amounts required for standard characterization can take 4–6 weeks , while integration into the chromosome via homologous recombination can take months . Therefore most studies have been limited to plasmid-based analysis , which requires continuous drug pressure in culture . Together these limitations have severely handicapped systematic analysis of transgenes in P . falciparum . To expedite the production of the stable cell lines expressing a fusion of the syntenic genes with gfp , we utilized the piggyBac transposition system ( see Fig . 3 ) [24] , [25] . This system is based on the integration of specific DNA sequences by the lepidopteran transposase into the sequence TTAA . In the P . falciparum strain 3D7 , this sequence is present 311 , 155 times ( 124 , 733 times in coding regions ) , presumably allowing for nearly random integration [25] . To make stably transfected parasites , P . falciparum strain 3D7 was transfected with two plasmids , pHTH [25] , which encodes the transposase , and a second plasmid that contains a drug marker ( human dihydrofolate reductase in this case ) and the gfp-fusion gene , flanked by the inverted repeats recognized by the transposase . Expression of the transposase then promotes the integration of the inverted repeats into the genome . The plasmid encoding the transposase does not contain a resistance marker and is presumed to be lost during propagation of the parasites . All the transgenic parasites obtained in this study were detected within fourteen days after initiation of drug selection and could be maintained in long-term culture over several months without addition of drug while retaining the gfp transgene ( Fig . 3 ) . Integration into genes required for export , resulting in parasites that are no longer capable of protein export , is extremely unlikely , as a subset of exported proteins is likely to fulfill an essential role in the erythrocyte ( see below ) . In principle , integration into essential genes can also occur , but since transposition is likely to occur at many different sites , the selection process after transfection will allow the growth of only those parasites that have no growth defect relative to other transfectants . While bioinformatics predictions are powerful in identifying candidates for export , they need to be validated in functional assays . We were interested in understanding the HT prediction , independent of the signal sequence prediction , and validated the export of the conserved proteins by making fusions with GFP at the C-terminus . Using the piggyBac system , we obtained stably resistant parasites for 10 out of the 11 genes listed in Table 1 within 14 days after transfection; only in the case of PF13_0218-GFP were we unable to obtain stable lines ( Fig . 3 ) . Each drug-resistant transformed culture displayed a uniform population of fluorescent parasites such that the export of GFP to the erythrocyte could be ascertained without subsequent cloning of the population . We determined the integration sites for three different clones , and found that transposition had occurred in intergenic sequences as well as open reading frames ( Table S6 ) . Eight of the ten transgenic parasite lines synthesized a fusion protein of the expected size , as judged by anti-GFP immunoblot ( Fig . S1 ) . In the two other cases , PF13_0317-GFP and PFC0555c-GFP , a large majority or all of the anti-GFP signal was detected in a band approximately the size of free GFP . Hence PF13_0317 and PFC0555c could not be analysed for export and were excluded from further analysis . The entire analytical procedure is outlined Fig . 3 . Five GFP-fusions , PFA0210c , PFL0600w , PF14_0607 , PFD0495c and PFC0435w , were found to be exported to the erythrocyte ( Fig . 4A ) . PFA0210c-GFP and PFL0600w-GFP were distributed evenly throughout the erythrocyte and were also detected at high levels in the parasitophorous vacuole , which may reflect the high level of expression in the transgene system that uses the strong calmodulin promoter for better visualization of the fusion protein . PFD0495c-GFP , a transmembrane protein was detected at the periphery of the erythrocyte , in intraerythrocytic membranes and the vacuolar parasite . PF14_0607-GFP and PFC0435w-GFP , which also contain transmembrane spanning regions , displayed punctate spots in the erythrocyte as well as closely associated with the PV ( Fig . 4A ) . PFC0435w ends with the sequence DEL but we do not think this C terminal sequence impacts its localization . This is because although xDEL at the C terminus can function as a retention signal for soluble proteins in the lumen of the ER , the ER retention of transmembrane proteins does not involve the xDEL sequence . It should be noted that PFC0435w is not a soluble protein . It is a single pass transmembrane protein with its C-terminus predicted to be on the cytoplasmic face of the ER . Hence PFC0435w is not likely to be an ER retained protein . Three fusion proteins , PFL1660c , PF10_0177 and PF13_0090 , were not detected within the erythrocyte ( Fig . 4B ) . The lack of export of PFL1660c-GFP is particularly surprising since it is also predicted to be exported by ExportPred [17] . The protein is annotated to be an aspartyl protease , was detected in small structures inside the parasite ( possibly the apicoplast; Fig . 4B ) and its N-terminus was recognized as an apicoplast targeting sequence by the PATS program [26] . Residue 5 of its HT-motif is a serine , which is an unusual amino acid for this position , and may be part of the reason the protein is not exported . The perinuclear staining of PF10_0177-GFP indicated the protein was not exported . It should be pointed out that for technical reasons , this fusion contained only the N-terminal 1015 residues ( out of 3162 , encompassing the first exon only ) . Although we think it unlikely , it is formally possible that lack of downstream sequences may have compromised export mediated by its HT motif . However , the two exons of the orthologues of PF10_0177 in RMP and P . vivax are annotated as two separate genes , in which case the entire gene was fused to gfp . PF13_0090 , which is annotated as a possible ARF family member , also appeared to associate with the parasite , with no detectable export to the erythrocyte ( Fig . 4 ) . When the ability of the N-terminal region of PF13_0090 to direct secretion was tested by fusing the first 59 codons to gfp , the resulting fusion protein accumulated inside the parasite rather than in the PV ( data not shown ) , suggesting that despite its high MaxS score , the predicted SS may not support recruitment into the secretory pathway . In summary , the data in this section suggest that six of the eight tested conserved proteins contain bona fide signal sequences that enable their recruitment into the secretory pathway . Further , 5 of the 8 could be detected exported to the erythrocyte ( either diffuse or in punctuate structures ) when fused to GFP . Thus we could confirm export with a prediction rate of ∼70% ( Fig . 3B ) . When the overlap of the Hiller , van Ooij and Sargeant secretomes is considered ( Fig . 3B ) , only three of the four conserved proteins predicted to be exported by all three secretomes are exported , while one ( PFL1660c ) may be transported to an internal secretory destination such as the apicoplast . So while the rate of successful prediction increases when the intersection of two predictions is considered , the success rate is still below 100% ( and likely not to exceed to 75% ) , indicating that there may be additional determinants that are currently not recognized by in silico prediction programs . Our data also show that two of Sargeant's predictions of conserved exported proteins [17] were false negatives . One of these proteins ( PF14_0607 ) is excluded from ExportPred because the N-terminal region is not able to promote export , even though the full-length protein is exported . In the other protein ( PFC0435w ) the HT-motif is separated from the signal peptide cleavage site by 68 residues , which likely exceeds the length probability of the spacer region between the N-terminal hydrophobic region and the HT motif ( PEXEL ) allowed by ExportPred . In order to learn more about the role of the conserved proteins during the intraerythrocytic cycle , we attempted to delete the genes encoding nine of the eleven of these proteins in the P . berghei system [27] , [28] . The deletions were attempted twice for each gene , and in no case were mutant parasites obtained , while transfections performed concurrently with unrelated deletion plasmids did produce mutant parasites . The inability to delete these genes strengthens the belief that these genes encode proteins essential for the growth of the parasite within the erythrocyte , which is not unexpected considering the high degree of conservation of these genes . Results of the deletions are listed in Table 1 . Most exported proteins have no in silico annotatable features . However transgenic parasites expressing GFP fusions potentially provide powerful reagents to enable functional characterization . This is especially so when the transcriptional profile of the gene of interest mimics the largely constitutive activity of the cam promoter with peak expression from 24–40 h of infection ( Fig . S2 ) . Many secretome genes are highly stage specific and show peak expression times at segmenters and early ring stages . However the expression profile of PFC0435w peaks at 24–40 h of infection and closely parallels that of cam in the second half of the asexual life cycle ( Fig . S2 ) . We confirmed expression of the fusion of the transgene at 24 , 36 and 42 h of infection demonstrating that a single fusion product is detected in rings and becomes prominent in the trophozoite and schizont stages ( Fig . S3 ) . These data confirmed that expression of PFC0435w-GFP closely mimicked that predicted for endogenous PFC0435w . We next examined transgenic lines expressing PFC0435w-GFP at the trophozoite stage by fluorescence microscopy . We confirmed that the fusion was detected in the periphery of the parasite as well as intraerythrocytic structures . However exported PFC0435w-GFP ( Fig . 5Aiii , arrow ) did not colocalize with Maurer's clefts ( Fig . 5Aiii , arrow head ) , major intraerythrocytic structures implicated in protein export to the erythrocyte membrane . Clefts are flat lamellar membranes exported from the parasite to the erythrocyte and our recent data suggest that they are targeted by the HT motif as conduits for protein export to the cytoplasm and membrane of infected erythrocyte [29] . It is possible that colocalization between one cleft ( of ∼10 ) and PFC0435w-GFP in the trophozoite stage in Fig . 5Aiii ( see asterisk ) may reflect transport of the fusion through a cleft , en route to a distinct intraerythrocytic destination . A chimeric gene containing just the HT motif and a transmembrane domain expressed by the cam promoter , drives efficient export to the clefts ( Fig 5Avi , arrow heads ) , confirming that expression of a transgene via the cam promoter does not preclude its quantitative localization to clefts . Together these data suggested PFC0435w-GFP was not a major resident protein of clefts . We were next interested in determining the relative distribution of PFC0435w-GFP in a tubovesicular import pathway that appears to be distinct from clefts [12] . To do this we determined the distribution of PFC0435w-GFP in relation to Rhodamine B , a fluorescent dye that does not freely diffuse into erythrocytes but is actively internalized into infected erythrocytes by the TVN ( Bhattacharjee and Haldar , unpublished data ) . As shown in Fig . 5B–D , Rhodamine B fluorescence accumulated to a high level within the parasite , as well as in tubovesicular structures that extend from the erythrocyte membrane to the parasite . Remarkably , PFC0435w-GFP could be detected in discrete localized regions of the TVN . The GFP-tagged protein apparently connects the vacuolar parasite to a large membrane loop of the TVN . Analysis of different optical sections from an infected cell confirmed that the protein appeared to form a bridge between the vacuolar parasite and the intraerythrocytic loops; curiously this bridge itself was relatively poorly stained with Rhodamine B ( Fig . 5B , D ) , suggesting it did not retain significant amounts of the internalized probe . PFC0435w-GFP was also detected in junctions between TVN structures closer to distal regions of the TVN near the erythrocyte membrane ( Fig . 5B , D and schematized in 5C ) . These data provide the first direct evidence of a parasite protein quantitatively located at the junction of the TVN and the PVM , and we therefore renamed the protein TVN-junction protein 1 ( TVN-JP1 ) . TVN-JP1-GFP was also seen at points where the PVM formed small buds ( see arrows in Fig . 5D ) , suggesting it may define junction sites of loop formation at the PVM . The TVN develops during ring to trophozoite development ( which occurs approximately 24 hrs after invasion of the erythrocyte , midway in the 48 hr developmental cycle ) . To further investigate the function of TVN-JP1 in TVN assembly , we examined its distribution in rings . At this stage TVN-JP1 was present in small vesicles that moved rapidly ( in seconds ) in the host cell cytosol ( see Video S1 , Fig . 6 ) . This was in sharp contrast with immobile GFP junctions connecting large membrane loops in association with the mature TVN in trophozoites . The movement of the vesicles appeared random , moving away from or towards the parasite equally , and no build-up of the protein at the periphery of the infected cell was detected , indicating that the vesicles did not fuse with compartments at the erythrocyte periphery . However movement of one vesicle ( above left-hand cell in Video S1 ) was highly restricted , suggesting that it may be attached in the erythrocyte . Remarkably , we also detect a membrane connection between this vesicular structure and the parasite , as indicated by the presence of GFP fluorescence spanning from this vesicle to the parasite . This type of vesicle movement has been detected previously in the form of acridine orange-labeled vesicles , which were also detected primarily during the ring stage [30] but their function was unknown . Our data suggest that they are precursors to TVN assembly . The acridine orange-labeled vesicles were detected in wild-type P . falciparum , making it unlikely that the appearance of PFC0435w-containing vesicles is an artifact resulting from overexpression of the protein . Together , these data suggest that export of highly mobile vesicles containing TVN-JP1 is an early step in the formation of the TVN . Since TVN-JP1 domains in the erythrocyte are immobile in trophozoites , anchoring of these vesicles and membrane connections between them and the vacuolar parasite are likely to precede sphingolipid-dependent budding of large vesicles and loops originally described as the first step of TVN biogenesis [31] . Thus although TVN-JP1 does not stain large domains of the TVN , its export is nonetheless expected to be important in erythrocyte remodeling and for proper development of the TVN . Our studies establish that rapid genetic methodologies enable identification of genes and mechanisms linked to formation and function of the P . falciparum TVN , suggesting they provide new targets for prophylaxis .
In this study we have provided empirical testing of the prediction of protein export from the malaria parasite P . falciparum and validation of the export prediction for a set of 11 proteins conserved throughout the genus Plasmodium . This is the first demonstration of export of proteins conserved in several species of Plasmodium . Relative to the size of the entire secretome this is a relatively small set of proteins to test , but their conservation across species confirms they are high-value determinants . Due to the limitations of genetic manipulation in P . falciparum , this kind of genome-wide analysis required utilization of a robust transgene expression system such as piggyBac , which shortened the time required for establishing stably transfected parasites and had a very high success rate ( 10 out of 11 genes could be investigated ) . Since we selected from a set that had the highest MaxS scores , we maximized chances that these proteins are recruited to the secretory pathway ( although the putative ARF is likely not ) . This increases the ability to evaluate the predictive value of the HT motif in mediating translocation beyond the PV to the erythrocyte Extrapolating the positive prediction rate of ∼70% of the tested proteins to the entire secretome predicts that well over 200 proteins will be exported , confirming original projections that host remodeling is a highly complex process . Nonetheless , due to the difficulty in predicting signal sequences in Plasmodium , we expect that in the more expansive secretomes ( such as the van Ooij and Sargeant secretomes ) , the positive prediction rate will be lower . As pointed out previously , two of the proteins shown to be exported were not recognized by the ExportPred algorithm , likely due to a relatively large number of residues separating the HT motif and the signal sequence ( PFC0435w ) and additional information in the remainder of the proteins ( PF14_0607 ) , indicating that parts of the algorithm limit the prediction of the secretome . It is difficult to explain fully why a fusion containing only the 90 N-terminal residues of PF14_0607 remained in the PV while the full-length proteins is exported , other than that the remainder of the protein must aid in exporting the protein across the PV . It is interesting to note that the N-terminal sequence of PF14_0607 , and that of 66 other secretome proteins , was recognized by the apicoplast-targeting prediction program PATS [26] ( Table S2 ) . This underscores the difficulty of predicting transport when multiple transport signals are present . How overlapping targeting sequences are resolved remains unclear . Only 11 proteins of the original Hiller secretome were conserved in the RMP and P . vivax . This constitutes 9 . 7% of the total secretome ( without the RIFINs and STEVORs ) , a surprisingly low number . In the Sargeant secretome the percentage of conserved proteins is even lower , 3 . 3% , while in the van Ooij secretome it is 12 . 9% . Since we found that the annotation of the orthologous genes of exported proteins often did not include the 5′ region ( which encodes the signal sequence; Table S5 ) , it is not yet possible to determine a complete secretome for the RMP or P . vivax to provide a direct comparison of the secretomes and identify all RMP or P . vivax-specific exported proteins . Even so , a large number of proteins in the P . falciparum secretome are species-specific , making it likely that they are important for P . falciparum-specific symptoms but possibly not survival of the parasite within the erythrocyte . Indeed , as shown in Table 2 , none of the published deletions of genes encoding exported proteins , all P . falciparum-specific , are lethal to the parasite . Thus P . falciparum contains species-specific mechanisms for sequestration involving PfEMP1 and KAHRP [10] , stabilization of the cytoskeleton through RESA [7] , [8] and MESA [32] , among other proteins , as well as formation of intraerythrocytic structures the Maurer's Clefts , by SBP1 [33] and MAHRP [34] , but these processes are not required for the growth of the parasites in culture . Since all the exported proteins studied to date are P . falciparum-specific , little is known about the functions shared by all Plasmodium species . It is likely that the conservation of the exported proteins , along with the likely function in host cell remodeling , will make these proteins essential factors for parasite survival with the host cell . Consistent with this is the finding that none of the genes encoding the conserved exported proteins could be deleted in P . berghei , indicating that they are indeed required for parasite growth ( Tables 1 , 2 ) . The results presented here suggest that at least one conserved exported protein is involved in formation of the TVN , an organelle of nutrient uptake . At the time of its original identification [12] , [31] , only a parasite sphingomyelin synthase activity was known to be important for TVN synthesis , but no report has described the localization of this sphingomyelin synthase activity . Members of the P . falciparum protein family ETRAMPS as well as the P . berghei protein UIS4 have been detected on vesicular elements budding off the PVM of blood stage and liver stage parasites , respectively [35]–[37] , but since they are detected primarily on the vacuole , they are not thought to be TVN resident proteins or have TVN-specific functions . This study reveals TVN-JP1 as the first TVN-specific protein marker . The distribution of the protein , in a bridge structure between large intraerythrocytic loops and the PVM , could be indicative of a structural role , which would be congruent with the finding that the protein is initially found in rapidly moving vesicles in the erythrocyte cytosol . The evidence that the TVN begins as small , mobile structures in the erythrocyte cytosol is highly unexpected since the mature TVN organelle is a relatively immobile structure in the trophozoite-infected erythrocyte as are the TVN-JP1 junction and large loops between these junctions ( Fig . 6 , 7 ) . The localization of TVN-JP1 at sites of PVM budding in trophozoites may reflect sites of budding that contribute to TVN development even at these later stages of growth ( summarized in Fig . 7 ) . Considering the size of the TVN within the erythrocyte cytosol and the functions in protein import it plays , there are undoubtedly more proteins involved in the formation of this organelle . The total number of conserved proteins that could be involved in TVN function and/or formation based on Hiller , Sargeant and van Ooij secretomes are expected to range from ∼10–29 . The total number of P . falciparum-specific proteins associated with TVN function is presently difficult to predict . The approach of a genome-wide screen for exported proteins combined with application of several criteria consecutively ( conservation in other Plasmodium species , validation of export , timing of expression and localization data ) has identified possible functions for a protein that was beyond annotation by in silico methods . By altering the criteria for selection , it should be possible to uncover the role of other pathogenic exported proteins of hypothetical function and their contribution to intracellular survival and pathogenesis .
Plasmodium falciparum parasites used in this study were of the 3D7 lineage and maintained in human erythrocytes , blood type A+ , in RPMI-1640 medium supplemented with 91 . 9 µM hypoxanthine , 11 mM glucose , 0 . 18% sodium bicarbonate and 10% human serum ( cRPMI ) . Plasmodium berghei , strain ANKA , parasites were maintained in Swiss mice . P . falciparum transfections were performed as described by Wu et al . [38] with modifications described by Deitsch et al . [39] . Briefly , erythrocytes were loaded with 100 µg of plasmid DNA containing the transgene , and where necessary , 100 µg of plasmid containing the piggyBac transposase ( pHTH ) , by electroporation using a BioRad GenePulser set at 0 . 310 kV and 950 µF . Transfected erythrocytes were washed three times with cRPMI and immediately mixed with Percoll-purified infected erythrocytes to a parasitemia of 1–5% . Drug selection was initiated by addition of WR99120 ( Jacobus Pharmaceuticals , Princeton , NJ ) to 2 . 5 nM 72–96 hours after infection . The medium was changed every other day until parasites were detected by Giemsa staining . Most transfected parasites could be detected within two weeks after drug selection . The P . falciparum strain expressing the HTTM-GFP fusion is described in Bhattacharjee et al . [29] . P . berghei transfection was performed according to published methods [27] , [28] . Briefly , 5–8 ml blood was extracted from an infected Wistar rat at a parasitemia of <3% . Blood was mixed with an approximately equal volume of RPMI with L-glutamine and HEPES , supplemented with 0 . 085% NaHCO3 and 25% fetal calf serum ( culture medium ) with 0 . 3 ml heparin stock solution ( 200 I . U . /ml ) and spun down . Cells were resuspended in 150 ml culture medium and allowed to mature to the schizont phase overnight by incubation at 37°C in an atmosphere of 5% O2 , 5% CO2 , 90% N2 . Mature parasites were harvested by density centrifugation on Nycodenz and subsequently transfected with 5–10 µg linearized plasmid DNA using the AMAXA device . To search for the presence of an ER-type signal sequence , proteins were analyzed using the SignalP-NN algorithm from the SignalP V2 . 0 . b2 program ( http://www . cbs . dtu . dk/services/SignalP-2 . 0/ ) [19] . This algorithm was trained on eukaryotic sequences and the input protein sequences were truncated at amino acid 100 . MeanS scores and MaxS scores for each sequence were plotted against each other . In our previously described secretome , all sequences with a maximum S-score above the default cutoff of 0 . 82 determined by the SignalP-NN software , were considered to have a signal peptide . For the revised secretome set defined in this paper , a MaxS-score of 0 . 58 was used as the discriminating factor for secretion signals . The HMMER software suite ( http://www . psc . edu/general/software/packages/hmmer/manual/main . html ) was used for making a hidden Markov model ( HMM ) . The initial model was built using a training set that consisted of five proteins known to be exported to the erythrocyte ( GBP130 , PfEMP2 , PfEMP3 , HRP1 , HRP2 ) , and this model was used to search a database of 3D7 proteins with a signal sequence predicted using SignalP 2 . 0 default maximum S-score as a discriminating factor [14] . The proteins identified from this search were used as a new training set to create the final HMM . This model was used to screen all proteins predicted to have a signal sequence according to the SignalP maximum S-score cutoff of 0 . 58 ( see Signal Sequence Prediction section below ) for sequences that contain the HMM . The chromosomal position of all 3D7 sequences in the original PlasmoHT set [14] were compared with the chromosomal position of their orthologues in the rodent malaria P . berghei and P . yoelii described in Kooij et al . [25] . e-values were used as a guide to predict functional class and synteny was used to select the functional orthologues . The syntenic genes were amplified with the primers listed in Table S7 ( which also list the amount of the gene amplified ) using P . falciparum 3D7 genomic DNA as template . The resulting DNA fragments containing pf13_0317 , pf13_0090 , pf13_0218 , pfa0210c , pfc0555c , pfl0600w , and pfd0495c were digested with the indicated restriction enzymes and cloned into pBLD70 to make a fusion with gfp . pBLD70 was made by cloning GFP mut2 followed by an XhoI site into pBluescript using the NheI and SacI sites . This fusion was liberated with XhoI and cloned into XhoI-digested pBLD194 . pfc0435w , pf10_0177 , pf14_0607 , and pfl1660c were amplified further using AttB1 and AttB2 primers to attach full-length Att-sites on the DNA fragments , and cloned into pDONR and subsequently transferred to the GFP-fusion expression vector pBLD194 using the GATEWAY technology ( Invitrogen , Inc , Carlsbad , CA ) . To make a truncated version of pf14_0607 , the 5′ 89 codons were amplified using the primers listed is Table S7 using plasmid pBLD201 , the pf14_0607 full-length expression vector described above , as template . The resulting DNA fragment was cloned into pDONR and subsequently moved to the GFP-expression vector pBLD194 . To make the F-A mutation or the replacement of the HT-motif , the 89 codon 5′ region was amplified in two parts using the primers listed in Table S7 ( which include the intended mutation ) . The resulting fragments overlapped by 47 bp ( including the introduced mutation ) , and were mixed with the 5′ upstream and 3′ downstream primers in a PCR to produce the entire 89-codon fragment that contains the intended mutation . This fragment was again amplified with the AttB primers to introduce full AttB sites and then cloned into pDONR and subsequently pBLD194 using the GATEWAY technology . Vectors for the targeted deletion of the syntenic genes in P . berghei were made as follows . A region of DNA upstream and downstream of the target gene was amplified , with each fragment in the range of 650–1100 bp ( with exception of the region upstream of PB000767 . 00 . 0 , which was approximately 3 kb ) , using 100 ng P . berghei genomic DNA as template . Primers used for amplification are listed in Table S7 . The amplified DNA fragments were inserted into the pDONR vector using the GATEWAY technology . The upstream fragments were released from the resulting vectors using SacII and SpeI and cloned into corresponding sites in pL0001 . Subsequently the corresponding downstream regions were cloned into the resulting vector with HindIII and KpnI to form the final deletion vector with the upstream and downstream regions flanking the hDHFR resistance cassette . For transfection the plasmids were linearized by digestion with KpnI . To obtain stable parasite lines with integrated copies of the transgene , parasites were transfected as described above with 100 µg of vector containing the transgene and 100 µg of vector containing the transposase , pHTH [24] , [25] . To identify the sites of integration , genomic DNA was extracted from ∼2–9×108 cloned parasites . Parasites were extracted from infected erythrocytes by lysis with 0 . 05% saponin in PBS . Parasites were pelleted and lysis was repeated . Parasites were washed with PBS twice and frozen at −80°C . Genomic DNA was isolated by standard procedure . Briefly , parasites were resuspended in lysis buffer ( 10 mM Tris , 20 mM EDTA , 0 . 5% SDS , 25 µg/ml Proteinase K ) and incubated at 37°C for 3 hours . The lysate was extracted once with phenol and once with chloroform and treated with RNase A ( 20 µg/ml ) for 15 minutes at 37°C . The DNA was subsequently extracted twice with phenol-chloroform and precipitated with ethanol . One µg of DNA was digested with Sau3AI overnight , ethanol precipitated , and self-ligated in a 100 µl-reaction overnight . The ligated DNA was precipitated with ethanol and resuspended in 20 µl . Of this , 1 µl was used for PCR using previously described primers [25] . The resulting DNA fragments were cloned into pGEM-T and sequenced using M13 forward and M13 reverse primers . Parasites were cloned by limiting dilution . Parasitemia of cultures was determined by Giemsa staining and parasites diluted to 3 parasites/ml in cRPMI . Of this dilution , 100 µl was added to one well of a 96-well plate , and 10 µl of 50% human erythrocytes and 90 µl of cRPMI were added . Medium was changed and erythrocytes added every four days . Parasites were detected approximately 18 days after dilution . Parasites were prepared for microscopy as described [21] . Briefly , ∼3×108 erythrocytes infected at a parasitemia of 1–10% were spun down and resuspended in 1 ml PBS with 11 mM glucose containing 10 µg/ml Hoechst 33342 and left at room temperature for five minutes . The cells were washed twice with 3 ml PBS-glucose and resuspended to a parasitemia of ∼30% . A 3 µl aliquot was placed on a slide and covered with a coverslip , which was sealed with nail polish . The sample was viewed on an Olympus IX inverted fluorescence microscope and images were collected with a Photometrix cooled CCD camera ( CH350/LCCD ) driven by DeltaVision software from Applied Precision Inc . ( Seattle , WA ) . For immunofluorescence staining , we followed the protocol of Apodaca et al . [40] . Briefly , cells were deposited on poly-L-lysine coated coverslips for thirty minutes at 37°C in PBS containing 11 mM glucose . Cells were washed twice with PBS and fixed with 1% formaldehyde in PBS for ten minutes , then quenched with 50 mM ammonium chloride in PBS for ten minutes . Then the cells were washed , permeabilized with 0 . 05% saponin in PBS and blocked with 0 . 7% fish skin gelatin , 0 . 01% saponin in PBS ( FSP ) for thirty minutes at 37°C . Primary antibody against SBP1 ( LWL ) and GFP diluted in FSP were added and the cells were incubated for one hour at 37°C . Cells were washed three times with FSP before addition of secondary antibody ( goat anti-mouse TRITC and goat anti-rabbit FITC ) and incubation at 37°C for one hour . Cells were then washed three times with FSP , twice with PBS , once with PBS containing 0 . 01% saponin , once with PBS , once with PBS containing 0 . 1% Triton X-100 and once with PBS . Cells were then fixed again with 4% formaldehyde in 0 . 1 M sodium cacodylate for thirty minutes . The cells were washed once with PBS , then stained with 10 µg/ml Hoechst 33342 in PBS for five minutes . The cells were washed two more times with PBS before being mounted on glass slides on a drop of DABCO and sealed with nail polish . Slides were stored at −20°C . Parasites were harvested from erythrocytes with 0 . 05% saponin as described above and frozen at −80°C . In some cases the parasites were resuspended directly in SDS-PAGE loading dye , boiled , and an equivalent of 1–5×107 parasites was loaded on an SDS-PAGE gel after boiling . In other cases the parasites were resuspended in 100 µl RIPA buffer , incubated on ice in the presence of protease inhibitors . An equivalent volume of SDS-PAGE loading dye was added , the samples were boiled and the samples run as above . Proteins were transferred to nitrocellulose . The resulting blot was blocked with 5% milk in PBS with 0 . 05% Tween-20 ( PBS-T ) for one hour at room temperature or at 4°C overnight , and incubated with anti-GFP antibody ( Santa Cruz , Molecular Probes ) at room temperature for 1 hour or 4°C overnight , followed by incubation at room temperature in the presence of HRP-linked secondary antibody . Protein was visualized using the ECL system ( SuperSignal West Pico chemiluminescence , Pierce ) . For labeling with Rhodamine B , ∼2×107 cells were pelleted and resuspended in PBS containing 11 mM glucose and 1 mM Rhodamine B and 10 µg/ml Hoechst 33342 . Subsequently cells were incubated at 37°C . Cells were washed five times with PBS-glucose and viewed . | The parasite Plasmodium falciparum causes malaria by replicating inside red blood cells of infected individuals . By exporting many different proteins into the host cell , the parasite changes many of its properties . Knowledge of the identity and function of all the exported proteins will both increase our understanding of the modifications required for parasite survival and provide us with targets that can be inhibited to block the growth of the parasites . Several years ago , a motif within the exported proteins was discovered that allowed them to be exported , which was used to predict the total set of proteins exported to the host cell ( the secretome ) . We show here that the earlier studies have either under- or overestimated the total number of proteins exported into the host cell , and derive a more accurate prediction of proteins exported to the host cell . We validate the predictions by making parasites that express a fusion of predicted exported proteins to the Green Fluorescent Protein ( which allows the localization of the protein to be determined visually ) . This revealed a positive prediction rate of ∼70% . In addition , several proteins were identified that are very likely to play an essential role in infection , with at least one involved in the formation of a structure required for nutrient import . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/genomics",
"cell",
"biology/membranes",
"and",
"sorting",
"microbiology/parasitology",
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/tropical",
"and",
"travel-associated",
"diseases"
] | 2008 | The Malaria Secretome: From Algorithms to Essential Function in Blood Stage Infection |
In Germany , rabies in bats is a notifiable zoonotic disease , which is caused by European bat lyssaviruses type 1 and 2 ( EBLV-1 and 2 ) , and the recently discovered new lyssavirus species Bokeloh bat lyssavirus ( BBLV ) . As the understanding of bat rabies in insectivorous bat species is limited , in addition to routine bat rabies diagnosis , an enhanced passive surveillance study , i . e . the retrospective investigation of dead bats that had not been tested for rabies , was initiated in 1998 to study the distribution , abundance and epidemiology of lyssavirus infections in bats from Germany . A total number of 5478 individuals representing 21 bat species within two families were included in this study . The Noctule bat ( Nyctalus noctula ) and the Common pipistrelle ( Pipistrellus pipistrellus ) represented the most specimens submitted . Of all investigated bats , 1 . 17% tested positive for lyssaviruses using the fluorescent antibody test ( FAT ) . The vast majority of positive cases was identified as EBLV-1 , predominately associated with the Serotine bat ( Eptesicus serotinus ) . However , rabies cases in other species , i . e . Nathusius' pipistrelle bat ( Pipistrellus nathusii ) , P . pipistrellus and Brown long-eared bat ( Plecotus auritus ) were also characterized as EBLV-1 . In contrast , EBLV-2 was isolated from three Daubenton's bats ( Myotis daubentonii ) . These three cases contribute significantly to the understanding of EBLV-2 infections in Germany as only one case had been reported prior to this study . This enhanced passive surveillance indicated that besides known reservoir species , further bat species are affected by lyssavirus infections . Given the increasing diversity of lyssaviruses and bats as reservoir host species worldwide , lyssavirus positive specimens , i . e . both bat and virus need to be confirmed by molecular techniques .
Lyssaviruses are non-segmented negative-strand RNA viruses of the order Mononegavirales , family Rhabdoviridae and causative agents of rabies in bats and other mammals as well as in humans [1] . While rabies in dogs and other carnivores has been known since antiquity , the first evidence of rabies in haematophagous and insectivorous bats was reported from the Americas in the first half of the 20th century [2] . Since 1954 , bat rabies cases have also been reported from other continents . Antigenic and genetic analyses revealed the diversity of different lyssavirus species , and to date , besides classical rabies virus ( RABV ) , thirteen additional lyssaviruses have been discovered , mostly in bats [3] . Beyond Europe , Lagos bat virus ( LBV ) , Mokola virus ( MOKV ) , Duvenhage virus ( DUVV ) , Shimoni bat virus ( SHBV ) , and Ikoma lyssavirus ( IKOV ) were found in Africa . In Asian bat species , Aravan virus ( ARAV ) , Khujand virus ( KHUV ) , and Irkut virus ( IRKV ) were isolated . With the exception of MOKV and IKOV , all of those viruses were detected in bats [3] . In Australia , which has a long history of freedom from classical rabies , Australian bat lyssavirus ( ABLV ) is found in insectivorous and pteropid bats [4] . In Europe , bat rabies is also caused by several lyssavirus species . Between 1977 and 2012 , a total of 1039 bat rabies cases were reported from European countries ( http://www . who-rabies-bulletin . org ) . The majority was characterized as European bat lyssavirus type 1 ( EBLV-1 ) isolated from Eptesicus bat species ( E . serotinus , E . isabellinus ) [5] . Genetically , EBLV-1 can be divided in two subtypes , EBLV-1a and 1b [6] , [7] . While the EBLV-1a subtype is predominantly found in Central and Eastern Europe ( France , The Netherlands , Denmark , Germany and Poland ) , EBLV-1b has been reported from Spain , France , Southern Germany , and central Poland [8]–[11] . European bat lyssavirus type 2 ( EBLV-2 ) has been isolated from Daubenton's bats in the UK , Switzerland , Finland and Germany , and from Pond bats ( M . dasycneme ) in The Netherlands [12] . As of today , three Natterer's bats ( Myotis nattereri ) infected with the novel Bokeloh bat lyssavirus ( BBLV ) have been found in Germany and France [13]–[15] . A single detection of the West Caucasian bat virus ( WCBV ) in a Schreiber's bent-winged bat ( Miniopterus schreibersii ) has been reported from Western Caucasus Mountains [16] . Interestingly , specific RNA from a putative new lyssavirus named Lleida bat lyssavirus ( LLEBV ) was detected in brain material from the same bat species collected in Spain [17] . The public health relevance of bat rabies in general is highlighted by the fact that most of the bat associated lyssaviruses have caused human rabies [18] . In Europe , both EBLV-1 and EBLV-2 were responsible for four confirmed human casualties [19] . Also , sporadic spill-over infections of EBLV-1 to terrestrial mammals have been reported , i . e . in sheep in Denmark , two cats in France and a stone marten ( Martes foina ) in Germany [20]–[22] . Because of the zoonotic character of bat lyssaviruses knowledge about distribution , abundance and epidemiology is important to estimate and subsequently reduce the public health risk posed by bat rabies . Guidelines for the surveillance of bat lyssaviruses in Europe were established by the European research consortium Med-Vet-Net [23] and supported by EUROBATS [24] . The investigation of sick or dead bats for lyssavirus antigen in brain samples ( passive surveillance ) and testing of oro-pharyngeal swab samples and serum samples from free-living indigenous bats ( active surveillance ) for the presence of viral RNA or virus neutralizing antibodies , respectively , were recommended . However , the levels of active and passive bat rabies surveillance in Europe are still very heterogeneous despite previous recommendations [5] . Based on published data , active surveillance provides only limited information and cannot replace passive bat rabies surveillance [25] . Comprehensive passive bat rabies surveillance was conducted in The Netherlands [26] , the United Kingdom [27] , France [28] and Germany [10] . With the exception of Germany , passive surveillance in these countries is realized by only one or two cooperating departments investigating all bats submitted from the whole country . In contrast , rabies diagnosis in Germany is the responsibility of the sixteen federal states [10] . Dead or diseased bats with symptoms suggestive of rabies , particularly after contact with humans ( bites and scratches ) have to be submitted and tested for lyssavirus infection in the regional veterinary laboratories . While cases of this notifiable disease in carnivores and bats were reported to the National Reference Laboratory for Rabies at the WHO Collaborating Centre for Rabies Surveillance and Research ( FLI Riems , Germany ) , the number of bats tested negative was only sporadically submitted . Furthermore , the identification of bats to species level is generally missing as in some other European countries [5] . Therefore , routine bat rabies surveillance in Germany has relied on limited and opportunistic sampling which may not be representative of the true epidemiological situation [10] . To overcome these limitations and to obtain further information on the epidemiology of bat rabies in Germany an enhanced passive retrospective surveillance study was started at FLI in 1998 . In this study , the focus was on dead bats excluded from routine diagnostic testing . This included bats obtained from ( private ) collections from different parts of Germany . Each sample was identified to species level , partly by molecular tools and tested for lyssavirus infection . Here , we present the data from this study and compare it with published data from routine diagnostic screening .
Dead bats were submitted under the prevailing laws of the respective federal states and following EUROBATS guidelines [24] . Because this study was in the frame of a surveillance programme conducted by the national reference laboratory for rabies no further permits were necessary . Starting in 1998 , on a federal state level bat conservationists , as well as various institutions and authorities ( e . g . Nature and Biodiversity Conservation Union ( NABU , Germany ) , Museum of Natural Science , wildlife care centers ) were requested and encouraged to submit dead bats for rabies diagnosis irrespective of the circumstances of acquisition . Archived or newly acquired bats were submitted from all 16 German federal states by local bat biologists during long-time monitoring or routine inspection of maternity roosts , wintering grounds or were killed by cats , wind turbines or unintentional removal of roosts . All bats were stored frozen prior to submission in a chilled state . Usually , bat carcasses were submitted with additional information , e . g . geographical origin , date found , sex , age and species identification . Bats without species information were determined to genus or even to species level using external morphological features [29] . Bat carcasses which were degraded or damaged and those suspected to represent cryptic bat species ( e . g . Myotis mystacinus , M . brandtii and M . alcathoe ) were identified by a mitochondrial cytochrome b ( cyt b ) gene specific PCR [30] if not restricted by museum specific preservation requirements . For this purpose , wing membrane samples were collected and stored separately in Eppendorf tubes at −80°C until analysis . For DNA preparation a small piece ( 1 . 0×1 . 0 mm ) of each sample was lysed overnight ( 56°C , 400 rpm ) using 3 µl proteinase K ( 10 mg/ml ) and 300 µl lysis buffer ( 50 mM KCL , 10 mM TRIS-HCL ( pH 9 . 0 ) , 0 . 45% nonidet NP 40 and 0 . 45% Tween 20 ) . After centrifugation ( 1 min , 13000 rpm ) the supernatant was stored at −20°C . For PCR amplification two primer pairs ( CytB Uni fw: 5′-CATCMTGATGAAAYTTYGG-3′ and CytB Uni rev: 5′-ACTGGYTGDCCBCCRATTCA-3′ [30]; HG for: 5′-CACTACACATCAGAYAC-3′ and HG rev: 5′-AAGGCGAAGAATCGRGT-3′ ) were used to obtain fragments of about 950 bp and 400 bp , respectively . The latter primer mix was developed based on reference material submitted to the AHVLA to identify all bat species indigenous to the UK ( data not shown ) . The PCR reaction mix ( total volume of 25 µl ) consisted of RNase-free water ( 17 . 65 µl ) , 25 mM of each dNTP ( 0 . 5 µl ) , 50 mM MgCl2 ( 0 . 75 µl ) , 10 pmol/µl of each primer ( 0 . 5 µl ) , 10x PCR RxN Buffer ( 2 . 5 µl ) , ( 0 . 1 µl ) Platinum-Taq DNA Polymerase ( Invitrogen , Darmstadt , Germany ) and 2 . 5 µl template DNA . The amplification was performed with the following temperature profile: 3 min at 94°C ( initial denaturation ) , followed by 50 cycles of 30 s at 94°C ( denaturation ) , 30 s at 47°C ( annealing ) , 1 min at 72°C ( elongation ) and a final extension at 72°C for 10 min . Amplification of the expected products was confirmed in a 1% agarose gel stained with ethidium bromide or SYBR safe DNA gel stain . PCR products were then purified ( NucleoSpin Gel and PCR Clean-up kit , Macherey-Nagel , Düren , Germany ) and sequenced using BigDye Terminator v1 . 1 Cycle Sequencing Kit ( Applied Biosystems/Life Technologies , Carlsbad , CA , USA ) . The cytochrome b sequences were compared with published sequences of European bat species ( GenBank ) using the Basic Local Alignment Search Tool ( BLAST , http://blast . ncbi . nlm . nih . gov/Blast . cgi ) and the species determination was finalized by identification of the species of the highest nucleotide sequence similarity ( ≥90% ) . Rabies diagnosis was performed on bat brain samples which were removed either by opening of cranium or , in case of natural scientific collections , by puncture of foramen occipitale magnum using a 26-gauge needle . Lyssavirus antigen was detected by standard fluorescent antibody test ( FAT ) using commercially available polyclonal fluorescein isothiocyanate ( FITC ) -labelled anti-rabies conjugates ( Behring , Marburg; SIFIN , Berlin , Germany ) following standard protocols [31] . Additional tests included virus isolation in cell culture , reverse-transcription quantitative real-time polymerase chain reaction ( RT-qPCR ) and sequencing following RT-PCR was performed to confirm positive FAT results . For virus isolation , FAT positive or inconclusive bat brain samples were homogenized in a volume of 1000 µl sterile minimum essential medium ( MEM-10 , with 2% Streptomycin ) . The resulting brain suspensions ( 500 µl ) were subjected to the RTCIT [32] , using a mouse neuroblastoma cell line ( MNA 42/13 , No . 411 , cell culture collection for veterinary medicine , FLI ) . Infected cells were incubated for three days at 37°C and 5% CO2 and then tested using FAT . A result was confirmed negative after the third consecutive cell passage . RNA was extracted from 200 µl brain suspension or RTCIT supernatant using TRIzol Reagent ( Invitrogen , Darmstadt , Germany ) /peqGOLD TriFast ( peqlab Biotechnologie GmbH , Erlangen , Germany ) method . The RNA pellet was re-suspended in a volume of 20 µl bidistilled water . Samples were analysed for the presence of viral RNA using quantitative real-time PCR ( RT-qPCR ) specific for EBLV-1/-2 as described [25] . In cases of inconclusive FAT results a conventional panlyssavirus RT-PCR was additionally performed [33] . All EBLV-isolates were further characterized by sequence analysis [34] . RNA was subjected to one-step RT-PCR using primers JW12 and JW6 E [33] followed by sequencing . Briefly , after amplification , PCR-products were run in a 1% agarose gel stained with ethidium bromide , excised and purified essentially as for the molecular bat species identification . Sequences were manually checked for quality , trimmed to the first 400 bp using SeqMan ( Lasergene , DNASTAR , Madison , WI , USA ) ) and submitted to NCBI GenBank ( Table S1 ) . Sequence alignment and subsequent phylogenetic analysis was performed using MEGA 5 . Further representatives of EBLV-1 and 2 were derived from GenBank for comparison ( Table S2 ) .
From 1998 to June 2013 a total of 5478 bats from all German federal states ( N = 16 , Figure 1b ) were investigated . The annual number of submissions to FLI of obtained specimens varied between 30 and 1200 individuals . The bats encompassed specimens from the entire study period and before , with the oldest sample originating from 1981 . Among all samples , 21 out of the 23 indigenous bat species in Germany were included ( Table 1 ) , although the proportion of bat species differed per federal state . The majority of bat samples originated from Lower Saxony ( N = 1252 ) , followed by Baden-Wuerttemberg ( N = 736 ) and Saxony-Anhalt ( N = 692 ) . In contrast , in three and two of the remaining federal states the sample size was less than 90 and 15 , respectively . With the exception of a single carcass of the Lesser horseshoe bat ( Rhinolophus hipposideros ) , all other species investigated belonged to the family Vespertilionidae . Among those , the most frequently tested bat species were the Common pipistrelle and Noctule bat followed by Serotine bat and Brown long-eared bat ( Table 1 ) . A total of 330 bats could not be identified to species level using external morphological criteria . Cyt b sequences were obtained from 119 bats , representing 15 different species ( Table 1 ) . The sequence similarity ranged between 92% and 100% when compared to publicly available sequences . Wing membrane samples from the remaining 211 individuals from natural scientific collections were not available . Most positive specimens were found in bats from Lower Saxony ( N = 27 ) , Saxony-Anhalt ( N = 10 ) and Berlin ( N = 5 ) ( Figure 1d ) . Bat rabies was detected in animals from additional 10 German federal states although only sporadically ( 1–3 cases ) . No lyssavirus infection was found in bats originating from Rhineland-Palatinate ( N = 108 ) , Baden-Wuerttemberg ( N = 736 ) and Bavaria ( N = 252 ) ( Figure 1b–d ) . Except for a single Serotine bat for which sufficient brain material was not available , lyssaviruses were successfully isolated and sequenced from 54 and 55 bats , respectively , which had been tested FAT-positive ( Table 1 ) . The presence of EBLVs was confirmed in five different bat species ( E . serotinus , P . pipistrellus , P . nathusii , Pl . auritus and M . daubentonii ) . The majority of viruses were identified as EBLV-1 , predominately isolated from E . serotinus ( N = 48 ) . Single lyssavirus infections in other species were also characterized as EBLV-1 ( Table 1 ) . The phylogenetic analysis of the N gene derived sequences identified the two lineages of EBLV-1 , i . e . five out of the 52 available sequences were characterized as EBLV-1b found in Serotine bats originating from Saarland ( N = 1 ) , Saxony-Anhalt ( N = 3 ) and Saxony ( N = 1 ) ( Figure 2a ) . Some clustering was observed for EBLV-1a isolates from the same or from neighbouring federal states , with occasional exceptions . The nucleotide sequence divergence within the EBLV-1a group was <1% . Of 160 Daubenton's bats tested , three ( 1 . 88% ) individuals were rabies positive , and EBLV-2 was isolated in each case ( Table 1 , Figure 2b ) . Those infected bats were submitted from Saxony-Anhalt [35] , Thuringia and Hesse . Overall , in 47 smears from different bat species investigated using the FAT small fluorescing structures indicative for lyssavirus antigen was found , but the infection could not be confirmed by other methods , e . g . RTCIT , EBLV-1/-2 specific RT-qPCR and conventional RT-PCR . Furthermore , 13 . 1% ( N = 718 ) of all submitted bats could not be investigated because the carcasses were mummified or organs had autolysed ( Table 1 ) . Of all animals tested by FAT and with a reference to a date ( month , N = 3714 ) the peak of bat finds were in July , August and September , with a second peak in February and March ( Figure 3 ) . Of those , the percentage of bats tested EBLV-positive was highest in July ( N = 11 , 1 . 95% ) and August ( N = 12 , 1 . 96% ) . Altogether , 50 Serotine bats tested rabies positive by FAT , of which 18 were males and 11 females . Four of the positive cases were juvenile animals whereas the remaining animals were sub-adults or adults .
With enhanced passive surveillance 56 additional bat rabies cases were detected also in federal states where rabies in bats had not been found previously . Considering the large number of animals tested the prevalence was lower than in routine surveillance and likely represents the true level of lyssavirus infections in indigenous bats in Germany . Although the vast majority of cases were found in the known reservoir species Eptesicus serotinus , spill-over cases were also observed . In conclusion , all bat species need to be sampled and identified , and , since some bat species are still underrepresented , the enhanced surveillance should be maintained . | According to the World Health Organization rabies is considered both a neglected zoonotic and a tropical disease . The causative agents are lyssaviruses which have their primary reservoir in bats . Although bat rabies is notifiable in Germany , the number of submitted bats during routine surveillance is rarely representative of the natural bat population . Therefore , the aim of this study was to include dead bats from various sources for enhanced bat rabies surveillance . The results show that a considerable number of additional bat rabies cases can be detected , thus improving the knowledge on the frequency , geographical distribution and reservoir-association of bat lyssavirus infections in Germany . The overall proportion of positives was lower than during routine surveillance in Germany . While the majority of cases were found in the Serotine bat and characterized as European bat lyssavirus type 1 ( EBLV-1 ) , three of the four EBLV-2 infections detected in Germany were found in Myotis daubentonii during this study . | [
"Abstract",
"Introduction",
"Material",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases",
"rabies",
"veterinary",
"diseases",
"zoonoses",
"medicine",
"and",
"health",
"sciences",
"infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"emerging",
"infectious",
"diseases",
"veterinary",
"microbiology",
"virology",
"emerging",
"vi... | 2014 | Enhanced Passive Bat Rabies Surveillance in Indigenous Bat Species from Germany - A Retrospective Study |
Scrub typhus is a mite-borne febrile disease caused by O . tsutsugamushi infection . Recently , emergence of scrub typhus has attracted considerable attention in several endemic countries in Asia and the western Pacific . In addition , the antigenic diversity of the intracellular pathogen has been a serious obstacle for developing effective diagnostics and vaccine . To understand the evolutionary pathway of genotypic diversification of O . tsutsugamushi and the environmental factors associated with the epidemiological features of scrub typhus , we analyzed sequence data , including spatiotemporal information , of the tsa56 gene encoding a major outer membrane protein responsible for antigenic variation . A total of 324 tsa56 sequences covering more than 85% of its open reading frame were analyzed and classified into 17 genotypes based on phylogenetic relationship . Extensive sequence analysis of tsa56 genes using diverse informatics tools revealed multiple intragenic recombination events , as well as a substantially higher mutation rate than other house-keeping genes . This suggests that genetic diversification occurred via frequent point mutations and subsequent genetic recombination . Interestingly , more diverse bacterial genotypes and dominant vector species prevail in Taiwan compared to other endemic regions . Furthermore , the co-presence of identical and sub-identical clones of tsa56 gene in geographically distant areas implies potential spread of O . tsutsugamushi genotypes . Fluctuation and diversification of vector species harboring O . tsutsugamushi in local endemic areas may facilitate genetic recombination among diverse genotypes . Therefore , careful monitoring of dominant vector species , as well as the prevalence of O . tsutsugamushi genotypes may be advisable to enable proper anticipation of epidemiological changes of scrub typhus .
Scrub typhus is an acute febrile illness caused by Orientia tsutsugamushi infection . The bacterium is an obligate intracellular pathogen maintained through transovarian and transtadial transmission in trombiculid mites that serve as vectors for the infectious disease [1 , 2] . The disease is endemic in Asia and the western Pacific area including northern Australia . The first description of a febrile disease thought to be scrub typhus , along with the morphology of the vector mites , appeared in Chinese literature in 313 A . D . [3] . Therefore , it seems to an ancient infectious disease that has long been confined to its endemic area , although several cases of suspected scrub typhus have been reported outside of the endemic region [4–7] . Global incidence of scrub typhus across the whole endemic region has been poorly defined due to the limited epidemiological data in many of the endemic countries . Nevertheless , it has been estimated that more than a million cases occur annually and a billion people are at risk [8] . In addition , there has been a rapid increase in scrub typhus cases , as well as sporadic local outbreaks during the last decade [9–13] , making it a serious public health issue in the endemic area . Recent epidemiological data available in various resources ( S1 Table ) , clearly demonstrates the gradual emergence of scrub typhus in several endemic countries ( Fig 1 ) . Even though the increasing number of reported cases of scrub typhus might be partly due to increased awareness and better surveillance systems in the developing countries [2 , 9] , environmental change and human activity might be important factors contributing to the emerging trend [14–17] . Given that vector mites maintain the intracellular pathogen , ecological changes of the vector species in local endemic regions could be the primary cause of the emergence of scrub typhus , as recently observed in South Korea [15 , 16] . However , the distribution of mite species associated with scrub typhus in the whole endemic region has been poorly monitored and the currently available vector map issued by the World Health Organization ( WHO ) is based on data before 1974 [18–20] . Another critical issue of scrub typhus is the apparent antigenic diversity of O . tsutsugamushi throughout the region of endemicity [1] . The antigenic heterogeneity has been a serious obstacle for developing effective diagnostic methods , as well as a universal scrub typhus vaccine [1 , 2] . Historically , the antigenic variation of O . tsutsugamushi was characterized by several serological techniques using whole bacterial antigens , such as complement fixation and immunofluorescence assay . Based on serological analyses , the bacterial pathogen has been classified into various strains , including Karp , Gilliam , and Kato prototypes [21] . The defined “serotypes” have been changed into “genotypes” since a 56 kDa type-specific antigen , tsa56 , occupying approximately 20% of whole bacterial proteome [22] , was identified to be a major bacterial surface antigen reactive to strain-specific antibodies [23] . Genetic analysis of the tsa56 gene has provided the most useful standard to differentiate the genotypes of O . tsutsugamushi [22 , 24 , 25] and a growing number of tsa56 sequences has been deposited in the international nucleotide database [26] . As the number of tsa56 sequences increases , more diverse genotypes of O . tsutsugamushi have been identified [1 , 27] . Nevertheless , the question of how the genetic diversity of tsa56 gene has evolved remains unsolved . In addition , the potential relationship between genotype variation and epidemiological changes has been poorly assessed in the global level . In order to understand the evolutionary pathway of genotypic diversification of O . tsutsugamushi , as well as the environmental basis associated with the epidemiological changes of scrub typhus , we collected and analyzed the data of tsa56 genes , including genetic sequences and their spatiotemporal information . We also searched and reviewed references containing information on the Leptotrombidium species , the primary vectors of scrub typhus , to update the vector map and examine ecological changes of the “natural host” of O . tsutsugamushi in the whole endemic region . The systemic analysis of genotype diversity and geographical distribution of the vector hosts may not only provide valuable insight into the factors affecting epidemiological changes , but also enhance our current knowledge required for developing better diagnostics and an effective vaccine for scrub typhus .
Annual incidences of scrub typhus in endemic countries were collected from various references and resources ( S1 Table ) . The number of scrub typhus cases reported in each country has been based on different diagnosis standards . The criteria for confirmed cases of scrub typhus in China , Japan , and Taiwan include clinical manifestations , and one of the following laboratory diagnosis criteria: serological analysis such as indirect immunofluorescence assay , detection of O . tsutsugamushi DNA by PCR , or isolation of the pathogen from clinical specimens . Data from other countries include both clinically suspected cases , which are solely based on epidemiological exposure histories and clinical symptoms , and confirmed cases by laboratory diagnosis as described above . Nucleotide sequences encoding the tsa56 gene were collected from the National Center for Biotechnology Information ( NCBI , http://www . ncbi . nlm . nih . gov/ ) . As of Dec . 31 . 2015 , 1 , 030 nucleotide sequences have been deposited in the sequence databases . Among them , we selected and analyzed 324 nucleotide sequences that covered at least 85% of the open reading frame . Other information of the selected tsa56 genes , such as isolation host , year of isolation , and isolated location , were also retrieved from the database or manually collected from the references citing the sequences , and summarized in S2 Table . Based on the phylogenetic analyses described below , we re-annotated the genotypes of the 324 sequences and selected a representative proto-genotype sequence for each genotype . The proto-genotype sequences were selected if genomic information was available or its genome sequencing was underway in Bioproject ( http://www . ncbi . nlm . nih . gov/bioproject/ ) . If there was no genomic information available , the proto-genotype was annotated to a sequence which was firstly reported among the genotype members . We also searched and reviewed literature for epidemiological data on the distribution of nine Leptotrombidium mite species , the primary vectors responsible for the transmission of scrub typhus , to construct and update the vector map . The spatiotemporal information of the mite species is summarized in S3 Table . The geographical distribution of the vector species was based on locations described in literature using the QGIS program ( http://qgis . org/ ) and map dataset available in Natural Earth ( http://www . naturalearthdata . com/ ) . If a geographical reference did not include specific decimal latitude and longitude information , we used area information ( generally province or county level ) described in literature . The sequence data of tsa56 genes were translated and analyzed using the MEGA6 software [28] . Since the length of nucleotide and amino acid sequences were quite variable , we first aligned the amino acid sequences using the MAFFT algorithm ( v . 7 . ) with the E-INS-i option [29] , manually checked the aligned sequences , and trimmed them to select the sequence region shared by all 324 genes [30] . The lengths of the selected amino acid sequences range from 417 ( 1251 bases ) to 467 ( 1401 bases ) ( S2 Table ) , and cover the majority of the extracellular region and excludes the signal peptide of the TSA56 protein . Among the 324 genes analyzed , 156 were found to share identical nucleotide sequences in at least two genes . Therefore , we annotated 206 sequence IDs to the gene set ( S2 Table ) and analyzed them for genetic relationships . Phylogenetic analysis of the aligned nucleotide sequences was performed using the MAFFT algorithm and the Randomized Axelerated Maximum Likelihood ( RAxML ) method as implemented in SeaView software ( v . 4 . 5 . 1 ) [31] . Shimodaira-Hasegawa-like ( SH-like ) test [32] was computed to measure the statistical support of ML tree , as implemented in RAxML . The values for SH-like branch support are presented at the nodes on the phylogenetic tree and values above 0 . 9 were considered as significant phylogenetic support . Pairwise identity and similarity matrices of amino acid sequences were constructed by the MatGAT2 . 1 program [33] , which aligned the sequences using the BLOSUM62 matrix . Intragenic recombination was screened within the aligned sequences using the Genetic Algorithm Recombination Detection ( GARD ) method [34] implemented in Datamonkey server [35] . This program identifies the number and location of breakpoints and sequences involved in putative recombination events . In addition , seven methods implemented in Recombination Detection Program ( RDP4 ) suite [36] were also applied to detect potential recombinant sequences , parental sequences , and recombination breaking points: 3Seq [37] , Bootscan [38] , Chimaera [39] , GENECONV [40] , MaxChi [41] , RDP [36] , and SisScan [42] . Analyses were performed with default settings for the detection methods and each potential event was considered significant when a support p value was less than 0 . 05 by more than six detection methods . The breakpoints and recombinant sequences inferred for every potential event were manually checked and adjusted using the phylogenetic and recombination signal analyses available in RDP4 suite . A similarity plot of tsa56 nucleotide sequences displaying the extent of genetic diversity between the significantly related genotypes was generated using a window of 200 nucleotides and a step of 20 nucleotides . In order to estimate and compare the degree of genetic diversity of tsa56 genes with other O . tsutsugamushi genes , we collected 53 bacterial genes including the tsa56 gene from two complete genome sequences ( Boryong [43] and Ikeda strain [44] ) and seven draft genomic contigs ( strain:Bioproject accession no . ; Gilliam:PRJNA212442 , Karp:PRJNA212456 , Kato:PRJNA212440 , TA716: PRJNA212457 , TA763:PRJNA212454 , UT76:PRJNA212456 , UT144:PRJNA232539 ) available in Bioproject ( http://www . ncbi . nlm . nih . gov/bioproject/ ) . All selected 53 genes are present in at least eight genomes . Nucleotide diversity ( π , the average number of nucleotide differences per site ) , mutation rate ( θ , Watterson’s mutation parameter ) , and recombination parameter ( ρ ) of the gene sets were estimated by LDHat package implemented in RDP4 suite . The number of non-synonymous ( Ka ) and synonymous ( Ks ) substitutions for each gene locus were calculated by using the SeqinR package in the R-project . We also identified genotype-specific insertions/deletions ( Indels ) in tsa56 genes by manual inspection of 206 aligned sequences and summarize them in S4 Table .
Phylogenetic analysis was performed using 206 unique tsa56 nucleotide sequences ( S2 Table ) and defined the genetic clusters based on branching supporting values ( SH-like value ≥ 0 . 90 ) and the relative branch length from a node . At least 17 genotypes were defined by the phylogenetic analysis of the nucleotide sequences and named after the prototype strains ( Fig 2 ) . Based on the phylogenetic distances , 17 genotypes were further classified into 5 groups: Karp , Gilliam , TA763 , Kato , and Shimokoshi . When we compared the 206 genes using protein sequences , the ranges of sequence similarity and identity in the amino acid level further support the grouping of genotypes ( Fig 3 and S5 Table ) . Within each genotype , minimum similarity and identity among the gene members are generally over 80 . 0% and 70 . 0% , respectively , with the exception of the Shimokoshi genotype ( min . similarity: 79 . 0% , min . identity: 68 . 2% ) . Among the gene members within a group , minimum similarity and identity are further reduced to 73 . 0% and 60 . 1% , respectively , as observed in the Shimokoshi group . Among all the collected genes , the minimum similarity ( 66 . 2% ) was observed between members of Saitama and Shimokoshi genotypes and the minimum identity ( 52 . 9% ) was detected between members of Kato_A and Shimokoshi genotypes . It is worth noting that gene members in the Gilliam genotype and in the TA763 group show relatively higher similarity to those of Karp group ( Fig 3 ) , although they are phylogenetically distant . Since we observed a relative conservation of amino acid sequences among the genotype members belonging to phylogenetically distant groups , such as Gillam and TA763 members with Karp members , and it has long been proposed that genetic recombination might be a driving force for generating genetic diversity of the intracellular pathogen [22 , 27 , 43 , 45–47] , we examined the genetic recombination of 17 prototype sequences of tsa56 using several different recombination detection programs . First , the evidence for recombination in the aligned 17 proto-genotype sequences was tested by the GARD method using the Datamonkey web server ( http://www . datamonkey . org ) [48] . This method detected evidence for recombination at multiple breakpoints predicted at nucleotide positions 295 , 575 , 958 , and 1406 ( p < 0 . 01 ) ( Fig 4A ) . Detection of potential recombinant sequences , identification of potential parental sequences , and localization of possible recombination break points were further determined using the 3Seq [37] , Bootscan [38] , Chimaera [39] , GENECONV [40] , MaxChi [41] , RDP [36] , and SisScan [42] methods embedded in RDP suite [49] . As shown in Fig 4B and S6 Table , significant recombination events were detected in 11 proto-genotypes with a high degree of confidence ( p < 0 . 05 for at least six out of seven recombination detection programs ) , but not in Saitama , Boryong , Kawasaki , TD , Shimokochi , and TA686 genotypes . Among the 11 proto-genotypes , 9 genotypes showed a single recombination event , whereas multiple recombination events were predicted in 2 genotypes , JG_A and Gilliam . The major parents , minor parents , and break points with statistical significance ( p < 0 . 05 ) confirmed by MaxChi and BOOTSCAN programs are summarized in S6 Table . Fig 4C shows four representative recombination events observed in Karp_A , JG_C , TA763 , and Kato_B genotypes . These results suggest that 6 genotypes , Saitama , Boryong , Kawasaki , TD , Shimokoshi , and TA686 , may be the ancestral parents of the 11 recombinant genotypes . Based on the number of recombination events observed in the diverse genotypes , we speculate that the recombinant genotypes might have been generated by sequential recombination events among the parental genotypes ( Fig 5 ) . The 6 genotypes which lack any evidence of recombination , might be the first generation that contributed to the second generation ( Karp_C , Karp_B , Karp_A , TA763_B , and Kato_B ) . Three members ( Kawasaki , Shimokoshi , and Boryong ) of the first generation also contributed to the third and fourth generations . Gilliam genotype , the sole member of the fourth generation , seems to have been generated by recombination between Boryong , Karp_B , and JG_C genotypes ( Fig 5B ) . These results suggest that the genotype diversification of O . tsutsugamushi may be an ongoing process driven by continuous recombination events among preexisting genotypes . Genetic diversification of a bacterial gene can be attributed to point mutations as well as genetic recombination . Therefore , we analyzed the relative contribution of recombination and point mutation to the diversification of tsa56 genes and compared these results to those of 52 other O . tsutsugamushi genes collected from nine genomes that are completely sequenced or undergoing sequencing ( Fig 6 and S7 Table ) . Average recombination rate per base pair ( ρ/bp ) of the 53 genes is 0 . 083 , with a range of 0 . 003 ( rpsB ) to 1 . 131 ( rpsT ) , and average mutation rate per base pair ( θ/bp ) is 0 . 020 , with a range of 0 . 010 ( trmU ) to 0 . 086 ( tsa56 ) . The detecting per site ρ/θ value for the overall gene sets is 4 . 071 , suggesting that recombination occurred more frequently than point mutation . Interestingly , the mutation rates of tsa56 and sca family genes , encoding outer membrane proteins [50] , are generally higher than other house-keeping genes , even though their recombination rates were near average value . When we recalculated per site recombination and mutation rates using 206 unique tsa56 genes , ρ/bp and θ/bp are 0 . 041 and 0 . 050 , respectively , indicating that mutation rate is slightly higher than recombination rate ( per site ρ/θ value = 0 . 812 , S7 Table ) . These results suggest that genetic diversification of the major outer membrane protein , TSA56 , might be driven by recombination as well as frequent point mutation . Based on the Ka/Ks ratio , an indicator of selective pressure acting on a protein-coding region , most O . tsutsugamushi genes , including tsa56 , predominantly evolve by purifying selection with the exception of rplM , scaA , and scaE , which may evolve under positive selection . In addition to the point mutations , Indels of nucleotide sequences have often been observed in Rickettsial genes [51 , 52] and may also contribute to the genotypic diversification of Orientia . Extensive analysis of the aligned 206 tsa56 genes revealed 4 , 771 Indels at over 108 sites , especially in regions encoding variable domains [22] . However , only a few of them are consistently detected in a specific set of genotype sequences ( S4 Table ) , indicating that , although they contributed to the diversification of tsa56 sequences , only fraction of Indels are conserved in a specific genotype . The geographical distribution of 324 tsa56 genes is presented in Fig 7 and S2 Table . The Karp group includes the largest number of isolates ( 175 genes ) and genotypes in this group are found in most endemic countries . Among the genotype members in the Karp group , each country has a specific predominant genotype , such as Boryong in South Korea , Karp_C in Japan , and Karp_A in Taiwan and Cambodia . Genotype members of the Gilliam group ( 78 isolates ) are also quite prevalent in the endemic countries . Among the genotype members in this group , the Kawasaki genotype is prevalent in South Korea and JG_C genotype is primarily found in Taiwan , Thailand , and Cambodia . The isolates belonging to the TA763 and Kato groups are mainly reported in Taiwan , but rarely in South Korea and China . The highly divergent Shimokoshi genotype is only reported in Japan . Even though the number of sequences isolated from each country is quite varied , the genotype diversity found in Taiwan located in the middle of endemic area of scrub typhus , is particularly notable . 14 genotypes out of 17 are found in Taiwan , compared to 5 genotypes in South Korea . When we compared the diversity and relative proportion of genotypes in Taiwan with those of the northern endemic area ( China , Japan , and Korea ) or southern endemic countries ( Cambodia , Malaysia , Myanmar , Papua New Guinea , Thailand , and Vietnam ) , it is clear that not only the diversity , but also the relative proportion of each genotype in Taiwan are quite distinct from other endemic regions ( Fig 7B ) . Prevalence of more divergent genotypes in a certain central locality than in countries at the boundary of the endemic region suggests that Taiwan might serve as a mixing ground for the diverse genotypes . The distribution of vector mites is the primary factor affecting the epidemiological features of scrub typhus . However , the currently available vector map , published by the World Health Organization in 1989 [18] , is primarily based on data before 1974 [1 , 3 , 19 , 20] . Therefore , we searched references and updated the geographical distribution of the Leptotrombidium species , the main vectors of scrub typhus ( Fig 8 and S3 Table ) . Three major Leptotrombidium species , L . palpale , L . pallidum , and L . scutellare , have been found in the northeastern area of endemicity [15 , 18 , 53–55] . In particular , L . scutellare has recently become the primary vector in northern China [56 , 57] , South Korea [15] , and Japan [53] . It is also notable that L . pavlovskyi , the primary vector responsible for scrub typhus in the Primorye region of Russia during the 1960s , was also prevalent in the 1990s , although O . tsutsugamushi could not be isolated from the vector species [58] . L . delicense is the primary vector in the southern parts of the endemic region , ranging from southern China to the north , Pakistan to the west , and northern Australia and western Pacific islands to the south . A number of studies have also reported the presence of L . delicense in south-eastern countries of the endemic region since 1974 ( S3 Table ) . It appears that two Leptotrombidium species , L . scutellare in the northern part and L . delicense in the southern area of endemicity , might be the primary vectors for current scrub typhus . However , there is local variation of major vectors such as L . imphalum in eastern Taiwan [59] , L . chiangraiensis and L . imphalum in northern Thailand [60] . In addition , L . scutellare has been recently reported to have expanded both northward to mainland China [56 , 57] and South Korea [15] , and the southward to southern China [55] and Taiwan [59 , 61] . As a result , the southern provinces of mainland China and Taiwan have become mixing grounds for the two primary vector species , L . scutellare and L . delicense . It is also notable that L . imphalum , the primary vector in northern Thailand [60] , has also been reported as the dominant mite in a local area of Taiwan [59] , such that diverse vector species prevail in islands of Taiwan . As mentioned above , there are 38 sets of identical sequences , including 156 sequences , among the selected 324 tsa56 genes ( S2 Table ) . Interestingly , among the identical sequence sets , 10 sets of genes include sequences from two different countries ( S8 Table ) , whereas 28 sets were only reported within a single country . For example , the same sequences belonging to Karp_C , Kawasaki , and JG_A have been reported from South Korea and Japan , even though the isolated years are different . Identical sequences in the Karp_A genotype have also been found in Taiwan and Thailand . In addition , we detected sequence pairs , showing only one or two base sequence differences ( S8 Table ) , originated from two different countries , such as Boryong genotype from Korea and Taiwan . The presence of identical or sub-identical ( 1 ~ 2 different bases ) tsa56 genes in geographically distant countries implies a potential migration or expansion of the bacterial clones , even though it needs to be confirmed whether other bacterial genes and/or the whole genomic sequences of the isolates are also identical . It is also intriguing that Taiwan emerged as a central node connected to northern countries ( China , Japan , and Korea ) , as well as southern countries ( Cambodia , Malaysia , and Thailand ) when we linked the countries where the identical and sub-identical sequences have been reported ( Fig 9A ) . Considering that prominent diversity in bacterial genotypes and mite vector species have been observed in Taiwan and its central location in the endemic area of scrub typhus , the subtropical region may potentially serve as a hub point mediating migration or expansion of vector mites , thereby contributing to the spread of diverse genotypes and their recombination .
The O . tsutsugamushi genome is quite unique in that up to 40% of its genome contains dispersed repeat sequences , primarily composed of mobile genetic elements including conjugative transfer ( tra ) components of a type IV secretion system and transposases [43] . Previously , sex pili-like cell surface appendages for conjugal DNA transfer were observed in Rickettsia belli , which encodes a tra cluster phylogenetically close to those of O . tsutsugamushi [45 , 62] . It is possible that genetic recombination occurs via a similar mechanism among O . tsutsugamushi . Coinfection of multiple genotypes in a single host might facilitate the genetic recombination among different genotypes . Indeed , mixed infection of multiple genotypes has often been reported in the vector mites [63 , 64] and human patients [46 , 47 , 65] . Given that humans are dead-end hosts , recombination between different strains more likely occurs in the mite vectors or in the rodent reservoirs . In addition to the conjugative transfer system , O . tsutsugamushi retains a relatively large repertoire of genes for recombination and repair processes , which may ensure genomic flexibility during recurrent host changes and induce genetic variation , especially in surface antigens for immune evasion [45 , 66] . Considering that tsa56 encodes the major outer membrane protein that plays a significant role in bacterial invasion into host cells [67 , 68] and the neutralizing antibodies against it provide protection [69] , immunological pressure on the bacterial antigen during mammalian host infection might be a crucial driver for genetic diversification via point mutations , Indels , and genetic exchange among different genotypes . Indeed , our current study revealed that the mutation rate of tsa56 gene is substantially higher than other house-keeping genes ( Fig 6 ) . This higher mutation rate was also observed in a group of outer membrane proteins encoded by sca family genes [50 , 70 , 71] . It is notable that the mutation rate of the scaA gene , which encodes a potential bacterial adhesin [72] , is the second highest compared to other included genes . In addition to the point mutations , intragenic recombination may also contribute to the genetic diversity of the outer membrane proteins . Even though the recombination rate of the tsa56 gene is similar to those of other Orientia genes ( Fig 6 ) , intragenic shuffling of the gene fragments has significantly contributed to the genotype diversification ( Fig 4 ) . Given that the extracellular domain of TSA56 includes highly variable domains [22] , as well as multiple antigenic domains [73] , intragenic recombination observed among the diverse genotype sequences can result in substantial shifts in genetic variation [27] and antigenicity [74] . In consistent with these , similar result was reported from a study using multiple tsa56 sequences isolated from human patients from three countries in Southeast Asia [75] . They also suggests that weak divergence in the core genome and ancestral haplotypes are maintained by permanent recombination in mites while the tsa56 gene is diverging in higher speed potentially due to selection by the mammalian immune system [75] . Due to the wide antigenic variation , immunity generated by vaccine trials , or even after natural infection , do not provide effective cross-reactivity among numerous genotypes , and reinfection with scrub typhus is relatively common in highly endemic areas [76] . Various studies also showed inter-genotype variation in virulence in humans and rodents , ranging from unapparent disease to consistent fatal infection when untreated [1] . Considering that each genotype generation ( Fig 5 ) classified based on the sequential recombination events in this study includes both virulent and avirulent genotypes and the relative virulence in challenged animals appears to be highly mouse-strain specific [77 , 78] , the genetic recombination of tsa56 gene may not be specifically associated with virulence for individual genotypes although detailed analysis on the degree of virulence need to be examined . Classification of O . tsutsugamushi genotypes has been primarily based on the sequence variation of tsa56 since it is unique to O . tsutsugamushi and highly variable in amino acid content due to multiple variable domains [1] . By the end of 2015 , more than a thousand tsa56 sequences ( size range , ~ 150 to > 2 , 000 bases ) were deposited in international sequence databases . Most sequences include the variable portions of the gene and were annotated as a specific strain name and/or genotype name based on sequence analysis . Our current results using sequences covering a complete or nearly complete coding sequence ( covering at least 85% of the open reading frame , ≥ 1 , 251 bases corresponding to 417 amino acids ) showed that genetic recombination might have occurred at multiple sites within the coding region . Therefore , genotypic classification using a small fragment , including only parts of variable domains , may not be sufficient to completely identify inter-strain variation within the target gene . Sequence analysis using the complete or nearly complete sequences including all the variable domains , as well as the region beyond the recombination break points near the 3’ and 5’ ends ( Fig 4 and S6 Table ) of the tsa56 gene might be required to clearly define O . tsutsugamushi genotypes . In a previous study , Kelly et al . reported that O . tsutsugamushi genotypes can be classified into at least 9 definable clusters when they analyzed 135 complete or nearly complete ( > 1 , 200 bases ) tsa56 genes [1] . Here , we identified at least 17 clusters of genotypes , belonging to 5 definable groups , when using 206 complete or nearly complete sequences ( ≥ 1 , 251 bases ) . Based on our sequence analysis , similarity and identity in amino acids also need to be considered to define the genotypes or groups since some genotypes , such as Gilliam and TA763 members , show unexpected higher similarity in amino acid sequences with phylogenetically distant genotypes ( Fig 3 and S5 Table ) . As the number of tsa56 sequences increases , the genetic variation is expected to further diversify when considering the high mutation rate and on-going recombination within the tsa56 gene . Currently , the geographical distribution of O . tsutsugamushi genotypes is a critical issue for the development of effective diagnostics and vaccine [2] . Antigenic variation generated by genetic diversification of the immunogenic major outer membrane protein , TSA56 , complicates diagnosis and efforts towards vaccine development . Therefore , an investigation of genotype diversity and prevalence in local endemic areas needs to be continued not only for the epidemiological monitoring of scrub typhus , but also for the improvement of diagnostic accuracy and vaccine development [47 , 74] . In this study , we examined the distribution of O . tsutsugamushi genotypes using the sequence data and the related geographical information . In addition , we also reviewed spatiotemporal changes of the primary vector species to assess the association with epidemiological changes of scrub typhus . Based on extensive data analyses , we found some compelling epidemiological features of scrub typhus . First , the prevalence of diverse genotypes of O . tsutsugamushi and multiple vector species in Taiwan is quite marked when compared to those of other endemic countries . The local prevalence of Leptotrombidium species is generally determined by multiple environmental factors such as temperature , precipitation , and host diversity [79 , 80] . Considering that L . deliense and L . scutellare are the major vectors of scrub typhus in the southern tropical area and northern temperate region , respectively ( Fig 8 ) , the presence of both mite species as dominant vectors might be a good indicator of vector diversity . The subtropical climate of Taiwan , as well as its location in the center of the endemic area , might provide a natural environment for such a vector diversity . Although L . deliense is a major vector throughout the islands of Taiwan , L . imphalum and L . pallidum , which are also primarily found in tropical area and temperate region , respectively , were more dominant in some Taiwanese provinces [80] . In addition , L . deliense was replaced by L . scutellare during the winter season in islands with lower winter temperature than the other areas , such that the former is responsible for summer scrub typhus and the latter for winter scrub typhus [80] . It is also interesting to note that the recent exponential increase of scrub typhus cases in mainland China has been primarily associated with regional clusters of the southern subtropical area [9 , 57] , which is geographically close to Taiwan . Recently , the presence of highly diverse mite species was reported in the Yunnan province , the main hotspot of scrub typhus in mainland China [9] , where L . scutellare and L . deliense are the major mite species [79 , 81] . The dominance of the two major vectors , as well as species diversity , are associated with local altitude and latitude gradients , suggesting an importance of climate and environmental conditions for codominance of mite species [79] . Although the genotype diversity of O . tsutsugamushi in endemic hotspots of southern China has been poorly defined , fluctuation and variety of the vector species due to environmental factors could also be associated with epidemiological features of scrub typhus in local endemic regions . Additionally , local changes in prevalent mite species harboring O . tsutsugamushi have been continuously reported in other subtropical and temperate area of endemic regions [16 , 80] . Ecological changes in the specific endemic locality may provide the environmental basis for the diversification of O . tsutsugamushi genotypes and/or their prevalence . Second , presence of identical or near-identical ( 1 ~ 2 different bases ) tsa56 genes in geographically distant countries suggests a potential of international migration of O . tsutsugamushi , even though the genomic identity of the clones needs to be further verified . Considering that O . tsutusgamushi are obligate intracellular bacteria , their migration is absolutely dependent on the associated host vectors and/or reservoir animals . Moreover , larval mites do not migrate more than a few meters from where they hatch and usually form ‘mite islands’ ranging from a few cm to meters [19] , so their ability to migrate on their own is very limited and their movement is mainly associated with the migration of hosts infested with chigger mites [3] . In addition to the wide spread of major vectors and diverse genotypes of O . tsutsugamushi over the endemic region including many islands in the Indian and Pacific Oceans [1] , continuous fluctuation in the distribution of chigger mites at the local level suggests that parasitized small rodents and birds may be potential phoretic hosts of the infected mites [3 , 19 , 82 , 83] . A recent study reported a potential role of an exotic rodent species introduced from Southeast Asia and Pacific islands , Rattus exulans , as a host for chiggers in Taiwan [83] . Even though exotic R . exulans appears to play a relatively minor role in supporting chigger species infected with O . tsutsugamushi in Taiwan , the fact that both prevalence and loads of chiggers in R . exulans vary greatly with environment , along with the abundance and the ecological flexibility of R . exulans , implies a potential health risk as this species expands to areas with more chiggers [83] . Since the role and influence of exotic rodent species in local diversity and spread of the vector-borne disease are important but poorly assessed thus far , further investigation on the role of invasive rodent hosts on the dynamics of scrub typhus needs to be followed . Additionally , the association of migratory birds in spreading vector-borne infectious agents , such as Borrelia burgdorferi [84] , Tick-borne encephalitis virus [85] , and severe fever with thrombocytopenia syndrome virus [86] , has been well documented . Considering that chigger mites attach and feed on host animals , including birds and rodents , for about 36–72 hours and withstand harsh environmental condition such as temperatures of -20°C for up to several weeks [87] , they can travel hundreds to thousands of kilometers while attached to migratory birds , to a new geographic area that they may colonize if environmental conditions are optimal for their survival [88] . Since O . tsutsugamushi has rarely been recovered from tissues of wild birds [19 , 89] , birds are more likely mechanical carriers for short- or long-distance transmission of chigger mites infected with O . tsutsugamushi rather than biological carriers . Based on our examination of the distribution of identical or near-identical tsa56 sequences , Taiwan was found to be the nodal point of clonal expansion to northern and southern parts of the endemic area ( Fig 9A ) . This further supports the idea that Taiwan , located in the center of the endemic area , may serve as a hub point mediating potential migration or expansion of vector mites , thereby enabling the generation and/or spread of diverse genotypes . Taiwan is also located at the center of the East Asia/Australasia Flyway migratory bird routes crossing the endemic countries of scrub typhus , extending from Arctic Russia to the southern limits of Australia ( Fig 9B ) . In addition , the major habitats of the migratory bird species appear to be correlated with the endemic area of scrub typhus . Therefore , consideration of avian migration patterns might be useful in understanding and predicting epidemiological changes , such as local outbreaks of scrub typhus , as well as spread of mite vectors and O . tsutsugamushi genotypes . Fluctuation and diversification of vector species harboring O . tsutsugamushi , potentially caused by environmental changes and influx from other endemic regions , could affect the epidemiological features of scrub typhus and facilitate the genetic recombination among the different genotypes , thereby enhancing the genotypic diversity of O . tsutsugamushi in local endemic regions . Careful monitoring of dominant mite species and the prevalence of O . tsutsugamushi genotypes associated with the vectors might be required to reveal the correlation of genotype diversification of O . tsutsugamushi with ecological vector changes . | Scrub typhus , caused by Orientia tsutsugamushi infection , is a mite-borne febrile illness endemic in the Asia-Pacific region . Recent emergence and continuous local outbreaks in many of the endemic countries make it a serious public health issue . In addition , the antigenic diversity of the tsa56 gene , encoding a major outer membrane protein , hampers the development of effective diagnostics and vaccine . Here , we extensively analyzed tsa56 sequences and their spatiotemporal information to elucidate the evolutionary pathway of genotypic diversification , as well as the environmental basis associated with the epidemiological changes of scrub typhus . Based on various informatics analyses , we found that genetic diversification of tsa56 might have been attained via frequent point mutations and subsequent genetic recombination among diverse genotypes . Prevalence of numerous bacterial genotypes and dominant vector species in Taiwan also suggest that the subtropical area located at the center of endemicity , may serve as a local mixing ground for genotype diversification . In addition , detection of identical and sub-identical clones of tsa56 genes in geographically distant countries indicates a potential spreading of bacterial genotypes . Continuous monitoring of dominant vector species and the associated O . tsutsugamushi genotypes might be required for developing better diagnostics and an effective vaccine for scrub typhus . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"biogeography",
"invertebrates",
"typhus",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"population",
"genetics",
"geographical",
"locations",
"microbiology",
"animals",
... | 2017 | Diversification of Orientia tsutsugamushi genotypes by intragenic recombination and their potential expansion in endemic areas |
Schlemm's canal ( SC ) plays central roles in ocular physiology . These roles depend on the molecular phenotypes of SC endothelial cells ( SECs ) . Both the specific phenotype of SECs and development of SC remain poorly defined . To allow a modern and extensive analysis of SC and its origins , we developed a new whole-mount procedure to visualize its development in the context of surrounding tissues . We then applied genetic lineage tracing , specific-fluorescent reporter genes , immunofluorescence , high-resolution confocal microscopy , and three-dimensional ( 3D ) rendering to study SC . Using these techniques , we show that SECs have a unique phenotype that is a blend of both blood and lymphatic endothelial cell phenotypes . By analyzing whole mounts of postnatal mouse eyes progressively to adulthood , we show that SC develops from blood vessels through a newly discovered process that we name “canalogenesis . ” Functional inhibition of KDR ( VEGFR2 ) , a critical receptor in initiating angiogenesis , shows that this receptor is required during canalogenesis . Unlike angiogenesis and similar to stages of vasculogenesis , during canalogenesis tip cells divide and form branched chains prior to vessel formation . Differing from both angiogenesis and vasculogenesis , during canalogenesis SECs express Prox1 , a master regulator of lymphangiogenesis and lymphatic phenotypes . Thus , SC development resembles a blend of vascular developmental programs . These advances define SC as a unique vessel with a combination of blood vascular and lymphatic phenotypes . They are important for dissecting its functions that are essential for ocular health and normal vision .
Although Schlemm's canal ( SC ) has central roles in ocular physiology and homeostasis , its development , mature phenotype , and molecular processes are poorly understood [1]–[3] . SC has a critical role in aqueous humor drainage ( AQH ) from the eye , a process that regulates the intraocular pressure ( IOP ) [1] , [2] , [4] , [5] . Abnormal resistance to AQH drainage results in IOP elevation , a key factor contributing to glaucoma [2] . Glaucoma is one of the most common neurodegenerative diseases and will affect an estimated 80 million people by the end of this decade [6] . SC is also important for anterior chamber associated immune deviation ( ACAID ) , a form of immune tolerance [3] . During ACAID , immune cells are exposed to an antigen in the eye and then exit the eye via SC . From SC they return to the systemic circulation via blood vessels to which SC is connected [7] , [8] . After exiting SC , these cells induce a systemic suppression of immune responses to that antigen . Thus , SC is a unique and important vessel that needs to be better understood . SC is a flattened tube made of endothelial cells , which encircles the anterior portion of the eye . It is embedded within the ocular wall in the region connecting the cornea and sclera that is known as the limbus . Specifically , SC is located in tissue of the iridiocorneal angle ( angle formed by the iris and cornea ) [2] , [9] . The inner wall of SC consists of morphologically specialized endothelial cells and their basement membrane , which provide a final barrier to the drainage ( outflow ) of AQH and the exit of immune cells from the eye [1] , [2] , [7] , [8] . SC endothelial cells ( SECs ) and their specialized basement membrane are likely to contribute a key source of resistance to AQH outflow . As immune cell behavior is modulated by interactions with endothelial cells , SECs are likely to have important molecular roles in immune tolerance . However , many mechanistic questions about the functions of SC remain unanswered . Determining the origin and phenotype of the SC and its endothelial cells is key to understanding its roles in ocular homeostasis and immune regulation . Based on a variety of features including marker expression , nature of cellular junctions , direction of fluid flow , and cellular morphology , SECs have similarities and differences to both blood endothelial cells ( BECs ) and lymphatic endothelial cells ( LECs ) and may be a unique endothelial cell type [5] . However , studies investigating the expression of lymphatic markers detected none in both human and mouse SC [10]–[12] . Thus , the molecular nature of SECs remains controversial . SC is proposed to develop from blood vasculature , but further investigation of its tissue origins is required as existing models of SC development differ significantly . In the first model , SC forms from a blood filled venous plexus anterior to the trabecular anlage ( the anlage that gives rise to the trabecular meshwork , which is adjacent to SC in mature eyes ) [13]–[17] . In the second model , SC forms from blood vessels originating from a more superficial limbal plexus [18] . Our previous studies suggested that SC forms by the penetration of existing vessels to a location adjacent to the trabecular anlage and that they anastomose to make SC [19] . These previous studies are limited by the use of techniques that sample small regions of tissue in two-dimensional sections ( using light and electron microscopy ) . They provide no molecular detail about mechanisms and have not considered or tested a lymphatic origin for SC . To allow a modern , more detailed and extensive analysis of the SC phenotype and its developmental origins , we developed a new limbal whole-mount procedure and applied lineage-specific fluorescent reporter genes , high-resolution confocal microscopy , and three-dimensional ( 3D ) rendering to study large regions of the developing limbus . We show that in addition to expressing markers of BECs , developing and mature SECs express PROX1 . PROX1 ( prospero-related homeobox1 ) is well established to be an important regulatory protein , which is necessary and sufficient for acquiring a lymphatic fate [20] . Furthermore , we discover that SC develops by a previously unknown process , which has commonalities and differences to the three described processes of vascular development—vasculogenesis , angiogenesis , and lymphangiogenesis .
The ocular drainage structures are delicate and easily damaged . This has impeded studies of their development in unsupported tissue preparations . Surmounting these impediments , we developed a whole-mount procedure for the limbus and anterior portion of the eye . This procedure allows us to study the entire mouse limbus and SC in 3D ( Figure 1 and Figure S1 ) . This method permits high-resolution in situ analysis of not only the entire adult SC but also the developing SC at a cellular level , and in the context of all limbal vessels and tissue . It overcomes substantial shortcomings of analyzing conventional sections , including the study of limited tissue expanses and the difficulties of accurately interpreting data from thin sections . It allows vascular connections between vessels types to be directly visualized . To clarify the orientation of the images presented throughout this article , we use a 3D-coordinate system to align the conventional view of the SC and limbus in sagittal section to whole-mount images ( Figure 1 and Figure S1 ) . Figure 1 and Figure S1 also introduce a depth coding procedure that is used in other figures , and relates tissue depths to the conventional view of SC . This procedure codes tissues in different colors based on their relative distance from the external ocular surface . Endomucin is an endothelial sialomucin . In humans , both BECs and LECs express it [21] . In contrast , in mice it is most highly expressed in BECs of veins and capillaries and expressed at low or undetectable levels in other mouse endothelial cell types including LECs [22] , [23] . Immunostaining with an endomucin antibody shows that adult mouse SECs express this sialomucin at readily detectable levels , though often lower than that in BECs ( Figure 1A and B ) . It is not detectable in mature mouse LECs ( Figure S2 ) . As far we are aware , this is the first report of endomucin expression in SECs . Adult SC is evident as a flat vessel running along the length of the limbus . In cross-section , its width varies from 50–200 µm . The vascular , basement membrane marker Collagen IV is also robustly produced by SC , and it highlights the characteristic anatomy of its inner and outer walls ( Figure 1C–E ) . When Z-depth is color-coded , SC codes as cyan to blue . A prominent vascular plexus runs parallel and superficial to SC ( codes magenta to red , Figure 1A , B ) , and we call it the limbal vascular plexus ( LVP ) . Branches of the LVP penetrate to the depth of SC with some connecting to SC as collector channels for AQH . Other vessels that exit the eye run at right angles to the LVP and connect with it . These vessels are often filled with red blood cells ( RBCs ) and are the episcleral vessels ( not shown ) . To facilitate studies of SC , and to further define the vascular markers expressed by it , we evaluated the expression of green fluorescent protein ( GFP ) transgenes controlled by vascular marker promoters and assessed other markers by immunolabeling . In addition to endomucin , SECs express a panel of vascular markers including Tie2 , Kdr ( Vegfr2 ) , EfnB2 , CD31 ( PECAM1 ) , CD34 , and VE-cadherin ( VECAD ) ( Figure 2A ) . Of these , CD31 , CD34 , and VECAD were previously identified in human SC [12] , [24] . Tie2 and CD34 are detected in BECs but not LECs under normal conditions [25] , [26] , whereas CD31 is expressed at much lower levels in LECs than BECs [25] . Resembling BECs , our data show robust expression of Tie2 , CD31 , and CD34 in SECs . KDR is an important vascular marker that is expressed in both BECs and LECs . VECAD is an adherence junction protein expressed in BECs and LECs and is critical in regulating junctional permeability [27] . VECAD localization between cells of blood vessels is continuous , contributing to an impermeable junction [28] . In contrast , VECAD has a discontinuous button-like localization in initial lymphatics , which are highly permeable [28] . In SC , VECAD localization resembles that of blood vessels and it serves as a useful marker for identifying SC with its distinct inner and outer wall morphology ( Figure 2B and C ) . Efnb2 is expressed in arterial endothelial cells and LECs but not in venous endothelial cells [29] , [30] . Existing models suggest that SC is derived from veins and so expression of Efnb2 may reflect differentiation towards a lymphatic phenotype . Our results thus show that the mouse SCE expresses common endothelial markers with important characteristics of BECs . PROX1 and LYVE1 are commonly used as markers of a lymphatic lineage [31] . Given morphologic similarities between SC and lymphatics , we assessed expression of these molecules in SECs . Using immunostaining , we discovered that PROX1 expression is readily detectable in SECs of the inner wall but is not detectable in SECs of the outer wall ( Figure 3A , Figure S3 ) . Thus , PROX1 , the master regulator of lymphatic phenotypes , may have a pivotal role in controlling the differentiation and maintenance of the functionally specialized inner wall SECs . To independently confirm Prox1 expression and to provide a fluorescent tool for analyzing SC , we produced transgenic mice expressing GFP from the Prox1 promoter ( Prox1-GFP ) [32] . Using these mice , we confirmed Prox1 expression in inner wall SECs ( Figure S4 ) . Again , expression was predominantly in the inner wall , but low levels of expression were detectable in some outer wall cells ( possibly due to less specificity of GFP expression from the transgene or greater sensitivity of this assay ) . These experiments indicate that the Prox1 promoter is transcriptionally active in adult SECs with predominant expression in the inner wall cells . These results show for the first time that PROX1 is expressed in the SC , although at lower levels than in lymphatics . FLT4 ( VEGFR3 ) is another key protein required for lymphatic development and also serves as a marker for lymphatics . FLT4 transcription is controlled by PROX1 [33] . Using immunostaining , we determined that FLT4 expression mirrors PROX1 expression , localizing predominantly in the inner wall of SC with low levels of expression in outer wall cells ( Figure 3B ) . In contrast to PROX1 and FLT4 , LYVE1 is not produced by SECs ( Figure 3C ) . SC also does not express a commonly used lymphatic marker , podoplanin ( Figure S5 ) . Overall , the above data indicate that SC is a unique vessel with a blend of blood vessel and lymphatic phenotypes . Given limitations of previous studies and the earlier erroneous belief that the eye lacked lymphatic vessels , it remains possible that SC is derived from lymphatics . We used a Lyve1-Cre transgene [34] along with the ROSA-mTmG reporter gene [35] to indelibly label cells that are derived from LYVE1-expressing cells . In cells expressing both the Lyve1-Cre and mTmG transgenes , the CRE recombinase deletes a constitutively expressed tdTomato transgene flanked by loxP sites . This irreversibly activates expression of a GFP transgene , and this expression is passed on to descendent cells ( Figure 4A ) . LYVE1 is first expressed in cells of the cardinal vein at embryonic day 9 . 5 and is the first indicator of these cells committing to a lymphatic fate [36] . Thus , the Lyve1-Cre mTmG mouse serves not only as an indicator of lymphatic expression but also of previous activation of a lymphatic developmental program . Our lineage tracing experiment clearly indicates that SECs and their progenitors do not express LYVE1 during development ( Figure 4B and Figure S6 ) . Thus , SC arises using a program different from embryonic lymphangiogenesis , and SECs are not derived from lymphatics . For some time , it has been suggested that SC is derived from mesoderm but not neural crest cells [37] . More recently , this was supported in mouse eyes using Cre-based lineage tracing [38] . Because a lacZ reporter and conventional sections were used , a mixed contribution of progenitor cell types including some neural crest cells could have been missed . Thus , we have revisited the possibility of a partial neural crest contribution by analyzing the entire SC following lineage tracing using both Wnt1-Cre and ROSA-mTmG transgenes . Our results confirm previous findings and clearly demonstrate that neural crest cells do not contribute to SC development , even though the surrounding tissues have a strong neural crest contribution ( Figure 5 , Figure S7 ) . Our data also confirm that neural crest cells do not contribute to the vasculature of the limbus ( Figure 5 ) , which is of mesodermal origin [38] . Together our results are consistent with a common mesodermal origin of both SC and the limbal vasculature , and they support a developmental model where SC arises from limbal blood vessels . Hypothesizing that SC develops from limbal blood vessels and revaluating the existing developmental models , we studied SC development in the context of all limbal vasculature . Because Kdr is expressed in both BECs and SECs , we used Kdr-GFP heterozygous mice [39] to visualize the limbal vessels and developing SC ( Kdr heterozygous mice are established to have normal vascular development [40] ) . Moreover , we found that the SC of Kdr-GFP heterozygous mice develops normally when compared to C57BL/6J mice ( e . g . , Figure 2 ) . We established that the LVP in a P1 eye is complex and runs completely around the eye just below the external ocular surface ( Figure 6 and Figure S8 ) . Deep within the limbus and close to the inner surface of the ocular wall there is another vascular bed . These vessels run radially in a direction perpendicular to the LVP . We named these vessels the “radial vessels” ( RVs , Figure 6 and Figure S8 ) . SC will form in the tissue between these vascular beds , a region that we name the “intermediate zone” ( IZ ) . At this age , the IZ is ∼40 µm thick ( Figure 6A and B ) . Signs of SC development are evident at P1 , as endothelial sprouts penetrating into the IZ . Importantly , these sprouts emanate from both the LVP and RV ( Figure 6B ) . Such sprouts are present all around the limbus and penetrate 10–20 µm into the IZ , the region where SC develops . Tip cells are specialized endothelial cells that are critical in angiogenesis . They are present at the leading edge of vascular sprouts and lead the migration of endothelial cells from the parental vessel into the surrounding tissue . They are characterized by long filopodia ( which integrate environmental cues ) and a characteristic polar morphology . They are followed by other specialized endothelial cells know as stalk and phalanx cells that connect the sprout to the parent vessel [41] . Tip and stalk cells are not stable cell fates but shuffle and interchange status during the sprouting process [42] . We discovered that the leading cells of the sprouts that penetrate into the IZ have a characteristic tip cell morphology with long filopodia ( Figures 6 and 7 ) . For clarity , and because they give rise to SC , we call these cells “SC-tip cells . ” It is worth noting that these sprouts form in the context of active angiogenic remodeling within the LVP ( see Figure S11 ) . Angiogenesis is initiated by interaction of VEGFA peptides with their receptor KDR . Here we show that KDR and its ligand VEGFA-164 localize to the developing SC at an age of active tip cell formation and interaction ( Figure S9 shows VEGFA-164 localization ) . In angiogenesis , pairs of tip cells interact with each other using their filopodia . Macrophages chaperone the filopodial interactions to accomplish anastomosis and the formation of a new vascular branch as the sprouts fuse [43] . Using endomucin and VECAD labeling of P2 eyes , we found that SC-tip cells interact with each other through their filopodia ( Figure 8 and Figure S10 ) . In most cases , the interacting filopodia have an attendant macrophage ( arrowheads , Figure 8A ) . Thus , similar to angiogenesis ( see Figure S11 ) , macrophages appear to chaperone filopodial interactions of SC-tip cells . In angiogenesis , VECAD interactions between tip cells promote adhesion [44] . Thus , filopodial VECAD is expected to make adhesive contacts between SC-tip cells . In contrast to angiogenesis , where tip cells interact with each other and their attached sprouts anastomose , groups of SC-tip cells interact with each other without anastomosis of their sprouts and without new tube formation . Instead , the SC-tip cells adhere to each other and interlace in the IZ , while maintaining a flattened morphology and giving rise to tip cell clusters ( Figure 9 ) . By P3 , the interacting SC-tip cells have formed tip cell clusters in the IZ throughout the limbus ( Figure 10 and Figure S12 ) . Many of these clusters are still attached to the LVP , RV , or both LVP and RV ( Figure 10 ) . However , not all of the cell clusters are connected to blood vessels . In these cases , it is not clear if connections to blood vessels were already pruned or if some of these clusters originated separately from other undefined progenitors . The number of tip cell clusters continues to increase . Between P3 and P4 , the clusters combine with each other to form a continuous flattened chain of cells ( Figure 11 ) . Although this chain lacks any tubular morphology , we have named it the rudimentary SC ( rSC ) because it encircles the entire limbus . By P3 . 5 to P4 , the number of cells in the rSC increases and it becomes more branched within its original tissue plane ( Figure 11 ) . This process resembles vasculogenesis in that cells divide and form branched chains prior to vessel formation . The rSC is still connected to the LVP and RV . As shown above , the adult SC expresses Prox1 . Because Prox1 is a potent developmental regulator , we have monitored its expression at all stages of SC development using the Prox1-GFP mice . During lymphatic development , one of the earliest detectable events is the activation of Prox1 expression . Unlike lymphatic development , Prox1 is not expressed at the earliest stages of SC development . The first detectable Prox1 expression occurs regionally in the rSC at late P4 to early P5 . Prox1-expressing regions are reorganized into flattened tube-shaped vessels ( Figure 12 ) . The level of Prox1 expression increases with an increase in tubular morphology ( compare region between arrowheads and arrows , Figure 12 ) . At this stage , most of the tubes are flattened , but regionally they are filled with RBCs , reflecting continued connection to blood vessels ( Figure 12 ) . Thus , expression of Prox1 correlates with the transition from a flattened chain of sprouting cells to a tube . Importantly , PROX1 expression remains lower than in lymphatics and is not easily detectable by immunostaining . As in the adult SC and discussed above , LYVE1 expression is absent in the developing SC . The developing SC increases in girth with obvious growth until P17 and with subtler increases afterwards . The continued development and growth of SC involves sprouting from the rSC . Sprouts from rSC are first detected at P4 . 5 . By P5 , the PROX1-expressing ( PROX1+ ) tubular regions have remodeled into a flattened core running centrally along the developing SC ( Figure 13 ) . By this stage , the flattened core has no detectable internal space and no longer contains RBCs . Sprouting continues and sprouts appear to loop back and connect to the main vessel or other sprouts . These sprouts do not end with an obvious tip cell but have long filopodia that are not restricted to their leading cell . Filopodia are also obviously evident on cells along the main body of the developing SC ( Figure S13 ) . This is different to both classic angiogenic sprouting and the initial sprouting into the IZ during SC development where filopodia are restricted to tip cells . The filopodia appear to be important for adhesion of sprouts to each other and to the main trunk of the developing SC . These sprouts lack detectable PROX1 expression . By P9 no sprouts are discernable . Cell proliferation , detected using the proliferation marker Ki67 [45] , is also important to the growth of SC and is readily detected at assessed ages from P3 to P7 ( Figure 14 ) . Cell proliferation is not restricted to the sprouting regions . By P9 almost all cells express PROX1 . At P10 , lumen formation and differentiation to a mature cellular architecture is occurring at some locations , with loss of PROX1 expression beginning to be evident in outer wall cells ( Figure 15 ) . SC has a largely mature appearance by P17 , but minor remodeling occurs later . To begin to define the molecules necessary for SC development , we tested the functional importance of the tyrosine kinase receptor KDR ( alias VEGFR2 , a receptor for vascular endothelial growth factor , or VEGF ) . VEGF is an important signaling protein in both vasculogenesis and angiogenesis [46] . VEGF interacts with KDR on the surface of tip cells to elicit the angiogenic program [47] . To inhibit KDR function , we injected a specific rat-derived inhibitory antibody called DC101 that binds KDR and competitively inhibits VEGF signaling [48] . This antibody is a valuable inhibitory tool not only because of its proven specificity but also because homozygous genetic disruption of KDR results in embryonic lethality [40] , whereas heterozygous disruption of KDR has no effect on SC development ( see above and Figure 2 ) . We injected DC101 and control antibodies into heterozygous Kdr-GFP mice . DC101 injection into these mice resulted in an almost complete absence of SC . However , this result is difficult to interpret because there was also serious disruption of the LVP and RV ( not shown ) . Given this drastic effect on the LVP and RV , it may not be possible to use homozygous conditional gene knockouts of KDR without significant confounding effects on the vasculature from which SC arises . To test the effects of KDR inhibition without these confounding complications , we tested two doses of DC101 in mice that have no mutations in KDR . DC101 was first injected into Tie2-GFP mice . KDR inhibition from P0 to P6 clearly disrupted SC development , with the higher of two doses of DC101 having the most profound effect ( Figure 16 ) . Injection of a nonspecific control antibody had no effect on SC . We next injected DC101 into C57BL/6J mice from P0 to P12 . Again , KDR inhibition substantially impeded SC development , whereas the control antibody had no effect ( Figure 17 ) . Importantly these doses of DC101 had no discernible effect on the LVP ( Figure S14 ) and RV ( not shown ) . Injection of DC101 after P5 had no obvious effect on SC development , indicating that the critical window of KDR signaling is at earlier ages . These results clearly demonstrate that KDR signaling plays a critical role in the early development of SC .
SC is an important and highly specialized vessel but is poorly understood . Here we present new tools that will facilitate future studies of SC , and we present valuable information about its development and molecular phenotype . Importantly we discover that SC has a unique molecular phenotype and develops by a previously unknown sequence of vascular development that we name “canalogenesis” ( Figure 18 ) . SECs have features resembling both blood and lymphatic vasculature , and its endothelial cells have molecular similarities to both cell types . For the first time , we establish that both the developing and mature SC express the lymphatic master controller PROX1 , which is likely of critical importance for inducing and maintaining key features of SC's functional specialization . We also demonstrate a critical requirement for KDR signaling early during canalogenesis . Canalogenesis has similarities to three well-characterized developmental processes of vessel formation—namely , angiogenesis , vasculogenesis , and lymphangiogenesis . Although early steps of canalogenesis have key similarities to angiogenesis , including endothelial sprouting and tip cells , later steps of canalogenesis are distinct . In angiogenesis , tip cells , at the end of endothelial sprouts , interact together resulting in anastomosis of sprouts producing patent , vascular branches . During canalogenesis , however , several tip cells interact and adhere together to form clusters of tip cells without either anastomosis or formation of a new tube . Cells in these clusters divide , producing a chain of cells prior to lumen formation . This process resembles stages of vasculogenesis , where endothelial cell clusters derived from angioblasts form a chain of cells [49] . In both canalogenesis and vasculogenesis , the chain of cells forms a tube ( a new capillary in the case of vasculogenesis ) . However , unlike vasculogenesis , PROX1 expression is activated in the endothelial cell chains during canalogenesis , and the PROX1+ cells remodel to form a tube . The branching of cellular chains that acquire a tube-like morphology also has similarities to the formation of the lymph sac during lymphangiogenesis [50] . Although the branching cells are PROX1− in canalogenesis , they are PROX1+ in lymphangiogenesis . Thus , the timing of PROX1 expression differs between these processes . Canalogenesis also differs from lymphangiogenesis in that LYVE1 is not expressed in the developing SC . Thus , canalogenesis combines elements of angiogenesis , vasculogenesis , and lymphangiogenesis but is significantly different from any one of them . At a molecular level , we have identified an early and critical requirement for KDR functions during SC development . Localization of VEGFA-164 to the developing SC indicates that this molecule is a ligand for KDR ( as it is in angiogenesis ) . Because KDR is expressed in the adult SC , VEGFA may regulate SC permeability or mediate other functional roles . This is consistent with a recent report that mice heterozygous for mutations in both Kdr and Flt1 ( Vegfr1 ) have elevated IOP [51] . Together with our study , this suggests that mutations in genes coding for proteins critical for SC development will cause high IOP and increase the risk of glaucoma . At this point , it is unclear if other VEGF receptors such as FLT1 ( VEGFR1 ) and FLT4 ( VEGFR3 ) play a role in development of the SC . Of these , FLT4 is important for lymphatic development and is a key marker of mature lymphatics . FLT4 expression is activated by PROX1 . Therefore , FLT4 may play a functional role in SC development . Importantly and supporting this , we demonstrate that FLT4 is expressed in SC with its expression mirroring the polar distribution of PROX1 ( greater in the inner wall compared to the outer wall of SC ) . Thus , the FLT4 ligands VEGFC and VEGFD may play key roles in SC development and maintenance/adult function . SC also expresses the angiopoietin receptor TIE2 . The TIE2/angiopoietin pathway is crucial in angiogenesis , lymphangiogenesis , blood vessel maturation , endothelial health , and regulating vascular homeostasis [52] . Thus , TIE2 and angiopoietins could play a role in both SC formation and regulation of SC permeability . Recently , the Caenorhebditis elegans homologue of PROX1 ( PROS-1 ) was shown to play a role in formation of the tubular excretory canal [53] . Given the concurrence of PROX1 expression with tube formation in the developing SC , it seems possible that PROX1 plays an important role in tubulogenesis in the rSC . The triggers for Prox1 expression remain unclear . Prox1 expression could be triggered by mechanical stimuli , as yet unidentified hemodynamic changes , cell morphologic changes , or cell–cell interactions in the branching rSC . Recently , both cell–cell interactions and mechanical stimulation by fluid flow have been shown to regulate the expression of Prox1 [54]–[56] . Thus , AQH outflow may be a factor maintaining PROX1 expression in the adult SC and at developmental stages when it functionally drains AQH . It does not seem possible that AQH flow initially induces PROX1 expression during P4 , however , as the trabecular anlage remains a compressed mass of cells [57] . Even if the still developing ciliary body [57]–[60] produces AQH at this stage , this cellular mass would not allow flow of this fluid to the developing SC . Although there are additional components , our findings are in line with earlier models from human and mouse tissues , where SC develops from superficial blood vessels that penetrate deep into the sclera [18] , [19] . A significant addition is that we establish a significant contribution to SC from two vascular beds , the superficial LVP and the deep RVs . In agreement with our previous study [19] , we show that lumen formation in the developing SC occurs discontinuously and initiates at P10 . Unique features of SC development are the polarization of Prox1 expression to the inner wall compared to the outer wall and the accompanying differentiation of the unique cell types in each wall . We are unaware of any other vascular-derived tubular structure developing this way . In adult SC , PROX1 is enriched in the inner wall cells and is likely important for its high degree of functional specialization . Inner wall cells are long and thin and readily deformed by pressure changes in the eye , likely reflecting specialization for mechanical sensing . The flow of AQH subjects the SECs to shear stress , especially near collector channels , which connect SC to the venous system . PROX1 is thus ideally positioned in inner wall cells to act as a regulator of AQH flow . PROX1 may control junctional permeability of the inner wall by controlling expression of VECAD as it does in lymphatic vessels [61] . PROX1 also regulates responses to fluid shear stress in LECs and is linked to mechanosensory signaling [54] . Modulation of PROX1 by pressure-dependent shear forces/strain could modulate the expression levels of downstream genes such as VECAD or alpha 9-integrin , which are produced by the inner wall cells of SC ( unpublished results ) . This mechanism would allow SC permeability or cell adhesion to be responsive to pressure changes in the eye . We also demonstrate that endomucin , a protein that plays a role in leukocyte rolling and is an early marker of vascular endothelial cells , is expressed in SC cells [62] , [63] . Other vascular cell surface molecules such as CD34 , CD31 are also expressed in SECs , and at levels similar to those in BECs . Along with other cell surface proteins , these molecules may modulate the behavior of immune cells that interact with SECs as they exit the eye . Thus , these molecules may modulate immune tolerance associated with the eye . Immune cells have been implicated in glaucoma and may mediate early neural damage [64]–[66] . It is possible that in addition to its influence on IOP , SC may influence glaucoma by modulating immunity . In summary , this study provides novel insights into SC , clearly demonstrating that it is a unique vessel with a combination of blood vascular and lymphatic phenotypes , which develops by a previously unknown sequence of vascular developmental events . Overall , it provides new molecular insights and new tools that will greatly facilitate our understanding of the complex functions of this important canal .
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 . All of the animals were handled according to approved institutional animal care and use committee ( IACUC ) protocols ( #99108 ) of The Jackson Laboratory . In addition , all experiments were conducted in accordance with the Association for Research in Vision and Ophthalmology's statement on the use of animals in ophthalmic research . All mice were housed in a 14-h light to 10-h dark cycle under previously described conditions [67] . The Jackson Laboratory's Institutional Animal Care and Use Committee approved all procedures described here . Prox1-GFP BAC transgenic mouse sperm ( Tg ( Prox1-EGFP ) KY221Gsat/Mmcd , cryo-archived ) [32] , [68] was purchased from the Mutant Mouse Regional Resource Centers ( MMRRC , UC Davis ) . A colony of these mice was established in our laboratory following rederivation by in vitro fertilization of C57BL/6J oocytes . The following mice were obtained from the Jackson Laboratory repository: Stock TgN ( Tie2GFP ) 287Sato/J ( JAX stock no . 003858 ) [69] , Stock KDR tm2 . 1Jrt/J ( JAX stock no . 017006 ) [39] , B6;129P2-Lyve1tm1 . 1 ( EGFP/cre ) Cys/J ( JAX stock no . 012601 ) [34] , B6;129P2 ( Cg ) - Gt ( ROSA ) 26Sortm4 ( ACTB-tdTomato , -EGFP ) Luo/J ( ROSA mtmG , JAX stock no . 07676 ) [35] , and STOCK Tg ( Wnt1-cre ) 11Rth ZTg ( Wnt1-GAL4 ) 11Rth/J [70] ( JAX stock no . 003829 ) . The eyes from B6;129S4-EfnB2tm2Sor/J mice [71] were obtained from Dr . Taija Makinen , Upssala University . Pups were dated based on visualization of vaginal plugs in females in timed pregnancy crosses and also by closely monitoring the cages thereafter for newly born pups . In all , at least 20 eyes were analyzed for each development stage . As image collection is very time-intensive , we did not collect images from all eyes . Images of either the entire limbus or partial limbal regions were collected for at least 8–10 eyes for each developmental stage . Although the mice were of different strain backgrounds , the details and developmental timing of SC formation were not different between backgrounds . The strain backgrounds were: C57BL/6J ( B6 ) , ICR/HaJ , FVB/N , and FVB/N X B6 . Adult SC was analyzed in >20 wild-type eyes and approximately 10 eyes with each GFP reporter gene genotype . For lineage tracing , six eyes were analyzed for both the Lyve1-Cre and Wnt1-Cre experiments , respectively . Enucleated adult eyes or heads of postnatal mice were fixed overnight at 4°C in 4% paraformaldehyde ( PFA , Electron Microscopy Science , Hatfield , PA ) prepared in phosphate buffered saline ( 1× PBS , 137 mM NaCl , 10 mM phosphate , 2 . 7 mM KCl , pH 7 . 4 ) . The postnatal eyes were dissected out following fixation . The anterior part of the eye was cut just posterior to the limbus ( see Figure S1 ) , and the iris , lens , ciliary body , and thin strip of retina were carefully removed to obtain the anterior eye-cup . The anterior cup includes the cornea , limbus , and small potion of retinal pigmented epithelium . Four centripetal cuts were made to relax the eye-cup and facilitate eventual mounting onto a slide after immunofluorescence . Microscopy was performed using an LSM SP5 or SP8 confocal microscope ( Leica ) using either a 20×0 . 7 NA multi-immersion objective or 63×1 . 4 NA glycerol immersion objective . The Mark and Find mode was used to automate collection of images encompassing the entire limbus and generated a folder full of Z stacks at various individual overlapping positions along the limbus . The SP8 confocal was also used when specialized settings such as the highly sensitive photon counting mode was required . For inhibition of KDR signaling in vivo , the anti-KDR function-blocking rat monoclonal antibody DC101 ( 6 . 33 mg/ml , BioXCell , West Lebannon , NH ) [48] or control rat IgG antibodies ( Jackson ImmunoResearch ) was injected subcutaneously as previously described [75] . The antibody was injected for a dose of either 25 or 50 mg/kg using a 33-gauge needle and Hamilton syringe . For most experiments , the antibody was injected daily at approximately 9 AM during the following periods: P0–P5 , P0–P11 , or P6–P11 . For P0–P11 injections , 20 wild-type eyes were analyzed and complete or partial image sets of the limbus were collected for 6–8 eyes for each treatment and time point . Six wild-type eyes were analyzed for each of the P0–P5 and P6–P11 injections . For the Kdr-GFP heterozygotes , 12 eyes were analyzed up to P12 . | Schlemm's canal serves as a drainage tube for fluid from the anterior chamber of the eye and is directly relevant to glaucoma , a disease that causes vision loss in over 70 million people . Aqueous humor enters the canal and then drains into connected veins . Molecular understanding of the development of Schlemm's canal and its drainage functions has remained limited . We provide a detailed characterization of Schlemm's canal development , and in so doing discover a novel process of vascular development that we name “canalogenesis . ” We show that although the process requires a functional KDR receptor , which is also critical in blood vessel development , the endothelial cells of Schlemm's canal have a unique hybrid molecular phenotype , expressing proteins that are characteristic of both blood and lymphatic vessels . Of note , the expression of Prox1 , a master regulator of lymphatic fate , and other lymphatic proteins are largely restricted to specialized cells of the inner wall of Schlemm's canal through which the aqueous humor passes as it exits the eye . Thus , Prox1 and other lymphatic proteins may be critical for the functional specialization of these cells for aqueous humor drainage . Schlemm's canal is thus a unique vessel with a combination of blood vascular and lymphatic characteristics . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"medicine",
"and",
"health",
"sciences",
"cell",
"biology",
"physiology",
"ophthalmology",
"biology",
"and",
"life",
"sciences"
] | 2014 | Schlemm's Canal Is a Unique Vessel with a Combination of Blood Vascular and Lymphatic Phenotypes that Forms by a Novel Developmental Process |
The glyoxylate bypass allows Escherichia coli to grow on carbon sources with only two carbons by bypassing the loss of carbons as CO2 in the tricarboxylic acid cycle . The flux toward this bypass is regulated by the phosphorylation of the enzyme isocitrate dehydrogenase ( IDH ) by a bifunctional kinase–phosphatase called IDHKP . In this system , IDH activity has been found to be remarkably robust with respect to wide variations in the total IDH protein concentration . Here , we examine possible mechanisms to explain this robustness . Explanations in which IDHKP works simultaneously as a first-order kinase and as a zero-order phosphatase with a single IDH binding site are found to be inconsistent with robustness . Instead , we suggest a robust mechanism where both substrates bind the bifunctional enzyme to form a ternary complex .
Robustness in biological systems has seen a renewal of research interest in recent years [1]–[12] . To define robustness , one needs to specify what feature is robust and with respect to which variations . Classic experimental studies have shown that metabolic fluxes are often insensitive to the levels of enzymes in the pathway , as reviewed in [13] . Metabolic control theory addresses this by suggesting that control of flux is distributed amongst many enzymes , and thus no single enzyme is rate limiting . In the last decade , studies have added a new level of understanding on robustness by providing detailed molecular mechanisms that can preserve the essential function of a system in the face of large variations in the protein levels . For example , specific mechanisms explain how exact adaptation in bacterial chemotaxis is robust with respect to chemotaxis protein levels [2] , [3] , and how patterning in drosophila embryos is robust with respect to morphogen production rates [12] , [14] , [15] . A recent review summarizes experiments and theoretical mechanisms for robustness [10] . Recently , an intriguing class of robust mechanisms has been found , based on bifunctional enzymes that carry out two opposing reactions ( such as both modifying a target protein , and removing the modification ) [8] , [11] . These robust mechanisms seem to apply to a class of bacterial two-component signaling system . These systems show robustness of input-output relations , in the sense that output responds to input signals in a way that is not disrupted by variations in protein levels . Here , we extend this line of research to one of the best studied regulation steps in E . coli metabolism , the IDHKP/IDH system . This system raised our interest because it employs a bifunctional enzyme that carries out two opposing reactions , hinting at a robust mechanism . However , it has several biochemical differences from previously studied systems [8] , [11] , suggesting that it may show a new type of robust mechanism . The need for precise regulation in the IDHKP/IDH system is evident from its biological function . The IDH system regulates the partitioning of carbon flux between the TCA cycle and the glyoxylate bypass ( Figure 1 ) . Precise regulation of flux to the glyoxylate bypass is essential when the bacterium grows on substances such as acetate that contain only two carbon atoms . Without the glyoxylate bypass , both carbon atoms would be converted to CO2 by the TCA cycle , thereby leaving no carbon available for biosynthesis of cell constituents . Hence , growth on acetate and other two-carbon compounds requires directing some of the carbon flux to the glyoxylate bypass , thereby avoiding carbon loss . The precise partitioning of carbon flux between the cycle and the bypass is achieved by regulating the activity of the enzyme IDH ( isocitrate dehydrogenase ) , which stands at the entry to the bypass . The activity of IDH is determined by its phosphorylation state: only unphosphorylated IDH is active . During growth on substances with more than two carbon atoms , IDH is mostly unphosphorylated and hence active . Thus , most of the carbon flux is directed to the more efficient TCA cycle . On the contrary , during growth on acetate , most of IDH is phosphorylated and hence inactive , so that a large part of the carbon flux is directed to the bypass [16]–[19] . To regulate the IDH phosphorylation level , E . coli employs a bifunctional enzyme . This enzyme catalyzes both the phosphorylation of IDH , and its dephosphorylation , and is called IDHKP ( IDH Kinase/Phosphatase ) [20] . IDHKP uses ATP as the phosphoryl donor for the kinase reaction , and also requires ATP as a cofactor for the dephosphorylation reaction [20]–[22] . The activity of IDHKP is allosterically regulated by the levels of various metabolites in the cell that act as the input signals to this system [21] . The robustness of IDH activity has been experimentally tested by Laporte et . al . [23] . It was found that during growth on acetate , the concentration of active ( unphosphorylated ) IDH is extremely robust: The level of active IDH changes by less than 20% upon 15-fold variation in total IDH concentration . What is the mechanism for this robustness ? It was suggested in [23] that the robustness of active IDH levels may result either from regulation of the activity of IDHKP by putative modulators sensitive to the metabolic state of the cell , or by a specific mechanism whereby the enzyme IDHKP works simultaneously as a first-order kinase and as a zero-order phosphatase . We quote from [23]: The simplest model corresponding to the argument above is as follows: ( 1 ) I denotes active IDH , Ip denotes phosphorylated IDH , and E denotes the bifunctional enzyme IDHKP . The arrows in ( 1 ) do not denote full chemical reactions . Rather , they symbolize enzyme catalysis steps . The behavior of the model depends on the details of the actual chemical reactions involved . The simplest mass-action kinetic system corresponding to ( 1 ) is ( 2 ) This system assumes a single binding site ( shared by I and Ip ) on the bifunctional enzyme E . An intuitive analysis of ( 1 ) would involve assigning , in the usual way , Michaelis-Menten rate functions f1 and f2 to the “reactions” and , respectively: ( 3 ) where ( 4 ) ( 5 ) and [E]T is the time-conserved total concentration of E: ( 6 ) Note that this analysis ignores the possibility of I and Ip competing for the active site of E . Subsequently , assume that the rate constants in ( 2 ) are such that E works as a first-order kinase and a zero-order phosphatase [18] , [24]–[27]: that is , ( 7 ) Then , using ( 7 ) in ( 3 ) gives ( 8 ) At steady-state , the rates of enzyme-catalyzed phosphorylation and dephosphorylation are equal: ( 9 ) Using ( 4 ) , ( 5 ) , and ( 8 ) in ( 9 ) yields ( 10 ) Inspection of ( 10 ) shows that , under the assumptions made , [I] is insensitive to changes in [I]T . Thus , the robustness of [I] with respect to [I]T is apparently explained . In the present study we show that the full mass-action kinetic model ( 2 ) cannot give rise to equation ( 10 ) , regardless of the choice of parameters . In fact , we demonstrate that for all parameter choices , the ratio [I]/[Ip] , not [I] , is robust at steady state . Thus , ( 2 ) cannot account for the experimentally observed robustness of [I] . From this it follows that the use of Michaelis-Menten rate functions ( 3 ) to derive ( 10 ) is inconsistent with the “parent” mass-action model ( 2 ) , due to competition of I and Ip for the active site of E . We then propose mass-action models that explain how robustness might arise in the IDHKP/IDH system . A common feature of these models is the formation of a ternary complex between I , Ip , and E .
Our goal , in this section , is to show that mass-action system ( 2 ) cannot give rise to equation ( 10 ) , and to robustness . This , once demonstrated , implies that there is a flaw in the derivation of equation ( 10 ) , namely , in the assumption that the Michaelis-Menten approximation without competition applies . We begin by writing the differential equations corresponding to mass-action system ( 2 ) . ( 11 ) Equations ( 11 ) are consistent with ( 6 ) , as well as with the conservation of total I: ( 12 ) From the second and fifth equations in ( 11 ) we have that ( 13 ) Considering the last two equations of ( 11 ) and equation ( 13 ) at steady-state we obtain the ratio between the active and the inactive forms of I: ( 14 ) where ( 15 ) Equation ( 14 ) shows that system ( 2 ) implies that the ratio [I]/[Ip] is robust: [I]/[Ip] = a−1 . Robustness of [I]/[Ip] obtains because it does not depend on protein levels , only on rate constants . This happens regardless of the choice of parameters in the system . Moreover , if we assume that enzyme E is rare compared to its substrate , ( 16 ) we can approximate ( 17 ) Using ( 17 ) in ( 14 ) we find that the unphosphorylated ( and thus active ) form I depends on the total I level: ( 18 ) In other words , not only system ( 2 ) fails to show robustness of I activity in the face of variations in the total I protein level , but also the dependence of [I] on [I]T at steady-state is linear . More generally , the inconsistency between ( 10 ) and ( 18 ) implies that the Michaelis-Menten approximation , which applies to each phosphorylation and dephosphorylation reaction alone , cannot apply when both reactions are simultaneously catalyzed by the same bifunctional enzyme with a single site for which I and Ip compete . Our aim , in this section , is to construct a mass-action model that can explain how a high degree of robustness of [I] with respect to variations in [I]T can be achieved . Following Goldbeter and Koshland [26] , we will view phosphorylation and dephosphorylation as irreversible modifications , and will not explicitly account for ATP , ADP and phosphoryl ions . This allows a clear understanding of the model . The model begins with the bifunctional enzyme E that phosphorylates I and dephosphorylates Ip , as described in the reactions in ( 2 ) . To obtain robustness requires several additional assumptions . Most importantly , we need to suppose that E has two distinct binding sites: one for I and one for Ip . This is suggested by the fact that mutant E . coli strains have been isolated where E has greatly reduced phosphatase activity but retains the kinase activity [22] . Kinetic studies on these mutants show that they have a 40-fold reduction in their affinity to Ip , whereas their affinity to I remains virtually the same as in the wild-type [22] . In addition , we assume that the ternary complex EIpI can form and has kinase activity . In our initial analysis , we shall also assume that the ternary complex EIpI has only kinase activity , and that the ternary complex forms in an ordered fashion , that is , first E binds Ip and then EIp binds I ( both assumptions will later be relaxed . ) Thus , we propose the following mass-action model: ( 19 ) The differential equations corresponding to the mass-action reactions of ( 19 ) are ( 20 ) Summing equations in ( 20 ) shows conservation over time of the total I protein level , ( 21 ) and total E protein level , ( 22 ) As before , we will consider the physiologically relevant case where the substrate I is much more abundant than the enzyme E and thus ( 23 ) Using ( 21 ) and ( 23 ) we see that equation ( 17 ) is valid in the present case . ( Note that here , as in the case of system ( 2 ) , the use of ( 23 ) is required for deriving the approximate conservation law ( 17 ) , and not for ensuring quasi steady-state . ) By summing the second , fifth and sixth equations in ( 20 ) we find that ( 24 ) We now consider the last three equations in ( 20 ) and equation ( 24 ) at steady state . This gives a balance of phosphorylation and dephosphorylation rates , ( 25 ) and a set of relations between the concentrations of complexes and the product of the concentrations of their constituent elements: ( 26 ) Using ( 26 ) and ( 17 ) in ( 25 ) we obtain a quadratic equation for the steady-state value of [I]: ( 27 ) where the parameters b and c are functions of the rate constants of the system: ( 28 ) Note for later use that ( 29 ) From ( 27 ) and ( 29 ) we see that there is a unique solution for the steady-state value of [I] , which satisfies the requirement ( 30 ) Inspection of ( 30 ) shows that [I] is robust with respect to changes in [E]T , because [E]T does not appear in the equation for [I] . In general , [I] depends on [I]T , but robustness results when ( 31 ) which implies , by ( 29 ) , that ( 32 ) As a result one obtains from neglecting b with respect to [I]T in ( 30 ) , and then Taylor-expanding the resulting expression with respect to the small parameter That ( 33 ) This shows that for large values of [I]T compared to b and c , [I] ( and thus I activity ) is highly robust with respect to variations in the total I level . What happens if we relax the assumptions that the ternary complex can form only in the ordered fashion of ( 19 ) and has only kinase activity ? We then need to consider the mass-action system ( 34 ) which gives rise to the ordinary differential equations ( 35 ) Here , an analytic expression for [I] as a function of [I]T is no longer obvious , even if ( 23 ) is used . Nevertheless , in cases where ( 23 ) and ( 31 ) apply , and the ternary complex has stronger kinase than phosphatase activity , numerical analysis of ( 35 ) suggests that the steady-state value of [I] is approximately robust over a large range of [I]T values ( see Methods ) . Moreover , if the ternary complex has much more kinase activity than phosphatase activity , we have that the steady-state value of [I] is well approximated by the leading term in ( 33 ) : ( 36 ) ( see Methods ) . Thus , even if the assumptions that the ternary complex must form in the ordered fashion of ( 19 ) and that the ternary complex has only kinase activity are relaxed , approximate robustness occurs over a large range of [I]T values . We note that if we maintain the assumption of ordered binding , but now with E binding I first and then EI binding Ip to form EIpI , simulations suggest that , subject to ( 23 ) , ( 31 ) and , robustness of [I] with respect to [I]T is lost , and in fact ( see Methods ) . Finally , we observe that ( 34 ) is symmetric with respect to exchanging the index “p . ” Thus , if I is exchanged with Ip , EI is exchanged with EIp , EIIp is identified with EIpI , and the rate constants are suitably relabeled , then ( 34 ) remains invariant . This implies that if the ternary complex has more phosphatase than kinase activity then Ip becomes the approximately robust species . Let us intuitively understand the origin of robustness in ( 19 ) . When [I]T is sufficiently large , that is , when , most of I is phosphorylated and found in the form Ip . Hence , E is saturated with Ip . This implies that most of the kinase activity is carried out by the abundant ternary complex EIpI , whereas the phosphatase activity is carried out only by the binary complex EIp . This situation can be approximately described by the mass-action system ( 37 ) Because at steady-state the rates of phosphorylation and dephosphorylation are equal , and because these rates are proportional to [EIpI] and [EIp] , respectively , it follows that at steady state ( 38 ) Because E is saturated with Ip and EI is neglected , the ternary complex is effectively formed in an ordered fashion , with Ip binding first and I binding second . This , through the second equation in ( 37 ) , constrains the equilibrium concentration of EIpI to be proportional to the product of the concentrations of its constituent species: that is , ( 39 ) Using ( 39 ) in ( 38 ) gives ( 40 ) which , following the cancellation of [EIp] from both sides of ( 40 ) , yields the robust result ( 41 ) This shows that [I] is independent of the total level of both proteins in the system ( [I]T and [E]T ) . The “cancellation principle” used above [11] is related to that which demonstrates how robustness [8] , [11] , [30] , [31] may arise in the EnvZ/OmpR system of E . coli . It is important to note that the activity of I is still a function of the allosteric inputs to the enzyme E , through the rate constants that determine the parameter c . One may say that equation ( 41 ) describes a robust input-output relation [11] between IDH activity ( output ) and the allosteric effectors of IDHKP ( inputs ) . This input-output relation is not affected by fluctuations in the total levels of the enzymes in the system .
This study suggests a new mechanism for the robustness in the glyoxylate bypass regulation of E . coli . The experimentally observed robustness of IDH activity in this system does not necessarily follow from intuitive arguments about first- and zero-order kinetics of the bifunctional regulator IDHKP . Rather , robustness requires specific biochemical features . These features work together to allow IDH activity to be highly insensitive to variations in the levels of the proteins in the system . While IDH activity is robust to protein levels , it is still responsive to input signals that affect rate constants . Thus the system may be said to have a robust input-output relation [11] , where IDH levels respond to input signals in a reliable way that is not disrupted by fluctuation in enzyme levels . The present mechanism for robustness relies , in addition to the known features of the system , on the assumption that the ternary complex EIpI exists . In addition , robustness in the present model requires that the ternary complex has more kinase than phosphatase activity . This role for a ternary complex in robustness adds to previous observations that relate ternary complexes to robustness and bistability [11] , [32] , [33] . The present model may explain a seemingly paradoxical aspect of the system . This effect occurs when E . coli is shifted from glycerol to acetate ( where robustness has been observed ) . Despite the fact that IDH activity decreases in acetate compared to glycerol , the total IDH protein level increases due to upregulated gene expression [19] . The present model may explain this puzzle by showing that robustness under acetate conditions requires that [I]T levels are sufficiently high . The present model is quite general: it may apply to other systems with a bifunctional enzyme that catalyzes antagonistic reactions . A possible example is the pyruvate , ortho-phosphate dikinase ( PPDK ) enzyme of plants [34] . The details of the proposed mechanism can be tested experimentally . To test for the existence of the ternary complex , one may construct two tagged versions of IDH , each with a different tag , and test if they co-immunoprecipitate only in the presence of the bifunctional enzyme IDHKP and ATP . Another experiment involves labeling in-vitro preparations of I with CFP ( cyan fluorescent protein ) and Ip with YFP ( yellow fluorescent protein ) , and adding saturating amounts of both to IDHKP and ATP . In the proposed mechanism , this should result in FRET ( fluorescent resonance energy transfer ) via the ternary complex . If the ternary complex is shown to exist , then the next step is to test whether it has more kinase than phosphatase activity . One possible way to do this is to prepare CFP-I and YFP-Ip where the phosphate is radioactive . One then adds saturating amounts of CFP-I and YFP-Ip to IDHKP and ATP where the γ-phosphate is radioactive . Then , we expect that CFP-Ip would form faster than YFP-I . This could be checked by immunoprecipitating and measuring the immunoprecipitates for radioactivity and color at several time points . Finally , the current robust model was derived using mass-action kinetics and not Michaelis-Menten approximations . For bifunctional enzymes , care must be taken to explicitly consider competitive and cooperative effects before applying Michaelis-Menten approximations , as standard Michaelis-Menten behavior will not necessarily arise from “parent” mass-action systems . Further research can aim to specify conditions where Michaelis-Menten approximations are applicable , and to define the general classes of systems that can show robust properties [35] .
We studied mass-action system ( 34 ) using numerical simulations . We considered three scenarios: ( a ) The ternary complex forms in a random order . ( b ) The ternary complex forms in an ordered fashion , with E binding Ip first and then EIp binding I . ( c ) The ternary complex forms in an ordered fashion , but now with E binding I first and then EI binding Ip . All simulations were performed using Matlab . In each iteration , we studied ( a ) , ( b ) and ( c ) in the following way: First , we chose each rate constant ( with the exception of k8 ) in mass-action system ( 34 ) from a lognormal distribution with ( natural ) log mean equal to 0 and ( natural ) log standard deviation equal to 1 . To ensure that , we chose k8 randomly from the interval [0 . 1k6 , 0 . 9k6] . The parameters b and c were calculated according to ( 28 ) . To ensure that ( 23 ) is met , the conserved total enzyme concentration [E]T was chosen randomly from the interval [0 . 1c , c] , and [I]T was assigned the values r1 = 1000b , 1100b , … , 2000b = r2 . For each value of [I]T , we chose the initial conditions in the standard way , with E ( 0 ) = [E]T , I ( 0 ) = [I]T , and the initial concentrations of the remaining chemical species set to 0 . The differential equations ( 35 ) , which correspond to scenario ( a ) of random binding , were then integrated for each value of [I]T using the “ode23s” differential equation solver . The corresponding steady-state values of [I] were extracted . To analyze case ( b ) , we repeated the exact same procedures as in ( a ) , but with k7 and k−7 set equal to 0 . Similarly , to analyze case ( c ) , we repeated the procedures in ( a ) , but now with k5 and k-5 set equal to 0 . We performed a total of 10 , 000 simulation runs for each of the three scenarios . In scenario ( a ) ( random binding ) , we find that over the range [r1 , r2] the steady-state value of [I] is well approximated by a linear function of [I]T: T . The goodness of fit as measured by R2 was greater than 0 . 95 for 9 , 987 of the 10 , 000 iterations . For the 13 cases where R2 was less than 0 . 95 , we repeated the simulations with [I]T in the range [10r1 , 10r2] . R2 exceeded 0 . 98 in all 13 cases . For each choice of parameters , the fractional change in the steady-state value of [I] with respect to [I]T was calculated as follows: ( M1 ) Thus , measures the percent change in the steady-state value of [I] as a result of doubling [I]T . ( corresponds to perfect robustness of [I] with respect to [I]T . ) We find that in over 95% of the simulations , In over 99 . 7% of the simulations . For all cases where was found to exceed 0 . 1 , we repeated the simulations with [I]T in the range [10r1 , 10r2] . In all cases , the approximate linear dependence of [I] on [I]T was maintained , and was now less than 0 . 1 . In 24 out of 25 cases We therefore conclude that in a large range of [I]T values , the steady-state value of [I] is approximately robust with respect to [I]T . Next , we focused on the case where the ternary complex has much more kinase than phosphatase activity To study the limiting value of [I] as [I]T grows large , we performed 1 , 000 simulations as in scenario ( a ) above , but with . For each simulation , we evaluated the mean deviation of the steady-state value of [I] from the value c , as predicted by the leading term in ( 33 ) . The deviation was calculated by the formula ( M2 ) where <[I]> is the mean of the steady-state value of [I] over the range [r1 , r2] of [I]T values . We found that in 983 out of 1000 simulations δ was less than 0 . 01 . For the 17 cases where δ exceeded 0 . 01 , we repeated the simulations with [I]T in the range [10r1 , 10r2] . In all 17 cases δ was now less than 0 . 01 . We therefore conclude that , in a large range of [I]T values , [I] is approximately equal to c , provided that the ternary complex has much more kinase than phosphatase activity . We note that repeating the simulations of scenario ( a ) , but with the ternary complex having more phosphatase than kinase activity , causes the robustness of [I] to be lost . In scenario ( b ) ( ordered binding with E binding Ip first and then EIp binding I ) , we find that over the range [r1 , r2] , [I] is approximately a linear function of [I]T . In all 10 , 000 cases , R2 exceeded 0 . 95 , and in all cases was less than 0 . 012 . We therefore conclude that approximate robustness obtains in scenario ( b ) . In scenario ( c ) ( ordered binding with E binding I first and then EI binding Ip ) , we find that over the range [r1 , r2] the steady-state value of [I] is a linear function of [I]T to very good approximation: In every case , R2 was greater than 0 . 998 . In all cases , was found to be in the range [0 . 69 , 1 . 58] , and in 9 , 998 of the 10 , 000 cases , was found to be in the range [0 . 9 , 0 . 1] . We therefore conclude that in case ( c ) robustness of the steady-state value of [I] with respect to [I]T is lost . Moreover , the fact that in the vast majority of cases was approximately equal to 1 indicates that [I] is roughly proportional to [I]T . In summary , we conclude that over the range of parameters tested , robustness of [I] requires that the ternary complex EIpI be assembled either in a random fashion or sequentially , with E binding Ip first and then EIp binding I , and that the ternary complex's kinase activity exceed its phosphatase activity . This is summarized in Table 1 . | To grow well , the cell needs to produce a balanced set of building blocks by means of its metabolic network . Regulatory circuits are used to maintain appropriate fluxes as metabolites flow through the branching pathways in the network . Here , we asked how such regulatory circuits can work precisely , despite the fact that they are made of proteins whose levels vary from cell to cell and in the same cell over time . We used a well-studied circuit , at a key branch point called the glyoxylate bypass , as a model system . Previous experiments showed that this system is remarkably robust to changes in the levels of its proteins . Here , we propose a mechanism to explain this robustness , based on a bifunctional enzyme that catalyzes two opposing reactions . We show that a simple explanation based on enzyme saturation is inconsistent with more rigorous mathematical analysis . Our proposed mechanism suggests several experimentally testable predictions . It shows how a systems-level feature ( robustness ) may arise from seemingly unrelated biochemical details . Because analogous designs with bifunctional enzymes are found in other systems in different organisms , the present mechanism might apply more broadly . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"computational",
"biology/systems",
"biology",
"physics/interdisciplinary",
"physics"
] | 2009 | Robustness in Glyoxylate Bypass Regulation |
The molecular details of Chlamydia trachomatis binding , entry , and spread are incompletely understood , but heparan sulfate proteoglycans ( HSPGs ) play a role in the initial binding steps . As cell surface HSPGs facilitate the interactions of many growth factors with their receptors , we investigated the role of HSPG-dependent growth factors in C . trachomatis infection . Here , we report a novel finding that Fibroblast Growth Factor 2 ( FGF2 ) is necessary and sufficient to enhance C . trachomatis binding to host cells in an HSPG-dependent manner . FGF2 binds directly to elementary bodies ( EBs ) where it may function as a bridging molecule to facilitate interactions of EBs with the FGF receptor ( FGFR ) on the cell surface . Upon EB binding , FGFR is activated locally and contributes to bacterial uptake into non-phagocytic cells . We further show that C . trachomatis infection stimulates fgf2 transcription and enhances production and release of FGF2 through a pathway that requires bacterial protein synthesis and activation of the Erk1/2 signaling pathway but that is independent of FGFR activation . Intracellular replication of the bacteria results in host proteosome-mediated degradation of the high molecular weight ( HMW ) isoforms of FGF2 and increased amounts of the low molecular weight ( LMW ) isoforms , which are released upon host cell death . Finally , we demonstrate the in vivo relevance of these findings by showing that conditioned medium from C . trachomatis infected cells is enriched for LMW FGF2 , accounting for its ability to enhance C . trachomatis infectivity in additional rounds of infection . Together , these results demonstrate that C . trachomatis utilizes multiple mechanisms to co-opt the host cell FGF2 pathway to enhance bacterial infection and spread .
Chlamydia trachomatis , an obligate intracellular bacterium , is the most common bacterial cause of sexually transmitted diseases and non-congenital infertility in Western countries and the leading cause of acquired blindness in developing countries ( reviewed in [1] ) . C . trachomatis serovars A–C cause eye disease , serovars D–K cause genital tract infections , and serovars L1–L3 are associated with lymphogranuloma venereum ( LGV ) , a more invasive genital tract disease . C . pneumoniae infections result in upper and lower respiratory tract infections and have been linked to a growing number of chronic diseases , including atherosclerosis , multiple sclerosis , and Alzheimer's disease . The capacity of Chlamydiae to lead to infertility and blindness , their association with chronic diseases , and the extraordinary prevalence and array of these infections make them public concerns of primary importance . All Chlamydia species share a dimorphic life cycle in which they alternate between an extracellular , spore-like form , the elementary body ( EB ) , and an intracellular , metabolically active but non-infectious form , the reticulate body ( RB ) [2] . Chlamydiae can productively infect most cultured cells , suggesting that the receptor ( s ) is widespread and/or that there are multiple receptors . Although neither the bacterial ligand nor the host receptor ( s ) have been definitively identified , it is thought that binding is a two step process that involves an initial reversible interaction between the EB and the host cell followed by high affinity irreversible binding to a secondary receptor ( reviewed in [3] ) . After binding , Chlamydiae induce their uptake into non-phagocytic cells through small-GTPase dependent reorganization of the actin cytoskeleton to form microvillus-like structures ( reviewed in [4] ) . Both bacterial and host factors are implicated in this process . These include ( i ) translocated actin recruiting phosphoprotein ( TARP ) , a chlamydial protein secreted into the host cell cytosol upon bacterial binding , ( ii ) the host tyrosine kinases platelet-derived growth factor receptor ( PDGFR ) -β and Abl kinase , and ( iii ) other host actin cytoskeleton regulatory proteins that are recruited to TARP and/or to PDGFR [5] , [6] , [7] . Following entry , EBs are sequestered within a membrane-bound compartment , termed a vacuole or inclusion , which quickly dissociates from the endo-lysosomal pathway and avoids fusion with phagosomes ( reviewed in [8] ) . Subsequently , EBs differentiate into RBs , which replicate by binary fission within the enlarging inclusion over a 12–72 hr time period . In response to an as yet unknown signal , RBs redifferentiate into EBs and are released from the host cell through cell lysis or active extrusion [9] , where they can initiate secondary rounds of infection in neighboring cells . Chlamydia infection alters the transcription of many host genes , including pro-inflammatory cytokines , and regulators of apoptosis , cell differentiation , and the cytoskeleton [10] , [11] . The combination of tissue destruction and host inflammatory responses is at least partially responsible for the devastating long-term consequences of Chlamydia infection [12] . For many C . trachomatis serovars , the initial reversible binding step is thought to involve binding to host heparan sulfate proteoglycans ( HSPGs ) , as addition of excess heparin or heparan sulfate but not chondroitin sulfate , inhibits binding [13] , [14] , [15] . HSPGs are extraordinarily heterogeneous structures that are composed of a linear array of repeating dissacharides covalently linked to various core proteins and are variably sulfated [16] . A large number of growth factors , cytokines , and differentiation factors , as well as various classes of cell surface receptors , extracellular matrix proteins , and enzymes , bind to HSPGs [17] . Prominent among these is fibroblast growth factor 2 ( FGF2; also known as basic fibroblast growth factor ) , one of the 23 FGF family members . FGF2 is critical during development and can also mediate many cellular responses in adult tissues by binding to and activating the receptor tyrosine kinases FGFR1-FGFR4 [18] . FGF2 itself is heterogeneous; it is expressed as five different isoforms ( 34 , 24 , 22 . 5 , 22 , and 18 kDa isoforms ) that result from alternative translational start sites within a single mRNA . The 18 kDa isoform can be further processed into a 16 kDa protein , which has identical properties [19] . Only the 16/18 kDa isoforms are secreted , where they bind to cell surface HSPGs , functioning in a paracrine and autocrine manner to activate FGF receptor ( FGFR ) family members . In contrast , the larger FGF2 isoforms , translated from non-canonical CUG start codons , have nuclear localization signals and are found primarily in the nucleus; their functions remain incompletely elucidated [20] , [21] . Although FGF2 dimers can bind to and activate FGFR , binding of HSPGs stabilizes activated ligand-bound dimeric FGFR [22] , [23] , [24] . FGFR activation leads to recruitment and tyrosine phosphorylation of the docking proteins FGFR substrate ( FRS ) 2α and FRS2β , as well as SHC1 , followed by recruitment and activation of Grb2 , SOS , Ras , mitogen-activated protein kinase ( MAPK ) , and phosphatidyl inositol-3 kinase ( PI3K ) [25] , [26] . In this work , we examined the effect of HSPG-associated growth factors on C . trachomatis infection . We report that C . trachomatis serovar L2 EBs bind to FGF2 , which facilitates bacterial binding , and that FGF2 promotes internalization of EBs via FGFR . In addition , C . trachomatis stimulates FGF2 transcription , production , and release to facilitate additional rounds of infection . Upregulation of FGF2 is independent of FGFR activation but involves bacterial protein synthesis and activation of the extracellular signal-regulated kinase ( Erk ) 1/2 pathway . Many of these findings were also observed with C . trachomatis serovar E , whose binding was also enhanced by FGF2 in an HSPG-dependent manner . Thus , C . trachomatis utilizes multiple mechanisms to co-opt the host cell FGF2 pathway to enhance bacterial infection and spread .
Given the previously published involvement of HSPGs on initial steps in C . trachomatis binding [5] , [27] , we tested whether HSPG-dependent growth factors enhance C . trachomatis binding to cultured cervical cells . HeLa cells were serum-starved for 2 hrs and then infected with C . trachomatis serovar L2 in serum-free media ( SFM ) , in the presence of 10% fetal bovine serum ( FBS ) , or HSPG-dependent growth factors . Binding and vacuole formation were measured at 1 hr post-infection ( hpi ) and 20 hpi , respectively . Addition of 10% FBS or purified FGF2 stimulated binding and vacuole formation ∼2–3 fold ( Figs 1A–C ) . In contrast , neither PDGF-BB , FGF-1 , FGF-10 , Vascular Endothelial Growth Factor ( VEGF ) , Epidermal Growth Factor ( EGF ) , nor heparin-binding EGF-like growth factor ( HB-EGF ) significantly enhanced EB binding or vacuole formation ( Figs 1A , B ) . FGF2 increased C . trachomatis binding in a dose-dependent , saturable manner , with half-maximal binding observed at ∼50 ng/mL of FGF2 ( Fig S1A ) . We also observed that FGF2 was sufficient to enhance C . trachomatis vacuole formation in H292 cells , a human lung epithelial cell line ( Fig S1B ) . As FGF2-FGFR interactions are stabilized through binding to HSPGs , we tested whether FGF2-stimulated C . trachomatis binding is HSPG-dependent ( Fig 1C ) . HeLa cells were treated with heparinase for 2 hrs in SFM . After washing to remove the heparinase , the cells were infected for 1 hr with C . trachomatis in the presence or absence of FGF2 or serum . Pre-treatment of cells with heparinase significantly decreased FBS- or FGF2-stimulated binding ( p<0 . 005 ) , but had no statistically significant effect on EB binding in SFM , ruling out non-specific effects ( Fig 1C ) . In control experiments , we verified that heparinase-treatment was effective , as surface staining with an anti-heparan sulfate antibody was decreased by 60% under the conditions of our experiments ( Fig S2 ) . The residual stimulatory effect of FGF2 after heparinase treatment may reflect incomplete enzymatic removal or may represent HSPG-independent FGF2 stimulation . Nonetheless , these results show that FGF2 is sufficient to enhance C . trachomatis binding in an HSPG-sensitive manner and that FGF2-mediated C . trachomatis binding leads to productive infection . As FGF2 is known to bind to both FGFR and to HSPGs on the cell surface , it is possible that FGF2 could facilitate EB binding through direct interactions with C . trachomatis and with FGFR , thereby functioning as a bridging molecule to bring EBs to the host cell surface . Alternatively , FGF2 could indirectly enhance C . trachomatis binding by upregulating a host cell receptor or stimulating its endocytosis as a consequence of FGFR activation . The first model predicts that FGF2 would co-localize with cell surface bound EBs , whereas in the second model , FGF2 binding could be spatially distinct from EB binding . We therefore examined whether FGF2 colocalizes with C . trachomatis at the cell surface . HeLa cells were serum starved and infected with C . trachomatis for 45 min in the presence of FGF2 , or as a negative control , with FGF1 , and co-localization of EBs with FGF1 or FGF2 was quantified by immunofluorescence ( IF ) microscopy . As shown in Fig 2 , ∼35% of surface bound EBs colocalized with FGF2 , whereas only ∼5% of surface bound EBs were found in association with FGF1 ( p<0 . 001 ) . We next examined whether purified growth factors could bind directly to EBs in vitro . Renograffin-purified EBs were incubated with FGF2 or FGF1 in SFM at 37°C for 1 hr , centrifuged onto coverslips , fixed , stained with antibodies to FGF1 or FGF2 , and examined by IF . In the absence of either growth factor , very little antibody to FGF1 or FGF2 bound to purified EBs ( Fig 3A and B ) . Surprisingly , addition of purified FGF2 from two different sources increased the frequency of anti-FGF2 staining of EBs from less than 10% to 50% ( Fig 3C and data not shown; p<0 . 001 ) . In contrast , only ∼5% of FGF1 associated with EBs , and this fraction was not enhanced upon addition of exogenous FGF1 ( Figs 3A and 3C ) . In control experiments , anti-FGF1 antibody recognized FGF1 bound to the cell surface ( Fig 2A ) . Co-localization of FGF2 with purified EBs was not diminished by pre-treatment of EBs with heparinase ( Fig 3D , P>0 . 05 ) . As a further test of specificity , we examined whether neutralizing antibodies against FGF2 decreased FGF2 binding to EBs . EBs were incubated with SFM containing 100 ng/mL FGF2 in the presence of the indicated neutralizing antibody for 1 hr and then examined for FGF2 binding to EBs by IF . As shown in Fig 3E , addition of FGF2 antibodies decreased FGF2 binding to EBs by approximately 2-fold compared to incubation with an isotype-matched control antibody or in the absence of antibody ( p<0 . 001 ) . Together , these results suggest that EBs can bind specifically to FGF2 to facilitate binding to the surface of host cells . We note that these experiments do not eliminate the possibility that FGF2 may also stimulate C . trachomatis binding through additional mechanisms . The ability of exogenous FGF2 to stimulate C . trachomatis binding led us to examine whether endogenous FGF2 contributed to C . trachomatis binding . Since FGF2 is expressed as 5 different isoforms that result from alternative translation initiation sites ( not from differential splicing ) , we used a short hairpin RNA ( shRNA ) directed against a region common to all isoforms . Cells were transfected with an shRNA against FGF2 or against GFP ( control ) ( Fig 4A , left panel ) . Infections were performed in SFM , as the presence of exogenous FGF2 in serum would likely mask the effects of RNAi depletion . Under conditions where total FGF2 was depleted by 60% as determined by densitometry , we observed a 40% decrease in C . trachomatis binding compared to cells treated with a control ( GFP ) shRNA ( p<0 . 01; Fig 4A , right panel ) . These results suggest that endogenous FGF2 contributes to C . trachomatis binding . Our results predict that FGF2-mediated binding of EBs to the host cell surface should result in activation of FGFR and downstream signaling pathways . FGF2 binds to and activates most isoforms of FGFR ( FGFR1 IIIb and IIIc , FGFR2 IIIc , FGFR3 IIIc , and FGFR4 ) [28] , [29] , [30] , resulting in activation of their tyrosine kinase activity , autophosphorylation of tyrosine 653/654 , and phosphorylation of the scaffolding protein , FRS2α , which is constitutively associated with FGFR . Upon tyrosine phosphorylation , FRS2α functions as a site for coordinated assembly of a multi-protein signaling complex , including Grb and Shp2 , leading to activation of the PI3K and the Ras/Erk pathways [26] . We used several approaches to determine whether FGFR is activated upon C . trachomatis binding . First , we examined by IF microscopy whether activated FGFR ( pFGFR ) co-localized with cell surface bound EBs , using a monoclonal antibody that specifically recognizes phosphorylation of the conserved tyrosines at 653/654 . IF analysis revealed that pFGFR was found with cell-associated EBs ( Fig 4B ) . We were unable to detect changes in the localization of total FGFR1 or FGFR2 ( data not shown ) , possibly because only a small fraction of total cellular FGFR is recruited and phosphorylated at the site of EB binding . Second , since activation of FGFR is required for FRS2α phosphorylation , we used IF microscopy to determine whether phosphorylated FRS2α is recruited to the site of C . trachomatis binding . Indeed , an antibody directed against phosphotyrosine 436 of FRS2α ( pFRS2α ) co-localized with cell-surface bound EBs ( Fig 4B , right panels ) . Third , we measured activation of FRS2α by immunoblotting with an antibody to pFRS2α . At 45 min pi , increased phospho-FRS2α was detected in C . trachomatis-infected HeLa cells ( Fig 4C , lower panel ) , at levels similar to what is observed at 5 min after addition of FGF2 to serum starved HeLa cells ( Fig 4C , upper panel ) . C . trachomatis-induced FRS2α phosphorylation was blocked by the FGFR inhibitor , PD173074 ( Figs 4C and 4D ) , confirming that FRS2α phosphorylation was a result of C . trachomatis-activation of FGFR . Depletion of endogenous FGF2 by shRNA reduced the percent of EBs colocalized with pFGFR from 35% to 15% , suggesting that recruitment of pFGFR to the site of C . trachomatis binding is FGF2-dependent ( Fig 4E ) . Together , these findings indicate that FGF2 mediated C . trachomatis binding is associated with the activation and recruitment of phosphorylated FGFR and FRS2α . There is evidence of cross-talk between growth factor receptors; for example , PDGFR has been shown to activate FGFR [31] . Since we recently reported that PDGFR contributes to C . trachomatis binding and internalization [5] , we examined the possible interrelationship between FGFR and PDGFR signaling pathways during C . trachomatis infection in SFM using pharmacological inhibitors . In control experiments , we established that AG1296 , a PDGFR inhibitor , blocked PDGF-dependent PDGFR activation ( data not shown ) but had no effect on FGF2 dependent activation of FRS2α ( Fig 4C , upper panel ) . Importantly , AG1296 did not inhibit C . trachomatis-induced FRS2α activation ( Figs 4C , lower panel , and 4D ) . These experiments confirm that FGFR activation during C . trachomatis infection is independent of PDGFR activation . Using PD173074 or AG1296 singly or in combination , we examined the effect of blocking FGFR or PDGFR activation on the efficiency of EB internalization . HeLa cells were pre-incubated in SFM containing PD173074 , AG1296 , or both drugs for 2 hrs , and then infected with C . trachomatis in SFM in the presence of drug ( s ) . At 1 hpi , the bacterial internalization efficiency ( the fraction of bound EBs that were internalized ) was quantified by inside-out staining as described in Materials and Methods . Treatment with either drug alone showed a small effect on internalization that failed to reach statistical significance . However , treatment with both drugs showed an additive effect , decreasing internalization by 40% ( p<0 . 001; Fig 4F ) . This result indicates that bacterial entry occurs through redundant pathways that involve activation of PDGFR and FGFR . As host microarray analyses previously revealed that C . pneumoniae induces fgf2 transcription [32] , [33] , we tested whether C . trachomatis induces fgf2 transcription . We performed qRT-PCR with fgf2 specific primers and found that C . trachomatis infection increased fgf2 mRNA expression 3-fold relative to gapdh at 12 hpi and 4–5 fold at 24 hpi ( Fig 5A ) ; in contrast , no significant increase in fgf1 expression was detected ( Fig S3 ) . Likewise , total FGF2 ( secreted and cell associated; measured by ELISA assay ) increased throughout infection ( Fig 5B ) . The increased FGF2 transcription and production was accompanied by a striking change by 12 hpi in the distribution of the FGF2 isoforms ( Figs 5C and 6A ) . The 22 kDa , 22 . 5 kDa , and 24 kDa forms became undetectable in cell lysates by immunoblot analysis , whereas the 18 kDa form and a slightly faster migrating form ( which likely corresponds to a previously described 16 kDa secreted form that has similar biological activities to the 18 kDa form [19] ) appeared to increase in the cell-associated fraction . By 18 hpi , a corresponding increase in the 16/18 kDa isoforms was detected in the culture medium ( Fig 5C , lower panel ) . Diverse stimuli induce fgf2 transcription , including FGF-mediated activation of FGFR [34] and Erk1/2 activation [35] . As C . trachomatis infection has been shown to activate Erk1/2 [36] , [37] , [38] , [39] , [40] , [41] , we tested the role of Erk1/2 in upregulation of fgf2 transcription . When the kinetics of C . trachomatis-induced Erk1/2 activation were examined by immunoblot analysis with an antibody that specifically recognizes the phosphorylated ( i . e . activated ) form of Erk1/2 , we detected an early peak of Erk1/2 phosphorylation at 45 min pi followed by a second peak of Erk1/2 phosphorylation beginning at 10 hpi ( Fig 6A ) , consistent with previously reported results [36] , [42] . We performed a more detailed time course of the induction of fgf2 mRNA and found a small but statistically significant ( p<0 . 05 ) enhancement of fgf2 transcription as early as 6–8 hpi , which further increased at 10 hpi ( Fig 6B ) . Thus , the kinetics of C . trachomatis-induced upregulation of fgf2 transcription is consistent with the involvement of Erk1/2 signaling . To determine whether Erk1/2 activation is necessary for up-regulation of fgf2 transcription in response to C . trachomatis infection , we monitored the effect of inhibiting Erk1/2 on fgf2 transcription during infection ( Fig 6C and D ) . The Erk/MAPK kinase ( MEK ) inhibitor U0126 was added for the first 3 hrs of infection ( 0–3 hpi; to prevent the first wave of Erk1/2 activation ) or for the last 4 hrs of infection ( 8–12 hpi; to prevent the second wave of Erk1/2 activation ) , and fgf2 transcription was assessed at 12 hpi . Inhibition of early Erk1/2 activation resulted in an approximately 30% reduction in fgf2 mRNA levels at 12 hpi , while blocking late Erk1/2 activation decreased fgf2 mRNA levels approximately 60% . These results suggest that both the early and late peaks of Erk1/2 activation contribute to the upregulation of fgf2 mRNA observed at 12 hpi . The reduction in fgf2 transcription by U0126 was not an indirect consequence of blocking early events in the life cycle , as short-term treatment with U0126 had no effect on C . trachomatis binding , uptake , or replication ( Figs S4A–C ) . We noted that the partial decrease in fgf2 transcription in the presence of the MAPK inhibitor ( Fig 6D ) was not immediately reflected in decreased levels of cell associated FGF2 ( Fig 6C ) . Post-transcriptional regulation could account for this observation [43] , [44] , [45] , suggesting that regulation of FGF2 is complex and subject to multiple levels of control . Since Erk1/2 activation was previously reported to be dependent upon C . trachomatis replication [36] , we examined whether bacterial protein synthesis was required for the early and/or late wave ( s ) of Erk1/2 activation . The addition of the bacterial protein synthesis inhibitor chloramphenicol ( CAM ) did not reduce Erk1/2 phosphorylation at 45 min pi , suggesting that bacterial protein synthesis is not required for the early peak of Erk activation ( Fig 6E , left panel ) . In contrast , addition of CAM to C . trachomatis-infected cells from 0–12 hpi , 3–12 hpi , or 6–12 hpi abrogated late Erk1/2 activation , as assessed at 12 hpi ( Fig 6E , right panel ) . Consistent with its role in late Erk1/2 activation , inhibition of bacterial protein synthesis partially reduced the increase in fgf2 transcription at 12 hpi ( Fig 6F ) . Although FGFR or PDGFR activation are known to activate Erk1/2 and to enhance fgf2 transcription [46] , [47] , [48] , pharmacologic inhibition of these growth factor receptors during C . trachomatis infection only partially blocked Erk1/2 activation at 45 min pi ( Fig S5A ) , did not inhibit Erk1/2 activation at 12 hpi ( data not shown ) , and had no statistically significant inhibitory effect on C . trachomatis-mediated induction of fgf2 transcription ( Fig S5B ) . Together , these results suggest that both waves of C . trachomatis-mediated Erk1/2 activation contribute to the induction of fgf2 transcription and that this transcriptional regulation is independent of FGFR receptor activation . We examined whether Erk1/2 activation , bacterial protein synthesis , and/or host protein synthesis is required for the change in FGF2 isoforms . Addition of U0126 to C . trachomatis-infected cells from 0–3 hpi or from 8–12 hpi had no effect on the change in FGF2 isoforms ( Fig 6C , lower panel ) . However , addition of CAM from 0–12 hpi or from 6–12 hpi prevented the change in FGF2 isoforms , suggesting that either de novo synthesis of a bacterial protein or bacterial growth was required for this FGF2 isoform change ( Fig 6E , right panel ) . Furthermore , the eukaryotic translation inhibitor cycloheximide ( CHX ) did not prevent the C . trachomatis-induced changes in FGF2 isoforms ( Fig 7A ) , indicating that these changes occurred at a post-translational step . The loss of the 22 kDa , 22 . 5 kDa , and 24 kDa isoforms could arise through host-mediated degradation or through a secreted chlamydial protease , such as Chlamydial Protease/proteasome-like factor ( CPAF ) , a broad spectrum protease [49] . We first tested whether recombinant CPAF could alter the FGF2 isoform distribution by incubating recombinant CPAF with uninfected HeLa cell lysates . No change in the FGF2 isoform distribution was observed ( Fig 7B ) , although CPAF was active against its known substrate vimentin ( Fig 7B ) [50] . We next determined whether host proteosomal enzymes might be involved in degradation of HMW FGF2 isoforms . Treatment of cells from 9–12 hpi with the proteosome inhibitors Lactacystin or MG132 prevented the change in FGF2 isoforms ( Fig 7C ) . Together , these results suggest that the abrupt change in FGF2 isoforms is independent of Erk1/2 activation , does not involve CPAF , and likely involves degradation by host proteosomal proteases activated during C . trachomatis intracellular replication . It is also possible that an as yet to be identified C . trachomatis protease that is secreted into the host cell cytosol and that is inhibited by Lactacystin or MG132 is responsible for the loss of the HMW FGF2 isoforms . We were particularly interested in our observation that C . trachomatis infection increases the secretion of the 16/18 kDa FGF2 isoforms , as these isoforms are important for FGFR activation and for C . trachomatis binding . These isoforms lack a canonical secretion signal , and their secretion does not involve the classical ER-Golgi secretion pathway [51] , [52] . The mechanism of FGF2 secretion is incompletely understood; there are reports that FGF2 release is associated with membrane blebs , caspase-1 activation , and/or the Na-K ATPase [53] , [54] , [55] . Recent studies reported that FGF2 can be directly translocated across the plasma membrane in a process which depends on the ability of FGF2 to bind HSPGs and a transient interaction between FGF2 and phosphatidylinositol-4 , 5-bisphosphate ( PI ( 4 , 5 ) P2 ) [21] . As C . trachomatis has been shown to induce caspase-1 activation [56] , we tested whether caspase-1 is involved in the C . trachomatis mediated release of FGF2 . C . trachomatis infected cells were stained with an antibody to p20 , one of the two cleavage products of activated caspase-1 . An increase in p20-positive cells was not detected until 18 hpi and became most marked at 24 hpi ( Fig S6A ) , consistent with previously published studies showing that C . trachomatis activation of caspase-1 is a late event in the intracellular life cycle [56] . Thus , activation of caspase-1 does not correlate with C . trachomatis-induced FGF2 release seen from 12 hpi . Likewise , neither of the caspase-1 inhibitors YVAD nor WEHD altered production ( data not shown ) or release ( Fig S6B ) of FGF2 during C . trachomatis infection . Although our studies do not rule out other active pathways of release , they suggest that the C . trachomatis-induced release of the FGF2 is likely independent of caspase-1 activation . As C . trachomatis infection ultimately leads to host cell death and lysis [2] , we considered the possibility that C . trachomatis-induced FGF2 release may be a consequence of host cell lysis . As shown in Fig 8 , the kinetics of FGF2 secretion in C . trachomatis-infected cells correlates with and is proportional to host cell lysis , as determined by LDH release . Together , these results suggest that one mechanism by which FGF2 may be released from C . trachomatis-infected cells is by passive release secondary to C . trachomatis-induced host cell lysis . Since C . trachomatis infection stimulates FGF2 production and release from the host cells , we hypothesized that the increase in released FGF2 may facilitate subsequent rounds of C . trachomatis infection . We collected filtered conditioned media ( CM ) from C . trachomatis-infected HeLa cells at 20 hpi ( CT-CM ) or from mock-infected cells ( mock-CM; see materials and methods and Fig 9A ) . The concentration of FGF2 in the CT-CM was ∼50 fold higher than mock-CM ( data not shown ) . EBs resuspended in CT-CM enhanced EB binding 3-fold compared to mock-CM ( p<0 . 001; Fig 9B ) . To determine whether FGF2 was responsible for the stimulatory effect of CT-CM , we immunodepleted FGF2 from the CT-CM using antibodies to FGF2 . EB binding in the presence of FGF2-immunodepleted CT-CM was approximately 50% less efficient compared to CT-CM that had not been immunodepleted or that had been immunodepleted with a control goat IgG antibody or with antibodies to an unrelated growth factor , EGF ( Figs 9C and 9D; P<0 . 005 ) . To test for the efficiency of immunodepletion , we measured FGF2 by quantikine assay . Compared to untreated CT-CM , immunonodepletion with anti-FGF2 decreased the concentration of FGF2 by 60% , ( data not shown ) . Together , these results suggest that C . trachomatis infection stimulates production and release of FGF2 , which can then be co-opted by C . trachomatis to facilitate additional rounds of infection and bacterial spread . As serovar E binding has been reported to be variably sensitive to heparan sulfate [27] , [57] , [58] , we investigated the role of the FGF2 pathway during epithelial cell infection . We first tested whether FGF2 can enhance serovar E binding to HeLa cells . Addition of FGF2 stimulated the binding of renograffin-purified serovar E to HeLa cells in SFM ( Fig 10A and B ) . FGF2-enhanced binding was HSPG-dependent since it was inhibited by pretreatment of cells with heparinase ( Fig 10A ) or by competitive blocking with heparin ( Fig 10B ) . Neither of these treatments significantly altered serovar E binding in the absence of exogenously added FGF2 . Although renograffin-purified serovar E bound to purified FGF2 ( Fig 10C ) , the binding was less efficient than serovar L2 ( 21% ( Fig 10C ) compared to 50% ( Fig 3D ) . Depletion of endogenous FGF2 failed to significantly reduce serovar E binding to HeLa cells in the absence of serum ( data not shown ) , suggesting that endogenous levels of FGF2 may not be sufficient to stimulate serovar E binding . Overall , the HSPG-dependence of serovar E appears qualitatively similar to that of serovar L2 , although there are some quantitative differences . We next examined whether serovar E infection induced FGF2 transcription , production , secretion , or processing . Serovar E infection of HeLa cells resulted in induction of fgf2 transcription within 9 hrs ( Fig 10D ) , biphasic activation of Erk1/2 phosphorylation ( with peaks at 45 min and at 12 hrs; Fig 10E ) , and processing of the FGF2 isoforms beginning ∼12 hpi ( Fig 10E ) . Conditioned medium harvested from serovar E infected cells at 72 hpi ( a time at which host cell lysis commences ) was enriched for the 16/18 kDa isoform of FGF2 ( Fig 10F ) and stimulated serovar E binding up to 70% compared to control conditioned medium ( Fig 10G; p<0 . 001 ) . Thus , at least two different serovars of C . trachomatis appear to be capable of co-opting the FGF2 pathway to facilitate bacterial spread .
The molecular details of Chlamydia trachomatis binding , entry , and spread are incompletely understood . HSPGs are thought to play a role in the initial binding interactions . Since cell surface HSPGs facilitate the interactions of many growth factors with their receptors , we investigated the role of HSPG-dependent growth factors in C . trachomatis infection . Here , we report the novel finding that FGF2 is necessary and sufficient to enhance C . trachomatis binding to host cells in an HSPG-dependent manner . Unexpectedly , we found that FGF2 binds directly to EBs , where it may function as a bridging molecule to facilitate interactions of EBs with FGFR on the cell surface . Upon EB binding , FGFR is activated locally and contributes to bacterial uptake into non-phagocytic cells . We show that C . trachomatis infection stimulates fgf2 transcription and enhances production and release of FGF2 through a pathway that requires bacterial protein synthesis and activation of Erk1/2 signaling but that is independent of FGFR activation . Intracellular replication of the bacteria results in host proteosome-mediated degradation of the HMW isoforms of FGF2 and increased amounts and release of the LMW isoforms . Finally , we demonstrate the in vivo relevance of these findings by showing that conditioned medium from C . trachomatis infected cells is enriched for FGF2 and that this accounts for its ability to enhance C . trachomatis infectivity in additional rounds of infection . Together , these results demonstrate that C . trachomatis utilizes multiple mechanisms to co-opt the host cell FGF2 pathway to enhance bacterial infection and spread ( Fig 11A ) . By several criteria , we found that the binding of FGF2 to EBs appears to be quite specific . We postulate that FGF2 functions as a bridging molecule , by binding simultaneously to EB surface proteins and to HSPGs and/or FGFR on the host cell ( Fig 11B ) . Co-localization of FGF2 with purified EBs was not diminished by pre-treatment of EBs with heparinase , suggesting that FGF2 binding to EBs was not mediated by HSPGs . However , EB-FGF2 binding may involve synergistic interactions with OmcB , a cysteine rich outer membrane protein found in most chlamydial species that contains a HS binding domain and mediates attachment to HSPGs [59] . We further show that a consequence of FGF2 binding to EBs is that activated FGFR and FRS2α are recruited to the site of bacterial binding , facilitating uptake . Activation of FGFR , however , is not required for the Chlamydia-induced upregulation of fgf2 transcription , production , processing , and release . In previous work , we have shown that phospho-PDGFR co-localizes with bound EBs , but other growth factor receptors , such as EGFR , are not recruited [5] , suggesting selectivity and specificity in growth factor receptor recruitment . Although PDGFR signaling has been shown to stimulate FGFR under some conditions , we did not find evidence for cross-talk in the setting of C . trachomatis-induced activation of FGFR in the absence of serum . However , using informative pharmacologic inhibitors , we found evidence that the PDGFR and FGFR pathways may function redundantly in C . trachomatis entry . Growth factor signaling may also be important at steps downstream of entry , for example by providing pro-survival signals for the host cell [60] . We found that C . trachomatis infection upregulates FGF2 transcription , production , and secretion . FGF2 transcription and production were upregulated within the first 12 hpi and continued for at least 24 hpi . This process was independent of FGFR activation , but involved biphasic activation of Erk1/2 kinases . Early Erk1/2 activation was independent of de novo bacterial protein synthesis . We speculate that the first wave of Erk1/2 activation may involve TARP , a chlamydial type III secreted protein that is present in EBs and then injected into the host cell cytoplasm upon bacterial binding . TARP has recently been shown to bind to SHC1 [42] , a Src homology-2 domain containing protein that is recruited to and phosphorylated by FGFR ( and EGFR ) upon its activation and that subsequently mediates Erk1/2 activation [61] , [62] , [63] . Thus , recruitment and activation of FGFR may facilitate or synergize with TARP and other chlamydial factors to activate Erk1/2 as well as to enhance bacterial internalization . Indeed , the failure of the FGFR inhibitor PD173074 to completely block C . trachomatis-induced Erk1/2 activation may result from the redundant involvement of chlamydial factors , such as TARP , together with the activation of FGFR . The second peak of Erk1/2 activation required active bacterial protein synthesis . This finding suggests either that a de novo synthesized chlamydial protein ( as opposed to an immediate early protein such as TARP ) is secreted from the vacuole to activate the Erk pathway , or that Erk is activated in response to vacuolar and/or bacterial intracellular growth . In any case , we conclude that the two waves of Erk1/2 activation , which occur through separate pathways , contribute to upregulation of FGF2 expression . Our work also reveals that midway through the chlamydial intracellular life cycle , there is a loss of the HMW FGF2 isoforms and a concurrent increase in the LMW isoforms ( 16/18 kDa ) . The change in the spectrum of FGF2 proteins was independent of Erk1/2 activation but required bacterial intracellular growth . We favor the idea that the HMW forms are degraded rather than processed into the 16 and/or 18 kDa form . The molecular identity of the 16 kDa isoform is currently under investigation , but may represent a previously reported pepstatin-sensitive acid proteinase cleavage product [19] . We considered several possible mechanisms for the change in FGF2 isoforms . First , the change in isoforms could result from a shift in the translation initiation sites . However , a modified pulse-chase experiment , in which we followed the isoform distribution after inhibiting host protein synthesis at 6–12 hpi with cycloheximide , demonstrated that the change in FGF2 isoforms still occurred , eliminating this possibility . Second , we tested whether the Chlamydia protease CPAF might be responsible for degrading or processing the FGF2 isoforms , but in vitro experiments using recombinant CPAF ruled out this notion . Third , and most likely , the change in FGF2 isoforms may be a consequence of C . trachomatis-induced activation of a host protease , as pretreatment with lactacystin or MG132 prevented the isoform change . In hematopoietic cells , thrombin has been reported to process the HMW FGF2 isoforms into an 18 kDa species [64] , though this process seems less likely in epithelial cells that lack thrombin . However , it is possible that an as yet-identified bacterial-encoded protease could account for the processing . Finally , we demonstrate that by enhancing secondary rounds of infection , C . trachomatis-induced up-regulation of FGF2 is physiologically important . Conditioned media from C . trachomatis-infected cells ( CT-CM ) stimulated EB binding . Two pieces of evidence provide support that FGF2 contributed to the activity of the CT-CM . First , there was an increase in FGF2 levels in CT-CM compared to CM isolated from mock-infected cells . Second , immunodepletion of FGF2 from the CT-CM decreased its ability to stimulate EB binding , whereas depletion with a control antibody or an irrelevant antibody was without effect . In addition to stimulating EB binding , we speculate that FGF2 production enhances secondary rounds of infection by its prosurvival activity . We found both similarities and differences in the HSPG-dependence and modulation of FGF2 signaling of serovar E compared to serovar L2 . As observed with L2 , serovar E binding to HeLa cells was stimulated by FGF2 in an HSPG-dependent manner but was not affected by depletion of host cell FGF2 . Serovar E bound to FGF2 in vitro , though perhaps less avidly . It is intriguing to speculate that the absence of a functional heparan sulfate binding domain in the OmcB surface protein of serovar E [59] may explain in part the decreased FGF2 binding ( Fig 11C ) . Nonetheless , serovar E stimulated transcription , production , and processing of FGF2 . Together these results suggest that serovar E activates the Erk pathway and FGF2 production similarly to serovar L2 and that it may utilize FGF2/HSPG-dependent pathway for binding . In the future , it will be interesting to determine whether FGFR signaling is activated upon serovar E binding . Modulation of growth factor expression or distribution is an emerging theme in bacterial infections . Neisseria gonorrhoeae infection induces expression , processing and release of amphiregulin , an epidermal growth factor ( EGF ) family member that is anti-apoptotic [65] . H . pylori infection stimulates HB-EGF production , which may contribute to cancer progression [66] . C . pneumoniae infection of cultured endothelial cells has been reported to increase FGF2 and PDGF production , which may be responsible for smooth muscle cell proliferation and intimal thickening in aortic tissues , and could account for its potential association with atherosclerosis [67] . In summary , our results demonstrate that C . trachomatis co-opts FGF2 to enhance infection and bacterial spread ( Fig 11A ) . Activation of the Erk1/2 pathway , either at the time of binding and entry or during subsequent intracellular growth , leads to increased fgf2 transcription and production . In addition , intracellular growth activates host protease ( s ) , resulting in alterations in the distribution of FGF2 isoforms and enhanced release of the secreted forms during host cell lysis . The released FGF2 serves as a bridging molecule to facilitate subsequent rounds of binding , entry , and intracellular development . This positive feedback loop amplifies secondary infection as well as promoting efficient bacterial spread . FGF2 may play additional roles in the pathogenesis of chlamydial infection , by potentiating the inflammatory response , by inhibiting apoptosis , or by modulating gene expression [68] , [69] , [70] . In the future , it will be interesting to determine whether FGF2 contributes to pelvic inflammatory disease and whether other human adapted chlamdyial species , such as C . pneumoniae , utilize FGF2 to enhance infection .
Recombinant human FGF1 and FGF2 were purchased from Invitrogen and Leinko Technology . FGF10 , PDGF , EGF , HB-EGF , VEGF were purchased from R&D systems . Heparin , Heparinase Cycloheximide , and MG132 ( Z-Leu-Leu-Leu-al ) were purchased from Sigma-Aldrich . PD173074 and AG1296 were purchased form Stemgent and Calbiochem , respectively . Carboxyfluorescein FLICA kit was purchased from ImmunocChemistry Technology . Caspase 1 inhibitors , YVAD and WEHD were purchased from Biovision and R&D systems , respectively . Chloramphenicol was purchased from Allstar . shRNA constructs specific for FGF2 or GFP were obtained from OriGene Technology . Quantikine kit used to measure FGF2 concentration was purchased from R&D systems . MEK inhibitor U0126 and the Cytotox 96 non-radioactive cytotoxicity assay kit used to measure LDH activity were purchased from Promega . QIAshredder , RNeasy kit , RNase-free DNase , and cDNA synthesis kit were purchased from Qiagen . SYBR GreenER qPCR SuperMix was purchased from Invitrogen . Antibodies were obtained from the following sources: mouse anti-Chlamydia FITC conjugate from Meridian Diagnostics; goat anti-C . trachomatis MOMP and rabbit anti-Chlamydia LPS from Fitzgerald; mouse anti-GAPDH from Chemicon; mouse anti-FGFR1 , mouse anti-FGFR2 , rabbit anti-phospho-FGFR ( Y653/654 ) , normal goat IgG , goat anti-FGF2 , and goat anti-EGF from R&D systems; rabbit anti-FRS2α from Santa Cruz Biotechnology; rabbit anti-phospho FRS2α ( Y436 ) , rabbit anti-Erk1/2 , and mouse anti-phospho Erk1/2 ( Y202/Y204 ) from Cell signaling technology; goat anti-human EGF from R&D systems; mouse anti-vimentin from Sigma-Aldrich; rabbit anti-caspase-1 from Biovision; HSPG-10E4 antibody from Seikagaku corp . ; HRP-rabbit anti-goat IgG from Zymed; HRP-goat anti-rabbit IgG and goat anti-mouse IgG HRP from Amersham Biosciences; all fluorescently labeled secondary antibodies and phalloidin from Molecular Probes . HeLa 229 cells and L929 cells were obtained from ATCC and passaged as previously described [71] . H292 cells were a gift from Dr . Lemjabbar-Alaoui ( UCSF ) . C . trachomatis serovar L2 ( LGV 434 ) was propagated in L929 cells grown in suspension culture and purified using a renograffin step-gradient as previously described [72] . C . trachomatis serovar E , a gift from Dr . Wyrick ( East Tennessee State University ) , was propagated and purified as described previously with the following modifications [27] . C . trachomatis serovar E was grown in semiconfluent HeLa cells for 48 hrs in MEM supplemented with 10% FBS and cycloheximide ( 2 µg/mL ) . Cell monolayers were scraped with plunger and then sonicated . Cell debris was removed by centrifugation , and chlamydiae were purified using a renograffin step-gradient as previously described [72] . The final L2 or serovar E pellet was resuspended in sucrose phosphate buffer ( SPG; 5 mM glutamine , 0 . 2 M sucrose . 0 . 2 M phosphate buffer ) and stored at −80°C . IF was carried out as previously described [5] . For each set of experiments , the exposure times were identical for all images . Images were analyzed with Metamorph ( Molecular devices ) or with Adobe Photoshop CS4 to count nuclei , EBs , or vacuoles . For all image analysis , a minimum of 8 fields was analyzed per treatment . Data were compiled from at least 3 independent experiments unless it is specified . To determine colocalization of phosphorylated FGFR or FRS2α with EBs , HeLa cells were grown on glass coverslips in 24-well plates and infected with C . trachomatis for 45 min . Cells were fixed in 4% paraformaldehyde ( PFA ) , permeabilized with 0 . 2% Triton , blocked in 1% bovine serum albumin ( BSA ) in phosphate-buffered saline ( PBS ) , and incubated with rabbit anti-phospho-FGFR or anti-phospho-FRS2α for 1 hr . Cells were washed with three times with PBS and then incubated for 1 hr with Alexa-488 conjugated secondary antibody and in some cases with fluorophore-conjugated phalloidin . HeLa cells were grown on glass coverslips in 24-well plates in MEM containing 10% FBS overnight , and were infected with C . trachomatis at an MOI of 2–3 ( for quantitation of vacuole formation ) or 10 ( for quantitation of EB binding ) in the presence or absence of FBS for 1 hr . For quantitation of binding , unbound bacteria were removed by washing three times with PBS , and the infected cells were fixed with 4% PFA for 30 min . For quantitation of vacuole formation , PBS-washed cells were incubated in fresh MEM containing 10% FBS for 20 hr . The cells were fixed and were permeabilized with 0 . 2% Triton X-100 for 15 min , followed by blocking in 2% FBS/1% Fish Skin Gelatin in PBS for 30 min . Bound EBs or vacuoles were visualized by staining with anti-Chlamydia MOMP antibody followed by staining with Alexa-488 conjugated secondary antibody . The host cell was visualized by staining the actin cytoskeleton with phalloidin-Alexa 594 . Images were acquired and analyzed as described above . Data was presented as number of cell-associated EBs per cell or number of vacuoles per cell . To analyze the effect of growth factors on C . trachomatis binding , HeLa cells were serum starved for 2 hrs and then infected with C . trachomatis in SFM in the absence or presence of growth factors ( 100 ng/mL ) . Binding and vacuole formation were measured at 1 hpi and 20 hpi , respectively . For heparinase treatment , HeLa cells grown on glass coverslips in 24-well plates were incubated with 1 unit of heparinase ( Sigma ) in 0 . 5 mL of MEM containing 0 . 1% BSA at 37°C for 2 hr . Enzyme-treated cells were washed three times with PBS , infected with C . trachomatis , and then analyzed as described above . The efficacy of heparinase treatment was examined by IF staining with 10E4 antibody , which recognizes HW N-sulfation . FGF1 or FGF2 ( 100 ng/mL ) was added to renograffin purified EBs ( 1×107 IFU ) suspended 1 mL of MEM containing BSA ( 0 . 1% ) , incubated with gentle agitation for 1 hr at 37°C , transferred to coverslips in 24-well plates , and centrifuged at 1000 rpm for 10 min . Unbound bacteria were removed by washing three times with PBS . EBs were visualized by staining with DAPI . FGF was visualized by staining with goat anti-FGF1 or anti-FGF2 antibody followed by staining with Alexa-488 conjugated secondary antibody . In some case , EBs ( 1×107 IFU ) was pretreated heparinase 1 unit in 0 . 5 mL of MEM containing 0 . 1% BSA at 37°C for 2 hr before incubating with FGF2 . To examine the specificity of the C . trachomatis-FGF2 interaction , goat IgG or goat anti-FGF2 antibody ( 2 µg/mL ) was added to the mixture of EBs-FGF2 and incubated for 1 hr rotating in 37°C . Data is presented as percentage of EB associated with FGF1 or FGF2 relative to total EB . HeLa cells grown in 6-well plates were transfected with the indicated shRNA according to manufacturer's protocol . At 48 hrs post transfection , the cells were trypsinized and reseeded onto glass coverslips in 24-well plates . At 72 hrs post transfection , cells were infected with C . trachomatis in SFM for 1 hr and then fixed . Lysates from shRNA-treated cells were immunoblotted with antibodies to FGF2 to determine the efficiency of FGF2 depletion . HeLa cells were lysed for 15 min on ice in Lysis Buffer ( 50 mM Tris HCl , pH 7 . 5 , 150 mM NaCl , 1% Triton X-100 , 1 mM EDTA , 50 mM NaF , 1% sodium deoxycholate , 0 . 1% SDS , 1 mM sodium orthovanadate , 0 . 1 mM okadaic acid , and Complete protease inhibitors ( Roche Diagnostics ) ) . Cell lysates were collected , centrifuged at 20 , 800 g for 15 min to remove cell debris , and the supernatant was boiled in NuPage 4X LDS sample buffer ( Invitrogen ) with 100 mM DTT for 10 min . Proteins in the supernatant were separated on 10% NuPAGE Novex Bis-Tris gels ( Invitrogen ) and transferred to 0 . 45 mm Trans-blot nitrocellulose membranes ( BioRad Laboratories ) . Membranes were rinsed in water , followed by blocking with 3% milk ( Upstate ) in Tris-buffered saline ( TBS ) for 1 hr . Each membrane was incubated with the indicated antibody in 3% milk in TBS with 0 . 02% Tween-20 ( TBST ) overnight at 4°C , followed by an incubation with the appropriate HRP-conjugated antibodies for 1 hr . HRP-conjugated antibodies were detected by ECL ( Amersham Biosciences ) according to the manufacturer's protocol . For quantification , band intensity was analyzed using Metamorph image analysis software ( Molecular devices , Sunnyvale , CA ) . Inside-out staining was performed as described previously [5] with the following modifications . HeLa cells grown overnight on glass coverslips in 24-well plates were infected with C . trachomatis for 1 hr at 37°C . Cells were washed three times with PBS to remove unbound bacteria and then fixed in 1% PFA for 15 min , conditions under which the host cell plasma membrane is not permeabilized . After fixation , cells were blocked in 2% FBS/1% FSG/PBS for 30 min and then incubated with goat anti-MOMP antibody for 1 hr followed by incubation with donkey anti-goat Alexa 488 antibody to stain external EBs . Cells were then permeabilized with 0 . 2% Triton X-100 for 15 min , blocked in 2% FBS/1% FSG/PBS for 30 min again , and incubated with rabbit anti-Chlamydia LPS antibody followed by incubation with goat anti-rabbit Alexa 594 antibody to stain both intracellular and extracellular EBs . The host cells were visualized by staining the actin cytoskeleton with phalloidin-Alexa 350 . All images were acquired and analyzed as described in Immunofluorescence studies . The percent efficiency of internalization was calculated as follows: ( number of total cell associated EB-number of extracellular EB ) / ( number of total cell associated EB ) ×100 . HeLa cells grown in T-75 flask at ∼100% confluency ( 8 . 4×106 cells ) were harvested and lysed in 200 µL 50 mM Tris buffer ( pH 7 . 5 ) with 1X protease inhibitory cocktail ( Roche ) and 5 µM PMSF . Cell lysates were sonicated gently and centrifuged at 10 , 600 X g for 5 min at 4°C . Supernatants were split into four samples ( ∼50 µL each ) . Purified recombinant CPAF ( a kind gift of Dr . Raphael Valdivia ) was added to three samples and incubated for 30 min at 4°C , 10 min at 37°C , or 30 min at 37°C . Incubation without CPAF for 30 min at 37°C served as a negative control . Samples were boiled with NuPage LDS sample buffer ( Invitrogen ) and immunoblotted with antibodies to FGF2 , vimentin , or GAPDH . HeLa cells grown in 24-well plate were infected with C . trachomatis for 1 hr , washed , and then incubated in 0 . 5 mL media containing 2% FBS for the indicated times . At the end of incubation , C . trachomatis-conditioned media ( CT-CM ) , control conditioned medium ( Mock-CM ) , and cell lysates were collected . The supernatants were centrifuged at 1000 rpm to remove cell debris and filtered through a 0 . 22 µm filter to remove any EBs . The mock or L2 infected cells were washed once with PBS , lysed with 0 . 5 mL lysis buffer , centrifuged for 10 min at 20 , 800 X g , and then the supernatants were collected . FGF2 concentration in CT-CM and Mock-CM were measured using the Quantikine kit according to manufacturer's protocol . Total FGF2 represents the FGF2 concentration of the cell lysate and the supernatant . The percentage of released FGF2 was calculated as released FGF2/total FGF2 . Portions of the supernatant and cell lysate were diluted and quantified for LDH activity using the Cytotox-96 kit according to manufacturer's protocol . Total LDH activity ( cell associated plus released ) or released LDH activity was calculated as described above . C . trachomatis infected HeLa cells were labeled with FAM-YVAD-fmk caspase-1 FLICA kit according to the manufacturer's protocol at 12 , 18 , or 24 hpi . Hoechst dye was used to stain the host cell nuclei during the last 5 min of incubation of FLICA reagents . Excess dye was removed by washing three times with PBS and IF images of live cells were taken immediately . The percentage of FLICA positive cells relative to the total cell number was calculated . To assess the effect of inhibition of Caspase-1 activation on FGF2 release , C . trachomatis-infected HeLa cells were incubated with DMSO , YVAD ( 100 µM ) , or WEHD ( 100 µM ) from 12 hpi to 24 hpi . The ratios of FGF2/LDH activity in the supernatant in the presence of inhibitors were measured as described above and compared to the ratio in control ( DMSO-treated ) cells . RNAs from uninfected and Chlamydia-infected HeLa cells were isolated using the QIAshredder and the RNeasy kit according to the manufacturer's instructions . RNA was treated with RNase-free DNase according to the manufacturer's instruction . RNA concentrations were measured using Nanodrop Spectrophotometer ( Thermo-scientific ) . One µg of RNA was reverse transcribed using cDNA synthesis kit in a 20 µL reaction . Quantitative PCR ( qPCR ) was performed with 2 µL of the cDNA preparation using SYBR GreenER qPCR SuperMix in a 25 µL reaction using DNA Engine Opticon-2 Real-Time PCR Detection System in the Opticon-2 Real-Time Cycler ( BioRad ) . Primers for human gapdh were 5′-CTTCTCTGATGAGGCCCAAG-3′ forward and 5′-GCAGCAAACTGGAAAGGAAG-3′ reverse . Primers for human fgf2 were 5′-CGTGCTATGAAGGAAGATGGA3′ forward and 5′-TGCCCAGTTCGTTTCAGT-3′ reverse . qPCR included initial denaturation at 94°C for 10 min , followed by 35 cycles of 94°C for 10 s , 53°C for 15 s , 72°C for 20 s , 72°C for 1 s and then 72°C for 10 min followed by a dissociation curve every 0 . 5°C from 55°C to 95°C . In a dissociation curve , a single peak was confirmed in each of the amplified sequences . For the quantification of fgf2 expression relative to gapdh in different samples , the threshold cycle ( Ct ) values of targets were expressed as 2−ΔΔCt ( fold ) as described previously [73] . Each sample was additionally amplified without reverse transcription reaction to confirm the absence of contaminating DNA in the RNA sample . For quantitative PCR of Chlamydia groEL relative to human gapdh , DNA from infected and uninfected Hela cells was isolated using the Gentra Puregene kit ( Qiagen ) according to the manufacturer's instructions . qPCR was performed with 100 ng of RNase treated DNA . The primers for Chlamydia groEL were 5′-GCTCATCTTCATTAGTCAACATTGG-3′ forward and 5′-CTCTCTGGTGGAGTAGCAGTCATT-3′ reverse . The qPCR cycle included 2 min at 50°C , 10 min at 94°C , followed by 35 cycles of 94°C for 15 s , 57°C for 45 s , and 72°C for 20 s . After the cycle , it was followed by a dissociation curve every 0 . 5°C from 55°C to 95°C . Relative gene copies of groEL to gapdh was expressed as 2−ΔΔCt ( fold ) as described above . HeLa cells grown in 6-well plates were mock-infected or infected with C . trachomatis for 1 hr . Unbound bacteria were removed by washing three times with PBS , and infected cells were incubated in 1 mL of media containing 2% FBS for 20 hrs . At the end of incubation , the conditioned media from mock-infected cells ( Mock-CM ) or C . trachomatis infected cells ( CT-CM ) was collected and centrifuged 5 min at 240 X g . The CM was passed through a 0 . 22 µm filter to remove any residual EBs . EBs were resuspended in filtered Mock-CM or CT-CM and vacuole formation was quantified at 18 hpi . For immunoprecipitation of FGF2 , the filtered MOCK-CM or CT-CM was incubated for 2 hrs with 4 µg of anti-FGF2 goat IgG preconjugated to 50 µL of Protein G Sepharose TM 4 Fast Flow ( GE HealthCare ) at 4°C . Immunoprecipitates were recovered by centrifugation at 1 min at 1000 rpm . The immunoprecipitates were boiled in LDS sample buffer containing 100 mM dithiothreitol and then subjected to immunoblotting to detect FGF2 . For immunodepletion of FGF2 , the filtered CT-CM was depleted with normal goat IgG ( R&D system ) , anti-FGF2 goat IgG , or anti-EGF goat IgG preconjugated to Protein G Sepharose TM 4 Fast Flow ( GE HealthCare ) . Supernatants were collected after centrifugation and used for further infection . Immunoprecipitates were recovered and subjected to immunoblotting as mentioned above to ensure FGF2 or EGF depletion . Data represented the mean ± standard error of at least experiments . Statistical analysis was performed using the software program InStat . The significance between groups was determined by ANOVA . p<0 . 05 was considered to be statistically significant . FGF1 ( 2246 ) , FGF2 ( 2247 ) , FGF10 ( 2255 ) , FGFR1 ( 2260 ) , FGFR2 ( 2263 ) , FGFR3 ( 2261 ) , FGFR4 ( 2264 ) , FRS2α ( 10818 ) , ERK-1 ( 5595 ) , ERK-2 ( 26413 ) , PDGFR-β ( 5159 ) , PDGF-B ( 18591 ) , EGF ( 1950 ) , HB-EGF ( 1839 ) , and VEGF ( 7422 ) . | Chlamydia trachomatis is an obligate intracellular bacterium that is an important cause of human disease , including sexually transmitted diseases and acquired blindness in developing countries . The inability to carry out conventional genetic manipulations limits our understanding of the mechanisms of C . trachomatis binding , entry , and spread . Previous studies have shown that heparan sulfate proteoglycans ( HSPGs ) play a role in early binding events . As cell surface HSPGs facilitate the interactions of many growth factors with their receptors , we investigated whether HSPG-associated growth factors affect C . trachomatis binding or entry . Here , we report the novel finding that Fibroblast Growth Factor 2 ( FGF2 ) , a ubiquitously expressed growth factor , enhances C . trachomatis binding to host cells in an HSPG-dependent manner . Furthermore , C . trachomatis infection stimulates production and release of FGF2 through distinct signaling pathways . Released FGF2 is sufficient to enhance the subsequent rounds of infection . Together , these results demonstrate that C . trachomatis utilizes multiple mechanisms to co-opt the host cell FGF2 pathway that sets up a positive feedback loop to enhance bacterial infection and spread . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbial",
"pathogens",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"bacterial",
"pathogens",
"pathogenesis"
] | 2011 | Chlamydia trachomatis Co-opts the FGF2 Signaling Pathway to Enhance Infection |
Nutritional immunity – the withholding of nutrients by the host – has long been recognised as an important factor that shapes bacterial-host interactions . However , the dynamics of nutrient availability within local host niches during fungal infection are poorly defined . We have combined laser ablation-inductively coupled plasma mass spectrometry ( LA-ICP MS ) , MALDI imaging and immunohistochemistry with microtranscriptomics to examine iron homeostasis in the host and pathogen in the murine model of systemic candidiasis . Dramatic changes in the renal iron landscape occur during disease progression . The infection perturbs global iron homeostasis in the host leading to iron accumulation in the renal medulla . Paradoxically , this is accompanied by nutritional immunity in the renal cortex as iron exclusion zones emerge locally around fungal lesions . These exclusion zones correlate with immune infiltrates and haem oxygenase 1-expressing host cells . This local nutritional immunity decreases iron availability , leading to a switch in iron acquisition mechanisms within mature fungal lesions , as revealed by laser capture microdissection and qRT-PCR analyses . Therefore , a complex interplay of systemic and local events influences iron homeostasis and pathogen-host dynamics during disease progression .
Iron is one of the most abundant elements on Earth and , as an important cofactor in metabolic conversions and a mediator of redox reactions , iron is an essential micronutrient for most organisms . On the other hand , excess iron is toxic mediating the formation of potentially deleterious free radicals [1] . Therefore the cellular acquisition , storage and release of iron are tightly controlled . Iron is usually present in the insoluble ferric form ( Fe ( III ) ) at neutral pH [1] . However , most iron in the mammalian body is found in the bloodstream bound to haemoglobin . It is estimated that 2×1011 erythrocytes are turned over every day in the adult human [2] , but dietary iron is insufficient to replenish the potential losses of this red blood cell senescence . Therefore , the iron from senescent red blood cells is recycled by macrophages of the reticulo-endothelial system [3] . During bacterial infection , iron availability is tightly regulated in the host with a view to limiting the availability of this micronutrient , which is essential for the pathogen [4] , [5] , [6] . This phenomenon has been dubbed ‘nutritional immunity’ [7] , [8] . Iron availability is regulated mainly through the hepcidin-ferroportin axis . The liver-derived peptide hormone hepcidin binds to the only characterised mammalian iron exporter , ferroportin , causing its internalisation and degradation , thereby leading to a drop in extracellular iron levels [3] . Modulation of hepcidin levels affects cellular iron pools to minimise pathogenic outgrowth [4] . Furthermore , host iron homeostasis interacts with the immune defences . For example , IL-6 induces hepcidin expression , while increased intracellular iron concentration activates NF-ΚΒ [3] , [9] . Iron is also required for the oxidative burst that enables phagocytic killing of pathogens [10] . In turn , iron partitioning within the host influences microbial niche tropism [6] . The impact of iron during bacterial infection is further illustrated by the observation that iron administration can increase mortality rates in infected patients [11] , [12] . Fungal pathogens represent an increasingly important clinical problem that affects the lives of millions of individuals worldwide and imposes an economic burden representing billions of pounds [13] . Evidence is mounting that iron availability in the host also has a significant impact upon fungal pathogens . The inactivation of iron assimilation mechanisms attenuates the virulence of major fungal pathogens such as Candida albicans and Cryptococcus neoformans [14] , [15] . The link between iron availability and disseminated mucormycosis is well established [16] . Similarly , iron overload exacerbates the progression of candidiasis in mice [17] . Nevertheless , the ability of the host to limit the progression of fungal infections via iron nutritional immunity has not been explored . Aside from reports of increased hepcidin levels during fungal infection [18] , little is known about the in situ competition between the fungal pathogen and host for this essential micronutrient . Candida albicans is a major fungal pathogen of humans . This yeast is a frequent cause of mucosal infection ( thrush ) , and a common cause of nosocomial bloodstream infections in severely immunocompromised hosts , which are associated with 40% mortality [13] , [19] . Most individuals carry C . albicans as a relatively harmless commensal organism in their microflora , and the transition from commensalism to pathogenicity can be difficult to detect . Therefore the early and accurate diagnosis of systemic candidiasis is particularly challenging . Furthermore , new antifungal therapies are being sought to complement the current armoury of clinically useful antifungal drugs [13] . Therefore significant efforts are being invested in the study of fungal metabolic versatility with a view to identifying novel therapeutic approaches [20] , . Blocking either reductive iron assimilation ( via FTR1 deletion ) [24] or haem-iron acquisition ( via HMX1 deletion ) [25] attenuates the virulence of C . albicans . Therefore , the inhibition of fungal iron assimilation presents a potential therapeutic strategy . Given the dynamic nature of fungus-host interactions [26] , our aim was to characterise the spatio-temporal distribution of iron in host niches during disease progression and the accompanying molecular responses of the invading fungus in situ . We reveal a complex interplay between systemic and localised events , which govern fungal iron availability during disseminated candidiasis . We also show that the fungus adapts to these changes in micronutrient availability by adjusting its iron acquisition strategies during disease progression .
Hepcidin is a hepatic hormone that exerts its effects at a system-wide level , regulating the saturation of iron pools within the body [3] , [4] . Disseminated candidiasis is accompanied by increased hepatic hepcidin levels [18] . We examined renal hepcidin levels during systemic candidiasis , revealing elevated hepcidin in the kidneys of infected mice ( Fig . 1A ) . Furthermore , histochemical analyses revealed that hepatic iron storage was more pronounced in animals with candidiasis , when compared to healthy controls ( Fig . 1B ) . This correlated with an increase in the amount of the haem iron-extracting enzyme , haem oxygenase 1 ( HO-1 ) in these organs ( Fig . 1C ) . At the same time , the distribution of iron in the spleen remained relatively unchanged ( Fig . 1D ) . We conclude that global iron homeostasis in the host is perturbed during systemic candidiasis . We focused on the organ with the highest fungal burdens in the classical murine model of systemic candidiasis [27] , and mapped renal iron ( 56Fe ) distributions . Mice were infected with the virulent clinical C . albicans isolate SC5314 [28] , and iron distributions were mapped across kidney sections from animals at various stages of infection ( Fig . 2 ) using laser ablation-inductively coupled plasma mass spectrometry ( LA-ICP MS ) [7] , [29] . Renal fungal burdens increased during disease progression , and this correlated with a redistribution of iron from the cortex in healthy control animals ( Fig . 2A ) to the medulla and medullary pelvis in infected animals ( Figs . 2B &2C ) . To quantify the iron loading in diseased kidneys , 56Fe intensities were normalized against the corresponding 13C intensities ( a proxy for biomass ) across the same tissue sections [29] . For animals with advanced candidiasis , regions of high iron loading ( 56Fe/13C ratio at least twice the background ) accounted for 30% of the total kidney section area , which was double that for healthy controls ( Fig . 2D ) . When the cut-off for normalized 56Fe/13C signal intensities was raised to three times background , a three-fold difference in intensity was observed in iron loaded areas between infected and uninfected kidneys ( ∼10% versus ∼3% ) ( Fig . 2D ) . This represented an average increase in iron levels from 95±19 µg/g in the healthy renal medulla , to 195±26 µg/g in an infected kidney medulla , respectively . Therefore , C . albicans infection affected the loading as well as the distribution of iron in infected kidneys . This redistribution of renal iron was replicated during infections with a second C . albicans clinical isolate from a different epidemiological clade [30] ( AM2003/0191: Fig . 2E ) . Furthermore , the degree of cortical-to-medullary redistribution of renal iron corresponded to the size and number of lesions detected in the kidney ( cf . Fig . 2E , more than 40 lesions 250–300 µm in diameter; and Fig . S1 , ∼15 lesions 50–100 µm in diameter ) . In addition , LA-ICP MS mapping of renal iron distributions in mice infected with C . albicans strains that display reduced virulence indicated that a sustained C . albicans infection must be established to trigger iron accumulation in the renal medulla ( Fig . S2 ) . This iron redistribution was accompanied by the establishment of iron exclusion zones in the renal cortex that correlated with the positions of fungal lesions ( Fig . 2F ) . Furthermore , the size of the iron exclusion zones around fungal lesions correlated with the mass of immune infiltrates encompassing these lesions ( Fig . 2F ) . This observation is consistent with previous studies pertaining to the active role of immune infiltrates in nutritional immunity [7] . It was recently shown that , during bacterial infection , neutrophil infiltration drives the redistribution of manganese and zinc in infected tissues around the infection foci [7] . Native MALDI MS imaging ( MALDI IMS ) [31] , [32] was used to examine the kidney proteome in situ . This revealed an ion of m/z 14981 that displayed significant differences in its spatial distribution across healthy versus infected kidneys ( Fig . S3 ) . The identity of this polypeptide was murine haemoglobin alpha ( HBA ) ( gi|122441 ) as revealed by LC-MS/MS in conjunction with MALDI-TOF MS ( see Materials and Methods ) . The spatial distribution of HBA replicated that of iron , in that HBA was predominantly located in the renal cortex of healthy animals and elevated in the renal medulla of mice with advanced candidiasis ( Fig . 3 , Fig . S4 ) . Furthermore , the distribution of a small species of m/z 617 . 6 , which was identified as haemoglobin haem B based on its LIFT fragmentation spectrum [33] ( Fig . S5C ) , also reflected iron distribution patterns during infection ( Fig . 3 , Fig . S4 ) . Comparisons of replicate MALDI TOF spectra of tryptic peptides from healthy and infected kidneys confirmed that HBA was significantly overrepresented in the renal proteomes of infected animals relative to the healthy controls . Indeed , the HBA peptides IGGHGAEYGAEALER ( ion m/z 1529 . 73 ) and TYFPHFDVSHGSAQV ( ion m/z 1819 . 88 ) were amongst the most discriminatory peptides between the infected and uninfected renal proteomes ( Fig . S5 , Table S1 ) . In contrast , the spectral abundance of tryptic peptides derived from haemoglobin beta subunits 1 and 2 ( HBB ) ( gi|122513; gi|17647499 ) did not differ significantly between the healthy and infected states ( Fig . S5; Table S1 ) . To test whether the elevated iron levels in the kidney were associated with renal tissue or the vasculature of this organ , kidneys were perfused with saline before LA-ICP MS imaging . This procedure depleted the cortical iron from healthy kidneys suggesting that this iron was primarily associated with red blood cells . In contrast , perfusion failed to displace the medullary iron from infected tissue ( Fig . 4 ) , suggesting that during candidiasis , excess iron was deposited in the medulla of infected kidneys . Iron accumulation in the kidney is often associated with severe tissue injury , haemolysis and haematuria [3] . It has been reported that animals with disseminated candidiasis succumb to progressive sepsis and can develop severe renal failure [34] . However , several observations indicated that the renal accumulation of iron during systemic candidiasis was not caused by kidney injury and haemolysis in our infection model . Firstly , histological analyses revealed no pronounced changes in kidney architecture . Secondly , changes in renal iron distribution were apparent early in the infection process before large fungal lesions had developed and significant tissue damage was incurred ( Fig . S1 ) . Thirdly , the differential accumulation of alpha and beta globins ( HBA and HBB ) was inconsistent with internal haemorrhage . Fourthly , only sporadic traces of haemolysed and non-haemolysed blood were detected in mouse urine , and these did not correlate with the progression of infection ( Table S2 ) . Thus we conclude that the progressive renal iron accumulation observed during systemic candidiasis was not mediated by kidney injury and haemolysis . Since the perfusion of infected kidneys failed to dislodge the medullary iron , we reasoned that this iron might be stored by the renal tissue . Hence the levels of renal ferritin were examined by immunohistochemistry . Normally , low levels of this major iron storage protein are present in most tissues , but ferritin abundance is up-regulated in response to iron [35] . Significantly , ferritin levels increased in infected kidneys ( Fig . 5A , bottom row ) , primarily in the renal medulla ( Fig . 5B ) . In contrast , the weaker ferritin signal observed in healthy kidneys localised primarily to the renal cortex ( Fig . 5A , top row , and 5B ) . The cellular receptor for transferrin-bound iron [36] was also more abundant in infected than healthy tissue . Like ferritin , the transferrin receptor was concentrated outside fungal lesions ( Fig . 5A ) . Taken together , our data suggest that some of the iron that accumulated in the renal medulla during disease progression was stored in ferritin complexes and some was bound by haem outside lesions . Next we examined the spatial distribution of haem-iron extracting enzymes – haem oxygenases ( HO ) – during systemic candidiasis . In mammals , HO-1 is inducible and involved in red blood cell recycling in the spleen and general responses to inflammation , whereas HO-2 is expressed at constitutively low levels in most tissues [37] , [38] . Immunohistochemical analyses of HO-2 revealed that its abundance was slightly increased in infected kidneys in areas outside the fungal lesions ( Fig . 5A ) . Meanwhile HO-1 accumulated in distinct bands that surrounded fungal lesions ( Fig . 5A ) . The external radii of HO-1 bands corresponded to the size of the immune infiltrates ( Fig . 5A , bottom row , and Fig . 5C; also Figs . S1D & S6 ) , analogous to the iron exclusion zones observed by LA-ICP MS ( Fig . 2F ) , suggesting that HO-1 might contribute to host nutritional immunity mechanisms . Interestingly , HO-1 expressing cells did not react with anti-Ly-6G or anti-F4/80 specific antibodies , suggesting that cells other than neutrophil granulocytes and tissue macrophages mediate the exclusion zones ( Fig . 5C ) . We conclude that renal iron retention and accumulation during fungal infection is a dedicated , rather than random process that involves specific host proteins ( Fig . 5A ) . Most mammalian iron is present in erythrocytes and is recycled via the HO-1 synthesized by splenic red pulp macrophages . Significantly , it has been suggested that the kidney plays a major role in the retention of iron from senescent red blood cells in HO-1−/− animals whose splenic iron recycling function is severely impaired [39] . Therefore we tested the hypothesis that fungal infection perturbs splenic function . The fungal burden ( CFU/g weight ) in the spleen is typically a hundred-fold lower than that of the kidney in our infection model [27] , and immunohistochemistry with the F4/80 macrophage marker [40] revealed that splenic red pulp macrophage numbers were not significantly reduced during infection ( Fig . 5D ) . However , the spleens of mice with systemic candidiasis contained less HO-1 than healthy controls ( Fig . 5D ) , indicating that splenic function was affected by the infection . Furthermore , selective partial chemical ablation of red pulp macrophages with clodronate [41] ( Fig . 6A ) replicated the renal iron redistribution phenotype observed during the early stages of candidiasis ( Fig . 6B , Fig . S1 ) . The clodronate treatment had no discernable impact on the F4/80 resident macrophage population in the kidney , as assessed by immunohistochemistry with specific antibodies ( data not shown ) . The observed changes in renal iron loading and distribution , iron regulatory and storage proteins , and splenic HO-1 function during systemic candidiasis were initially observed in BALB/c mice . Since immune responses to C . albicans can differ in some mouse strains , we repeated the experiments in a different mouse genetic background . All of our observations were recapitulated in C57BL/6 mice ( Fig . S6 ) . We conclude that the fungal infection affects renal and splenic iron homeostasis of the host . C . albicans has three characterised pathways of iron acquisition: xenosiderophore-mediated , non-haem iron ( reductive pathway ) , and haem-iron acquisition [14] , [42] , [43] , [44] . Transcript analyses from C . albicans lesions harvested at various stages of development by laser capture microscopy ( Fig . 7A and 7B ) indicated that the fungus relies primarily on the FTR1-dependent reductive pathway of iron acquisition during the early stages of renal infection . However , as fungal lesions matured , C . albicans induced the expression of the HMX1-dependent haem iron acquisition pathway . Based on the regulation of these processes [14] , [44] , these data suggest that the fungus increasingly experiences iron limitation as the disease progresses . Indeed , we determined the average iron content across fungal lesions to be roughly half that of healthy renal cortex ( 24±6 µg/g versus 61±13 µg/g , respectively ) . This is consistent with the idea that , despite the elevated haemoglobin and haem iron loading in the renal medulla , the host limits the availability of iron to the fungal lesions in the renal cortex via nutritional immunity , akin to what has been observed for bacterial and viral infections [8] .
Our paper addresses the question of iron homeostasis and nutritional immunity during fungal pathogenesis , an area of increasing medical importance . In vivo data concerning host iron homeostasis during systemic candidiasis is essentially limited to a small number of reports that focus on hepcidin and ferroportin [18] . For example , hepcidin has been reported to possess some anti-fungal properties [45] . Also , C . albicans infection has been shown to stimulate hepcidin production in the liver and cause a decrease in serum transferrin levels in the murine model of disseminated Candida infection [18] . We detected increased hepcidin levels in the kidneys of infected animals , and this was accompanied by elevated iron storage in the liver and increased HO-1 activity in that organ ( Fig . 1 ) . These results are analogous to the host response to infection by extracellular bacterial pathogens , i . e . , increased intracellular iron storage [4] , [6] . We report here that disseminated candidiasis causes systemic perturbation of iron homeostasis by major internal organs ( Fig . 1 ) , both in organs where the fungus proliferates ( kidney ) and in those with lower fungal burdens ( spleen ) . Candida infection triggered dramatic changes in the renal iron landscape , with the iron shifting from the renal cortex to the medulla , in a process involving specific host proteins ( Figs . 2 & 5 ) . It is known that the kidney participates in iron retention [46] , but the extent of its involvement in host iron homeostasis has been far from clear . Analyses of transgenic mice lacking the main haem oxygenase , HO-1 , have suggested that the kidney helps to salvage systemic iron when splenic red blood cell recycling is dysfunctional [39] . However , this relationship had not previously been demonstrated during microbial infection . Our data indicate the possibility of the iron homeostatic cross-talk between spleen and kidney under pathological conditions , when the host is suffering disseminated candidiasis . The pharmacological attenuation of splenic red pulp macrophage function by clodronate administration recapitulated the renal iron loading phenotype observed during the early stage of infection ( Fig . 6 ) . This suggested a plausible relationship between perturbed red blood cell recycling by splenic red pulp macrophages and iron accumulation in the kidney , suggesting that systemic C . albicans infection affects splenic function . Interestingly , Qian et al [47] reported that clodronate treatment increases the susceptibility of mice to systemic candidiasis . These authors suggested splenic involvement in the control of the infection through clearance of the pathogen from the bloodstream . Our data suggest that the effects of clodronate might instead be mediated through the exacerbation of defects in host iron homeostasis triggered by the fungal infection . We also show that nutrient immunity operates during systemic candidiasis , further contributing to the iterative and dynamic relationship between host iron homeostasis and microbial micronutrient assimilation . Our data indicate that HO-2 promotes the assimilation of renal iron , which is then stored by ferritin in the renal medulla . Meanwhile , iron accumulation is limited locally around fungal lesions in the renal cortex via a combination of infiltrating immune cells and local increases in host HO-1 expression . We did not associate HO-1 activity with F4/80-expressing macrophages ( Fig . 5C ) . It has been proposed that elevated HO-1 activity exerts anti-inflammatory effects through the modulatory roles of carbon monoxide ( a haem breakdown product ) in intracellular signalling [3] , [25] . Therefore it is not inconceivable that the nutritional immunity that operates during systemic candidiasis has the dual function of depriving the pathogen of iron , and limiting inflammation at infection sites via HO-1 activity . C . albicans has three characterised iron acquisition systems: ( a ) the FTR1-dependent reductive pathway [24]; ( b ) the HMX1-dependent haem-iron acquisition pathway [48]; and ( c ) the SIT1-mediated xenosiderophore acquisition pathway [14] , [42] . However , it was unclear which of these pathways operates during disease onset and progression , and whether there is functional redundancy between these pathways in tissues . We show here that the fungus responds to the dramatic changes in the renal iron landscape that occur during infection by adjusting its iron acquisition strategies . We observed that FTR1 pathway genes are expressed throughout the infection , but that the HMX1 pathway is induced in the latter stages of infection . No expression of the SIT1 siderophore receptor gene was detected . This is consistent with the observation that the SIT1 pathway contributes to epithelial invasion in the presence of exogenous siderophores , presumably during dissemination from the gut [42] . On the other hand , FTR1 appears to be essential for virulence and organ colonisation during bloodstream dissemination [24] , and HMX1 seems to be required for the maintenance of developing fungal lesions in the kidney [25] . It is conceivable that the HMX1 pathway induction observed in latter stages of the infection was a consequence of the haemoglobin accumulation in the kidney ( Fig . 2 ) . Whatever the mechanism , HMX1 induction could reflect the successful imposition of nutritional immunity by the host , further supported by our in situ iron measurements ( see above ) . In addition , this response might be hard-wired into fungal defence mechanisms against the immune system of the host , as suggested by Navarathna and Roberts [25] . Taken together , our data show that systemic events in the host influence iron homeostasis and iron levels in major organs during fungal infection . Meanwhile , the host limits the availability of this essential micronutrient to the invading fungus through local mechanisms involving nutrient immunity ( Fig . 7C ) . Our study provides unprecedented insights into the host-fungus interactions that revolve around iron homeostasis in the host and microbial iron assimilation during systemic fungal infection .
C . albicans inocula ( 104–105 CFU/g body mass ) were injected into the lateral tail veins of 6–10 week old female BALB/c or C57BL/6 mice . Infections were allowed to proceed for up to 4 days . Typically , early infections were analysed on day 2 , and advanced infection on days 3–4 . Advanced infection was characterised by a prevalence of large renal fungal lesions ( one dimension ≥200 µm ) , and early infection by small fungal clusters ( circa 60 µm in diameter , or smaller ) , as assessed by periodic acid/Schiff staining [49] ( vide infra ) . Harvested organs were either submerged in RNAlater solution ( Qiagen , Crawley , UK ) for transcript profiling , or immediately placed on dry ice and stored at −80°C . For perfusion experiments , animals were sacrificed and their blood displaced by perfusion with 1–2 mL of a sterile saline . C . albicans SC5314 , AM2003/0191 , and J981301 are clinical isolates [28] , [30] . The construction of the C . albicans ftr1 null mutant ( Caftr1 [24] ) and the C . albicans hmx1 null mutant ( DLR2 [25] ) has been described . For BALB/c infections , C . albicans strains were grown overnight at 30°C in NGY medium [30] , and for C57BL/6 infections strains were grown in Saburaud dextrose medium ( Oxoid , Basingstoke , UK ) . The cells were grown to early stationary phase , washed and resuspended in saline , and cell density determined by hemocytometer counting before i . v . injection . Actual inoculum levels were confirmed by viable cell counts . Mouse urine samples were collected , snap frozen and kept at −20°C until use . Samples were analysed using Siemens Healthcare Diagnostics dipsticks ( Dipstix ) according to the manufacturer's instructions . All tissues were cryosectioned in a Leica CM 1850 cryostat ( Leica Biosystems , Newcastle Upon Tyne , UK ) . Tissue pathology was assessed with the periodic acid/Schiff reagent/hematoxylin stain [49] . Perls reagent ( Polysciences , Inc . , Warrington , PA ) was used to visualise non-haem iron in tissues , according to the manufacturer's instructions . Specific antigens were detected in tissue sections using VECTASTAIN Elite ABC system ( VectorLabs , Orton Southgate , UK ) according to the manufacturer's recommendations . The following mouse-specific primary antibodies were used: FTH1 ( ferritin , heavy polypeptide 1 ) , rabbit , polyclonal ( Aviva Systems Biology , San Diego , CA ) ;F4/80 , rat , monoclonal ( AbDSerotec , Kidlington , UK ) ; hepcidin-25 , rabbit , polyclonal ( Abcam , Cambridge , UK ) ; HO-1 , rabbit , polyclonal ( Abcam , Cambridge , UK ) ; HO-2 , rabbit , polyclonal ( Abcam , Cambridge , UK ) ; Ly-6G , rabbit , polyclonal ( Abcam , Cambridge , UK ) ; transferrin receptor , rabbit , polyclonal ( Abcam , Cambridge , UK ) . The secondary antibodies used were biotinylated horse anti-rabbit IgG ( H+L ) ( VectorLabs , Orton Southgate , UK ) , or goat anti-rat IgG2b:horse radish peroxidase conjugate ( AbDSerotec , Kidlington , UK ) , as required , using 3 , 3′-diaminobenzidine as the substrate ( VectorLabs , Orton Southgate , UK ) . All images are representative of numerous replicates from at least three independent biological replicates . Brown colour indicates positive reaction; blue , no staining . Sequential sections from the same tissue were prepared for LA-ICP MS , MALDI IMS ( vide supra ) , and histology ( vide infra ) . Cryosections ( 22 µm thick ) were mounted on conventional glass slides and elemental distribution mapping was performed using laser ablation system ( UP-213 , New Wave ) coupled to an Agilent 7500c ICP-MS , largely as described previously [7] , [50] . Typically , the scan speed was 25 to 50 µm/s , with 80 µm laser spot size , and 20 µm spacing between the lines . Data were normalised as 56Fe to 13C intensity ratios [29] and were plotted using Microsoft Excel v14 . 2 . 4 . Unless otherwise specified , the images are representative of at least two biological replicates . Quantitative iron measurements were conducted as described [51] and are averages from two biological replicates . Fungal lesion iron content was determined based on measurements from six fungal lesions . The numbers are given as µg per g dry organ weight and errors are standard deviations from the mean . Before processing , optical images of 12 µm thick tissue cryosections were acquired with NikonCoolScan , V ED slide scanner ( Nikon , Kingston upon Thames , UK ) for co-registration with subsequent MALDI IMS experiments . The sections were ethanol washed as per manufacturer's instructions and spray-coated ( ImagePrep , BrukerDaltonics ) with the appropriate matrix∶sinapic acid for native MALDI IMS ( 10 µg/mL ) ( Fluka , Dorset , UK ) ; in-house recrystallized cyano-4-hydroxy-cinnamic acid ( 7 µg/mL ) for peptide and haem imaging ( Sigma Aldrich , Dorset , UK ) . For non-native imaging , tissue sections were spray-coated with sequencing-grade trypsin ( Promega , Southampton , UK ) solution ( 40 µg in 200 µL of freshly prepared 50 mM ammonium bicarbonate buffer ) , incubated overnight in humidified chamber ( 50% methanol , 38°C ) and then spray-coated with matrix . MALDI TOF data were acquired with ultrafleXtreme MALDI TOF/TOF ( BrukerDaltonics ) at 50 , 60 , 70 or 100 µm resolution , in positive ion mode , and with pre-set manufacturer protocols . The data were subsequently analysed with flexImaging 2 . 1 and ClinProTools2 . 2 software ( Bruker ) . Peptides of sufficient abundance and haem were sequenced in situ either from the imaged sections , or from sequential sections , using the LIFT protocol [31] . For peptide identification , the following Mascot ( http://www . matrixscience . com/search_form_select . html ) parameters were used for analysis of the MALDI TOF/TOF spectra: enzyme , trypsin; up to 2 missed cleavages allowed; unless otherwise indicated parental ion error ±2 . 0 Da; fragment ion error ±1 . 0 Da; preset variable modification , oxidation of methionine residues; database searched , UniProtKB/TrEMBL . Typically , peptide sequencing and identification of a given m/z species was performed from at least three biological replicates to confirm peptide identity . To identify the protein present as a dominant peak in the average infected native MALDI IMS spectrum ( 14981 Da ) ( vide supra ) , sixteen 12 µm-thick kidney sections were collected , washed in ethanol ( 2×1 min 70% , 1×1 min 100% ) , and air dried . Proteins were extracted three times with 60∶40 ACN/0 . 2% TFA into a total volume of 240 µL , the extracts evaporated to dryness and residues resuspended in 0 . 1% TFA . 100 µL were submitted to HPLC fractionation on Brownlee Aquapore RP-300 column ( C8 , 7 µm , 30×2 . 1 mm , at 40°C ) , using Proteineerfc ( BrukerDaltonics ) system , operated by HyStar 3 . 2 . 44 . 0 software . The flow rate was 100 µL/min , with the following solvents: A , 0 . 1%; B , 70% ACN/0 . 085% TFA . Proteins were fractionated using a gradient of 5–75% solvent B in 70 min , holding the gradient for 10 min , followed by a column wash in 100% solvent B . 96×100 µL fractions were collected , dried , and resuspended in 10 µL 50% ACN/0 . 1% TFA . After mixing 1∶1 with matrix solution ( sinapic acid , vide supra ) , 1 µL was spotted on a MALDI target plate ( BrukerDaltonics ) . MALDI TOF spectra were automatically acquired with a native MALDI IMS method on ultrafleXtreme MALDI TOF/TOF ( BrukerDaltonics ) . Fractions containing the m/z species of interest were prepared and digested with sequencing-grade trypsin ( Promega , Southampton , UK ) following standard protocols . This was followed by liquid chromatography/tandem mass spectrometry using BrukerDaltonics HCTultra mass spectrometer with Dionex UltiMate 3000 LC system . Finally , the acquired spectra were submitted to a Mascot search , using the following parameters: enzyme , trypsin; up to 2 missed cleavages allowed; parental ion error ±1 . 5 Da; fragment ion error ±0 . 5 Da; charge: +2 and +3; preset modifications , oxidation of methionine residues ( variable ) , carbamidomethylation of cysteine residues ( fixed ) ; the database searched was UniProtKB/TrEMBL . Transcript profiling from laser capture microdissected material was done as described elsewhere [52] with some modifications . Briefly , a Zeiss PALM system was used for microdissection of RNAlater ( Qiagen , Crawley , UK ) fixed tissues . This was followed by RNA extraction and amplification with Arcturus RiboAmp HS Plus Two-Round RNA Amplification Kit ( Life Technologies Ltd . , Paisley , UK ) . Roche LightCycler 480 and Universal Probes for monocolour hydrolyses reactions were employed in qRT-PCR , according to the manufacturer's instructions . Primer pairs and specific probes are listed elsewhere [52] . For the assessment of relative abundance of PGA10 ( orf19 . 5674 ) gene transcript , the following were used: PGA10 . left primer , 5′-CTGGTTGTTTGTGTGTCATGC-3′; PGA10 . right primer , 5′-GTTTTTAGCAACACAGTCACCAAT-3′; probe #119 ( http://www . roche-applied-science . com/sis/rtpcr/upl/index . jsp ) . Gene expression was normalised to ACT1 , and differences were considered statistically significant for p≤0 . 05 in two-tailed t-test using the Mann-Whitney U test . Uniprot Knowledge Base ( http://www . uniprot . org ) accession numbers for C . albicans proteins mentioned in the text: ACT1 ( orf19 . 5007 ) , P14235; ALS3 ( orf19 . 1816 ) , O74623; CSA1 ( orf19 . 7114 ) , G1UB63; CSA2 ( orf19 . 3117 ) , Q5A0X8; FTR1 ( orf19 . 7219 ) , Q59ZX2; HMX1 ( orf19 . 6073 ) , Q5AB97; PGA10 ( orf19 . 5674 ) , Q59UP6; SIT1 ( orf19 . 2179 ) , Q5A2T6 . All animal experiments were conducted in compliance with United Kingdom Home Office licenses for research on animals ( project license number PPL 60/4135 ) , and were approved by the University of Aberdeen ethical review committee . Animal experiments were minimised , and all animal experimentation was performed using approaches that minimised animal suffering and maximised our concordance with the 3Rs . | Microbial pathogens must assimilate essential micronutrients to establish infections . During bacterial infection , mammals limit the availability of micronutrients to inhibit the growth of the pathogen – a phenomenon termed ‘nutrient immunity . ’ Nutrient immunity has not been examined during disseminated candidiasis . Yet micronutrient assimilation , and iron assimilation in particular , is required for fungal virulence , and life-threatening disseminated fungal infections are recognised as a major medical threat for patients with compromised immune systems . We show that nutrient immunity operates during disseminated Candida albicans infections in mice . Over time immune cells congregate around the fungal lesions in the kidney cortex , driving nutrient immunity and reducing iron availability for the pathogen . The fungus responds by tuning its iron assimilation strategies to the reduced iron levels . Paradoxically , iron levels increase in other parts of the kidney as Candida infections progress . We show that the fungal infection disturbs global iron homeostasis in the host by perturbing red blood cell recycling in the spleen and this is associated with increased iron storage in the kidney medulla . Therefore , fungal infection exerts system-wide effects upon iron homeostasis in the mammalian host , whilst triggering local nutrient immunity to limit the infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Fungal Iron Availability during Deep Seated Candidiasis Is Defined by a Complex Interplay Involving Systemic and Local Events |
Hydatidosis is a zoonotic disease of worldwide distribution caused by Echinococcus granulosus . Our study aimed to determine the prevalence of human and canine echinococcosis as well as the associated risk factors in a rural area of the Limarí province in northern Chile . A cross-sectional study was conducted between August and November 2009 using a stratified sampling design in each of the five districts of the province . In the selected villages , up to 10 households were sampled . Serum and fecal samples from an adult family member and a dog were collected from each participating household . Risk factors were assessed by standardized questionnaires . Seroprevalence was assessed using a multi-step approach: an ELISA for screening , IFA , IHA and western blot for confirmation of results , respectively . The prevalence of echinococcal infection in dogs was determined by coproantigen genus specific ELISA . Chi-square , Fisher tests and logistic regressions were used to assess risk factors for human seropositivity and dog copropositivity . A seroprevalence of 2 . 6% ( 10/403 ) and coproprevalence of 28% ( 26/93 ) was recorded for humans and dogs respectively . Contact with dogs and dog feces were risk factors for human seropositivity while dog copropositivity was associated with home slaughter of livestock ( OR = 3 . 35; CI 90%: 1 . 16–6 . 85 ) and households de-worming dogs ( OR = 2 . 82; CI 90%: 1 . 33–8 . 43 ) . Echinococcal infection of humans and their dogs is common in Limarí province . Risk factors for human seropositivity were related to contact with domestic dogs and their feces , whereas those for dogs were home slaughter of livestock and the practice of de-worming dogs .
Hydatidosis or cystic echinococcosis ( CE ) is a chronic zoonotic parasitic disease of almost worldwide distribution caused by the cestode parasite Echinococcus granulosus . Cystic echinococcosis accounts for 95% of the estimated 2–3 million cases of human echinococcal infections worldwide and represents a major public health problem in many parts of the world [1] . It is listed by the World Health Organization as a neglected zoonotic disease since this disease mainly affects poor and marginalized populations in low-resource settings ( www . who . int/neglected_diseases/zoonoses ) . The life cycle of this helminth includes carnivores , mostly dogs , as definitive hosts and herbivores such as sheep and goats as intermediate hosts . Humans become infected after accidental ingestion of eggs excreted with carnivore feces . Cystic echinococcosis in humans and animals is characterized by the development of metacestode larval stages in the liver and other organs . Known key factors of persistence , emergence or re-emergence of hydatid disease in a given human population are among others ( i ) the presence of large numbers of dogs harbouring E . granulosus worms , ( ii ) access of dogs to infected offal ( iii ) inadequate facilities for slaughter and destruction of infected viscera , ( iv ) slaughtering of livestock in homesteads [2] , [3] . The practice of feeding dogs with infected offal is by far one of the most important factors for the persistence of this disease [4] . The age of dogs is another relevant factor , since young dogs eliminate higher numbers of echinococcal eggs in their feces than older dogs [5]–[7] . In addition , restrictions regarding dog ownership are of epidemiological influence , since street dogs have an increased risk of acquiring E . granulosus infection through uncontrolled access to infected carcasses [7] . Other factors include the lack of regular deworming of dogs and the absence of knowledge pertaining to infection and disease [5]–[7] . Echinococcus granulosus is hyperendemic in the southern parts of South America , e . g . Argentina , Chile , Peru , and Uruguay , where it has an important economic and public health impact [8] . Sheep are the main intermediate host for the G1 genotype of E . granulosus , which has a worldwide distribution including South America [9]; to the best of our knowledge no studies of the genetic strain circulating in Chile has yet been published . The epidemiological situation in South America is complex and not fully understood and comprehensive epidemiological data is lacking . In Chile , echinococcosis mostly affects humans and their livestock in rural and poorly developed areas . According to national surveillance data , the surgical incidence has remained stable at around 2 cases per 100 , 000 inhabitants since the early 1990s [10] , [11] . Although in Chile hydatid disease is a relevant public health problem , data regarding local distribution and risk factors is limited . One study carried out in the Coquimbo region which assessed risk factors for the presence of E . granulosus eggs in dog feces revealed that juvenile dogs from households performing home slaughter , which had not been de-wormed in the two previous months , were at highest risk of contracting echinococcosis [12] . To the best of our knowledge , epidemiological studies of E . granulosus infection in both humans and dogs at a household level , which are essential to implement control programs , have not been carried out in Chile and are rarely reported elsewhere [eg . 13] , [14] , [15] . Our study aimed to determine the prevalence of human echinococcosis as well as the associated risk factors including canine echinococcosis in rural areas of the Limarí province of the Coquimbo region in northern Chile .
A cross-sectional study was conducted from August to November 2009 within the Limarí province of the Coquimbo region in northern Chile ( Figure 1 ) . Stratified sampling depended on the number of rural villages in each of the five municipalities . To estimate sample sizes , we used an echinococcosis prevalence of 3% in humans [16] and 28% in dogs [17] . For target values of 90% for confidence intervals and ±2 . 5% and ±4% for errors of human and dog populations , respectively , a sample size of 480 human samples was estimated using Epi Info 6 . 0 ( wwwn . cdc . gov/epiinfo/ ) . This number was approximated to 500 samples , of which 84 , 67 , 46 , 217 , and 86 were assigned to the municipalities of Ovalle , Punitaqui , Rio Hurtado , Monte Patria , and Combarbalá , respectively , according to the proportion of villages of each municipality and the overall number of villages in the province . In each village/settlement , ten households were randomly selected to be visited . During field visits epidemiological data was collected by a standardized questionnaire and blood samples were obtained from one adult per household ( keeping an even sex ratio of the total samples ) . Furthermore , a fecal sample was collected from one dog per household in randomly selected households of two municipalities . The study was approved by the Institutional Review Board ( IRB ) of the Ministry of Health at the Coquimbo region . Information regarding the study was initially communicated to potential participants prior to their signing an informed consent . The study included a questionnaire survey to determine the potential risk factors for transmission of E . granulosus in humans and dogs [eg . 5] , [12] . The questionnaire included basic demographical data of the dog owner and his household , data on education and occupation , living standards including waste management and water supply as well as slaughtering practices and knowledge of the disease ( using graphic material ) . Furthermore , the questionnaire covered data about the sampled dogs , dog-keeping practices , such as contact with dogs ( i . e . high: grooming , petting , sleeping with dogs , dogs allowed to enter into the house; low: dogs sleeping outside the household with close contact on few occasions ) and other factors that could influence contact between humans and the parasite . Complex questions were asked as open format questions to reduce bias . The questionnaire took between 30–40 minutes to complete . After each interview , a blood sample was taken from each participant by peripheral venipuncture . Specimens were centrifuged on the same day using a portable centrifuge ( Mobilspine , Vulcon Technologies , Richmond , USA ) , serum was separated and kept at −20°C until further analysis . Additionally , a fresh fecal sample from one dog within each household was collected either rectally or from the ground as previously described [14] . A small amount ( approx . 1 gram ) of feces was placed in 5% phosphate-buffered saline formalin and thoroughly mixed . The supernatant was transferred into Eppendorf tubes and maintained at 4°C until further analysis [5] . Human serum samples were all screened by a commercial enzyme-linked immunosorbent assay ( ELISA ) detecting IgG antibodies against E . granulosus antigens ( Echinococcus IgG ELISA classic , Serion Immundiagnostica , Würzburg , Germany ) . Positive or indeterminate samples were retested at the Institute of Medical Microbiology , Immunology and Parasitology , Bonn , Germany , by additional serological techniques such as indirect hemagglutination assay ( IHA ) and immunofluorescence assay ( IFA ) , which have been described elsewhere [18] . Samples with positive or borderline results in either in-house assay were confirmed by Western Blot ( WB ) ( Echinococcus Western Blot IgG , LDBIO Diagnostics , Lyon , France ) . Seropositivity was defined by means of a WB positive result . To diagnose echinococcal infections in dogs , an Echinococcus genus-specific coproantigen ELISA was used ( Cestode Zoonoses Research Group , University of Salford ) , as previously described [19] , [20] . Data was analyzed using Stata 10 . 0 ( Stata Corporation , Texas , USA ) . To compare frequencies of confirmed seropositive human samples between municipalities , a Fisher exact test was used . Additionally , risk factors associated with seropositivity to E . granulosus in people inhabiting rural areas of the Limarí province was assessed by two tailed Fisher or Chi-square tests . To determine risk factors for canine echinococcosis , fixed effect logistic regression analyses were performed , using adjustment by sample numbers in each municipality . For risk analysis , univariate logistic regression was carried out to select variables with p values≤0 . 250 , which were included in further multivariate models .
A total of 393 households were visited ( Figure 1 ) . An overall of 3 . 5 persons and 2 . 2 dogs per household were included , with a ratio of 1 . 7 human per each dog per household . Livestock herding was the occupation reported by 23% of interviewees . Only 64% of interviewees reported to have finished their primary education and 52% of them said to have potable water . In 3% of all households , at least one member reported to suffer from hydatid disease . High contact rates with dogs and regular contact with dog feces were reported by 41% and 43% of the participants , respectively . The majority of individuals ( 75% ) stated that they had never dewormed their dogs with antiparasitic drugs . Home slaughter or purchase of undisemboweled animals for consumption was reported in 63% of households , of which 75% had noticed the presence of fluid-filled structures compatible with hydatid cysts . Furthermore , 61% reported the presence of household dogs during slaughter or disembowelment of livestock , 50% reported feeding of dogs with viscera , and 34% feeding of dogs with hydatid cysts . Regarding the knowledge of zoonotic diseases , 55% of participants quoted that they knew of diseases transmitted from animals to humans , however only 17% had heard about human hydatidosis ( Table 1 ) . The initial ELISA screening revealed positive results in 47 out of 403 serum samples , resulting in a seroprevalence of 12% ( 90% CI 9–14% ) . When the 47 initially positive samples were retested by IHA , IFA , and WB 10 samples were confirmed positive by this multi-step analysis , resulting in a seroprevalence of 2 . 6% ( 90% CI 1 . 6–4 . 4% ) . Prevalence in different municipalities ranged from 0% to 6 . 8% , but remained without significant differences ( Table 2 ) . Statistical analyses revealed that a higher seropositivity was found in people from households reporting high dog contact , those that allowed dogs to defecate in orchards , and those that did not regularly collect their dog feces ( Table 3 ) . Due to the small number of positives it was not possible to run a multivariable analysis . A further study with a larger sample size is recommended . A total of 93 canine fecal samples from the Combarbalá and Monte Patria municipalities were collected and analysed . Of those , 26 were positive resulting in an overall prevalence of 28% ( 90% CI 21–36% ) . The prevalence in the municipalities of Combarbalá and Monte Patria was 29% ( 15/52 ) and 27% ( 11/41 ) , respectively . Variables ( n = 15 ) from the epidemiological questionnaire were tested for the risk of the presence of E . granulosus antigen in canine samples in the respective household using a univariable logistic regression analysis ( see Table 4 ) . Four variables were selected for further multivariable logistic regression model: ‘de-worming of dogs’ , ‘regular veterinary care’ , ‘home slaughter’ , and ‘feeding dogs with viscera’ ( selected variables in bold in Table 4 ) . Final conditional regression analysis revealed that ‘dogs from households reporting home slaughter of livestock were 3 . 35 ( 90% CI 1 . 16–6 . 85 ) times more likely to shed E . granulosus antigen than those from households without home slaughter . Furthermore , dogs from households reporting de-worming of pets had a 2 . 82 ( 90% CI 1 . 33–8 . 43 ) times higher risk of carrying E . granulosus than dogs from households that did not de-worm their dogs .
Our study was designed to obtain an accurate picture of human and dog echinococcosis in rural areas of the Limarí province in the Coquimbo region through the stratification and random selection of villages and households in each municipality . Based on the seroprevalence of 2 . 6% it may be assumed that about 1560 individuals in this province would have been in contact with the parasite and could suffer from hydatidosis . This value is several orders of magnitude higher than that officially reported [21] and should be considered when planning control programs . Our findings confirm that monitoring the seroprevalance by means of an ELISA which applies native , cyst fluid antigens of E . granulosus might easily over estimate human exposure to E . granulosus in areas where contact or infection with other helminths is possible . Using a one-step approach with an ELISA of low specificity , as done in many epidemiological studies , the seroprevalence rate would have increased to 12% in our population . Through the implementation of a combination of confirmatory techniques , we attempted to eliminate false positive results caused by cross-reactivity with other parasites [4] , as suggested by the WHO [22] . Serological studies have been used to assess prevalence and risk factors of cystic echinococcosis in humans worldwide [eg . 14] , [23] , [24] . Still , this approach does not detect all cases of CE in a population due to its lower sensitivity compared to field studies using imaging techniques such as portable ultrasound [14] , [25] , [26] . Therefore , we were not able to calculate the number of false negative and positive individuals in this specific area . Future research on CE in the Coquimbo region should include those tests . Using coproantigen testing we detected a 28% prevalence of canine infection in two municipalities of the Limarí province . A previous study reported a coproprevalence of 7 . 2% in dogs originating from both rural and urban settings from the Elqui province in the northern part of Coquimbo region [12] . These dogs could be at a lower risk of infection than those in this study and/or are regularly de-wormed by their owners . Nevertheless , the prevalence reported in this study is lower than that found in other rural areas in Latin America , for example in the central Andes of Perú ( 51% ) [25] and in Río Negro in Argentina ( 42% ) [27] . Two risk factors for E . granulosus infection of dogs were identified . The first , home slaughter , has been previously reported in Chile [12] and elsewhere [28] . The second factor was previous de-worming of dogs in the respective households . The significance of this finding is unclear; one explanation could be that households which deworm dogs actually have higher risk of parasitic infections and that treatment was irregular and inadequate , e . g . by using pyrantel , which does not eliminate E . granulosus tapeworms . Further studies are required to verify this finding such as cross-sectional studies with a larger sample size or longitudinal studies using different de-worming strategies [eg . 29] . In most areas of our study , community level risk factors for the persistence of the parasite within the environment were present such as home slaughter ( 60% ) , feeding dogs with cysts ( 21–45% ) , and high rates of echinococcosis in dogs ( 28% ) . In an endemic area such as the Limarí province , where the main factors for seropositivity are those linked to contact with dogs , it is extremely important for intervention activities to prioritize on the interruption of the chain of infection from dog to human . Thus , strategies should be based on education to promote proper hygienic measures such the management of waste , waste handling , washing hands , the use of plastic gloves when cleaning homesteads , reduction of dog grooming to prevent contact in a highly polluted environment and regular deworming of dogs , repeated at least every 45 days to be effective against E . granulosus [1] . This latter strategy is rarely adapted mainly due to economic constraints . Therefore an important intervention would be to increase the frequency of antiparasitic treatment of dogs by governmental sponsored programs , according to international criteria . Still , only a comprehensive program that includes various measures including education and animal management would allow disruption of the cycle of the parasite [30] . Currently , the Chilean Ministry of Health has taken these recommendations into account and instigated a control program with a focus on public education and de-worming and sterilization of dog populations . This latter measure could reduce the contamination of the environment particularly by young dog populations that are known from previous studies to shed large numbers of echinococcal eggs [7] . The results of our questionnaire survey showed that the crucial factors for the maintenance of the life cycle of E . granulosus were widely present throughout the rural areas of the Limarí province . In the Coquimbo region , hydatid disease is endemic mainly by the existence of large numbers of goats and sheep maintained in rural areas [31] . Due to poverty and poor animal health management , these areas provide ideal conditions for the maintenance of the life cycle of this parasite . A general limitation of our analysis of risk factors was that the questionnaire did not clearly identify current and past practices , a fact which might confound our results and interpretations . Cystic echinococcosis is a relevant public health and an economic problem worldwide [15] , [32]–[34] as well as in many areas of South America [8] including Chile . [21] . However , due to the lack of solid epidemiological data , difficulties in diagnosis and the chronic nature of infection and the complicated treatment required , it often has low priority and is therefore part of the group of neglected diseases [22] . The epidemiology of human cystic echinococcosis is complex and depends on the presence of the parasite in zoonotic cycles . Prevention and control of infection therefore requires careful mapping of regional epidemiological data and risk factors to tailor intervention strategies to specific situations . Accordingly , this study provides epidemiological data on prevalence and risk factors to both human and canine echinococcal infection that were determined at the household level . | Hydatidosis is a hyperendemic zoonotic disease in Chile caused by the dog tapeworm , Echinococcus granulosus . In Chile as in many other countries in South America , this disease has been largely neglected with few exceptions . Chile's growing economy and the interest of health authorities has lead to an increase in the number of studies investigating the epidemiology of echinococcosis and the factors related to infections of the main definitive host , the domestic dog and humans . In this study , we determined the prevalence of human and canine echinococcosis as well as the associated risk factors in a rural area of the Limarí province in northern Chile . We undertook a household questionnaire survey in rural areas of the five municipalities of the Limarí province in Coquimbo region . For each household serum of an adult family member and fecal samples from a dog were taken . Results of our study indicate that infection occurs in 2 . 6% of humans and 28% of dogs and is primarily due to feeding of dogs with contaminated offal and high dog-human contact . As a result of this study , the Chilean Ministry of Health instigated a control program aimed to control the infection in dogs and avoid new infections to humans . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"public",
"and",
"occupational",
"health",
"medicine",
"and",
"health",
"sciences",
"epidemiology",
"parasitology",
"biology",
"and",
"life",
"sciences",
"tropical",
"diseases",
"parasitic",
"diseases",
"veterinary",
"science"
] | 2014 | Prevalence and Risk Factors for Echinococcal Infection in a Rural Area of Northern Chile: A Household-Based Cross-Sectional Study |
Polyploidy , a state in which the chromosome complement has undergone an increase , is a major force in evolution . Understanding the consequences of polyploidy has received much attention , and allopolyploids , which result from the union of two different parental genomes , are of particular interest because they must overcome a suite of biological responses to this merger , known as “genome shock . ” A key question is what happens to gene expression of the two gene copies following allopolyploidization , but until recently the tools to answer this question on a genome-wide basis were lacking . Here we utilize high throughput transcriptome sequencing to produce the first genome-wide picture of gene expression response to allopolyploidy in fungi . A novel pipeline for assigning sequence reads to the gene copies was used to quantify their expression in a fungal allopolyploid . We find that the transcriptional response to allopolyploidy is predominantly conservative: both copies of most genes are retained; over half the genes inherit parental gene expression patterns; and parental differential expression is often lost in the allopolyploid . Strikingly , the patterns of gene expression change are highly concordant with the genome-wide expression results of a cotton allopolyploid . The very different nature of these two allopolyploids implies a conserved , eukaryote-wide transcriptional response to genome merger . We provide evidence that the transcriptional responses we observe are mostly driven by intrinsic differences between the regulatory systems in the parent species , and from this propose a mechanistic model in which the cross-kingdom conservation in transcriptional response reflects conservation of the mutational processes underlying eukaryotic gene regulatory evolution . This work provides a platform to develop a universal understanding of gene expression response to allopolyploidy and suggests that allopolyploids are an exceptional system to investigate gene regulatory changes that have evolved in the parental species prior to allopolyploidization .
Polyploidization refers to events that result in a sudden increase in the number of chromosome sets carried by an organism . Polyploidy is a major force in evolution , and has led to the emergence of new lineages in many major eukaryotic groups [1]–[8] . Unlike the incremental series of small changes that characterize the usual evolutionary process , polyploidization has the potential to form new species nearly instantaneously [9] . There are two classes of polyploidization: autopolyploidy is the duplication of a genome; while allopolyploidy is caused by interspecific hybridization between different species or genera resulting in the union of two or more dissimilar genomes . Such allopolyploids are often ecologically competitive , in many cases showing improved adaptability relative to parental species [10] . This is thought to arise from masking of deleterious mutations , fixed heterosis ( ‘hybrid vigor’ ) , and/or greater evolutionary plasticity resulting from the duplicated gene copies [11]–[13] . Genome shock describes changes in genome organization and behavior that occur in response to the sudden appearance of multiple genome copies [14] . Several manifestations of genome shock as a consequence of polyploidization are known , including gene loss , chromosome mis-pairing , transposon activation , altered methylation , and rearrangements between the genomes [1] , [11] , [15]–[21] . Gene loss has been particularly well studied in the hemiascomycetous yeasts , where substantial loss of gene duplicates has occurred following a whole genome duplication [22]–[24] . Gene loss has also been observed in a number of plant polyploids [2] , [25]–[27] suggesting that it is a general feature of polyploidy . Nevertheless , some plant polyploids , such as cotton , retain remarkably stable parental genome complements [28] , [29] . Moreover , different classes of genes are more prone to duplicate loss or retention following changes in ploidy [30] , although the overall trend is great malleability in genomic responses to polyploidization . Another feature of allopolyploids that has generated great interest is transcriptome shock , the sudden change in gene expression following the mixing of two dissimilar genomes , each with their own set of transcription factors and their own chromatin profiles [6] , [31] . To date , most studies that have examined the response of gene expression to allopolyploidization have focused on plant systems , ranging from evolutionarily old allopolyploidy events through to synthetic plant allopolyploids [32]–[40] . Two transcriptional phenomena are emerging from these studies [41]–[43] . The first is called “homeolog expression bias” ( here we use the term homeolog for the different parental copies of a gene within an allopolyploid; see Figure 1 ) . This refers to cases where the homeologs in the allopolyploid show an expression pattern different to that observed in the parents . The second phenomenon , termed “expression-level dominance” , is where a gene that shows a difference in expression between the two parents has a combined level of expression from the two homeologs in the allopolyploid that resembles one of the parental expression levels , rather than being an average of the two . However , because the single gene methods and probe-based assays ( such as microarrays ) traditionally used to study allopolyploid expression lack sufficient resolution to reveal the full suite of genome-wide gene expression patterns at an individual homeolog level [44] , the generality of these phenomena and how they are manifested has not been clear . High throughput mRNA sequencing technologies can overcome these limitations by resolving the expression levels of each homeolog in allopolyploids [45] . Recent studies have started to utilize these sequencing technologies to address allopolyploidization , and these are beginning to reveal the complex and multi-faceted transcriptional changes that occur during and after this event [39] , [45] , [46] . While much research has focused on plants , polyploidization is also an important feature of the evolutionary history of other lineages , including the fungi [47]–[49] . A model system for studying fungal allopolyploidy is the epichloë endophytes , an ecologically and economically important group of fungi from the family Clavicipitaceae [50] that includes the genera Epichloë and Neotyphodium . As systemic , obligate symbionts of cool-season grasses , the epichloë endophytes underpin the productivity of most global pastoral economies [51] . Most epichloë endophytes produce a range of biologically active secondary metabolites that can be economically beneficial by protecting against insect damage , but can also be economically detrimental by preventing mammalian herbivory , resulting in livestock productivity losses through toxic conditions called ryegrass staggers and fescue toxicosis . Interestingly , many species are natural polyploids [5] , [52] , which are usually referred to as “hybrids” in the fungal literature . Epichloë polyploids are often formed from very different parents , with nucleotide divergence between parent species reaching as high as 8% [49] , a level of divergence comparable to that between humans and macaques [53] . One well-characterized epichloë polyploid , designated as Lp1 , is an economically important asexual interspecific hybrid ( hereafter referred to as an allopolyploid to conform with the extensive plant literature ) between a haploid sexual species , E . typhina , and a haploid asexual species , N . lolii [50] , [54]–[56] . Although the mechanism of allopolyploidization is not known , it is thought to be similar to normal fungal mating , where cells first fuse to create a dikaryon ( a cell containing two nuclei ) , followed by nuclear fusion [49] . Lp1 cells have been shown microscopically to contain only a single nucleus [50] , proving that it is a true allopolyploid and not simply a fungal dikaryon , and because Lp1 is asexual , this nuclear arrangement is a permanent state . The allopolyploid nature of Lp1 is reinforced by genetic data: most genes studied to date in Lp1 have both parental copies maintained [50] , [54] , with the notable exception of the ribosomal DNA repeats ( rDNA ) and the mitochondrial DNA , each of which derive from just one parent [54] , [55] , [57] . The existence of species closely related to the original parents [50] , coupled with it being a naturally occurring allopolyploid , make Lp1 an ideal system to explore the consequences of allopolyploidy on transcription within the fungi . Here , we describe an RNA-seq based analysis of gene expression in the Lp1 allopolyploid . We develop a new computational pipeline to determine gene expression levels of both homeologs using next-generation mRNA sequencing in Lp1 without the benefit of a close reference genome sequence , and compare these to the parental species' transcriptomes . We document the four possible relative gene expression level outcomes for orthologs that become united in an allopolyploid , as well as a total expression level outcome , expression-level dominance ( Figure 1 ) . Our results show remarkable concordance of allopolyploid gene expression outcomes with those seen in plants , suggesting there are common transcriptional responses to allopolyploidy that reflect the underlying systems of gene regulation . We also show that almost all genes are retained in duplicate in Lp1 , implying that little gene loss has occurred since the polyploidization event , which we calculate occurred no more than 300 , 000 years ago .
To investigate the fate of gene expression following allopolyploidization in the well-characterized Neotyphodium lolii×Epichloë typhina allopolyploid endophyte , Lp1 , we performed Illumina mRNA sequencing on Lp1 and its putative parental species ( Figure 2 ) . The closest extant E . typhina strain is believed to be E8 [50] . Previous studies employed strain Lp5 as the closest N . lolii parent [e . g . , 57] . However , in this study we used N . lolii strain AR5 as the parental isolate because ( a ) it was isolated from the same germplasm as Lp1 ( B . Tapper , AgResearch New Zealand , pers . comm . ) , and ( b ) AR5 and the N . lolii component of the Lp1 genome are identical over a 430 bp region of the PYR4 gene ( results not shown ) . Replicate cultures of Lp1 , AR5 and E8 were grown on rich medium and then transferred to a defined nutrient-limited medium to stimulate transcription of a wide range of genes [58] . After 24 h of growth , mRNA was extracted from culture mycelium and sequenced using Illumina HiSeq sequencing . The numbers of reads obtained are shown in Table 1 . As a quick validation of data quality , reads were mapped to the E2368 reference genome . Mapping rates varied between 66–80% , with most unmapped reads subsequently identified as rRNA derived . To allocate Lp1 reads to the AR5- and E8-like homeologs , we developed a pipeline ( Figure S1 ) that utilizes an existing catalogue of well annotated gene models developed from the genome sequence of a related epichloë species , E . festucae strain E2368 [59] . We created two separate reference gene sets , one each for the AR5- and E8-like homeologs . To achieve this , we first identified AR5- and E8-specific SNPs by mapping reads from the two parental species to the E2368 gene models at low stringency . Relative to E2368 , we identified 44 , 665 and 336 , 802 SNPs for the AR5 and E8 parental species , respectively . This order-of-magnitude difference reflects the close evolutionary relationship between AR5 and E2368 , both of which have an E . festucae ancestry , compared to the more divergent E . typhina strain E8 . The median distance between SNPs in AR5 was 244 bp ( 95% confidence interval: 3–1 , 725 bp ) , while the median distance between SNPs in E8 was 22 bp ( 95% CI: 1–141 bp ) . Because Lp1 carries additional SNPs ( i . e . , polymorphisms that have arisen since the allopolyploidization event ) , we also mapped the Lp1 reads at low stringency to the E2368 gene models . 374 , 931 SNPs were identified , most of which ( 94% ) were shared with one or other of the parent species . Of the 22 , 543 new SNPs identified , 3 , 241 ( 14% ) could be assigned to either the AR5 homeolog ( 1 , 745 ) or the E8 homeolog ( 1 , 496 ) on the basis of linkage disequilibrium with known AR5 and E8 parental SNPs . All SNPs ( i . e . , those in both Lp1 and its parents ) were divided into diagnostic classes that are informative for determining homeolog ancestry ( Table 2 ) and these were used to create the two reference gene sets . After culling genes with insufficient sequencing coverage ( Protocol S1 ) , the reference gene sets contained 6 , 698 ( 55% ) of the 12 , 199 predicted gene models from the E2368 strain reference . We used this SNP dataset to genetically date the allopolyploidization event , with the results suggesting this occurred within the last 300 , 000 years ( see Text S1 ) . This first estimate of the age of Lp1 suggests that it is comparatively young , at least compared to the better-known autopolyploid event in hemiascomycetous yeast that occurred millions of years ago [22] . The two biological replicate datasets from Lp1 were mapped separately to the AR5- and E8-like reference gene sets with high stringency ( i . e . zero mismatch mappings ) to determine from which homeolog each read originates . This resulted in 12 , 283 , 690 reads mapping uniquely to the AR5-like reference , and 11 , 769 , 529 reads mapping uniquely to the E8-like reference . Previous investigations on just ten genes suggested that , apart from the special cases of rDNA and mtDNA ( see Text S1 ) , Lp1 contains both parent genomes with little evidence of gene loss [50] , [54] . To explore gene loss on a genome-wide scale , we determined the number of genes with reads mapped to the respective AR5- and E8-like homeologs . Of the 6 , 698 genes for which we are able to distinguish homeologs , at least one read mapped to 6 , 654 genes ( 99 . 3% ) , thus under our study conditions only 44 genes had no detectable expression of either homeolog in Lp1 . In addition , 35 genes showed expression solely from the AR5 homeolog , while 8 genes showed expression solely from the E8 homeolog . Therefore only 87 genes are candidates for gene loss events ( see later ) , suggesting that Lp1 has not experienced widespread gene loss following allopolyploidization . Genes that have the largest expression differences in Lp1 may represent biologically important functions that can shed light on the adaptive response of gene expression to allopolyploidization . We defined extreme differentially expressed ( EDE ) genes as those with very large expression differences in the allopolyploid – either a 50-fold or greater difference in expression level , or no expression from one of the parental homeologs . Excluding genes with fewer than 5 reads in the Lp1 dataset , 58 genes ( 0 . 9% ) fit these criteria . To test for common functions in EDE genes , we performed a gene ontology ( GO ) slim analysis to determine whether certain classes of genes identified in the analyses above were enriched for particular functional categories . No strong patterns were detected ( results not shown ) , although this outcome is tempered by the small number of genes analyzed . To determine whether EDE gene expression patterns result from allopolyploidy or instead derive from regulation differences that exist between the parents , we constructed a heat map showing the ratio of gene expression levels between homeologs in Lp1 and the corresponding orthologs in the AR5 and E8 parents ( Figure 3A and 3B ) . Expression ratios were determined in two ways . First , we calculated expression ratios between the two homeologs in Lp1 , and between the corresponding AR5 and E8 parental orthologs ( the two left-hand columns ) to determine whether the biased expression observed in the allopolyploid is also present in the parents . Second , we calculated expression ratios between the AR5 parental ortholog and the AR5 homeolog , and between the E8 parental ortholog and the E8 homeolog ( the two right-hand columns ) , to determine which homeolog is responsible for any expression change . Interestingly , about half the EDE genes where only one homeolog is expressed in Lp1 ( 21 of 41 using a 2-fold difference in expression as the cutoff ) are due to changes in the allopolyploid ( Figure 3A ) . A similar picture is seen for EDE genes with >50-fold expression difference between homeologs in Lp1 ( Figure 3B ) . In both cases most expression changes have occurred in the AR5 homeolog . For completeness , we also looked at the reciprocal situation: determining the fate of genes with high expression differences between the E8 and AR5 parents ( Figure 3C ) . Only 12 orthologs showed 50-fold or greater difference in expression between the AR5 and E8 parents ( our mapping approach automatically excludes any genes with expression in only one parent ) . Again , for about half of these genes ( 5/12 ) , the differential expression is lost in Lp1 . Loss of gene expression from one homeolog in the allopolyploid may result from chromatin changes or gene deletion , and may affect a broader genomic region than just a single gene . We therefore looked to see whether the EDE genes are physically linked in the reference genome [59] . Strikingly , 18 of 59 EDE genes ( 30 . 5% ) are found in five physically contiguous clusters of two or more genes ( Figure 3A , B ) . This pattern is particularly common in genes where only one homeolog is expressed in Lp1 , with almost half of these genes occurring in clusters . To distinguish between these regions becoming heterochromatinized , thus silencing blocks of genes , or being a result of genomic deletions , PCR-RFLP of genomic DNA was employed . We chose three examples of clustered EDE genes to investigate ( Figure 4 ) . In one region , a block of nine genes on Supercontig 98 all show exclusively E8 homeolog expression in Lp1 . PCR-RFLP analysis of the first and penultimate genes in this cluster shows that the AR5 homeolog has been deleted in both cases , and the most parsimonious interpretation is that a single , large deletion encompassing this entire block of genes has occurred in the AR5-derived genome . In the second example , a cluster of three genes is located on Supercontig 148 , two of which show E8 homeolog-specific expression and one shows highly biased E8 homeolog expression . PCR-RFLP analysis indicates that the AR5 homeolog of the middle gene in this block has been deleted . We propose that this three-gene cluster has been deleted from the AR5-derived genome , with some reads emanating from a fragment of the weakly-expressed AR5 homeolog that remains . Interestingly , both these deletion events are adjacent to AT-rich regions at the end of the contig , which are usually indicative of transposon-rich repeat regions in epichloë species [59] . The third example is a cluster of two genes on Supercontig 19 that show E8 homeolog-specific expression . PCR-RFLP analysis also indicates deletion of the AR5 homeolog . Additionally , PCR-RFLP on non-clustered EDE genes shows that some , but not all , are deleted ( Figure S2 ) . These results illustrate that gene loss accounts for some of the most extreme homeolog expression biases observed in Lp1 , and suggests that at least 20 genes have been deleted from Lp1 since the allopolyploidization event . The true number is likely to be larger , as biased homeolog expression from partial gene deletions is difficult to diagnose , and because our mapping procedure excluded about half the genes in the genome . This pattern of clustered gene loss resembles the chromosomal deletions observed in allopolyploid species of the plant genus Tragopogon [26] , [60] . Given the low level of gene loss , we next investigated the fate of gene expression following the allopolyploidization event on a genome-wide basis . The enormous number of reads in our study gave us great statistical power , meaning that for highly expressed genes a small proportional difference in read counts is statistically significant . Therefore we also employed a biological criterion of a 2-fold difference in expression between the AR5- and E8 homeologs ( after normalization for different numbers of reads per sample ) as a conservative cut-off for differential expression . Surprisingly , the majority of genes ( 4 , 515; 67 . 4% ) did not differ from the null expectation of equal expression of the two homeologs in Lp1 . The above analysis ignores the relative expression level of the parental orthologs . To compare gene expression between the allopolyploid and the parental species , we calculated the relative expression levels of orthologs from the parental transcriptome data ( cultured at the same time under identical conditions ) . We plotted the cumulative distribution of gene expression ratios between the AR5 and E8 homeologs in Lp1 , and between the parental orthologs , to reveal how expression differences between homeologs/orthologs are distributed ( Figure 5 ) . Figure 5A shows that the majority of homeologs in Lp1 are not differentially expressed . Furthermore , there is no trend towards higher expression of one homeolog over the other: there are similar numbers of genes where the E8 homeolog is dominant and where the AR5 homeolog is dominant . In other words , Lp1 exhibits balanced homeolog expression bias [43] . In contrast , comparison of the parental species shows that more orthologs have higher relative expression in E8 than AR5 ( Figure 5B ) , particularly genes that show relatively minor expression differences ( 2–10 fold ) . This pattern is offset by a small proportion of genes that have much higher relative expression in AR5 , thus balancing total normalized read counts between E8 and AR5 . Together , these results suggest that the predominant transcriptional response to allopolyploidy in Lp1 has been an overall reduction in differential homeolog expression relative to the parent species . To investigate the fate of gene expression in Lp1 further , we used the framework developed in cotton [41] , [46] , where genes are divided into different classes representing their transcriptional response following allopolyploidy . For this analysis , the key is a comparison of the expression ratio between the two orthologs in the parent species with that of the two homeologs in the allopolyploid , and thus looks at relative transcriptional responses . The E8-like gene can either show higher expression , lower expression , or similar expression to the AR5-like gene , both between orthologs in the parents , and between homeologs in Lp1 . The nine possible combinations of these states that derive from Table 3 of Yoo et al . [46] are shown in Figure 6 . We binned each gene into one of these nine categories , again using a 2-fold difference in expression ratio as the cut-off . To analyze these results we used the scheme of Yoo et al . , who grouped the nine possible combinations of states ( shown in Figure 6 ) into three broad outcomes for relative gene expression responses in the allopolyploid [46] . We put genes that show a reversal in the dominant homeolog into a separate group , thus giving four outcomes for gene expression that are defined in Figure 1 . The first outcome ( that includes the first three categories from Figure 6 ) is no change in expression patterns of AR5- and E8-like copies between the parents and the allopolyploid ( for example , if a gene has greater expression in one parent , this pattern is maintained in the allopolyploid ) , something we call “parental expression inheritance” . A majority ( 56 . 1% ) of genes display this behavior , with most being genes that show no expression difference between homeologs or orthologs . The second outcome contains genes displaying differential expression in the parents that has been lost in the allopolyploid . We call this ‘homeolog expression blending’ , and about one-quarter ( 25 . 1% ) of all genes show this pattern . The third outcome is the opposite pattern , homeolog expression bias , where differential expression has arisen in the allopolyploid [43] . The number of genes exhibiting homeolog expression bias is fewer ( 15 . 6% ) than those showing homeolog expression blending . These results again illustrate the tendency towards reduced expression bias in the allopolyploid , and corroborate the picture obtained from the global analysis above . The final outcome is where an expression bias in the parents has been reversed in the allopolyploid ( “homeolog expression reversal” ) . Unsurprisingly this outcome is relatively rare ( 3 . 2% of genes ) . We next wondered whether the genes with altered expression patterns in the allopolyploid are responding to natural selection . We performed a GO-slim analysis on genes in the four outcome groups described above , as well as for all genes showing allopolyploid differential expression ( Figure S3 and Text S2 ) . No strong patterns of gene classes preferentially changing their transcription levels were identified , suggesting that if selective forces have shaped the transcriptional response in Lp1 , it has only been for a small minority of genes . Recent studies have found evidence for expression-level dominance in allopolyploids , where genes that show differential expression between the parents have a total combined homeolog expression level that is similar to one or other of the parental levels ( the ‘dominant’ ortholog ) in the allopolyploid , rather than simply approaching the average of the orthologs [37] , [41] , [42] , [46] . To investigate whether this phenomenon also occurs in Lp1 , we calculated total expression for each gene in Lp1 that had 2-fold or greater expression difference in the parents , and classified these genes into three bins where allopolyploid expression more closely resembles: the highly-expressed ortholog; the lowly-expressed ortholog; or the mean of the orthologs ( Figure 7 ) . Almost three quarters of genes that show differential expression in the parents had an expression level similar to one ortholog ( the dominant ortholog ) . Interestingly , a similar number of genes matched the low expression ortholog as matched the high expression ortholog . Therefore , in keeping with other studies , we find that expression-level dominance is a major factor in the transcriptional response to allopolyploidy in Lp1 . Expression-level dominance can be achieved by various combinations of altered homeolog expression . In the one previous study where this was investigated , changes in the non-dominant homeolog were the driving force for most expression-level dominance [46] . To investigate the situation in Lp1 , we took all genes that exhibited expression-level dominance and determined how the two homeologs contribute to this pattern ( Figure 7 ) . About three-quarters of genes showing expression level dominance involve a change in expression of the non-dominant homeolog . Therefore , the majority of genes that were differentially expressed in the parents have undergone a more complex set of transcriptional changes than just simple inheritance of parental gene expression patterns .
Here we investigated the fate of gene expression in an established , but relatively young ( considerably less than 1 million year old ) , naturally occurring fungal allopolyploid . We developed a novel analytical pipeline that allowed us to utilize the power of high throughput mRNA sequencing to determine the expression levels of both homeologs in the allopolyploid on a genome-wide scale , and to compare these with the expression of the same genes in the closest known relatives to the parents . This analysis provides the first snapshot of global allopolyploid gene expression outside the plant kingdom , and illustrates a conservative transcriptome response to allopolyploidy , implying that genomic shock is largely buffered at the transcriptional level . The strongest gene expression pattern we found was fewer differentially expressed genes in the allopolyploid than the parents , with more than two-thirds of all the genes we investigated not being differentially expressed in the allopolyploid . This loss of differential expression is the net result of two phenomena: a larger proportion of genes that lose differential expression following allopolyploidization ( homeolog expression blending ) ; and a smaller proportion of genes that gain it ( homeolog expression bias ) . Therefore , at the global level , allopolyploidy is associated with conservative gene expression regulation , as has been observed previously [36] , [39] , [46] . Our analysis pipeline required us to remove around half the genes from the analysis , primarily because they were not ( or barely ) expressed in one or both parents , showed too little genetic variation to distinguish the two parental types , or were too short . This is likely to preferentially retain housekeeping genes and exclude genes that are expressed in response to specific environmental conditions . Given that housekeeping genes are likely to have predominantly conservative expression patterns , we may be over-estimating the proportion of genes that are not changing their expression levels in Lp1 , and under-estimating those with significant changes in expression . Our results are remarkably consistent with those of the pioneering RNA-seq study of cotton allopolyploids [46] , despite these fungal and plant allopolyploids being different ages , of different ploidies , from different eukaryotic kingdoms , and having been analyzed with different algorithms/analytical methods . As outlined in Table 3 , there are striking similarities in the proportions of genes in the major relative gene expression pattern outcomes we identified , including parental expression inheritance , homeolog expression blending , and homeolog expression bias . Additionally , the widespread expression-level dominance that we find , where combined total homeolog expression for a given gene in the allopolyploid is similar to that of one of the parental orthologs , is very similar to previously reported results [46] ( Table 3 ) . Also consistent with previous studies , we find that expression-level dominance is largely driven by transcriptional changes in the non-dominant homeolog , and that there is no particular bias towards dominance of the highly-expressed copy over the lowly-expressed copy [41] , [46] . Cross-kingdom similarities in gross transcriptional change were previously noted [61] , therefore we propose that the conserved transcriptional responses we observe are a general feature of allopolyploidy . Further RNA-seq studies from allopolyploids across the tree of life will be required to determine how general these allopolyploid transcriptional responses are . We document only a small amount of gene loss in Lp1 . Most losses involve the AR5 copy , contrary to the general trend of greater relative expression of AR5 homeologs within Lp1 , suggesting that the E8-derived genome may have remained relatively inert . While our results do not address the mechanism behind gene loss , the proximity of several deletions to AT-rich regions implicate polyploidization-induced transposon activity in these deletions , as AT-rich regions are often associated with transposable elements in epichloë genomes [59] . Previous studies have shown that loss of duplicates is a routine phenomenon during and following polyploidy [30] , and the low level of loss suggests that Lp1 is either a very young allopolyploid , or recalcitrant to gene loss ( a property displayed by cotton allopolyploids [29] ) . Our current data do not allow us to resolve this question , as while we estimate the maximum time for allopolyploidization , we cannot provide a lower bound on its age . The strongest evidence that Lp1 is an established allopolyploid is the homogenization of the rDNA to a single parental type ( Table S1 and Text S1 ) , which is likely to have required a number of generations [62] , [63] . The GO-slim analyses we performed provided little support for selection having played a major role in shaping the expression patterns of most genes in Lp1 . If these expression patterns are overwhelmingly not the result of strong gene-specific selective forces , what is causing them ? We suggest that our results are primarily explained by intrinsic gene regulation factors [32] , [64] , [65] , as originally proposed by Roose and Gottlieb [66] . Gene expression is the net outcome of a complex interplay of cis- and trans-acting factors [67] , [68] , including chromatin structure , transcription factors , effectors , RNAi , and nuclear position , that we collectively term a ‘modulon’ ( Figure 8 and Figure S4 ) . Changes anywhere in a modulon can impact gene expression , be it cis-acting changes ( e . g . mutations in the promoter ) , epigenetic changes , and/or changes in the constellation of trans-acting factors . Expression differences between orthologs in the parents result from differences that were established in the modulon systems as the two parents evolved separately following speciation , while orthologs that are not differentially expressed are either governed by the same modulon system in both parents or coincidentally achieve the same expression through different modulon systems . Under this framework , the allopolyploid expression patterns we observe can be explained as follows: homeologs that faithfully inherit expression differences from their parents have modulon systems exhibiting little cross talk between the homeologs ( Figure 8A ) . Conversely , genes that show homeolog expression blending result from modulon systems that partially or fully cross talk between homeologs ( Figure 8B ) . Finally , genes that gain expression bias following allopolyploidy result from modulon systems that preferentially regulate one homeolog over the other , resembling classical dominance ( Figure 8C ) . Therefore we propose that the allopolyploid gene expression patterns we observe are predominantly the net outcome of the modulon features that existed in the parents , although it will be important to distinguish the effects of genome doubling [69] . If these patterns of allopolyploid expression fate are predominantly the result of regulatory differences present in the parents , they open a window into the evolution of gene regulation following speciation [61] , [66] . Genes where the homeologs have faithfully inherited expression differences from the parents have developed full transcriptional independence in the time since the two parents speciated , whereas those showing homeolog expression blending have not evolved independence since speciation . We can use the results from Figure 6 to estimate the proportions of genes that have developed independence , partial independence and have not developed independence via the number of genes that show simple inheritance of biased expression , homeolog expression bias , and homeolog expression blending , respectively . Depending on how we treat genes with non-biased expression in both the parents and the allopolyploid ( see ) , our results suggest that 13–28% of genes have evolved independent regulation , 19% have evolved partial independence , and 53–68% have not evolved independence . The results presented in Figure 7 can also be used to estimate these proportions , as the first and third rows represent independent regulation , and the second and fourth rows represent partial and no independent regulation ( combined ) . These totals ( 25 . 9% and 74 . 1% respectively ) both fall within the ranges estimated from Figure 6 . We are proposing that the transcriptional responses to allopolyploidy are largely a passive outcome of regulatory evolution that has occurred in the parents following speciation . In principle , they could also be explained by more saltational events that are specific to the allopolyploid , such as genome-wide chromatin reprograming [70] . However , we believe that the striking cross-kingdom conservation in transcriptional responses supports the view that the majority of these responses are outcomes of parental regulatory evolution . It is not clear why saltaic mechanisms should result in conserved patterns of transcriptional responses between species with highly diverged genome structures , global expression regulators , and chromatin networks . However , if these conserved gene expression response patterns largely result from mutations accumulating in modulons over time , then a roughly monotonic increase in the proportions of genes that evolve partial and full transcriptional independence would be expected ( Figure S5 ) . One prediction of this hypothesis is that the more genetically diverged the parents of two allopolyploids are , the more different the allopolyploid transcriptional responses will be . Clearly more allopolyploid transcriptome studies are required to determine whether this is true and how conserved the transcriptional responses are . It will also be of great interest to determine whether similar genes evolve similar regulatory patterns in different allopolyploids , and how the effects of homeolog interference influence the transcriptional response [71] . In conclusion , we present the first study of global gene expression in a fungal allopolyploid species . We show that most genes are still retained in duplicate , suggesting that Lp1 is either a young allopolyploid or is resistant to the gene loss process that often accompanies allopolyploidization . We find a mixture of expression patterns , with the homeologs for many genes retaining the gene expression patterns seen in the parents , fewer showing less biased expression than seen in the parents , and fewer still developing biased expression . Strikingly , these expression patterns are remarkably concordant with those recently ascertained for allopolyploid cotton , suggesting there exists a general pattern of interspecific allopolyploid gene expression fate that is largely independent of taxonomic kingdom , gene repertoire or local environment . We conclude that the transcriptional response to allopolyploidy is conservative and conserved , reflecting the stochastic nature of genetic regulatory evolution . Our work suggests that the fate of allopolyploid gene expression follows general principles that apply across eukaryotes , and that allopolyploid transcriptomes are a novel and powerful way to unmask the regulatory changes that evolve following speciation .
Three filamentous fungi from the ascomycete family Clavicipitaceae were used in this study: the asexual , diploid interspecies fungal allopolyploid Neotyphodium lolii×Epichloë typhina Lp1 ( syn . AR6 ) [50] , [72] , the haploid asexual species Neotyphodium lolii AR5 , and the haploid sexual species Epichloë typhina E8 . Cultures were grown in 2 . 4% potato dextrose ( PD ) media until maturity , washed twice with double distilled water , then resuspended in a defined medium ( CDGN ) comprised of Czapek Dox salts [73] containing 100 mM glucose and 10 mM ammonium sulphate . The fungal cultures were filtered and washed , and ∼100 mg of mycelium ( wet weight ) per sample was used for RNA extraction , which was performed using an RNeasy Plant RNA extraction kit ( Qiagen ) according to the manufacturer's instructions . Total RNA samples were treated with DNAse I to remove contaminant DNA , mRNA was extracted using polyA selection , and Illumina sequencing libraries were prepared using standard protocols . Libraries were sequenced on the HiSeq 2000 – 100 bp single end sequences for AR5 and E8 , and 100 bp paired end sequences for Lp1 . Two biological replicates were sequenced for each . Bases were called using CASAVA ( v . 1 . 7 . 0 , Illumina , Hayward , CA , USA ) . Sequencing quality control was performed using the SolexaQA package ( v . 1 . 10 ) [74] . In all six datasets , >79% of bases were sequenced to Q30 or higher ( i . e . , bases have a probability of error , P<0 . 001 ) . Full details of the analyses that were performed to process the sequences , to create the reference gene sets that allowed us to bin Lp1 reads as coming from the AR5-like and E8-like homeologs , and to perform these allocations , are described in Figure S1 , Figure S6 and Protocol S1 . Briefly , two E8 and AR5-like reference gene sets were created by modifying an existing annotated set of gene models from the closely related Epichloë festucae E2368 using SNPs that were generated from the parental and Lp1 transcriptome data . Lp1 reads were then mapped with high stringency to the informative sites of these two reference gene sets to determine from which parental copy each Lp1 sequence read derived . Genes that had fewer than five reads in one or both of the parental transcriptomes were excluded from both E8 and AR5 reference gene sets , as were genes where the informative regions were less than 150 bp . The statistical significance of gene expression differences between AR5-like and E8-like homeologs in the Lp1 allopolyploid data was determined using Fisher's Exact Test [75] as implemented in the R [76] package DEGseq v . 1 . 8 . 0 [77] . A correction for multiple testing was applied using the False Discovery Rate ( FDR ) approach described by Storey and Tibshirani [78] . The fold difference for each gene i was calculated as ( 1 ) where gene counts were converted to ‘reads per million’ to normalize for unequal numbers of reads between biological replicates , as well as unequal numbers of reads mapping to the AR5-like and E8-like homeolog references . Subsequent analyses were performed in Excel . Differential gene expression was defined throughout this work as statistical significance and a 2-fold or greater difference in normalized gene counts , except where otherwise noted . Gene ontology analysis was performed as described in Text S2 . RFLPs were identified in candidate genes using the read data in IGV [79] , and primers were designed to flank them ( Table S2 ) . Genomic DNA was extracted , using the protocol of Peintner and colleagues [80] , from the same material used to make the RNA . End-point PCR was performed using standard procedures with Ex-Taq polymerase ( Takara ) . PCR products were precipitated with isopropanol and then digested with the appropriate restriction enzyme according to manufacturers' instructions . The products were run on 2% agarose or 8% native polyacrylamide gels . Approximate estimates of lineage ages were made using a mutation rate of 1×10−9/site/year [81] and a cumulative haploid sequence length of 7 , 123 , 190 bp in the final masked AR5-like and E8-like homeolog references . See Text S1 , Figure S6 and Figure S7 for details . | Organisms are complex biological systems that must continue to function even as their genomes evolve . While evolution is usually gradual , the formation of new species by the hybridization of different parents—allopolyploidization—occurs nearly instantaneously . A key question is what happens to expression of the two parental gene copies following genome merger . To determine this , we focused on a fungal allopolyploid from a group that dominates many of the world's pastoral economies . To investigate the fate of gene expression in this system , we developed a novel pipeline to assign high throughput RNA sequence reads to the two parental gene copies , thus allowing quantification of expression . We found transcriptional responses to be predominantly conservative: most gene copies either inherit parental expression patterns , or if differentially expressed in the parents , that difference is lost in the hybrid . Moreover , we identified an extraordinary level of concordance in the fate of genome-wide allopolyploid gene expression with that seen in cotton . The very different nature of these two allopolyploids suggests that there is a set of universal rules for the transcriptional response to genome merger . We propose a mechanistic model whereby this conserved response reflects similarities in mutational processes that underlie gene regulatory evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome",
"expression",
"analysis",
"mycology",
"fungi",
"gene",
"expression",
"genetics",
"gene",
"regulation",
"molecular",
"genetics",
"biology",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"microbiology",
"computational",
"biology",
"dna",
"transcri... | 2014 | An Interspecific Fungal Hybrid Reveals Cross-Kingdom Rules for Allopolyploid Gene Expression Patterns |
Mimivirus is the largest known virus whose genome and physical size are comparable to some small bacteria , blurring the boundary between a virus and a cell . Structural studies of Mimivirus have been difficult because of its size and long surface fibers . Here we report the use of enzymatic digestions to remove the surface fibers of Mimivirus in order to expose the surface of the viral capsid . Cryo-electron microscopy ( cryoEM ) and atomic force microscopy were able to show that the 20 icosahedral faces of Mimivirus capsids have hexagonal arrays of depressions . Each depression is surrounded by six trimeric capsomers that are similar in structure to those in many other large , icosahedral double-stranded DNA viruses . Whereas in most viruses these capsomers are hexagonally close-packed with the same orientation in each face , in Mimivirus there are vacancies at the systematic depressions with neighboring capsomers differing in orientation by 60° . The previously observed starfish-shaped feature is well-resolved and found to be on each virus particle and is associated with a special pentameric vertex . The arms of the starfish fit into the gaps between the five faces surrounding the unique vertex , acting as a seal . Furthermore , the enveloped nucleocapsid is accurately positioned and oriented within the capsid with a concave surface facing the unique vertex . Thus , the starfish-shaped feature and the organization of the nucleocapsid might regulate the delivery of the genome to the host . The structure of Mimivirus , as well as the various fiber components observed in the virus , suggests that the Mimivirus genome includes genes derived from both eukaryotic and prokaryotic organisms . The three-dimensional cryoEM reconstruction reported here is of a virus with a volume that is one order of magnitude larger than any previously reported molecular assembly studied at a resolution of equal to or better than 65 Å .
Mimivirus , Acanthamoeba polyphaga Mimivirus , is the largest known virus [1–3] and a putative human pneumonia agent [4] . It has an icosahedral shape with a 0 . 75-μm diameter [3] and a ∼1 . 2-Mbp genome that contains most of the genes found in small bacteria [5] . The external morphology of Mimivirus had initially led to its false identification as a bacterium [1 , 4] . Initial cryo-electron microscopy ( cryoEM ) studies [3] had shown that Mimivirus has a diameter of about 5 , 000 Å , with multiple layers of proteins and lipid membranes that surround a nucleocapsid . In addition , there is a dense layer of 1 , 250-Å-long fibers that cover the viral surface , making the total diameter of the particles about 7 , 500 Å . The outermost layer of the capsid is about 70 Å thick and corresponds to the major capsid protein ( MCP ) . There is an irregularly shaped nucleocapsid , which itself is enveloped by a 70-Å-thick layer , and is separated from the capsid by a distance that varies from 300 to 500 Å . Thus , the large size of Mimivirus , its gene content , and its functional complexity as described here and elsewhere [2–6] stretch the definition of a virus [7] . The capsomer structures of some large double-stranded DNA ( dsDNA ) viruses—including adenovirus [8] , Paramecium bursaria Chlorella virus 1 ( PBCV1 ) [9 , 10] , the bacteriophage PRD1 [11] , Sulfolobus turreted icosahedral virus [12] , and the marine bacteriophage PM2 [13]—have been determined by x-ray crystallography and shown to be similar . Although these viruses infect a wide variety of hosts covering the prokaryotic , eukaryotic , and archaeal domains of life , the similarity of their MCP structures suggest that they have , in part , evolved from a common precursor [9 , 12 , 14] . Each monomer in the trimeric capsomers consists of two successive “jelly-roll” folds , producing a pseudo-hexameric structure with a thickness of ∼75 Å and a diameter varying between 74 Å in PBCV1 [9 , 10] and about 85 Å in adenovirus [8] . One or other of the two jelly-roll motifs within the monomer often has a large insertion in the DE and FG loops ( the β strands along the polypeptide of each jelly-roll are named A to H ) ( Figures 1 and 2 ) , creating a “tower” on top of each of the three monomers within a capsomer . These towers give capsomers a triangular appearance on the surface while maintaining a pseudo-hexagonal shape below the towers , appropriate for packing into hexagonal arrays [8 , 15 , 16] . The Mimivirus MCP is homologous to the MCP of PBCV1 with 31% amino acid identity ( Figure 2 ) . Therefore , it is highly likely that the structure of Mimivirus capsomers are similar to the aforementioned capsomers in large dsDNA icosahedral viruses [9 , 12 , 14] . However , in Mimivirus , there are about 190 additional amino acids inserted into the DE loop of the second jelly-roll motif that are similar to the large tower insertions in adenovirus ( Figures 1 and 2 ) . The forest of long fibers on the Mimivirus surface increases the ice thickness , creating difficulties for cryoEM [3] . The random scattering of the electrons by the additional ice thickness and by the disordered fibers reduced the signal-to-noise ratio . Here we report that we were able to partially overcome this problem by digesting the fibers with lysozyme and proteases . Both atomic force microscopy ( AFM ) and cryoEM were then used to analyze the structure of untreated as well as defibered Mimiviruses . The viral capsid surface was found to have a hexagonal array of depressions separated by about 140 Å . These were interpreted as systematic vacancies within a hexagonal array of double jelly-roll capsomers , accounting for the absence of one-third of all capsomers . Furthermore , the previously recognized starfish-like feature [17] was well-resolved and associated with a unique pentameric vertex on mature particles below the forest of surface fibers . We also show that the Mimivirus nucleocapsid has a defined shape surrounded by an envelope , which is separated from the viral capsid by a space whose size and dimensions are conserved in all particles .
A large number of cryoEM particle images were collected to improve the previously computed [3] , icosahedrally averaged , three-dimensional reconstructions of Mimivirus . However , increasing the number of images beyond about 30 , 000 failed to show the anticipated hexagonal arrays of capsomers as found in PBCV1 and other related dsDNA viruses [12 , 18 , 19] . Hence , AFM was used in an endeavor to obtain better-resolved structural information . AFM images of defibered Mimivirus showed hexagonal arrays of depressions covering the surface of the virus ( Figure 3A ) and a starfish-like structure associated with one of the vertices on many of the particles ( Figure 4 ) , as had also been observed on some previous EM micrographs [17] . The presence of a structural feature on only one of the 12 vertices demonstrated a significant departure from icosahedral symmetry . Therefore , further reconstructions were based on only 5-fold , rather than icosahedral symmetry ( see Materials and Methods ) , using about 700 lysozyme- and protease-treated defibered virus particles . This reconstruction clearly showed a unique pentameric vertex with a starfish-like attachment , but failed to visualize the hexagonal arrays of depressions seen on the AFM images . Because about 31 , 000 images of the mature fibered particles had been collected , a further reconstruction was calculated—using these particles and assuming only 5-fold symmetry—which was initialized with the newly reconstructed model from the defibered particles . The resultant 65 Å resolution cryoEM map of Mimivirus showed that surface depressions , separated by 140 Å , were arranged in hexagonal arrays ( Figure 3B ) , which was consistent with the AFM observations . Each equilateral triangular face of the virion consisted of 19 rows of depressions parallel to each edge , with each row containing one less depression than the previous row . The Mimivirus genome [5] contains four genes , including L425 and R441 , that are homologous to the double jelly-roll PBCV1 Vp54 , and to the MCPs of other large dsDNA viruses [14] . Although a homology model of the R441 gene product was built by Benson et al . [14] , the actual MCP of Mimivirus was found to be the gene product of L425 [5] . The limited resolution of the cryoEM reconstruction barely resolves individual capsomers , but the array of large depressions suggests that these are missing capsomers ( vacancies ) in the hexagonal arrays of PBCV1-like capsomers . There is one 190-amino-acid-long insertion in the DE loop of the second jelly-roll along the polypeptide of the Mimivirus MCP ( Figures 1 and 2 ) , which is located on the external edge of each of the three monomers in a capsomer . The systematic vacancies in Mimivirus could arise as a consequence of steric conflict between these insertions in three neighboring capsomers and would be relieved by creating the systematic vacancies . Each of the depressions on the Mimivirus surface is surrounded by six barely resolved triangular shapes ( Figure 3C–3E ) , which are similar in appearance to the triangular external surface of capsomers in other viruses with double jelly-roll MCPs [8 , 15 , 16] . The orientations of neighboring trimeric capsomers surrounding each depression differ by about 60° , thus generating a 6-fold symmetry axis in the center of each depression ( Figure 3C–3F ) . However , the trimeric shapes are barely resolved from each other so that each of the three “towers” that form the triangular shape at the top of a capsomer merges with the towers of the neighbouring capsomers ( Figure 3G ) . A simulation using the known PBCV1 capsomer structure [9] , assembled into hexagonal arrays as found for Mimivirus , demonstrated that the proposed arrangement mimics the observed pattern of depressions with poorly resolved surrounding trimeric capsomers at the resolution attained for the Mimivirus reconstruction ( Figure 3F ) . Given that the distance between depressions is 140 Å , the center-to-center distance between adjacent triangular capsomers will be 81 Å ( Figure 3G ) . This is in the range expected for trimeric capsomers assembled from double jelly-roll monomers [10] .
To our knowledge , the arrangement of protein subunits in an icosahedral capsid was first discussed by Crick and Watson [20] . Their concepts were extended by Caspar and Klug , who suggested that arrays of hexagonal capsomers could be interspersed with pentameric capsomers at the icosahedral 5-fold vertices , resulting in only quasi-equivalent environments for monomers at the 5-fold vertex compared with those in hexagonal arrays [21] . The organization of pseudo-hexameric capsomers in large dsDNA icosahedral viruses , for which the triangulation number ( T ) expresses the number of jelly-rolls rather than monomers in the icosahedral asymmetric unit , is , therefore , a further extension of the concept of quasi-symmetry . If all the depressions were filled by capsomers , and as there are 19 depressions between neighboring pentameric vertices , the coordinates of the nearest vertex would be h = 19 ± 1 and k = 19 ± 1 , where h and k are the number of capsomers along the hexagonal axes of the array ( Figure 3G ) . The uncertainty arises because it is not clear whether there is a depression or a capsomer on each pentameric vertex . Thus , the triangulation number , given by T = h2 + hk + k2 [21] , would be 3 × ( 19 ± 1 ) 2 or have one of nine possible values in the range of 972 ≤ T ≤ 1200 . The previously predicted value of around 1 , 180 jelly rolls [3] was based on an estimate for the center-to-center distance between capsomers being 75 Å . The above observations show that this distance is 81 Å , which would have given T = 1 , 012 , which is still within the range of the above determination . However , this statement further extends the definition of T , because it not only considers the depressions being filled by capsomers , but also tacitly assumes that all the capsomers are similarly oriented . The p6 plane group arrangement of capsomers in Mimivirus allows 3/2 times as much area per capsomer—compared with a completely filled hexagonal array of capsomers in a p3 plane group—as , for instance , in PBCV1 ( Figure 3G ) . Thus , the actual number of jelly-rolls will be 2T/3 , and the number of capsomers per icosahedral asymmetric unit will be T/9 or about 120 for Mimivirus . The p6 plane group organization of the capsomers in Mimivirus is essentially the same as that of trimeric “packing units” observed by cryoEM for infectious bursal disease virus ( IBDV ) [22 , 23] which has a T = 13 ( h = 1 , k = 3 ) surface lattice . The structure of the IBDV major capsid protein has been determined [24] and shown to have three domains ( B , S , and P ) , of which the S and P domains have jelly-roll folds . However , the domain organization within the IBDV trimeric capsomers is different to the pseudo-hexagonal capsomer structures found in PBCV1 and some other large dsDNA viruses . Thus , although the p6 organization of capsomers in Mimivirus resembles the capsid of a dsRNA virus , the amino acid sequence of the major capsid protein of Mimivirus has greatest similarity to other dsDNA viruses such as PBCV1 . CryoEM studies of Mimivirus recognized that some particles had a special vertex [3] . More recently , transmission electron microscopy ( TEM ) of sectioned Mimivirus-infected amoeba found a starfish-shaped feature associated with one vertex on many Mimiviruses [17] . Starfish-shaped density features were also observed with cryoEM on some fiberless immature Mimivirus particles that occurred in purified samples [17] . Here we show , using AFM , that a starfish-shaped feature can be seen on many defibered Mimivirus particles ( Figure 4 ) . Furthermore , the 5-fold–averaged cryoEM reconstruction of Mimivirus was initiated with a simplified model ( see Materials and Methods ) that did not have a starfish-shaped feature . However , the resulting reconstruction ( Figure 5 ) showed a starfish-shaped feature similar to what was observed with AFM ( Figure 4C ) and confirmed the existence of the starfish-shaped feature on each virus . The 5-fold–averaged cryoEM results showed that the arms of the starfish have a thickness of about 400 Å , a width of about 500 Å , and extend about 2 , 000 Å almost all the way towards the neighbouring 5-fold vertices . The exceptional clarity of the starfish-shaped feature in cryoEM reconstructed map ( Figure 5 ) demonstrated that it must exist on almost every fibered particle . Both AFM and cryoEM showed that the arms of the starfish are inserted and open a gap between the neighbouring faces that are associated with the special vertex ( Figure 4D ) . The five faces associated with the special vertex are inclined by about 5° to what would be expected if the virus were completely icosahedral , accounting for the gap between faces ( Figure 5E ) . The arms of the starfish-shaped feature do not show the hexagonal arrays of depressions ( Figure 4D ) , suggesting that the starfish-like feature is not assembled from the MCP . Evidence for the starfish-shaped feature being a separate entity was also found in cryoEM images of defibered Mimivirus samples in which there were objects that had five arms of appropriate size radiating from a common center ( Figure 6A ) . CryoEM images of thin sectioned samples [17] and AFM images of mature Mimivirus ( Figure 4B ) showed that there are star-shaped crevices between the long surface fibers , implying that the starfish-shaped feature is not covered by fibers . It had been suggested that the “starfish”-associated vertices might be the portal for DNA release based on its location further from the associated virus factory [17 , 25] . Thus , if the long cross-linked fibers of Mimiviruses [3] were to cover the complete viral surface , they would be an obstacle for genome delivery into a host . However , the star-shaped crevice between the fibers could provide an exit portal for the genome . Scanning electron microscopy [17] , traditional TEM of thin sections [25] , and cryoEM studies ( Figure 6B ) show that defibered particles missing the starfish-shaped feature are associated with membrane-like “puffs” at their special vertices . Furthermore , cryoEM showed that particles that had lost their genome ( Figure 6C ) had also lost the starfish-shaped feature . In addition , AFM showed that the ejected DNA is unprotected by any surrounding proteins ( Figure 6D ) . Thus , the starfish-shaped feature might be acting as a seal to hold together the five faces associated with the special vertex . Therefore , the first step of genome delivery would be the release of the starfish-shape feature , allowing the DNA to exit through the special vertex . Special vertices for genome delivery have also been observed in some other large dsDNA viruses [26 , 27] , in tailed bacteriophages [28–31] , and in herpes virus [32] . The presence of a special vertex in tailed bacteriophages or in herpes virus whose MCPs have a HK97-like fold [33] or in viruses that have a double jelly-roll fold in their capsids , suggests convergent evolution to a common solution for genome delivery . AFM images show that a number of external fibers of Mimivirus are frequently attached to a single central feature at one end with their free end being associated with a globular terminus ( Figure 7A and 7B ) . However , there is no indication where the fibers attach to the capsid on the viral surface . The surface fibers are resistant to proteases unless first treated with lysozyme , suggesting that the fibers are protected by peptidoglycan ( as previous suggested [5] ) , which is consistent with Mimivirus being Gram-positive [1 , 4] . CryoEM images of Mimivirus that had been partially treated with bromelain show successive rings of density on the fibers separated by 200–500 Å , representing different structural segments along their lengths ( Figure 7C ) . AFM images show murky material surrounding the fibers ( Figure 7B ) that might be peptidoglycan cross-linking neighboring fibers . Fibers with peptidoglycan components perhaps act as a decoy for attracting amoeba [34] . The central slice of the cryoEM reconstruction , perpendicular to the unique 5-fold axis , showed that the genome is surrounded by a membrane-like envelope ( Figure 5F ) . A central slice , containing the unique 5-fold axis , showed that the nucleocapsid had a concave depression facing the “starfish”-associated vertex ( Figure 5E ) , which suggests a specialized organization that might be required for host infection . The clarity of these features after five-fold averaging implies that the nucleocapsid has a defined shape and also a fixed position relative to the external capsid . Unlike many other viruses in which the genome is closely surrounded by the capsid , Mimivirus has a 300–500 Å gap between the enveloped genome and the outer capsid . Thus , there must be supports across the gap that accurately position the genome relative to the viral capsid and internal membrane , although apparently they are too few or lack symmetry to make them visible in the cryoEM reconstruction . Long internal fibers were observed by AFM after applying mechanical force to the virus that broke the outer capsid layers ( see Materials and Methods ) . These internal fibers have a diameter of about 60 Å with repeat units at intervals of about 70 Å ( Figure 7D ) . The nucleocapsid might be supported by these fibers but , at this time , there is no further evidence for this suggestion . The enveloped genome within the larger viral capsid , perhaps supported by fibers ( Figure 7D ) , has some similarity to eukaryotic cells . In contrast , the external peptidoglycan component mimics bacterial cell walls ( Figure 7A–7C ) . In addition , the existence of a unique vertex in Mimivirus , possibly for genome delivery [17 , 25] , is reminiscent of tailed bacteriophages . These observations are consistent with other results [2 , 35] , implying that Mimiviruses and some other large icosahedral dsDNA viruses have gathered genes from eukaryotic , prokaryotic , as well as archaeal origins . The three-dimensional cryoEM reconstruction reported here , which was made possible in part by relaxing the icosahedral symmetry , is of a virus whose volume is an order of magnitude larger than has previously been reported . Thus , the detection of a unique vertex may have been missed in other structural studies in which strict icosahedral symmetry had been imposed [36] .
The Mimivirus fibers were digested by sequential application of lysozyme and bromelain . The Mimivirus was pelleted by centrifuging at 1 , 000g for 30 min . Each volume of pelleted virus was incubated with four volumes of 10 mg/ml lysozyme in TES buffer ( 0 . 05 M N-[Tris ( hydroxymethyl ) methyl]-2-aminoethanesulfonic acid , pH 7 . 5 , 0 . 01% NaN3 ) at room temperature for at least one day . The sample was washed twice with TES and digested with five volumes of 14 mg/ml bromelain from pineapple stem ( Sigma ) in TES buffer ( 0 . 035M TES , pH = 7 . 5 , 0 . 3M KCl , 0 . 02M DTT ) at room temperature for at least one day . CryoEM data of untreated and defibered Mimivirus were collected as described previously [3] . Micrographs were scanned on a Nikon Coolscan 9000 with a final pixel size of 15 . 9 Å . The cryoEM reconstruction was performed assuming 5-fold symmetry using programs FREALIGN [37] and a modified version of XMIPP [38] ( V . A . Kostyuchenko et al . , unpublished data ) . The reconstruction was initiated with a model in which the density of the five faces around one pentameric vertex were pushed outwards along the associated icosahedral 5-fold axis by about 300 Å . Of a total of 1 , 378 boxed defibered Mimivirus particles , 691 were selected to produce a map of 120 Å resolution . The resolution was determined using a Fourier shell correlation threshold of 0 . 5 . The map shows a clear starfish-shaped feature . This map was used as a starting model for reconstruction of untreated , fibered Mimivirus . Of a total of 53 , 640 boxed fibered Mimivirus particles , 30 , 919 were selected to achieve a 5-fold-averaged reconstruction with a resolution of 65 Å . The cryoEM map has been deposited with the EBI and has been given the accession number of EMDB 10623 . Mimivirus particles , both native and those treated with enzymes , were spread on freshly cleaved mica that was coated with poly-l-lysine and scanned under buffer . Capsids , which were pretreated with lysozyme and bromelain , were , in some experiments , further exposed to 1 mg/ml solutions of proteinase K and 1% SDS at 37 °C for 30 min to 2 h , washed with water , and then imaged . No fixation of any kind was used . Two methods were used to expel the DNA and other internal structures from the virus . In the first method , virus solution was dried on mica , rehydrated with a small amount of water , and then pressed between two surfaces of mica . In the second method , very concentrated virus solution was placed in small wells , crushed with a glass stick , diluted in water , and then deposited on mica . The Mimivirus DNA was recognized by comparing the AFM images with DNA extracted from PBCV1 [16] , T4 phage , vaccinia viruses [39] , plasmid DNA , and thymus DNA . All these images had the same tangled appearance and had the same height ( thickness ) above substrate . Furthermore , the material could not come from the host or some other source , because almost all the images show these fibers to be closely associated with isolated virions , not just spread out randomly on the substrate . AFM analysis was carried out using a Nanoscope III multimode instrument ( Veeco Instruments ) . Samples were scanned at 25 °C using oxide-sharpened silicon nitride tips in a 75-μl fluid cell containing buffer or in air . For scanning in air , silicon tips were used . The images were collected in tapping mode [40] with an oscillation frequency of 9 . 2 kHz in fluid and 300 kHz in air , with a scan frequency of 1 Hz . Procedures were fundamentally the same as described for previous investigations of viruses [16 , 41] . In the AFM images presented here , height above substrate is indicated by increasingly lighter color . Thus , points very close to the substrate are dark and those well above the substrate are white . Because lateral distances are distorted due to an AFM image being a convolution of the cantilever tip shape with the surface features scanned , quantitative measures of size were based either on heights above the substrate or on center-to-center distances on particle surfaces . The AFM instrument was calibrated to the small lateral distances by imaging the 111 face of a thaumatin protein crystal and using the known lattice spacings [42] as standard . | Mimiviruses are larger than any other known virus , yet despite their size , the capsid has been shown to be a regular icosahedron . Using cryo-electron microscopy and atomic force microscopy , we show that the icosahedral symmetry is only approximate , in part because one of the 5-fold vertices has a unique “starfish-shaped” feature and because a better three-dimensional reconstruction was obtained by assuming only 5-fold symmetry . Contrary to expectations , the arrangement of the capsomers on the Mimivirus surface is not as that in many other large icosahedral dsDNA viruses . Instead , the faces of Mimivirus have systematic vacant sites that are surrounded by six capsomers with alternative orientations which differ by about 60° . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biophysics/macromolecular",
"assemblies",
"and",
"machines",
"virology/virion",
"structure,",
"assembly,",
"and",
"egress",
"virology",
"biophysics"
] | 2009 | Structural Studies of the Giant Mimivirus |
The early steps of retrovirus replication leading up to provirus establishment are highly dependent on cellular processes and represent a time when the virus is particularly vulnerable to antivirals and host defense mechanisms . However , the roles played by cellular factors are only partially understood . To identify cellular processes that participate in these critical steps , we employed a high volume screening of insertionally mutagenized somatic cells using a murine leukemia virus ( MLV ) vector . This approach identified a role for 3′-phosphoadenosine 5′-phosphosulfate synthase 1 ( PAPSS1 ) , one of two enzymes that synthesize PAPS , the high energy sulfate donor used in all sulfonation reactions catalyzed by cellular sulfotransferases . The role of the cellular sulfonation pathway was confirmed using chemical inhibitors of PAPS synthases and cellular sulfotransferases . The requirement for sulfonation was mapped to a stage during or shortly after MLV provirus establishment and influenced subsequent gene expression from the viral long terminal repeat ( LTR ) promoter . Infection of cells by an HIV vector was also shown to be highly dependent on the cellular sulfonation pathway . These studies have uncovered a heretofore unknown regulatory step of retroviral replication , have defined a new biological function for sulfonation in nuclear gene expression , and provide a potentially valuable new target for HIV/AIDS therapy .
The Retroviridae are a large viral family that includes the human pathogens Human Immunodeficiency Viruses 1 and 2 ( HIV-1 and HIV-2 ) , the causative agents of acquired immune deficiency syndrome ( AIDS ) . Due to their small coding capacity and requirement for integration into the host cell genome , retroviruses are heavily dependent upon host cell machinery for efficient replication . The retroviral lifecycle can be divided into two distinct phases . The early stage consist of virus binding to a cellular receptor , fusion of viral and cellular membranes leading to delivery of the viral core into the cytoplasm , reverse transcription of the positive strand RNA genome to generate a dsDNA product , translocation of viral nucleoprotein complexes to the nucleus , and provirus establishment through integration of the viral DNA into the host cell genome . The late stage consists of transcription of the viral genome by host RNA pol II , RNA processing and export to the cytoplasm , translation of viral proteins , viral assembly , egress and maturation . While progress has been made on the identification of many of the cellular proteins involved in the late stage of the retroviral lifecycle , particularly in transcription , RNA processing and egress , less is known about the contribution of cellular factors to the early stage of the retroviral lifecycle . In particular , the contribution of cellular factors to steps subsequent to virus:cell membrane fusion and that lead to proviral DNA establishment are only partially understood [1] . A number of cellular factors that facilitate early steps in infection have been identified , although in some cases the roles of these factors are controversial . These factors include the actin cytoskelton and microtubule network [2]–[7] , LAP-2α , barrier-to-autointegration factor ( BAF ) , and emerin [8]–[15] , SUMOylation factors [16] , importins [17]–[19] , tRNAs [20] and LEDGF [21]–[28] . Although a recent genome-wide siRNA screen uncovered a number of cellular genes that contribute to various stages of HIV infection , it was notable that only a few additional factors were described that are associated with either viral DNA synthesis or integration [17] . It therefore seems likely that other , as yet unidentified , cellular factors participate in early retroviral replication . To identify other cellular factors that are involved , we have employed a somatic cell mutagenesis-based approach . This study led to the identification of the 3′-phosphoadenosine 5′-phosphosulfate synthase 1 ( PAPSS1 ) gene as playing an important role in retroviral replication . PAPSS1 and PAPSS2 are homologous enzymes that synthesize 3′-phosphoadenosine 5′-phosphosulfate ( PAPS ) , the high energy sulfate donor used in all known sulfonation reactions catalyzed by cellular sulfotransferases [29] . Golgi sulfotransferases catalyze the sulfonation of lipids , of carbohydrates , and of tyrosines in proteins [29]–[33] . Cytoplasmic sulfotransferases lead to the sulfonation of a wide variety of peptides , hormones and xenobiotics [29] , [34] . The data described in this report reveal a novel role for the cellular sulfonation pathway in retroviral replication during provirus establishment , one that modulates the subsequent transcriptional competency of the provirus .
A schematic of the proviral forms of the MLV constructs used in this paper is provided in Figure S1 . The viral genome plasmids pMMP-nls-LacZ , pCMMP-eGFP and pCMMP-IRES-GFP , pCMMP-CD4-eGFP , pHIV-TVA800-hcRED , pRET and the ASLV-A genome plasmid RCASBP ( A ) -AP have been previously described [35]–[37] . The MLV vectors pLEGFP ( Clontech , Palo Alto , CA ) and pQCLIN ( Clontech , Palo Alto , CA ) as well as the HIV-1 self inactivating ( SIN ) pLenti6/V5-GW/lacZ ( Invitrogen , Carlsbad , CA ) were obtained commercially . The HIV-1 vector pNL4-3 . Luc . R-E- [38] was obtained from the NIH AIDS Research and Reference Reagent Program , Division of AIDS , NIAID , NIH ( deposited by Dr . Nathaniel Landau ) . To construct the MLV vector pCMMP-CD4 ( expressing human CD4 from the viral LTR ) , the previously described pCMMP-CD4-eGFP vector [39] was digested with PmlI and HpaI to remove the IRES-eGFP cassette and then the plasmid was re-ligated . The MLV vector pCMMP-HcRED ( encoding the red fluorescent protein HcRED from the viral LTR ) was generated by removing the multiple cloning site and IRES from pCMMP-IRES-GFP by AgeI/HpaI digestion and inserting an AgeI/StuI fragment containing the HcRED coding sequence from pHcRED1 ( Clontech , Palo Alto , CA ) . The MLV vector pCMMP-SEAP-IRES-GFP ( encoding SEAP and GFP ) was generated by inserting the SEAP gene from pSEAP-control ( Clontech , Palo Alto , CA ) upstream of the IRES in pCMMP-IRES-GFP . HIV-1 vectors for stable expression of PAPS synthases or control cDNAs were generated by PCR amplification of coding sequence ( PAPSS1: IMAGE#3869484 , PAPSS2: IMAGE#2988345 , control cDNA ZNF639:IMAGE#4794621 ) from commercially available cDNAs ( Open Biosystems ) and cloning into MluI/EcoRV digested pLenti6/V5-GW/lacZ . Chinese hamster ovary cells ( CHO-K1 , ATCC CCL-61 ) were cultured in F-12 media supplemented with 10% bovine calf serum ( BCS ) ( Invitrogen , Carlsbad , CA ) . Human embryonic Kidney 293T cells ( ATCC CRL-11268 ) were cultured in DMEM supplemented with 10% fetal calf serum ( FCS ) ( Hyclone , Logan , UT ) . Chicken DF-1 cells ( ATCC CRL-12203 ) were cultured in DMEM supplemented with 10% FCS . Jurkat cells ( ATCC TIB-152 ) were cultured in RPMI-1640 supplemented with 10% FCS . CHO cells expressing the receptor for ASLV ( TVA-800 ) were generated as previously described [39] by infection with HIV-1-TVA800-hcRED[VSV-G] at an approximate moi of 0 . 5 hcRED transducing units for 2 hours . Cells infected with this virus express: TVA800; HcRED; and the blasticidin S deaminase ( BSD ) gene . Infected cells were selected for two weeks in the presence of 3 µg/ml blasticidin ( Invitrogen , Carlsbad , CA ) . Cells expressing either PAPSS1 , PAPSS2 or the control cDNA , ZNF639 , were generated by infecting CHO-K1 and IM2 cells with VSV-G pseudotyped HIV-1 vectors encoding the appropriate ORF ( see above ) at an approximate moi of 0 . 5 blasticidin transducing units for 2 hours . Infected cells were selected for two weeks in 3 µg/ml blasticidin . MLV VSV-G and EnvA pseudotyped viruses were generated by calcium phosphate transfection of 293T cells as previously described [35] , [39] , [40] . VSV-G pseudotyped HIV-1 vector was produced by a similar proceedure except the genome plasmid used was pNL4-3 . Luc . R-E- . The VSV-G pseudotyped self-inactivating HIV-1 vector was made using the Virapower kit ( Invitrogen , Carlsbad , CA ) following manufacturers instructions . DF-1 cells were transfected using the calcium phosphate method with the subgroup A-specific ASLV-A vector , RCASBP ( A ) -AP , encoding alkaline phosphatase [37] . Media from transfected cells was collected 2 days post transfection to 7 days post transfection and filtered through a 0 . 45 µm bottle top filter . Virus was stored at 4°C through the collection period , combined and then frozen at −80°C for long-term storage . Virus for use in Quantitative PCR amplification studies was treated with DNaseI ( Roche Applied Science , Indianapolis , IN ) to remove contaminating plasmid DNA from the virus preps . DNaseI was added as a powder to a final concentration of 1 µg/ml when the virus containing supernatants were collected . The supernatants were incubated 1 hr at room temperature before filtration . The titer of VSV-G and envA vector stocks were determined by assaying for transduction of a marker gene following infection of either WT CHO-K1 cells or WT CHO-K1 cells that had been engineered to express TVA800 [39] . For viruses that lack a cell-based reporter gene assay , immunoblot analysis of viral capsid protein ( CA ) levels ( α-p24 for HIV or α-p36 for MLV ) in the extracellular supernatants of producer cells was used to equalize the amounts of input virus as compared to those associated with viral vectors that contain reporter genes ( Lenti6/V5-GW/lacZ[VSV-G] for HIV and MMP-nls-LacZ[VSV-G] for MLV ) . CHO-K1 cells ( 1×108 ) were mutagenized by infection with VSV pseudotyped pRET at an approximate MOI of 0 . 01 GFP transducing units . Cells were selected in 900 µg/ml G418 for two weeks . A pool of 2×107 insertionally mutagenized CHO-K1 cells were challenged with CMMP-CD4 [VSV-G] at an approximate m . o . i of 1 CD4 transducing units for two hours at 37°C 37 in the presence of 4 µg/ml polybrene . Unbound viruses were then removed and fresh medium was added . At 48 hours post infection ( hpi ) the cells were removed from the plate with phosphate buffered saline ( PBS ) containing 5 mM EDTA . Cells were pelleted ( 200×g , 5 min ) and resuspended in 500 µl PBS containing 2 mM EDTA and 2% bovine serum albumin ( BSA ) ( Sigma-Aldrich , Inc . , St . Louis , MO ) . The cells were incubated with anti-human CD4 iron-conjugated antibody ( Miltenyi Biotec Inc . , Auburn , CA ) at 20 µg/107 cells for 15 minutes at 4°C . Large cell ( LC ) columns ( Miltenyi Biotec Inc . , Auburn , CA ) were applied to a magnetic field and washed with 2 ml PBS containing 2 mM EDTA and 2% BSA . Cells were filtered through a 30 µm mesh ( Miltenyi Biotec Inc . , Auburn , CA ) and applied to the LC column . Cells were washed twice with 2 ml PBS containing 2 mM EDTA and 2% BSA . Column flow through and washes were collected and the cells were pelleted , resuspended in medium and replated . Cells were allowed to recover for at least 16 hours before the next viral challenge . When necessary , the cells were expanded between each round of virus challenge to a minimum of 5×105 cells per sort . The challenge and selections were repeated five times . The population was challenged a final time with CMMP-HcRED[VSV-G] and the HcRed negative cells were single cell cloned after high speed FACS ( University of Wisconsin Comprehensive Cancer Center ) . Single cell clones from the sorted insertional mutant pools were grown for 14 days post sorting , trypsinized and then plated onto duplicate assay plates . The assay plates were incubated for 2 hours with pMMP-nls-LacZ [VSV-G] at an approximate m . o . i . of 1 LacZ transducing unit ( LTU ) in the presence of 4 µg/ml polybrene . Unbound virus was then removed and fresh medium was added . At 48 hpi , one plate was assayed for β-galactosidase activity using the Galacto-Star chemiluminescent kit ( Applied Biosystems , Foster City , CA ) according to the manufacturers instructions and the other plate was assayed for cell number and cell viability using CellTiter-Glo reagent ( Promega , Madison , WI ) following the manufacturers instructions to control for variations in cell number among the clones . Quantitative chemiluminescent infection assays were performed as previously described [39] , briefly , 8 wells of a 96 well plate were seeded at 1×104 cells/well for each cell line tested . The cells were incubated for 2 hours with an approximate m . o . i of 1 transducing unit ( based on marker gene expression for β-galactosidase and alkaline phosphatase , or CA equivalents for luciferase , as described above ) , in the presence of 4 µg/ml polybrene . Unbound virions were removed and fresh medium was added . At 48 hpi , four wells were assayed for β-galactosidase activity using the Galacto-star Kit ( Applied Biosystems , Foster City , CA ) , for alkaline phosphatase activity using the Phospha-Light Kit ( Applied Biosystems , Foster City , CA ) or for luciferase activity using the Britelite ( PerkinElmer , Boston , MA ) according to the manufacturer's instructions . The other four wells were assayed for cell number and cell viability using CellTiter-Glo reagent ( Promega , Madison , WI ) as described above . The results obtained were normalized for relative cell number . To determine the absolute fold-resistance to viral infection , X-Gal staining was performed on cells that were infected with serial dilutions of viruses . For these experiments , cells were seeded at 1×104 cells/well in triplicate rows for each cell line tested . The cells were then infected for 2 hours with ten-fold serial dilutions of MMP-nls-LacZ [VSV-G] in the presence of 4 µg/ml polybrene as described before and the cells were subsequently stained with X-gal as previously described [41] . The blue cells contained in wells that had between 20 and 200 β-galactosidase positive cells were counted to give an accurate measure of the viral titer . PAPS assays were performed as previously described [42] . Briefly , cells were lysed by three freeze thaw cycles in PAPS lysis buffer [20 mM Tris pH 8 , 20% sucrose , 1 mM EDTA , 1 mM DTT] in the presence of 1× protease inhibitor cocktail ( RPI , Mt . Prospect , IL ) . Cell lysate ( 1 µl ) was mixed with 5 mM ATP and 10 μCi [35]S labeled sulfate in reaction buffer [50 mM Tris pH 8 , 25 mM MgCl2 , 0 . 9 M EDTA , 13 . 5 mM DTT ) and incubated for 30 minutes at room temperature . Thin layer chromatography ( TLC ) was used to separate PAPS , APS and SO4 on PEI cellulose TLC plate ( EMD Chemicals , Gibbstown , NJ ) in 0 . 9 M LiCl . TLC plates were dried , exposed to phosphoimager plates , and quantified using the Imagequant software volume method . Mobility positions were confirmed with commercial PAPS preparations ( PerkinElmer , Waltham MA , Cat# NEG010100UC ) . Each sample was normalized for µg of total protein in the lysate determined by Bradford assay using the Quick Start Bradford Dye reagent ( Bio-rad , Hercules , CA ) . To measure the amounts of reverse transcription intermediates in infected cells , cells were seeded in triplicate wells at 5×105/well in a 6 well plate and then infected at 4°C on a rocking platform at an m . o . i . of 1 GFP transducing unit ( GTU ) for 2 hours with an MLV vector ( pLEGFP; Clontech , Palo Alto , CA ) pseudotyped with VSV-G that was treated with DNaseI as described above . Virus derived from pLEGFP was used for these assays because the 3′ viral LTR varied enough from pCMMP so realtime PCR primers could be designed that specifically recognized the pLEGFP derived test virus but not the pCMMP derived screen virus . DNA was harvested from infected cells 24 hpi ( hpi ) using the DNeasy Kit ( Qiagen , Valencia , CA ) . For the nuclear fractionation studies nuclei were harvested from infected cells 24 hpi using the Nuclei EZ Prep Kit ( Sigma-Aldrich , Inc . , St . Louis , MO ) following the manufacturers instructions and DNA was isolated from nuclei as described above . To measure integrated proviral DNA copy number , cells were seeded and infected as described above and then passaged for 18 days . DNA was then harvested from 1×106 cells as described above . DNA concentration was calculated by measuring the A260 on a SPECTRAmax Plus 96 well UV spectrophotometer ( Molecular Devices , Sunnyvale , CA ) . Quantitative , real time PCR ( QPCR ) analysis was performed on an ABI 9600 ( Applied Biosystems , Foster City , CA ) using the standard cycling conditions of 50°C 10 min , 40 cycles of 95°C 30 s , 60°C 2 minutes . DNA ( 10 µl/25 µl reaction ) was amplified in TaqMan Universal PCR Mastermix ( Applied Biosystems , Foster City , CA ) with 1 µM each primer and 0 . 1 µM 5′ , 6-FAM , 3′TAMRA labeled probe . Each primer probe set was tested on each cell line in a minimum of 3 independent experiments . The number of molecules in each reaction was determined by comparison to standard curves generated from amplification of plasmid DNA containing the target sequence . The primers used are specific for the U3-U5 region of the LEGFP vector and are shown along with the viral LTR feature and the bp position recognized in pLEGFP are: OJWB39 ( 5′-CAGTTCGCTTCTCGCTTCTGTTC-3′ ) [U3 , bp 523–535] , OJWB47 ( 5′-GTCGTGGGTAGTCAATCACTCAG-3′ ) [R and U5 , bp 697–719] and OJWB38 ( 5′-6-FAM- ATCCGAATCGTGGTCTCGCTGTTC-TAMRA-3′ ) [R , bp 657–680] . Templates for RNA probes to MLV were generated by PCR amplification using 1 µg total DNA from CHO-K1 cells infected with CMMP-GFP[VSV-G] along with the oligonucleotide primers OJWB7 ( 5′-GAACAGATGGTCCCCAGATGC-3′ ) and OJWB8 ( 5′-CGGTGGAACCTCCAAATGAA-3′ ) . ExTaq polymerase ( Takara , Madison WI ) was used with cycling conditions of [95°C 5 min , 30 cycles of 95°C 30 S , 50°C 30 S 72°C 1 min] . This resulting LTR fragment was cloned into pGem T-easy ( Promega , Madison , WI ) and spanned 192 bp upstream of the transcription start ( +1 , the start of R ) to 139 bp downstream of +1 , which results in a 140 bp protected fragment in the RNase protection assays . The template for RNA probes to hamster actin RNA were generated by reverse transcription of the hamster β-actin cDNA cloned by reverse transcription PCR amplification of 1 µg total RNA isolated from CHO-K1 cells with OJWB313 ( 5′- TCACCCACACTGTGCCCATCTATGA-3′ ) and OJWB314 ( 5′-CAACGGAACCGCTCATTGCCAATGG-3′ ) and MasterAmp tTh polymerase ( Epicenter , Madison WI ) using cycling conditions of [60°C 5 min , 30 cycles of 95°C 30 S , 50°C 30 S 72°C 1 min] . The resulting PCR amplified product was cloned into pGem T-easy ( Promega , Madison , WI ) and generates a 294 bp protected fragment in RNase protection assays . Anti-sense RNA probes were generated by digesting the plasmids with SpeI and performing performing in vitro transcription reaction using the Riboscribe Kit ( Epicenter , Madison WI ) with T7 polymerase and 50 μCi α−[32]P-UTP . To measure the amounts of transcription from integrated proviruses in infected cells , cells were seeded in triplicate wells at 5×105/well in a 6 well plate and then infected at 4°C on a rocking platform at an m . o . i . of 1 GTU for 2 hours with an MLV vector ( pCMMP-GFP ) pseudotyped with VSV-G . RNA was isolated form cells 24 hpi using the RNeasy kit ( Qiagen , Valencia , CA ) following manufacturers instructions . RNase protection assays were performed by mixing 2 µg ( viral transcripts ) or 0 . 5 µg ( β-actin ) of total RNA with 5×104 cpm probe , hybridizations and digestions were done using the RPA III kit ( Ambion , Ausin TX ) . Protected fragments were separated on a 6% PAGE-Urea gel , dried and exposed to a phosphoimager plate . Phosphorimage units were measured using the Imagequant software volume method .
Chinese hamster ovary ( CHO-K1 ) cells were used for insertional mutagenesis by a retroviral vector since these cells are functionally hypodiploid at numerous loci [43] and therefore insertion of the viral vector into a single allele of a given cellular gene can be sufficient to produce a genetically-null phenotype . The insertional mutagenesis was performed with the murine leukemia virus ( MLV ) -based vector pRET , which encodes green fluorescent protein ( GFP ) , as well as a neomycin phosphotransferase ( NPT ) mRNA that contains an instability element downstream of a canonical splice donor site [44] . Integration of pRET upstream of a cellular exon gives rise to a NPT mRNA transcript in which the instability element is removed by mRNA splicing , thereby conferring G418 resistance on the mutagenized cells ( Figure S1A ) . Approximately 1×106 colonies of G418-resistant cells were generated by challenging CHO-K1 ( 1×108 ) cells with VSV-G pseudotyped pRET at an moi of 0 . 01 ( note: at this moi only a small fraction of these cells are “infected” ) to ensure only one integration event per cell . Mutagenized cells were selected in medium containing 900 µg/ml G418 for two weeks , after which the population was expanded and pooled . In order to identify cells in the population that were resistant to retroviral infection , a pool of 2×107 insertionally mutagenized cells were subjected to five rounds of challenge with a second , replication-defective , VSV-G pseudotyped MLV vector which contains a human CD4 gene that is expressed from the viral promoter ( Fig . 1 ) . Infected cells that expressed human CD4 on their surface were removed from the population at each round by magnetic cell sorting ( MACS ) using an iron-conjugated CD4-specific antibody ( Fig . 1 ) . Each round of infection and sorting resulted in an approximate 3-fold enrichment of CD4-negative cells relative to the preceding round , with a total enrichment of 47-fold . The resultant cell population , which exhibited an overall 2 . 5-fold resistance to MLV infection , was then challenged a final time with another VSV-G pseudotyped MLV vector encoding the far-red fluorescent protein HcRed . A total of 264 single cell clones of HcRed-negative cells were then isolated by FACS ( Fig . 1 ) and tested for their susceptibility to infection by a VSV-G pseudotyped MLV vector encoding β-galactosidase . One cell line , designated IM2 , that was judged to be one of the most resistant ( approximately 12-fold ) to challenge by that viral vector , based upon viral reporter gene expression ( Fig . 2A ) , is characterized in detail in this report . To determine if the defect associated with the IM2 cell line is specific for the MLV vector , wild-type CHO-K1 cells and mutant IM2 cells were engineered to express TVA800 , the cellular receptor for an avian retrovirus , subgroup A avian sarcoma and leukosis viruses ( ASLV-A ) [45] , [46] . The TVA800 expressing cells were then challenged with either the MLV vector encoding β-galactosidase vector pseudotyped with the ASLV-A envelope protein ( EnvA ) or instead with an ASLV-A vector that encodes heat-stable alkaline phosphatase [37] . Viral reporter gene expression following infection of IM2-TVA800 cells by the EnvA-pseudotyped MLV vector was 9 . 7-fold reduced as compared with CHO-K1-TVA800 cells ( Fig . 2B ) . This effect mirrored that seen with VSV-G pseudotyped MLV vectors ( e . g . Fig . 2A ) . Thus , the defect seen with IM2 cells is independent of the nature of the viral glycoprotein used to pseudotype the MLV vector . By contrast , the level of viral reporter gene expression following infection by the ASLV-A vector was comparable between IM2-TVA800 and CHO-K1 cells ( Fig . 2B ) . Since both vectors utilized EnvA to mediate entry , these observations indicate that the defect associated with the IM2 cell line is specific for protein or RNA components of the MLV core . To identify which cellular gene was disrupted by the mutagenic pRET vector , total RNA was isolated from IM2 cells and reverse transcription PCR amplification was performed using primers anchored on the virally encoded NPT gene and the poly ( A ) tail . DNA sequence analysis of the PCR amplification products and a comparison with the sequenced mouse genome revealed that the pRET provirus had integrated upstream of exon 12 of the 5′ phospho-adenosine , 3′phosphosulfate synthase 1 gene ( PAPSS1 ) ( Fig . 3A ) . The full sequence of hamster PAPSS1 gene , and its corresponding mRNA product , have not yet been reported . However , comparison with the cognate mouse gene indicates that , in IM2 cells , the pRET-encoded NPT open reading frame is fused by mRNA splicing to the third base of the codon encoding amino acid residue 579 of PAPSS1 ( Fig . 3A ) . PAPSS1 and the highly related PAPSS2 enzyme catalyze the formation of the high energy sulfate donor 3′ phospho-adenosine , 5′phosphosulfate ( PAPS ) [42] , [47] , [48] used for all sulfonation reactions in the cell . Consistent with the prediction that IM2 cells have less PAPS available for sulfonation reactions , IM2 cells incorporated 17% less [35]SO4 into macromolecules than CHO-K1 cells in bulk labeling experiments ( Figure S2 ) . However , the readout of these experiments is several steps downstream of PAPS synthase and represents the summation of multiple enzyme/substrate interactions . To directly determine if IM2 cells were deficient in PAPS synthase activity , an in vitro PAPS assay was performed . ATP and [35]SO4 were mixed with cell lysates prepared from CHO-K1 cells , IM2 cells , or IM2 cells engineered to express human cDNA clones of either PAPSS1 or PAPSS2 . The reaction products were separated on PEI cellulose TLC plates in 0 . 9 M LiCl . ( Fig . 3B ) Inorganic sulfate exhibits the greatest mobility , followed by the reaction intermediate adenosine phosphosulfate ( APS ) , with PAPS being retained closest to the origin [47] . Mobility positions were confirmed with commercial PAPS preparations ( data not shown ) . TLC plates were exposed to phosphoimager plates and the levels of PAPS synthesized were measured . These studies demonstrated that IM2 cells have five-fold lower levels of PAPSS activity per µg of protein than do the parental CHO-K1 cells ( Fig . 3B and 3C ) . PAPS synthase activity in IM2 cells was significantly increased by stable expression of either human PAPSS1 or PAPSS2 cDNA clones ( Fig . 3B and 3C ) although not to full WT levels . These data indicate that the pRET vector disrupted the function of the PAPSS1 gene in IM2 cells . To investigate whether the deficiency in PAPS synthase activity in IM2 cells was responsible for the block to MLV infection , CHO-K1 and IM2 cells engineered to express either human PAPSS1 or PAPSS2 were challenged with the VSV-G pseudotyped MLV vector encoding β-galactosidase and infected cells were enumerated by X-gal staining . Expression of either PAPSS enzyme complemented the MLV infection defect of the IM2 cell line ( Fig . 4A ) . By contrast , a control cDNA , containing an ORF unrelated to sulfonation , did not rescue virus infectivity in these cells ( Fig . 4A ) . These data confirm that the deficiency in PAPS synthase activity is responsible for the virus infection-resistant phenotype of IM2 cells . To further investigate a role for the sulfonation pathway , CHO-K1 cells were treated with either chlorate , a substrate analog of sulfate and a competitive inhibitor of PAPS synthases [48]–[51] , or with the sulfotransferase inhibitor guaiacol [50] , [51] , prior to challenge with the MLV vector . As compared to untreated cells , chlorate-treated , guiacol-treated , and chlorate/guaiacol dual-treated cells gave rise to approximately 6 . 7-fold , 3 . 4-fold , and 23-fold less blue cells , respectively ( Fig . 4B ) . Only the dual inhibitor treatment led to a significant ( 2 . 3-fold ) reduction in viable cell number ( Fig . 4C ) , which was still considerably less than the effect on infection . Similarly , chicken DF1 cells treated with chlorate were approximately 9 . 3-fold less susceptible to infection by this viral vector as judged by reporter gene expression ( Fig . 4D ) . However , this treatment did not influence infection of these avian cells by an ASLV-A vector . Treatment of IM2 cells with chlorate reduced MLV infection an additional 2-fold ( Figure S3 ) , which is consistent with the observation that these cells contain some residual PAPS synthase activity ( Fig . 3B and 3C ) . These data further show a role for the cellular sulfonation pathway in infection by MLV , but not ASLV , vectors and indicate that the mechanism ( s ) responsible are shared between different host cell species . Real time PCR amplification was used to monitor the effect of the sulfonation pathway on the levels of reverse transcription products and integrated viral DNA . Cells were infected with an MLV vector ( pLEGFP ) and either total DNA or nuclear DNA was subsequently harvested . Since these cells potentially contain both the mutagenic pRET vector , and the pCMMP derived vector utilized in the screen , the primer/probe set was chosen to amplify the plus strand strong stop replication intermediate [52] , [53] and annealed specifically to the unique 3 ( U3 ) and unique 5 ( U5 ) long terminal repeat ( LTR ) region of only the pLEGFP MLV vector ( data not shown ) . This primer probe set exhibits an excellent dose response over 6 orders of magnitude ( Figure S4A ) and a very low background , such that the signal from infected cells at 24 hpi is 400-fold higher than from cells where the virus is bound but not internalized ( 0 hpi , Fig . 5A ) The difference between infected and uninfected cells is even greater ( Fig . 5D and Figure S4B ) . The levels of total and nuclear reverse transcription products were found to be the same in IM2 cells as in CHO-K1 cells ( Fig . 5A and 5B ) . Furthermore , treatment of CHO-K1 cells with chlorate had no effect on the accumulation of viral reverse transcription products , confirming that the sulfonation pathway does not influence viral DNA synthesis ( Fig . 5C ) . Importantly , this is not due to saturation of the assay as dilution of input genomic DNA showed a proportionate decrease in both CHO-K1 and IM2 samples , even when a ten-fold higher multiplicity of infection was used ( Figure S4B ) . To investigate the possible role of this pathway in viral DNA integration , IM2 cells were infected with the same MLV vector , passaged for 18 days to allow loss of episomal forms of viral DNA [54] , [55] , and the levels of total viral DNA were then measured . IM2 cells and chlorate treated CHO-K1 cells contained nearly the same amounts of integrated viral DNA as untreated CHO-K1 cells ( 1 . 1 and 2 . 6-fold less , respectively ) ( Fig . 5D ) , which is insufficient to explain the approximately 10-fold decrease in infectivity ( Fig . 2A and 4B ) . By comparison at 18 days post-infection , nearly 400-fold lower levels of viral vector DNA were detected in MCL7 cells , a chemically mutagenized CHO-K1 cell line that exhibits a strong block to MLV DNA integration [39] ( Fig . 5D ) . These data demonstrate that the cellular sulfonation pathway does not influence either the levels of viral DNA that are synthesized in the target cell or that become integrated into the host cell genome . Since the sulfonation pathway did not influence the level of integrated viral DNA , we next determined if it impacts subsequent provirus gene expression . In these studies , the level of MLV LTR-driven transcription from the MMP-nls-LacZ vector was compared to that from the internal CMV promoter contained in QCLIN , a commercially available , self-inactivating ( SIN ) MLV vector with promoter defective LTRs [56] . The levels of β-galalactosidase from the QCLIN vector were the same in infected IM2 and CHO-K1 cells ( Fig . 6A ) , a result that supports our observation that the sulfonation pathway does not influence the overall level of viral DNA integration . By striking contrast , MLV LTR-driven reporter gene expression following infection was reduced 5 . 6-fold in IM2 cells as compared with CHO-K1 cells ( Fig . 6A ) . Consistently , a combination of chlorate and guaiacol treatment reduced β-galalactosidase levels produced from the MMP-nls-lacZ vector by 7 . 3-fold , following infection of CHO-K1 cells , but this treatment did not influence gene expression from the SIN vector ( Fig . 6B ) . These data suggest that the target of action for the cellular sulfonation pathway is contained within the MLV LTR . To directly examine the influence of the sulfonation pathway upon MLV LTR-driven mRNA transcription , total RNA was isolated from CHO-K1 and IM2 cells that were infected with a VSV G-pseudotyped MLV vector encoding EGFP . RNase protection assays were performed with a probe that hybridizes to the primary viral mRNA transcript ( hybridizing to the R-U5 region ) . IM2 cells accumulated 3 . 5-fold less primary transcript than CHO-K1 cells ( Fig . 6C & E ) . Similarly , the levels of viral-derived transcript were reduced in CHO-K1 cells treated with inhibitors of the cellular sulfonation pathway ( Chlorate 11- fold , guaiacol 17- fold , and chlorate and guaiacol 40- fold ( Fig . 6D & E ) . All values were normalized to hamster β-actin levels , which varied less than 2-fold in all cases ( Fig . 6C and 6D ) . These data indicate that the sulfonation pathway influences a step that impacts the transcriptional competency of the provirus . To determine the time point during infection when the cellular sulfonation pathway is involved , CHO-K1 cells were incubated with the VSV-G pseudotyped MLV vector encoding β-galactosidase at 4°C , and infection was then initiated by a temperature shift to 37°C . Chlorate was then added at various times post-infection and the effect of this treatment on the establishment of viral vector in these cells was then measured by quantitating β-galalactosidase expression . Chlorate addition up to 16 hpi led to a reduction in subsequent viral reporter gene expression ( Fig . 7A ) . However , addition of the inhibitor at time points 18 hpi , or later , had no effect ( Fig . 7A ) . This timing coincides with maximal levels of viral DNA integration [57] , suggesting that the cellular sulfonation pathway might influence a step during or shortly after provirus establishment . To explore this possibility further we compared the effect of chorate treatment on proviral gene expression from resident , versus newly acquired , proviruses . A CHO-K1 cell line was established that contains a resident MLV vector encoding secreted alkaline phosphatase ( Fig . 7B ) . These cells were then challenged with the MLV vector encoding β-galactosidase in the presence of chlorate to generate newly acquired MLV proviruses under conditions where the sulfonation pathway was inhibited . These experiments showed that the chlorate treatment affected gene expression from the newly acquired , but not the resident proviruses ( Fig . 7B ) . In an independent experiment , chlorate treatment was shown not to influence β-galalactosidase expression from a resident MLV vector ( data not shown ) , confirming that the effect seen was not reporter gene-specific . Taken together with the timing of the sulfonation requirement during infection ( Fig . 7A ) , these results strongly imply that this cellular pathway influences MLV replication at a step during provirus establishment , one that impacts subsequent viral gene expression . The previous experiments showed that the sulfonation pathway affects LTR-driven gene expression from newly acquired MLV , but not ASLV , proviruses . To test the influence of this pathway on HIV-1 LTR-driven gene expression , CHO-K1 and IM2 cells were challenged with either of two VSV-G pseudotyped HIV-1 vectors , one with luciferase expressed from the viral LTR and the other a SIN vector with β-galactosidase expressed from an internal CMV promoter . Reporter gene expression from the HIV-LTR was reduced 5-fold in IM2 versus CHO-K1 cells whereas that from the internal CMV promoter was the same in both cell types ( Fig . 8A ) . Consistently , treatment of CHO-K1 cells with chlorate , guaiacol , or with both inhibitors resulted in 10- , 8- , and 12-fold reductions in HIV LTR-driven reporter gene expression , respectively . By contrast , internal CMV promoter-driven reporter gene expression was unaltered or was slightly enhanced by these treatments ( Fig . 8B ) . Similar results were observed using human Jurkat T cells infected with VSV-G pseudotyped HIV or MLV viral vectors that express luciferase from the viral LTRs ( Fig . 8C ) . Therefore as for MLV , HIV LTR-driven gene expression is also regulated by the cellular sulfonation pathway .
Here we have presented multiple lines of evidence that the host cell sulfonation pathway influences retroviral infection by affecting a step during provirus establishment , one that modulates gene expression from the viral LTR promoter . First , insertional mutagenesis and genetic complementation studies identified PAPSS1 as a cellular gene that is important for MLV infection . Second , a similar defect was seen with cells treated with the PAPS synthetase inhibitor , chlorate , or with the sulfotransferase inhibitor , guaiacol . Third , inhibition of the sulfonation pathway had no impact on the levels of integrated MLV DNA but influenced downstream MLV LTR-driven gene expression from newly formed proviruses . Fourth , MLV was sensitive to inhibitors of the sulfonation pathway at time points up to that associated with maximal levels of viral DNA integration [57] . Finally , the observations made with MLV held true for HIV-1 , the causative agent of AIDS , since the sulfonation pathway also influenced LTR-driven transcription from that virus . These data suggest that sulfonation may play an important role in the regulation of nuclear gene expression . Consistent with this , PAPSS1 localizes to the nucleus , which implies there is a requirement for high levels of PAPS , and by extension sulfonation , in the nucleus [58] . Thus , these studies have uncovered a heretofore unknown regulatory step of retroviral replication , one that is potentially important for HIV/AIDS therapy . The data in this report are consistent with either one of two models . In the first model , the sulfonation pathway might influence viral DNA integration site specificity so that when this pathway is impaired , the virus is targeted to regions where the provirus is less transcriptionally competent . This model is consistent with the observation that viruses sensitive to the sulfonation pathway , HIV and MLV , both share a strong preference for integration into genes , although MLV shows a much stronger preference for integration near the viral promoter regions [59]–[62] . By contrast , ASLV , which is not influenced by this pathway , shows little or no preference for integration into genes [60] , [63] . In the second model , the sulfonation pathway might have no impact upon integration site specificity but , during integration or shortly thereafter , the sulfonation pathway might influence the nature of epigenetic modifications introduced onto the viral DNA . These modifications could , in turn , regulate the transcriptional competency of the provirus . Sulfonation induced changes in DNA methylation , histone acetylation , methylation or positioning are all potential processes which could affect the transcriptional activity of the provirus [64] , [65] . Indeed the importance of epigenetic modifications in HIV transcription is apparent in a recent large scale analysis of HIV integration sites which revealed a positive correlation between integration and epigenetic modifications favoring transcription and a negative correlation with modifications that silence transcription [66] . We are currently performing experiments aimed at distinguishing between these two models . The host cell sulfonation pathway involves a set of golgi and cytoplasmic sulfotransferases ( SULTs ) that transfer the sulfonate from PAPS to target substrates . In humans there are thirteen distinct cytosolic SULTs , arranged into three different families , and these enzymes are involved in the metabolism of steroids , bile acids , neurotransmitters , and xenobiotics [67] . Golgi sulfotransferases are involved in sulfonating carbohydrates , generating the glycosaminoglycans ( GAGs ) , heparan sulfate , chondroitin/dermatan sulfate , and keratan sulfate [68] , as well as glycolipids [29] . Two golgi tyrosylprotein sulfotransfrerases ( TPST-1 and TPST-2 ) are responsible for sulfonation of tyrosine residues on proteins and peptides . Tyrosyl sulfonation can have important regulatory effects on cell surface proteins including an influence on protein-protein interactions [69] , as exemplified by the requirement for sulfonation of tyrosine residues at the amino-terminus of the CCR5 chemokine receptor for high affinity interaction with both its natural ligands , MIP-1α and MIP-1β , as well as with HIV-1 gp120 [70] . This entry effect seen previously is distinct from our observation that sulfonation also affects a post entry step coinciding with provirus establishment . Since inhibition of sulfonation can block HIV at multiple stages of the viral lifecycle , the cellular sulfonation pathway is an intriguing target for the development of novel antivirals . Future work will be aimed at identifying the specific components of the sulfonation pathway that are critical for modulating MLV and HIV-1 infection . We expect that this information will help to uncover precisely how the sulfonation pathway regulates retroviral infection at a step coincident with provirus establishment and that influences the subsequent transcriptional competency of the provirus . | A genetic screen was used to identify host cell functions important for the replication of retroviruses , including human immunodeficiency viruses . These studies have uncovered a heretofore unexpected role for the cellular sulfonation pathway in an intracellular step of retroviral replication . Through the addition of sulfate groups , this pathway is responsible for modifying and regulating different types of cellular factors including proteins , lipids , carbohydrates and hormones . The role of this pathway was further confirmed by using specific chemical inhibitors . The sulfonation requirement was mapped to a step during viral DNA integration into the host genome that has a subsequent effect upon the level of expression of viral genes . These studies have uncovered a new regulatory mechanism of retroviral replication and suggest that components of the host cell sulfonation pathway might represent attractive targets for antiviral development . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"virology/immunodeficiency",
"viruses",
"virology",
"cell",
"biology/cell",
"signaling"
] | 2008 | The Host Cell Sulfonation Pathway Contributes to Retroviral Infection at a Step Coincident with Provirus Establishment |
Establishing the general and promoter-specific mechanistic features of gene transcription initiation requires improved understanding of the sequence-dependent structural/dynamic features of promoter DNA . Experimental data suggest that a spontaneous dsDNA strand separation at the transcriptional start site is likely to be a requirement for transcription initiation in several promoters . Here , we use Langevin molecular dynamic simulations based on the Peyrard-Bishop-Dauxois nonlinear model of DNA ( PBD LMD ) to analyze the strand separation ( bubble ) dynamics of 80-bp-long promoter DNA sequences . We derive three dynamic criteria , bubble probability , bubble lifetime , and average strand separation , to characterize bubble formation at the transcriptional start sites of eight mammalian gene promoters . We observe that the most stable dsDNA openings do not necessarily coincide with the most probable openings and the highest average strand displacement , underscoring the advantages of proper molecular dynamic simulations . The dynamic profiles of the tested mammalian promoters differ significantly in overall profile and bubble probability , but the transcriptional start site is often distinguished by large ( longer than 10 bp ) and long-lived transient openings in the double helix . In support of these results are our experimental transcription data demonstrating that an artificial bubble-containing DNA template is transcribed bidirectionally by human RNA polymerase alone in the absence of any other transcription factors .
It is generally acknowledged that the structure and dynamics of DNA at the eukaryotic promoter play important roles in gene regulation , but the nature of this relationship is unclear . From a structural perspective , RNA polymerases require single stranded DNA , or the formation of a ‘transcriptional bubble’ at the transcriptional start site ( TSS ) to initiate transcription [1] , [2] . Eukaryotic transcription initiation often proceeds from a negatively supercoiled template in the absence of helicases [3]–[6] , implicating spontaneous local melting of dsDNA as a key feature of promoter sequences . Furthermore , introduction of few mismatched bases to unzip the DNA at the start site allows transcription in the absence of supercoiling [6] , [7] . It is likely that locally enhanced breathing dynamics of the DNA are a common feature of the TSS , required to seed the formation of the transcriptional bubble . We previously showed a correlation between transcriptional start site location , single strand nuclease sensitivity , and transient dsDNA strand separation predicted by statistical calculations with the nonlinear Peyrard-Bishop-Dauxois ( PBD ) model of DNA [8] , [9] . This one-dimensional model , originally designed to explain DNA melting profiles , has successfully reproduced thermodynamic parameters for DNA phase transitions [10] , helicase unwinding force calculations [11] , mechanical unzipping [12] and DNA bubble nucleation experiments [13] . Statistical thermodynamic implementations of PBD are fast enough to allow recently the calculation of the local melting ( bubble ) probability profile of the entire Adenoviral genome ( 30 Kb ) [14] . Such calculations , however , require pre-defined bubble size thresholds and yield probability values that contain no information about bubble lifetimes and the frequency of DNA breathing motions . In search of the distinguishing dynamic features of gene promoter TSS sequences , we performed PBD-based Langevin molecular dynamic ( LMD ) simulations [8] , [15] of eight experimentally characterized mammalian core promoters . From the LMD trajectories we extracted three distinct dynamic characteristics: bubble probability , bubble lifetime , and the average strand separation coordinates . The calculated dynamical profiles suggest that a relatively large , long-lived DNA bubble commonly forms at the transcription start site .
The PBD model is a one-dimensional nonlinear model that describes the transverse opening motion of the opposite strands of dsDNA . The Hamiltonian of the model is ( 1 ) where the sum is over all N base pairs of the DNA . yn denotes the relative displacement from equilibrium of the complementary bases of the n-th base pair , divided by √2 . The first term of the Hamiltonian is the Morse potential which represents the base pair hydrogen bonds together with the electrostatic repulsion of the backbone phosphates . The parameters Dn and an depend on the nature of the base pair ( A-T vs . G-C ) at site n . The second term represents a harmonic potential approximation but with a nonlinear coupling constant , which takes into account the influence of the stacking interactions between consecutive base pairs on the transverse stretching motion . The exponential term effectively decreases the harmonic spring constant K when one of the base pairs is displaced away from its equilibrium position in the double helix: Kmax = k ( 1+ρ ) ; when yn+yn−1 = 0 , a condition met , e . g . , at equilibrium , and Kmin = k; when yn or yn−1→∞ , i . e . , when at least one of the base pairs is out of the double helix stack . This term is essential for simulating long-range cooperative effects important for sharp DNA melting [16] . The parameters of the model have been previously obtained by fitting simulations to DNA UV melting curves [10] . Langevin molecular dynamics simulations were performed at T = 310 K , by numerically integrating systems of stochastic equations based on the Peyard-Bishop-Dauxois ( PBD ) model . Periodic boundary conditions were applied in order to avoid terminal base pair effects , effectively circularizing the DNA sequence ( but without any torsional effects ) . Each DNA sequence ( Figure 1 ) was simulated in 1000 separate realizations for 1 ns , using 1 fs timesteps and a 200 ps preheating time . Simulations were performed on Linux clusters at LANL and Harvard Medical School . The probability Pn ( l , tr ) for the existence of a bubble ( collective opening ) of a certain length l base pairs and amplitude threshold ( tr , Å ) ( Figure 2 ) [15] was calculated as ( 2 ) where < >M denotes averaging over M simulations and ts is the time of the simulation . qkn ( l , tr ) enumerates the bubbles of duration Δt[qkn ( l , tr ) ] with amplitude tr [Å] and length l base pairs , beginning at the nth base pair in the kth simulation . The average bubble duration τLifetime was calculated as the average lifetime of a bubble of a given shape , i . e . , with amplitude tr [Å] and length l [bp] , over all occurrences of that bubble . ( 3 ) The average displacement of each base pair from its equilibrium double stranded conformation was calculated for the adeno-associated virus P5 promoter in two ways: using Metropolis Monte Carlo algorithm [13] and by averaging over all Langevin dynamics trajectories obtained in the above MD simulations . The average lifetime of all bubbles ( see Eq . 3 ) of a given shape , i . e . , with amplitude tr [Å] and length l [bp] containing a given base pair was calculated from the Langevin dynamic trajectories , and plotted as a function of bubble length and bubble amplitude . The sequence of the DNA promoter template , assembly of the run-off transcriptional reactions , purification of human RNA polymerase II , RNA product separation , and visualization have been previously described [6] . The control nonpromoter sequence ( 80 bp ) is part of the published sequence for the human collagen intron ( NW_927317 ) GCAAACGCCGTCGTCCGCACCGGTCGCGACTCGGCAAGGGAGCGGGCGGAAGCTGACTCG CGGCGGAGG GGGGTCACTC . All figures are assembled using Photoshop , FreeHand , Mathematica and MATLAB .
Figure 3 shows the probability for the formation of bubbles above a certain amplitude tr as a function of bubble length ( Figure 3 , panel a ) , as well as above a certain length l as a function of amplitude ( Figure 3 , panel b ) . The observed profiles differ significantly between promoters , both in probability values ( color scale ) and overall peak distribution , especially when bubbles of any size are considered ( not shown ) . However , bubble length l ( panel a ) and strand separation amplitude values ( panel b ) can be found for each promoter , above which the TSS displays the maximum probability . These thresholds vary between promoters , but in all cases except the HSV UL11 and snRNA , bubbles longer than 10 bp and with larger than 2 Å amplitudes are most likely to be present at the TSS . In comparison , the UL11 and snRNA promoters are very active across the entire simulated promoter segments , and the TSS only become predominant for very large bubbles ( panels a , b insets ) . The human ABF-1 promoter is the least dynamically active , with bubbles of l>10 bp and tr>1 Å ( panel b ) , an order of magnitude less likely than similar size bubbles in the other promoters , but a very well pronounced TSS bubble . Overall , the probability for the occurrence of bubbles longer than 10 bp varies between ∼10−4 and ∼10−3 for bubbles with larger than 1 Å amplitudes , and is in the order of 10−5 for tr>3 Å . Interestingly , NMR studies estimated comparable probabilities ( ∼10−5 ) for single base pair openings that lead to exchange between base paired hydrogens and water [19] . Comparison between the probability plots and the promoter element distributions ( Figure 1 ) reveals intriguingly that ‘classic’ promoters that contain well-known sequence motifs exhibit ‘clean’ dynamic profiles with strong peaks at the TSS , while the dynamic profiles of two promoters without known elements have poorly defined start site bubbles . Such difference could arise from higher G/C content of these two promoters , causing a bias in the simulations , as discussed in the last section . Alternatively , the observed probability differences may reflect differences in transcriptional regulation . To further characterize the DNA dynamics of the selected promoters , we used the simulated Langevin trajectories to derive the average lifetime of a given opening as a function of base pair length and amplitude ( Eq . 3 ) . Figure 4 shows the lifetimes of bubbles above certain amplitude , as a function of bubble length . The bubble lifetime profiles are more closely related among the studied promoters than the probability profiles ( Figure 3 ) . The longest-lived openings are clearly present at the transcriptional start site in most cases . Exception is again the mouse snRNA promoter , where the TSS is only slightly predominant as well as the UL11 promoter , where bubbles of similar size and stability are also present 25 bp up- and downstream of the TSS . Overall , the most stable bubbles are ∼10 bp long , with the exception of the snRNA promoter ( 5 bp ) . A notable feature of the plots is that in some cases longer bubbles are significantly more stable than smaller ones at the same location . As previously pointed out in the literature [14] , [15] , [20] , [21] , statistical probability calculations do not always predict the most likely opening to be at the TSS , and regulatory sites 20–30 bp up- or downstream of the TSS , such as a TATA box often exhibit a higher probability for opening that the start site in such calculations . In the present study , the probability for strand separation of the collagen promoter is similar at the TATA box region and the transcription start site ( Figure 3 ) , but a remarkably stable ( 5 ps ) concerted opening of 10–15 bp is seen only at the TSS ( Figure 4 ) . In contrast , the UL11 promoter displays three bubbles that are similar both in terms of probability and lifetime , at the TSS and flanking regions . According to our results the TSS and TATA-box in the collagen promoter exhibit distinct dynamic behavior . Namely , the TSS displays a lower frequency of opening but forms relatively stable bubbles , while the TATA-box region is characterized by higher frequency motions , forming bubbles of low duration . As previously reported [15] , [20] , the adenoassociated virus ( AAV ) P5 promoter displays a higher probability for opening at the TATA box than at the TSS . A detailed profile of the bubble lifetimes at individual base pare promoter positions is shown in Figure 5 , panel b . Analogous to the collagen promoter , bubbles around the AAV P5 TATA box again have significantly shorter lifetimes ( −30 , Figure 5 , panel b ) than bubbles formed around the TSS ( +1 ) . The calculated bubble lifetimes ( Figure 4 ) are in the order of few picoseconds , a number that is somewhat dependent on the choice of the PBD parameters . PBD is a phenomenological representation of DNA melting behavior , and water collisions are implicitly modeled in the Langevin simulations , necessarily yielding a qualitative description of dynamic lifetimes . Our focus here is therefore on relative but not absolute timescales . To verify that the observed DNA dynamic profiles are relevant to transcription initiation , we performed identical PBD-LMD simulations on nonpromoter DNA sequences . The simulation results for the intron sequence of the human collagen gene are shown in Figure 6 . The intron sequence was chosen to exclude transcription factors binding sites , as we previously showed that such sites are often dynamically active ( 14 ) . As shown ( Figure 6 ) the intron sequence displays significantly lower propensity for strand separation both in terms of probability for opening with given amplitude ( panel a ) , probability for opening with given length ( panel b ) , and bubble lifetime ( panel c ) . The profiles of other examined sequences containing the repeats: [ATATATATAT]7 , [GCGCGCGCGC]7 , [GCATGCATGC]7 , [GCGCGATATA]7 , [GCGATA]12 also lacked localized bubbles ( not shown ) of the size and lifetime observed for the studied core promoters . Our data support the conclusion that nonpromoter sequences lack the characteristic signature of strand separation dynamics of the gene promoters . That bubbles , such as those predicted by the simulations , are coupled to biochemical DNA events is suggested not only by the successful reproduction of DNA melting [10] and unzipping [12] data by the PBD model , but also by single strand nuclease sensitivity and in vitro transcription experiments . We previously reported such experiments for the AAV P5 and adenoviral major late ( AdMLP ) promoters [8] . The role of DNA local melting in eukaryotic transcription is supported by the fact that inserting a promoter in a supercoiled plasmid allows transcription to proceed in the absence of helicase activity [3] , [4] , and even in the absence of the TATA box binding protein TBP in a TATA box promoter [5] , [6] . Here we demonstrate that human RNA polymerase II ( RNAP2 ) bidirectionally initiate transcription in the absence of any transcription factors , if an artificial long-lived bubble of >/ = 5 bp is introduced at the start site of the AAV P5 promoter ( Figure 5 , panel a , lanes 1 , 2 , and 3 ) . When the DNA template is linear and unzipped , transcription does not proceed ( panel a , lane 4 ) , even though the promoter sequence DNA is intact ( panel a , schematic diagram ) . These results could explain our previously reported experimental data with linear and supercoiled AAV P5 promoter DNA templates [6] . They suggest that some structural aspect of the DNA sequence is favorably enhanced by the external unwinding force of supercoiling in the promoter region . The transcriptional data here ( panel a ) , together with the previously published results by us and also by others , clearly suggests that the aspect in question is most likely local DNA melting , remarkably enabling bidirectional transcription by RNAP2 alone . The calculated bubble lifetime profile of the P5 promoter ( panel b ) is consistent with the idea that a transient local bubble in the dsDNA at the promoter , amplified and stabilized by negative supercoiling , is necessary for transcription initiation by RNAP2 . The role of transcription factors including YY1 in this case appears to be to further assist bubble formation , and direct the transcription reaction only downstream of the TSS [6] . Besides the statistical probability and lifetimes of the open states , the Langevin dynamic trajectories can be used to derive the average displacement of the dsDNA base pairs from their equilibrium closed state . Figure 7 shows the average displacements of bp −47 to +22 of the adeno-associated virus P5 promoter and a transcriptionally silent A>G/T>C mutant [8] . We previously reported a dramatic difference in the bubble probability at the mutated site in those two sequences [8] , [15] , matching the dramatic difference in transcriptional activity of the promoters . The average displacements calculated by Monte Carlo ( MC ) simulations are also shown for comparison with the Langevin data . The results from the LMD and MC simulations are virtually identical , as should be expected from properly conducted simulations . The strongest signals in the P5 promoter are again at the TATA box and TSS , but in contrast to the probability distributions ( Figure 3 ) , and average lifetimes ( Figure 4 ) , the average coordinates of the TATA box and the TSS do not stand out so clearly . Curiously , the simulations predict differences as large as 0 . 2 Å in the average base pair length at different positions of AAV P5 . Such significant differences should be experimentally detectable by NMR measurement of residual dipolar couplings in a weakly oriented medium [22] . The slightly lower average displacement of the TSS region compared to the TATA box is consistent with the idea that bubbles there are formed more rarely but persist longer and have higher amplitudes . A comparison between the average displacement profiles of wild-type P5 promoter and the transcriptionally silent mutant ( Figure 7 ) reveals a rather small difference in the average displacement of the TSS position , in contrast to the dramatic difference in the bubble lifetime profiles ( Figure 5 , panel b ) . This result supports the notion that bubble lifetime , probability , and average amplitude are distinct dynamic properties with nontrivial dependence on DNA sequence . The data suggest that the studied TSS are more easily distinguished by lifetime and bubble probability , than average displacement . Nevertheless , if the average strand displacements predicted here are accurate , variations of such magnitude in the double helix width may have a functional effect on protein-DNA recognition in general . Despite the differences ( Figure 1 ) in type of regulation ( e . g . , always turned ON ‘housekeeping’ vs . highly regulated between low and high level of expression mammalian oncogene vs . viral ) and promoter class ( e . g . , TATA/Inr , non-Inr ) , six of the eight studied promoters display TSS bubbles that are remarkably similar in length ( ∼10 bp ) and lifetime ( 5–10 ps ) , according to the simulations . As noted , those are ‘classical’ promoters , in the sense that they represent examples of the familiar TATA box and Inr sequence elements . Among those , it might be speculated that the constitutively expressed collagen and keratocan promoters , which exhibit strong and well pronounced bubbles at the TSS , may require less assistance with DNA unwinding during transcription initiation than the less transcriptionally active , inducible gene ABF-1 [23] . PU . 1 gene is another tightly regulated gene , but the experimental evidence suggests that this gene is constitutively active and is down-regulated post-transcriptionally [24] , [25] . Interestingly , it has been proposed that most housekeeping genes have CpG island promoters that transcribe from multiple TSS ( reviewed in [18] ) . In this study , the HSV-1 UL11 and the snRNA are more G/C-rich than the rest of the simulated promoters ( 75% and 69% G/C , respectively ) and both contain CpG islands upstream of the TSS ( not shown ) . Whether the observed broad dynamic activity across these promoters corresponds to a distinct mode of regulation through the presence of multiple TSS remains to be established . In addition to the eight promoters shown in Figure 1 , we tested several promoters with very high G/C-content ( 80%–95% ) in the TSS region . These promoters did not display any significant probability of opening at the start site ( data not shown ) . The observed dynamic profiles of G/C-rich promoters may result from a bias of the PBD model against G/C-rich sequences , introduced by the sequence independence of the stacking potential ( Eq . 1 ) . Experimental evidence by us and also by others suggests that G/C tracks exhibit unusual base pair opening [26] and melting [27] behavior and we are currently modifying the stacking term [28] to incorporate such effects ( Alexandrov et al . , submitted ) . It should be emphasized that the PBD model performs well for ‘mixed’ sequences and a heterogeneous stacking term should not introduce significant changes in the majority of the shown profiles . We believe that establishing the general mechanistic features of transcription initiation requires detailed understanding of both the sequence and the structure/dynamics of promoter DNA . PBD Langevin dynamic ( LMD ) simulations occupy a unique niche between fast bioinformatic methods and all atom simulation techniques . We have used PBD LMD to derive three different criteria describing the strand separation dynamics of promoter DNA sequences . The results suggest that the most stable dsDNA openings do not necessarily coincide with the most probable openings or with the highest average strand displacement , underscoring the advantages of proper molecular dynamic simulations . According to the simulations , each promoter exhibits distinct DNA dynamic characteristics , but the transcriptional start site is often distinguished by large , relatively stable openings in the double helix . Such local openings are likely to be recognized and engaged by the transcriptional machinery , and may then be amplified , stabilized , or suppressed by DNA-protein interactions as part of gene transcriptional regulation . Data from in vitro transcription experiments directly support the stable bubble requirement for DNA transcription by RNA polymerase in the absence of any transcription factors . | Accessing the information encoded in DNA requires that RNA polymerases recognize the core promoter , a sequence that marks the start of a gene . Statistical analysis of known promoter sequences has failed to reveal a simple code for identifying promoters , leading to the suggestion that promoter DNA is distinguished by certain structural/dynamic properties encoded in nonobvious ways by the literal sequence . Because the DNA strands at the promoter need to be separated for transcription to begin , we previously proposed that promoter sequences exhibit a propensity for spontaneous strand separation . Here , we conduct simulations of the ultrafast , small-scale strand separation motions of eight mammalian promoters and show that start sites tend to form larger and more stable openings in the double helix compared to other sequences . Experimentally , we show that an artificial permanent opening in the double helix is sufficient for transcription in the absence of sequence-specific protein–DNA contacts . These findings support a view of DNA as a structurally active participant in gene expression , rather than the commonly envisioned passive digital storage device . Our analysis suggests that functionally relevant structural variation in genomic DNA occurs at the level of fast motions not readily observed by traditional molecular structure analysis . | [
"Abstract",
"Introduction",
"Methods",
"Results/Discussion"
] | [
"computational",
"biology/transcriptional",
"regulation"
] | 2009 | Toward a Detailed Description of the Thermally Induced Dynamics of the Core Promoter |
Mutualistic cooperation often requires multiple individuals to behave in a coordinated fashion . Hence , while the evolutionary stability of mutualistic cooperation poses no particular theoretical difficulty , its evolutionary emergence faces a chicken and egg problem: an individual cannot benefit from cooperating unless other individuals already do so . Here , we use evolutionary robotic simulations to study the consequences of this problem for the evolution of cooperation . In contrast with standard game-theoretic results , we find that the transition from solitary to cooperative strategies is very unlikely , whether interacting individuals are genetically related ( cooperation evolves in 20% of all simulations ) or unrelated ( only 3% of all simulations ) . We also observe that successful cooperation between individuals requires the evolution of a specific and rather complex behaviour . This behavioural complexity creates a large fitness valley between solitary and cooperative strategies , making the evolutionary transition difficult . These results reveal the need for research on biological mechanisms which may facilitate this transition .
It is well known that , in the absence of genetic relatedness , altruistic behaviours in which individuals pay a fitness cost for the benefit of others cannot evolve by natural selection [1 , 2] . However , it is often assumed that mutualistic behaviours , wherein individuals collectively gain a common benefit [28 , 29] , do not pose such a problem , and are therefore of limited interest to evolutionists: they simply evolve because they benefit the individuals who express them . However , mutualistic behaviours do often pose a different kind of evolutionary problem than altruism: they require coordination [3 , 5–28] . Many collective traits are only mutually beneficial if several individuals express them together in a coordinated fashion . That is , it would not be beneficial for a single individual to express the cooperative trait if others did not express it as well . Consequently , whereas altruistic behaviours pose a problem of stability , which can only be solved by genetic relatedness , many forms of mutualistic behaviours pose a problem of evolution . These collective strategies are stable equilibria but their evolution is complex . This problem has been formalized in game theory as the stag hunt game [6] . In the stag hunt , two hunters are confronted with the choice of either hunting a hare alone for a small but guaranteed benefit , or coordinating to hunt a stag cooperatively for a bigger reward , with the risk of not being rewarded at all if they hunt the stag alone . There are two evolutionarily stable Nash equilibria in this game: ( 1 ) simultaneous defection ( i . e . both players hunt hares ) , which is risk-dominant as it maximizes the minimum payoff an individual can expect , and ( 2 ) simultaneous cooperation ( i . e . both players hunt stags ) , which is payoff-dominant as it maximizes the total payoff at equilibrium . One of the aims of evolutionary analyses of the stag hunt is to characterize the mechanisms that facilitate the transition from the solitary equilibrium to the cooperative equilibrium . The difficulty is that cooperation can only be favoured by selection when a sufficient proportion of individuals in the population also cooperate . The transition from a population with a majority of solitary individuals to one with a majority of social individuals requires the rise of cooperation above an invasion threshold , which must occur for non-selective reasons . In game-theoretic analyses , the hunting strategy of individuals is generally assumed to be encoded by a single genetic locus with two alleles: solitary or social [6] . In this case , random mutations and/or demographic stochasticity can lead to the appearance of a subpopulation of mutants playing the social strategy which is sufficient to overcome the invasion threshold . Moreover , Skyrms [6] showed that this cooperation can be further facilitated in a spatially structured population in which individuals tend to interact more with genetically related partners . However , this approach makes a very strong assumption about the underlying mechanistic nature of behaviour: that a single mutation is sufficient to transform an individual playing a solitary strategy into an individual playing a perfectly efficient social strategy . In reality , hunting socially implies several novel behavioural abilities . In particular , it implies the ability to coordinate with others in order to focus on the same prey , which is unlikely to occur with only a single random mutation . In this paper , we postulate that critical aspects of coordinated cooperation have been neglected by game-theoretic analyses and investigate the mechanistic constraints which interfere with the evolution of coordination in a more realistic setting where the mapping between genotype and phenotype is not limited to a strict binary encoding . Evolutionary robotics is a useful methodology for the simulation and study of this more realistic conception of behaviour and its genetic underpinnings [7 , 8] . This approach allows to simulate the evolution of complex genotypes and observe the resulting behaviours in robotic agents . Such simulations also make it possible to investigate the complex mechanistic constraints at play in the translation from genotype to phenotype [24] . A considerable body of work has already been dedicated to modeling social evolution with robotic approaches [14] . These studies have been interested in a large diversity of issues: the evolution of swarms [11] , the mechanics of division of labour in social insects [15 , 16] or the evolution of communication [10 , 17–19] . The evolution of cooperation in particular has been addressed in numerous papers . In the vast majority of this literature , however , social partners are genetically related [20] , whether motivated by design [21 , 22] or to study the evolution of altruism [9 , 23] . Few articles , in comparison , have been interested in the evolution of mutualistic cooperation between genetically unrelated individuals [19] . Moreover the specific problem posed by the stag hunt game , where cooperation is not the only evolutionarily stable strategy and a non-collective solution acts as a stable attractor , has never been studied in evolutionary robotics . In this paper , we use an experimental model where simulated robotic agents interact in a situation equivalent to the stag hunt and compare the results of our model to those of standard game-theoretic analyses . Our results shed new light on the influence of mechanistic constraints in the evolution of coordinated actions . We then use this model to explore realistic mechanisms that could drive the transition to collective behaviours .
We consider an environment with two hunters and several prey , both hares and stags . Hunters can choose to hunt either of these prey , earning different food rewards depending on whether they hunt alone or cooperate ( see Fig 1 ) . Food rewards for killing a prey are shown in Table 1 . A hare yields a reward of 50 , regardless of whether it is hunted in a solitary or cooperative fashion . A stag yields a reward of 500 for each hunter only if it is hunted cooperatively . If a stag is killed by a single hunter , it is still removed from the arena but is considered a failed hunt and rewards nothing . None of the rewards are split between cooperators . Simulated robotic agents are evaluated in an 800 by 800 unit square arena , which has four solid walls and is devoid of any obstacles aside from other agents . Each circular-shaped agent , with a diameter of 14 units , is equipped with two independent wheels and a collection of sensors . Hunters can use the information provided by 12 proximity sensors and a front camera . Proximity sensors have a range of approximately twice the diameter of the agent’s body , and provide the agent with the proximity of the nearest obstacle . They are evenly distributed around the agent’s body . The front camera consists of 12 rays with infinite range spread out in a 90 degree cone in front of the body . Each ray in the camera provides two different pieces of information about the first target it intersects with: the type of target ( hunter , hare , or stag ) and its proximity . This robot model facilitates the evolution of basic walls avoidance and agents recognition behaviours , which we consider not to be of interest here . Hence we separate obstacles recognition ( by the proximity sensors ) from agents’ recognition ( by the camera ) . Only the hunters are capable of movement; prey remain at their initial positions . ( Complementary experiments with moving prey capable of avoidance behaviours did not produce significantly different results; not shown . ) A prey is caught if any hunter remains close enough during a fixed amount of time steps ( 800 steps , in a simulation lasting 20 . 000 time steps ) . Cooperative hunting is defined as a prey with two hunters in catching distance at the time of its capture . Therefore , cooperation happens even if only one of the two hunters is in catching distance of the prey for most of the time , as long as the two hunters are there in the final step . The prey is then immediately replaced at a random position in the arena , thus keeping a fixed number of agents and prey during the whole simulation . The hunters’ behaviour is computed by an artificial neural network which maps sensory inputs to motor outputs . The neural network is a fully connected multi-layer perceptron with a single hidden layer of 8 neurons . The inputs of this network are the perceptions of the agent , with 12 neurons for the proximity sensors and 48 for the camera ( 4 for each of the 12 rays ) plus a bias neuron ( whose value is always 1 ) , for a total of 61 input neurons . The two outputs of the network control the speed of each of the agent’s wheels and the mapping function between inputs and outputs is a sigmoid function ( see Fig 2 ) . Changing the number of hidden neurons did not yield significantly different results ( not shown ) . To simulate evolution , we use an evolutionary algorithm to evolve the genome of the hunters . This genome is comprised of a collection of 410 real values in the range [0 , 1] , one for each of the neural network’s weights , and is initially randomized for each individual in the population . In order to obtain its fitness , each individual is successively paired five times with a partner randomly chosen each time ( except itself ) in the arena presented in the Experimental Setup subsection , for an evaluation round of 20 . 000 time steps . The payoff of the evaluated individual at the end of a round is given by the total amount of food it has managed to obtain by killing prey in this round . As this quantity depends heavily on the initial conditions ( random initial positions of the prey ) , five simulations are performed for each pair of individuals . The individual’s fitness is then obtained by computing the sum of payoffs averaged over the total number of simulations for the individual . In this case the number of simulations is 25 , with 5 partners and 5 simulations with each partner . Experiments were conducted using a Wright-Fisher model [12] with constant population size ( 20 individuals ) , which is commonly known as a fitness-proportionate selection method in evolutionary robotics [13] . Using this model , the population of the next generation is formed by a random sampling of offspring from the previous generation , with the probability of sampling a particular parent proportional to the parent’s fitness . Each offspring is simply a mutated clone of its parent; recombination is not included in our simulation . Consequently , new genotypes appear only through mutation . These mutations are performed using a Gaussian function , with a standard deviation of 2 × 10−1 and a mutation probability of 5 × 10−3 . Each experiment lasted 3000 generations . All simulation parameters are summarised in Table 2 .
In order to explore the evolutionary transition between the risk-dominant equilibrium ( hare hunting ) and the payoff-dominant equilibrium ( cooperative stag hunting ) , individuals first evolved in an environment composed solely of hares . This ensured that the populations initially reached the solitary equilibrium . Only then did we add stags and study the dynamics of evolution . Fig 3 ( a ) shows the evolution of the mean percentage of stags hunted successfully ( i . e . , hunted cooperatively ) out of the total number of prey hunted over time for 30 independent runs . Fig 3 ( b ) shows the mean proportion of each type of prey hunted during the last generation of each run . Stag hunting evolved in only one run out of 30 and even in that run accounted for less than 30% of the total number of prey hunted . In the other 29 runs , the individuals hunted only hares as they had previously evolved to do . These simulations demonstrate that the evolution of collective hunting is very unlikely when the population is composed of individuals who are already efficient solitary hunters . For comparison we simulated the same scenario using the standard game-theoretic version of the stag hunt , where the expression of the two types of behaviour was encoded by a single binary locus . Each individual in the population initially possessed the allele for hare hunting ( Fig 4 ) . Here the transition to collective hunting occurred in each of the 30 independent runs and this strategy then remained stable . This result differs drastically from the results of our robotic simulations in which this transition never fully occurred ( Mann-Whitney U test on the proportion of stags hunted successfully during the last generation , p-value <0 . 001 ) . In a second experiment , we wanted to investigate the evolution of hunting strategies “from scratch” , with the individuals’ genotypes initialized with random values , rather than evolved with a specific hunting strategy . Fig 5 shows the mean percentage of stags hunted over time and the mean number of prey hunted during the last generation . We observed the transition to a clearly cooperative strategy in a single run , while in two other runs , 50% of prey hunted were stags . In the 27 remaining runs the proportion of stags hunted was less than 25% . In comparison , in simulations using the standard game-theoretic version of the stag hunt where individuals are initially unable to hunt , stag hunting evolved and remained stable in every run ( see supporting information S1 Fig ) . The above experiments show that mechanistic constraints have a critical effect on the evolution of coordinated collective actions . In a simple game-theoretic analysis in which the hunting strategy is encoded by a single binary gene , collective behaviour systematically evolved . However , in a setting where the hunting strategy was determined by a more complex artificial neural network , cooperative behaviour evolved in fewer than 10% of cases . These results encourage further exploration into the evolutionary origin of coordinated collective actions and the mechanisms which may facilitate their evolution . In the following section , we explore two such mechanisms . In the next experiment , food was also rewarded for hunting a stag in a solitary fashion so that cooperative behaviour did not entail a risk . We wanted to study whether hunting a stag alone could act as a transition towards the evolution of the collective strategy . Hunting a stag alone was given the same reward as hunting a hare ( Table 3 ) , differing from classical models of the stag hunt . Fig 6 shows the results of robotic simulations where individuals initially evolved to hunt hares ( as in Fig 3 ) . As expected , the evolution of collective hunting was significantly facilitated when the risk of hunting stags alone was removed ( Mann-Whitney , p-value <0 . 001 ) . The populations completely switched to hunting stags in two runs out of 30 , and in three other runs , more than 50% of the prey hunted were stags , with a large part of the prey hunted cooperatively in each of these runs . However , in most of the runs ( 25 out of 30 ) , the evolved strategy was to hunt both types of prey in a solitary fashion . From these results , it entails that the individuals are still hindered by the evolution of a successful coordination strategy . Genetic relatedness among social partners is known to influence the evolution of many types of social traits [1] . In particular , [6] showed how it can facilitate the evolution of cooperation in a stag hunt game [6 , chapetr 3] . It can yield more frequent interaction between cooperators , which in turn increases their probability of benefiting from cooperative behaviour . In order to include this mechanism , we considered an extreme situation in which each individual is always paired with a clone of itself , known as “clonal selection” in robotics , ensuring a maximal genetic relatedness of 1 . These results show that genetic relatedness has a positive effect on the evolution of cooperation ( Fig 7 ) . In four out of 30 runs the population evolved the cooperative strategy . Moreover , in two other runs , stags accounted for more than 75% of prey hunted , as compared to less than 25% without relatedness ( Mann-Whitney , p-value <0 . 005 ) . When the initial population was random , rather than only hare hunters ( see supporting information S2 Fig ) , the positive effect of genetic relatedness was also observed in 12 out of 30 runs , where more than 50% of prey hunted were stags .
There is a profound difference between evolutionary game-theoretic and robotic simulations of the stag hunt . Using identical model parameters , the transition from the solitary equilibrium to the social equilibrium always occurred in game-theoretic simulations , but was extremely unlikely in robotic simulations , occurring in 1 run out of 30 . The complexity of the mapping between genotype and phenotype is responsible for much of this contrast . Individuals involved in a coordination game such as the stag hunt face a chicken & egg problem: the cooperative behaviour must be beneficial in order to evolve , but no individual can benefit from this behaviour unless the behaviour is already expressed by other individuals . When binary variation at a single genetic locus encodes the expression of the solitary or cooperative strategy , a single mutation is sufficient for a cooperative mutant to appear in a resident population of solitary individuals . In a finite population , demographic stochasticity can then lead to the rise of cooperators above the invasion threshold , at which point natural selection leads to their fixation , switching from a solitary equilibrium to a social one . In contrast , in our robotic simulations , the mapping between genotype and phenotype is more complex . Adopting the social strategy entails both a modification of the preferred hunting target and the ability to coordinate with others . Thus , several mutations are necessary for the appearance of full-fledged cooperative behaviour . As several individuals must carry these multiple mutations for the behaviour to become beneficial , the transition to the cooperative equilibrium is nearly impossible . In particular , in our robotic simulations we were able to observe that coordination entails a specific and rather complex behaviour . Fig 8 ( see also supporting information S1 Movie ) shows the behaviours evolved by the best individuals in the cooperative run shown in Fig 5 ( Run 9 ) . The solution they evolved for coordination was to circle around one another , allowing each of them to constantly see their partner while both moving closer to a stag . This behaviour was replicated in every cooperative run . We thus observed the evolution of an ingenious ( given the agents’ limited capabilities ) and complex hunting strategy . These findings demonstrate that the practical mechanics of behaviour can have important evolutionary consequences , and that models which ignore these properties may lead to misleading predictions . Moreover , the evolution of cooperation is also strongly impacted by ecological features . Social hunting poses a bootstrapping problem because it entails both a modification of the preferred hunting target and an ability to coordinate with others . Its evolution can be facilitated , therefore , if hunters have a reasonable probability of hunting the same prey as their partner , just by chance , with no need of active coordination . Biologically , this could occur if hunters live in a dense social environment ( with many other hunters in the vicinity ) , and/or if the density of prey is low , such that the likelihood of ending up on the same prey is large . To test this possibility , we conducted additional experiments where the density of prey was varied . The number of prey was whether ( 1 ) decreased from 18 to 6 or ( 2 ) increased from 18 to 30 . The population was initially constituted of hare hunters and we kept the same ratio of prey as in previous experiments ( i . e . 50% of hares and 50% of stags ) . We show ( see supporting information S3 Fig ( a ) ) that when the number of prey is decreased ( 6 ) the transition to a cooperative strategy is facilitated ( Mann-Whitney , p-value <0 . 05 ) as in 9 runs out of 30 , more than 30% of the prey hunted are stags . In comparison , a higher density of prey ( 30 ) entails that it is impossible to evolve cooperation ( see supporting information S3 Fig ( b ) ) . These results reinforce our claim that the practical mechanics of coordination are crucial in understanding the evolution of cooperation . In particular , here , the precise ecological situation faced by individuals plays a key role in the transition to the collective equilibrium . Finally , the complexity of coordination suggests that the recycling of a previously evolved trait could be necessary for the transition to cooperation , i . e . individuals could coordinate thanks to behavioural features that may not have been selected for cooperation at first . Such features could include the evolution of communication , or a leader-follower strategy . The role of both of these behaviours has already been studied in real-life stag hunt type interactions in chimpanzees and human children [25 , 26] , and there is an already extensive literature in evolutionary robotics on their role in the evolution of collective actions [16 , 19 , 22 , 27] . This offers some directions for future works on this problem . | Mutualistic behaviours wherein several individuals act together for a common benefit , such as a collective hunt , are often deemed of minor interest by theoreticians in evolutionary biology . These behaviours benefit all the individuals involved , and therefore , so the argument goes , their evolution is straightforward . However , mutualistic behaviours do pose a specific kind of evolutionary problem: they require the coordination of several partners . Indeed , a single individual expressing a preference for cooperation cannot benefit if others wish to remain solitary . Here we use simulations in evolutionary robotics to study the consequences of this problem . We show that it constitutes a far more serious obstacle for the evolution of cooperation than was previously thought on the basis of game theoretical analyses . We find that the transition from solitary to cooperative strategies is very unlikely , and we also observe that successful cooperation requires the evolution of a specific and rather complex behaviour , necessary for individuals to coordinate with each other . This reveals the critical role of the practical mechanics of behaviour in evolution . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"ecology",
"and",
"environmental",
"sciences",
"engineering",
"and",
"technology",
"predator-prey",
"dynamics",
"applied",
"mathematics",
"population",
"dynamics",
"robotic",
"behavior",
"simulation",
"and",
"modeling",
"algorithms",
"evolutionary",
"computation",
"mathemat... | 2016 | To Cooperate or Not to Cooperate: Why Behavioural Mechanisms Matter |
Most genomes of bacteria contain toxin–antitoxin ( TA ) systems . These gene systems encode a toxic protein and its cognate antitoxin . Upon antitoxin degradation , the toxin induces cell stasis or death . TA systems have been linked with numerous functions , including growth modulation , genome maintenance , and stress response . Members of the epsilon/zeta TA family are found throughout the genomes of pathogenic bacteria and were shown not only to stabilize resistance plasmids but also to promote virulence . The broad distribution of epsilon/zeta systems implies that zeta toxins utilize a ubiquitous bacteriotoxic mechanism . However , whereas all other TA families known to date poison macromolecules involved in translation or replication , the target of zeta toxins remained inscrutable . We used in vivo techniques such as microscropy and permeability assays to show that pneumococcal zeta toxin PezT impairs cell wall synthesis and triggers autolysis in Escherichia coli . Subsequently , we demonstrated in vitro that zeta toxins in general phosphorylate the ubiquitous peptidoglycan precursor uridine diphosphate-N-acetylglucosamine ( UNAG ) and that this activity is counteracted by binding of antitoxin . After identification of the product we verified the kinase activity in vivo by analyzing metabolite extracts of cells poisoned by PezT using high pressure liquid chromatograpy ( HPLC ) . We further show that phosphorylated UNAG inhibitis MurA , the enzyme catalyzing the initial step in bacterial peptidoglycan biosynthesis . Additionally , we provide what is to our knowledge the first crystal structure of a zeta toxin bound to its substrate . We show that zeta toxins are novel kinases that poison bacteria through global inhibition of peptidoglycan synthesis . This provides a fundamental understanding of how epsilon/zeta TA systems stabilize mobile genetic elements . Additionally , our results imply a mechanism that connects activity of zeta toxin PezT to virulence of pneumococcal infections . Finally , we discuss how phosphorylated UNAG likely poisons additional pathways of bacterial cell wall synthesis , making it an attractive lead compound for development of new antibiotics .
Almost all prokaryotic genomes encode toxin–antitoxin ( TA ) systems [1] . These loci consist of bicistronic operons that encode for a bacterial toxin and its cognate inhibitor , which neutralizes the toxin under dormant conditions . Once de novo synthesis from the operon is impaired , continuous proteolytic degradation of the antitoxin eventually releases the toxin and , depending on the functional mechanism , induces cell stasis or cell death . Thus , TA systems have been linked with numerous cellular functions , including programmed cell death , maintenance of mobile genetic elements , stress response , persistence , and biofilm formation [2]–[5] . To understand how TA systems fulfill such a variety of tasks , it is crucial to unravel the molecular principles that define the mode of action of the toxins . Moreover , understanding these systems is likely to disclose new strategies for the development of antibiotic agents [6] , [7] . Members of the epsilon/zeta TA family have been shown to stabilize resistance plasmids in major human pathogens such as Streptococcus pyogenes , Enterococcus faecium , and Enterococcus faecalis [8]–[10] . The epsilon/zeta system is encoded from a bicistronic operon , which is regulated by the repressor protein omega [11]–[13] . Upon failure of epsilon biosynthesis , the zeta toxin is released from epsilon by continuous antitoxin degradation through AAA+ proteases . Eventually , zeta becomes freed , leading to cell death [11] , [14] . In addition to these plasmid-encoded epsilon/zeta systems , chromosomally encoded systems ( PezAT for pneumococcal epsilon/zeta ) have recently been identified on different integrative and conjugative elements of Streptococcus pneumoniae [15]–[17] . PezT and zeta toxins share 42% sequence identity and are structurally highly homologous [17] . In contrast , the antitoxin PezA is a multidomain protein and its C-terminal domain is similar to epsilon in its primary as well as tertiary structure . Although , the N-terminal helix-turn-helix domain of PezA acts similarly to the omega protein as a transcription repressor , the proteins are not evolutionaryily related [17] . The epsilon/zeta and the PezAT system form similar heterotetrameric complexes in which two antitoxin molecules inhibit two toxins . In addition to these pronounced structural similarities , mutational studies have shown that the two systems are functionally equivalent [17] , [18] . PezAT systems are found encoded on pneumococcal pathogenicity islands that support their host with virulence factors and resistance to different antibiotics [15]–[17] . Notably , the PezT toxin of such systems was reported to accelerate the progression of pneumococcal infections , implying a role as a possible virulence factor [19] . Although epsilon/zeta systems were thought to be exclusively found in Gram-positive bacteria , recent reports have described homologous systems in Gram-negative pathogens such as Neisseria gonorrhoeae and enterotoxigenic Escherichia coli B7A [20] , [21] . The broad distribution of epsilon/zeta TA systems within the bacterial kingdom suggests that they utilize a ubiquitous bacteriotoxic mechanism . Structures of zeta toxins as well as mutational studies suggested that their toxicity is connected to an ATP-dependent phosphorylation event [17] , [18] . However , whereas all other TA systems known to date poison macromolecules involved in translation or replication [22]–[26] , the target of the cytosolic zeta toxin family remained inscrutable [8] , [14] , [27] . Here , we reveal the mechanism used by zeta toxins to induce programmed cell death in bacteria . Since expression of wild-type zeta toxins leads to either instantaneous cell death or spontaneous mutation of the open reading frame [8] , [17] , [27] , we established a system with which we could follow formation of the toxic phenotype at moderate time scales . This system enabled us to show that zeta toxins provoke an autolytic phenotype as a consequence of impaired cell wall integrity and breakdown of the osmotic barrier . We found that zeta toxins represent a novel family of kinases that manipulate a central metabolic branch point of bacterial cell wall synthesis . The crystal structure of a zeta toxin bound to its target allowed us to map the enzyme–substrate interactions and revealed that the kinase activity is indeed responsible for the toxic function in vivo . In fact , we were able to show that the phosphorylated product inhibits MurA , the enzyme responsible for the first step of peptidoglycan synthesis in bacteria .
Genetic manipulations of the full-length zeta or homologous PezT toxins without the cognate antitoxin are unfeasible because of the high toxicity of the proteins . Moreover , toxin variants that can be isolated from surviving clones are generally inactive because of spontaneous mutations [8] , [17] . We found that a carboxy-terminally truncated variant lacking the last 11 amino acids ( PezTΔC242 , henceforth referred to simply as PezT ) can be stably maintained in E . coli . Importantly , sequencing of the plasmid DNA isolated from E . coli cells after prolonged expression experiments showed that this variant did not accumulate any spontaneous mutations . Nevertheless , this variant still retained the toxic phenotype . Half an hour after induction of PezT expression , we found that the cells in such cultures formed midcell-positioned bulges following membrane permeabilization and lysis ( Figures 1A , S1A , and S1B ) . Notably , cells that had survived to that point were apparently unable to undergo cytokinesis , even though chromosome replication was complete ( Figure S1B ) . One hour after induction , massive cell death had occurred and the few intact cells that remained were characterized by small size and an ovoid morphology ( Figure S1C ) . To exclude the possibility that the observed phenotype is due to general overexpression , we performed control experiments with cells bearing the expression plasmid of the nontoxic variant PezT ( D66T ) [17] . We did not observe bulge formation or lysis after induction of PezT ( D66T ) ( Figure S1D ) . Therefore , we concluded that the observed phenotype was caused by the specific action of PezT . Lysis through bulge formation implied that cells poisoned by PezT suffered from defects in their cell wall integrity . Bacterial growth and binary fission demand a tightly controlled balance between murein synthesis and degradation . Perturbations of this balance are known to cause uncontrolled peptidoglycan degradation by murein hydrolases following lysis because of the intracellular turgor pressure [28] . Since PezT appeared to kill cells predominantly prior to cytokinesis , which requires the build-up of septal murein , we speculated that the toxin targets cell wall synthesis . To test this hypothesis , we induced toxin expression in adherently growing cultures , which are less exposed to osmotic and mechanic stress than liquid cultures [29] . One hour after induction , we observed that these cells adopted a bloated , misshapen morphology that was clearly different from the normal rod-like shape of cells expressing nontoxic PezT ( D66T ) ( Figure 1B ) . In fact , these cells resembled spheroplasts that form after treatment with β-lactam antibiotics such as ampicillin , which are authentic inhibitors of bacterial peptidoglycan synthesis [30] . To further corroborate our hypothesis that PezT targets cell wall synthesis , we probed whether toxin expression resembles β-lactam treatment in general . We found that the onset of cell death caused by PezT expression was preceded by a strong rise in membrane permeability , by measuring influx kinetics of the membrane-impermeable dye propidium iodide ( Figure 1C ) . Eventually , membrane permeability increased enough to allow the 31-kDa periplasmatic ribonculease I to enter the cytosol , resulting in massive rRNA degradation ( Figure S1E ) . Strikingly , we could provoke almost identical characteristics of cell death in E . coli expressing nontoxic PezT ( D66T ) by additional treatment of the culture with ampicillin ( Figures 1C and S1F ) . In contrast , we did not observe any of these symptoms when cells were treated with tetracycline , which solely targets protein biosynthesis ( Figures 1C and S1F ) . In summary , we concluded that the toxic mechanism of PezT causes inhibition of cell wall synthesis that eventually provokes bacterial autolysis . Nevertheless , the identity of the molecular target of PezT remained enigmatic , since inhibition of cytosolic steps of bacterial cell wall synthesis can occur on a multitude of levels [31] . However , recent reports had shown that the zeta protein from S . pyogenes is toxic in eukaryotes such as Saccharomyces cervisiae [32] . This suggested the presence of a putative ubiquitous substrate whose modification by the toxins corrupts a shared metabolic branch point found in prokaryotes and eukaryotes . Such a fundamental metabolite is uridine diphosphate-N-acetylglucosamine ( UNAG ) , which is produced in the hexosamine biosynthesis pathway and is found in all kingdoms of life [33] . This nucleotide sugar is essential for the formation of a plethora of glycoconjugates , among them the peptidoglycan macromolecule in prokaryotes . We next set out to determine whether UNAG is indeed the substrate of PezT . Given the structural similarity of PezT and zeta toxins with phosphotransferases [17] , [18] , we speculated that they modify UNAG by phosphorylation . Therefore , we purified our toxic PezT variant from poisoned E . coli and probed its activity in vitro . Indeed , using anion exchange chromatography , we observed an ATP- and Mg2+-dependent modification of UNAG ( Figures 2A and S2A–S2C ) . In addition to adenosine diphosphate formation , we noticed a product that was more negatively charged than the substrate UNAG and that eluted close to the remaining ATP . This strongly suggested that the PezT toxin had phosphorylated UNAG . We could exclude that the observed UNAG modification was the result of a contaminating activity , since addition of stoichiometric amounts of the cognate antitoxin PezA completely inhibited turnover of both UNAG and ATP ( Figures 2A and S2B ) . Further , we showed that the PezT activity was specific for the presence of the 2′-N-acetyl group on the sugar moiety and the stereoisomeric form of UNAG , since selectivity for uridine diphosphate ( UDP ) –glucose and UDP-N-acetylgalactosamine was dramatically reduced ( Figure S2E and S2F ) . Most importantly , we also showed that the zeta toxin from S . pyogenes can catalyze the same reaction as PezT and modifies UNAG using ATP ( Figure 2B ) . This strongly suggests that the enzymatic function described is a conserved activity of the entire family of zeta toxins . We investigated next whether PezT indeed phosphorylates UNAG . To this end , we performed electrospray ionization spectrometry experiments , which showed that the product of the PezT toxin and UNAG differ by the mass of a phosphoryl group ( Δm/z = 80; Table S1 ) . Using fragmentation by tandem mass spectrometry we could cleave the phosphorylated UNAG molecule into two main fragments , one with a mass corresponding to UDP and one corresponding to N-acetylglucosamine with a phosphoryl group attached to a hydroxyl group ( Table S1 ) . Ultimately , we showed by nuclear magnetic resonance ( NMR ) ( Figure S3 ) that the PezT toxin had attached a phosphoryl group to the 3′-hydroxyl group of the N-acetylglucosamine moiety . In conclusion , we have demonstrated that zeta and PezT toxins are UNAG kinases that form UDP-N-acetylglucosamine-3′-phosphate ( UNAG-3P ) from UNAG using ATP and Mg2+ as a cofactor ( Figure 2C ) . Site directed mutagenesis studies of PezT and zeta toxins yielded several inactive toxin variants that have been linked with binding of the at-that-time-unknown substrate [17] , [18] . We therefore were interested in a structural characterization of a toxin bound to the substrate . Previous structural reports on epsilon/zeta and PezAT complexes showed that the toxin is structurally similar to poly- and mononucleotide kinases [17] , [18] . Based on these findings , the active site of zeta and PezT toxin has been identified and a mechanism of toxin inhibition has been proposed . Once bound to the antitoxin , side chain groups of the antitoxin block the ATP binding site [17] , [18] . The second substrate binding site , however , remains unaffected by the antitoxin . Thus , we surmised that PezT and zeta toxins are still able to bind UNAG even if they are inhibited by the antitoxin . Indeed , we could determine the crystal structure of the epsilon/zeta TA complex from S . pyogenes bound to UNAG at 2 . 7 Å resolution ( Figure S4; Table S2 ) . UNAG binds to a deep cleft at the molecular surface of the zeta toxin ( Figure 3 ) . The side chain group of Asp67 forms a hydrogen bond ( 3 . 2 Å ) to the 3′-hydroxyl group of the amino sugar moiety of UNAG . Based on superposition with structurally related phosphotransferases , this particular side chain of the zeta toxin had been suggested to be the catalytic base , which deprotonates the substrate [18] . For PezT toxin , Asp66 is the functionally equivalent residue , and , as mentioned above , its mutation to a threonine residue totally abolishes toxicity . Since we could not detect any activity of PezT ( D66T ) in our kinase assay ( Figure S2D ) , we conclude that the phosphoryltransfer reaction is required for the bacteriotoxic mechanism of zeta toxins . Thr118 of the zeta toxin formed a hydrogen bond with an oxygen atom of the β-phosphate group of UNAG ( 2 . 8 Å ) , and exchange of this conserved residue in the PezT toxin to a valine residue also yielded a nontoxic variant [17] . The conserved Thr121 , which is in close proximity to the bound UNAG but did not contact the substrate directly , has been shown to interfere with toxicity of PezT only mildly [17] . Additional specific contacts between the zeta toxin and UNAG are electrostatic interactions of Arg120 with the phosphate groups of UDP . Apart from van der Waals interactions , specific hydrogen bonds are formed between Thr128 and the uracil base and between Glu100 and the 2′-hydroxyl group of the UDP ribose . Most likely , mutations of these residues will affect toxicity as well . Thus , our data show that UNAG binding by the zeta toxin is mediated by residues that yield nontoxic and kinase-deficient zeta and PezT variants in vivo and in vitro . In accordance with our in vitro results , we also showed that UNAG-3P is the main product of the PezT toxin activity in vivo . In fact , we identified enriched UNAG-3P in low-molecular-weight-metabolite pools of PezT-poisoned cells obtained by aqueous acetonitrile extraction and HPLC ( Figure 4 ) . We confirmed the presence of UNAG-3P in these fractions by chromatography of authentic standards ( Figure 4 ) and by mass spectrometry ( Table S1 ) . Our experiments suggested that cells that contained active PezT toxin constantly accumulated UNAG-3P , whereas the pool of UNAG was largely depleted . Modification of the amino sugar 3′-hydroxyl group by PezT and zeta toxins suggests several possible scenarios by which UNAG-3P can interfere with peptidoglycan synthesis . One of these is inhibition of the conserved enolpyruvyl transferase MurA , which catalyzes the initial step of muramic acid synthesis . Subsequent to the hexosamine biosynthesis pathway , MurA modifies the amino sugar 3′-hydroxyl group of UNAG ( Figure 5A ) and thereby provides the starting point for peptidoglycan biosynthesis [34] . MurA catalyzes the transfer of the enolpyruvyl moiety of phosphoenolpyruvate to the 3′-hydroxyl group of N-acetylglucosamine , forming enolpyruvyl-UNAG and inorganic phosphate ( Figure 5A ) . Due to modification of the 3′-hydroxyl group by the phosphate group , we speculated that MurA would be unable to utilize UNAG-3P , which renders the phosphorylated nucleotide sugar unusable for peptidoglycan synthesis . Indeed , we did not observe any MurA-dependent turnover of UNAG once it had been phosphorylated by PezT in an in vitro activity assay designed to measure the release of inorganic phosphate ( Figure 5B ) . During the native reaction of MurA , a negatively charged tetrahedral adduct is formed [35] . Because of the resemblance of the MurA tetrahedral intermediate and UNAG-3P ( Figure 5A ) , we hypothesized that UNAG-3P could bind to MurA in a fashion similar to that of the normal substrate UNAG . In this hypothesis zeta toxins would not only produce a stable dead-end metabolite but would also halt cell wall synthesis directly by inhibition of MurA . In fact , we found that increasing UNAG-3P concentrations decreased the apparent turnover rate of UNAG by MurA ( Figure S5 ) . This strongly suggests that UNAG-3P binds to MurA , forming an inactive , non-productive complex . Furthermore , we observed that increasing UNAG-3P concentrations led to an increase of the apparent Km of MurA for UNAG without affecting the maximal reaction rate Vmax when we analyzed the initial rate of product formation at varying UNAG and UNAG-3P concentrations ( Figure 5C ) . This implies that the phosphate group of UNAG-3P mimics the tetrahedral intermediate during MurA catalysis and is a reversible inhibitor that eliminates turnover of phosphenolpyruvate via the modification of the 3′-hydroxyl group of the N-acetylglucosamine moiety . The Michaelis-Menten dataset could be fitted globally by a competitive inhibition mechanism that yielded a Ki of 7 µM for UNAG-3P in the presence of physiological concentrations of phosphoenolpyruvate . ( Figure 5A ) . Thus , zeta toxins not only produce a metabolic dead-end product but additionally form a competitive inhibitor for peptidoglycan synthesis . Most antibiotics that target peptidoglycan synthesis are known to rapidly kill bacteria during exponential growth but fail to kill slowly growing and stationary cells [36] . Therefore , we considered whether slow growth was also an option to evade cell killing through the mechanism utilized by zeta toxins . When we induced toxin expression at low optical densities , where the general growth rate was rapid , we observed the lytic phenotype described above . In contrast , induction of PezT in cells approaching stationary phase did not result in lysis ( Figure 6A ) . This indicated that UNAG-3P was predominantly toxic for fast-dividing cells , whereas the toxic effect was less severe for cells proliferating more slowly . In order to provoke slow growth , we grew cells in a nutritionally deprived medium . Generally , cells growing in such a medium were much smaller and had an increased doubling time ( Figures 6A and S6A ) . As expected , lysis caused by PezT expression in nutritionally deprived medium was less severe and observable only after induction at the lowest optical density tested ( OD600 = 0 . 2 ) ( Figure 6A ) . We could exclude that this finding was caused by differential expression levels of PezT as the cytosolic concentrations of the recombinant protein were comparable to those of cultures grown in fresh Luria broth ( LB ) medium ( Figure S6B ) . When we inspected cells of cultures surviving toxin expression in nutritionally deprived medium 3 h after induction , the majority of cells had accumulated membrane protrusions and lost their characteristic rod-like shape ( Figure 6B ) . Thus , PezT was still performing its disruptive activity within peptidoglycan metabolism . However , under the growth conditions tested , the damage done to the cell wall was not sufficient for autolysis of the majority of cells . Therefore , we conclude that minimizing the demand on cell wall precursors through small size and slow growth enables bacteria to survive toxin action for an extended period of time .
The synthesis of muramic acids is regulated by a negative feedback loop in which MurA is inhibited by its downstream product UDP-N-acetylmuramic acid in E . coli [37] . Thus , PezT and zeta toxin activity impairs muramic acid synthesis in two different ways: first , it depletes the pool of UNAG precursors and , second , UNAG-3P is a competitive inhibitor of MurA . Thus , increasing UNAG-3P levels favor UNAG-3P production by inhibition of the competing murein synthesis pathway . UNAG-3P will interfere with peptidoglycan synthesis in Gram-negative and Gram-positive organisms , since MurA is highly conserved among prokaryotes [34] . Furthermore , since UNAG is an essential precursor for teichoic and lipoteichoic acid synthesis in Gram-positive bacteria [38] , formation of UNAG-3P is also likely to be detrimental to their synthesis . UNAG phosphorylation will also compete with lipid A synthesis of Gram-negative bacteria , since condensation of the lipid anchor and UNAG is performed via the amino sugar 3′-hydroxyl group of UNAG . Thus , zeta toxins ubiquitously interfere with synthesis of a variety of cell wall components , independent of the cell wall architecture , and cells harboring an epsilon/zeta or PezAT TA system are provided with a potent killing system . UNAG is an abundant metabolite in bacteria [39] , and we therefore do not expect the activity of PezT and zeta toxins to immediately kill the host . This is in agreement with the previous finding that depending on the dose and exposure time to free toxin , its activity either leads to growth arrest or cell lysis [14] . We argue that the toxin acts as a killer when the toxin activation rages out of control . Such a situation is realized when , for instance , a plasmid encoding an epsilon/zeta system TA is lost and the zeta toxin is released via degradation of the epsilon antitoxin . Moreover , we speculate that the bactericidal action of zeta toxin release will be higher during fast growth , when cells are more vulnerable to inhibition of cell wall synthesis . This matches our finding that cells react differently to the toxin activity , depending on the environmental conditions . We surmise that cells can readjust their cell wall metabolism during situations where toxin activation is temporary and reversible , for instance during transcriptional or translational arrest . Our in vivo experiments suggest that a subpopulation of cells can adapt to the drain in cell wall precursor components by adopting a state of dwarfism paired with slow growth . Such different scenarios can explain the unresolved ambiguity as to why zeta toxins perform postsegregational killing in one situation or trigger cell stasis in another [8] , [14] . The pneumococcal PezT toxin has been suggested to support virulence of its pathogenic host during infection , since strains in which the PezT gene has been deleted are attenuated in mouse models of infection [19] . One of the major pneumococcal virulence factors that accelerate infection progress is the pore-forming toxin pneumolysin [40] , [41] . Intriguingly , pneumolysin is located in the cytosol of S . pneumoniae and requires a bacteriolytic activity in order to be released into the lumen [42] , [43] . It is therefore tempting to speculate that pneumococcal strains bearing a PezAT gene cassette are equipped with an additional option to trigger lysis and pneumolysin release in a subpopulation of cells ( Figure 7 ) . Release of PezT toxin from PezA antitoxin might appear under conditions that lead to either prolonged down-regulation of protein biosynthesis or enhanced PezA degradation . Similar to other TA systems , down-regulation of PezA biosynthesis could occur , for instance , during antibiotic treatment or amino acid starvation , when general transcription or translation is impaired [44] . Under these conditions , constant proteolytic degradation of PezA by constitutive proteases would eventually activate PezT . On the other hand , different TA systems within a single host were shown to be specifically activated under different stress conditions [45] . Whereas in case of the epsilon/zeta system , constant degradation of epsilon by the constitutive protease Lon and ClpXP [14] leads to toxin activation , PezT release seems not to be performed by a housekeeping protease [46] . Thus , it is rather likely , that activation of the PezAT system in its host organism is tightly controlled and requires a specific event such as stress response [46] . The ability of TA systems to induce cell lysis or cell stasis has also been linked to biofilm and persister cell formation in pathogens [47] . Biofilm formation of S . pneumoniae , for example , is thought to be initiated by the formation of cell aggregates that require autolysis of a subpopulation [48] , [49] . Most probably , moderate activation of PezT and zeta toxins will support autolysis of rapidly dividing cells , and thus PezT and zeta toxins can be beneficial for the entire cellular population despite the suicide of individual cells . Cells with reduced metabolism and growth , such as persisters , might survive the toxins' activity and are thus selected ( Figure 7 ) . Additionally , UNAG-3P formation is also likely to interfere with hyaluronic acid capsulation of streptococci , which needs to be down-regulated and reduced during biofilm formation [50] , [51] , since condensation of the polysaccharide in this pathway is linked via the amino sugar 3′-hydroxyl group of UNAG . Eventually , survivors will be protected in a biofilm and can recover from stress conditions and toxin activation by synthesis of new cognate antitoxin . UNAG-3P is a suicide antibiotic , because bacteria are harmed by their self-inflicted enzymatic activity depending on environmental conditions . This is in contrast to common antibiotics that bacteria produce and target against other species . Additionally , UNAG-3P is a naturally derived lead compound for the development of novel antibiotics . Our results imply that either activation or inhibition of epsilon/zeta and PezAT systems will interfere with the fate of the host bacteria and thus make them a potent Achilles' heel for microbes .
For microscopy , cultures of E . coli BL21-CodonPlus ( DE3 ) -RIL ( Stratagene ) bearing pET28b ( pezTΔC242 ) or pET28b ( pezTΔC242 ( D66T ) ) were grown in 100 ml of LB medium supplemented with kanamycin ( 50 µg/ml ) and chloramphenicol ( 34 µg/ml ) at 37°C in unbaffled flasks with mild shaking . Protein expression was induced by addition of IPTG to 1 mM at an OD600 of 0 . 4 . To evaluate the effects of toxin expression on cell membrane integrity , the BacLight live/dead bacterial viability kit ( Molecular Probes ) was used according to the supplier's instructions . Phase contrast and fluorescence images were captured with a ProgRes C3 CCD camera ( Jenoptic ) on an Axiovert 135 microscope ( Zeiss ) using an oil immersion objective lens ( 100×/N . A . 1 . 3 ) . The viability of the culture was assessed using filters with 450-nm to 490-nm excitation and 520-nm long pass emission for green fluorescence ( live ) or 545-nm band pass excitation and 590-nm long pass emission for red fluorescence ( dead ) . To additionally improve visualization of membrane defects , 200 µl of culture was withdrawn and immediately fixed in 1 ml of a 1∶3 ( v/v ) mixture of acetic acid and methanol . Cells were pelleted by centrifugation , resuspended in 0 . 9% ( w/v ) NaCl , and spotted on an LB agar slab . Subsequently , cells were inspected by phase contrast microscopy using an Axiovert 405M ( Zeiss ) equipped with a 100×/N . A . 1 . 25 oil immersion objective lens . Adherent cultures were derived from a liquid culture grown overnight and surviving PezTΔC242 expression . Cells were spotted on a microscope slide covered with LB agar and incubated for 1 h at 37°C . Protein expression was induced by diffusion of IPTG into the agar drop sandwiched between cover slides and subsequently incubated for 1 h at 37°C before inspection . Unfixed cells were inspected directly by phase contrast microscopy as described above . Equivalent growth experiments using preconditioned LB medium are described in Text S1 . To measure time-resolved breakdown of the osmotic barrier , cells were grown in LB medium at 37°C to an OD600 = 0 . 2 in absence of any inducing agent . Next , 75 µl of culture was diluted with an equal volume of LB containing 1 mM IPTG and 20 µM propidium iodide . Control cultures expressing PezT ( D66T ) were additionally supplemented with ampicillin ( 50 µg/ml ) or tetracycline ( 15 µg/ml ) . Samples for baseline correction were not inoculated with bacteria . All samples were transferred to a black 96-well microtiter plate ( Corning ) and were incubated at 37°C on a Thermomixer comfort ( Eppendorf ) equipped with a MTP sample holder . Breakdown of the osmotic barrier and staining of cytosolic DNA/RNA was monitored by measuring fluorescence at 620 nm in a Varioskan Flash ( Thermo Scientific ) with fluorescence excitation set to 520 nm at 20-min intervals . The optical density was recorded from the same samples in absorbance mode . Between measurements , the microtiter plates were covered with an AirPore tape sheet ( Qiagen ) . For each time series , fluorescence intensities were baseline corrected and averaged ( n = 6 ) . Experiments monitoring breakdown of the osmotic barrier by breakdown of ribosomal RNA are described in Text S1 . The kinase activity of PezTΔC242 and its inhibition by PezA were investigated by incubating 1 µM recombinant toxin in buffer R ( 25 mM HEPES-NaOH [pH 7 . 5] , 100 mM NaCl , 5 mM MgCl2 ) supplemented with 1 mM ATP and 0 . 25 mM of the nucleotide sugars UNAG , UDP-N-acetylgalactosamine , or UDP-glucose at 25°C for the times mentioned in the respective figure captions ( Figures 2 and S2 ) . All nucleotide sugar species were purchased from Sigma . After incubation , samples were diluted 1∶2 with H2O , filtered and applied to a 1-ml MonoQ column ( GE Healthcare ) equilibrated in 50 mM Tris-HCl ( pH 8 . 0 ) at 8°C . The different nucleotide/nucleotide sugar species were eluted with a gradient to 1 M NaCl . Inactivity of the nontoxic variant PezTΔC242 ( D66T ) was confirmed by performing the same assay using 3 µM protein . Since inhibition of the S . pyogenes zeta toxin by the epsilon antitoxin is much weaker than in the PezAT system found in S . pneumoniae [11] , [46] , we could verify the phosphoryltransfer reaction of the zeta toxin directly by extended incubation ( 24 h ) of 1 µM purified complex with the reaction mixture described above . Quantitative production of UNAG-3P as well as electrospray ionization tandem mass spectrometry and NMR experiments verifying its identity are described in Text S1 . The epsilon/zeta complex was purified as described previously [18] . The complex was crystallized using the hanging-drop vapor diffusion method by mixing 1 µl of protein—12 mg/ml protein in buffer ( 25 mM NaCl , 50 mM Tris-HCl [pH 7 . 5] ) —and 1 µl of reservoir solution containing 0 . 2 M sodium acetate ( pH 4 . 8 ) , 16% ( w/v ) PEG 1500 , and 7% ( v/v ) 2-methyl-2 , 4-pentanediol . Crystals were soaked for 30 min in mother liquor supplemented with 10 mM UNAG and flash cooled in liquid nitrogen . Diffraction data were collected at the Swiss Light Source ( Villigen , Switzerland ) , beamline X10SA at 100 K . Phases were refined by molecular replacement methods with REFMAC [52] using the apo-structure of epsilon/zeta ( Protein Data Bank accession number 1GVN ) as starting model . Detailed descriptions of the methods applied are found in Text S1 . Atomic coordinates and structure factor amplitudes have been deposited in the Protein Data Bank ( http://www . pdb . org/ ) under accession number 3Q8X . E . coli cells were grown as described for the microscopy experiments . Expression of the PezT proteins was induced at an OD600 of 0 . 4 for 1 h . Cells were harvested by centrifugation and resuspended in ice-cold 80% ( v/v ) aqueous acetonitrile solution . The suspension was incubated for 30 min on ice and with regular periods of gentle agitation . Cells were pelleted by centrifugation ( 20 min , 11 , 000g , 4°C ) , and the supernatant shock frozen in liquid nitrogen and stored at −80°C . For HPLC analysis , the extract was concentrated by vacuum concentration and resuspended in deionized water . Insoluble compounds were removed by high-speed centrifugation and filtration with Ultrafree-MC centrifugal filter units ( Millipore ) . The A260 of the extracts was adjusted to a value of 10 AU , and a 30-µl sample volume was applied to a Partisil-5 SAX RAC II column ( Whatman ) equilibrated with 5 mM KH2PO4 . Bound metabolites were eluted with a binary gradient ( 1 ml/min , 36 column volumes ) to 500 mM KH2PO4 , and the absorbance monitored at 260 nm . The cell extracts obtained from cells expressing PezTΔC242 showed an additional peak eluting at the same retention time as purified UNAG-3P . Fractions containing this peak were pooled , desalted , and concentrated . Analysis by electrospray ionization tandem mass spectrometry demonstrated the presence of UNAG-3P in this sample ( Table S1 ) . The enolpyruvyl activity of E . coli MurA was monitored by coupling phosphate release to cleavage of fluorescent 7-methylguanosine ( Sigma ) by bacterial purine nucleoside phosphorylase ( Sigma ) [53] . Each reaction mixture of 400 µl contained 0 . 3 U purine nucleoside phosphorylase , 50 µM 7-methylguanosine , and varying concentrations of UNAG and UNAG-3P in buffer R2 ( 50 mM NaCl , 50 mM HEPES-NaOH [pH 7 . 5] , 1 mM phosphoenolpyruvate ) . UNAG turnover was started by addition of MurA to 0 . 5 µM . The reactions were monitored by the decrease in fluorescence at 400 nm with excitation set to 300 nm in a FluoroMax-3 spectrofluorometer ( Horiba Jobin Yvon ) with excitation and emission bandwidths of 5 nm . In order to determine the Ki of UNAG-3P for MurA , initial velocities were plotted against the UNAG concentration and fitted globally assuming competitive inhibition of MurA by UNAG-3P using the equation ( 1 ) where Vmax is the maximal rate in fluorescence decrease at the given MurA concentration , Km the Michaelis-Menten constant of UNAG for E . coli MurA , and Ki the inhibition constant of UNAG-3P . While the Ki for UNAG-3P converged to 7 µM , the best fit parameter for Km was 15 µM , which is the same as reported previously [54] . | Most bacteria harbor toxin–antitoxin ( TA ) systems , in which a bacterial toxin is rendered inactive under resting conditions by its antitoxin counterpart . Under conditions of stress , however , the antitoxin is degraded , freeing the toxin to attack its host bacterium . One such TA system , PezAT , has been difficult to study in the past because the PezT toxin is so toxic without its antitoxin counterpart that bacteria die before any useful measurements can be made . Here , we use a truncated version of PezT that kills bacteria more slowly than normal , allowing us to examine the mechanisms of how this TA system operates . We find that zeta toxins convert an essential building block of bacterial cell walls ( known as UNAG ) into a form that prevents normal cell wall growth , causing distortions in bacterial shape that leave the bacteria vulnerable to the hydrostatic pressure of its contents . Consequently , the bacteria burst , similar to what happens when they are treated with penicillin . These results may serve useful for designing new antibiotics . Additionally , our results support the hypothesis that activation of PezT during bacterial infections may be a method by which rapidly growing bacteria can instigate a suicide program , which would promote the release of virulence factors that facilitate spread of infections . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biomacromolecule-ligand",
"interactions",
"microbial",
"metabolism",
"streptococci",
"small",
"molecules",
"enzymes",
"microbiology",
"proteoglycans",
"bacterial",
"pathogens",
"enzyme",
"classes",
"transferases",
"biology",
"drug",
"discovery",
"biochemistry",
"glycobiology"
... | 2011 | A Novel Mechanism of Programmed Cell Death in Bacteria by Toxin–Antitoxin Systems Corrupts Peptidoglycan Synthesis |
In yeast , the G1 cyclin Cln3 promotes cell cycle entry by activating the transcription factor SBF . In mammals , there is a parallel system for cell cycle entry in which cyclin dependent kinase ( CDK ) activates transcription factor E2F/Dp . Here we show that Cln3 regulates SBF by at least two different pathways , one involving the repressive protein Whi5 , and the second involving Stb1 . The Rpd3 histone deacetylase complex is also involved . Cln3 binds to SBF at the CLN2 promoter , and removes previously bound Whi5 and histone deacetylase . Adding extra copies of the SBF binding site to the cell delays Start , possibly by titrating Cln3 . Since Rpd3 is the yeast ortholog of mammalian HDAC1 , there is now a virtually complete analogy between the proteins regulating cell cycle entry in yeast ( SBF , Cln3 , Whi5 and Stb1 , Rpd3 ) and mammals ( E2F , Cyclin D , Rb , HDAC1 ) . The cell may titrate Cln3 molecules against the number of SBF binding sites , and this could be the underlying basis of the size-control mechanism for Start .
The budding yeast Saccharomyces cerevisiae commits to cell-cycle entry at a point called “Start , ” equivalent to the restriction point in animals . Start depends on cell growth to critical size [1]–[5] . At the molecular level , Start coincides with and depends on a Start-specific burst of transcription of over 100 genes including the G1 cyclins CLN1 and CLN2 , the S-phase cyclins CLB5 and CLB6 , and many genes for budding and DNA synthesis [6] , [7] . The burst of transcription at Start depends on two closely related transcription factors SBF and MBF , each of which contains the transcriptional modulator Swi6 , and a sequence-specific DNA binding protein , Swi4 in SBF , and Mbp1 in MBF [8] . SBF is most important for the transcription of CLN1 and CLN2 , and these G1 cyclins are most important for propelling the cell cycle forward . SBF is found bound to the CLN1 and CLN2 promoters in early G1 , well before Start , but at this time does not induce any transcription [9]–[12] . Indeed , in early G1 , SBF may repress transcription . When cells have grown to critical size , the SBF is somehow converted to a transcriptional activator , and induces transcription of many genes including CLN1 and CLN2 . The G1 cyclin Cln3 , in combination with the cyclin dependent kinase Cdc28 ( or Cdk1 ) , is a key regulator of Start [13] , [14] , and is critical for the size-dependent activation of SBF and MBF [7] , converting them from their early G1 repressive forms into the late G1 activating forms . Consistent with the idea that CLN3 is a critical activator of Start , hyperactive alleles of CLN3 ( e . g . , WHI1-1 ) , and over-expression of CLN3 , accelerate Start to smaller cell sizes , whereas deletion of CLN3 delays Start to much larger cell sizes [13] , [14] . A cln3 null mutant , despite having a delayed Start and large cells , is viable because there are alternative methods of inducing transcription of CLN1 and CLN2 . The most important alternative route depends on the mysterious gene BCK2 . The mechanism by which the Bck2 protein activates CLN1 and CLN2 transcription is still largely unknown [15] , [16] . A cln3 bck2 mutant is inviable in most genetic backgrounds precisely because it does not express sufficient amounts of CLN1 or CLN2 , and inviability can be suppressed by the expression of CLN2 from a heterologous promoter [17] , [18] . An obvious model for the CLN3-dependent activation of SBF is that the Cln3-Cdc28 kinase complex might phosphorylate SBF , thus activating it . However , no evidence for this model has been found [19] . Instead , there has been an accumulation of evidence that Cln3 works , at least in part , by inhibiting a repressor of SBF . Costanzo et al . and de Bruin et al . have identified Whi5 as one such repressor [20] , [21] . The Whi5 protein associates with SBF on the CLN2 promoter to repress transcription , and Cln3-Cdc28 phosphorylates and antagonizes Whi5 [20] , [21] . Furthermore , deletion of WHI5 , like over-expression of CLN3 , accelerates Start to smaller cell sizes [22] , and the whi5 null mutant suppresses the inviability of the cln3 bck2 double mutant [20] , [21] . Although Whi5 is clearly an important target of Cln3 , and an important regulator of SBF , it may not be the only target . Costanzo et al . found some evidence that whi5 null mutants were still responsive to CLN3 , suggesting that CLN3 was also acting by at least one alternative pathway . An enduring mystery has been the link between cell size and the activation of SBF . Cln3-Cdc28 activates SBF only when cells have grown to a critical size . But Cln3-Cdc28 is present even in very small cells . At least in slowly growing G1 cells , Cln3 abundance increases through G1 as the cell grows more-or-less in proportion to cell size and total cell protein . That is , its absolute abundance increases , but its relative abundance ( relative to cell volume , or relative to protein content ) does not , or at least not by very much [7] , [23] . How does a small increase in abundance trigger Start at a critical size ? One possibility is that Cln3 is titrated against something that is constant per cell . Here , we suggest that increasing amounts of Cln3 are titrated directly against the SBF bindings sites in genomic DNA , which are of course constant in number through G1 phase . At a sufficiently high Cln3/SBF site ratio , SBF is activated , and Start ensues . Finally , it is remarkable how well eukaryotic cell cycle control mechanisms have been conserved , with the functional replacement of fission yeast cdc2 by human cdc2 ( CDK1 ) [24] an early and striking example . The yeast system for promoting Start is analogous and perhaps homologous to the mammalian system , with SBF , Cln3 , and Whi5 playing roles similar to those of E2F-Dp , Cyclin D , and Rb , respectively [25] . Our results suggest that the analogy goes even deeper , with both the yeast and mammalian system making critical use of the Rpd3 histone deacetylase to repress transcription of S-phase genes .
The only known role of Cln3 is to activate SBF and MBF; evidence for this is that swi6 mutants ( which lack both SBF and MBF ) are completely nonresponsive to CLN3 [19] . That is , the cell size of swi6 mutants is unaffected by over-expression or under-expression of CLN3 . If Whi5 is the one and only target of Cln3 , then the size of whi5 mutant cells , like that of swi6 mutant cells , should also be nonresponsive to CLN3 . Whether or not this is true is unclear , although Costanzo et al . [20] found some evidence that whi5 mutants did respond to CLN3 . To address this issue in a more sensitive way , we used strains containing a bck2 mutation . Since Bck2 is a redundant with Cln3 for expression of CLN1 , CLN2 , and other genes [17] , [18] , bck2 mutants are even more sensitive than wild-type cells to the effects of CLN3 . Thus we compared bck2 WHI5 cells with bck2 whi5 cells with respect to the effect of CLN3 on cell size . Results ( Figure 1 , top two panels ) show clearly that both genotypes are still responsive to CLN3 . Thus , Cln3 can affect cell size , and presumably SBF/MBF activation , even in a whi5 null strain , suggesting it has some target in addition to Whi5 . As described above , a cln3 bck2 double mutant is inviable in many strains , because SBF cannot be activated , and so CLN1 , CLN2 , CLB5 , and CLB6 cannot be expressed at sufficient levels . However , a whi5 mutation relieves some of the repression of SBF , and so a cln3 bck2 whi5 mutant is viable . Mutations in other putative repressors of SBF might also suppress the inviability of a cln3 bck2 strain . Thus we constructed a cln3 bck2 strain kept alive by plasmid-borne MET-CLN2 ( a construct where CLN2 expression is repressed by methionine ) . This strain is viable in the absence of methionine , but dies with a G1 arrest in the presence of methionine . The strain was mutagenized using a transposon library ( so that mutant genes could be identified ) , and spread on +met plates to select suppressors . This screen yielded two classes of mutants irrelevant to our studies . First , there were a variety of mutants ( many cis-acting ) that derepressed the MET promoter , and thus mis-expressed the plasmid-borne MET-CLN2 . Second , there were mutants that by one means or another increased the expression of the RME1 gene , which encodes a transcription factor that , among other things , binds directly to the CLN2 promoter and increases CLN2 expression [26] . We identified these irrelevant mutants using secondary screens; they were not further analyzed . The screen also yielded four complementation groups that may be of direct relevance: chd1 , hda2 , pho23 , and stb1 . The chd1 and stb1 mutations were obtained many times each , whereas hda2 and pho23 were obtained only once each . No whi5 mutation was obtained , but subsequent examination of the mutagenic transposon library by PCR showed that this library did not contain even one disrupted copy of WHI5 . CHD1 is CHromoDomain 1 , a nucleosome remodeling factor containing a chromodomain , ( which can mediate binding to histones bearing methylated lysines ) , a helicase domain , and a DNA binding domain [27] . It is a component of both the SAGA and SILK complexes [27] . It is a likely mediator of SBF activity , but its relevance will be considered in a separate report . Interestingly , Rb-binding protein 1 ( RBP1 ) , a mediator of E2F repression in mammalian cells , also contains a chromodomain . HDA2 is Histone DeAcetylase 2 , a member of the Hda1 histone deacetylase complex [27] . Its function is poorly understood . Although we do not further consider Hda2 here , it could well be a repressor at the CLN2 promoter . PHO23 encodes a component of the Rpd3 histone deacetylase complex [28] . The Rpd3 histone deacetylase is a major histone deacetylase activity in yeast [29] , [30] , and moreover is the yeast ortholog of mammalian HDAC1 , the histone deacetylase that interacts with E2F and Rb . Finally , STB1 ( Sin Three Binder 1 ) was originally isolated as an interactor with Sin3 [31] , and Sin3 is a targeting subunit for the Rpd3 histone deacetylase [32] . Stb1 has also been isolated as a protein binding to the Swi6 component of SBF and MBF , and modulating transcription [33] , [34] . Thus Stb1 could be a link between SBF and the Rpd3 histone deacetylase complex . The involvement of both Stb1 and Pho23 implicated the Rpd3 histone deacetylase complex at the CLN2 promoter . Furthermore the Rpd3 complex has previously been implicated in the repression of various cell cycle genes , especially SBF or MBF dependent genes [35] , [36] . Therefore we asked whether mutations in RPD3 ( encoding the catalytic subunit ) or SIN3 ( encoding the targeting subunit ) could , like mutations in STB1 , PHO23 , or WHI5 , suppress the inviability of the cln3 bck2 double mutant . Indeed , both rpd3 and sin3 did suppress the inviability of the cln3 bck2 mutant ( Figure 2 ) . Consistent with this , D . Huang , S . Kaluarchchi , and B . Andrews ( personal communication ) have also found that rpd3 can suppress the cln3 bck2 double mutant . As judged by growth rate of the various mutants ( Figure 2 ) , whi5 is the strongest suppressor . Because RME1 slightly activates CLN2 transcription directly , and because one class of suppressor over-expressed RME1 , we wondered about the relationship , if any , between pho23 , sin3 , rpd3 , etc . , and RME1 . Therefore we analyzed the suppressors ( chd1 , pho23 , stb1 , sin3 , rpd3 , and whi5 ) in cln3 bck2 strains that were either RME1 or rme1 . In the RME1 background , all the suppressors could suppress inviability , and could lose the MET-CLN2 plasmid . In the rme1 background , the suppressors could again suppress inviability of the cln3 bck2 MET-CLN2 strain on +met plates; however , strains of these genotypes could not lose the MET-CLN2 plasmid ( with the whi5 strain being an exception , and able to lose the plasmid ) . This result suggested that the slight , residual expression from the repressed MET-CLN2 construct was important for viability . The inability of the suppressed rme1 strains to lose the MET-CLN2 plasmid meant there were two possible explanations for the suppression . First , it could be that some or all of the suppressors de-repressed the native , genomic CLN2 locus , allowing viability , but that the degree of de-repressed CLN2 expression was modest , and viability also required a trace of additional expression , which could come either from RME1 ( driving genomic CLN2 ) , or from repressed MET-CLN2 ( expressing low , residual levels of CLN2 ) . Second , it could be that some or all of the suppressors were activating RME1 ( thereby inducing native CLN2 ) and also de-repressing MET-CLN2 . Two lines of experimentation showed that the first possibility is correct . First , the transcript from the native CLN2 locus differs in length from the MET-CLN2 transcript . We used quantitative ( q ) PCR to show that the suppressors increase transcription of the native CLN2 locus , but have no effect on transcription of MET-CLN2 ( Figure 3 ) . The two strongest suppressors , stb1 and whi5 , activated CLN2 transcription to similar extents ( Figure 3 ) . Expression of CLN1 was also increased . Second , we integrated a second copy of CLN2 at the CLN2 locus , using a large restriction fragment that included sequences up to and including the flanking genes . The tested suppressors ( sin3 , stb1 , and whi5 ) were able to suppress inviability of the resulting cln3 bck2 rme1 2×CLN2 strain , and these strains were able to lose the MET-CLN2 plasmid ( unpublished data ) . Thus , in a 2×CLN2 strain , neither RME1 nor MET-CLN2 is required for suppression; the suppressors must act by de-repressing the native CLN2 locus . These results establish that Stb1 , Sin3 , and Rpd3 , like Whi5 , play a role in the repression of SBF target genes . However , they do not establish whether Stb1 , Sin3 , and Rpd3 are additional components of the Whi5 pathway ( i . e . , Whi5 might act by attracting the Rpd3 complex ) , or whether some or all of these new repressors constitute the second pathway that allows whi5 mutant cells to respond to CLN3 . To address this , we did epistasis analysis . We constructed double mutants with whi5 ( i . e . , stb1 whi5 , sin3 whi5 , rpd3 whi5 ) , and asked whether any of these double mutants would reduce or eliminate responsiveness to CLN3 ( which would indicate that the new repressors are in the second new pathway ) . Unlike either of the single mutants , a whi5 stb1 double mutant is almost nonresponsive to CLN3 ( Figure 1 ) . Thus STB1 likely defines a second pathway by which CLN3 controls activity of SBF . Epistasis analysis of rpd3 and its targeting subunit sin3 with whi5 and with stb1 gave complex results . The whi5 sin3 and the whi5 rpd3 mutants are still responsive to CLN3 with respect to size ( Figure 1 , fifth and sixth panels ) , although the whi5 sin3 mutant does not show any responsiveness with respect to budding . This suggests that sin3 ( in particular ) and rpd3 may be partially but not fully blocking the Stb1 pathway . But stb1 sin3 double mutants are responsive to CLN3 ( unpublished data ) , suggesting that the sin3 mutation is not fully blocking the Whi5 pathway . Previous experiments have established links between Sin3 , Rpd3 , and Stb1 [29] , [31] . We feel there are several alternative interpretations of these data ( see Discussion ) , the most likely being that Whi5 , Stb1 , and Swi6 all interact to some extent with the Rpd3 histone deacetylase complex . Consistent with this , Huang , Kaluarchchi and Andrews have recently found an association between Whi5 and Rpd3 by co-immunoprecipitation ( personal communication ) . To further characterize the mechanisms by which Cln3 promotes transcription of SBF target genes , we used chromatin immunoprecipitation ( ChIP ) to build on the earlier work of Cosma , Nasmyth , and coworkers [9] , [10] and observe events at the CLN2 promoter ( an important SBF target gene ) as a function of Cln3 abundance . This is a challenging goal , since once cells have passed through Start , they repress transcription of SBF target genes by additional mechanisms [37] . Thus activation of SBF target genes under normal conditions is transient and difficult to characterize . Therefore , we constructed a strain with genotype GAL-CLN3 bck2 cdc34-2 ( i . e . , CLN3 is expressed from the GAL promoter ) . This strain can be synchronized in G1 before Start by growing in raffinose ( i . e . , without galactose , CLN3 expression is off , its target G1 cyclins CLN1 and CLN2 are not expressed , and Start does not occur ) . When these G1 cells are then switched to raffinose plus galactose medium at 37°C ( the restrictive temperature for the cdc34-2 mutation ) , CLN3 is turned on , SBF targets are transcribed , but progress through the cell cycle ( and the consequent repression of SBF targets ) does not occur because of the cdc34-2 defect . Thus we can follow a cell population from a state where SBF genes are fully repressed in all cells ( in raffinose medium ) to a state where SBF genes are fully induced in all cells ( in galactose medium at 37°C ) . Figure 4 shows the fate of some relevant proteins at the CLN2 promoter as a function of CLN3 expression . As expected from previous work , Swi4 is at the SBF binding sites of the CLN2 promoter at all times [10]–[12] , regardless of the presence or absence of CLN3 . RNA polymerase II is initially absent , but is recruited to the TATA box of CLN2 ( and to the TATA box of BBP1 , the divergently transcribed , SBF-controlled gene at the other end of the intergenic region ) 5 to 10 min after induction of CLN3 . Northern analysis shows that the production of CLN2 mRNA almost exactly coincides with recruitment of RNA pol II ( Figure 4 ) . Previous studies have shown that this recruitment of RNA pol II depends on Cdc28 kinase [9] . Stb1 is also found near the SBF binding sites , and its presence is not affected by induction of Cln3 . Finally , Whi5 , Sin3 , and Rpd3 are all initially present near the SBF binding sites on the CLN2 promoter , and each of these proteins is lost after CLN3 is induced . Consistent with the relative early and CLN3-dependent loss of Rpd3 , Huang , Kaluarchchi and Andrews have recently found that CLN3 can reduce the amount of the Whi5-Rpd3 complex seen by co-immunoprecipitation ( personal communication ) . Several of these results were also repeated , with the same results , on the YOX1 promoter , which is also regulated by CLN3 and is coregulated with CLN2 ( Figure 4C ) . Surprisingly , the loss of the repressive proteins is not obvious until 5 to 15 min ( for Sin3 and Rpd3 ) , or 15 to 25 min ( for Whi5 ) after CLN3 induction; that is , recruitment of RNA pol II , and the appearance of CLN2 transcript , occur before all the repressive proteins are lost . There are at least three nonexclusive explanations for these kinetics: first , it could be that the repressive proteins are quickly phosphorylated and thereby inactivated as repressors by the Cln3-Cdc28 kinase complex; loss of the proteins from the promoter could be a secondary event . Second , it could be that Cln3-Cdc28 is promoting some positive event that directly induces transcription , and this precedes full loss of the repressive activities . Perhaps phosphorylation of Stb1 or Swi4 or Swi6 , for instance , could directly promote transcription even in the presence of repressors . Third , there could be some systematic bias in our ChIP assay such that it is easier to see new proteins arriving at CLN2 than to see old proteins leaving . Additional experiments will be required to distinguish these possibilities . The GAL-CLN3 bck2 cdc34-2 strain used in the experiments above allowed us to look at events at the CLN2 promoter in a way that is powerful and sensitive , but also contrived . Therefore , we repeated some of the experiments in a different genetic background . We used a strain carrying a cdc20 mutation and a galactose inducible CDC20 gene ( GAL-CDC20 ) to arrest cells at the cdc20 block ( mitosis; pre-anaphase ) , then release them synchronously . Although this approach is less sensitive than the GAL-CLN3 bck2 cdc34-2 method , we were able to reproduce several of the main results . For example , Figure 4E shows Sin3 being recruited to the CLN2 promoter early in G1 , then leaving as cells exit G1 ( in this experiment , budding begins at about 50 min ) . We began to characterize the binding dependencies of some of the proteins at the CLN2 promoter ( Figure 5; Table 1 ) ; one obvious question is whether binding of the Sin3-Rpd3 complex depends on Whi5 or Stb1 . Sin3 still binds to the CLN2 promoter in a whi5 mutant and also in an stb1 mutant , and also in a whi5 stb1 double mutant . To see if Sin3 binding was SBF dependent , we used both swi4 and swi6 mutations , and found that the association of Sin3 with the CLN2 promoter is dependent on SWI6 ( p = 4×10−3 ) , but only slightly if at all dependent on SWI4 ( Figure 5; Table 1 ) . Presumably in the absence of Swi4 , MBF ( Mbp1+Swi6 ) is binding to the CLN2 promoter and recruiting Sin3-Rpd3 . The dependence of Rpd3 binding on Swi6 correlates with previous findings that Swi6 contains Cln3-modulated repressive domains [19]; these could be the regions responsible ( directly or indirectly ) for recruiting the Rpd3 complex . Stb1 and perhaps Whi5 could further influence the recruitment or activity of the Sin3-Rpd3 complex . Robert et al . [35] previously found that Rpd3 associates with the promoters of CLB6 and PCL1 , which are regulated by SBF and by MBF . In contrast to our finding of Swi6 but not Swi4 dependence at the CLN2 promoter , Robert et al . found that the association of Rpd3 with CLB6 and PCL1 required both Swi6 and Swi4 . The reason for the difference between the studies with respect to the requirement for Swi4 is unclear , but in any case both studies agree that SBF is involved in the recruitment of the Rpd3 complex . While it is clear that that Cln3-Cdc28 protein kinase complex somehow promotes the loss of Sin3 , Rpd3 , and Whi5 from the CLN2 promoter , it is not clear how directly Cln3 acts , or exactly what proteins the Cln3-Cdc28 complex phosphorylates . We found that Cln3 co-immunoprecipitates with the Swi6 component of SBF , and that this co-immunoprecipitation depends on Swi4 ( Figure 6 ) . Thus , there is a relatively direct interaction between Cln3 and SBF . ChIP showed that Cln3 is found on the CLN2 promoter close to the SBF binding sites ( Figure 6 ) , the same location as SBF , Whi5 , Sin3 , Stb1 , and Rpd3 . Cln3-Cdc28 is thus in a location suitable for the direct phosphorylation of these and other associated proteins . Both Whi5 and Stb1 have a very high density of consensus phosphorylation sites for the Cdc28 kinase ( 2 . 7 or 2 . 6 consensus and near-consensus sites per 100 amino acids , respectively ) , and previous work suggests that they are very likely substrates for Cln-Cdc28 kinase [20] , [21] , [34] . Swi6 and Swi4 also have multiple potential Cdc28 phosphorylation sites , and could be Cln3-Cdc28 substrates in vivo . The fact that Cln3 is at the CLN2 promoter raises another issue . Yeast has 100 to 200 genes under the control of SBF and the related transcription factor MBF , and these genes typically have two , three , or more SBF/MBF binding sites each . Thus the total number of functional SBF and MBF binding sites in the cell is in the vicinity of 400 . But the average number of Cln3 molecules in a haploid cell is only about 100 [38] . Of course there is considerable uncertainty in these measurements , but nevertheless it is likely that cellular SBF/MBF binding sites are in excess over Cln3 . This excess of binding sites could provide a basis for the critical size requirement for Start . As cells grow in mass , ribosome content , and protein synthetic capacity , they contain increasing numbers of Cln3 molecules [7] . Indeed , growth in the number of Cln3 molecules may be faster than the growth in mass [39] . Yet the number of SBF binding sites is fixed by DNA content . Thus , as the cell grows , it could titrate an increasing number of Cln3 molecules against a fixed number of SBF binding sites , which are initially in excess . At some ratio , the bound Cln3 could activate SBF , resulting in Start . If this model were correct , then an increase in the number of SBF sites in the cell would increase the requirement for Cln3 , and so would cause an increase in cell size at Start . We transformed otherwise wild-type cells with a high copy number plasmid containing four tandem , perfect SBF binding sites ( an SBF binding site is called an “SCB” ) . Since the plasmid has a copy number of about 30 , this provides about 120 extra sites , or roughly a 20% increase over the wild-type number of sites . We used elutriation to collect small G1 phase cells carrying the 4×SCB plasmid ( or a control plasmid lacking the 4×SCB insert ) , then let these cells grow . We assayed cell volume , and the percentage of budded cells as an assay of Start . As shown in Figure 7 , cells lacking the 4×SCB insert went through Start at about 32 fl , while cells containing the 4×SCB insert went through Start at about 38 fl , roughly a 20% increase . This experiment is consistent with the idea that the cell is setting the critical size for Start by titrating some molecule against the number of available SBF binding sites . This experiment was done using several independent pairs of transformants a total of five times ( i . e . , five pair-wise comparisons ) . In every case , the strain with the 4×SCB plasmid had a larger critical size than the strain with the control plasmid . The differences in critical size in the five experiments were 2 . 6 , 3 . 7 , 6 . 1 , 9 . 2 , and 10 . 6 fl , with a mean of 6 . 4 fl ( p<0 . 005 for a test of the hypothesis that the difference is 0 using a paired sample one-tailed Student's t-test; also statistically significant by nonparametric tests ) . The experiment shown in Figure 7 is the median experiment , with a 6 . 1 fl difference . One issue with this titration experiment is that the 4×SCB plasmid might increase critical size through some irrelevant pathology . If this were so , then the 4×SCB plasmid would cause roughly the same percentage increase in size regardless of any changes we might make to the CLN3/WHI5/SBF system . A second issue is that even if the activator titration model is correct , it is not clear what activator is being titrated; it might be Cln3 , but Swi4 , Swi6 , and even Stb1 are also possibilities . To address these issues , we first repeated the experiment in a strain carrying two copies of CLN3 ( 2×CLN3; a second copy is tandemly integrated at the wild-type CLN3 locus ) . If the titration hypothesis is correct , then the second copy of CLN3 should largely compensate for the ∼20% increase in SCB sites . Indeed , the increased size caused by the 4×SCB plasmid in a 2×CLN3 background is only 0 . 6 fl ( Figure 7 ) . A 2×CLN3 4×SCB strain had almost exactly the same size as a wild-type ( i . e . , 1×CLN3 ) strain bearing the control plasmid . ( We note that a second copy of CLN3 causes only ∼10% decrease in critical size in a wild-type strain [38] . Presumably when Cln3 is sufficiently abundant , some other molecule becomes limiting for Start . ) In addition , we did the titration experiment in a cln3 deletion strain . If the effect of the 4×SCB plasmid is irrelevant pathology , then in the cln3 strain , the same irrelevant pathology should occur , and the 4×SCB plasmid should again increase critical size . On the other hand , if Cln3 is the activator being titrated , then in the cln3 strain , the 4×SCB plasmid should have no effect on critical size , since it has nothing to titrate . In fact , to our great surprise , we got neither of these results . Instead , cln3 cells actually got smaller when we added the 4×SCB plasmid ( Figure 7 ) . This surprising result was confirmed with two additional experiments ( unpublished data ) , using independently constructed strains . ( The experiment shown has the median difference of the three experiments . ) This result tells us two things: first , the results are not irrelevant pathology , because the results change in a specific way with changes in the CLN3/WHI5/SBF system . Second , a likely interpretation is that Cln3 is indeed the activator being titrated ( otherwise a deletion of CLN3 would make no difference ) , but that the 4×SCB plasmid is also titrating repressors . That is , Cln3 is most limiting ( so high copy 4×SCB causes bigger cells in a wild-type background ) , and repressors are next most limiting ( so when there is no Cln3 anyway , then the high copy 4×SCB plasmid decreases size by titrating repressors ) . If this is true , then a strain that lacks both Cln3 and also the repressors ( Whi5 and Stb1 ) should not be affected by high copy 4×SCB . And this proves to be the case ( Figure 7 , bottom right ) ; a whi5 stb1 cln3 strain is not affected by the high copy 4×SCB plasmid . Note that the whi5 stb1 mutant is not responsive to CLN3 ( Figure 1 ) ; the fact that it is also not responsive to extra SCBs is the expectation from the titration model . The whi5 stb1 cln3 strains shown in Figure 7 lack all known regulation of the Cln3 size control pathway . Yet , these strains have a size at budding similar to that of wild-type , and have a sigmoidal budding curve suggesting a dependence of budding on size . That is , although the best-characterized size control mechanism is missing , the cells apparently exhibit some form of size control . This suggests the existence of a redundant size control mechanism . The same phenomenon has been observed previously in different circumstances [39] , where the redundant size control was attributed to a translational mechanism . In addition , Jorgensen et al . [22] found many size control mutants that were not in the CLN3 pathway .
Here we have found that the Whi5 pathway is not the sole link between Cln3-Cdc28 and SBF activity . We have found several mutants that , like whi5 , relieve the repression of SBF , and render its activity somewhat independent of Cln3-Cdc28 . These mutants include chd1 , hda2 , pho23 , sin3 , rpd3 , and stb1 . Of these , pho23 , sin3 , stb1 , and rpd3 , are members of , or have been physically linked to , the Rpd3 histone deacetylase complex , a repressive histone deacetylase orthologous to mammalian HDAC1 . Although we do not know the exact relationship between these proteins and Whi5 , we have found that the stb1 mutation is synergistic with whi5; that is , in the context of a bck2 mutation , the stb1 whi5 double mutant , unlike either single mutant , has little ability to respond to Cln3-Cdc28 . Thus in some sense Stb1 identifies a pathway for regulating SBF that is separate from the Whi5 pathway . While we have identified STB1 in a screen for repressors of SBF , others have previously identified STB1 as an activator of SBF or MBF [33] , [34] . While paradoxical at first sight , it is quite common for transcription factors to have both positive and negative roles in transcription . An example is Fkh2 , which collaborates with Mcm1 and Ndd1 transcription factors , and with Clb-Cdc28 kinase activity , to regulate mitotic genes . In this context , Fkh2 appears to be an activator in late G2 and mitosis , but a repressor at other times [40]–[44] . Similarly , we imagine that Stb1 helps repress SBF in the absence of Cln3 and Bck2 ( the situation in which we found it as a repressor ) , but helps activate SBF in the presence of Cln3 or Bck2 ( the situation in which it was characterized as an activator ) . Consistent with this , cell cycle expression analysis of stb1 mutants shows that target genes are less repressed at troughs , and less induced at peaks; i . e . , they are less regulated and more constitutive ( e . g . , Figure 3 in [34] ) . The fact that CLN3 can induce expression of CLN2 even before Sin3 , Rpd3 , and Whi5 are lost from the promoter ( Figure 4 ) is consistent with the idea that the initial expression of CLN2 depends on activation , perhaps via Stb1 , rather than on loss of repression . If Stb1 is both a repressor and an activator , then some of our assays may preferentially see one of these activities , and some may see the other . Presumably it is the lack of repression by Stb1 that allows the stb1 mutation to suppress the lethality of the cln3 bck2 mutant . But the cell size assay for responsiveness to CLN3 ( Figure 1 ) may be more sensitive to Stb1 as an activator; in particular , the synergistic defect between whi5 and stb1 may be due to the lack of repression in the whi5 mutant , plus the lack of activation in the stb1 mutant . We note that the combinations of mutations that include stb1 tend to have relatively large cell sizes after induction of GAL-CLN3 ( Figure 1 ) , perhaps showing that STB1 is needed for full induction of CLN2 . While Whi5 and Stb1 seem to define two pathways of regulation of SBF , it is still unclear how the Sin3-Rpd3 histone deacteylase complex is recruited to the CLN2 promoter . Previously , the Rpd3 complex has been linked to Stb1 [29] , [31] . More recently , Huang , Kaluarchchi and Andrews have found evidence for an association between Rpd3 and Whi5 ( personal communication ) . Despite these associations , we found that even whi5 stb1 double mutants had at least some Sin3 ( and so presumably Rpd3 ) at the CLN2 promoter , whereas swi6 mutants had little or no Sin3 . Thus although one could imagine various relationships between these proteins , one model is that SBF has some ability to recruit each of Whi5 , Stb1 , and Sin3-Rpd3 , but that these proteins in addition interact with each other ( Figure 8 ) . Later , in a size- and growth-dependent fashion , Cln3-Cdc28 also joins the complex , and phosphorylates Whi5 and Stb1 and probably Swi6 and possibly Swi4 . This causes the loss of the Rpd3 complex; a somewhat slower loss of Whi5 ( Figure 4 ) ; and perhaps allows phosphorylated Stb1 to help activate transcription ( Figure 8 ) . That Swi6 , along with Whi5 and Stb1 , is probably a target of Cln3-Cdc28 phosphorylation is strongly suggested by the fact that over-expression of a mutant Whi5 lacking CDK phosphorylation sites is lethal in a mutant where Swi6 is likewise lacking CDK phosphorylation sites [20] , [45] . The involvement of Swi6 as a likely target of Cln3-Cdc28 , and as a recruiter of Sin3-Rpd3 , may explain why even whi5 stb1 double mutants seem to have some slight residual Cln3-responsiveness ( Figure 1 ) ; that is , this residual responsiveness could be through direct phosphorylation of Swi6 . Results reminiscent of ours with regard to Sin3 and Rpd3 were previously obtained by Veis et al . [36] , who found that Sin3 and Rpd3 associate with the promoter of the CLB2 gene , which encodes a mitotic cyclin . Although CLB2 is most highly expressed in G2/M , the association of Sin3 and Rpd3 with the CLB2 promoter was lost in late G1 , at about the same time we see loss of Sin3 and Rpd3 from the CLN2 promoter . Veis et al . interpreted their results in terms of the association between Sin3/Rpd3 and the Fkh2 ( forkhead ) transcription factor , and suggested that this association was sensitive to Start . However , we note that CLB2 , despite being most strongly up-regulated in G2/M , is a client of SBF as well as a client of Fkh2 . The CLB2 promoter contains at least three clustered SBF/MBF binding sites , at least two of which are conserved in other species of yeast [46] . In ChIP experiments , CLB2 is a target of SBF or MBF binding [47] , [48] . Thus the loss of Sin3/Rpd3 from the CLB2 promoter in late G1 as seen by Veis and coworkers could involve SBF at the CLB2 promoter , and so could be related to the phenomenon we see at the CLN2 promoter . Another protein we find at the CLN2 promoter is Cln3 . However , demonstrating this association was difficult , and required a special genetic background and over-expression of Cln3 . Part of the difficulty in ChIPing Cln3 to the CLN2 promoter is presumably because Cln3 is a nonabundant protein , and does not bind DNA directly . But in addition , Cln3 may not be a stoichiometric member of the complex . Instead , it may bind weakly and transiently , phosphorylate its substrate ( s ) , and leave . The two proteins we find to be essential for Cln3 responsiveness , Whi5 and Stb1 , are both very likely substrates of Cln-Cdc28 [20] , [21] , [34] . Cln3 is present at only about 100 molecules of protein per cell , and yet there are in the vicinity of 400 functional binding sites for SBF and the related factor MBF . The fact that Cln3 is sub-stoichiometric with respect to binding sites could provide a partial solution to the size control problem: Perhaps the amount of Cln3 in the cell , which is a function of cell size and growth rate , is titrated against the number of binding sites . And indeed we found that cells containing extra SCBs had to grow to a larger size to accomplish Start , and this effect could be compensated by one extra dose of CLN3 . Extra SCBs did not enlarge a cln3 null mutant , and extra SCBs had no effect whatever on cln3 stb1 whi5 triple mutants . These findings are all supportive of the titration model . Even though larger G1 cells contain more Cln3 molecules than smaller cells , the increase in Cln3 content with size is probably quite moderate , possibly only linearly correlated with cell size . Thus even at cell sizes adequate for Start , Cln3 may still be sub-stoichiometric with respect to binding sites . Thus we imagine that at any and all physiologically reasonable concentrations of Cln3 , there will only be fractional occupancy of SBF sites , especially if Cln3 is a weak and transient binder . But as the amounts of Cln3 rise , and are titrated against a fixed number of SBF sites , that fractional occupancy will rise , until at some occupancy ( i . e . , at some critical cell size ) , CLN2 and other targets are expressed , and the cell passes through Start . The issue is , how to convert a relatively small change in total Cln3 into a large change in fractional occupancy , or , alternatively , how to convert a small change in occupancy into a large effect ? Although we do not know the answers to either of these questions , Ferrell and coworkers have described many mechanisms by which such “super-sensitivity” can occur [49]–[56] . One mechanism would use the fact that SBF target genes have multiple SBF binding sites . Perhaps the binding of Cln3 to SBF is cooperative; or perhaps the Cln3 molecules , once bound , cooperate to do something else , such as phosphorylate a substrate . Cooperativity of any kind between multiple sites will give exponential sensitivity to Cln3 amounts , so this is one possible mechanism . A second mechanism is multisite phosphorylation . That is , perhaps the substrates of Cln3-Cdc28 have to be phosphorylated at multiple sites , and this can only happen when fractional occupancy of SBF sites by Cln3 is relatively high . Since phosphorylation is in a dynamic equilibrium with dephosphorylation , a requirement for multisite phosphorylation ( at , say , five sites ) imposes a super-sensitive threshold on the amount of kinase required [52] , [57] , [58] . Multisite phosphorylation can give extreme sensitivity to the amounts of a protein kinase [52] , [57] , [58] . A third possible mechanism is to consider the relationship between the complexes at the multiple SBF sites . There are three sites at CLN2; if all three have repressive proteins , is enough Cln3 needed to fill all three sites simultaneously , even though occupancy of any one site is always transient ? At any rate , although we do not know how supersensitivity works in this situation , there are lots of ways it could work in theory , as cited above . There are remarkable parallels between the SBF/Cln3/Whi5 , Stb1/Rpd3 regulatory module in yeast , and the E2F-Dp/Cyclin D1/Rb/HDAC1 regulatory module in mammalian cells . To begin with , the cluster of regulated genes is highly conserved: In S . cerevisiae [6] , in the distantly related yeast S . pombe [59] , in mammalian cells [60] , and probably in most or all other eukaryotes , there is a highly conserved cluster of genes needed for DNA replication , and expressed around the G1/S transition . In both yeasts and mammals , the motifs regulating these genes contain a core “CGCG” element . In both yeasts and mammals , the transcription factors recognizing this element ( SBF/MBF in the yeasts , E2F-Dp in mammals ) contain a DNA binding domain with a “winged helix” fold [61]–[63] . There is no apparent sequence homology between the yeast and mammalian DNA binding domains , but the domain is small , the evolutionary distance vast , and there are other examples where structure but not sequence has been preserved across time . In E2F-Dp , the transactivator is repressed by binding of Rb and its family members . There are two mechanisms of repression [64]–[70] . First , the transactivation domain is masked . Second , Rb family members ( but possibly not Rb itself-[69] ) recruit mSin3B and HDAC1 which deacetylate and otherwise modify chromatin so as to be inhospitable towards expression . Here , we likewise show that there are at least two pathways of regulation , one of them involving recruitment of a histone deacetylase . In mammals , the transactivation domain is unmasked when a cyclin-CDK complex such as cyclin D-CDK4 phosphorylates Rb and family members , disrupting binding to E2F-Dp , and allowing Sin3m and HDAC1 to leave the chromatin . Similarly , in yeast , Cln3-CDK phosphorylates Whi5 and probably Stb1 . Whi5 , Sin3 , and Rpd3 all leave the chromatin . Interestingly , expression of the target gene Cln2 precedes the loss of the repressive proteins , consistent with a dominant activation , possibly due to Stb1 . In any case , it is clear that there are deep , well-conserved parallels between the SBF/Cln3/Whi5 , Stb1/Rpd3 regulatory module in yeast , and the E2F-Dp/Cyclin D1/Rb/HDAC1 regulatory module in mammals . It is possible that these modules have regulated the cluster of genes for DNA synthesis since early in eukaryotic evolution .
Strains are shown in Table 2 . For GAL-CLN3 cdc34-2 block and release experiments , cells growing in YEP with 2% raffinose+2% galactose ( YEPRG ) at 25°C were arrested in G1 by washing with YEP+2% raffinose ( YEPR ) and incubating in YEPR for four hours at 25°C . Cells were then shifted to 37°C for 1 h , and then cultures were split in two; one half remained in YEPR and to the other half galactose was added to 2% final concentration . Both cultures were incubated at 37°C and samples were taken every 5 min . For elutriations , cells containing plasmids ( pBA70 and pMT3579 ) were grown in Synthetic Complete ( SC ) medium with 2% filter-sterilized sucrose as the carbon source . Small unbudded G1 cells were isolated by centrifugal elutriation and grown in preconditioned SC+2% sucrose at 30°C . Cell size distributions were obtained on a Z2 Coulter Counter and budding indexes were determined by counting cells . Bud counts were done “blind” on randomized samples . Immunoprecipitations and Northern and Western blots were carried out essentially as described previously [71] , [72] . Cell lysates were obtained by vortexing cell suspensions in lysis buffer ( 0 . 1% NP40 , 250 mM NaCl , 50 mM NaF , 5 mM EDTA , and 50 mM Tris [pH 7 . 5] ) in the presence of glass beads . Cell debris was pelleted by centrifugation for 10 min at 4°C . Protein concentration in the lysate was determined by the DC protein assay ( Bio-Rad ) . Lysates typically contained 20–50 mg/ml of total protein . For immunoprecipitation , 5 mg of cell lysate mixed with 2 µl of 9E10 anti-Myc ascites fluid or 5 µg of anti FLAG antibody ( Sigma ) were rotated for 1–2 h at 4°C . Immune complexes were collected on protein A-Sepharose beads by rocking at 4°C for 1 h . For detection of immunoprecipitated proteins , beads were pelleted by very gentle , brief , low-speed centrifugation , washed four times with lysis buffer , and boiled in protein sample buffer immediately before SDS-PAGE . Early exponential phase cells were collected and formaldehyde was added to 1% final concentration . Cells were fixed at room temperature for 15 min . Cross-linking was quenched by the addition of glycine to 125 mM . Cells were pelleted at 3 , 000 g for 5 min and washed twice with ice-cold TBS ( 150 mM NaCl , 20 mM Tris-HCl [pH 7 . 6] ) . To break cells , cell suspensions in lysis buffer ( 50 mM HEPES-KOH [pH7 . 5] , 140 mM NaCl , 1 mM EDTA , 1% Triton X-100 , 0 . 1% sodium deoxycholate ) were mixed with glass beads and vortexed at 4°C for 45 min . Chromatin was sheared by sonication at power 3 ( W-380 Sonicator , Heat Systems-Ulrasonic , INC ) ten times , 10 s each time , and tubes were kept on ice throughout sonication . Cell debris was removed by maximal speed centrifugation for 15 min at 4°C . Whole-cell extracts were prepared for use in ChIPs . Protein concentration for each sample was determined by DC protein assay ( Bio-Rad ) . Immunoprecipitations were performed with 1 mg of extract . Lysates were rotated with 25 µl IgG Sepharose beads at 4°C overnight . Immune complex beads were washed with lysis buffer , lysis buffer 500 ( 50 mM HEPES-KOH [pH 7 . 5] , 500 mM NaCl , 1 mM EDTA , 1% Triton X-100 , 0 . 1% sodium deoxycholate ) , and LiCl/detergent ( 0 . 5% sodium deoxycholate , 1 mM EDTA , 250 mM LiCl , 0 . 5% NP-40 , 10 mM Tris [pH 8 . 0] ) twice for each buffer and washed once with cold TE . Bead washing was performed at 4°C . DNA was eluted by incubating beads at 65°C with elution buffer ( 10 mM EDTA , 1% SDS , 50 mM Tris . Cl [pH 8 . 0] ) for 10 min , and crosslinks were reversed by incubating samples at 65°C overnight . PCR was carried out for 30 cycles and products were separated using 2 . 4% agarose gels . A SuperScript III Platinum SYBR green one-step q ( real-time ) RT-PCR kit ( Invitrogen ) was used for the detection and quantification of RNA . 5 ng RNA was used for the RT-PCR reaction . Total RNA were purified with RiboPure-Yeast kit ( Ambion ) . Cells were grown overnight in either YEPR or YEPRG so that cell densities were between 1 and 2×107 cells/ml . Cultures were placed on ice , sonicated to separate mothers from daughters , and cell sizes were measured on a Z2 Coulter Counter . Cells were then photographed at 40× and brightness and contrast adjusted in Adobe Photoshop . All data shown are from cells in the S288c genetic background; cells in both the W303 and BF305 backgrounds were also tested , and gave identical results . Cells were grown overnight in SC-Met , sonicated briefly , and 1∶4 serial dilutions were plated onto either SC-Met or SC+2 mM Met plates . Cells on SC-Met plates were grown for 3 d at 27°C before being photographed , whereas cells grown on SC+2 mM Met plates were grown for 5 d at 27°C . | Cells seem to divide only after they have grown “big enough . ” Entry into the cell cycle , at a point called Start in budding yeast , is triggered by activation of the Cln3 cyclin-dependent kinase ( CDK ) , which in turn activates downstream transcription . We find that the Cln3-CDK acts through a histone deacetylase , as well as through the previously discovered repressor Whi5 , to activate the SBF transcription factor and trigger entry into the cell cycle . The system is strikingly similar to the one in mammalian cells , which relies on Cyclin D , CDK , the transcription factor E2F , its repressor Rb , and the histone deacetylase system . There is preliminary evidence that as the yeast cell grows in size , the increasing number of Cln3 molecules is titrated against the fixed number of Cln3-CDK-SBF binding sites in genomic DNA , and that this cell size-dependent titration could be the mechanism that makes cell cycle entry dependent on cell size . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/histone",
"modification",
"molecular",
"biology/transcription",
"initiation",
"and",
"activation",
"genetics",
"and",
"genomics/gene",
"expression",
"cell",
"biology/cell",
"growth",
"and",
"division",
"genetics",
"and",
"genomics/gene",
"function",
"b... | 2009 | Recruitment of Cln3 Cyclin to Promoters Controls Cell Cycle Entry via Histone Deacetylase and Other Targets |
The prevalence of type 2 diabetes in the United States is projected to double or triple by 2050 . We reasoned that the genes that modulate insulin production might be new targets for diabetes therapeutics . Therefore , we developed an siRNA screening system to identify genes important for the activity of the insulin promoter in beta cells . We created a subclone of the MIN6 mouse pancreatic beta cell line that expresses destabilized GFP under the control of a 362 base pair fragment of the human insulin promoter and the mCherry red fluorescent protein under the control of the constitutively active rous sarcoma virus promoter . The ratio of the GFP to mCherry fluorescence of a cell indicates its insulin promoter activity . As G protein coupled receptors ( GPCRs ) have emerged as novel targets for diabetes therapies , we used this cell line to screen an siRNA library targeting all known mouse GPCRs . We identified several known GPCR regulators of insulin secretion as regulators of the insulin promoter . One of the top positive regulators was Gpr27 , an orphan GPCR with no known role in beta cell function . We show that knockdown of Gpr27 reduces endogenous mouse insulin promoter activity and glucose stimulated insulin secretion . Furthermore , we show that Pdx1 is important for Gpr27's effect on the insulin promoter and insulin secretion . Finally , the over-expression of Gpr27 in 293T cells increases inositol phosphate levels , while knockdown of Gpr27 in MIN6 cells reduces inositol phosphate levels , suggesting this orphan GPCR might couple to Gq/11 . In summary , we demonstrate a MIN6-based siRNA screening system that allows rapid identification of novel positive and negative regulators of the insulin promoter . Using this system , we identify Gpr27 as a positive regulator of insulin production .
Nearly 13% of American adults have diabetes and these numbers continue to rise , mostly from an increase in type 2 diabetes [1] , [2] . Although insulin resistance is a cardinal feature of type 2 diabetes , most people with insulin resistance do not develop diabetes because their pancreatic beta cells are able to compensate by increasing insulin production . However , if insulin production cannot match the increased demand imposed by insulin resistance , hyperglycemia and frank diabetes ensues . Over time , beta cell function further declines in most people with type 2 diabetes , resulting in the eventual failure of oral medications and the necessity of insulin therapy [3] . Improving insulin production and beta cell function is therefore a universal goal of diabetes therapeutics . We reasoned that an unbiased search for regulators of insulin production might reveal new diabetes drug targets . Therefore , we constructed a novel screening system to screen for genes important for insulin promoter activity . By screening siRNAs targeting all GPCRs , we identify several GPCRs that regulate insulin promoter activity and specifically characterize Gpr27 as a novel regulator of insulin production .
To allow rapid evaluation of insulin promoter activity , the MIN6 mouse beta cell line was infected with a lentivirus that stably expresses destabilized GFP under the control of the proximal 362 base pairs of the human insulin promoter ( Figure 1A ) [4] . This insulin promoter fragment maintains a substantial proportion of promoter activity and tissue specificity while being compact enough to allow lentiviral delivery [5] . To favor single copy integration , the construct was delivered at a low multiplicity of infection ( MOI ) and a clonal line was selected . To generate an internal control reporter , the GFP positive subline was subsequently infected at a low MOI with a second lentivirus containing mCherry under the control of the constitutive rous sarcoma virus promoter ( RSV ) ( Figure 1A ) . A stable clone expressing both constructs was isolated . In these cells , the ratio of GFP to mCherry fluorescence indicates human insulin promoter activity . When transfected into this reporter line , siRNAs targeting activators of insulin gene transcription would be expected to reduce insulin promoter activity and reduce the GFP/mCherry ratio , while siRNAs targeting negative regulators of the insulin promoter should increase the GFP/mCherry ratio ( Figure 1B ) . Indeed , transfection of an siRNA targeting the insulin gene transcription factor Pdx1 reduced the GFP/mCherry ratio by 80% as compared to a non-targeting siRNAs ( Figure 1C and 1D ) [6] . An RNAi library containing four independent siRNAs targeting the mouse GPCR-ome and selected GPCR related genes was transfected into the reporter cell line . The ratio of the GFP to mCherry fluorescence five days after transfection was calculated for each siRNA and the data were then analyzed using the redundant siRNA analysis ( RSA ) software [7] . To avoid off-target effects , each siRNA was transfected separately and only genes with more than one siRNA hit were selected for further analysis . The top genes judged by RSA were then ranked by unsupervised clustering of each gene's RSA p value and its expression level in primary mouse islets , since only those genes expressed in primary cells are of biological interest ( Figure 2A ) . Two publically available mouse islet mRNA-Seq data sets were used . One of these data sets has been previously published and consists of approximately four million mapped reads from islets isolated from female non-pregnant mice and approximately four million mapped reads from islets isolated from pregnant mice [8] . The second , submitted to the NCBI Short Read Archive by Merck , contains approximately 120 million reads from mouse islets ( see methods ) . Because of these modest read numbers , some low abundance transcripts may be erroneously reported as being not expressed using this analysis [9] . siRNAs to the top six genes ( Ffar2 , Gpr27 , Grk5 , p2ry6 , Gpr109a , Bdkrb2 ) that reduced insulin promoter activity and had detectable expression in primary mouse islets were transfected into the screening cell line for confirmation ( Figure 2B ) . All six genes had at least 2 siRNAs confirm . For the siRNAs that increased GFP/mCherry , we retested the top three genes with high RSA scores and detectable expression in mouse primary islets – Adra2a , Cckar , and Aplnr . Of these three , only the known negative regulator of insulin secretion , Adra2a , confirmed with two independent siRNAs ( Figure 2C ) . Several of the positive regulators of the insulin promoter we identified were already known to stimulate insulin release in beta cells . Of particular interest was the orphan GPCR , Gpr27 , which had no known role in insulin production but was previously found to be enriched in the mouse and human pancreatic islet [10] , [11] . We subsequently tested all four siRNAs targeting Gpr27 in the library set on an independently generated MIN6 reporter line expressing stable GFP under the control of the insulin promoter and mCherry under the control of the RSV promoter . All four siRNAs reduced GFP/mCherry fluorescence ( Figure S1A ) . Furthermore , all 4 siRNAs efficiently reduced expression of the Gpr27 mRNA ( Figure S1B ) . We also confirmed that Gpr27 is enriched in beta cell lines ( beta TC and MIN6 ) compared to an alpha cell line ( alpha TC ) ( Figure S2A ) and is expressed in primary mouse beta cells ( Figure S2B ) . Since the screen was based on a human insulin promoter fragment , we measured the effect of Gpr27 knockdown on the endogenous mouse Ins2 and Ins1 genes . Because mature insulin mRNAs have a half-life of nearly 80 hours , we measured insulin pre-mRNAs as previously described [6] , [12] . MIN6 cells infected with a Gpr27 shRNA expressing adenovirus ( Ad-shGpr27 ) had a 40–60% reduction in pre-ins2 and pre-ins1 levels compared to control adenovirus ( Ad-control ) ( Figure 3A ) . To confirm these findings in primary beta cells , we infected intact primary mouse islets with these same adenoviruses . At a high MOI , we were only able to obtain 50% infection rates as measured by flow cytometry , presumably reflecting poor adenovirus penetration into the core of the mouse islet [13] . Therefore , intact islets were dissociated prior to adenovirus infection . Three days after infection , cells were isolated by flow cytometric sorting for GFP and RT-qPCR was performed . Knockdown of Gpr27 produced a significant ∼30% reduction of pre-ins2 ( p = 0 . 03 ) . Concomitantly , there was a nearly significant 30% reduction in the less abundant pre-ins1 message ( p = 0 . 055 ) ( Figure 3B ) . While insulin production requires insulin promoter activity , minute-to-minute changes in plasma insulin levels are controlled by insulin secretion . Therefore , we asked if Gpr27 knockdown would affect glucose stimulated insulin secretion . Infection of MIN6 cells with Ad-control at an MOI necessary to get >90% infection inhibited glucose stimulated insulin secretion ( data not shown ) . Therefore , we infected MIN6 cells at a lower MOI to achieve approximately 60% infection and measured glucose stimulated insulin secretion from this mixed population by batch incubation . Ad-shGpr27 infected MIN6 cells secreted ∼40% less insulin at 20 mM glucose compared to Ad-control infected cells ( Figure 3C ) . There was no statistically significant difference at 2 mM glucose . Notably , we did not detect a difference in total insulin as normalized to total protein concentration ( Ad-control = 27 . 9+/−1 . 1 mg insulin per g of total protein; Ad-shGpr27 = 29 . 4+/−0 . 94 mg insulin per g of total protein , p value = 0 . 13 ) . This was not unexpected since the half-life of insulin mRNA is ∼80 hours and the knockdown of Gpr27 was limited to 72 hours due to adenovirus toxicity after that time point . We conclude that Gpr27 plays a measurable role in insulin secretion in addition to insulin promoter activity . To define the mechanism of Gpr27 action , we measured transcript levels of selected regulators of the insulin promoter by RT-QPCR in MIN6 cells after Ad-shGpr27 infection . Glis3 , Pax6 , Nkx6 . 1 , HNF4a , and Pdx1 were reduced after Gpr27 knockdown while others including MafA , NeuroD1 , and Pax4 were unchanged ( Figure 4A ) . Concordant with this expression data , Gpr27 knockdown reduced the transcriptional activity of mini-enhancers that bind to Glis3 and Pdx1 ( Z , E1/A1 , E2/A3 ) while Gpr27 knockdown had no effect on mini-enhancers that bind to MafA and NeuroD1 ( C1/E1 ) ( Figure 4B and 4C ) . Since Pdx1 is required for insulin promoter activity and insulin secretion [14] , [15] , we asked if Pdx1 is required for the effect of Gpr27's on the insulin promoter . By luciferase assay , we found that the single knockdown of Pdx1 reduced insulin promoter activity by 90% and Gpr27 knockdown alone reduced insulin promoter activity by 40% . However , the knockdown of both Gpr27 and Pdx1 had no additional effect over the single knockdown of Pdx1 , showing that Pdx1 is important for the effect of Gpr27 on the insulin promoter ( Figure 4D ) . Importantly , double knockdown of both Gpr27 and Pdx1 was as efficient as single knockdown ( Figure S3 ) . We then asked if Pdx1 was required for the effect of Gpr27 on insulin secretion . The knockdown of Pdx1 reduced fractional insulin secretion at 20 mM glucose and total insulin content ( Figure 4E and Figure S4 ) . As with the adenoviral knockdown of Gpr27 , an siRNA to Gpr27 reduced glucose stimulated insulin secretion . However , the knockdown of Gpr27 in addition to Pdx1 did not further reduce insulin secretion at 20 mM glucose . We conclude that Gpr27 plays a measurable role in insulin secretion and insulin promoter activity via a mechanism involving Pdx1 . G protein coupling software analysis predicts that Gpr27 could function via Gi or Gq/11 signaling pathways [16] . Since Gpr27 is already expressed in MIN6 cells , we ectopically expressed mouse Gpr27 in HEK293T cells . Robust expression of FLAG-tagged Gpr27 was detected by 24 hours on the surface of the majority of cells ( Figure 5B ) . We then measured cAMP and IP1 – higher cAMP would indicate Gs coupling , lower cAMP would indicate Gi coupling and higher IP1 would indicate Gq/11 coupling ( Figure 5A ) . Gpr27 expression resulted in a 2-fold elevation of IP1 levels while leaving cAMP levels unchanged ( Figure 5C and 5D ) showing that in this heterologous cell type , Gpr27 may activate the Gq/11 pathway . If Gpr27 activates Gq/11 in beta cells , then IP1 levels should be reduced in MIN6 cells after knockdown of Gpr27 . Therefore , we measured IP1 levels and cAMP levels in MIN6 cells after Gpr27 knockdown . Indeed , knockdown of Gpr27 resulted in reduced IP1 levels while cAMP levels were not significantly changed ( Figure 5E and 5F ) . Taken together , these data show that Gpr27 positively regulates inositol phosphate levels , supporting a role for Gpr27 in activating the Gq/11 pathway .
To identify new regulators of the insulin promoter , we developed a novel siRNA screening system in MIN6 cells that allows rapid measurement of insulin promoter activity . As an initial test of the system , an siRNA screen of the GPCR-ome was performed . The RSA algorithm was used to select hits in order to capitalize on the four fold redundancy of the siRNA library [7] . To further increase the specificity of the screen , at least 2 siRNAs must have been identified for a gene to be a hit . The top RSA hits were then prioritized by expression level in mouse primary islets . Besides filtering out genes expressed in MIN6 but not in primary islets , this step also eliminates off-target hits . On the other hand , hit genes with low expression may have been erroneously eliminated because they were below the limit of detection of the mRNA-seq data available at this time [17] . Nonetheless , this filtering step allowed us to focus on genes with reasonable expression in primary cells . While the confirmation rate for siRNAs to positive regulators was 100% , the confirmation rate for negative regulators was only 33% . This is likely due , in part , to the more modest effect of these siRNAs ( ∼20–30% increase in GFP/mCherry ratio ) as compared to the reconfirmed Adra2a ( ∼50% ) , a known negative regulator of insulin secretion . We identified several other known regulators of insulin secretion as regulators of the insulin promoter . The bradykinin receptor 2 mediates increases in insulin secretion in beta cells [18] , [19] . Pyrimidinergic receptor 6 ( p2yr6 ) agonists augment insulin release and this receptor participates in an autocrine feedback loop that potentiates insulin secretion [20] , [21] . The free fatty acid receptor 2 , which has been hypothesized to play a role in beta cells , was also identified as a positive regulator of the insulin promoter in our screen [22] . Several other receptors were identified in the screen that have no known role in beta cells and these may merit further investigation . Of note , Glp1r was not identified in this screen for a trivial reason; siRNAs targeting this gene were not included in the commercial screening set . Given the nature of our screen , hits would be predicted to either have basal activity or have ligand present in the culture conditions as has been described for p2yr6 [21] . We were most intrigued by the orphan GPCR , Gpr27 [23] . Previous studies have shown that it is enriched in the pancreatic islets of both human and mouse [10] , [11] . Detailed mouse tissue profiling of Gpr27 expression by RT-QPCR shows high expression in the mouse brain with lower expression in the islet and heart [24] . Furthermore , Gpr27 mRNA is up-regulated in Neurogenin3 positive endocrine precursors in the developing mouse pancreas [11] . Conversely , Gpr27 is 8-fold down regulated in the Neurogenin3 knockout pancreas [25] , [26] . Taken together , these data suggest Gpr27 is an endocrine pancreas specific gene . We confirmed that knockdown of Gpr27 reduces the activity of human insulin promoter reporters , levels of endogenous mouse Ins2 pre-mRNA , and glucose stimulated insulin secretion . Importantly , Gpr27 knockdown also reduces the levels of endogenous Ins2 pre-mRNA in dissociated primary mouse islets . We also found that the mRNAs for multiple transcription factors that activate the insulin promoter ( Glis3 , Pdx1 , HNF4a ) were reduced by Gpr27 knockdown . Other transcription factors critical for beta cell development were also reduced including Nkx6 . 1 and Pax6 . In agreement with the reduction in their expression , only Pdx1 and Glis3 binding mini-enhancers were affected by Gpr27 knockdown ( Figure 4C ) . Finally , there was no further reduction in insulin promoter activity when adding Gpr27 knockdown to Pdx1 knockdown . A limitation of this double knockdown experiment is that given the very strong effect of Pdx1 knockdown alone on insulin promoter activity , a further reduction with Gpr27/Pdx1 double knockdown may be either below our limit of detection or simply reflect no remaining insulin promoter activity . How might Gpr27 affect both insulin transcription and glucose stimulated insulin secretion ? The Gq/11 pathway was an obvious candidate as the expression of Gpr27 in HEK 293T cells increased IP1 levels while the knockdown of Gpr27 reduced IP1 levels in MIN6 cells . Furthermore , triggering of an engineered Gq/11-coupled GPCR in beta cells increases steady state insulin mRNA levels and insulin secretion [27] . However , Pdx1 levels did not change after triggering this Gq/11-coupled GPCR [27] and Gq/11 knockout beta cells have normal levels of Ins1 and beta cell transcription factor mRNAs [21] . Therefore , even if Gpr27 directly couples to Gq/11 , Gpr27 may affect insulin secretion and insulin promoter activity independent of Gq/11 as has recently been demonstrated for the M3 receptor [28] . Another candidate for mediating the effects of Gpr27 on insulin promoter and insulin secretion was Pdx1 since it is known to positively regulate both insulin transcription and insulin secretion [14] , [15] . We found that the double siRNA knockdown of Gpr27 and Pdx1 produced no further reduction in insulin secretion over Pdx1 knockdown alone , suggesting that Pdx1 is important for Gpr27's effect on insulin secretion . In combination with the reduction in Pdx1 mRNA by Gpr27 knockdown , these data suggest Gpr27 functions upstream of Pdx1 . However , the double knockdown data do not exclude the possibility that a Pdx1 lies in a parallel pathway to Gpr27 and these two pathways intersect upstream of insulin secretion . Taken together , these data suggest that a linear pathway connecting Gpr27 to a single G protein and a single regulatory element in the insulin promoter is overly simplistic . Indeed , a single GPCR can trigger multiple G proteins ( reviewed by [29] ) , can trigger a combination of G protein dependent and independent pathways [30] , and can function as heterodimers [31] . Likewise , the insulin promoter contains multiple elements that are both redundant and cooperative [32] . The complexity of these systems highlights the advantage of using a broad , unbiased approach to finding new and unexpected regulators of the insulin promoter . Here , we used such a system to identify a novel GPCR regulator of both insulin secretion and insulin promoter activity – Gpr27 . Based on its islet expression and its positive effects on the insulin promoter and insulin secretion , we suggest that Gpr27 may be a novel target for diabetes therapies .
MIN6 cells were a gift from Dr . Miyazaki . Alpha TC and beta TC were a gift from Dr . Hanahan . Cells were maintained in high glucose DMEM with 10% fetal bovine serum , and 71 . 5 mM beta-mercaptoethanol . Sublines were isolated by limiting dilution . Original passage lines were used between passage 25–40 . Sublines were used at passages 5–10 . Human insulin promoter deletions have been previously described [5] . Promoters were subcloned from pFoxCAT into pFoxLuc [33] . For the lentiviral reporter , the human −362 promoter region was cloned upstream of destabilized GFP or GFP and this cassette was used to replace the U6/CMV-EGFP in pSicoR . pSicoR-RSV-mCherry was created by replacing the U6/CMV of pSicoR mCherry with the RSV promoter [5] . Mini-enhancer reporter constructs have also been previously described [5] , [34] . They were subcloned upstream of a minimal thymidine kinase promoter-firefly luciferase reporter . Approximately 5 , 000 MIN6 cells were transfected in 96 well plates using HiPerfect ( Qiagen ) with a final siRNA concentration of 25 nM . Cells were analyzed by flow cytometry ( LSRII , BD ) 5 days after transfection and the geometric mean fluorescence intensity of GFP was normalized to that of mCherry . If the knockdown of GFP by an anti-GFP siRNA was not >80% , the transfection of that plate was considered to be a technical failure and the plate was discarded . This occurred on 1 out of 20 plates and for this reason some genes were only targeted by 3 siRNAs ( including Gpr27 ) . Each well was normalized to the negative control siRNA on that 96 well plate . For the confirmation assay for Gpr27 siRNAs , a distinct MIN6 human insulin promoter-GFP/RSV-mCherry reporter line was transfected with the indicated siRNAs with Lipofectamine RNAiMax for 5 days and GFP and mCherry were measured . Mouse islet mRNA-seq data was downloaded from the NCBI Short Read Archive ( SRP000752 and SRP002569 ) and FPKM values were calculated using the TopHat and Cufflinks software using the NCBI RefSeq as the reference . Log FPKM and negative log RSA p values were clustered using Cluster 3 . 0 and heat maps were plotted with JavaTreeView . siRNAs were obtained from Qiagen . All custom Taqman probes had a confirmed PCR efficiency of between 95–110% . Samples without reverse transcriptase did not amplify . See Text S1 for sequences of custom probes . Taqman probes to mouse Glis3 , MafA , Pdx1 , NeuroD1 , Pax4 , Nkx6 . 1 , HNF4a were obtained from Applied Biosystems . Negative control siRNAs for the reconfirmation assay were All-Stars Negative Control ( 1027280 ) , Negative Control ( 1022076 ) , Unspecific-Luciferase-1 ( 1022070 ) , Unspecific-Luciferase-2 ( 1022073 ) , Hs_LMNA_11 ( 1022050 ) , Mm_Lmna_5 ( SI02655450 ) , Hs_GAPD_5 ( SI0253266 ) , Hs_ACTB_1 ( 1022168 ) . Total RNA was isolated by Trizol ( Invitrogen ) . The RNA was DNase I treated ( Turbo DNase , Ambion ) and reverse transcription was performed ( Superscript III , Invitrogen ) using a combination of random hexamers and oligo dT primers . For cell line experiments , each qPCR reaction used between 10–30 ng of total RNA equivalent . To convert to arbitrary linear units , the following formula was used: ( 2∧15 ) * ( 2∧ ( deltaCT to beta-glucuronidase ) . Islets from 12–30 week old MIP-GFP mice were isolated by the UCSF Islet Production Core . Islets were digested with trypsin until single cell suspensions were obtained . Cells were sorted by flow cytometry ( Aria II , BD or MoFlo , DakoCytomation ) into GFP positive and negative fractions and total RNA was isolated . 20 ng of total RNA equivalent was loaded per QRT-PCR reaction . 140 , 000 MIN6 cells were transiently transfected in 24 well plates with the relevant siRNA ( 5 pmoles ) , the indicated insulin reporter firefly luciferase plasmid ( 100 ng ) , and pRL-TK ( Promega ) ( 25 ng ) using Lipofectamine 2000 ( Invitrogen ) . For double siRNA knockdowns , 5 pmoles of each siRNA or 10 pmoles of control ( anti-GFP ) were used . Two days after transient transfection , firely and renilla luciferase were measured using the Dual Luciferase Assay ( Promega ) . Gpr27 was cloned by PCR from mouse genomic DNA downstream of a viral signal sequence and amino terminal FLAG epitope tag [35] . This cassette was used to replace the EGFP in pSicoR . HEK 293T cells were transiently transfected with either pSicoR-EGFP or pSicoR-FLAG-Gpr27 using LT1 ( Mirus ) . 293T cells were dissociated with PBS without Ca or Mg , stained with M1 anti-FLAG antibody and a Goat anti-mouse secondary antibody coupled to Alexa-594 ( Invitrogen ) . For 293T , one day after transient transfection in 24 well plates , cells were placed in stimulation buffer ( HTRF ) for 30 minutes at 37 degrees . The stimulation buffer was then removed and the cells were lysed using the kit lysis buffer . IP1 and cAMP were then measured as directed by the protocol in 384 well plates ( HTRF ) . IP1 and cAMP levels were normalized to live cells numbers counted from duplicate wells . Viable cells counts from Gpr27 transfection were within 20% of control plasmid transfection . For MIN6 cells , 125 , 000 cells were infected with Gpr27 shRNA or control adenovirus at an MOI of 200 and grown in 24 well dishes ( resulting in nearly ∼95% infection ) . Three days after infection , the cells were placed in stimulation buffer ( HTRF ) for 30 minutes at 37 degrees . The stimulation buffer was then removed and the cells were lysed in 1% Triton-X100 , 50 mM HEPES pH 7 . 0 , NaF 15 mM . The lysate was pre-cleared by centrifugation at 14 , 000 rpm for 10 minutes . A fraction of the lysate was taken for protein quantitation ( micro-BCA , Pierce ) , IP1 or cAMP measurement ( HTRF ) . Data were normalized to total protein content . The Gpr27 shRNA was cloned into a modified version of pSicoR with a BstXI site replacing the HpaI site . The mouse U6 promoter and Gpr27 shRNA were then subcloned from pSicoR and placed upstream of the CMV-GFP marker in pAdTrack [36] , [37] . Adenovirus was prepared and tittered as previously described [38] . Islets were isolated by the UCSF Islet Production Core Facility from 8–12 week old C57Bl/6 male mice . After 24 hours of culture in RPMI and 10% FBS , islets were trypsinized until single cell suspensions were obtained . The dissociated islet cells were resuspended in RPMI+10% FBS and infected with adenovirus at multiplicity of infection ( MOI ) of 25 . Three days after infection , the cells were sorted by flow cytometry ( Aria II , BD ) for GFP positive cells ( 50–75% of the live population ) and RT-qPCR was performed . The knockdown of pre-ins2 , pre-ins1 or Gpr27 from Gpr27 shRNA adenovirus infected cells was calculated by the delta-delta CT method compared to the control adenovirus infection . For the adenovirus assays , approximately 500 , 000 MIN6 cells were infected with the indicated adenoviruses at an MOI of 100 in 6 cm dishes in complete media . Three days after infection , the infection rate was ∼60% by FACS for GFP . Cells were washed 5 times in KRBH buffer ( 10 mM HEPES pH 7 . 4 , 130 mM NaCl , 5 mM KCl , 1 . 25 mM KH2PO4 , 1 . 25 mM MgSO4 , 2 . 68 mM CaCl2 , 5 . 26 mM NaHCO3 ) with 2 mM glucose and rested for 2 hours at 37 degrees . Cells were then washed an additional 3 times with 2 mM glucose KRBH and incubated in 3 mL of 2 mM glucose KRBH for 1 hour at 37 degrees . This supernatant was collected and replaced with 20 mM glucose KRBH for 1 hour at 37 degrees . Cells were washed with PBS before lysis in 50 mM Tris-HCl pH 8 . 0 , 150 mM NaCl , 1% Triton X-100 with protease inhibitors . Lysates were spun at 14 , 000 rpm for 10 minutes and supernatants were spun at 5000 rpm for 5 minutes before analysis by an Ultrasensitive Insulin ELISA ( Mercodia ) . Total protein was measured by Micro-BCA ( Pierce ) . Total insulin was normalized to total protein in the lysate . For the siRNA transfections , 20 , 000 MIN6 cells were transfected per well of a Corning CellBIND 96 well plate with 25 nM of each siRNA ( or 50 nM of control siRNA ) using Lipofectamine RNAiMax . 5 days after transfection , cells were washed in KRBH with 2 mM glucose twice , then incubated for 2 hours at 37 degrees , then washed again with KRBH 2 mM glucose twice , then incubated for one hour with KRBH 2 mM , then KRBH with 20 mM glucose for 1 hour . Lysates were prepared in 75 uL of lysis buffer as above . Due to the lower cell numbers in the 96 well plate assay , total insulin was normalized to total genomic DNA measured by Qubit High Sensitivity DNA kit ( Life Technologies ) . For siRNA primary confirmation assay , an independent , two sample , one tailed t-test was used . For the primary islet adenovirus knockdown of Gpr27 an independent , one sample , two tailed t-test was used . All other p values were calculated with an independent , two sample , two tailed t-test . Animal experiments were approved by the UCSF Institutional Animal Care and Use Committee ( Protocol AN082433-02 ) with care taken to avoid any unnecessary suffering . Animals were maintained in accordance with the applicable portions of the Animal Welfare act and the DHHS Guide for the Care and Use of Laboratory Animals . | Pancreatic beta cells are the only physiologic source of insulin . When these cells are destroyed in type 1 diabetics , there is uncontrolled hyperglycemia from complete insulin deficiency . In type 2 diabetes , these same cells fail to increase insulin secretion to compensate for peripheral insulin resistance leading to relative insulin deficiency . We constructed a novel screening system to find new regulators of insulin production in this critical cell type . Here , we describe a screen of the G protein coupled receptors ( GPCRs ) and show a role for orphan GPCR , Gpr27 , in insulin promoter activity and insulin secretion . We propose that Gpr27 is a novel target for diabetes therapeutics . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"diabetic",
"endocrinology",
"endocrinology",
"endocrine",
"system",
"physiology",
"biology",
"anatomy",
"and",
"physiology",
"diabetes",
"and",
"endocrinology",
"endocrine",
"physiology"
] | 2012 | An siRNA Screen in Pancreatic Beta Cells Reveals a Role for Gpr27 in Insulin Production |
Domestic dog breeds exhibit remarkable morphological variations that result from centuries of artificial selection and breeding . Identifying the genetic changes that contribute to these variations could provide critical insights into the molecular basis of tissue and organismal morphogenesis . Bulldogs , French Bulldogs and Boston Terriers share many morphological and disease-predisposition traits , including brachycephalic skull morphology , widely set eyes and short stature . Unlike other brachycephalic dogs , these breeds also exhibit vertebral malformations that result in a truncated , kinked tail ( screw tail ) . Whole genome sequencing of 100 dogs from 21 breeds identified 12 . 4 million bi-allelic variants that met inclusion criteria . Whole Genome Association of these variants with the breed defining phenotype of screw tail was performed using 10 cases and 84 controls and identified a frameshift mutation in the WNT pathway gene DISHEVELLED 2 ( DVL2 ) ( Chr5: 32195043_32195044del , p = 4 . 37 X 10−37 ) as the most strongly associated variant in the canine genome . This DVL2 variant was fixed in Bulldogs and French Bulldogs and had a high allele frequency ( 0 . 94 ) in Boston Terriers . The DVL2 variant segregated with thoracic and caudal vertebral column malformations in a recessive manner with incomplete and variable penetrance for thoracic vertebral malformations between different breeds . Importantly , analogous frameshift mutations in the human DVL1 and DVL3 genes cause Robinow syndrome , a congenital disorder characterized by similar craniofacial , limb and vertebral malformations . Analysis of the canine DVL2 variant protein showed that its ability to undergo WNT-induced phosphorylation is reduced , suggesting that altered WNT signaling may contribute to the Robinow-like syndrome in the screwtail breeds .
Morphological differences have been one of the primary drivers of dog breed formation since wolf domestication and subsequent selection to create dog breeds [1] . In many cases , the morphological traits are also genetically linked to disease-predisposition traits [2–6] . Therefore , it is of great interest to determine the breed specific genetic variations , as this knowledge could provide insights into not only the evolution of dog breeds , but also the mechanisms of tissue morphogenesis and disease pathogenesis . Some subsets of dog breeds share distinctive morphologies . A shortened and kinked tail—which is referred to as a “screw tail”—is one of the distinctive morphological traits that characterizes Bulldogs , French Bulldogs and Boston Terriers , which were historically developed from the Bulldog breed and , thus , closely related [7] . These breeds also share a craniofacial morphological phenotype referred to as brachycephaly , which includes both profound shortening of the muzzle and widening of the skull . Possibly due to the width of their heads , these three breeds have the highest rate of cesarean section among dog breeds [8] . The three breeds are also all relatively short in stature being less than 17 inches tall at the shoulder . In addition to morphological features , certain diseases are found at a high prevalence across all 3 breeds such as vertebral malformations [9–13] , cleft lip and cleft palate [14–16] , congenital heart disease [17 , 18 , 19] , and glioma [5 , 20 , 21] . Molecular characterization of the cause of short tails in dogs demonstrated that a synonymous mutation in the T box transcription factor gene confers a short tail ( bob tail ) phenotype in the Pembroke Welsh Corgi breed [22] . Although this mutation is common across a wide range of dog breeds , it is not responsible for the short tail seen in Bulldogs and Boston Terriers [23] . Screw tail is distinguishable from the short tail phenotypes in other breeds due to presence of vertebral malformations and fusion of the caudal vertebrae in addition to absence of caudal vertebrae . Additional conformational sequelae of vertebral malformation in screw tail breeds includes kyphosis , lordosis or scoliosis most commonly affecting the thoracic vertebral column , and typically resulting from wedge , hemi-vertebrae and butterfly vertebrae [24 , 25] . Although vertebral malformations affecting thoracic vertebrae in screw tail breeds are common , they are rarely directly associated with clinical signs [9] . However , it may increase their risk of developing other diseases such as intervertebral disc disease , to which the French Bulldog and Boston Terrier are predisposed [6 , 26] . Brachycephalic skull morphology has been previously investigated in dogs and has a complex etiology . Multiple different chromosomal regions are associated and may differ between breeds . In a Genome Wide Association ( GWAS ) for skull shape across dog breeds , significant associations were identified on CFA ( Canis familiaris ) 1 , CFA5 , CFA24 , CFA30 and CFA32 for brachycephaly . A missense mutation ( p . F452L ) in the bone morphogenetic protein 3 ( BMP3 ) gene was identified as the skull modifying locus on CFA 32 . This allele is fixed in Bulldogs and French Bulldogs and has an allele frequency of 0 . 99 in Boston Terriers [27] . The mutations underlying the associations on CFA5 , CFA24 and CFA30 have not been reported . A second group mapped proportional snout length in a GWAS across dog breeds using size as a covariate and identified both the CFA1 and CFA5 loci; however underlying causative mutations were not defined [28] . Recently , a LINE-1 insertion in the SPARC-related modular calcium binding 2 ( SMOC2 ) gene was identified as the causative mutation underlying the significant associations to brachycephalic head morphology on CFA 1 [29] . In humans , a rare genetic disorder called Robinow syndrome shares phenotypic similarities with the screw tail breeds . Robinow syndrome is characterized by mesomelic-limbed dwarfism and abnormalities of the head , face , genitalia , and vertebral column [30] . Patients with Robinow syndrome have hypertelorism with a broad nasal root and broad forehead and fusion of thoracic vertebrae with frequent hemivertebrae [31] . Mutations in genes from WNT pathways , including ROR2 , FZD2 , WNT5A , DVL1 , and DVL3 , have all been found to cause Robinow syndrome in humans [30 , 32–34] . WNT pathways control embryonic morphogenesis by regulating crucial developmental processes including cell-fate determination , proliferation , and morphogenetic cell/tissue movement [35–40] . Since the development of SNP genotyping arrays for dogs , the standard approach to mutation identification has been GWAS followed by Sanger or whole genome sequencing to identify causative variants [41] . Compared to other species , long linkage disequilibrium ( LD ) within the domestic dog allows successful associations with few samples and few SNPs [42] . However the long LD within dogs also presents challenges for mutation identification . In this paper , we by-pass the use of SNP genotyping arrays and utilize variant calls from whole genome paired end sequences from 100 canine samples of 21 breeds to perform whole genome variant association in the screw tail breeds . We report the identification of a frameshift mutation in the DISHEVELLED 2 ( DVL2 ) gene that segregates with the breed defining phenotype of screw tail and vertebral malformations . The frameshift mutation preserves the majority of the DVL2 protein but replaces the last 49 amino acids in the C-terminus with a novel 26-amino acid sequence . Through biochemical analysis of the bulldog DVL2 variant protein in cultured cells , we further demonstrate that the mutant protein has a reduced capacity to undergo WNT-dependent phosphorylation , suggesting that aspects of WNT signaling might be compromised . Our data suggest that the bulldog-related breeds share similar genetic , morphological and pathological origins with human Robinow syndrome .
Bulldogs , French Bulldogs and Boston Terriers are the only American Kennel Club recognized dog breeds characterized by a screw tail ( Fig 1A ) . The ‘screw tail’ is caused by a variety of malformed and fused vertebrae and lack of approximately 8 to 15 caudal vertebrae , which normally form the canine tail ( Fig 1B 4 ) . In addition to deformities of the caudal vertebrae , these breeds may also have variable morphological deformities of vertebrae along the vertebral column including hemivertebrae , wedge vertebrae , butterfly vertebrae and fused vertebrae ( Fig 1B 1–3 ) . Based on breed standards , the three breeds share physical characteristics including short stature , head shape and include eyes to be as wide apart as possible ( hypertelorism ) and for them to have a broad muzzle ( Fig 1A and S1 Table ) . These breeds are considered brachycephalic meaning that they have a short muzzle length . The degree of brachycephaly in the three screw tail breeds is more prominent than in some other brachycephalic breeds due to an even shorter and broadened maxilla , broadened frontal bone , and an increased curvature to the zygomatic bone creating a wider and more extreme orbit as compared to , for example , the Boxer breed ( Fig 1C ) . Due to the similarity in phenotype and historical relationships between these breeds , we hypothesized that they would share a mutation responsible for their morphology . To identify variants responsible for the screw tail phenotype , we used paired end whole genome sequence data generated from 100 dogs from 21 breeds followed by association analysis of all biallelic variants identified . There were 6 trios included in the dataset . The twenty-one breeds included 5 Bulldogs , 3 French Bulldogs and two Boston Terriers . The remainder of the dogs did not have screw tails; however , two of the breeds were brachycephalic ( Boxer and Pug ) . A complete list of breeds is available in S2 Table . The variant calling pipeline identified 15 , 353 , 085 SNPs and 8 , 514 , 447 indels within the 100 canine genomes . After quality filtration and exclusion of variants of uncharacterized chromosomes , 13 , 591 , 986 SNPs and 7 , 126 , 341 indels passed our filters . The average rates of Mendelian errors per meiosis were calculated to be 4 . 1% in the six dog trios . A dendrogram was constructed to examine the historical relatedness of the breeds . Clustering of dogs by their breed and the relatedness in the case of the trios confirms the sensitivity of the technique to detect genetic relatedness ( S1 Fig ) . The dendrogram highlights a common origin of all screw tail breeds ( Bulldogs , French Bulldogs and Boston Terriers ) and shows the bifurcation from Boxer dogs and Pugs , the other two brachycephalic breeds ( Fig 2A and S1 Fig ) . A whole genome association analysis was performed using the 10 dogs of the screw tail breeds compared to 84 dogs belonging to 21 pure and 4 mixed breeds ( related dogs were removed ) . The association analysis started with 20 , 342 , 844 biallelic variants identified through the whole genome sequence ( excluding 375 , 483 multi-allelic loci ) with a total genotyping rate of ~93 . 6% . 3 , 023 , 667 variants were removed due to missing genotype data and 4 , 852 , 941 variants were removed for low minor allele frequency . Association testing was performed using the remaining 12 , 461 , 460 variants . Genomic inflation was high ( estimated lambda = 3 . 2761 ) due to the expected population stratification . To minimize false positive associations caused by genomic inflation , a simple genomic control approach was used instead of other modeling approaches that correct for relatedness to avoid the exclusion of variants that were identical by descent . Since there is relatively long linkage disequilibrium in dogs , statistical thresholds were further corrected for multiple testing of 587 , 159 variants representing common haplotypes in the tested population . After adjustment , a PBonferroni of 0 . 05 is equivalent to a p = 3 . 99 X 10−22 and PBonferroni of 0 . 01 is equivalent to a p = 1 . 96 x 10−24 ( Fig 2B , Fig 2C ) . A list of all the significantly associated variants is shown in S3 Table . The most associated variant was CFA5: 32195043_32195044del , p = 4 . 37 X 10−37 ( uncorrected ) ; however there were many associated regions ( Fig 2C ) . This variant remained significantly associated at a PBonferroni of 0 . 01 corrected for all 12 , 461 , 460 variants as well ( p = 1 . 5 x 10−28 ) . In addition to being highly associated , we predicted that variants responsible for breed defining characteristics would have a high allele frequency in affected breeds and an extremely low allele frequency in unaffected breeds . We therefore selected variants with more than 90% allelic difference between cases and controls . Long regions of fixed candidate variants were then selected for further analysis and included CFA5 , CFA26 and CFA 32 ( Fig 2D ) . There were two additional significant loci on CFA 25 and 29 that were not evaluated further since they were less significant and did not show long regions of homozygosity to support Identical by descent inheritance [43] . The two peaks on CFA 26 ( 6772912–10171935 ) and CFA 32 ( 4533724–6620950 ) were previously reported to be associated in these breeds with canine glioma and brachycephaly respectively [5 , 27] . The highest association was to the region on CFA 5 , which also had a long region of homozygosity ( 5:29243555–34607475 ) . The single most significantly associated variant found was a frame shift mutation in the Dishevelled 2 ( DVL2 ) gene ( g . 32195043_32195044del ) . The mutation was homozygous in all cases and absent from controls except a single Labrador retriever that was called as heterozygous in the absence of supporting read coverage ( S3 Table ) . This dog was later confirmed by Sanger sequencing to be wild-type at this location . The single base deletion found on CFA 5 ( g . 32195043_32195044del ) that was homozygous in the three screw tail breeds ( 5 Bulldogs , 3 French Bulldogs and 2 Boston Terriers ) was located within the 15th and penultimate exon of the canine DVL2 gene . DVL2 cDNA was sequenced from the skeletal muscle of a dog with a normal tail and a screw tail Bulldog ( Fig 3A ) to confirm the presence of the mutation in the mRNA in the Bulldog sample ( DVL2c . 2044delC ) . In orde to determine if there was a difference in transcript level between animals with the DVL2c . 2044delC mutation and without semiquantitative RT PCR was performed . Mutant transcript levels were comparable to wildtype ( S2 Fig ) . This deletion is predicted to lead to a frameshift mutation , causing a premature stop codon that truncates the translated protein by 23 amino acids ( p . Pro684LeufsX26 ) . In addition , 26 altered amino acids are predicted to be present in the highly conserved C-terminus of the mutant protein ( Fig 3B , Fig 3C ) . The location of the truncation of the protein is remarkably similar to the effect of mutations within the other Dishevelled family members , DVL1 and DVL3 , that lead to Robinow syndrome in people . Alteration of the amino acid sequence in the highly conserved C-terminal region as well as truncation of 21 to 50 amino acids also occurs in the human Robinow DVL mutations ( Fig 3D and S4 Table ) . To confirm the association of the DVL2c . 2044delC mutation with the screw tail phenotype , 667 dogs , from 49 breeds , were genotyped for the DVL2 mutation ( Table 1 ) . 177 dogs were from the screw tail breeds including 33 Bulldogs , 79 French Bulldogs and 65 Boston Terriers . All were homozygous for the mutant allele except 6 of the Boston Terriers ( 4 heterozygous , 2 wildtype ) . In addition , we identified dogs from several other breeds , including Pit bulls , Staffordshire Bull Terrier , Shih Tzu and mixed breeds , that are heterozygous or homozygous for the DLV2 mutation . The Pug breed has sometimes been classified with the screw tail breeds due to its curled tail; however , the tail is full length and does not have caudal vertebral malformations ( S3 Fig ) . 29 Pugs tested were wild-type for the DVL2 mutation . Likewise , the Pug dogs do not share the high MAF with the screw tail breeds around the DVL2 mutation ( S4 Fig ) . Three hundred and eighty five dogs from 43 other breeds were also tested and were all wild-type ( S5 Table ) . In order to determine the penetrance of vertebral malformations , dogs homozygous for the DVL2 variant from the screw tail breeds were evaluated for thoracic and caudal vertebral malformations ( Table 2 ) . The penetrance of the thoracic malformations varied between the three breeds from 45–100% , while the caudal vertebral malformations were 100% penetrant . Since the DVL2 variant is virtually homozygous in these breeds , it was important to evaluate segregation of the variant with phenotype . In order to confirm the association of the DVL2 mutation with the vertebral column malformations , additional dogs were identified that had radiographs or imaging available and segregated the DVL2 mutation . Segregation of the caudal vertebral column malformations and genotype at DVL2 was consistent with a fully penetrant recessive mode of inheritance in Boston Terriers , Shih Tzus , Pit Bulls and mixed breeds ( Table 3 ) . Segregation of the thoracic vertebral malformations was consistent with a recessive mode of inheritance with variable penetrance between breeds . In order to evaluate the molecular contribution to brachycephalic skull shape , the DVL2 mutant breeds were genotyped for the previously identified brachycephaly associated Line insertion affecting SMOC 2 splicing . Bulldogs and French Bulldogs were homozygous for the variant and the Boston Terriers had an allele frequency of 90 . 3% ( S6 Table ) . Similar allele frequency for the BMP3 missense mutation was also reported for these three breeds indicating that they are homozygous or have a high allele frequency for three mutations that affect head shape [27] . Based on the significant changes to the DVL2 C-terminus caused by the frameshift mutation , we hypothesized that the expression and/or biochemical properties of the mutant variant DVL2 might be altered . To test this hypothesis , we synthesized full-length cDNA of both the wild-type and mutant variant dog DVL2 gene , N-terminally tagged them with the MYC epitope , and expressed them via lentiviral vectors in NIH/3T3 cells . We chose to use NIH/3T3 cells because they have been used previously to study WNT pathways and DVL regulation [44 , 45] . Western analysis of lysates from cell lines expressing a wild-type or mutant variant of DVL2 showed that both proteins were expressed at similar levels ( Fig 4A ) , suggesting that the mutant protein is properly synthesized and its novel C-terminus does not affect the stability of the DVL2 protein . It has been previously established that Wnt stimulation results in the phosphorylation of DVL2 , which can be observed on western blots as gel motility shifts and is often used as an indicator of pathway activity [46–49] . To compare the capacity of wild-type and mutant variant DVL2 to respond to WNT signals , we stimulated cells expressing these proteins with purified , recombinant WNT5A or WNT3A and analyzed the extent of DVL2 gel mobility shifts on western blots . We observed prominent DVL2 gel motility shifts in cells expressing the wild-type DVL2 protein after WNT5A or WNT3A treatment ( Fig 4A ) . However , DVL2 gel motility shifts were reduced in cells expressing the mutant protein , suggesting that phosphorylation of the mutant variant DVL2 may be impaired . To demonstrate that the gel motility shifts observed were indeed due to phosphorylation , we treated cell lysates with calf intestinal phosphatase ( CIP ) to remove any potential phosphate groups . CIP treatment resulted in loss of the slower migrating DVL2 bands and an increased amount of unphosphorylated DVL2 in all conditions ( Fig 4B ) . These results indicate that Wnt-dependent phosphorylation of the mutant variant DVL2 protein is reduced compared to wild-type DVL2 . Additional studies conducted in human and mouse cells have demonstrated that WNT-dependent phosphorylation of DVL2 is dependent on casein kinase 1 ( CK1 ) [50 , 51] . To assess if this mechanism also drives phosphorylation of the canine DVL2 protein , we pretreated cells with D4476 , a small molecule inhibitor of CK1 , and then stimulated with either WNT5A or WNT3A . D4476 treatment resulted in the loss DVL2 phosphorylation in all conditions , thereby demonstrating that CK1 also mediates Wnt-dependent phosphorylation of the canine DVL2 protein ( Fig 4C ) . Collectively , these results demonstrate that the canine mutant variant DVL2 protein exhibits reduced WNT- and CK1-dependent phosphorylation and further suggest that reduced WNT signaling may contribute to the Robinow-like phenotype of bulldogs and associated screw tail breeds .
Whole genome sequence data from 100 dogs was utilized to interrogate the molecular cause of the breed defining trait screw tail . Using over 12 million bi-allelelic variants and comparing 10 cases to 84 controls , a frame shift mutation in the DVL2 gene was found to be the most strongly associated of all variants . The mutation leads to a 23 amino acid truncation of the protein in the last exon within the highly conserved C-terminal domain . Analogous truncations in the same regions of human DVL1 and DVL3 proteins result in Robinow syndrome in humans . This shared genetic signature , taken together with the similar anatomical changes , strongly suggests that the bulldog-related breeds share common pathological origins with human Robinow syndrome . DVL2 is part of an evolutionarily conserved cytoplasmic scaffolding protein family that also includes DVL1 and DVL3 in vertebrates [52 , 53] . The three mammalian DVL homologs display significant sequence identity to each other [54 , 55] as well as across species [56 , 57] . Our results suggest that the mutant variant DVL2 protein likely has reduced capabilities to mediate WNT signaling . Although DVL proteins are key players in a variety of WNT pathways , the Robinow-like phenotypes of bulldogs and human patients coupled with mouse developmental expression patterns lead us to believe that the frameshift mutations in DVL2 in screw tail dog breeds primarily affect a noncanonical branch of WNT signaling , the WNT5A-ROR pathway . In addition to the bulldog DVL2 frameshift mutation , human mutations causing Robinow syndrome have been identified in WNT5A , ROR2 , FZD2 , DVL1 , and DVL3 , which are all major components of the WNT5A-ROR pathway [32 , 33 , 58–61] . Further , WNT5A and ROR2 are highly expressed in the facial primordia , limb buds , and vertebrae of mice , all areas of the body that are affected in Robinow syndrome patients and bulldogs; additionally , Wnt5a and Ror2 knockout mice exhibit Robinow-like features , including truncated limbs and broad , flat faces [44 , 46] . Collectively , this suggests that screw tail dog breeds and Robinow syndrome patients at least partly share similar underlying pathophysiology . In both bulldogs and human patients , frameshift mutations in the penultimate or ultimate exon of DVL proteins result in the substitution of the highly conserved C-terminus region of the proteins with a new stretch of amino acids . Given that the frameshift mutation leaves most of the protein intact , it remains plausible that the more N-terminal modular domains of DVL2 can still function in other Wnt pathways , such as canonical WNT/beta-catenin signaling , which primarily uses the DIX domain , and planar cell polarity , which primarily uses the DEP domain [53] . Beyond these modular domains , however , few roles have been assigned to the DVL2 C-terminus; work by Bernatik et al has shown that human DVL3 C-terminus contains some CK1 phosphorylation sites that are conserved in DVL2 and would be lost by the frameshift mutation [62] . This correlates with our observations that the mutant DVL2 variant exhibits reduced CK1-dependent phosphorylation in response to WNT stimulation and is consistent with a recessive mode of inheritance . Given the importance of DVL2 in WNT5A-ROR signaling and the potential roles that these pathways play in defining the unique phenotypical and pathological characteristics of the screw tail bulldog breeds , additional follow up studies are required to define the molecular mechanism ( s ) by which the DVL2 variant protein affects downstream WNT signaling activity during development . In addition to the Dvl2 frameshift mutation , the three Bulldog related breeds harbor other mutations in developmentally important genes that could affect their craniofacial morphology . For example , these breeds also carry a missense mutation in BMP3 and a Line 1 insertion that affects splicing of the SMOC2 gene [27 , 29] . The absence of homozygosity for these mutant alleles in breeds fixed for the brachycephalic head phenotype indicates that this is a complex trait influenced by multiple loci . Screw tail breeds are distinguished from some of the other brachycephalic dogs by additional shortening and broadening of the muzzle , broadening of the skull , and hypertelorism suggesting the presence of more variants affecting their skull morphology ( Fig 1A , S1 Table and [27 , 63] ) . We did not undertake the evaluation of head morphology in this work since recruiting pet dogs for head CT that is not medically warranted is ethically challenging since it requires general anesthesia and carries a risk to dogs with brachycephalic head conformation . However , previous across breed genome wide associations for head morphology identified this locus on CFA 5 [27 , 28] . Boyko et al used breed-based averages for a large number of linear measurements followed by QTL mapping [28] . Schoenebeck et al performed principal component analysis of geometric morphometry of museum specimen skulls from breeds followed by GWAS using DNA samples from the same breeds [27] . Both groups identified the same highly associated SNP ( CFA5:32359028 ) within 160 Kb of the DVL2 mutation as having a significant contribution to skull shape in dogs . A third study on canine skull morphology that was based on individual measurements taken from skull CTs did not identify the CFA 5 locus; however , the screw tail breeds only made up ~2% of the sample sets used in this across breed study to identify QTLs that affect head morphology [29] . Since the study was designed to capture common loci across many breeds it is not surprising that the DVL2 locus would not be identified . Based on our allele frequency measures , the mutation is rare across breeds and virtually homozygous in affected breeds . Additional , as yet undiscovered , deleterious variants could be present in these breeds that affect their skull shape considering the extremely strong selection based on head phenotype applied to these breeds by dog breeders . Normal development of the skull requires coordinated development of cranial sutures , skull base synchondroses and brain , and it is likely that genetic abnormalities may affect both suture and synchondrosis development directly , or indirectly due to secondary effects [64 , 65] . WNT signaling has been shown to regulate cranial base development and growth , and abnormalities in WNT signaling have been implicated in cranial synostosis in humans , most commonly with brachycephaly-associated coronal synostosis [66–68] . Given the apparent polygenic pathogenesis of brachycephaly in dog breeds , and the essentially fixed nature of the DVL2 mutant allele in the Bulldog and Boston breeds , assigning specific skull morphometric sequelae to the DVL2 mutation is challenging . The cranial dysmorphology seen in Robinow patients and the more extreme nature of the brachycephaly in the DVL mutant dog breeds is , however , highly suggestive of DVL2’s involvement in the brachycephalic phenotype . A more detailed study of skull morphology , particularly in animals segregating the various associated genes isnecessary to define specific gene contributions to the brachycephalic phenotype in dogs . Although caudal vertebral malformations appear to be a consistent finding within DVL2 mutant dog breeds , thoracic vertebral malformations are more variable in their presence and severity . Similar findings are observed in mouse models and human patients with WNT pathway abnormalities where penetrance of vertebral anomalies is variable . Even in highly inbred Dvl2 knockout mice , only 90% were reported to have vertebral anomalies [69] , and hemivertebrae and associated scoliosis/kyphosis are seen variably in >75% and <25% of cases with recessive and dominant forms of Robinow syndrome respectively [70] . In addition to vertebral abnormalities , Dvl2 knockout mice exhibit 50% perinatal lethality due to cardiac defects , 25% have tail kinks and 4% have tail truncations [69] . Based on the presence of protein product found in MYC tagged experiments , we propose that the canine DVL2 mutation is not a null mutation but rather a hypomorphic mutation with respect to vertebral malformations . This is also consistent with the more severe phenotype seen in DVL2 knockout mice . The possibility exists that there are different effects in different developmental pathways or tissues , and possibly , polygenic effects due to the presence of many segregating mutations that are already known to affect size and skull shape in domestic dogs . The ability to perform whole genome association in dogs allowed the elimination of the fine structure mapping step in causative variant identification . This whole genome variant association approach successfully replicated two previously identified loci that were known to be fixed within the screw tails breeds namely the BMP3 missense mutation ( CFA32 ) associated with head morphology and the glioma susceptibility locus ( CFA26 ) [5 , 27] . This approach is particularly tractable in the dog where deleterious variants are shared within and across breeds and only a single causative variant is expected . However , it should be noted that only SNPs and small insertion deletions were identified in this analysis , and there are many examples of disease causing variants that would not have been identified using this approach [6 , 29] . As whole genome sequencing costs continue to decrease and our abilities to call variants improves , whole genome variant association provides an efficient method to define disease causing variants .
The following application was reviewed and approved by the UC Davis IACUC on January 19 , 2018 . Title: Canine DNA collection from privately owned animals . Principal Investigator: Danika L . Bannasch ( Protocol # 20356 ) Institution: University of California , Davis . Active protocols are reviewed annually . This institution is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care , International ( AAALAC ) . This institution has an Animal Welfare Assurance on file with the Office of Laboratory Animal Welfare ( OLAW ) . The Assurance Number is A3433-01 . The IACUC is constituted in accordance with U . S . Public Health Service ( PHS ) Animal Welfare Policy and includes a member of the public and a non-scientist . Buccal swabs or blood samples were collected from privately owned dogs through the William R . Pritchard Veterinary Medical Teaching Hospital at UC Davis . Owners specified the breed of each dog . Samples used for whole genome sequencing included 96 dogs from 21 pure breeds and 4 dogs from mixed breeds ( S2 Table ) . Images of screw tail breeds were obtained during the course of necessary veterinary procedures and 3D computed tomography reconstructions were perfomed on selected images . The American Kennel Club breed standards were used as a guide to describe typical dogs of each breed . Vertebral column phenotypes defined as a ) thoracic vertebral malformation and b ) caudal malformation ( “screw tail” ) were based on assessment of imaging ( radiographs , computed tomography or magnetic resonance imaging ) by a board certified veterinary radiologist and a board certified veterinary neurologist . Cases were considered affected if there was evidence of thoracic vertebral malformations characterized by the presence of wedged vertebrae , hemi-vertebrae or butterfly vertebrae , or the presence of similar malformations as well as shortening and vertebral fusions affecting the caudal vertebrae . Inclusion required the availability of imaging for all thoracic vertebrae in lateral and dorso-ventral /ventro-dorsal planes for thoracic vertebrae , and imaging of a minimum of 6 caudal vertebrae . Phenotyping of 4 cases for the presence of caudal malformation ( “screw tail” ) was based on visual and physical examination by a veterinarian . Genomic DNA was extracted using the Qiagen kit ( QIAGEN , Valencia , CA ) . 96 biological samples ( including 6 trios ) were subjected to next generation sequencing using Illumina paired end cycles . The whole genome sequencing of 4 Pug samples was publically available by Tgen company ( https://www . tgen . org/ ) . The metadata table contains details about the sequencing libraries and coverage ( S2 Table ) . Adaptors and low quality sequences were removed using the Trimmomatic software ( V 0 . 36 ) [71] . Adaptor trimming was done using recommended parameters of simple matching ( threshold of 10 ) and palindromic matching ( threshold of 30 and minimum adapter length of 1 ) . High quality reads were aligned to the dog reference genome CanFam3 [72] using the BWA-MEM algorithm of the BWA software package ( v0 . 7 . 7 ) [73] . Duplicate reads were excluded using the Picard tool MarkDuplicates ( v2 . 2 . 4 ) ( http://broadinstitute . github . io/picard ) . Variant calling was performed with the GATK HaplotypeCaller ( v3 . 5 ) [74] using joint genotyping across all sequenced samples . Known variants from the Ensembl variation database ( release 82 ) [75] and canine annotation of Broad institute ( https://www . broadinstitute . org/ftp/pub/vgb/dog/trackHub/canFam3/variation/final . Broad . SNPs . vcf . gz ) were used for variant annotation . Candidate variants were filtered using the following thresholds: QualByDepth ( QD ) < 2 . 0 , FisherStrand ( FS ) > 60 . 0 , StrandOddsRatio ( SOR ) > 4 . 0 , ReadPosRankSum < -8 . 0 , and depth of coverage ( DP ) > 3105 for both SNPs and indels , RMSMappingQuality ( MQ ) < 40 . 0 , MQRankSum < -12 . 5 for SNPs , and InbreedingCoeff < -0 . 8 for indels . After quality filtration and exclusion of variants of uncharacterized chromosomes , 13 , 591 , 986 SNPs and 7 , 126 , 341 indels passed our filters . The PLINK software for GWAS is designed to deal with SNPs for association studies[76] . To allow PLINK to deal with indels as well , we developed a script which changed the bi-allelic indels into SNPs . Here , we excluded multi-allelic variants then replaced multi-character alleles by a single character ( A or T ) chosen to maintain allelic variation between reference and alternatives ( https://github . com/dib-lab/dogSeq ) . An IBS matrix was calculated to document the breed-based population stratification and examine the similarities between the breeds used . A subset of autosomal and X chromosome variants genotyped in >95% of samples with MAF > 5% was selected and then subjected to linkage disequilibrium based pruning using a threshold of variance inflation factor ( VIF ) equals 2 . Pruning recursively removed SNPs within a sliding window of 50 SNPs , with a window step size of 5 SNPs producing 645 , 697 variants for IBS calculation . The distance matrix was constructed using the ‘—distance 1-ibs’ function of PLINK 1 . 9 and plotted as a dendrogram using the ‘ape’ package in R . The “—mendel” option in PLINK 1 . 9 was used to calculate the rate of Mendelian errors per meiosis in the sequenced trios using the pruned subset of variants . The average rate of Mendel errors was used as an index for the genotyping accuracy . Among the 100 dogs sequenced , there were 6 trios whose offspring were excluded from further analysis . All variants were subjected to mild filtration to exclude those failing to genotype in more than 10% of all sequenced samples as well as those with MAF of less than 1% . Following this , we used PLINKv1 . 9 to perform a case/control association analysis ( S3 Table ) . Statistical probabilities were adjusted for genomic inflation using the Genomic Control ( GC ) approach GC correction is based on the assumption that most of variants are not associated with the trait of interest and thus the chi-square values of statistical tests should have a mean of one . Genomic inflation increases this mean and to correct for this , all test statistics values are divided by the mean of the test statistics to restore the expected distribution [77] . Bonferroni correction for multiple testing was performed using pruned variants as an index for independently tested haplotypes to obtain a list of candidate loci [78] . Pruning was done as described above for IBS calculations but after excluding the offspring of the 6 trios . All variants belonging to the same haplotype are dependent and should have similar association probabilities . Correction of multiple testing should be applied to the “independent” statistical trials . In our experiment , correcting for all tested variants did not prevent the detection of the causative variant ( p = 4 . 37 X 10−37 uncorrected ) even with a threshold of 0 . 01 after Bonferroni correction . Fixed variants in affected breeds that were approaching absence from unaffected breeds were selected as those with more than 90% allelic differences between cases and controls . The Variant Effect Predictor ( VEPv85 ) tool from Ensembl [79] was used to annotate possible effects of all detected variants . Variant annotation was done using the NCBI dog genome annotation ( last modified on 9/18/15 ) [80] . Primers were designed using Primer3 [81] . Primers to amplify the DVL2 mutation produced a 297 base pair product ( Forward Primer: CGGCTAGCTGTCAGTTCTGG; Reverse Primer: CAGTGAGTCTGAGCCCTCCA ) . PCR products were sequenced using the Big Dye termination kit on an ABI 3100 Genetic Analyzer ( Applied Biosystems , Foster City , CA ) . Segregation analysis was evaluated by Fisher’s exact test . Sequences were evaluated using Chromas ( Technelysium , South Brisbane , QLD , Australia ) . The sequences were aligned to the Boxer dog reference sequence ( CanFam 3 . 1 ) using BLAT ( UCSC Genome Browser ) . Primers described by Marchant et al . [29] were used to evaluate the SMOC2 mutation status for 152 dogs . PCR products sizes were visualized via gel electrophoresis . All primers were designed using Primer 3 ( Forward Primer: CCACGAGCTGTCATCCTACA; Reverse Primer: CAACTGACAGGGCAGACAGA ) [81] . RNA was isolated from skeletal muscle and spleen using Qiagen QIAamp Blood Mini Kit tissue protocols ( QIAGEN , Valencia , CA ) . RNA was reverse transcribed into cDNA using Qiagen QuantiTect Reverse Transcription Kit . DVL2 and RPS5 [82] cDNA was PCR amplified from skeletal muscle tissue from one Bulldog and one Labrador Retriever . RPS5 was amplified in skeletal muscle to ensure equivalent amounts of cDNA were produced . The PCR products were sequenced on an ABI 3500 Genetic Analyzer and analyzed using Chromas ( Technelysium , South Brisbane , QLD , Australia ) . The sequences were aligned to the reference Boxer dog genome ( Can Fam 3 . 1 ) , using BLAT ( UCSC Genome Browser ) , to confirm sequences matched DVL2 . Semiquantitative RT PCR was performed by PCR amplification ( Forward Primer: CGAGCTGTCATCCTACACCT , Reverse primer: TGACGAGCCTCTGGAAGG ) of cDNA from Spleen from two different dogs of each genotype ( homozygous mutant and wildtype for DVL2c . 2044delC ) . Reduced PCR cycle number ( to 28 ) were used to estimate the transcript differences . PCR products were visualized on agarose gels . Due to extremely high GC-content , attempts to clone the canine DVL2 open reading frame ( ORF ) were unsuccessful . We therefore commercially synthesized the wild type and the bulldog variant of DVL2 and subcloned them into a modified pENTR-2B vector containing an N-terminal Myc tag ( MT ) using the FseI and AscI restriction sites . The ORFs were verified by Sanger sequencing . To generate lentiviral transfer vectors , the pENTR-2B-MT constructs were recombined with the pLEX_307 vector ( a gift from David Root; Addgene plasmid #41392 ) using LR clonase ( Thermo Fisher Scientific , Hanover Park , IL ) . Transgene expression from pLEX_307 is driven by the EF1 promoter . Lentiviruses were generated in HEK293T cells via co-transfection of lentiviral vectors with the following third generation packaging plasmids: pMD2 . G ( Addgene plasmid # 12259 ) , pRSV-rev ( Addgene plasmid #: 12253 ) and pMDLg/pRRE ( Addgene plasmid #: 12251 ) [83] . 0 . 75mL of viral supernatant was used to infect NIH/3T3 cells plated at 20% confluency in 24-well plates . Puromycin selection ( 0 . 002mg/mL ) was carried out for 4 days . To block synthesis and production of endogenous Wnt proteins , the porcupine inhibitor Wnt-C59 ( 100nM final concentration ) was added to cells 24 hours prior to lysis . For WNT stimulation , cells were incubated for 6 hours with Wnt5a ( R&D , catalog #645WN010 , 200ng/mL final concentration ) or Wnt3a ( R&D , catalog #1324-WN-002 100ng/mL final concentration ) . For casein kinase 1 inhibitor treatment , cells were pre-treated with D4476 ( APEXBIO catalog #A3342 , 100nM final concentration ) for 1h prior to WNT5A or WNT3A treatment . D4476 was maintained in the culture during the 6-hr WNT stimulation period . Cells were washed in 1X cold PBS and lysed in 200μL RIPA buffer supplemented with Halt™ Protease Inhibitor Cocktail ( 100X ) ( Prod # 186127 , Thermo Fisher Scientific ) . BCA analyses were conducted to determine the absolute concentrations of protein in lysate samples . For phosphatase treatment , 17 . 3μg of protein from cell lysates were treated with 7 U of CIP ( NEB , catalog #M0290S; final concentration of 350U/mL ) at 37C for 30 minutes . Lysates subjected to mock treatment were incubated at the same temperature and duration without enzyme . Protein concentrations were normalized and lysates were mixed with 1/3 the volume of 4X LDS sample buffer ( NP0008 , Thermo Fisher Scientific ) supplemented with 2-mercaptoethanol ( 4 . 25% final concentration ) . Lysates were heated at 95°C for 5 minutes before SDS-PAGE and western blots were generated . For detecting exogenous DVL2 , a commercially purchased monoclonal anti-c-Myc antibody ( clone 9E10 , Thermo Fisher Scientific , catalog #9801 ) was used as the primary antibody at a dilution ratio of 1/1000 , and a goat anti-rabbit IgG polyclonal antibody ( conjugated to IRDye 800CW; catalog # 926–32211 , Li-cor Biosciences , Lincoln , NE ) was used as the secondary antibody at a dilution ratio of 1/30 , 000 . For detecting α-tubulin , the DM1A mouse monoclonal antibody ( catalog # SC-32292 , Santa Cruz Biotechnology , Dallas , TX ) was used as the primary antibody at a dilution ratio of 1/1000 , and a goat anti-mouse IgG polyclonal antibody ( conjugated to the IRDye 800CW; catalog # 926–32210 , Li-cor Biosciences ) was used at a dilution ratio of 1/30 , 000 ) . Imaging of the western blots was performed using the Odyssey infrared imaging system ( Li-cor Biosciences ) according to the manufacturer’s instructions . Non-saturated protein bands were quantified using Odyssey software , with a gamma level of 1 . | Some dog breeds are characterized by extreme morphological differences from their ancestor , the wolf . One group of three breeds ( Bulldog , French Bulldog and Boston Terrier ) is characterized by a wide head , short muzzle , widely spaced eyes , small size and abnormalities of the vertebral bones of the back and tail . These breeds are referred to as the screw tail breeds since the characteristic that is unique and easy to see in these breeds is their shortened and kinked tails . These breed have become increasingly popular as pet dogs , although they have health issues associated with their morphology . We analyzed the genome sequences of 100 dogs , including 10 screw tail dogs , and identified all the genetic differences between those dogs . We then compared these differences to identify changes in the DNA sequences associated with screw tail . The mutation and the affected gene identified are very similar to the types of mutations that have been shown to be responsible for a rare human disorder with similar clinical abnormalities , called Robinow syndrome . We demonstrate that the dog mutation makes an altered protein that affects an important cell-cell communication system crucial for tissue development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"phosphorylation",
"animal",
"types",
"medicine",
"and",
"health",
"sciences",
"spine",
"vertebrates",
"pets",
"and",
"companion",
"animals",
"animals",
"mammals",
"dogs",
"mutation",
"vertebrae",
"animal",
"anatomy",
"skeleton",
"mammalian",
"genomics",
"frameshift",
... | 2018 | Whole genome variant association across 100 dogs identifies a frame shift mutation in DISHEVELLED 2 which contributes to Robinow-like syndrome in Bulldogs and related screw tail dog breeds |
Endogenous retroviruses ( ERVs ) arise from retroviruses chromosomally integrated in the host germline . ERVs are common in vertebrate genomes and provide a valuable fossil record of past retroviral infections to investigate the biology and evolution of retroviruses over a deep time scale , including cross-species transmission events . Here we took advantage of a catalog of ERVs we recently produced for the bat Myotis lucifugus to seek evidence for infiltration of these retroviruses in other mammalian species ( >100 ) currently represented in the genome sequence database . We provide multiple lines of evidence for the cross-ordinal transmission of a gammaretrovirus endogenized independently in the lineages of vespertilionid bats , felid cats and pangolin ~13–25 million years ago . Following its initial introduction , the ERV amplified extensively in parallel in both bat and cat lineages , generating hundreds of species-specific insertions throughout evolution . However , despite being derived from the same viral species , phylogenetic and selection analyses suggest that the ERV experienced different amplification dynamics in the two mammalian lineages . In the cat lineage , the ERV appears to have expanded primarily by retrotransposition of a single proviral progenitor that lost infectious capacity shortly after endogenization . In the bat lineage , the ERV followed a more complex path of germline invasion characterized by both retrotransposition and multiple infection events . The results also suggest that some of the bat ERVs have maintained infectious capacity for extended period of time and may be still infectious today . This study provides one of the most rigorously documented cases of cross-ordinal transmission of a mammalian retrovirus . It also illustrates how the same retrovirus species has transitioned multiple times from an infectious pathogen to a genomic parasite ( i . e . retrotransposon ) , yet experiencing different invasion dynamics in different mammalian hosts .
Viral cross-species transmission ( CST ) represents a major threat to both human and animal populations . Most viral diseases of humans are zoonotic: they stem from CST of viruses from domestic or wild animals [1] . The explosion and development of human society , including modern transportation , over the last 100 years has exposed us to an increasing number of pathogens [2] . AIDS , which has caused more than 25 million deaths over the past ~30 years ( aids . gov ) , is one of the most notorious examples of a pandemic initiated by viral CST [3 , 4] . The pathogens causing AIDS ( HIV-1 and HIV-2 ) are retroviruses , a family of RNA viruses that use reverse transcription to replicate their genome [5] . Other retroviral CST events have been documented within primates , felids and ruminants , suggesting that retroviral CST represents a continuous threat to human and animal health [6–10] . Retroviruses are unique amongst animal viruses in that chromosomal integration of so-called proviruses is an obligatory step in their replication cycle [5] . As a consequence , retroviral infection of germ cells or their progenitors result in proviruses that may be vertically inherited along with the host genome . Such inheritable proviruses are called endogenous retroviruses ( ERVs ) . Under some circumstances , which are still poorly understood , ERVs can further propagate within the genome and spread in the population , resulting in the formation of large families of interspersed repeats in the host genome [11] . Despite the potentially deleterious consequences associated with the genomic propagation of ERVs , the process has been remarkably pervasive during mammalian evolution . Indeed every mammalian genome thus far examined harbor a great abundance and diversity of ERVs , which are mostly lineage-specific . For example , 8% of the human genome is composed of ERV sequences derived from a wide variety of retroviruses acquired at different time points during primate evolution [12–14] . Once integrated and endogenized , most ERVs appear to evolve at the host’s neutral mutation rate , which is much slower than the mutation rate of exogenous retroviruses ( XRVs ) [15] . Therefore ERVs provide a valuable fossil record of past retroviral infections and a unique opportunity to investigate retroviral evolution at a deep time scale , including CST events [16–20] . Many ancient CST events have been inferred by comparing ERV sequences across species [21–28] . Most of the well-documented cases of retroviral CST events involve closely related host species ( e . g . from the same order ) . Indeed , it is thought that viral CST is often constrained by the evolutionary distance between donor and recipient species [19 , 20 , 29] . The observation that all retroviruses known to infect humans have been acquired from other primates is consistent with this notion [7] . However , retroviral CST events can also occur between distantly related species . For example , the cat RD114 gammaretrovirus is a recombinant containing an envelope domain mostly closely related to Baboon endogenous virus ( BaEV ) , and is thought have been acquired by the domestic cat from an Old World monkey [30 , 31] . Also , the koala retrovirus ( KoRV ) , which is currently spreading and undergoing endogenization in the wild , is very closely related to gibbon ape leukemia virus ( GALV ) and to ERVs found in Asian rodents , from which it was most likely acquired [32] . It has also been reported that reticuloendotheliosis virus ( REV ) was likely transmitted from mammals to birds [10] . Recent phylogenomics surveys of ERVs across a wide variety of vertebrate species suggested that CST between widely diverged species ( i . e . from different orders or classes ) may be more common than initially anticipated [19 , 20 , 33 , 34] . However , the evidence remains limited and more detailed case studies are needed to confirm this idea . Bats ( order Chiroptera ) are increasingly regarded as exceptionally potent reservoirs of zoonotic viruses [35–40] . Indeed , a variety of bat species have been implicated in the spillover of diverse and highly pathogenic RNA viruses such as Rabies , Nipah , Hendra , SARS , Marburg , and Ebola viruses in the human population [41] . Very recently , one potential case of CST of an endogenous betaretrovirus involving phyllostomid bats , rodents and New World monkeys was reported [28] . We previously produced a comprehensive catalog of ERVs in the vespertilionid bat Myotis lucifugus [42] ( referred to as MLERVs hereafter ) , documenting a rich and recent history of retroviral infections in this species lineage . Here , we have taken advantage of this resource to seek evidence of CST events implicating MLERVs . We identified an intriguing case of a gammaretrovirus that colonized independently the genomes of vespertilionid bats , felids and pangolin but followed a different fate and amplification dynamics in these lineages .
To detect possible CST events involving M . lucifugus ERVs , we used the sequence of the reverse transcriptase domain ( RVT_1 ) ( 642 nt ) from members of each of the 86 MLERV subfamilies previously identified [42] as queries in megaBLAST searches of all mammal genomes deposited in the NCBI whole genome shotgun ( WGS ) database as of February 2015 ( 107 mammal species ) . Excluding hits to M . lucifugus , the most significant hits ( >80% nucleotide identity over the entire domain; e-value < 10−80 ) were obtained with a query representing the MLERV1 family [42] against the genome assemblies of the domestic cat ( Felis catus ) [43] , Amur tiger ( Panthera tigris ) [44] and Chinese pangolin ( Manis pentadactyla ) . In addition , and less surprisingly , many highly significant hits to MLERV1 were also obtained in the genomes of vespertilionid bat species closely related to M . lucifugus ( Brandt’s myotis , Myotis brandtii [45]; David’s bat , Myotis davidii [46]; big brown bat , Eptesicus fuscus ) . Further examination revealed that the hits in the feline genomes corresponded to an endogenous gammaretrovirus family initially described in the domestic cat . Two proviruses of this family were initially documented in cat as FERVmlu1 and FERVmlu2 [47] . In 2011 , this ERV family was also reported in Repbase [48] as ERV1-1_Fca . In a more recent and more systematic inventory of ERVs in the cat genome [49] , this family was designated as FcERV_γ6 , a nomenclature we will adopt hereafter . Most recently , this family was identified as part of “lineage VII” by Mata et al . [34] who also reported the presence of closely related gammaretroviral elements in several wildcat species , including jaguar , puma , jaguarundi and tiger . To our knowledge , the related elements in the pangolin have not been previously characterized elsewhere . Hereafter we refer to this novel ERV family as MPERV1 for Manis pentadactyla ERV1 and deposited its consensus sequence in Repbase . For simplicity , we refer to all the elements detected in vespertilionid bats as MLERV1 and all the elements in different felids as FcERV_γ6 . To determine the ERV copy number in each species , we used the LTR sequences to mask their corresponding genome assembly using the Repeatmasker program and parsed the positional output to infer the number of putative full-length proviruses ( i . e . containing two LTRs ) and solitary ( solo ) LTR ( see Methods ) . The results of this analysis ( Table 1 ) show that each species harbors a relatively small number of full-length proviruses ( 2–50 ) but often numerous solo LTRs ( up to 1600+ in M . lucifugus ) . It should be noted that the vast majority of proviruses we inferred to be full-length ( based on the occurrence of a pair of LTRs within 10 kb ) contain sequencing/assembly gaps . Thus we cannot ascertain whether they contain all the coding domains of a complete provirus . To illustrate the exceptional level of sequence similarity among MLERV1 , FcERV_γ6 and MPERV1 , we generated nucleotide pairwise alignments of FcERV_γ6 and MLERV1 and of FcERV_γ6 and MPERV1 using the most closely related full-length proviruses from each family and performed a sliding window analysis of nucleotide identity across the two pairwise alignments ( Fig 1A ) . As a comparison , we performed the same analysis for proviral sequences representative of HIV-1 ( Group M subtype B ) and its closest relative from the chimpanzee SIVcpz [50 , 51] . The results show that the two representatives of the MLERV1 and FcERV_γ6 families and two representatives of FcERV_γ6 and MPERV1 are highly similar throughout their entire length , with an average level of nucleotide identity ( ~85% ) comparable to that between HIV-1 and SIVcpz ( Fig 1A ) . The most divergent segment corresponds to the predicted surface ( SU ) domain of the envelope protein ( ~50% identity in the N-terminal region ) . Elevated divergence in the SU region is also apparent between the two lentiviruses , as previously documented [52] , and is thought to reflect the rapid adaptation of retroviral envelope to diverged host cell receptors [53] . In summary , MLERV1 , FcERV_γ6 and MPERV1 are just as closely related to each other as HIV-1 and SIVcpz , and thus these three elements and their relatives in the bat , cat and pangolin genomes can be considered as endogenous elements descended from the same retrovirus . The overall level of nucleotide similarity between MLERV1 , FcERV_γ6 and MPERV1 is strongly incongruent with a scenario of vertical inheritance of an ancestral ERV present in the common ancestor of chiropterans , felids and pangolins , which dates back to ~85 million year ago ( MYA ) [54 , 55] . Furthermore , we could not find any close relative of MLERV1 or FcERV_γ6 ( no megaBLAST hit with sequence identity >80% ) in the genome assemblies of species representative of other chiropteran ( e . g . flying fox , Pteropodidae ) or carnivore families ( e . g . dog , Canidae; bear , Ursidae; ferret , Mustelidae; seal , Phocidae; walrus , Odobenidae ) . The Chinese pangolin genome is the only available representative of the order Pholidota , which is considered sister to Carnivora , and thus equally related to Perissodactyla ( horse , rhino ) and Cetartiodactyla ( cow , pig , hippo , whales ) , all of which appear to lack related ERVs ( Fig 1B ) . Thus , the taxonomic distribution of MLERV1/FcERV_γ6 elements is extremely patchy , being detected in four vespertilionid bats , two feline species ( cat and tiger ) , and one pangolin , but not in any of the numerous phylogenetically intermediate species represented in the NCBI WGS database ( Fig 1B ) . This taxonomic distribution suggests that the retrovirus that gave rise to MLERV1 , FcERV_γ6 and MPERV1 underwent at least two CST events and was endogenized at least 3 times independently in the vespertilionid , felid , and pangolin lineages . To gain further insights into the evolutionary history of these ERVs , we next sought to estimate when they first infiltrated their host genomes . Given that the likelihood of the same endogenous retrovirus to integrate at the same exact genomic location independently in different lineages is negligible , the presence of an element at orthologous position in different species can be interpreted as having inserted prior to their divergence time [56 , 57] . Conversely , since ERVs are not known to excise from the genome , the absence of an element in one species at a genomic location occupied by an ERV in another species strongly suggests that the ERV integrated after the split of the two species [58 , 59] . Such ‘empty’ sites can be corroborated by the presence of a single copy of the host target sequence duplicated upon proviral integration ( typically 4-bp target site duplication for gammaretroviruses ) . This cross-species presence/absence approach has been widely applied to date a variety of mobile element insertions , including ERVs [14 , 42 , 59 , 60] . It is possible that the age of some integration events may be underestimated because of incomplete lineage sorting . Therefore , orthologous insertion analysis should be interpreted with caution when applied to rapidly radiating species such as the three Myotis considered here . We first examined the sharing of FcERV_γ6 elements between the cat and tiger , which diverged ~10 . 8 MYA [61] . Out of a total of 1 , 419 putative full length proviruses and solo LTRs detected in the current whole genome assemblies of the two species , we were able to ascertain that 256 occupy orthologous positions , while 261 and 201 are specific to the cat and tiger lineages , respectively . None of these elements were detectable in other available carnivore genome assemblies ( e . g . dog , panda , ferret , seal ) , while some of their flanking host sequences were readily detected ( e . g . , the flanking sequence of FcERV_γ6–68 is found in dog chromosome 14 ) . These data indicate that FcERV_γ6 first invaded a felid ancestor sometime between ~10 . 8 million years ( MY ) and ~55 MYA and has continued to amplify to generate many insertions specific to the cat and tiger lineages ( Fig 2 ) . Our previous phylogenetic analysis [42] has shown that the MLERV1 family of the little brown bat M . lucifugus can be divided into 3 subfamilies . Here we performed a systematic analysis of the presence/absence of MLERV1 elements ( including solo LTRs ) from the 3 subfamilies in the genome assemblies of three other vespertilionid bats currently available: Brandt’s myotis ( Myotis brandtii ) , David’s bat ( Myotis davidii ) and the big brown bat ( Eptesicus fuscus ) , which have been estimated to diverge from M . lucifugus ~10 MYA , ~13 MYA and ~25 MYA , respectively [62–65] . The vast majority of MLERV1 elements and their close relatives were found to be species-specific ( Fig 2 ) . Only 3 elements were present at orthologus loci across the 3 Myotis genomes ( Fig 2 ) and we could not find a single insertion shared between E . fuscus and any of the Myotis . Another interesting observation is that members of the MLERV1_3 subfamily , which contributes the vast majority ( >80% ) of MLERV1 elements in the 3 Myotis genomes , could not be identified at all in the E . fuscus genome . Indeed , all 29 elements detected in E . fuscus cluster with either one of the other two subfamilies ( S6 Fig ) . Together these data suggest that the MLERV1 family expanded independently in the Myotis and Eptesicus lineages , but achieved a much higher copy number in the Myotis lineage due to the amplification of the MLERV1_3 subfamily , which has generated numerous species-specific insertions ( Fig 2 ) . Another widely applied method to date retroviral and other LTR-bearing retroelement insertions relies on the divergence of the 5’ and 3’ LTR of individual elements . This is because their retrotransposition mechanism results in two identical LTRs at the time of chromosomal integration . Given that most ERV LTR sequences are assumed to evolve neutrally once integrated in the host chromosome , the age of a provirus can be estimated based on LTR divergence by applying the host neutral substitution rate [49 , 59 , 66 , 67] . To eliminate the inflated divergence caused by hypermutable methylated CpG sites [68] , we excluded all the CpG sites from our calculation of LTR-LTR divergence . We applied this method to calculate the age of all complete ( i . e . with two LTR ) proviruses detected in cat , M . lucifugus and pangolin genome assemblies . We use previously estimated neutral substitution rates of 2 . 7×10−9 and 1 . 8×10−9 per year for vespertilionid bats and felids respectively [44 , 69] , and an “average” mammal neutral substitution rate of 2 . 2×10−9 per year [70] for the pangolin . The results of these calculations predict that the oldest MLERV1 and FcERV_γ6 proviruses would be ~10 MY and ~20 MY respectively ( Fig 3A ) . The amplification of the bat MLERV1 family would have peaked sharply in the last 2 MY , while the cat FcERV_γ6 elements inserted more continuously over the past ~15 MY ( Fig 3A ) . The two MPERV1 proviruses identified in the pangolin genome are estimated to be ~10 and ~18 MY based on this approach . While these estimates are consistent with independent ERV invasions of the vespertilionid , felid and pangolin lineages , we noticed that the age of individual insertions based on LTR divergence were generally lower than those estimated based on their presence/absence at orthologous position across species . For instance , we found that 27 FcERV_γ6 proviruses were orthologous in cat and tiger , which indicates that all must have inserted prior to speciation of these felids , which has been robustly estimated at ~10 . 8 ( 8 . 4–14 . 5 ) MY [61] . However only 13 of these 27 insertions were estimated to be older than 10 MY based on LTR divergence ( S1 Table ) . Similarly , M . lucifugus and M . brandtii are thought to have diverged ~10 MYA [62 , 64] , but the age of the four MLERV1 insertions orthologous between these two species was estimated to be 10 . 5 , 6 . 8 , 4 . 5 and 1 . 2 MY based on LTR divergence ( S1 Table ) . One possible explanation for these discrepancies between the two dating methods is the phenomenon of gene conversion between two LTRs adjacent in the genome , which essentially erases some of the divergence accumulated over time through point mutations occurring in each of the LTRs , causing to underestimate the date of proviral insertion [71 , 72] . Indeed , a phylogenetic analysis of the LTR sequences from the four MLERV1 proviruses orthologous in M . lucifugus and M . brandtii , shows topologies consistent with LTR homogenization through gene conversion events for at least two of the proviruses examined ( corresponding to the two upper trees in Fig 3B ) : their 5’ and 3’ LTR cluster together rather than by species ( Fig 3B ) . This is the topology predicted if conversion events in one or both of the species lineages had removed nucleotide divergence accumulated between 5’ and 3’ LTR prior to speciation [72] . Thus , estimates of the age of proviruses based on LTR divergence should be interpreted with caution , as they are likely to be underestimates . Nonetheless , the results are in agreement with the other lines of evidence that the vespertilionid , felid , and pangolin lineages were independently infiltrated by the same ERV during an evolutionary timeframe ranging from ~25 to ~13 MYA . To further characterize the evolution history of MLERV1 , FcERV_γ6 and MPERV1 , we examined the phylogenetic relationship of elements within these families using a maximum-likelihood tree built from an alignment of their 3’ LTR sequences . We used only ‘complete’ proviruses ( 30 in bats , 43 in cats and 2 in pangolin ) , since we observed that including tiger provirus and solo LTRs did not yield any new major clade in the phylogeny ( S1 Fig ) . Also the general topology of the tree is identical if the 5’ LTR sequences are included ( S2 Fig ) . Trees generated using internal coding sequences also displayed the same general topology ( S3 Fig ) , but offered less phylogenetic resolution due to the more constrained nature of retroviral coding sequences relative to LTRs [73 , 74] . The unrooted tree resulting from the phylogenetic analysis ( Fig 4 ) clearly shows that FcERV_γ6 and MPERV1 elements are more closely related to each other than to the bat MLERV1 elements . Another striking observation is that elements within the FcERV_γ6 family fall within a single clade with uniformly short branches , whereas the MLERV1 elements , as we previously reported [42] can be divided into 3 distinct subfamilies separated by long branches , with MLERV1_2 and MLERV1_3 being closer to each other and more distant from FcERV_γ6 than MLERV1_1 ( Fig 4 ) . These data are consistent with a scenario whereby the FcERV_γ6 family was amplified from a single infectious progenitor , while MLERV1 elements might have originated from at least three distinct infectious progenitors . To further explore the history of the FcERV_γ6 and MLERV1 families , we next turn to an analysis of selection regimes that have acted on their coding sequences during their amplification . Such analysis can help discern whether ERVs have spread primarily through reinfection or retrotransposition events because the latter mechanism , which is strictly intracellular , is predicted to be associated with the loss of envelope function . Indeed , the envelope protein binds to host cell membrane receptor to promote virion entry in the host cell and therefore is required for most retroviral infection [5] . Thus , proviruses that originate from infection events should show evidence of functional constraint on envelope domains [75 , 76] , Magiorkinis:2012gy} . To perform this analysis , we used all MLERV1 ( n = 30 ) and cat FcERV_γ6 ( n = 43 ) proviruses with complete ( or nearly complete ) coding capacity . Given that only 2 proviral MPERV1 copies could be identified , we did not perform selection analysis for this family . To evaluate how natural selection may have constrained the different coding regions of MLERV1 and FcERV_γ6 , we computed the dN/dS ratio ( ω ) applying the branch model implemented in PAML , where dN denotes the non-synonymous substitution rate and dS denotes the synonymous substitution rate , along the branches of the phylogeny of FcERV_γ6 and MLERV1 elements for each of their predicted coding domains [77] . ω values significantly smaller than 1 are indicative of purifying selection acting to maintain a functional protein sequence , while ω values not significantly different from 1 are indicative of neutral evolution or relaxed functional constraint . To test for significant deviation of ω from 1 , we apply a likelihood ratio test [78] . Within the FcERV_γ6 family , the analysis reveals that purifying selection has acted on all coding domains ( ω value ranging from ~0 . 6 to ~0 . 9 , p < 0 . 05 ) , with the notable exception of Gag matrix and envelope domains ( Fig 5A ) . The ω value is not significantly different from 1 ( neutral evolution ) for the Gag matrix domain ( Gag_MA ) . Besides , all but nine of the 43 FcERV_γ6 proviruses lack an envelope domain ( TLV_coat ) . The nine copies that have retained a recognizable envelope domain occupy basal branches in the phylogeny ( Fig 4 ) and have orthologs in the tiger genome suggesting that they predate the envelope-less copies ( S1 Table ) . Furthermore , 29 of the 43 FcERV_γ6 proviruses examined , including all cat-specific copies , share the same deletion breakpoint removing most of envelope gene ( Fig 5B ) . These data suggest that FcERV_γ6 copies potentially coding an envelope were inserted prior to the speciation of cat and tiger ( ~10 . 8 MYA ) , while copies integrated more recently lacked the envelope domain . In addition , the envelope open reading frames of these nine ancient FcERV_γ6 elements accumulated multiple indels or missense substitutions . Thus , none of the FcERV_γ6 elements in the cat genome appear to have retained a functional envelope domain . These data suggest that FcERV_γ6 rapidly lost its infectious capacity in the cat lineage but has continued to amplify primarily via retrotransposition amplified primarily via retrotransposition . By contrast , selection analysis suggests that the MLERV1 family has experienced a more complex amplification history . We focused our analysis on the MLERV1_2 and MLERV1_3 subfamilies because they are the two best-supported monophyletic subfamilies with sufficient number of proviruses to draw solid conclusions . First , we observe that generally the signature of purifying selection is more pronounced on the bat elements than on the cat elements , as indicated by much lower ω values ( Fig 5A ) . The only exception is the Gag matrix domain of the MLERV1_3 subfamily , which exhibits relatively higher ω value ( ω = 0 . 67 , p = 0 . 02 ) ( Fig 5A ) . In addition , all MLERV1_3 elements appear to have lost their envelope domain through the same deletion event ( Fig 4 ) . This pattern contrasts with elements within the MLERV1_2 subfamily , for which all coding domains , including envelope , have evolved under strong purifying selection during the spread of these elements ( ω from 0 . 15 to 0 . 27 , p<0 . 001 ) ( Fig 5A ) . These data suggest a scenario whereby MLERV1_3 has amplified primarily by retrotransposition , while the spread of MLERV1_2 has been driven by multiple infection events . We also observe that in both FcERV_γ6 and MLERV1_3 , the losses of envelope coincided with the elevation of the dN/dS ratio in their Gag matrix domain ( Fig 5A ) . To evaluate whether this reflects a loss of function ( neutral evolution ) or a relaxation of purifying selection , we further examined the integrity of open reading frames ( ORFs ) in each ERV family by computing the frequency of stop codons and frameshift mutations occurring in each of the domains ( see Methods ) . Overall the results indicate that the coding integrity of the Gag matrix domains of FcERV_γ6 and MLERV1_3 elements is not significantly different from that of the other ERV subfamilies or that of the other protein domains ( S4 Fig ) . These results suggest that the Gag matrix domain is not dispensable for retrotransposition , as previously demonstrated functionally for IAP elements [79] , but appears to evolve faster in retrotransposing ERVs .
Until recently , most retroviral CST events that have been documented rigorously have implicated closely related species [6 , 9] , suggesting that the phylogenetic distance between species is an important determinant of the host range of a retrovirus [7 , 29] . Indeed , many previous studies have illustrated how the divergence of host cellular factors that either facilitate or restrict viral replication can modulate the host range of a virus [80–82] . The systematic analysis of retroviruses fossilized in the genome as ERVs is progressively revealing a more nuanced picture whereby some retroviruses appear to have been capable to infect widely diverged species ( i . e . belonging to different orders ) without seemingly much changes occurring in their own sequences . Recent large-scale pylogenomics analyses have suggested that cross-order transmission may actually be fairly common for some groups of retroviruses , including gammaretroviruses [19 , 20 , 34] and IAP betaretroviruses [33] . While these studies disclosed phylogenetic patterns suggestive of multiple CST events , they did not explicitly rule out alternative hypotheses , such as vertical persistence and stochastic loss of the ERV in some lineages , and thus they generally await confirmation through more detail analyses such as the one presented here . Our study provides multiple lines of evidence supporting the notion that a gammaretrovirus infiltrated independently the germline of bat , cat and pangolin species representing three mammalian orders ( Chiroptera , Carnivora , Pholidota , respectively ) . First , elements found in these species display a level of nucleotide sequence similarity ( ~85% ) along their entire length that is comparable to that observed between closely related retroviruses that have undergone very recent CST ( such as SIVcpz and HIV-1 ) . Such a level of sequence similarity between ERVs inhabiting species diverged by ~85 MYA [54] is incompatible with a scenario of vertical descent from an ERV inherited from their common ancestor . The CST hypothesis is also bolstered by the highly discontinuous taxonomic distribution of this particular ERV family . Out of 107 mammal species for which whole genome assemblies are publicly available , we could only detect members of this ERV family in vespertilionid bats , felids and pangolin , but not in several species representing related mammal families ( 6 additional Chiroptera species from 4 families and 7 additional Carnivora species from 5 families ) . Thus , a scenario evoking a single introduction of this ERV family in the common ancestor of bats , cats and pangolins followed by vertical inheritance would necessitate at least 5 independent losses ( Fig 1B ) to account for its current taxonomic distribution . A more parsimonious scenario is that this ERV family was acquired horizontally and independently in each of the three species lineages where it is currently detected . It is also possible that the ancestral retrovirus infected other species but failed to endogenize in their genomes , or it could also be that additional species lineages hosted this ERV family but have gone extinct . Finally , our estimation of the dates at which these elements first entered their host genomes , which relies on two independent approaches ( cross-species comparison of orthologous ERV loci and LTR-LTR divergence ) , converges to a bracket of 13 to 25 MYA , which far postdates the divergence of their host species ( ~85 MYA ) . Together these data indicate that a progenitor gammaretrovirus infiltrated the germlines of ancestral vespertilionid bat , felid and pangolin species . It is conceivable that this retrovirus could have transferred directly between these ancestral species because their geographic distribution likely overlapped in Eurasia during the estimated period of initial ERV infiltration ( ~13–25 MYA ) [61 , 62 , 83] . Given that cats are known to prey on both bats [84–87] and pangolins [88] , a direct transfer from bat or pangolin to cat is plausible . Indeed , predation has been put forward as the most likely explanation for the spillover of bat lyssaviruses ( rabies ) into domestic cats [89] . On the other end , both bats and pangolins are capable of surviving a cat attack , which makes the transfer from predator to prey conceivable as well . Nonetheless , multiple lines of evidence indicate that MLERV1 colonized these bat genomes more recently ( Figs 2 and 3 ) , which may suggest a CST from cat to bat . Furthermore , we cannot rule out that one or several intermediate hosts were involved in the introduction of the retrovirus in these species . Our data suggest that , shortly after infiltration of the felid genome , FcERV_γ6 lost the capacity to infect cells and transformed into a retrotransposon . Envelope domain remnants are only found in the basal branches in their phylogeny , and all the FcERV_γ6 elements amplified in the domestic cat lineage clearly derive from a progenitor that lacked coding capacity for a functional envelope protein ( Figs 4 and 5B ) . Together these data suggest that FcERV_γ6 lost its infectious capability soon after it became endogenous , but continued to propagate by retrotransposition , much like the IAP elements in the mouse genome [90 , 91] . Coincided with the envelope loss , we found gag matrix domain evolves at a relaxed rate in FcERV_γ6 family . A recent study showed that a FcERV_γ6 insertion in the KIT gene currently segregating in domestic cats is responsible for the “Dominant White” and white spotting pigmentation phenotypes [92] , which supports our findings that some FcERV_γ6 insertion activity is very recent and likely ongoing . Interestingly , the FcERV_γ6 element inserted at the KIT locus lacks envelope domain and clusters with other recently active FcERV_γ6 copies in our phylogenetic analysis ( Fig 4 ) . Collectively these data suggest that FcERV_γ6 has morphed into a successful retrotransposon that may still be active in the domestic cat . In contrast to FcERV_γ6 , the sequence diversity and phylogenetic structure of MLERV1 elements in the vesper bat genomes are indicative of a more complex amplification history characterized by a mixture of retrotransposition and reinfection events . Our phylogenetic analysis delineates at least three highly diverged MLERV1 subfamilies . The separation between three subfamilies suggested that this family stemmed from at least three related infectious progenitors independently , which is conceivable considering the independent introduction of multiple HIV-1 strains in human population [93] . Furthermore , selection analyses suggest that different subfamilies have adopted different evolutionary trajectories . The MLERV1_2 subfamily is characterized by a signature of intense purifying selection acting on all coding regions throughout the whole clade ( Fig 5A ) . These data strongly suggest that elements within that subfamily have retained their infectious capacities for extended period of time and most likely spread primarily through reinfection events . It is even possible that MLERV1_2 is still active and infectious: most insertions are very recent ( Fig 2 and S1 Table ) and at least one copy ( MLERV1 . 80 ) contains apparently full-length and intact gag , pol , and env genes . The MLERV1_3 subfamily appears to have followed a different evolutionary path whereby the divergence of the elements was accompanied by a strong signature of purifying selection in all coding regions with the notable exception of the Gag matrix domain which has been evolving faster than other domains ( Fig 5A ) and the envelope domain which was apparently deleted altogether . This selection pattern resembles that of FcERV_γ6 and is indicative of proliferation primarily via retrotransposition as opposed to reinfection . Consistent with this hypothesis and the so-called superspreader model [33] , the MLERV1_3 family has been by far the most successful at spreading during Myotis evolution: it has the highest copy number , including many species-specific insertions ( Table 1 ) . Interestingly , none of the 29 MLERV1 elements identified in the big brown bat E . fuscus belong to the MLERV1_3 subfamily . This is consistent with the idea that the MLERV1_3 subfamily originated after the split of Eptesicus-Myotis split ~25 MYA and amplified during the diversification of the Myotis lineage . At present it remains unclear whether the MLERV1 elements present in E . fuscus and Myotis descend from element ( s ) introduced in their common ancestor or if they result from independent acquisition of the same retrovirus . On the one hand , the observation that both E . fuscus and Myotis harbor elements from two diverged subfamilies may be interpreted as evidence that these subfamilies descend from a single progenitor ERV acquired in the common ancestor of these species . On the other hand , the fact that none of the MLERV1 insertions are shared ( orthologous ) between E . fuscus and any of the 3 Myotis genomes ( Fig 2 ) and that none of the provirus insertions dated in any of these bat species appear older than 13 MY ( considerably less than the estimated divergence between the two genera , 25 MYA ) ( Fig 3 ) supports a scenario of multiple , independent acquisition . This scenario , while requiring at least two CST events , is conceivable because Eptesicus and Myotis bats likely occupied a widely overlapping geographic distribution at the estimated time of MLERV1 invasions [62] and these congeners are currently known to frequently come into contact within the same roost [94 , 95] . Regardless of the origin of MLERV1 , the data summarized above illustrate how the same retrovirus has infiltrated widely diverged mammals and transitioned multiple times ( at least twice: FcERV_γ6 and MLERV1_3 ) from an infectious pathogen to a genomic parasite ( i . e . a retrotransposon ) . The biological factors and sequence of events underlying such transition remain poorly understood . In a seminal study , Ribet et al . showed that the loss of envelope gene combined to the gain of an endoplasmic reticulum targeting signal were apparently sufficient for an infectious progenitor of the mouse IAP elements to turn into a highly active retrotransposon [79] . Magiorkinis et al . [33] have extended this paradigm and proposed that the passive loss of envelope lead ERVs to become “superspreaders” in the genome . Through a study of IAP-like elements across a wide range of species , these authors observed that envelope-less elements generally achieve much higher copy numbers than those maintaining a functional envelope . Our results support this model . First , envelope-less FcERV_γ6 elements have proliferated to high copy numbers in the domestic cat ( n = 832 ) and tiger ( n = 730 ) . In addition , in the bats the only subfamily of MLERV1 elements that has attained similarly high copy number is MLERV1_3 , which conspicuously lack a functional envelope gene ( Fig 5 ) . MLERV1_3 elements have generated many species-specific insertions consistently outnumbering the MLERV1_2 subfamily , which appears to have spread primarily by reinfection ( 659 vs . 14 in M . lucifugus genome , 350 vs . 12 in M . brandtii and 331 vs . 19 in M . davidii ) ( Fig 2 ) . Thus , our study is consistent with the notion that the loss of infectious capacity correlates with ERV expansion by retrotransposition , as proposed previously for the rodent IAP families [33 , 79] . One important difference is that the shift between infection and retrotransposition in the MERV1 family was apparently accompanied by little changes in the sequence of MLERV1 elements . Indeed , members of the MLERV1_2 and MLERV1_3 subfamilies diverge by ~15–20% in their RT domain nucleotide sequences with Kimura correction . By comparison , infecting and retrotransposing IAP subfamilies diverge more substantially ( ~65% in RT domain ) . Thus our findings suggest that the transition between the two modes of ERV amplification can occur relatively fast during ERV evolution . An intriguing finding of this study is that the same or very similar retrovirus was endogenized in three different mammalian hosts , but followed quite different evolutionary trajectories in the three species lineages . In the pangolin lineage , the ERV family failed to amplify ( only 2 detectable copies ) and was essentially ‘dead-on-arrival’ . In the cat lineage , the ERV progenitor apparently lost its infectious capacity shortly after endogenization and subsequently amplified to high copy numbers by retrotransposition through an extended period of time ranging from at least 10 MYA ( 256 insertions orthologous in cat and tiger ) to modern times ( KIT insertion segregating in domestic cats ) . Meanwhile , in the bat lineage , the ERV followed a more complex evolutionary path characterized by multiple episodes of reinfection , and at least one burst of amplification by retrotransposition . These observations beg the question whether the loss of infectious capacity of an ERV and its conversion to a retrotransposon is a purely stochastic process , largely owing to the stochastic mutation of gag matrix and loss of envelope functions , or possibly the characteristics of different proviral ancestors , or if it can be influenced by some biological characteristics of the host species ? For instance , it has been recently reported that the level of endogenous retroviral activity may be partly governed by host body size [96] . The pattern of sustained reinfection of MLERV1 in the bat lineage is particularly intriguing in light of the growing appreciation that bats seem to frequently act as reservoir for viruses otherwise lethal to other mammals [35 , 39] . The reasons for bats’ propensity to support high and diverse loads of viral pathogens are poorly understood , but it is thought that some physiological ( e . g . immunopathological tolerance ) and/or ecological features ( e . g . flight , roosting ) allow these animals to tolerate higher level of viral replication and/or facilitate viral transmission [39 , 97 , 98] . By the same token , it is tempting to speculate that the same properties might predispose bats to support higher level of ERV reinfection compared to other mammals such as cats . However , we only investigated one ERV family in three mammal species here . It is possible that MLERV1 is just a unique ERV family in bat genome , and overall the ERV replication in bats is not significantly different from other mammals . Testing this hypothesis will necessitate a more systematic examination of the amplification dynamics of ERVs in a wide range of mammals to assess whether the tendency toward maintenance of infectious capacity is a general trademark of bats or possibly other groups of mammals .
Nucleotide sequences of all RVT_1 domains of previously identified MLERVs [1 , 42] were used as queries to search the whole genome sequence database from the National Center for Biotechnology Information ( NCBI ) using default MegaBLAST parameters [99] . An 80% similarity over 80% region was used as filter to exclude non-specific hits . Complete MLERV1 and FcERV_γ6 proviruses in the M . lucifugus and cat genomes were collected from previous publications [42 , 49] . To ensure we only considered elements from these families , we only retained elements with 80% nucleotide similarity to another family member , a procedure which resulted in the exclusion of the FcERV_γ6_46 copy from the FcERV_γ6 family ( S3 Fig ) . To identify complete proviruses in other vesper bat genomes , the RVT_1 domain sequence of MLERV1 . 71 in M . lucifugus was used as query in blastn search of the M . brandtii , M . davidii and E . fuscus genome assemblies available in NCBI . In parallel , we applied LTRharvest [100] and LTRdigest [101] as described previously [42] to identify all putative proviruses in each of the three bat genome assemblies . We then used BEDTools to intersect the coordinates of RVT_1 domain blastn hits with that of the candidate proviruses [102] . All the candidate proviruses intersecting with a MLERV1 RVT_1 hit were ‘manually’ inspected to refine their termini and confirm their identity as members of the MLERV1 family . To comprehensively retrieve all proviruses and solo LTRs related to the FcERV_γ6/MLERV1/MPERV1 families in each of their respective genomes , we run RepeatMasker [103] with default setting and a custom repeat library with representitives from all MLERV1/MPERV1/FcERV_γ6 subfamilies against each genome assembly . The RepeatMasker output was then parsed using script parse_RMout_count_solo_and_full . pl to produce bed files of all complete solo LTRs and full length ERVs . We define a complete solo LTR as a sequence matching the LTR with missing less than 150 bp at their 5’ termini and missing less than 10 bp at their 3’ termini . We identified elements as putative proviruses those delimited by two LTRs in the same orientation separated by 3 kb to 10 kb of intervening sequence . Manual inspection of a subset of putative proviruses identified by this approach confirmed that most contained typical ERV coding sequences , though frequently interrupted by large sequence/assembly gaps . The LTR libraries and PERL scripts used for these analyses have been deposited on Github ( https://github . com/xzhuo/orthologusLTR . git ) . In the pangolin genome , our initial MEGAblast search yielded only 2 significant hits to the MLERV1 RVT_1 domain , but many more related ERVs could be retrieved using the RT domain from these two initial hits in reiterative blast searches against the pangolin genome assembly . To examine the relationship of these RT elements to each other and to the MLERV1 and FcERV_γ6 families , we conducted a phylogenetic analysis using the Maximum Likelihood package PhyML3 . 1 with the GTR+Γ model [104] . The resulting tree ( S3 Fig ) revealed that only the two initial hits clustered with FcERV_γ6/MLERV1 and were considered part of the MPERV1 family . The other elements form a distinct family we called MPERV2 . MPERV1 and MPERV2 elements share less than 80% nucleotide sequence similarity in their coding regions , but still retain substantial level of sequence similarity in their LTRs . Thus , to correctly estimate the number of solo LTRs for the MPERV1 family , we had to examine their position on a phylogenetic tree of LTRs ( S5 Fig ) . Using this approach , 27 solo LTRs could be assigned to the MPERV1 family in the pangolin genome assembly ( as reported in Table 1 ) . Because MPERV2 was much more distantly related to FcERV_γ6 and MLERV1 ( <80% sequence similarity ) , we did not analyze further MPERV2 in this study . Reference sequences for MPERV1 and MPERV2 have been deposited in Repbase [48] . All identified ERVs are available as bed format ( S1 File ) . To generate the sliding window analysis shown in Fig 1A , we used MUSCLE [105] to align the nucleotide sequence for three pairs of proviruses: MLERV1_77 vs . FcERV_γ6_62; FcERV_γ6_62 v . s MPERV1_ltr106; SIVcpz ( AF115393 ) vs . HIV-1 ( NC_001802 ) and used SEAVIEW [106] to manually adjust each alignment . Each pairwise alignment was then split into 300 bp windows with step size 50bp ( i . e . = 170 segments for the MLERV1_77 vs . FcERV_γ6_62 alignment ) and the percentage of sequence similarity was computed and corrected using kimura 2 parameter model for each window [107] . Orthologous ERV loci were detected similarly as we described previously [42] . Briefly , we used the Perl script extract_flanking_fasta . pl to extract 200 bp at both ends of each query element along with 200 bp of flanking sequences . The output file is then used as query in a batch blastn search against the target genome assembly with default parameters . The csv format blast output is then parsed using orthoblast_finder . pl to pair 5’ end hits with 3’end hits . Finally , the paired hits output was parsed using the script final_annotation . pl to infer the presence/absence of each element in the target genome . All these perl scripts were deposited on Github ( https://github . com/xzhuo/orthologusLTR . git ) . Sequence divergence between 5’ and 3’ LTRs from the same provirus was computed as previously described [42] . To infer insertion dates from LTR divergence of MLERV1 and FcERV_γ6 elements , we used previously estimated lineage-specific neutral substitution rates of 2 . 7 × 10−9 yr-1 [69] and 1 . 8 × 10−9 yr-1 [44] for the vespertilionid and felid lineages respectively . Since no substitution rate has yet been estimated for the pangolin lineage , we used the ‘average’ mammal neutral substitution rate of 2 . 2 × 10−9 yr-1 [70] to infer the age of MPERV1 insertions . The maximum-likelihood phylogenies presented for LTR sequences were built using PhyML3 . 1 and the support for each node is determined with an approximate likelihood ratio test [108] . The multiple alignment of LTR sequences was constructed using MUSCLE and PRANK with default nucleotide parameters and manually adjusted using SEAVIEW [105 , 106 , 109] . Nucleotide substitution model was chosen using AIC criterion in jmodeltest2 . 1 . 6 [110] ( GTR + Γ ) . Dendroscope 3 was used for tree visualization [111] . ERV coding regions were predicted using HMMER3 in all 6 reading frames [112] to delineate Gag_MA ( matrix ) , Gag_p30 ( capsid ) , RVP ( protease ) , RVT_1 ( reverse transcriptase ) , RnaseH , rve ( integrase ) and TLV_coat ( envelope transmembrane domain ) domains in MLERV1 , FcERV_γ6 and MPERV1 proviruses . A multiple codon alignment was generated for each set of coding domains using MUSCLE and manually adjusted with SEAVIEW [106] . The program codeml from the PAML4 . 8 package [77] was used to estimate dN/dS ratio with branch model = 2 . A maximum likelihood phylogeny of the LTR sequences was used as the guide tree in codeml . To test for purifying selection on each coding domain , we calculated the control lnL value by running codeml with ω fixed to 1 . Then likelihood ratio test was performed as suggested by PAML to test if ω is significantly different from 1 [78] . Coding region integrity was assessed by calculating the frequency of stop codon or frameshift indels per codon . We calculated the total length of each domain in every subfamily , and counted the occurrence of stop codon and frameshift indels . The mean frequency and 95% confidence interval is calculated using the Poisson distribution [113] . | The cross-species transmission of viruses poses a continuous threat to public health . Bats are increasingly recognized as a major reservoir for zoonotic RNA viruses , including rabies , Ebola , and possibly MERS , but little is known about their capacity to harbor and transmit retroviruses . Here we investigated past incidents of cross-species transmission involving bat retroviruses , by screening for the presence of endogenous retroviruses ( ERVs ) previously identified in the genome of the little brown bat in more than 100 diverse mammal species . This screen revealed an intriguing case of a gammaretrovirus that independently infiltrated the germ line of species belonging to three mammalian orders: vesper bat , felid cat and pangolin . We found that the ERV initiated its genomic invasion of the three lineages around the same timeframe ~13–25 million years ago , but experienced a different fate in each lineage . In the pangolin lineage , the ERV’s genomic propagation stalled shortly after endogenization , while it amplified continuously throughout felid and vesper bat evolution to generate hundreds of species-specific insertions in each lineage . Furthermore , in the cat lineage genomic amplification appears to have occurred predominantly via retrotransposition; while in bats the ERV has expanded via a mixture of retrotransposition and reinfection activity that may still be ongoing . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Cross-Species Transmission and Differential Fate of an Endogenous Retrovirus in Three Mammal Lineages |
The digenetic trematode Schistosoma mansoni is a human parasite that uses the mollusc Biomphalaria glabrata as intermediate host . Specific S . mansoni strains can infect efficiently only certain B . glabrata strains ( compatible strain ) while others are incompatible . Strain-specific differences in transcription of a conserved family of polymorphic mucins ( SmPoMucs ) in S . mansoni are the principle determinants for this compatibility . In the present study , we investigated the bases of the control of SmPoMuc expression that evolved to evade B . glabrata diversified antigen recognition molecules . We compared the DNA sequences and chromatin structure of SmPoMuc promoters of two S . mansoni strains that are either compatible ( C ) or incompatible ( IC ) with a reference snail host . We reveal that although sequence differences are observed between active promoter regions of SmPoMuc genes , the sequences of the promoters are not diverse and are conserved between IC and C strains , suggesting that genetics alone cannot explain the evolution of compatibility polymorphism . In contrast , promoters carry epigenetic marks that are significantly different between the C and IC strains . Moreover , we show that modifications of the structure of the chromatin of the parasite modify transcription of SmPoMuc in the IC strain compared to the C strain and correlate with the presence of additional combinations of SmPoMuc transcripts only observed in the IC phenotype . Our results indicate that transcription polymorphism of a gene family that is responsible for an important adaptive trait of the parasite is epigenetically encoded . These strain-specific epigenetic marks are heritable , but can change while the underlying genetic information remains stable . This suggests that epigenetic changes may be important for the early steps in the adaptation of pathogens to new hosts , and might be an initial step in adaptive evolution in general .
The interaction of hosts and parasites is one of the best-studied examples of evolution in a changing environment [1] . Their reciprocal antagonistic co-evolution can be illustrated by an arms race in which host and parasite develop mechanisms to circumvent counter-measures developed by their opponents [2] , [3] . Under certain conditions , parasite virulence and host defence can be in equilibrium leading to a phenomenon called compatibility . Compatibility occurs in a host-parasite system when the parasite species is capable of infection and transmission through the host species [4] . The phenomenon that some parasite strains are compatible with certain host strains but not with others ( and vice versa ) is called compatibility polymorphism . This phenomenon was described in the platyhelminth Schistosoma mansoni and its intermediate host , the mollusc Biomphalaria glabrata [5] . S . mansoni is a human parasite whose life cycle is characterised by the passage through two obligatory sequential hosts: the fresh-water snail B . glabrata ( or dependent on the geographical location other Biomphalaria species ) for asexual replication , and humans or rodents as hosts for sexual reproduction [6] . The molecular mechanisms underlying compatibility polymorphism between S . mansoni and B . glabrata were recently investigated by comparing the proteomes of two S . mansoni laboratory strains: one strain that is compatible ( the C strain ) and one that is incompatible ( the IC strain ) with the same reference B . glabrata strain from Brazil [7] . The study identified S . mansoni Polymorphic Mucins ( SmPoMucs ) as key markers for compatibility ( see [4] for a recent review ) . SmPoMuc glycoproteins have a mucin-like structure with an N-terminal domain containing a variable number of tandem repeats ( VNTR ) [8] . SmPoMuc proteins are highly polymorphic [8] and interact with the Fibrinogen RElated Proteins ( FREPs ) of the mollusc [9] . FREPS are diversified antigen recognition molecules playing a central role in the secondary immune response to digenetic trematodes [10] , [11] , [12] . The extraordinary level of SmPoMuc polymorphism is generated by a complex cascade of mechanisms , a “controlled chaos” , acting at the transcriptional , translational and post-translational level [8] . SmPoMucs are encoded by a multigene family with at least 10 members that are organised in 4 clusters on the genome . They recombine frequently and generate new alleles [8] . Each individual miracidium ( the larva that infects the mollusc ) expresses only a specific subset of SmPoMuc genes . The mechanisms controlling this expression polymorphism of SmPoMucs remained unclear . Our recent finding that Trichostatin A , a modifier of chromatin structure , influences SmPoMuc transcription patterns [13] suggests that epigenetic mechanisms participate in transcription control . Epigenetic information is information on the status of gene activity that is heritable , for which changes are reversible and that is not based on the DNA sequence [14] , [15] , [16] . The scientific debate about the reason of the evolution of an epigenetic inheritance system ( EIS ) in most organisms is intense . Others and we have suggested that EIS provides a basis for modifications in the reaction norms that do not require changes of genotypes [17] , [13] , resulting in increased phenotypic plasticity at the individual level or increased phenotypic variability at the population level . If EIS influences the capacity to generate different phenotypes , both the better adapted phenotype and the capacity to generate this phenotype will be selected for and carried into the next generation . This hypothesis has been largely validated in the malaria parasite Plasmodium falciparum which displays “Clonally Variant Gene Expression” ( CVGE ) [18] . Genes that show CVGE are present in multicopy , such that individual parasites within an isogenic population express these genes at very different levels , often fully active or completely silenced . Their transcriptional patterns are clonally transmitted to the next generations through asexual multiplication , and stochastic changes of the transcription level occur at low frequency . This bet hedging strategy allows for stochastic generation of phenotypic diversity and can be controlled by epigenetic based events , similar to those described for the var gene family . The var genes encode the red blood cell surface antigen P . falciparum erythrocyte membrane protein 1 ( PfEMP-1 ) and their “CVGE” regulation strategy is responsible for surface antigen variation that ultimately results in immune evasion . In this context , the EIS that leads to “CVGE” allows for rapid adaptation to the ever-changing vertebrate immune environment . In S . mansoni miracidia , we have shown that epigenetic-based events influence the phenotypic plasticity in populations [13] and particularly regulate SmPoMuc gene expression . To gain further insight into the precise mechanism of regulation of these genes , in the present study we investigated the genetic and epigenetic changes that occurred during the evolution of the phenotypic compatibility polymorphism in two S . mansoni strains . We focused on the sequences of the promoters of active SmPoMuc genes and investigated whether there exist differences in the promoter sequences between S . mansoni compatible and incompatible strains . Our study reveals that IC and C strains display very little within strain genetic variability , and limited nucleotide differences between promoter sequences of the two strains , but show strong chromatin structure differences . These chromatin structures are heritable throughout the life cycle and transmitted to the next generation , therefore demonstrating that EIS can control a heritable adaptive trait , such as compatibility polymorphism .
SmPoMuc genes are classified into 4 groups ( Roger et al . 2008 ) according to their 3′region: group 1 to 4 . Group 3 is itself divided into subgroups ( 3 . 1 , 3 . 2 , 3 . 3 and 3 . 4 ) . SmPoMucs genes have a 5′ region containing a variable number of tandem repeats ( exon2 ) , which have been previously called r1 and r2 [8] . r2 exclusively occurs in the group1 and 2 and the intermingled r1–r2 exclusively occurs in the subgroup 3 . 1 , which is present in several copies with either the r1–r2 intermingled repeats or with r1 . Due to the very high degree of sequence similarity between the SmPoMuc groups , specific transcriptional analyses of the different SmPoMuc groups were only possible for groups 1 , 2 and 3 . 1 ( r1–r2 ) . The transcription levels of these groups were compared between miracidia of the IC and C strains . SmPoMuc gene groups 1 , 2 and 3 . 1 ( r1–r2 ) are 2 . 2 to 4 . 9 , 2 . 5 to 6 . 7 and 18 . 6 to 59 . 7 fold more transcribed in the IC than in the C strain , respectively ( fig . 1 ) . The 3 . 1 subgroup containing intermingled r1–r2 repeats is highly transcribed in the IC strain but was practically undetectable in the C strain . This result is consistent with a previous study on individuals of the IC and C strains , which showed that variants containing the r1–r2 combinations are only expressed in the IC strain [8] . To investigate the mechanisms underlying differences of transcription between SmPoMuc groups and subgroups , we characterized the minimal promoter region of the SmPoMuc genes . We sequenced a region spanning 1 . 04 to 2 . 00 kb upstream of the transcriptional start site ( TSS ) for 4 groups of SmPoMuc ( Groups 1 , 2 , 3 . 1 and 3 . 1 ( r1–r2 ) . We produced a PCR product of a 996 bp of the region of the promoter of the group 3 . 1 ( r1–r2 ) and a PCR product of 1002 bp of the group 3 . 1 just upstream of the transcriptional start site . Plasmids containing these sequences upstream of a reporter gene ( EGFP ) were transfected into HeLa cells and fluorescence was observed under a microscope ( fig . 2 ) . These experiments showed that these sequences are sufficient to drive the heterologous expression of the reporter gene and contain the minimal promoter sufficient for transcription . As a first approach to investigate a putative genetic basis for the difference in transcription levels between strains , we investigated the paralogous and orthologous relationships between the four groups of SmPoMuc gene promoters and between the two S . mansoni IC and C strains using phylogenetic analysis , reciprocal BLAST dot-plots and comparison of repetitive elements , duplication , recombination events and gene conversions ( fig . 3 ) . We annotated the sequences and visualised them by colour-coding of blocks with less than 95% identity ( fig . 3 ) . A recombination event was detected using BootScan [19] , [20] , Maximum Chi Square [21] , [22] and Sister Scanning [23] methods in RDP3 and the recombination break points were putatively identified ( fig . 3 ) . In both strains we observed one duplication in group 3 . 1 ( r1–r2 ) promoters resulting in an insertion , several insertions/deletions ( indels ) including one large deletion in group 3 . 1 promoters and probably a recombination event from the group 2 to group 1 promoter . High similarity to a repeated DNA element was detected in the group 2 promoter; however , it constituted only a small fragment of the complete repeat – 61 bp out of 385 bp of the DIVER2 LTR ( Drosophila ) . The estimated divergence time between the IC and C S . mansoni strains is about 400 years [6] and the promoter sequences between the two strains are highly conserved ( 0 . 000–0 . 004 net substitutions per site , Table 1 ) . The number of fixed differences between the two strains varied between 0 in the promoter region of SmPoMuc group 2 genes , to 3 in group 3 . 1 , 4 in group 1 and 8 in group 3 . 1 ( r1–r2 ) ( Table 1 ) . No substitution was observed in the TATA signal , nor in the TSS regions or in putative regulator binding sites of the promoters between the two strains . SmPoMuc promoter sequences were divided into four paralogous sequence groups and sequence differences between strains ( orthologous relationships ) within groups were much less than the differences observed between groups of the SmPoMuc gene family - net substitutions per site varied from 0 . 000–0 . 004 within groups of promoter sequences between strains compared to 0 . 024–0 . 041 between promoter groups ( Table 1 ) . The number of SmPoMuc promoter sequence differences between strains was equal to or slightly higher than the number of sequence differences for the promoter of the single copy gene SmFTZ-F1 [24] which shows no difference between strains ( Table 1 ) . Six of 14 microsatellite loci also showed no sequence differences between the two strains ( one unique allele ) . The two strains share the molecular evolution and phylogeny of the promoter region of the four groups of the SmPoMuc gene family ( fig . 3 ) – indels , recombination and duplication events . These findings indicate that the divergence between groups of the SmPoMuc gene family from a common gene ancestor is ancient and largely predates the time of separation between the IC and C strains . At this stage of the study we hypothesized that SmPoMuc expression differences in C and IC strains could be due to nucleotide differences in the promoter regions of the genes . The sequencing of 1 . 4 kb of SmPoMuc group 1 promoter region for 20 and 18 individuals of the IC and C strains respectively , revealed a very low number of alleles and genotypes ( Table 2 ) – one genotype in the IC strain and 3 genotypes in the C strain . In the C strain , sequence variation was minimal , with the three alleles differing by only one base pair from each other , resulting in insignificant nucleotide diversity ( Table 2 ) . All individuals were homozygotes . The IC strain allele of the SmPoMuc promoter group 1 differed from the three C strain alleles by four to five base pairs , a sequence divergence of 0 . 29 to 0 . 36% . In summary , nucleotide sequence differences between the two strains are surprisingly small . Promoter diversity within strain and divergence between strains of SmPoMuc group 1 genes were similar to those of 14 microsatellite loci that can be used to reflect genome-wide diversity and divergence [25] . The promoter diversity of SmPoMuc group 1 was 0 . 00 ( one allele ) in the IC strain compared to 0 . 22 ( 3 alleles ) in the C strain ( Table 2 ) , while expected heterozygosity was 0 . 000 ( one allele ) for both strains for 14 microsatellite loci ( Data not shown ) . All individuals were homozygotes . Six out of 14 microsatellite loci showed no divergence between the two strains . At eight microsatellite loci , the IC strain alleles differed from the C strain alleles by one to eleven microsatellite repeats . The promoter region of the single copy SmFTZ-F1 gene displayed a unique sequence common to the two strains . We estimated extremely high and significant genetic differentiation between the two strains for both SmPoMuc group 1 promoter sequences and microsatellite loci using θ , ΦST and RST estimators ( Table 3 ) . However , we detected almost no heterozygotes and highly significant inbreeding coefficients f in both strains and for both SmPoMuc group 1 promoter sequences and the microsatellite loci ( Table 3 ) . Therefore the high values of divergence are likely the result of the bottleneck induced during the care of the life cycle in the laboratory in the two strains as discussed previously [25] . Nonetheless , the distribution of alleles matched the pattern of differentiation as we detected fixed alleles that were different in the two strains . We reasoned that the small genetic differences in the promoter region are simply a by-product of clonality and not the reason for expression differences . We therefore explored an alternative hypothesis , i . e . that the expression differences are due to dissimilarity in the epigenetic information . As the difference in SmPoMuc transcription phenotype cannot easily be explained by genetic differences in the promoter region , we investigated the putative implication of epigenetic mechanisms . As a previous study had shown that histone modifications are clearly involved in S . mansoni epigenetic mechanisms [13] , [26] , we tried to influence the epigenotype and phenotype ( SmPoMuc expression pattern ) of S . mansoni using trichostatin-A ( TSA ) that is a specific and reversible inhibitor of class I and II histone deacetylases ( HDAC ) . Treatment with this drug prevents histone deacetylation and is expected to increase the overall acetylation of histones and therefore gene expression [26][27] . The influence of TSA treatment on the transcription of SmPoMuc genes ( group 1 , 2 and 3 . 1 ( r1–r2 ) of both C and IC strains was tested in miracidia larvae exposed during 4 h to the drug . A Friedman non-parametric test was performed to test the significance of the TSA effect ( Figure S1 ) . We observed a statistically significant increase in transcription of groups 1 and 2 after TSA treatment in the IC strain only ( p-value = 0 . 05 ) . This indicates that changes in histone acetylation correlate with increased expression for SmPoMuc group 1 and 2 in the IC strain and has no effect in the C strain . Control genes were also tested for their response to TSA in order to determine that its effect was not pleiotropic . No effect of TSA was observed for these genes ( GAPDH , Smp_011030 , Smp_152710 . 1 , Smp_054160 , Smp_158110 . 1 , GST . B , Glyaxalase , data not shown ) . Since the TSA treatment influences overall histone acetylation , it could not be excluded that the observed effect is an indirect one and that SmPoMuc expression control is posttranscriptional and/or posttranslational such as selective RNA or protein degradation . We reasoned that in the offspring of crosses between the IC and C strains transcriptional control would produce an additive pattern of SmPoMuc proteins , while control by selective degradation of gene products would produce a subtractive pattern . Western blots show that in miracidia that are produced from crosses between the strains an additive pattern of the C and IC specific bands can be observed ( fig . 4 ) . This indicates that regulation operates at the transcriptional and not the post-transcriptional level and further supports the view that chromatin structure plays a role in the generation of specific SmPoMuc profiles for each strain . Since all our experiments had delivered results in favour of a difference in chromatin structure of the SmPoMuc locus between strains , we decided to investigate the chromatin status in these loci . The occurrence of DNA methylation in S . mansoni is currently debated [28][29] . To test for DNA methylation in the promoter region of SmPoMucs we performed bisulfite genomic sequencing of DNA from miracidia using in-vitro methylated DNA as a positive control . We did not detect any methylated cytosine in the target region while 98% of the CpGs of in-vitro methylated DNA scored methylation positive . Our results are in line with earlier results showing that DNA methylation is rare from genes in S . mansoni [29][28] . We then performed Chromatin ImmunoPrecipitation ( ChIP ) experiments to check for histone modifications in the promoter regions . Due to the high similarity between the different groups of SmPoMuc promoters , ChIP-qPCR ( quantitative Polymerase Chain Reaction ) analysis was possible only in degenerate regions . Therefore , the chromatin structure analysis was performed on the promoter regions of SmPoMuc groups 1 , 3 . 1 and 3 . 1 ( r1–r2 ) . ChIP experiments were performed using an antibody that recognised Histone 3 acetylated on lysine 9 ( H3K9Ac ) and Histone 3 tri-methylated on lysine 4 ( H3K4Met3 ) which are euchromatic marks and an antibody that recognised H3 tri-methylated on lysine 9 ( H3K9Met3 ) , which is a heterochromatic mark . Immunoprecipitation with the antibody that targets H3K4Met3 did not show any enrichment in the SmPoMuc region tested for either the IC or C strains whereas controls , αTub ( Smp_090120 . 2 ) and 28S ( Z46503 . 1 ) were positive ( data not shown ) . The H3K4Met3 mark is usually very sharp and difficult to localise by target approach . . Both SmPoMuc group 1 and 3 . 1 ( r1–r2 ) from the IC strain displayed a higher level of H3K9Ac compared to the C strain ( fig . 5 ) . Consistent with this result , the C strain displayed a higher level of the heterochromatic mark ( H3K9Met3 ) for group 1 and 3 . 1 ( r1–r2 ) . These results have been obtained with several generations of the parasite , demonstrating that the phenotype is transmitted to the next generation . In the IC strain , epigenetic marks showed differences among SmPoMuc groups 1 , 3 . 1 and 3 . 1 ( r1–r2 ) ( Figure S2 ) . The promoter of group 3 . 1 ( r1–r2 ) is the most acetylated and the least heterochromatic . This result is consistent with expression analysis after TSA treatment where no effect of TSA was observed for the expression of group 3 . 1 ( r1–r2 ) . This absence of an effect of TSA may be explained by the fact that acetylation on this promoter is already saturated and cannot be further increased as previously observed for H4 acetylation in the promoter region of HDAC1 in S . Mansoni [26] . The chromatin status in the promoter sequence of SmPoMuc groups 1 , 3 . 1 and 3 . 1 ( r1–r2 ) was also investigated in the IC strain in cercaria and adults where SmPoMuc genes are not expressed . The level of the heterochromatic and euchromatic marks was the same as in miracidia and this level was maintained through several generations ( Figure S3 ) .
The host-parasite arms race determines that variability-generating processes are crucial for survival on both sides of the interaction ( red queen hypothesis , [2] ) . The mechanisms that are responsible for these ( heritable ) phenotypic variations are a current and fundamental question in evolutionary biology . Traditionally , random genetic changes are seen as the sole source of phenotypic variation . But the picture is probably more complex: heritable adaptive phenotypic shifts could be partly controlled by epigenetic factors that were underrated until recently [30] , [31] . A high rate of heritable epigenetic changes would generate phenotypic variation , which in turn could allow a rapid response to selection pressures [13]; [32] . This could allow for a transient and efficient response to changes in the environment , and could subsequently be followed by stabilization through genetic changes [33] , [34] . Epigenetic modifications affect the transcription status of a gene in a heritable way without changes in the DNA sequence [14] , [15] , [16] and epigenetic information can be based on a chromatin marking system . Chromatin exists either as a relaxed structure that is permissive to gene expression and is called euchromatin , or as a condensed structure that is typically silent and is called heterochromatin [35] . Therefore , these different chromatin states alter gene expression and , ultimately , influence phenotypic outcomes without changes to the DNA sequence . The evolutionary implications of epigenetic inheritance systems and their potential link to stress-induced phenotypic variation have been discussed in several models [36] , [37] , [38] , [31] , [39] , [40] , [41] as well as in the specific context of host-pathogen interaction [42] . While it is clear now that induced epigenetic modifications are heritable [43] , there are very few reports that show that epigenetic events lead to modification of gene expression profiles , production of new phenotypes and adaptation to the environment [44] . In the present work , we addressed the question of the relative importance of genetic and epigenetic differences between two strains of S . mansoni that show clear differences in an ecological important adaptive trait: the capacity to infect their intermediate host . We had previously identified the SmPoMuc genes as surface molecular markers important for host compatibility . These markers encode mucins that display an extraordinary level of polymorphism , although they are produced from a relatively small number of very similar genes . As we had shown that nucleotide differences in the coding region could not explain differences in transcription , we focused therefore on the promoter regions in the present work . Our comparative survey of sequence variation in the different groups of SmPoMuc gene family from IC and C strains revealed a high level of conservation of the promoter sequences of SmPoMuc genes between the two strains . The molecular evolution of SmPoMuc promoters was uniform between all strains analysed , IC , C and NMRI . The sequence differences between the IC , C and NMRI strains within each group of SmPoMuc promoter were small , and the number of substitutions between the IC and C strains was equal or slightly higher than in the monomorphic single-copy gene SmFTZ-F1 and consistent with sequence differences at 14 microsatellite loci . To assess whether substitutions between the two strains could have an effect on transcription , we searched for functional regions of the active promoters . None of the substitutions between the IC and C strains occurred in the TATA signal , putative transcription factor binding sites or TSS regions . The nucleotide differences between the two strains consisted of zero in group 2 to eight substitutions in group 3 . 1 ( r1–r2 ) , resulting in net nucleotide substitutions per site similar or lower than the ones observed in presumably neutral SmPoMuc introns ( Table 2 ) . At the population level , our analysis of SmPoMuc group 1 promoters in the IC and C strains revealed very low allelic and nucleotide variability within strain and high allele frequency differences between the IC and C strains due to fixed substitutions . All individuals were homozygotes at SmPoMuc group 1 promoter , similarly to the genotypes at 14 microsatellite loci , suggesting that S . mansoni strains present genome-wide homozygosity . Both strains are characterised by a high significant inbreeding coefficient , resulting from high clonality in the two strains [25] , which may have arisen because of the bottleneck due to the strain maintenance in laboratory conditions . Despite the lack of diversity within strains , alleles fixed in each strain for the SmPoMuc group 1 promoter and nine microsatellites were different , resulting in high genetic differentiation between the two strains as estimated by FST . This contrasted with the promoter of the single-copy gene SmFTZ-F1 and six microsatellite loci , which displayed a unique sequence common to the two strains . In summary , our analysis of the genetic information shows that ( i ) both strains are genetically monomorphic , including the SmPoMuc promoter regions , ( ii ) both strains are different in terms of alleles , i . e . they do not share the same alleles , but ( iii ) these alleles are similar or display low number of base substitutions ( outside functional regions ) . It could be argued that the small nucleotide differences observed between the two strains are sufficient to provoke modulation of histone modification . Such a leverage effect of SNPs cannot be excluded but has so far not been observed in heavily studied models such as human , Drosophila melanogaster and Arabidopsis thaliana . It could also be the case that strain-specific loci exist that regulate the chromatin structure of the SmPoMuc genes in trans or in cis ( upstream of the minimal functional promoter ) . However , previous work has compared the proteomes of both C and IC strains [7] and did not pinpoint any major regulators that may be responsible for such a phenotype . In view of these results , we argue that genetic differences between sequences within each group of SmPoMuc promoters were unlikely to solely dictate the high level of variation in SmPoMuc transcription and compatibility polymorphism phenotypes . We therefore further investigated the epigenetic basis for such phenotypes . TSA treatment was used to study the impact of overall acetylation status of histones on miracidia larvae where SmPoMuc is expressed . This drug is known to be a specific histone deacetylase ( HDAC ) inhibitor and has been previously shown to influence phenotypic traits in S . mansoni [13] . A dose dependant effect of TSA was observed for SmPoMuc expression ( all groups taken together ) in the IC strain whereas no effect was observed in the C strain . This result suggests that the acetylation status of histones in the promoter sequences is differentially regulated between the IC and C strains . HDACs seem to play a more prominent role in regulating the acetylation level in the IC strain that allowed us to pinpoint a TSA effect in this strain . More specifically , we report a TSA effect on groups 1 and 2 of the IC strain whereas no effect is observed for group 3 . 1 ( r1–r2 ) for which acetylation is the strongest . This also suggests that a differential regulation by HDAC exists between the SmPoMuc groups in the same strain . Further support for regulation on transcriptional level comes from a crossing experiment in which strain hybrids were produced . Western blots show that in the hybrids , both the C-specific and the IC-specific SmPoMucs are expressed . One could hypothesize that production of SmPoMuc variants is due to post-transcriptional strain-specific regulation . In this scenario all genes would be expressed , but the gene products would be processed in a strain-specific form . In the hybrids , in which the hypothetical post-transcriptional regulation pathway for both strains is present , we should have seen a diminution of non-IC and the non-C SmPoMuc forms . This was not the case . In summary , all lines of evidence point towards a chromatin-based regulation of SmPoMuc expression . The chromatin configuration was further investigated by ChIP analysis using antibody that recognises heterochromatic and euchromatic marks . ChIP results clearly demonstrate that different epigenetic marks occur on the SmPoMuc promoter of group 1 and group 3 . 1 ( r1–r2 ) between the IC and C strains likely resulting in a different chromatin configuration . On these loci , chromatin is indeed more enriched in H3 acetylated on lysine 9 in the IC compared to the C strain and less enriched in the opposite mark , H3 trimethylated on lysine 9 . Therefore , the local chromatin structures differ between the two strains for groups 1 and 3 . 1 ( r1–r2 ) and are consistent with expression data as stronger acetylation correlates with enhanced expression . Importantly , H3K9Met3 and H3K9Ac marks are maintained through the cercarial and adult stages at which the genes are not expressed . This persistence of the chromatin mark throughout other stages of the S . mansoni life cycle is a crucial result as this is a necessary condition for the epigenetic mechanism to act as a heritable trait . Similarly , several CVGE genes of P . falciparum that display a bistable chromatin state to regulate their expression in the intraerythrocytic stages have been shown to maintain their epigenetic marks during trophozoite and schizont stages , the other asexual stages at which these genes are not expressed [45] . It is now established that the phenotype is not onlya product of genetic processes , but expression of an ensemble that is composed of genetic and epigenetic components . Others and we have proposed that this additional system allows for rapid adaptive evolution without necessarily changing the genotype initially . A theoretical framework for this model was provided by Pal and Miklos ( 1999 ) [17] , and more recently by Klironomos , Berg and Collins ( personal communication ) . Essentially , these authors propose that a higher rate of random changes in epigenetic marks compared to genetic mutations transmitted from one generation to the next in a population generates increased phenotypic variations that can be selected for if the environment changes . In this sense , epigenetic modifications provide a source of rapid and reversible phenotypic variation and are therefore expected to be major players in the context of host-pathogen interaction where selection pressures are strong and evolution is fast [42] , [18] . In this context , epigenetic based events to generate variability of surface antigens of parasites perfectly matched to this theory . For exemple , VSP diversification of Giardia sp . likely occurs by epigenetic mechanisms involving the histone acetylation status [46] and/or RNAi [47] . Chromatin remodeling proteins and histone modifications have been shown to play a role in VSG expression site silencing [48] and Plasmodium Var diversification is orchestrated by multiple epigenetic factors including monoallelic transcription at separate spatial domains at the nuclear periphery , differential histone marks on otherwise identical var genes , and var silencing mediated by telomeric heterochromatin [49] . On the host side , genetic and epigenetic crosstalks have been previously demonstrated in the generation of a high level of polymorphism of the receptors of the adaptative immune system [50] , [51] . Therefore , all these variability generating mechanisms are examples of local adaptation to an ever-changing environment where epigenetic based events are used to rapidly produce new phenotypes and potentially induce rapid evolutionary change of genes that are under pressure . In our work , we show that two population of S . mansoni with distinct phenotypic traits , in particular their compatibility with a reference host , show low nucleotide differences in both coding sequence and promoters of SmPoMuc but high epigenetic differences in the promoter regions . Both parasite populations are in a situation where the fitness value of genetically encoded phenotypes has not changed significantly , but epigenetic variations have produced phenotypic variants that are adapted to different environments ( compatible hosts ) . While we have compared only South American strains , our observations suggest a scenario for the adaptation of S . mansoni to the new world host: in the 15th–16th century the ancestral strain of contemporary strains IC and C migrated via the slave trade from Africa to the West Indian Islands and the South American continent , respectively [6] . There , they had to adapt to a new intermediate host . The initial bottleneck resulting from the migration of only a limited number of parasites and the expected strong selective pressure acting on both genetic and epigenetic variants of the key-molecules for compatibility with the new snail hosts , SmPoMucs , may have significantly reduced genetic and epigenetic variation in the newly formed laboratory IC and C strains compared to the ancestral strain . Now , it is likely that epigenetic variation retained from the ancestral strain and the higher rate of occurrence of epigenetic changes in subsequent generations , rather than the strain genetic variation , enabled the parasite to adapt rapidly to their host and new environment . A conundrum with the “epigenetic mutation system first” hypothesis is that epigenetic information concerns the transcriptional activity of a gene but not its coding potential , in other words , a gene can be switched on and off by the surrounding chromatin but the resulting protein cannot be changed . Loss of function of genes can easily be imagined through an epigenetic mechanism , but for gain of function a complex inhibitor-based mechanism would be necessary . The classical Ohno hypothesis of gene duplications as way to provide material for evolution [52] could deliver a solution . Rodin and Riggs have shown that duplicated genes have a tendency to be heterochromatic [53] . It is interesting to note that the SmPoMuc proteins , essential for host compatibility , are encoded by duplicated genes . Our analysis shows that the duplication events predate the IC/C separation and occurred in the strain's common ancestor , i . e . gene duplication was not a result of divergence of the two strains . We postulate that SmPoMuc duplicated genes provide an additional system for phenotypic variation . Duplicated genes are randomly modulated in their relative transcriptional activity through chromatin structure changes as evidenced by our current and previous results [13] , resulting in new combinations of expressed SmPoMuc genes and subsequent increased phenotypic variation . If the parasite encounters new intermediate hosts , the probability for the phenotypes to match is increased , thus allowing for adaptive evolution . Therefore , our work shows that in a gene family that codes for an adaptive phenotypic trait , epigenetic changes are more important than genetic changes . This finding provides support for theoretical models of adaptive evolution in which epimutations occur more rapidly than mutations .
The French Ministère de l'Agriculture et de la Pêche and French Ministère de l'Education Nationale de la Recherche et de la Technologie provided permit A 66040 to our laboratory for experiments on animals and certificate for animal experimentation ( authorization 007083 , decree 87–848 ) for the experimenters . Housing , breeding and animal care followed the national ethical requirements . A compatible strain ( C ) ( Brazilian strain ) , an incompatible S . mansoni strain ( IC ) ( Guadeloupean strain ) , the reference NMRI S . mansoni strain ( Puerto Rican strain ) and a reference mollusc strain ( B . glabrata BRE isolated from Brazil ) were used in this study . For initial breeding , each strain was maintained in its sympatric ( compatible ) B . glabrata strain , and in hamsters ( Mesocricetus auratus ) as described previously [54] . Adult worms and miracidia were obtained as described previously [8] . Individual B . glabrata snails were infested with a single miracidium to obtain cercarial clonal populations . Subsequently the sex of the cercariae was determined as described previously [55] . Strain hybrids of S . mansoni were produced by infection of mice or hamster with 300 cercariae: 200 males from a clonal cercarial population combined with 100 females from another clonal cercarial population . Different combinations of parental cercariae of the IC and C strains were used , thus generating worm couples in which the male is C and the female is IC or vice versa . Eggs were recovered from infected ( 3 to 6 ) mice ( Mus musculus ) 12 weeks post-infection . Livers were collected and homogenized , and eggs were filtered and washed . Miracidia were allowed to hatch in spring water and were concentrated by sedimentation on ice for 15 minutes . 1000 Miracidia were incubated in 350 µl UTCD buffer ( ultrapure urea 8 M , Tris 40 mM , DTT 65 mM , CHAPS 4% ) , two hours at room temperature . The extract was cleared by centrifugation for 30 minutes at 1500 g , and the supernatant was collected . Total proteins ( 5 µg per sample ) were separated by 10% SDS-PAGE gel electrophoresis before being blotted on a nitrocellulose membrane ( Trans-Blot turbo , Bio-Rad ) . The membrane was blocked with 5% non-fat dry milk in TBST ( TBS buffer containing 0 . 05% tween 20 ) one hour at room temperature , and incubated with the primary antibody “anti-SmPoMuc” diluted 1/500 in TBST for 90 minutes at room temperature . This rabbit polyclonal antibody was produced according to standard procedures and was shown to recognise all the SmPoMuc groups [9] . Then , the membrane was incubated with secondary antibody ( peroxidase conjugated , purified anti-rabbit IgG ) diluted 1/5000 in TBST for 1 hour . After washing 3 times for 10 minutes in TBST , the detection was carried out using the ECL reagents and the ChemiDoc MP Imaging system – BioRad ) . We searched for sequences of promoter regions of SmPoMuc genes in the genomic database of the S . mansoni NMRI strain ( assembly version 3 . 1 ) using BLAST searches . Contigs matching to SmPoMuc genes were assembled with the Sequencher software ( Gene Codes Corporation ) to recover the sequences of the promoter regions of the genes . From the BLAST search and manual assemblage of relevant contigs , scaffolds of promoter regions were constructed for the different SmPoMuc genes in groups 1–4 . Primers were designed on these contigs to amplify the promoter regions of the different SmPoMuc genes in the C and IC strains of S . mansoni . The DNA templates to generate PCR products were either genomic DNA ( C and IC strains ) , a BAC library ( NMRI strain ) or a phage library ( IC strain ) . Genomic DNA was extracted from adult worms as described previously [8] . The production of the phage library is described below . Promoter regions were amplified using the Advantage 2 PCR Enzyme System ( Clontech ) ( Table S1 for primer sequences , amplified fragment lengths and sources of DNA ) . PCR products were either cloned into pCR-XL-TOPO ( TOPO TA Cloning kit for sequencing , Invitrogen ) and plasmid DNA was purified using the Wizard Plus SV Miniprep DNA purification system ( Promega ) , or sequenced directly . We sent PCR amplificons or plasmids containing the promoter regions to GATC ( GATC Biotech , Germany ) for cycle sequencing in both directions and performed primer walking up to 2 . 0 kb upstream of the transcription start sites ( TSS ) of SmPoMuc genes ( for primer sequences see Table S2 ) . We checked trace data and aligned nucleotide sequences manually using the BioEdit software . We scanned the promoter sequences for putative regulator binding sites using the web based interface Program NSITE ( Softberry Inc . ) ( http://linux1 . softberry . com/berry . phtml ? topic=nsite&group=programs&subgroup=promoter ) . The presence of multiple copies of some SmPoMuc genes sometimes prevented the amplification of a single copy and assembly of a gene with its corresponding promoter . To address this problem , we constructed a phage library of the IC strain using the Lambda Fix II vector system from Stratagene . The expected size of inserts was 15 to 23 kb corresponding to the size range of SmPoMuc genes ( 10–30 kb ) . Details of the construction of the phage library and screening are available at http://methdb . univ-perp . fr/epievo/ . Genome coverage of the library was four fold . The library was screened for SmPoMuc genes using as a probe UR1 , a highly conserved intronic sequence spanning the region between two repeat units of the SmPoMuc genes [8] . The probe was labeled with the DIG High Prime DNA Labeling and Detection Starter Kit II using Random primed DNA labeling with digoxigenin-dUTP , alkali-labile and chemiluminescence with CSPD ( Roche ) . Screening was performed according to the manufacturer's instructions . Secondary and tertiary screening rounds were performed with the same probe to isolate individual phage clones . Phages that scored positive for SmPoMuc repeat units were screened by PCR using a combination of diagnostic primers for each group of SmPoMuc genes ( Table S2 ) with the Advantage 2 PCR Enzyme System ( Clontech ) . Selected phages were subsequently purified and used as templates to PCR amplify SmPoMuc group 3 . 1 ( r1–r2 ) as described in the section “PCR screening for promoters of SmPoMuc genes , cloning and sequencing” . We used DnaSP to characterise promoter sequence variation within and between groups of SmPoMuc promoter sequences as the number of polymorphic sites , number of mutations between strains , net number of substitutions per site between strains and between groups of SmPoMuc promoter sequences . We amplified and sequenced the promoter region of the SmFTZ-F1 gene . This gene encodes the nuclear receptor fushi tarazu-factor 1alpha and its promoter has been fully characterised [24] in 1 and 2 individuals of S . mansoni strains IC and C , respectively , from genomic DNA with primers Smftzf1-F ( 5′-ATGAGATGTTTCTGAGCAATGGC-3′ ) and Smftzf1-R ( 5′-TCTTCTCGTAGCTGAATCTGACC-3′ ) using the Advantage 2 PCR Enzyme System ( Clontech ) . PCR amplicons were then sequenced and analysed for sequence variation and gene diversity as described above . We amplified 996 kb of the SmPoMuc group 3 . 1 ( r1–r2 ) promoter and 1002 kb of the SmPoMuc group 3 . 1 promoters . These sequences are located just upstream of the transcriptional start site and have been amplified from the IC strain . These sequences were amplified using primers containing SacI and BamHI restriction sites ( Table S2 ) . The PCR product was gel-purified ( Wizard SV gel and Clean-Up system , Qiagen ) , digested with both restriction enzymes and cloned into a SacI and BamHI digested pEGFP-1 reporter vector with T4 DNA ligase ( New England Biolabs ) . The construct was verified by sequencing both DNA strands . Plasmids pEGFP-1 and pCMV-EGFP driving EGFP expression , under the control of the CMV-promoter , were used as negative and positive controls in the transfection assay . A 3 . 3 kb region of the SmPoMuc group 1 gene promoter region was amplified using primers SmpomucpromGP3 . 1 . f2 and BR2 ( Table S1 ) in individuals of each of S . mansoni IC and C strain . The PCR products span from 1 . 8 kb upstream of the TSS to the first repeat unit of the SmPoMuc gene and cover the promoter region . 1 . 4 kb of the promoter region was sequenced for 20 and 18 individuals of the IC and C strains , respectively , by primer walking ( Table S2 ) . We used Arlequin 3 . 1 to characterise SmPoMuc group 1 promoter diversity within the two strains as the expected unbiased gene diversity , the nucleotide diversity , corrected for sample size and incorporating nucleotide information [62] . We tested for sequence variation between the two strains using population comparisons and differentiation in Arlequin 3 . 1 . Estimations incorporated Tamura-Nei distances between sequences and allele frequencies ( Nei's Φ-estimator of FST ) . The significance of genetic differentiation was tested by permuting the alleles among all samples 2 , 000 times . We also estimated the inbreeding coefficient in each strain using f and genetic differentiation between the two strains using FST estimator θ ( [63] , incorporating allele frequencies only ) . Inbreeding coefficients and genetic differentiation for departure from the null hypothesis ( f = 0 , θ = 0 ) were tested using 2 , 000 permutations in GENETIX 4 . 05 [64] . Nineteen individuals of each of the IC and C strains were genotyped using 14 microsatellite loci [25] . We estimated genetic diversity of microsatellite loci as the mean number of alleles per locus ( A ) and observed and expected unbiased heterozygosities ( HO and Ĥ ? respectively ) under the assumption of Hardy–Weinberg equilibrium [62] . We estimated the inbreeding coefficient f in each strain , genetic differentiation between the two strains RST estimator [65] , [66] and the FST estimator θ as above . Trichostatin-A ( TSA ) ( invivoGen met-tsa-5 ) was dissolved in ethanol to 20 mM and added to the 1000 IC or C miracidia pool at 20 µM and 200 µM during 4 h . We had shown previously the effect of TSA at these concentrations on development , morphology , mobility and gene expression without any cytotoxicity for the larvae [13] , [27] . To the untreated control , an equal volume of ethanol was added ( mock treatment ) . After 4 h , metamorphosis arrest was observed for larvae treated with TSA at 200 µM as expected for a positive effect with this drug [27] . Miracidia were then spun down at 12 , 000 g during 5 min and suspended in 100 µl of lysis buffer ( Dynabeads mRNA DIRECT Micro kit , Dynal Biotech ) in RNase-free tubes and stored at −80°C . Messenger RNAs were extracted using the Dynabeads mRNA isolation Kit according to the manufacturer's instructions . mRNA poly-A residues were eluted from the surface of the paramagnetic beads by a final denaturation step of 10 min at 75°C in 20 µl of Tris-HCl 10 mM . cDNA synthesis was carried out using 10 µl of mRNA in a final volume of 20 µl according to manufacturer's instructions ( 0 . 5 mM dNTPs , 0 . 01 mM DTT , 1× first strand buffer , 2 U RNase out , 10 U SuperScript II RT ( Invitrogen ) during 50 min at 42°C ) . After reverse transcription , the cDNAs were purified with the PCR clean-up system ( Promega ) and eluted into 100 µl 10 mM Tris/HCl ( ph 7 . 5 ) . Specific primers for qPCR from groups 1 , 2 and 3 . 1 ( r1–r2 ) were designed based on sequence alignment performed on cDNA variant representative of each group ( Table S2 ) . Their specificity was tested using as template a plasmid in which a cDNA variant of group 1 , 2 or 3 . 1 ( r1–r2 ) was cloned . Group 4 genes contain a STOP codon in exon 8 of the gDNA sequence and their cDNA has never been detected . Therefore , transcripts of the group 4 genes were not targeted in this study . Other subgroups were not studied as it was not possible to design specific primers to amplify them . qPCR amplifications were performed as described below . Results were normalised with the αTub gene . The 2ΔCt value was calculated . Statistical tests were performed on at least 3 different biological samples . Native chromatin immunoprecipitation was performed as described before [67] . Briefly , antibodies against histone isoforms were used to precipitate chromatin in miracidia from IC and C strains ( Table S3 ) . DNA was extracted from the precipitated complex and analysed by qPCR using specific primers of SmPoMuc groups 1 , 3 . 1 and 3 . 1 ( r1–r2 ) . Primers specifically targeting these genes were designed based on sequence alignment of SmPoMuc promoter sequences ( Table S2 ) . We tested their specificity using as templates plasmids with promoters of group 1 , 3 . 1 or 3 . 1 ( r1–r2 ) . It was not possible to design primer sets that would hybridize specifically to the promoter sequences of the other groups or subgroups because conservation in the sequences resulted in cross-amplification between these groups . The amount of target DNA recovered in the immunoprecipitated fraction was quantified by calculating the percent input recovery ( % IR ) normalised with the percent input recovery obtained with a reference locus ( αTub ) as previously described [67] . Bisulfite genomic sequencing was carried out as described in [68] ) on gDNA extracted from miracidia from the NMRI strain . Amplification was performed using primers BS . IC-1-Group1/1111-1715 . 48f GATATGTTTTAAGAAGTAGAAAAGAATATT , BS . IC-1-Group1/1111-1715 . 508r ATAAAAATTTTACAACCACCTACTC and BS . IC-1-Group3 . 1/421-952 . 29f ATTGTTTTTTTTAATTTTAGATATGTTTTA and two rounds of PCR . 1 µl of each PCR products were cloned into the TOPO TA vector ( Invitrogen ) and sequenced . In-vitro methylation with M . SssI ( NEB ) was done as recommended by the supplier . A total of 20 sequences ( 7 M . SssI treated positive controls and 13 target miracidial gDNA ) were aligned with the genomic sequence from GenBank ( Bioedit ) to visualise the sites of methylated cytosine . qPCR amplifications were performed with 2 . 5 µl of immunoprecipitated DNA or cDNA in a final volume of 10 µl on a LightCycler® 480 II Real Time instrument ( 1 . 5 µl H20 , 0 . 5 µM of each primer , 5 µl of master mix ) . The following protocol was used: denaturation , 95°C for 10 minutes; amplification and quantification ( 40 times ) : 95°C for 10 seconds , 60°C for 10 seconds , 72°C for 20 seconds; melting curve , 65–97°C with a heating rate of 0 . 11°C/s and continuous fluorescence measurement , and a cooling step to 40°C . For each reaction , the cycle threshold ( Ct ) was determined using the “2nd derivative” method of the LightCycler® 480 Software release 1 . 5 . PCR reactions were performed in duplicate and the mean value of Ct was calculated . Correct melting curves were checked using the Tm calling method of the LightCycler® 480 Software release 1 . 5 . The amplification of a unique band was verified by electrophoresis separation through a 2% agarose gel for each qPCR product . JQ615951–JQ615966 . | Schistosoma mansoni is a parasitic worm and agent of a disease that causes a considerable economic burden in African and South American countries . The propagation of the parasite requires passage through a freshwater snail of Biomphalaria genus . In the field , actually very few snails are infected . This is due to the fact that specific strains of the parasite can infect only specific strains of the snail . Comparative studies have shown that this so-called compatibility is based on the expression of a family of genes that are called SmPoMucs . We have shown previously that all parasites strains possess the repertoire of all SmPoMuc genes but every strain and even every individual parasite expresses only a subset . These differences could be due to DNA sequence differences in the regions that control gene expression , but here we show that these regions are nearly identical . Instead , the chromatin structure shows strain-specific characteristics . This means that the parasite can adapt to different snail strains simply by changing its chromatin structure and not necessarily the DNA sequence . If this holds true for other parasites , then we have to rethink the way parasite evolution is currently imagined but this also provides a new potential entry point to control the spread of diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology"
] | 2013 | Schistosoma mansoni Mucin Gene (SmPoMuc) Expression: Epigenetic Control to Shape Adaptation to a New Host |
During a dengue outbreak on the Caribbean island Aruba , highly elevated levels of ferritin were detected in dengue virus infected patients . Ferritin is an acute-phase reactant and hyperferritinaemia is a hallmark of diseases caused by extensive immune activation , such as haemophagocytic lymphohistiocytosis . The aim of this study was to investigate whether hyperferritinaemia in dengue patients was associated with clinical markers of extensive immune activation and coagulation disturbances . Levels of ferritin , standard laboratory markers , sIL-2R , IL-18 and coagulation and fibrinolytic markers were determined in samples from patients with uncomplicated dengue in Aruba . Levels of ferritin were significantly increased in dengue patients compared to patients with other febrile illnesses . Moreover , levels of ferritin associated significantly with the occurrence of viraemia . Hyperferritinaemia was also significantly associated with thrombocytopenia , elevated liver enzymes and coagulation disturbances . The results were validated in a cohort of dengue virus infected patients in Brazil . In this cohort levels of ferritin and cytokine profiles were determined . Increased levels of ferritin in dengue virus infected patients in Brazil were associated with disease severity and a pro-inflammatory cytokine profile . Altogether , we provide evidence that ferritin can be used as a clinical marker to discriminate between dengue and other febrile illnesses . The occurrence of hyperferritinaemia in dengue virus infected patients is indicative for highly active disease resulting in immune activation and coagulation disturbances . Therefore , we recommend that patients with hyperferritinaemia are monitored carefully .
Outbreaks of dengue virus ( DENV ) infection have become more frequent in the American and Caribbean region , even threatening to spread in the United States [1] . DENV is a flavivirus , which is transmitted by the bite of an Aedes mosquito . Brazil is the country with most reported dengue cases in the Americas . A large DENV-2 outbreak in 2010 caused more than 34 . 000 cases and 64 deaths in the State of São Paulo , Brazil [2] . On the Caribbean island Aruba , there was an epidemic from September 2011 till April 2012 , in which DENV-1 and DENV-4 were both co-circulating . The symptoms of DENV infection are mild and self-limiting in the majority of cases , consisting of fever , headache , retro-orbital pain , myalgia , arthralgia , thrombocytopenia , minor mucosal bleeding and skin manifestations . Some patients develop severe symptoms , such as shock , severe bleeding or organ impairment . These symptoms usually develop three to five days after the onset of disease around the time of defervescence . It has been hypothesized that severe dengue is caused by a cytokine storm inducing systemic inflammatory effects ( Reviewed in [3] ) . The pathophysiological mechanisms that cause this cytokine storm are not fully unravelled and represent an important focus for dengue research . In addition to the current laboratory markers for dengue , ferritin levels were described to be associated with clinical disease severity in children [4] . In many cases ferritin levels higher than 500 µg/L were detected , defined as hyperferritinaemia [5] . Ferritin is an acute-phase reactant and highly expressed by cells of the reticulo-endothelial system in response to infection and inflammation . Ferritin binds iron , limiting its availability in the circulation . Because many pathogenic microorganisms need iron for their proliferation , this mechanism is favourable for the host . Moreover , iron deficiency enhances the immunological performance of lymphocytes , neutrophils and macrophages ( reviewed in [6] ) . Hyperferritinaemia is a hallmark of diseases , characterized by extensive immune activation , including haemophagocytic lymphohistiocytosis ( HLH ) and macrophage activation syndrome ( MAS ) . HLH can be congenital or triggered by an external stimulus , such as malignancy or viral infection , including dengue [7] . NK cells and CD8+ T lymphocytes are impaired in their cytotoxic function in patients with HLH , which results in reduced clearance of infected and antigen-presenting cells from the circulation . This may lead to an exaggerated immune response with proliferation of dendritic cells , tissue macrophages and T-cells , contributing to a cytokine storm ( reviewed in [8] ) . The symptoms of HLH consist of ongoing fever , hepatosplenomegaly , cytopenia ( affecting more than 2 cell lineages ) , hypofibrinogenaemia , hypertriglyceridaemia , hyperferritinaemia , increased levels of sIL-2R and coagulopathy ( reviewed in [5] ) . Although clinical symptoms of DENV infection are usually rather mild compared to HLH , some patients develop severe symptoms . These are most probably caused by extensive immune activation and show similarities with the clinical hallmarks of HLH and MAS , suggesting similar pathophysiology . The aim of this study was to investigate the association between hyperferritinaemia , immune activation and coagulation disturbances in DENV infected patients . We showed that the presence of hyperferritinaemia could discriminate between dengue and other febrile diseases . Moreover , we found an association between increased ferritin levels and severe clinical disease , thrombocytopenia , liver enzyme and coagulation disturbances and a pro-inflammatory cytokine profile .
In order to determine the infecting serotype a semi-quantitative RT-PCR ( Taqman ) was performed . Primers and probes directed against the capsid were derived from Sadon et al . [13] . Briefly , 4× TaqMan Fast Virus 1-step Master Mix ( Invitrogen ) was used with 20 pmol of primers and 10 pmol of probes . The cycling program consisted of 5 minutes at 50°C , then 20 seconds at 95°C followed by 40 cycles of 3 seconds at 95°C and 30 seconds at 60°C . Another quantitative RT-PCR was performed to determine the viral copy number . The primers and probes directed against the 3′UTR were derived from Drosten et al . [14] . Briefly , 4× TaqMan Fast Virus 1-step Master Mix was used with 15 pmol of primers and 10 pmol of probes and an additional 25 mM of MgCl2 was added [15] . The cycling program was similar to the serotype quantitative RT-PCR . Plasma ferritin concentrations were determined at the Landslaboratorium in Aruba within a few hours after blood sampling . The assay was performed using the ‘Access’ ( Beckman Coultier , USA ) under standardized conditions . Serum sIL-2Rα and IL-18 levels were determined at the department of experimental internal medicine from the Radboud University . sIL-2Rα was measured using a commercially available luminex kit ( ‘Milliplex’ , Merck Millipore , Germany ) . Samples were diluted 1∶5 and the assay was performed according to the manufacturer's instructions and run on a Luminex 200 dual laser detection system . The sensitivity limit was 15 pg/ml . Levels of IL-18 were measured using a commercially available ELISA kit ( MBL , Japan ) according to the manufacturer's instructions . All markers of coagulation were determined in citrate samples . Activated partial thromboplastin time ( APTT ) and prothrombin time ( PT ) were determined at the Landslaboratorium in Aruba within a few hours after blood withdrawal . PT ( Dade Innovin ) and APTT ( Dade Actin FSL ) were determined on a Sysmex CA-1500 System ( Siemens Healthcare Diagnostics , USA ) . All other coagulation parameters were determined at the department of Experimental Vascular Medicine from the Academic Medical Centre . Von Willebrand Factor ( vWF ) was measured using a home-made ELISA with antibodies from DAKO ( Glostrup , Denmark ) . In vitro thrombin generation was assayed by measuring peak thrombin levels with the Calibrated Automated Thrombography ( CAT ) as described previously [16] . In vivo thrombin generation was determined by detecting thrombin-antithrombin complexes ( TAT ) using a commercially available ELISA ( Enzygnost ) . Levels of the fibrinolytic markers plasminogen activator inhibitor type 1 ( PAI-1 ) and plasmin-α2-antiplasmin ( PAP ) complexes were measured with commercially available ELISAs according to the manufacturer's instructions ( PAI-1 , Hyphen BioMed; PAP complexes , DRG Diagnostics ) . D-dimer levels were determined with a particle-enhanced immunoturbidimetric assay ( Innovance D-dimer , Siemens Healthcare Diagnostics ) . Serum ferritin levels were determined with a commercially available ELISA ( Biolisa ferritina , Bioclin , Brazil ) performed according to the manufacturer's instructions . The measurement of cytokines and the cluster analysis have been described in a previous publication [12] . Briefly , levels of thirty cytokines were measured using a multiplex immunoassay kit with spectrally encoded antibody-conjugated beads ( Human Cytokine 30-plex panel , Invitrogen , USA ) . Twenty-three cytokines were used in a cluster analysis procedure , which was adapted from van den Ham et al . [17] . Briefly , cytokine values were log-transformed and subjected to hierarchical correlation clustering ( i . e . , with distance measure 1 – pearson's pairwise correlation value ) using Ward's method . IBM SPSS Statistics v . 20 was used to calculate statistical significance . The Mann Whitney U test was used to compare the difference between two groups . The Spearman's correlation coefficient was applied to calculate correlations . The Chi-Squared test was used to calculate differences in proportions between groups and the Fisher's exact to determine whether one distribution was unequally distributed over the groups . Using bonferroni correction the p-value was adjusted for multiple testing .
Ferritin levels were determined in patients with dengue and OFI to identify any association with disease severity . Using the 2009 WHO dengue case classification , ferritin levels were significantly increased in WS+ patients compared to OFI at each time point and in WS− patients compared to OFI at day 4–5 ( Figure 1A ) . At day 4–5 and 6–8 the highest ferritin levels were observed and a tendency was shown towards higher ferritin levels in WS+ patients compared to WS− , although these differences were not statistically significant . In clinical practice and according to the official HLH-criteria , ferritin levels ≥500 µg/L are considered hyperferritinaemia [5] . A larger proportion of males showed hyperferritinaemia compared to females in this cohort , which approached statistical significance ( Table 1 ) . It is known that baseline ferritin levels are higher in the male than in the female population [18] . We calculated a fold change by dividing the absolute values of ferritin by the median ferritin levels for males and females from the autologous control group ( Females:/37 µg/L and males:/154 µg/L ) , which were similar to levels previously described [18] . The ferritin fold change was significantly increased in WS+ dengue patients compared to OFI at each time point ( Figure 1B ) . Another marker of disease severity is the hospitalization rate . Absolute levels and the ferritin fold change were significantly increased in hospitalized and outpatients compared to OFI at almost all time points ( Figure 1C and 1D ) . The absolute ferritin levels as well as the fold change showed a tendency of increased values in hospitalized patients compared to outpatients . Because the difference in ferritin levels between patients with dengue and OFI was significant , we calculated an odds ratio for the occurrence of hyperferritinaemia and a confirmed diagnosis of DENV infection . In dengue patients , 19 out of 43 had hyperferritinaemia compared to two out of 17 patients with OFI . This resulted in a sensitivity of 44% , a specificity of 88% and an odds ratio of 6 . The high values of the specificity and odds ratio suggest that the occurrence of hyperferritinaemia may serve as a discriminatory marker between dengue and OFI . The presence or absence of viraemia in the early phase was linked to ferritin levels during the course of disease . Patients were considered viraemic if they had detectable virus titres at day 2–3 and day 4–5 . Patients with undetectable levels at these days were considered non-viraemic . The absolute ferritin levels were significantly elevated in viraemic patients compared to non-viraemic patients at day 6–8 ( Figure 2A ) . The ferritin fold change was significantly elevated in viraemic patients at all time points ( Figure 2B ) . There were no strong correlations between the viral load and the levels of ferritin at the same day of disease . However , absolute levels of ferritin at day 6–8 correlated significantly with the viral copy number at day 2–3 ( ρ = 0 . 5; P = 0 . 008 ) and day 4–5 ( ρ = 0 . 5; P = 0 . 002 ) ( Figure S1 ) . The ferritin fold change at day 6–8 also showed a significant correlation with the viral load at day 2–3 ( ρ = 0 . 5; P = 0 . 003 ) and day 4–5 ( ρ = 0 . 6; P<0 . 0001 ) . This suggests that viral replication in the early phase of disease may cause an increase in ferritin levels in the convalescent phase . Hyperferritinaemia is a prominent symptom of patients with HLH . To investigate whether the clinical picture of DENV infection shows more similarities , the official diagnostic criteria for HLH [5] were linked to hyperferritinaemia in dengue patients . In each patient the occurrence of hyperferritinaemia was evaluated at each time point . Severe cytopenia in at least two cell lineages is a prominent feature of HLH due to the increased phagocytic activity of macrophages . In our cohort the platelet count was significantly decreased in patients with hyperferritinaemia compared to patients with no hyperferritinaemia and OFI at each time point ( Figure 3A , Table S2 ) . No significant differences in the leukocyte count were detected ( Data not shown ) . Another criterium is the presence of hypertriglyceridaemia and/or hypofibrinogenaemia . Levels of fibrinogen were significantly decreased in patients with hyperferritinaemia compared to patients without hyperferritinaemia at day 6–8 , but levels were still in the range of the autologous control group ( Figure 3B ) . The triglyceride levels were in the normal range of the autologous control group in both dengue as well as OFI patients ( data not shown ) . MAS is characterized by hepatosplenomegaly and liver dysfunction . The liver also plays an important role in the pathogenesis of DENV infection . Levels of the liver enzyme ASAT were significantly increased in patients with hyperferritinaemia compared to patients with no hyperferritinaemia and OFI at each time point ( Figure 3C ) . ALAT levels were also significantly increased in patients with hyperferritinaemia at day 4–5 and 6–8 ( Figure 3D ) . sIL-2R is a marker of T-cell activation and IL-18 of macrophage activation . sIL-2R was significantly increased in patients with hyperferritinaemia compared to OFI at day 2–3 ( Figure 3E ) . Levels of IL-18 were significantly elevated in patients with no hyperferritinaemia compared to OFI at day 2–3 and in patients with hyperferritinaemia compared to OFI at day 4–5 ( Figure 3E ) . Altogether , we can conclude that hyperferritinaemia in uncomplicated dengue patients is strongly associated with thrombocytopenia and elevated liver enzymes , but these patients had no hypertriglyceridaemia , hypofibrinogenaemia or cytopenia in another lineage than the platelets . Hyperferritinaemia was investigated in association with parameters , indicating the activation of coagulation and fibrinolysis . The APTT and PT showed no significant differences between any of the groups ( data not shown ) . vWF is released upon endothelial cell activation and plays an important role in the formation of the thrombus . Significantly increased levels were found in patients with hyperferritinaemia compared to OFI at day 2–3 and in both dengue groups compared to OFI at day 4–5 and 6–8 ( Figure 4A ) . Activation of the coagulation cascade starts with thrombin generation after which it is bound by antithrombin . Thrombin-antithrombin ( TAT ) complexes are a marker for activation of the coagulation cascade in vivo . Levels were significantly elevated in dengue patients with hyperferritinaemia compared to OFI at day 2–3 and 4–5 and also in patients without hyperferritinaemia compared to OFI at day 4–5 ( Figure 4B ) . The ability of plasma to generate thrombin in vitro can be investigated by the calibrated automated thrombrogram measuring peak thrombin levels . Interestingly , while the levels of TAT were significantly increased , the peak thrombin levels were significantly decreased in patients with hyperferritinaemia compared to OFI at day 2–3 and 4–5 ( Figure 4C ) . Thrombin generation will lead to fibrin formation and eventually fibrinolysis , resulting in the formation of plasminogen-α2-antiplasmin ( PAP ) complexes . PAP showed increased levels in patients with hyperferritinaemia compared to patients with OFI at each time point ( Figure 4D ) . Plasminogen activator inhibitor-1 ( PAI-1 ) can counteract the fibrinolytic system . Levels of PAI-1 were significantly elevated in patients with hyperferritinaemia compared to OFI at day 4–5 ( Figure 4E ) . Activation of the coagulation and fibrinolytic systems eventually result in the production of D-dimers ( Figure 4F ) . Levels of D-dimers were significantly increased in patients with hyperferritinaemia compared to patients with no hyperferritinaemia and OFI at day 2–3 and levels were significantly elevated in patients with hyperferritinaemia compared to OFI at day 4–5 and in patients without hyperferritinaemia compared to OFI at day 6–8 . The coagulation and fibrinolytic systems are highly activated in dengue patients and dengue patients with hyperferritinaemia in particular . The strongest activation was shown at day 2–3 and 4–5 after onset of fever with increased levels of vWF , TAT , PAP and D-dimer . To confirm our findings concerning ferritin levels in the cohort from Aruba , we studied ferritin in a previously published dengue cohort obtained during the 2010 DENV outbreak in Brazil . This cohort consisted of 50 WS− , 49 WS+ and 33 severe dengue patients ( More clinical details about this cohort are described in the previous publication and in Tables S1 , S3 and S4 [12] ) . In this cohort the ferritin fold change was calculated with the same formula as described for the cohort of Aruba , because the autologous control group in Aruba was much larger ( N = 45 ) than the healthy control group in Brazil ( N = 14 ) . The ferritin fold change was significantly elevated in patients with severe dengue according to the 2009 WHO classification , as well as in patients with shock and severe haemorrhage compared to patients with uncomplicated dengue ( Figure 5A , B and C ) . In non-survivors levels were significantly elevated compared to survivors ( Figure 5D ) . The absolute values of the ferritin fold change were on average higher in the Brazilian than in the Aruba cohort . This could be due to the presence of more severe disease in the cohort from Brazil and the use of a different assay . Patients were clustered based on the expression of the determined cytokines as has been previously described ( Figure 5E and heatmap ) [12] . Cluster A contained mainly healthy controls , cluster B mild to moderately ill dengue patients and cluster C contained severely ill dengue patients . Severe dengue ( P = 2 . 2×10−16 ) , shock ( 3 . 4×10−5 ) , severe haemorrhage ( P = 0 . 007 ) and death ( P = 0 . 03 ) occurred significantly more often in cluster C than the other two clusters . Cluster C showed a pro-inflammatory cytokine profile with increased expression of IL-6 , IL-8 , IL-10 , IL-15 , IL-1RA , sIL-2R , HGF , VEGF , G-CSF , MCP-1 , IP-10 , and MIG . Levels of ferritin were significantly increased in cluster C compared to the other two clusters and levels were also significantly elevated in the ‘dengue’ clusters B and C compared to healthy control cluster A ( Figure 5E ) . In summary , we can conclude that levels of ferritin were significantly associated with clinical disease severity and a pro-inflammatory cytokine profile .
In the cohort from Aruba , increased concentrations of ferritin were significantly associated with a confirmed dengue diagnosis and viraemia . Moreover , hyperferritinaemia in dengue was strongly associated with thrombocytopenia and increased levels of liver enzymes and both activation of the coagulation and the fibrinolytic systems . The findings were confirmed in a cohort from Brazil , in which increased levels of ferritin were associated with severe disease and a pro-inflammatory cytokine profile . Ferritin is an acute-phase reactant and a significant amount is produced by monocytes , macrophages and hepatic cells . It has been shown that synthesis of ferritin can be induced by cytokines and iron [19] , [20] . We showed that increased levels of ferritin were associated with a pro-inflammatory cytokine profile . Lipopolysaccharide ( LPS ) was shown to induce iron retention in human monocytic cells , which may subsequently induce ferritin expression [21] . Interestingly , increased levels of LPS have been reported in patients with dengue and were also associated with a pro-inflammatory cytokine profile , suggesting that cytokines , LPS and ferritin all play a role in immune activation in severe dengue [12] , [22] . Interestingly , it has been shown in vitro that ferritin can bind to high-molecular-weight kininogen and block the release of bradykinin [23] . Bradykinin is a potent vasoactive agent and plays an important role in the induction of vascular permeability and even hypotension ( reviewed in [24] , [25] ) . This may suggest that ferritin during DENV infection is induced in an effort to protect the host . It is well known that infectious diseases in general cause hyperferritinaemia ( reviewed in [6] ) . We showed that even in mild dengue , the occurrence of hyperferritinaemia could serve as a discriminatory marker between dengue and other febrile illnesses . Increased levels of cytokines and LPS have been reported in several infectious diseases and therefore these mechanisms cannot solely explain these extremely high ferritin levels . Macrophages , monocytes and lymphocytes in the peripheral blood are the major target cells of DENV replication in vivo [26] , [27] . Monocytes and macrophages are also important producers of ferritin and therefore direct infection and subsequent viral replication in these cells may activate them and increase the ferritin production . In agreement with this , ferritin levels in the convalescent phase correlated strongly with the viral load in the early phase . Interestingly , a high viral load in the early phase of DENV infection has previously been associated with the development of severe symptoms around the time of defervescence [28] . Hepatocytes can also synthesize ferritin and in our study liver enzymes were significantly elevated in patients with hyperferritinaemia . DENV replicates very well in hepatic cell lines in vitro , but whether DENV replicates well in the liver in vivo is still a matter of debate [26] . However , it is likely that liver cells are also indirectly activated by cytokines and/or activated immune cells to produce high amounts of ferritin . It has been shown that DENV infection in mice resulted in NK and CD8+ T cell infiltration of the liver [29] . HLH is characterized by extensive activation and proliferation of NK and CD8+ T cells . CD8+ T-cells can be infected by DENV in vitro [30] . Moreover , apoptosis of CD8+ T cells plays an important role in immune modulation during DENV infection [31]–[33] . sIL-2R is a marker of T-cell activation and increased levels of sIL-2R have been detected in dengue patients with severe disease [28] , [34] , [35] . In a previous study with patients from the Brazilian cohort , increased levels of sIL-2R were associated with mortality [12] . In this study , using the same cohort , levels of ferritin were also significantly associated with mortality , suggesting that extensive activation of monocytes and macrophages with subsequent T-cell activation may be detrimental for the host during DENV infection . Certain HLH-criteria , such as hypertriglyceridaemia , hypofibrinogenaemia and cytopenia in at least two cell lineages were not found in this study , most probably because the patients in this cohort only suffered from uncomplicated dengue . Increased triglyceride levels and hypofibrinogenaemia have been reported in patients with dengue shock syndrome and non-survivors [36] , [37] . Therefore , we cannot exclude that HLH-like disease occurs in dengue patients with severe symptoms . In our study thrombocytopenia was strongly associated with hyperferritinaemia . Thrombocytopenia is a hallmark of DENV infection and it is hypothesized that it can be caused by binding of platelets to activated endothelial cells [38] . Platelets are most probably bound by vWF multimers , which were increased in patients with hyperferritinaemia in this study . Because the cytopenia was limited to the platelet count in DENV infection , it is not very likely that phagocytosis by highly activated macrophages is the cause of thrombocytopenia as in the case of HLH . Coagulopathy is one of the criteria of HLH and also described in severe dengue ( reviewed in [39] ) . In our cohort , the coagulation and fibrinolytic systems were highly activated in patients with hyperferritinaemia at day 2–3 and 4–5 after the onset of fever resulting in increased levels of vWF , TAT , PAP and D-dimer . Levels of TAT were increased in patients with hyperferritinaemia , while levels of peak thrombin were decreased . TAT is a marker of thrombin generation in vivo , while peak thrombin is a marker for the potential of plasma to generate thrombin in vitro . From these results we may conclude that coagulation activation , thrombin formation and the consumption of coagulation factors decrease the ex vivo capacity for clotting during DENV infection , which may result in clinical bleeding symptoms . In addition to activation of the coagulation cascade , increased levels of PAP , PAI-1 and D-dimer showed that the fibrinolytic system was also highly activated in patients with hyperferritinaemia . Based on the collective results presented in the manuscript , hyperferritinaemia can be considered as a clinical marker for DENV infection , which can discriminate between dengue and other febrile illness . Moreover , ferritin can also serve as a marker for highly active disease resulting in extensive immune activation , coagulation disturbances and severe clinical symptoms . Therefore , we suggest that patients with hyperferritinaemia are monitored carefully , as they are at higher odds to develop severe disease . | Ferritin is an acute-phase reactant and produced by reticulo-endothelial cells in response to inflammation and infection . In general , ferritin levels are increased in inflammatory conditions , but in this study we found that ferritin levels were much higher in dengue virus infected patients than in patients with other febrile illnesses . This indicates that ferritin could be used as a marker to discriminate between dengue and other febrile diseases . Moreover , the presence of hyperferritinaemia ( ferritin levels≥500 µg/L ) was associated with markers of immune activation and coagulation disturbances and clinical disease severity , suggesting that it could serve as a marker of activity of disease . Clinical markers to determine the presence and severity of dengue virus infection are important for diagnostic and treatment purposes . Our results indicate that increased ferritin levels could be used to increase the likelihood on a positive dengue diagnosis . Moreover , patients with hyperferritinaemia should be monitored carefully , because they are at risk to develop severe disease due to extensive immune activation . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"biochemistry",
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"protein",
"complexes",
"human",
"ferritin",
"proteins",
"dengue",
"fever",
"ferritin",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases"
] | 2014 | Hyperferritinaemia in Dengue Virus Infected Patients Is Associated with Immune Activation and Coagulation Disturbances |
In this paper we report a quantitative laser Biospeckle method using VDRL plates to monitor the activity of Trypanosoma cruzi and the calibration conditions including three image processing algorithms and three programs ( ImageJ and two programs designed in this work ) . Benznidazole was used as a test drug . Variable volume ( constant density ) and variable density ( constant volume ) were used for the quantitative evaluation of parasite activity in calibrated wells of the VDRL plate . The desiccation process within the well was monitored as a function of volume and of the activity of the Biospeckle pattern of the parasites as well as the quantitative effect of the surface parasite quantity ( proportion of the object’s plane ) . A statistical analysis was performed with ANOVA , Tukey post hoc and Descriptive Statistics using R and R Commander . Conditions of volume ( 100μl ) and parasite density ( 2-4x104 parasites/well , in exponential growth phase ) , assay time ( up to 204min ) , frame number ( 11 frames ) , algorithm and program ( RCommander/SAGA ) for image processing were selected to test the effect of variable concentrations of benznidazole ( 0 . 0195 to 20μg/mL / 0 . 075 to 76 . 8μM ) at various times ( 1 , 61 , 128 and 204min ) on the activity of the Biospeckle pattern . The flat wells of the VDRL plate were found to be suitable for the quantitative calibration of the activity of Trypanosoma cruzi using the appropriate algorithm and program . Under these conditions , benznidazole produces at 1min an instantaneous effect on the activity of the Biospeckle pattern of T . cruzi , which remains with a similar profile up to 1 hour . A second effect which is dependent on concentrations above 1 . 25μg/mL and is statistically different from the effect at lower concentrations causes a decrease in the activity of the Biospeckle pattern . This effect is better detected after 1 hour of drug action . This behavior may be explained by an instantaneous effect on a membrane protein of Trypanosoma cruzi that could mediate the translocation of benznidazole . At longer times the effect may possibly be explained by the required transformation of the pro-drug into the active drug .
Chemotherapy is the most widely used method to control , prevent and treat parasitic infections . In order to understand the action of a drug on a parasite , in vitro assay techniques that test its activity , as well as the physical factors that affect the parasites are performed under the best conditions that allow the detection and quantification of the effect , avoiding or decreasing random factors and using fast , linear , sensitive , accurate , precise , inexpensive and reproducible methods [1] . The activity of most anti-parasitic molecules is based on their ability to permeate , to reach a specific receptor inside the parasite , to obtain delivery of effective drug concentrations in sufficient time to cause the therapeutic effect before it is degraded , transformed or eliminated [2] . Understanding the relationship of the anti-parasitic effect with concentration and time is crucial for this rational . Moreover , recently , efforts have focused on combination of drugs in order to aim at new therapies as a more effective alternative in the treatment of Chagas’ disease [3] . The design of in vitro assays must take into account that parasites are very sensitive to physical , chemical and biological factors such as handling , temperature , evaporation , osmolarity changes , nutritional factors , doubling time , etc . The assays that are based on viability require a low dose of the test drug and several hours and even days to develop [4] . Therefore , the use of methods that require a short time are useful in avoiding undesired factors such as degradation , biotransformation and other effects on the test drug . A fast assay could be useful for the detection of the first stages of a drug’s pharmacological activity . Image processing using new algorithms has already been proposed for T . cruzi detection and drug testing [1][5][6] . However , it is desirable to develop algorithms , programs and free and open source software , to quantify microorganisms , especially in the case of Neglected Tropical Diseases that are frequently addressed under limited conditions . In the present experimental work , we describe a novel assay methodology for the in vitro testing of parasite activity using a VDRL plate . The methodology is based on laser dynamic speckle interferometry , called Biospeckle interferometry [7][8] . Different Biospeckle descriptors have been proposed and evaluated from the performance point of view [9] and others have explored the combination of spatial—temporal patterns for the analysis of the dynamics of slow-varying phenomena [10] . In the present work the Biospeckle technique is applied to several samples , following the time evolution of the activity of the Biospeckle patterns . Epimastigotes of Trypanosoma cruzi M/HOM/ VE/67 /EP-67 were used as the experimental model . In order to standardize the method we evaluated variable amounts of the parasite and benznidazole as a test drug with a combination of high concentration and short time , as well as different computational approaches for the processing of the laser Biospeckle interferometry images , using flat test wells of VDRL plates as a novel sample container for this technique . The VDRL plate was recommended as a specific and reproducible system by the World Health Organization in 1980 [11] . The design and dimensions of the VDRL flat test wells are adapted to optical microscopes and magnifiers and they are available in most health centers worldwide . Similarly , other authors have adapted flat-bottomed 96 well micro-plates to quantifying the effects of drugs , antibodies and gene modifications on parasite fitness and replication rates of the human malaria parasite Plasmodium falciparum , another agent of a tropical disease [12] .
VDRL Plates were used with the following supplier’s information: each plate of pressed glass has the dimensions 89x57x4 . 5mm , with 12 numbered cavities of approximately 1 . 5mm depth and 15mm external diameter with overflow grooves . In order to test reproducibility , the internal diameter of the cavities , 60 wells ( 5 plates ) were measured under a calibrated magnifier , obtaining a Mean internal diameter of 12 . 7+/- 0 . 1 mm . With these dimensions and the approximate size of the parasite ( length 20μm and width 5μm , mean 12 . 5μm ) , the object’s plane ( surface disk containing the most exposed parasites ) available for the Biospeckle technique has a mean height of 0 . 0125 mm , a mean volume of 1 . 66μL and contains approximately 670 parasites when the density is 4x105 parasites per mL ( 4x104 parasites per well ) . Unless otherwise stated , all the assays were performed with an initial volume of 100μL . Trypanosoma cruzi strain M/HOM/ VE/67 /EP-67 was used . The parasites were grown in Liver Infusion Tryptose ( LIT ) medium with 5% Neonatal Bovine Serum [13] at 28°C . Just before performing each experiment , parasites were counted in a Neubauer chamber and used when cultures were at exponential growth ( usually 2-6x105 parasites/mL ) [14] . In order to estimate desiccation of the LIT medium at room temperature ( 20–25°C ) , each one of 12 VDRL test wells in 4 different plates , was filled with 100μL of LIT medium , thus each plate had a total volume of 1200μL . At four different times ( 0 , 55 , 120 and 270 minutes ) one of the plates was used to measure the total volume and estimate the desiccation process . The results were compared with two separate experiments in which the activity of the Biospeckle pattern of a well with parasites in LIT medium and a well with LIT medium without parasites was taken at different times , but within a similar period of time as the experiment for the desiccation of the liquid . Assay 1 was performed with and without 2 . 4x104 parasites per well and the activity of the Biospeckle pattern was taken at 8 different times ( 1 , 18 , 35 , 52 , 70 , 90 , 107 and 120 min ) , recording 16 videos , two at each time point . Assay 2 was performed with 4x104 parasites per well and the activity of the Biospeckle pattern was taken at 4 different times ( 1 , 61 , 128 and 204 min ) . The data of this assay correspond with the benznidazol experiment discussed below . The ability of the method to discriminate different parasite quantities was tested in two ways . In the first experiment , parasites were adjusted to 4x104 parasites/mL , and each well of the VDRL plate was filled with 100μL of LIT medium containing different amounts of the parasite ( from 0 to 4x103 parasites/100μL ) , varying the density of the parasites within the 100μL of each well . In the second experiment , parasites were adjusted to 4x105 parasites/mL , and each well of the VDRL plate was filled with different amounts of LIT medium , changing the volume of liquid within each well ( from 13 . 3 to 133μL ) , thus varying the amount of parasites ( from 0 . 532x104 to 5 . 32x104 total parasites per well ) , while keeping the density constant ( 4x105 parasites/mL ) . In each case a video was recorded ( 10 total videos ) . This assay was designed as is shown in Table 1 . In both experiments , the activity of the Biospeckle pattern was taken using the three approaches for image processing described below . Since the object’s plane has a constant volume of 1 . 66μL due to the parasite’s dimensions ( length 20μm and width 5μm , mean 12 . 5μm ) , the percentage that it represents in the constant density experiment increases as the total volume of liquid in the well decreases . With the results of the constant density experiment , the activity of the Biospeckle pattern was evaluated in terms of the percentage of the object’s plane . In order to assess the statistical robustness of the quantitative assay , the Biospeckle pattern of repetitions of three parasite concentrations was evaluated and compared to the Biospeckle pattern of LIT medium without parasites . For this purpose , 6 different wells with LIT medium and 2x104 parasites per well , 6 different wells with LIT medium and 3x104 parasites per well , 9 different wells with LIT medium and 4x104 parasites per well and 11 different wells with LIT medium without parasites from different experiments , were evaluated . In each case a video was recorded ( 32 total videos ) and the activity of the Biospeckle pattern was calculated in each case . The statistical analysis was performed with ANOVA , Tukey post hoc and Descriptive Statistics using R and R Commander . For the ANOVA model the null hypothesis states that the mean of the four groups is equal . Therefore , the alternative hypothesis states that at least the mean of one of the four groups is statistically different . The Tukey post hoc test is performed if the null hypothesis of the ANOVA Model is rejected , in order to find the combinations by pairs that have statistically different means . R and R Commander were used to perform these tests . A similar analysis was performed with the data of the activity of the Biospeckle pattern as a function of the concentration of benznidazole , with and without parasites: descriptive statistics , ANOVA , Tukey post hoc , Bartlett’s test of homogeneity of variances and Levene's test for homogeneity of Variances . Benznidazole was used as a trypanocidal reference drug . It was provided as a generic drug by the Health Public System in Venezuela . Each well of the VDRL plate was filled with 100μL of LIT medium in the presence or absence of 4x104 epimatigotes per well , containing 7 different concentrations of benznidazole: 0 , 1 . 95x10-2 , 7 . 81 x10-2 , 31 . 25 x10-2 , 1 . 25 , 5 and 20 μg/mL ( ranging from 0 . 075 to 76 . 8μM ) and at different times ( 1 , 61 , 128 and 204 minutes ) a video was recorded ( 28 videos with parasites and 24 videos without parasites ) and in each case the activity of the Biospeckle pattern was calculated . In the plots with a logarithmic scale the 0 concentration values are omitted . In some cases the concentrations are presented with either two or more decimals . Fig 1 shows the scheme of the experimental setup . The laser Biospeckle imaging system consists of a 1mW He-Ne laser ( unpolarized ) operating at 632 . 8nm which is coupled to a convex divergent lens to form a spot of approximately 10mm onto individual wells in the VDRL plate . The laser is located at a distance of 50cm and the beam has an angle of incidence of 72° . Since the wells have an internal diameter of 12 . 7+/- 0 . 1mm and the laser beam makes a spot that has a maximum diameter of 10mm , the beam illuminates the center of the well thus avoiding edge effects . The VDRL plate is put on top of an opaque black cardboard fixed on an anti-vibration table . A CCD camera ( Thorlabs USB . 2 , 30 fps , 6 . 45-μm Square Pixels ) is located at 30 cm from the sample and is connected to a PC which records 0 . 5–1 min videos . The resolution of the camera is 1280 × 1024 Pixels , and its optical system consists in fixed focal lengths of 3 . 5–75mm with maximum aperture of up to f/0 . 95 , as well as an 18–108 mm f/2 . 5 zoom lens . The high magnification zoom lenses are made up of a modular system that features magnification from 0 . 07 to 28 . The speckle video data is sent to the computer for video and image processing . With this set up , the illuminated region completely occupies the image in a homogeneous manner so that the whole image is subject to analysis and processing . Fig 1 shows the dimensions of the VDRL well . In order to assign numerical values to measure the activity of the dynamic Biospeckle patterns , three approaches were taken , which are based on modifications of the Temporal Difference Method [8] . In all cases , whole successive images of a video were processed . This could be achieved because the setup was designed so that the active region of the image occupies the whole frame in a nearly homogeneous manner so as to be able to consider and compare whole images . In the first approach , Eq ( 1 ) was developed as an algorithm of averages that analyzed all the frames in one video sequence . The frames were downloaded using a frame-by-frame mode in the image processing unit and transformed to a gray scale . Eq ( 2 ) was used for the purpose of comparison with Eq ( 3 ) . In both cases , A and B are measures of mean intensity ( I ) , A being expressed as an averaged mean intensity and B being expressed as the sum of the mean intensities: A=∑n=1N〈|In−In+1|〉/N−1 ( 1 ) B=∑n=1N〈|In−In+1|〉 ( 2 ) where | . | is the absolute value , < . > is the mean value , I is intensity in gray scale and N represents 1500 total frames analyzed . In the case of Eq ( 1 ) , the mean intensity of the absolute value of a subtraction pixel by pixel of whole successive images , is averaged to obtain A , when a video has 1500 images , there will be 1499 mean differences ( N-1 ) . With this algorithm , successive images are subtracted , transformed to the absolute value of the difference , the mean for the absolute values of the intensities of the differences is obtained and finally the means are averaged . It should be noted that with this algorithm , the final number corresponds to the average of means . In the case of Eq ( 2 ) the same algorithm is performed adding the mean intensities of the differences , without the final average . In the case of Eq ( 1 ) , this was achieved with a free software program , Image Delta Processor ( named in this work as ImageDP ) , written for this purpose in Java , which is provided as Supporting Information ( S1 Text Documentation ImageDP , S2 Text Source ImageDP , S3 Text Settings ImageDP , S4 Text Project ImageDP , S5 Text Classpath ImageDP ) . Eq ( 2 ) , since it was only used for testing the linearity of the method , it was worked out using the program ImageJ . In the second approach , Eq ( 3 ) , a modified version of Eq ( 2 ) was used applying the program ImageJ ( a public domain Java image processing program developed at NIH , USA ) . In the third approach , a script using R Commander/RSAGA/SAGA GIS ( named in this work as R/SAGA ) was designed with the same Eq ( 3 ) . R is a free software environment , R commander is its graphical user interface and SAGA-GIS ( System for Automated Geoscientific Analysis-Geographic Information System ) is a Free Open Source Software . This software is provided as Supporting Information ( S6 Text Script R/SAGA , S7 Text Manual R/SAGA ) . C is a measure of the mean intensity ( I ) , being expressed as the sum of the intensities , as explained below: C=∑n=1N−1|In−In+1| ( 3 ) With ImageJ , a set of frames was downloaded , typically 11 whole frames were taken . The addition of the difference between the consecutive frames leads to the construction of a new image from which the mean intensity ( I ) is taken . It should be noted that in this case , 11 successive whole frames are subtracted creating 10 new images containing the absolute values of the intensities of the differences , then the 10 new images are added to obtain a resulting image from which the mean intensity is taken . The same algorithm is carried out with R/SAGA for which a set of 11 frames is imported with SAGA and processed in a Linux environment . Then , in a second step with R Commander 10 new images are created by taking the absolute value of the difference between consecutive frames . It should be noted that only “Channel 1” is taken to calculate the absolute value of the difference , so as to exclude two of the color channels or bands and therefore , to make the calculation in gray scale . Finally , the 10 difference images are imported and added in SAGA , generating a final image from which the mean intensity is taken . In this work the activity of the Biospeckle pattern refers to the values of A , B or C , depending on the equation that is selected for calculation . Also , since very different programs and algorithms are used , the absolute values may be different but the relations are comparable . The frames for calculation were sometimes selected from a segment of the video avoiding vibrations from stirring or external interferences but they usually corresponded to the region that goes from more than 3 seconds ( to allow stabilization ) to less than 25 seconds . In order to test the linearity of each equation , in one case up to 100 frames of the same video were processed by the algorithm of the three equations with ImageJ and by the algorithm of Eq ( 3 ) with R/SAGA . In all these cases , the same set of frames was processed from a video with a high activity of the Biospeckle pattern from the experiment with parasites and benznidazole ( 0 . 3125μg/mL or 1 . 2μM ) taken at 1min . Also , as a comparison , a set of frames was taken from a video with a low activity of the Biospeckle pattern from the experiment with benznidazole ( 0 . 3125μg/mL or 1 . 2μM ) without parasites taken at 1min , and processed with Eq ( 3 ) with R/SAGA and with ImageJ . In order to illustrate the difference in the activity of the Biospeckle pattern , 2 , 10 or 30 frames were included in the calculation of 1 , 9 or 29 image differences with ImageJ and Eq ( 3 ) for the videos of high and low activity of the Biospeckle pattern , respectively and six surface plots were constructed .
The effect of parasite quantity on the activity Biospeckle pattern was evaluated by conducting two experiments ( constant density and variable density ) and analyzing in each case , the same video with the three approaches for handling of image data . In the experiment of constant density 5 different test wells had 4x104 parasites/mL but the volume in each well varied from 13 . 3 to 133μL and the parasite number varied from 0 . 532 to 5 . 32x104 parasites per well . As is shown in Fig 2 , the three approaches are able to detect an increase in the activity of the Biospeckle pattern as a function of parasite quantity . Moreover , the plots are nearly linear using the three approaches . In the three cases , an increase in the number of parasites obtained by increasing the volume and thus the height of the liquid column in the assay well ( constant density ) , shows linearity . This indicates that if the density is kept constant , the Biospeckle pattern is proportional to the quantity of parasites in the test wells within the evaluated range . In other words the Biospeckle pattern increases as a function of parasite quantity , despite the fact that the density of the parasites in the test wells and in the object’s plane is constant . Therefore , the system is detecting the parasites in the whole well . In the experiment of variable density 5 different test wells had 100μL of LIT medium containing from 0 to 4 x103 parasites per well . With two of the approaches ( R/SAGA and ImageJ with Eq ( 3 ) ( Fig 3 ) , a linear relationship is obtained up to 3x103 parasites in the assay well . Moreover , comparing the constant and variable density experiments , a linear behavior is observed in both cases . In contrast with this finding , with the approach ImageDP , an increase in the quantity of parasites by increasing the density and thus the amount of parasites in the object’s plane shows a non-linear behavior . This lack of linearity of ImageDP was used as one of the criteria to exclude this approach as the selected image processing method for the following experiments . With the other two methods , either ImageJ or R/SAGA with Eq ( 3 ) , the activity of the Biospeckle pattern of both experiments is a function of the quantity of parasites within the well . Since the experiment for variable density is performed with 0 to 4 x103 parasites in 100μL per well and the experiment for constant density is performed with 0 . 532 to 5 . 32x104 parasites in a variable volume per well , and if the method is able to detect the total number of parasites in the well irrespective of the volume , then it should be possible to find a good relation between the activity of the Biospeckle pattern as a function of the number of parasites . Fig 4 shows that the number of parasites of both experiments can fit a linear regression . In order to further examine the proposed equations , a comparison of algorithms and software was performed , as is shown in Figs 5 and 6 . Since ImageDP uses all the photograms in a video ( 750–1500 frames ) , in order to compare the result of using the equations on a selected set of photograms ( up to 100 frames ) , only ImageJ or R/SAGA and not ImageDP , were used to analyze the same segment of a video . Figs 7 and 8 show two frames taken from a video with high Biospeckle activity and low Biospeckle activity , respectively . Using the program ImageJ , Eqs ( 2 ) and ( 3 ) show similar curves when the calculation includes up to 40 frames ( Fig 5 ) but if more frames are included there is a saturation effect when using Eq ( 3 ) . Since both equations were worked out with the same program ( ImageJ ) , the difference between the curves must be due to the algorithms . In the case of Eq ( 2 ) , B corresponds to the sum of numbers , each one representing the mean obtained after subtracting consecutive frames . In the case of Eq ( 3 ) , C corresponds to the mean intensity of an image that was constructed by adding the images that resulted by taking the difference between consecutive frames . Therefore , in the former , there is the addition of numbers while in the latter there is the addition of image rasters , showing that in this case , there is a saturation with non-linear behavior as more frames are included in the calculation , possibly due to the fact that this is an image processing program with a radiometric resolution with a limit of intensity at a value of 256 . Fig 5 , also shows that if the video has a lower activity of the Biospeckle pattern as would be expected in the case of LIT medium without parasites , there is no saturation with the same number of frames , suggesting that the number of frames required to reach this saturation effect depends on the activity of the Biospeckle pattern of the selected video . This effect is further analyzed in Fig 9 which shows the surface plots of the images constructed for the curves obtained with ImageJ and Eq ( 3 ) of Fig 5 , with 2 , 10 and 30 frames ( 1 , 9 and 29 image differences ) included in the calculation , avoiding the saturation that occurs after 40 frames . A marked difference is observed when comparing the surface plots on the right ( Fig 9 . d , 9 . e and 9 . f ) which were obtained with LIT medium without parasites , with the surface plots on the left ( Fig 9 . a , 9 . b and 9 . c ) which were obtained with LIT medium with parasites , with the same concentration of benznidazole , in both cases . As is shown in the left panel of Fig 9 , the presence of parasites increases the activity of the Biospeckle pattern , which explains the saturation that is obtained in Fig 5 . Conversely , as is shown in the right panel of Fig 9 , the absence of parasites shows low activity of the Biospeckle pattern , which explains that there is no saturation ( Fig 5 ) , at least in the tested interval . As shown in Fig 5 , with Eq ( 2 ) a linear behavior is obtained up to at least 80 frames , however , if this algorithm is worked out as Eq ( 1 ) as would be the case for the program ImageDP , performing the average , the value tends to decrease as more frames are included in the calculation for image processing ( Fig 6 ) . Since the program ImageDP was designed for Eq ( 1 ) and for processing all the frames in a video , we assume that the final value ( A ) will be affected by this decrease and will be a fraction of the value obtained at the beginning of the video . This observation could explain the shape of the curve of variable density of parasites ( Fig 3 ) . Therefore , up to this point the results indicate that we are obtaining only approximate values because the algorithm of Eq ( 1 ) with ImageDP calculates decreased values and the algorithm of Eq ( 3 ) with ImageJ has the tendency to reach saturation . However , as is shown in Fig 5 , Eq ( 3 ) works well with R/SAGA , in both cases with and without parasites , showing a linear relation up to 100 frames . Since usually no more than 20 frames are selected for the calculation , it can be expected to provide an adequate representation of the Biospeckle pattern and this is in agreement with the curves for constant and variable density of parasites of Figs 2 and 3 . Therefore , in the following experiments of this work , all the calculations were carried out using Eq ( 3 ) with R/SAGA . In order to test the relationship between the object’s plane and the activity of the Biospeckle pattern , we used the data of Table 1 and the constant density curve of Fig 2 . Fig 10 shows the activity of the Biospeckle pattern ( with R/SAGA Eq 3 ) as a function of the increase of the calculated percentage of the volume of the disk of the object’s plane with respect to the total volume of the liquid in the well , in the experiment in which the total volume decreases . It shows that the intensity decreases as the percentage of the disk increases , indicating that as it was stated above , since the density is constant and the percentage of the object’s plane increases , the system is detecting the total amount of parasites in the well . Applying an ANOVA Model on the activity of the Biospeckle pattern on 32 values of four groups of parasites per well ( 0 , 2x104 , 3x104 and 4x104 ) , the null hypothesis stating that the mean of the four groups is equal , is rejected because the p-value ( 2x10-16 ) is less than the significance level of 0 . 05 . As is shown in Supporting Information ( S1 Data Statistical Analysis ) , the assumptions for normal distribution , homoscedasticity of the residuals and absence of autocorrelation , are satisfied . Thus a Tukey post hoc test was carried out in order to explore the differences among the four groups of data , compared in pairs . As Table 2 shows , all the p-values for the six combinations are less than the significance level of 0 . 05 which indicates that the mean value between pairs , is statistically different . Finally , Descriptive Statistics tests were performed on each group of data , obtaining the results shown in Table 3 , where parameters such as Mean and Standard Deviation of each group can be obtained , as well as the indicators of the type of distribution of the data . The desiccation of the liquid in the air exposed VDRL wells and its implications on the activity of the Biospeckle pattern was evaluated in several ways . Fig 11 shows the desiccation in terms of remaining volume , after 0 , 55 , 120 and 270 minutes have elapsed . In the same Figure two assays ( Assay 1 and Assay 2 ) show the activity of the Biospeckle pattern in separate wells with and without parasites and processed under R/SAGA and Eq ( 3 ) . The results indicate that although there is a marked desiccation effect , the Biospeckle pattern remains stable up to around 200 minutes , indicating that as was seen in Figs 2 , 3 and 4 ( Effect of Parasite Quantity ) , the Biospeckle pattern under the conditions of this work , reveals the total number of parasites within the well of the VDRL plate . However , it should be noted that Assay 1 shows a more variable profile than Assay 2 , suggesting that the lower number of parasites and the repeated irradiation of the same sample well with the laser beam may have an effect on the activity of the Biospeckle pattern , in both cases , with and without parasites . These assays allowed the selection of the conditions of the method using VDRL plates . We selected the following conditions: 100μL in the test well , parasites in exponential phase , 4x104 parasites per well , a maximum time of evaluation of the activity of the Biospeckle pattern of 204 minutes . The recommended method for image processing is R/SAGA with Eq ( 3 ) . With these conditions , the effect of varying amounts of benznidazole on the activity of the parasites was tested and compared to the pattern obtained from the LIT medium with benznidazole and without parasites . Videos of each well were taken at different times . Figs 12 , 13 , 14 , 15 and 16 show the curves obtained for the activity of the Biospeckle pattern of 4x104 parasites per well , at times of 1 , 61 , 128 and 204 minutes with six concentrations of benznidazol ranging from 0 . 0195 to 20μg/mL , and as a comparison , the curves obtained with benznidazole without parasites . Fig 12 shows that there is a distinct difference between the curves with and without parasites , as would be expected from the statistical analysis ( Table 3 ) in which there was a significant difference among the means for samples with these two conditions . In this case the assays in the presence of growing benznidazole concentrations with and without parasites , show activities of the Biospeckle pattern that are statistically different with means of 46 . 02±5 . 27 and 22 . 76±3 . 95 , respectively ( see S1 Data ) . The effect of benznidazole on T . cruzi was analyzed from two perspectives , concentration and time of action of the drug . The activity of the Biospeckle pattern as a function of benznidazole concentration shows curves that fit similar Polynomial regressions at all the tested times ( Fig 13 ) with a decrease of the activity as a function of concentration and a clustering of the values ( mean of the clustered low concentration values of 48 . 39±2 . 91 , see S1 Data for details ) of the four time curves up to a concentration of benznidazole of 1 . 25μg/mL . Fig 14 which includes the activity of the Biospeckle pattern in the absence of benznidazole , shows that at the low concentrations the four time curves have a similar profile . However at greater concentrations of benznidazole ( Fig 13 ) , there is a dispersion of the values ( mean of the dispersed high concentration values of 40 . 12±5 . 33 , see S1 Data for details ) of the activity of the Biospeckle pattern . In the absence of parasites the mean values are 22 . 50±4 . 53 and 23 . 28±2 . 61 for the lower and higher concentrations of benznidazole , respectively . As shown in S1 Data , in the presence of parasites the mean value of the activity of the Biospeckle pattern at the high concentrations of benznidazole is statistically different from the mean value of the activity of the Biospeckle pattern at the low concentrations of benznidazole . Conversely , without parasites the means for both groups ( low and high concentrations of benznidazole ) , are statistically equal . All this indicates that the growing concentrations of benznidazole cause an activity window that can be detected by the statistically lower mean value only in the presence of parasites . Moreover , the variability of the four groups is statistically equal as is shown by Bartlett’s test and Levene's test for homogeneity of variances which show p-values that are higher than 0 . 05 , indicating that the variances are statistically equal . On the other hand the behavior of the activity of the Biospeckle pattern at the shorter times was explored plotting only the curves for 1min and 61min for all the concentrations of benznidazole . As shown in Fig 15 , there is an instantaneous effect at 1min that remains with similar activity values at least for one hour for all the benznidazole concentrations . The profile of these two early curves is different from the two late curves ( 128min and 204min ) only at the higher concentrations ( 5 and 20μg/mL ) , Fig 13 . Analyzing the time of action of the drug , the same data were used in Fig 16 to evaluate the effect of the elapsed time on the activity of the Biospeckle pattern for each concentration of benznidazol , with and without parasites . In this case the lower values show a higher activity of the Biospeckle pattern than the higher values of benznidazol at all the tested times . Moreover , at the lower concentrations , the activity remains stable up to 204 minutes . At the higher concentrations ( 5 and 20μg/mL ) , the activity of the Biospeckle pattern is stable up to 61 minutes and then decreases reaching final values that approach the activity of LIT medium without parasites . The results of Figs 12 , 13 , 14 , 15 and 16 suggest that benznidazole produces an instantaneous effect at 1min on T . cruzi that occurs at all the tested concentrations , with a profile that remains almost the same up to at least 61min , as shown by the polynomial tendencies ( Fig 13 ) and the resulting profile of Fig 15 . Furthermore , the activity of the Biospeckle pattern is similar for all the tested times up to a concentration of 1 . 25μg/mL ( 4 . 8μM ) . At the higher concentrations of 5 and 20μg/mL ( 19 . 2 and 76 . 8 μM , respectively ) the elapsed time seems to have a more relevant effect on this activity . This reveals that there are two effects on T . cruzi , one at the lower concentrations and another at the higher concentrations . The former is less affected by the elapsed time than the latter , which could be explained by the combination of elapsed time and higher concentrations of the drug . It can be assumed that desiccation may interfere by increasing drug concentration at the longer times . However the activity of the Biospeckle pattern at 1 . 25μg/mL is similar at all the tested times in spite of the fact that desiccation affects each time curve differently . The highest benznidazole concentrations ( 20μg/mL , 76 . 8μM ) could be interfering with the assays that were performed at 1 , 61 and 128min . At 204min it is possible that this interference caused by the high concentration of the drug may be neutralized by a decrease of the activity promoted by the elapsed time . As these results show , the present method reveals in a relatively short time , that the effect of benznidazole is dependent on the elapsed time and concentration of the drug , as has been demonstrated by other authors in assays that require a much longer time [3] . A closer look at Figs 12 , 13 , 14 , 15 and 16 reveals that all the time curves have the smallest value of the activity of the Biospeckle pattern in a range that goes from 5 to 20μg/mL ( 19 . 2 to 76 . 8μM , respectively ) . This range of values represents the concentration of benznidazole that gives maximal inhibition under the assayed conditions . However , 1 . 25μg/mL ( 4 . 8μM ) could be considered as the inflexion point that suggests the existence of a minimal concentration that affects the activity of the Biospeckle pattern , separating the effect of benznidazole at low concentrations from the effect at higher concentrations . Traditional IC50 determination requires a measure of the viability of the organism as a function of concentration of the tested drug . In this work the Biospeckle pattern is taken as a measure of the activity of the parasites and the relation of this pattern to viability and drug action has to be determined . Rodrigues et al [3] report approximately 12μM for the lowest survival index and an IC50 value of 8 . 8±0 . 04μM for benznidazole acting on epimastigotes in LIT culture medium in a period of 96 hours . It is important to point out that in the traditional in vitro assays with Trypanosoma cruzi [4] , benznidazole is usually used in a range of 0 . 156 to 5μg/mL ( 0 . 6 to 19 . 2μM ) , and the parasites are counted after 96 hours . In the present work , the concentration of benznidazole is used in a range of 0 . 0195 to 20μg/mL ( from 0 . 075 to 76 . 8μM ) and the activity is detected in a minimum time of 1 minute .
In this work we are describing a quantitative research aiming at parametric calibration . In this case , the number of observations n is greater than the number of estimated parameters which is in agreement with Gujarati and Potter [15] for the adjustment of a parametric model . A quantitative method for the evaluation of the activity of Trypanosoma cruzi , using relatively small volumes and short times has been designed . VDRL plates were chosen because they are widely known and used for other purposes and assays and because the liquid contained in the assay well is flat shaped and therefore , has a minimal distortion with the optical setup and the recorded video . For image processing two levels of complexity have been designed . The use of an image processing program such as ImageJ to work out Eq ( 3 ) provides a readily available and easy to use system for the analysis of Biospeckle patterns because this is software released by NIH that is in the public domain . However , due to its tendency to reach saturation with the designed algorithm , this method provides only approximate values depending on the intensity of the Biospeckle pattern . Hence , the upper limit of the linear tendency ( lower limit of saturation ) , could be used to qualify the relative Biospeckle activity recorded in a video . Furthermore , the combination of ImageJ with Eq ( 3 ) gives approximate values when the Biospeckle pattern is high , but it could be the appropriate choice if the activity of the Biospeckle pattern is low , such as in the case of a low parasite density . On the other hand , ImageDP and R/SAGA can be used for calculating the values of the Biospeckle pattern , but in both cases a program had to be designed for this purpose . In the case of Eq ( 1 ) and ImageDP , the program is suitable for high parasite densities and significant differences among samples , due to its tendency to decrease the real value of the Biospeckle pattern . In contrast , with Eq ( 3 ) and R/SAGA a linear profile is obtained . Due to this and to the fact that there is no distortion up to 100 processed frames , R/SAGA is the approach recommended for the evaluation of the activity of T . cruzi using VDRL plates with the selected conditions that are reported in this work . However , the three approaches tested in this work are useful for the evaluation of the activity of the Biospeckle pattern of T . cruzi in the VDRL plate . ImageJ with Eq ( 3 ) works well if the Biospeckle pattern is low , ImageDP with Eq ( 1 ) is suitable for videos with a high activity and R/SAGA with Eq ( 3 ) has shown up to now , the best way to describe slight differences of high and low activity videos suggesting that it is the best to test the effect of a drug on a population of parasites . It should be noted that with the combination of the optical setup and R/SAGA with Eq ( 3 ) , the system is able to detect all the parasites that are present in the test well , being able to quantify the differences among at least two orders of magnitude . Furthermore , in the tested range , the activity of the Biospeckle pattern is proportional to the number of parasites present within the well , regardless of the amount of liquid which can vary as a result of assay volume or as a result of desiccation . This observation is relevant when considering the action of drugs on the parasites due to the fact that different drugs will have different modes of action , altering the motility , the shape , the aggregation and even the number of live parasites within the well . Since the drug treated parasites will still be present in the well and may still be motile and active , it may be necessary to use an image processing method that is able to detect subtle changes . In the case of R/SAGA since the presence of parasites is significantly different to the absence of parasites as more frames are included in the calculation ( Fig 5 ) , it is possible that using up to 80 frames the small differences that are caused by a drug , may be more effectively detected . Although statistically the method has shown to be reproducible and acceptable for the quantification of T . cruzi epimastigotes and the differential effect of high and low concentrations of benznidazole on the activity of the Biospeckle pattern , there is much need of improvement in areas such as technical refinement , automation of the algorithms and interaction of interdisciplinary orientations . Focusing on the parasite , it is important to consider that two modes of motion have been described for T . cruzi epimastigotes , a persistent ( a quasi rectilinear fashion ) and a tumbling mode ( more complex and difficult to describe ) [16] . The selection of the best assay and image processing conditions must take into consideration the assumption that the activity of T . cruzi epimastigotes could be described by both types of motion under regular culture conditions . The switching between different modes of motion is associated to chemotaxis or some other types of taxis [16] . Thus , although it has not been experimentally proved , the findings in this work suggest that T . cruzi epimastigotes experience some kind of taxis in the presence of benznidazole which could explain the effect of the drug on T . cruzi epimastigotes , in a very short time . This statement is in agreement with the suggestion that deviations from a usual motility pattern could be employed as an indicator of drug activity in a drug test [17] . Benznidazole is a pro-drug that has to be converted to the active drug . In T . cruzi benznidazole is metabolized by an NADH-dependent , mitochondrial localized type I nitro-reductase rendering the cytotoxic and mutagenic agent glyoxal [18] [19] . However , it has also been shown that T . cruzi has a TcABCG1 transporter protein , located on the plasma membrane for which a nucleotide binding , a membrane and a transmembrane domain have been described . This protein is involved in a variety of translocation processes including benznidazole [20] . Transport processes are rapid and efficient mechanisms and could respond to intracellular and extracellular signals [21] . This probably constitutes the first line of response for pro-drug uptake and could represent a type of taxis which acts in a very short time . From the activity curves for benznidazole on T . cruzi , we could presume that we are in the presence of a two-step mechanism , one that is an almost immediate response to the presence of benznidazole and another that occurs after the pro-drug has been internalized . Whether the activity of the Biospeckle pattern is detecting these two steps has to be further investigated . Before this , it would be necessary to establish the relationship between viability and the activity of the Biospeckle pattern , the possible optical interferences of high drug concentrations which should be avoided especially in those cases in which desiccation is also an issue to be taken care of , the requirement of this method to measure activity in a period of time that may not be long enough to allow the expression of the complete mechanism of action in order to be comparable with the viability tests performed at 96 hours and the requirement for a metabolic condition of the parasites that should be standardized since this may have implications on motility and aggregation , thus on the activity of the Biospeckle pattern . Also it would be interesting to perform a parametric analysis on trypomastigote/amastigote stages with the proposed method . Finally , it is observed that with this method it might be difficult to design an experiment to determine an IC50 due to the fact that the meaning of the activity of the Biospeckle pattern with respect to viability is unknown . However , what is known is that , at concentrations of benznidazole above 1 . 25μg/mL ( 4 . 8μM ) , there are important changes in the profiles of the responses , suggesting that this can be considered as the minimal concentration that affects the activity of the Biospeckle pattern , under these conditions .
A method has been designed for the evaluation of the activity of the Biospeckle pattern of Trypanosoma cruzi , using relatively small volumes , short times , VDRL plates , an optical setup with a laser and a camera and an image processing method with R/SAGA . It is concluded that the activity of the Biospeckle pattern is proportional to the quantity of parasites present in the assay well and although it is susceptible to desiccation , the measure is stable up to around 200 minutes . With this quantitative method it is possible to statistically differentiate the amount of parasites along two orders of magnitude and although with certain limitations , it is possible to evaluate the effect of benznidazole on T . cruzi . In spite of the limitations , an effect of benznidazole on the activity of the Biospeckle pattern of T . cruzi epimastigotes can be detected at concentrations higher than 1 . 25μg/mL which is statistically different from the effect at lower concentrations . This effect is better detected after 1 hour of drug action . Further studies are necessary to elucidate how exactly the activity of the Biospeckle pattern is related to viability and drug effect . | Biospeckle refers to a pattern that occurs when a laser beam illuminates a dynamic surface , such as a liquid that contains microorganisms . The movement or the roughness of its surface causes the wave fronts to interfere and produce a pattern of moving dots that resemble boiling water . This research describes the application of Biospeckle to Trypanosoma cruzi , the parasite that causes Chagas disease . The purpose was to observe the movement of the Biospeckle dots and to detect differences depending on the presence of the parasite , the quantity of the parasite and the conditions of the parasites when they are affected by a drug . We designed a method using VDRL plates where the sample has a relatively small volume and is flat shaped , a laser , a camera and a lens . The Biospeckle pattern is recorded in a video in a computer and shows the Biospeckle dots which move rapidly as the concentration of parasites increases and less rapidly as the concentration decreases or as the parasites are affected by a drug such as benznidazole . We designed algorithms which take the difference between successive frames and expressed them in a program in Java , in a script in R Commander and SAGA and in ImageJ . Thus we obtained a quantitative description of the movement of T . cruzi . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion",
"Conclusions"
] | [
"medicine",
"and",
"health",
"sciences",
"engineering",
"and",
"technology",
"signal",
"processing",
"applied",
"mathematics",
"lasers",
"microbiology",
"parasitic",
"diseases",
"protozoan",
"life",
"cycles",
"parasitic",
"protozoans",
"drug",
"screening",
"simulation",
... | 2016 | Quantitative Laser Biospeckle Method for the Evaluation of the Activity of Trypanosoma cruzi Using VDRL Plates and Digital Analysis |
In sub-Saharan Africa , non-typhoidal Salmonella ( NTS ) are emerging as a prominent cause of invasive disease ( bacteremia and focal infections such as meningitis ) in infants and young children . Importantly , including data from Mali , three serovars , Salmonella enterica serovar Typhimurium , Salmonella Enteritidis and Salmonella Dublin , account for the majority of non-typhoidal Salmonella isolated from these patients . We have extended a previously developed series of polymerase chain reactions ( PCRs ) based on O serogrouping and H typing to identify Salmonella Typhimurium and variants ( mostly I 4 , [5] , 12:i:- ) , Salmonella Enteritidis and Salmonella Dublin . We also designed primers to detect Salmonella Stanleyville , a serovar found in West Africa . Another PCR was used to differentiate diphasic Salmonella Typhimurium and monophasic Salmonella Typhimurium from other O serogroup B , H:i serovars . We used these PCRs to blind-test 327 Salmonella serogroup B and D isolates that were obtained from the blood cultures of febrile patients in Bamako , Mali . We have shown that when used in conjunction with our previously described O-serogrouping PCR , our PCRs are 100% sensitive and specific in identifying Salmonella Typhimurium and variants , Salmonella Enteritidis , Salmonella Dublin and Salmonella Stanleyville . When we attempted to differentiate 171 Salmonella Typhimurium ( I 4 , [ 5] , 12:i:1 , 2 ) strains from 52 monophasic Salmonella Typhimurium ( I 4 , [5] , 12:i:- ) strains , we were able to correctly identify 170 of the Salmonella Typhimurium and 51 of the Salmonella I 4 , [5] , 12:i:- strains . We have described a simple yet effective PCR method to support surveillance of the incidence of invasive disease caused by NTS in developing countries .
In industrialized countries , non-typhoidal Salmonella ( NTS ) constitute a well recognized public health problem that in healthy subjects is overwhelmingly encountered clinically as self-limited gastroenteritis [1] , [2] . In immunocompromised and debilitated hosts , NTS can become invasive , leading to bacteremia , sepsis and focal infections ( e . g . , meningitis ) [2] , [3] . Among infants less than three months of age who become infected with NTS in industrialized countries , invasiveness is also occasionally observed , resulting in bacteremia and focal infections [4] . Interestingly , whereas systematic blood culture-based surveillance of febrile pediatric patients in Asia has clearly highlighted the high incidence of bacteremia associated with Salmonella enterica serovars Typhi and Paratyphi A in children residing in crowded urban settings [5]–[7] , isolation of NTS has not been common . In striking contrast , systematic blood culture-based surveillance and clinical studies of hospitalized and ambulatory pediatric patients <60 months of age with fever or focal infections in sub-Saharan Africa have documented the important role of NTS as invasive bacterial pathogens [8]–[17] . NTS constituted one of the three most common invasive bacterial pathogens in all these studies . Importantly , two serovars , Salmonella Typhimurium ( and Typhimurium variants ) and Salmonella Enteritidis have been reported to account for 79–95% of all bacteremic non-typhoidal Salmonella infections in sub-Saharan Africa [9] , [11]–[13] , [15] , [16] , [18] , [19] . Salmonella Dublin has been associated with a few percent of cases in some studies [12] , [13] but with a more substantial proportion in Mali [18] , where a fourth serovar , Salmonella Stanleyville , also accounted for a notable proportion of all isolates [18] , bringing the cumulative total to >95% of all strains . We previously developed a multiplex polymerase chain reaction ( PCR ) -based approach to identify the three main pathogens responsible for typhoid ( Salmonella Typhi ) and paratyphoid ( Salmonella Paratyphi A and Salmonella Paratyphi B ) fevers [18] . Three sequential PCRs identify strains of Salmonella serogroups A , B or D ( and Vi positive or negative ) ; strains that express Phase 1 flagellar ( H ) antigen types H:a , H:b or H:d; and strains incapable of fermenting d-tartrate ( d-T ) . By means of this PCR technology , Salmonella Typhi ( O serogroup D , Vi+; H:d ) , Salmonella Paratyphi A ( O serogroup A; H:a ) and Salmonella Paratyphi B ( O serogroup B; H:b; d-T non-fermenter ) strains were identified with 100% sensitivity and 100% specificity . Classical Salmonella serotyping methods identified the serovars of 336 NTS isolates from blood cultures of febrile children <16 years of age in Bamako , Mali , obtained in the course of systematic surveillance of children admitted to hospital or seen in the Emergency Department with fever or invasive infection syndromes [20]–[22] . Salmonella Typhimurium and “variants” ( mainly I 4 , [5] , 12:i:- ) , Salmonella Dublin , Salmonella Enteritidis and Salmonella Stanleyville were the most commonly isolated NTS [18] . Herein , we describe PCRs that when used in conjunction with the O serogrouping PCR described by Levy et al . [18] can identify Salmonella Typhimurium and variants ( O serogroup B; H:i ) , Salmonella Enteritidis ( O serogroup D , Vi-; H:g , m ) , Salmonella Dublin ( O serogroup D , Vi+ or Vi-; H:g , p ) and Salmonella Stanleyville ( O serogroup B; H:z4 , z23 ) with 100% sensitivity and 100% specificity . We anticipate that this methodology will be useful in reference laboratories and major clinical microbiology laboratories to identify Salmonella isolated from blood and other sterile sites in developing countries where robust PCR-based typing techniques are becoming increasingly popular and because high quality H typing sera are difficult to obtain , expensive and technically demanding to use .
The surveillance protocol and consent form were reviewed by the Ethics Committee of the Faculté de Médecine , Pharmacie et Odonto-Stomatologie , Université de Bamako , and by the Institutional Review Board of the University of Maryland , Baltimore . For any patient eligible for laboratory surveillance to detect invasive bacterial disease , informed consent was obtained prior to their enrollment; ∼95% of eligible subjects agreed to participate . Since the literacy rate in Bamako is <30% , as is customary practice for CVD-Mali clinical studies [20]–[22] , the consent form was translated into Bambara and several other local languages and the translations recorded on audiotape [20] . CVD-Mali personnel explain the study , including the objectives and risks and benefits associated with participation . The audiotaped version of the consent form is then played and any questions posed are answered . Once the parent or patient has had all questions answered and agrees to participate , this is documented on a printed consent form written in French . If the participant is illiterate , a witness who is present throughout the consent procedure completes the necessary portions and signs the consent form; the parent/participant marks the consent form ( either fingerprint or other notation ) . If the person is literate , then he/she may read and sign the consent form . This standard method of obtaining consent practiced by CVD-Mali was approved by ethics commitees in Mali and at the University of Maryland . Since July 2002 , clinical staff of the Centre pour le Développement des Vaccins du Mali ( CVD-Mali ) and l'Hôpital Gabriel Touré ( HGT ) have been conducting systematic surveillance to detect invasive bacterial disease among hospitalized children <16 years of age [20]–[22] . Age-eligible children presenting to the emergency department with fever ( ≥39°C ) or focal clinical findings suggestive of invasive bacterial infection ( meningitis , septic arthritis , etc . ) and requiring hospitalization are referred to CVD-Mali staff by the evaluating clinicians . A CVD-Mali physician obtains informed consent , records clinical and epidemiologic data , and obtains blood ( and other relevant fluids ) for culture in the HGT Clinical Bacteriology Laboratory . The child's clinician is promptly notified when a culture yields a bacterial pathogen . Salmonella Typhimurium strain 81 . 23500 , Salmonella Enteritidis strain CVD SE and Salmonella Dublin strain 06-0707 were used to develop the multiplex PCR . Twenty-four control strains which came from the Salmonella Reference Laboratory of the Centers for Disease Control and Prevention ( CDC ) , Atlanta , GA or the Center for Vaccine Development , Baltimore , MD have previously been described [18] . These strains were Salmonella serovars of various O serogroups ( B , C1 , C2 , D , E1 , O28 and O38 ) and H types ( b , c , d , h , i , g , k , l , m , p , s , t , v , y , z10 and z29 ) . Nine O serogroup B , Phase 1 flagella antigen H:i reference strains from the CDC were used to develop a PCR that discriminates between Salmonella Typhimurium and I 4 , [5] , 12:i:- ( Table 1 ) . The NTS test strains consist of 327 Salmonella serogroup B and D isolates that were originally obtained from the blood cultures of febrile patients at l'Hôpital Gabriel Touré in Bamako , Mali . These strains were identified by conventional microbiological and classical serotyping methods by the CVD and CDC , as previously described [18]; 69 isolates were O serogroup D , including 37 Salmonella Dublin and 32 Salmonella Enteritidis , and 258 isolates were O serogroup B . PCR was performed in 1× PCR buffer , 3 . 5 mM MgCl2 , 0 . 2 mM of dNTPs and 0 . 2 U of Invitrogen Taq DNA polymerase ( final volume of 25 µl ) in an Eppendorf Mastercycler® . The primer mixes contained primers at a concentration of 5 µM each ( final concentration of 0 . 2 µM ) except for FFLIB and RFLIA that were used at a concentration of 10 µM each and the positive control primers ( 16SF and DG74 ) that were used at a concentration of 2 . 5 µM each . For each PCR reaction , 1 . 0 µl of primer mix was used . Crude DNA was prepared by suspending 3 colonies in 100 µl water and boiling for 10 min followed by centrifugation at 16 , 000×g for 30 sec and purified DNA was prepared using a GNOME DNA kit ( QBIOgene , Irvine , CA ) according to the manufacturer's instructions , and 5 µl of DNA was used in each PCR . The cycling parameters of the multiplex PCR that detects H:i , H:g , p and Sdf I involved denaturation at 95°C for 2 min , followed by 25 cycles comprised of heating to 95°C for 30 sec , 64°C for 30 sec and 72°C for 15 sec , and a final step of 72°C for 5 min . The cycling parameters of the PCR that discriminates between Salmonella Typhimurium and I 4 , [5] , 12:i:- involved denaturation at 95°C for 2 min , followed by 25 cycles of 95°C for 30 sec , 64°C for 30 sec and 72°C for 1 . 5 min , and a final step of 72°C for 5 min . PCR products were separated on 2% ( w/v ) agarose gels , stained with ethidium bromide and visualized using a UV transilluminator .
Figure 1 shows that the primers within the multiplex PCR were able to clearly identify the appropriate NTS alleles . A 779-bp product was amplified from Salmonella Dublin ( fliC-gp ) , a 551-bp product was amplified from Salmonella Typhimurium ( fliC-i ) and a 333-bp product was amplified from Salmonella Enteritidis ( Sdf I ) . The internal positive control primers ( universal 16S rRNA gene primers ) amplified a 167-bp product from each strain . To preliminarily assess the specificity of the multiplex PCR assay , we tested 24 control Salmonella strains consisting of a range of serovars ( previously described in [18] ) in a blinded fashion ( Figure 2 ) . The multiplex PCR correctly identified Salmonella Typhimurium and Salmonella Cotham as H:i , Salmonella Dublin as H:g , p and Salmonella Enteritidis as containing Sdf I ( Figure 2 ) . Faint products of the size of Sdf I were observed for Salmonella Meleagridis and Salmonella Livingstone . However , Salmonella Meleagridis is O serogroup E1 and Salmonella Livingstone is O serogroup C1 , so when also tested by our previously described O serogrouping PCR [18] , these serovars would not be mistaken as Salmonella Enteritidis . The same is true for Salmonella Cotham , which although it possesses fliC-i , is not O serogroup B and would not be mistaken as Salmonella Typhimurium . Therefore , the new multiplex PCR was sensitive in terms of its ability to identify serovar Cotham as H:i and was specific , when combined with the O-serogrouping PCR , in showing that the strain was not serovar Typhimurium . We also blind-tested a sample of Salmonella Typhi and Salmonella Paratyphi A and B strains to ensure that the PCR would not detect these strains . The multiplex PCR correctly identified fliC-i of six Salmonella Typhimurium , Sdf I of four Salmonella Enteritidis , and fliC-g , p of five Salmonella Dublin strains but only the 16S rRNA gene was amplified from five strains each of serovars Typhi , Paratyphi A and Paratyphi B ( data not shown ) . We blind-tested 69 non-Typhi serogroup D Salmonella and 258 serogroup B strains that were originally obtained from the blood cultures of febrile patients at l'Hôpital Gabriel Touré in Bamako , Mali [18] with the multiplex PCR designed to identify Salmonella Typhimurium ( I 4 , [5] , 12:i:1 , 2 ) and variants ( monophasic I 4 , [5] , 12:i:- and non-motile ( NM ) I 4 , [5] , 12:NM ) , Salmonella Enteritidis and Salmonella Dublin . This PCR was performed in parallel to serotyping . We correctly identified all the serogroup D isolates ( 37 Salmonella Dublin and 32 Salmonella Enteritidis ) and all 232 Salmonella Typhimurium and variant strains ( Table 3 ) . If the Salmonella Typhimurium-like strains ( i . e . , I 4 , [5] , 12:i:- and I 4 , [5] , 12:NM ) are included in the target group then the PCR is 100% sensitive and 100% specific in identifying Salmonella Typhimurium , Salmonella Enteritidis , Salmonella Dublin and Salmonella Typhimurium-like organisms . The remaining 26 serogroup B isolates were negative for the tested targets . During the course of this study , we decided to determine the prevalence of I 4 , [5] , 12:i:- in Mali . Levy et al . [18] identified 220 Salmonella Typhimurium , four I 4 , [5] , 12:i:- and eight I 4 , [5] , 12:NM strains . However , in this previous study , Phase 2 flagella typing was not performed on all of the strains . We re-examined the 220 Salmonella O serogroup B , H:i isolates that had been previously been presumptively identified as Salmonella Typhimurium and used classical methods ( i . e . , sera against the Phase 2 H1 , 2 flagella ) to determine that 48 isolates were in fact I 4 , [5] , 12:i:- ( bringing the total number of isolates of this serovar to 52 ) and one isolate was I 4 , [5] , 12:NM ( bringing the total number of isolates of this serovar to nine ) . The remaining 171 strains were confirmed as Salmonella Typhimurium . We have combined previously described primers in a PCR to discriminate between Salmonella Typhimurium and I 4 , [5] , 12:i:- . Primers FFLIB and RFLIA amplify the fliB-fliA intergenic region of the flagellin gene cluster [24] . Salmonella Typhimurium strains possess an IS200 fragment in this region [25] . Burnens et al . [25] showed that 21 of 23 isolates of Salmonella Typhimurium and none of 85 isolates of 37 other Salmonella serovars contained IS200 in this region . Primers FFLIB and RFLIA have been reported to amplify a 1-kb product from Salmonella Typhimurium and I 4 , [5] , 12:i:- strains and a 250-bp product from all other serovars [24] . However , when validating these primers , we found that a 1-kb fragment was amplified from Salmonella Farsta ( not tested by Echeita et al . [24] ) suggesting that this serovar also possesses IS200 in the fliB-fliA intergenic region ( Figure 3 ) . Primers Sense-59 and Antisense-83 amplify the fljB allele [26] . Primer Sense-59 binds at position 258 and primer Antisense-83 binds at position +100 of the 5′-3′ consensus fljB1 , 2 sequence . These primers amplify a 1389-bp product from strains that possess a Phase 2 flagellar antigen and no product from strains that lack a Phase 2 flagellar antigen such as I 4 , [5] , 12:i:- . As shown in Figure 3 , the PCR was able to discriminate between Salmonella Typhimurium and I 4 , [5] , 12:i:- strains and other serogroup B , H:i serovars except Salmonella Farsta . We tested all the Salmonella Typhimurium , I 4 , [5] , 12:i:- and I 4 , [5] , 12:NM strains identified in Mali and found that 170 of 171 Salmonella Typhimurium strains were correctly identified ( i . e . , possessed a 1-kb fliB-fliA intergenic region product and fljB1 , 2 ) , and 51 of 52 I 4 , [5] , 12:i:- strains were correctly identified ( i . e . , possessed a 1-kb fliB-fliA intergenic region product and lacked fljB1 , 2 ) ( Table 4 ) . The nine I 4 , [5] , 12:NM strains produced mixed results in that all nine strains produced a 1-kb fliB-fliA intergenic region product but three strains possessed fljB1 , 2 . Since Salmonella Stanleyville was found to be fairly common among the Mali NTS isolates , we added primers to detect fliC-z4 , z23 of Salmonella Stanleyville to the multiplex PCR containing primers H-for , Hi , sdfF , sdfR , 16SF and DG74 . The primers were first tested on Salmonella Stanleyville by themselves and produced a 427-bp amplicon . The fliC-z4 , z23 primers were then added to the multiplex primer mix and PCR was performed ( using the previously optimized conditions ) on all 26 Salmonella Stanleyville strains , and a sample of 10 Salmonella Typhimurium , 10 Salmonella Dublin and 11 Salmonella Enteritidis strains . Correct amplicons were observed for all the strains tested . Figure 4 shows amplicons from a sample of three Salmonella Stanleyville strains and the control Salmonella Typhimurium , Salmonella Enteritidis and Salmonella Dublin strains .
We have combined published primers and new primers in a multiplex PCR that , following the application of a previously described O serogrouping multiplex PCR [18] , can identify Salmonella Typhimurium ( and variants ) , Salmonella Enteritidis , Salmonella Dublin and Salmonella Stanleyville . Detection of Salmonella Typhimurium , Salmonella Dublin and Salmonella Stanleyville is based on amplification of the respective fliC alleles . We were unable to design primers to detect fliC-g , m of Salmonella Enteritidis due to the high nucleotide identity between fliC-g , m and fliC-g , p ( of Salmonella Dublin ) . We therefore used primers to detect “Salmonella difference fragment I” ( Sdf I ) , a segment of Salmonella Enteritidis DNA that was reported to be absent from 73 non-Enteritidis Salmonella enterica isolates comprising 34 different serovars as determined by PCR [27] . We confirmed the utility of Sdf I , with the exception of serovars Meleagridis and Livingstone . We found that Salmonella Livingstone yielded a weak PCR product using the same Sdf I primers that were previously reported [27] . The disparity could be due to a difference in the amplification method ( different polymerases and cycling conditions were used ) . From the epidemiologic and public health perspective , being able to detect strains that are genetically similar to Salmonella Typhimurium yet that constitute distinct serovars ( i . e . , I 4 , [5] , 12:i:- and I 4 , [5] , 12:NM ) is important ( e . g . , for outbreak investigations ) . In the USA and Europe such strains are increasingly being reported [28]–[30] . In Spain , I 4 , [5] , 12:i:- was the fourth most commonly isolated Salmonella serovar from humans from 1998–1999 [29] and several studies suggest that this monophasic serovar is a variant of Salmonella Typhimurium [24] , [31]–[33] . The PCR that we have described can generally discriminate the diphasic Salmonella Typhimurium serovar ( I 4 , [5] , 12:i:1 , 2 ) from monophasic ( I 4 , [5] , 12:i:- ) variants . Only one Salmonella Typhimurium was misidentified as I 4 , [5] , 12:i:- and vice versa . It is possible that our PCR will not be able to detect some serologically monophasic I 4 , [5] , 12:i- strains as lack of Phase 2 flagellar antigen expression can be due to a variety of mechanisms ranging from point mutations to partial or complete deletions in fljB1 , 2 and adjacent genes . Additionally , if there is a deletion in the first 250 bp of fljB1 , 2 , the primers we have chosen will not identify the strain as I 4 , [5] , 12:i- . Furthermore , our PCR scheme cannot differentiate between Salmonella Typhimurium and Salmonella Farsta . However , in practical terms , this is unlikely to pose a problem as Salmonella Farsta is extremely rare . One small set of strains where our PCR gives differing results from traditional serological methods are Salmonella Typhimurium-like non-motile variants ( I 4 , [5] , 12:NM ) . Notably , all nine Malian strains identified by serotyping methods as I 4 , [5] , 12:NM were found to possess the fliC-i allele and three of the strains also possessed the fljB1 , 2 gene . Two quite distinct explanations can account for these observations . One is that in some strains lack of motility is not due to loss of flagellar genes but rather to other factors ( e . g . , regulation ) that keep expression turned off . Alternatively , it may be that our genetic identification of these strains is correct and that the failure to detect flagella phenotypically is merely a consequence of not knowing how to grow the bacteria under conditions optimal for expression of those flagella . We assume that the I 4 , [5] , 12:NM strains from Mali are Salmonella Typhimurium variants as they possess fliC-i and IS200 in the fliB-fliA intergenic region . It is also possible , albeit unlikely , that they could be the very rarely isolated Salmonella Farsta . Soyer et al . [34] have reported that there are at least two common clones of I 4 , [5] , 12:i:- with different genomic deletions ( an ‘American’ deletion genotype and a ‘Spanish’ deletion genotype ) . Both I 4 , [5] , 12:i:- clones completely lack fljB and fljA . Preliminary analysis of the deletion using a variety of primers that amplify different sections of the fljB1 , 2 gene indicates that the I 4 , [5] , 12:i:- strains from Mali appear to possess the 3′ end of fljB and the entire fljA ORF . At least 250 bp of fljB ( including the Sense-59 binding site ) has been deleted at the 5′ end ( data not shown ) . This suggests that these strains are genetically different from both the Spanish and American I 4 , [5] , 12:i:- isolates . We are sequencing the deletion in several Malian I 4 , [5] , 12:i- strains to determine the exact deletion . It will be interesting to see whether I 4 , [5] , 12:i:- strains from other African countries are genetically similar to the Malian strains . Several other DNA-based Salmonella typing methods have been described [35]–[41] . However , some of these do not identify the breadth of enteric fever and NTS serovars of our multistep , multiplex PCR or fail to include an internal positive control . An O serogroup-specific Bio-Plex assay to detect serogroups B , C1 , C2 , D , E and O13 and serovar Paratyphi A [42] and a DNA sequence-based approach to serotyping have also been described [43] . However , these methods require greater financial and technical resources over those required for our method . Our PCRs are novel because they use as few primers as possible to identify the most common non-typhoidal Salmonella serovars isolated from blood and other invasive sites in sub-Saharan Africa , including Salmonella Typhimurium ( and several variants ) , Salmonella Enteritidis , Salmonella Dublin and Salmonella Stanleyville . Since the late 1980s , the majority ( 85 to 95% ) of NTS associated with invasive disease in sub-Saharan Africa belong to these serovars [9] , [11]–[13] , [15] , [16] , [18] , [19] . Therefore , we do not believe that there is a need for multiplex PCRs that detect more serovars unless the epidemiologic picture changes . We have tried to keep the PCRs as simple as possible so that they can be performed easily and the results interpreted correctly in laboratories in Africa that may be new to PCR . If a large outbreak or otherwise frequent isolation occurred of a serovar not presently recognized or contained within our multiplex , this serovar would not be identifiable using our PCR and would have to be identified in a reference laboratory using antisera or by molecular serotyping . We are currently evaluating various PCR reagents that are stable at room-temperature and can be readily obtained by laboratories in Africa . Depending on the prevalence of certain serovars in a given country , either typhoidal or non-typhoidal Salmonella ( or both ) can be identified using our primer sets ( Table 5 and Figure 5 ) . For example , one may wish to test all Salmonella isolates in the O serogrouping PCR , then screen serogroup A , B and D Vi+ strains using the first H typing multiplex PCR to identify Salmonella Typhi , Salmonella Paratyphi A and Salmonella Paratyphi B . The d-tartrate fermentation PCR can be performed to differentiate Salmonella Paratyphi B sensu stricto strains from Salmonella Paratyphi B Java . Any serogroup B isolates not identified by the 1st H typing PCR can be tested along with non-Typhi O serogroup D strains in the second H typing/Sdf I multiplex PCR to identify serovars Typhimurium ( and related strains ) , Dublin ( which can be Vi+ or Vi- [44] ) , Enteritidis and Stanleyville . The O serogroup B H:i strains can be tested using the Typhimurium/I 4 , [5] , 12:i:- PCR to identify Salmonella Typhimurium and I 4 , [5] , 12:i:- . It should be stressed that the O serogrouping PCR described by Levy et al . [18] needs to be performed in conjunction with the PCRs described here to ensure that Salmonella Enteritidis and Salmonella Typhimurium are identified correctly and not mistaken as Salmonella Meleagridis and Salmonella Livingstone; and Salmonella Cotham , respectively . The surveillance experience in Mali is the first to show that Salmonella Dublin and Salmonella Stanleyville can constitute important serovars associated with invasive non-typhoidal Salmonella disease , along with Salmonella Typhimurium ( and variants ) and Salmonella Enteritidis . Previously , Salmonella Dublin and Salmonella Stanleyville were recovered only occasionally from blood cultures of patients in Africa [12] , [13] , [45] , [46] . We thought it useful to be able to detect these serovars by PCR in future surveillance studies in Africa . In conclusion , we have described a series of PCRs based on O serogrouping and H typing that can identify the causative agents of enteric fever ( Salmonella Typhi and Salmonella Paratyphi A and Salmonella Paratyphi B ) , the three most commonly isolated serovars that cause invasive disease in young children in sub-Sahara African ( Salmonella Typhimurium [and Typhimurium-like] , Salmonella Enteritidis and Salmonella Dublin ) and Salmonella Stanleyville , an invasive pathogen that may be of regional importance in West Africa . | The genus Salmonella has more than 2500 serological variants ( serovars ) , such as Salmonella enterica serovar Typhi and Salmonella Paratyphi A and B , that cause , respectively , typhoid and paratyphoid fevers ( enteric fevers ) , and a large number of non-typhoidal Salmonella ( NTS ) serovars that cause gastroenteritis in healthy hosts . In young infants , the elderly and immunocompromised hosts , NTS can cause severe , fatal invasive disease . Multiple studies of pediatric patients in sub-Saharan Africa have documented the important role of NTS , in particular Salmonella Typhimurium and Salmonella Enteritidis ( and to a lesser degree Salmonella Dublin ) , as invasive bacterial pathogens . Salmonella spp . are isolated from blood and identified by standard microbiological techniques and the serovar is ascertained by agglutination with commercial antisera . PCR-based typing techniques are becoming increasingly popular in developing countries , in part because high quality typing sera are difficult to obtain and expensive and H serotyping is technically difficult . We have developed a series of polymerase chain reactions ( PCRs ) to identify Salmonella Typhimurium and variants , Salmonella Enteritidis and Salmonella Dublin . We successfully identified 327 Salmonella isolates using our multiplex PCR . We also designed primers to detect Salmonella Stanleyville , a serovar found in West Africa . Another PCR generally differentiated diphasic Salmonella Typhimurium and monophasic Salmonella Typhimurium variant strains from other closely related strains . The PCRs described here will enable more laboratories in developing countries to serotype NTS that have been isolated from blood . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/bacterial",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"infectious",
"diseases",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2010 | Identification by PCR of Non-typhoidal Salmonella enterica Serovars Associated with Invasive Infections among Febrile Patients in Mali |
The corn smut fungus Ustilago maydis requires the unfolded protein response ( UPR ) to maintain homeostasis of the endoplasmic reticulum ( ER ) during the biotrophic interaction with its host plant Zea mays ( maize ) . Crosstalk between the UPR and pathways controlling pathogenic development is mediated by protein-protein interactions between the UPR regulator Cib1 and the developmental regulator Clp1 . Cib1/Clp1 complex formation results in mutual modification of the connected regulatory networks thereby aligning fungal proliferation in planta , efficient effector secretion with increased ER stress tolerance and long-term UPR activation in planta . Here we address UPR-dependent gene expression and its modulation by Clp1 using combinatorial RNAseq/ChIPseq analyses . We show that increased ER stress resistance is connected to Clp1-dependent alterations of Cib1 phosphorylation , protein stability and UPR gene expression . Importantly , we identify by deletion screening of UPR core genes the signal peptide peptidase Spp1 as a novel key factor that is required for establishing a compatible biotrophic interaction between U . maydis and its host plant maize . Spp1 is dispensable for ER stress resistance and vegetative growth but requires catalytic activity to interfere with the plant defense , revealing a novel virulence specific function for signal peptide peptidases in a biotrophic fungal/plant interaction .
Microbial cells modulate conserved signaling pathways to adjust their intracellular physiology to constantly changing environmental conditions [1–3] . The smut fungus Ustilago maydis is a highly adapted biotrophic pathogen of its host plant maize that establishes a compatible fungal/plant interaction without evoking obvious plant defense responses [4] . The switch from saprophytic to biotrophic growth of U . maydis is initiated after fusion of two compatible haploid sporidia , generating an infectious filamentous dikaryon that strictly depends on living host tissue for further propagation [5] . Filaments elongate by tip-growth on the leaf surface and do not actively divide , as their cell cycle is arrested [6 , 7] . After appressoria-mediated plant penetration , the cell cycle arrest is released and mitotic growth of the dikaryotic filament initiated , followed by massive proliferation of dikaryotic hyphae and tumor formation [8] . The central regulator of pathogenic development in U . maydis is the bE/bW transcription factor complex , encoded by the b-mating type locus [9 , 10] . The b-dependent transcriptional network comprises 345 genes and controls the early steps of pathogenic development including filament formation , maintenance of the G2-cell cycle arrest , appressoria formation and plant penetration [11 , 12] . Further development in planta , however , requires modulation of the b-dependent regulatory network by the Clp1 protein , mediating cell cycle release and mitotic proliferation in planta [8] . To cope with the plant immune system and establish a compatible interaction , biotrophic pathogens secrete effector proteins that suppress the plant defense and reprogram host metabolism [13 , 14] . In U . maydis , the apoplastic effectors Pep1 and Pit2 actively interfere with the plant defense response by inhibiting the host peroxidase POX12 , and cysteine proteases , respectively [15–18] . In addition , Tin2 and the chorismate mutase Cmu1 manipulate the plant host metabolism to foster colonization by U . maydis [19 , 20] . Effector encoding genes are expressed in an organ and tissue specific manner and are coordinately upregulated during biotrophic growth resulting in effector waves [21 , 22] that impose enormous stress on the protein folding machinery in the endoplasmic reticulum ( ER ) . In the budding yeast Saccharomyces cerevisiae , ER stress is sensed by the ER-membrane localized kinase/RNase Ire1 , leading to Ire1 activation and endonucleolytic cleavage of the constitutively expressed HAC1 mRNA . The processed mRNA encodes the bZIP transcription factor Hac1 that is functionally conserved in most eukaryotes including mammals ( Xbp1 ) and U . maydis ( Cib1 ) [23 , 24] . Elevated expression of Hac1 target genes promotes the restructuring of the secretory pathway by increasing ER folding capacity , ER expansion and degradation of irreversibly misfolded proteins via the ER associated degradation ( ERAD ) pathway [25–27] . One mechanism to target proteins for ERAD is mediated by signal peptide peptidases ( SPPs ) which are ER membrane-localized aspartyl-proteases that cleave type II oriented transmembrane proteins and remnant signal peptides [28–31] . Although conserved in their catalytic activity , SPPs exert diverse physiological roles including antigen representation in human cells [32] , transcription factor activation during hypoxia in Aspergillus species [33] or regulation of embryonic development in Caenorhabditis elegans [34] . A functional UPR is crucial for virulence in various human and plant-pathogenic fungi , including U . maydis [35–41] and important for efficient secretion of hydrolytic enzymes involved in decomposition of complex polysaccharides including cellulose and plant cell walls in the fungal saprophytes Trichoderma reesei and Neurospora crassa [42–45] . In U . maydis , the UPR is connected to the regulation of biotrophic development and specifically activated after plant penetration . Expression of the UPR regulator Cib1 and its interaction with the developmental regulator Clp1 leads to accumulation of Clp1 protein and triggers re-initiation of mitotic growth in planta [36] . Moreover , Cib1 is required for efficient secretion of effector proteins and participates as well in their transcriptional regulation [46] . Apparently , UPR gene expression is adapted to the lifestyle of the fungus and altered by the interaction between Cib1 and Clp1 . Both proteins are constantly expressed during biotrophic growth in planta [8 , 11 , 36] , and their interaction renders U . maydis hyper-resistant towards ER stress . It is conceivable that this enables long-term UPR activity in planta , which can be otherwise deleterious [36] . Although initial data suggests that this is connected with dampening of UPR activity , the detailed consequences on UPR gene expression have not been addressed , yet . In addition to the central role of Hac1-like proteins in fungal pathogens , only few UPR regulated genes have been identified that contribute to fungal virulence . Importantly , none of these factors exerts virulence specific functions . In strains lacking the ER chaperone Lhs1 in the hemibiotrophic blast fungus Magnaporthe oryzae ( Pyricularia oryzae ) , the ER co-chaperone Dnj1 in U . maydis or the protein disulfide isomerase Pdi1 in the necrotrophic plant pathogen Botrytis cinerea , reduced virulence is invariably connected to erroneous protein folding and reduced ER stress resistance [47–49] . Consequently , UPR regulated factors with virulence functions unrelated to ER stress resistance remain to be discovered . Here , we show that modulation of UPR gene expression by Clp1 correlates with Cib1-phosphorylation and stabilization . We identify the UPR-regulated signal peptide peptidase Spp1 as a novel key virulence factor that is dispensable for vegetative growth or filament formation but is of crucial importance to establish a compatible biotrophic interaction and cause disease . Importantly , the loss of virulence cannot be attributed to altered protein secretion , defects in ER stress resistance or ERAD-mediated protein degradation , but is connected to the catalytic activity of Spp1 itself to interfere with plant defense responses .
Previously , we have demonstrated that the physical interaction between Cib1 and Clp1 leads to drastically increased ER stress resistance ( Fig 1A ) , by preventing deleterious UPR hyperactivation and adaptation of the UPR for long-term activity during biotrophic development of U . maydis [36] . To identify UPR regulated genes and address their modulation by Clp1 on a global scale we performed RNAseq analysis of U . maydis under non-stress and tunicamycin ( TM ) -mediated ER stress conditions . Tunicamycin inhibits N-glycosylation of proteins , leading to accumulation of misfolded proteins in the ER and induction of ER stress . To avoid crosstalk with the b-dependent signaling cascade we used U . maydis strain JB1 ( WT ) , in which the b-mating type locus is deleted [11] . The JB1 derivative UVO151 allows for arabinose-inducible expression of clp1 by the crg1-promoter ( Pcrg:clp1 ) [11] and strain JB1Δcib1 ( Δcib1 ) [36] was used as additional control to identify TM-induced side effects unrelated to UPR signaling . The experiment was performed with three biological replicates for each strain and condition ( n = 3 ) . Genes were considered UPR regulated when expression was induced in response to TM-mediated ER stress and no alteration in gene expression was observed in the Δcib1 control strain . Differentially expressed genes were filtered at a 4-fold cut-off ( log2FC ≥2 ) and a false discovery rate ( fdr ) of <0 . 05 . Using these criteria , we identified 204 genes differentially expressed in response to TM treatment ( 103 upregulated , 101 downregulated ) in WT ( JB1 ) background ( WT vs . WT +TM ) . 115 genes of the 204 TM-regulated genes required cib1 for differential expression and were not affected by TM in the Δcib1 control ( WT +TM vs . Δcib1 +TM ) . Of these 115 genes , 65 were induced and 50 were repressed in response to TM treatment ( Fig 1B and S1 Table ) . Since all Hac1-like UPR regulators are transcriptional activators , Cib1-dependent expression of the 50 UPR-repressed genes is likely to be indirect or affected by additional regulators . We thus defined the 65 upregulated as UPR core genes and focused on them in our subsequent analyses . In this set of genes we identified bip1 , lhs1 , mpd1 and dnj1 , all of which have been identified as UPR regulated genes in previous studies [36 , 49] . Under less stringent filtering criteria ( log2FC ≥1 ) 269 UPR core genes were identified , that include cib1 , ost3 , cne1 and pit1 in addition to the previously mentioned genes [36 , 46] ( S1 Table ) . UPR core genes were subjected to enrichment analysis of functional categories using the MIPS functional catalogue database ( http://mips . helmholtz-muenchen . de/funcatDB/ ) ( S2 Table ) . As expected , functional categories significantly overrepresented within the 65 UPR core genes are associated with the UPR ( total of 23 categories ) , including "protein folding and stabilization" ( p = 1 . 02E-08 ) , "unfolded protein response" ( p = 2 . 15E-07 ) , "non-vesicular ER transport" ( p = 3 . 87E-03 ) , "ER to Golgi transport" ( p = 4 . 85E-03 ) , "stress response" ( p = 8 . 57E-03 ) , "transport routes" ( p = 1 . 12E-02 ) and "protein binding" ( p = 1 . 19E-02 ) ( Fig 1C and S2 Table ) . Using GeneOntology ( GO ) analysis ( www . geneontolgy . org ) 37 of the 65 UPR core genes were functionally classified , revealing significant enrichment only for the GO term GO:0034976 "response to endoplasmic reticulum stress” ( p = 1 . 54E-03 ) . Consistently , UPR core genes encode proteins with conserved functions in ER protein folding ( Bip1/UMAG_15034 , Lhs1/UMAG_00904 , Dnj1/UMAG_05173 , Dnj2/UMAG_10099 , Mpd1/UMAG_05352 , Pdi1/UMAG_10156 , Ero1/UMAG_05219 , Dnj1/UMAG_10099 ) , ER-associated calcium transport ( Ena5/UMAG_00204; Pmr1/UMAG_20218 ) , ER-associated degradation ( ERAD ) ( Hrd1/UMAG_00542 , Hrd3/UMAG_04355; Der1/UMAG_05898 ) , protein secretion and protein transport processes ( Sec11/UMAG_00481 , SPC3/UMAG_15029 , Sec63/UMAG_06175 ) . No significant enrichment of functional categories was observed for the 50 UPR repressed genes . Since the UPR is specifically induced after plant penetration ( 2 days post inoculation ( dpi ) ) and remains constitutively active during the fungal/plant interaction , we compared our UPR core gene set with the genes differentially expressed during biotrophic development ( 2 , 4 and 6 dpi ) [22] . This revealed that 51 of the 65 core genes were consistently upregulated by the UPR and during biotrophic development . Interestingly , 6 genes were UPR-induced but repressed in planta ( Fig 1D ) ( S3 Table ) , suggesting that expression of these genes is not solely regulated by the UPR and is most likely affected by other , potentially plant-related cues . By contrast , of the 50 UPR-repressed genes 17 showed significantly increased levels , 6 unchanged and 27 significantly reduced expression levels during biotrophic development [22] ( S3 Table ) . The large overlap of the UPR- and plant-induced genes provides further evidence for the central role of the UPR during biotrophic development of U . maydis . To address the effects of Clp1 on the UPR we compared UPR core gene expression between WT ( JB1 ) and Pcrg:clp1 ( UVO151 ) strains after TM-mediated UPR activation . We reasoned that monitoring gene expression changes of the UPR core gene set might be most informative as their regulation is unlikely to involve additional factors . Expression of only 10 of the 65 UPR core genes was affected more than 2-fold ( log2FC ≥+/-1 ) ( 7 repressed , 3 induced ) by clp1 . In addition , expression of 32 . 3% ( 21 of 65 ) of the UPR core genes trended to be higher in UVO151 ( log2FC = 0 . 92 to 0 . 1 ) , whereas expression of 35 . 4% ( 23 of 65 ) of the genes trended to be decreased ( log2FC = -0 . 1 to -0 . 8 ) and expression of 17% ( 11/65 ) remained completely unaffected ( log2FC = -0 . 1 to 0 . 1 ) ( Fig 2A and S1 Table ) . Although these effects appear marginal , comparison of normalized read counts per kilobase of transcript , per Million of mapped reads ( RPKM ) of all 65 UPR core genes revealed a drop of more than one fourth ( 26% ) in UVO151 ( 23919 ) compared to JB1 ( 32380 ) . Overall differences in UPR gene expression were modest between both strains . We thus analyzed by qRT-PCR expression of 8 UPR core genes ( increased by clp1: UMAG_12178 , UMAG_03404; not affected: UMAG_05009; UMAG_02944 , UMAG_11513 , UMAG_02729-spp1; decreased by clp1: UMAG_00904-lhs1 , UMAG_5352-mpd1 ) and 2 additional UPR marker genes ( UMAG_10287-cne1 and UMAG_05352-ost3 ) using cDNA that was generated in independent experiments ( n = 3 ) . Expression of all genes was significantly increased by TM treatment ( p<0 . 05 ) in strain JB1 and remained at basal levels under identical conditions in the Δcib1 background ( Fig 2B ) . All genes showed a similar modulation of gene expression by Clp1 in qRT-PCR and RNAseq analysis , confirming our initial results . Protein-protein interactions between accessory proteins and bZIP transcription factors can influence gene expression in many different ways [50] , including modulation of the protein-DNA interaction . To identify direct Cib1 targets and to test if the interaction with Clp1 affects DNA binding by Cib1 , we performed chromatin immunoprecipitation followed by massively parallel DNA sequencing ( ChIP-seq ) analysis . We expressed a functional Cib1-3xHA fusion protein [46] under the control of the endogenous promoter from the endogenous genomic locus in the WT ( JB1 ) and Pcrg:clp1 strain ( UVO151 ) . Four hours after TM-mediated UPR induction chromatin was isolated , followed by immunoprecipitation ( ChIP ) using anti-HA beads and sequencing of enriched DNA . The experiment was performed with two biological replicates for each strain ( n = 2 ) . In addition , input DNA and DNA unspecifically bound to HA-agarose beads ( mock ) was sequenced to identify and correct for binding or sequencing bias . Peak calling was performed using peakZilla [51] , providing a peak-score ( Δ normalized reads ( IP-input ) x distribution score ) conflating true positive probability of DNA binding and estimation of DNA binding strength . Scores of individual peaks were accumulated when identified on a single promoter ( 1 . 5 kb upstream of translation start site ( tss ) ) and at least one peak score was ≥40 , to yield promoter scores . Promoters were filtered with a promoter score cut-off of ≥100 . 217 promoters were bound by Cib1-3xHA in both strains , and 63 and 188 promoters were only bound in the WT ( JB1cib1-3xHA ) or Pcrg:clp1 ( UVO151cib1-3xHA ) strain , respectively . 41 and 46 promoters of UPR core genes had scores ≥100 , in WT and Pcrg:clp1 , respectively . Importantly , within the top twenty list of promoters with the highest scores we identified in both strains the known UPR genes bip1 , cib1 , pdi1 , ero1 , lhs1 , pmr1 , lhs1 and dnj1 ( S4 Table ) . We further tested binding of Cib1 to the promoters of cib1 and ero1 by ChIP-qPCR analysis ( S2 Fig ) . In comparison to the negative control eIF2b , amplicons corresponding to the respective promoter regions were highly enriched , confirming binding to both promoters and the previously postulated autoregulation of cib1 gene expression by the Cib1 protein [36] . MEME-ChIP analysis of peaks derived from promoters specifically bound by Cib1-3xHA in WT or the Pcrg:clp1 strain did not reveal a clear binding motif . By contrast , when peaks were derived from UPR core gene promoters that were bound in both , WT and Pcrg:clp1 strains ( log2FC≥1 , n = 114 peaks , n = 91 promoters ) , an overrepresented CRE3-like binding motif T/CGACGTGGAAG ( E = 4 . 6e-44 ) was identified in WT background ( Fig 3A ) [52] . This motif is highly similar to the unfolded protein response element ( UPRE ) bound by Xbp1 in human ( GATGACGTGGC , E = 2 . 0e-8 ) [52 , 53] and almost perfectly matches the Cib1-binding site consensus sequence ( reverse complement: CGACGTGGCA ) in the promoter regions of pit1/2 and tin1-1 [46] . However , by ChIPseq only binding to the tin1-1 promoter was detectable ( S3 Fig ) . When MEME-ChIP analysis was performed on peaks derived from the clp1 expressing Pcrg:clp1 strain a largely overlapping TGACGTGG motif was highly enriched ( E = 9 , 9e-47 ) , lacking only the terminal AAG triplet of the motif identified in the WT strain ( Fig 3A ) . Consistently , visual inspection of ChIPseq data using the integrated genomics viewer tool ( IGV ) revealed that shapes and locations of peaks were conserved between strains as illustrated by the representative examples ( Fig 3B ) . Of the 65 UPR core gene ( log2FC = 2 ) promoters 41 were bound by Cib1-3xHA in WT and 46 were bound in the Pcrg:clp1 strain , with an overlap of 37 promoters ( S4 Table ) . Comparison of promoter scores revealed that promoters with the four highest scores in WT ( bip1 , cib1 , pdi1 , ero1 ) showed the strongest score decrease in the Pcrg:clp1 strain in conjunction with reduced expression levels in UVO151 ( Pcrg:clp1 ) in comparison to the JB1 ( WT ) control ( Fig 3C and S1 Table ) . Promoter scores and expression levels of spp1 , predicted to encode a signal peptide peptidase , correlated as well when compared between both strains . In contrast , for UMAG_11799 ( cak1 ) these were inversely correlated ( Fig 3B ) . In summary , our data suggest that modulation of UPR core gene expression is not related to an altered DNA binding specificity of Cib1 and that modulation of UPR gene expression by Clp1 might involve other mechanisms . To further explore how Clp1 affects UPR gene expression we first analyzed the effect of Clp1 on Cib1 protein localization . We visualized the subcellular localization of a functional Cib1-GFP fusion protein expressed under the control of its endogenous promoter from the endogenous genomic locus in strains JB1cib1-GFP ( WT ) and UVO151cib1-GFP ( Pcrg:clp1 ) , by fluorescence microscopy . Four hours after TM treatment Cib1-GFP was detectable in the nucleus of WT cells and was in addition localized to the cytoplasm when clp1 was expressed ( Pcrg:clp1 ) ( Fig 4A ) . Next , we tested by Western hybridization whether Cib1-GFP levels were altered by clp1 expression . Cib1-GFP was detectable after TM treatment and higher levels of Cib1-GFP were detected in the clp1 expressing strain ( Pcrg:clp1 ) in comparison to the WT control ( Fig 4B ) . To test if increased abundance of the fusion protein correlated with increased expression or splicing of cib1 mRNA we used qRT-PCR analysis to determine transcript levels of the spliced cib1s mRNA . This revealed a TM-dependent increase of cib1s levels in both strains . However , contrary to the Cib1 protein levels , transcript abundance was significantly lower in the clp1-expressing strain ( Pcrg:clp1 ) when compared to the WT control ( Fig 4C ) . We therefore concluded that Clp1 affects abundance of Cib1 posttranscriptionally . To test if the increased abundance of Cib1-GFP is related to altered protein stability , we performed promoter shut-off assays using the doxycycline-repressible Tet-off system [54] . To this end , we expressed Cib1-GFP under the control of the Tet-promoter from the native genomic locus . We induced ER stress with TM for 4 hours , blocked transcription of cib1-GFP with doxycycline ( 10 μg/ml ) and monitored Cib1-GFP levels over time . We quantified protein levels relative to T0 and observed that Cib1-GFP levels decreased significantly slower in the clp1 expressing strain when compared to the WT control ( Fig 4D ) . This observation was as well confirmed by cycloheximide chase assays ( S1 Fig ) . This indicates that Cib1-GFP is stabilized by Clp1 and that this accounts for the increased abundance of Cib1-GFP . In addition to the differences in Cib1-GFP levels we noticed that the fusion protein displayed mobility shifts in Western hybridization experiments that might be caused by posttranslational modifications ( Fig 4B ) . Protein phosphorylation often precedes ubiquitin-dependent protein degradation [55] . Since mobility shifts were more pronounced in the WT when compared to the clp1 expressing strain ( Pcrg:clp1 ) we speculated that stability of Cib1-GFP might be related to phosphorylation of the fusion protein . We tested this assumption by phosphatase treatment of protein extracts prepared from WT and clp1 expressing cells . Phosphatase treatment prevented the mobility shift in extracts derived from WT ( JB1 ) cells , and this could be blocked by addition of phosphatase inhibitor . By contrast , only minor effects were observed when protein extracts were derived from the clp1 expressing ( Pcrg:clp1 ) strain ( Fig 4E ) . These results suggest that Cib1-GFP is phosphorylated and that phosphorylation is reduced by Clp1 . In summary , our data demonstrate that Clp1 influences phosphorylation , stability and localization of Cib1-GFP . We assume that these processes are interdependent and connected to the Clp1-mediated modulation of UPR gene expression . Modulation of the UPR by Clp1 enables long-term activity during biotrophic growth of U . maydis [36] . Cib1 and Clp1 are specifically expressed after plant penetration and throughout all subsequent stages of biotrophic growth . Hence , it is conceivable that the UPR is constitutively modulated in planta and that the modulation of UPR core gene expression represents a promising read-out to identify potential virulence or ER stress related factors . We selected 32 candidate genes on basis of an increased or stable expression in clp1 expressing strains when compared to the WT ( Fig 2A , genes marked in red and white ) and screened potential candidates for altered virulence and ER stress resistance . To this end , we deleted the open reading frame ( ORF ) of a total of 29 UPR core genes in the solopathogenic SG200 strain background [56] . We were not successful in our attempt to generate deletion strains for three additional genes ( UMAG_00481 , UMAG_06089 , and UMAG_15029 ) , all of which are predicted to encode signal peptidase components , suggesting that these might be essential for growth . Surprisingly , none of the 29 deletion strains showed reduced ER stress resistance ( S4 Fig ) and 26 of the 29 deletion strains displayed no alteration of virulence ( S5 Fig ) . Single deletions of the two genes UMAG_12178 , encoding a protein related to 5-carboxyvanillate-decarboxylases and UMAG_11083 encoding a p24-like protein , resulted in slightly reduced virulence . Strikingly , only deletion of UMAG_02729 , predicted to encode a signal peptide peptidase resulted in the loss of virulence and complete absence of tumor formation ( S5 Fig ) . This suggests that UMAG_02729 ( hereafter referred to as Spp1 ) represents a major virulence factor that is connected to the UPR in U . maydis . The Spp1 protein is predicted to function as signal peptide peptidase ( SPP ) . SPPs are ER-membrane localized aspartyl-proteases that are found in all eukaryotes including fungi , protozoa , plants and animals and catalyze intramembrane proteolysis at the ER membrane . SPP substrates include signal peptide remnants ( cleavage products of signal peptidases ) , ER membrane-bound transcription factors or high-affinity transporters [28 , 29 , 33 , 57 , 58] . spp1 belongs to the UPR core genes identified under stringent filtering criteria ( 7 . 9-fold increased by TM ) ( Fig 1B and S1 Table ) and expression of spp1 is as well strongly induced in planta ( 11-fold at 2 dpi vs . axenic ) ( Fig 1D ) [22 , 36] . spp1 levels were not affected by Clp1-mediated UPR modulation ( Fig 2A and 2B ) and promoter scores derived from ChIPseq of WT and clp1 expressing strains were almost identical ( promoter score: 261 ) ( Fig 3B and 3C and S4 Table ) . Hence , we discovered a direct regulation of spp1 expression by the UPR , which to the best of our knowledge , has not been described in other organisms . Spp1 displays the characteristic SPP domain structure with 9 predicted trans membrane ( TM ) domains ( http://www . cbs . dtu . dk/services/TMHMM/ ) [59] , harboring the highly conserved YD and GLGD motifs important for catalytic activity and the QPALLY motif putatively conferring substrate specificity [29] ( Fig 5A and S6 Fig ) . To visualize the subcellular localization of Spp1 we expressed the protein as a C-terminal mCherry ( mC ) fusion under the control of the native spp1 promoter in the SG200Δspp1 background . In accordance with the predicted function , fluorescence microscopy revealed that 2 hours after TM-mediated UPR induction Spp1-mC localized to structures resembling the perinuclear and cortical ER ( Fig 5B ) . By contrast , no accumulation of Spp1-mC fusion protein was observed in the Δcib1 background under these conditions ( S7 Fig ) , corroborating that the accumulation of Spp1-mC requires cib1-dependent UPR induction and is not a side effect of TM treatment . Expression of spp1-mC under the control of the constitutive active otef promoter revealed that localization of Spp1-mC is similar in WT and Δcib1 strain background ( S7 Fig ) . Phylogenetic analysis revealed that outside of the Ustilaginales Spp1 is related to human HM13 ( identity: 35% , similarity: 52% , E-value: 2e-72 ) and Plasmodium falciparum PfSPP ( identity: 33% , similarity: 49% , E-value: 6e-55 ) and more distantly related to SppA from Aspergillus species ( identity: 28% , similarity: 43% , E-value: 9e-50 ) and yeast Ypf1p ( identity: 32% , similarity: 52% , E-value: 2e-32 ) ( Fig 5C ) . To retest our screening results we performed plant infection assays with WT ( SG200 ) , Δspp1 and the Δspp1-spp1-mC complementation strain , in which the spp1-mCherry fusion construct was integrated into the ip-locus of strain SG200Δspp1 and expressed under the control of the native spp1 promoter . The loss of virulence in the Δspp1 mutant was fully complemented by expression of Spp1-mCherry ( Spp1-mC ) , indicating that the fusion protein is functional ( Fig 5D ) . Analysis of vegetative growth under axenic conditions , ER stress resistance , cell wall stress resistance or the ability to form b-dependent filaments on charcoal containing solid medium did not reveal differences between WT ( SG200 ) and the Δspp1 ( SG200Δspp1 ) derivative ( Fig 5E and 5F and S8 Fig ) . Since formation of infectious filaments was not affected in Δspp1 strains we investigated at which stage of pathogenic development Δspp1 mutants are blocked . To this end , we quantified the fungal biomass at 2 and 4 dpi , using the mfa1 gene as fungal marker , in maize plants inoculated with either the WT ( SG200 ) or the Δspp1 deletion mutant . This revealed that at both time points significantly less fungal biomass was produced in plants inoculated with the Δspp1 mutant in comparison to the WT ( SG200 ) control ( Fig 6A ) . To assess the fungal morphology during biotrophic development , we stained infected leaf tissue 3 dpi with Chlorazol Black E and analyzed the samples by microscopy ( Fig 6B ) . While the WT ( SG200 ) strain showed extensive proliferation , hyphal branching and clamp cell formation , deletion mutants of spp1 were severely impaired . Δspp1 strains displayed filamentous growth and formed appressoria that penetrated the leaf surface . However , proliferation of fungal hyphae after plant penetration was highly reduced and in most cases restricted to the epidermal cell layer . In addition , hyphae of the Δspp1 strain showed altered intercellular growth with bulbous enlargements formed prior and after plant cell wall traversal and constrictions at the point of plant cell wall passage ( Fig 6B , white arrows ) . Together , this demonstrates that Spp1 is required for proliferation in planta , which might be connected to defects in intercellular growth . To test for functional conservation between Spp1 and potential orthologs in Sporisorium reilianum , Ustilago hordei , Aspergillus nidulans , S . cerevisiae and Homo sapiens , respective genes were expressed as C-terminal mCherry ( mC ) fusion under the control of the constitutive otef promoter in U . maydis strain SG200Δspp1 ( Δspp1 ) . In addition , we tested if catalytic activity of Spp1 is required for virulence of U . maydis by expressing the Spp1D279A mutant , harboring a D>A exchange in the highly conserved GLGD motif that is known to abolish catalytic activity [29] , in the Δspp1 strain . Expression of the fusion proteins was tested by Western hybridization using anti-mCherry antibodies . Expression of all fusion proteins was detectable with the exception of S . cerevisiae Ypf1p-mC and Aspergillus nidulans SppA-mC ( S9 Fig ) . Previous studies revealed premature polyadenylation of transcripts in U . maydis if GC content or codon usage was explicitly different from U . maydis [54] . It appears likely that the failure to express Ypf1p-mC and SppA-mC might be prevented by this mechanism . Expression of Spp1 orthologs from S . reilianum and U . hordei fully rescued the virulence defect of the Δspp1 mutant , whereas expression of H . sapiens HM13-mC suppressed the virulence defect of the Δspp1 mutant only partially , but in a dose-dependent manner ( Fig 7 ) . Importantly , expression of Spp1D279A did not rescue the virulence defect of the Δspp1 mutant , as infected plants were indistinguishable from those infected by the Δspp1 progenitor strain ( Fig 7 ) . In conclusion , our data strongly suggest that Spp1 is a bona fide SPP , homologous to HM13 and that cleavage of SPP substrates is essential for virulence of U . maydis . SPPs have been implicated in ERAD-dependent processes in various organisms , including fungi , protozoa and mammals [28 , 30 , 31] . To test if the virulence function of Spp1 might be connected to ERAD-dependent protein degradation we deleted individually and in combinations the genes encoding the conserved ERAD components Hrd1 ( UMAG_00542 , ubiquitin-protein ligase ) , Doa10 ( UMAG_10911 , ubiquitin-protein ligase ) , Der1 ( UMAG_05898 , derlin-like ) and Der2 ( UMAG_11402 , derlin-like ) in the solopathogenic SG200 ( WT ) background . Single deletions of genes encoding Hrd1 and Der1 were already generated in the course of the UPR core gene deletion screen . Surprisingly , neither ER stress resistance nor virulence was affected in any of the single , double or triple mutants ( Fig 8 ) . This implies that ERAD does not play a major role in virulence of U . maydis and that the virulence function of Spp1 is not connected to ERAD-mediated protein degradation . In A . nidulans and the human opportunistic pathogen A . fumigatus the SPP SppA mediates cleavage of the basic helix loop helix transcription factor SrbA ( sterol regulatory element binding protein ) under hypoxic conditions [33] . We speculated that the single protein related to SrbA in U . maydis ( UMAG_05721/Srb1 , E = 2 . 0e-14 ) might potentially be involved in adaptation to reduced oxygen levels in planta . To this end , we generated srb1 deletion mutants in the solopathogenic strain SG200 ( WT ) , but did not observe altered ER stress resistance or virulence of Δsrb1 mutants in comparison to the WT control ( S10 Fig ) . These results suggest that the virulence function of Spp1 is not related to Srb1-dependent processes . In human embryonic kidney cells ( HEK293 ) the heme oxygenase 1 ( HO1 ) is subject to SPP-dependent cleavage under hypoxic conditions [60] . To test whether UMAG_00783 , the single gene predicted to encode a heme oxygenase in the U . maydis genome , is important for virulence we generated deletion mutants in the SG200 strain ( WT ) . However , plant infection experiments revealed that deletion of UMAG_00783 did not affect virulence when compared to the WT control ( S11 Fig ) , suggesting that the virulence defect of Spp1 mutants is not related to the function of UMAG_00783 . A critical step in the infection process of biotrophic pathogens is the establishment of a compatible interaction , which is mediated by secreted effector proteins . To test if spp1 mutants are impaired in effector secretion we deleted spp1 in the strain SG200Δpit2-Potef:pit2-mCherry [46] , expressing the Pit2-mCherry fusion protein under the control of the constitutive otef promoter . As additional control we used strain SG200Δpit2-Potef:pit2-mCherryΔcib1 , which is defective in secretion of the Pit2-mCherry under ER stress conditions [46] . Secretion of Pit2-mCherry was visualized by Western hybridization 4 hours after TM-mediated ER stress induction . In the Δcib1 mutant , secretion of Pit2-mCherry was strongly reduced under UPR-inducing ER stress conditions , but not in the WT control or the Δspp1 mutant ( S12 Fig ) . We employed the same strategy to test for altered secretion of the effectors Pep1 [15] , Tin2 [19] and Cmu1 [20] , expressed as mCherry ( mC ) fusion proteins in the Δspp1 mutant in comparison to the WT control . Consistent with the results obtained for Pit2-mC we did not observe obvious differences in effector secretion between WT and the Δspp1 mutant ( S13 Fig ) . Thus , our data suggests that Spp1 is dispensable for effector secretion in axenic culture , Notably , under the conditions tested secretion of Pep1-mC and Tin2-mC was strictly dependent on TM-induced ER stress , whereas secretion of Cmu1-mC was not . Moreover , Pep1-mC and Cmu1-mC showed altered migration patterns , when cells were incubated under ER stress inducing conditions in comparison to the untreated control ( S13 Fig ) , indicating that these proteins are selectively processed under these conditions . In summary , this suggests that induction of the UPR , but not expression of Spp1 , might be critical for secretion and/or processing of a subset of effectors . We reasoned that although Spp1 is not required for secretion of effectors under axenic conditions , Spp1 might be important to cope with the parallel expression and high amounts of effector proteins during establishment of the biotrophic interaction ( second and strongest effector wave at 2 dpi ) [22] . As this cannot be directly investigated , we used an indirect assay to address this assumption . If our assumption were correct , we would expect strongly elevated ER stress levels in the Δspp1 mutant at this stage . Hence , we used the expression levels of cib1s and the fungal UPR marker genes bip1 , lhs1 , cne1 and UMAG_11594 as a read out , to compare ER stress levels in plants inoculated with Δspp1 and the spp1D279A strains relative to the WT control 2 dpi . Expression of all UPR marker genes was similar in all strains ( S14 Fig ) , suggesting that spp1 deletion mutants do not suffer from increased ER stress during pathogenic development of U . maydis . In addition , this implicates that Δspp1 mutants are not impaired in effector secretion at this stage of biotrophic growth . Since Δspp1 mutants displayed problems during plant cell traversal ( Fig 6 ) , we wondered whether this might be caused by elevated plant defense responses . The failure to establish a compatible host-pathogen interaction results in the formation of plant derived reactive oxygen species ( ROS ) to counter plant invasion by the pathogen . We visualized ROS formation by 3 , 3’-diaminobenzidine ( DAB ) staining of leaf tissue derived from plants inoculated with the Δspp1 mutant , the spp1D279A expressing strain and the SG200 ( WT ) control . Surprisingly , Δspp1 mutants showed not only local accumulation of DAB , but large areas of DAB precipitates ( Fig 9A ) , indicative for the formation of ROS and the induction of a hypersensitive response . These precipitates were absent in tissue infected with the SG200 ( WT ) strain , and were slightly more increased in strains expressing the catalytically inactive Spp1D279A instead of wildtype Spp1 ( Fig 9A ) . We next asked whether Δspp1 mutant strains might be hypersensitive towards ROS . To this end , we performed drop plate assays using different concentrations of H2O2 , but did not observe any differences between the WT , Δspp1 and Δspp1-spp1 complementation strains ( S15A Fig ) . In addition , we performed plant infection experiments using the NADPH oxidase inhibitor Diphenyleneiodonium ( DPI ) as a supplement to counter plant-derived ROS formation as described previously [61 , 62] . However , DPI treatment did not restore virulence or hyphal morphology of Δspp1 mutant strains in planta ( S15B and S15C Fig ) . Overall , these data suggest that the loss of virulence in Δspp1 mutant strains cannot be attributed to hypersensitivity against ROS . To assess the plant defense responses elicited by Δspp1 strains in a quantitative manner we analyzed expression of defense related plant genes by qRT-PCR analysis . As a read-out for salicylic acid ( SA ) -related defense responses we used the classical marker genes PR1 and PR5 , as well as ATFP4 encoding a SA-induced metal-binding protein , POX12 encoding an apoplastic peroxidase triggering ROS generation in response to pathogen attack as well as PR3 and PR4 ( Fig 9B , and S16 Fig , dark grey bars . ) . For jasmonic acid ( JA ) -related responses we used CC9 , encoding the cysteine protease inhibitor cystatin and BBI encoding the bax-inhibitor protein 1 as marker genes ( Fig 9B , light grey bars ) . Typically , biotrophic pathogens attempt to suppress SA and induce JA-related defense responses to prevent programmed plant cell death [63] . Strikingly , expression of all SA marker genes was strongly induced in plants infected by Δspp1 mutants or the Spp1D279A strain in comparison to plants infected by the SG200 ( WT ) or the spp1-complementation strain ( Fig 9B and S16 Fig ) . The highest induction was observed for PR1 with 78-fold and 151-fold increased levels in plants infected with Δspp1 or the spp1D279A expressing strain in comparison to the WT control , respectively . By contrast , expression of both JA marker genes was significantly reduced in plants infected with the Δspp1 mutant or the spp1D279A strain ( Fig 9B ) . Hence , the catalytic activity of Spp1 is crucial for the suppression of SA-related plant defense responses , such as ROS generation and the hypersensitive response , revealing an unexpected function for Spp1 that is required for the establishment of a compatible interaction between a biotrophic fungus and its host . In summary our study exploited Clp1-mediated UPR modulation to identify Spp1 as a novel UPR regulated virulence factor that is functionally conserved in related plant pathogens and higher eukaryotes . Since the enzymatic activity of Spp1 is essential for virulence , generation of cleavage products is essential for interference with plant defense responses . The discovery of SPP-like proteins as an inducible platform for suppression of plant defense responses by a fungal pathogen identifies a novel physiological role for SPPs and highlights how fungal pathogens adapt conserved pathways for specialized functions in host pathogen interactions .
In this study , we identified a set of 65 UPR core genes by RNAseq based transcriptome analysis . Modification of Cib1 functionality is based on Clp1-dependent posttranscriptional effects including reduced phosphorylation and increased stability of Cib1 . Expression of UPR core genes is differentially affected , and not related to an altered DNA-binding specificity of Cib1 . The UPR regulated signal peptide peptidase Spp1 represents a novel virulence factor that is crucial for plant defense suppression and the establishment of a biotrophic interaction . In line with previous transcriptome analysis of the UPR in budding yeast and other filamentous fungi , the predicted gene functions of most UPR core genes identified in this study involve ER associated process ( Fig 1C ) important for the adaptation of the secretory pathway during ER stress [25 , 26 , 64 , 65] . A direct regulation of the majority of UPR core genes by Cib1 was deduced from ChIPseq analysis . Hac1-like proteins are highly divergent with respect to their amino acid sequence , but conserved in their bZIP domain and function and are thus expected to bind similar UPREs in different species [66] . Here , we identified an UPRE in the promoters of Cib1-regulated UPR core genes that closely resembles the binding site of CRE3 and the Hac1-homolog XBP1 in higher eukaryotes [52 , 53] ( Fig 3A ) . Based on the functional complementation of Δhac1 strains by expression of the spliced cib1s mRNA , we previously used Hac1 DNA binding data [67] to predict UPREs in the promoter region of effector encoding genes in the U . maydis genome [46] . Validation of Cib1 binding to tin1-1 and pit1/2 promoter regions by ChIP-qPCR suggested a Cib1 binding site that is highly similar to the UPRE identified by genome wide ChIPseq analysis . However , in the current study we only observed differential gene expression for pit1 and reproducible binding to the tin1-1 promoter ( S1 Table and S3 Fig ) , suggesting that this might result from the different strain backgrounds used ( SG200 vs . JB1 ) . The solopathogenic SG200 strain harbors an active b-pathway that is inactive in JB1 due to the full deletion of the b-mating type locus [11] . As the b-dependently expressed transcription factor Hdp2 is involved in the regulation of pit1/2 expression [68] , it appears possible that binding of Cib1 to the pit1/2 promoter might involve Hdp2 . Modulation of the UPR is observed in diverse organisms , and a logical consequence of the multitude of interacting pathways and different lifestyles of individual species [43 , 45 , 69–73] . Modulation of the UPR can occur upstream of transcriptional regulation by regulation of Ire1-phosphorylation [74] , membrane alterations [75] , iron abundance [76] , as well as downstream by modulation of HAC1 mRNA or protein stability [77–79] . The increased complexity of the UPR in higher eukaryotes , in which besides the evolutionary conserved Ire1/Hac1 ( XBP1 ) axis , two additional arms of the UPR regulated by the bZIP transcription factors ATF6 and ATF4 exist , provides additional means of UPR modulation . ATF6 transcriptionally activates expression of XBP1 [80] and forms heterodimeric transcription factor complexes with XBP1 [81] , resulting in regulation of overlapping and specific subsets of target genes by the individual complexes [82] . ATF4 is the homolog of the central regulator of general amino-acid control Gcn4 in budding yeast , and is involved in execution of cell death if ER stress cannot be resolved [24] . Although both pathways are mutually interconnected , the anticipated physical interaction between Gcn4 and Hac1 has not been demonstrated [83 , 84] . The reduced phosphorylation of Cib1 in Clp1 expressing strains correlated with increased protein stability and strongly elevated ER stress resistance [36] ( Figs 1A and 4D ) . In budding yeast , Hac1 and Gcn4 are phosphorylated during transcription initiation by the Srb10 kinase , a component of the SRB/mediator module of RNA polymerase II [78 , 85–88] . Phosphorylation targets the bZIP proteins for SCFCdc4-dependent ubiquitylation and proteasomal degradation , generating a negative feedback loop referred to as the "black widow" model [87] . Stabilization of Hac1 by deletion of SRB10 or CDC4 , or by mutation of the phospho-sites targeted by Srb10 , resulted in increased ER stress resistance [78] . In contrast to yeast , where stabilization of Hac1 led to increased expression of the UPRE reporter , Clp1-mediated stabilization of Cib1 resulted in a more complex regulation of UPR gene expression including repression of a subset of UPR core genes . This suggests , that although the underlying principle and effects on ER stress resistance might be similar between yeast and U . maydis , the consequences on gene regulation are different . Although our localization studies revealed an altered subcellular localization pattern of Cib1-GFP when clp1 is co-expressed , it remains to be determined whether this is directly related to the modulation of Cib1 function . Our ChIPseq data do not provide evidence for alterations of the Cib1-binding specificity that might account for the Clp1-dependent effects on UPR gene expression . However , we observed that the most strongly regulated UPR core genes showed higher promoter scores in JB1 ( WT ) in comparison to the UVO151 ( +clp1 ) strain ( Fig 3C ) . In N . crassa , activity of the white-collar complex ( WCC ) , involved in the light-dependent regulation of the circadian clock , correlates with reduced WCC stability that is triggered by DNA-binding and counteracted by frequency-dependent nuclear exclusion [89 , 90] . Thus , it is well possible that a similar mode of action applies to the Cib1/Clp1 interaction . In conclusion , our results support a hypothetical model in which Clp1 negatively affects transcriptional activity of Cib1 , leading to reduced phosphorylation and increased stability of Cib1 . To this end , binding of Clp1 might either i ) mask the Cib1 transactivation domain , ii ) interfere with Cib1 homodimer formation or iii ) alter the subcellular localization of Cib1 . The estimated half-life of Clp1 ( t1/2<30 minutes ) [36] is much shorter than of Cib1 ( t1/2>60 minutes ) ( Fig 4D and S1 Fig ) . Accordingly , this inhibitory effect would be transient , thereby preventing excessive effects on UPR gene expression . Clp1 and Cib1 are expressed throughout all stages of fungal development in planta [8 , 22 , 36] , suggesting that the UPR is continuously modulated . In this respect , Clp1 could buffer fluctuating demands on the secretory pathway e . g . as imposed by effector waves and maintain stable UPR gene expression during the different stages of biotrophic development . RNAseq analysis revealed that factors previously shown to be important for ER stress resistance in U . maydis or other fungi such as the co-chaperone Dnj1 ( UMAG_05173 ) [49] , the protein disulfide isomerase Pdi1 ( UMAG_10156 ) [48] , the ER oxidoreductase Ero1 ( UMAG_05219 ) [91] and the ER chaperones Lhs1 ( UMAG_00904 ) [47] and Bip1 ( UMAG_15034 ) [92] , showed reduced expression levels during clp1-modulated UPR . With the aim to identify novel factors for ER stress resistance and virulence , UPR core genes were chosen for gene deletion based on increased or unchanged expression levels during clp1-modulated UPR . This set of genes comprised mainly factors that were previously not connected to any of these functions . To our surprise , we identified not a single gene important for ER stress resistance ( S4 Fig ) . Thus , contrary to our expectations , in this group of 29 UPR core genes no enrichment of genes involved in either ER stress resistance or virulence was observed . This finding might be explained by the existence of proteins with redundant or partially overlapping functions , which might support the robustness of the pathway [93] or help in fine-tuning and adapting the pathway during changing environmental conditions in planta . Alternatively , the physiological role of clp1-dependent UPR modulation might be primarily to prevent the deleterious overexpression of the above-mentioned known UPR genes , as previously hypothesized [36] . Deletion of spp1 did not affect ER stress resistance but completely abolished virulence ( S5 Fig ) , suggesting that Spp1 is a key factor for fungal virulence . Spp1 is specifically required during biotrophic growth since neither vegetative growth , nor resistance to ER- or cell wall stress was affected in spp1 deletion strains . The conserved domain architecture and catalytic YD/GLGD and substrate binding ( QPALLY ) motifs , as well as the subcellular localization pattern and the observation that deletion of spp1 can be complemented with the human SPP HM13 , strongly suggests that Spp1 is a bona fide SPP , cleaving type 2 transmembrane domains . The catalytic activity of Spp1 is required for its virulence specific function , since expression of Spp1D279A harboring a mutation of the conserved aspartate in the catalytic YD/GLGD motif [29 , 94] did not rescue the virulence defect of spp1 deletion strains . Spp1 is the single SPP encoded in the U . maydis genome , arguing against the possibility that functionally redundant proteins encoded in the U . maydis genome may mask specific functions of Spp1 during normal growth or stress adaptation . Although SPP proteins are implicated in ER homeostasis [28 , 30] , transcriptional regulation of SPPs by the UPR has not been reported for other organisms . However , we observed strong conservation of the Cib1 binding site and its location in the spp1 promoter of related smut species ( S17 Fig ) . Together with the increased expression of spp1 orthologs during biotrophic growth of U . hordei and Ustilago bromivora [95 , 96] , this suggests that the connection between the UPR , Spp1 and fungal virulence might not be restricted to U . maydis . It is tempting to speculate that in U . maydis cleavage of specific SPP substrates generates products with virulence specific functions . SPP substrates include but are not limited to subsets of signal sequence remnants in the ER membrane after processing of the precursor protein by the signal peptidase complex ( SP ) and type II transmembrane proteins [28 , 31 , 97 , 98] . Most substrates are subsequently degraded and pharmacological inhibition of the SPP activity in P . falciparum interferes with ERAD and prevents intraerythrocytic development [99] . In budding yeast and humans , SPP cleavage of the high-affinity zinc transporter Zrt1 and the UPR repressor XBP1u , respectively , targets the cleaved proteins for degradation via ERAD as well [28 , 31] . Since ERAD is not of major importance for ER stress resistance and virulence in U . maydis , the virulence phenotype of spp1 deletion strains is most likely not connected to this pathway . In the human pathogen A . fumigatus , only multiple deletions of genes encoding ERAD components led to reduced ER stress resistance but did not affect pathogenicity [100] . Importantly , SPP cleavage products are not always degraded but also exert biological functions . In humans , SPP cleavage of the hepatitis core protein C promotes trafficking of the core protein to lipid droplets [97] , and processing of MHC class I signal sequences generates cell surface epitopes that protect against the attack of natural killer cells [32] . However , similar mechanisms do not exist in fungi or plants and a major function for hypoxia adaptation through Spp1-mediated cleavage of the SREBP transcription factor appears unlikely . The establishment of a compatible interaction between biotrophic pathogens and their host requires plant defense suppression by secreted effectors [14] . However , Spp1 is not important for protein secretion as evidenced by monitoring secretion of Pit2 , Pep1 , Tin2 and Cmu1 under axenic conditions or expression of UPR marker genes during biotrophic development , contrasting the central role of a functional UPR for efficient secretion and processing of effector proteins [36 , 46 , 49] . Remarkably , secretion of Pep1 and Tin2 and processing of Pep1 and Cmu1 is strictly dependent on ER stress , providing obvious clues on how effector secretion and function might be coordinated and potentially modulated by the UPR and altered ER stress levels . Since U . maydis depends on the conserved catalytic activity of Spp1 but not on ERAD or other previously identified SPP substrates to cause disease we postulate that plant defense suppression is either achieved by specific activation of Spp1 , or , more likely , by cleavage of substrates that are specifically expressed during the biotrophic stage . Collectively , our data revealed a novel pathway of a biotrophic pathogen to suppress the plant defense and establish a compatible host-pathogen interaction .
Escherichia coli strain TOP10 ( Invitrogen ) was used for cloning purposes and amplification of plasmid DNA . Ustilago maydis cells were grown at 28°C in yeast-extract-peptone-sucrose ( YEPS ) light medium [101] , complete medium ( CM ) [102] , or yeast nitrogen base ( YNB ) medium [103] , supplemented with 1% glucose or 1% arabinose . Induction of Pcrg1 driven gene expression was performed as described by [104] . Filamentous growth was induced on potato dextrose plates containing 1% ( w/v ) activated charcoal [102] . ER stress tolerance was tested on yeast-nitrogen-base ( YNB ) media containing the indicated concentrations of tunicamycin ( Sigma-Aldrich ) . Sensitivity towards H2O2 , congo-red or calcofluor white was tested in drop-plate assays on YNB solid media containing the indicated concentrations of respective supplements . U . maydis strains used in this study are listed in ( S5 Table ) . Molecular methods used in this study followed described protocols [105] . DNA isolation and transformation procedures for U . maydis were performed as described previously [9] . Linearized plasmids or PCR amplified DNA was used for homologous integration into the U . maydis genome . All constructs were verified by Sanger sequencing prior transformation or PCR amplification . Correct integration of constructs into the U . maydis genome was verified by Southern hybridization . Q5 polymerase ( NEB ) was used for PCR amplification . Primers used in this study can be found in ( S6 Table ) . RNA was prepared from cells during exponential growth in axenic culture or from infected maize plants using Trizol reagent ( Invitrogen ) according to the manufacturer’s instructions [11] , followed by removal of residual DNA with Turbo DNase ( Ambion/Lifetechnologies ) . If RNA was used for RNAseq analysis samples were further column-purified with the RNeasy Kit ( Qiagen ) . Integrity of RNA was checked by ethidium bromide staining or with a Bioanalyzer equipped with an RNA 6000 Nano LabChip kit ( Agilent ) . All gene deletions were performed using a PCR based approach [106] . For the cib1-GFP fusion , the 5 . 5 kb SfiI 3xGFP-HygR fragment of plasmid pcib1-3xGFP [8] was replaced with the 2 . 5 kb SfiI GFP-NatR fragment from pUMa389 [107] to generate the plasmid pcib1-GFP . The resulting vector was used to generate plasmid pcib1-3xHA by exchanging the SfiI GFP-NatR cassette with a 1 . 8 kb SfiI 3xHA-NatR fragment from pUMa793 [108] . For replacement of the cib1 promoter with a tetracycline-regulated promoter , 1kb upstream of the cib1 start codon and 1kb of the cib1 open reading frame ( ORF ) were PCR amplified from genomic DNA , ligated to the SfiI cassette of pUMa707 [54] and integrated in the pCR2 . 1 TOPO vector ( Invitrogen ) generating plasmid pPtef-tTA-tetO:cib1 . To generate the spp1-mCherry fusion , the ORF of spp1 ( UMAG_02729 , UM521 ) lacking the stop codon was PCR amplified from genomic DNA introducing a BamHI site at the 5’ end and a BspHI site at the 3’ end and integrated into p123-mCherry [109] , to yield p123:spp1-mCherry . Cloning of orthologous genes from Sporisorium reilianum Srspp1 ( sr13785 , strain SRZ1 ) , Ustilago hordei Uhspp1 ( UHOR_04354 , strain Uh4857-4 ) and Aspergillus nidulans sppA ( ANID_08681 , strain AGB551 ) followed the same procedure , generating plasmids p123:Srspp1-mCherry , p123:Uhspp1-mCherry and p123:sppA-mCherry , respectively . For cloning of S . cerevisiae YPF1 the ORF ( YKL100C , strain sigma 1287 ) was PCR amplified from genomic DNA introducing BamHI sites at the 5’ and 3’ end removing the stop codon , and integrated into p123-mCherry to yield plasmid p123:YPF1-mCherry . The cDNA of the human HM13 ( BC062595 , cDNA clone ) was PCR amplified from the vector pCS6 ( BC062595 ) -TCH1303-GVO-TRI ( BioCat ) introducing a XmaI site at the 5’ end , a NcoI site at the 3’ end and removing the stop codon and subsequently ligated into p123-mCherry to yield p123:HM13-mCherry . To replace the otef promoter in p123:spp1-mCherry with the spp1-promoter , a 1 . 4kb spp1 promoter fragment was PCR amplified introducing a NdeI site at the 5’ end and a BamHI site at the 3’ end . The PCR fragment was integrated into p123:spp1-mCherry to generate pPspp1:spp1-mCherry . To generate the catalytically inactive version of spp1 ( spp1D279A ) , a point mutation was introduced into the ORF of spp1 by standard PCR procedures . Cloning of the PCR fragment followed the procedure as described for p123-spp1:mCherry , yielding plasmid p123:spp1D279A-mCherry . To generate the pep1-mCherry fusion , the ORF of pep1 ( UMAG_01987 , UM521 ) lacking the stop codon was PCR amplified from genomic DNA introducing a BamHI site at the 5’ end and a NcoI site at the 3’ end and integrated into p123-mCherry [109] , to yield p123:pep1-mCherry . To generate the tin2-mCherry fusion , the ORF of tin2 ( UMAG_05302 , UM521 ) lacking the stop codon was PCR amplified from genomic DNA introducing a BamHI site at the 5’ end and a NcoI site at the 3’ end and integrated into p123-mCherry [109] , to yield p123:tin2-mCherry . To generate the cmu1-mCherry fusion , the ORF of cmu1 ( UMAG_05731 , UM521 ) lacking the stop codon was PCR amplified from genomic DNA introducing a XmaI site at the 5’ end and a NcoI site at the 3’ end and integrated into p123-mCherry [109] , to yield p123:cmu1-mCherry . Microscopic analysis was performed using an Axio Imager . M2 equipped with an AxioCam MRm camera ( ZEISS ) or an Axio Imager . M1 ( ZEISS ) equipped with a CoolSNAP HQ2 CCD camera ( PHOTOMETRICS ) . All images were processed with ZEN 2 . 3 blue edition ( ZEISS ) . Chlorazol Black E staining was performed according to [104] . For microscopic analysis of cells after TM treatment U . maydis strains were grown in CM to an OD600 of 0 . 35 . TM was added to a final concentration of 5 μg/ml and cells were incubated for the indicated time to induce the UPR . For detection of reactive oxygen species ( ROS ) in infected leaf tissue , 3 , 3’-diaminobenzidine ( DAB ) was used as described previously [61] . Briefly , leaves ( third leaf ) were detached with a razor blade 1 cm above and 2 cm below the injection site 24h post infection and incubated for 12h in 1 mg/ml DAB solution under darkness at room temperature . For decolorization , leaves were immersed in ethanol ( 96% ) for 48h . For storage of the specimens , the leaves were transferred into 10% ( v/v ) glycerol . Brown polymerization products resulting from the reaction of DAB with ROS were microscopically identified using a binocular microscope ( Keyence Digital Microscope VHX-500F ) . qRT-PCR analysis was carried out as described before [46] . For all qRT-PCR experiments , mRNA was isolated from three independent biological samples , subjected to cDNA synthesis ( RevertAid First Strand cDNA Synthesis Kit , Thermo Scientific ) and analyzed in two technical repeats using MESA Green qPCR Mastermix Plus ( Eurogentec ) . qRT-PCR was performed on CFX Connect Real-Time PCR Detection System and the CFX Manager Software ( BioRad ) . Statistical significance was calculated with Student’s t-test . For RNAseq , strains were grown in YNB supplemented with 1% glucose and 0 . 2% ammonium sulfate ( YNBG ) overnight to an OD600 of 0 . 25 and shifted to YNB supplemented with 1% arabinose and 0 . 2% ammonium sulfate ( YNBA ) to induce Clp1 expression ( Pcrg1:clp1 ) . To induced the UPR , TM was added to a final concentration of 5 μg/ml and cells were further incubated for 4 hours at 28°C . Cells were harvested and quick-frozen in liquid nitrogen . RNA extraction followed the procedure as described above . 5 μg of total RNA was used to enrich mRNA using the NEB Next Poly ( A ) mRNA Magnetic Isolation Module ( NEB ) according to the manufacturers instructions . Strand-specific cDNA libraries were constructed with the NEBNext Ultra directional RNA library preparation kit for Illumina ( NEB ) . To assess quality and size of the libraries samples were run on an Agilent Bioanalyzer 2100 using an Agilent High Sensitivity DNA Kit as recommended by the manufacturer ( Agilent Technologies ) . Concentration of the libraries was determined using the Qubit dsDNA HS Assay Kit as recommended by the manufacturer ( Life Technologies ) . Sequencing was performed using the HiSeq4000 instrument ( Illumina Inc ) and the HiSeq 3000/4000 SR Cluster Kit for cluster generation and the HiSeq 3000/4000 SBS Kit ( 50 cycles ) for sequencing in the single-end mode , running 1x 50 cycles . A minimum of 15 Million raw reads were generated for individual samples . Raw RNAseq reads were aligned to the Ustilago maydis genome from Ensembl Genomes 33 [110] using STAR [111] version 2 . 4 . 1 . Read counts and RPM ( reads per million ) were calculated using custom Python scripts . Differential expression was assessed with DESeq2 [112] at an FDR threshold of 0 . 05 and a log2 fold change threshold of 1 or 2 . RNAseq data was deposited at EBI ArrayExpress ( https://www . ebi . ac . uk/arrayexpress/ ) under accession E-MTAB-7463 . Heat map of identified UPR core genes was visualized using ClustVis Web Tool [113] . Hierarchical clustering was performed using Euclidean distance and complete linkage for genes . UPR core genes were further analyzed using the Functional Catalogue annotation of the MIPS U . maydis database ( http://mips . gsf . de/funcatDB/ ) . ChIP analysis was done essentially as described before [46] , with the modification that chromatin was sheared in a Covaris S200 set to yield a DNA average size of approximately 100–300 bp . DNA was recovered by column purification ( PCR Purification Kit , Qiagen ) and subjected to library preparation . For ChIPseq experiments the libraries were prepared from 1 ng of enriched DNA or input DNA using the NEBNext Ultra II DNA Library Prep with Beads as recommended by the manufacturer ( New England BioLabs ) . To assess quality and size of the libraries , samples were run on an Agilent Bioanalyzer 2100 using an Agilent High Sensitivity DNA Kit as recommended by the manufacturer ( Agilent Technologies ) . Concentration of the libraries was determined using the Qubit dsDNA HS Assay Kit as recommended by the manufacturer ( Life Technologies GmbH ) . Libraries were sequenced on a HiSeq4000 instrument ( Illumina Inc ) using the HiSeq 3000/4000 SR Cluster Kit for cluster generation and the HiSeq 3000/4000 SBS Kit ( 50 cycles ) for sequencing in the single-end mode , running 1x 50 cycles . A minimum of 40 Million raw reads were generated for the ChIPseq experiments . Raw ChIPseq reads were aligned using Bowtie2 [114] version 2 . 0 . 0-beta7 to the Ustilago maydis genome from Ensembl Genomes 33 [110] . Peak calling was performed using PeakZilla [51] , GitHub commit version 7167f084e024676bcb34e5b5c3e1281910423c25 . ChIPseq data was deposited at EBI ArrayExpress ( https://www . ebi . ac . uk/arrayexpress/ ) under accession E-MTAB-7460 . Peak calling was performed individually for both biological replicates . Only peaks identified in both replicates were used for further analyses . Assignment of peaks to genes in case of divergent promoters was based on the relative distance to the translational start site ( tss ) and on gene expression after TM treatment . Peak scores were accumulated to promoter scores if more than one peak was identified in the promoter of a single gene and at least one peak score was above 40 . Promoter scores were filtered by a cut-off of 100 . Promoters harboring more than four peaks could never be assigned to differentially expressed genes and were thus discarded from further analysis . Normalized bigWig files were generated from BAM files derived from both replicates and visualized using the Integrative Genomics Viewer ( IGV ) [115] . For identification of possible binding motifs of Cib1 , sequences of assigned ChIP peaks derived from the UPR core gene set were subjected to the MEME ( Multiple EM for Motif Elicitation ) -ChIP analysis [116] . For plant infection studies the maize ( Zea mays ) cultivar Early Golden Bantam was used under controlled conditions using a CLF Plant Climatics GroBank with a 14 h ( 28°C ) /10 h ( 22°C ) day/night cycle . U . maydis strains were incubated in YEPSlight at 28°C to a final OD600 of 0 . 8–1 . 0 , washed with H2O and concentrated to an OD600 of 1 . 0 in H2O . 300–500 μl of the cell suspension were injected into the basal stem of 7 day-old maize seedlings . DPI treatment was performed as described previously [61 , 62] . All infection experiments were repeated three times or as otherwise indicated . Disease rating was performed according to [56] . Protein isolation and Western hybridization experiments were performed as described previously [117] . Commercially available rabbit anti-GFP ( Sigma-Aldrich ) ( 1:4000 dilution ) or anti-RFP [6G6] ( Chromotek ) ( 1:1000 dilution ) , were used to detect GFP- or mCherry-fusion proteins , respectively . Horseradish peroxidase-conjugated anti-mouse IgG ( Promega ) were used as secondary antibody . The Luminata Crescendo Western HRP substrate ( Merck Millipore ) was used for chemiluminescence detection . Secretion assays of Pit2-mC , Pep1 , Tin2 and Cmu1 were performed as described previously [46] , with small alterations . Instead of DTT , TM ( 5μg/ml ) was used to induce ER stress for 4 h prior to protein preparation . Stability of Cib1-GFP in response to Clp1 expression was determined with a doxycycline ( DOX ) based promoter shut-off system ( PtetO:cib1-GFP ) [54] . U . maydis strains were grown in CM supplemented with 1% glucose ( CMG ) to an OD600 of 0 . 35 , and shifted to CM supplemented with 1% arabinose ( CMA ) to induce Pcrg1-driven clp1 expression . For UPR activation , TM was added to a final concentration of 5 μg/ml . 4 h after UPR induction DOX ( 10 μg/ml ) was added ( T0 ) and protein extract was prepared from samples taken at the indicated time points [1 h ( T1 ) , 2 h ( T2 ) , 3 h ( T3 ) and 4 h ( T4 ) ] after DOX treatment . Cycloheximide ( CHX ) -based determination of Cib1-GFP stability was performed as described before [36] . Briefly , cells were grown as described for promoter shut-off assays . Protein biosynthesis was inhibited using CHX ( 100 μg/ml ) and cells were sampled directly before ( T0 ) , or at the indicated times after CHX treatment [30 min ( T1 ) , 60 min ( T2 ) or 90 min ( T3 ) ] . Cib1-GFP levels ImageJ were quantified using ( https://imagej . nih . gov/ij/ ) and normalized to Ponceau S stained bands . Stability of proteins was calculated relative to T0 . Experiments were performed in three biological replicates . Statistical significances ( P value ) were calculated using Student’s t test . Protein phosphatase assays were performed after immunoprecipitation of Cib1-GFP followed by on-bead treatment with lambda-phosphatase ( NEB ) . Cells were incubated as described for promoter shut-off experiments . Briefly , 4 hours after TM-mediated UPR activation ( 5 μg/ml ) equal culture volumes were centrifuged , cell pellets were washed once with tris buffered saline ( 20 mM Tris-HCl , 137 mM NaCl , pH 7 . 6 , supplemented with 2x cOmplete proteinase inhibitor ( ROCHE ) ( PI ) and phosphatase inhibitor cocktail ( 1 mM NaF , 0 . 5 mM Na3VO4 , 8 mM β-glycerophosphat , PhI ) . The pellet was resuspended in 750 μl buffer B-300 ( 300 mM NaCl , 100 mM Tris , pH 7 . 5 , 10% Glycerol , 1 mM EDTA , supplemented with 2x PI and PhI ) , shock frozen in liquid nitrogen and disrupted in a cell mill ( Retsch MM400 , 30Hz , 2min ) . After cell lysis , 750 μl of B+300 buffer ( 300 mM NaCl , 100 mM Tris ( pH 7 . 5 ) , 10% Glycerol , 1 mM EDTA , 0 . 1% NP40 , supplemented with 2x PI and PhI ) was added and the whole cell lysate was centrifuged at 45 , 000 rcf for 30 minutes at 4°C . The supernatant was added to 60 μl of magnetic agarose GFP-Trap beads ( Chromotek ) and incubated for 3h at 4°C on a rotating wheel . After washing the beads 2x with 500 μl of B-300 buffer and removing the supernatant , beads were resuspended in 600 μl of buffer B-300 ( supplemented with 2x PI ) and evenly distributed in 200 μl aliquots . The supernatant was discarded and 1x lambda phosphatase buffer ( NEB ) was added to each sample . 1200U of lambda phosphatase were added . Control samples were left untreated or supplemented with 2x PhI . After incubation for 30 minutes at 30°C the supernatant was discarded and 30 μl 1x Roti Load 1 ( Carl-Roth ) was added to the beads and boiled at 98°C for 3 min . Samples were run on a 10% SDS-PAGE and subjected to Western hybridization . All steps were performed in Protein LoBind Tubes ( Eppendorf ) . Experiments were repeated at least three times . Sequence data from this article can be found at the Munich Information Center for Protein Sequences Ustilago maydis database ( http://mips . helmholtz-muenchen . de/genre/proj/ustilago/ ) and the National Center for Biotechnology Information database under the following accession numbers: UMAG_00258 , XP_011386180 . 1; hrd1 , UMAG_00542 , XP_011386378 . 1; UMAG_00783 , XP_011386557 . 1; lhs1 , UMAG_00904 , XP_011386916 . 1; UMAG_01025 , XP_011387002 . 1; UMAG_01112 , XP_011387066 . 1; UMAG_01232 , XP_011387166 . 1; pit2 , UMAG_01375 , XP_011387264 . 1; pep1 , UMAG_01987 , XP_011387901 . 1; clp1 , UMAG_02438 , XP_011388726 . 1; UMAG_02487 , XP_011388764 . 1; spp1 , UMAG_02729 , XP_011389095 . 1; UMAG_02944 , XP_011389351 . 1; UMAG_03404 , XP_011389878 . 1; UMAG_03507 , XP_011389952 . 1; UMAG_03541 , XP_011389978 . 1; UMAG_03665 , XP_011390151 . 1; ost3 , UMAG_04198 , XP_011390684 . 1; UMAG_04605 , XP_011390905 . 1; eIF2b , UMAG_04869 , XP_011391708 . 1; UMAG_04896 , XP_011391221 . 1; UMAG_05009 , XP_011391306 . 1; tin2 , UMAG_05302 , XP_011392015; mpd1 , UMAG_05352 , XP_011392054 . 1; srb1 , UMAG_05721 , XP_011391469 . 1; cmu1 , UMAG_05732 , XP_011391476 . 1; der1 , UMAG_05898 , XP_011392243 . 1; UMAG_10006 , XP_011386596 . 1; cln1 , UMAG_10287 , XP_011389173 . 1; UMAG_10686 , XP_011391738 . 1; doa10 , UMAG_10911 , XP_011390969 . 1; UMAG_10921 , XP_011386693 . 1; UMAG_11083 , XP_011390090 . 1; UMAG_11190 , XP_011392333 . 1; der2 , UMAG_11402 , XP_011388858 . 1; UMAG_11513 , XP_011390555 . 1; UMAG_11594 , XP_011392502 . 1; UMAG_11651 , XP_011387452 . 1; UMAG_11763 , XP_011391003 . 1; cib1 , UMAG_11782 , XP_011390112 . 1; UMAG_12149 , XP_011386842 . 1; UMAG_12178 , XP_011388139 . 1; UMAG_12304 , XP_011391965 . 1; UMAG_12318 , XP_011392356 . 1; UMAG_12332 , XP_011388414 . 1; bip1 , UMAG_15034 , XP_011387505 . 1; Srspp1 , sr13785 , CBQ73124 . 1; Uhspp1 , UHOR_04354 , CCF52970 . 1; Ubspp1 , UBRO_04354 , SAM82079 . 1; HM13 , BC062595 , NP_110416 . 1; sppA , ANID_08681 , XP_681950 . 1; sppA , AFUA_6G02150 , XP_747862 . 1; YPF1 , YKL100C , NP_012822 . 1; Pfspp , PF3D7_1457000 , XP_001348717 . 2; SPPL2B , NP_694533 . 1; PR1 , AAC25629 . 1; PR3 , NP_001340366 . 1; PR4 , NP_001130495; PR5 , NP_001105702 . 2; CC9 , BN000513 . 1; BBI , EU955113 . 1; POX12 , ACG36543 . 1; ATFP4 , NP_001152411 . 1; GAPDH , NP_001105413 . 1 | Biotrophic pathogens establish compatible interactions with their host to cause disease . A critical step in this process is the suppression of plant defense responses by secreted effector proteins . In the maize infecting fungus Ustilago maydis expression of effector encoding genes is coordinately upregulated at defined stages of pathogenic development in so-called effector waves . Efficient secretion of the multitude of effectors relies on the unfolded protein response ( UPR ) to maintain homeostasis of the endoplasmic reticulum . Activation of the UPR is connected to the control of fungal proliferation through direct protein-protein interactions between the UPR regulator Cib1 and the developmental regulator Clp1 . Here , we show that this interaction leads to functional modification of Cib1 and modulation of UPR gene expression to adapt the UPR for long-term activity in the plant . Within a core set of UPR regulated genes we identify the signal peptide peptidase Spp1 as a key factor for fungal virulence . We show that Spp1 requires its conserved catalytic activity to suppress the plant defense and cause disease . The virulence specific function of Spp1 does not involve pathways previously known to be associated with Spp1-like proteins or plant defense suppression , suggesting a novel role for Spp1 substrates in biotrophic interactions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"ustilago",
"maydis",
"pathology",
"and",
"laboratory",
"medicine",
"fungal",
"genetics",
"pathogens",
"plant",
"physiology",
"fungi",
"plant",
"science",
"model",
"organisms",
"experimental",
"organism",
"systems",
"plant",
"p... | 2019 | Signal peptide peptidase activity connects the unfolded protein response to plant defense suppression by Ustilago maydis |
Transposable elements represent a large proportion of the eukaryotic genomes . Long Terminal Repeat ( LTR ) retrotransposons are very abundant and constitute the predominant family of transposable elements in plants . Recent studies have identified chromoviruses to be a widely distributed lineage of Gypsy elements . These elements contain chromodomains in their integrases , which suggests a preference for insertion into heterochromatin . In turn , this preference might have contributed to the patterning of heterochromatin observed in host genomes . Despite their potential importance for our understanding of plant genome dynamics and evolution , the regulatory mechanisms governing the behavior of chromoviruses and their activities remain largely uncharacterized . Here , we report a detailed analysis of the spatio-temporal activity of a plant chromovirus in the endogenous host . We examined LORE1a , a member of the endogenous chromovirus LORE1 family from the model legume Lotus japonicus . We found that this chromovirus is stochastically de-repressed in plant populations regenerated from de-differentiated cells and that LORE1a transposes in the male germline . Bisulfite sequencing of the 5′ LTR and its surrounding region suggests that tissue culture induces a loss of epigenetic silencing of LORE1a . Since LTR promoter activity is pollen specific , as shown by the analysis of transgenic plants containing an LTR::GUS fusion , we conclude that male germline-specific LORE1a transposition in pollen grains is controlled transcriptionally by its own cis-elements . New insertion sites of LORE1a copies were frequently found in genic regions and show no strong insertional preferences . These distinctive novel features of LORE1 indicate that this chromovirus has considerable potential for generating genetic and epigenetic diversity in the host plant population . Our results also define conditions for the use of LORE1a as a genetic tool .
A large proportion of the eukaryotic genome is composed of transposable elements ( TEs ) . In flowering plants , Long Terminal Repeat ( LTR ) retrotransposons have been regarded as the largest order of TEs [1] , [2] and it has been suggested that the ratio between propagation and exclusion of LTR retrotransposons may have affected the size of host genomes [3] , [4] . In line with this notion , large plant genomes usually contain substantially more LTR retrotransposons than small plant genomes [5] , [6] . However , data from a wide range of flowering plants strongly suggest that LTR retrotransposons are not distributed evenly in genomes . Biased accumulation has led to the formation of LTR retrotransposon-rich , gene-poor heterochromatic blocks , which separate gene-rich euchromatic regions [7] . Thus , the activity of LTR retrotransposons has contributed remarkably towards generating the basic structure of current plant genomes . In flowering plants , the LTR retrotransposons have been classified into two superfamilies , Gypsy and Copia , according to their structural features [8] . In many plants , Gypsy outnumbers Copia [9]–[13] . An exception is grapevine , in which the number of Copia elements exceeds that of Gypsy [14] . Chromovirus is a most widely-distributed lineage of Gypsy , characterized by a chromodomain at the carboxyl terminal of the ORF [15] , [16] . It has been proposed that the insertion site preference of chromoviruses is controlled by the chromodomain [15] , [16] , and this suggestion has been supported by functional characterization of MAGGY , identified in the rice blast fungus Magnaporta grisea [17] , [18] . The MAGGY chromodomain was shown to interact with histone H3 di- and tri-methyl K9 , which are hallmarks of heterochromatin [18] . When it was fused to the integrase of Tf1 retrotransposon , the modified Tf1 preferentially transposed into heterochromatic regions in Schizosaccharomyces pombe genome [18] . In flowering plants , chromoviruses are phylogenetically distinct from the lineage containing MAGGY and they are classified into four clades , Reina , Tekay , Galadriel and CRM [16] , [19] . Members of CRM were originally known as Gypsy elements which accumulate in centromeric and pericentromeric regions in plant genomes [20]–[23] . Since all four clades have been identified in both dicots and monocots , and Reina and CRM elements have been found in angiosperms and gymnosperms , these elements are likely to have an ancient origin within the seed plants [16] , [19] . In order to complement these evolutionary studies , a precise characterization of retrotransposon transpositional activity is now being pursued by experimental analyses , and this activity represents one of the subjects that must be addressed if we are to develop a deeper understanding of plant genome dynamics and evolution . Previously , most experimental studies of transpositional activity and the regulation of plant LTR retrotransposons were conducted using three Copia elements , Tnt1 and Tto1 in tobacco , and Tos17 in rice . Transpositions of these elements were observed only in cultured cells , where their transcriptional up-regulation occurs [24]–[26] . Since transpositional activity is immediately repressed in regenerated plants due to a decrease in transcription , transpositions in intact plants have not been well characterized . Thus far , transposition of Tos17 has been observed in intact transgenic plants in which the transcriptional level of a gene encoding histone H3K9 specific methylase was downregulated by RNA interference [27] , but the spatio-temporal pattern of transposition remained unclear . Furthermore , little is known about the transpositional activity of plant Gypsy elements , including chromoviruses , despite their high abundance in plant genomes . In more than a decade of studies , the model legume Lotus japonicus has facilitated dissection of the molecular mechanisms governing symbiotic nitrogen fixation with rhizobia . The L . japonicus genome has been sequenced and sequence data covering 67% of the genome ( 472 Mb ) , corresponding to 91 . 3% of the gene space , is now available [13] . From this model legume , we have identified two transpositionally active LTR retrotransposon families designated as LORE1 and LORE2 ( Lotus Retrotransposon 1 and 2 ) [28] , [29] . Both belong to the Gypsy superfamily and were first identified as insertions in symbiotic mutants isolated from a transgenic plant population established by tissue culture- mediated transformation [28]–[31] . However , the machinery underlying their activation remained to be characterized . Both LORE1 and LORE2 encode unique long open reading frames ( ORFs ) with a chromodomain at the carboxyl terminal ends , which suggests that they are chromoviruses ( Figure 1A ) [29] . Although this chromodomain was overlooked in the original characterization of LORE1 [28] , Novikova et al . re-classified LORE1 as a member of the Reina clade of chromovirus [19] . Previously , we estimated the number of “preexisting copies” ( insertions that were already present in a plant accession ) of LORE1 in the Gifu accession as ten , and obtained full or partial sequences for nine out of the ten preexisting LORE1 copies [28] . Nucleotide sequence polymorphisms among the nine copies enabled us to distinguish them from each other , and we designated them in alphabetical order as LORE1a , b , c , d , e , f , g , h , and i [28] . In this report , we show that in the Gifu accession , the preexisting LORE1a can be epigenetically de-repressed in standard tissue culture . However , transpositions per se occur primarily in pollen , i . e . , male gametophytes , of regenerated intact plants , and so far new insertions generated in cultured cells have not been detected . We assume that the pollen-specific LTR promoter of LORE1a regulates the spatio-temporal pattern of transposition . Although LORE1 is a chromovirus , it does not appear to have a strong insertional preference for heterochromatin . These distinctive features of LORE1 underlie its ability to generate insertional polymorphisms , leading to a wide range of genetic and epigenetic diversity in a population . The results also define conditions for using LORE1 for insertion mutagenesis .
The transpositional activity of LORE1 was first demonstrated by the identification of four symbiotic mutant alleles , nin-7 , symrk-1 , nup133-3 and nap1-1 , in which gene inactivation was caused by the insertion of LORE1 [28] , [32] . As all four mutants were isolated from the same Ac/T-DNA tagging population established using the L . japonicus Gifu accession [30] , [31] , we screened other plants of the same population for LORE1 transpositions . Sequence-specific amplified polymorphism ( SSAP ) analysis of LORE1 insertion sites detected new transpositions in 32 plants out of a sub-population of 41 plants ( Population 1 in Table 1 ) , indicating that LORE1 was widely active in this population . Next , we investigated whether LORE1 transpositions were present in four transgenic or non-transgenic regenerated plant populations created using the Gifu accession . To detect new insertion sites of LORE1 , we used SSAP to analyze the T1 and R1 progeny of primary transformants ( T0 ) and of primary non-transgenic regenerated plants ( R0 ) . In addition to population 1 ( Table 1 ) , transpositions were detected in three of the other four populations . Importantly , transposition was detected in transgenic plants generated using six different constructs , as well as in non-transgenic regenerated plants . These results suggest that the simple process of in vitro tissue culture can activate LORE1 in a stochastic manner that is independent of the presence or absence of transgenes , antibiotic selection , and of the composition and contents of transgene constructs . The newly transposed LORE1 copies observed in R1/T1 plants might have resulted from transpositions in cultured cells and/or in the parental R0/T0 plants . However , LORE1 transposition was absent , infrequent , or below the detection levels of the SSAP method in a total of 27 plants from the initial R0/T0 plants from populations 2 and 3 ( Table 1; data not shown ) . Previously , we observed the absence of obvious transcriptional or transpositional activation of LORE1 in cultured cells [28] , [29] . These results suggest that even though LORE1 was apparently de-repressed in tissue culture , the transpositions per se appear to have occurred in regenerated intact plants , rather than in the cultured cells ( see details in the next section ) . To gain more precise information about transposition of LORE1 in intact plants , eight independent T0 plants were randomly selected from population 3 ( Table 1 ) and investigated together with their T1 progeny . LORE1 transpositions were detected in the T1 progeny from 6 of the 8 T0 plants . A typical result of a genomic Southern blot analysis and SSAP analysis of a T0 and its 10 T1 progeny plants ( in this instance plant line no . 30 ) are shown in Figure 1B and Figure S1 , respectively . Notably , the banding pattern in the T0 plants was the same as in the control Gifu , again indicating absent or infrequent LORE1 transposition in the primary regenerated plants ( Figure 1B ) . However , additional bands corresponding to newly transposed LORE1 copies were detected in the T1 progeny ( Figure 1B and Figure S1 ) . The highly polymorphic banding pattern indicates the occurrence of frequent independent transpositions of LORE1 in T1 plants ( Figure 1B and Figure S1 ) . Next , we determined whether the new insertions of LORE1 found in the T1 plants were the result of transmission of previous transpositions in somatic cells from T0 forming sectors or of de novo transposition . We analyzed T1 plants originating from two seed pods at the top of the same shoot of the parental T0 plant ( Figure 1C ) . We did not detect any new bands that were shared by the two neighboring pods , or T1 plants originating from the same pod . This result indicates that the majority of LORE1 transpositions occurred at late developmental stages in T0 plants . Reciprocal crosses between plant no . 30 ( from the Gifu accession ) and plants from the MG20 accession were used to determine if the new transposed copies detected in the T1 plant were transmitted via male or female gametes . Five F1 plants obtained from each reciprocal cross were analyzed for LORE1 copy number ( Figure 1D ) . In total , 21 bands corresponding to new LORE1 transpositions were detected among the 5 F1 plants obtained from the MG20 ( female ) × no . 30 ( male ) cross . In contrast , only 1 newly transposed LORE1 copy was detected in 5 F1 plants from no . 30 ( female ) × MG20 ( male ) cross . We conclude that although LORE1 is active in both male and female gametophytes , its activity is much higher in male tissues . Next , we used parent-specific single nucleotide polymorphisms ( SNPs ) in the flanking regions to determine the parental origin of the seven new insertion sites in MG20x30 F1 plants . This analysis showed that all the new transpositions originated from Gifu , the pollen donor . Hence , the majority of LORE1 transpositions detected in the F1 plants seemed to occur before fertilization . Altogether , LORE1 was revealed to be robustly active especially in male gametophytes . Previous reports indicate that activated retrotransposons can be re-silenced again by activities such as copy number-dependent establishment of epigenetic silencing [33] , [34] . However , LORE1 was still active in three T1 plants that already possessed an increased number of LORE1 copies ( Figure S2A ) . This finding indicates that once activated , LORE1 was able to transpose over at least two successive generations . On the other hand , we also observed that LORE1 was inactivated in the nup133-3 mutant , in which a single new transposition was detected in the Nup133 gene ( Figure S2B ) . Since the newly inserted LORE1 copies identified in the three symbiotic mutant alleles ( nfr5-2 , symrk-2 , and nup133-3 ) were identical to one of the nine preexisting copies , LORE1a , we suspected that LORE1a was preferentially activated [28] . LORE1a-specific SNPs were identified in regions 1 and 2 ( Figure 1A ) and in all eight of the newly-transposed LORE1 fragments from population 3 . This observation is consistent with our suggestion that LORE1a is responsible for the majority of LORE1 transpositions described here . The transpositional activity of retrotransposons is often controlled at the transcriptional level [1] , [2] , [24]–[26] . We used RT-PCR to compare the levels of LORE1 transcription in mature flowers containing both male and female gametophytes ( where LORE1 transposition presumably occurs ) . Among the eight T0 plants , including no . 30 from population 3 ( Table 1 ) , higher levels of LORE1 transcription were observed in the six T0 plants that possessed active LORE1 elements , compared to the control Gifu plant ( Figure 2A ) . This finding indicated a correlation between the transcriptional and transpositional activities of LORE1 . To determine which LORE1 family members were present in the transcript pool , RT-PCR products were TA cloned and sequenced . RT-PCR products spanning regions 1 and 2 were amplified separately from flowers of the control Gifu plant and two T0 plants ( nos . 30 and 45 ) that exhibited LORE1 activity . LORE1a-specific SNPs were present in the region 1 of 7/16 , 15/15 and 15/16 clones from the control Gifu , no . 30 and no . 45 plants , respectively . For region 2 , LORE1a-specific SNPs were present in 3/12 , 16/16 and 16/16 clones from the control Gifu , no . 30 and 45 plants , respectively . These data suggest that transcriptional activation is responsible for the preferential transposition of LORE1a among the family members . This expectation is supported by the following lines of evidence: i ) a generally increased level of LORE1 transcripts in flowers of active lines; ii ) a clear increase in LORE1a transcripts in two activated plants; and iii ) all transposition events detected thus far are of LORE1a origin . The pattern of LORE1a activation via tissue culture is different from that of other well-characterized retrotransposons such as Tos17 , Tto1 , and Tnt1 , which are activated and transpose during tissue culture , resulting in a copy number increase in the primary regenerated plants ( R0 ) [24]–[26] . We hypothesized that tissue- or cell-specific transcription determines the unique spatio-temporal pattern of LORE1 transposition . To test this hypothesis , we compared LORE1 transcript levels in leaves and flowers among four T0 plants and a control Gifu plant , as well as its transcriptional level in cultured cells ( Figure 2B ) . We found that there were no detectable differences in LORE1 transcript levels in the leaves of the four T0 plants or in the control Gifu plant . In contrast , high levels of LORE1 transcripts accumulated in the flowers of plant nos . 3 and 30 compared to nos . 11 and 42 , or the control Gifu plants and cultured cells . Furthermore , high LORE1 transcript levels were detected in pollen from the two T0 plants exhibiting LORE1 activity , compared to the two T0 plants without LORE1 activity or the control Gifu plant ( Figure 2C ) . These observations suggest that LORE1 has transpositional activity in pollen and that tissue specificity is controlled at the transcriptional level . Since the 5′ LTR is known to function as a promoter for LTR retrotransposons [1] , we determined promoter activity of the LORE1a LTR using a transgenic L . japonicus Gifu accession carrying LORE1a LTR fused to a GUS reporter gene . GUS activity was detected in mature pollen grains that were released from anthers and had accumulated at the tip on the inside of the keel ( Figure 3A ) , as well as in isolated pollen grains ( Figure 3B ) . We could not detect LTR-driven GUS activity in any other tissues ( data not shown ) . A similar pattern of GUS activity was observed in three out of six independent transgenic plant lines . These results are in good agreement with the RT-PCR analyses , which indicate up-regulation of LORE1 transcription in pollen grains ( Figure 2C ) . To investigate the LTR promoter activity in a heterologous system , we generated transgenic Arabidopsis plants carrying the same construct . Four out of the seven Arabidopsis transgenic lines showed GUS activity in hydrated pollen grains on stigmas and in pollen tubes ( Figure 3F and 3G ) . Prolonged staining for GUS activity detected weaker expression in developing young anthers ( Figure S3A ) . In the youngest anthers showing activity , GUS was detected primarily in cell layers around the developing pollen , rather than in the developing pollen grains ( Figure S3C and S3E ) . No GUS activity was detected in other tissues . Taken together , these results indicate that the LORE1 LTR specifically promotes transcription in pollen and that the tissue specificity of the cis-elements may be operational in a wide range of flowering plants . The reported locations of several chromoviruses in the host plant genomes suggest that chromoviruses preferentially accumulate in heterochromatic regions [18] , [20]–[23] . Of the nine preexisting LORE1 copies so far identified , the insertion sites of LORE1d , e , f , h , and i were found in genomic clones containing highly repetitive sequences , which were potential heterochromatic regions . However , the remaining four , LORE1a , b , c and g , were found in contigs that did not display any apparent heterochromatic characteristics ( S . S . unpublished data ) . To investigate whether LORE1 exhibits a strong insertion site preference for heterochromatic regions , we used SSAP to obtain flanking sequences located immediately 5′ of new insertions in the T1 and R1 populations . A total of 97 SSAP fragments longer than 40 bp were analyzed by homology search using public databases including the L . japonicus genome sequence data obtained from the MG20 accession [13] . The absence of the 97 LORE1 insertions in the wild-type Gifu accession was confirmed by PCR ( data not shown ) . In this analysis , only sequences showing homology higher than 77% , along stretches longer than 40 bp and with bit scores larger than 58 , were considered homologous sequences . For the 75% of the LORE1 flanking sequences ( 73 out of the 97 ) , homologous sequences including possible identical ( allelic ) sequences were identified from the published L . japonicus genome sequences ( Table 2 ) . The percentage ( 75% ) is close to the coverage of the whole genome reported for the genome sequence project ( 67% ) [13] . Among the 73 sequences , 37 were protein coding cellular genes or expressed sequence tags ( ESTs ) , 11 were homologous to transposable elements ( TEs ) , and the residual 25 did not show homology to genes or TEs and were categorized as unknown ( Table 2 ) . On the other hand , among the 24 fragments that did not show significant homology with L . japonicus sequences , 6 were classified as genes or ESTs , one was categorized as a TE , and the remaining 17 were classified as unknown ( Table 2 ) . Thus , a total of 43 sequences were assumed to be in genic regions . Among the 43 , 31 were predicted to be exonic , since the insertion site was positioned in a region homologous to protein coding sequences and/or deposited ESTs . In contrast , 12% of the 97 LORE1 flanking sequences showed homology with TEs , which is lower than the predicted TE content of the L . japonicus genome ( 36% ) derived from end-sequencing data of randomly selected BAC clones ( S . S . unpublished data ) . Finally , we physically mapped 24 of the 73 SSAP sequences , and 4 of the 9 preexisting LORE1 members whose positions could be uniquely assigned , to the latest version of L . japonicus chromosome pseudo molecule [13] ( Figure 4 ) . This mapping indicated that the new insertion sites were distributed across the Lotus genome and no strong preference for LORE1 insertion sites was observed from those data . Because of the frequent but stochastic derepression of LORE1a in regenerated plant populations ( Table 1 ) , we predicted that LORE1a activation accompanying tissue culture was induced epigenetically rather than genetically . We examined the status of cytosine methylation around the 5′ end of LORE1a by Southern blot analysis using two restriction enzymes , Hind III and Alu I , which are sensitive to cytosine methylation at residues inside their recognition site [35] . We examined genomic DNA from five T0 plants ( nos . 3 , 11 , 30 , 42 and 45 ) , together with the control Gifu ( Figure 5A ) . When Hind III was used to digest genomic DNA from leaves , we observed distinct bands ( approximately 1 . 5 kb ) in all of the five plants , suggesting the absence of cytosine methylation at the two Hind III sites surrounding the region complementary to the DNA probe used in this analysis ( Figure 5B ) . When genomic DNA samples were digested with Alu I , signals corresponding to approximately 300 and 650 bp DNA fragments were detected in each plant ( Figure 5A ) . We assumed that the lower band signals represented a mixture of three Alu I fragments of 263 , 284 , and 306 bp , resulting from the digestion of the Alu I site 5′ adjacent to LORE1a and one of three Alu I sites in the 5′ LTR ( Figure 5B ) . Thus , detection of the smaller hybridizing bands indicates the presence of hypomethylated Alu I sites in the 5′ LTR . On the other hand , the larger band was assumed to correspond to the 640 bp Alu I fragment , resulting from the absence of hypomethylated cleavable Alu I sites in the 5′ LTR ( Figure 5B ) . Detection of signals from both high and lower sized DNA fragments indicates heterogeneity of the methylation status at the three Alu I sites in the 5′ LTR of each of the six investigated plants . However , the relative signal intensity of these DNA fragments showed variation among the five plants . The intensity of lower bands ( corresponding to a hypomethylated status ) was predominant in plant nos . 3 and 30 , which have active LORE1a . However , the higher band ( corresponding to hypermethylated alleles ) was more intense in plant no . 11 , which did not have active LORE1a . In plants nos . 42 and 45 , both higher and lower bands were detected , with intensities similar to that of the control Gifu ( Figure 5A ) . These trends in the relative signal intensity between the large and smaller sized bands were reproducible in independently extracted genomic DNA ( Figure S4 ) . The banding patterns observed in flowers , where transcriptional activation of LORE1a was observed , were similar to those observed in leaves ( Figure S4 ) . This finding suggests that no obvious changes in cytosine methylation pattern can be correlated to changes in LORE1 transcriptional level between the two tissues . Altogether , it would appear that T0 plants have a variable epigenetic status for LORE1a , and that it is different from Gifu control plants . An independent determination of the cytosine methylation status was obtained by bisulfite sequencing of the 5′ LTR of LORE1a in the five T0 plants and control Gifu . The same genomic leaf DNA samples used in the Southern blot in Figure 5 were analyzed , and twenty to twenty-four amplicons were sequenced from each plant line . This analysis revealed that cytosine residues in U3 , the promoter region of LORE1a containing the three Alu I sites , are frequently methylated in control Gifu DNA , especially at CG and CHG sites ( Blue bars in Figure 6A–6E ) . Graphical representation of the methylation status obtained from twenty amplicons showed some heterogeneity in the cytosine methylation patterns of the control Gifu ( Figure S5A ) . This correlates with data obtained from the Hind III and Alu I digestion patterns ( Figure 5 and Figure S4 ) . LORE1a is activated in plant no . 30 , and compared with control Gifu , this line showed a dramatic decrease in the cytosine methylation level throughout the investigated region ( Figure 6C and Figure S5D ) . Plant no . 3 possesses activated LORE1a and it showed a general decrease in the methylation level in the U3 region; in three of twenty-three amplicons a complete loss of cytosine methylation in U3 was observed ( Figure 6A and Figure S5B ) . LORE1a remains inactive in plant no . 11 , and methylation at CG and CHG sites was maintained , as well as being very evident in the U3 ( Figure 6B and Figure S5C ) . Plant nos . 42 and 45 showed similar methylation patterns when averaged among clones ( Figure 6D and 6E ) . However , two amplicons corresponding to alleles that were completely demethylated in U3 were observed in plant no . 45 , in which LORE1 is active , but not in no . 42 , in which LORE1 remains inactive ( Figure S5E and S5F ) . Among the T0 plants analyzed , these data support the idea that there may be a correlation between LORE1a activation and the presence of LORE1a alleles that have totally lost cytosine methylation in U3 . To determine whether alteration in the methylation pattern occurs in the same region of other LORE1 loci , we used bisulfite sequencing to determine the methylation status of two LORE1 loci , LORE1b and LORE1f , which contain 5′ LTRs identical to that of LORE1a . This analysis revealed that the cytosine methylation profile of LORE1f is similar to that observed for LORE1a in control Gifu ( blue bars in Figure 6K–6O ) . Specifically , it shows a higher level of methylation in the U3 region compared with the remaining regions in the investigated areas . However , in contrast to LORE1a , the methylation profile of LORE1f was largely unchanged among the five T0 plants investigated ( red bars in Figure 6K–6O ) . LORE1b showed a moderate level of methylation throughout the investigated region in control Gifu , resulting in a flatter profile of methylation compared with LORE1a and LORE1f ( blue bars in Figure 6F–6J ) . The significant decrease in methylation levels in LORE1b was observed in all the 5 T0 plants , even though the level of decrease differed ( red bars in Figure 6F–6J ) . Taken together , the bisulfite sequencing unveiled variation of epigenetic status at LORE1 loci in control Gifu plants and indicated alteration of this status in the five T0 plants investigated . A characteristic observed with LORE1a was the variability of epigenetic changes among the T0 plants , whereas LORE1b and LORE1f exhibited stability or rather similar changes among the five T0 plants .
After the submission of this article , Tsukahara et al . reported the identification of a Gypsy element transposed in intact ddm1 mutant plants of Arabidopsis thaliana [51] . Precise characterization of the behavior of the Gypsy element , together with that of LORE1 , will facilitate our understanding of the interaction between LTR retrotransposons and plant genomes .
The Gifu accession of Lotus japonicus was used to generate both the transgenic and regenerated populations . The MG20 accession was used in the reciprocal crosspollination experiment with the T0 plant exhibiting LORE1 activity . For promoter analysis of LORE1 LTR using Arabidopsis thaliana , the ecotype Columbia was used to generate transgenic plants . Transgenic and regenerated plant populations were produced from the Gifu accession using two different protocols . Populations 1 and 2 were generated according to the method described in [52] . Populations 3 , 4 , and 5 were generated following the method described in [53] . Antibiotic selection was not used when populations 2 and 5 were produced . The 225 bp LTR of LORE1a , corresponding to the region from 137 bp to 361 bp of the AJ966990 sequence , was cloned into a multi-cloning site upstream of an intron-containing GUS gene in the binary vector pZN-GUS [54] . The resulting plasmid was introduced into Agrobacterium tumefaciens strain EHA105 . Arabidopsis thaliana ecotype Colombia was subsequently infected to generate transgenic plants following the method described in [55] . L . japonicus Gifu accession was infected with the same Agrobacterium strain and transgenic plants were generated following the method described in [53] . Genomic Southern blots were carried out following the method described in [29] . Hind III was used to digest genomic DNA in the Southern blot analyses shown in Figure 1 and Figure S2 . Hind III and Alu I were used to digest genomic DNA in Southern blot analyses shown in Figure 5 and Figure S4 . Washes were performed at high stringency ( 65°C , 0 . 1x SSC , 0 . 1% SDS ) . The DNA probe used in Figure 1 and Figure S2 was generated by PCR using the primer pair LORE1gagF ( 5′-GTTGCCAGTATCGCCATGGACG-3′ ) and LORE1gagR ( 5′-GGATTGAGGCCTCCAAGATAAC-3′ ) , and BAC DNA containing LORE1a [28] . The DNA probe used in Figure 5 and Figure S4 was generated by PCR using the primer pair 5′FLKF ( 5′-TTGACCTGCTCTTCAGTGCATG-3′ ) and 5′FLKR ( 5′-GAATCCGGGTATAAGGGTTCC-3′ ) . The Megaprime DNA Labeling System ( GE Healthcare ) was used for labeling the DNA probes with alpha-32P-dCTP . SSAP analyses to detect new LORE1 insertions were conducted as described in [28] . In brief , genomic DNA was digested with Mse I ( New England Biolabs ) , and ligated with Mse I adapters . The first PCR was conducted using a primer annealing to a internal region of LORE1 and oriented outward , and a primer specific to the Mse I adapters . A nested PCR was conducted using the first PCR reaction as template . The amplified SSAP fragments were electrophoresed on polyacrylamide sequencing gels , and detected by silver staining . Bands for putative new insertions , i . e . , absent from control Gifu analyses , were excised using a scalpel , boiled in 1x PCR buffer , and then used as a template to reamplify the fragment using the same primer pairs as in the nested PCR of the SSAP reaction . The reamplified fragments were electrophoresed on 1% agarose gels , excised , and extracted from the gel using Wizard SV Gel and PCR Clean-up System ( Promega ) . Cleaned fragments were sequenced using a BigDye Terminator v3 . 1 Cycle Sequencing Kit ( Applied Biosystems ) . The reamplified fragments would be expected to contain the junction sequence between LORE1 and its flanking DNA , in which Mse I sites are absent . Sequences that contained Mse I sites were regarded as artifacts and not subjected to further analyses . To amplify junctions between the flanking DNA and LORE1 , we designed primers specific to the flanking sequences obtained and oriented toward LORE1 . When genome sequences corresponding to flanking DNA were available on the database , they were utilized to design primers . We confirmed that amplifications were successful for plants from which the SSAP fragments were recovered , but not from the parent plant or control Gifu accession . DNA sequences corresponding to regions 1 and 2 in newly transposed LORE1 elements were obtained by direct sequencing of PCR products . These were amplified by primers specific to the 5′ flanking sequences of each LORE1 element and primer 4 ( 5′-CAACAGTAGTATCAAATGTAGG-3′ ) , as indicated in Figure 1A , using a BigDye Terminator v3 . 1 Cycle Sequencing Kit . The primers used for sequencing region 1 were Reg1F ( 5′-AGTAGCACCTGTAACAGTGGAG-3′ ) and Reg1R ( 5′-CATTAAGAGAGACTTTAGGAAC-3′ ) , and those for region 2 were Reg2F ( 5′-CCTCCAACATTGTCAGTGATAG-3′ ) and Reg2R ( 5′-TAGCTGTAAAGCTCCTGTCCAC-3′ ) . In the reciprocal cross analysis shown in Figure 1D , PCR reactions were performed using Primer 1 ( 5′-GACTAAGTGCCTCTTCAACTGC-3′ ) and Primer 2 ( 5′-GACTAAGTGCCTCTTCAACTGC-3′ ) to amplify LORE1a from Gifu , and Primer 1 and Primer 3 ( 5′-CACCTGACGATGCTAGCCTTGG-3′ ) to amplify the region allelic to LORE1a ( absence of LORE1 ) from MG20 ( see Figure 1 legend ) . Genomic DNA samples were extracted from the leaves of T0 plants . Sodium bisulfite treatment of the DNA was conducted using a BisulFast Methylated DNA Detection Kit ( TOYOBO ) , following the manufacturer's instructions . Briefly , 1 µg of column-purified genomic DNA was digested with Eco RI , treated with Proteinase K , and then subjected to bisulfite modification . Bisulfite-treated DNA ( 1 µl ) was used as template for PCR reactions . Primary and nested PCR reactions were conducted for each LORE1 locus . The following primers were used for the primary PCR reactions: BSF R1 ( 5′-CTCTRAAACCTTRTTRCTTCARCCAT-3′ ) in combination with BSFa F ( 5′-TAAAAGAGAATYTGGGTATAAGGGAA-3′ ) for LORE1a; BSFb F ( 5′-TTYAAAGGTGYAGTYTYAATTGTATT-3′ ) for LORE1b; and BSFf F ( 5′-AGGGAGAYGAYAGTGATGGTGTTTT-3′ ) for LORE1f . For nested PCR reactions , 1 µl of the primary PCR reaction was used as template , with the following primers: for LORE1a , BSF R2 ( 5′- CCATRATTCRCTCCTCCRCTTCAC-3′ ) and BSFa F; for LORE1b , BSF R2 and BSFb F; and for LORE1f , BSF R2 and BSFf F . PCR reactions ( 20 µl ) were conducted as follows: incubation at 94°C for 2 min as an initial denaturation step , 30 cycles of 30 s at 94°C , 45 s at 55°C , and 45 s at 72°C for amplification , and incubation at 72°C for 5 min . Amplified fragments were TA cloned using the pGEM-T Easy Vector System ( Promega ) . For LORE1a , 6 to 8 TA clones were obtained from each of three PCR reactions and , in total , between 20 and 24 clones were sequenced for each plant analyzed . For LORE1b , 12 clones obtained from a PCR reaction were analyzed for each plant examined . For LORE1f , 11 or 12 clones obtained from a PCR reaction were analyzed for each plant examined . A method modified from [56] was used for RNA isolation from plant tissues . Ground tissues ( ∼0 . 1 g ) were incubated with 700 µl of extraction buffer ( 2% ß-mercaptoethanol , 2% hexadecyltrimethylammonium bromide , 100 mM Tris-HCl [pH 8 . 0] , and 25 mM EDTA ) at room temperature for less than 5 min . The recovered RNA was treated with 5 U DNase I at 37°C for 30 min in a 100 µl reaction . DNase-treated RNA was purified and recovered using an RNeasy Mini Kit ( QIAGEN ) , with additional DNase treatment performed on a column , following the manufacturer's instructions . For RT-PCR , cDNA was synthesized by ReverTra Ace α ( TOYOBO ) using 1 µg of purified total RNA and oligo ( dT ) 20 primer in a 20 µl reaction . A 5× dilution of the cDNA reaction ( 2 µl ) was used as template for semi-quantitative RT-PCR in a 20 µl PCR reaction using Ex Taq ( TaKaRa ) and 5 pmoles of each primer . The primers LORE1gagF and LORE1gagR were used for detection of LORE1 transcripts and products were amplified with 28 PCR cycles . As a control , the primers EF1αF ( 5′-GTGAGGGACATGAGACAGACTG-3′ ) and EF1αR ( 5′-AAATAGCAGTGTAGGACAAGTC-3′ ) were used for detection of transcripts of elongation factor 1 alpha , and these reactions required 24 PCR amplification cycles . To identify the transcription of LORE1 members , RT-PCR amplifications of regions 1 and 2 were conducted using the primer pairs Reg1 F and Reg1 R , or Reg2 F and Reg2 R , respectively . BLAST searches were used to identify sequences homologous to SSAP fragments . These were conducted using Miyakogusa jp ( http://www . kazusa . or . jp/lotus/ ) , NCBI BLAST ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) , and Phytozome Glycine max ( http://www . phytozome . net/soybean ) . Pfam was accessed at http://pfam . sanger . ac . uk/ . Bisulfite sequencing data was analyzed using QUMA [57] and CyMATE [58] . | In contrast to animals , where germline differentiation initiates early in embryogenesis , germline differentiation in plants starts in the adult phase during reproductive development . Transpositions of transposable elements in both somatic and gametic cells can be transmitted to the next generation . As a result , plant genomes may contain transposable elements exhibiting a variety of tissue-specific activities . Thus far , the spatio-temporal activity of LTR retrotransposons , the most abundant class of transposable elements in plants , has not been well characterized . Here , we report a detailed analysis of the spatio-temporal transposition pattern of a plant LTR retrotransposon in the endogenous system . Using the model legume Lotus japonicus , we found that LORE1a , a member of the chromovirus LORE1 family that belongs to the Gypsy superfamily , was epigenetically de-repressed via tissue culture . Activation was stochastic and derepression was maintained in regenerated plants . This feature made it possible to trace the original spatio-temporal activity of the retrotransposon in the intact plants . We determined that the plant chromovirus retrotransposes mainly in the male germline , without obvious insertional preferences for chromosomal regions . This finding suggests that the tissue specificity of transposable elements should be taken into account when considering their impact on the host genome dynamics and evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"plant",
"biology/plant",
"genomes",
"and",
"evolution",
"genetics",
"and",
"genomics/plant",
"genomes",
"and",
"evolution",
"genetics",
"and",
"genomics/epigenetics",
"plant",
"biology/plant",
"genetics",
"and",
"gene",
"expression",
"genetics",
"and",
"genomics/plant",
... | 2010 | Derepression of the Plant Chromovirus LORE1 Induces Germline Transposition in Regenerated Plants |
The majority of the heritability of coronary artery disease ( CAD ) remains unexplained , despite recent successes of genome-wide association studies ( GWAS ) in identifying novel susceptibility loci . Integrating functional genomic data from a variety of sources with a large-scale meta-analysis of CAD GWAS may facilitate the identification of novel biological processes and genes involved in CAD , as well as clarify the causal relationships of established processes . Towards this end , we integrated 14 GWAS from the CARDIoGRAM Consortium and two additional GWAS from the Ottawa Heart Institute ( 25 , 491 cases and 66 , 819 controls ) with 1 ) genetics of gene expression studies of CAD-relevant tissues in humans , 2 ) metabolic and signaling pathways from public databases , and 3 ) data-driven , tissue-specific gene networks from a multitude of human and mouse experiments . We not only detected CAD-associated gene networks of lipid metabolism , coagulation , immunity , and additional networks with no clear functional annotation , but also revealed key driver genes for each CAD network based on the topology of the gene regulatory networks . In particular , we found a gene network involved in antigen processing to be strongly associated with CAD . The key driver genes of this network included glyoxalase I ( GLO1 ) and peptidylprolyl isomerase I ( PPIL1 ) , which we verified as regulatory by siRNA experiments in human aortic endothelial cells . Our results suggest genetic influences on a diverse set of both known and novel biological processes that contribute to CAD risk . The key driver genes for these networks highlight potential novel targets for further mechanistic studies and therapeutic interventions .
Coronary artery disease ( CAD ) remains a leading cause of death worldwide despite a variety of available interventions to reduce cardiovascular events . CAD is partly familial [1] , [2] , which motivates genetic studies to elucidate novel pharmacological targets . However , large-scale genome-wide association studies ( GWAS ) have revealed a complex genetic architecture of CAD susceptibility with modest effect sizes for the single nucleotide polymorphisms ( SNPs ) detected to date [3] , [4] . The heritability explained by the top SNPs is approximately 10% , whereas the estimates of total heritability from family studies are substantially higher , between 30% and 50% [1] , [2] . Furthermore , the SNP associations themselves rarely provide evidence on their downstream functional consequences , which has prompted the need to integrate DNA variants with functional data to better understand the pathogenic processes . Genes and their downstream products comprise a complex regulatory machinery that sustains the delicate homeostasis of an organism in a changing environment [5] . Genetic variants can perturb parts of this regulatory network and its ability to restore and maintain homeostasis in the presence of environmental pressure . Consequently , the dysregulated biological processes such as cholesterol metabolism and transport can eventually lead to CAD [6] . To elucidate additional as yet unidentified CAD-related processes , regulatory and functional data on the intermediate tissue-specific molecular phenotypes are essential [7]–[9] . Regulatory networks between genes can be captured by various network reconstruction algorithms [10]–[13]; functional information of genetic variants can be derived from expression quantitative trait loci ( eQTLs; contain expression SNPs or eSNPs ) that inform on the downstream target genes of genetic variants [8] , [14]–[16] . Integration of these empirical data allows us to aggregate eSNPs from multiple interacting genes into eSNP sets that collectively perturb a part of the regulatory network . Subsequently , the eSNP sets can be directly compared with SNP-to-disease associations from a GWAS to connect gene networks to disease . In this study , we apply an integrative genomics framework ( illustrated in Figure 1 ) to identify the genetically perturbed regulatory networks that contribute to CAD . We make use of four distinct types of data sources . First , associations between SNPs and CAD were determined in 16 independent GWAS – 14 from the CARDIoGRAM Consortium and two from the Ottawa Heart Institute [4] , [17] . Second , the effects of SNPs on gene regulation were determined according to eQTLs in multiple tissue-specific genetics of gene expression studies of CAD-related tissues or cell types in humans . As a result , we were able to link the CAD SNPs from the GWAS with their empirically defined target genes . Third , we downloaded known metabolic and signaling pathways ( in the form of gene sets ) from public repositories , and complemented these known pathways with data-driven network modules of co-expressed genes from multiple transcriptomic studies , to investigate the collective genetic risk via multiple functionally related genes . Finally , we overlaid the identified CAD-associated gene sets onto causal network models of gene-gene interactions from multiple genomic studies to pinpoint the most central regulatory genes . This combination of human genetics , functional genomics , tissue-specific gene networks from empirical data , and biological knowledge in this integrative genomics framework provides us with further insights into the known and hitherto unknown pathogenic processes that are relevant for CAD .
Our first aim was to test if any of the known biological pathways curated in Reactome , Biocarta and Kyoto Encyclopedia of Genes and Genomes ( KEGG ) databases [18] , [19] was more likely to harbor tissue-specific eSNPs that were also associated with CAD in GWAS ( Figure 1A ) . To reduce false discovery and identify the most robust signals , we implemented a multi-stage design that utilized two independent sets of CAD meta-analysis ( Stage 1 and Stage 2 ) each involving distinct sets of CAD cohorts , as well as the overall meta-analysis of all 16 CAD cohorts ( combined Stage 1+2 set; see Methods for details ) . The QQ plots of these three sets of meta-analysis are shown in Figure S1 . As majority of the cohorts were from the CARDIoGRAM consortium , the GWAS results from our new meta-analysis closely resembled those of CARDIoGRAM reported previously [4] and there were no new loci identified at genome-wide significance level ( Table S1 ) . Tissue-specific eSNPs from CAD related human cells or tissues including adipose tissue , liver , human aortic endothelial cells ( HAECs ) , blood , as well as a pooled eSNP set from multiple tissues and cell types ( denoted as ‘All eSNPs’ ) , were used for eSNP-to-gene mapping , yielding five sets of eSNPs mapped to each pathway ( Materials and Methods ) . Each eSNP set representing a pathway was then compared with random eSNP sets drawn from the background eSNPs of matching tissue to look for enrichment of low p value associations with CAD in GWAS using SSEA . Enrichment was measured by a score calculated as the mean -log P-value from Kolmogorov-Smirnov test and Fisher's exact test ( see Methods ) in SSEA . We defined a pathway to be significantly associated with CAD when permutation-based false discovery rates ( FDRs ) from Stage 1 , Stage 2 , and combined Stage 1+2 analyses simultaneously reached 20% , 20% , and 5% . Considering that Stage 1 and Stage 2 GWAS datasets were independent , the combined FDR from these two sets of analysis was expected to be <5% ( 20% * 20% = 4% ) . An additional requirement for FDR<5% from the combined Stage 1+2 analysis further ensured low FDR . To test whether our method can pick up true positive signals , we selected two predefined CAD gene sets based on the GWAS Catalog [20] and CADgene database [21] ( details in Methods ) as positive controls . These positive controls exhibited strong and consistent signals across multiple sets of SSEA using eSNPs from individual tissues ( 7<score<29; equivalent to 1e-7<p<1e-29 ) , supporting the sensitivity of our approach . A total of 79 out of 833 canonical pathways tested were associated with CAD in at least one of the five sets of SSEA using different eSNP sets ( Table 1; full results in Table S2 ) . Among these , the lipid and lipoprotein pathway from Reactome was significant in adipose eSNP analyses ( score 9 . 8 ) . On the other hand , the bile acid recycling pathway was strongly indicated by the liver eSNPs ( score 8 . 5 ) . The next large group of CAD-associated pathways was related to the immune system: ‘Immunoregulation with lymphoid and non-lymphoid cells’ was the top Reactome pathway ( score 11 . 4 for adipose eSNPs ) , ‘Antigen processing and presentation’ from KEGG was significant in the liver ( score 10 . 8 ) , and ‘Adhesion and diapedesis of granulocytes’ from Biocarta was a significant pathway when using the HAEC eSNPs ( score 3 . 7 ) . A number of pathways that were directly related to the vascular system or heart , and four pathways that were related to blood coagulation were also associated with CAD . For instance , the SSEA implicated the vascular endothelial growth factor ( VEGF ) , hypoxia and angiogenesis pathway , and the platelet/endothelial cell adhesion molecule 1 ( PECAM1 ) pathway . In essence , the analysis of the curated pathways was able to detect genetic links between CAD and its classical risk factor dyslipidemia , and other suspected CAD processes such as inflammation and vascular dysfunction . More importantly , the use of tissue-specific eSNPs sets helped implicate the most relevant tissues for the significant pathways . Our analyses also revealed pathways that so far have not been clearly implicated in CAD development . These included four pathways related to the nervous system ( such as ‘Erythropoietin mediated neuroprotection through NF-kB’ and ‘TrkA receptor signaling pathway’ from BioCarta ) , 13 cell cycle and proliferation pathways ( such as ‘NRAGE signals death through JNK’ from Reactome and ‘EGF signaling’ from BioCarta ) , and ten DNA or RNA pathways ( such as ‘Spliceosome’ from KEGG and ‘E2F regulation of DNA replication’ from Reactome ) . Furthermore , we observed pathways such as ‘Phase I functionalization’ from Reactome and ‘Proteasome’ from KEGG that have a role in the disposal and neutralization of harmful molecules . Pathways that covered amino acids and peroxisome proliferator-activated receptors were also among the significant signals . We compared the top knowledge-driven pathways detected with our eSNP-based SSEA to those detected by several widely used location-based gene-to-SNP mapping methods for gene set enrichment analysis including iGSEA4GWAS [22] , MAGENTA [23] and GSA-SNP [24] , and observed considerable variation in the results between methods , with relatively greater consistency between our SSEA approach and iGSEA4GWAS ( data not shown ) . The results from iGSEA4GWAS are reported in a separate manuscript by Ghosh et al ( under review ) . We found that signals from SSEA such as lipid metabolism , immune and inflammatory pathways , PDGF signaling , NOTCH signaling , and PPAR signaling could be replicated in one or more of the other methods tested . Nonetheless , SSEA and iGSEA4GWAS each yielded additional biologically plausible pathways . Our current analysis included majority of the large-scale CAD-related eQTL sets published before mid 2013 . During the revision of this manuscript , several additional blood eQTLs became available [25]–[27] . We tested the pathways identified in our study using the updated blood eQTLs and found minimal impact on our main results , with Pearson correlation coefficient of the pathway scores being 0 . 96 ( comparison between scores before and after incorporating the new blood eQTLs is shown in Table S3 ) . To uncover hitherto unknown biological processes , we augmented the set of canonical pathways with data-driven co-expression modules ( empirical sets of tightly co-regulated genes ) from multiple previous human and mouse studies ( detailed in Materials and Methods ) . A total of 341 of the 2 , 706 modules tested satisfied the FDR<20% in both Stage 1 and Stage 2 , and FDR<5% in the combined meta-analysis ( Table S4 ) . Given that canonical pathways defined by different databases may overlap and also co-expression modules may overlap with known biological pathways , we collected all CAD-associated gene sets regardless of their source ( 79 canonical pathways + 341 co-expression modules = 420 in total ) , and analyzed their overlap structure ( Figure 1B; overlap matrix in Figure S2 ) . After merging CAD gene sets with overlap of >20% in their member genes ( details in Materials and Methods ) , 62 non-overlapping merged supersets remained . To ensure the merged supersets still captured the features of the significant pathways , we performed a second round of SSEA on the supersets and applied a stringent Bonferroni-corrected statistical cutoff ( correcting for the total number of pre-merged gene sets , n = 3539 , not 62 supersets ) to focus on the most reliable signals . Therefore , although the second round of SSEA was mainly confirmatory to ensure that we did not lose the signals during merging/trimming , a highly stringent Bonferroni-correction that considered multiple testing of 3539 original gene sets ( not 62 supersets after merging ) further ensures that the signals passed the threshold were truly robust . Note that this level of Bonferroni correction is highly conservative because we are treating all 3539 gene sets as independent . In reality , the highly overlapping structures among these gene sets make the number of truly independent gene sets much smaller . Out of the 62 non-overlapping supersets , 22 were confirmed to be genetically associated with CAD in the second round of SSEA ( Table S5 ) and the top six supersets are summarized in Table 2 . The data-driven supersets implicated lipid metabolism ( ‘Lipid I’ and ‘Lipid II’ ) , the immune system ( ‘Immunity’ and ‘Antigen’ ) and coagulation processes ( ‘Lipid II’ ) , consistent with the findings from the canonical pathways . Eight supersets did not significantly overlap with any known pathway or process and could not therefore be annotated by functional categories ( hence named “Unknown”; Table S5 ) . In order to determine the regulatory genes ( referred to as key drivers ) at the center of the CAD-associated supersets as a means to further explore regulatory mechanisms and prioritize disease genes , we performed KDA using tissue-specific Bayesian network models constructed from transcriptomic and genetic datasets from multiple human and mouse studies ( Figure 1C; details in Materials and Methods ) [12] , [28] , [29] . The topology of these Bayesian networks captures detailed gene-gene regulatory relationships and can help infer key network drivers . The KDA results for the top six supersets are summarized in Table 3 and full list of key drivers are in Table S6 . To test if the key driver genes were also responsible for the enrichment of CAD genetic signals in each of the CAD superset , we also ranked and selected the member genes whose eSNPs ( i . e . , SNPs that are associated with the expression levels of the member genes ) showed the strongest CAD association within the superset , termed “GWAS signal genes” , for comparison ( Table 3 ) . Interestingly , the key driver genes were mostly different from the GWAS signal genes , which supports a previously observed phenomenon [30] that important regulatory genes may not harbor common susceptibility polymorphisms by natural selection and , conversely , that a majority of common disease susceptibility loci ( as captured by GWAS ) do not involve key regulatory genes but are situated in the periphery of biological networks . Both the ‘Lipid I’ and ‘Lipid II’ supersets fell under the general category of lipid and fatty acid metabolism ( Figure 2A and 2B ) . They share 14% of their members , including seven apolipoproteins , but were considered non-overlapping according to our a priori overlap threshold of 20% . In fact , the key drivers for ‘Lipid I’ comprise genes important for fatty acid metabolism ( DCI , ETFDH and EHHADH ) and cholesterol biosynthesis ( SQLE ) , whereas ‘Lipid II’ was regulated by coagulation ( PLG and HRG ) and carrier proteins ( GC and PZP ) , which confirms non-overlapping functionality between the two supersets . Of note , two critical genes involved in lipoprotein metabolism , LPL and LDLR , were among the top GWAS signals genes for ‘Lipid I’ . Two supersets of immune system genes – ‘Antigen’ and ‘Immunity’ - were significantly enriched for CAD loci in adipose tissue and the liver ( Figure 2D and 2E ) . The ‘Antigen’ superset owes its annotation to the human leukocyte antigen ( HLA ) and mouse HLA orthologs ( H2 genes ) that comprise 21 of the 221 member genes . We found the key drivers for this superset such as GLO1 , PPIL1 and DECR2 to be highly consistent across Bayesian networks from multiple tissues ( Table S7 ) . The ‘Immunity’ superset contains a variety of immune response genes including six HLA genes and 18 cytokines or their receptors such as CCL2 , CD antigen genes , CXCL10 , IL2RB and TLR2 . Four of the top five key drivers for the ‘Immunity’ superset ( PTPRC , FYB , FCGR1A and FCER1G ) participate in the immune response , and three ( PTPRC , FYB and FCER1G ) have been previously identified as key drivers of an inflammatory gene signature underlying multiple diseases ( including CAD ) [12] . We could not annotate two out of our six top supersets ( ‘Unknown I’ and ‘Unknown II’ ) . These supersets , however , have consistent network key drivers across multiple tissues ( Figure 2C and 2F , Table S7 ) . ‘Unknown I’ contained a diverse set of key driver genes such as SGK1 , SIK1 and SLC10A6 ( sodium metabolism and hypertension ) , MT2A and TSC22D3 ( glucocorticoid signaling ) , GADD45G , ERRFI1 , GPRC5A , and EGFR ( cell growth and apoptosis ) , and CEBPB , CEBPD , and KCNA5 ( heart development and function ) . Possible functions of ‘Unknown II’ include RNA metabolism ( ZC3H7B ) , protein methylation ( PRMT1 ) , glycosylation ( ALG8 ) , chaperone recycling ( DNAJC7 ) and ubiquitination ( UBE2S ) , and similar annotations could also be found for the network neighboring genes of these key drivers . Of note , the gene CYP39A1 which converts cholesterol into bile acid was shared between ‘Lipid I’ , ‘Antigen’ and ‘Unknown II’ supersets . The ‘Antigen’ superset had the highest combined CAD association score across the five sets of SSEA using eSNP sets from different tissues ( Table 2 ) and their key driver genes identified were highly consistent across the Bayesian networks used for KDA ( Table S7 ) . For these reasons , we tested the effects of silencing three of the key drivers , glyoxalase I ( GLO1 ) , peptidylprolyl isomerase I ( PPIL1 ) and peroxisomal 2 , 4-dienoyl CoA reductase 2 ( DECR2 ) , on the expression of member genes in the ‘Antigen’ superset in HAECs , aiming to validate the role of these key drivers in regulating this CAD superset . HAECs were chosen based on their critical role in maintaining a healthy vessel wall and knowledge that endothelial dysfunction is observed early in the development of atherosclerosis [31] . Of note , the number of HAEC specific eSNPs was relatively low due to the limited sample size in the original study [32] , which could explain the lack of significant pathway enrichment signals in this cell type in this study . However , this statistical power issue should not be misconstrued as a lack of relevance of this cell type in the pathogenesis of CAD . Three separate siRNAs against the GLO1 transcript NM_006708 ( Qiagen Catalog Numbers SI04175892 , SI04206244 , SI04266052 ) resulted in 86% , 88% and 91% reduction in GLO1 expression , and siRNA SI04284224 against the PPIL1 transcript NM_016059 resulted in 87% reduction . DECR2 expression level was too low in HAECs to be informative . Whole genome transcript expression in response to GLO1 and PPIL1 suppression was measured using microarrays and then compared with that from scrambled siRNAs ( null control ) to detect genes with significant changes . A total of 485 and 656 genes were affected by GLO1 and PPIL1 suppression , respectively ( P<0 . 001 for both ) . Due to substantial overlap ( 281 genes were affected by both GLO1 and PPIL1 ) , we pooled the 860 unique genes that were significantly affected by either GLO1 or PPIL1 . We determined how many of these top genes were neighbors by two edges to any of the five key drivers ( GLO1 , PPIL1 , DECR2 , VPS52 and GFER ) of the ‘Antigen’ superset when considering all available Bayesian networks . We found the 547 neighbor genes to be enriched for the differentially expressed genes by 1 . 7 fold ( 37 observed vs . 21 . 5 expected , P = 5 . 5×10−4 by Fisher's exact test ) , indicating that the suppression of GLO1 and PPIL1 perturbed the ‘Antigen’ superset in HAEC . To verify the connection of GLO1 and PPIL1 perturbations to CAD , we tested the two pre-defined CAD positive control gene sets ( Table 1 ) against the differential P-values from the siRNA experiments . The CADgene positive control set was significantly associated with altered expression due to either GLO1 ( PFisher<0 . 0001 , PK-S = 0 . 0042 ) or PPIL1 suppression ( PFisher<0 . 0001 , PK-S = 0 . 0085 ) . These results indicate that the expression levels of genes from CAD-related processes are significantly more affected by GLO1 and PPIL1 knockdown compared to a random set of genes in HAEC .
We performed an integrative genomics study that combined association signals from a large GWAS , tissue-specific eQTL datasets , known canonical pathways , and data-driven regulatory networks to gain insights into the causal molecular mechanisms of CAD . Our approach identified both established and novel biological processes supported by functional evidence; specifically , the expression levels of the member genes within these processes were controlled by multiple CAD-associated SNPs . To dissect the key regulatory mechanisms , we derived a network representation of the central genes involved in the pathogenic processes and investigated how the affected genes were related to known biological pathways and metabolic cascades and how the processes were inter-connected in multi-tissue regulatory networks . Our study revealed a highly complex and multifactorial genetic basis for CAD , and implicated several known and novel causal pathways along with their potential regulators deserving of further study . Several aspects of this study distinguish it from previous pathway and network studies of CAD . First , we used data from eQTL studies of CAD-related tissues or primary cell types to assign eSNPs to genes , whereas previous approaches have primarily utilized genomic location for assignment of SNPs to genes [3] , [22] , [23] , [33] , [34] . Our method incorporates empirical functional support and tissue specificity into the analyses to increase the sensitivity of detecting tissue-specific molecular events that would have been missed by conventional methods [22] , [23] , [33] and to enhance the biological and mechanistic interpretability of the disease-related signals [9] , [35]–[37] . Our partial re-analysis of data incorporating recently reported blood eSNPs suggests that the addition of new eQTLs reinforces the significance of the pathways identified thus far . Second , we tested both knowledge-driven and data-driven gene sets to expand the coverage of novel biological processes . Third , we used two large independent CAD GWAS meta-analyses , merged and trimmed overlapping CAD-associated gene sets , and imposed a strict Bonferroni threshold for final statistical evaluation of the CAD signals to avoid false positives and focus on the most reliable core processes for CAD . Fourth , we utilized scores of empirically-derived gene networks from diverse CAD-related tissues to extract the CAD network architecture and the key driver genes , whereas previous studies have relied on literature-based topologies , protein-protein interaction networks , or single-tissue networks [3] , [38]–[40] . Lastly , we performed targeted siRNA studies in HAEC to provide experimental support for our in silico findings . Of note , the KDA approach we utilized has been recently demonstrated by multiple studies to have the capacity to identify hidden novel regulatory genes that are missed by traditional analysis , and novel predictions from each of these studies have been experimentally validated [12] , [28] , [41] . The key drivers identified , however , are not necessarily GWAS hits . In contrast , it is their downstream or peripheral genes that are more likely to be identified in GWAS and the expression of these genes are more likely to be cis-regulated by GWAS SNPs . This may explain why a majority of the GWAS hits uncovered to date only have small effects on complex disease phenotype . As elucidated previously by Goh et al . [30] , the lack of GWAS signals from key driver genes can be explained by evolutionary constraints imposed on important regulators because strong genetic perturbations in these key regulators are more likely to be deleterious . If certain genetic polymorphisms within key regulatory genes ( e . g . , transcription factors ) indeed successfully segregate in general population and can be identified in GWAS , these polymorphisms tend to be cis-eSNPs of the regulators themselves and then trans-eSNPs of additional disease genes [25] . Our results support the role of genetic perturbation to lipid metabolism , immune response and inflammation , coagulation , and vascular wall function in the etiology of CAD . Apart from cholesterol metabolism and transport , a causal role for many of these processes in pathogenesis of CAD has been debated for years . For instance , recent Mendelian randomization studies as well as randomized control trials of cardiovascular drugs have demonstrated that a number of known key genes within these pathways ( e . g . CRP , fibrinogens ) are not causally associated with CAD [42]–[44] . Our results suggest that a critical mass of causal variants may be inherited within many of the genes in these pathways even if the pathway includes some genes that have no causal role in the pathogenesis of CAD . We identified novel biological processes such as neuroprotection , cell cycle and proteolysis that were perturbed by the CAD-associated genetic variants . Furthermore , the data-driven network models identified CAD-associated gene sets that did not overlap with any known biological processes . We merged the knowledge-based biological pathways and data-driven functional units of genes derived from expression patterns to bridge the knowledge gaps , and focused on six gene sets that were not only strongly associated with CAD in human GWAS but also exhibited a consistent causal network topology around a limited number of key regulatory genes . The supersets ‘Lipid I’ and ‘Lipid II’ are involved in cholesterol and lipid biosynthesis or degradation . The ‘Lipid II’ superset appears to have more diversified functions beyond lipid biosynthesis and transport , as it contains multiple additional genes from coagulation and complement pathways . If ‘Lipid I’ and ‘Lipid II’ were simultaneously perturbed , one can speculate that the lipid transport system would become overwhelmed , wound healing processes and the complement cascade would become over-activated , and the lipid-rich debris would feed the accumulation of plaque on the vessel wall . Furthermore , two of the central genes in Lipid II , plasminogen ( PLG ) and the histidine-rich glycoprotein ( HRG ) regulate the fine balance between clotting and fibrinolysis , which can affect the propensity of thrombosis after plaque rupture . The strongest overall CAD association was observed for the ‘Antigen’ superset . The key drivers were highly consistent across tissues but , surprisingly , none of them have been directly implicated in antigen processing . In fact , many of the network driver genes appear to be involved in protein processing , and endosomal and lysosomal functions . For instance , glyoxalase 1 ( GLO1 ) plays a critical part in the enzymatic defense against dysfunctional glycated forms of proteins [45] , the vacuolar protein sorting 52 homolog ( VPS52 ) is involved in the transport and sorting of proteins from the plasma membrane to the lysosome via mannose-6-phosphate receptors [46] , the N-acetylglucosamine-1-phosphate transferase gamma subunit ( GNPTG , top 10 key driver , between GLO1 and GFER in Figure 2 ) is part of mannose-6-phosphate synthesis [47] , and peptidylprolyl isomerase ( PPIL1 ) is a member in the cyclophilin family that regulates protein folding and immune responses [48] . Of note , there may also be a direct link to lipid metabolism: the peroxisomal 2 , 4-dienoyl CoA reductase ( DECR2 ) , which participates in the beta-oxidation of unsaturated fatty acids [49] , is a key driver of this ‘Antigen’ superset and a member of ‘Lipid I’ . GLO1 is an interesting candidate for a causal CAD gene . Diabetes , kidney disease , and diabetic kidney disease in particular increase the risk and severity of CAD dramatically [50] , [51] , and the glyoxalase system is an important protective mechanism against the formation and subsequent accumulation of advanced glycation end products that are believed to promote diabetic end-organ damage . In a mouse study , a Glo1 knock-down model spontaneously developed kidney disease even without diabetes [52] , and a single case of human GLO1 deficiency exhibited both end-stage renal disease and severe atherosclerosis [53] . A recent study demonstrated a protective role of Glo1 in restoring neovascularization of ischemic tissue in diabetic rats [54] . In our study , knock-down of GLO1 in HAECs perturbed the expression of many of the same genes that were affected by CAD-associated SNPs in the human GWAS . Therefore , the glyoxalase system may represent one common pathway responsible for both the microvascular and macrovascular complications observed in subjects with diabetes . The ‘Immunity’ superset may represent the same core inflammatory signature that we have observed across species , tissues , and multiple diseases [12] . Four of the top key drivers ( HCK , TYROBP , NCKAP1L and AIF1 ) identified from the previous multi-disease inflammatory signature were also detected as key drivers for the ‘Immunity’ superset in this study ( Figure 2D ) . Although we cannot provide definitive answers on the sequence of events , one may hypothesize that the ‘Immunity’ superset executes the downstream machinery that is recruited in response to the antigen presentation in cells under metabolic stress . The central genes in the ‘Unknown I’ superset include the serum glucocorticoid regulated kinase 1 ( SGK1 ) , a mitogen-activated protein kinase ( MAP3K6 ) , a sodium/bile acid co-transporter ( SLC10A6 ) and a C/EBP transcription factor ( CEBPD ) . Of these , SGK1 has been studied the most and is believed to be important for renal sodium absorption , salt-sensitivity to hypertension and glycemia , cardiac repolarization , and numerous other processes [55] . MAP3K6 regulates VEGF expression [56] , and its expression was altered in a mouse model of cardiomyopathy [57] . The SLC10 gene family contains three sodium-dependent transporters , of which SLC10A6 transports sulfoconjugated steroid hormones . One of the shared genes between ‘Antigen’ and ‘Unknown I’ in Figure 2 , salt-inducible kinase 1 ( SIK1 ) , is thought to be part of the sodium-sensing network [58] . Cyclin-dependent kinase inhibitors ( CDKN ) may be causally related to atherosclerosis [3] , [59] , although the exact role of CDNK1A ( adjacent to MAP3K6 and SLC10A6 in Figure 2 ) remains unclear . Thus , the superset ‘Unknown I’ appears bring together processes quite relevant to CAD including glucocorticoid signaling , vascular stress response , cell growth , and blood pressure control . ‘Unknown II’ could not be annotated by known processes and the functions of the core genes remain poorly understood . There were a number of genes that may regulate methylation , histones , chromatin , and splicing ( ING3 , CBX6 , LMNB1 , SFRS5 ) , post-translational protein modifications and activation ( PRMT1 , ALG8 , PDIA3 , DNAJC7 ) , ubiquitination ( UBE2S , RNF25 , RNF146 ) , cytoskeleton organization ( ARPC3 , MAP1LC3A , DYNLL2 , ARL6IP5 ) , and cell cycle ( CD82 , ING3 , UBE2S , MDFI , TRAF4 ) . It is possible that the genes within ‘Unknown II’ participate in the stress-induced epigenetic and proteomic changes that contribute to atherogenic processes . If this is true , it may explain why the curated pathways , many of which consist of chemical reactions between metabolites , missed the predominantly regulatory gene networks such as ‘Unknown I’ and ‘Unknown II’ . We had a wealth of data sources at our disposal in this study to derive a comprehensive view of the complex mechanisms of CAD . Nevertheless , we acknowledge the following limitations . First , our study cannot distinguish pathways or gene subnetworks that are more relevant to specific subtypes of CAD from those which cause CAD through more general mechanisms involving relatively well-understood cardiometabolic processes . Future studies involving sample sets that include more refined subtyping of cases may help further advance our understanding in this respect . Second , although the concept of eQTL as an empirical alternative to the traditional location-based gene-SNP mapping in pathway analysis is appealing from a biological perspective as it carries functional implications and allows detection of tissue-specific signals , in practice , however , the lack of comprehensive and large enough genetics of gene expression studies may limit the power and the biological coverage of the approach , as the total number of eSNPs is typically lower and eSNPs from additional CAD-related cell types of tissues are not necessarily available . On the other hand , emerging resources such as ENCODE [60] , [61] and GTEx [62] are likely to improve the situation in the future . In conclusion , we used an integrative genomics framework to shed light on the key genes and regulatory processes involved in the pathogenesis of CAD . We detected genetically driven perturbations of several pathways with a strong a priori evidence of involvement in CAD ( cholesterol synthesis , inflammation , and blood coagulation ) , as well as novel processes ( neuroprotection , epigenetic and post-translational modifications , intracellular transport , proteolysis , and cell cycle ) . The data suggest that many genes in these biological processes are causally associated with CAD even if this may not be the case for all the pathway or network members . We verified the importance of the key drivers in the top-scoring gene set using an experimental gene expression model . Thus , the CAD associated gene networks and key drivers identified in this study warrant further validation in additional population genetic and mechanistic studies . Further knowledge gained through such studies has the potential to lead to major advances in the development of therapeutic strategies to reduce the risk of CAD .
The overall integrative framework is depicted in Figure 1 . First , we applied a modified SNP set enrichment analysis ( SSEA ) [36] , [37] to find sets of functionally related genes that were associated with CAD ( Figure 1A ) . In this analysis , we used knowledge-based canonical pathways and data-driven co-expression network modules as the functional units of genes that were tested for CAD association , tissue-specific eQTL studies to connect the genes to SNPs , and CAD GWAS to provide the associations between SNPs and CAD . To reduce false discovery and identify the most robust signals , we implemented a multi-stage design that leveraged two independent GWAS meta-analyses of CAD . Next , we investigated the statistically significant CAD-associated pathways and co-expression modules for shared genes , and merged any overlapping gene sets into non-overlapping supersets ( Figure 1B ) . Lastly , the key regulator genes for each superset were determined by integrating multiple tissue-specific Bayesian causal network models of gene interactions with the CAD-associated gene supersets ( Figure 1C ) . The design , clinical classification and genotyping within the Coronary ARtery DIsease Genome-wide Replication And Meta-Analysis ( CARDIoGRAM ) Consortium have been described previously [4] , [63] . The dataset used in this study comprised 25 , 491 cases with coronary artery disease , myocardial infarction or both and 66 , 819 controls from the 14 cohorts within CARDIoGRAM and two GWAS by the Ottawa Heart Institute in collaboration with Cleveland Clinic and Duke University [17] . The 16 GWAS were split into two independent sets ( Table S8 ) : The Stage 1 set combined the results from the Ottawa Heart Genomics Study with the Cleveland Clinic Gene Bank ( OHGS_A and OHGS_CCGB_B ) , the CAD component of the Wellcome Trust Case Control Consortium ( WTCCC ) , the Duke CATHGEN Study , and the German Myocardial Infarction Family Studies I , II , III with Collaborative Health Research in the Region of Augsburg ( GerMIFS1 , GerMIFS2 , GerMIFS3/KORA ) . The remaining CARDIoGRAM cohorts formed the Stage 2 set and included Atherosclerotic Disease VAscular functioN and genetic Epidemiology Study ( ADVANCE ) , CADomics , Cohorts for Heart and Aging Research in Genomic Epidemiology ( CHARGE ) , deCODE CAD , Ludwigshafen Risk and Cardiovascular Health Study ( LURIC/AtheroRemo 1 , LURIC/AtheroRemo 2 ) , MedStar , Myocardial Infarction Genetics Consortium ( MIGen ) and the PennCath Study . Genotyping of the individuals was performed by Affymetrix or Illumina platforms and imputed to 2 . 5 million SNPs prior to meta-analyses [4] . Ancestry was restricted to European origin by self-reporting or principal component analysis of genotypes or both . The Ottawa cohorts were imputed separately using IMPUTE2 and MACH software , and a reference panel that included 112 European genomes from the 1000 Genomes Project ( August 2009 ) and 298 additional subjects from a separate CEU/TSI reference panel [17] . Correcting of population stratification was performed as described previously [17] . The smartPCA ( principal components analysis ) program from EIGENSOFT v3 . 0 [64] was used to identify and remove subjects of admixed or non-European ancestry . Study subjects were processed with 270 HapMap2 subjects for PCA ( 90 CEU , 90 JPT+CHB , 90 YRI ) . In the resulting first 2 dimensions from PCA , k-means was used to ascertain the center of each of the CEU , JPT+CHB , and YRI clusters , and the original 2 PC dimensions were projected onto these axes . Subjects were removed if they fell outside an oval whose major axes were 10 times the standard deviation of the CEU cluster along the 2 transformed axes . The SNP-level associations were estimated by a meta-analysis approach similar to that used for CARDIoGRAM [4] . SNPs with minor allele frequency below 1% , significant Hardy-Weinberg equilibrium ( P<0 . 0001 ) , imputation quality below 50% or call rate below 75% were excluded . Rare SNPs that were present in less than three Stage 1 GWAS or less than five Stage 2 GWAS were also excluded . The GWAS were analyzed jointly by a fixed-effect inverse-variance weighted model within the Stage 1 and Stage 2 sets , respectively . Heterogeneous SNPs with significant Q and I statistics ( P<0 . 0001 ) were analyzed by DerSimonian and Laird inverse variance model of random effects . We also used all available cohorts to create an overall meta-cohort ( denoted as Stage 1+2 ) . We included curated pathways from the Reactome , Biocarta and Kyoto Encyclopedia of Genes and Genomes ( KEGG ) databases [18] , [19] . The Reactome database is based on reactions between diverse molecular species rather than limiting the pathways to protein-protein interactions or other types of non-biological categories . Also , the nested structure of the Reactome database helps to increase the coverage to multiple levels of gene set organization . The KEGG database represents carefully curated and experimentally validated pathways of metabolic processes and gene sets of human diseases , while BioCarta is a community based effort to describe interactions that arise from proteomic and other similar studies . In total , 833 gene sets were included in the analyses , collectively referred to as knowledge-driven pathways . We constructed two positive control gene sets using previously known CAD candidate genes . The first positive control gene set was based on the GWAS Catalog [20] . SNPs with P<5 . 0×10−8 for the traits ‘Coronary heart disease’ , ‘Coronary artery calcification’ and ‘Myocardial infarction’ were collected from the catalog , and the reported genes for these loci were included in the control set . Another positive control was formed from the CAD candidate genes curated in the CADgene database [21] . Consistent expression patterns among specific sets of genes were investigated in previous studies to define co-regulated gene sets , commonly referred to as co-expression network modules . These modules can be considered data-driven “pathways” of gene regulation that typically operate upstream of the classical pathways of chemical signaling and enzymatic action . We utilized co-expression modules constructed using the weighted Gene Co-expression Network Analysis [13] from ∼10 human and mouse studies that involved multiple CAD-related tissues ( details and references in Table S9 ) . Human modules were obtained from HAEC , adipose tissue , blood , and liver . Mouse modules were obtained from adipose tissue , liver , muscle , brain , heart , islet cells and kidney . A total of 2706 co-expression modules representing data-defined functional units of genes were used in this study . Human eQTL studies are analogous to GWAS of quantitative traits , except that the traits are tissue-specific gene expression levels rather than biomarkers or clinical measures . eQTL studies constitute an important source of empirically supported mappings from a genetic variant ( eSNP ) to its gene target and these mappings can be reversed to convert a gene set back into the respective eSNP set for direct testing of disease associations in GWAS . The human eSNP data used in this study were collected previously from adipose tissue , liver , HAEC , blood , fibroblasts , lymphoblasts , and monocytes ( details and references in Table S9 ) . Both cis-eSNPs ( within 1 Mb distance from gene region ) and trans-eSNPs ( beyond 1 Mb from gene region ) at false discover rate <10% were included . We chose adipose tissue , liver , blood and HAEC for tissue-specific analyses due to their direct relevance to CAD and relative abundance of eSNPs , while all eSNP resources were pooled into a tissue-independent set denoted as ‘All eSNPs’ , yielding five sets of gene-eSNP mapping for each gene set . We observed a high degree of linkage disequilibrium ( LD ) between eSNPs . If left uncorrected , redundant eSNPs can inflate the disease association scores of a pathway or gene network and increase the number of false positives . For this reason , we devised an algorithm to remove eSNPs in LD while preferentially keeping those with a strong statistical association with gene expression ( Text S1 ) . The preferential treatment based on these expression association P-values was motivated by previous observations that disease loci are enriched among eSNPs [36] , [65] . We used eSNPs from multiple human cohorts with different sample sizes and study designs . Simply comparing the raw expression association P-values in the LD pruning algorithm could potentially skew the selection according to the study characteristics rather than biological relevance . Therefore , we ranked the expression association P-values and scaled the ranks to the interval between 0 and 1 within the study-specific eSNP dataset before pooling the studies and applying the LD pruning algorithm . The reduction in the number of accepted eSNPs after LD pruning was smooth over a wide range of LD thresholds ( Figure S3 ) . We chose a moderate LD cutoff ( R2<0 . 7 ) that lead to the rejection of approximately 50% of eSNPs . This cutoff was chosen because it preserves statistical power while removing eSNPs in high LD . We used the LD structure of the CEU HapMap population [66] for the eSNP pruning given our CAD GWAS included only subjects of white/European descent . We applied a modified SSEA to identify gene sets associated with CAD [36] , [37] ( Figure 1A ) . We collected gene sets from knowledge-based pathway databases , or defined them according to data-driven co-expression network modules ( Figure 1A , left ) . We also determined the specific sets of eSNPs that perturb the expression of the member genes in each gene set based on the tissue-specific eSNP sets described above ( Figure 1A , middle ) . We retrieved the CAD association P-values for the eSNPs from the CAD GWAS ( Stage 1 , Stage 2 , Stage 1+2 , separately ) , compared the P-values against the random expectation , and summarized the observed difference as a single enrichment score , as detailed below ( Figure 1A , right ) . Importantly , the study involved two levels of P-values . The first level of P-values for each SNP in the GWAS was calculated according to the genotypes of the participating individuals in relation to their CAD phenotype . For our purposes , these trait association P-values represent the statistical strengths of CAD associations , and produce the ranking of eSNPs according to their relevance to CAD . It has been previously observed that eSNPs are enriched for disease associations [36] , [65] . Therefore , simply using eSNPs to determine pathway signals typically leads to false positives . In our study , we first removed SNPs that were not eSNPs , and used the remaining pool of eSNPs as the null background for subsequent enrichment tests . For instance , the background for adipose tissue comprised all the 59 , 979 non-redundant adipose eSNPs . This procedure was adopted for each tissue-specific eSNP set separately . The gene set enrichment P-values represent the second level of P-values which reflect the degree of enrichment of high ranking disease-associated eSNPs within a given gene set as compared to the null distribution of randomly expected uniform distribution of all ranks . The enrichment of CAD association signals for each gene set ( pathway or co-expression module ) was estimated by the Kolmogorov-Smirnov ( K-S ) test and Fisher's exact test . The K-S test takes into account the total deviation of the observed ranks from the expectation and is therefore sensitive to a large number of weak GWAS signals . The Fisher's exact test detects if the top 5% of eSNPs based on their CAD association strength is over-represented among the eSNPs representing a gene set of interest ( sensitive to a few strong signals ) . The final enrichment score was defined as the mean −log10 of K-S and Fisher P-values . False discovery rates ( FDR ) were estimated by randomly permuting the CAD association P-values of the background eSNPs while keeping all other data structures intact . For a single permutation , FDR was estimated as the ratio between the number of permuted gene sets that exceeded a given enrichment score , and the observed number of gene sets that actually exceeded the score threshold . The final FDR was averaged over 1000 permutations for each tissue-specific eSNP set separately . Gene sets that satisfied FDR<20% in both Stage 1 and Stage 2 GWAS sets , and FDR<5% in the combined Stage 1+2 were considered statistically significant . As Stage 1 and Stage 2 were independent , the request for simultaneous satisfaction for these FDR cutoffs ensured the overall FDR to be <5% . SSEA was performed in R . A substantial number of gene sets overlapped based on their shared member genes , given that similar functional units of genes could be captured by the pathway databases and gene expression studies used in the study . To reduce the redundancy of the dataset , we collapsed overlapping CAD-associated gene sets into non-overlapping “supersets” ( Figure 1B ) . When multiple gene sets are merged , the size of the resulting superset can grow very large . For this reason , we included only the core genes ( that were shared with most of the constituent gene sets ) in the final supersets . For two gene sets A and B with different numbers of member genes , two overlap ratios were calculated: the proportion of genes in A that were also in B ( rAB ) , and the proportion of genes in B that were also in A ( rBA ) . We chose the formula r = ( rAB×rBA ) 0 . 5 to describe the degree of overlap . Importantly , r is small whenever the sizes of A and B are substantially different , which discourages the merging of nested gene sets . We also required that Fisher's exact test for the number of shared genes was statistically significant ( P<0 . 05 after Bonferroni correction ) . We employed hierarchical clustering to define blocks of overlapping gene sets . First , the overlap matrix of the CAD-associated gene sets was estimated and all non-significant elements were set to zero . The overlaps were then converted to distances d = ( 1 - r ) . Clusters of overlapping gene sets were identified by the hclust ( ) function in the R programming environment with a static cutoff at zero overlap . In the last step , the gene sets within clusters were merged and trimmed into supersets . The above procedure was repeated two times to reduce the maximum observed overlap below 20% between any two resulting supersets . A size limit of 200 genes was chosen to trim the raw supersets down to the core genes that were shared across overlapping gene sets . This choice of optimal size was motivated by earlier SSEA analyses [37] . The least shared genes were successively removed until the next removal would have reduced the superset size below the 200-gene limit . Overlap ratios were re-calculated between the trimmed supersets before the next round of hierarchical clustering . The functional categorization of each superset was based on the known pathways from the Gene Ontology and KEGG databases . We evaluated the over-representation of a functional category within the member genes of a superset with the Fisher's exact test . Significant functional categories ( P<0 . 05 after Bonferroni correction ) were used to annotate the functionality of each superset . If no significant annotation could be found , we labeled the superset as ‘Unknown’ . A second round of SSEA was performed for the merged supersets to confirm that they captured the features of the pre-merged gene sets . Significance was determined at SSEA P<0 . 05 after Bonferroni-correction for the number of all original gene sets ( n = 3539 , not the number of supersets after merging ) to ensure stringency . The above SSEA analysis is able to identify a gene set that is likely to contain disease-causing genetic variation . To pinpoint the most influential regulatory genes , we utilized Bayesian network models of gene-gene interactions that take into account both the genotypes that affect gene expression ( causal direction known ) , and the statistical relationships between gene expression levels ( causal direction uncertain ) , using the established method by Zhu et al . [29] , [67] . Bayesian network models from human and mouse studies were constructed based on genetics and gene expression data generated from multiple tissues from multiple previously published studies , each involving hundreds of individuals ( details and references in Table S9 ) . Human networks were obtained from adipose tissue , blood , and liver . Mouse networks were obtained from adipose tissue , liver , muscle , brain , and kidney . Bayesian networks are directed acyclic graphs in which the edges of the graph are defined by conditional probabilities that characterize the distribution of states of each gene given the state of its parents [68] . The network topology defines a partitioned joint probability distribution over all genes in a network . The likelihood of a Bayesian network model given observed genomic data is determined using Bayes formula . For each dataset , 1000 Bayesian networks , each using different random seeds , were reconstructed using Monte Carlo Markov Chain simulation [69] . Bayesian Information Criteria was used to determine the model with the best fit for each network . From the resulting set of 1000 networks , edges that appeared in greater than 30% of the networks were used to define a consensus network for a given dataset . To infer causal directions between genes in a network , genetic information was used as priors by allowing genes with cis-eSNPs to be parent nodes of genes without cis-eSNPs and preventing genes without cis-eSNPs to be parents of genes with cis-eSNPs [70] . Bayesian network provides a natural framework for integrating diverse data and reconstruct biological causal networks . We used the key driver analysis ( KDA ) to determine the key regulatory genes of the CAD-associated gene supersets [12] , [28] , [29] . We defined a key driver as a gene that is connected to a large number of genes from a CAD-associated superset , compared to the expected number for a randomly selected gene within a Bayesian causal network . The basic idea of KDA is depicted in Figure 1C . First , one needs a network topology that defines links between pairs of genes . In this study , we used tissue-specific Bayesian networks that were constructed from large-scale genetic and genomic datasets from multiple previously published studies as described above . Second , a disease-related gene set is needed . In our study , this comes from the CAD-associated gene supersets ( call it Gene Set S ) . We then tag each of the member genes in the Gene Set S within the network , as shown by the colored nodes in Figure 1C . For a gene in the network ( call it Gene A ) , we then ask the following question: How many of Gene A's neighbors are members in the Gene Set S ? If the proportion of member genes is higher than what could be expected for a random gene set , we define Gene A as a key driver for Gene Set S . The statistical significance of a key driver for a given CAD superset in a particular Bayesian network is determined by Fisher's exact test which assesses the enrichment of CAD genes in the candidate key driver's network neighborhood . Bonferroni-corrected p<0 . 05 was used to determine key drivers . As multiple networks were available for a number of tissues , we used two criteria to prioritize the five most important key driver genes . Firstly , we counted how many times a gene was a key driver in multiple networks ( denoted as N ) . The consistency across networks was expressed as ( N - 0 . 99 ) to strongly favor key genes that could be identified in at least two networks . The second criterion was based on the statistical significance of the key driver . In particular , the significance value was calculated as mean ( -log P ) , where P denotes the KDA significance P-values from each of the networks . The final ranking of genes was based on the product of the consistency and significance criteria . KDA was performed using R . To test whether perturbing key drivers identified in our study indeed result in perturbations of CAD gene networks , we used siRNAs to knockdown the expression of novel key drivers in HAECs . HAECs were grown to 80% confluency on 0 . 1% gelatin coated culture plates in MCDB-131 complete medium ( VEC technologies ) . Cells were transfected with siRNAs against each candidate key driver gene under investigation and a negative control ( Cat . No . 1027280 ) at a final concentration of 40 nM with Lipofectamine 2000 reagent for 4 hours in Opti-MEM medium ( Invitrogen ) . Three HAEC lines from different donors were used as biological replicates for each siRNA . Media was replaced with MCDB-131 and cells were lysed for RNA isolation after 24 hours . Whole transcriptome expression was assessed with Illumina HumanHT-12 v4 Expression BeadChip . We identified the differential expression of genes between cells transfected with the siRNAs targeting the candidate genes and the control siRNA using the limma package in R ( 2 . 14 . 0 ) . Overlap between the differentially expressed genes in siRNA experiments and a CAD network of interest was assessed using Fisher's exact test . | Sudden death due to heart attack ranks among the top causes of death in the world , and family studies have shown that genetics has a substantial effect on heart disease risk . Recent studies suggest that multiple genetic factors each with modest effects are necessary for the development of CAD , but the genes and molecular processes involved remain poorly understood . We conducted an integrative genomics study where we used the information of gene-gene interactions to capture groups of genes that are most likely to increase heart disease risk . We not only confirmed the importance of several known CAD risk processes such as the metabolism and transport of cholesterol , immune response , and blood coagulation , but also revealed many novel processes such as neuroprotection , cell cycle , and proteolysis that were not previously implicated in CAD . In particular , we highlight several genes such as GLO1 with key regulatory roles within these processes not detected by the first wave of genetic analyses . These results highlight the value of integrating population genetic data with diverse resources that functionally annotate the human genome . Such integration facilitates the identification of novel molecular processes involved in the pathogenesis of CAD as well as potential novel targets for the development of efficacious therapeutic interventions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"systems",
"biology",
"genomics",
"gene",
"regulatory",
"networks",
"genome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"genetics",
"of",
"disease",
"human",
"genetics"
] | 2014 | Integrative Genomics Reveals Novel Molecular Pathways and Gene Networks for Coronary Artery Disease |
TRIM proteins play important roles in the innate immune defense against retroviral infection , including human immunodeficiency virus type-1 ( HIV-1 ) . Rhesus macaque TRIM5α ( TRIM5αrh ) targets the HIV-1 capsid and blocks infection at an early post-entry stage , prior to reverse transcription . Studies have shown that binding of TRIM5α to the assembled capsid is essential for restriction and requires the coiled-coil and B30 . 2/SPRY domains , but the molecular mechanism of restriction is not fully understood . In this study , we investigated , by cryoEM combined with mutagenesis and chemical cross-linking , the direct interactions between HIV-1 capsid protein ( CA ) assemblies and purified TRIM5αrh containing coiled-coil and SPRY domains ( CC-SPRYrh ) . Concentration-dependent binding of CC-SPRYrh to CA assemblies was observed , while under equivalent conditions the human protein did not bind . Importantly , CC-SPRYrh , but not its human counterpart , disrupted CA tubes in a non-random fashion , releasing fragments of protofilaments consisting of CA hexamers without dissociation into monomers . Furthermore , such structural destruction was prevented by inter-hexamer crosslinking using P207C/T216C mutant CA with disulfide bonds at the CTD-CTD trimer interface of capsid assemblies , but not by intra-hexamer crosslinking via A14C/E45C at the NTD-NTD interface . The same disruption effect by TRIM5αrh on the inter-hexamer interfaces also occurred with purified intact HIV-1 cores . These results provide insights concerning how TRIM5α disrupts the virion core and demonstrate that structural damage of the viral capsid by TRIM5α is likely one of the important components of the mechanism of TRIM5α-mediated HIV-1 restriction .
TRIM5α is an important component of the innate immune defense against retroviral infection , including human immunodeficiency virus type -1 ( HIV-1 ) [1] , [2] , and numerous studies suggest that TRIM5α interacts with assembled capsids and induces premature capsid disassembly ( uncoating ) , before reverse transcription takes place [3]–[6] . TRIM5α is a 56 kD protein comprising a tripartite motif ( TRIM; with RING , B-box 2 , and coiled-coil ( CC ) domains ) followed by a C-terminal B30 . 2/SPRY domain [7]–[9] . Each of these domains plays distinct roles in the antiviral function of TRIM5α . The B30 . 2/SPRY domain binds to the viral capsid and determines the specificity of restriction , with sequence variation within this domain greatly impacting binding specificity [6] , [10]–[16] . For example , a single amino acid change in human TRIM5α ( TRIM5αhu ) , R332P , renders the protein capable of binding the HIV-1 capsid , causing it to behave like rhesus TRIM5α ( TRIM5αrh ) with regard to HIV-1 restriction [11] , [17] . The CC domain is necessary and sufficient for TRIM5α homo-dimerization , and this is important for capsid binding and restriction [12] , [18]–[20] . In vitro , specific recognition and binding to a hexagonal CA lattice requires both the CC and SPRY domains [19] . The B-box 2 domain is thought to be involved in higher-order structure formation and self-association , and its presence in the protein enhances TRIM5α binding to the capsid , compared to the CC-SPRY domains alone [21] , [22] . Several mutations in the B-box 2 domain abrogate HIV-1 restriction by TRIM5αrh [22]–[24] . The N-terminal RING domain is the least explored domain of TRIM5α . In general , RING domains are components of a particular class of E3 ubiquitin ligases that are involved in proteasome-mediated protein degradation ( reviewed in [25] ) . TRIM5α exhibits E3 activity , but the role of the ubiquitin ligase activity in retrovirus restriction is unclear . Deletion of the N-terminal RING domain reduces , but does not abolish antiviral restriction [23] , [26] , and treatment of cells with proteasome inhibitors does not prevent restriction by TRIM5α [27] . However , proteasome activity is necessary for the TRIM5α-mediated block to reverse transcription [27] , and engagement of restriction-sensitive virus cores results in proteasome-dependent degradation of TRIM5α [28] . Together , these data suggest that TRIM5α action in host restriction of retroviruses involves all of its domains . The negative influence of TRIM5α on viral reverse transcription is well established [1] , [3] , [4] , [6] , [29] , [30] , however , the detailed mechanism of restriction has not been elucidated . TRIM5α binds to assembled complexes composed of the CA-NC region of Pr55gag , but does not significantly interact with monomeric or soluble CA protein [31] . Furthermore , mutations in CA that decrease capsid stability appear to reduce TRIM5α binding in target cells , as HIV-1 particles with unstable cores are less effective at saturating TRIM5α-mediated restriction [5] . Finally , recent studies using a recombinant TRIM5αrh chimera , containing the RING domain of TRIM21 , demonstrated that the hybrid protein binds to CA-NC tubular assemblies and causes shortening of the tubes [32] , [33] , or self-assembles into higher-order structures , enhanced by binding to a preformed CA-NC hexagonal template [34] . Here , we employed cryoEM to investigate the direct interactions of tubular HIV-1 capsid assemblies and purified HIV-1 cores with the TRIM5αrh CC-SPRY protein and the structural consequences of TRIM5αrh CC-SPRY binding . We demonstrate that TRIM5αrh binding disrupts the tubes and creates non-random fragments . Specific inter-hexamer interfaces are preferentially broken , resulting in strings of subunits that are held together by the CA-CTD dimer . We further demonstrate that disruption by TRIM5αrh of purified HIV-1 cores also occurred preferentially at the inter-hexamer interfaces . Our data suggest that TRIM5αrh-mediated HIV-1 restriction involves direct engagement of the viral capsid , and structural damage to the capsid is likely one of the key components in this event .
To investigate the direct interactions between TRIM5αrh and the HIV-1 capsid , we generated purified recombinant proteins . Full length , wild-type TRIM5αrh has been quite difficult to obtain in sufficient quantities for biophysical and structural studies [32] , [35] . Therefore , we tested the expression and solubility of a number of different TRIM5αrh constructs , including one that comprises the CC-SPRY portion , by performing transient expression in SF9 insect cells . TRIM5αrh CC-SPRY ( residues 134–497 ) and TRIM5αhu CC-SPRY ( residues of 132–493 ) exhibited sufficient protein levels and solubility and , therefore , were selected for production in SF21 insect cells , using recombinant baculoviruses . The quaternary state of the purified recombinant human and rhesus TRIM5α CC-SPRY proteins was assessed by size exclusion chromatography in conjunction with in-line multi-angle light scattering , confirming that these proteins were dimers . The observed molecular masses extracted from the light scattering analyses are 92 kDa and 89 kDa , respectively ( Fig . 1A ) , compared to the theoretical values of 46 . 1 kDa and 45 . 6 kDa , respectively , based on amino acid sequences . Both proteins gave rise to almost identical CD spectra with a predominantly α-helical signature ( Fig . 1B ) . It is widely accepted that the restriction specificity of TRIM5α resides in its SPRY domain and that this domain interacts with retroviral capsids [1] , [3] , [11] , [14] , [15] , [36] . However , only recently has direct binding been demonstrated for a TRIM5-21R fusion chimera with CA-NC assemblies [32] , [34] . We used recombinant TRIM5α CC-SPRY proteins to examine direct binding to CA and CA-NC assemblies . Incubation of preassembled HIV-1 CA or CA-NC tubes with TRIM5αrh resulted in co-sedimentation of TRIM5αrh CC-SPRY/CA or TRIM5αrh CC-SPRY/CA-NC complexes , respectively , in the pelleted fractions ( Fig . 2 , Fig . S1 ) . More TRIM5αrh was observed bound to CA assemblies than to CA-NC assemblies ( Figs . 2 & 3 ) . In contrast , we observed negligible binding of TRIM5αhu CC-SPRY to HIV-1 CA or CA-NC complexes under the same assay conditions ( Fig . 2 ) . These data are consistent with previous results that demonstrated the inability of TRIM5αhu to bind and restrict HIV-1 , but a capacity for the same protein to recognize N-tropic murine leukemia virus ( MLV ) capsid [3] , [4] , [12] . A more quantitative analysis of TRIM5αrh binding was carried out by measuring molar ratios of CA and CA-NC-bound TRIM5αrh CC-SPRY over a range of TRIM5αrh concentrations . Dose-dependent binding was observed for both CA and CA-NC assemblies ( Fig . 3 ) . Consistently , at all concentrations , TRIM5αrh CC-SPRY bound CA more efficiently than CA-NC . This could be due to differences in CA and CA-NC structures on the surfaces of the assemblies , or differences in the flexibility of these assemblies , as CA-NC tubes were assembled in the presence of oligonucleotide . The binding ratios were 0 . 41 for TRIM5αrh CC-SPRY/CA and 0 . 21 for TRIM5αrh CC-SPRY/CA-NC , respectively , for the highest concentration of TRIM5αrh CC-SPRY ( 18 µM ) . When a lower concentration of TRIM5αrh CC-SPRY ( 1 µM ) was used for binding to the CA-NC tubular assemblies ( 10 µM ) , a molar ratio of 0 . 034 was obtained . This ratio is somewhat lower than the value reported by Langelier et al . for TRIM5-21R by immunoblotting [32] . The lower binding ratio for TRIM5αrh CC-SPRY is expected , since it lacks the self-associating B-box 2 domain , compared to the TRIM5-21R fusion protein . Furthermore , incubation with CC-SPRYrh did not alter the fraction of pelletable CA and CA-NC , even at the highest TRIM5αrh CC-SPRY concentrations ( Figs . 2&3 ) . These results are in accord with those reported for TRIM5-21R [32] and a binding study with CA-NC assemblies using TRIM5αrh-containing lysates [37] . Taken together , the data indicate that dimeric TRIM5αrh CC-SPRY directly interacts with tubular CA and CA-NC assemblies and that binding of TRIM5αrh does not dissociate these assemblies into soluble monomeric CA protein . Although no dramatic effect of purified TRIM5αrh on uncoating has been observed in vitro using CA-NC assemblies [32] , a substantial decrease in intact CA-NC tubes was noted when TRIM5αrh-containing cell lysates were mixed with CA-NC tubular assemblies [37] . To investigate this apparent dichotomy , we carried out cryoEM structural analyses of the samples that were used in the TRIM5α CC-SPRY/CA tubular assembly binding assays ( Figs . 2A&3A ) . CryoEM micrographs showed well-ordered CA tubular structures after incubation with binding buffer only ( Fig . 4A ) or TRIM5αhu CC-SPRY ( Fig . 4B ) , similar to our previously described assemblies [38] . In contrast , incubation of CA tubular assemblies with TRIM5αrh CC-SPRY ( 18 µM ) resulted in a massive structural break-down of the tubes ( Fig . 4C ) , accompanied by the appearance of distinct fragments composed of strings of hexamers ( Fig . 4C inset ) [38] . The remaining tubes had generally lost the regularity of the hexagonal lattice . Some TRIM5αrh CC-SPRY densities apparently remained on several of the fragments ( Fig . 4C inset ) . Gold-labeling of TRIM5αrh CC-SPRY in complex with CA tubular assemblies confirmed that TRIM5αrh CC-SPRY bound to the CA assemblies ( Fig . S2 ) . These break-down fragments were primarily present in the pellet fraction after centrifugation ( Fig . 3A ) , confirmed by cryoEM imaging of the pellet samples ( Fig . S3 ) , explaining why no effect on uncoating was detected in assays that measure soluble CA [32] , [37] . These results suggest that the predominant effect of TRIM5αrh is the break down of HIV-1 capsids into fragments and not the dissociation into soluble monomers . We further examined the effect of CA mutations on TRIM5αrh disruption . Several CA mutants , including A92E , which was used in our previous structural study [38] , and the E45A mutant , which produces hyperstable capsids , were analyzed . The effect of TRIM5αrh CC-SPRY binding to A92E CA tubular assemblies was similar to that observed with wild-type CA ( Fig . S4A&B ) . The CA tubular assemblies carrying the capsid-stabilizing E45A mutation [46] also experienced structural damage by TRIM5αrh , but to a lesser degree ( Fig . S4C&D ) . This suggests that the overall stability of HIV-1 capsid assemblies may modulate or interfere with TRIM5αrh function , consistent with findings that hyperstable capsid core mutants effectively saturate TRIM5α-mediated restriction [5] . To determine which interface in the capsid lattice is disrupted by CC-SPRYrh , we tested the effect of TRIM5αrh CC-SPRY on cross-linked CA tubular assemblies . In previous work , we showed that introduction of a pair of cysteines , P207C/T216C , at the pseudo three-fold inter-hexamer interface , efficiently cross-linked three neighboring CA molecules into trimers upon oxidation ( Fig . 5A&B ) . The interactions at this interface are mediated by the CA-CTD , predominantly helices H10 and H11 [38] . Such cross-linked P207C/T216C CA tubular assemblies are expected to contain stronger hexamer-hexamer interactions , stabilizing the lattice . The P207C/T216C mutant assembles into tubular structures very similar to the wild-type CA ( Fig . S5 ) . Both oxidized and non-oxidized P207C/T216C CA tubular assemblies bound TRIM5αrh CC-SPRY , without any significant difference between them ( Fig . 5B , lanes 1-4 ) . However , cryoEM analysis revealed that TRIM5αrh CC-SPRY exerted very little structural damage onto the cross-linked tubes , whereas the non-oxidized tubular assemblies exhibited similar structural breakdown as seen for wild type CA tubes ( Fig . 5C&D ) . These data suggest that TRIM5αrh CC-SPRY engages in inter-hexamer binding , most likely pulling apart the trimer interface , thereby disrupting the assembled tubes . We further tested this possibility by measuring the cross-linking efficiency of P207C/T216C CA assembly after TRIM5αrh CC-SPRY treatment . As can be seen from the results illustrated in Fig . 5B ( lanes 5&6 ) , the level of cross-linked trimers was significantly reduced after incubation with TRIM5αrh CC-SPRY . The percentage of the cross-linked CA trimer over total CA in the reduced sample is 3-fold less in the TRIM5αrh CC-SPRY treated sample , compared to untreated sample , confirming that the trimer interface between three neighboring hexamers is disrupted by TRIM5αrh CC-SPRY . An alternative scenario could involve binding of the TRIM5α CC-SPRY dimer within a CA hexamer , with TRIM5αrh CC-SPRY dimers pushing apart the hexamers . However , simple geometric considerations make this a very unlikely scenario if TRIM5αrh SPRY binds near the cyclophilin A binding loop in CA [39] , since the distance between two sites ( >110 Å ) is too large for the TRIM5α CC-SPRY dimer protein to span . Nonetheless , we tested for this possibility using a A14C/E45C CA double cysteine mutant , which can cross-link CA within hexamers [40] . Following incubation with TRIM5αrh CC-SPRY , crosslinked A14C/E45C CA assemblies exhibited only a slight reduction in CA hexamers ( Fig . S6 , compare lanes 2 & 5 ) , compared to the dramatic reduction of the trimer in the P207C/T216C CA assemblies ( Fig . 5B , right panel ) . This small effect on the CA hexamer could be caused by minor perturbations at the intra-hexamer CA interfaces upon TRIM5αrh CC-SPRY binding . Small amounts of CA dimer ( ∼50kD , Fig . S6 , lanes 1 , 3 , 5 , 7&9 ) in the non-oxidized assemblies and dimer of hexamers ( ∼280kDa , Fig . S6 , lanes 2&8 ) in the oxidized A14C/E45C CA assemblies were observed by SDS-PAGE , possibly due to the CA CTD dimer interaction . Interestingly , the amount of hexamer dimers was greatly diminished in the TRIM5αrh CC-SPRY treated sample ( Fig . S6 , lane 5 compared to lane 2&8 ) . Again , these data further support that TRIM5αrh CC-SPRY binding perturbs the CA inter-hexamer interface . To extend the above in vitro studies to biological HIV-1 capsids , we examined the effect of TRIM5αrh CC-SPRY on isolated HIV-1 cores . For this purpose , we purified cores from the HIV-1 CA mutants A14C/E45C and P207C/T216C for two reasons; first , the mutant cores appeared to be more stable through the isolation procedure , and second , A14C/E45C and P207C/T216C cores bear the same cysteine mutations that we used for the in vitro analysis described in the previous section . A14C/E45C and P207C/T216C cores were isolated from virions in high yield ( average of 44% of virion-associated CA , vs . ∼15% typically observed for wild type ) by brief detergent treatment and sucrose gradient sedimentation . The CA protein in A14C/E45C cores was readily cross-linked into hexamers , as shown by non-reducing SDS-PAGE analysis ( Fig . S7 ) . Despite the extensive CA hexameric crosslinking in A14C/E45C cores , incubation with TRIM5αrh CC-SPRY resulted in a dramatic loss of intact cores observed by cryoEM , compared to the samples treated with the same amount of human TRIM5α CC-SPRY ( Fig . 6A-C ) . In contrast , no significant reduction in the number of P207C/T216C cross-linked cores was seen upon TRIM5αrh incubation ( Fig . 6E and F , Fig . S7 , +oxidizer samples ) . However , without ensuring effective cross-linking at the trimer interface ( Fig . S7 , -oxidizer ) , a four-fold decrease in the number of P207C/T216C cores was seen upon TRIM5αrh treatment , compared to incubation with TRIM5αhu ( Fig . 6D and F , oxidizer samples ) . Although very few , a small number of P207C/T216C cores were observed in TRIM5αrh treated samples , presumably due to low levels of spontaneous crosslinking of isolated P207C/T216C cores at the trimer interface . Furthermore , similar protofilament fragments as seen for the in vitro assemblies were also observed after TRIM5αrh treatment of cores ( Fig . 6D , arrows and inset ) . The above data demonstrate , for the first time , that TRIM5αrh CC-SPRY is capable of exerting direct structural damage on the isolated HIV-1 cores and TRIM5αrh binding preferentially disrupts the inter-hexamer interfaces in the HIV-1 capsid . Examination of the fragments present in the cryoEM images revealed predominantly curved linear structures ( Fig . 4C ) . These structures resemble fragments of protofilaments in CA helical assemblies . Our results are consistent with previous studies that TRIM5αrh binding to CA-NC assemblies did not increase soluble CA-NC monomers and dimers [32] , [37] , and further suggest that binding of TRIM5αrh disrupts the hexamer-hexamer interfaces , thereby releasing protofilaments along one of the three principal helical directions . A model based on the above findings is depicted schematically in Figure 7 . CA assembles into helical tubes in vitro with a hexagonal surface unit formed by CA NTDs that is connected by CTD-CTD dimer and trimer interfaces on the inner surfaces of the three-dimensional tube or cone [38] , [40] , [41] . In these helical tubes , three slightly different inter-hexamer interactions were observed ( see Fig . 3 in [38] ) . Binding of TRIM5αrh may disrupt these interactions differentially , weakening the CTD-CTD interfaces between hexamers . In turn , this causes release of CA protofilament fragments , such as those illustrated in Figure 7 , and , indeed , similar types of fragments were observed in the cryoEM images ( Fig . 4C ) . For TRIM5-21R interacting with CA-NC , shortening of tubes was observed in vitro , in addition to fragmentation [32] . Examples of this type of fragmentation of helical tubes have also been observed in other biological systems , including microtubules in vivo and in vitro [42] , [43] , actin filaments [44] and dynamin spirals and tubes [45] . Thus , the disassembly of the CA tubes into helical-type fragments is not unprecedented . Importantly , the use of two mutants , A14C/E45C and P207C/T216C , containing engineered disulfide bonds , allowed us to assign the site of TRIM5αrh action to the inter-hexamer interface ( vs . the intra-hexamer interface ) , both , for in vitro assemblies and isolated HIV-1 cores , providing compelling evidence for specific structural disruption of the trimer interface of the HIV-1 capsid upon TRIM5α binding . In this manner , key insights into the mechanistic aspects of TRIM5αrh - capsid interaction were obtained . Retroviral uncoating is a poorly characterized process , generally defined as viral capsid disassembly following release of the viral core into the target cell . Studies using HIV-1 CA mutants indicate that the stability of the viral core is optimally balanced for efficient viral replication [46] . Therefore , a plausible mechanism for restriction by TRIM5α involves binding to the viral capsid , capsid destabilization , and perturbation of uncoating . Here , we show by cryoEM that TRIM5αrh CC-SPRY binding to CA assemblies causes massive destruction of assembled HIV-1 CA complexes . A similar effect was observed with purified HIV-1 cores . Intriguingly , this effect was seen with the TRIM5αrh CC-SPRY domain construct lacking the RING and B-box domains , albeit at high concentrations , even though TRIM5α protein devoid of RING and B-box domains was reported to lack restriction activity when expressed in cells [23] , [24] . These seemingly inconsistent results could be due to several factors , including: ( 1 ) reduced binding to the viral capsid in the cell due to lack of self-association mediated by the B-box that can be overcome at high protein concentration in vitro; ( 2 ) improper intracellular localization of the deletion protein; or ( 3 ) altered association with host cell factors . We favor the first explanation , since the CC-CypA protein has been shown to restrict HIV-1 and FIV when expressed in target cells [47] , and oligomerization of CypA appears sufficient to induce HIV-1 restriction [48] . Given the ability of the B-box domain to promote higher-order TRIM5α association [21] , it seems plausible that this domain in intact TRIM5αrh may potentiate the effects observed here for TRIM5αrh CC-SPRY . Most importantly , while the CC-SPRY from rhesus TRIM5α was active on our in vitro assemblies and isolated cores , the corresponding human TRIM5α fragment was inactive . Thus , binding of the CC-SPRY domain to CA is essential for TRIM5α retroviral restriction and for structural disruption of the capsid . However , our current results do not exclude the possibility of additional structural consequences induced by higher-ordered oligomerization of TRIM5α on the viral capsid . Although the molecular mechanism of TRIM5α restriction is not fully understood , current models hypothesize that after capsid release into the target cell , TRIM5α binds and triggers premature capsid disassembly . Our results suggest that direct binding of TRIM5α to the capsid is sufficient to inflict direct structural damage . Yet , cellular proteasome activity is clearly involved in the block to reverse transcription induced by TRIM5α[27] . Recruitment of proteasomes , most likely via the TRIM5α RING domain , may further disaggregate capsid fragments and also degrade TRIM5α [28] , thereby mediating the irreversible block to infection . In contrast to TRIM5α-mediated restriction , Fv1 restriction of MLV does not result in inhibition of reverse transcription , yet both TRIM5α and Fv1 target the retroviral capsid . We speculate that the common feature in TRIM5α and Fv1 restriction is the structural damage to the capsid , with the major mechanistic difference involving recruitment of the proteasome in the case of TRIM5α-dependent restriction . The findings presented here represent the first detailed structural analysis of TRIM5α disruption of the CA lattice to date . Additional structural studies of TRIM5α effects , especially with regard to the CTD-CTD interfaces in CA assemblies and HIV-1 cores , as well as the involvement of the RING and B-box domains , will further aid to elucidate the molecular mechanisms of TRIM5-mediated HIV-1 restriction and may offer insights into the HIV-1 virus-cellular interplay as well as lead to novel approaches in antiviral therapy .
cDNAs encoding the coiled-coil and SPRY domains of human and rhesus TRIM5α ( TRIM5α CC-SPRY; residues 132-493 and 134-497 , respectively ) were amplified and cloned into the pENT-TOPO vectors ( Invitrogen ) , modified to encode a Strep-tag at the N-terminus and a His6-tag at the C-terminus of the proteins . The cDNAs encoding HIV-1 capsid ( CA ) and capsid- nucleocapsid ( CA-NC ) were amplified from pNL4-3 and cloned into the pET21 vector ( Invitrogen ) . All clones were verified by sequencing of the entire coding region . Baculoviruses expressing human and rhesus TRIM5α CC-SPRY were prepared using the Baculdirect C-term ( Invitrogen ) according to the manufacturer's protocols . Proteins were expressed in SF21 insect cells by infecting cells with recombinant baculoviruses at a MOI of 2 for 40 h . Cells were lysed by sonication in a buffer containing 25 mM sodium phosphate , pH 7 . 5 , 250 mM NaCl , 10 mM beta-mercaptoethanol , and 0 . 02% sodium azide . Soluble proteins were purified over a 5 mL Ni-NTA column followed by passage over a Hi-Load Superdex 200 16/60 column ( GE Healthcare ) in a buffer containing 25 mM sodium phosphate , pH 7 . 5 , 150 mM NaCl , 2 mM DTT , 10% glycerol , and 0 . 02% sodium azide . The fraction containing TRIM5α CC-SPRY was further purified over a 5 mL Hi-Trap QP column ( GE Healthcare ) using a gradient of 0–1 M NaCl or a 5 mL StrepTrap-HP column ( GE-Healthcare ) using 2 . 5 mM desbiotin for elution . CA-NC proteins were expressed in E . coli Rosetta 2 ( DE3 ) , cultured in Luria-Bertani medium , using 0 . 4 mM IPTG for induction and growth at 18°C for 23 h . The proteins were purified as described in Ganser et al [49] . Briefly , soluble proteins were precipitated with 40% ( w/v ) ammonium sulfate after DNA was removed by precipitation with polyethylenimine . The precipitates were dialyzed against a buffer containing 25 mM TrisHCl , pH 7 . 5 , 50 mM NaCl , 1 µM ZnSO4 , 10 mM beta-mercaptoethanol , and 0 . 02% azide . Proteins were separated by column chromatography over a 5 mL Hi-Trap SP ( GE Healthcare ) with a 0–1 M NaCl gradient and Hi-Load Superdex75 26/60 columns , equilibrated with a buffer containing 25 mM TrisHCl , pH 7 . 5 , 150 mM NaCl , 1 µM ZnSO4 , 10 mM beta-mercaptoethanol , and 0 . 02% azide . CA proteins were prepared as described in Byeon et al [38] . HIV-1 cores were isolated from virions by a modification of the “spin-thru” method previously described [50] . HIV-1 viruses were derived from the R9 molecular clone [51] and mutants thereof . CA mutations were created by overlap PCR . SpeI-ApaI fragments were transferred into R9 , and the transferred region was verified by PCR . HIV-1 viruses were produced by transient transfection of sixty dishes of 6×106 293T cells with 10 µg plasmid DNA ( using 10 µg of HIV-1 construct R9 , R9 . Env- , or R9 . A14C/E45C ) using polyethylenimine ( 3 . 6 µg/ml , Polysciences ) [52] in each 10 cm dish . Two days after transfection , virus-containing supernatants were collected and clarified by filtration ( 0 . 45 µm pore-size ) . Particles in clarified supernatants ( 600 ml ) from 293T cells were pelleted through 3ml cushions of 20% sucrose ( 120 , 000 ×g , 2 . 5 h ) in a Beckman SW32Ti rotor then gently suspended in a total of 1 . 2 ml STE buffer ( 10 mM Tris-HCl [pH 8 . 0] , 100 mM NaCl , 1 mM EDTA ) for 2 h at 4°C . The concentrated virus suspension was subjected to equilibrium ultracentrifugation ( 120 , 000 × g , 16 h , 4°C , Beckman SW-32Ti rotor ) through a layer of 1% Triton X-100 into a linear gradient of 30%–70% sucrose in STE buffer . Twelve 1-ml fractions were collected . CA concentrations were determined by p24 ELISA [53] . The peak p24 fractions near the bottom of the gradient were pooled and concentrated to ∼100 µl by diafiltration with an Ultracel-10K protein concentrator ( Amicon ) . The sample was diluted with STE buffer and reconcentrated to reduce the final sucrose concentration in the sample to less than 0 . 5% . The concentrated samples of cores were then assayed for p24 by ELISA . Light-scattering data were obtained using an analytical Superdex200 column ( 1 cm ×30 cm , GE Healthcare ) with in-line multi-angle light scattering ( HELEOS , Wyatt Technology ) , variable wavelength UV ( Agilent 1100 Series , Agilent Technology ) and refractive index ( Optilab rEX , Wyatt Technology . ) detection . Approximately 100 µL of 2 mg/mL protein solutions were injected into the pre-equilibrated column using 25 mM sodium phosphate buffer ( pH 7 . 5 ) , 250 mM NaCl , 10% glycerol , and 0 . 02% ( w/v ) sodium azide at a flow-rate of 0 . 5 ml/minute for equilibration and elution . Molecular masses were determined from the scattering data using the ASTRA program ( Wyatt Technology ) . CD spectra of TRIM5α CC-SPRY ( 5 . 4 µg/mL ) were collected in a buffer containing 1 mM sodium phosphate , pH 7 . 5 , 14 mM NaCl with a Jasco-810 CD spectrophotometer ( Easton , MD ) . Data were collected with a scan rate of 1 nm/sec from 260 to 200 nm at a constant temperature of 12°C and averaged over 40 scans . CA and CA-NC tubes were assembled containing 80 µM ( 2 mg/ml ) CA , 1 M NaCl and 50 mM Tris-HCl ( pH 8 . 0 ) at 37°C for one hour or 300 µM CA-NC , 60 µM TG50 oligonucleotide in 250 µM NaCl , 50 mM Tris-HCl ( pH 8 . 0 ) buffer at 4°C for 19 hr , respectively . For the TRIM5α CC-SPRY binding assays , the binding buffer , 10 mM Tris pH 7 . 5 , 330 mM NaCl , 1 mM TCEP , 0 . 02% Azide , 5% Glycerol , is also the stock buffer for TRIM5α CC-SPRY proteins . Briefly , binding buffer containing different concentrations of TRIM5α CC-SPRY was added to preassembled CA and CA-NC tubes . CA concentration was slightly reduced to 64 µM in the binding assays . The CA-NC assemblies were diluted to final concentrations of 80 µM ( comparable to the amount of total protein used with CA ) or 10 µM ( comparable to the number of tubes seen with CA ) with assembly buffer prior to the binding assays . TRIM5αhuCC-SPRY or TRIM5αrhCC-SPRY aliquots from 4 mg/ml stock solutions were added to preassembled CA and CA-NC tubes . The reaction mixture was incubated on a rocking platform at room temperature for 1 hr with gentle mixing at 10 min intervals . At the end of incubation , 5 µl samples were withdrawn from the reaction mixtures and immediately used for cryoEM analysis . 6 µl samples from the same reaction mixtures were mixed with 4X LDS loading buffer ( Invitrogen ) supplemented with 10 mM DTT for SDS-PAGE analysis ( t ) . The remaining sample was pelleted at 20 , 000 g with an Eppendorf centrifuge 5417R for 15 min and supernatants ( s ) and pellets ( p , resuspended in 1/3 of volume ) were mixed with 4X LDS loading buffer for gel analysis . Total , supernatant , and pellet samples , without boiling , were loaded on 10% SDS-PAGE and stained with Coomassie Blue . Each experiment was carried out at least three times . His-tagged TRIM5α proteins at the C-terminus , TRIM5αhuCC-SPRY and TRIM5αrhCC-SPRY , were labeled using 5 nm Ni-NTA-Nanogold gold beads from Nanoprobes ( Yaphank , NY ) . For gold labeling , wild type CA protein was assembled into tubes using 80 µM ( 2 mg/ml ) CA in the assembly buffer ( 1 M NaCl and 50 mM Tris-HCl ( pH 8 . 0 ) ) at 37°C for one hour . TRIM5αhuCC-SPRY or TRIM5αrhCC-SPRY ( 2 µl ) was added to the assembly mix ( 20 µl ) to a final concentration of 18 µM and incubated on a rocking platform at room temperature for 1 hr with gentle mixing at 10 min intervals . 2 . 7 µl of 5 nm Ni-NTA-Nanogold gold beads ( stock concentration , 0 . 5 µM ) in 100 mM imidazole ( pH 8 . 0 ) was added to the assemblies and allowed to incubate at room temperature for 20 minutes . The mixture was then centrifuged at 3 , 000 g and the pellet was resuspended in assembly buffer . Samples were immediately applied to glow-discharged EM grids for negative staining with 1% uranyl acetate solution after resuspension . Images were acquired on an FEI Tecnai TF20 electron microscope at a nominal magnification of 50 , 000 and with underfocus values about 2 µm , using a Gatan ultrascan 4KX4K CCD camera ( Gatan Inc . , Pleasanton , CA , U . S . A . ) . The cross-linking experiment was set up as previously described [54] . Briefly , 30 µl P207C/T216C or A14C/E45C CA were preassembled in the presence of 50 µM DTT under the conditions described above . The assembled material was then subjected to centrifugation at 20 , 000 g at room temperature in an Eppendorf centrifuge 5417R for 15 minutes . The pellet was resuspended in 30 µl assembling buffer and oxidized with 1 µl of 30x oxidizer mix ( 60 µM CuSO4 , ( Sigma ) dissolved in water , and 267 µM 1 , 10-Phenanthroline ( Sigma ) dissolved in 100% ethanol in a 1∶1 ratio ) for 5 seconds , immediately followed by quenching with 20 mM iodoacetamide ( Sigma ) and 3 . 7 mM Neocuproine ( Sigma ) . For the dose-dependent TRIM5αrh CC-SPRY binding assay , the SDS-PAGE gels were scanned ( Epson 4990 scanner ) and the integrated intensities of CA , CA-NC , and TRIM5αrh protein bands in pellet fractions were measured using Image J 1 . 40 g program ( NIH ) . The molar ratios were calculated according to the formula ( TRIM5αrh band intensity/TRIM5αrh molecular weight ) / ( CA band intensity/CA molecular weight ) . Aliquots from the binding assays ( above ) were subjected to cryoEM analysis . 2 µl were applied to the carbon side of a glow discharged perforated Quantifoil grids ( Quantifoil Micro Tools , Jena , Germany ) , and 2 . 5 µl binding buffer was added to the back side of the grids . Grids were blotted and plunge-frozen in liquid ethane using a manual gravity plunger . Low dose ( 10∼15 e−/Å2 ) projection images were collected on an FEI Tecnai TF20 electron microscope at a nominal magnification of 50 , 000 and with underfocus values ranging from 1 . 0 to 2 . 5 µm , using a Gatan ultrascan 4KX4K CCD camera ( Gatan Inc . , Pleasanton , CA , U . S . A . ) . The effect of Rhesus TRIM5α CC-SPRY on HIV-1 cores was examined and quantified using cryoEM . 18 µM rhesus or human TRIM5α CC-SPRY proteins were added to a solution of isolated HIV-1 A14C/E45C or P207C/T216C cores ( ∼11 µg/ml ) . After one hour incubation at room temperature with gentle agitation , the samples were subjected to cryoEM analysis . For each sample , about 80 low dose projection images were collected at 19 , 000x magnification . Each field of view covers about 5 µm2 . The image areas were chosen randomly , owing to the nature of cryoEM imaging . The number of cores in each sample was quantified using average number of cores per image frame . Mean values from four totally independent experiments are plotted in Fig . 6 with the standard deviation indicated . | The cellular protein TRIM5α is a host cell restriction factor that blocks HIV-1 infection in Rhesus macaque cells by targeting the viral capsid . Here , we show that direct binding of a TRIM5α protein , consisting of the coiled-coil and B30 . 2/SPRY domains , to the viral capsid results in disruption of the surface lattice and fragmentation of the capsid , specifically at inter-hexamer interfaces . Our results reinforce the notion that structural damage of the viral capsid by TRIM5α is central to the mechanism of TRIM5α-mediated HIV-1 restriction . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunodeficiency",
"viruses",
"biochemistry",
"protein",
"interactions",
"viral",
"core",
"proteins",
"virology",
"macromolecular",
"assemblies",
"biology",
"microbiology",
"viral",
"structure",
"biophysics"
] | 2011 | Rhesus TRIM5α Disrupts the HIV-1 Capsid at the InterHexamer Interfaces |
Infectious complications are a common cause of morbidity and mortality in cancer patients undergoing chemotherapy due to increased risk of oral and gastrointestinal candidiasis , candidemia and septicemia . Interactions between C . albicans and endogenous mucosal bacteria are important in understanding the mechanisms of invasive infection . We published a mouse intravenous chemotherapy model that recapitulates oral and intestinal mucositis , and myelosuppression in patients receiving 5-fluorouracil . We used this model to study the influence of C . albicans on the mucosal bacterial microbiome and compared global community changes in the oral and intestinal mucosa of the same mice . We validated 16S rRNA gene sequencing data by qPCR , in situ hybridization and culture approaches . Mice receiving both 5Fu and C . albicans had an endogenous bacterial overgrowth on the oral but not the small intestinal mucosa . C . albicans infection was associated with loss of mucosal bacterial diversity in both sites with indigenous Stenotrophomonas , Alphaproteobacteria and Enterococcus species dominating the small intestinal , and Enterococcus species dominating the oral mucosa . Both immunosuppression and Candida infection contributed to changes in the oral microbiota . Enterococci isolated from mice with oropharyngeal candidiasis were implicated in degrading the epithelial junction protein E-cadherin and increasing the permeability of the oral epithelial barrier in vitro . Importantly , depletion of these organisms with antibiotics in vivo attenuated oral mucosal E-cadherin degradation and C . albicans invasion without affecting fungal burdens , indicating that bacterial community changes represent overt dysbiosis . Our studies demonstrate a complex interaction between C . albicans , the resident mucosal bacterial microbiota and the host environment in pathogenesis . We shed significant new light on the role of C . albicans in shaping resident bacterial communities and driving mucosal dysbiosis .
Oropharyngeal ( OPC ) and gastrointestinal candidiasis are common infections in patients on high dose cancer chemotherapy , mostly attributed to Candida albicans . In these populations prevalence rates of OPC range between 25–40% [1] . Cytotoxic chemotherapy also causes an inflammatory form of oral and gastrointestinal injury known as mucositis [2] . Mucosal injury combined with the myelosuppressive effects of chemotherapy , promote bacterial and fungal translocation through mucosal barriers leading to bloodstream infections , a major cause of morbidity and mortality in this patient population [3 , 4] . There is some evidence that the oral and intestinal bacterial microbiota in humans may be altered by cytotoxic chemotherapy , although the effects on the fungal microbiota are less clear [5 , 6] . In a healthy host , unperturbed resident commensal bacterial communities play an important role in limiting C . albicans colonization in mucosal sites [7] . However , when the microbial equilibrium is changed by immunosuppression certain bacterial species may overgrow and form mutualistic relationships with C . albicans . This in turn may lead to a well-coordinated dysbiosis which amplifies mucosal damage . We recently revealed mutualistic relationships between C . albicans and commensal oral streptococci in a mouse model of OPC [8 , 9 , 10 , 11] . However , these studies were performed with bacteria that are not part of the indigenous mouse microbiota . A role for indigenous bacterial community-mediated dysbiosis in fungal pathogenesis has never been examined . Furthermore , the effects of C . albicans in modulating commensal bacterial community composition have only been studied in the mouse gastric and intestinal mucosa . In these studies C . albicans was shown to favor growth of endogenous enterococci post-antibiotics treatment [12 , 13 , 14] . Studies focusing on the interplay between C . albicans and resident oral mucosa bacteria in health and disease are nonexistent . Chemotherapy-associated mucosal candidiasis studies use high doses of chemotherapeutic agents and focus exclusively on the development of gastrointestinal or disseminated candidiasis . The vast majority of these studies also used antibiotics aimed at increasing gastrointestinal fungal burdens [15 , 16 , 17 , 18] . We recently developed a mouse model of low dose intravenous 5Fu administration that recapitulates the histopathologic changes associated with cancer chemotherapy-induced mucositis [19] . Using this model we tested the hypothesis that mucosal injury combined with peripheral neutropenia induced by 5Fu , are sufficient for the development of oropharyngeal and intestinal candidiasis in mice with unperturbed resident bacterial microbiota . For the first time we also examined the influence of 5Fu and candidiasis on the mucosal bacterial microbiome and the reciprocal effect of the resident microbiota on C . albicans virulence .
In the first series of experiments we tested whether intravenous , low dose 5Fu administration increases the susceptibility of mice to oral and intestinal candidiasis . Fig 1A shows that mice receiving four doses of 5Fu ( 50 mg/kg every other day ) developed tongue papillary atrophy , epithelial desquamation and erosion , consistent with early stages of mucositis [19 , 20] . While this treatment caused gradual depletion of mature neutrophils in the bone marrow and blood , a steady infiltrate of CD11b+/LyG+ cells was found in the tongue mucosa reflecting local inflammation secondary to mucosal injury ( S1A , S1B and S1C Fig ) . Mice receiving 5Fu+C . albicans developed thick white biofilms covering the posterior tongue surface , associated with epithelial ulcerations ( Fig 1A ) . In these mice C . albicans fungal burdens increased significantly over time in all mucosal surfaces ( Fig 1B ) . This group also lost significantly more weight than mice treated with 5Fu alone ( Fig 1C ) . We also noted almost complete absence of neutrophils in the tongues of this group ( S1A , S1B and S1C Fig ) . This was not due to absence of neutrophil activating cytokines since KC and IL-6 were significantly increased in infected tissues ( S1D Fig ) . In mice treated with 5Fu and infected with C . albicans , tongue biofilms were composed by C . albicans and indigenous bacteria ( Fig 1A , lower panel ) . The PBS control group receiving C . albicans showed complete absence of biofilms and pathology and an early increase in tongue neutrophils consistent with other reports ( Figs 1A and S1B ) [21] . No bacterial biofilms were seen on the surface of 5Fu-only or C . albicans-only treated mice using a pan-eubacterial FISH probe ( Fig 1A , lower panel ) . The presence of endogenous bacteria in biofilms with C . albicans prompted a closer examination of the resident bacterial microbiota . We first compared viable ( CFU ) and total ( qPCR ) bacterial biomass across all experimental groups at the end of the infection period . As seen in Fig 1D there was a significant increase in the viable and total bacterial biomass in mice treated with 5Fu and infected with C . albicans , compared to all other groups . We also noted a significant increase in endogenous bacteria with 5Fu treatment alone , compared to untreated control and C . albicans alone groups . However , since the increase in bacterial biomass with 5Fu did not lead to the development of biofilms ( Fig 1A ) , this suggested that in the absence of C . albicans bacteria could not organize in biofilm community structures on the tongue surface . A positive correlation was found in fungal and bacterial loads on the same tongues at the end of the experimental period ( S2A Fig ) , suggesting that fungal burdens increased in concert with endogenous bacterial burdens . To confirm this we examined time-dependent changes in the viable and total bacterial biomass on the tongue mucosa of mice in this group . S2B Fig shows a gradual increase in bacterial biomass over time , confirming that endogenous bacteria increased as C . albicans burdens rose ( Fig 1B ) . Collectively these data show that C . albicans infection in 5Fu-treated mice promotes bacterial overgrowth on the oral mucosa and that 5Fu treatment contributes to this effect . In contrast , mucosa-associated bacterial loads did not change significantly after 8 days of infection in the jejunum of mice with candidiasis ( S2C Fig ) , suggesting that the effects of C . albicans infection on commensal bacteria are mucosal site-specific . Strain SC5314 colonized the jejunum of PBS control mice in low numbers ( not shown ) and this was associated with a significant reduction in resident bacterial CFUs ( S2D Fig ) . A smaller decrease in the viable bacterial biomass was also noted in mice receiving both 5Fu and C . albicans . These results suggested that growth of C . albicans in this mucosal site displaces endogenous bacterial communities . Cytotoxic chemotherapy elevates the risk for bloodstream infections [22] , we thus asked whether C . albicans disseminated in distant organs . We found a time-dependent increase in fungal burdens in kidneys and livers , accompanied by increased bacterial burdens in the same organs ( Fig 1E ) . Collectively these data suggest that 5Fu creates favorable conditions for C . albicans and endogenous bacterial growth in the upper and lower alimentary tract mucosae and systemic dissemination of bacteria and fungi . Importantly , mice receiving both 5Fu and C . albicans had an endogenous bacterial overgrowth in the oral mucosa that exceeded 5Fu treatment alone . The increase in oral bacterial burdens observed in mice with OPC raised the possibility of global mucosa-associated bacterial microbiota changes . We thus performed high throughput sequencing of the V4 hypervariable region of the 16S rRNA gene in DNA extracted from tongues . We also analyzed the bacterial microbiome of the jejunum in the same mice , for comparison . To explore differences in bacterial community composition within each treatment and control groups we first analyzed alpha diversity as reflected by Shannon index ( Fig 2A ) and richness estimates ( S3A Fig ) . Compared to the untreated group the 5Fu group was not significantly different in community diversity when either the tongue or jejunum were examined . However , inoculation with C . albicans alone induced a decrease in bacterial diversity in the oral mucosa , whereas diversity in the jejunum increased ( Fig 2A ) , consistent with other reports [23] . This illustrates that daily inoculation with this strain had an impact on oral biodiversity even though colonization was below the sensitivity limit of the CFU assay . A dramatic drop in bacterial diversity was noted in the tongues of mice receiving C . albicans and 5Fu ( Fig 2A ) , consistent with the reduction in the number of species ( OTUs ) observed ( S3A Fig ) . A significant reduction in the average number of species was also observed in the jejunum of the same mice ( S3A Fig ) . To further explore the impact of C . albicans on oral bacterial diversity we performed regression analysis of the Shannon index in relation to C . albicans CFUs in the same tongues . This analysis showed a negative correlation between oral fungal burdens and bacterial diversity ( Fig 2B ) . In summary , these data show that in a disease-permissive host environment C . albicans reduces the bacterial diversity both in the upper and lower GI tract mucosa . To better understand the effect of C . albicans in chemotherapy-treated mice we performed beta diversity analyses comparing the 5Fu and 5Fu+C . albicans groups . Non-metric multidimensional scaling ( NMS ) analysis of Bray-Curtis dissimilarities between the two treatments showed that the bacterial microbiome composition was distinct in mice receiving 5Fu compared to mice receiving 5Fu+C . albicans in both sites ( Fig 2C ) . Community structure from most tongues of the group receiving C . albicans alone , clustered closer to the two uninfected control groups , suggesting that C . albicans alone does not alter the composition of the oral microbiome significantly . As expected , microbial communities also clustered by site indicating that they harbor microbiota with distinct global community structures and composition . Time-dependent analysis of beta diversity changes in the microbiomes of the two treatment groups further showed that C . albicans caused a profound disruption of the tongue and small intestinal community structure after 6 days of chemotherapy ( S3B Fig ) . Analysis of the top 10% prevalent bacterial OTUs on the tongue mucosa revealed distinct genus level differences among the four experimental groups ( Fig 2D ) . Specifically , an increase in the relative abundance of the genus Enterococcus was noted on the oral mucosa in both groups receiving C . albicans , with the most dramatic increase in mice also receiving 5Fu ( representing ~99% of all OTUs ) , explaining the almost complete loss of diversity in this group ( Fig 2A ) . A smaller increase in the enterococcal OTUs was also noted with 5Fu treatment alone ( Fig 2D ) . The increase in enterococcal biomass in all groups over untreated control , with the 5Fu+C . albicans group being the highest , was validated by a genus-specific qPCR assay ( Fig 2E ) . Importantly , endogenous enterococci were identified in mixed tongue biofilms with C . albicans using genus and species ( E . faecalis ) -specific FISH probes . In these biofilms C . albicans was noted invading into the submucosal tongue compartment ( Fig 2F , demarcated area ) . Furthermore , 98% of the OTUs in bacterial cultures from these tongues were identified as enterococci by 16S rRNA gene sequencing ( Fig 2G ) and all isolates PCR-amplified with E . faecalis-specific primers ( not shown ) . A similar analysis of the most abundant OTUs in the jejunum mucosa of the same mice showed the most dominant taxa to be Stenotrophomonas , Alphaproteobacteria and to a lesser extent Enterococcus ( S3C Fig ) . A reduction in the relative abundance of Lactobacillus species was observed in both the tongue and intestinal mucosa of infected mice ( Figs 2D and S3C ) . This further shows that despite the distinct mucosal microbiota shifts in the two sites C . albicans infection led to Enterococcus growth and Lactobacillus reduction in both sites . C . albicans SC5314 does not stably colonize the oral mucosa of healthy mice when inoculated via the drinking water and viable counts are below the sensitivity of the CFU assay ( Fig 3A ) . In order to further explore the relative contributions of 5Fu treatment and C . albicans in oral bacterial changes we asked whether strains that stably colonize the oral mucosa of healthy mice could induce similar bacterial changes . We thus tested a clinical strain ( 529L ) , isolated from a patient with oral candidiasis [24] , which has been previously reported to colonize the oral cavity of healthy mice for over 5 weeks [25] . As seen in Fig 3A strain 529L colonized the tongues of immunocompetent mice at levels similar to strain SC5314 in mice receiving chemotherapy . In healthy mice this strain was found mainly in the superficial epithelial layers ( Fig 3C , upper panel ) and did not form a visible white biofilm ( not shown ) . We also noticed that this strain formed mostly pseudohyphae under hyphal-inducing conditions in vitro ( i . e . RPMI , 10%FBS ) . Despite colonizing the mucosa of healthy mice , this strain was not associated with an increase in viable bacterial counts ( Fig 3A ) and caused a lower increase in the enterococcal biomass above the untreated control , compared to other experimental groups ( Fig 3B ) . When combined with 5Fu , strain 529L was more invasive ( Fig 3C , lower panel ) and was associated with a significantly higher increase in enterococcal biomass compared to 529L alone , or 5Fu alone groups ( Fig 3B ) . Enterococci were also detected in sites of tissue invasion with this organism ( Fig 3C , lower panel ) . Although invasive infection was detected histologically , the majority of these mice did not form a visible biofilm lesion ( Fig 3C and 3D ) . These observations indicated that the increased virulence of this strain under conditions of 5Fu-induced immunosuppression and mucosal injury is associated with enterococcal changes . To confirm these findings we next tested the tup1Δ/Δ mutant which forms pseudohyphae and is avirulent in intestinal candidiasis models [17] . Importantly , this mutant acquires virulence in the GI tract with increased dissemination and mortality in a mouse model of immunosuppression combined with intestinal damage [17] . Similar to strain 529L , this strain colonized the tongues of healthy mice ( Figs 3A and 1C , upper right panel ) , and did not cause an increase in the viable bacterial biomass , but was associated with an enterococcal increase similar to 5Fu treatment alone ( Fig 3A ) . Consistent with the GI tract phenotype this strain formed tissue-invasive biofilms in 5Fu-treated mice [17] . Similar to other strains , this mutant co-localized with endogenous enterococci on tongue biofilms ( Fig 3C , lower right panel ) , which were visible as white plaques in some , but not all , mice ( Fig 3D ) . Consistent with the ability to cause invasive biofilms in 5Fu-treated mice , this strain was able to induce an increase in viable bacterial counts ( Fig 3A ) and a further enterococcal biomass increase , comparable to strain SC5314 under the same conditions ( Figs 2E and 3B ) . In summary these results show that C . albicans virulence in a host permissive environment is associated with further enterococcal increases in the oral mucosa . Since the 5Fu-induced host permissive environment for C . albicans entails both mucosal injury and neutropenia we next examined their relative contributions . We first asked whether mucosal injury alone promotes bacterial , fungal or mixed biofilm growth and invasion , using our published organotypic model of 5Fu-induced mucosal toxicity [26] . In this model , oral mucosal constructs were infected with the tup1Δ/Δ mutant and an E . faecalis isolate from mice with 5Fu-associated OPC . As seen in Fig 4 , pre-treatment with 5Fu did not significantly affect enterococcal or fungal biofilm growth on the mucosal surface , arguing against a direct role of mucosal injury in promoting biofilms , similar to observations with other bacterial species in this model [26] . However , 5Fu pre-treatment increased invasion of the tup1Δ/Δ mutant ( Fig 4 , right panel ) . Growth of this strain with E . faecalis in biofilms promoted invasion into the submucosal compartment of untreated and 5Fu-treated tissues . Finally we asked whether immunosuppression alone , in the absence of mucosal injury , was sufficient to induce bacterial biomass and enterococcal changes on the oral mucosa . For this we used the cortisone model , which is not associated with mucosal injury [27 , 9] . Cortisone treatment alone caused a small but statistically significant increase in the total bacterial biomass on mouse tongues ( Fig 5A ) . Although the total bacterial biomass increased , the enterococcal biomass decreased up to 40% compared to untreated control mice , suggesting that the total biomass increase was due to overgrowth of other bacterial species ( Fig 5A and 5B ) . Importantly , infection with C . albicans under this type of immunosuppression caused a further increase in total bacterial burdens , while enterococci rebounded at or slightly above untreated levels ( Fig 5A and 5B ) . Taken together these data show that while both types of immunosuppression increase oral bacterial burdens , they influence bacterial biodiversity in different ways . We also conclude that C . albicans infection favors the growth of enterococci under different immunosuppressive states , regardless of pre-existing mucosal injury . Since enterococci comprised the vast majority of the bacteria in the tongue mucosa of mice with candidiasis , co-localized with C . albicans in tissue invasive biofilms , and promoted invasion of the tup1Δ/Δ strain in the organotypic mucosa , we asked whether these bacteria are involved in augmenting fungal virulence . We thus hypothesized that an antibiotics regimen which depletes enterococci from the oral mucosa would attenuate fungal virulence . This regimen drastically reduced bacterial CFUs in all groups ( S4A Fig ) . No bacterial DNA amplicons were obtained from tongues in any antibiotics group and few isolates from the group with OPC receiving antibiotics were not enterococci , as determined by 16S sequencing ( S4A and S4B Fig ) . Treatment of immunocompetent mice with antibiotics increased C . albicans burdens on the tongue , as previously observed by others [reviewed in 28] , but did not lead to development of white mucosal biofilms or weight loss ( Fig 6A , 6B and 6C ) . In contrast , in 5Fu-treated mice inoculated with C . albicans antibiotics did not cause changes in fungal loads , or in the biofilm surface area , compared to mice not receiving antibiotics . Even though the biofilm surface area and fungal loads were similar in the two chemotherapy groups tissue invasion of C . albicans in mice receiving antibiotics was significantly attenuated ( Fig 7A and 7B ) . However , both chemotherapy groups had diarrhea and lost a significant amount of weight , consistent with an absence of a protective effect of antibiotics on the intestinal mucosa [17] ( Fig 6C ) . Our earlier work demonstrated that oral mucosal invasion of C . albicans is associated with E-cadherin degradation from epithelial adherens junctions [29] and that one way whereby commensal streptococci can augment tissue invasion is by synergistically activating host enzymatic pathways to degrade this protein [8] . Since antibiotics attenuated the invasive phenotype of C . albicans on the tongues we reasoned that they would attenuate E-cadherin degradation . As we showed previously in this model [19] , 5Fu treatment alone induced a significant loss of the E-cadherin signal in the oral mucosa ( Fig 7C and 7D ) . Invasive infection in 5Fu-treated mice was associated with almost complete E-cadherin dissolution , whereas protein integrity in adherens junctions was partially preserved with antibiotics treatment ( Fig 7C and 7D ) . In summary , these results suggest that the dysbiotic bacterial microbiota present in the oral mucosa of mice with OPC may contribute to the E-cadherin degradation and increased C . albicans invasion in the 5Fu model . One of E . faecalis virulence attributes which contributes to loss of intestinal barrier function , is secretion of the metalloproteinase GelE , involved in extracellular domain E-cadherin degradation [30] . Intestinal epithelial cell permeability by this enzyme was shown to be mediated by via protease activated receptor 2 ( PAR2 ) [31] . Although most E . faecalis strains have a copy of this gelatinase gene , expression varies significantly from strain to strain [32] . Thus we tested two E . faecalis isolates ( isolates #13 and #14 ) from chemotherapy-treated mice with OPC for gelE expression under standard growth conditions and asked whether conditioned media from these isolates could degrade recombinant E-cadherin . GelE was assessed at the transcript level and compared to strain OG1RF which expresses very high levels of this enzyme [31 , 30] . Both oral isolates expressed this gelatinase , albeit at much lower levels than strain OG1RF , consistent with other reports on other E . faecalis isolates ( Fig 8A ) [30 , 32] . In accordance with GelE expression levels , recombinant E-cadherin was degraded by conditioned media from E . faecalis isolates in variable degrees , whereas strain OG1RF completely degraded this protein . Treatment of recombinant E-cadherin with conditioned media from E . faecalis isolates and C . albicans together , further degraded this protein ( Fig 8B ) . In the next series of experiments we asked whether conditioned media of the two organisms alone or in combination could increase the permeability of oral epithelial cells in a transwell monolayer assay . Following pre-treatment of the cells with E . faecalis conditioned media we first measured the flux of FITC-labeled dextran , added on the upper chamber , across the monolayer . E . faecalis conditioned media promoted the permeability of the epithelial monolayer . Importantly , a PAR2 antagonist partially but significantly inhibited the permeability induced by E . faecalis , suggesting a role for gelatinase E in this process ( S5 Fig ) . Finally , we found that C . albicans translocation across the epithelial layer was greater in cells pre-treated with E . faecalis conditioned media and that this was attenuated by the PAR2 antagonist ( Fig 8C ) . Taken together these results support the hypothesis that E . faecalis at least partially contributes to mucosal barrier breach .
In this study we investigated the influence of C . albicans infection on the composition of the oral and intestinal mucosa-associated bacteria in the context of cytotoxic chemotherapy . We demonstrated that C . albicans infection led to a profound taxonomic imbalance on the oral mucosa that contributed to pathology . We also discovered that antibiotics that clear the dominant Enterococcus taxon during infection ameliorate invasive candidiasis , further demonstrating that bacterial community changes in OPC represent a dysbiotic shift promoting C . albicans virulence . Importantly , although antibiotics alone significantly increased oral fungal burdens , virulence required a chemotherapy-modulated host environment . Mucosal injury and immunosuppression caused by 5Fu played a major role in increased invasive infection in this model , with dysbiotic communities playing an accessory role . Thus our studies support a novel pathogenesis framework in the oral mucosa which includes the fungus , the resident bacterial microbiota and a host-permissive environment ( Fig 9 ) . One of the limitations of the 5Fu model is that bone marrow and mucosal toxicity are simultaneously occurring in the same host such that it is extremely difficult to dissect their independent contributions in dysbiosis . A comparison of the 5Fu and cortisone models showed that both types of immunosuppression allow overgrowth of endogenous bacterial organisms on the oral mucosa . However the two treatments had different effects on enterococci , with 5Fu promoting and cortisone curtailing their growth , supporting qualitatively different effects on biodiversity . This is not surprising since mice given cortisone have an increased number of functionally competent infiltrating neutrophils , whereas mice given cytotoxic chemotherapy are severely neutropenic [19 , 9 , 33 , 34] . Given the central role of neutrophils in the control of endogenous enterococcal overgrowth in mice [35] , the finding that a neutropenic state induced by 5Fu led to overgrowth of these organisms was anticipated . Alternatively , mucosal injury could be involved in the differences observed between the two types of immunosuppression . Although studies in the organotypic model did not support this alternative , the model lacks vascular and immune components and cannot fully recapitulate in vivo conditions . On the other hand , C . albicans infection increased growth of enterococci in both immunosuppressive states , suggesting a mutualistic relationship of the two organisms , independent of immunosuppression . This was further supported by the fact that healthy mice inoculated with C . albicans daily via the drinking water had an increase in the endogenous enterococcal biomass on the oral mucosa . We conclude that both immunosuppression and Candida play a direct role in bacterial changes but increased fungal burdens due to 5Fu treatment further amplify bacterial biomass changes . This is one of the few studies that compared the effects of C . albicans inoculation on oral and intestinal sites of the same mice , and the only study to perform comparative global microbiome analyses in health and immunosuppression . We found that in healthy mice inoculated with C . albicans bacterial community diversity decreased in the oral and increased in the small intestinal mucosa . This may reflect differences in the resident bacterial composition prior to inoculation , or differences in the interaction of C . albicans with distinct innate mucosal host environments . In contrast , in a disease-permissive host environment the most important driver of community changes was C . albicans since infection led to reduction in bacterial diversity and enterococcal expansion both in the oral and intestinal mucosa . A previous study of changes in the intestinal microbiota in response to 5Fu in mice used a single high dose intraperitoneal injection and analyzed stool and not intestinal mucosa-associated bacteria [36] , thus our analyses are not directly comparable . In our model emphasis was placed on mucosa-associated bacteria with the potential to form mucosal biofilms with C . albicans . Our focus on jejunum instead of other parts of the intestinal mucosa was due to the relevance of this site to mouse intestinal mucositis models [37] . Our taxonomic profiles showed a dominance of Enterococcus on the oral mucosa , with all of our isolates being E . faecalis . Like C . albicans , Enterococcus species are a major concern in critical care areas due to resistance to multiple antibiotics [38 , 39 , 40] . In the human oral cavity enterococci are generally considered transient commensals and carriage rates in healthy adults are below 10% . However , the oral carriage rate of Enterococcus species ( predominantly E . faecalis ) in patients with underlying systemic disease such as diabetes or cancer rises up to 60% [41 , 42 , 43 , 44 , 45] . These are also some of the most high-risk populations for OPC [reviewed in 46] . In fact , there is mounting evidence that C . albicans and Enterococci co-exist in human disease samples [reviewed in 47] . A large-scale retrospective analysis showed that E . faecalis was twice as likely to be isolated in Candida-positive sputum and even more likely in sepsis [48] . Other studies have shown E . faecalis and C . albicans co-isolated in >10% of root canal infections [49] and in >40% of human tongue mucosal lesions [50] . Whether , in addition to pathogenic synergy , this co-isolation reflects similar host adaptation strategies remains unclear . The only published model of cytotoxic chemotherapy that assessed oral and gastrointestinal candidiasis in the same mice included wide spectrum antibiotics , which precluded assessment of the role of candidiasis on bacterial communities or the role of bacteria in disease progression [51] . One of the known consequences of mucosal injury and neutropenia in chemotherapy models is systemic dissemination of C . albicans to distant organs through the gastrointestinal mucosa [17 , 18] . However , studies that examined this effect of chemotherapeutic agents also used antibiotics which by themselves reduce gastrointestinal mucosal barrier function [52 , 53] . Our model thus offered the opportunity to examine the effects of chemotherapy on Candida dissemination in mice with unperturbed commensal microbiota . As one regulator of mucosal barrier function E-cadherin is targeted by both mammalian and microbial metalloproteases [54] . In proof of principle experiments we tested the proteolytic activity of our isolates and showed their ability to partially degrade E-cadherin in vitro . Compromised epithelial junction integrity mediated by GelE activity is a key step in E . faecalis intestinal pathogenesis [30 , 55 , 56] . Despite the epithelial junction integrity consequences , these studies observed no substantial penetration of E . faecalis in the inner mucus layer of the intestinal mucosa , consistent the lack of oral mucosal invasion by enterococci in our model . Oral isolates from mice with OPC showed low levels of gelE expression , suggesting this enzyme does not play a major role in breaching the oral mucosal barrier . Activation of host metalloproteinases by dysbiotic bacteria may represent an alternative mechanism of synergistic mucosal barrier destruction [57] . In addition , high amounts of reactive oxygen metabolites produced by enterococci can amplify epithelial damage [58] . In intestinal pathogenesis models E . faecalis has also been shown to have proinflammatory consequences by signaling through TLR2 to induce IL-6 [59 , 60 , 61] or by increasing expression of TLR4 , which in theory would increase proinflammatory signals by Gram ( - ) bacteria [62] These proinflammatory attributes of E . faecalis may contribute to the increased proinflammatory cytokine production in the tongues of mice with OPC . In conclusion we identified distinct bacterial microbiota profiles associated with C . albicans infection in the context of cytotoxic chemotherapy . We also discovered that changes in the resident bacterial microbiota contribute to OPC pathogenesis , thus representing a dysbiotic state . This is the only study to experimentally and systematically dissect the role of C . albicans in shaping resident bacterial communities in an immunosuppressed host and the reciprocal role of mucosal bacteria in fungal pathogenesis .
All animal studies were performed in compliance with the federal regulations as described in the Animal Welfare Act ( AWA ) , the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health , and the guidelines of University of Connecticut Institutional Animal Use and Care Committee ( IACUC ) . All protocols used in this study were approved by the IACUC committee of UCONN Health , IACUC protocol #101640–0720 . C . albicans SC5314 ( kindly provided by Dr . Aaron Mitchell ) , is a laboratory strain originally isolated from a patient with bloodstream infection [63] . C . albicans strain 529L , is a strain originally isolated from a patient with oral candidiasis ( kindly provided by Dr . Marc Swidergall , UCLA ) . C . albicans tup1Δ/Δ homozygous deletion mutant is from the SC5314 parental background ( kindly provided by Dr . Alexander Johnson , UCSF ) . Candida strains were routinely maintained in yeast extract-peptone-dextrose ( YPD , BD Difco ) agar and overnight stationary phase cultures were prepared in YPD broth prior to each experiment [64] . E . faecalis OG1RF ( kindly provided by Dr . Margaret Vickerman , SUNY Buffalo ) , and mouse bacterial isolates were cultivated to early log phase in brain heart infusion medium ( BHI , BD Difco ) at 37°C under aerobic conditions , prior to each experiment . We used an intravenous chemotherapy mouse model that recapitulates mucosal and bone marrow changes in cancer patients receiving 5-fluorouracil [19] . In this model , we tested the ability of C . albicans supplied in the drinking water to cause oral , gastrointestinal and disseminated candidiasis ( animal protocol 101640–0720 ) . Six to nine-week-old female C57BL/6 mice ( Jackson Labs ) were used ( animal protocol 101640–0720 ) . Mice received 50 mg/kg 5Fu ( Sigma ) , intravenously ( IV , via lateral tail vein ) every 48 hours , for 8 days . This mode of administration was previously optimized to induce oral and small intestinal mucositis recapitulating cancer chemotherapy-associated mucositis in humans [19 , 65] . Control groups received PBS IV . Mice were inoculated with an overnight C . albicans suspension culture added daily in the drinking water ( 6 × 106 yeast/ml of water ) . Chemotherapy-naive mice were inoculated with C . albicans strain 529L according to a sublingual protocol that allows stable colonization in a commensal state [25] and a suspension culture was also added daily in the drinking water . In some experiments 5Fu-treated mice received a combination of three antibiotics ( Penicillin 1 . 5 mg/mL , Streptomycin 2 mg/mL and Gentamicin 0 . 1 mg/ml ) in their drinking water , starting three days prior to fungal inoculation and continuing throughout the experimental period . For cortisone immunosuppression mice received two subcutaneous injections with cortisone acetate ( 225 mg kg−1 ) on the first and third day of the experiment . On the second day mice were anaesthetized by an intramuscular injection of ketamine: xylazine ( 90–100 and 10 mg kg−1 of body weight , respectively ) and a small cotton pad soaked with 100 μl of a C . albicans SC5314 cell suspension ( 6 × 108 yeast ml−1 ) was placed under the tongue for 2 hours . During the following 5 days animals were also given drinking water containing a daily-fresh suspension of C . albicans as above . Organs were retrieved after sacrifice at each time point . Experiments were repeated at least twice with a minimum of 4 mice per group . DNA extracted from tissues was quantified using the Quant-iT PicoGreen kit ( Invitrogen ) . 16S rRNA genes were amplified in triplicate using 30ng of extracted DNA as template . The V4 region was amplified using 515F and 806R primers with Illumina adapters and bar codes on the 3’ end [71] . PCR products were pooled for quantification and visualization using the QIAxcel DNA Fast Analysis kit ( Qiagen ) . Pooled PCR products were processed using the Mag-Bind RxnPure Plus kit ( Omega Bio-tek ) according to the manufacturer’s protocol , to include only sequences between 250–400 bp . The cleaned pool was sequenced on the MiSeq using v2 2x250 base pair kit ( Illumina ) . Sequences were processed following a standard pipeline [72] and classified using Mothur’s version of the Ribosomal Database Project classifier ( Mothur 1 . 39 . 5 ) [73] . For operational taxonomic unit ( OTU ) analyses , sequences were clustered using a 97% similarity cutoff and classified up to genus level based on the consensus taxonomy . Alpha and beta diversity statistics were calculated by subsampling 1000 reads per sample . Alpha-diversity was measured via Shannon diversity index , and community richness as average OTUs ( species ) . The relative abundance of OTUs of main genera was determined to assess the effect of treatments . NMS was used to observe community clusters associated with Bray-Curtis dissimilarity calculations in different tissues of the same mice undergoing different treatments . Association between Shannon diversity index and bacterial CFUs was tested using linear regression . NMS plots were used to survey bacterial OTU heterogeneity relative to each treatment at each time point and permutational ANOVA ( PERMANOVA ) comparisons of the Bray Curtis dissimilarity distances were performed . Community structures across groups were visualized in standard graphing packages within R , version 3 . 2 ( http://www . r-project . org ) . Isolates were selected on the basis of colony and cell morphology after Gram staining , and stored in 20% glycerol at -80°C . Isolates were identified at the genus level by sequencing , as described above . Species level identification was performed for Enterococcus isolates by PCR using species-specific primers as described elsewhere [74] . Organisms were visualized by immunofluorescence staining combined with FISH , as we describe in detail elsewhere [75] . Briefly , tissues were stained with a FITC-labeled anti-Candida polyclonal antibody ( Meridian Life Science ) , the oligonucleotide probe EUB338 labeled with Alexa 546 , targeting all bacteria [76] , the Enterococcus/Lactobacillus probe LAB158 labeled with Alexa 546 , and the E . faecalis probe ENFL84 labeled with Alexa 405 [77] . To evaluate the integrity of the mucosal barrier , E-cadherin was assessed by immunofluorescence staining using a polyclonal antibody ( BD Biosciences ) followed by a fluorescein isothiocyanate-conjugated secondary antibody ( DyLight 488; Vector Laboratories , Burlingame , CA ) as described previously [19] . Neutrophils were visualized in fresh frozen sections with a monoclonal antibody ( NIMP-R14 , Hycult ) , followed by a secondary antibody conjugated with Alexa 555 ( A-21434 , Invitrogen ) . Tissues were counter-stained with the nucleic acid stain Hoechst 33258 ( Invitrogen ) . Images were obtained using a Zeiss Axio Imager M1 microscope and an EC-Plan-Neofluar 920-NA 0 . 5 air-objective . ImageJ software was used to quantify signal intensity in a minimum of 5 sections/condition . Oral mucosal organotypic constructs that mimic non-keratinized stratified human oral mucosa have been described elsewhere [78] . Briefly , the constructs consist of SCC15 oral keratinocytes ( American Tissue Culture Collection , ATCC ) seeded on collagen type I–embedded fibroblasts ( 3T3 , ATCC ) . Tissues are airlifted for 2–3 weeks prior to infection . Tissues were inoculated with 106 cells of C . albicans SC5314 , 107 cells of bacteria , or a combination , and incubated for 20h . Pre-treatment of some tissues with 10 μM 5Fu overnight was performed prior to microbial inoculation as detailed elsewhere [26] . Prior to inoculation 5Fu was removed by washing tissues three times in PBS . For CFU determinations of invading organisms , superficially growing biofilms were removed by gentle washing with sterile PBS prior to tissue weighing , followed by homogenizing and plating . E . faecalis isolates from 5Fu-treated mice with candidiasis ( #13 , and #14 ) and E . faecalis strain OG1RF ( positive control ) [30] were grown overnight to stationary phase in BHI broth . Concentrated culture-conditioned media ( CCM ) were concentrated 25 times and 50μl of each were incubated with purified recombinant human E-cadherin ( 3μg ) ( Sigma ) at 37°C for 1h . As control , E-cadherin protein was incubated with BHI broth ( 50μl ) . Protein degradation was detected by Western blot using E-cadherin intercellular junction marker antibody ( ab40772 , Abcam ) and goat anti-rabbit IgG-HRP ( ab6721 , Abcam ) . E . faecalis cultures were grown overnight in BHI broth to OD = 1 . Bacterial pellets were harvested by centrifugation and washed with sterile PBS . RNA was isolated with the RNeasy mini kit ( Qiagen ) followed by DNAse digestion using the Turbo DNA free kit ( Invitrogen ) . Equal amounts of RNA were converted to cDNA using Superscript III first strand synthesis kit ( Invitrogen ) . Q-PCR was performed using CFX96- Real time system ( Biorad ) and data were normalized using recA as internal control . The primers used were as follows: recA: FP- GCAACGAAATGGTGGAACAG , RP- AAGGCATCGGCAATCTCTAAG [79] , gelE: FP- CGGAACATACTGCCGGTTTAGA , RP- TGGATTAGATGCACCCGAAAT [80] . SCC15 cell monolayers were seeded at 5 x104 cells/well in 0 . 2 ml of KSFM media on the upper chamber of the transwell ( Millicell 0 . 45 μm pore size , Millipore ) , and 0 . 5 ml of media were added in the bottom of the well . Epithelial cells were allowed to adhere and form a monolayer at 37°C in 5% CO2 . Next day , cells were exposed to PAR2 antagonist ( FSLLRY-NH2; Tocris , Bristol , United Kingdom ) for 24 hours , at a final concentration of 20 μM [31] . Concentrated conditioned media from E . faecalis isolates were prepared as described elsewhere [31] and added apically for 16 hours . Permeability was measured by fluorescein isothiocyanate ( FITC ) -labeled dextran ( 78 × 103 ng/ml ) added to the upper chamber , and monolayers were incubated for 4 hours . Fluorescence was measured in the lower chambers of the transwell plates using a fluorometer ( Synergy 2 , by BioTek with Gen5 Software ) at an excitation wavelength of 485 nm and an emission wavelength of 530 nm to determine the flux of FITC-dextran across the monolayer , based on a previously prepared standard curve . Candida albicans transmigration assay through the monolayer was performed under the same conditions , with a 5 . 0 μm pore size transwell insert ( Costar , Corning 3421 ) . Following treatments with PAR2 inhibitor and conditioned media from E . faecalis as above , transwells received 104 cells/well of C . albicans which were allowed to translocate to the lower chamber for 4 hours . C . albicans migration through the SCC15 monolayer was quantified by CFU counts from the lower chamber of the transwells . Statistical significance was determined by two-tailed t-test , assuming equal variances , or the Mann Whitney test when data were not normally distributed . | In cancer patients receiving high dose chemotherapy mucosal candidiasis is common and can lead to invasive , systemic fungal infection with mortality ranging 25–30% . We showed that in chemotherapy-immunosuppressed mice C . albicans induces a dysbiotic switch that favors growth of enterococci on mucosal surfaces , particularly on the oral mucosa . Overgrowth of these organisms led to enhanced oral mucosal barrier breach by C . albicans . Like C . albicans , Enterococcus species are a major concern in critical care patients due to their involvement in sepsis and resistance to multiple antibiotics . There is mounting evidence that C . albicans and Enterococci co-exist in human disease samples . Our discoveries provide the first experimental evidence of pathogenic synergy between these microorganisms in the context of dysbiosis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"enterococcus",
"infections",
"pathogens",
"drugs",
"microbiology",
"animal",
"models",
"bacterial",
"diseases",
"fungi",
"model",
"organisms",
"tongue",
"antibiotics"... | 2019 | Candida albicans induces mucosal bacterial dysbiosis that promotes invasive infection |
Cutaneous leishmaniasis ( CL ) is the most frequent form of leishmaniasis , with 0 . 7 to 1 . 2 million cases per year globally . However , the burden of CL is poorly documented in some regions . We carried out this review to synthesize knowledge on the epidemiological burden of CL in sub-Saharan Africa . We systematically searched PubMed , CABI Global health , Africa Index Medicus databases for publications on CL and its burden . There were no restrictions on language/publication date . Case series with less than ten patients , species identification studies , reviews , non-human , and non-CL focused studies were excluded . Findings were extracted and described . The review was conducted following PRISMA guidelines; the protocol was registered in PROSPERO ( 42016036272 ) . From 289 identified records , 54 met eligibility criteria and were included in the synthesis . CL was reported from 13 of the 48 sub-Saharan African countries ( 3 eastern , nine western and one from southern Africa ) . More than half of the records ( 30/54; 56% ) were from western Africa , notably Senegal , Burkina Faso and Mali . All studies were observational: 29 were descriptive case series ( total 13 , 257 cases ) , and 24 followed a cross-sectional design . The majority ( 78% ) of the studies were carried out before the year 2000 . Forty-two studies mentioned the parasite species , but was either assumed or attributed on the historical account . Regional differences in clinical manifestations were reported . We found high variability across methodologies , leading to difficulties to compare or combine data . The prevalence in hospital settings among suspected cases ranged between 0 . 1 and 14 . 2% . At the community level , CL prevalence varied widely between studies . Outbreaks of thousands of cases occurred in Ethiopia , Ghana , and Sudan . Polymorphism of CL in HIV-infected people is a concern . Key information gaps in CL burden here include population-based CL prevalence/incidence , risk factors , and its socio-economic burden . The evidence on CL epidemiology in sub-Saharan Africa is scanty . The CL frequency and severity are poorly identified . There is a need for population-based studies to define the CL burden better . Endemic countries should consider research and action to improve burden estimation and essential control measures including diagnosis and treatment capacity .
Cutaneous leishmaniasis ( CL ) is the most common clinical manifestation of leishmaniasis , a parasitic neglected tropical disease ( NTD ) [1] . Caused by an obligate intracellular protozoa from the Leishmania species and transmitted by the bite of Phlebotomine sand flies , the clinical presentations of CL include localized skin nodules ( often called oriental sores ) , diffuse non-ulcerated papules , dry or wet ulcers , and , in the mucocutaneous form , extensive mucosal destruction of nose , mouth , and throat . Transmission of CL may involve animal reservoir hosts ( e . g . , rodents , hyraxes ) in zoonotic foci , while anthroponotic CL ( where humans are the main parasite reservoir ) occurs in urban or periurban settings [2] . Environmental changes in rural contexts such as agricultural activities , irrigation , migration , and urbanization may increase the exposure risk for humans and result in epidemics . Likewise , outbreaks in densely populated cities or settlements have occurred , especially in conflict-affected zones such as Afghanistan or Syria [3 , 4] , in refugee camps and contexts of large-scale forced migration of populations . Globally , the World Health Organization ( WHO ) considers CL as endemic in 20 countries in the New World ( South and Central America ) and in 67 countries in the Old World ( southern Europe , Africa , the Middle East , parts of southwest Asia ) [5] . Between 700 , 000 to 1 , 200 , 000 CL cases are estimated to occur annually worldwide , with >70% of cases in 2014 reported from Afghanistan , Algeria , Brazil , Colombia , Costa Rica , Ethiopia , the Islamic Republic of Iran , Peru , Sudan and the Syrian Arab Republic [5 , 6] . Multiple parasite species cause CL: in the Old World , these are L . major , L . aethiopica , L . tropica , and , rarely , the viscerotropic L . donovani ( in Sudan ) , resembling similar a phenomenon more known for L . infantum [7–10] . Though CL is often considered self-healing , the duration varies for different species and can take months , or years [11] . Due to the clinical and epidemiological diversity in CL , its geographic clustering and lack of reliable surveillance data , estimating the CL burden are challenging [12] . The most widely used measure of disease burden known as the Disability Adjusted Life Year ( DALY ) combines estimated prevalence , incidence , and mortality , with an assigned disability weight for each disease [13] . However , the disability weights are defined using different approaches with regards to the expert panel composition , health state description , and valuation methods [14 , 15] . The specific stigma and psychosocial distress generated by a non-fatal condition are often overlooked [16 , 17] , although the social impact of CL is potentially severe and has been well-documented [18 , 19] . Moreover , in sub-Saharan Africa ( SSA ) , not only the disability but also the number of CL cases is largely underestimated . A recent global burden analysis listed 19 countries in SSA in the top 50 high burden countries [20] . The passive epidemiological surveillance system that prevails in these countries leads to the patchy data from this region . According to WHO , only Sudan and Ethiopia reported cases of CL [21] . The objective measures of burden such as prevalence and incidence of CL are scarce in this region , making it hard to advocate for funding and resources to tackle the disease . Whereas attention has been given to CL in Northern Africa ( Algeria , Libya , Morocco , Tunisia , Egypt ) and the Middle East [22–24] , data for sub-Saharan Africa is critically lacking , particularly in countries where CL is not a notifiable disease . This study focuses on SSA because it is a blind spot on the CL epidemiological burden map and the overall picture of what has been documented on CL is not known . We undertook a systematic review of the literature to synthesize current knowledge on CL burden in SSA .
We searched the following electronic databases: National Library of Medicine through Pubmed , Cochrane Register , Web of Science , CABGlobal Health , African Index Medicus and Google Scholar . We did an initial keyword search and subsequent searches based on Medical Subject Headings ( MeSH ) with various combinations of search terms “cutaneous leishman*” AND “Africa , South of the Sahara” ( which also included “Africa , Western”; “Africa , Eastern”; and “Africa , Southern” ) OR “Leishmaniasis , cutaneous” OR “Leishmaniasis , diffuse cutaneous” OR “Leishmaniasis , mucocutaneous” AND each individual sub-Saharan countries . The World Bank classification was used to define sub-Saharan African countries and to group them according to the region ( i . e . , southern , eastern , western , and middle Africa- see Box 1 ) . No language restrictions were set for searches , while we limited the publication date until 31 May 2018 . We hand-searched the reference lists of all recovered studies for additional references . We also explored and summarized information from the Global Health Observatory for leishmaniasis maintained by WHO for CL . We included studies if they are reporting primary data that help to determine the burden of CL in countries in SSA . The burden is defined as elements of 1 ) severity of the problem ( clinical , disability , case fatality , … ) in human patients; 2 ) frequency ( prevalence , incidence , … ) and 3 ) economic cost ( from patient , societal or health system perspective ) . We excluded animals or vector studies , studies on pathogenesis , immunology , histopathology , or on Leishmania species only , studies on diagnostic tests or treatment for CL and cases of Post Kala Azar Dermal Leishmaniasis ( PKDL ) –skin sequelae of VL . Case reports and case series of fewer than ten patients were also excluded . Sub-Saharan Africa as the main geographical interest refers to the settings where the studies were performed/conducted . Reviews about CL in a specific country or region without original data were excluded . The systematic review was conducted in line with PRISMA guidelines [25 , 26] . The review protocol was registered in PROSPERO , an international prospective register of systematic reviews , in July 2016 , number 42016036272 [27] . We selected the articles in a two-step process . In a first stage , titles and abstracts of all retrieved records were independently reviewed by two investigators ( TS and KV ) . In a second stage , the selected full-text articles were again reviewed ( by TS , KV , and a third person ) for eligibility . When full-text articles were excluded , the reason for exclusion was registered and reported . Any discordances were resolved through discussion or seeking consensus with a third investigator ( MB ) .
The data were extracted in parallel by two independent readers , using a specific data form , including information on the published record ( year , author ) , setting ( country ) , aim , study design , and main outcomes . We sought data on prevalence or incidence of CL among patients in health facilities and the community; demographic and clinical characteristics of CL patients , and the association between CL and other morbidities , notably Human Immunodeficiency Virus ( HIV ) . We attempted to use the STROBE checklist ( for reporting epidemiological studies ) to assess the ‘risk of bias , ’ but could not continue due to a large number of historical studies that are not in line with current reporting standards . The data analysis thus resulted in a narrative , qualitative synthesis of the included studies .
The flow diagram in Fig 1 shows the selection process: we identified 340 published articles , and after removing duplicates , we screened the title and abstracts of 289 articles , and exclude 184 . The full-text articles of the remaining 105 were assessed for eligibility , after which a further 51 were excluded . The remaining 54 articles were included . ( See Supporting Information 1 for all the included studies and the key information ) . The studies were published between 1955 and 2016; with only 12 ( 22% ) after 2010 . The studies were conducted in 13 out of the 48 countries in Sub-Saharan Africa: in eastern Africa ( Ethiopia , Kenya , Sudan ) , western Africa ( Burkina Faso , Cameroon , Chad , Ghana , Guinea , Niger , Nigeria , Mali , Senegal ) and southern Africa ( pre-independent Namibia ) . More than half of the studies were from western Africa ( 30/54 ) , notably Senegal ( 6 ) , Burkina Faso ( 5 ) and Mali ( 5 ) . Twenty-three studies studied CL in the community ( including three among school-children ) , and 28 used data collected in health facilities ( including 18 dermatology specialized services ) . The remaining three studies were mixed . All 54 studies were observational: 29 ( 54% ) were descriptive case series ( numbering a total of 13 , 257 cases ) , and 25 ( 46% ) followed a cross-sectional design , usually survey with various tools employed such as clinical screening or questionnaires . In eastern Africa , CL has been known for more than a century , with the first indigenous CL case recorded in 1911 in Sudan [28] . In Ethiopia , CL has been known since 1913 , and diffuse CL ( DCL ) clinical form was documented in 1960 in the highlands [29] . The first report of L . aethiopica as a distinct taxonomic entity was published in 1978 [30 , 31] , and since then , the species has also been found in the mountainous region of Kenya [32] . L . tropica was later reported from certain areas in Kenya during the 1990s , and since then considered to have a more restricted distribution than L . major [33 , 34] . In western Africa , only L . major has been thought to circulate in this region . The oldest case reports of CL come from Niger in 1911 [35] , then from Nigeria in 1924 , and from Senegal in 1933 [36] . Later more cases were reported from Cameroon , Mali , Mauritania , Burkina Faso and Guinea [37 , 38] . During the first half of the 20th century , the colonial medical officers documented sporadic case reports from an area that later became recognized as the ‘CL belt’ [38] . Several comprehensive ecological and epidemiological studies took place in suspected hyperendemic foci in Senegal [39–42] , Mali and Niger [43] . Current Namibia ( previously South West Africa ) , reported dozens of CL cases in the 1970s [44] , but the disease was not considered as a public health problem by the authorities [45] . Twelve studies ( Table 1 ) reported prevalence estimated by the Leishmanin Skin Test ( LST ) —also known as Montenegro test—to detect exposure to the parasites in CL foci . Through intradermal injection of Leishmania antigens , the induration is being read 48–72 hours later as a demonstration of a delayed type hypersensitivity reaction , much like a tuberculin skin test [11] . LST does not differentiate between past and present infection and not species specific , yet it is often used as a marker for cellular immunity against CL [46] . These studies were conducted at the community level in CL foci , and have shown fluctuation over time ( Table 1 ) . Changes from 4% to 91% in LST positivity rate were observed in the same villages following an outbreak in Sudan [47 , 48] . High variability across foci within one country has also been reported , for example in Ethiopia: in Ocholo , 57% of school children without CL lesions were LST positive [49] , while another study in the central-Ethiopian Rift Valley , LST positivity was maximum 5% . A study conducted in two neighboring villages in central Mali also demonstrated high variability: prevalence of Leishmania infection in Kemena was 45% , with the incidence of 19% and 17%; higher than Sougoula with 20% , 6% and 6% for the same years [50] . Reasons for these discrepancies are not known but possibly linked with hyper-clustering of reservoirs and vectors , leading to different intensity of peridomestic transmissions in Kemena [50] . A 2014 study from Mali complemented LST surveys with PCR and finger prick blood sample to measure antibody levels to sand fly saliva in endemic districts [54] . The results showed uneven prevalence of LST positivity across three different climatic areas ( 49 . 9% , 24 . 9% and 2 . 6% in Diema , Kolokani , and Kolondieba respectively ) , linked with north-south declining vector density . PCR was used to confirm L . major as the causative agent . LST positivity was also shown to be correlated to higher levels of antibodies to sand fly salivary proteins [54] . Across the studies , a consistent finding is that the proportion of positive LST increased with age and areas where CL transmission is active , at least a third of the population have had exposure to the Leishmania parasite [37 , 43 , 47–51 , 54–56] . Twenty-one studies reported estimates of CL prevalence or incidence; five were using medical records from hospitals , and the remaining were population estimates obtained through active screening for CL lesions and scars at the community level . All diagnosis was based on clinical examination . Though additional confirmatory methods ( microscopy/smear , histology , culture in NNN or combination of these ) were mentioned in all studies but two , it is unclear whether these were used in some or all or none of the patients . Among the five studies that were hospital-based , two used the number of dermatology consultations as the denominator , and the CL cases proportion found is 2% in Ouagadougou , Burkina Faso [57] and 14% in Addis , Ethiopia [58] . If suspected cases were to be denominator to calculate the CL cases proportion , they were found to be 78% ( 251/320 ) in Mali [59] and 93% ( 74/80 ) in Burkina Faso [60] . In most of the studies in the community , the prevalence of active CL was less than 5% . In endemic areas , the frequency of CL scars usually exceeds that of CL active lesions , except in a few special settings ( Table 2 ) . In Utut , Rift Valley in Kenya , a higher lesion versus scar rate ( 50% vs . 18% ) in migrant charcoal workers suggested a non-immune population’s encounter with the disease in an area where transmission occurs [34] . Also during an outbreak in a new focus in Silti , Ethiopia , the frequency of CL lesions was considerably more than that of CL scars [63] . In Sudan , 36% of the community were found to harbor active lesions during an outbreak [68] . To complement the findings from published studies , we also examined the data from the country official reporting system to WHO . The system record data from 1996 onwards , but clearly there are missing data ( Fig 2A and 2B ) . The absolute number of CL cases reported from eastern Africa is always higher than from western Africa , with Sudan bearing most of the burden . In western Africa , the number of cases reported from different countries is highly variable , and recurrent outbreaks were occurring in a 5–7 years cycle [74] . The increased cases in Ghana during 2002–2003 was prominent , yet there was a vacuum between 2007 and 2010 , and cases were reported again starting in 2011 . Other countries contribute little , with <100 cases per year ( Nigeria , Senegal ) . No data was reported from this region during 2015–2017 [75] . The majority ( n = 28 ) of the included records are clinical case series based on medical files from dermatology clinics or hospitals as the main data source . These studies describe a cohort of CL patients over a certain period , ranging from two to nine years . Chronologically , 10 studies reported CL cases in periods before 1980 [41 , 45 , 47 , 52 , 74 , 76–80] , 11 described patient groups observed between 1980–2000 [35 , 57 , 59 , 67 , 69 , 81–87] , and seven between 2000 and 2013 [58 , 60 , 88–92] . Hospitals reported that CL patients mainly came from surrounding areas or outside the cities or capital , such as Dakar , Senegal [74 , 88 , 93] or Niamey , Niger [84] . Eighteen studies report cases seen in specialized dermatology services . The proportion of CL cases among patients seen in those dermatology clinics is consistently less than 5% [59 , 69 , 94] . In the context of an outbreak , CL patients who seek care in specialized services represent only the tip of an iceberg , as shown in Burkina Faso ( further described below ) . Between 1999 and 2005 , a total of 7444 cases were recorded from various health centers in the capital Ouagadougou [95 , 96] , but during the same period , the dermatology hospital had only seen 251 CL cases [57] . Diagnosis in all the case series is obtained through clinical examination and smears or histopathology . In Chad , a hospital close to the Sudanese border reported a very high proportion of CL confirmed cases ( 580 out of 680 cases between 2008–2012 ) [89] . Three countries have published studies on CL outbreaks: Sudan , Ethiopia , and Ghana . The first ever epidemics in Sudan were reported in 1976–1977 along the Nile , in Shendi-Atbara north of Khartoum [68] , while the second and third outbreaks occurred in 1985 and 1986–1987 , respectively [97] . The last epidemic in Sudan was in Tuti island , and it affected at least 10 , 000 people in 7 months . Underestimation is likely mandatory reporting only started after the epidemic reached its peak [86] . People of both sexes , all age groups and all socio-economic classes were affected , which is suggestive of a disease ravaging in a non-immune population . The causal parasite was L . major LON-1 [98] and the outbreak was attributed to various factors such as immigration from west Sudan , the heavy rainfall in the year of the outbreak after a long period of drought—which led to increase in sandfly density as well as the rodent reservoir population—and waning of herd immunity of migrants from CL endemic areas in western Sudan ( Sayda el-Safi , personal communication ) . In Ethiopia , a CL outbreak occurred in 2005 in a district 150 km south of Addis . A survey then established an overall prevalence of 4 . 8% ( 92/1907 ) , and 1 in 5 cases had mucocutaneous lesions [63] . In Ghana , an outbreak of localized skin lesion consistent with CL occurred in Ho municipality , Volta region in 2003 [90] . The usual triggers of CL epidemics such as intrusion of humans into vector habitat through deforestation , road construction , wars or migration were not at work here . Previously , only one CL case had been reported from the country in 1999 , although the arid , Sahelian area of northern Ghana is considered to be part of the West African CL belt . Through passive case detection ( with biopsy as a confirmatory diagnosis ) with medical records review and active case finding , it was estimated that there were about 8876 CL cases between 2002 and 2003 in Ghana ( Fig 2A ) . All age groups were affected , and since then CL is considered endemic in this area . A study in the same district later found 60% parasite-confirmed cases among active CL suspects ( 41/68 ) . A phylogenetic analysis identified this Ghanaian parasite as new member of Leishmania enriettii complex , a possible new subgenus of pathogenic human Leishmania parasites [99] . Thirty-two studies described the clinical presentations of CL lesions . The most commonly used categories of the lesions are as followed: the localized CL or LCL , otherwise known as the classic oriental sore , refers to the lesion at the site of sand fly bites that may get ulcerated . LCL may appear as dry , papular forms with crust , or the wet , ulcerative forms with indurated edges . LCL can be singular or multifocal . When the nodules are multiple and nonulcerative , this is typically called a diffuse CL or DCL . In Sudan , mucosal leishmaniasis is described as lesion ( s ) that involves destructive mucosal inflammation which does not always start with a cutaneous lesion . This differs from New World mucocutaneous leishmaniasis ( MCL ) , which refers to a metastatic dissemination to the mucosal tissues starting from a distal cutaneous lesion [52 , 100] . Bacterial superinfection is common along with pain , itchiness , fever and the secondary inflammation often complicates clinical diagnosis [11 , 101] . The diagnosis documented in the medical files are often missing . A dermatology hospital in Addis , Ethiopia reported that among 234 confirmed CL cases , only 22% were categorized—consisting of 9% DCL , 10% MCL and 3% LCL [58] . The higher proportion of complicated or atypical lesions are frequently reported from teaching hospitals or specialized services . This includes sporotrichoid CL with painless subcutaneous nodules along the lymphatic vessels in Sudan [80 , 87] , or the diffuse CL in Ethiopia , which appear pseudo-lepromatous and can result in fungating or tumor-like lesions [52 , 80] . In the majority of the studies , the natural history of the lesions is only briefly described ( n = 51 ) . The duration between the first bite to lesion formation for LCL varied between 3–12 weeks [62 , 90] . Although CL can heal spontaneously , this seems to be dependent on the reported parasite species: L . major heals within approximately 2 to 12 months and L . tropica within 15 months , with a terminal scar appearing after about 24 months [11] . The description of diffuse CL caused by L . aethiopica suggests that it presents initially with nodules which do not heal or ulcerate but can metastasize widely [76] and are known to be very difficult to treat . In the case of DCL , spontaneous cure almost never happens . Mucocutaneous leishmaniasis is rare in Africa , but cases have been reported from Sudan and Ethiopia [52 , 80 , 100] . The lesions tend to be infiltrative and result in chronic edematous inflammation involving the lips , nose , buccal mucosa and larynx are . With regard to the locations of CL lesions , there appears to be a regional difference . CL lesions from eastern Africa are mostly found on the head ( i . e . , face including cheek , nose , forehead , ears , lips ) and less on the arms , legs or trunk , while from western Africa the highest proportion of lesions are on the upper and lower extremities . Amongst the 42 studies reporting the sex ratio of the patients ( Fig 3 ) , only 12 recorded more females than males affected [49 , 50 , 56 , 63 , 70 , 72 , 82 , 95 , 102] while the remaining described male preponderance , either due to hypothesized occupational exposure or males’ easier access to seek care in a health facility . Thirty-six out of the 54 studies reported the age of the CL cases: people of all ages are affected . However , when stratification according to age was reported , there is a broad tendency towards younger age groups ( between 10–30 years old . CL and HIV co-morbidities has been described in Burkina Faso [57 , 60 , 103] , Cameroon [70] , Mali [59] , and Ethiopia [91] , while sporadic cases have also been reported from Guinea , Ghana , Senegal , Nigeria , Ivory Coast and Sudan . Burkina Faso has recorded 13 . 5% ( 10/74 ) HIV positivity in a cohort of CL patients in 2000 , and another cohort of 32 CL/HIV patients was described in 2003–2004 [60 , 103] . Six out of 10 DCL cases in Ouagadougou were co-infected with HIV [57] . In Bamako , Mali , the prevalence of HIV among CL patients was 2 . 4% [59] . In Tigray , Ethiopia , a study reported an HIV prevalence of 5 . 6% , which increased to 8% two years later in 167 CL patients [92 , 104] . The only study reporting CL/HIV prevalence in the community was done in Cameroon in 2008 . Here , a total of 32 466 subjects were clinically screened , and amongst 146 active CL patients , seven ( 4 . 8% ) tested positive for HIV-1 and/or HIV-2 [70] . The consistent finding is that the clinical forms of CL are more diverse and complex in HIV co-infected patients , posing significant challenges in diagnosis and treatment . The lesions tend to be more severe: there are reports of infiltrative , leprosy-like , diffuse , psoriasis-like , verrucous , sporotrichoid , and angiomatous or Kaposi-like . Patients are more likely to have more than one lesion and more than one clinical forms [103] . Also , the time to lesion healing was longer in immunosuppressed individuals [70] , and particularly in atypical and severe CL patients with poor response to treatment [91] .
The epidemiological burden of cutaneous leishmaniasis in sub-Saharan Africa appears to be poorly documented . There is a paucity of robust evidence on prevalence and incidence on CL in this region . The diversity of CL epidemiological characteristics in endemic countries is not yet fully investigated . Nevertheless , the burden of CL morbidity remains important and most likely to be underestimated . Surveillance and mapping should be improved to mitigate outbreak risk and address dual co-infection with HIV . The current fragmented knowledge should be approached regionally , and awareness must be raised . In addition to population-based studies that better define the CL burden in sub-Saharan Africa , health systems should consider studies and action to improve CL essential diagnosis and care . | Cutaneous leishmaniasis ( CL ) is the most common form of this group of parasitic diseases , transmitted by sandflies . In sub-Saharan Africa , its extent of the problem is unknown , while elsewhere its disfigurement and stigma may cause a severe impact . This study systematically searched the literature to find evidence on the epidemiological data on human CL in this part of the world . Historically , CL has been present for decades in both western and eastern Africa , but unfortunately , in the last decades , the data are irregular and patchy . The estimated burden , relying on detected cases , may only capture part of the true number of cases . This article shows that there is insufficient evidence to have accurate figures; the diversity of the disease , along with poor surveillance have resulted in unprecedented CL outbreaks in the past . Many knowledge gaps remain , and we highlight the importance of improving the current fragmented knowledge by increasing commitments to tackle CL and conduct better population studies . CL in sub-Saharan Africa appears to be a blind spot and should not remain so . | [
"Abstract",
"Introduction",
"Methods",
"Data",
"extraction",
"and",
"synthesis",
"Results",
"Discussion"
] | [
"dermatology",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"geographical",
"locations",
"tropical",
"diseases",
"sudan",
"parasitic",
"diseases",
"signs",
"and",
"symptoms",
"global",
"health",
"neglected",
"tropical",
"disease... | 2018 | Uncharted territory of the epidemiological burden of cutaneous leishmaniasis in sub-Saharan Africa—A systematic review |
DNA double-strand breaks ( DSBs ) represent one of the most deleterious forms of DNA damage to a cell . In cancer therapy , induction of cell death by DNA DSBs by ionizing radiation ( IR ) and certain chemotherapies is thought to mediate the successful elimination of cancer cells . However , cancer cells often evolve to evade the cytotoxicity induced by DNA DSBs , thereby forming the basis for treatment resistance . As such , a better understanding of the DSB DNA damage response ( DSB–DDR ) pathway will facilitate the design of more effective strategies to overcome chemo- and radioresistance . To identify novel mechanisms that protect cells from the cytotoxic effects of DNA DSBs , we performed a forward genetic screen in zebrafish for recessive mutations that enhance the IR–induced apoptotic response . Here , we describe radiosensitizing mutation 7 ( rs7 ) , which causes a severe sensitivity of zebrafish embryonic neurons to IR–induced apoptosis and is required for the proper development of the central nervous system . The rs7 mutation disrupts the coding sequence of ccdc94 , a highly conserved gene that has no previous links to the DSB–DDR pathway . We demonstrate that Ccdc94 is a functional member of the Prp19 complex and that genetic knockdown of core members of this complex causes increased sensitivity to IR–induced apoptosis . We further show that Ccdc94 and the Prp19 complex protect cells from IR–induced apoptosis by repressing the expression of p53 mRNA . In summary , we have identified a new gene regulating a dosage-sensitive response to DNA DSBs during embryonic development . Future studies in human cancer cells will determine whether pharmacological inactivation of CCDC94 reduces the threshold of the cancer cell apoptotic response .
After cells undergo genotoxic stress , multiple DNA-damage response ( DDR ) pathways are essential for the faithful replication and transmission of chromosomes to subsequent generations . Depending on the type of lesion , different pathways are engaged to repair the DNA [1] . One of the most detrimental lesions to occur upon exposure to ionizing radiation ( IR ) and certain chemotherapies is the DNA double-stranded break ( DSB ) . Immediate cell cycle arrest following DNA DSBs plays a critical role in promoting efficient DNA repair before cells enter mitosis . When exposed to excessive amounts of DNA DSBs that overwhelm their repair machinery , cells that are competent to do so will undergo p53-dependent apoptosis [2] . While the exact events that determine how this decision is made are not well understood , it is clear that p53-mediated transcriptional induction of the BH3-only protein Puma is critical for IR-induced apoptosis [3]–[5] . Puma induction triggers the activation of Bax and Bak [6] leading to mitochondrial outer membrane permeabilization , release of apoptotic factors including cytochrome C , and activation of the Caspase cascade of proteolytic degradation . Once Caspases are activated , an irreversible program of cellular destruction ensues . Anti-apoptotic members of the Bcl-2 family of proteins , like Bcl-2 and Bcl-xL , can inhibit this process by binding and sequestering Puma ( and other BH3-only proteins ) to prevent activation of Bax/Bak . Consequently , mutations that lead to the overexpression of Bcl-2 or to the impairment of the p53 pathway play pivotal roles not only in the development and progression of cancer , but also in the resistance to chemo- and radiotherapy that develops in established tumors [2] , [7] . Interestingly , a number of genes with prominent functions in the DSB-DDR pathway are also required for normal development of the nervous system [8] . Ataxia-Telangiectasia ( A–T ) was one of the earliest recognized diseases that arise from defects in the DSB-DDR pathway and is characterized by severe ataxia , radiosensitivity , defective immune function , sterility and predisposition to cancer [9] . A–T is caused by homozygous recessive mutations in ataxia-telangiectasia mutated ( ATM ) [10] , a gene encoding a kinase that plays pivotal roles in sensing DNA DSBs and coordinating a complex cellular signaling response that mediates the commitment to undergo cell cycle arrest , DNA repair and apoptosis [11] . Developing neurons are highly proliferative and the associated increase in oxidative stress likely exposes them to excessive DNA damage , thereby explaining their unique sensitivity to defects in the DSB-DDR pathway [8] . Since p53-dependent apoptosis is a common consequence of excessive DNA damage in this tissue [2] , developing neurons are selectively dosage-sensitive to IR . Indeed , we and others have shown that the developing nervous system in zebrafish represents an excellent system to identify genes required for the DSB-DDR [3]–[4] , [12] . The DSB-DDR pathway is complex and remains incompletely understood [13] . ENU-based genetic screens provide an unbiased method for identifying new components of the DSB-DDR pathway whose inactivation could kill cells that have become resistant to DNA-DSB-inducing therapies . To date , there are no published accounts of forward genetic screens performed in vertebrate in vivo models designed to identify novel radioprotective genes . Here we describe a rapid , thirty-hour zebrafish screen to identify mutations that enhance apoptosis after exposure to moderate levels of IR . One of the mutants we identified from this screen , which we named radiosensitizing mutation 7 ( rs7 ) , causes a severe sensitivity of zebrafish embryonic neurons to IR-induced apoptosis and is required for the proper development of the central nervous system . Hypersensitivity to DNA-DSBs by rs7 arises from an increase in p53 mRNA expression and activity . We have mapped the rs7 mutation to an early stop codon within ccdc94 , a gene that encodes a protein with few known functions or informative domains but that is highly conserved from yeast to humans [14] . Using biochemical and genetic approaches , we demonstrate that Ccdc94 is a functional component of the Prp19 complex in vertebrate cells . Prp19 complex members have established roles in pre-mRNA splicing [15]–[16] and DNA repair [17] . We show that depleting components of this complex renders cells more sensitive to DNA damage because of inappropriately high p53 mRNA and protein levels , but this effect does not arise from a global splicing defect . In summary , by taking advantage of the powerful embryonic and genetic attributes of the zebrafish system , we have identified a new gene regulating a dosage-sensitive DSB-DDR pathway during development .
To discover novel radioprotective genes using the zebrafish genetic model , we first sought to identify an obvious bright-field phenotype in embryos that distinguishes different levels of apoptotic response to IR . High doses of IR , such as 15 Gy , administered to a transparent zebrafish embryo at 24 hours post-fertilization ( hpf ) cause extensive apoptosis in neural tissue resulting in the accumulation of opaque tissue in the head . This phenotype is very consistent and readily observable by bright-field microscopy by six hours post-IR ( hpIR , Figure 1A , arrowheads ) . We treated wild-type zebrafish embryos with different levels of IR and found that the opaque tissue in the head was not observed using doses less than or equal to 8 Gy . These data show that a threshold of IR exposure exists between 8 and 15 Gy in wild-type embryos that gives rise to the obvious opaque neural tissue phenotype . We then reasoned that any mutation that inactivates a radioprotective gene should sensitize the embryonic neural tissue such that exposure of embryos to 8 Gy would cause a phenotype reminiscent of 15 Gy . Based on this logic , we performed a recessive genetic screen using 8 Gy IR ( Figure S1 ) and identified a number of mutations that sensitize embryos to IR . The first mutation we characterized was rs7 ( Figure 1B ) because these mutants showed a pronounced response to 8 Gy IR . The rs7 mutation gives rise to a homozygous recessive radiosensitive phenotype . Embryos that are heterozygous for the rs7 mutation are phenotypically identical to homozygous wild-type embryos with reference to all of the assays performed in this study . To verify that the observed neural cell death in irradiated rs7 mutants was due to apoptosis , we fixed the embryos at six hpIR and performed immunofluorescence with an antibody to detect activated Caspase-3 ( Figure 1C ) . Wild-type embryos showed moderate levels of activated-Caspase-3-positive apoptotic cells in the brain and spinal cord in response to 8 Gy IR . By comparison , irradiated rs7 mutants showed a dramatic increase in cell death in the neural tissue when compared to wild-type irradiated embryos ( Figure 1C , arrowheads ) . We also occasionally noticed an increase in IR-induced apoptosis in the intermediate cell mass ( ICM ) where primitive hematopoietic tissue resides ( Figure 1C , arrows ) , but we remained focused on analyzing the consistent neural radiosensitive phenotype . Unirradiated rs7 mutant embryos exhibited a level of apoptosis in neural tissue that appeared similar to that of irradiated wild-type embryos indicating that the rs7 mutation also causes apoptosis in neural tissue independent from IR . For clarity , this will subsequently be referred to as “rs7-mediated neurodegeneration . ” Since radiosensitization is characterized by a multiplicative , rather than additive , effect on the response to IR , we questioned whether the enhanced apoptosis in the irradiated rs7 mutants represented a true radiosensitization or simply an addition of rs7-mediated neurodegeneration to the apoptosis that is normally induced by 8 Gy IR . To address this question , we quantified levels of fluorescence in each of the four experimental groups from Figure 1C and plotted the values in Figure 1D , normalizing the response of irradiated wild-type embryos to one . As suggested by Figure 1C , unirradiated rs7 mutants showed levels of fluorescence that were not significantly different from those of irradiated wild-type embryos . Notably , compared to irradiated wild-type embryos , irradiated rs7 mutants have a 95-fold increase in activated-Caspase-3 staining . These data show that rs7 is a bona fide radiosensitizing mutation . While identification of mutations like rs7 that cause neurodegeneration is a relatively common event in zebrafish ENU-mutagenesis screens [18]–[19] , we found that only 11% ( 4/36 ) of the neurodegenerative mutants we identified also gave rise to a radiosensitizing phenotype . Conversely , we have also identified radiosensitizing mutations that do not affect the survival of neural tissue at 30 hpf ( our unpublished data ) . Indeed , morpholino knockdown of known components of the DSB-DDR pathway also leads to radiosensitization of neural tissue with or without associated developmental neurodegeneration ( [20]–[21] , our unpublished observations ) . This suggests that sensitivity to IR is not a frequent consequence of compromised survival in neural cells . Genetic linkage analysis of the rs7 phenotype to 239 microsatellite markers revealed that markers z58296 and z7358 flanked a 2 . 1 centimorgan region on chromosome 2 that harbored the rs7 mutation ( Figure 2A ) . Only 40 genes were annotated between these markers in the available zebrafish genome assembly Zv7 ( http://www . ensembl . org/Danio_rerio ) , and none of these genes were implicated in the DSB-DDR response pathway . We reasoned that the gene harboring the rs7 mutation should be expressed in the brain and spinal cord because these tissues were selectively radiosensitized in rs7 mutants . Using published gene expression data for the 40 genes in the interval , we narrowed our list of candidates to six genes based on high levels of RNA expression in neural tissues ( http://zfin . org/cgi-bin/webdriver ? MIval=aa-pubview2 . apg&OID=ZDB-PUB-040907-1 ) . We sequenced five of these genes and found a premature stop codon in coiled-coil domain containing gene 94 ( ccdc94 , R125Stop , Figure 2B ) . The Ccdc94 gene is present in genomes from yeast to humans ( http://www . ensembl . org/Danio_rerio/Gene/Summary ? g=ENSDARG00000026185r=2:52746237-52760438 ) and has previously been shown to regulate pre-mRNA splicing [14] . While the zebrafish and human protein sequences share 67% overall identity ( Figure S2 ) , the first 175 amino acids are nearly an exact match ( 94% identity ) . The Ccdc94 protein contains three predicted coiled-coil domains ( http://www . ensembl . org/Danio_rerio/Transcript/Summary ? g=ENSDARG00000026185r=2:52746237-52760438t=ENSDART00000036813 ) , all of which would be either eliminated or disrupted by the R125Stop mutation , suggesting that the R125Stop mutation interrupts a highly conserved function of Ccdc94 . To determine if the R125Stop mutation caused the rs7 radiosensitivity phenotype , we tested whether wild-type ccdc94 mRNA could rescue the excessive IR-induced apoptosis in rs7 mutants . We injected one-cell stage wild-type embryos , or embryos derived from a cross between rs7 heterozygotes , with mRNAs encoding either zebrafish ccdc94 , human CCDC94 or egfp ( control ) . We irradiated the clutches with 8 Gy and found that both zebrafish ccdc94 and human CCDC94 mRNA rescued the rs7 bright-field radiosensitization phenotype ( Figure 2C ) . We also followed the development of rs7 mutant embryos in the absence of IR and found that the rs7 mutation causes major deterioration of neural tissue resulting in a small head and curled tail phenotype by day 2 ( Figure 2D , Figure S3 ) and death by the end of day 3 . To prove that this phenotype was also caused by the mutation in ccdc94 , we performed the rescue experiment described above , but instead of irradiating the embryos at 24 hpf , we allowed them to develop unperturbed until 48 hpf . Figure 2D demonstrates that both zebrafish ccdc94 and human CCDC94 mRNA rescued the rs7 morphological phenotype at 48 hpf . Ultimately , however , this transient rescue ( injected mRNA generally lasts for up to three days ) did not rescue these embryos from lethality , as they eventually succumbed to complications from developing edema and overall dysmorphic effects by day 6 ( data not shown ) . Nonetheless , these data demonstrate that the early embryonic radiosensitizing and neurodegenerative phenotypes of rs7 mutants are specifically due to the loss of a highly conserved function of Ccdc94 . To independently show that loss of the ccdc94 gene causes the rs7 phenotype , we knocked-down the endogenous ccdc94 in wild-type animals with an anti-sense translation-blocking morpholino . Figure 2E–2F demonstrates that ccdc94 knockdown radiosensitizes embryos as measured by whole-mount immunofluorescence to detect activated Caspase-3 . Since the ccdc94 knockdown only increases Caspase-3 activity by 34-fold ( compared to 95-fold in rs7 mutants , Figure 1D ) , the morpholino likely induces a partial knockdown of Ccdc94 . While the “curly-up” phenotype seen in the rs7 mutants ( Figure 1B , Figure 2C–2D ) is not obvious at 27 hpf in ccdc94 morphants , it becomes prominent by 2 days-post-fertilization ( data not shown ) . As such , we have confirmed by three independent assays that the rs7 phenotype is due to the effects of a recessive loss-of-function mutation in the ccdc94 gene . The anti-apoptotic oncoprotein Bcl-2 has been shown in a number of studies to confer cancer-cell resistance to IR and chemotherapy [22]–[24] . To determine whether the rs7 mutation could overcome bcl-2 overexpression and restore apoptosis after exposure to IR , we injected one-cell stage wild-type or rs7 mutant embryos with 5 pg of bcl-2 mRNA . Figure S4 shows that while 5 pg of bcl-2 mRNA completely abolishes all apoptosis in irradiated wild-type embryos , it cannot achieve the same response in rs7 mutant animals , which exhibit typical levels of IR-induced apoptosis despite overexpression of bcl-2 mRNA . However , since rs7 mutants injected with 5 pg of bcl-2 mRNA exhibit less apoptosis than control-injected rs7 mutants ( Figure S4 ) , we hypothesized that higher levels of bcl-2 expression would fully block rs7-mediated radiosensitivity . Indeed , injection of 50 pg of bcl-2 mRNA ( or mRNA encoding bcl-xL , another anti-apoptotic member of the Bcl-2 family ) was able to completely block the rs7-mediated radiosensitization . These experiments indicate that Ccdc94 is a dose-dependent modifier of the anti-apoptotic function of bcl-2 in a manner that is genetically upstream of bcl-2 in the DSB-DDR pathway . Since the only described function for Ccdc94 involved the regulation of splicing in yeast [14] , we investigated if it played a similar role in vertebrates . We sequenced the transcriptome of 30 hpf wild-type , rs7 siblings and rs7 mutants by Illumina RNA-Seq analysis and compared gene expression profiles . We found no obvious global differences in splicing since greater than 96% of genes showed no significant differences ( i . e . , p>0 . 05 ) in mRNA expression , and there was no evidence of changes to alternative splicing . Instead we found a remarkable increase in p53 mRNA levels that were 3 . 4 and 5 . 1-fold higher in rs7 mutants compared to rs7 siblings and wild-type embryos , respectively . These results were confirmed by quantitative real-time reverse transcriptase PCR ( qPCR ) and whole-mount in situ hybridization with a probe complementary to p53 mRNA ( Figure 3A , 3B ) . Interestingly , the high expression of p53 in both the neural and hematopoietic tissue in rs7 mutants ( Figure 3B ) mirrors the high expression of ccdc94 in these same tissues in wild-type embryos ( http://zfin . org/cgi-bin/webdriver ? MIval=aa-pubview2 . apg&OID=ZDB-PUB-040907-1 and data not shown ) . To determine if elevated p53 mRNA levels were likely due to either increased p53 transcription or enhanced mRNA stabilization , we measured levels of intronic p53 sequence as a representation of pre-mRNA . Indeed , rs7 mutants show a similar increase in p53 pre-mRNA ( Figure 3C ) suggesting that the elevated levels of p53 are likely due to increased transcription . Robu et al [25] have demonstrated the activation of a p53-dependent “general stress response” driven by morpholino off-target effects . While this pathway likely required the post-translational activation of p53 , rather than upregulation of p53 transcription , we were concerned that induction of full-length p53 mRNA in rs7 mutants could be a downstream feed-forward mechanism of the p53-dependent general stress response . We tested this possibility by performing the same experiments in a p53 mutant background ( p53e7/e7 , [26] ) to eliminate the transcriptional activity of the p53 protein . Analysis of rs7;p53 double mutants showed that p53 mRNA remains highly elevated in this context and is not simply an indirect consequence of a general p53-dependent stress response ( Figure 3D ) . We next sought to clarify whether the increased p53 mRNA in rs7 mutants was due to specific regulation of p53 transcription or a general activation of the DSB-DDR pathway . To answer this question , we first tested wild-type embryos for IR-mediated induction of p53 mRNA . We found that IR exposure in wild-type embryos leads to an increase in p53 mRNA expression ( Figure S5A ) that is modest compared to unirradiated rs7 mutants ( Figure 3A ) . We next analyzed the rs7 mutants for upregulation of genes that are known to be induced by IR-induced activation of E2F1 in a p53-independent manner , such as apaf1 , caspase7 , and p73 [27] . To measure p53-independent gene expression , we performed the experiment in the p53e7/e7 background . Figure S5B demonstrates that none of these genes are significantly upregulated by the rs7 mutation . These experiments suggest that the rs7 mutation specifically induces p53 expression through a selective mechanism that is not simply due to general activation of the DSB-DDR pathway . To determine whether the increased p53 mRNA in rs7 mutants translated to increased p53 protein , we analyzed protein levels in rs7 siblings and mutants by western analysis using a previously described antibody to zebrafish p53 ( ZFp53-9 . 1 , [28] ) . Indeed , we found that the rs7 mutation caused a dramatic increase in p53 protein compared to siblings ( Figure 3E ) . A previously characterized p53 morpholino [26] was included to demonstrate antibody specificity . RNA-Seq analysis showed that a reduction in Mdm2 expression was likely not contributing to increased p53 protein levels since mdm2 mRNA levels were 1 . 5- and 1 . 2-fold higher in rs7 mutants than in rs7 siblings and wild-type embryos , respectively . These experiments define a new role for Ccdc94 in embryonic development as a negative regulator of p53 mRNA and protein expression . We next questioned whether an increase in p53 protein activity accounts for the extreme radiosensitivity of the rs7 mutants to IR . Since p53-dependent puma induction is essential for IR-induced apoptosis in mammals and zebrafish [4]–[5] , [29] , we analyzed expression of puma mRNA as a measure of p53 transcriptional activity . As expected , after exposure to IR , puma expression is much stronger in rs7 mutants than siblings ( Figure 3F ) , and the expression of puma in irradiated mutants and siblings requires wild-type p53 ( data not shown ) . To validate the requirement for p53 in rs7-mediated radiosensitization , we tested the rs7 mutation in the p53 mutant background . We irradiated embryos at 24 hpf , analyzed them three hours later for apoptosis , and found that wild-type p53 is required to execute the rs7-mediated radiosensitivity ( Figure 3G ) . These experiments suggest that the rs7-dependent increase in p53 protein and IR-induced activity accounts for the rs7-mediated radiosensitivity of developing neural tissue . In the absence of IR , rs7 mutants show increased puma expression in neural tissue ( Figure 3F , arrowheads ) compared to siblings , a sign that increased p53 mRNA expression translates to an increase in p53 pro-apoptotic activity even in the absence of an exogenous DNA-damage signal . Overexpression of puma has been shown to have a potent pro-apoptotic effect in zebrafish embryos [3] , [12] , so we questioned whether this might contribute to the developing neurodegenerative phenotype that becomes strikingly evident by day 2 in rs7 mutants ( Figure 2D , Figure S3 ( arrowheads ) ) . We found that expression levels of puma were extremely high in rs7 mutants as measured at 32 hpf and 48 hpf by qPCR , and this aberrant expression was entirely dependent on the presence of wild-type p53 ( Figure S6A ) . In p53 mutants , where puma expression was absent , there was a marked reduction in neural cell death and an obvious improvement in brain development in rs7 mutants by brightfield microscopy ( Figure S6B , arrows ) . While loss of wild-type p53 prolonged the life of rs7 mutants by 2–3 days , it ultimately failed to rescue the developmental lethality ( data not shown ) indicating that Ccdc94 also has essential p53-independent roles in development . However , these data show that inactivation of p53 significantly rescues the developmental neurodegeneration in rs7 mutants , and the pro-apoptotic activity of Puma likely contributes to rs7-mediated neurodegeneration . To gain further insight into the molecular function of Ccdc94 in the DSB-DDR pathway , we performed a tandem-affinity purification followed by mass spectrometry ( TAP/MS ) of human CCDC94 in mammalian cells . Specifically , CCDC94 cDNA was cloned into the pGlue vector encoding a dual-affinity tag containing streptavidin-binding protein , calmodulin-binding protein , and the hemagglutinin epitope . Lines of 293T cells expressing low levels of the tagged-bait fusion proteins were generated , detergent-solubilized , subjected to two rounds of affinity purification , trypsinized and analyzed by liquid chromatography–tandem mass spectrometry . The top hit was CCDC94 ( Figure 4A ) which serves as a positive control for the analysis . Interestingly , some of the other highly significant hits ( PRP19 , CDC5L , PLRG1 , BCAS2 ) are core members of the PRP19 complex [30]–[32] which has previously been shown to be required for pre-mRNA splicing in yeast [15]–[16] and for DNA repair in both yeast and human cells [33]–[37] . As an independent validation that Ccdc94 interacts with Prp19 complex members , we analyzed whether ccdc94 could genetically interact with prp19 or plrg1 in vivo . We reasoned that targeting each gene by morpholino knockdown would allow us to titrate levels of each protein to best reveal a potential genetic interaction . We designed a splice-blocking morpholino targeting prp19 ( called prp19 e3i3 , Figure S7 ) and made use of a previously characterized translation-blocking plrg1 morpholino [38] . We injected the highest dose of each morpholino that gave rise to minimal apoptosis or developmental abnormalities . At 22 hpf , injection with ccdc94 or prp19 morpholino alone gave rise to mostly normally developing embryos whereas the plrg1 morpholino alone led to an accumulation of cell death in the central nervous system in about half of injected embryos with varying severity ( represented in Figure 4B and quantified in Figure 4D ) , similar to plrg1 mutants [39] and previously characterized morphants [38] . Strikingly , these same developmental abnormalities were highly abundant among embryos co-injected with ccdc94 morpholino and either prp19 or plrg1 morpholinos ( Figure 4B–4D ) . These experiments indicate that ccdc94 genetically interacts with both prp19 and plrg1 in vivo . To determine whether the radioprotective function of Ccdc94 is derived from its interaction with the Prp19 complex , we questioned whether loss of prp19 or plrg1 could phenocopy the rs7 mutation . We injected the prp19 morpholino into wild-type embryos and analyzed expression of p53 mRNA by qPCR . To ensure that we were analyzing full-length p53 mRNA and not a truncated isoform that has been shown to be non-specifically induced by morpholino off-target effects [25] , we used primers that were designed to amplify across exons 1 and 2 of the p53 mRNA . The ccdc94 morpholino and a mismatch morpholino were injected as positive and negative controls for induction of p53 mRNA , respectively . The prp19 morpholino independently caused a significant increase in p53 mRNA and pre-mRNA expression , similar to knockdown of ccdc94 ( Figure 5A–5B ) . Of note , upregulation of p53 transcripts in the ccdc94 and prp19 morphants was reduced compared to that observed in the rs7 and plrg1 mutants , likely due to incomplete knockdown by the morpholinos . We next took advantage of a plrg1 mutant zebrafish line [plrg1 ( hi3174aTg ) , [39]] . This line contains a retroviral insertion in intron 1 of the plrg1 gene that leads to severely reduced mRNA levels ( Figure S8A ) . Embryos that are homozygous for this retroviral insertion exhibit major developmental cell death in the central nervous system and usually die by the end of day two [39] . Injection of wild-type plrg1 mRNA completely rescued this developmental phenotype at 24 hpf ( Figure S8B ) . We analyzed p53 mRNA and pre-mRNA levels in the plrg1 mutants and found that , similar to the rs7 mutants , they express abnormally high levels of p53 mRNA and pre-mRNA ( Figure 5C–5D ) . We reasoned that the increase in p53 expression in response to knockdown or loss of prp19 and plrg1 , respectively , should lead to an increase in IR-induced apoptosis , similar to knockdown or loss of ccdc94 ( Figure 2E–2F and Figure 1C , respectively ) . Since the plrg1 ( hi3174aTg ) mutants have severe neurodegeneration that interfered with our ability to evaluate irradiated embryos for an increase in apoptosis , we elected to use the plrg1 morpholino [38] to titrate plrg1 expression levels . We first analyzed whether knockdown of prp19 and plrg1 would phenocopy the rs7-mediated increase in p53 protein expression seen in Figure 3E . Figure 5E shows that knockdown with either the prp19 e3i3 or plrg1 atg morpholinos caused an increase in p53 expression that was specifically rescued by overexpression of the respective wild-type mRNAs . Knockdown of either prp19 or plrg1 also sensitized the embryos to IR-induced apoptosis ( Figure 5F–5G ) . Together these results suggest that Ccdc94 is a component of the Prp19 complex which functions to protect proliferating embryonic neural cells from genotoxic stresses such as IR by modulating the levels of p53 mRNA expression ( Figure 5H ) .
We have previously shown that zebrafish embryonic neural tissue is an excellent model system to dissect the DSB-DDR pathway since it faithfully recapitulates many of the complex molecular signaling events elucidated in mammalian systems [4] , [12] . With the notion that conserved embryonic pathways are exploited by cancer cells to promote survival , we embarked on a unique approach to use an unbiased genetic screen in zebrafish to identify novel radioprotective genes in the DSB-DDR pathway . The goal of these studies is to better understand the role of the DSB-DDR pathway in development , knowledge that can be ultimately translated to improved therapies for cancer treatment . Our efforts led us to identify a novel radioprotective gene called ccdc94 . The Ccdc94 S . cerevisiae ortholog ( named Yju2 ) has previously been shown to interact with the Prp19 complex [14] , [30] , [32] and is required for the first catalytic step in pre-mRNA splicing in yeast [14] . Our experiments confirm that the Ccdc94/Prp19 complex interaction is conserved in vertebrate cells; however , we have identified a new function for this complex in repressing p53 mRNA expression during development . The Prp19 complex has a well-established role in pre-mRNA splicing in yeast [15]–[16] , but whether the complex has an equivalent role in vertebrates is not clear [17] . Our studies show that the Prp19 complex normally represses p53 mRNA expression in neural cells to promote cell survival . There is a large body of evidence documenting the regulation of p53 activity by post-translational mechanisms [40] but a relative dearth of studies addressing potential mechanisms by which p53 is regulated at the transcriptional level despite the fact that potent transcriptional up-regulation could conceivably overcome post-translational mechanisms that restrict p53 activity . Indeed , we found that the 9-fold up-regulation of p53 mRNA transcripts in rs7 mutants translates to increased p53 activity . To date , there is no evidence to suggest that the Prp19 complex directly regulates transcription . An analysis of down-regulated genes in the rs7 mutants could yield novel , direct mechanisms by which the p53 gene is transcriptionally regulated . The Prp19 complex has also been linked to DNA repair [17] . Indeed , the prp19 gene was originally identified in S . cerevisiae in a screen for radiosensitizing mutations [33] . The prp19 mutant identified in this screen , initially called xs9 , was subsequently renamed pso4-1 since it was shown to have much greater sensitivity to the DNA interstrand cross-linking ( ICL ) agent PAVA ( psoralen plus UVA light therapy ) than to IR [34] . Further analysis of this thermoconditional mutant revealed that Prp19 functions in repair of DNA ICLs in a pathway that appears to be genetically separable from its role in pre-mRNA splicing [33]–[34] , [36] . Studies of human PRP19 confirmed the role of the PRP19 complex in ICL repair [37] and also suggested a role for PRP19 in the repair of DNA DSBs [35] . Therefore , the increase in p53 expression resulting from loss of Prp19 complex gene expression could ultimately stem from a lack of DNA repair . The neural tissue is one of the most radiosensitive tissues in the early developing zebrafish embryo . One likely explanation arises from the supposition that developing neurons are already poised to undergo apoptosis if they fail to receive neurotrophic survival factors from their synaptic targets [41] . As such , readiness to undergo DNA-damage-induced apoptosis is a feature that embryonic neural tissue shares with many cancers [42] . The neural tissue is also one of the most highly proliferative tissues in the developing embryo ( our unpublished observations ) , another feature in common with cancer [43] . By exploiting the similarities between embryonic neural tissue and cancer cells , we have been able to use an unbiased genetic approach to identify novel and conserved genes involved in the DSB-DDR that represent potentially important targets in both neurodegenerative disease and radio-resistant cancers . Sensitivity to IR-induced apoptosis by the rs7 mutation appears to be restricted to neural tissue and possibly primitive hematopoietic tissue in the ICM ( Figure 1C ) . These areas are consistent with disproportionately high levels of ccdc94 mRNA expression during zebrafish embryonic development ( http://zfin . org/cgi-bin/webdriver ? MIval=aa-pubview2 . apg&OID=ZDB-PUB-040907-1 and data not shown ) as well as regions of high p53 mRNA expression ( Figure 3B ) in rs7 mutants . In general , tissues other than neural or ICM have modest expression of ccdc94 by comparison suggesting that Ccdc94-independent mechanisms evolved to regulate p53 expression in these other tissues . However , CCDC94 appears to be ubiquitously expressed in tissues of adult humans ( http://www . ncbi . nlm . nih . gov/UniGene/clust . cgi ? ORG=Hs&CID=21811 ) suggesting that it may have a broader role in the regulation of p53 expression beyond development . In conclusion , we have identified and analyzed the role of a novel radioprotective gene , ccdc94 , in the first genetic screen in a vertebrate system designed to identify radiosensitizing mutations in vivo . We found that in vertebrate cells Ccdc94 interacts with core members of the Prp19 complex both biochemically and genetically and that protection from IR-induced apoptosis by this complex is mediated by inhibition of p53 expression . Upregulation of p53 expression could potentially overcome mechanisms that evolve during cancer progression to restrict its protein activity . Future experiments will determine whether CCDC94 and PRP19 complex components will be useful targets for sensitizing p53 wild-type cancer cells to IR therapy .
All experiments involving zebrafish conformed to the regulatory standards and guidelines of the Dana-Farber Cancer Institute and University of Utah Institutional Animal Care and Use Committee . Zebrafish were maintained , mutagenized and bred as described [44] . Wild-type embryos and the rs7 mutant line were derived from the AB strain . The rs7 mutant line was outcrossed to wild-type AB fish seven times and all phenotypes described herein are representative of crosses between at least seventh generation rs7 heterozygous fish . The plrg1 ( hi3174aTg ) line was obtained from the Zebrafish International Resource Center ( http://zebrafish . org/zirc/home/guide . php ) . We have previously described the p53e7/e7 zebrafish line that carries a homozygous M214K mutation in the p53 coding sequence [26] . IR was administered with either a Cs-137 gamma irradiator ( Gammacell 1000 ) or an X-ray irradiator ( RadSource RS2000 ) . Irradiators were completely interchangeable such that 8 Gy of X-rays gave rise to identical embryonic phenotypes described in this study as 8 Gy of gamma rays . Zebrafish one-cell stage embryos were injected with 500 picoliters of mRNA or the indicated concentration of morpholino . In every experiment , total RNA or morpholino concentrations were kept constant through the addition of egfp mRNA or a mismatch morpholino , respectively . Morpholinos were designed and created by GeneTools , and sequences are listed in Text S1 . For mRNA microinjection , zebrafish cDNAs were sub-cloned into pCS2+ , and mRNA was made by 1 ) linearization of each construct with NotI , 2 ) SP6 Message Machine kit ( Ambion , AM1340 ) and 3 ) purification for microinjection with NucAway Spin Columns ( Ambion , AM10070 ) . For morpholino rescue experiments , the plrg1 coding sequence was mutated to prevent direct binding of plrg1 mRNA to the translation-blocking plrg1 atg morpholino while creating only silent changes in regard to Plrg1 amino acid sequence ( relevant primers are listed in Text S1 ) . For in situ hybridization , zebrafish cDNAs were sub-cloned into the pGEM-T-Easy ( Promega ) . The pGEM-puma vector and generation of antisense probe was described previously [46] . The full-length zebrafish tp53 coding sequence was amplified with primers based on the published GenBank sequence ( NM_131327 and Text S1 ) . Full-length antisense and sense p53 RNA was generated by digesting with Apa1 and EcoRI , respectively , and transcribed in vitro with Sp6 and T3 polymerase , respectively . Embryos were dechorionated , staged , and fixed overnight in 4% paraformaldehyde at 4°C . Fixed embryos were washed in 1× PBST ( 1× PBS plus 0 . 1% Tween-20 ) at room temperature ( RT , 3×10 minutes ) and incubated in 100% methanol at −20°C for a minimum of two hours . Embryos were rehydrated in 1× PBST ( 3×10 minute washes at RT ) and incubated in Hyb-minus ( 50% formamide , 5× SSC , and 0 . 1% Tween-20 ) for one hour at 68°C , then transferred to Hyb-plus ( Hyb-minus , 5 mg/ml torula RNA type VI , 50 ug/ml heparin ) for three hours at 68°C . RNA probe was added to embryos in Hyb-plus at 1 ng/uL and incubated overnight at 68°C . Embryos were then subjected to 20-minute washes at 68°C in the following order: twice with 2× SSCT-formamide ( 2× SSC , 0 . 1% Tween-20 , 50% formamide ) , once with 2× SSCT , twice with 0 . 2× SSCT . At RT , embryos were then washed ( 3×10 minutes ) with MABT ( 100 mM maleic acid , 150 mM sodium chloride , 100 mM Tris pH 9 . 5 , 0 . 1% Tween-20 ) , then incubated in block [MABT , 2% Blocking Reagent ( Roche #11096176001 ) , 1% fetal bovine serum] at RT for one hour . Embryos were incubated overnight in 1∶5000 anti-digoxigenin-fluorescein Fab fragments ( Roche #11207741910 ) in block at 4°C . Embryos were then washed in 1× MABT ( 3×10 minutes at RT ) and 0 . 1 M Tris pH 9 . 5 ( 3×10 minutes at RT ) and stained with Vector BCIP/NBT alkaline phosphatase ( Roche #11697471001 ) . Upon completion of staining , embryos were washed in 1× PBST ( 3×10 minutes at RT ) to stop the reaction . Embryos were imaged in either 80% glycerol or 3% methylcellulose ( to allow for subsequent genotyping ) . Whole-mount activated Caspase-3 immunofluorescence was performed and quantified as described previously [12] . Quantitation was performed after immunofluorescence experiments by removing embryo tails at the level of the yolk . Tails were laid flat on a petri dish in 80% glycerol and activated Caspase-3 staining was documented by fluorescence microscopy using a Nikon Digital Sight DS-2MBWc black and white camera . All fluorescent pictures were taken at exactly the same exposure , gain , and magnification . Pictures were then cropped in Adobe Photoshop to include the same size region ( using the ruler function on Photoshop ) of the spinal cord using the end of the yolk extension as a reference point for all measurements , and to exclude any fluorescence arising outside of the spinal cord . Quantitation of fluorescence was performed with Volocity software by using the “Find Objects Using Intensity” option . The same exact parameters for eliminating the inclusion ( and therefore quantitation ) of background fluorescence were applied to all pictures equally . Six to twelve embryos from each group were included for all quantifications . GraphPad Prism software was used to plot the data , and error bars represent the standard error of averaged data from the embryos in a single experiment . Statistical analyses were performed using GraphPad Prism software using an unpaired student's T test . Representative quantitations from at least three experiments are shown for all data . Quantification represents measurements of fluorescence intensity which is directly related to Caspase-3 activity . However , fluorescence intensity is likely to fluctuate within cells . Therefore , changes in fluorescence intensity likely represent both increasing apoptotic cell number as well as increasing Caspase-3 activity within cells . The transcriptome of 30 hpf wild-type , rs7 siblings and rs7 mutants was analyzed by Illumina RNA–Seq analysis . Total RNA samples were prepared for sequencing using an Illumina mRNA Seq Sample Prep Kit and were sequenced using standard protocols on an Illumina Genome Analyzer IIx . Paired 36 base pair reads were obtained from the ends of each sequenced fragment . Reads were aligned to the Danio rerio Zv8 genome build ( release date December 2008 , obtained from UCSC Genome Bioinformatics , http://genome . ucsc . edu ) with the SOAPAligner software ( release 2 . 19 , http://soap . genomics . org . cn/index . html ) . The SOAPAligner was run to allow an insert range of 100 to 140 bp , as the mean library insert size was 120 bp . Sequencing and alignment of the samples yielded the following in terms of sample , read pairs , aligned reads , and percent aligned: 1 ) rs7 muts , 19 , 034 , 491 , 18 , 358 , 098 , 96 . 4% , 2 ) rs7 sibs , 20 , 036 , 828 , 19 , 206 , 760 , 95 . 9% , 3 ) wild-type , 16 , 604 , 110 , 12 , 630 , 153 , 76 . 1% . RNA–Seq data from the rs7 mutant , sibling , and wild-type samples was analyzed for differential expression and differential splicing using the DefinedRegionScanSeqs method of the USeq software [45] . Transcript coordinates for this analysis were collected from the UCSC Genome Browser database [46] . RNA was isolated from embryos ( 10–25 embryos/sample ) using the Qiagen RNeasy kit ( 74104 ) . One microgram of purified RNA was used to generate cDNA using the Invitrogen Thermoscript RT-PCR kit ( 11146-024 ) . For mRNA analysis , RNA was reverse transcribed using oligo-dT primers . For pre-mRNA analysis , RNA was reverse transcribed with random hexamers . cDNA was diluted 1∶20 in nuclease-free water and three technical replicates were analyzed using the LightCycler 480 Probes Master PCR mix ( 4707494001 ) and the Roche 480 Light Cycler or Eppendorf Realplex . Primers were designed by Roche to be used with their Universal Probe Library and are listed in Text S1 . Primers were designed to cross exons 1 and 2 of the p53 gene to ensure specific amplification of full-length p53 mRNA , and expression of the gapdh gene was analyzed to normalize p53 mRNA levels . Introns 4 and 9 of the p53 gene were analyzed by qPCR to determine levels of p53 pre-mRNA , and expression of 28S RNA ( which does not undergo splicing ) was analyzed to normalize p53 pre-mRNA levels . All data was averaged from at least three independent experiments . GraphPad Prism software was used to plot the data , and error bars represent the standard error of averaged data . Statistical analyses were performed using GraphPad Prism software using an unpaired student's T test . CCDC94 cDNA was cloned into the pGlue vector encoding a dual-affinity tag containing streptavidin-binding protein , calmodulin-binding protein , and the hemagglutinin epitope . Lines of 293T cells expressing low levels of the tagged-bait fusion proteins were generated , detergent-solubilized , subjected to two rounds of affinity purification , trypsinized , and analyzed by liquid chromatography–tandem mass spectrometry . Tryptic peptides were separated by reverse phase nano-HPLC using a nanoAquity UPLC system ( Waters Inc ) . Peptides were first trapped in a 2 cm trapping column ( 75 µm ID , C18 beads of 2 . 5 mm particle size , 200 Å pore size ) and then separated on a self-packed 25 cm column ( 75 µm ID , C18 beads of 2 . 5 mm particle size , 100 Å pore size ) at room temperature . The flow rate was 350 nl/min over a gradient of 5% buffer B ( 0 . 1% formic acid in acetonitrile ) to 40% buffer B in 200 minutes . The identity of the eluted peptides was determined with a Velos-Orbitrap mass spectrometer ( Thermo-Scientific ) . Specifically , following a FT full scan , MS2 spectral data were acquired by collision induced dissociation ( CID ) on the 9 most intense ions from the full scan , taking into account dynamic exclusion . The polysiloxane lock mass of 445 . 120030 was used throughout . All raw data were converted to mzXML format before a semi-tryptic search of the resultant spectra using Sequest and the Transproteomic Pipeline ( TPP ) on a Sorcerer 2 . 0 platform ( Sage N Research , Milpitas , CA ) . Approximately 100 30-hpf dechorionated embryos were rinsed three times at RT with 1× PBS . Ice-cold PBS was added , and embryos were incubated on ice for 5 minutes . Embryos were then de-yolked by pipetting 40 times through a thin bore plastic bulb pipet ( Samco 235 ) in ice-cold PBS . Tubes were returned to ice for two minutes to let embryo bodies sink to the bottom of the tube . Supernatant was removed and embryo bodies were washed three times in ice-cold PBS , pelleted by centrifugation ( 1 second at top speed ) , and lysed with RIPA buffer ( 1% Nonide P-40 , 0 . 5% sodium deoxycholate , 0 . 1% sodium dodecyl sulfate ) containing 1% protease inhibitors ( Sigma P8340 ) and 1% benzonase ( Novagen 70746-3 ) . BCA protein assay kit ( Thermo Scientific 23227 ) was used to determine protein concentration . Ten micrograms of protein was loaded onto a denaturing gel ( Novex NP0301BOX ) and transferred to pre-hydrated PVDF membrane ( GE PV4HYA0010 ) . Membrane was blocked in 10% non-fat milk diluted in tris-buffered saline plus 1% tween-20 ( TBST ) for anti-p53 antibody . Zebrafish p53 antibody was kindly provided by Dr . Lane at A*STAR , Singapore and used at 1∶30 in 10% milk with overnight incubation at 4°C . The blot was washed ( 4×5 minutes each ) in 1× TBST . Anti-mouse-horseradish peroxidase secondary antibody ( Cell Signaling 7076S ) was used at 1∶5000 in the same block for primary antibody . The blot was washed ( 4×5 minutes each ) in 1× TBST before chemiluminescent horseradish peroxidase substrate ( Millipore WBKLS0500 ) and film ( Thermo Scientific 34090 ) were used to detect signal . Blot was stripped for 45 minutes rotating at 60°C ( 62 . 5 mM Tris , 2% sodium dodecyl sulfate , 100 mM 2-mercaptoethanol ) and washed ( 8×15 minutes at RT ) in TBST . The blot was then re-blocked in 3% bovine serum albumin ( Amresco 0332 ) in TBST and probed using anti-GAPDH antibody ( Abcam 9484 ) at 1∶2000 or anti-α-Tubulin antibody ( Sigma T9026 ) at 1∶10 , 000 . The mouse secondary antibody and detection was performed as outlined above . Developed films were electronically scanned at 600 dots per inch and quantified using ImageJ software . | Radiation therapy and most chemotherapies elicit cancer cell death through the induction of excessive DNA damage . However , cancer cells can harbor genetic defects that confer resistance to these therapies . To identify cellular components whose targeted therapeutic inactivation could potentially enhance the sensitivity of treatment-resistant cancer cells to DNA–damaging therapies , we have chosen an unbiased genetic approach in live whole zebrafish embryos to identify genes that normally protect cells from the lethal effects of DNA damage . This approach has yielded the discovery of a novel radioprotective gene called ccdc94 . Upon inactivation of ccdc94 , cells become more sensitive to radiation-induced cell death . Our further analysis revealed that the Ccdc94 protein functions in the Prp19 complex , which is known to regulate gene expression and repair of damaged DNA . We found that this complex normally represses radiation-induced cell death by inhibiting the expression of the p53 gene , a critical mediator of DNA damage–induced cell death . Future experiments that inactivate Ccdc94 and Prp19 complex proteins in human cancer cells will determine if inactivation of this complex represents a novel therapeutic strategy that could increase p53 expression to enhance sensitivity to DNA damaging therapies in chemo- and radio-resistant cancer cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"model",
"organisms",
"genetics",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics",
"radiobiology"
] | 2012 | Ccdc94 Protects Cells from Ionizing Radiation by Inhibiting the Expression of p53 |
In vivo , antibiotics are often much less efficient than ex vivo and relapses can occur . The reasons for poor in vivo activity are still not completely understood . We have studied the fluoroquinolone antibiotic ciprofloxacin in an animal model for complicated Salmonellosis . High-dose ciprofloxacin treatment efficiently reduced pathogen loads in feces and most organs . However , the cecum draining lymph node ( cLN ) , the gut tissue , and the spleen retained surviving bacteria . In cLN , approximately 10%–20% of the bacteria remained viable . These phenotypically tolerant bacteria lodged mostly within CD103+CX3CR1−CD11c+ dendritic cells , remained genetically susceptible to ciprofloxacin , were sufficient to reinitiate infection after the end of the therapy , and displayed an extremely slow growth rate , as shown by mathematical analysis of infections with mixed inocula and segregative plasmid experiments . The slow growth was sufficient to explain recalcitrance to antibiotics treatment . Therefore , slow-growing antibiotic-tolerant bacteria lodged within dendritic cells can explain poor in vivo antibiotic activity and relapse . Administration of LPS or CpG , known elicitors of innate immune defense , reduced the loads of tolerant bacteria . Thus , manipulating innate immunity may augment the in vivo activity of antibiotics .
Antibiotics are of great importance for treating bacterial infections . However , the resistance of bacteria against antibiotics has remained a significant problem of global concern [1] . Resistance can be conferred by resistance determinants encoded in the pathogen's genome , as well as by “noninherited resistance” ( also termed “phenotypic tolerance” or “persistence”; [2] ) . Such tolerance is a phenotypic adaptation allowing survival of genotypically susceptible bacteria at antibiotic concentrations exceeding the “minimal inhibitory concentration” ( MIC ) . Although the molecular basis of phenotypic tolerance is still not entirely clear , the bacterial growth rate is often a cardinal factor [3] . Most ( if not all ) genetically susceptible bacteria are exquisitely susceptible during exponential growth , but display tolerance against diverse classes of antibiotics in the stationary phase [2] , [4] . Early hints and a growing body of anecdotal observations suggest that slow pathogen growth rates in vivo may explain why antibiotics therapy in vivo takes much longer and is much less efficient than predicted from ex vivo analysis of exponentially grown cultures [5]–[7] . To verify this hypothesis , we would need robust experimental systems quantifying the growth rates of tolerant bacteria in vivo . To study bacterial tolerance to antibiotics in vivo , we have chosen the pathogenic bacterium Salmonella enterica serovar Typhimurium ( S . Tm ) . In humans , the majority of cases develop “noncomplicated , ” self-limiting diarrhea , where the pathogen remains restricted to the gut lumen , gut tissue , and the gut-associated lymphatic tissue . However , in complicated cases ( i . e . , young children , elderly , immunocompromised patients ) , S . Tm spreads beyond the gut-draining lymph nodes and to systemic sites , thus causing a life-threatening infection . In these cases , antibiotics ( e . g . , fluoroquinolones like ciprofloxacin; typically two doses of approximately 7 mg/kg per day ) are used for therapy [8] , [9] . Fluoroquinolones are broad-spectrum gyrase inhibitors , interfere with bacterial DNA replication , enhance bacterial DNA fragmentation , and display bactericidal activity against many Gram-negative and Gram-positive bacteria [10] . However , in spite of exquisite in vitro activity ( within minutes to hours ) , and excellent tissue penetration characteristics of fluoroquinolones [11] , the in vivo activity is generally much lower , requiring treatment for at least 5–10 d with frequent relapses [12] , increased risks of long-term carriage [8] , and long-term persistence of the pathogen in blood and bone marrow [13] . Before the introduction of efficient antiretroviral therapy , AIDS patients displayed high susceptibility to complicated S . Typhimurium infection . Antibiotic therapy did relieve the symptoms . However , due to high rates of relapse , many AIDS patients underwent life-long antibiotic therapy ( [14] , [15] ) . Intriguingly , the bacteria re-isolated from relapses do generally remain genetically susceptible to the respective antibiotic . Thus , tolerance might play a significant role . It has remained enigmatic why the pathogen is tolerant to antibiotic therapy in vivo , how this may impact the pathogen–host interaction , and how tolerant bacteria can be eliminated .
We analyzed antibiotic tolerance of S . Tm in vivo in an intragastric mouse infection model using C57BL/6 mice , which are susceptible to enteropathy and systemic spread of the pathogen [16] . This model displays the typical features of complicated Salmonellosis—that is , diarrhea accompanied by pathogen spread to the cecal lymph node ( cLN ) and systemic sites . Without antibiotic treatment , the mice would succumb to systemic infection by wild-type S . Typhimurium within 5 to 6 d . In this model , therapy with two doses of 15 mg/kg ciprofloxacin per day is known to reduce fecal pathogen shedding and systemic pathogen loads . However , viable pathogens could be recovered from the cLN and relapse occurs at high frequency soon after the discontinuation of the therapy [17] . To establish the kinetics of pathogen clearance by ciprofloxacin , we infected mice with S . Tm ( strain SL1344 , 5×107 cfu by gavage ) and began high-dose ciprofloxacin treatment from day one postinfection ( 2×62 mg/kg per day by gavage ) . This high dosage of ciprofloxacin was chosen to achieve systemic antibiotic concentrations of ≥50-fold the MIC throughout the duration of the experiment ( MIC≤0 . 03 µg/ml as determined by the in vitro plating assay; Table S1 and Figure S1 ) and to ensure pathogen clearance from the gut lumen of all mice ( in spite of reduced gut luminal bioavailability—e . g . , through antibiotic absorption by food particles ) . This allowed us to focus on the tolerant bacteria in the host tissue . Within 3 h of ciprofloxacin treatment , the pathogen was cleared from the gut lumen and pathogen spread to the spleen was prevented by the antibiotic ( Figure 1A , B and Figure S2A ) . In contrast , ciprofloxacin did not clear the pathogen from the gut-draining cLN ( Figure S2B ) . cLN-loads declined with fast kinetics by 10-fold during the first 2 h after ciprofloxacin administration . This indicated that ciprofloxacin is bio-active at this site . After the first 2 h , the killing kinetics slowed down and 50–1 , 000 viable bacteria remained for up to 10 d ( Figure 1C , Figure S2B ) . Such bi-phasic killing is a typical feature if pathogens form a tolerant subpopulation [3] . Indeed , bi-phasic killing is also observed if S . Tm bacteria taken from the logarithmic growth phase are exposed to ciprofloxacin ( Figure S2C ) . However , in this case , only <0 . 1% of the bacteria survived after 6 h . This provided a first hint that pathogen–host interactions in vivo might affect poor pathogen elimination from the cLN ( Figure 1C ) . Subsequent experiments established that the cecal gut tissue and the cecal patch may also harbor S . Typhimurium cells surviving the antibiotic treatment ( Figure S3 ) . Please note that these numbers have to be interpreted with caution , as gut luminal contamination [i . e . , from relapses ( see , below ) or bacteria closely associated with the mucosal surface] cannot be completely ruled out . We also observed some antibiotic-tolerant bacteria in the spleens at 10 d after infection or even earlier if the ciprofloxacin treatment was started on days 2 or 3 postinfection ( Figure S4 ) . In that case , the starting loads in the spleen were equivalent , but the fraction of surviving S . Typhimurium cells was much lower than in the cLN ( 0 . 5%–5% as compared to ∼20%; compare Figure S4B with Figure S4A , Figure 1C ) . These data established that S . Typhimurium cells can survive antibiotic treatment at several sites within the mouse . The cLN was chosen for subsequent analyses , as this site harbors a high fraction of surviving bacteria , is easily retrievable without the risk of contamination , and allows fast and efficient tissue dissociation for FACS . In order to verify that the surviving S . Typhimurium cells were indeed “tolerant , ” we have performed a number of control experiments . Pharmacokinetic analysis showed that the ciprofloxacin concentration always remained ≥50-fold above the minimal concentration required for inhibiting Salmonella growth ex vivo ( MIC<0 . 03 µg/ml; Table S1 and Figure S1 ) . Furthermore , tissue culture infection experiments and confocal fluorescence microscopy verified that ciprofloxacin is indeed efficiently penetrating into infected host cells ( Figure S5 ) . In line with the fast initial killing rate in the cLN of an antibiotic-treated mouse , these data verified a high bioavailability of ciprofloxacin in the tissues of interest . Furthermore , pathogen survival was not attributable to genetic changes , as all re-isolated colonies tested remained ciprofloxacin sensitive ( same MIC as parental strain; Table S1 ) and fully virulent when transferred into naïve hosts ( Figure S6 ) . Thus , the cLN provides an environment that selects for or induces antibiotic-tolerant bacteria . Tolerant bacteria were observed in the cLN no matter when the treatment was started ( Figure S4 ) , or which mouse line or pathogen strain was used ( Figure S7 , nramp-positive 129SvEv mice; Figure S8 , S . Tm ATCC14028 ) . Thus , in our mouse model , the cLN are a site harboring phenotypically tolerant S . Tm . However , it had remained unclear if these bacteria were capable of causing a relapse , as observed a few days after the end of such an antibiotic treatment [17] . The capacity of tolerant bacteria from the cLN to cause a relapse-like infection was analyzed in cell transfer experiments . We have chosen the cLN population for this experiment , because it harbors a high density of tolerant cells and allows cell retrieval with minimal risk of contamination ( e . g . , from gut luminal contents ) . This cell transfer technique allowed us to focus on the bacteria located within the cLN , as all pathogens detected in the recipient animal must originate from this organ , while contributions of other bacterial populations ( e . g . , gut mucosa , cecal patch , spleen , or other , nonidentified organs ) , which might serve as additional reservoirs of tolerant bacteria , could be excluded . Donor mice ( C57BL/6 ) were infected with S . Tm as described in Figure 1 , while the acceptor mice remained uninfected during this initial phase of the experiment . One day later , high-dose ciprofloxacin therapy was applied to both groups . After 2 d of therapy , the infected mice were sacrificed and we recovered the cecum content , the cLN , and the spleens . Single cell suspensions were prepared from each sample and pathogen loads were verified by plating an aliquot . In line with the findings above , no bacteria were detected in the cecum contents or the spleens of the donor mice , while a total of ∼1 , 000 bacteria had survived therapy in the cLN . The remainder of the cell suspension was injected into the peritoneal cavity of noninfected recipient mice . As a control for viable , but nontolerant bacteria , we infected a fourth group of mice with S . Tm ( 500 cfu i . p . ) from an early log-phase culture . These bacteria should be highly susceptible to ciprofloxacin levels present in the recipient mice ( Figure S1 , Figure S9; [18] ) . After the cell transfer , the ciprofloxacin treatment of the recipients was discontinued so that antibiotic concentrations drop gradually ( Figure S1 ) . Four days later , recipient mice were sacrificed and pathogen loads in the respective organs were determined . Transfer of cLN cells leads to significantly higher infection rates compared to transferred spleen cells , cecum content , or S . Tm cultured in LB ( Fisher's exact test: p = 0 . 0048; Figure 2 , left part; Figure S10 ) . This is consistent with the high fraction of tolerant S . Typhimurium cells observed in the cLN ( Figure 1 ) . The loads in spleens and livers of the recipient mice were equivalent to those observed upon discontinuation of ciprofloxacin treatment in the donor mice ( “relapse control”; Figure 2 , right part; Figure S10; [17] ) . In contrast , cLN and gut luminal colonization were much lower in the recipient animals than in the relapse control . This may suggest that the site of re-seeding affects the colonization patterns during a relapse . In conclusion , the cell transfer experiments verified that the tolerant pathogens surviving in the cLN are indeed capable of re-initiating infection and can therefore account for relapses . However , it remained unclear whether the in vivo tolerance was linked to a particular intracellular niche of the cLN . To identify cellular niches harboring the antibiotic tolerant S . Tm , we performed microscopy and FACS . Initially , we focused on dendritic cells , as these are important host cells during intestinal infection that can engulf and transport viable bacteria within the gut and the associated lymphatic tissue [19]–[25] . CD11c-YFP mice [26] , which express a fluorescent marker protein in dendritic cells , were infected with S . Tm ( pRFP; red fluorescence ) for 1 d and treated ( or not ) with ciprofloxacin . In the cLN of untreated mice , the bacteria were located within dendritic cells ( CD11c+; 53% ) and other cells ( CD11c−; 47% ) , as detected by fluorescence microscopy . Upon ciprofloxacin therapy , most remaining bacteria were located within CD11c+ cells ( >80%; Figure 3A ) . It should be noted that this method did not allow us to establish if all detected bacteria were indeed still alive . Nevertheless , these data suggested that dendritic cells might represent a niche harboring tolerant bacteria . In a complementary approach , we have analyzed the cell types harboring viable S . Tm by fluorescence activated cell scanning ( FACS ) . As bacterial loads in the cLN of ciprofloxacin treated mice were very low compared to the numbers of host cells ( 102–2×103 cfu versus 107 cLN cells; Figure 1 , Figure S4 ) , the background noise of standard FACS was too high for a reliable analysis . Therefore , we have developed a FACS sorting protocol with subsequent plating of the sorted pools ( Materials and Methods ) . This allowed highly sensitive detection of viable bacterial cells within subpopulations of cLN cells . For the experiment , we used CX3CR1gfp/+ transgenic mice that express gfp in particular monocyte populations [27] . These animals were infected for 3 d with S . Tm and treated ( or not ) during the last day with ciprofloxacin ( 2×62 mg/kg by gavage ) . Our data above indicated that about 1 , 000 S . Tm cells survive the antibiotic treatment in the cLN . The cLN was harvested , dissociated , and stained as described in Materials and Methods . Gating for CD45+CD3−CD19− cells allowed us to focus on two particular cLN cell populations , the CD103−CX3CR1+ interstitial dendritic cells ( iDCs ) and the “classical” CD103+CX3CR1−CD11c+ dendritic cells ( cDCs ) , which transport antigens from the mucosa to the cLN and are thought to be key antigen-presenting cells of the gut-associated immune system ( Figure 3B; [28] , [29] ) . Ciprofloxacin treatment affected the frequency of iDCs and cDCs in the cLN by no more than 2-fold ( Figure 3C , left panel ) . The same was true for the S . Tm loads of the cDC ( Figure 3C , right panel ) . In contrast , the S . Tm loads in the iDC plummeted by >10-fold in the ciprofloxacin-treated mice . These data are consistent with the microscopic analysis and indicated that cDC represent a key reservoir for tolerant S . Tm in the cLN of antibiotic-treated mice . The S . Tm population localized in cDC seems to feature a particularly high fraction of tolerant bacteria . In order to assess the importance of the dendritic cells as a niche for tolerant S . Tm by independent means , we manipulated the number of dendritic cells via two well-established experimental systems . In the first experiment , dendritic cell numbers were reduced . For this purpose , we used a transgenic mouse ( CD11c-DTR+/−; [30] ) that allows dendritic cell depletion via diphtheria toxin ( DTX; Materials and Methods ) . These mice ( or nontransgenic controls ) were infected with S . Tm , treated for 2 d with ciprofloxacin , and dendritic cells were depleted during the last day of the therapy . This reduced the load of S . Tm surviving ciprofloxacin treatment in the cLN ( p<0 . 05; Figure 3D , left side ) . Conversely , increasing dendritic cell numbers by treating with Flt-3 ligand , a well-characterized agent expanding the dendritic cell population ( i . e . , cDC [31] ) , resulted in elevated S . Tm loads in the cLNs of ciprofloxacin-treated mice ( Figure 3D , right side ) . These data confirmed dendritic cells as an important niche for antibiotic-tolerant bacteria in the cLN . It had remained unclear how tolerant S . Tm cells do emerge . Is there a preexisting tolerant fraction of S . Tm cells that is enriched in the host's dendritic cells ? Alternatively , the intracellular environment within a host cell might induce slow growth and tolerance in a fraction of the phagocytosed bacteria . In a first approach to address this question , we have performed in vitro infection experiments in bone-marrow–derived dendritic cells ( BMDCs ) and analyzed the fraction of S . Tm cells surviving ciprofloxacin treatment . We have compared the in vitro killing kinetics of the internalized bacteria with those of the inoculum . Bacteria residing within BMDCs displayed a higher fraction of tolerant cells than the bacteria of the inoculum ( ∼3% versus ∼0 . 06%; Figure S11 ) . This may suggest that physiological changes during phagocytosis and/or phagosome maturation may induce tolerance . Alternatively , the BMDC may “enrich” tolerant bacteria from the inoculum—for example , by killing fast-growing intracellular S . Tm populations or by cell death of those BMDCs phagocytosing fast-growing bacteria . Anyway , our observations are in line with earlier work detecting slow-growing S . Tm cells during ex vivo infection of macrophages and BMDCs [32] , [33]–[36] . However , it had remained unclear if this would also occur in vivo . We hypothesized that a slow in vivo growth rate may explain the pathogen's antibiotic tolerance in the cLN . This seemed plausible as earlier work had shown that many bacterial species can become antibiotic tolerant in the stationary phase when bacterial growth is extremely slow [1]–[7] , [18] , [37] , [38] . Moreover , tolerant Salmonella populations were indeed detectable in fluoroquinolone-treated mice ( [17] , [39]; this work ) . Though suggestive , none of these studies was able to establish that tolerant pathogen populations in an infected host were indeed growing slowly . Therefore , we needed evidence linking pathogen growth rates in vivo to the observed recalcitrance to antibiotic treatment . Determining the replication rate of bacteria in vivo constitutes a challenge . Simply counting the number of bacteria in an anatomical compartment does not allow disentangling replication , migration , and clearance [40] . However , parameters characterizing these processes can be estimated with a population dynamical analysis of infection experiments conducted with mixtures of differentially tagged strains . In theory , fluctuations in the proportions of the differentially tagged strains are indicative of population bottlenecks during migration between and replication within host compartments , and can be used to estimate the population dynamical parameters ( Materials and Methods ) . To that end , we devised a novel approach extending our recent work on the population dynamics of cLN colonization by S . Tm in the mouse model for complicated Salmonellosis [40] . The original method employed a defined mixture of differentially tagged , isogenic S . Tm strains—that is , wild-type S . Tm ( SL1344 S . Tm ) spiked with limiting amounts of phenotypically identical strains ( S . TmWITS ) . These strains are isogenic to wild-type S . Tm , except for a 40-nucleotide sequence , which serves as an identifiable neutral marker that can be quantified by real-time qPCR [40] , [41] . The inoculum contained a total of seven different S . TmWITS , each carrying a unique 40 nucleotide sequence ( mixed in a 1∶1∶1∶1∶1∶1∶1 ratio ) and a 20-fold excess of untagged wild-type S . Tm ( [40] , this work; Figure S12 ) . In our initial work , this approach was used to estimate the net bacterial replication rate in the cLN by analyzing the S . TmWITS infection data with a stochastic birth-death model extended by immigration [40] . The model predicted the fraction of strains that successfully migrated to the cLN and their population size as a function of the rate of immigration , μ , replication , r , and clearance , c ( Figure 4A ) . Fitting this model to our experimentally determined number of each tagged strain in the cLN , we could infer rates at which bacteria immigrate into the cLN ( 298 bacteria during the first 24 h p . i . ) and replicate therein . The validity and robustness of this approach was verified analyzing S . TmWITS infections in wild-type and knockout mice [40] . In our present analysis of the tolerant S . Tm cells in the cLN , we used these parameters to estimate the composition of the S . Tm population at the beginning of the ciprofloxacin treatment ( Figure 4A , B ) , and extended this approach to quantify the growth rate of the S . Tm cells surviving the ciprofloxacin treatment in the cLN ( Figure 4A , Figure S12 , Materials and Methods , Text S1 ) . The population dynamics experiment had two phases . First , infection was established for 24 h with a mixture of S . TmWITS ( n = 35–49 data points per group; 7 differentially tagged S . TmWITS strains ) . Second , high-dose ciprofloxacin treatment was started at 24 h p . i . This rapidly diminished pathogen loads in the gut lumen ( Figure S2A ) , and thus , de novo immigration into the cLN can be neglected in our analysis of this second phase of the experiment . The number of each S . TmWITS in the cLN was determined experimentally at days 1 , 3 , 5 , and 10 postinfection ( Table S2 ) . From these data , we could determine both the doubling time and the clearance rate of the surviving bacteria in the cLN for each time period separately ( i . e . , days 1–3 , 3–5 , and 5–10; Figure S12 ) . During the first period of the treatment ( days 1–3 ) , susceptible bacteria were killed and the persistent bacteria remained in the lymph node . Thus , the population dynamics will represent the aggregate of both processes ( that is , the average over the dynamics of the two subpopulations , not the aggregate of replication and clearance ) , and the apparent in vivo doubling time was about 3 . 6 h . In the subsequent two periods , we could analyze the population dynamics of the tolerant S . Tm cells in isolation . Indeed , the doubling time increased to 8 . 8 h ( days 3–5 ) and 44 h between days 5 and 10 of the therapy ( Figure 4B , C ) . In this period , the clearance rate dropped to extremely low levels ( <0 . 01/h ) , indicating that the half-life of bacteria that survived the ciprofloxacin in the cLN exceeded 100 h . We do not know if S . Tm forms different types of tolerant subpopulations . This might explain why the doubling time and the half-life of the tolerant S . Tm cells increased between days 3 and 10 of the treatment . In any case , these data established that the antibiotic-tolerant bacteria enter a persistent state characterized by extremely slow replication and virtually no clearance . To confirm the slow growth rates of tolerant S . Tm cells in vivo , we employed a plasmid-dilution strategy ( Figure 5 , left side ) . pAM34 is an IPTG-addicted plasmid that stops replication as soon as IPTG is removed from the environment , including all mouse tissues analyzed ( pAM34 , AmpR , [42] ) . This type of plasmid can be used to verify the slow growth rate of the ciprofloxacin-tolerant S . Tm subpopulation . Such tolerant cells should survive the antibiotic and should feature a high fraction of cells retaining pAM34 . Three groups of mice were infected with S . Tm ( pAM34 ) . One group was sacrificed at day 2 p . i . to assess cLN loads and the fraction of bacteria still retaining the plasmid . Here we observed the typical pathogen loads of approximately 10 , 000 cfu in the cLN ( Figure 5; compare to Figure S4 ) . Approximately 100 of these bacteria were still carrying pAM34 ( Figure 5 , open symbols ) . The second group was left untreated , analyzed at day 3 p . i . , and served as a control . The third group was treated with ciprofloxacin from day 2 on and analyzed at day 3 p . i . Here , we observed significantly reduced total pathogen loads in the cLN ( 1 , 000–2 , 000 cfu; p<0 . 05 ) , while the total loads of pAM34 carrying S . Tm cells did not differ significantly from those at day 2 or the control mice at day 3 p . i . ( 50–100 bacteria; p≥0 . 05 ) . These findings are consistent with our data presented above and indicate that S . Tm does form a susceptible and a tolerant subpopulation in the cLN . Based on that data , we reasoned that approximately 80%–90% of the total cLN bacteria displayed the fast-growing/susceptible phenotype , while 10%–20% belonged to the slow-growing/tolerant phenotype . If slow-growing bacteria would exist in the cLN even in the absence of ciprofloxacin , most of the pAM34 should reside in such slow-growing/tolerant S . Tm cells at day 2 p . i . , before the onset of ciprofloxacin treatment . These should survive the ciprofloxacin treatment and thereby account for the high loads of pAM34-positive S . Tm at day 3 p . i . This is supported by the data from the group of mice infected for 2 d with S . Tm ( pAM34 ) and treated with ciprofloxacin during the third day of the infection ( Figure 5 ) . Here the susceptible S . Tm cells were eliminated , while tolerant cells prevailed and the number of bacteria retaining pAM34 did not drop significantly . This confirmed that the ciprofloxacin-tolerant cells were harboring the bulk of the pAM34 plasmid observed at day 2 p . i . and provided in vivo evidence that the slow-growing/tolerant subpopulation exists within the cLN ( presumably within classical dendritic cells ) even in the absence of ciprofloxacin . These findings are in line with the ex vivo data by Helaine and Holden , who observed by elegant fluorescence microscopy techniques that some fraction of internalized S . Typhimurium cells grow at a strikingly low rate [32] . From the pAM34 data of the ciprofloxacin-treated mice we could roughly estimate the doubling time of the tolerant bacteria . We assumed that the bacteria had reached the cLN by 12 h p . i . , resided at this site for 60 h , and displayed the tolerant , slow-growing phenotype throughout these 60 h . On this basis we could take the pAM34 frequencies of the inoculum and of the cLN resident bacteria at day 3 p . i . to estimate a doubling time of 8 . 9 h for the tolerant phenotype . This value is strikingly similar to that observed in our population dynamics experiments [8 . 8 h ( 7 . 1–11 . 8 h ) at days 3–5; Figure 4C] and further supports that tolerance is growing indeed quite slowly in vivo . A second phenomenon observed in the ciprofloxacin treatment was a reduced pro-inflammatory gene expression signature of the cLN ( reduced cxcl2 , ifng , s100a9 , lcn2 expression; Figure S13 ) . Overall , these data suggested that the antibiotic treatment may have two effects in parallel . First of all , it reduces the total bacterial burden in the cLN by killing the susceptible bacteria . This dampens the pro-inflammatory response in the tissue . Second , it might be of importance that the tolerant bacteria do not just survive the ciprofloxacin treatment . They may in fact face a reduced pro-inflammatory defense in the cLN ( Figure S13 ) . One might speculate that this further enhances the survival of tolerant S . Tm cells in the tissue . It will be an important task for future work to dissect which environmental signals do induce slow growth/tolerance by S . Tm and whether/how this is affected by pro-inflammatory signaling in the infected tissue . Finally , we were interested to identify a strategy for reducing loads of ciprofloxacin-tolerant pathogens in wild-type hosts . Earlier work had indicated that antibiotics and the host's immune system may cooperate in eliminating pathogens during antibiotic therapy [43] , [44] . Furthermore , the antibiotic-treated lymph node displays a signature of reduced innate immune defense ( Figure S13 ) . Based on such observations , it has been speculated that triggering innate responses during the antibiotic treatment may allow reducing the number of persistent bacteria [1] . However , to the best of our knowledge , this has not been substantiated by experimental data in vivo . To test this hypothesis , we applied a PRR ligand . Mice were infected for 1 d with S . Tm and treated with ciprofloxacin alone ( 2×62 mg/kg per day ) , LPS alone ( one dose of 5 µg at 48 h p . i . ) , or ciprofloxacin and LPS . Indeed , a single dose of LPS applied during the ciprofloxacin treatment elicited within 2 h an innate immune response in the cLN and significantly reduced the number of tolerant bacteria ( Figure 6 , Figure S13 , Figure S14 ) . In six animals the pathogen loads were reduced below the limit of detection ( red arrowhead; Figure 6 ) . Loads of tolerant bacteria were also reduced in the cecum wall and the cecal patch ( Figure S15 ) . However , it had remained unclear whether elimination of tolerant bacteria was attributable to direct or to indirect ( paracrine ) signaling to the infected host cell . In a first approach , we have generated mixed bone marrow chimeric mice . These mice displayed 50%–70% tlr4−/− cDC and 30%–50% wt cDC . Only the latter should directly respond to LPS , while both should respond to paracrine signaling . When these mice ( or wt control animals ) were infected with S . Tm ( pAM34 ) and treated with ciprofloxacin and LPS , we observed the same strong reduction of the cLN loads ( and pAM34 retention ) , as in wild-type mice ( Figure S16 ) . This suggested that LPS can affect tolerant S . Tm cells via an indirect , paracrine mechanism . A second PRR-ligand , CpG , was also capable of reducing the tissue load of tolerant S . Tm cells ( Figure 6 ) . These data demonstrated that the persistent bacteria lodged within the cLN dendritic cells are indeed susceptible to innate immune responses and that the PRR-agonists LPS and CpG can supplement antibiotic treatment .
In conclusion , high-dosed ciprofloxacin is capable of reducing S . Tm populations from most sites of the infected host . However , a fraction of the bacteria survives the treatment . In the cLN , approximately 10% survive in a persistent state and the “classical” cDCs seem to represent a niche harboring a very high fraction of ciprofloxacin-tolerant S . Tm cells . It is interesting to note that the cDCs play an important role in antigen sampling by the gut-associated immune system . Based on these observations , it is tempting to speculate that it is exactly the dendritic cells' function in adaptive mucosal immunity—that is , microbe transport to the draining lymph nodes and prolonged antigen presentation [25] , [45]–[47] , which might generate ( e . g . , induce or select for ) the tolerant S . Tm subpopulation . It remains to be shown whether this also holds true for human patients treated for complicated infections by Salmonella spp . or other gut pathogens . Such pathogen–host interactions and their effects on antibiotic therapy will be of broad importance for infection biology and represent an interesting topic for future work . The experimental framework developed using the infection model for complicated Salmonellosis could enable such studies . Considering the broad variety of pathogens , the diversity of niches within the host , the multitude of virulence factors , and the vastly differing average growth rates within hosts displayed by pathogens , it seems likely that such work will reveal pathogen-specific adaptations and common mechanisms explaining recalcitrance to antibiotic therapy . Clearly , deciphering the population dynamics of pathogens within their hosts will be of key importance for understanding the infection process and developing improved therapeutic strategies . It is still not entirely clear how bacteria can obtain an antibiotic-tolerant phenotype , whether different bacteria might employ different strategies , and how host–cell interactions might affect tolerance . Originally , a tolerant subpopulation was discovered in S . aureus [48] . Later , this phenotype has been observed in most bacterial species analyzed [3] , [49] . Bacterial populations of identical genotypes do generally feature a fraction of cells that are more resistant to a given antibiotic than the others . The size of this fraction can increase in response to external stimuli , including nutrient limitation , quorum sensing signals , antibiotics , or oxidative stress . This engages signaling mechanisms within the bacterial cell , including the SOS response and toxin–antitoxin systems , which are finally thought to reduce the bacterial growth rate and/or elicit tolerance [3] . In conjunction with earlier observations [20] , [32]–[36] , our data suggest that slow growth and concomitant antibiotic tolerance of S . Typhimurium in vivo are attributable to the peculiar interaction with dendritic cells . The high prevalence of tolerant S . Tm cells in cDC might result from the combination of long-term survival of S . Tm in this cell type ( but not in other monocytic phagocytes and granulocytes; [47] , [50] ) and the slow growth of the pathogen within these cells in vivo . Formally , we cannot exclude that cDC might also “select” for slow-growing , tolerant S . Tm cells . Anyhow , upon ciprofloxacin exposure , this bacterial subpopulation survives and might even benefit from the antibiotic-inflicted reduction of the overall tissue burden . Future work will have to identify the host cellular features and the bacterial responses establishing tolerance . This will be of importance as such cells can cause relapses after the end of an antibiotic therapy . The presence of ciprofloxacin-tolerant cells in gut mucosa , the cecal patch , the cLN , and the spleen may suggest that relapses can be “seeded” from various reservoirs . The site of reseeding may affect the relapsing disease progression . This will be a challenging topic for future work , as site specificity of the “seeding” step is difficult to establish , at least by end point assays as used in our current study . It is interesting to note that cell transfer into the peritoneal cavity of the recipient mouse led to seeding of the spleen and the liver ( Figure 2 ) . In contrast , no ( or few ) bacteria were detected in the cLN or the gut lumen of the recipient mice within 4 d . This is consistent with the route of drainage of the peritoneal cavity , which should initially seed the spleen and the liver , while gut lumen and the gut draining mesenteric lymph nodes would represent sites of subsequent dissemination at later stages of the relapse . Finally , our data establish that persistent bacteria are not invulnerable . Triggering of innate immune responses can decimate the loads of persistent S . Tm in the cLNs of antibiotic-treated mice . The TLR4 mixed bone marrow chimeras indicated that this may occur via indirect , paracrine signals . Clearly , more work is required to establish the underlying cellular mechanisms . Furthermore , as Toll-like receptor agonists may evoke significant unwanted side effects , alternative strategies for inducing innate immune responses should be explored . CpG-based drugs are presently tested in clinical trials for cancer therapy [51] . This may foster future work optimizing the augmentation of antibiotic treatment by PRR ligands and may lead to novel antibacterial therapies with increased in vivo efficacy .
The wild-type strain SB300 used in this study is an SL1344 derivative and has been described previously [52] . This strain was resequenced and found to be 100% identical to the published wt SL1344 sequence ( GenBank accession FQ312003 . 1 ) except for two nucleotide changes at positions 635606 ( T→A , Glu→Val in a putative mannose-specific PTS system enzyme IIAB ) and 1756092 ( G→A , Asp→Asn in yciT , a putative regulatory protein ) , respectively [53] . S . Tm strain ATCC14028 has been described [54] , [55] . Wild-type isogenic-tagged strains ( S . TmWITS , SB300 background [41] ) and the RFP expression plasmid pDsRed ( pACYC184 backbone , lac promotor ) have been described [23] . For infection experiments , the bacteria were grown overnight in LB broth ( 0 . 3 M NaCl ) , subcultured for 4 h , and suspended in cold PBS as described previously [56] . A stationary culture ( 37°C , aerated , in LB ) was diluted 1∶100 , 000 and 10 µl were spotted on an LB agar plate containing ciprofloxacin . The first concentration to inhibit visible growth was considered to be the MIC , as described in [57] . The MIC was determined to be <0 . 03 µg/ml , comparable to earlier results [17] . B6-Tg ( Itgax-EYFP ) mice expressing EYFP under the CD11c promoter ( CD11c-YFP; [26] ) , CD11c-DTRtg heterozygous transgenic mice ( B6 . FVB-Tg ( Itgax-DTR/EGFP ) 57Lan/J; CD11c-DTRtg; [30] ) , CX3CR1gfp/+ heterozygous transgenic mice ( CX3CR1tm1Litt ) [27] , TLR4−/− ( TLR4lps-del ) mice [58] , and wild-type C57BL/6 mice were bred and kept specified pathogen free in individually ventilated cages ( RCHCI , ETH Zürich ) . Ciprofloxacin ( ciproxine 500 , Bayer Schering Pharma ) was dissolved in water , sterile-filtered , and the concentration was quantified by UV spectrometry ( A271 nm = 30 , 614 l×mol−1×cm−1; [59] ) . It was applied intragastrically twice daily , as indicated ( 2×62 mg/kg per day ) . LPS from S . Typhimurium ( Sigma L2262-5MG; 5 µg per mouse ) was dissolved in PBS and injected intraperitoneally 24 h after the beginning of ciprofloxacin treatment . All animal experiments were approved ( licences 201/2007 , 223/2010 , Kantonales Veterinäramt Zürich ) and performed as legally required . Streptomycin-pretreated mice ( 20 mg/animal ) were infected by gavage ( 5×107 CFU; [16] , [23] ) . Live bacterial loads ( colony forming units , cfus ) in cecum draining lymph node ( cLN ) , spleen , liver , and cecal content were determined by plating [60] . The small intestine draining lymph node has not been analyzed , as it is only occasionally colonized by a small number of bacteria during the first 1–2 d of infection . Minimal detectable levels were 10 cfu/cLN , 10 cfu/spleen , and 100 cfu/g stool/cecal content . CD11c-DTRtg heterozygous transgenic mice and nontransgenic littermate controls were treated with DTX ( Sigma-Aldrich , dissolved in PBS , 100 ng/25 g body weight; i . p . injection ) 24 h before termination of the experiment . The depletion efficiency has previously been shown to be about 80% in the intestinal lamina propria and more than 90% in spleen and mesenteric lymph nodes [23] . It was verified by FACS analysis of cLNs . To expand dendritic cell numbers , mice were treated for 3 subsequent days with 10 µg of hFLT3-L ( kindly provided by Christian Münz or from eBioscience ) , starting 2 d prior to infection . Expansion of the DC population was verified by FACS of peripheral blood 24 h after the last injection ( see Materials and Methods , FACS ) . Cecum lymph node tissue samples from CD11c-YFP mice infected with S . Tm pDsRed were incubated for 1 h in 4% paraformaldehyde dissolved in PBS at 4°C , equilibrated 4 h in 20% sucrose dissolved in PBS at 4°C , and snap-frozen in O . C . T . compound ( Sakura ) . The 20 µm cryosections were air-dried , blocked ( 10% goat serum , PBS ) , stained with DAPI ( Sigma-Aldrich ) , and mounted ( Vectashield; Vector Laboratories ) . Images from immunofluorescence stainings were recorded with a Zeiss Axiovert 200 microscope , an Ultraview confocal head ( PerkinElmer ) , and a krypton argon laser ( 643-RYB-A01 , Melles Griot , Didam , The Netherlands ) . Infrared , red , and green fluorescence was recorded confocally , and blue fluorescence by epifluorescence microscopy . Images were transformed to the colors indicated , superimposed , and analyzed with Volocity 5 . 0 . 3 . ( Improvision , Coventry , UK ) . Statistical analysis was performed using the exact Mann–Whitney U Test ( Prism Version 5 . 04 ) . p<0 . 05 ( two-tailed ) was considered as statistically significant as described [16] . The exact Fisher's test ( R 2 . 15 . 1 ) was used to determine significance of reinfection/relapse efficiency in Figure 2 . For analysis of the WITS experiment , please refer to the dedicated Materials and Methods section . Mice were infected according to our standard infection protocol [16] with a uniform mixture of 7 S . TmWITS [41] , diluted with a 20- or 100-fold excess of untagged S . Tm . Mice were killed by cervical dislocation and the cLN was aseptically removed and homogenized in 500 µl of ice-cold PBS ( 0 . 5% Tergitol , 0 . 5% bovine serum albumin ) by using a Potter homogenizer . We inoculated 250 µl of the cLN homogenate into an LB overnight culture containing 50 µg/ml Kanamycin to enrich for S . TmWITS , and 150 µl were plated on MacConkey agar plates to determine the cfu . Chromosomal DNA from enrichment cultures was isolated using QIAamp DNA Mini Kit ( Qiagen , Cat . No . 51306 ) and subjected to rtqPCR using FastStart Universal SYBR Green Master ( Rox ) ( Roche , 13206900 ) with primers and temperature profiles as described [41] . The ratio of present WITS was multiplied with the number of cfus recovered to determine the absolute number of each tagged strain in the cLN . S . Tm ( pAM34 ) was cultured for 5 h in the presence of IPTG ( 1 mM ) , subcultured twice 1∶20 into LB w/o IPTG for 3 h , spun down ( 10 , 000 rpm , 4°C , 5 min ) , resuspended in ice-cold PBS , and inoculated orally into mice ( 5×107 cfu ) . In order to estimate the number of pAM34 plasmids per bacterial cell of the inoculum , serial 1∶100 dilutions of the inoculum were inoculated into fresh LB w/o IPTG , grown to stationary phase , and plated onto MacConkey plates containing streptomycin or ampicillin and IPTG . The observed percentage of ampicillin-resistant colonies obtained from each differentially inoculated culture was used to generate a standard curve linking pAM34 loss to the numbers of divisions of a given S . Tm population that had grown out from the same inoculum . Mice were infected and treated with antibiotics as described under mouse experiments . At the end of the experiment , we aseptically removed the cLN and determined the total loads of viable S . Tm by plating on MacConkey agar ( 50 µg/ml streptomycin ) and the loads of S . Tm ( pAM34 ) by plating on MacConkey agar ( 100 µg/ml ampicillin , 1 mM IPTG ) . For estimating the growth rate of the ciprofloxacin-tolerant S . Tm cells in vivo ( Figure 5 ) , we used the standard curve described above for estimating the numbers of generations that these bacteria had undergone during the infection process . Femur and tibia of C57BL/6 mice were excised , cleaned , and flushed with RPMI 1640 ( 10% FCS , 50 µM β-mercaptoethanol , 50 µg/ml streptomycin ) . Cell suspension was incubated for 30 min ( 37°C , 5% CO2 ) , supernatant was removed , spiked with GM-CSF ( 200 U/ml ) , and seeded into 24-well plates . Medium containing GM-CSF was replaced after 2 d . On day 3 , all nonadherent and loosely adherent cells were removed , and medium was replaced . BMDCs were used after 5 d in culture . BMDCs were infected for 20 min with S . Tm ( 4 h subculture of 12 h overnight culture in 0 . 3 M NaCl LB ) , washed , and incubated for 13 h under gentamycin . We added 10 µg/ml ciprofloxacin , incubated as indicated at 37°C , 5% CO2 . Similarly , an OD = 0 . 2 subculture of S . Tm was treated with 10 µg/ml ciprofloxacin , incubated at 37°C , and surviving bacterial cfus were determined each hour by plating . Single-cell suspensions of cLNs were generated by incubating the tissue in DMEM ( 50 µg/ml Liberase and 20 µg/ml Dnase ) for 30 min at 37°C and subsequently run through a 100 µm cell strainer . For sorting , samples were stained with CD90 . 2-PercP ( BioLegend , 105322 ) , CD11b-APC-Cy7 ( BioLegend 101220 ) , and B220-APC ( Becton Dickinson 553093 ) . We sorted 2×106 lymph node cells into CD90 . 2−CD11c− or CD90 . 2−CD11c+ using a FACSAria cell sorter ( Becton Dickinson ) , washed them , and lysed them , and bacterial loads were determined by plating . To determine the efficiency of Flt3-L mediated expansion of DCs ( CD4−CD8−B220+CD11c+: PBS , 0 . 3%; hFLT3-L , 0 . 6% ) , blood samples were stained with B220-APC ( Becton Dickinson 553092 ) , CD11c-APC-Cy7 ( BioLegend 117324 ) , CD8α-PacB ( BioLegend 100725 ) , and CD4-PE ( Becton Dickinson 553049 ) . To asses efficiency of DC depletion by DTX injection ( CD45 . 2+CD4−CD8−B220−CD11c+: wt mice , 5 . 3%; DTR+ mice , 2 . 6% ) , cLNs were stained with CD45 . 2-APC ( BioLegend 109813 ) , CD11b-APC-Cy7 ( BioLegend 101226 ) , CD11c-FITC ( Becton Dickinson 557400 ) , CD4-PercP ( Becton Dickinson 553052 ) , CD8-PercP ( Becton Dickinson 553036 ) , and B220-PercP ( Becton Dickinson 553093 ) and FACS analyzed . The following antibodies of Biolegend were used for cell surface staining to sort myeloid subsets for Figure 3: CD45 ( 30-F11 ) , CD3 ( 17A2 ) , CD19 ( 6D5 ) , CD103 ( 2E7 ) , CD11c ( N418 ) , and Gr-1 ( RB6-8C5 ) . Live cells were determined as SYTOX− cells ( Invitrogen ) . At least 10×106 lymph node cells were analyzed and 104 to 105 cells sorted into ( 1 ) CD45+ CD3− CD19− CD103− CX3CR1+ myeloids , ( 2 ) CD45+ CD3− CD19− CX3CR1− CD103+ CD11c+ Gr1− conventional DCs , and ( 3 ) CD45+ CD3− CD19− CD103− CX3CR1− CD11c− Gr-1+ myeloids using a BD FACSAria III cell sorter . Sorted lymph noded cells were pelleted , washed , lysed , and bacterial cfus were determined by selective plating . The excised cLN was placed in 600 µl RNAlater ( Qiagen ) and shock frozen at −80°C . Total RNA extraction was done using the RNeasy mini kit ( Qiagen ) with RNase-free DNase digest ( Qiagen ) . For reverse-transcription of 1 µg mRNA aliquots , the RT2 HT First Strand cDNA Kit ( Qiagen ) was used . Custom RT2 Profiler PCR Arrays ( Qiagen ) were run with RT2 SYBR Green ROX FAST ( QIAGEN ) on an Applied Biosystems 7900HT Fast Real-Time PCR System to amplify the resulting cDNA . Relative mRNA levels ( 2-ΔCq ) were determined by comparing the PCR quantification cycle ( Cq , determined with the Software SDS 2 . 2 . 1 ) for genes related to inflammation and defense against S . Typhimurium infection ( the selection is based on [61] ) with the reference gene Actb . The differences in their Cq cycles were calculated ( ΔCq ) . In all experiments , the upper limit of Cq was fixed to 35 cycles . Then , the fold actin was calculated and plotted . Each sample was controlled for mouse genomic DNA contamination . All DNA-positive data were excluded from further analysis . The inoculum in our infection experiments contained a minority population consisting of seven distinct WITS . Measuring the population sizes of each WITS in the cLN reveals the stochasticity that is inherent in the colonization dynamics but is “averaged out” in measurements of the majority population . By revealing the stochasticity , WITS allow us to disentangle dynamical parameters that could not be estimated individually . In a previous study [40] we showed that a minority population of WITS in the inoculum allows us to disentangle immigration and net replication rates . Here , we can exploit the stochasticity to separate replication and clearance rates in the cLN during antibiotic treatment . There are many potential ways of using the complex experimental data we generated for mathematical modeling . We chose to use only the WITS data to estimate the rates of replication and clearance . We also have measurements of the total bacterial population size in the cLN , which were determined independently from the population sizes of the WITS . Although we could have used the data on the total bacterial population size to further refine our parameter estimates , we decided to use them to validate our model fit instead ( see Figure 4B ) . The consistency between experimental data and model prediction corroborates our parameter estimates for replication and clearance rates under antibiotic treatment . Mathematically , we describe the colonization of the cLN by S . Tm by a stochastic birth-death-immigration model ( [62] see also [40] ) . Let L be a random number that denotes the number of a unique WITS in the cLN . In the model , WITS are assumed to migrate from the cecum to the draining lymph node at a rate μ , replicate in the lymph node at a rate rL , and be cleared at a rate cL . All these processes are assumed to be stochastic . The transition probabilities of our model are then given by the following equations:A diagram showing the processes incorporated into our model is shown in Figure 4A . In [40] , we derived a likelihood function and used it to estimate parameters of our model for cLN colonization within the first day after intragastric infection . We estimated that during the first day , 298 bacterial cells enter the cLN , and that the bacterial population grows at a rate of 2 . 82 per day ( corresponding to a population doubling time of 5 . 9 h; no ciprofloxacin applied during this period ) . To estimate replication and clearance parameters during antibiotic treatment , we combined the above estimates ( characterizing the early colonization dynamics ) with new experiments . In these experiments , mice were infected with mixed inocula containing seven uniquely tagged S . TmWITS . Treatment was started 1 d after infection and maintained until day 3 , 5 , or 10 . Because treatment kills almost all bacteria in the gut , we set μ = 0—that is , we assume that , once treatment has started , the influx of bacteria into the cLN is negligible . As a useful side effect of having only two parameters to estimate , we do not face any problem of identifiability as for the early colonization parameters [40] . As treatment starts 1 d after inoculation , the cLN is not empty but contains approximately 1 , 000 bacterial cells . To derive a likelihood function , we therefore need to calculate the probability generating function for arbitrary distributions of initial bacterial population sizes . ( In contrast , for the derivation of the likelihood function for early colonization , we could assume an empty cLN . ) For the general initial condition f ( s , t = 0 ) = g ( s ) that describes an arbitrary distribution of initial bacterial population sizes , the probability generating function is: ( 1 ) We first estimated replication and clearance rates from the WITS data sampled at day 3 after infection . We denoted these rates as r3 and c3 , respectively . We assume that the number of WITS in the cLN at day 1 after infection is distributed according to the probability distribution: Hereby , the parameters μ1 , r1 , and c1 characterize the dynamics from day 0 to day 1 after inoculation without antibiotic treatment . This assumption leads to better estimates than assuming a fixed bacterial population size in the cLN 1 d after inoculation . From Equation ( 1 ) with g ( s ) = f1 ( s ) , we can determine the probability generating function under treatment . Unfortunately , the state probabilities pk , which are needed for likelihood-based inference of the model parameters , cannot be determined in closed form . This is in contrast to the scenario considered in our previous study [40] , in which the process started with an empty cLN . We therefore calculated the state probabilities numerically using Fourier transformation of the characteristic function . The characteristic function can be obtained from a probability generating function by substituting eiθ for the dummy variable s . The state probabilities pk are related to the characteristic function as . We can invert this relation and calculate the state probabilities pk using the normalized discrete inverse Fourier transform ( see also [63] ) : The parameter N should be a large number . The larger N , the more accurate the state probabilities . In our implementation , we have chosen N = 214 . With these values , the likelihood and maximum likelihood estimates of the parameters reach a relative accuracy of 10−8 . To obtain maximum likelihood estimates from the state probabilities pk , we followed the same procedure as in our previous study [40] . The resulting estimates for r3 and c3 constitute averages of the replication and clearance rate over the period from day 1–3 . We follow an analogous procedure to obtain parameter estimates for the average replication and clearance rates over the periods day 3–5 and day 5–10 from the experimental enumeration of the S . TmWITS at days 5 and 10 , respectively . To that end , we set g ( s ) in Equation ( 1 ) to the probability generating function describing the state of the system at the end of the preceding period . These parameters are denoted as r5 , r10 , c5 , and c10 . All the probability-generating functions and likelihoods described here were implemented in the R language of statistical computing [64] . To obtain the maximum likelihood estimates , we used the R-function optim ( ) . Confidence intervals for the estimates were obtained with a bootstrap routine from 200 replicates . To simulate the process as shown in Figure 4B , we used the R-package GillespieSSA [65] . In these simulations , we used the parameter estimates of r1 , r3 , r5 , r10 , c1 , c3 , c5 , and c10 that we had obtained from the data . An R-package containing the datasets , likelihood , and simulation functions is provided as Protocol S1 and described in Text S1 . A systems biology markup language ( SBML ) version of the simulation model was also deposited in BioModels Database and assigned the identifier MODEL1312170001 . Mice were treated by gavage with 2 . 5 mg of unlabeled ciprofloxacin spiked with 75 µg of C-14 labeled ciprofloxacin ( kindly provided by Bayer HealthCare ) . Mice were sacrificed 2 , 8 , and 12 h later , and the indicated tissues were dissolved in 1 ml Solvable ( PerkinElmer 6NE9100; 55°C , overnight ) . We added 100 µl of 30%–35% H2O2 and incubated it for 1 h at 37°C . Samples were mixed with 15 ml Ultima Gold scintillation cocktail ( PerkinElmer: 6013329 ) , and C-14 was measured using a PerkinElmer Tri-Carb scintillation counter . | Antibiotics that are known to kill bacteria in vitro can be less efficacious in vivo . The reasons for this have remained poorly understood . Using a mouse model for Salmonella diarrhea , we found that bacterial persistence occurs in the presence of the antibiotic ciprofloxacin because Salmonella can exist in two different states: as a fast-growing population that spreads in the host's tissues and as a slow-growing “persister” subpopulation . The slow-growing bacteria infect and hide out inside dendritic cells of the host's immune system and cannot be attacked by the antibiotic—they are thereby rendered “tolerant , ” despite their genetic susceptibility to the drug . These tolerant bacteria form a reservoir of viable cells that are able to reinitiate the infection on cessation of antibiotic therapy . Fortunately , however , these tolerant Salmonella cells are not invincible , and can be killed by adding agents that directly stimulate the host's immune defense . Combining innate immune stimulants with antibiotic treatment may offer new opportunities to improve antibacterial therapies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"computer",
"science",
"computer",
"modeling",
"immunology",
"population",
"biology",
"biology",
"microbiology"
] | 2014 | Cecum Lymph Node Dendritic Cells Harbor Slow-Growing Bacteria Phenotypically Tolerant to Antibiotic Treatment |
The biophysical nature of the interaction between a transcription factor and its target sequences in vitro is sufficiently well understood to allow for the effects of DNA sequence alterations on affinity to be predicted . But even in relatively simple in vivo systems , the complexities of promoter organization and activity have made it difficult to predict how altering specific interactions between a transcription factor and DNA will affect promoter output . To better understand this , we measured the relative fitness of nearly all Escherichia coli binding sites in different promoter and environmental contexts by competing four randomized promoter libraries controlling the expression of the tetracycline resistance gene ( tet ) against each other in increasing concentrations of drug . We sequenced populations after competition to determine the relative enrichment of each −35 sequence . We observed a consistent relationship between the frequency of recovery of each −35 binding site and its predicted affinity for that varied depending on the sequence context of the promoter and drug concentration . Overall the relative fitness of each promoter could be predicted by a simple thermodynamic model of transcriptional regulation , in which the rate of transcriptional initiation ( and hence fitness ) is dependent upon the overall stability of the initiation complex , which in turn is dependent upon the energetic contributions of all sites within the complex . As implied by this model , a decrease in the free energy of association at one site could be compensated for by an increase in the binding energy at another to produce a similar output . Furthermore , these data show that a large and continuous range of transcriptional outputs can be accessed by merely changing the , suggesting that evolved or engineered mutations at this site could allow for subtle and precise control over gene expression .
While we have a reasonable understanding of the biophysical forces that determine the affinity of a transcription factor to its target sequences [1]–[4] , we still have a poor understanding of how the affinity of a factor for a site affects the output of the promoter in which it sits . The major challenge is that these relationships are highly context dependent . A high affinity site tightly bound in isolation will have no function in that it will not affect the rate of transcription of a gene , whereas a low affinity site weakly bound in the context of the initiation complex will . More subtly , a single base pair difference in the spacing between sites can affect the function of those sites [5] , [6] . Here , we attempt to better understand how binding site affinity and context relate to promoter output by determining the relative fitness of binding sites within specific variations of an engineered promoter in the bacteria Escherichia coli . The engineered promoter that we use contains three binding sites: one for the transcriptional activator MarA [6] , and another for the and the that are recognized by [7] . In the simplest thermodynamic model of transcriptional regulation in prokaryotes , the rate of transcriptional output varies as a direct function of the stability of the initiation complex [8]–[11] . The stability of the initiation complex in turn is dependent upon the cooperative binding of multiple DNA-binding transcription factors , each of which recognizes a degenerate set of sequences with different affinities [4] . The binding strengths of these sites are distributed such that there is a single optimal site that is bound with the highest affinity ( the consensus site ) and an increasing number of sequences that are bound with lower affinities as the sequences deviate from the consensus [1]–[3] . At some point the deviation becomes so great , that the site is no longer specifically bound and all remaining sequences have the same non-specific binding energy . The general assumption has been that the greater the affinity that the factor has for a site , the greater the occupancy at that site and the greater the probability that it will affect transcription [10] . This has only recently been tested for large libraries of sequences , and indeed much of the variance in expression can be explained by differences in binding site affinity [12] . Given this relationship , the distribution of binding energies for a factor defines the range of regulatory phenotypes that can be selected [2] , [13] , the number of possible DNA sequences that can be used to generate that phenotype , and subsequently the likelihood of a sequence of that strength evolving . How multiple binding sites combine to determine the stability of the initiation complex is poorly understood , mainly because there are a large number of proteins that can cooperate to regulate transcription through a variety of mechanisms [9] , [14] , including direct stabilization or destabilization of the initiation complex through protein-protein interactions or occlusion [15] , [16] or by perturbations of DNA structure that affect promoter-DNA binding [17] , [18] . MarA has been shown to modulate transcription through multiple mechanisms depending on its binding context [6] . Here we use MarA as a Class I activator that increases the rate of expression by stabilizing interactions with the carboxy-terminal domain of the alpha subunit ( CTD ) [6] , [9] , [19] . The ordering , spacing and orientation of binding sites can also mediate transcriptional regulation [11] , [20] . Differences in the spacing between the and the [5] , [21] and between MarA and the have been shown to affect the rate of transcription [6] . Here , we examine the effects of varying a binding site on promoter output by measuring the relative fitness of binding sites in different promoter and environmental contexts . To do this we placed the tetracycline resistance gene under control of the MarA-activated promoter on the plasmid pBR322 . We generated four libraries that contained different strength and MarA binding sites , to yield four varied energetic contexts for selection . By increasing the tetracycline concentration , we can change the range of selected viable transcriptional outputs . We competed variants within a library in liquid culture for 24 hours , and sequenced the competed population with an Illumina Solexa sequencer . Using this approach , we were able to map the fitness of a large population of binding sites in multiple promoter and environmental contexts relatively easily .
We generated four plasmid libraries that contained the tetracycline resistance gene ( tet ) under the control of a MarA-activated promoter with a randomized binding site . Each library contained a different combination of and MarA binding sites ( Figure 1 ) . The was either the consensus ( TATAAT ) or the weaker variant ( TTTAAT ) . The MarA binding site was either the one that regulates the mar operon [22] , or the anti-consensus site , which is not expected to bind or be activated by MarA . We will refer to each library based on which MarA binding site ( Mar or Anti ) , and which binding site ( TAT or TTT ) it contains . The four libraries therefore are named Mar:TAT , Anti:TAT , Mar:TTT and Anti:TTT . To test the dependency of cell growth in tetracycline on the sequence at the , we created promoters that contained either the consensus TTGACA or the anti-consensus GCCGGC in the Mar:TTT context . The anti-consensus site did not allow growth at as low as 5 g/ml of tetracycline , where the consensus allowed for growth in tetracycline concentrations at least as high as 100 g/ml suggesting that cell survival is dependent upon the binding site ( data not shown ) . Promoter competitions were performed as described in Materials and Methods . Briefly we transformed each library into E . coli cells and grew the cells overnight . The following morning , fresh LB cultures containing increasing concentrations of tetracycline were inoculated with the overnight cultures . Cells were competed for 24 hours and the competed populations were sequenced on a Solexa sequencer to determine the relative frequency of each hexamer . We sequenced 24 competed populations that covered 20 distinct selection conditions . Each competed population is named based on the competed library and on the concentration of tetracycline used in the competition . We carried out two independent competitions with the Mar:TAT and Mar:TTT libraries . The first was performed over the range of 5 to 30 g/ml tetracycline . We expanded the range to 50 g/ml tetracycline for all other experiments . To distinguish between different competitions with the same library , each culture that came from the same starter is given a common number ( 1 or 2 ) . For example , Mar:TAT Tet-5 ( 1 ) and Mar:TAT Tet-10 ( 1 ) came from the same Mar:TAT overnight culture , but Mar:TAT Tet-50 ( 2 ) came from a different one . The number of sequencing reads are given in Table S1 . Differences in read numbers are most likely a result of sample loss in the Solexa prep and to the lower cell density in higher tetracycline concentrations , especially with libraries containing the TTT . All but four of the sequenced competed populations had at least 25 , 000 reads . As expected , Mar:TAT Tet-5 ( 1 ) was the most variable , and appeared to show only a slight preference for the sequence at the binding site . We observed 3918 of the 4096 possible hexamers in this population , suggesting that the coverage of all sequences in our library is essentially complete . We sequenced Anti:TAT Tet-5 ( 1 ) and Mar:TTT Tet-5 ( 2 ) on two independent sequencing runs to determine if the number of sequenced DNA molecules gave an accurate and reproducible representation of the competed promoter populations . These runs generated 29 , 803 and 93 , 863 reads for the Anti:TAT Tet-5 ( 1 ) library and 33 , 229 and 11 , 263 reads for the Mar:TTT Tet-5 ( 2 ) library . We compared the relative frequency of each as determined from sequencing run 1 against run 2 and observed an for both samples ( data not shown ) . This suggested that for the more degenerate TAT libraries , as few reads sufficiently covers the distribution of binding sites . As few as reads are sufficient for the TTT libraries . Sequence logos are shown for the population of binding sites from each promoter context at 5 , 10 , 20 and 50 g/ml tetracycline ( Figure 2 ) . Logos generated from the Mar:TAT ( 1 ) and Mar:TTT ( 1 ) competitions over the smaller range of 5 to 30 g/ml were similar ( data not shown ) . We observed a decrease in the variability for each library as the amount of tetracycline used for selection was increased , with the population converging towards the consensus binding site TTGACA , suggesting that only stronger sites ( those closer to the consensus ) are viable under more stringent selection conditions . We observed a similar decrease in variability as we decreased the energetic contribution of the other components in the promoter , strongly suggesting that a decrease in the affinity of the or MarA binding sites can be compensated by an increase in the strength of the . The single base-pair mutation in the had a major effect on the population variability . Whereas completely destroying the MarA binding site by replacing it with the anti-consensus affected the population variability considerably less . For most populations , the first position of the hexamer is the least variable , and the site increases in variability towards the end . The first three positions are much more conserved than the last three , and position 6 appears to be relatively non-specific for most populations . This is consistent with the logo made from naturally occurring sites [11] . Only at the most stringent selective condition ( Anti:TTT Tet-50 ) does the consensus sequence dominate . We compared the information content ( ) [23] for each competed population as a function of tetracycline concentration for the Mar:TAT and Mar:TTT libraries ( Figure 3 ) . This figure includes data for both competition series with these libraries . Both libraries show a linear increase in information content from 5 to 30 g/ml , with a leveling at 50 g/ml . As apparent from the sequence logos in Figure 2 , the information content of the Mar:TTT library is much greater than that of the Mar:TAT library at all concentrations of tetracycline , suggesting that a weaker needs to be compensated for by a stronger for the promoter to be viable . Duplicate selections at 5 and 10 g/ml showed similar information contents for both libraries . We predicted the relative affinity ( ) of to each using the information theory based approach described in [2] , [4] and the model presented in [11] ( see Materials and Methods ) . The sites ranged in strength from to bits of information . Conventionally , sites with more than 0 bits are thought to be specifically bound [24] . 418 of the 4096 binding sites were bits . The relative fitness of each in the population was calculated by dividing the number of occurrences of that by the number of occurrences of the most frequently observed . We ranked all binding sites according to their , and compared the relative frequency for each in each experiment in Figure 4 , and only those sites with an bits in Figure S1 . The majority of hexamers were present in all libraries that contained the sequence TATAAT . As seen in Figure 2 , there is a decrease in the variability of observed binding sites as we increased the concentration of tetracycline used in selection and as the strengths of the and MarA sites are decreased in the promoter . We also observed a convergence of the viable sites towards those with higher information ( sites closer to the consensus sequence ) . Several competitions contained scattered low affinity sites with significantly higher fitness than the sites around them . We ordered all hexamers alphabetically ( AAAAAA , AAAAAC , AAAAAG … TTTTTT ) to see if there were sets of binding sites close in sequence space that had a high relative fitness , but not a high predicted affinity ( Figure S2 ) . We identified clusters of hexamers that contained a strong shifted one base to right ( orange boxes in Figure S2 and Figure 5 ) . That is , the second base of the randomized hexamer was the first base of the binding site . Differences in spacing between the and have been shown to affect the rate of initiation [5] . We tried to limit the number of binding sites with sub-optimal spacings from our libraries by placing bases disfavored by the model at the positions flanking the randomized hexamer [11] ( see Materials and Methods ) . Since the last two bases of the hexamer are fairly non-specific , it is difficult to exclude viable s with shorter spacings . The fitnesses of the binding sites were reduced at shorter spacings compared to the larger optimal spacing , and only the strongest sites were viable and only under the mildest selection conditions ( Figure 5 ) . To quantify this , we calculated the average relative fitness of four sets of hexamers that had shifted binding sites ( Table 1 ) . These sets of binding sites contained the 16 sites that had the consensus ‘TTG’ at the first three positions ( positions 2–4 of the randomized hexamer ) and a ‘G’ at the sixth position ( TTGNNG ) . This ‘G’ is the base immediately of the randomized region , and is therefore fixed . The four sets only varied in which base was of the , and should be the highest affinity sites at this spacing according to the binding site model [11] . The average relative fitness was calculated across all experiments for these sequences ( Table 1 ) . The four sets had a similar average fitness to each other and a significantly higher fitness relative to 100 , 000 randomly chosen 16 hexamers ( p ) , but on average were half as fit as the same set of sites at the optimal spacing ( TTGNNG ) and one third as fit as the 16 binding sites closest to the consensus ( TTGANN ) ( Table 1 ) . To directly compare sequence activity to and relative fitness , we measured the transcriptional output of 8 binding sites in the Mar:TTT promoter context and 7 in the Mar:TAT context by quantitative PCR ( Figure 6 ) . The sequences of these sites , their predicted affinities and their transcriptional activities are reported in Table 2 . For both libraries , output generally increased with . The data was best fit by a single exponential curve , but weakly; and for Mar:TTT and Mar:TAT respectively ( these values were only calculated for sites with an bits ) ( Figure 6A ) . Sites similar in sequence produced almost equivalent outputs . In the Mar:TTT context , TTGCGT , TTGCAG and TTGCTT vary only at their last two bases , and have similar activities ( Table 2 ) . In the Mar:TAT context , TGGAGC and TGGCTA vary at the last three bases and have the same output , and TTGCTC , TTGATG and TTGCTT have similar outputs . We suspect the model is slightly overestimating the contributions of the last 3 bases of the hexamer , and this can account for inconsistencies between our predicted affinity and transcriptional output . Expression from the Mar:TAT context was much greater than from the Mar:TTT context . The weak TAGACG in conjunction with the consensus TATAAT produced an output greater than the strongest that we assayed in the Mar:TTT context , TTGACT . Additionally , the activity of the same sequence ( TTGCTT ) in both contexts was 2 . 8 fold greater with the stronger . As seen in Figure 2 and Figure 3 , these results indicate that differences in the have a significant effect on transcriptional activity . Two of the binding sites in the Mar:TTT context had an bits , and both produced the same weak expression level ( Table 2 ) . We expect all non-specifically bound s to have this same output . One of these sites ( CTTGAC ) contained a strong that was shifted one base closer to the , but showed no activity ( blue triangle in Figure 6 ) . Additionally , we characterized two hexamers in the Mar:TAT context with an bits . One of these sequences ( CCGTTC ) showed a significantly reduced output relative to all other Mar:TAT sequences , but a high output relative to the Mar:TTT sequences . We expect this to be the transcriptional output for all non-specific s in this context . The other sequence ( CTTGCC ) contained a strong that was shifted one base to the right ( orange triangle in Figure 6 ) , but unlike the shifted site in the Mar:TTT context displayed high activity . This suggests that s with shorter spacings are only functional with the stronger , as seen in Figure 5 . There was a strong correspondence between transcriptional output and relative fitness for the 8 characterized s in the Mar:TTT context ( Figure 6B ) . At 5 g/ml of tetracycline , fitness increased as a function of output for the 5 lowest expressing s and then slightly decreased for the 3 highest expressing . At 10 g/ml , the increase in fitness extended to all but the strongest , and at 20 and 50 g/ml , fitness increased with output for all sequences . The cellular advantage for producing more of the tetracycline resistance protein may be outweighed by the cellular cost in low concentrations of drug [25] . This may explain this decrease in the overall fitness at greater outputs . The relationship between output and fitness for the Mar:TAT characterized s was less striking ( Figure 6C ) . At 5 and 10 g/ml of tetracycline , we observed an initial increase in fitness from the lowest to the second lowest expressing , and then no consistent trend . It is important to note that the differences in fitness between variants in this context are relatively small , especially compared to the Mar:TTT examples , and there could possibly be no effect on fitness at these high expression levels in these low concentrations of drug . More data points are needed to determine this . At 20 and 50 g/ml of tetracycline , we observed a general increase in fitness with output . Unlike in the Mar:TTT context , there was a gradual increase in fitness across these sites . Fitness landscapes for individual hexamers across 16 different conditions are shown in Figure 7 . We chose a series of five hexamers that decrease in predicted binding affinity from the consensus TTGACA , and differ from their neighboring sequence by a single nucleotide mutation . We also show a fitness landscape for the anti-consensus binding site GCCGGC . As expected the anti-consensus is not viable under any condition . There is an interesting contrast in the fitness landscape of the consensus sequence ( TTGACA ) to the weaker site TTGTTG . The consensus sequence shows a general increase in fitness to more stringent selective conditions , with a relatively low fitness in weak selective conditions . Conversely , TTGTTG is most fit in the weakest conditions and not viable at stringent conditions . TTGACG like TTGACA shows low fitness in the TATAAT libraries , but has a greater fitness for most of the selections with the weaker TTTAAT binding site , except for the most stringent . The fitness profile for TTGATG is weaker than expected for a site of that strength suggesting that its actual affinity may be lower than predicted . Regardless of our prediction of site strength , the difference between the TTGACG and TTGATG landscapes is large , illustrating how a single nucleotide mutation can radically change the fitness landscape of a binding site . To better understand how binding site strength correlates with relative fitness in different promoter and environmental contexts , we calculated the average relative fitness for all sites within 1 bit bins ( Figure 8 ) . For the Mar:TAT library ( Figure 8A ) , we observed that the range that has the greatest average fitness is not the highest one . We did observe an increase in the strength of the optimal fitness range as we increased the selection concentration of tetracycline , but for all tetracycline concentrations we saw a decrease in fitness at the highest range of binding sites . For the Mar:TTT library , we observed a general increase in relative fitness as a function of binding site strength for all tetracycline concentrations . Interestingly we did not observe the decrease here as we observed in Figure 6B . We did observe a similar decrease in fitness at higher information sites for the Anti:TAT library at 5 g/ml tetracycline , but not at higher concentrations . The Anti:TTT library only showed an increase in fitness at higher binding site strengths ( data not shown ) .
The fitness of the transcriptional output of a binding site is a complex function of the cellular gain and cost associated with the production of expressed gene [25] . The cellular gain in our synthetic system is the increased ability to export tetracycline from the cell . The cellular cost is the toxic effect of over-expressing the tetracycline efflux pump [26] , [27] . While we do not fully understand the absolute relationship between binding site strength , transcriptional output and the fitness of that output , clearly these things are related ( Figure 8 , Figure 6 ) and highly context dependent ( Figure 7 ) . The relative frequency of recovery of a binding site in a competed population is dependent upon two variables , ( Minimum Viable Stability ) and ( Optimal Stability ) . is the minimum stability of the initiation complex needed to produce enough of the tet gene to survive . is the stability of the initiation complex that produces the maximally fit output given a concentration of tetracycline . For a to be viable in our selection , it must have an affinity that in combination with the other binding sites produces an initiation complex stability that is stronger than . As the strength of the other sites or the output requirement changes , so does the boundary of the minimum viable binding site strength . This is indeed what we observe in Figure 4 and Figure S1 . As we increased the concentration of tetracycline ( decrease ) or as we decreased the strength of the or MarA binding sites , only stronger s remained in the selected population . This is also illustrated in Figure 2 and Figure 3 as a decrease in the variability of the population and a convergence on the consensus sequence at more stringent ( energetically demanding ) selection conditions . Compensation in binding energies between sites to produce similar stabilities has been previously predicted computationally for binding sites [11] and is shown clearly here . Interestingly , the information content of the competed populations increases linearly as a function of tetracycline concentration over the range of 5 to 30 g/ml and levels off at 50 g/ml for both the Mar:TTT and Mar:TAT libraries ( Figure 3 ) . We are not sure why the information content levels off . One possibility is that we are approaching the maximum stability where the transcriptional initiation rate is limited by the stability of the closed complex . The most fit in a given context should have an affinity , that in combination with the other binding sites , equals . We expect that fitness will increase with the overall stability of the initiation complex from to . We observe this qualitatively for libraries containing the weaker TTTAAT binding site or libraries selected at high concentration of tetracycline . Here , sites generally increase in fitness as a function of binding site strength ( Figure 4 , Figure S1 ) . Some sequences show an unexpected high or low fitness compared to their neighboring sequences with similar predicted affinities . These could be partially explained by insufficient sequencing depth , but we expect to a small degree since technical replicates suggest that for most conditions our depth gives an accurate representation of the population . Another possibility could be that some promoters may be under or over-represented in the initial library . We expect that to some extent these discrepancies are due to inaccuracies in the binding model that we used . A comparison between and transcriptional output suggests that the model may be slightly overestimating the energetic contributions of the last three bases of the hexamer to binding site strength ( Figure 6 ) . A large number of sequence anomalies can also be attributed to binding sites with shifted spacings relative to the ( Figure 5 ) . When the average fitness is calculated for binding sites with similar affinities ( reducing the effects of anomalous s ) , we see a smooth relationship between fitness and binding site strength ( Figure 8 ) . In strong selection conditions ( high tetracycline concentration , weak ) , exceeds the maximum stability that can be accessed by only varying the binding site , so here an increase in binding affinity always increases fitness ( Figure 6B and 6C , Figure 8B ) . In weak selection conditions ( low tetracycline , strong ) , the optimal binding site does not appear to be the strongest ( Figure 6B , Figure 8A ) . That is , is within the range of affinities that can be accessed by changing the . The additional energy from the presumably shifts the distribution of outputs for the binding sites into a range where there is no longer an increased advantage or even a disadvantage for transcribing that much tet . Overall , we observed a large and continuous range of fitnesses suggesting a similar scope of potential outputs can be evolved or engineered by solely mutating the . Fitness landscapes of individual sequences illustrate the large effect on fitness by even a single mutation ( Figure 7 ) . It is not clear what the maximum stability of the initiation complex is where increases in stability will no longer increase output ( closed-complex stability is not limiting ) . It has been shown for some promoters that too strong of an interaction can actually decrease transcriptional output , presumably because it is difficult for the polymerase to dissociate from the DNA [28] . A decrease in fitness from the highest affinity consensus binding site compared to a single base pair mutation of the consensus in the Anti:TTT context ( Figure 7 ) , suggests that the range of affinities of binding sites alone does not exceed that maximum . There may have been selection on to keep the range of affinities below this maximum , to maximize its output range . The relative contributions of the and MarA binding sites do not appear to be equivalent . A single mutation in the second position of the consensus greatly reduces the variability of the binding site populations . Whereas completely removing the MarA binding site has a significantly reduced effect . This suggests that binding at the contributes more to the stability of the initiation complex than does binding by MarA . The decrease in effect from the MarA site could be related to the energetics in the contact with the CTD which we do not understand [11] , or MarA expression could be low resulting in a low occupancy of the site . The significant effect of mutating the on transcript production is clearly shown in Figure 6A . The expression levels of all s in the Mar:TAT context , except for the non-specifically bound one , are greater than the expression from the most active in the Mar:TTT context that we characterized . This suggests that differences in the may contribute more than differences in the to the overall output . Open complex formation occurs through melting at the [11] , [29] , [30] . A mutation in the sequence could have a greater effect on the rate of initiation because it could lead to both a change in promoter stability and the rate of open complex formation . We expect that regardless of whether differences in the affect the stability of the closed complex or open complex formation , selection on the will be on its binding site strength . The larger range of outputs in the Mar:TAT context compared to the Mar:TTT context suggests some cooperativity between sites ( Figure 6A ) . We do not have enough data to determine to what extent . As previously mentioned , the spacing between the and can affect the rate of initiation [5] . While we tried to minimize the number of binding sites with alternative spacings from our library , this proved difficult because the last two positions of the hexamer are fairly non-specific . We observed that binding sites were viable with a 1 bp shorter spacing relative to the , but only in weak selective conditions ( low tetracycline , strong and MarA binding sites ) and only the strongest sites ( Figure 5 ) . This was confirmed by quantitative PCR , where we observed that only in the Mar:TAT context , could shifted sites produce an output above that of a non-specifically bound ( Figure 6 ) . The additional energy of the may be able to compensate for the energetic cost of binding the with a sub-optimal spacing [11] . We observed a similar average fitness for related sets of binding sites with a shifted ( Table 1 ) , suggesting that differences in the position of the do not affect transcriptional initiation . These sets of binding sites were on average about half as fit as the same set of sites with the larger optimal spacing , suggesting that differences in spacing significantly decrease transcriptional activity .
We placed the tetracycline resistance gene ( tet ) under control of a MarA-activated promoter on the E . coli plasmid pBR322 . pBR322 has several advantages: ( 1 ) It confers resistance to both ampicillin and tetracycline , allowing for maintenance of the plasmid to be either independent of or dependent on the promoter of tet . ( 2 ) It is a relatively low copy plasmid ( 15–20 copies per cell ) [31] . This eliminates the high expression of tet associated with large copy numbers . We generated four promoter libraries where the was randomized and contained either one of two MarA and binding sites ( Figure 1 ) . Variability in the relative spacing between binding sites can affect the rate of transcription [5] , [6] . We designed the promoter insert to strongly favor a single spacing between the and the to avoid having to consider spacing effects on the fitness of the promoter in the analyses . We used the optimal spacing between the and [11] , where deviations from this spacing would result in a decrease in binding affinity . Additionally , the two bases immediately ( ‘CA’ ) and the two bases immediately ( ‘GC’ ) of the hexamer are disfavored at the first and last two positions of the respectively [11] , further reducing the possibility of strong binding sites with different relative spacers . The sequence between the and MarA binding site is a slight variant of the sequence found between the MarA site and the in the mar promoter [22] . We shortened the spacer by one base at the end to have the disfavored ‘CA’ immediately adjacent to the . Martin et al . showed that this shortened spacing has a minimal effect on the degree of MarA activation [6] . We also changed three bases in the spacer to create a BstBI site ( TTCATT is now TTCGAA ) . The weaker ( TTTAAT ) in the promoter of the tet gene was mutated to the consensus ( TATAAT ) by QuickChange according to Zheng et al . [32] . These two pBR322 variants , pBR322 and pBR322 , were used for subsequent library construction . The of the tet gene on pBR322 is flanked by two unique restriction sites , EcoRI and ClaI . These sites were used to clone in MarA binding site and variants as described below . The randomized library inserts were created by DNA synthesis ( Integrated DNA Technologies ) . Variation of the binding site was done by mixing equal quantities of each base at those positions . Two library inserts were synthesized that contained either the stronger mar MarA binding site [22] , or the non-specific anti-consensus MarA binding site . The latter has the least frequently observed base at each position based on the MarA binding model ( model not published but generated from sequences in [6] ) and should not be bound . These inserts will be referred to as Ins and Ins . The DNA was made double stranded by second strand synthesis with Klenow ( NEB ) , and the fragments were purified with a QIAquick PCR purification kit ( Qiagen ) . pBR322 , pBR322 , Ins , and Ins were cut with EcoRI and ClaI ( New England Biolabs ) for two hours at C and gel purified using a QIAquick gel extraction kit ( Qiagen ) . All four combinations of plasmids and inserts were mixed and ligated overnight at C with T4 DNA ligase ( NEB ) generating 4 libraries ( Mar:TAT , Anti:TAT , Mar:TTT and Anti:TTT ) . The ligated libraries were transformed by electroporation into DH10B cells ( Gibco BRL ) , and plated on 100 ml LB+30 g/ml ampicillin plates . The number of transformants for each library was ca . . The colonies were suspended from the plate in 10 ml LB , and mini-prepped using a QIAquick miniprep kit ( Qiagen ) . Libraries were transformed by electroporation into the E . coli strain DH10B ( Gibco BRL ) . The number of transformants was ca . as determined by plating . After transformation , cells were recovered in 500 l LB for 1 hour , and grown further in 5 ml LB+30 g/ml of ampicillin overnight at C , with shaking at 225 RPM . Fresh 5 ml LB cultures containing from 5 to 50 g/ml of tetracycline were inoculated with 100 l of the promoter libraries grown overnight . Promoter libraries were competed against each other for 24 hours at C , with shaking at 225 RPM . Plasmids were purified from the competed libraries using a QIAquick miniprep kit . The Mar:TTT and Mar:TAT libraries were plated on LB agar plates containing 0 to 100 g/ml of tetracycline . Individual colonies were sequenced from these plates , and 8 variants in the Mar:TTT context and 7 variants in the Mar:TAT context were chosen that covered a large range of predicted binding strengths for further analysis . 5 ml LB cultures containing 30 g/ml of ampicillin were inoculated with E . coli containing a single binding site variant and grown overnight . A fresh 5 ml LB+30 g/ml ampicillin culture was started at and grown to an . cells were added to RNAprotect Bacteria reagent ( Qiagen ) , and RNA was purified using the RNeasy Mini kit with on-column DNase digestion ( Qiagen ) . cDNA was made from 2 g of RNA using the Superscript III RT kit ( Invitrogen ) . QPCR was performed with the SYBR green mix from NEB . QPCR primers specific to the tet and gyrA gene were both used . The relative expression of the tet gene was determined by the ratio of tet transcript abundance over gyrA transcript abundance for each sample . A serial dilution of the Mar:TTT , TTGACT sample was used as a standard for both primer sets . The expression of the tet gene for all variants was calculated relative to this . All sequences used , their predicted affinity ( ) and the expression values are reported in Table S2 . Conversion of pBR322 to pBR322 destroyed a HindIII site that overlapped the first two bases of the hexamer . Libraries that contained the wild type ( TTTAAT ) were digested with HindIII and PvuI ( NEB ) for 2 hours at C . pBR322 libraries were digested with ClaI and PvuI ( NEB ) for 2 hours at C . base pair fragments were gel purified for all four libraries using the QIAquick gel extraction kit . Excised fragments from all four promoter libraries , selected at a single tetracycline concentration , were mixed at equal concentration . Solexa libraries were then generated from this mixed population . The Illumina genomic library protocol was slightly modified ( Illumina , Inc . ) . We used a 1∶10 dilution of the Solexa genomic adapter , and ran the PCR for 16 rounds . We gel purified the final product after the PCR step instead of before as suggested . This allowed the removal of potential adapter contaminants . Sample purity and concentration were measured using a Bioanalyzer ( Agilent Technologies ) . A 45 bp single-end run was performed on a GAII machine according to the Illumina protocol . For each tetracycline concentration , the reads were identified as originating from one of the four promoter types . We used only those sequences that had an exact match to 14 or 21 specific bases that flanked the −35 region for the TAT and TTT libraries respectively . We did this to ensure that this sequence was not mutated , the spacing between the −10 and −35 was not changed , and to increase our confidence in the accuracy of the −35 sequence . We used 7 additional bases for the TTT libraries because those libraries were cut 7 bases further from the −35 than the TAT libraries . These additional bases were used to determine which variant was present for that sequence . Additionally , we required an additional 10 bases before and overlapping the MarA binding site to exactly match to confidently distinguish between the Mar and Anti libraries . The number of reads for each competition that pass these criteria are reported in Table S1 . Each was counted for each competed library at a tetracycline concentration . To determine the relative fitness of a in a competed population , the number of reads containing that was divided by the number of reads of the most frequently observed . For two of the competitions , Anti:TAT Tet-5 and Anti:TAT Tet-10 , three hexamers ( TGCCCA , TCCATT and CTGGAT ) were disproportionally high relative to the others . Interestingly , if two of these hexamers are put in the context of the promoter sequence , CA-TCCATT-G is only one base different from the reverse complement of CA-CTGGAT-G ( C-ATCCAG-TG ) . The hexamer sequence is separated from surrounding sequence by ‘-’ . These sequences may encode for the binding site of some unknown factor which may explain their increased fitness . At greater tetracycline concentrations though , these were observed much less frequently . For these competitions , the fitness of the hexamers were calculated relative to the fourth most frequently observed hexamer . Sequence logos were generated from the alignment of all reads for a single library at a single tetracycline concentration using the delila software [33] . We used the program scan to predict the relative affinity ( ) of to each hexamer . Briefly , scan compares an individual sequence to an information theory based weight matrix and sums the information contribution of each base across all positions in a site [2] . The weight matrix that we used is the one generated from 401 experimentally verified promoters in E . coli presented in [11] and is given in the supplemental materials of this paper ( Table S3 ) . There are several advantages to this approach . First , the weight matrix is generated from a large number of experimentally verified promoters , and should not be skewed by binding site selection biases [34] . Second , has been shown experimentally to be directly proportional to and more specifically [4] . Third , the information theory approach predicts a clear demarcation between specifically and non-specifically bound sites at 0 bits [24] . | A major challenge in molecular genetics has been to understand how cis-regulatory information is integrated to determine the amount of transcript generated . The difficulty has been that there are a large number of variables ( known and unknown ) that combine through an extensive array of possible mechanisms . Differences in the affinity of a binding site for its cognate binder within the initiation complex are known to account for significant differences in promoter output , but data for the activity of binding site variants in vivo has been limited . Here , we were able to map the fitness of nearly all E . coli binding sites in multiple promoter and environmental contexts using a novel method that utilizes the sequencing power of a next generation DNA sequencer . These data for the first time show the phenotypic range and continuity of a nearly complete set of possible binding targets in vivo , and they are useful in our ability to understand the mechanism , evolution , and designability of gene regulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/transcription",
"initiation",
"and",
"activation",
"computational",
"biology/genomics",
"computational",
"biology/transcriptional",
"regulation"
] | 2010 | The Fitness Landscapes of cis-Acting Binding Sites in Different Promoter and Environmental Contexts |
Catalytic loop motions facilitate substrate recognition and binding in many enzymes . While these motions appear to be highly flexible , their functional significance suggests that structure-encoded preferences may play a role in selecting particular mechanisms of motions . We performed an extensive study on a set of enzymes to assess whether the collective/global dynamics , as predicted by elastic network models ( ENMs ) , facilitates or even defines the local motions undergone by functional loops . Our dataset includes a total of 117 crystal structures for ten enzymes of different sizes and oligomerization states . Each enzyme contains a specific functional/catalytic loop ( 10–21 residues long ) that closes over the active site during catalysis . Principal component analysis ( PCA ) of the available crystal structures ( including apo and ligand-bound forms ) for each enzyme revealed the dominant conformational changes taking place in these loops upon substrate binding . These experimentally observed loop reconfigurations are shown to be predominantly driven by energetically favored modes of motion intrinsically accessible to the enzyme in the absence of its substrate . The analysis suggests that robust global modes cooperatively defined by the overall enzyme architecture also entail local components that assist in suitable opening/closure of the catalytic loop over the active site .
An issue yet to be resolved is the extent to which the intrinsic dynamics of proteins predispose them to ligand binding . Is there any correlation between local functional events such as loop rearrangements involved in ligand binding and the collective motions intrinsically accessible to the protein prior to ligand binding ? To what extent do the structure-encoded global modes of motions ( e . g . domain opening/closing , exposure or burial of active sites , cooperative conformational switches in allosteric proteins ) simultaneously engage loop motions that facilitate functional interactions ? Or , are loop reconfigurations mainly induced on a local scale by the ligand ? Notably , two different views have been advanced in recent years in linking protein dynamics and function: ( i ) enzyme structural flexibility affects its catalytic reactivity [1]–[4] , ( ii ) catalysis is independent of collective dynamics [5]–[7] . The second view is supported by the limited mobility of catalytic residues in the collective motions of the protein ( due to the requirement of precise positioning for chemical reactivity ) . Recent studies show that the preorganization of the active site is a rate-limiting factor in catalysis , while conformational dynamics help reorganize structural elements near the catalytic site [8] . The global motions of enzymes , also called slowest or softest due to their low frequency or small effective force constants , have been shown in numerous studies [9] , [10] to be robustly defined by the evolutionarily selected fold . It is conceivable that these structure-encoded modes play a role in facilitating the enzymatic activity , for example , by favoring structural changes that enable efficient recognition and binding of the substrate/ligand [11] . There is experimental evidence that the loss of conformational motion affects the enzymatic mechanism , even though the structure and electrostatics are preserved [4] , while recent work showed that electrostatic preorganization , not conformational motions , makes the largest contribution to catalysis [7] . Our examination of triosephosphate isomerase ( TIM ) collective dynamics suggests that there is a coupling between the global dynamics of the molecule and the local motions of the catalytically active loop 6 [12] , [13] . As illustrated in the Supporting Information ( SI ) Figure S1 , the experimentally observed closure of loop 6 over the ligand is in accord with the essential/principal mode of motion observed in molecular dynamics ( MD ) simulations of TIM; furthermore this first principal mode extracted from MD by essential dynamics analysis ( EDA ) [14] is in agreement with the global ( softest , lowest frequency ) mode predicted for the dimer using the anisotropic network model ( ANM ) [15] , [16] . Collective monomeric counter-rotations , which are not evident in experimental data , appear to be coupled to the functional loop's opening/closure over the active site . Moreover , experiments for TIM indicate that loop closure is not ligand-gated and emerges as an intrinsic motion of the apo enzyme [17] . While these observations signal a role of global dynamics in facilitating functional loop motions , there has been no systematic study of enzyme dynamics in relation to loop motions to establish the generality of these observations , apart from a recent study by Jernigan and coworkers where attention has been invited to the importance of slow modes for functional loop motions [18] . With the rapid accumulation of both apo and liganded structures ( usually open and closed forms , respectively ) for a given protein in the Protein Data Bank ( PDB ) [19] , and with the development of analytical models and tools for rapid estimation of intrinsic dynamics , we are now in a position to ( i ) critically examine the structural changes undergone in recognition loops and/or catalytic sites based on structurally resolved proteins in the presence/absence of a ligand and ( ii ) examine to what extent those motions are correlated with , or driven by , the global modes that are predictable using simplified , physics-based models . To this aim , we focus on a series of enzymes , where loop motions relevant to function have been experimentally detected ( Table 1 ) . In each case , we have a set of structures containing the apo and ligand-bound forms , which differ particularly in their loop regions . As listed in Table 1 , the root-mean-square deviation ( RMSD ) between the open and closed forms varies in the range 0 . 9–3 . 9 Å ( after optimal superposition of the open and closed structures to eliminate rigid-body translations and rotations ) , while the loop RMSD varies between 3 . 5 and 14 . 5 Å; and the tip residues of the loops are displaced by 6 . 7 to 25 . 0 Å between the open and closed conformations . On the other hand , the internal RMSDs of the loops , obtained after structural alignment of the isolated loops , are lower than 5 . 5 Å ( Table 1 ) , suggesting that the large displacements of the loops on the proteins are to a large extent due to the rigid-body displacements , which may be coupled to the collective motions of the enzymes . Notably , four out of ten enzymes ( TIM , protein tyrosine phosphatase ( PTP ) , L-lactate dehydrogenase and 3-dehydroquinase ) exhibit almost a rigid-lid type closure with a loop internal RMSD less than 2 Å ( Table 1 ) . The approach we undertake is the following: ( i ) to determine the dominant conformational changes of the functional loops by performing a principal component analysis ( PCA ) of the available crystal structures for each enzyme as well as by direct examination of two structures representatives of the open and closed forms ( see Table S1 in Text S1 ) , ( ii ) to determine collective modes of motion of a representative unliganded member using the ANM , and ( iii ) to examine the overlap between functional loop reconfiguration derived from experimental data and structure-based motions predicted by the ANM , as explained in previous work [11] . Additionally , we will extract essential motions from MD simulations for PTP and TIM as two case studies to further establish the correlation , if any , between computationally predicted loop motions , and those experimentally observed . We will show that a few well-defined , energetically accessible collective modes of motions encoded by the entire architecture , not by the local binding site only , favor suitable repositioning of the catalytic loop , which in turn , enable the predisposition of the active site to catalytic activity .
Calculations were performed for a dataset of 117 structures from the PDB corresponding to 10 enzymes ( Tables 1 and S1 ) , with 2 to 28 structures resolved in different forms per enzyme . Among them , HhaI methyltranferase ( M . HhaI ) is a DNA-binding enzyme; and all others bind ligands of various sizes . They contain s = 10–21 residue long loops that close over the active site during reaction . By this means , a catalytic residue located on the loop is correctly positioned in the active site and the site is protected from solvent during catalysis . We compare two sets of data generated for each enzyme: experimental , derived from the structures known for the enzyme; and computational , predicted for a representative unliganded structure ( indicated as open structure in Table S1 in Text S1 ) . Of interest is to assess the correspondence , if any , between the experimentally observed ( local ) loop motions , and the predicted loop motion as driven by the soft ( global ) ANM modes . As a metric , we use the overlap O1j≡|p ( 1 ) . u ( j ) | between the dominant motion inferred from experiments ( expressed by 3N-dimensional unit directional vector , p ( 1 ) , also called PC1 if obtained by PCA or deformation vector if calculated from the difference between open and closed forms; see Methods ) and the jth eigenmode u ( j ) predicted by the ANM . O1j varies by definition in the range [0 , 1] . An overlap close to 1 means that the experimentally observed structural change is essentially driven by the mode j . Another metric is the cumulative overlap , a summation over a subset of p modes ( see Methods ) , describing the fractional contribution of p modes to the ( experimentally ) observed deformation . Figure S2 displays the O1j values for the slowest 40 modes ( bars ) and their cumulative overlap ( curve ) for each enzyme . In six out of ten enzymes , there is at least one mode with an O1j>0 . 4 , and a cumulative overlap of 0 . 7 or higher is attained in 7/10 cases , suggesting that the soft modes facilitate , if not enable , functional loop motions . We further made a direct assessment of the orientational correlation between the loop motions observed in experiments and those predicted by computations . To this aim , we evaluated the correlation cosine , between the 3s-dimensional subvectors ps ( 1 ) and us ( j ) corresponding to the loop regions of p ( 1 ) and u ( j ) . O1jloop will be shortly called loop overlap . Table 2 shows that a loop overlap of 0 . 57≤O1jloop≤0 . 86 is achieved by at least one mode ( among the softest 10; written in parentheses ) in each examined enzyme ( column 2 ) . Column 3 lists the softest mode that yields a loop overlap higher than 0 . 5; and column 4 , the modes , among the softest 10 , that yield a loop overlap of 0 . 5 or higher . We also calculated the weighted-average overlaps , <O|s>p , averaged over p = 10 modes ( see Eq . 1 in Methods ) evaluated for segments of s consecutive residues . Figure S3 displays <O|s>p for the catalytic loop ( s-residue long ) , calculated for successive sets of 10 modes ( shifting windows along the abscissa of 3N-6 ANM modes ) . A general trend of decreasing loop overlap with increasing mode number is observed for all enzymes . As a further test , we compared weighted-average loop overlaps , calculated for the softest 10 modes , to those accomplished by randomly generated modes . To that aim , we evaluated the difference Δ<O|s>p = <O|s>pANM−<O|s>prandom at the loop region of s residues , and repeated the calculations for all successive windows of s residues along the protein sequence . The goal was to test whether the resulting ‘difference profiles’ as a function of residue ( sliding window ) index would distinguish the loop regions as regions of high overlap with ANM softest modes ( e . g . p = 10 of them ) . The difference profiles presented in Figure S4 clearly indicate that for the most part the catalytic loop regions ( the positions of which along the sequences are indicated by red stars and dashed vertical lines ) are distinguished by their high overlap with slow modes , in support of the correlation between structure-encoded soft modes and functional loop reconfigurations . The last column in Table 2 shows that the enhancement factor calculated as the ratio <O|s = loop>pANM/<O|s = loop>prandom . Notably , the enhancement factor varies between 1 . 8–16 . 1 , with PTP exhibiting the smallest enhancement , and L-lactate dehydrogenase , the largest . In summary , in each studied protein , at least one of the top-ranking ( energetically favorable ) 10 collective modes predicted by the ANM yields a high loop overlap , and the weighted-average overlap achieved by these soft modes at the loop region is enhanced by a factor of 6 . 0 on average ( over 10 proteins ) compared to randomly generated modes . These data further support the view that the seemingly ‘local’ loop reconfigurations inferred from experimental data are not decoupled from the global modes intrinsically encoded by the overall structure . On the contrary , global modes generally exhibit higher overlaps with the functional loop motions than local ( high frequency ) modes ( Figure S3 ) or random modes ( Figure S4 ) . Below we describe in more details the results for four enzymes . Protein tyrosine phosphatases form a superfamily of enzymes that regulate the tyrosine phosphorylation levels in signal transduction pathways together with the action of protein tyrosine kinases . Specifically , PTPs catalyze the hydrolysis of phosphate moiety in phosphotyrosine-containing proteins . Class I cytoplasmic PTPs include human PTP ( PTP1B ) and Yersinia PTP ( YopH ) , which show low sequence identity ( ∼20% ) [20]–[22] , but share a structurally conserved catalytic domain of ∼280 residues [21] , namely an eight-stranded mixed β-sheet wrapped by seven α-helices ( Figure 1A ) . One important feature of PTPs is the WPD loop , which carries the conserved catalytic residue Asp356 ( in YopH ) and closes over the active site upon binding of the substrate ( open and closed conformations of the loop are shown in Figure 1A ) . Similar to TIM , loop closure correctly positions the functional residues around the ligand and shields the site from bulk solvent during catalysis [21] . Standard and targeted MD simulations [23] , [24] on PTP1B have identified important regions ( S-loop , R-loop ) that are possibly related to ligand binding and closure of the WPD loop , respectively ( Figure 1A ) . M . HhaI catalyzes the methylation of cytosine residues located in specific DNA sequences with the aid of a cofactor ( S-adenosyl-L-homocysteine ) . M . HhaI is a monomeric enzyme , which positions the DNA between its large ( Rossmann fold ) and small domains [27] ( Figure 3A ) . The catalytic nucleophile Cys81 is located on a long , flexible loop , with tip residue displaced by 25 Å when it binds the DNA [27] . This is accompanied by a flip of the target cytosine out of the DNA helix into the active site . The collective motions of the M . HhaI have been proposed to facilitate the base flipping observed in the ternary complex in a study using elastic network model [28] . Orotidine 5′-phosphate decarboxylase is a homodimeric enzyme with the classic TIM-barrel fold [31] . It catalyzes the conversion of orotidine 5-monophosphate ( OMP ) to uridine 5-monophosphate ( UMP ) in the biosynthesis of primidine nucleotides . The active site is located at the dimer interface . A flexible loop located at the C-terminal end is associated with substrate binding and release of the product in the last step of the reaction . The loop ( Figure 4A ) is in open conformation when it is ordered and in closed conformation at the active site contacting the ligand [31] . Hur and Bruice performed MD simulations and found that the loop changes conformation during the catalytic reaction [32] . The homodimeric enzyme TIM plays a crucial role in the glycolytic pathway by catalyzing the interconversion of dihydroxyacetone phosphate and glyceraldehyde 3-phosphate . Each subunit adopts the TIM-barrel fold as in OMP decarboxylase . Loop 6 that carries a catalytic glutamate closes over the active site and protects it from solvent during catalysis . However , this loop closure is not ligand-gated , i . e . it also takes place in the apo state [17] . Aligned apo and ligand-bound structures of chicken TIM in Figure 5A indicate the conformational change in loop 6 ( colored blue and magenta ) . The results for other proteins are displayed in Figures 6 and S5 . Additional data provided in Tables 1 and 2 , Figure 2 , Tables S1 , S2 in Text S1 and Figures S2 , S3 , S4 essentially consolidate the results described in detail for the four cases .
Proteins undergo a broad range of motions under physiological conditions , spanning from local to global changes in conformations . Among them , the most probable motions , also known as the softest modes , are usually highly collective , i . e . , they drive the cooperative motions of entire domains/subunits [10] , [33] . Many activities of proteins are achieved , on the other hand , by relatively localized motions , such as loop reconfigurations that accompany ligand binding . A common behavior in all enzymes studied here was the occurrence of the catalytic loop reconfiguration based on the available apo and bound structures . This observation has commonly leaded to the hypothesis that loop motions are triggered by ligand binding . Given that loop motions are not collective in nature , but seemingly confined to short segments on the backbone , they might be attributed to local , rather than global dynamics . Many studies focused on such ‘regions of interest’ implicitly assuming that the loop reconfiguration observed is predominantly determined by local interactions . Our analysis demonstrates , however , that the local conformational changes observed in experiments at functional loops are not independent of the soft modes of motions intrinsically favored by the architecture . On the contrary , at least in the examined dataset , the soft modes do contribute ( more than local high frequency modes ) to the reconfiguration of the loops along directions stabilized upon ligand binding . The top-ranking ANM modes are by definition collective modes of motions known to be highly robust against sequence and structure variations . The correlation between experimentally observed structural changes at the catalytic loops and these modes suggests the evolution of the enzymatic architecture to facilitate the predisposition of the catalytic loop to enzymatic activity . Our previous and current MD simulations on TIM from two different species consistently indicate high mobility and almost full opening/closure of loop 6 in both subunits of the homodimer . In contrast , only half-closure and restricted mobility is observed for the WPD loop during PTP simulations . Evidently , there are other factors that also affect catalytic loop dynamics in terms of reaching the closed state , or the state ‘pre-disposed’ to catalytic activity . One factor may be the favorable electrostatic interactions provided by the substrate ( not included in our simulations ) . Another factor proposed to facilitate loop closure is the presence of the conserved , glycine-rich loops interacting with the active-site loop in previous MD simulations on enolase , β 1 , 4-galactosyltransferase and lipase [34] , [35] . It is important to note that our study does not contradict the critical role of electrostatic interactions in catalysis pointed out earlier [5]–[7] , and in fact , our earlier work [36] showed that the catalytic site , once assuming the ‘active’ conformation , is mechanically constrained to maintain its precise geometry required for chemical reactivity . On the other hand , conformational flexibility comes into play , and plays an important role to our view , in facilitating the binding of the substrate , and in favoring the reconfiguration of the active site into its form prone to catalysis , hence the significant role of conformational flexibility in accomplishing catalysis observed in previous work [1]–[4] . In a sense , the structure-encoded flexibility , or the suitable reorientation of the catalytic loop ( as shown here to be favored by intrinsic collective motions ) is a prerequisite for the ensuing catalytic activity which requires the appropriate chemical ( and , in particular , electrostatic ) organization . It is worth noting that the ANM modes are purely based on native contact topology , or geometry . No residue-specific interactions are taken into consideration . The collective dynamics is essentially controlled by uniform spring-like potentials; and these potentials in turn account for the Gaussian fluctuations/distributions of inter-residue distances- the underlying assumption of the theory of elastic networks , as originally set forth for polymer networks [37] . As such , the directions of motions predicted by the ANM are those favored by elastic entropic effects ( for a recent review see ref [10] ) , and the structural changes initiated/favored by these entropic effects are likely to be complemented by enthalpic effects , including in particular electrostatic interactions with the bound ligand to shape and stabilize the final closed conformer . Yet , the he correlation with experimentally observed deformations suggests that these entropic effects play a significant role in defining the accessible mechanisms of ligand binding .
The experimental data for each protein composed of N residues are generated as follows: ( i ) the ensemble of structures is superimposed using an iterative Kabsch algorithm ( see SI ) , ( ii ) mean positions <Ri> = [<xi > <yi> <zi>]T are determined for α-carbons 1≤i≤N ( or those residues with known coordinates ) , ( iii ) deviations from mean position , ΔRis = [Δxis Δyis Δzis]T ( where Δxis = xis−<xi> ) are organized in a 3N-dimensional deformation vector ΔRs for each structure s in the ensemble; ( iv ) the cross-correlations between these deviations , averaged over the entire set are written in a 3N×3N covariance matrix C ( exp ) , and ( v ) C ( exp ) is diagonalized to determine the principal modes of structural variations , where p ( i ) p ( i ) T . Here p ( i ) and σi , are the respective ith eigenvector and eigenvalue , and m is the number of structures resolved for the studied protein . The principal modes are rank-ordered: PC1 ( p ( 1 ) ) refers to the direction of maximal variance , succeeded by PC2 , etc . The fractional contribution of p ( i ) to structural variance is given by fi = σi/Σj σj where the summation is performed over all modes . The Hessian matrix , H , forms the basis of ANM approach . H can be written in terms of N×N submatrices , H ( ij ) , each of size 3×3 , given byfor , and . Here is the magnitude of the distance vector Rij0 between α-carbons i and j ( observed in the PDB ) , and , , and are the components . is the ijth element of the Kirchhoff matrix equal to 1 if i and j are connected ( within a cutoff distance of rcut ) in the network , 0 otherwise . A uniform force constant , γ , is used for all pairwise interactions . H decomposed into 3N-6 nonzero eigenvalues λi , and corresponding eigenvectors u ( i ) , as u ( i ) u ( i ) T . ANM covariance is CANM = H−1 , where H−1 is pseudo inverse , such that 1/λ1 is the counterpart of the PCA σ1 , and u ( i ) is the counterpart of p ( i ) . The overlap between PCA and ANM modes is given by the absolute value of the correlation cosine Oij = |p ( i ) . u ( j ) | [9] . For enzymes with two available structures only , p ( 1 ) is equivalent to the 3N-dimensional deformation vector ( normalized ) between open and closed crystal structures ( Table S1 in Text S1 ) . Cumulative overlap is defined as [38] . Note that CO1J = 1 for J = 3N-6 , i . e . , the 3N-6 ANM eigenvectors form a complete set of orthonormal basis vectors . The orientational correlation between the s-residue long loop motions experimentally observed and computationally predicted is measured by the overlap between the loop elements ( 3s-dimensional subvectors ) of p ( 1 ) and u ( j ) . We further define the weighted-average overlap for any segment of length s based on p modes . ( 1 ) This definition takes account of the magnitudes of loop motions , in addition to their orientations . Calculations were performed using the software package ProDy [39] . PTP was simulated for 50 ns in explicit TIP3 [40] water using NAMD [41] with CHARMM force field [42] ( see SI for details ) . Langevin dynamics and Langevin piston Nose-Hoover [43] , [44] methods were used to keep the temperature and pressure constant at 300 K and 1 atm . EDA [14] was performed after iterative superposition of the MD trajectory onto the crystal structure . TcTIM simulations were performed using AMBER [45] , [46] with the ff03 force field parameters [47] , and the protocol described in previous work [13] . | Protein loops have critical roles in ligand binding and catalysis . An unresolved issue in this context is the extent to which the intrinsic dynamics of proteins predispose loops to perform their molecular function . In this work , we ( i ) critically examine the structural changes undergone by functional/catalytic loops based on a set of enzyme crystal structures in the presence/absence of a ligand , and ( ii ) examine to what extent those motions are correlated with , or driven by , the global modes that are predictable using simplified , physics-based models . Using a dataset of 117 structures for ten enzymes of different sizes and oligomerization states , we show that the collective modes defined by the protein topology favor loop rearrangements in reasonable agreement with those experimentally observed upon activation . These results suggest that simple but robust motions encoded by the entire architecture , not the local binding site only , assist in binding of the ligand , positioning of the catalytic loop , and/or sequestration of the catalytic site , which in turn , enable efficient catalysis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biophysic",
"al",
"simulations",
"biology",
"computational",
"biology"
] | 2012 | Coupling between Catalytic Loop Motions and Enzyme Global Dynamics |
Candida albicans is a common commensal in the human gut but in predisposed patients it can become an important human fungal pathogen . As a commensal , C . albicans adapts to low-oxygen conditions and represses its hyphal development by the transcription factor Efg1 , which under normoxia activates filamentation . The repressive hypoxic but not the normoxic function of Efg1 required its unmodified N-terminus , was prevented by phosphomimetic residues at normoxic phosphorylation sites T179 and T206 and occurred only at temperatures ≤35°C . Genome-wide binding sites for native Efg1 identified 300 hypoxia-specific target genes , which overlapped partially with hypoxic binding sites for Ace2 , a known positive regulator of hypoxic filamentation . Transcriptional analyses revealed that EFG1 , ACE2 and their identified targets BCR1 and BRG1 encode an interconnected regulatory hub , in which Efg1/Bcr1 act as negative and Ace2/Brg1 act as positive regulators of gene expression under hypoxia . In this circuit , the hypoxic function of Ace2 was stimulated by elevated CO2 levels . The hyperfilamentous phenotype of efg1 and bcr1 mutants depended on Ace2/Brg1 regulators and required increased expression of genes encoding Cek1 MAP kinase and its downstream target Cph1 . The intricate temperature-dependent regulatory mechanisms under hypoxia suggest that C . albicans restricts hyphal morphogenesis in oxygen-poor body niches , possibly to persist as a commensal in the human host .
Candida albicans is a regular fungal inhabitant of the human gastrointestinal tract and the skin [1–3] but in predisposed patients it can also cause life-threatening systemic disease [4] . Systemic candidiasis occurs if resident fungi translocate to the blood and proliferate massively in extraintestinal organs [5 , 6] . Currently , the requirements for C . albicans commensalism are being investigated using murine models of colonization , in which fungi are fed orally and monitored during transit and following exit of the gut [7–11] . In all studies , C . albicans cells growing in the gut lumen were found to propogate in the yeast form and not in the alternative hyphal form . Transciptomal analyses revealed that C . albicans adapts to conditions in the mouse gut or in internal organs by upregulation of genes related to growth , stress-resistance and cell surface components [12] . Several proteins required for gut colonization were identified by their defective mutant phenotypes [9 , 10 , 12] . In contrast , mutants lacking the transcription factor Efg1 or its homologue Efh1 were found to hyperproliferate in the murine gut [7–9] , while overproduction of the Efg1-antagonist Wor1 stimulated excessive proliferation [9] . These results suggested that C . albicans limits its gastrointestinal growth by the repressive transcriptional activity of the Efg1 protein [8] . Gut mucosal damage , deficiencies in immune defenses and defects of the gut probiotic microbiome have been described as essential preconditions to allow translocation and systemic dissemination of C . albicans originating from the gut [11 , 13] . Despite this knowledge , the environmental cues and signaling pathways favouring commensal growth of C . albicans and its transition to the invasion and dissimination states are largely unclear . Oxygen-poor locations are frequent in the human host and some niches including the gut may be anoxic [14 , 15] , while other tissues including tissue of exposed skin are hypoxic [16 , 17] . Hypoxia has also been verified in the mouse gastrointestinal tract [18] . C . albicans adapts to hypoxia by increasing glycolytic and decreasing respiratory metabolism; furthermore , increased expression of genes required for the oxygen-dependent biosynthesis of compounds including ergosterol and unsaturated fatty acids procures maximal use of residual oxygen [19–21] . Under hypoxia , genes required for ergosterol biosynthesis are induced by the transcription factor Upc2 [20 , 22] , while the transcription factors Efg1 and Ace2 both upregulate glycolysis and downregulate oxidative activities [19 , 23 , 24] . Efg1 is required for rapid transcriptomal adaptation to hypoxia [25] , it controls the regulation of many hypoxic genes and prevents inappropriate hypoxic regulation of normoxic genes [14 , 19] . Besides their hypoxic metabolic functions , Efg1 and Ace2 also regulate the yeast-to-hypha transition , an important virulence trait of C . albicans , in an oxygen-dependent manner . Under normoxia , efg1 mutants are unable to form hyphae indicating that Efg1 acts as an inducer of morphogenesis [26 , 27] . In contrast , Efg1 represses hyphal growth under hypoxia , which is apparent by hyperfilamentous growth of efg1 mutants during hypoxic growth on an agar surface [19 , 23 , 28] or during embedment in agar [4 , 29] but not during growth in liquid media . The increased hyperfilamentous phenotype of an efg1 efh1 double mutant demonstrated further that the Efg1 homolog Efh1 acts synergistically with Efg1 [23] . The function of Efg1 as a hypoxic repressor was strikingly temperature-dependent since efg1 mutants were hyperfilamentous at temperatures ≤ 35°C , while at 37°C they were unable to form hyphae under both hypoxia and normoxia [19] . In contrast to Efg1 , the Ace2 protein was found to be largely dispensable for hyphal morphogenesis under normoxia [30–32] but it was required for filamentation under hypoxia [30 , 32] . Thus , Efg1 and Ace2 have opposing functions under hypoxia and recent results suggested that Efg1 represses ACE2 expression [32] , as well as expression of Ace2 target genes [32 , 33] under normoxia . Both described hypoxic functions of Efg1 , i . e . to regulate yeast proliferation of C . albicans in vitro and in the mouse gut in vivo , may be directed by similar if not identical regulatory circuits . In support of this notion , as compared to the wild-type strain , an efg1 mutant not only was hyperproliferative in the mouse gut [7–9] but it also showed increased extraintestinal dissemination in animals exposed to hypoxia [34] and increased virulence in orally-inoculated mice [35]; in contrast , the virulence of the efg1 mutant was strongly reduced in the systemic model of bloodstream-infection , i . e . under increased oxygen levels [27 , 35] . Hypoxia by decreased blood flow in individual gut villi had previously been shown to favor invasion and translocation of C . albicans across enterocytes [36] . Conceivably , the hypoxic repressor functions of Efg1 are relevant not only at temperatures <37°C , i . e . for fungal colonization of exposed skin tissue but also for translocation across gut epithelia . Here we identify a transcriptional regulatory hub describing the functions of Efg1 under hypoxia that controls the proliferation and morphogenesis of C . albicans in oxygen-limiting environments .
Under hypoxia , Efg1 has been described as a temperature-dependent regulator of morphogenesis because it suppresses filamentation during surface growth at temperatures ≤35°C [19] , while at 37°C it is required for hyphal growth [19] as under normoxia [26 , 27] . Properties of the efg1 mutant were re-confirmed by growth on the surface of YPS agar , which under normoxia does not induce hypha formation in C . albicans [37] . Under hypoxia ( 0 . 2% O2 ) , the efg1 mutant was unable to filament at 37°C , while it showed vigorous hyphal outgrowth at 34 , 30 and 25°C; in contrast , under normoxia no filamentation was observed at 25°C ( Fig 1A ) . At 37°C under hypoxia , the defective filamentation of an efg1 homozygous mutant on YPS agar was fully restored not only by native Efg1 but also by an N-terminally HA-tagged Efg1 variant ( Fig 1A ) or by an Efg1 variant carrying a N-terminal deletion ( Fig 1B ) , as under normoxia [38] . In contrast , at 25 , 30 or 34°C the synthesis of HA-Efg1 did not prevent hyperfilamentation in an efg1 genetic background , while native unmodified Efg1 had this activity ( Fig 1A ) . The repressing function of authetic Efg1 was slightly reduced by deleting residues 9 to 74 ( ΔN-Efg1 ) since colonies grown at 25°C ( but not at 30 or 34°C ) showed residual filamentation ( Fig 1B ) . Thus , the morphogenetic repressor function of Efg1 requires its native N-terminus . The structural requirements for hypoxic Efg1 functions were explored further by single-site mutated variants mutated for residues T206 and T179 . T206 fits the consensus sequence for phosphorylation by PKA [39] and T179 is considered as the phosphorylation site of the Cdc28-Hgc1 kinase complex [33]; phosphorylation of both residues is needed for efficient hypha formation under normoxia [33 , 39] . EFG1 versions encoding T206A , T206E , T179A and T179E variants were integrated into the genome of an efg1 mutant and filamentation phenotypes of transformants were examined . All Efg1 variants were produced at similar levels during hypoxic growth ( Fig 1C ) as under normoxia [38] , which was assayed by immunoblotting of cell extracts using a newly generated anti-Efg1 antiserum ( S1 Fig ) . Interestingly , at 25 , 30 and 34°C , both non-phosphorylatable variants T206A and T179A effectively repressed filamentation , while Efg1 variants mimicking phosphorylation ( T206E and T179E ) were unable to act as repressors ( Fig 1B ) . This result suggests that opposite to its normoxic functions , the phosphorylated forms of Efg1 are inactive in hypoxic repression , while the non-phosphorylated forms are active . Collectively , the results suggest that normoxic and hypoxic functions of Efg1 have different structural requirements . An unmodified native N-terminus of Efg1 and the lack of T206/T179 phosphorylation appear essential for its repressive functions under hypoxia at temperatures slightly below 37°C . The above experiments had shown that native but not HA-tagged Efg1 is active for morphogenetic repression under hypoxia at 25°C and 30°C . Next , we sought to identify genes that are hypoxically repressed by Efg1 but not HA-Efg1 under hypoxia . For this purpose genomic binding sites for both Efg1 versions were determined by ChIP chip analyses and compared . Binding of native Efg1 was determined in the Efg1+ strain CAF2-1 using anti-Efg1 antibody for immunoprecipitation and the efg1 mutant HLC52 [27] as the background control strain; binding of HA-Efg1 was established using anti-HA antibody and strain CAF2-1 as the reference control [40] . Cells were grown at 30°C ( i . e . a temperature compatible with the hypoxic repressor function of Efg1 ) in liquid glucose-containing YPD medium . This experimental setup was chosen to focus on hypoxia-regulated targets under clearly defined conditions using uniformly exponentially-growing yeast cells and to exclude other targets , e . g . related to differential filamentation . Furthermore , normoxic targets for HA-Efg1 had been previously determined in identical conditions [40] and provided a useful dataset for comparisons . Genomic binding sites for native Efg1 ( 221 sites corresponding to 300 ORFs ) greatly outnumbered those for HA-tagged Efg1 ( 100 sites corresponding to 118 ORFs ) and surprisingly little overlap was found for sites binding both proteins ( 23 sites ) ; 198 sequences were exclusively bound by native Efg1 under hypoxia ( Fig 2A ) . Little overlap was also detected between targets of HA-Efg1 under hypoxia and normoxia [40] ( S2 Fig ) . Binding sites for both proteins are specified in S1 Table and are available at ( http://www . candidagenome . org/download/systematic_results/Desai_2014/ ) . Hypoxic binding occurred exclusively in promoter regions upstream of ORFs , marking these genes as potential regulatory targets ( in case of divergently transcribed genes both ORFs were considered as regulatory targets ) . A significant subset of identified genes has a known or suspected role in hyphal growth of C . albicans ( shaded area in Fig 2A ) . Genes binding both Efg1 and HA-Efg1 included EFG1 itself [40] , as well as seven genes encoding general morphogenetic regulators comprising NRG1 , TCC1 and TYE7 . Forty-one genes were only bound by native Efg1 but not by HA-Efg1 under hypoxia including BCR1 , CEK1 , CPH2 , CYR1 , STE11 and TPK1 . Consensus sequences in promoters binding Efg1 proteins were calculated using the program RSAT dyad analysis [41] and revealed an enrichment for CA-containing motifs for both Efg1 and HA-Efg1 ( Fig 2B ) that may represent binding sites . This result suggests that although target promoters differ , untagged and tagged Efg1 proteins bind to identical sequences under hypoxia . Interestingly , the binding sequences resemble CA-containing sequences bound by HA-Efg1 during hyphal induction under normoxia but differ from the major binding site ( EGR-box TATGCATA ) in normoxically grown , yeast-form cells [40] . Gene ontology ( GO ) analysis of genes binding native Efg1 under hypoxia revealed an enrichment for genes involved in filamentation and transcription factor activity ( Fig 3 ) , as expected from the Venn diagram ( Fig 2A ) . Several of these genes had previously been identified as targets of HA-tagged Efg1 grown under normoxia in liquid [40] or during biofilm formation [42] . Gene ontology assignments for HA-Efg1 are shown in S3 Fig . The above results had indicated a subset of genes bound by native but not HA-tagged Efg1 , which are known to be involved in the yeast-to-hypha transition . Efg1 binding in promoter regions of these genes suggested that they are transcriptionally regulated by Efg1 , explaining the hypoxic repressor function of Efg1 on hyphal morphogenesis . To verify this notion , transcript levels of selected genes were monitored during the shift from normoxia to hypoxia in Efg1+ cells ( CAF2-1 ) and efg1 mutant cells ( HLC52 ) . Genes STE11 , CEK1 and CPH1 encode members of the Cek1 MAP kinase cascade , which is needed for hypha formation of C . albicans mainly during surface growth [43–45]; Efg1 binding occurs in the promoter region of STE11 and CEK1 genes at the CA-type consensus binding sequences ( Fig 2C ) . Transcripts of genes CYR1 and TPK1 were also analyzed that encode adenylate cyclase and PKA isoform 1 , respectively , which are members of the cAMP-dependent pathway of filamentation [46 , 47] . In addition , the KIC1 transcript encoding a presumed regulator of the Ace2 transcription factor [48] was assayed since Ace2 stimulates hypoxic filamentation [30] ( see below ) . In the control strain , transcripts for the Cek1 MAP kinase , its downstream transcription factor Cph1 and for the Kic1 protein were present at low levels but increased temporarily at 10–20 min following the hypoxic shift ( Fig 4 ) . Remarkably , in efg1 mutant cells , these transcripts were strongly upregulated suggesting hypoxic repression by Efg1 . A completely different pattern of regulation was detected for the STE11 gene encoding a kinase upstream of Cek1 , as well as for the CYR1 adenylate cyclase gene . Transcript levels for both of these genes decreased strongly during the hypoxic shift in the Efg1+ strain and were significantly downregulated in the efg1 mutant . Thus , the regulation of both STE11 and CYR1 did not fit the pattern of an Efg1-repressed but rather of an Efg1-induced gene; in addition , expression of both genes was down- and not upregulated during the course of hypoxic exposure . The TPK1 transcript also was downregulated under hypoxia but the absence of Efg1 did not affect its levels . Collectively , these results suggest that under hypoxia Efg1 downregulates CEK1 , CPH1 and KIC1 transcript levels to suppress filamentation , which becomes evident by the efg1 mutant phenotype ( Fig 1 ) . Efg1 binding to STE11 , CYR1 and TPK1 promoters may have other functions that are not directly related to repression of hypoxic filamentation . To verify the transcriptional data we analyzed levels of the Cek1 MAPK protein and of its phosphorylated form by immunoblotting in wild-type and mutant strains grown under hypoxia and normoxia . Under hypoxia , the total amount of Cek1 and of its phosphorylated form ( Cek1-P ) was strongly increased in the efg1 mutant as compared to the wild-type strain , while under normoxia Cek1 levels were unaltered in the mutant and Cek1-P was not detected ( Fig 5A ) . This result matches the observed increase in the CEK1 transcript level in the efg1 mutant ( Fig 4B ) . Activation of MAP kinase activity was specific for Cek1 since the Mkc1 phosphorylation status was unaffected by the presence of Efg1 ( Fig 5A ) . To test if during hypoxic surface growth , excessive filamentation by the Cek1-Cph1 pathway is suppressed by the Efg1 protein we examined filamentation phenotypes under hypoxia in strains lacking or overproducing potential regulator proteins . We observed that a cph1 mutant and colonies of an efg1 cph1 double mutant did not form hyphae , unlike the hyperfilamentous efg1 mutant ( Fig 5B ) . In addition , overexpression of STE11 and CPH1 genes encoding members of the Cek1 signaling pathway by an anhydrotetracyclin-inducible promoter stimulated filamentation in the wild-type genetic background ( Fig 5C ) . In this experiment , the failure of overexpressed CEK1 to induce filamentation may reflect low activity of non-phosphorylated Cek1 in the absence of activation by an upstream kinase . Collectively , the results provide strong evidence that Efg1 represses hypha formation of C . albicans under hypoxia by repressing the biosynthesis and activity of the Cek1 MAP kinase pathway . Ace2 is a transcription factor that under hypoxia , unlike Efg1 at lower temperatures , is required for filamentation [30] . Efg1 represses the transcription of ACE2 and of Ace2-dependent genes and binds to the ACE2 promoter under normoxia [31 , 32] . On the other hand , Ace2 and Efg1 have similar functions to stimulate glycolysis and to repress oxidative metabolism [23 , 30] . These results suggested that Ace2 and Efg1 regulatory circuits overlap to jointly control filamentation of C . albicans under hypoxia . To verify this notion we compared genomic binding sites of Ace2 and Efg1 in cells grown under hypoxia . Strain CLvW004 ( ACE2-HA/ace2 ) was constructed , which synthesizes the Ace2 protein with an added C-terminal triple HA-tag . This strain did not show any of the known ace2 mutant phenotypes for antimycin A resistance , wrinkled colony growth [24 , 30] and sensitivity to the Pmt1 O-mannosylation inhibitor [49] indicating that the Ace2-HA fusion protein is functional ( S4 Fig ) . For the identification of hypoxic Ace2 genomic binding sites , based on preliminary results , a broad ChIP chip screening strategy was chosen to identify all hypoxic targets including those targets requiring the presence of CO2 . For this purpose genomic binding sites for Ace2-HA in strain CLvW004 were determined following growth in 0 . 2% O2/ 6% CO2 and related to results of strain BWP17 synthesizing unmodified Ace2 for background correction; in parallel , normoxic binding sites were determined . Binding sites for Ace2 are listed in S2 and S3 Tables and deposited at http://www . candidagenome . org/download/systematic_results/Desai_2014/ . 296 significant genomic Ace2 binding sites were identified in C . albicans promoter regions , while no binding occurred within ORFs ( Fig 6A ) . The majority of binding sites ( >80% ) was identical in cells grown under hypoxia or normoxia ( S5 Fig ) . Analysis with the RSAT program dyad analysis [41] revealed potential consensus binding sequences CAACAA , CACCAC , CAGCW and ATCAT for Ace2 ( Fig 6B ) . The sequence CAGCW is similar to the CCAGC motif deduced from transcriptomal analyses of Ace2 [30] and matches the binding sequence of S . cerevisiae Ace2 [50] . Interestingly , the CAACAA and CACCAC motifs had also been observed as potential binding sites for native and tagged Efg1 ( Fig 2B ) suggesting that these sequences are targeted by both Ace2 and Efg1 . Genomic positions for both proteins correspond to binding motifs , mostly to the CACCAC sequence , in a selected group of promoters ( Fig 6C ) . 53 promoters bound both Ace2-HA and Efg1 and/or HA-Efg1 under hypoxia ( Fig 7A ) . Gene ontology analysis of the corresponding genes revealed their preferential function as transcription factors to regulate processes of cell adhesion , biofilm formation and morphogenesis ( Fig 7B ) . The transcription factors Brg1 [51] and Bcr1 [52] are known to regulate morphogenesis under normoxia but they also appear to function under hypoxia because they are under joint control of both Ace2 and ( HA- ) Efg1 in this environment ( Fig 7B ) . Other common hypoxic Ace2/Efg1 target genes encode regulators with more specific functions including Aaf1 [53] , Adh1 [54] , Eed1 [55] , Tye7 [56] , Rfg1 [57] , Wor2 [58] , Wor3 [59] and Zcf21 [60] . In control experiments , binding targets were verified by qPCR following ChIP demonstrating strong enrichment of binding sites for Efg1 , HA-Efg1 or Ace2-HA in a selected group of target promoters ( S6 Fig ) . By this sensitive method , binding of HA-Efg1 ( in addition to Efg1 ) was also detected at the BCR1 promoter; furthermore , these data demonstrated the specificity of antibodies used for immunoprecipitation since anti-Efg1 antibody precipitated both Efg1 and HA-Efg1 , while anti-HA antibody was specific for HA-Efg1 and Ace2-HA ( S6 Fig ) . In addition , 242 genes were identified that only bound Ace2-HA but not ( HA- ) Efg1 . This group was enriched for genes involved in glycolysis and oxidative metabolism ( e . g . PFK2 , ACO1 , LSC1 ) confirming previous transcriptomal analyses [30] . Interestingly , genes involved in mitochondrial translation ( NAM2 , ORF19 . 4929 , ORF19 . 4705 , EAF7 , PIM1 ) were also identified among Ace2 targets . The promoter of the SCH9 gene encoding a kinase repressing hypha formation under hypoxia if CO2 is present [37] , was also identified as a hypoxic binding target of Ace2 . Collectively , the group of “Ace2-only” genes appears to regulate metabolism and growth but also contains some genes for some relevant morphogenetic regulators including FLO8 [61 , 62] , CAS5 [63] , SFL1 [64 , 65] and WOR1 [58] . Conceivably , under hypoxia activation of FLO8 , which is known to be required for CO2 sensing [62] , may be mediated by Ace2 ( see below ) . Joint binding of Efg1 and Ace2 to target promoters under hypoxia suggested that both proteins regulate the respective genes on the transcriptional level . To clarify a specific role of hypoxia on gene regulation the transcript levels of selected Ace2-Efg1 target genes were determined under hypoxia and normoxia . Wild-type cells , as well as ace2 and efg1 mutant cells , were grown under normoxia and hypoxia ( 0 . 2% O2 ) both in the absence or presence of CO2 ( 6% CO2 ) ; hypoxia in combination with elevated CO2 levels was tested because previous results had suggested that this environment triggers specific patterns of gene expression [37] . Transcript levels were determined for Ace2-Efg1 target genes under hypoxia and normoxia ( Fig 8 ) . First , mutual regulation of EFG1 and ACE2 was examined . In the wild-type strain , the ACE2 transcript was strongly upregulated under hypoxia but only in the presence of CO2; upregulation did not require CO2 in the efg1 mutant ( Fig 8A ) suggesting that Efg1 strongly represses ACE2 under hypoxia also in the absence of CO2 . The EFG1 transcript level was upregulated about twofold under hypoxia in the wild-type strain; this occurred even in the ace2 mutant in the absence but not in the presence of CO2 . This mutual regulatory pattern of both genes indicated that under hypoxia , Efg1 acts as a transcriptional repressor independently of CO2 , while Ace2 requires CO2 for its induction activity . In a similar manner we analysed the hypoxic expression of two Ace2-Efg1 target genes encoding key morphogenetic regulators during biofilm formation under normoxia ( BCR1 , BRG1 ) ( Fig 8B ) . Bcr1 is a positive regulator of biofilm formation , cell surface composition and filamentation [52 , 66 , 67] , while Brg1 ( Gat2 ) promotes hypha-specific gene expression during hyphal elongation [51 , 68 , 69] and promotes ACE2 expression under normoxia [32] . The BCR1 transcript was downregulated in the wild-type strain but upregulated in the efg1 mutant under normoxia but more strongly under hypoxia revealing Efg1 as a strong hypoxic repressor of BCR1 . In the ace2 mutant the BCR1 transcript was downregulated more strongly in the presence than in the absence of CO2 suggesting that Ace2 upregulates BCR1 in this environment , counteracting Efg1-mediated repression . In contrast to BCR1 regulation , BRG1 expression was upregulated in the wild-type strain under hypoxia and this upregulation was even enhanced in an efg1 mutant but reduced in the ace2 mutant . Thus , although BCR1 and BRG1 genes are regulated differently under hypoxia , Efg1 and Ace2 regulate these genes similarly , with Efg1 acting as repressor and Ace2 as an inducer of gene expression . Other genes encoding relevant transcription factors , including Tye7 [56] , Aaf1 [53] and Zcf21 [60 , 70] , were also controlled by Efg1/Ace2 ( S7 Fig ) ; these proteins regulate glycolysis , biofilm formation and/or commensalism of C . albicans . In a previous report Bcr1 and Brg1 had been shown to bind to the EFG1 promoter [42] suggesting feedback regulation between EFG1-ACE2 and BCR1/BRG1 genes under normoxia . To clarify the regulation under hypoxia ACE2 and EFG1 transcript levels were determined in bcr1 or brg1 mutants . The ACE2 transcript was strongly downregulated in the brg1 mutant in all conditions and largely increased in the bcr1 mutant ( not further increasing the already elevated level under hypoxia/CO2 ) ( Fig 8C ) . Thus , Brg1 activates and Bcr1 represses ACE2 transcript levels . Brg1 also functions as an activator of the EFG1 transcript under hypoxia , which did not increase in the brg1 mutant in this condition . On the other hand , gene products of BCR1 and BRG1 also mutually acted as negative regulators since the hypoxic downregulation of the BCR1 transcript did not occur in a brg1 mutant ( showing even transcript upregulation ) and the BRG1 transcript was upregulated in the bcr1 mutant under hypoxia ( Fig 8D ) . Under normoxia , however , the BRG1 transcript was strongly reduced in the bcr1 mutant indicating that Bcr1 is a normoxic inducer but a hypoxic repressor for BRG1 . Collectively , the results indicate that EFG1 , ACE2 , BCR1 and EFG1 genes form an interconnected regulatory hub , in which each participant regulates expression of the co-regulators . The transcriptional output of this unit is specific for hypoxia and is influenced significantly by CO2 levels . The above transcript analyses had revealed that both CEK1 and CPH1 genes are repressed by Efg1 under hypoxia ( Fig 4 ) . Because Efg1 is part of an interconnected regulatory hub we re-examined the hypoxic/normoxic ratios of both transcripts in the respective mutant backgrounds ( Fig 8E ) . In the wild-type strain , CEK1 and CPH1 transcripts were lowered in a hypoxic atmosphere without CO2 but reached normoxic levels in the presence of CO2 ( Fig 8E ) . The repressive effect of Efg1 on both genes in this environment was clearly evident by strongly increased transcript levels in the efg1 mutant . Under normoxia the Efg1 co-regulators Ace2 , Bcr1 and Brg1 did not greatly influence CEK1 or CPH1 transcript levels . However , under hypoxia these regulators all repressed the CEK1 transcript , while the CPH1 transcript was downregulated only by Bcr1 ( and Efg1 ) . Consistently , protein levels of Cek1 and its phosphorylated form Cek1-P was upregulated under hypoxia ( Fig 5A ) . Collectively , these results confirm the conclusion that Efg1 and its co-regulators control the Cek1 MAP kinase pathway . To examine if and how members of the Efg1-Ace2 regulatory hub influence hyphal morphogenesis the colony phenotypes of control and mutant strains were recorded . Cells were grown on YPS agar under hypoxia or normoxia , in the absence or presence of 6% CO2 and at 25°C or 37°C . Under hypoxia , the control strain CAF2-1 showed no or sparse filamentation at 25°C but strong hypha formation at 37°C ( Fig 9A ) . The efg1 mutant was hyperfilamentous at 25°C but non-filamentous at 37°C verifying the previously reported dual repressor/activator role of Efg1 [19 , 23 , 29] . Strong hypha formation was also observed for the bcr1 mutant at 25°C but unlike the efg1 mutant , this mutant had a hyperfilamentous phenotype at 37°C . Both ace2 and brg1 mutant were defective in filamentation; this defect occurred for the ace2 mutant in all conditions , whereas the brg1 mutant was able to filament at 37°C in the presence of CO2 . Interestingly , under normoxia the wild-type strain presented vigorous filamentation at 37°C , while all single mutants showed complete or partial ( ace2 mutant ) filamentation defects ( S8 Fig ) . Thus , the Efg1 , Ace2 , Bcr1 and Brg1 regulators determine morphogenesis under both hypoxia and normoxia . To establish if the hyperfilamentous phenotype of the efg1 and bcr1 mutants at 25°C requires Ace2 and/or Brg1 proteins double mutants were constructed and tested ( Fig 9B ) . The construction of a homozygous efg1 ace2 double mutant failed repeatedly suggesting that a C . albicans strain lacking both Efg1 and Ace2 is not viable; therefore , an ace2/ace2 efg1/EFG1 heterozygous mutant ( CLvW047 ) was constructed . Filamentation of the efg1/EFG1 heterozygote was slightly but reproducibly increased at 25°C as compared to the control strain , while filamentation was reduced in strain CLvW047 . The hyperfilamentous phenotype of the bcr1 mutant ( especially in the presence of CO2 ) was also reduced in the bcr1 ace2 double mutant . These results suggest that the increased hypha formation of efg1 and bcr1 mutants requires Ace2 . The Brg1 protein is also needed for this phenotype because efg1 brg1 and bcr1 brg1 double mutants were completely defective for filamentation at 25°C; at 37°C , Brg1 was also needed for the bcr1 phenotype in the absence of CO2 . In summary , the results indicate that hyphal morphogenesis of C . albicans under hypoxia is effectively repressed by Efg1 and Bcr1 , counteracting the stimulatory effects of the Ace2 and Brg1 proteins .
C . albicans is an opportunistic pathogen that inhabits the human host as a harmless commensal but that also can turn into a serious pathogen , which causes tenacious superficial and deadly systemic fungal disease . Candidiasis is typically caused by the strong proliferation of the same C . albicans strain that had inhabited the patient as a commensal before [6] raising questions about the molecular events that occur in the pathogen during the commensal-to-pathogen transition . As a commensal , C . albicans colonizes the gut and partly also mucosal surfaces [2 , 71–73] . Recent results have suggested that the fungus actively restrains its proliferation in the gut by transcriptional regulators Efg1 and Efh1 [7 , 8 , 74] , while the Wor1 protein enhances gut colonization [9] . Events in the gut occur in oxygen-poor conditions ( partly under anoxia ) and mostly at elevated carbon dioxide concentrations [14 , 15] . Here we describe a transcriptional hub that downregulates filamentous growth of C . albicans and favors proliferation of its yeast form under hypoxic conditions . Surprisingly , this repressive activity involves regulators including Efg1 , which positively regulate filamentation under normoxia . Efg1 directs several aspects of morphogenesis and metabolism in C . albicans . It has an important transcriptional role under hypoxia since it contributes to but also prevents hypoxic regulation of many genes [14 , 19] . The change to a hypoxia-specific pattern of gene expression requires Efg1 at an early time-point following a shift to hypoxia [25] . It is known that Efg1 has a dual role on hyphal morphogenesis: under hypoxia it acts as a hyphal repressor during growth on agar at temperatures ≤ 35°C [19 , 23 , 28] , while under normoxia , Efg1 is a strong inducer of hypha formation [26 , 27] . The Efg1 signaling pathway under normoxia comprises adenylate cyclase Cyr1 activity that increases cAMP levels [46 , 47] , which activates PKA isoforms Tpk1/Tpk2 and in turn Efg1 by phosphorylation of residue T206 [39]; an additional phosphorylation of T179 by the Cdc28-Hgc1 complex was also described to occur during hyphal morphogenesis [33] . Here we report that the hypoxic repressor function of Efg1 has specific structural requirements . Efg1 lost its repressor activity , when its N-terminal end was modified by extension and partially by deletion , while under normoxia such variants were active in hyphal induction [38] . Interestingly , chlamydospore formation , which is induced by oxygen limitation , also was found to require an undeleted N-terminus of Efg1 [28] . In addition , phosphomimetic residues at Efg1 phosphorylation sites ( T179E , T206E ) blocked the hypoxic repressor activity , while the corresponding alanine replacement variants were fully active in repression but inactive for the normoxic induction of hyphae [33 , 39] . This result corresponds to the lowered CYR1 and TPK1 transcript levels under hypoxia , which predicts lowered PKA activity and reduced T206 phosphorylation , thus resulting in enhanced hypoxic repressor activity of Efg1 . With regard to Efg1 target sequences , the deduced CA-rich hypoxic binding sites did not match the major normoxic binding site TATGCATA for the yeast growth form , although Efg1 binds to CA- sequences shortly after hyphal induction [40] . Thus , the different functions of Efg1 as a hypoxic repressor involve different recognition and target sequences , as compared to normoxia . Previously , synergistic and antagonistic functions of Efg1 and Ace2 transcription factors have been described . Both proteins enhance glycolytic and oxidative patterns of gene expression [23 , 30] and positively influence filamentation under normoxia [26 , 27 , 30–32] . In contrast , under hypoxia Efg1 represses hypha formation [19 , 23 , 28] , while Ace2 acts as an inducer [30 , 32] . Efg1 represses ACE2 transcript levels , possibly by direct binding of Efg1 to the ACE2 promoter [32] . To further characterize the functional intersection of both regulators we used ChIP chip analyses to compare their genomic binding patterns under hypoxia . A significant overlap of target genes was identified and the deduced Ace2 binding sequences in promoters included sequences resembling the ACCAGC motif for S . cerevisiae Ace2 [50] but also the above discussed CA-sequences representing hypoxic binding sites for Efg1 . Interestingly , the group of genes targeted by both ( HA- ) Efg1 and Ace2 included important regulators of hyphal growth , biofilm formation and cell adhesion . Targets included the EFG1 promoter , which thereby was confirmed not only as an autoregulatory target for Efg1 [40] but also identified as an Ace2 target . Confirming this result , Ace2 was required for upregulation of the EFG1 transcript in a hypoxic CO2-containing atmosphere , while Efg1 repressed the ACE2 transcript as under normoxia [32] . We analyzed the mode of joint target gene regulation by the Efg1/Ace2 proteins by focusing on BCR1 [52 , 66 , 67] and BRG1 ( GAT2 ) [51 , 68 , 69] , which regulate filamentation and were found to get hypoxically down- and , respectively , upregulated . Surprisingly , these genes were not only targets but also regulators of Efg1/Ace2 and they negatively regulated each other , thereby generating an interconnected regulatory loop ( Fig 10A ) . Efg1 acted as hypoxic repressor of BCR1/BRG1 and also of two other target genes ( TYE7 , ZCF21 ) , while it was an inducer of AAF1 expression ( S3 Fig ) . In general , Ace2 activated hypoxic expression of all of these genes , especially in the presence of CO2 . Transcripts of hypoxia-upregulated genes including EFG1 , BRG1 and TYE7 and of hypoxia-downregulated genes including BCR1 , AAF1 and ZCF21 were all reduced if ace2 mutant cells were grown hypoxically in the presence of CO2 . Interestingly , Ace2 bound strongly to the promoter of the FLO8 gene , which encodes a CO2 sensor interacting with Efg1 [38 , 62] that is required for white-to-opaque switching and for filamentous growth [61] . Thus , the lack of oxygen combined with an increased level of CO2 generates an environment that elicits a specific regulatory response in C . albicans . These results are reminiscent of and confirm previous results for the hypoxia-specific , CO2-dependent functions of Sch9 kinase [37] and the Ume6 regulator [75] . The morphogenetic output of the described hypoxic regulatory hub was tested by examining hypha formation on agar . The results confirmed that Ace2 is a positive factor for filamentation under hypoxia and partly also under normoxia [30 , 32] , while Efg1 has a dual repressor/activator function under hypoxia/normoxia . Similar to Efg1 , Bcr1 acted as a repressor of hypha formation at 25°C , especially in the presence of CO2 . Transcript data suggested that in this environment , Ace2 stimulates BCR1 expression to ensure efficient blockage of hypha formation . Brg1 was needed for hypha formation at 37°C under hypoxia , but only in the absence of CO2; in its presence , Brg1 was dispensable for filamentation . The BRG1 transcript was repressed by Efg1 and activated by Ace2 , largely independent of CO2 . As shown by the phenotypes of double mutants the increased filamentation of efg1 and bcr1 mutants depended on the activity of both Brg1 and Ace2 . Thus , under hypoxia C . albicans restrains the stimulatory actions of both Brg1 and Ace2 on hyphal morphogenesis , using Efg1 and Bcr1 as repressors . While this study puts its focus on hypha formation , it is likely that other adaptation processes occurring under hypoxia , especially re-direction of metabolism to a fermentative mode , are regulated by transcription factors that link to the Efg1-Ace2 regulatory hub . Relevant transcription factors for this function could include Tye7 [56] , Aaf1 [53] and Zcf21 [60] , proteins that regulate glyolysis , biofilm formation and/or commensalism . Which signaling pathway of morphogenesis is downregulated by Efg1 or Bcr1 proteins ? Transcript analyses revealed that the genes encoding MAP kinase Cek1 , its downstream target Cph1 and the kinase Kic1 are downregulated specifically under hypoxia by Efg1 . The Cek1/Cph1 pathway is known to permit filamentation under normoxia during surface growth of C . albicans [43–45] , while in S . cerevisiae , Kic1 is part of the RAM pathway that activates Ace2 activity [48] . The repressive action of Efg1 and its co-regulators on the Cek1 pathway was verified by demonstrating that Efg1 , Ace2 , Bcr1 and Brg1 all act as repressors of CEK1 transcript levels , while the CPH1 transcript was especially repressed by Bcr1 and Efg1 . Furthermore , Efg1- and Ace2-mediated repression of the Cek1 protein in its non-phosphorylated and phosphorylated form was also demonstrated . Confirming these results , an efg1 cph1 double mutant was unable to filament , while overexpression of CPH1 in a wild-type genetic background triggered hypha formation under hypoxia . These results clearly indicate that C . albicans actively suppresses filamentation mediated by the Cek1 pathway in certain oxygen-poor conditions to promote proliferation of the yeast form ( Fig 10B ) . We have discovered this suppression in vitro slightly below the core body temperature ( i . e . < 37°C ) but this activity may also occur in the special molecular environment of the gastrointestinal tract or in hypoxic skin tissue [16 , 17] . Nevertheless , this scenario does not exclude that upregulation of hypoxic filamentation may occur under hypoxia , e . g . if the repressive action of Efg1 is blocked by the Czf1 protein [4] . In this situation , other regulators including Mss11 and Rac1 , which were identified by their embedded growth phenotypes [76 , 77] , may also promote hypoxic filamentation . A similar hyperfilamentous phenotype was also reported for mutants lacking the kinase Sch9 , although elevated CO2-levels were required in this case [37] . In the human gut , locally increased filamentation could favor anchoring of C . albicans to the epithelium and trigger strong fungal proliferation and systemic invasion mediated by invasion-specific regulators including Eed1 [8 , 55] . These events may initiate the pathogenic stage of fungal colonization , which will ultimately become apparent by the symptoms of disease . It appears that as a commensal C . albicans attempts to avoid immune responses and to maintain its residency by downregulation of filamentation and possibly other virulence traits . Strengthening of fungal commensalism , e . g . by novel therapeutic molecules or by probiotic microbes could become a promising strategy to combat serious fungal disease .
C . albicans strains are listed in S4 Table . Strains were grown in liquid YP medium ( 1% yeast extract , 2% peptone ) containing 2% glucose ( YPD ) or 2% sucrose ( YPS ) ; solid media contained 2% agar . An Invivo200 hypoxia chamber ( Ruskinn ) was used for hypoxic growth under 0 . 2% O2 [37]; liquid media were pre-equilibrated overnight under hypoxia before inoculation . Strains overexpressing genes using tetracyclin-inducible promoters were grown in/on YPS medium containing 3 μg/ml anhydrotetracycline . Oligonucleotides are listed in S5 Table . Plasmid pTD38-HA contains promoter and coding region for an N-terminally hemagglutinin ( HA ) -tagged Efg1 [38] . In this plasmid , HA-encoding sequences reside on a BglII and BamHI fragment , which were removed in plasmid pPRDEFG1 , in which the native EFG1 ORF is preceded by a BamHI site . Other plasmids carrying EFG1 genes encoding Efg1 variants without HA tag were constructed in two steps , as in the case of pPRDNEFG1 , in which nucleotides 25 and 222 of the EFG1 ORF are deleted encoding a variant lacking residues Y9 to G74 of Efg1 . Primers EFG1BamHIFor and EFG1BamHIRev were used for PCR amplification of the ORF of this variant , using plasmid pBI-HAHYD-D1 [38] as the template . The BamHI-digested PCR fragment was inserted downstream of the EFG1 promoter by ligation with the large BglII fragment of pTD38-HA to generate plasmid p2621NΔEFG1 . The PacI-SpeI fragment of this plasmid carrying the junction of the EFG1 promoter and its ORF was then used to replace the corresponding fragment in pTD38-HA . By this procedure , the mutated EFG1 ORF was joined to its 3´-UTR . Similarly , the mutated EFG1 alleles encoding the T206A and T206E mutations were transferred from plasmids pDB1 and pDB2 [39] into pTD38-HA to generate plasmids pPDEFG1T206A and pPDEFG1T206E . Analogous plasmids encoding T179A and T179E Efg1 variants were constructed by mutating the EFG1 ORF by primer pairs EFG1179AFor/Rev and T179Efor/rev , respectively , using site-directed mutagenesis ( QuikChange kit , Stratagene ) . The resultant plasmids pPDEFG1T179E and pPDEFG1T179A encode T179A and T79E variants of Efg1 , respectively . All constructs were sequenced using primers EFG1seqFor , EFG1seqRev and Efg1SeqM to confirm the presence of the mutations within EFG1 . Plasmids were integrated into the chromosomal EFG1 locus of efg1 mutant HLC67 by transformation following digestion with PacI in the EFG1 promoter [38] . Their correct chromosomal integration was verified by PCR of gDNA using primers UTREfg1For and Efg1seqM . For localization studies of Ace2 a heterozygous mutant strain was constructed first . One ACE2 ORF was replaced with a lacZ-ACT1p-SAT1 cassette . The cassette was amplified from plasmid pStLacZ-SAT . All PCR products were separated on an agarose gel and purified using the QIAquick Gel Extraction Kit ( Qiagen ) before transformation . The oligonucleotides ( lacZACE2for/lacZACE2rev ) used for amplification , tagged the cassette with 90 bp of flanking sequence complementary to the ACE2 ORF . The strain BWP17 was transformed and transformants were screened for nourseothricin resistance . Integration of the lacZ-ACT1p-SAT1 cassette and replacement of the ACE2 ORF was verified by Southern blot analysis using a probe for SAT1 . The resulting heterozygous strain CLvW001 ( ace2-lacZ-ACT1p-SAT1/ACE2 ) was used for C terminal tagging of Ace2 . Oligonucleotides ACE2HAFor and ACE2HARev were used to amplify the triple HA-encoding sequence from plasmid p3HA-URA3 thereby adding homologous sequences of the 3‘- end of the ACE2 ORF and 5‘-UTR of the ACE2 allele to the amplicon . Strain CLvW001 was transformed for C terminal tagging of the remaining ACE2 allele . Correct chromosomal integration at the ACE2 locus was confirmed by colony-PCR using oligonucleotides ACE2For and HARev and by Southern blot analysis . The resulting strain CLvW004 , expressing C-terminal HA-tagged Ace2 at the native locus was used to study chromosomal localization of Ace2-HA . For the deletion of ACE2 in the bcr1 and efg1 mutant background and in strain CAI4 , plasmid pSFS5 was used , containing a modified SAT1 flipper cassette [78] . The ACE2 upstream and downstream regions were amplified with the oligonucleotide pairs CAF1ApaI/CAF2XhoI and CAR1SacII/CAR2SacI , for construction of the inner deletion cassette , and oligonucleotide pairs CAF3ApaI/CAF4XhoI and CAR3SacII/CAR4SacI for construction of the outer deletion cassette; the ace2 mutant strain MK106 described by Kelly et al . [24] was used as the reference strain . gDNA of strain SC5314 was used to generate PCR products , which were digested with the indicated restriction enzymes and cloned on both sites of the SAT1 flipper cassette in pSFS5 [78] resulting in plasmids pCLvW90 ( inner cassette ) and pCLvW91 ( outer cassette ) . The deletion cassettes in plasmids pCLvW90 and pCLvW91 were released by digestion with SacI and ApaI and used for transformation of strains using first the fragment containing the outer cassette and , after excision of the SAT1 flipper cassette , with the second fragment containing the inner cassette . Integration of the deletion cassette was confirmed by colony PCR using oligonucleotides FLP1 , which binds within the FLP1 gene , and ACE2UTR5 , which binds upstream of the ACE2 ORF . Transformants containing the ACE2 deletion cassette were grown overnight in liquid YCB-BSA medium ( 20 g yeast carbon base , 4 g bovine serum albumin and 2 g yeast extract per liter ) to induce excision of the cassette by FLP-mediated recombination; corresponding strains were identified by their small colony size on YPD plates containing 25 μg/ml nourseothricin . After excision of the second SAT1 flipper cassette , deletion of both ACE2 alleles was confirmed by Southern blot analysis . With this approach both ACE2 alleles were deleted in the bcr1 mutant strain CJN702 [52] resulting in strain CLvW024 and in CAI4 resulting in strain CLvW008 . However , deletion of the second ACE2 allele in the efg1 mutant HLC52 [27] was not successful . In an additional approach we tried to generate the ace2 efg1 double knockout strain by deleting EFG1 in the ace2 mutant strain CLvW008 with the Ura-blaster disruption technique [79] but again we were not able to obtain a homozygous mutant strain suggesting that the double knockout is lethal . The URA3 deletion cassette used for this approach was released from plasmid pBB503 [80] by digestion with HindIII and KpnI . The fragment was purified and transformed into strain CLvW008 and transformants were selected for uridine prototrophy on SD agar . Integration of the cassette was confirmed by colony-PCR . In the resulting strain CLvW041 one chromosomal copy of EFG1 was replaced by the sequence of URA3 flanked by hisG sequences ( efg1::hisG-URA3-hisG/EFG; ace2::FRT/ace2::FRT ) . Attempts to delete the second EFG1 allele after removal of URA3 [79] were not successful . For deletion of BRG1 gene in efg1 ( HLC52 ) and bcr1 ( CJN702 ) mutants the upstream region of the BRG1 ORF was amplified by genomic PCR using primers BRG1 5UTRKpnIFor/KpnIRev and cloned into the KpnI site of plasmid pSF5S to generate pSFS5-B5 . The BRG1 downstream region was amplified using primers BRG13UTRNot1For/SacIRev and cloned into NotI and SacI sites of pSFS5-B5 to generate pSFS5-B5B3 . The KpnI-SacI fragment of this plasmid was used to transform efg1 and bcr1 mutant strains , selecting transformants on YPD agar containing nourseothricin ( 200 μg/ml ) . Correct genomic integration was confirmed by colony PCR using oligonucleotides FLP1 , which binds within the FLP1 gene , and BRG1UpFor , which binds upstream of the BRG1 promoter region . Verified heterozygous transformants were grown in YCB-BSA medium to evict the disruption cassette and retransformed with the disruption fragment , as described [78] . Deletion of both BRG1 alleles was confirmed by negative colony PCR using primer BRG1UpFor , which binds upstream to the BRG1 ORF and BRG1midrev that is specific for the BRG1 ORF . BRG1 alleles were deleted in efg1 ( HLC52 ) and bcr1 ( CJN702 ) mutant strains resulting in strains PDEB4 ( efg1 brg1 ) and PDBB4 ( bcr1 brg1 ) respectively . The plasmids pClp10TETSTE11 , pClp10TETCEK1 and pClpTETCPH1 [81] encoding Ste11 , Cek1 and Cph1 proteins were linearized with StuI within the RPS1 sequence and transformed into C . albicans CEC2907 [81] selecting for uridine prototrophy; the resultant strains were named CECSTE11 , CECCEK1 and CECCPH1 . Correct plasmid integration at the RPS1 locus was confirmed by colony PCR using primers ClpUL and ClpUR . A rabbit polyclonal anti-Efg1 antiserum was generated using His10-tagged Efg1 produced in E . coli . The EFG1 ORF ( allele ORF19 . 8243 ) residing on a XhoI-BamHI fragment was subcloned into pET19b ( Novagen ) , downstream of the T7 RNA polymerase promoter . The resulting plasmid encoded a His10-Efg1 fusion but contained a single CUG codon ( residue 449 ) that encodes serine in C . albicans but leucine in E . coli . This codon was changed to a UCG serine codon by site-directed mutagenesis using oligonucleotides pET19Serinhin/her , resulting in plasmid pET19-His-Efg1Kodon , which was transformed into E . coli Rosetta 2 ( DE3 ) pLysS ( Merck ) . Transformants were grown and the T7 promoter was induced according to instructions of the manufacturer . Cells were resuspended in buffer ( 20 mM CAPSO pH 9 . 5 , 1 M NaCl , 1 mM EDTA , 20 mM imidazol , 0 . 1% Triton X100 ) and broken using 3 passages through a French press cell ( Slaminco Spectronic Instruments ) . Crude extracts were cleared by centrifugation and applied to HisTrap columns connected to an ÄKTA prime plus fraction collector ( GE Healthcare ) . The His10-Efg1 fusion protein was eluted using CAPSO buffer containing 250 mM imidazol . Purified protein ( 100 μg ) was injected on days 1 , 14 , 28 and 56 in 2 New Zealand White rabbits ( performed by Eurogentec , Belgium ) . One rabbit generated high anti-Efg1 titers in ELISA tests and in immunoblottings ( dilution 1:5000 ) . YPD precultures were grown under normoxia overnight at 30°C in YPD medium and were used to inoculate 40 ml of YPD medium , which had been preincubated overnight under hypoxia ( 0 . 2% O2 ) . Starting with an initial density of OD600 = 0 . 1 cells were grown at 30°C under hypoxia to an OD600 = 1 . Cells were harvested , frozen at -70°C for 1 h and then thawed by addition of 500 ml of CAPSO buffer ( 20 mM CAPSO pH 9 , 5 , 1 M NaCl , 1 mM EDTA , 20 mM imidazole , 0 , 1% Triton X-100 ) containing protease inhibitor ( Cocktail Complete , Mini , EDTA-free/Roche ) . Cells were broken at 4°C by shaking with one volume of glass beads ( 0 . 45 mm ) in a FastPrep-24 shaker ( MP Biomedicals ) using 4–6 cycles for 40 s at 6 . 5 ms-1; between cycles cells were placed on ice for 5 min . Debris was removed by centrifugation at 13 , 000 rpm for 5 min and protein in the supernatant was determined using the Bradford assay . 45 μg of the crude cell extract was separated by SDS-PAGE ( 8% polyacrylamide ) and analysed by immunoblotting using anti-Efg1 antiserum ( 1:5 , 000 ) or anti-histone H4 ( Abcam; 1:5 , 000 ) to detect histone H4 as loading control . Total Cek1 levels were detected by immunoblotting using anti-Cek1 antiserum [10] , while phosphorylated Cek1 was detected using monoclonal rabbit anti-phospho-p44/42 antibody ( Cell Signaling Technology ) . Anti-rabbit-IgG-HRP conjugate ( 1:10 , 000 ) was used as secondary antibody in all blottings . Signals generated by the chemiluminescent substrate ( SuperSignal West Dura; Pierce ) were detected by a LAS-4000 mini imager ( Fujifilm ) and evaluated by the Multi Gauge Software ( Fujifilm ) . The ChIP chip procedure was performed as described by Lassak et al . [40] , except that the strains and antibodies used for immunoprecipitation were different . Two independent cultures were assayed for each combination of strains . Precultures were grown overnight under normoxia at 30°C in YPD medium and were shifted to YPD medium precalibrated under hypoxia ( 0 . 2% O2 , 30°C ) . The cells were allowed to grow from OD600 = 0 . 1 to 1 . Two sets of strains were analysed: ( 1 ) wild-type strain CAF2-1 as test strain and efg1 mutant HLC52 as control strain were compared to determine the genomic localization of untagged Efg1 , using anti-Efg1 antibody for chromatin immunoprecipitation; ( 2 ) Strain HLCEEFG1 producing HA-tagged Efg1 as test strain and DSC11 ( Efg1 producing ) as control strain were compared to determine the genomic localization of HA-Efg1 using anti-HA antibody for chromatin immunoprecipitation . C . albicans genomic tiling microarrays were probed pairwise by immunoprecipitated chromatin as described previously [40] . For localization studies of Ace2 , precultures were grown overnight under normoxia at 30°C in YPD medium and were used to inoculate medium preincubated under normoxia or hypoxia ( 0 . 2% O2 ) with addition of 6% CO2 . Strain CLvW004 , producing HA-tagged Ace2 from its native locus and control strain BWP17 were used to determine the genomic localization of Ace2-HA using anti-HA antibody for immunoprecipitation . Significant binding peaks were defined as probes containing four or more signals above background in a 500 bp sliding window; the degree of significance depended on the FDR value . Results were visualized using the program SignalMap ( version 1 . 9 ) . The most significant binding peaks ( FDR ≤ 0 . 05 ) for Efg1 ( 202 peak genomic binding sites ) , HA-Efg1 ( 106 peak genomic binding sites ) and Ace2-HA ( 272 peak genomic binding sites ) , which coincided in both replicates , were analysed by the program RSAT dyad-analysis to predict DNA binding sequence [40] . | Candida albicans is an important cause of human disease that occurs if the fungus proliferates strongly on skin surfaces or in several internal organs causing superficial and systemic mycosis . Remarkably , at low cell numbers , C . albicans is also a normal inhabitant of mucosal surfaces and the gut and it is believed that its transition from the commensal to the virulent , highly proliferative state is a key event that initiates fungal disease . In the gut and other body niches , C . albicans adapts to an oxygen-poor environment , which downregulates its virulence traits including the ability to form hyphae . We report on a set of four transcription factors in C . albicans that form an interdependent regulatory circuit , which downregulates filamentation specifically under hypoxia at slightly lowered body temperatures ( ≤ 35°C ) . Disturbance of this circuit is expected to initiate the fungal virulence and proliferation in predisposed patients . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Hypoxia and Temperature Regulated Morphogenesis in Candida albicans |
Despite large vaccination campaigns , measles virus ( MeV ) and canine distemper virus ( CDV ) cause major morbidity and mortality in humans and animals , respectively . The MeV and CDV cell entry system relies on two interacting envelope glycoproteins: the attachment protein ( H ) , consisting of stalk and head domains , co-operates with the fusion protein ( F ) to mediate membrane fusion . However , how receptor-binding by the H-protein leads to F-triggering is not fully understood . Here , we report that an anti-CDV-H monoclonal antibody ( mAb-1347 ) , which targets the linear H-stalk segment 126-133 , potently inhibits membrane fusion without interfering with H receptor-binding or F-interaction . Rather , mAb-1347 blocked the F-triggering function of H-proteins regardless of the presence or absence of the head domains . Remarkably , mAb-1347 binding to headless CDV H , as well as standard and engineered bioactive stalk-elongated CDV H-constructs treated with cells expressing the SLAM receptor , was enhanced . Despite proper cell surface expression , fusion promotion by most H-stalk mutants harboring alanine substitutions in the 126-138 “spacer” section was substantially impaired , consistent with deficient receptor-induced mAb-1347 binding enhancement . However , a previously reported F-triggering defective H-I98A variant still exhibited the receptor-induced “head-stalk” rearrangement . Collectively , our data spotlight a distinct mechanism for morbillivirus membrane fusion activation: prior to receptor contact , at least one of the morbillivirus H-head domains interacts with the membrane-distal “spacer” domain in the H-stalk , leaving the F-binding site located further membrane-proximal in the stalk fully accessible . This “head-to-spacer” interaction conformationally stabilizes H in an auto-repressed state , which enables intracellular H-stalk/F engagement while preventing the inherent H-stalk’s bioactivity that may prematurely activate F . Receptor-contact disrupts the “head-to-spacer” interaction , which subsequently “unlocks” the stalk , allowing it to rearrange and trigger F . Overall , our study reveals essential mechanistic requirements governing the activation of the morbillivirus membrane fusion cascade and spotlights the H-stalk “spacer” microdomain as a possible drug target for antiviral therapy .
Measles virus ( MeV ) is a major human pathogen leading to more than 120 , 000 deaths per year [1] . The disease can be prevented by vaccination , and global eradication is considered feasible in principle , but requires maintenance of a 95% herd immunity . However , sub-optimal vaccine delivery in developing countries and non-compliance in western countries continue to foster measles outbreaks . In order to achieve global measles eradication , post-exposure prophylaxis has recently been proposed as a synergistic strategy to complement vaccination programs by filling herd immunity gaps [2] and newly available morbillivirus infection inhibitors have established proof-of-concept for the efficacy of this approach in animal models [3–5] . However , to minimize the possibility of emergence of drug-resistant mutants , development of additional candidate compounds for combined therapy is indicated . MeV belongs to the Morbillivirus genus within the Paramyxovirus family , which also contains important animal pathogens such as canine distemper virus ( CDV ) or peste des petits ruminants virus ( PPRV ) [6] . CDV is one of the major infectious agents of carnivores and often induces severe neurological disorders [7] . Importantly , CDV exhibits a very broad host range that even extends to non-human primates [8–11] , which raises concerns that the virus could eventually adapt to humans . Therefore , the development of a panel of broad-spectrum morbillivirus inhibitors might be important to augment measles eradication and suppress the emergence of future zoonotic morbilliviruses . Both MeV and CDV entry systems rely on two surface glycoproteins for infection: the receptor-binding protein H and the fusion protein F [6] . Both proteins tightly associate to execute membrane fusion at neutral pH . It is assumed that H-protein binding to a specific cell surface receptor is translated into the triggering of the F-protein [12 , 13] . Subsequently , F undergoes a series of irreversible conformational changes that lead to merger of the viral envelope with a host cell membrane , resulting eventually in the formation of a fusion pore [6 , 14 , 15] . Recent structural and biochemical studies revealed that tetramers represent the physiological oligomer of the morbillivirus H-protein [16 , 17] . Each H-monomer contains a short luminal tail , a single transmembrane domain and a large ectodomain . The extracellular region is composed of a membrane-proximal stalk section supporting a membrane-distal cuboidal head domain with a six-beta propeller fold [16 , 18–20] , which is responsible for binding to multiple receptors ( such as SLAM and Nectin-4 ) [18 , 21–30] . The H-stalk is further divided into three modules: ( i ) a central section consisting of a candidate F-contacting segment ( aa 110–118 ) [31 , 32] , which partially overlaps with an F-triggering region ( aa 84–117 ) , ( ii ) a compact intermediate “spacer” section ( aa 122–137 ) with unknown function , and ( iii ) two C-terminal dimeric “linker” regions ( aa 139–154 ) that may connect the four globular head domain to the stalks [33] . Although the precise structure of the morbillivirus H-stalk domain remains to be determined , the atomic structures of the related parainfluenza virus type 5 ( PIV5 ) and Newcastle disease virus ( NDV ) attachment protein ( HN ) -stalks were partially resolved and revealed a conserved four-helical bundle ( 4HB ) with an upper straight and lower supercoiled conformation [34–36] . Successful engineering of covalent bonds trapping dimers and/or tetramers throughout the CDV and MeV H-stalks indicate that the 4HB-like conformation is presumably a conserved theme among members of the Paramyxovirus family [31 , 33 , 37 , 38] . The most recent model for triggering the paramyxovirus entry machinery is based on discrete crystal structures of soluble form of the PIV5 and NDV receptor-binding proteins ( HN ) . In these structures , either one ( PIV5-HN ) or two ( NDV-HN ) dimeric head units backfold onto the C-terminal region of the stalk , thereby covering the putative F-activation/binding site ( referred to as “heads down” conformation ) [34 , 36] . Alternative structures were also reported in which both dimeric head units are assembled into tetramers ( referred to as “heads up” conformation ) [39] . These static atomic structures inferred the possibility that receptor-binding may “shift” the dimeric head units from the “down” to the “up” configuration , hence unmasking the F-activation/binding site . In turn , the stalks are freed to mediate interaction with and activation of prefusion F-complexes ( referred to as the “stalk-exposure” and “induced fit” models ) [13 , 34 , 36 , 40] . Overall , the models illustrate the essential role of the attachment protein’s “heads down” conformational state , which prevents F-binding prior to receptor engagement . Several lines of evidence indicate that the “stalk-exposure” model may act as a general mechanism regulating paramyxovirus entry: ( i ) paramyxovirus attachment protein stalk domain is involved in short-range interaction with the F-protein [31 , 32 , 41–45] , ( ii ) although unregulated , PIV5 HN “headless” constructs remain bioactive [46]; a phenotype that equally extends to MeV [47 , 48] , Nipah ( NiV ) [49] , NDV and Mumps ( MuV ) attachment protein stalk regions [40] , ( iii ) PIV5 HN and F do not assemble intracellularly [46 , 50 , 51] , ( iv ) biochemical and structural data indicate that MeV H can adopt discrete conformations [16 , 38] and ( v ) a polyclonal antibody generated against the stalk domain of the Nipah virus attachment protein ( G ) can sense a receptor-induced conformational change in the stalk [49] . However , several studies challenge that the “stalk-exposure” model: ( i ) the morbillivirus and henipavirus’ glycoproteins preassemble intracellularly [48 , 52–56] , ( ii ) stalk-elongated MeV H-mutants remain bioactive [57] , ( iii ) only tetrameric MeV H-heads conformations ( in complex with SLAM ) were crystalized [16] , and ( iv ) human parainfluenza type 3 ( HPIV3 ) glycoproteins may also associate prior to receptor-binding [58 , 59] , where the attachment protein ( HN ) would exert a stabilizing role on prefusion F-structures [60] . These findings suggest that the entry system of these paramyxoviruses either does not rely on the “stalk-exposure” model or that variations of the latter must exist . To elucidate the morbillivirus F-triggering machinery , we have characterized a monoclonal antibody ( mAb-1347 ) that very efficiently ablated morbillivirus-mediated virus-to-cell and cell-to-cell fusion activities . Molecular mapping of the mAb-1347 epitope revealed that the H-stalk C-terminal “spacer” region contains the putative antibody-docking site . Strikingly , mAb-1347 spotlighted a yet unrecognized receptor-induced “head-stalk” conformational change in morbillivirus H , which occurs prior to the “opening” of the central stalk section required for F-activation . Furthermore , alanine-scanning mutagenesis highlighted a key functional role of the “spacer” microdomain in enabling H-tetramers to fold into a putative critical pre-receptor-bound state . These findings are discussed in the context of a further refined model of morbillivirus membrane fusion triggering , which illustrates how paramyxoviruses may have evolved diverging mechanisms to initiate the F-protein refolding cascade necessary for cell entry .
To better understand the mechanism of morbillivirus F triggering , we characterized a panel of monoclonal antibodies that was previously raised against CDV [61] . We first classified these mAbs to identify those that are most efficient in blocking both CDV-mediated cell entry ( virus-to-cell fusion ) and spread ( cell-to-cell fusion ) . While virus-to-cell fusion inhibition was investigated by virus neutralization assays , block of cell-to-cell fusion was assessed by determining the capacity of the different mAbs to efficiently ablate H/F-mediated fusion of receptor-positive Vero-cSLAM cells . Our screen demonstrated that the anti-CDV-H monoclonal antibody 1347 ( mAb-1347 ) strongly inhibited both types of fusion ( Fig 1A–1C ) . To further investigate whether fusion inhibition correlated with prevention of receptor binding , a previously developed semi-quantitative SLAM-binding assay was employed . Results shown in Fig 1D illustrate that mAb-1347 did not alter SLAM binding activity , whereas the anti-CDV-H mAb-2267 used for comparison impaired H-SLAM binding by approximately 40% . These data demonstrate that the anti-CDV-H monoclonal antibody 1347 effectively inhibits viral cell entry and spread by a mechanism other than blocking receptor binding . To further characterize the mechanism of mAb-1347 fusion inhibition , we attempted to identify its binding site on CDV H . Towards this goal , we employed a soluble H-form ( sH-ecto; containing the full ectodomain and carrying a GCN4 motif fused N-terminally ) [31] . In addition , we constructed a soluble protein consisting of the H-stalk that was flanked by GCN4 ( N-terminally ) and RFP ( C-terminally ) . The whole cassette was cloned in frame with the IgK signal peptide to engineer a secreted version of the chimeric protein ( sH-stalk-RFP ) . Both soluble H-constructs also carry a supplementary N-terminal hexahistidine tag ( Fig 2A and 2B ) . To determine whether H-stalks or H-heads were recognized by mAb-1347 , both soluble proteins were expressed in 293T cells and culture supernatants subsequently incubated with different anti-CDV-H mAbs ( 1347 , 1C42 and 2267 ) . Immunoprecipitated proteins were subjected to western blot analysis and the results compared to the H-antigenic materials detected by a polyclonal anti-GCN4 antibody . Our data revealed that mAbs 1C42 and 2267 efficiently interacted with sH-ecto but not with sH-stalk-RFP , whereas mAb-1347 immunoprecipitated both soluble constructs ( Fig 2C ) . Furthermore , when all four mAbs were tested for their reactivity against H-proteins gel fractionated under denaturing and reducing conditions , only mAb-1347 detected the H-antigenic material ( S1A Fig ) . These data inferred that mAbs 1C42 and 2267 recognize conformational epitopes located in the H head domain , whereas mAb-1347 binds to a linear epitope located in the H-stalk section . To further refine the region within the H-stalk recognized by mAb-1347 , deletion mutants were generated based on the sH-stalk construct ( Fig 2B ) . Whereas all soluble H-stalk mutants were efficiently immunoprecipitated by an anti-histidine epitope mAb , mAb-1347 completely lost reactivity with deletion mutant 4 ( del4 ) ( Fig 2D ) . Since the del4 mutant lacked only 20 residues ( H-stalk section 120–139 ) compared to the del3 mutant , we engineered an additional construct containing only these 20 residues ( del5 ) . When subjected to pull-down analysis , del5 was efficiently immunoprecipitated by mAb-1347 ( Fig 2D ) . Furthermore , when residues 120–122 , 123–125 or 135–137 of del5 were substituted for alanines ( Fig 2E ) , the resulting H-stalk mutants S1 , S2 and S6 ( scans 1 , 2 and 6 , respectively ) were still recognized by the mAb , in contrast to S3 , S4 and S5 ( scans 3 to 5 spanning the 126–134 H-stalk region ) ( Fig 2F ) . Lastly , we transferred the eight core residues ( 126–133 ) of this segment into a location in the CDV F globular head domain that we have previously demonstrated to tolerate incorporation of short epitope-tags without substantially altering F bioactivity ( F-wt-tagFLAG , Fig 2G ) . The resulting F-variant ( F-wt-tag1347 ) was recognized by mAb-1347 as efficiently as by a control mAb ( 4941 ) , which detects the prefusion conformation of the CDV F-trimer ( Fig 2H ) . To confirm these data in the context of membrane anchored H-protein , we expressed an H construct that lacked the head domains ( headless CDV H; residues 1–159 ) , which were replaced by FLAG epitope tags ( S1B Fig ) . Headless CDV H was very efficiently recognized by mAb-1347 , as demonstrated by immunofluorescence ( IF ) staining followed by flow cytometry ( S1C Fig ) . Remarkably , and in contrast to H-wt , MFI values obtained with mAb-1347 and headless H were very similar to those recorded with the αFLAG mAb . Several truncated membrane-anchored headless CDV H versions were next engineered ( all containing the FLAG tags ) ( S1B Fig ) . While mAb-1347 interacted efficiently with H stalk 1–159 and 1–139 , a slightly decreased binding activity was recorded for headless 1–132 when compared to surface expression of this variant ( monitored with an anti-FLAG mAb ) . Consistent with the results of our initial epitope mapping , mAb-1347 binding to headless H 1–122 was completely lost , despite efficient surface expression of this mutant ( S1C Fig ) . Taken together , our data identified residues 126–133 , located in the C-terminal part of the H-stalk domain , as the critical amino acids governing mAb-1347 binding . Moreover , surface-exposed H-tetramers lacking the head domains interact more efficiently with mAb-1347 than standard H . It was recently reported that headless attachment protein constructs of other paramyxoviruses ( PIV5 , MeV , NiV , MuV and NDV ) spontaneously trigger F-trimer refolding and membrane fusion , albeit to various extents [40 , 46 , 47 , 49] . We therefore anticipated that headless CDV H may likewise trigger CDV F . Nevertheless , our initial attempts with wt CDV F remained unsuccessful , despite proper oligomerization and surface expression ( Figs 3A and S1C and S2A–S2C ) . However , we recently demonstrated that prefusion F-trimers derived from the neurovirulent CDV A75/17 strain are conformationally highly stable [62] and identified a single substitution ( V447T ) that substantially destabilizes wt CDV F complexes to a level very similar to that of MeV F Edmonston . Using a previously established F-triggering assay [63] , co-expression of CDV H stalk 1–159 with F-V447T led to efficient F activation ( Fig 3B ) . Conversely , F-triggering was not induced by H stalk 1–159 with an additional L111A substitution ( Fig 3B ) , which reportedly impairs F binding [32 , 47] . To determine whether membrane fusion can be mediated by headless CDV H , we co-expressed Hstalk 1–159 and F-V447T in receptor-positive and-negative cells ( Vero-SLAM and Vero cells , respectively ) and assessed fusion qualitatively . Regardless of the cell line used , syncytia were detected , albeit at a low level ( Fig 3C ) . Of note , fusion activity induced by CDV H stalk 1–159 remained below that observed for the heterologous MeV system . Interestingly , MeV H-stalk constructs required complex engineering to achieve proper stabilization and fusion promotion [47] . Hence , it is possible that headless CDV H naturally folds into a more stable conformation that enables fusion triggering , but tighter oligomerization subsequently interferes with optimal bioactivity . Using the fusion-competent headless H/F-V447T combination , we asked whether mAb-1347 would still be capable of blocking F-triggering . Because this glycoprotein combination induced only very limited cell-to-cell fusion , we investigated fusion promoted by H stalk 1–159 qualitatively and quantitatively . Remarkably , headless H/F-mediated membrane fusion was completely inhibited by the mAb , whereas fusion proceeded unperturbed when F was triggered by an even shorter H 1–122 construct , regardless of the presence or absence of mAb-1347 ( Fig 3D and 3E ) . Of note , CDV H-stalk 1–122 also required engineering of a stabilizing domain to achieve productive fusion triggering ( 1–122 opGCN4 ) , analogous to the previous experience with headless MeV H proteins [47] . Overall , these data confirm that headless H 1–122 lacked the epitope recognized by mAb-1347 , and illustrate that fusion inhibition by the antibody is independent of the presence of the H-head domains . We next aligned the amino acid sequence of the CDV H-stalk 126–133 microdomain with those of several other prototypic canine distemper and morbillivirus strains . In agreement with a previously generated phylogenetic tree of morbillivirus H proteins , this alignment showed very good conservation between the selected CDV and PDV strains , whereas divergence was noticeable when compared to other morbillivirus H sequences ( S3A Fig ) . As expected based on this observation , mAb-1347 likewise blocked fusion of other CDV strains ( S3B Fig ) . To determine whether mAb-1347 also blocks the related MeV H protein that contains two point mutations in the proposed epitope , MeV H and F ( Edm and ICB323 strains ) were co-expressed in Vero-hSLAM cells and fusion promotion was monitored 24h post-transfection in the presence or absence of the antibody . S3C and S3F Fig illustrate that mAb-1347 does not cross-react with MeV H , since cell-to-cell fusion remained essentially unaltered in the presence of the antibody . Since the putative mAb-1347 epitope in MeV H exhibited two substitutions compared to the sequence of CDV H , we generated a MeV H double mutant by changing these residues to their CDV counterparts . The MeV H-D128N/Y131F variant was bioactive when expressed with MeV F in receptor-positive cells but , strikingly , fusion promotion was completely inhibited by mAb-1347 ( S3D and S3F Fig ) . When mapped in our previously generated 4HB-stalk structural model [31] , the side chains of both residues are predicted to be solvent exposed , suggesting a direct contribution to the interaction with the mAb ( S3E Fig ) . The successful transfer of mAb-1347 membrane fusion inhibition to MeV H through epitope reconstruction implies that mechanistic insight into CDV fusion triggering can be extrapolated to related morbillivirus family members . The F-contact zone in MeV and CDV H supposedly includes residues encompassed in a linear section of the membrane-proximal stalk domain ( 110–118 ) [17 , 31 , 32 , 57] . Our epitope mapping experiments place the mAb-1347 binding site ( 126–133 ) slightly membrane-distal from this candidate F-interaction domain . To investigate whether F-trimers sterically interfere with mAb-1347 binding activity , we performed IF and subsequently flow cytometry analyses on Vero cells expressing H-wt either alone or in combination with F . Results in Fig 4A document that regardless of the presence or absence of F , very similar MFI values were recorded . When we determined mAb reactivity to headless H ( H-stalk 1–159 ) expressed alone or combined with F , identical MFI profiles were obtained ( Fig 4A ) . Remarkably however , we noted a substantially higher mAb-1347 reactivity to headless CDV H than to full-length CDV H-proteins ( Fig 4A ) , suggesting that the presence of the H-heads in a pre-receptor-bound conformation of the H tetramer may impair mAb-1347 binding . To further test the notion that mAb-1347 can interact with H-tetramers even when complexed with F-trimers , we conducted H/F co-immunoprecipitation assays with mAb-1347 in receptor-negative Vero cells . Consistent with our cytometry results , F1+2 complexes were efficiently co-immunoprecipitated by mAb-1347 ( recognizing the H-stalk ) as well as the control 2267 and FLAG monoclonal antibodies ( targeting the H-head ) ( Fig 4B ) . Co-immunoprecipitation experiments were repeated with the H-stalk 1–159 construct and results shown in Fig 4C indicate that F1+2 was again pulled down by both mAb-1347 and anti-FLAG . As anticipated , mAb-2267 could not co-immunoprecipitate F1+2 when co-expressed with headless CDV-H ( Fig 4C ) . Altogether , these data strongly support that mAb-1347 recognizes an F binding-competent H conformation that exists prior to receptor contact , but does not sterically interfere with the H/F hetero-oligomerization . Enhanced mAb-1347 reactivity to headless CDV-H constructs suggested that standard H-tetramers may assume a pre-receptor-bound conformational state in which at least one head domain partially shields the upper stalk region , blocking the epitope . Receptor binding may then induce conformational changes in the head domain that fully reveal the epitope . To test this hypothesis , we expressed wild-type H-proteins in receptor-negative Vero cells . One day post-transfection , H-expressing cells were concomitantly treated with mAb-1347 and overlaid with Vero cells expressing a morbillivirus receptor ( SLAM or Nectin-4 ) , or standard receptor-negative Vero cells ( 1 hour at 4°C or 37°C ) . The reactivity of mAb-1347 with H-proteins was then monitored through immunostaining followed by flow cytometry . MFI values were first normalized for total cell surface expression ( recorded by anti-FLAG mAb staining and flow cytometry ) and then standardized to H-wt treated with regular Vero cells . Values remained mostly unaltered when cells were kept at 4°C ( Fig 4D ) , but exposure to 37°C enhanced the mAb-1347 binding efficiency compared to values obtained for cells never exposed to receptor-positive cells ( referred to as receptor-induced mAb-binding-enhancement ( RBE ) phenotype ) ( Fig 4D ) . We previously reported that cell-cell fusion induction was dramatically enhanced when H/F-complexes were expressed in cells expressing SLAM compared to Nectin-4 [62] . Of note , we recorded a more pronounced RBE phenotype when H-tetramers were treated with SLAM than with Nectin-4 , suggesting a direct correlation between fusion-support efficacy and the ability of H to exhibit the RBE phenotype . Consequently , our results support the hypothesis that mAb-1347 senses a significant conformational modification occurring in H upon receptor engagement . Since H-tetramers are recognized to a certain extent by mAb-1347 prior to receptor binding ( regardless of the presence of F ) an H subpopulation may spontaneously assume a receptor-induced-like conformation . Alternatively , the pre-receptor-bound H conformation could mask some mAb-1347 epitope ( s ) , allowing only partial recognition by the mAb . In either case , our data demonstrate that mAb-1347 efficiently detects a yet uncharacterized receptor-induced conformational change in H . We next asked whether the identified H-stalk microdomain may also impact the overall bioactivity of the H-tetramer . To address this question , 13 single alanine H-variants spanning residues 126–138 located in the upper “spacer” region of the H-stalk were generated ( Fig 5A ) and their ability to trigger F tested qualitatively and quantitatively in Vero-cSLAM cells . Although microscopically-detectable cell-to-cell fusion was observed with most H-mutants , quantitation revealed that only H mutants N128A , E130A and H137A induced fusion to levels comparable to those recorded for wild type H , whereas the remaining 10 mutants showed substantial functional deficiencies ( Fig 5C ) . Anti-FLAG-based monitoring of cell surface expression confirmed that all H-variants were properly surface-expressed ( Fig 5C ) , suggesting that the loss of bioactivity was not due to gross protein misfolding . Assessment of SLAM binding revealed that most mutants likewise exhibited H-wt-like receptor binding . However , some showed slight ( <50% of standard H ) alterations , suggesting that parts of the 126–138 stalks section may affect the overall H conformation and influence receptor binding through long range effects ( Fig 5C ) . Mutants H-N128A , E130A and F131A were substantially compromised in mAb-1347 binding , whereas all other mutants exhibited strong reactivity ( Fig 5D ) . Combined with our previous results , these data support the conclusion that residues N128 , E130 , and F131 are directly recognized by mAb-1347 . We next tested the H alanine mutants for the mAb-1347 RBE phenotype . As before , H-variants were expressed in receptor-negative Vero cells and exposed to mAb-1347 and Vero cells expressing SLAM molecules or standard Vero cells at 37°C . For this experiment , we selected two hypo-fusogenic H-mutants ( P127A and W138A ) and one with wild-type-like fusion triggering activity ( H137A ) ( Fig 5B and 5C ) , based on their ability to efficiently react with mAb-1347 ( Fig 5D ) despite the differences in F-triggering activity . Strikingly , both hypo-fusogenic variants exhibited a reduced RBE phenotype ( Fig 5E ) , whereas the mutant associated with wt-like fusion-promotion ( H-H137A ) also displayed wt-like RBE ( Fig 5E ) . Although the RBE-deficient phenotype of H mutant P127A may be partially attributed to slightly reduced SLAM binding activity , the H-W138A variant exhibited wt-like receptor binding ( Fig 5C ) , suggesting an alternative mechanism blocking the RBE phenotype . We have previously demonstrated that structural rearrangements within the central region of the H-stalks are strictly required for F-triggering . We therefore thought to determine whether the newly identified “head-stalk” conformational change occurred prior to , or after this “in-stalk” rearrangement . To address this question , we took advantage of H-mutants I98C and I98A , which are deficient in rearranging the central stalk section [17 , 31 , 33 , 38 , 64 , 65] , and assessed their ability to undergo receptor-induced “head-stalk” conformational changes . Strikingly , both mutants exhibited a wt-like RBE phenotype ( Fig 5E ) , indicating that exposure of the 126–133 stalk section occurs prior to the structural rearrangements of the central region of the H-stalks . Taken together , these data suggest that ( i ) the 126–138 section contributes to the folding of H-tetramers into fusion triggering-competent states , ( ii ) receptor-induced conformational changes exposing the H 126–138 upper stalk section are required for productive F-triggering , and ( iii ) the “head-stalk” rearrangements precede the “in-stalk” conformational changes . Whereas MeV H-variants lacking the head domains retained receptor-independent F-triggering bioactivity , stalk-elongated ( heads-carrying ) versions did not [47 , 57] . Rather , these stem-elongated H-constructs still required the presence of receptor to trigger F [57] . Since insertions in MeV H were added N-terminal to the spacer region ( between residues 118 and 119 ) , we speculated that the bioactivity of such H-elongated mutants remained dependent on the presence of the receptor because the extended stalks remain able to assume proper pre-F-triggering folds and therefore still rely on the coordinated “head-stalk” conformational changes to activate F . To test this hypothesis , we inserted a segment of 11 residues ( putatively representing 1 complete turn of the helical wheel of this stalk portion ) into the CDV H-stalk between positions 122 and 123 ( H-elong ( +11 ) ) . We also constructed a second mutant in which stalk residues 125–135 are deleted , thus removing the mAb-1347 epitope from H ( H-short ( -11 ) ) ( Fig 6A ) . Slight mobility shifts were found when fractionating both engineered H proteins in SDS-PAGE followed by immunoblotting ( Fig 6B ) . The results summarized in Fig 6 indicate that H-short ( -11 ) completely lacked bioactivity regardless of the presence or absence of the receptor ( Fig 6C ) , despite efficient cell surface expression and SLAM binding activity ( Fig 6D and 6E ) . In contrast , the elongated H version , which likewise remained intracellular transport-competent and bound SLAM and mAb-1347 efficiently ( Fig 6D and 6E ) , promoted membrane fusion in the presence of SLAM ( Fig 6C ) . However , bioactivity was entirely inhibited by mAb-1347 ( Fig 6C ) . Co-immunoprecipitation experiments revealed that H-elong ( +11 ) exhibited proper F-binding activity , whereas H/F interactions were impaired when F was co-expressed with the shortened version of H ( Fig 6F and 6G ) , as previously demonstrated for an equivalently shortened MeV H mutant [56] . Importantly , H-elong ( +11 ) displayed wt-like mAb-1347 RBE ( Fig 6H ) . Overall , the fact that H-short ( -11 ) remained transport competent but fusion promotion-inactive confirms the essential role of the stalk region in ensuring sterical compatibility of morbillivirus H and F . Efficient RBE recorded with H-elong ( +11 ) confirmed that the stalk-elongated construct can proceed to productive F-triggering while fully relying on the “head-stalk” conformational change to trigger F refolding . The above results suggest that at least one mAb-1347 epitope must remain inaccessible to mAb binding in the pre-receptor-bound H conformation . This inferred that a “less-than-parity” stoichiometry of mAb to H-monomer may be sufficient to inhibit fusion . On the other hand , we cannot exclude the possibility that receptor binding-dependent unmasking of mAb-1347 epitope ( s ) opens a window of opportunity for additional mAb-1347 interaction ( leading to a critical 1:1 stoichiometry ) . To shed light on mAb-1347 to H stoichiometry requirements for fusion-inhibition , we adapted to the CDV system a previously reported H trans-complementation assay first developed for MeV ( [38] and Fig 7A and 7B ) . We based the assay on two reportedly inactive H-mutants: a FLAG-tagged headless CDV-H ( unable to activate the highly stable F-wt ) and the HA-tagged H-I98A variant ( F-trigger-defective , [17 , 64] ) . While cell-to-cell fusion remained absent when Vero-SLAM cells expressed F-wt together with either one of the H-mutants , fusion was fully restored when cells co-expressed F and both H-mutants ( Fig 7C and 7F ) . Cell surface co-immunoprecipitation experiments confirmed that mixed oligomers readily formed ( S4A Fig ) . As expected , restored fusion activity was efficiently blocked by mAb-1347 ( Fig 7C and 7F ) . These results indicate that headless and I98A H-mutants efficiently trans-complemented their fusion-deficiencies when assembled into mixed oligomers . In addition , these data confirmed the earlier finding that receptor binding to less than four H-heads per tetramer is sufficient to initiate F-refolding [38 , 66] . We next introduced the N128D and F131Y dual substitutions into the H-I98A background and repeated the trans-complementation assay . In H-wt , these substitutions ablated mAb-1347 binding ( S4B and S4C Fig ) . As anticipated from our earlier finding that mutations D128N and Y131F do not impair H bioactivity ( S3C Fig ) , co-expression of headless H with H-I98A/N128D/F131Y resulted in restoration of membrane fusion activity through trans-complementation ( Fig 7D and 7G ) . Remarkably , fusion activity promoted by functional hetero-oligomeric H-proteins was drastically reduced in the presence of the mAbs ( Fig 7D and 7G ) . As noted previously for MeV H hetero-oligomers , complementation in this setting could be based on mixing of H monomers ( resulting in hetero-dimers ) or dimers ( resulting in homo-dimers/hetero-tetramers ) . To distinguish between these alternatives , we added a supplementary C139A substitution to the H-I98A/N128D/F131Y triple mutant , following the previously established disulfide bond engineering approach to selectively test trans-complementation on the homo-dimer/hetero-tetramer level [38] . This strategy reportedly generates a population of hetero-oligomers with only two available mAb-1347 epitopes , both located in the H stalk 1–159 monomers . Although trans-complementation efficiency under these conditions was reduced , mAb-1347 nevertheless completely blocked membrane fusion ( Fig 7E and 7H ) . Together with the result that mAb-1347 epitopes are only partially accessible in a pre-receptor-bound H conformation ( based on RBE and coIP assays ) but sufficient for blocking viral cell entry ( based on neutralization assays ) , these findings suggest that ( i ) a “less-than-parity” mAb-to-H monomer stoichiometry leads to fusion inhibition , and ( ii ) the H-heads in a putative “heads down” conformation mask some mAb-1347 epitopes .
In the present study , we report the characterization of a monoclonal antibody ( mAb-1347 ) , which targets the C-terminal “spacer” module ( aa 126–133 ) of the CDV H-stalk domain . Binding of this mAb potently inhibited bioactivity of both standard and “headless” CDV H-variants without interfering with either receptor-binding or F-interaction abilities . Remarkably , our data demonstrated that mAb-1347 exhibited an enhanced reactivity not only with head-truncated H-constructs , but importantly also with standard and stalk-elongated CDV H-tetramers after exposure to receptors ( RBE phenotype ) . Although we cannot entirely exclude the possibility that a conformational change in the stalk is triggered by the mAb itself , the fact that enhanced antibody binding activity coincides specifically with receptor binding argues against this hypothesis . Our data therefore spotlight an as yet unrecognized receptor-induced structural rearrangement of the morbillivirus H-protein . Since previously reported “open stalk”-defective H-mutants likewise displayed the RBE phenotype , our findings furthermore reveal the sequence of H conformational changes that are required to translate receptor binding into F-activation . While intracellular assembly of morbillivirus H//F complexes [48 , 67] directly challenged the “stalk exposure/induced fit” hypotheses of paramyxovirus cell entry [40 , 46] , the recently proposed model of membrane fusion activation mediated by CDV [68] and MeV [48] ( referred to as the “safety catch” model in the latter study ) reconciled the different datasets . In the present study , we not only deliver tangible mechanistic insight in support of the safety catch hypothesis , but also substantially extend the model by unravelling the molecular nature governing the dynamics of morbillivirus membrane fusion triggering; prior to receptor-binding , H-tetramers initially fold into an auto-repressed conformational state , where the inherent F-triggering activity of the stalk is silenced by a specific positioning of the head domains . Importantly , we propose that this H auto-repressed state does not necessarily require covering of the F-binding/activation sites in order to prevent untimely H/F interaction and subsequent premature F-activation . Rather , receptor-contact at the cell surface of intracellularly preformed H/F complexes leads to re-positioning of the head domains ( referred to as the “head-stalk” conformational change ) , which consequently releases the conformational lock on the stalk domain . As a result , the stalk is freed to spontaneously undergo the change to the “open stalk” conformation , which is required to trigger the initiation of the F-refolding cascade ( Fig 8A ) . It is tempting to speculate that the proposed auto-repressed state of H resembles the “heads down” configuration obtained for PIV5 and NDV HNs [34 , 36] . Indeed , the physical contacts of the “lower” heads with the membrane-distal stalk region may be key to stabilize the auto-repressed conformational state . In contrast to morbillivirus H-proteins , however , HN-stalks do not display an analogous stalk spacer section located C-terminal to the 4HB portion , but short helices and/or flexible linkers that likely directly connect the globular head domains to the 4HB section [34] . Therefore , variation of the attachment protein stalk length among different members of the paramyxovirus family may govern discrete modes of envelope glycoprotein interactions and mechanisms of membrane fusion activation , despite a very similar overall structural organization . Indeed , attachment proteins characterized by extended ( “spacer-carrying” ) stalks can accommodate intracellular H/F association and thus may depend on the “safety catch” model for F-activation ( Fig 8A ) . Conversely , attachment proteins with a short ( “spacerless” ) stem feature head domains directly covering the F-binding/activation sites , which consequently prevent intracellular F interaction . These attachment proteins may therefore rely on “stalk-exposure” and subsequent “induced fit” models to trigger F [40] ( Fig 8B ) . Viral escape mutants to a fusion-inhibitory synthetic antibody targeting the PIV5 HN-stalk , which presumably blocks interaction with F , consistently exhibited mutations in the candidate F-binding/activation sites , thereby confirming this section as a key functional microdomain in spacerless HN-stalks [69] . The engineered spacerless morbillivirus H-variant and alanine H-mutants spanning the spacer domain were impaired in fusion-promotion . Strikingly , the spacer alanine mutants additionally exhibited a deficiency to properly undergo the RBE phenotype . These data thus spotlight a critical dual function of the morbillivirus H-stalk spacer microdomain: ( i ) residues within this section contribute to the folding of H-tetramers into the correct pre-receptor-bound conformation and ( ii ) they govern proper initiation of the newly identified head-stalk conformational change . These findings thus favor the hypothesis of a critical head-to-spacer interaction , which is instrumental to stabilize the auto-repressed state . Mutations in spacer may modulate the avidity of interactions with the contacting head ( s ) , consequently steadily locking or de-activating the stabilized state . In either case , membrane fusion activation will be inhibited , since H-proteins are either unable to refold or will experience the second “open stalk” conformational change prematurely . Collectively , the C-terminal H-stalk spacer microdomain emerges as an essential molecular determinant controlling the activation and de-activation stages of the auto-silenced pre-triggering conformation of H . It was demonstrated that elongated MeV H-constructs with segments inserted C-terminal to stalk residue 118 remain bioactive [57] . Although their membrane-proximal stem regions ( aa 59–118 ) are constitutively exposed , such mutants still required receptor binding to activate the fusion process . Likewise , the stalk-elongated CDV H-variant generated in this study remained fully bioactive and experienced the receptor-induced head-stalk conformational change to trigger F . In both the MeV and CDV elongated H-constructs , the insertions were located N-terminal to the spacer region . Since both the heads and the functional spacer microdomain were thus moved by the insertion as a single module , we hypothesize that stalk-elongated H-tetramers still efficiently assume the auto-repressed state , which further strengthens the hypothesis of the head-to-spacer interaction exerting an essential role in the spatiotemporal regulation of the membrane fusion process . Liu and colleagues recently proposed a three-step receptor-induced mechanism of NiV G-mediated membrane fusion activation , which interestingly also included a final head-stalk-like conformational change as a critical step leading to F-triggering [49] . Although these data suggest a common mechanism used by henipaviruses and morbilliviruses for cell entry , the authors identified the presumptive head-stalk-like conformational change using a polyclonal antibody raised against an engineered full-length G-stem construct . Hence , in contrast to our work , the microdomain of the G-stalk that might be exposed upon receptor binding remained undetermined , which complicated the interpretation of the model . Different from the morbillivirus safety catch model , the authors suggested a bi-dentate attachment protein/F interaction mechanism to exclude premature F-activation upon intracellular G/F interaction: F-trimers may initially bind to the attachment protein head domains and switch to interaction with the stalks upon receptor binding [49] . While this could be true for the henipaviruses , data obtained for CDV and MeV H-proteins rule out this hypothesis for the morbillivirus fusion complexes . Firstly , H-variants with an elongated stalk ( and demonstrated head-lift ) remained capable of F binding [57] . Secondly , the H-stalk mutant F111A lost binding competence to proteolytically matured F-trimers even in receptor-negative cells [57] . Thirdly , head-truncated H-constructs retained intracellular F interaction [48] . While these findings collectively support a single binding interface governing morbillivirus envelope glycoprotein interactions , additional experiments are needed to determine whether NiV glycoproteins either rely on a bi-dentate mechanism for cell entry , or more closely mimic the morbillivirus entry machinery through intracellular prefusion F binding directly to the G-stalk , which is auto-silenced until receptor contact . Based on EM data , Giu and colleagues recently proposed that HPIV3 HN/F complexes can form prior to receptor engagement and that the heads up configuration itself is insufficient to activate F [70] . Upon receptor contact , the HN-heads may transmit a signal for F-activation , possibly through an oligomerization mechanism . While possible , this hypothesis is largely supported by data from the same group suggesting clustering of preformed HN/F complexes upon receptor engagement [58 , 59] . At present , however , direct experimental support is lacking to confirm that monomeric , dimeric or tetrameric HN-structures indeed switch to higher order oligomers in a receptor binding-dependent manner to achieve complexes productive for F-triggering . In conclusion , our data highlight two sequential conformational changes in morbillivirus H-tetramers that together constitute the mechanistic core of the molecular link between receptor binding and F-triggering . Additionally , our findings provide first molecular evidence that both of these rearrangements are triggered upon receptor-induced de-activation of an auto-repressed conformational state of the H-stalk assumed prior to receptor binding . This locked state likely originates from a critical head-to-spacer interaction that allows intracellular H/F assembly , but prevents premature F-activation . Lastly , beyond having identified the spacer module as the key element in the H-stalk coordinating the dynamics of the safety catch mechanism , our data further reveal that despite possible strong overall structural conservation between different paramyxovirus envelope glycoproteins , the presence or absence of the spacer domain in the attachment protein stalk emerges as a key indicator for the F-triggering strategy applied by individual paramyxovirus family members .
293T cells ( ATCC ) , Vero ( ATCC CCL-81 ) and derivative Vero cells expressing the canine SLAM ( Vero-cSLAM , kindly provided by Yusuke Yanagi , Kyushu University , Japan ) or the Nectin-4 ( Vero-Nectin-4 , [62] ) receptors were grown in Dulbecco's modified Eagle's medium ( Gibco , Invitrogen ) with 10% fetal calf serum at 37°C in the presence of 5% CO2 . Cells were transfected using TransIT-LT1 ( Mirus ) . The MVA-T7 recombinant vaccinia virus was used for a quantitative cell-cell fusion assay and was kindly provided by B . Moss , NIH , Bethesda , MD . The recombinant A75/17-CDV , containing an additional RFP gene ( recA75/17red ) , was amplified and titrated in Vero-cSLAM cells . All single ( and multiple ) substitutions performed in pCI-CDV-H and pCI-CDV-F ( derived from hemagglutinin of the A75/17 CDV strain [71] ) were obtained using the Quick Change lightning site-directed mutagenesis kit ( Stratagene ) . Mutations were also performed in measles virus attachment protein-expression vectors ( pCI-MeV-H derived from the ICB323 and Edmonston strains ( kindly provided by Jürgen Schneider-Schaulies , Würzburg , Germany ) ) . The pCI-H301F vector was previously described and encodes the hemagglutinin protein derived from a fox-isolated CDV strain ( W10/301/red-fox/Ch/2010-JF810106 and referred in this study to as 301F ) [72] . To produce a soluble form of H , the complete ectodomain was cloned in-frame with the IgK signal peptide . In addition , to maximize the prospect of preserving the H-tetramer oligomeric state in its soluble form , the GCN4 peptide was fused N-terminally as well as a hexahistidine tag ( 6xHis ) [31] . FLAG-tag insertions ( DYKDDDK ) at the C-terminal region of the CDV H-proteins were performed by site directed mutagenesis , using the same kit as described above . All primers are available upon request . Vero cells , in 6-well plates at 90% confluency , were co-transfected with 2 μg of different pCI-F constructs , 1 μg of the various pCI-H plasmids with 9 μl of TransIT-LT1 ( Mirus ) . Vero cells in 24 wells were transfected with 1μg of the various pCI-H plasmids with 3 μl of TransIT-LT1 ( Mirus ) . All transfections were performed according to the manufacturer’s protocol . In some experiments , phase contrast pictures were taken 24 h post-transfection with a confocal microscope ( Olympus , Fluoroview , FV1000 ) . The quantitative fusion assay was performed as described previously [73 , 74] with subtle modification . Briefly , Vero cells were co-transfected with the F and H expression plasmids and 0 . 1 μg of pTM-Luc ( kindly provided by Laurent Roux , University of Geneva ) . In parallel , separate 6-well plates of Vero-cSLAM cells at 30% confluency were infected with MVA-T7 at a multiplicity of infection ( MOI ) of 1 . After overnight incubation , both cell populations were mixed and incubated at 37°C . 2 . 5 hours later , the cells were lysed using Bright Glo Lysis Buffer ( Promega ) , and the luciferase activity was determined using a luminescence counter ( PerkinElmer Life Sciences ) and the Britelite reporter gene assay system ( PerkinElmer Life Sciences ) . 293T cells were transfected with 10μg of various expression vectors . 72h post-transfection , the supernatant was harvested and concentrated using 10 kDa cutoff filtration columns ( Millipore ) . Subsequently , equal aliquots of supernatants were immunoprecipitated for 2 hours with either mAb 1347 , 1C42 or 2267 ( 1:1000 ) [61] or anti-histidine mAbs ( AbD Serotec ) ( 1:500 ) . This was followed by adding protein G-Sepharose beads overnight ( GE-Healthcare ) and subsequently fractionated in 10% NuPAGE Bis-Tris Mini gels ( life technologies ) under regular reducing conditions . Immunoprecipitated H-proteins were finally revealed by western blotting , as described , using a polyclonal anti-H antibody or anti-GCN4 antibody ( Santa Cruz Biotechnology ) ( 1:1000 ) . Western blots were performed as previously described [29] . After overnight incubation on a rotor at 4°C , the lysates were cleared by centrifugation for 3min at 4°C , and washed three times with RIPA ( 10mM Tris , pH7 . 4 , 150mM NaCl , 1%deoxycholoate , 1% Triton X-100 , 0 . 1% sodium dodecyl sulfate ( SDS ) ) containing protease inhibitor ( Roche , complete mix ) . The supernatant was mixed with an equal amount of 2x Laemmli sample buffer ( Bio-Rad ) containing 5% β-mercaptoethanol , subsequently boiled at 90°C for 5 min and fractionated on NuPAGE Bis-Tris Mini gels ( life technologies ) under regular reducing conditions . Separated proteins were transferred to nitrocellulose membranes by electroblotting . The membranes were then incubated with the polyclonal anti-CDV-H , anti-CDV-F [71] or anti-GCN4 antibody ( Santa Cruz Biotechnology ) ( 1:1000 ) . Following incubation with a peroxidase-conjugated secondary antibody , the membranes were subjected to enhanced chemiluminescence ( ECL ) kit ( Amersham Pharmacia Biotech ) according to the manufacturer's instructions . CDV F/H co-immunoprecipitations , initially developed by Paal and colleagues [57] , were performed as previously described [28] with the following modifications . Vero cells in a 6-well plate format were transfected in duplicate with 2μg of CDV H-protein expressing plasmids ( or derived variants ) and 2μg of CDV F-protein expressing plasmids . 24h post-transfection , the cells were washed with cold PBS and treated with DTSSP ( 1mM final concentration in PBS , ProteoChem ) . Cells were subsequently lysed in RIPA buffer ( 10mM Tris , pH 7 . 4 , 150mM NaCl , 1% deoxycholate , 1% Triton X-100 , 0 . 1% sodium dodecyl sulfate ( SDS ) ) containing protease inhibitor ( Roche , complete mix ) . Cleared lysates ( 20 , 000 X g; 20 min , 4°C ) were incubated 2–4 hours with the indicated monoclonal antibody ( 1:500 ) , followed by overnight incubation with immunoglobulin G-coupled Sepharose beads ( GE-Healthcare ) . The samples were then subjected to Western blot analysis as described above using either a polyclonal anti-HA antibody ( Covance ) , a polyclonal rabbit anti-F antibody or a polyclonal anti-H antibody . Vero cells were transfected with 1μg of various H-expressing proteins alone or combined with 1μg F-expressing DNA plasmids . One day post transfection , unfixed and unpermeabilized cells were washed twice with cold phosphate buffered saline ( PBS ) and subsequently stained with the various antibodies ( 1:1000 ) for 1h at 4°C . The anti-CDV F mAbs 4941 and 3633 , anti-CDV H mAb 1347 , anti-FLAG mAb ( Sigma Aldrich ) or anti-HA mAb ( Covance ) ( 1:1000 ) were employed . This was followed by washes with cold PBS and incubation of the cells with Alexa-fluor 488-conjugated secondary antibody ( 1:500 ) for 1h at 4°C . Cells were subsequently washed 2 times with cold PBS and consequently detached from the wells by adding PBS-EDTA ( 50μM ) 20min at 37°C . The mean fluorescence intensity ( MFI ) of 10’000 cells was then measured by using a BD LSRII flow cytometer ( Becton Dickinson ) . 10μg of a soluble HA-tagged SLAM protein-expressing plasmid ( of canine origin ) was transfected in 293T cells seeded in petri dishes . Four days later , 10ml of the supernatant was harvest and concentrated using 10 kDa cutoff filtration columns ( Millipore ) . Then , H-expressing receptor-negative Vero cells were incubated 1h at 4°C with concentrated soluble SLAM molecules ( at a dilution of 1:20 ) , followed by two PBS washes and treatment with monoclonal anti-HA ( 1:1000 ) ( Covance ) 1h at 4°C . Cells were then incubated with Alexa-fluor 488 conjugated secondary antibody ( 1:500 ) ( Invitrogen ) 1h at 4°C and were submitted to flow cytometry as described above . Relative SLAM-binding activity of the H-proteins was determined by MFI values recorded with the anti-HA antibody ( SLAM-binding efficiency ) that were normalized to MFI values obtained with anti-FLAG antibody ( cell surface expression ) of each H variants . Values were finally normalized to the H-wt/SLAM binding efficiency ( set at 100% ) . One hundred infectious units of recA75/17red ( titer 4 , 7x106 IU/μl ) were incubated with the indicated dilution of the neutralizing mAb-1347 for 1h on ice . The virus-antibody mixture was then added to Vero cells and plates were incubated at 37°C . Four hours post-infection , medium was replaced by DMEM containing 1% agar . The numbers of syncytia were counted under a fluorescence microscope 72 hours of post-infection . Values were then normalized to recA75/17red-induced number of syncytia obtained in the absence of any mAb treatments ( set at 100% ) . | With the ultimate aim to develop pan-morbillivirus fusion inhibitors , we here characterized a potent neutralizing monoclonal antibody . The antibody recognizes the ectodomain of the membrane-bound tetrameric attachment ( H ) protein , which together with the fusion protein and a host cell receptor executes plasma membrane fusion to deliver the viral genetic information into the cell . The H-ectodomain consists of a short F-binding/activating stalk region supporting receptor-binding head domains . Molecular characterization of the identified mAb epitope ( which locates in the membrane-distal stalk module called “spacer” ) , enabled us to unravel two sequential conformational changes occurring in CDV H-tetramers that stand at the core of the molecular mechanism translating receptor binding to F-triggering . We additionally propose that both rearrangements are triggered upon receptor-induced “de-activation” of an auto-repressed state assumed by H prior to receptor binding . This locked state enables H/F interaction while preventing premature F-activation . Furthermore , although paramyxovirus attachment proteins may fold into very similar “pre-receptor-bound” conformational states , the presence of the “spacer” module in the stalk emerges as a key determinant leading to distinct mechanisms of membrane fusion triggering . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Sequential Conformational Changes in the Morbillivirus Attachment Protein Initiate the Membrane Fusion Process |
Sepsis is characterized by a dysregulated host-pathogen response , leading to high cytokine levels , excessive coagulation and failure to eradicate invasive bacteria . Novel therapeutic strategies that address crucial pathogenetic steps during infection are urgently needed . Here , we describe novel bioactive roles and therapeutic anti-infective potential of the peptide EDC34 , derived from the C-terminus of tissue factor pathway inhibitor-2 ( TFPI-2 ) . This peptide exerted direct bactericidal effects and boosted activation of the classical complement pathway including formation of antimicrobial C3a , but inhibited bacteria-induced activation of the contact system . Correspondingly , in mouse models of severe Escherichia coli and Pseudomonas aeruginosa infection , treatment with EDC34 reduced bacterial levels and lung damage . In combination with the antibiotic ceftazidime , the peptide significantly prolonged survival and reduced mortality in mice . The peptide's boosting effect on bacterial clearance paired with its inhibiting effect on excessive coagulation makes it a promising therapeutic candidate for invasive Gram-negative infections .
Sepsis is a major cause of death in the western world , with mortality ranging between 30 and 70% [1] . The disease is characterized by an excessive and dysregulated immune and coagulation response to microbial infections , leading to capillary leakage , lung damage , and finally multiple organ failure [2] , [3] , [4] . Several clinical trials targeting coagulation as well as pro-inflammatory responses have been conducted , including administration of activated protein C [5] , [6] , antibodies against TNF-α [7] , [8] , [9] , interleukin-1 receptor antagonists [10] , [11] , interleukin-6 antagonists [12] , anti-endotoxin antibodies [12] , PAF receptor antagonists [13] , antithrombin III [14] , [15] , [16] and other agents [17] , [18] , [19] , [20] . Even though animal experiments showed promising results , all drug candidates tested so far have failed to show clinical efficiency . Consequently , today's treatment for sepsis is largely based on antibiotics in combination with supportive measures , illustrating the need for new therapeutic approaches . Antimicrobial peptides constitute one group of compounds , which have recently attracted attention as new anti-infectives . Due to their preferential interactions with prokaryotic and fungal membranes , these peptides provide a rapid and broad-spectrum response towards both Gram-negative and Gram-positive bacteria , as well as fungi [21] , [22] , [23] , [24] . Antimicrobial peptides also mediate diverse immunomodulatory roles , including anti-inflammatory effects as well as stimulation of chemotaxis , chemokine production , wound healing and angiogenesis [25] , [26] , [27] , motivating the term host defense peptides ( HDP ) . The molecular diversity of HDPs has provided a wide range of structures of potential interest for the anti-infective field . For example , immunomodulatory peptides such as IDRs ( Innate Defense Regulator ) selectively protect against bacterial infection by chemokine induction and neutrophil recruitment , while reducing pro-inflammatory cytokine responses [26] , [27] . Further , the lactoferrin-derived peptide hLF1-11 differentiates monocytes , which enhances clearance of pathogens , a feature currently utilized in the development of therapies for infections in immunosuppressed patients [28] . Additionally , C-terminal peptides of human thrombin , exerting anti-inflammatory , anti-coagulative and antimicrobial effects , are effective in ameliorating LPS-induced shock and Pseudomonas sepsis in experimental settings in animals [29] , [30] , [31] , further exemplifying that endogenous HDPs may have therapeutic potential . The present work is based on the finding that the highly positively charged C-terminus of tissue factor pathway inhibitor-2 ( TFPI-2 ) encodes for antimicrobial activity [32] . Previous data demonstrated that C-terminal TFPI-2 fragments are released in human wounds , and can be generated by neutrophil elastase in vitro . A direct antimicrobial effect of a prototypic 34 amino acids long C-terminal TFPI-2 peptide , EDC34 was furthermore demonstrated in vitro [32] . In this work , utilizing various in vitro and in vivo models aimed at characterizing effects on complement , coagulation , and bacterial clearance , we demonstrate a therapeutic potential of the peptide for the treatment of Gram-negative infections .
Initial experiments utilizing E . coli and P . aeruginosa isolates demonstrated that EDC34 displayed significant antibacterial activity in physiological buffer , and enhanced activity in the presence of human plasma ( Figure 1A and Figure S1A and B ) . Heat inactivation of human plasma abolished this potentiating effect . A control peptide ( DAA14 ) derived from the N-terminal part of TFPI-2 did not show any antimicrobial effects in buffer or plasma ( data not shown ) . In contrast , the cathelicidin LL-37 was partially inhibited by the addition of native as well as heat-inactivated plasma ( Figure S1A and B ) , which was compatible with previous observations showing that the peptide's activity is compromised in presence of plasma due to protein binding [33] . Thus , these data suggested that the antibacterial effect of EDC34 is dependent on additional bactericidal systems in normal plasma , such as complement . As described for LL-37 , it was observed that EDC34 showed reduced antibacterial activity in heat-inactivated plasma , particularly against the E . coli strains 25922 , 49 . 1 and 47 . 1 when compared to buffer ( Figure 1A and Figure S1A ) , possibly related to scavenging effects of plasma proteins . In human whole blood , EDC34 demonstrated potent antimicrobial effects against various E . coli and P . aeruginosa isolates at doses as low as 0 . 3 µM , except for P . aeruginosa PA01 , which was killed at 3 µM ( Figure S1C and D ) . Notably , 10–100 times higher doses were required for LL-37 mediated killing ( Figure S1C and D ) . Kinetic studies in the presence of plasma demonstrated that the bacterial killing mediated via EDC34 ( 3 µM ) occurred within 10–20 min for E . coli , and 40–60 min for P . aeruginosa ( Figure S2A and B ) . Contrary to results for a complement-susceptible E . coli strain ( Figure 1A , left panel ) , no EDC34-mediated enhancement of bacterial killing was observed for the E . coli O18:K1 strain , previously shown to be resistant to complement-mediated lysis [34] ( Figure 1A , right panel ) . Next , antibacterial and hemolytic effects of EDC34 were simultaneously analyzed in human blood . Consistent with the complement-dependent action of EDC34 , this peptide was active against Gram-negative E . coli and P . aeruginosa in whole blood , while exhibiting little or no effects against Gram-positive S . aureus and S . pyogenes AP1 ( Figure 1B , left panel ) . No significant increase in hemolysis was observed at peptide doses up to 120 µM ( Figure 1B , right panel ) . Next , the fact that the human EDC34 sequence differs from the related murine sequence prompted us to compare the effects of human EDC34 with those of the related peptide of murine origin ( DAC31 ) , derived from a corresponding C-terminal region of murine TFPI-2 ( Table S1 ) . Both peptides exerted similar antimicrobial effects in buffer as well as in human and mouse plasma ( Figure 1C ) . Taken together , the data imply that bacterial killing by EDC34 in plasma and blood is enhanced by the presence of an intact complement system . In order to study the complement boosting effect of EDC34 further , western blot and FACS analyses were used . EDC34 enhanced binding of C1q to E . coli ( Figure 2A ) , and increased the formation of the membrane attack complex ( MAC ) , as evidenced by increased binding of C9 and related high molecular weight compounds ( Figure 2A ) . A significant generation of C3a was also observed after addition of EDC34 ( Figure 2B–D and Figure S3 ) . After subjecting P . aeruginosa to plasma , in contrast to the results with E . coli above , an activation of complement by the bacteria per se was observed ( Figure 2A–D and Figure S3 ) . Nevertheless , quantification of C1q after addition of EDC34 detected an increase of this molecule in association with P . aeruginosa ( Figure 2C and Figure S2 ) . Additionally , fragments corresponding to bacterial-bound C3a , as well as peptides of higher molecular weight containing the C-terminal epitope of C3a , were detected , particularly after addition of EDC34 ( Figure 2B–D ) . Electron microscopy studies on fibrin slough from a patient with a chronic wound infected by P . aeruginosa were furthermore performed to explore a possible co-localization between the C-terminal TFPI-2 region and C3a in vivo . Using immunogold-labeled antibodies against the C-terminal part of TFPI-2 and against C3a , the evaluation of 50 bacterial profiles indeed showed that ∼70% of C-terminal TFPI-2 peptide epitopes were closely associated with C3a ( Figure 2D ) . Taken together , these results indicate that EDC34 promotes complement activation in response to E . coli and P . aeruginosa . TFPI-2 inhibits TF-VII-mediated coagulation and affects factor XIa and plasma kallikrein [35] , [36] , effects which are thought to be mediated by the Kunitz-type domains of TFPI-2 [37] . However , no evidence has so far been presented that the C-terminal region of TFPI-2 may influence coagulation . Clotting assays using human citrate plasma revealed that EDC34 interfered with the intrinsic pathway of coagulation , as illustrated by a dose-dependent prolongation of the activated partial thromboplastin time ( aPTT ) ( Figure 3A ) . At 50 µM , the increase in aPTT was about 4-fold in human plasma ( Figure 3A ) and 20-fold in mouse plasma ( Figure 3C ) . In contrast , the other parts of the coagulation system , such as the extrinsic pathway of coagulation , monitored by measuring the prothrombin time ( PT ) , and thrombin-induced fibrin-network formation ( thrombin clotting time; TCT ) , were not affected or to a lower extent at doses of 20–50 µM ( Figure 3B and C , and data not shown ) . The activation of the intrinsic system takes place at negatively charged surfaces , such as kaolin or bacteria , and involves activation of FXII , which then leads to the activation of plasma kallikrein ( PK ) and FXI [38] . Analysis of PK activity at the surface of kaolin showed that EDC34 , but not the control peptide from the N-terminal region of TFPI-2 ( DAA14 ) , was able to block the PK activity ( Figure 3D ) . In order to determine whether EDC34 inhibits PK activity also on bacterial surfaces , P . aeruginosa and E . coli bacteria , sensitive to the peptide in plasma and blood , were chosen for further studies . Bacteria were pre-incubated with EDC34 , followed by incubation with plasma , and the effect of EDC34 and the control peptide recorded by measuring the PK activity . The results showed that only EDC34 was able to inhibit PK activation ( Figure 3E ) . Since high molecular weight kininogen ( HMWK ) is a substrate for PK [38] , western blots were utilized to assess possible degradation of HMWK in plasma . Compatible with the results of the PK assay , EDC34 blocked the degradation of HMWK ( Figure 3F ) . PK-cleaved HMWK releases bradykinin ( BK ) , a potent pro-inflammatory mediator [38] , and corresponding to the above , EDC34 also significantly inhibited the generation of bradykinin ( Figure 3G ) . The N-terminal TFPI-2 control peptide DAA14 neither inhibited PK activation nor bradykinin generation ( Figure 3D–G ) . These results show that EDC34 mainly inhibits activation of the intrinsic pathway of coagulation , leading to reduced HMWK degradation and bradykinin release . To assess possible anti-coagulative effects in vivo , 12 mg/kg of LPS was injected into the mice followed by intraperitoneal administration ( i . p . ) of 0 . 5 mg of EDC34 after 30 min . Measurements of whole blood clotting , aPTT , PT , and TCT at 4 h and 8 h after LPS challenge clearly demonstrated an anti-coagulant effect of EDC34 in this model ( Figure 4A ) . This effect was confirmed by determining thrombin-antithrombin complexes ( TAT ) in mouse plasma 8 h after LPS injection . TAT complexes were reduced in EDC34-treated animals ( Figure 4B ) . Previous studies have demonstrated that EDC34 binds LPS similarly to LL-37 , compatible with its direct antibacterial effects on bacteria [32] . To test whether LPS-binding confers any anti-inflammatory effects to EDC34 , mouse macrophages were stimulated with LPS in presence or absence of EDC34 . GKY25 , a thrombin-derived LPS-binding and anti-inflammatory peptide , was used as positive control [29] , [31] . In contrast to GKY25 , which blocked the LPS response at 1–5 µM , EDC34 did not exert any endotoxin-blocking effects , even at significantly higher concentrations ( Figure S4 ) . Previously published work showed that GKY25 ( at 0 . 5 mg ) significantly improved survival as well as inhibited cytokine responses in a mouse model of endotoxin shock [29] , [31] . In contrast , and corresponding to the absence of anti-inflammatory effects in vitro , the same dose ( 0 . 5 mg ) of EDC34 , when administrated i . p . 30 minutes after endotoxin exposure , was not able to block IL-6 , MCP-1 , and TNF-α responses , although it affected the production of the anti-inflammatory IL-10 at an early time point ( Figure 4C ) , and reduced IFN-γ after 8 h . Thus , these results show that EDC34 largely modulates coagulation during LPS-induced shock in vivo . Taken together , the above data provided a rationale for using the human peptide EDC34 in murine Gram-negative infection models . Hence , mice were infected i . p . with E . coli , followed by i ) immediate i . p . treatment with EDC34 , ii ) delayed i . p . treatment with the peptide given after 1 h , or iii ) s . c . treatment after 1 h , in order to separate peptide and bacteria , and minimize direct peptide effects during i . p . administration . Notably , in these models , immediate as well as delayed EDC34 treatment administered i . p . or s . c . yielded significant improvements in survival when compared to controls ( Figure 5A ) . Although all treated mice displayed a significant weight loss , it was observed that the i . p . treated animals recovered faster ( Figure 5B ) . The i . p . model using immediate treatment was selected in order to evaluate peptide effects in greater detail . Mice were infected i . p . with E . coli , followed by treatment with EDC34 . Two hours post-infection , a significant reduction of bacterial levels in the peritoneal cavity was detected only in peptide-treated animals ( Figure 5C ) . To assess the importance of complement activation for the observed antibacterial effects of EDC34 in vivo , C3a levels were measured in peritoneal fluid and plasma ( Figure S5A and B ) . In peritoneal fluid , treatment with EDC34 indeed yielded increased C3a levels , in particular after 2 hours post-infection ( Figure S5A ) . These results were compatible with the observed reduction of bacterial counts in the peritoneum of these EDC34-treated animals ( Figure 5C ) . Concomitantly , C3a levels in plasma of peptide-treated animals were reduced , reflecting the local boosting effect of EDC34 at the site of infection leading to overall reduced bacterial levels and hence , relatively less complement activation systemically ( Figure S5B ) . The importance of an intact complement system was further demonstrated in experiments employing pretreatment with cobra venom factor ( CVF ) , used in order to deplete animals of complement factors [39] before infection and peptide treatment . In this infection model , the antibacterial effect of EDC34 was significantly compromised in CVF-treated mice when compared to control mice ( Figure S5C ) . Taken together , these experiments are compatible with the previous in vitro experiments presented in Figure 2 , and firmly demonstrate that EDC34 mediates its antibacterial effect in vivo by boosting of complement activation . Having shown this , a second set of in vivo experiments with E . coli , aimed at studying peptide effects on bacterial dissemination , cytokines , and coagulation parameters after 2 , 4 , and 8 h post-infection were performed . The results showed that treatment with EDC34 yielded significantly lower bacterial levels in the spleen , kidney , liver and blood when compared to the controls ( Figure 5D ) . Notably , infected mice showed a reduced clotting capacity and exhibited prolonged aPTT and PT times , along with an increase in TAT complexes in their plasma . Treatment with EDC34 however , resulted in an improved clotting function , as evidenced by reduced coagulation times ( Figure 5E ) and TAT complex formation ( Figure 5F ) . These data implied that the peptide , by enhancing bacterial clearance and blocking bacteria-induced coagulation , reduced the excessive consumption of coagulation factors in this animal model , thus improving hemostasis function . It is of note that time-dependent cytokine responses were detected also in EDC34 treated animals , although at lower levels when compared with non-treated infected animals . ( Figure 5G ) . However , this reduction of cytokine responses was not unexpected , since the bacterial levels were reduced 10–100 times by EDC34 ( Figure 5D ) . Hence , the lower , but retained cytokine response was compatible with the noted absence of significant anti-inflammatory effects of EDC34 in the LPS-shock model above ( Figure 4C ) . To further investigate the functions of EDC34 in a mouse model of Pseudomonas-induced sepsis , two strains of P . aeruginosa , PA01 and 15159 ( the latter a clinical isolate ) , were used . The bacteria were injected i . p . , and the peptide was immediately administered either by i . p . injection ( 1×1 ) or s . c . , either 1 h ( 1×1 ) or 1 and 7 h ( 1×2 ) after bacterial injection . EDC34 yielded significant improvements in survival after immediate treatment ( Figure 6A ) , but failed to improve survival after delayed treatment ( not shown ) . Corresponding to the survival results , a reduced number of bacteria was observed in spleen , kidney , and liver compared to control mice , for both strains , after immediate treatment with EDC34 ( Figure 6B ) . Delayed treatment s . c . yielded no significant reduction of bacterial levels . However , and compatible with the peptide's anti-coagulative actions in vitro and in vivo , scanning electron microscopy ( SEM ) analyses of the lungs from mice infected with P . aeruginosa PAO1 showed that EDC34 , irrespective of administration route , reduced fibrin deposition and pulmonary leakage of proteins and red blood cells ( Figure 6C ) . Although EDC34 lowered initial bacterial levels after i . p . administration , the peptide did not completely eradicate bacteria in the above infection models , particularly noted for the clinical isolate . The activation of the intrinsic coagulation system by bacteria underlies the excessive coagulation and bradykinin-induced vascular leakage , and is therefore of interest to target during P . aeruginosa infection . Since antibiotics do not inhibit these bacterial effects on coagulation , we decided to explore the multiple effects of EDC34 in combination with ceftazidime , an antibiotic often used in Gram-negative infections . In an infection model mimicking a potential clinical situation , bacteria were injected i . p . , followed by treatment 90 min and 4 . 5 h after bacterial challenge with either ceftazidime ( 300 mg/kg ) or EDC34 ( 0 . 5 mg s . c . ) alone , or the antibiotic in combination with EDC34 ( doses as above ) . In contrast to animals treated with ceftazidime alone , the combination treatment significantly improved survival of the animals ( Figure 6D ) . In contrast , the peptide alone did not increase survival in this model ( not shown ) . Further , bacterial levels were monitored in the spleen , kidney , and liver . Ceftazidime-treated animals had 100-fold less bacterial levels , which were further slightly reduced after addition of EDC34 ( Figure 6E ) , while EDC34 alone did not significantly reduce bacteria . Nevertheless , in spite of similar bacterial levels ( Figure 6E ) , the generation of TAT complexes was reduced in EDC34-treated mice compared to controls ( Figure 6F ) , and thrombocyte levels were increased after peptide treatment . This increase in thrombocytes was also noted in combination with the antibiotic , indicating a reduced activation of the coagulation system due to peptide treatment ( Figure 6G ) . This observation was supported by reduced fibrin deposition and pulmonary leakage of protein and red blood cells in the lungs of animals treated with EDC34 alone , or in combination with ceftazidime as judged by SEM ( Figure 6H ) and confirmed by histochemistry and scoring of lung damage ( Figure 6I and Figure S6 ) . It was noted that animals treated with ceftazidime alone showed lung damage changes , involving reduction of alveolar space , increased cell infiltration and wall thickness , and formation of thrombi , similar to those observed in infected untreated animals ( Figure 6I and Figure S6 ) . EDC34 did not significantly reduce cytokine levels when given alone . Nevertheless , cytokine levels were reduced after treatment with the antibiotic as well as with the combination with EDC34 ( Figure 6J ) . EDC34 at a total dose of 2 mg did not significantly affect coagulation ( aPTT , PT , TCT ) , thrombocytes , and cytokines after 12 h when administrated into healthy mice ( Figure S7 ) .
The development of novel anti-infective treatments has been largely addressing the bacteria itself , or the subsequent dysregulated host response , while efforts to control the dysregulated host response have so far failed . Therefore , there is a need for new therapies that address new targets in the complex interplay between microbes and the host . HDPs with multiple roles , such as those here defined for the TFPI-2-derived EDC34 , targeting both bacteria and coagulative mechanisms , should therefore be of therapeutic interest . The main objective in this work was to clarify bioactive effects of EDC34 in vitro , and to what extent these can be utilized in anti-infective therapy in vivo . Challenging in this perspective was the fact that the in vitro effects , as observed in isolated experimental systems , were not independent of each other in vivo . For example , reduced bacterial levels could also be linked to concomitant reductions in pro-inflammatory cytokines in vivo . In such cases , an attempt was made to reduce complexity , and to study isolated factors or events , such as the peptide's anti-inflammatory effects in vitro in relation to LPS-stimulation of macrophages , or in vivo , in relation to endotoxin shock . In other cases , we highlight unique and peptide-dependent effects , such as abolished fibrin deposition irrespective of bacterial load and antibiotic usage in the animal models , clearly linking the anti-coagulative effects in vitro including blocking of bacteria-induced kallikrein activation and bradykinin release , to those observed in vivo . Furthermore , interference by EDC34 with the coagulation system in vivo was demonstrated not only by reduced fibrin deposition but also evidenced by reduced TAT levels , along with increased thrombocyte levels . This is of importance , since sepsis- and coagulation-related acute lung injury is considered to be a critical feature compromising the clinical outcome during sepsis [40] , [41] , [42] . In response to LPS challenge , the coagulation cascade is activated , leading to an excessive activation of the coagulation system , followed by consumption of coagulation factors in the blood resulting in prolonged clotting times [43] . In line with this , LPS-injected mice showed a reduced clotting capacity and exhibited prolonged aPTT and PT times in their plasma ( Figure 4 ) . Treatment with EDC34 however , resulted in a partially normalized clotting function , as evidenced by reduced coagulation times ( Figure 4 ) . Hence , these data showed that the peptide , by blocking coagulation as shown in Figure 3 , reduced the excessive consumption of coagulation factors in this animal model . Also of importance , was the observation that no signs of bleeding or prolongation of coagulation was noted in animals treated with EDC34 only ( Figure S7A and B ) . This should be of value , since other agents tested in the clinic , such as activated protein C , mainly affect the extrinsic or primary pathway of coagulation , and thus , may increase the risk of bleeding complications [5] , [6] . As mentioned above , EDC34 did not present any significant anti-endotoxin effects in vitro or in vivo , in spite of its binding to LPS [32] . Such absence of direct anti-inflammatory effects may be advantageous , particularly in situations where an anti-coagulative action is of importance , while maintaining a normal LPS immune response . The complement system is crucial for bacterial clearance , indicating that strategies based on manipulating complement activation and thus bacterial clearance , may be of therapeutic interest . It is therefore notable that EDC34 boosts complement activation rapidly ex vivo in relation to the Gram-negative E . coli and P . aeruginosa . Thus , E . coli did not significantly induce activation of the classical complement pathway in absence of EDC34 in plasma in vitro , and the boosting effects were dependent on EDC34-binding to the bacteria . Although EDC34 was active against Gram-positive S . aureus in low salt buffer conditions [32] , the results in human blood , where EDC34 was particularly active against E . coli and P . aeruginosa , further underscores the importance of the peptide's dependence of an active complement system for its actions in the in vivo models . Notably , EDC34-mediated formation of C3a , an anaphylatoxin exerting antimicrobial effects [44] , was increased not only in vitro , but also locally at the site of infection in the animal E . coli infection model . In this context , it was interesting to note that although C3a was increased intraperitoneally , the overall levels of C3a in blood of the animals were reduced after peptide treatment . At a first sight , this may appear paradoxical , however , it is not unexpected when considering the significant reduction of bacteria systemically in peptide-treated animals . Also relevant in this context is that analyses of levels of C3 in E . coli infected animals showed that the total levels of this complement factor were unaffected after treatment i . p . with EDC34 ( not shown ) , indicating that the modulation takes place locally , and that the total levels of C3 may buffer a potential local , initial consumption of the protein in this particular animal model . Furthermore , animals with a compromised complement function due to CVF treatment [39] did not exhibit a boosting of E . coli clearance by EDC34 in vivo . These results were compatible with the in vitro studies using heat-inactivated ( hence complement-inactivated ) plasma ( Figure 2 ) , elegantly demonstrating the importance of an intact complement system for the action of EDC34 in vitro and in vivo . From a biological perspective , it remains to be investigated whether TFPI-2 adds to host defense in vivo . Although these studies are currently initiated and TFPI-2−/− mice have been generated and are currently validated and characterized , this work is clearly beyond the scope of this present work aimed at utilizing the TFPI-2 peptide as a therapeutic molecule . Nevertheless , it is notable that emerging evidence suggests an involvement of TFPI-2 in host defense . Thus , in vitro , stimulation of human endothelial cells with inflammatory mediators such as LPS , and TNF-α significantly increases TFPI-2 expression [45] . Analogously , in vivo in a murine model , TFPI-2 expression is dramatically upregulated in the liver and in the lungs during LPS stimulation [46] , [47] . These data , together with previous findings on release of C-terminal fragments by neutrophil elastase and their presence in human wounds [32] , and as shown herein , particularly in association with bacteria ( Figure 2D ) , suggest that TFPI-2 fragments may exert physiological roles during infection . From a clinical perspective infections with Gram-negative bacteria such as E . coli and P . aeruginosa contribute to morbidity and mortality during abdominal surgery , peritoneal dialysis , burn wounds or in immunocompromized patients [48] , [49] , [50] , [51] . An initial phase of effective bacterial clearance is crucial for patient outcome , and therefore , approaches based on substitution with EDC34 in combination with antibiotics may be of therapeutic interest . In order to verify such possibilities , experimental models are crucial , and it is notable that EDC34 exerted similar antibacterial effects in mouse and in human plasma , which should facilitate further development of experimental models incorporating readouts involving lung and organ damage . From a toxicological perspective , many HDPs cause side effects involving damage to eukaryotic cell membranes . It is therefore notably that EDC34 did not show toxicity at therapeutic doses in vitro [32] , that the peptides complement boosting activity was only recorded in presence of , and in relation to , bacteria , and that EDC34 was well tolerated in non-infected animals , neither significantly affecting coagulation parameters nor thrombocyte and cytokine levels ( Figure S7 ) . In summary , the effects of EDC34 , involving blocking of contact activation , and bacterial killing directly or by boosting of complement activation , while maintaining a functional cytokine response ( Figure S8 ) , represents a potential new approach for boosting bacterial clearance while inhibiting the deleterious effects of excessive pro-coagulative responses during infection with Gram-negative bacteria .
The use of human blood was approved by the Ethics Committee at Lund University , Lund , Sweden ( Permit Number: 657-2008 ) . Written informed consent was obtained from the donors . The animal experiments were conducted according to national guidelines ( Swedish Animal Welfare Act SFS 1988:534 ) and were approved by the Laboratory Animal Ethics Committee of Malmö/Lund , Sweden ( Permit Numbers: M228-10 , M252-11 ) The peptides EDC34 ( EDCKRACAKALKKKKKMPKLRFASRIRKIRKKQF ) , DAC31 ( DACHRACVKGWKKPKRWKIGDFLPRFWKHLS ) and DAA14 ( DAAQEPTGNNAEIC ) were synthesized by Biopeptide Co . , San Diego , CA , whereas LL-37 ( LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES ) was obtained from Innovagen AB , Lund , Sweden . The purity ( >95% ) of these peptides was confirmed by mass spectral analysis ( MALDI-ToF Voyager ) . The bacterial isolates E . coli DH5α , E . coli ATCC 25922 , P . aeruginosa ATCC 27853 , Staphylococcus aureus ATCC 29213 and S . pyogenes AP1 were obtained from the American Type Culture Collection . The P . aeruginosa strain PA01 was a generous gift from Dr . B . Iglewski ( University of Rochester ) . E . coli O18:K1 was a kind gift from Dr . C . van 't Veer ( University of Amsterdam ) . The clinical isolates E . coli 49 . 1 , E . coli 47 . 1 and P . aeruginosa 15159 were obtained from the Department of Bacteriology , Lund University Hospital , Sweden . Animals were housed under standard conditions of light and temperature and had free access to standard laboratory chow and water . The animals were purchased from Charles River or the animal facility at Lund University . E . coli , S . aureus and S . pyogenes AP1 strains were grown to mid-exponential phase in Todd-Hewitt ( TH ) broth . P . aeruginosa strains were grown in TH broth overnight . Bacteria were washed and diluted in 10 mM Tris , pH 7 . 4 , containing 0 . 15 M NaCl , either alone or with 20% normal or heat-inactivated citrate-plasma or 50% citrate blood . Fifty µl bacteria ( 2×106 cfu/ml ) were incubated at 37°C for 2 h with the C-terminal TFPI-2 derived peptide EDC34 , LL-37 or DAC31 at the indicated concentrations . Serial dilutions of the incubation mixture were plated on TH agar , followed by incubation at 37°C overnight and cfu determination . Human citrate blood was diluted ( 1∶1 ) with PBS prior to addition of 2×108 cfu/ml bacteria . The mixture was incubated with end-over-end rotation for 1 h at 37°C in the presence of peptides ( 60 and 120 µM ) . Two percent Triton X-100 ( Sigma-Aldrich ) served as positive control . The samples were then centrifuged at 800×g for 10 min and the supernatant was transferred to a 96-well microtiter plate . Hemoglobin release was determined by measuring the absorbance at 540 nm and is expressed as % of Triton X-100 induced hemolysis . Bacteria ( 1–2×109 cfu ) were incubated for 30 min or 1 h at 37°C with human plasma alone or supplemented with TAMRA-labeled EDC34 ( at 3 µM ) . Samples were then prepared for FACS analysis as previously described [52] . For visualization of the complement proteins , rabbit polyclonal antibodies against either LGE27 , a C-terminal epitope of human C3a , or rabbit polyclonal antibodies against C1q ( both at 1∶100 ) in combination with a secondary goat anti rabbit IgG FITC-labeled antibody ( 1∶500 , Sigma ) were used . Flow cytometry analysis ( Becton-Dickinson , Franklin Lakes , NJ ) was performed using a FACS Calibur flow cytometry system . The bacterial population was selected by gating with appropriate settings of forward scatter ( FSC ) and sideward scatter ( SSC ) . Controls without primary antibodies were included . Total positive cells present and fluorescence index ( FI ) ( positive cells present multiplied with mean ) are presented in the figures . Human citrate plasma was supplemented with bacteria ( 1–2×109 cfu ) and incubated alone or with EDC34 at 3 µM for 30 min or 1 h at 37°C , centrifuged and supernatants and the bacterial cells were collected . The pull down assay , to extract bound proteins from the bacteria was performed as described previously [52] , except two additional wash steps with acetone prior to sample separation by SDS page using 16 . 5% Tris-tricine gels ( C . B . S Scientific ) under reducing conditions . Proteins and peptides were transferred to nitrocellulose membranes ( Hybond-C ) , blocked by 3% ( w/v ) skimmed milk , washed , and incubated with rabbit polyclonal antibodies against the C-terminal part of C3a ( LGE27 antibodies ) ( 1∶1000 ) , rabbit polyclonal antibodies to C1q ( 1∶1000 ) ( Dako ) or rabbit polyclonal antibodies to C5b-9 ( 1∶1000 ) ( Abcam , England ) . The proteins were detected by using HRP-conjugated secondary antibodies ( 1∶2000 ) ( Dako ) and an enhanced chemiluminesent substrate ( LumiGLO ) developing system ( Upstate cell signaling solutions ) . All clotting times were analyzed using a coagulometer ( Amelung , Lemgo , Germany ) . For determination of prothrombin time ( PT , thromboplastin reagent ( Trinity Biotech ) ) and thrombin clotting time ( TCT , Thrombin reagent ( Technoclone ) ) , 50 µl of fresh citrate plasma , together with indicated concentrations of EDC34 or DAA14 were pre-warmed for 60 sec at 37°C before clot formation was initiated by adding 50 µl clotting reagent . To record the activated partial thromboplastin time ( aPTT ) , 50 µl of a kaolin-containing solution ( Dapttin , Technoclone ) was added to the plasma-peptide mix and incubated for 200 sec before clot formation was initiated by adding 50 µl of 30 mM fresh CaCl2 solution . To determine the blood clotting time , 50 µl of citrated blood were pre-warmed to 37°C for 60 sec , before 50 µl of 30 mM fresh CaCl2 were added to initiate coagulation . Thrombin/antithrombin complexes ( TATc ) were determined in mouse citrate-plasma by ELISA ( USCN Life Sciences Inc . ) . Bacteria were grown in TH broth to mid-exponential phase ( OD620∼0 . 5 ) , washed twice in 50 mM Tris/HCl , pH 7 . 4 and resuspended in 50 mM Tris/HCl , pH 7 . 4+50 µM ZnCl2 to a final concentration of 2×109 cfu/ml . Hundred microliter of bacteria were incubated with 10 µl of EDC34/DAA14 or buffer for 60 sec before the addition of 100 µl human citrate plasma . Samples were incubated for 35 min at 37°C on rotation followed by centrifugation . The bacterial pellets were washed once in 50 mM Tris/HCl , pH 7 . 4 and resuspended in 100 µl 50 mM Tris/HCl , pH 7 . 4+50 µM ZnCl2 buffer containing 2 mM of the chromogenic substrate S-2302 ( Chromogenix ) . Samples were incubated for 30–60 min at 37°C , centrifuged and the absorbance was measured at 405 nm in the bacterial supernatants . In other experiments 100 µl of Dapttin ( Technoclone ) were incubated with 10 µl of EDC34/DAA14 for 60 sec prior to the addition of human citrate plasma . Samples were incubated for 3 min at RT , centrifuged and the pellet was washed twice in 50 mM Tris/HCl , pH 7 . 4 before suspension in 100 µl 50 mM Tris/HCl , pH 7 . 4+50 µM ZnCl2 buffer containing 2 mM of the chromogenic substrate S-2302 . After 30 min incubation at RT samples were centrifuged and the absorbance of the supernatant was determined ( A405 nm ) . Samples containing 50 mM Tris/HCl , pH 7 . 4 instead of citrate plasma served as negative controls . Bacteria and Dapttin samples were prepared as described in the chromogenic substrate assay . Samples were incubated for 15 min at RT , with shaking . Dapttin samples were centrifuged at 10 . 000 rpm for 2 min and supernatants were stored at −20°C . Bacterial samples were washed twice in 50 mM Tris/HCl , pH 7 . 4 and resuspended in 55 µl 50 mM Tris/HCl , pH 7 . 4+50 µM ZnCl2 buffer followed by 15 min incubation at RT before centrifugation at 10 . 000 rpm for 2 min and storage of supernatants at −20°C . Samples were separated by SDS page and analyzed by western blot using antibodies against HK and its degradation products as previously described [53] . Human citrate plasma was incubated with 50 µM of EDC34 or DAA14 for 2 min at RT before the addition of Dapttin and incubation for 2 min at 37°C . Samples were kept on ice , diluted with deproteinising buffer ( 1∶5 ) , centrifuged ( at 4°C ) and the pellet was mixed with equal amounts of assay buffer . Samples with H2O instead of peptide served as positive controls . All buffers were provided together with the ELISA kit used to quantify the released bradykinin according to manufactures instructions ( Markit-M-Bradykinin Kit; DS Pharma Biomedical co . Ltd ) . Male C57BL/6 mice ( 8 weeks ) were i . p . injected with 5 mg/kg of E . coli O111:B4 LPS ( Sigma ) . Thirty minutes after LPS challenge mice were treated with 0 . 5 mg EDC34 ( i . p . ) . For analysis of cytokines and coagulation parameters , mice were sacrificed 4 and 8 h post-LPS injection and the blood was collected by cardiac puncture . E . coli DH5-α bacteria were grown to mid-exponential phase ( OD620∼0 . 5 ) , harvested , washed in PBS and diluted in the same buffer to 2×109 cfu/ml . Hundred fifty microliter of the bacterial suspension was injected intraperitoneally ( i . p . ) into male BALB/c mice immediately followed by 0 . 5 mg EDC34 or PBS . Mice were sacrificed 0 . 5 , 2 , 4 and 8 h post-infection to evaluate cfu , cytokines , coagulation parameters , blood counts and histology of the lungs . In another set of experiments mice were treated with 0 . 5 mg EDC34 injected i . p . , immediately , or after 1 h , or subcutaneously 1 h post-infection . The animal status and weight was followed daily for up to 7 days . Mice sacrificed before day 7 , according to predefined endpoint criteria , were counted as non-survivors . In another experiment , male BALB/c mice were depleted of complement factors by i . p . injection of 4 . 8 U of cobra venom factor ( CVF ) ( Quidel ) [39] . After 16 h mice were infected with E . coli DH5-α ( i . p . ) and immediately treated with 0 . 5 mg EDC34 or PBS . Mice were sacrificed 6 h post-infection and cfu were evaluated in spleen and liver , respectively . P . aeruginosa 15159 or P . aeruginosa PA01 bacteria were grown to mid-exponential phase ( OD620∼0 . 5 ) , harvested , washed in PBS , diluted in the same buffer to 2×109 cfu/ml , and kept on ice until injection . Hundred microliter of the bacterial suspension was injected ( i . p . ) into male C57BL/6 mice . EDC34 ( 0 . 5 mg ) or buffer alone was administered i . p . immediately after bacterial injection , or s . c . either as one dose after 1 h , or two doses at 1 h and 7 h . In another experiment , mice were treated twice s . c with a combination of 300 mg/kg ceftazidime and 0 . 5 mg of EDC34 , 1 . 5 h and 4 . 5 h after bacterial infection . Data from three independent experiments were pooled . The survival data were obtained by following the animals daily up to 7 days monitoring status and weight . Mice reaching the predefined endpoint-criteria were sacrificed and counted as non-survivors . For histological evaluation of lungs derived from the in vivo P . aeruginosa infection model , tissues were collected at indicated time points , fixed in 4% formaldehyde for 24 h , embedded in paraffin , sectioned and stained with hematoxylin and eosin . Assessment of differences in alveolar space , cell infiltration , thickness of alveolar septa ( cell wall thickness ) and thrombi ( Figure S6 ) was performed by scoring of at least five view fields per section by three blinded independent observers ( score 1–4; where 1 indicates no change , 2 minor , 3 medium , and 4 significant change ) . For transmission electron microscopy analysis of the presence of TFPI-2 fragments in vivo , fibrin slough from a patient with a chronic venous ulcer ( CWS ) was fixed and processed as previously described [52] . For immunostaining [52] , rabbit polyclonal antibodies against the C-terminal of TFPI-2 ( CAKALKKKKKMPKLRFASRIRKIRKKQF ) alone , or in combination with rabbit polyclonal antibodies against the C-terminal part of C3a ( LGE27 antibodies ) ( 1 µg/ml ) ( Innovagen AB ) were utilized . Controls without primary antibodies were also included . For simultaneous detection of TFPI-2 and C3a , 1 µg/ml EM rabbit anti-goat IgG 20 nm Au ( BBI ) and 1 µg/ml EM goat anti-rabbit IgG 10 nm Au ( BBI ) were used . All samples were examined with a Jeol JEM 1230 electron microscope operated at 80 kV accelerating voltage . Images were recorded with a Gatan Multiscan 791 charge-coupled device camera . For scanning electron microscopy , lungs were collected at 12 h after injection of bacteria and fixed in 2 . 5% ( v/v ) glutaraldehyde in 0 . 15 M sodium cacodylate buffer , pH 7 . 4 , overnight at room temperature and further treated as described previously [31] . Specimens were examined in a JEOL JSM-350 scanning electron microscope . To quantify pulmonary lesions , lung samples from 30 different fields covering an entire lung section were made , and the percentage of fibrin deposits and fields exhibiting hemorrhage were determined . The cytokines IL-6 , IL-10 , MCP-1 , IFN-γ , and TNF-α were measured in mouse plasma using the Cytometric bead array; Mouse Inflammation Kit ( Becton Dickinson AB ) according to the manufacturer's instructions . The number of platelets in mouse blood anti-coagulated with EDTA was determined using the VetScan HM5 ( TrioLab ) . Values are shown as mean with SEM . For statistical evaluation of two experimental groups the Mann-Whitney U-test was used and for comparison of survival curves the log-rank test . To compare more than two groups One-Way or Two-Way ANOVA with Bonferoni post-test were used . Viable count data are presented as mean with SD . All statistical evaluations were performed using the GraphPad Prism software 5 . 0 . with *p-<0 . 05 , **<0 . 01 and ***p<0 . 001 and ns = not significant . | Bacterial infections , especially sepsis , are worldwide a major cause of morbidity and mortality . Sepsis is characterized by an excessive and uncontrolled immune and coagulation response caused by bacteria and bacterial products , which eventually leads to multiple organ failure . Despite supportive treatments and administration of antibiotics , the incidence of sepsis is rising . Development of antibiotic resistance among bacteria , and the inability of antibiotics to target dysregulated host responses during severe infections and sepsis , motivates the search for novel anti-infective treatment modalities . Here , we describe a therapeutic potential of the peptide EDC34 , derived from the C-terminus of tissue factor pathway inhibitor-2 ( TFPI-2 ) . The peptide's boosting effect on bacterial clearance paired with its inhibiting effect on excessive coagulation makes it a promising therapeutic candidate for invasive Gram-negative infections . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | The TFPI-2 Derived Peptide EDC34 Improves Outcome of Gram-Negative Sepsis |
Stochastic channel gating is the major source of intrinsic neuronal noise whose functional consequences at the microcircuit- and network-levels have been only partly explored . A systematic study of this channel noise in large ensembles of biophysically detailed model neurons calls for the availability of fast numerical methods . In fact , exact techniques employ the microscopic simulation of the random opening and closing of individual ion channels , usually based on Markov models , whose computational loads are prohibitive for next generation massive computer models of the brain . In this work , we operatively define a procedure for translating any Markov model describing voltage- or ligand-gated membrane ion-conductances into an effective stochastic version , whose computer simulation is efficient , without compromising accuracy . Our approximation is based on an improved Langevin-like approach , which employs stochastic differential equations and no Montecarlo methods . As opposed to an earlier proposal recently debated in the literature , our approximation reproduces accurately the statistical properties of the exact microscopic simulations , under a variety of conditions , from spontaneous to evoked response features . In addition , our method is not restricted to the Hodgkin-Huxley sodium and potassium currents and is general for a variety of voltage- and ligand-gated ion currents . As a by-product , the analysis of the properties emerging in exact Markov schemes by standard probability calculus enables us for the first time to analytically identify the sources of inaccuracy of the previous proposal , while providing solid ground for its modification and improvement we present here .
Ion channels are the fundamental elements underlying neuronal excitability and information transfer , inter- and intracellularly . These protein pores , found also in other excitable cell types , undergo fast conformational modifications ( hereafter referred to as channel gating ) induced by a change in the electric field or by the binding of ligand molecules . By doing so , channels selectively affect the ionic conductances of the membrane and enable ions to flow according to their electrochemical potentials [1] . The impact of the first quantitative deterministic description of conductance gating [2] was extremely significant , as testified by its wide use up to today [3] . Since the 1970s however , the stochastic nature of the single ion channels gating has been fully recognised . The resulting random fluctuations in the membrane conductances ( which are known as channel noise ) have been the subject of intense theoretical and experimental research [4]–[12] . Nevertheless , only recently channel noise was emphasised to have a significant impact on neuronal signals generation , propagation and integration , and it was suggested for consideration in realistic models of single neurons [13]–[19] . In some parts of the peripheral nervous system , channel noise has been demonstrated to be prominent for information transfer and perception ( e . g . , see [20] and references therein ) . However , in the central nervous system whether or not channel noise plays a role at the level of large networks of interacting neurons , how heterogeneous ion channel types contribute to spontaneous network firing , and whether channel noise combines or interferes with other sources of noise ( synaptic , for instance ) remain open questions . Towards addressing these questions , the increasing availability of cheap parallel computing resources and improved algorithms [21] , [22] allow one to approach in silico the study of networks of thousands of morphologically detailed multi-compartmental model neurons [23] . In addition , a diversity of voltage- and ligand-gated ion channel types can be included in these large models with biophysical realism [24] . Unfortunately , channel noise is rarely considered for large network simulations or detailed multi-compartmental models [25] , due to its heavy computational load . Implementing single-channel stochastic models explicitly , for each of the thousands of channels per ion conductance type and per neuron , requires Montecarlo simulation techniques [5] , [14] , [15] , [26] that are computationally intensive even for single compartmental neurons , regardless of excellent speed-up techniques [14] . Throughout this paper , we refer to such explicit and exact simulation methods by the term microscopic , regardless of the details of their actual numerical implementation [27] . For the specific case of the Hodgkin-Huxley ( HH ) equations , Fox and collaborators proposed an alternative approximate method to mimic channel noise , avoiding a microscopic description of the individual channels [28] , [29] . This method relies on the use of stochastic differential equations to macroscopically account for the fluctuations in the overall conductance of sodium and potassium channels , with formal analogies to the Langevin equation [14] , [30] . Although this approach is very attractive and was employed widely in the literature ( see references in [31] ) , its accuracy was recently challenged and debated by several authors [27] , [31]–[33] . These authors compared numerical simulations of the exact microscopic descriptions of the HH model with those obtained by Fox's , finding some inconsistencies . It was however only with the work by Bruce ( 2009 ) , that a straightforward test and framework were proposed to quantify the accuracy of Fox's algorithm . Simulating a voltage-clamp experiment , while recording ion currents , clearly shows that Fox's approximation does not capture correctly the microscopic statistical properties , regardless of how large the number of single ion channels to be approximated is . An ad hoc partial correction of Fox's algorithm - based on the simultaneous Montecarlo simulations of single channels - was also proposed for some activity regimes [31] , but it cannot be generalised to arbitrary simulation conditions . In this paper we introduce and operatively define a general method , based on the diffusion approximation [30] , to transform any deterministic model neuron into its effective stochastic version , for an arbitrary set of ion conductances . As in previous studies , we focus on discrete Markov processes [8] , [34] , routinely employed in the experimental identification of voltage-gated channels and synaptic receptors . Our purpose is to reintroduce channel noise in deterministic conductance-based models with limited computational overhead . We also aim at accurately replicating the statistical properties of ion conductances , as predicted by the exact microscopic description , while avoiding the use of any ad hoc correction or heuristics in the choice of the parameters [35] . Our approach is related to previous Langevin-based formulations , although with a significant difference in the way channel fluctuations are reintroduced in model equations . It can be considered as an accurate and systematic generalisation of Fox's algorithm , to the case of voltage- , ion- , and ligand-gated channels with arbitrary complexity . We numerically compare our approach to that by Fox and we provide , as a Supporting Information , some analytical results showing where it fails . We validate our approach for single-compartmental neuronal simulations , incorporating HH fast inactivating sodium channels and delayed rectifier potassium channels , analogously to previous works . By comparing our effective method to the exact simulations of the stochastic channel kinetic schemes , we obtain satisfying agreement .
We consider a single-compartmental conductance-based neuron model [36] . For this class of models , the membrane potential obeys the following current balance equation [1]where is the specific membrane capacitance and is an externally applied current density ( expressed in ) . These models comprise a leak current and a number of intrinsic ( as well as synaptic ) currents that can be similarly expressed by an ohmic relationship , which links the current to the membrane potential . Each ionic conductance is completely determined by the fraction of corresponding channels in the open state ( see Fig . 1A–D ) . For reference to previously published papers [9] , [10] , [12] , [15] , [18] , [29] , we consider here the HH voltage-gated currents and with standard parameters [2] . Therefore , we consider and . In the deterministic model , and are expressed phenomenologically as a product of activation and inactivation deterministic variables [37]–[40]:Each of these variables obeys a first-order ordinary differential equation of the form ( 1 ) where and , are kinetic parameters . All the model parameters are summarised in Table 1 . Montecarlo methods represent the most commonly adopted way to simulate the random temporal evolution of ion conductances in a membrane patch , populated by a set of identical independent channels . Due to spatial proximity , channels are assumed to be coupled together by a common gating variable , such as the membrane potential or the local neurotransmitter concentration . Then , the full knowledge of the Markov kinetic scheme ( see Fig . 1A–D ) describing the distinct conformational states of each ion channel , as well as the transition probabilities across states , are needed [41] , [42] . The kinetic scheme is employed to simulate the random transitions of the state of each individual ion channel , by repeated pseudo-random number generation ( see [5] , [14] , [15] , [26] and references therein ) . Although refined fast-computation techniques have been proposed [14] , we employ here a basic numerical implementation . Briefly , instead of tracking the state of each channel , the number of channels in a given state is tracked and updated at each time step ( ) , with conditional probabilities that depend on the transition rates of the Markov scheme , as exemplified in Fig . 1A . We recall that simulating the occurrence of a random event with probability can be achieved by generating a pseudo-random number , uniformly distributed between and , and testing whether or not [43] . In the simulations reported here , we set the single-channel conductance for both sodium and potassium channels to , unless specified otherwise , and we consider a fixed channel density of and for sodium and potassium currents , respectively . In all simulations , a cylindrical single compartment was used with length and diameter equal to , unless otherwise noted . Albeit conceptually simple , these algorithms require a great amount of computational power , which increases with the number of channels that are to be simulated and with the probability of their activation . Simulation code and analysis scripts , developed in C++ and in NEURON [44] , are available from ModelDB [45] at http://senselab . med . yale . edu/modeldb via accession number 127992 . We examine the case of a ion current whose microscopic correlate is represented by a population of ion channels . The single-channel kinetics is a 2-state scheme: open and closed , as shown in Fig . 1A . This is the simplest kinetic scheme and is often employed , for instance , for the minimal description of ionotropic AMPA-receptors [46] . The symbols and in Fig . 1A represent the transition probabilities between states , expressed per time unit ( i . e . , as rates ) . They are functions of the channel gating variable ( s ) – such as membrane voltage , intracellular calcium concentration , extracellular magnesium concentration , extracellular glutamate concentration , etc . [8] – and are experimentally identified by routine electrophysiological techniques [7] and optimisation methods [34] . For the definition of our effective simulation technique for channel noise , we consider five realistic assumptions: ( i ) the channels are identical and statistically independent; ( ii ) for simplicity , only one conformational state is associated to a non-zero ion conductance ; ( iii ) is large and is known; ( iv ) the single-channel kinetics is described by a Markov process , where transition probabilities depend only on the current state and on the gating variable ( s ) , and not on the previous occupancy history; and ( v ) the gating variables ( e . g . , ) change slowly , compared to the channel kinetics , with time constant [1] . Because of ( i ) – ( ii ) , the maximal ion conductance associated to the channels can be expressed as . Then , the time-varying conductance depends only on , the fraction of channels in the open state: ( 2 ) Since individual channels undergo random transitions between states [7] , is a non-stationary random variable , whose instantaneous value is distributed according to a binomial probability function: the number of open channels , ( with constant ) , is a binomial random variable . As a consequence , the statistical properties of are fully specified by , the probability of occupancy of the open state [6] . By assumption ( iii ) , the binomial distribution of can be approximated by a Gauss distribution , invoking the de Moivre-Laplace ( or central limit ) theorem , valid when [47] . By ( iv ) , can be numerically computed as the solution of the following linear differential equation [6] , formally equivalent to the deterministic kinetic Eq . 1 [48]: ( 3 ) with and . Finally , under assumption ( v ) , Eq . 3 can be solved analytically and is expressed as an exponential function . Under these approximations , is Gauss-distributed and completely described by its mean and by its ( auto ) covariance function , which at the steady-state has an exponentially decaying profile: [6] , [49] . In the theory of stochastic processes , is called a diffusion process , with and its steady-state variance and autocorrelation time constant , respectively [30] . By these considerations , it follows that can be approximated and computer-simulated by an efficient method , alternative to the exact Montecarlo simulation of the discrete kinetic scheme [14] . This method consists in generating a realisation of an Ornstein-Uhlenbeck's process [30] , with time-varying mean , steady-state variance , and autocorrelation time constant : ( 4 ) ( 5 ) where is a -correlated Gauss-process with zero mean [47] ( see also Eq . 20 ) . Since [6] , [49] , the deterministic component of evolves as Eq . 3 , which is the familiar equation one expects by the mass-action law ( i . e . , Eq . 1 ) , while interpreting as deterministic the scheme of Fig . 1A [2] , [38] . For clarity , we rewrite such an equation as ( 6 ) with , and . As opposed to the deterministic HH formalism however , the stochastic nature of is now explicitly captured by , algorithmically generated as a pseudo-random process by iterating the discrete-time version of Eq . 5 [50] , reported for the sake of completeness in Eqs . 23–24 . Thus , by setting , Eqs . 4 , 5 , and 6 reproduce both the time-varying mean and the steady-state covariance of . More precisely , and the covariance of the term relax to the same analytical expression , after a transient of the order of . Finally , the clipping of negative conductance values for may be necessary but , if lacking , it will not affect by accumulation the numerical integration of in the present form of Eq . 4 . We remark that we do not ( heuristically ) add a noise term in the right-hand-side of Eq . 6 , as in previous Langevin-based algorithms . Instead , a precise approximation procedure is employed to statistically mimic the effect of channel noise fluctuations in . Although for 2-state channels Eqs . 4–6 are indeed equivalent to Fox's formulation ( see the Text S1 ) , our approach differs considerably from that by Fox as soon as multiple-state channels are considered , e . g . , the sodium fast-inactivating and the potassium delayed-rectifier channels . We now generalise the diffusion approximation ( Eqs . 4–6 ) to the more general case of a large population of identical independent channels , whose single-channel dynamics is described by an M-state Markov scheme . Under the same assumptions ( i ) – ( v ) , the probability of occupancy of the open state fully describes the fraction of open channels ( see Eq . 2 ) . However , now is a particular ( say , the k-th ) element of the probability vector of state occupancy , and each element of corresponds to a distinct state of the kinetic scheme . By assumption ( iv ) , satisfies a system of M linear ordinary differential equations , which can be written in compact form as ( 7 ) The transition matrix is filled with the appropriate combinations of the individual transition rates between all the possible states [51] . is a vector with only one ( the k-th ) non-zero element , set to . Under assumption ( v ) , can be computed analytically as a linear combination of a steady-state value and of M-1 exponentials with time constants , each being the inverse of the absolute value of a non-zero eigenvalue of [51] . As for the 2-state kinetics , the statistical properties of the fraction of open channels are fully specified by and by the binomial distribution [6] . By assumption ( iii ) , the distribution of can be approximated by a Gauss-distribution [47] , and can be numerically simulated by an equivalent diffusion process . However , differently from the previous case , the steady-state covariance contains a weighted sum of M-1 exponentials [6] , [49] and not a single term: ( 8 ) Therefore , Eq . 4 no longer approximates , and it must be extended to a linear combination of M-1 Ornstein-Uhlenbeck's independent processes , with appropriate coefficients and time constants: ( 9 ) ( 10 ) As for the 2-state model , . Then , one always recovers the deterministic description of the -state channels , formally coincident with Eq . 7 . The derivation of the analytical expressions for and is necessary , as they depend on the values of the gating variable ( s ) ( e . g . , ) , and requires the full expression of [6] , [49] , ( 11 ) which can be obtained by Laplace-transforms or linear algebraic methods [52] . We remark that , for our purposes , the derivation of Eq . 11 is important mainly to introduce Eqs . 8–10 . Indeed , Eq . 11 considerably simplifies in the case of ion channels whose -state kinetics can be mapped into , or have been experimentally identified as , the composition of several 2-state subunits . For instance , the scheme of Fig . 1B can be mapped into the equivalent kinetic scheme shown in Fig . 1E . This is very common in the computational neuroscience literature for voltage- and ligand-gated ion channels , whose single-channel open state corresponds to the simultaneous active state of a multiple number of independent subunit types . To illustrate how Eq . 11 simplifies , we discuss a specific example where three different subunit types are present [37] , [38] , although our considerations hold for any number of different subunit types . We name these three subunit types as m , h , and s , and for each of them we compute the steady-state probabilities of the active state and the gating time constants , following from the solution of Eq . 3: ( 12 ) We further assume that the overall single-channel conductance results from the composition of a given number of elements of each subunit type: say , q , r , and w subunits of the type m , h , and s , respectively . For instance , in the kinetic scheme of Fig . 1E , we have , , and . Since each subunit is described by 2-state kinetics , the total number M of states is . By this definition , the process is binomial and described by the joint probability that all subunits are simultaneously in their open state . Because of the statistical independence of each subunit , the joint probability is the product of elementary probabilities [6] . Under the same assumptions of previous section , can be approximated by a diffusion stochastic process , combining deterministic and stochastic terms , as in Eq . 4 . Being , we can rewrite Eq . 9 as follows: ( 13 ) ( 14 ) Since in this case the covariance of a product is the product of covariances , Eq . 11 reduces to [6] , [49] ( 15 ) with , and . Expanding the powers and products of Eq . 15 and obtaining the expressions for the distinct coefficients and time constants , needed for Eqs . 9 and 10 , is easier than manipulating the matrix exponential of Eq . 11 . In the specific case of HH fast-inactivating sodium ( i . e . , , , and ) and delayed rectifier potassium channels ( i . e . , , and ) ( Fig . 1B–C ) , and take the expressions reported in Table 2 . In order to further gain in computational efficiency , while numerically implementing our diffusion approximation of channel noise ( Eqs . 9–10 ) , it is possible to reduce to one the number of required independent Ornstein-Uhlenbeck's stochastic processes . This additional approximation consists in interpolating the covariance of by a single decaying exponential , by replacing Eq . 9 with Eq . 4 . Indeed , since Eq . 8 is the weighted sum of exponentials , one should not privilege any of those terms a priori and appropriately choose ( in Eq . 4 ) and ( in Eq . 5 ) as best-fit parameters for each value of the gating variable ( s ) , so that ( 16 ) Alternatively , by expanding both sides of Eq . 16 by the Taylor series , extended to the first-order ( or higher ) , the dominant term around can be approximated by setting ( 17 ) In investigating the impact of channel noise on the computational properties of single-neurons and networks , such a systematic and controlled reduction procedure should replace heuristic methods and may be extremely useful to dissect whether or not each of the terms is needed in accounting for a particular observation . Following Eqs . 9–10 and Table 2 , we now formulate the effective stochastic model , corresponding to the deterministic HH model introduced earlier: ( 18 ) The deterministic gating variables still obey Eq . 1 , while each of the new stochastic variables ( and ) is described by Eqs . 9 and 10: ( 19 ) where , , , and are the coefficients given in Table 2 , while are independent , identical , -correlated , Gauss-distributed processes with zero means and unitary variances ( see Eqs . 23–24 ) . We emphasise that the procedure leading to Eq . 18 is general and can be easily applied to more complex ( single- and multi-compartmental ) neuron models , which incorporate arbitrary ionic currents . Since the Ornstein-Uhlenbeck's stochastic process has been referred to repeatedly in the previous sections , we concisely review its definition and its practical numerical simulation . A realisation of this process , say , can be operatively defined as the exponential filtering of a Gauss-distributed white noise . Abusing the notation of ordinary differential equations , is the solution of ( 20 ) The term represents a stationary Gauss-distributed stochastic process , which is a white-noise , fully specified by its mean and covariance . By linearity , is also Gauss-distributed [47] and characterised by non-stationary mean and covariance : ( 21 ) ( 22 ) These quantities converge to stationary values after a time of the order of , so that at the steady-state has mean and variance equal to zero and , respectively , and an exponentially-decaying autocorrelation function , with time constant . For the purpose of obtaining independent realisations of in computer simulations , a discrete-time equivalent of Eq . 20 must be employed to generate a sequence of values . A simple iterative update formula is available , ( 23 ) which requires the generation of a Gauss-distributed pseudo-random number at each iteration , with zero mean and unitary variance [43] . Such an iterative expression is exact , in the sense that neither needs to be uniform nor infinitesimal for to approximate the statistical properties of [50] . For very small compared to , Eq . 23 can be also approximated by a first-order Taylor expansion , leading to ( 24 )
We keep the membrane voltage fixed in time , while numerically simulating Eqs . 18 , 19 . We then study the dependence of the fraction of open channels on at the steady-state , computing mean , variance and autocorrelation time length of , . The results confirm that our effective reduction allows one to match accurately the statistical features of the microscopic models , obtained by Montecarlo simulations of the Markov-schemes . Fig . 2 summarises these results for a range of clamped membrane potentials and different total numbers of ion channels . Panels A–C refer to the steady-state properties of HH potassium currents and panels D–F refer to sodium currents . In each panel , black and red markers refer to the actual numerical simulation of the microscopic and the effective models , respectively , whereas solid lines represent the theoretical steady-state values . The mean of the fraction of open channels accurately matches the theoretical predictions ( and for panels A , D - see Eqs . 13–14 ) and , as expected , it is independent of the number of channels . The variance inversely depends on and no difference is evident by comparing microscopic and effective simulations . The solid lines of panels B , E are obtained by plotting and ( see Table 2 ) . For each value of , the covariance has a decaying profile characterised by multiple time constants ( see Eq . 8 and Table 2 ) . In order to represent concisely how such a decaying profile changes for distinct values of , panels C and F show ( magenta curves ) the values obtained by best fitting with a single exponential function the autocorrelation function of . The agreement between microscopic and effective simulations is satisfying and demonstrates that , when predicting and mimicking the autocorrelation properties of channel-noise fluctuations , the kinetic terms , , and , emerging in previous Langevin-based approaches as effective autocorrelation time constants , fail significantly . When a single Ornstein-Uhlenbeck process is used to increase the computational efficiency , the single noise term approximation given in Eqs . 16–17 turns out to be more accurate than the heuristics based on the kinetic time constants , , and or the submultiples , and ( see also Text S1 ) . In the lower part of Fig . 2 ( panels G–L ) , the same analysis is repeated , comparing the microscopic Markov-scheme simulations and the results obtained by the Langevin-based approximation proposed by Fox and coworkers [28] , [29] . According to the mathematical expressions reported in the Supporting Information , numerical simulations of the Fox's model show that , regardless of the number of channels , the variance of potassium currents is overestimated ( panel H ) , whereas the variance of sodium currents is underestimated ( panel K ) . Because of the inherent limitations of the Langevin-based approach , where a single noise term is added to the differential equations describing activation and inactivation variables , the autocorrelation properties of channel noise fluctuations ( panels I , L ) are mismatched . Finally , Fig . 3 illustrates for the agreement between the microscopic model and our effective approximation ( panels A–F ) , as well as the mismatch of Fox's algorithm ( panels G–L ) , displaying sample time series of channel noise . Both histograms of fluctuations amplitude ( panels B , E , H , K ) and autocorrelation functions ( panels C , F , I , L ) confirm and further support the results of Fig . 2 . As the steady-state properties of the fractions of open channels are equivalent in the microscopic and effective models , we tested the full model as in a current-clamp experimental protocol . In this case , the gating variable is not clamped to a fixed value and both passive and active membrane properties arise by the interplay between ion currents . Once injected with a weak depolarising DC current , both the microscopic and the effective model neurons fire irregular action potentials [14] , as shown in Fig . 4A . In the absence of channel noise ( i . e . , for and ) , is not strong enough to elicit spiking activity as it is below threshold for ( deterministic ) excitability . In order to quantify more accurately this phenomenon , we show in Fig . 4B the coefficient of variation ( CV ) of the interspike interval distribution obtained simulating the microscopic , effective and Fox's models ( black , red and blue traces , respectively ) , for increasing values of the membrane patch area ( i . e . , of the number of ion channels ) . Note that Fox's model exhibits no spontaneous activity for larger cell sizes . On the other hand , the CV of the microscopic and effective models are very close . Fig . 4C shows the corresponding spontaneous mean firing rates: the presence of an “offset” in the results obtained by the effective model is evident , which is greatly reduced as the membrane area increases . This is due to the small number of channels in the membrane patch when the area is very small , against assumption ( iii ) . In order to perform a direct comparison with the analysis carried out in [27] , a monophasic current pulse of fixed duration and increasing amplitude was applied 10000 times to probe the impact of channel noise on neuronal evoked responses . In Fig . 5 , panel A displays the firing efficacy ( i . e . , the fraction of trials where a spike was elicited ) , panel B shows the average latency of the evoked action potential with respect to the stimulation time , and panel C displays the standard-deviation ( i . e . , the jitter ) of the firing latency . Black and red traces and dots result from the simulations of the exact kinetic schemes and from our diffusion approximation , respectively , while in blue we indicate the results from the simulation of the Langevin-approximation introduced by Fox . The satisfactory agreement between microscopic and effective models is apparent , whereas simulations according to Fox's algorithm differ considerably . Panel D shows the distribution of spike occurrence times , evoked by a biphasic stimulus over 10000 trials . The distributions of spike times obtained by the microscopic and effective models almost overlap , while Fox's distribution has a significantly different shape . The results we present here for the microscopic and Fox's models are in close agreement with those discussed in greater detail in [27] . The results shown in Fig . 5 refer to the application of either a mono- or biphasic stimulus of short duration , in the order of milliseconds . Here , we extend the previous analysis to the case of significantly longer stimulations: our objective is to study the so-called reliability of spike timing along the lines of the experimental protocol defined in [13] . It is well known that , as a consequence of channel noise , the reliability of evoked spike timing is higher for current stimuli fluctuating in time than for DC current pulses [13] , [15] , [17] . Indeed , larger fluctuations induced in the membrane potential by the driving stimulus transiently hyperpolarise the cell , thus reducing the variance of channel noise ( see Fig . 2B , E ) . A similar phenomenon has been described in the case of inhibitory autapses in the cerebral cortex [53] and it could also be represented at microcircuit-level by the role of disynaptic inhibition [54] . A single-compartmental model simulation incorporating channel noise can replicate this effect [15] and constitutes a further benchmark to compare microscopic and effective models . We note that for this analysis , we have chosen the neuron parameters in order to reproduce the results presented in [13] . The agreement between models is very good as shown in Fig . 6 , where black ( red ) traces and markers refer to the microscopic ( effective ) model . The spike responses to two repeated identical stimuli were considered: a DC pulse ( panel A ) and a realisation of an exponentially-filtered white noise ( panel B ) . The raster diagrams of the spike times ( upper plots ) , as well as the corresponding time histograms ( lower plots ) , demonstrate that the two models perform in the same way as the spread and latency of the spike times , in response to the repeated identical stimulation , are practically identical . Finally , a quantitative measure of both precision and reliability ( computed according to [13] ) provides values similar to those measured in in vitro experiments ( see figure caption ) . For stronger depolarising DC currents , the firing of both the microscopic and the effective models becomes more regular . The mean firing rate , as a function of was studied to test the agreement between their evoked response properties . Fig . 7 shows the curves computed over -long evoked spike-trains . For each current amplitude , the simulation was repeated 10 times , and firing rates obtained in each repetition were averaged . Error bars indicate the standard deviation of the firing rate across repetitions . Responses of both the microscopic and the effective models result in almost identical variability across repetitions and in both cases the type-II behaviour , typical of the deterministic HH model , fades away . This is a known consequence of the presence of channel noise , which smooths what would be an abrupt transition from a quiescent to a spiking regime . These irregular transitions occur for both models in the very same range of input currents ( green-shaded region in the figure ) , where the membrane potential repeatedly switches between a resting equilibrium point and a spiking limit cycle ( see [15] for an extended discussion ) . We finally compare the power-spectral densities of subthreshold membrane potential trajectories , obtained in simulations of the microscopic and effective models . We followed closely the numerical analysis of [18] , where a comparison between the microscopic model and a quasi-active linearised model with phenomenological inductances was instead presented . Once more , the agreement between the two models is satisfactory: in Fig . 8 we show the results , indicating by thick shaded curves the power spectra computed from the microscopic model , and by thin solid lines the power spectra computed from the effective model . The agreement is good over the entire frequency domain , reproducing some of the features that have been experimentally measured in cortical neurons and related to channel noise [19] .
In this paper , we introduced the systematic generalisation and improvement of previous Langevin-based channel-noise effective simulation techniques . By the diffusion approximation of ion channels population dynamics , we aimed at efficient and accurate computer simulation of channel noise . Our method approximates correctly the statistical properties of individual ion conductances ( their mean and autocovariance function ) , matching those emerging from the Montecarlo simulation of their corresponding Markov schemes . In addition , both passive and active properties of neuron model simulations are replicated with satisfying accuracy . While simulating of model time by a conventional Montecarlo algorithm takes about 22 hours for completion , the same simulation with very similar statistical features is replicated by the effective model in only 124 seconds , on a machine equipped with a Intel Core i7 , with of RAM , running Ubuntu Linux 9 . 10 . When relating the computation times to the benchmarking provided by [27] , our diffusion approximation is only 1 . 5 times slower than Fox's algorithm and therefore more than 4 . 5 times faster than the fastest available algorithm for exact microscopic simulations [14] . Our results have been obtained by custom C++ and NEURON model simulations ( see the Materials and Methods section ) , but the implementation of the method in other languages ( MATLAB , Python ) or other simulation environments ( Genesis , NEST , Brian ) is straightforward . Besides the speed increase , the value of our contribution is threefold: i ) mean , variance and spectral properties of fluctuations induced by the stochasticity of individual ion currents are correctly approximated , regardless of the number of channels; ii ) our method is presented operatively , allowing any deterministic neuron model , whose ion conductance kinetics is described by a Markov scheme , to be quickly converted into an equivalent stochastic version without involving any heuristics on the choice of the parameters for extra noise sources; iii ) the underlying assumptions for the validity of our approximation are also indicated with full details . The earlier proposals of [28] , [29] , recently challenged for their accuracy , are indeed very similar to our method , although focused only on the HH model . In these papers , the equations that state variables , , and obey to are modified by adding a single noise term , as follows: ( 25 ) where and is a Gauss-distributed noise term with zero mean and covariance given by ( 26 ) By direct inspection and comparison of Eqs . 4 , 5 , 6 , and 20 , it is possible to show that Eq . 25 and Eqs . 4–5 are equivalent ( see Text S1 ) . In other words , for 2-state kinetics the approximation given by Eq . 25 is correct but fails when the powers , , and are computed and when they are combined in the product . Under these circumstances , mean , variance and covariance function indeed deviate considerably from the correct dependence on , emerging from the microscopic simulations or computed analytically ( see Text S1 ) . Briefly , the potassium current simulated by the fourth power overestimates the correct variance , does not share the correct mean and has qualitatively different autocorrelation properties . The sodium current simulated by the third power and the product by instead underestimates the correct variance , does not share the correct mean and has quantitatively different autocorrelation properties . The interested reader can find all the details in the Supporting Information . We believe that the reason for the success of our approximation , compared to Fox's approach , lies not only in the correct agreement of fluctuations mean and variance , validated by direct comparison with the theoretical and numerical results of the microscopic description [31] , but also in the fact that the covariance function of those fluctuations must be precisely matched and should be approximated by a sum of white-noise terms and not by adding noisy terms to the deterministic kinetic equations for activation and inactivation variables . However , we note that under current-clamp condition , there is no a priori guarantee that any Langevin-based approach , including our diffusion approximation , works faithfully [55] . In fact , our assumption ( v ) , that the gating variable ( e . g . , ) changes slowly compared to channel kinetics , may not be instantaneously satisfied during very fast transients . Although the same condition is anyway employed for obtaining numerical speed-up in deterministic conductance-based models [1] , [56] , [57] , the instantaneous channel noise fluctuations might lag behind what predicted by microscopic exact models ( see Figs . 11–12 in Text S1 ) . Nevertheless , owing to the satisfying results we obtained in terms of firing-rate properties , firing time reliability , precision , efficacy , latency , jitter as well as subthreshold membrane fluctuations , we speculate that inaccuracies during very fast transients might still be compatible with accurate model performances ( perhaps due to the low-pass properties of the membrane ) , provided that first- and second-order voltage-clamp statistics are correctly matched . A very similar reduction procedure is implicitly mentioned in [18] , where the authors developed a quasi-active membrane potential equation employed only for the spectral analysis of subthreshold voltage noise , but not for its actual numerical simulations . The authors state clearly that their approximation can be viewed as a linearised approximation of the Fokker-Planck master equation [29] . As opposed to our method , which requires adding multiplicative noise terms to the membrane potential equation , their quasi-active model includes only additive noise , upon linearisation , resulting in the definition of electrical circuit analogs ( capacitances and inductors ) useful for the intuitive understanding of channel noise for subthreshold passive membrane properties , and for the analytical prediction of the spectral properties of membrane potential fluctuations . The authors , however , do not explicitly provide any derivation of their approach and do not test it for the excitable response neuronal properties as a replacement of microscopic simulations . One further approach to channel noise modelling has been proposed in [35] . We share the motivation of performing accurate and fast simulation by a Langevin-based approach , but we use stochastic processes with precise and defined statistical properties , coincident with those emerging from the microscopic description of the stochastic behaviour of channels . In the proposal by [35] , the effective stochastic term is modelled as Brownian motion , i . e . , as a Gauss-distributed process with independent increments and heuristically fixed constant variance , ignoring its voltage-dependence and the variety of autocorrelation time constants . Since the analytical derivation of the accurate statistical properties of channel noise is possible , and its implementation straightforward as we showed here , there is no need to use arbitrary parameters for simulating the stochastic components of ion currents gating . It is worth mentioning that population density approaches , proposed for integrate-and-fire as well as conductance-based models [58]–[61] , share to some extent the motivations of our work: exploring the impact of endogenous or exogenous noise sources while developing tools to capture or effectively simulate population-level dynamics [62] , [63] . Those works also aim at correctly mimicking actual network interactions in terms of an equivalent stochastic additive input to a generic unit of the network [64] , as in the mean-field approximation of synaptic interactions [65] . Since our work provides an accurate effective description of an intrinsic ( multiplicative ) noise source , our formulation could be very relevant for those approaches , in extending population density descriptions to incorporate endogenous channel noise . In conclusion , we believe that our method could open new possibilities for the investigations of channel noise impact in morphologically detailed conductance-based model neurons , as well as in large networks models , where realism cannot be compromised by computational parsimony . Spike timing computation in neural networks [66] with specific microcircuit architectures [54] might be for instance easily complemented by stochastic components of neural excitability , employing detailed neuron models . Finally , the possibility of further increasing the level of approximation , involving only a modification of the spectral properties of channel noise without affecting the accuracy of its variance , may lead to an in depth understanding of what temporal correlation properties are relevant for specific computational neuronal properties and how channel noise interacts with other noise sources . | A possible approach to understanding the neuronal bases of the computational properties of the nervous system consists of modelling its basic building blocks , neurons and synapses , and then simulating their collective activity emerging in large networks . In developing such models , a satisfactory description level must be chosen as a compromise between simplicity and faithfulness in reproducing experimental data . Deterministic neuron models – i . e . , models that upon repeated simulation with fixed parameter values provide the same results – are usually made up of ordinary differential equations and allow for relatively fast simulation times . By contrast , they do not describe accurately the underlying stochastic response properties arising from the microscopical correlate of neuronal excitability . Stochastic models are usually based on mathematical descriptions of individual ion channels , or on an effective macroscopic account of their random opening and closing . In this contribution we describe a general method to transform any deterministic neuron model into its effective stochastic version that accurately replicates the statistical properties of ion channels random kinetics . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"neuroscience/theoretical",
"neuroscience",
"computational",
"biology/computational",
"neuroscience"
] | 2011 | Accurate and Fast Simulation of Channel Noise in Conductance-Based Model Neurons by Diffusion Approximation |
Mother-to-infant transmission ( MTIT ) of HIV is a serious global health concern , with over 300 , 000 children newly infected in 2011 . SIV infection of rhesus macaques ( RMs ) results in similar rates of MTIT to that of HIV in humans . In contrast , SIV infection of sooty mangabeys ( SMs ) rarely results in MTIT . The mechanisms underlying protection from MTIT in SMs are unknown . In this study we tested the hypotheses that breast milk factors and/or target cell availability dictate the rate of MTIT in RMs ( transmitters ) and SMs ( non-transmitters ) . We measured viral loads ( cell-free and cell-associated ) , levels of immune mediators , and the ability to inhibit SIV infection in vitro in milk obtained from lactating RMs and SMs . In addition , we assessed the levels of target cells ( CD4+CCR5+ T cells ) in gastrointestinal and lymphoid tissues , including those relevant to breastfeeding transmission , as well as peripheral blood from uninfected RM and SM infants . We found that frequently-transmitting RMs did not have higher levels of cell-free or cell-associated viral loads in milk compared to rarely-transmitting SMs . Milk from both RMs and SMs moderately inhibited in vitro SIV infection , and presence of the examined immune mediators in these two species did not readily explain the differential rates of transmission . Importantly , we found that the percentage of CD4+CCR5+ T cells was significantly lower in all tissues in infant SMs as compared to infant RMs despite robust levels of CD4+ T cell proliferation in both species . The difference between the frequently-transmitting RMs and rarely-transmitting SMs was most pronounced in CD4+ memory T cells in the spleen , jejunum , and colon as well as in central and effector memory CD4+ T cells in the peripheral blood . We propose that limited availability of SIV target cells in infant SMs represents a key evolutionary adaptation to reduce the risk of MTIT in SIV-infected SMs .
Worldwide , over 30 million people are infected with HIV , including 3 . 3 million children . Transmission through breastfeeding can account for almost half of pediatric HIV infections [1] . Recent World Health Organization recommendations support breastfeeding by HIV-infected women along with antiretroviral therapy for mother or infant in areas where alternative feeding is not feasible . However , the high cost of anti-retroviral medications and limited access to therapy in developing countries along with unknown long-term consequences of the use of ART during breastfeeding demand further investigations into novel interventions to prevent breast milk-related transmission of HIV . A major barrier to the development of these interventions is a lack of understanding of the mechanisms that mediate breastfeeding transmission of HIV . Many species of African nonhuman primates , including the sooty mangabey ( Cercocebus atys ) are naturally infected with simian immunodeficiency virus ( SIV ) . In contrast to SIV infection of Asian macaques and HIV infection of humans ( non-natural hosts ) , natural SIV infections are typically non-pathogenic ( reviewed in [2] ) . These differential outcomes of SIV/HIV infection are consistently observed despite high-level virus replication in both natural and non-natural hosts . We have recently demonstrated that mother to infant transmission ( MTIT ) is rare in naturally SIV-infected sooty mangabeys ( SMs , <7% ) , compared to the much higher rates seen in SIV-infected rhesus macaques ( RMs , 25–75% ) and HIV-infected humans ( ∼40% ) [3]–[7] . The difference between natural and non-natural hosts is particularly striking in terms of breastfeeding transmission , given that post-partum experimental SIV infection of lactating mandrills ( natural hosts ) resulted in no MTIT events compared to the 60–75% transmission rate seen in breastfeeding RMs ( non-natural hosts ) [4] , [6] , [8] . Thus , over thousands of years of virus-host interaction , natural SIV hosts have evolved mechanisms to render SIV infection non-pathogenic and to restrict SIV transmission from mothers to infants . The rarity of MTIT in natural hosts has been hypothesized to be due to low levels of CD4+CCR5+ target cells for SIV infection found in peripheral blood and mucosal sites [8] , [9] . However , an in depth comparative analysis of target cell availability in multiple tissues had not previously been conducted in nonhuman primates . In the current study , we investigated several potential mechanisms to explain the differential rates of MTIT in natural and non-natural hosts . Specifically , we asked whether the low rate of MTIT observed in SMs could be due to i ) low levels of virus present in breast milk; ii ) enhanced inhibitory properties of breast milk , and/or iii ) limited availability of SIV target cells in infants . By directly comparing SMs and RMs we found similar properties in the breast milk of these natural and non-natural host nonhuman primate species . However , we observed strikingly lower levels of CD4+CCR5+ SIV target cells in infant SMs as compared to infant RMs . Based on these data we propose that limited target cell availability , rather than breast milk specific factors , is responsible for the rarity of mother-to-infant transmission of SIV in SMs .
To investigate the virologic and immunologic features of breast milk that may influence transmission of SIV in nonhuman primates , we pharmacologically induced lactation in female non-pregnant chronically SIV-infected RMs and SMs . Clinical features of the study animals are listed in Table 1 . RMs and SMs were experimentally infected with SIVmac239 and SIVsmm , respectively . Viral loads prior to pharmacologic induction of lactation ( during chronic infection ) ranged from 5 . 7×103 to 3 . 8×105 copies/ml plasma . In keeping with the nonpathogenic nature of SIV infection of SMs , CD4+ T cell counts were higher in SMs than RMs ( mean of 450 cells/mm3 compared to 227 cells/mm3 ) . The protocol used to induce lactation has previously been tested in RMs and African green monkeys and the breast milk produced is immunologically similar to that of milk collected from naturally-lactating monkeys [10] , [11] . This protocol involves treating the animals with medroxyprogesterone and estradiol to mature the mammary glands , haloperidol ( a dopamine receptor antagonist ) to raise endogenous prolactin levels , and oxytocin given prior to milk expression to simulate the milk ejection reflex . Lactation was successfully induced in all animals by weeks 4–6 and was sustained for 5 months . Throughout the lactation period , we monitored for effects of the used treatments on mononuclear cells in peripheral blood . Specifically , we measured the relative percentage of mononuclear cell subsets , the differentiation status of CD4+ and CD8+ T cells , as well as the level of T cell activation by flow cytometry . Minor perturbations were seen at various time points , but the pharmacologic agents did not induce a sustained change in any of the measured parameters ( data not shown ) . The level of HIV in plasma and breast milk of HIV-infected women correlates with mother to child transmission ( MTCT ) of HIV [12]–[17] . To investigate the viral loads in the milk of SMs and RMs undergoing hormone-induced lactation we first measured the level of cell-free SIV RNA by RT-PCR in de-fatted milk as well as plasma from these animals . As shown in Figure 1A , we found that plasma viremia was stable over the period of lactation and was similar in both RMs and SMs . Figure 1B shows that there was no overall difference in the level of cell-free SIV RNA in milk from RMs and SMs . At week 20 , temporally associated with a case of clinical mastitis , the macaque RRw9 did exhibit an increased level of SIV RNA in milk compared to other animals . We next measured cell-associated SIV DNA in breast milk cells and PBMCs from RMs and SMs , as several researchers have proposed that the level of cell-associated HIV , rather than cell-free HIV , is the main mediator of MTCT via breast milk [12] , [15] , [16] . We were able to detect SIV DNA in PBMCs from both RMs and SMs , but levels were more variable in RMs ( Fig . 1C ) . Interestingly , SIV DNA was infrequently detected in milk cells from RMs , but was more consistently detected in milk cells from SMs ( Fig . 1D ) . For each virologic comparison , area under the curve analysis followed by the nonparametric Mann-Whitney test showed no significant difference between cell-free or cell-associated viral loads in RMs vs . SMs . Taken together , these comparative data on virus load in the milk of RMs and SMs do not support the hypothesis that lower virus levels ( cell-free or cell-associated ) are responsible for reduced SIV transmission to infant SMs . As we found that the viral loads in the milk did not differ between RMs and SMs , we next sought to determine whether the immunologic features of milk might account for the differential rates of MTIT of SIV in these two species . To this end , we analyzed the relative proportions of CD4+ and CD8+ T cells and their subsets in breast milk from RMs and SMs ( see Fig . 2 for representative flow plots ) . CD45 was used to distinguish cells of hematopoietic origin in milk . We found that CD4+ T cells comprised 19 . 9–36 . 2% of CD45+CD3+ live cells in SMs and 1 . 7–22 . 9% in RMs ( Table 2 ) . We next measured the T cell differentiation status of CD4+ and CD8+ T cells derived from RM and SM milk . As expected , the vast majority of CD4+ and CD8+ T cells in milk from SMs and RMs possessed a memory phenotype ( >93% , Table 2 ) . Memory CD4+ T cells from SMs were predominantly comprised of effector memory cells ( CD28+/−CD95+CCR7− ) , consistent with the possibility that these cells are the major contributors to the observed level of cell-associated SIV DNA in milk . To further investigate the immunologic profile of milk , we measured the level of cytokines and chemokines in de-fatted milk from SIV-infected RMs and SMs following pharmacologic induction of lactation . We found that all of the tested immunological factors were present at similar levels in SIV-infected RMs and SMs ( Table 3 , left columns ) . Although not statistically significant ( likely related to sample size ) , milk from SIV-infected RMs contained higher levels of RANTES and IL-15 compared to milk from SIV-infected SMs . Higher milk levels of IL-15 have been associated with protection from HIV MTCT [18]; in contrast , RANTES in milk has been positively correlated with MTCT of HIV [19] , [20] , thereby confusing our interpretation of the elevated levels of both of these factors in RMs . We also compared the levels of cytokines/chemokines in milk and plasma within each species and found that , with the exception of IL-12 p40 in RMs and Eotaxin in SMs , milk contains higher levels of all tested immune factors ( data not shown ) . Recognizing that our sample size of infected animals was small , we further characterized immune mediators in milk of seven naturally-lactating , SIV-uninfected RMs and SMs . We were not able to perform the same comparison among naturally-lactating , SIV-infected animals as infected RMs are excluded from the breeding colony . Milk was collected by manual expression between 65 and 255 days post-partum and the de-fatted milk was used in a multiplex panel . We found that milk from uninfected SMs had elevated IL-8 compared to uninfected RMs ( p<0 . 001 , Table 3 , right columns ) ; SIV-infected SMs also had higher levels of this chemokine compared to SIV-infected RMs ( as well as uninfected SMs ) , although this comparison was not statistically significant . When statistical significance was assessed at the p<0 . 01 level , uninfected RMs had higher levels of IL-10 compared to uninfected SMs . At the p<0 . 05 level , uninfected RMs had higher levels of IL-1β , IL-6 , IL-15 , EGF , VEGF , and Eotaxin , but a lower level of MCP-1 . We did not detect a difference between RMs and SMs in the levels of the CCR5 ligands MIP-1α , MIP-1β , or RANTES in the SIV-uninfected animals ( Table 3 , right columns ) . Overall , while some differences were seen between RMs and SMs , a particular pattern that could be linked to enhanced or reduced transmission did not emerge in either the naturally-lactating or pharmacologically-lactating animals . Many cross-sectional and in vitro studies using human milk support a role for innate factors in conferring either resistance to or promotion of transmission ( reviewed in [21] ) . Additionally , previous work has identified specific factors in semen ( SEVI ) or blood ( VIRIP ) that respectively promote or restrict HIV infection [22] , [23] . While we did not find candidate inhibitors ( or enhancers ) of breastfeeding transmission of SIV in milk from SMs or RMs among the studied cytokines or chemokines , these experiments did not directly rule out the hypothesis that other factors , including innate milk constituents such as lactoferrin or mucin , may directly influence the infectivity of SIV in milk . Therefore , we sought to test directly whether milk from SMs was more able to inhibit in vitro SIV infection compared to milk from RMs . For these studies only milk from naturally-lactating , SIV-uninfected animals was used . Viral pseudotypes expressing envelope ( Env ) glycoproteins from two SMs naturally infected with SIVsmm subtypes 1 ( FFv18NOV04PLENV2 . 1 ) and 2 ( FWk12AUG04ENV4 . 1 ) and neutralizing antibody-sensitive and -resistant SIVmac251 Envs ( SIVmac251 . 6 and SIVmac251 . CS , respectively ) were used to infect the Tzm-bl reporter cell line . As shown in Figure 3 , we found that de-fatted milk diluted 1∶10 and 1∶50 from both SMs and RMs generally inhibited in vitro SIV infection by 40–50% , with RMs exhibiting more efficient inhibition at the 1∶10 dilution ( p = 0 . 0238 , Mann-Whitney test ) . SMs were able to equally inhibit infection of genetically divergent SIVsmm and SIVmac251 Envs , suggesting that this inhibition was not virus strain-specific and potentially not Env-dependent . The same was seen for RM milk inhibition of both the SIVmac251 and SIVsmm Envs . Thus , while we did observe a modest inhibition of in vitro SIV infection in the presence of SM or RM milk , there was no difference in inhibitory capacity between these two species . As we did not find specific differences in the breast milk of frequently-transmitting RMs vs . rarely-transmitting SMs , we next investigated whether these differential rates of MTIT could be better explained by infant factors . Previous work from our group has shown restricted expression of CCR5 on central memory CD4+ T cells from adult SMs [24] , and a paucity of CCR5 on circulating CD4+ T cells from infant mandrills was proposed as a mechanism to explain the absence of breastfeeding transmission in this species [8] . Therefore , we next sought to measure the availability of SIV target cells in infant SMs and RMs , focusing on tissues along the gastrointestinal tract as well as lymph nodes , by performing multicolor flow cytometry on cells isolated from nonhuman primate infant tissues at necropsy . Table 4 shows the ages and causes of death of the SM and RM infants whose tissues were studied . All infants were born to SIV negative mothers and , for the SMs , all parents were homozygous wild type for CCR5 . It should be noted that this is the first immunological analysis of infant SM blood and tissues . As such , we performed a detailed characterization of the CD4+ T cells obtained . Naïve and memory T cells were defined by expression of CD28 and CD95 as previously described [25] . As expected based on the age of these animals , the majority of CD4+ T cells had a naïve phenotype ( Fig . 4A ) , but a greater percentage of memory CD4+ T cells was found in the upper intestine compared to other sites ( mean of 27% in the jejunum ) . At lymphoid sites , RM infants had a similar predominance of naïve over memory CD4+ T cells , but the gastrointestinal tissues and spleen of RMs had an increased proportion of memory CD4+ T cells compared to the same sites in SMs ( Fig . 4A ) , a finding that could be due in part to the slightly older age of the RM infants available for study . We then measured the percentage of proliferating cells in infant RMs and SMs at these tissue sites using the marker Ki-67 . We found robust levels of proliferation in CD4+ T cells from both RMs and SMs ( SMs: range 3–18% Ki-67+; RMs: range 3–40% Ki-67+ ) ( Fig . 4B ) . These levels of CD4+ T cell proliferation are higher than previously seen in adult SMs [26]–[28] and are likely due to the developing infant immune system . To better identify and quantitate SIV targets in infant SMs and RMs , we then determined the frequency of CD4+ T cells expressing CCR5 in blood , lymphoid tissues and the gastrointestinal tract . As shown in Fig . 5A , we found minimal expression of CCR5 on total and CD28+/−CD95+ memory CD4+ T cells in SM infants ( median of <4% for all tissues ) . By comparison , RM infants had statistically significantly higher levels of CCR5+ total and memory CD4+ T cells at all sites . The difference was most pronounced in the RM spleen ( median of 22 . 1% CCR5+CD4+ memory T cells ) , jejunum ( median of 34% CCR5+CD4+ memory T cells ) , and colon ( median of 43% CCR5+CD4+ memory T cells ) . Due to the small numbers of cells obtained from some tissues , we were not able to further distinguish the central and effector memory populations to assess their levels of CCR5 in all animals . When sufficient cell numbers could be analyzed ( primarily in the lymph nodes and spleen , but also in the gastrointestinal tract of several animals ) , we consistently noted very low levels of CCR5 on central and effector memory subsets of SMs ( Fig . 5B ) , but higher levels in RMs ( Fig . 5C ) . Finally , we performed a direct comparison of the levels of SIV target cells within T cell memory subsets found in the peripheral blood of infant SMs and RMs . Among circulating CD28+CD95−CCR7+ naïve CD4+ T cells the levels of CCR5 were low and not different between RMs and SMs ( Fig 5D ) . However , we found significantly higher levels of CCR5 on CD28+CD95+CCR7+ central memory CD4+ T cells from RMs ( median 6 . 4% , range 4 . 5–9 . 7% ) compared to SMs ( median 0 . 6% , range 0 . 2–2 . 8%; p<0 . 001 ) . CCR5 was also expressed by a greater percentage of CD28+/−CD95+CCR7− effector memory CD4+ T cells from RMs ( median 32 . 7% , range 21 . 8–48 . 2% ) compared to SMs ( median 3 . 3% , range 0–5 . 4%; p<0 . 001 ) . These data confirm that infant RMs have ample target cells for SIV infection [29] . In summary , the level of memory CD4+CCR5+ T cells that are preferential targets for SIV infection is significantly lower in SM infants as compared to RM infants . We next sought to assess SIV target cells at sites that are the first to encounter HIV/SIV transmitted through breast milk , specifically the buccal mucosa , tonsils , and esophagus . Figure 6A shows representative data from two SM infants , one who was euthanized on the 10th day of life and another who was euthanized at 107 days . We found a similar pattern of robust proliferation with little CCR5 expression on CD4+ T cells from these sites , although in the older infant the level of CCR5 was slightly increased in the gastrointestinal tissues ( 4 . 9–5 . 7% of CD4+ T cells ) , but not in tonsillar tissue ( Fig . 6A ) . In Figure 6B , we present corresponding data from RM infants . Interestingly , we found low levels of CCR5 on CD4+ T cells in the buccal mucosa , esophagus , and tonsil in an RM infant euthanized in utero ( secondary to maternal complications ) and on the first day of life . However , we observed a massive expansion of CD4+CCR5+ T cells ( 31–62% ) in all three sites before one month of age , clearly indicating that high levels of SIV target cells are present in RM infants early in life . Due to a concern that coreceptors other than CCR5 have the potential to mediate SIV infection of SM infants , we measured the expression of CXCR6 and GPR15 by real time PCR in sorted CD4+ T cells from infant SMs and RMs . These coreceptors have previously been demonstrated to mediate SIVsmm entry in transfected cell lines and may be involved in SIV infection of SMs with CCR5 deletions [30] , [31] . As shown in Figure 7 , we found very low levels of these coreceptors in the SM infants ( whose parents all had wild type CCR5 genotypes ) . In addition , we measured cell surface GPR15 by flow cytometry on infant SM CD4+ T cells both pre- and post-activation with Concanavalin A and IL-2 and did not find significant expression ( data not shown ) . CXCR6 staining was not performed due to lack of commercially available monoclonal antiboides that cross-react with SMs and the minimal RNA expression we observed ( Fig . 7 ) . In summary , these data show that SM infants , unlike RM infants , have limited availability of SIV target cells in multiple tissues , thus providing a potential mechanism of resistance from MTIT in this natural host species .
Mother to child transmission of HIV still accounts for approximately 370 , 000 new pediatric infections per year , with close to 50% of infections attributed to breastfeeding [1] . Despite similar plasma viremia and breastfeeding practices ( e . g . , frequency and duration ) , natural hosts and non-natural hosts for SIV demonstrate highly divergent rates of MTIT in the wild and in captivity [4]–[8] , [32] , [33] . We have previously shown that the rate of MTIT of SIV in the natural host sooty mangabey ( SM ) is significantly lower than in the non-natural host rhesus macaque ( RM ) and , critically , is also significantly lower than the rate of MTCT of HIV in the absence of any preventative intervention [3] . In this study , we investigated the physiologic mechanisms underlying the rarity of MTIT in SIV-infected SMs . We examined three potential mechanisms to explain the low rates of MTIT of SIV in SMs . The first potential mechanism is that breast milk of natural hosts is less infectious because of lower viral loads . It should be noted that this study represents the first ever analysis of breast milk from SMs . The evidence did not support this hypothesis , as we have shown here that SMs and RMs have similar levels of cell-free SIV RNA and cell-associated SIV DNA in breast milk over a 5-month period , with SMs actually demonstrating more consistent virus shedding into the breast milk compartment ( Fig . 1 ) . Similar milk viral loads were reported in a comparison of RMs with another natural host species , the African green monkey , from a single sampling day [11] . Thus , the resistance to MTIT in natural hosts does not appear to be related to the inability of virus to traverse the mammary epithelium or replicate locally within the breast . The second potential mechanism to account for restricted MTIT in SMs is that the breast milk of natural hosts contains inhibitory factors that limit virus infectivity at the level of infant mucosa . There have been a number of innate factors described in human milk that either inhibit HIV infection in vitro or that have been epidemiologically linked to lower rates of transmission , such as lactoferrin , long chain polyunsaturated fatty acids , oligosaccharides , SLP1 , and a number of cytokines/chemokines [21] . We hypothesized that SMs might possess a specific factor ( or factors ) that interferes with SIV infection , perhaps similar to the VIRIP isolated from human blood [23] . However , while we found that milk from both SMs and RMs partially inhibits infection of Tzm-bl cells at the 1∶10 and 1∶50 dilutions , SMs did not more efficiently block infection ( Fig . 3 ) . When specific cytokines and chemokines were quantified in milk from both species , SMs consistently expressed more of the pro-inflammatory chemokine IL-8 than RMs ( Table 3 ) . IL-8 is increased in cases of subclinical mastitis [34] and milk containing high-level IL-8 ( in addition to other pro-inflammatory cytokines ) has been shown to enhance cell-associated HIV infection [35] , a finding that is difficult to reconcile with protection from MTIT in SMs . In addition , levels of specific factors with known antiviral activity , such as IFN-β , IFN-γ , TNF-α , MIP-1α , MIP-1β , RANTES , and MDC , did not differ between the natural and non-natural hosts for SIV ( Table 3 and data not shown ) . The third mechanism we investigated is that infant SMs have limited numbers of CD4+CCR5+ target cells for SIV infection . We favored this hypothesis as we have previously shown that adult SMs express lower levels of CCR5 on CD4+ T cells compared to RMs , and particularly so in central memory CD4+ T cells [24] , [36] . However , the role of CCR5 expression on infant CD4+ T cells as a mechanism to prevent MTIT of SIV in SMs had not been previously investigated . In recent work , intra-rectal SIV infection of African green monkeys and pigtailed macaques was found to be dependent on the level of CD4+CCR5+ T cells at the site of exposure , with a significant positive correlation between the percentage of CD4+CCR5+ T cells and the number of transmitted founder viruses [9] . Moreover , this study reported low levels of CCR5+CD4+ T cells in juvenile African green monkeys , with an increased percentage of CCR5+CD4+ T cells found in the blood and jejunum of sexually mature animals . In the present study , we greatly expanded the number of mucosal and lymphoid sites interrogated in a natural host model , examined CCR5 expression in CD4+ T cell subsets , and provided the first ever characterization of SIV target cells in infant SMs . Consistent with the hypothesis that MTIT of SIV is dictated by the level of CD4+CCR5+ target cells , we found extremely low percentages of these cells at all sites in infant SMs , including those that are in first contact with ingested virus such as the oral and esophageal mucosa ( Fig . 6 ) . In contrast , infant RMs had a much higher percentage of circulating CD4+CCR5+ T cells within both the central and effector memory subsets ( Fig . 5D ) as well as higher levels of CD4+CCR5+ memory T cells in all examined tissues , which may explain their increased susceptibility to MTIT of SIV compared to natural hosts . Interestingly , while cord blood CD4+ T cells do not express CCR5 in humans , an abundant presence of CD4+CCR5+ memory T cells has been found in the gastrointestinal tract of human infants and fetal tissue [37] . These cells were highly susceptible to HIV infection in vitro and probably represent the main targets for infection following ingestion of HIV in breast milk . We observed lower levels of SIV target cells in the oral mucosa , esophagus , and tonsils of RMs euthanized in utero and on the first day of life compared to an RM euthanized at 27 days of life ( Fig . 6 ) , suggesting rapid upregulation of CCR5 expression post-partum in this non-natural host . This increase in CCR5 expression correlates with high rates of breastfeeding transmission of SIV in RMs [4] , [6] . In this context , the major biological feature that may explain the differential rates of SIV MTIT in natural hosts as compared to non-natural host RMs or humans is the availability of target cells at mucosal and lymphoid sites . We have long known that SIV-infected SMs maintain high viral loads and , in chronic infection , that CD4+ effector memory T cells that express higher levels of CCR5 ( compared to central memory T cells ) likely produce the majority of virus [24] . SIV infection of a small percentage of sooty mangabeys homozygous for a deletion mutation in CCR5 ( delta 2 or delta 24 ) that abrogates its cell surface expression may be mediated by the alternative coreceptors GPR15 and CXCR6 [38] . Interestingly , viral loads in these CCR5 knock-out SMs are approximately 0 . 5 to 1-log lower than in animals wild type for CCR5 . We here investigated the potential role of the alternative SIV coreceptors GPR15 and CXCR6 in mediating SIV transmission to SM infants and found their expression to be very low in CD4+ T cells ( Fig . 7 ) , indicating that SIV infection of target cells is unlikely to be mediated by these pathways in infant SMs . One limitation of this study is that , due to federal regulations , we could only perform opportunistic collection of tissues from euthanized SM infants , and several of our study animals died from sepsis-related causes that might have influenced the overall level of CD4+ T cell activation . In light of this limitation , however , the very low level of CCR5 we observed in SM infants is perhaps even more striking . A second limitation is that we did not specifically investigate the role of the adaptive immune response , present either in milk or in the circulation of SM infants , as a potential mechanistic factor that limits MTIT of SIV . The rationale to prioritize other virological and immunological analyses in the samples that were available from the animals included in this study was that ( i ) SIV-infected SMs show circulating virus-specific CD8+ T cell responses that are of modest magnitude and are ineffective in controlling virus replication as well as low in vitro titers of autologous neutralizing antibodies in the plasma [39] , [40] , and ( ii ) the robust levels of virus , both cell-free and cell-associated , in the milk of lactating SIV-infected SMs that argue against immune control . While a neutralization response against the challenge virus was detected in both plasma and milk from SIV-infected African green monkeys , the importance of this finding remains unclear given the robust viral loads ( 104–105 copies / ml plasma ) observed in both body fluids [11] . Interestingly , in the rare SMs who become infected via vertical transmission as well as experimentally-infected neonatal African green monkeys , viral loads are 1- to 2-log lower than those seen in animals who become infected later in life by the horizontal route [3] , [41] . This low-level viremia is also consistent with the limited availability of SIV target cells in natural host infants and supports the hypothesis that low expression of CCR5 on CD4+ T cells has evolved as a major mechanism of protection from vertical transmission of SIV . The limited expression of CCR5 in natural hosts differs from what is seen in human infants ( as described above ) and it should be remembered that perinatally HIV-infected infants maintain high viral loads in absence of therapy and rapidly progress to AIDS . One could speculate that , thousands of years ago , natural host infants displayed a similar phenotype to perinatally HIV-infected children , resulting in poor clinical outcomes and “dead-end” infection in terms of transmission . Coevolution of virus and host may have selected for limited MTIT by decreased expression of CCR5 on all CD4+ T cell subsets in infants with widespread infection of older animals due to moderate expression of CCR5 on effector memory T cells . The consequent lack of pathogenesis in adult mangabeys is attributed to sparing of the central memory compartment from SIV infection [24] , but it should be noted that the exact mechanism that determines cell surface expression of CCR5 on different cell subsets in SMs remains unknown . In this scenario , infants are spared a highly pathogenic and universally fatal infection and thus propagation of the species continues . At the same time , the virus is able replicate for many years in adult natural hosts and is readily horizontally transmitted . Efforts to harness this mode of protection from MTIT in natural hosts for therapeutic purposes may have a major impact on HIV prevention , treatment , and cure strategies .
This study was conducted in strict accordance with USDA regulations and the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health , and were approved by the Emory University Institutional Animal Care and Use Committee ( AWA# A3180-01 ) . SIV-infected animals were housed in standard non-human primate cages , received standard primate feed as well as fresh fruit and enrichment daily , and had continual access to water . Cages also contained additional sources of animal enrichment including objects such as perching and other manipulanda . Animal welfare was monitored daily . Appropriate procedures were performed to ensure that potential distress , pain , or discomfort was alleviated . The sedatives Ketamine ( 10 mg/kg ) or Telazol ( 4 mg/kg ) were used for blood draws and expression of breast milk . Euthanasia of infant SMs or RMs , using Pentobarbital ( 100 mg/kg ) under anesthesia , was performed only when deemed clinically necessary by veterinary medical staff and according to IACUC endpoint guidelines . For the studies of induced lactation , three chronically SIV-infected SMs ( Cercocebus atys ) and RMs ( Macaca mulatta ) were followed . As part of earlier studies conducted prior to 2006 [42] , [43] , SMs were infected intravenously with 1 ml of plasma obtained from a naturally SIVsmm-infected SM ( viral load 4×106 copies / ml plasma ) . RMs were infected intravenously with 200 TCID50 SIVmac239 and were Mamu-B*08 and –B*17 negative . Milk was also obtained from SIV-uninfected SMs and RMs in breeding colonies . For the studies of SIV target cells in tissues , SM and RM infants who died within the first year of life underwent necropsy ( Table 4 ) . For the measurement of SIV target cells in CD4+ T cell memory subsets in peripheral blood , SMs and RMs <24 months of age were used . All RM and SM infants were all born to SIV-uninfected parents . All parents of the SM infants studied were homozygous wild type for the CCR5 gene . All animals were housed at the Yerkes National Primate Research Center . Nonpregnant female SMs and RMs were given depot medroxyprogesterone ( 3 mg/kg ) and increasing doses of estradiol ( 0 . 1–0 . 25 mg/kg ) intramuscularly to induce mammary gland maturation . Endogenous prolactin production was stimulated by oral administration of the dopamine antagonist haloperidol ( 0 . 25–0 . 45 mg/kg twice daily ) . Milk was collected three times per week via manual expression following intramuscular injection of 4–10 units of oxytocin . Doses were optimized for maximum milk collection and modeled after [10] , [11] . Serum levels of progesterone , estradiol , and prolactin were monitored by ELISA ( ALPCO Diagnostics ) . Levels of mononuclear cells in peripheral blood , including T cell subsets , as well as their proliferation , activation , and SIV coreceptor expression were measured by flow cytometry throughout the period of induced lactation . Milk was separated into the lipid layer , de-fatted ( “skim” ) milk , and cellular portion by centrifugation at 710×g for 20 min . The cell pellet was washed three times prior to staining for flow cytometry or freezing in RLT plus ( Qiagen ) for subsequent viral quantification . Peripheral blood mononuclear cells ( PBMCs ) were isolated by gradient centrifugation . Procedures for processing of lymphoid and gastrointestinal tissues and isolation of lymphocytes were performed as described in [28] . Tonsils were processed in the same fashion as lymph nodes . Buccal mucosa was processed similar to other gastrointestinal sites . Immunophenotyping was performed according to standard procedures using multicolor flow cytometry and monoclonal antibodies that were originally designed for humans or macaques and have been found to be cross-reactive in SMs . For analysis of SIV target cells in infants the following antibodies were used: Live/Dead Fixable Aqua from Invitrogen; CD4-Pacific Blue ( OKT4 ) from BioLegend; CD28-PE-Texas Red ( CD28 . 2 ) from Beckman-Coulter; CD95-PE-Cy5 ( DX2 ) from eBioscience; CD8-APC-Cy7 ( SK1 ) , CD3-Alexa 700 ( SP34-2 ) , CCR7-PE-Cy7 ( 3D12 ) , Ki-67-FITC ( B56 ) , CCR5-PE ( 3A9 ) all from Becton Dickinson . For analysis of RM and SM PBMCs the following additional antibodies were used: CD62L-PE ( SK11 ) , CCR5-APC ( 3A9 ) , CD14-Pacific Blue ( M5E2 ) , CD11c-APC ( S-HCL-3 ) , HLA-DR-APC-Cy7 ( G46-6/L243 ) all from Becton Dickinson; CD95-PE-Cy7 ( DX2 ) and CD123-PE-Cy7 ( 6H6 ) from eBioscience; CD20-PE-TexasRed ( B9E9 ) from Beckman-Coulter . For analysis of RM and SM breast milk cells , the following additional antibodies were used: CD45-V450 ( D058-1283 ) , CD28-PE-Cy5 ( CD28 . 2 ) , CD3-PerCP-Cy5 . 5 ( SP34-2 ) , CD14-FITC ( MphageP9 ) , Ki-67-Alexa 700 ( B56 ) all from Becton Dickinson; CD8-Pacific Orange ( 3B5 ) and CD4-PE-Cy5 . 5 ( S3 . 5 ) from Invitrogen . Flow cytometric acquisition was performed on an LSRII cytometer driven by the FACS DiVa software ( Becton Dickenson ) . Analysis of the acquired data was performed using FlowJo software ( TreeStar ) . The QIAamp viral RNA kit ( Qiagen ) was used to extract viral RNA from SM and RM skim milk and SM plasma , according to the manufacturer's recommendations . RM plasma was extracted with the Magna Pure LC ( Roche Diagnostics; Indianapolis , IN ) , according to the manufacturer's recommendations . Plasma viral RNA quantification was performed as previously described [26] . RM skim milk RNA was quantified using SIVgag primers and probe along with conditions described in [44] with a sensitivity of 153 copies/ml . SM skim milk RNA was quantified with primers and probe along with conditions described in [45] with slight modification: 27 µl of RNA were loaded into a 100 µl two-step reaction with a sensitivity of 53 copies/ml . Skim milk spiked with known quantities of either SIVmac239 or SIVsmm were used as positive controls . Samples with undetectable SIV RNA were assigned a level of half of the lower limit of detection for graphical purposes . Cell-associated proviral DNA from SM and RM breast milk cells and PBMC was obtained by extracting total DNA with the AllPrep DNA/RNA kit ( Qiagen ) . Then , 80 ng per 15 µl of DNA were loaded into a 50 µl reaction using an SIVgag primer/probe set ( RM ) or an SIVutr primer/probe set ( SM ) as described in [44] , [45] . Albumin was used as an internal control to quantify cell number against an external standard curve . Albumin primers and probe along with qPCR conditions were previously described in [46] . The proportion of SIV+ cells was determined as previously reported in [11] . The sensitivity of the assay is 5 SIV DNA copies per 105 cells . Samples with undetectable SIV DNA were assigned a level of half of the lower limit of detection for graphical purposes . Plasma and milk levels of twenty-eight cytokines , chemokines , and growth factors were measured using a sandwich immunoassay-based protein array system , the Monkey Magnetic 28-Plex Panel ( Invitrogen ) as instructed by the manufacturer and then read by the Luminex 100 reader ( Luminex Corp ) , which uses fluorescent bead-based technology . IFN-β in milk was measured using a simian ELISA kit ( Ucsn Life Sciences Inc . ) Virions , pseudotyped with one of four SIV envelopes ( FFv18NOV04PLENV2 . 1 , FWk12AUG04ENV4 . 1 , SIVmac251 . 6 , or SIVmac251 . CS ) , were created via co-transfection with an Env-deficient subtype B HIV-1 proviral plasmid ( pSG3Δenv ) in 293T cells using FuGENE HD ( Promega ) . Viral titers were determined via infection and β-galactosidase staining of the Tzm-bl indicator cell line . Five-fold serial dilutions ( beginning at 1∶10 ) of de-fatted milk from RMs ( 8 animals ) or SMs ( 7 animals ) were then tested for inhibitory potential against these viral pseudotypes . Roughly 6 , 000 Tzm-bl cells per well were plated and cultured overnight in flat-bottomed 96-well plates . Pseudovirus ( 2 , 000 IU ) in DMEM with ∼3 . 5% FBS ( HyClone ) , 40 µg/ml DEAE-dextran was incubated with diluted milk for approximately 1 hour before 100 µl of this mixture were added to plated Tzm-bl cells for a 48 hr infection . Cells were subsequently lysed and evaluated for luciferase activity , and raw data was retrieved from a BioTek Synergy HT multi-mode microplate reader with Gen 5 , v2 . 00 software . The average background luminescence from a set of uninfected wells was subtracted from each experimental well , infectivity curves were generated using GraphPad Prism v5 . 0d where values from experimental wells were compared against a well containing virus only with no milk , and IC50 values were determined using linear regression in Microsoft Excel for Mac 2011 , v14 . 2 . 3 . CD4+ T cells were negatively selected from frozen PBMC using magnetically labeled microbeads and subsequent column purification according to the manufacturer's protocol ( Miltenyi Biotec ) . The purity of enriched CD4+ T cells was evaluated by flow cytometry and was >80% in all cases . Sorted CD4+ T cells were lysed in RLT buffer ( QIAGEN ) and stored at −80°C . Total RNA from PBMC was purified using RNeasy mini kit ( QIAGEN ) according to manufacturer's protocol utilizing on-column DNAse digestion . RNA quantity was measured using Nanodrop analysis . RNA samples ( 0 . 3–1 . 0 µg ) were reverse transcribed in a volume of 20 µl using random hexamers and the Super Script II kit ( Invitrogen ) as previously described [47] and 0 . 2 µl of cDNA was used for real time SYBR green PCR analysis using an ABI 7900 HT instrument ( Applied Biosystems ) . Primer sequences for PCR were CXCR6: Fwd 5′-ACCCTGTGCTCTATGCCTTTGTCA-3′ , Rev 5′-AAGGGAGACAGCCAATGTCCTTCA-3′; GPR15: Fwd 5′-ACTGCAGTGTCTTCCTGCTCACTT-3′ , Rev 5′-AAACCAGATGCTGGCGCAAACTAC-3′ . Serial dilutions of pcDNA3 . 1-smGPR15 , and CXCR6 plasmids ( kindly provided by R . Collman , [30] , [47] ) were used for the quantification . Results were normalized to input RNA . Comparisons between the level of SIV target cells in RMs and SMs as well as the level of cytokines and chemokines in milk of RMs and SMs were made using the nonparametric Mann-Whitney test . The inhibitory capacities of RM and SM milk against in vitro SIV infection were compared at the 1∶10 and 1∶50 dilutions by nonparametric Mann-Whitney test . Comparisons between cell-free and cell-associated virus levels in blood and milk from RMs vs . SMs over time were made using area under the curve analyses followed by nonparametric Mann-Whitney test . In all cases , significance was attributed at p<0 . 05 . All analyses were conducted using GraphPad Prism 4 . 0c or 5 . 0d . | Currently 2 . 5 million children are infected with HIV , largely as a result of mother-to-child transmission , and there is no effective vaccine or cure . Studies of Simian Immunodeficiency Virus ( SIV ) infection of nonhuman primate species termed “natural hosts” have shown that mother-to-infant transmission of SIV in these animals is rare . Natural hosts are African monkey species that are naturally infected with SIV in the wild but do not develop AIDS . We sought to understand the mechanism by which natural hosts are protected from mother-to-infant transmission of SIV , aiming to translate our findings into novel strategies to prevent perinatal HIV infection . We found that natural host sooty mangabey infants have extremely low levels of target cells for SIV infection in lymphoid and gastrointestinal tissues . Direct comparison of infant sooty mangabeys and infant rhesus macaques ( non-natural host species with high SIV transmission rates ) confirmed that natural hosts have significantly lower levels of SIV target cells compared with non-natural hosts . Analysis of the breast milk of sooty mangabeys and rhesus macaques revealed similar levels of virus and ability to inhibit SIV infection . Our study provides evidence for target cell restriction as the main mechanism of protection from mother-to-infant SIV transmission in natural hosts . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"immune",
"cells",
"viral",
"transmission",
"and",
"infection",
"retrovirology",
"and",
"hiv",
"immunopathogenesis",
"immunology",
"host-pathogen",
"interaction",
"microbiology",
"animal",
"models",
"immunodeficiency",
"viruses",
"mechanisms",
"of",
"resistance"... | 2014 | Target Cell Availability, Rather than Breast Milk Factors, Dictates Mother-to-Infant Transmission of SIV in Sooty Mangabeys and Rhesus Macaques |
Alternative transcriptional initiation ( ATI ) refers to the frequent observation that one gene has multiple transcription start sites ( TSSs ) . Although this phenomenon is thought to be adaptive , the specific advantage is rarely known . Here , we propose that each gene has one optimal TSS and that ATI arises primarily from imprecise transcriptional initiation that could be deleterious . This error hypothesis predicts that ( i ) the TSS diversity of a gene reduces with its expression level; ( ii ) the fractional use of the major TSS increases , but that of each minor TSS decreases , with the gene expression level; and ( iii ) cis-elements for major TSSs are selectively constrained , while those for minor TSSs are not . By contrast , the adaptive hypothesis does not make these predictions a priori . Our analysis of human and mouse transcriptomes confirms each of the three predictions . These and other findings strongly suggest that ATI predominantly results from molecular errors , requiring a major revision of our understanding of the precision and regulation of transcription .
The transcription start site ( TSS ) is the first nucleotide transcribed in a run of transcription , while the surrounding genomic region of the TSS is often referred to as the core promoter [1] . Owing to the strong association between TSSs and core promoters , these terms are sometimes used interchangeably [1] . Under an appropriate external signal , the core promoter forms a transcription preinitiation complex with a number of accessory proteins including RNA polymerase and transcription factors to initiate transcription [1–5] . Needless to say , regulation of transcriptional initiation is a crucial step in the control of gene expression [6 , 7] . The transcription of a gene may start from one of several TSSs , a phenomenon known as alternative transcriptional initiation ( ATI ) ; the different core promoters used are referred to as alternative promoters [8 , 9] . It has been reported that ATI occurs to most eukaryotic protein-coding genes [6 , 7 , 10–12] . For example , over 50% of all human genes have alternative promoters [13] , and on average , a human gene has four TSSs [7] . ATI allows the production from the same gene of transcripts differing in the 5′ untranslated region ( 5′ UTR ) or even the protein-coding region . ATI-dependent variations of the 5′ UTR may impact the translational efficiency of the transcript [14] . One example is the human runt-related transcription factor 1 gene RUNX1 , which can be transcribed from two different TSSs; the mRNA produced from the distal TSS mediates cap-dependent translation , whereas that from the proximal TSS contains a functional internal ribosome entry site ( IRES ) and mediates cap-independent translation [15] . ATI-dependent variations of the coding region may affect protein function . For instance , human LEF1 , encoding lymphoid enhancer binding factor 1 that regulates the transcription of Wingless/Integrated ( Wnt ) /β-catenin genes , produces two different protein isoforms by using alternative TSSs; the longer isoform recruits β-catenin to Wnt target genes , whereas the shorter isoform cannot interact with β-catenin and instead suppresses the Wnt regulation of target genes [16] . A number of case studies showed that the TSS choice may vary among tissues [17] , across developmental stages [18 , 19] , or during cell differentiation [20] and that aberrations in the TSS choice can lead to various diseases [21–23] . Such findings led to the adaptive hypothesis that ATI is a widely used , regulated mechanism to expand the transcriptome and/or proteome diversity [7–9 , 24–26] . Nevertheless , alternative TSSs with verified benefits account for only a tiny fraction of all known TSSs , while the vast majority of TSSs have unknown functions . More than 90 , 000 TSSs are annotated for approximately 20 , 000 human protein-coding genes in ENSEMBL genome reference consortium human build 37 ( GRCh37 ) . Recent surveys using high-throughput sequencing methods such as deep cap analysis gene expression ( deepCAGE ) [27] showed that human TSSs are much more abundant than what has been annotated [7] . Are most TSSs of a gene functionally distinct , and is ATI generally adaptive ? While this possibility exits , here we propose and test an alternative , nonadaptive hypothesis that is at least as reasonable as the adaptive hypothesis . Specifically , we propose that there is only one optimal TSS per gene and that other TSSs arise from errors in transcriptional initiation that are mostly slightly deleterious . This hypothesis is based on the consideration that transcriptional initiation has a limited fidelity [28] , and harmful ATI may not be fully suppressed by natural selection if the harm is sufficiently small or if the cost of fully suppressing harmful ATI is even larger than the benefit from suppressing it [29] . The error hypothesis makes a series of distinct predictions about patterns of ATI that are not expected a priori under the adaptive hypothesis . By analyzing high-throughput mRNA 5′-end sequencing data from multiple cell lines and tissues of humans and mice , we provide unequivocal evidence for the error hypothesis . This finding echoes a number of recent discoveries that mechanisms thought to adaptively increase transcriptomic and/or proteomic diversities , such as alternative splicing , alternative polyadenylation , and RNA editing , are all largely manifestations of molecular errors .
Under the error hypothesis of ATI , TSS diversity arises from imprecise transcriptional initiation that could be harmful for several reasons . First , the transcript generated may miss certain regulatory sequences for translation , influencing the protein production . Second , the transcript may lack part of the coding sequence , leading to a reduction , loss , or alteration of the protein function . Third , the transcript may have altered upstream open read frames ( uORFs ) , interfering with normal protein synthesis . Because some of these harms , such as the toxicity of the dysfunctional proteins produced or energy waste owing to the synthesis of functionless proteins , increase with the number of protein molecules synthesized , the overall deleterious effect of imprecise transcriptional initiation of a gene is expected to rise ( but not necessarily linearly ) with its mRNA concentration [30] . Consequently , natural selection against the transcriptional initiation error intensifies with the rise of the gene expression level . As a result , the error rate and TSS diversity should decline with the rise of the gene expression level . By contrast , this trend is not predicted a priori by the adaptive hypothesis , under which the TSS diversity of a gene depends on the specific function and regulation of the gene . To distinguish between the error hypothesis and the adaptive hypothesis of ATI , we analyzed the TSSs identified by 5′-end sequencing ( CAGE-seq and TSS-seq; see Materials and methods ) of numerous cell lines and tissues from humans and mice ( S1 Table ) . Following our recent study of alternative polyadenylation [31] , we used the Simpson index [32] and Shannon index [33] to quantify the TSS diversity of each protein-coding gene in each sample ( see Materials and methods ) . Both indices are commonly used in biodiversity research and tend to rise with the number of TSSs in a gene as well as the evenness of the relative uses of these TSSs , but the Simpson index gives more weight to the frequently used TSSs than does the Shannon index . Because CAGE-seq is regarded as the best among many different 5′-end RNA sequencing ( RNA-seq ) methods [34] , our analyses mainly used TSSs identified by the CAGE-seq high-throughput data of functional annotation of the mouse/mammalian genome project ( FANTOM ) [7] . Let us define a gene by the DNA segment between its 5′-most TSS annotated and its 3′-most end among all annotated transcripts . We concentrated on so-called permissive TSSs ( see Materials and methods ) , which are located within a gene or within 500 bp upstream of the 5′-most end of the gene . Because TSS diversity is poorly estimated when the number of sequencing reads mapped to a gene is too small , only genes with at least 10 reads in a sample were analyzed . By counting the CAGE-seq reads corresponding to each TSS in a gene , we calculated the gene’s TSS diversity . The gene expression level was measured by the number of CAGE-seq reads mapped to the gene per million reads ( RPM ) mapped in the entire sample . We started by analyzing the human universal sample , which is a mixture of 10 cell lines originating from different human tissues [35] . Consistent with the prediction of the error hypothesis , the rank correlation ( ρ ) between the expression level of a gene and its Simpson index of TSS diversity is significantly negative , and this negative correlation is apparent across the entire expression range ( Fig 1A ) . The magnitude of ρ appears small , likely because estimates of gene expression levels and TSS diversities have relatively large sampling errors , especially for lowly expressed genes . When the genes are grouped into 100 bins with the same expression range size or the same gene number , the rank correlation between the mean expression level and mean TSS diversity across the 100 bins becomes –0 . 96 and –0 . 84 , respectively , suggesting that gene expression level is a major determinant of TSS diversity . To rule out the possibility that the observed negative correlation is an artifact of our analysis , we performed a computer simulation . Briefly , we randomly generated genes whose expression level and relative TSS usages respectively follow the gene-expression–level distribution and relative TSS usage distribution of the real genes . We analyzed the simulated genes as if they were the actual data but found a weak , positive correlation between expression level and TSS diversity ( S2 Table ) , confirming that the negative correlation observed in the actual data is not an artifact of our statistical analysis . To investigate the robustness of the above results , we performed several additional analyses . First , because annotated TSSs are generally considered genuine , we focused on TSSs that are within 500 bp from each annotated TSS in a gene [7] , and the obtained result ( S1A Fig ) is highly similar to that from all TSSs . Second , using so-called robust TSSs ( see Materials and methods ) should reduce false-positive TSSs . The obtained result , however , remains qualitatively unchanged ( S1B Fig ) . Third , because some genuine TSSs identified by CAGE-seq may be far upstream from the most upstream TSS currently annotated for a gene [36] , we included TSSs that are located within 1 , 5 , or 10 kb upstream of the most upstream TSS annotated for the gene . But the results were qualitatively unchanged ( S1C–S1E Fig ) . Fourth , while CAGE-seq has been demonstrated to be as reliable as the canonical mRNA sequencing ( RNA-seq ) in measuring gene expression levels [37] , it is valuable to examine whether our observation holds when gene expression levels are measured by RNA-seq . To this end , we replaced the gene expression level in Fig 1A with that obtained from canonical RNA-seq of the same sample [38] . As expected , the result is similar ( S1F Fig ) . Because sequencing depth and the precision of TSS survey for a gene rise with its expression level , it is possible that the correlation in Fig 1A originates from unequal TSS surveys of genes of different expression levels . To eliminate this potential bias , we down-sampled our data by randomly picking 10 CAGE-seq reads per gene for all genes with at least 10 reads and then re-estimated the Simpson index . The correlation ( ρ′ ) between the gene expression level and the re-estimated Simpson index remains negative ( Fig 1B; see S2 Table for the simulation result ) . Similar patterns were observed in other human cell lines and tissues ( Figs 1B and S1 ) . To examine the robustness of our results from the down-sampled data , we down-sampled CAGE-seq reads to as few as 5 and as many as 80 reads per gene from genes with at least that many reads and found our results to remain qualitatively unchanged ( S2 Fig ) . Using the Shannon index to measure TSS diversity similarly yielded a negative correlation between gene expression level and TSS diversity , as shown in Fig 1C for the human universal sample , Fig 1D for the other human cell lines and tissues , and S2 Fig for different levels of down-sampling . To further examine the robustness of the above CAGE-seq–based results , we analyzed two different and independent 5′-end RNA-seq data: TSS-seq and 5′ global run-on sequencing ( GRO-cap ) ( see Materials and methods ) . We examined 42 human cell line/tissue data ( S1 Table ) generated by TSS-seq and observed significant , negative correlations between TSS diversity and gene expression level in the vast majority of these samples ( S3A Fig ) . In addition , we analyzed the GRO-cap data from the human cell line K562 , allowing direct comparison of the result with our CAGE-seq–based result from the same cell line . The GRO-cap data show a significantly negative correlation between TSS diversity and gene expression level ( ρ = –0 . 16 for both the original and down-sampled data; S3B Fig ) , similar to that from the CAGE-seq data ( ρ = –0 . 19 for both the original and down-sampled data; the fourth cell line in Fig 1B ) . To minimize the influences of potential confounding factors in the above analyses , we compared between human paralogous genes of different expression levels because paralogous genes are similar in gene structure , DNA sequence , regulation , and function [39] . Consistent with the error hypothesis , Simpson and Shannon indices tend to be lower for the relatively highly expressed gene than the relatively lowly expressed one in a pair of paralogous genes , and this trend generally holds after CAGE-seq reads are down-sampled to 10 per gene ( S4 Fig ) . To examine whether the negative correlation between gene expression level and TSS diversity observed in humans also holds in other mammals , we analyzed the CAGE-seq–based TSS data of 11 mouse tissues ( S1 Table ) available at FANTOM5 phase1 . 3 [7] . The mouse results ( S5 Fig ) resembled the human results , indicating that the negative correlation is not a human-specific phenomenon . The above analyses suggest that many TSSs of a gene are suboptimal such that the overall TSS diversity declines with the gene expression level as a result of natural selection against suboptimal TSSs , but how many of the TSSs are suboptimal and which ones are suboptimal are unclear . To address these questions , we ranked all TSSs of a gene by their fractional usages . The fractional usage of a TSS is the number of CAGE-seq reads mapped to the TSS divided by the total number of reads mapped to all TSSs of the gene . For a given gene , the TSS with the highest fractional usage ( i . e . , ranked #1 ) is referred to as the major TSS , while all others are referred to as minor TSSs . Intuitively , the major TSS should be a preferred TSS . Because natural selection against transcriptional initiation error intensifies with the gene expression level , the fractional usage of each preferred TSS should increase while that of each unpreferred TSS should reduce as the expression level increases . We first tested this prediction in the human universal sample . Again , we considered only genes with at least 10 CAGE-seq reads to ensure a certain level of accuracy in TSS usage estimation . Indeed , the fractional usage of the major TSS in a gene increases with its expression level ( upper-left plot in Fig 2A ) . By contrast , each minor TSS examined shows the opposite trend , suggesting that none of them is preferred . For example , among all genes with at least two TSSs , the fractional usage of the second most frequently used TSS in a gene decreases with gene expression level ( lower-left plot in Fig 2A ) . A similar negative correlation is observed for the third most frequently used TSSs among genes with at least three TSSs ( upper-right plot in Fig 2A ) and for the fourth most frequently used TSSs among genes with at least four TSSs ( lower-right plot in Fig 2A ) . These trends remain unchanged when only TSSs within 500 bp around each annotated TSS of each gene are considered ( S6A Fig ) ; when only robust TSSs of each gene are considered ( S6B Fig ) ; when TSSs located within 1 , 5 , or 10 kb upstream of the most upstream TSS annotated for a gene are considered ( S6C–S6E Fig ) ; or when gene expression levels are measured by RNA-seq ( S6F Fig ) . We also observed a negative correlation when the analysis in Fig 2A is extended to the fifth , sixth , seventh , and eighth most frequently used TSSs among genes with at least five , six , seven , and eight TSSs , respectively . We further verified the results in Fig 2A by down-sampling the original data to 10 CAGE-seq reads per gene and reranking TSSs using the down-sampled data ( Fig 2B ) . Computer simulations confirmed that these trends are not statistical artifacts ( S2 Table ) . CAGE-seq data from other human tissues and cell lines ( Figs 2B and S6 ) as well as TSS-seq data from multiple tissues and cell lines ( S7 Fig ) show similar patterns . We also verified that the statistical trends in Fig 2 generally hold even when we limited the analysis to the common set of genes with at least four TSSs ( S8A Fig ) . Analysis of mouse tissues yielded similar results ( S8B and S8C Fig ) . These observations strongly suggest that , for most genes , in any tissue or cell line , only the major TSS is preferred while all other TSSs are unpreferred . We also validated the above human results using paralogous genes , which should be more comparable as mentioned . For the human universal sample , in 56% of the 1 , 962 pairs of paralogous genes analyzed , the major TSS is used more often in the relatively highly expressed paralog than in the relatively lowly expressed one , significantly more than the random expectation of 50% ( S9A Fig ) . By contrast , for the second , third , and fourth most frequently used TSSs , respectively , significantly smaller than 50% of gene pairs show higher fractional usages in the relatively highly expressed gene than in the relatively lowly expressed one ( S9A Fig ) . Other tissues and cell lines show similar patterns ( S9B Fig ) . These trends generally hold in down-sampled data ( S9B Fig ) . The above analyses of ATI in each tissue or cell line support our hypothesis that there is only one optimal TSS in each tissue or cell line and that all other TSSs are suboptimal . Now , we compare TSS usage among cell types . Because a tissue is usually composed of many different cell types , comparison among tissues is less precise than that among cell lines . We thus analyzed the CAGE-seq data of five human cell lines . Under the error hypothesis of ATI , any difference in TSS usage between cell types is due to the stochastic nature of transcriptional initiation error . Hence , the hypothesis predicts that this difference decreases as the expression level of the gene rises , because of the reduced transcriptional initiation errors of more highly expressed genes . By contrast , no such prediction is made a priori by the adaptive hypothesis because the difference in ATI between cell types would depend on the specific cell types and genes . To this end , we measured the distance in the fractional uses of TSSs of a gene between two cell types ( see Materials and methods ) and then correlated the distance with the gene's mean expression level in the two cell types compared . We found that in all 10 pairs of cell lines compared , the correlation is significantly negative ( Fig 3A ) , supporting the error hypothesis . To avoid the influence of different sequencing depths of different genes , we sampled 10 CAGE-seq reads per gene from each cell line for all genes that have at least 10 reads in each of the five cell lines and confirmed that all correlations remain negative despite some that become statistically insignificant ( Fig 3A ) . The error hypothesis further predicts that , when the expression level of a gene varies among cell types , the TSS diversity of the gene in a cell type decreases with the rise of its expression level in the cell type . To verify this prediction , we calculated for each gene the correlation between its expression level and TSS diversity across the five human cell lines . We focused on down-sampled data to guard against the influences of unequal sequencing depths of a gene across cell lines . Indeed , significantly more genes exhibit negative correlations than expected by chance regardless of whether we used the Simpson or Shannon index of TSS diversity ( Fig 3B ) . The error hypothesis further predicts a negative correlation across cell types between the expression level of a gene and the fractional use of each minor TSS . To verify this prediction , for each gene , we defined the major and minor TSSs in each cell line separately and then computed the across-cell–line rank correlation between the expression level of a gene in a cell line and the fractional use of a TSS of a certain rank in the cell line ( based on down-sampled data ) . Indeed , 53% of genes show a positive correlation for the TSSs of rank #1 , significantly more than the random expectation of 50% , whereas only 45–46% of genes show a positive correlation for each TSS of ranks #2 , #3 , and #4 , significantly lower than the random expectation ( Fig 3C ) . Although our evidence so far suggests that , for most genes , each cell type has only one preferred TSS , it remains possible that the optimal TSS varies among cell types such that the ATI variation among cell types is adaptive . To assess this possibility , we first repeated the analysis in Fig 3C by defining the global major and minor TSSs for each gene using the combined CAGE-seq reads from all five cell lines . Interestingly , the patterns observed are similar to those in Fig 3C . For example , 53% of genes show a positive correlation between the fractional usage of the TSS of global rank #1 in a cell line and the expression level of the gene in the cell line , significantly more than the random expectation ( S10 Fig ) . But for all global minor TSSs examined , only 46–47% of genes show a positive correlation , significantly less than the random expectation ( S10 Fig ) . Along with the observations in Fig 3C , these results suggest that only a small fraction of genes may have different optimal TSSs in different cell types . To estimate this fraction , we first counted the number of different major TSSs observed in each gene across the five cell lines ( N; Fig 3D ) because if all five cell lines share the same major TSS , it is most likely that they all share the same optimal TSS . We found that , of 7 , 793 genes examined , 6 , 400 ( or 82 . 1% ) have the same major site in all five cell lines ( i . e . , N = 1 ) . We also examined the maximum possible number of different major TSSs in the five cell lines ( M ) for each gene , which would be the smaller of 5 and the total number of TSSs observed in the five cell lines for the gene ( Fig 3D ) . When M ≥ 2 , 99 . 9% of genes show N < M , suggesting that cell-type–specific optimization of transcription start sites is far less than what ATI could potentially offer , consistent with the hypothesis that among-cell–type variations in ATI are largely nonadaptive . Even when different cell lines show different major TSSs , the optimal TSS could still be the same in these cell lines because the observation could be due to sampling error caused by limited sequencing depths . To examine this possibility , for each gene , we randomly shuffled its CAGE-seq reads among the five cell lines without altering the number of reads in each cell line and then used the shuffled data to count the number of different major TSSs in the five cell lines . We repeated this process 10 , 000 times and estimated the mean number of major TSSs for the gene in the shuffled data ( n ) ( Fig 3E ) and the fraction of times ( f ) when the number of major TSSs observed in the shuffled data equals to or exceeds that in the actual data . Here , f is an estimate of the one-tailed P-value in testing the null hypothesis that all cell lines share the same major TSS . We converted the P-values to Q-values to control for multiple testing and found that 343 genes have Q < 0 . 05 ( red dots in Fig 3E ) . Thus , approximately 4 . 4% of genes examined appear to have at least two different optimal TSSs in the five cell lines . The above analysis assumed that the major TSS of a gene in a cell line is the optimal TSS in the cell line , but this may not always be the case because using the optimal TSS more than any other TSS in every cell type for every gene could be difficult because of the limited power of transcription start regulation . Thus , it is possible that , even after the exclusion of sampling error , the observed major TSS is still not the optimal site for some genes in some cell types . This kind of high usage of suboptimal sites should have a fitness cost that rises with the gene expression level . Consequently , this phenomenon should have lower occurrences in more highly expressed genes as a result of natural selection . To this end , for each gene , we sampled 10 CAGE-seq reads from each cell line and re-estimated N and n as was done for the original data . We then divided all genes into two groups: those with N > n ( regardless of statistical significance of this inequality ) and the rest . The expression level is significantly lower for the former group of 1 , 582 genes than the latter group of 6 , 211 genes ( Fig 3F ) . This observation supports the idea that a sizable proportion of genes with N > n do not necessarily have different optimal sites in different cell types; rather , they cannot use the single optimal TSS in all cell types as the major TSS . In other words , the fraction of genes with evidence for adaptive differential ATI among cell types is lower than the above estimate of 4 . 4% . Notwithstanding , statistical power for detecting differential optimal TSSs among cell types rises with the sequencing depth , and the potential for the existence of at least two different optimal TSSs among cell types increases with the number of cell types examined . Hence , the above value of 4 . 4% is tentative , and this question should be revisited in the future with a larger and better data set . The error hypothesis predicts that the interspecific difference in fractional uses of various TSSs of a gene should decrease as the average expression level of the gene in the two species rises because highly expressed genes have reduced TSS diversity in each species . By contrast , no such trend is predicted a priori by the adaptive hypothesis . To distinguish between the two hypotheses , we measured the distance in the fractional uses of TSSs of one-to-one orthologous genes between human and mouse in the same tissue ( see Materials and methods ) . In each tissue , we correlated between this distance and the mean expression level of the gene in the two species across genes . We considered only genes with at least 10 CAGE-seq reads in the pair of human and mouse tissues concerned to ensure relatively accurate measures of TSS usage . In support of the error hypothesis , a significantly negative correlation is found in each of the six tissues examined ( Fig 4A ) , and these trends hold in down-sampled data ( Fig 4A ) . Because a gene may have different expression levels between human and mouse in a given tissue , the error hypothesis predicts that its TSS diversity should be lower in the species in which its expression level is higher . Indeed , using down-sampled data , we observed this trend in all six tissues ( Fig 4B ) . Furthermore , for most genes , the fractional use of the major TSS identified in a species increases with the expression level of the gene in the species ( Fig 4C ) . For each minor TSS , the opposite is true ( Fig 4C ) . That a nucleotide site is used as a TSS is because of the existence of a nearby core promoter , which is commonly composed of four well characterized cis-elements: the TATA box , the initiator ( INR ) , the TFIIB recognition element ( BRE ) , and the downstream promoter element ( DPE ) [2] . These cis-elements jointly determine the activity of the core promoter and hence the use of the corresponding TSS ( Fig 5A ) . Our finding that , for most human genes , the major TSS is likely optimal while all minor TSSs are likely suboptimal predicts that the cis-elements corresponding to the major TSS should be evolutionarily conserved , while those corresponding to minor TSSs should not be conserved and may even be selected against . To test this prediction , we merged all CAGE-seq reads of 15 independent cell lines and tissues of humans to determine the global major and minor TSSs of each gene . For comparison , we identified the cis-elements from the corresponding segment of the complementary strand of DNA; they are referred to as pseudoelements because they are not expected to be functional . Because some cis-elements of a core promoter may overlap with the coding , intron , or UTR of a transcript corresponding to another promoter , we considered only the cis-elements in regions that have never been annotated as coding sequence ( CDS ) , intron , or 3′ UTR in any transcript , as well as the pseudoelements in the corresponding regions . Because the DPE is located in the transcribed region , we could not examine its evolutionary constraint that is purely due to the DPE function . Therefore , we focused on the other three cis-elements . We used PhastCons scores across 46 mammals as a measure of evolutionary conservation [40] . We found that the PhastCons scores are substantially greater for the cis-elements of major TSSs than for the cis-elements of minor TSSs or pseudoelements ( Fig 5B–5D ) . As predicted , the PhastCons scores are similar between the cis-elements of minor TSSs and pseudoelements ( Fig 5B–5D ) . Specifically , INRs exhibit a slightly but significantly lower conservation in minor TSSs than in pseudoelements ( Fig 5B ) , probably reflecting natural selection against the existence of minor TSSs and/or attributable to weak purifying selection on pseudoelements associated with functional antisense transcription . The conservations of BREs and TATA boxes are slightly but significantly higher for minor TSSs than for pseudoelements ( Fig 5C and 5D ) . Because some minor TSSs are located downstream of the major TSS such that their conservations could be due to their positions in the 5′ UTR of the primary transcript instead of their promoter activities , we further analyzed only the minor TSSs upstream of the major TSS . Now INRs ( S11A Fig ) and TATA boxes ( S11C Fig ) both show substantially lower conservations in minor TSSs than pseudoelements , whereas BREs show slightly higher conservations than pseudoelements ( S11B Fig ) . These results generally support the prediction of the error hypothesis .
The prevalence of ATI of human and mouse genes has been known for years , and the prevailing view is that ATI is mostly adaptive , although evidence supporting this view exists in only a small number of genes [8 , 9 , 24 , 25] . In this study , we proposed that ATI is largely a manifestation of deleterious transcriptional initiation error . By analyzing multiple 5′-end sequencing data and cis-elements corresponding to TSSs , we provided strong evidence for the above error hypothesis in mammals . While most of our evidence was based on the tissues and cell lines analyzed ( S1 Table ) , the analysis of the evolutionary conservation of cis-elements was based on genome sequences , and hence our conclusion that ATI is largely nonadaptive is not restricted to the specific tissues and cell lines examined . Assuming that deleterious ATI has not been selectively purged in genes of the lowest expressions but has been completely removed in those of the highest expressions , we can treat ATI of lowly expressed genes as the total ATI ( T ) and ATI of highly expressed genes as nondeleterious ATI ( ND ) . Thus , the fraction of ATI that is deleterious is ( T-ND ) /T = 1-ND/T . We used the Simpson index of TSS diversity to measure the amount of ATI because both the number of TSSs and their relative usages are considered . In the human universal sample depicted in Fig 1A , the 20 most weakly expressed genes have a mean Simpson index of 0 . 530 ( T ) , while the 20 most highly expressed genes have a mean Simpson index of 0 . 069 ( ND ) . Hence , the fraction of deleterious ATI is 1 –ND/T = 1–0 . 069/0 . 53 = 87% . To examine the robustness of the above estimate , we used another method to estimate ND and T . We divided all genes in Fig 1A into four equal-interval bins according to their gene expression levels ( i . e . , all bins have the same log10RPM interval ) . Using the leftmost bin to estimate T and the rightmost bin to estimate ND yielded 1 –ND/T = 1–0 . 05/0 . 42 = 88% . Note that both of the above estimates of the fraction of deleterious ATI are conservative because T is an underestimate of all ATI before selection ( because some deleterious ATI could have been removed in lowly expressed genes ) and ND is an overestimate of all nondeleterious ATI ( because some deleterious ATI may not have been removed in highly expressed genes ) . The above finding is broadly consistent with our estimate that only approximately 4 . 4% of genes have evidence for different optimal TSSs in different cell types and explains why cis-elements of minor TSSs are generally evolutionarily unconserved . Some authors distinguished between two types of TSSs: sharp and broad TSSs [1 , 26 , 41] . Transcription initiates almost exclusively from one of a few neighboring nucleotides at a sharp TSS but from any of a large segment of nucleotides at a broad TSS . In the GRO-cap data of human K562 cell line [41] that we analyzed , 53 . 6% of sharp TSSs and 88 . 6% of broad TSSs are minor TSSs . Because we found that almost all minor TSSs are nonadaptive , the above numbers suggest that most sharp TSSs as well as the vast majority of broad TSSs are nonadaptive . Even though ATI arises primarily from molecular error and is thus stochastic , the magnitude of this error does not have to be entirely random . Our results provide unequivocal evidence for multiple forms of regulation of the magnitude . For instance , because of the common trans-environment for transcriptional initiation in a tissue/cell line , the negative correlation between the expression level of a gene and its TSS diversity in a tissue/cell line indicates that the magnitude of transcriptional initiation error is regulated by cis-factors such as various cis-elements analyzed . Because the cis-factors for a gene are constant across cell lines , the observation that the same gene has lower TSS diversities in cell lines where its expression levels are higher indicates that the magnitude of transcriptional initiation error is also regulated by trans-factors . Because not all genes have lower TSS diversities in one cell line than in another cell line , the magnitude of transcriptional initiation error must also be regulated by interactions between cis- and trans-factors . The various trends we observed ( Figs 1–4 ) strongly suggest that these regulations have been shaped by natural selection against transcriptional initiation error . The disadvantage of using a particular minor TSS of a gene when compared with the use of the major TSS not only depends on the gene expression level but may also vary by the position of the minor TSS relative to the major TSS . One might think that using minor TSSs upstream of the major TSS is less deleterious than using minor TSSs downstream of the major TSS because using an upstream TSS makes the transcript longer , so it is unlikely to cause a disruption of any regulatory sequence that is supposed to be in the transcript from the major TSS . This prediction , however , may not be correct , because extending the 5′ UTR could cause the appearance of uORFs , which often interfere with the translation of the functional ORF [42 , 43] . In the future , it will be interesting to study how fitness is altered by the use of minor TSSs at different positions . From the FANTOM TSS data , a substantial proportion of TSSs are found in RNA genes or intergenic regions , most of which are at the 5′ ends of genes encoding long noncoding RNAs [44] or microRNAs [45] . It will be important to verify whether our results on TSSs of protein-coding genes extend to RNA genes . Our results on ATI echo recent findings about a number of phenomena that increase transcriptome diversity , including alternative polyadenylation [31] , alternative splicing [46 , 47] , and several forms of RNA editing [48–50] . They have all been shown to be largely the results of molecular errors instead of adaptive regulatory mechanisms . Together , these findings reveal the astonishing imprecision of key molecular processes in the cell , contrasting the common view of an exquisitely perfected cellular life [51] . Interestingly , protein synthesis , the step of gene expression following transcription , also shows prevalent variations such as the non-AUG translational initiation [52] , use of uORFs [53] , and stop-codon readthrough [54] . It remains to be seen whether these variations that increase proteome diversity are also largely molecular errors . That ATI is generally nonadaptive does not preclude the existence of a minority of cases of ATI that are beneficial [16 , 20 , 25] . However , identifying these few beneficial ATI cases will be challenging . Our past studies of RNA editing [49 , 55] suggest that evolutionarily conserved TSSs may be beneficial , and future studies can prioritize those TSSs in the search for adaptive ATI .
The CAGE-seq data from human cell lines and tissues and mouse tissues were downloaded from FANTOM5 phase1 . 3 ( http://fantom . gsc . riken . jp/5/datafiles/phase1 . 3/ ) . We analyzed five human cell lines , as well as 11 tissues ( 10 specific tissues plus a mixed one ) from each of the two species ( S1 Table ) . FANTOM5 phase 1 . 3’s TSS annotation already clustered original TSSs into TSS peaks ( a peak is a short genomic region that could include more than one original TSS ) by the decomposition peak identification ( DPI ) algorithm upon the removal of technical noise [7] . For protein-coding genes , we found that only 0 . 0082% of adjacent peaks are <5 nucleotides apart , and only 0 . 14% of adjacent peaks are <10 nucleotides apart . The mean distance between adjacent peaks is 1 , 846 nucleotides , and the median is 135 nucleotides . The permissive and robust CAGE peaks identified by the original authors through rigorous filtering [7] were used in our analyses . Both types of peaks were obtained by removing technical noise followed by applying DPI [7] . As a result , both peaks are reliable , with the only difference being that the tag evidence thresholds are higher for robust peaks . In total , there were 1 , 048 , 124 permissive and 184 , 827 robust peaks identified from 975 human samples and 652 , 860 permissive and 116 , 227 robust peaks identified from 399 mouse samples , respectively . Because the robust set misses most TSSs ( e . g . , >80% in human ) identified by high-throughput sequencing , we primarily used the permissive TSSs unless noted . TSS was defined as the position within a CAGE peak region with the highest total number of CAGE tags across all samples in FANTOM5 phase1 . 3 . Human ( hg19 and hg38 ) and mouse ( mm9 ) genomic annotations were downloaded from Ensembl ( http://useast . ensembl . org/index . html ) . We focused our analysis on protein-coding genes , including 20 , 745 human genes and 22 , 745 mouse genes . A CAGE peak whose 5′ end is within 500 bp of the 5′ end of a gene is considered to belong to the gene if the peak is on the same strand as the gene [7] . Peaks mapped to more than one gene were removed . The total CAGE tags within a peak were considered as the tags of the representative TSS in that peak . The major TSS of a gene in a tissue or cell line is defined as the most frequently used TSS of the gene in the given tissue or cell line , while all other TSSs are considered minor . In addition , we analyzed two independent 5′-end RNA-seq data: GRO-cap and TSS-seq . GRO-cap is the global nuclear run-on sequencing method that enriches for 5′-7meGTP-capped RNAs [41] . It differs from CAGE-seq in that it can also detect the TSSs of unstable RNAs . We analyzed the GRO-cap data from the human cell line K562 [41] to allow a direct comparison with the CAGE-seq-based result on the same cell line . To compare CAGE-seq and GRO-cap results fairly , we assigned GRO-cap reads to TSSs annotated in FANTOM5 phase 1 . 3 before computing TSS diversity . TSS-seq is another independent method for identifying TSSs and measuring their fractional usages [56] . TSSs and reads from TSS-seq [56] for 26 human cell lines and 16 human tissues ( S1 Table ) were downloaded from DBTSS ( https://dbtss . hgc . jp/ ) . The analyses of CAGE-seq and TSS-seq reads were the same except that the reference genome hg38 was used in TSS-seq read mapping by the original authors [56] . Simpson and Shannon indices of TSS diversity for a gene are respectively defined by 1–∑i=1Spi2and –∑i=1Spilnpi where S is the number of TSSs in the gene and pi is the proportion of RNA molecules of the gene that use the ith TSS . To measure the difference in TSS usage for a gene between CAGE-seq samples A and B , we used a net correlational distance defined by dAB− 0 . 5dA− 0 . 5dB . Here , dAB equals 1 minus Pearson’s correlation coefficient between samples A and B in the fractional uses of all TSSs of the gene , dA is the same as dAB except that the two samples used are two CAGE-seq bootstrap samples derived from sample A , and dB is the same as dAB except that the two samples used are two CAGE-seq bootstrap samples derived from sample B [31] . Because CAGE-seq sequences only the 5′ end of an mRNA , the expression level of a gene is proportional to the number of CAGE-seq reads mapped to the gene [57] . The expression level of a gene is computed by the total number of CAGE-seq reads mapped to all TSSs of the gene multiplied by 106 and then divided by the total number of reads mapped to all TSSs of all genes in the sample . This is referred to as RPM . Although CAGE-seq data are comparable with the canonical RNA-seq in measuring gene expression , we also measured gene expression levels by canonical RNA-seq [38] for the same human universal sample included in the CAGE-seq data . To remove the potential influence of unequal sequencing depths of different genes in a sample on our analyses , we conducted down-sampling analyses . Briefly , we randomly picked the same number of CAGE-seq reads from all genes . Unless otherwise noted , we randomly picked 10 CAGE-seq reads per gene for all genes with at least 10 reads . Although using original and down-sampled data usually yielded qualitatively similar results , results from down-sampled data are more reliable because of equal surveys of ATI among genes . To rule out the possibility that the trends observed in Figs 1A , 1C and 2A are statistical artifacts , we performed computer simulations of the same number of random genes as the actual number of genes analyzed . We randomly generated each gene whose expression level ( reflected by the total number of reads mapped to all TSSs of the gene ) and relative TSS usages ( reflected by the numbers of reads mapped to various TSSs of the gene ) respectively follow the gene-expression–level distribution and relative TSS usage distribution of the real genes . Specifically , the total read number of a simulated gene was sampled from the collection of actual read numbers of all genes with replacement , while the numbers of reads mapped to the TSSs of the gene were multinomial random variables drawn according to the TSS usages of a randomly picked real gene . We then analyzed the read data from the simulated genes as if they were the actual data . Human paralogous genes as well as human–mouse one-to-one orthologous genes were downloaded from Ensembl ( release 89; May 2017 ) . We obtained 3 , 678 human gene families , including 51 , 657 pairs of human paralogous protein-coding genes . When comparing between paralogs , we randomly selected from each gene family only one paralogous pair that has at least a 2-fold expression difference to allow a sufficient statistical power . Because of the expression level variation among tissues or cell lines , the included paralogous pairs in our analysis vary by tissue or cell line . The number of orthologous genes between human and mouse is 16 , 805 . The one-to-one orthologous CAGE peaks ( i . e . , orthologous TSSs ) between human and mouse were obtained from FANTOM5 phase 1 . 3 ( http://fantom . gsc . riken . jp/5/datafiles/phase1 . 3/extra/CAGE_peaks_Cross_species_projection/ ) . When comparing TSSs between human and mouse orthologs , we set zero usage in mouse for TSSs that are found only in human and vice versa . The structure of a human core promoter with positions and consensus sequences of cis-elements [58] analyzed in this study is shown in Fig 5A . Following a previous study [58] , we used the consensus sequences to search three cis-elements around TSS ( +1 ) ; BRE was searched in [–50 , –1] , TATA box was searched in [–40 , –1] , and INR was searched in [–6 , +7] . Only 12% of TSSs were found to have at least one of the three motifs because most minor TSSs do not have any detected motif . The fraction of TSSs with at least one motif tends to decrease with the TSS usage rank . All detected copies of a motif associated with a TSS were considered . To examine the evolutionary conservation of the cis-elements , we downloaded from UCSC ( http://hgdownload . cse . ucsc . edu/goldenpath/hg19/phastCons46way/placentalMammals/ ) PhastCons scores computed from genome alignments of 46 placental mammals , including the human ( hg19 ) . | Multiple surveys of transcriptional initiation showed that mammalian genes typically have multiple transcription start sites such that transcription is initiated from any one of these sites . Many researchers believe that this phenomenon is adaptive because it allows production of multiple transcripts , from the same gene , that potentially vary in function or post-transcriptional regulation . Nevertheless , it is also possible that each gene has only one optimal transcription start site and that alternative transcriptional initiation arises primarily from molecular errors that are slightly deleterious . This error hypothesis makes a series of predictions about the amount of transcription start site diversity per gene , relative uses of the various start sites of a gene , among-tissue and across-species differences in start site usage , and the evolutionary conservation of cis-regulatory elements of various start sites , all of which are verified in our analyses of genome-wide transcription start site data from the human and mouse . These findings strongly suggest that alternative transcriptional initiation largely reflects molecular errors instead of molecular adaptations and require a rethink of the precision and regulation of transcription . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"sequencing",
"techniques",
"ecology",
"and",
"environmental",
"sciences",
"gene",
"regulation",
"dna",
"transcription",
"gene",
"sequencing",
"molecular",
"biology",
"techniques",
"rna",
"sequencing",
"simpson",
"index",
"research",
"and",
"analysis",
"methods",
"ecolo... | 2019 | Evidence that alternative transcriptional initiation is largely nonadaptive |
The ascomycete fungus Tolypocladium inflatum , a pathogen of beetle larvae , is best known as the producer of the immunosuppressant drug cyclosporin . The draft genome of T . inflatum strain NRRL 8044 ( ATCC 34921 ) , the isolate from which cyclosporin was first isolated , is presented along with comparative analyses of the biosynthesis of cyclosporin and other secondary metabolites in T . inflatum and related taxa . Phylogenomic analyses reveal previously undetected and complex patterns of homology between the nonribosomal peptide synthetase ( NRPS ) that encodes for cyclosporin synthetase ( simA ) and those of other secondary metabolites with activities against insects ( e . g . , beauvericin , destruxins , etc . ) , and demonstrate the roles of module duplication and gene fusion in diversification of NRPSs . The secondary metabolite gene cluster responsible for cyclosporin biosynthesis is described . In addition to genes necessary for cyclosporin biosynthesis , it harbors a gene for a cyclophilin , which is a member of a family of immunophilins known to bind cyclosporin . Comparative analyses support a lineage specific origin of the cyclosporin gene cluster rather than horizontal gene transfer from bacteria or other fungi . RNA-Seq transcriptome analyses in a cyclosporin-inducing medium delineate the boundaries of the cyclosporin cluster and reveal high levels of expression of the gene cluster cyclophilin . In medium containing insect hemolymph , weaker but significant upregulation of several genes within the cyclosporin cluster , including the highly expressed cyclophilin gene , was observed . T . inflatum also represents the first reference draft genome of Ophiocordycipitaceae , a third family of insect pathogenic fungi within the fungal order Hypocreales , and supports parallel and qualitatively distinct radiations of insect pathogens . The T . inflatum genome provides additional insight into the evolution and biosynthesis of cyclosporin and lays a foundation for further investigations of the role of secondary metabolite gene clusters and their metabolites in fungal biology .
Fungi are prolific producers of secondary metabolites and an important source of novel and commercially important pharmaceuticals , mycoinsecticides , and antibiotics . Cyclosporin A ( CsA; CAS ID: 59865-13-3 ) , the well-known immunosuppressant drug which revolutionized organ transplantation from an experimental to a relatively routine lifesaving procedure [1] , was first discovered in the insect pathogenic and ubiquitous soil fungus , Tolypocladium inflatum . CsA targets and binds with high affinity human cyclophilin A ( hCypA , peptidylprolyl isomerase A , EC: 5 . 2 . 1 . 8 ) , a conserved immunophilin found across eukaryotes [2] , [3] . The CsA-hCypA complex suppresses the vertebrate immune system by binding to and inhibiting calcineurin , a conserved calcium-calmodulin activated serine/threonine-specific protein phosphatase ( EC: 3 . 1 . 3 . 16 ) [4] , [5] . Inhibition of calcineurin blocks activity of NF-AT ( nuclear factor of activated T-cells ) , a regulator of transcription of Interleukin 2 in T-lymphocytes [6] . CsA also impairs the immune response in insects and shows both antifungal [7] and antiviral activity [8] . CsA and related cyclosporins ( B-Z and isoforms ) form a family of cyclic undecapeptides produced by nonribosomal peptide synthetases ( NRPSs ) , a class of large multimodular enzymes that produce peptides via a nonribosomal mechanism [9] . While the 45 . 8 kb locus encoding the NRPS synthetase ( simA ) responsible for biosynthesis of cyclosporin was cloned in 1994 [10] , the complete biosynthetic cluster remains uncharacterized . T . inflatum belongs to the fungal order Hypocreales , containing fungi known to produce a high diversity of bioactive secondary metabolites [11] . In addition to cyclosporins , T . inflatum synthesizes a number of other products via both NRPSs and polyketide synthetases ( PKSs ) , another class of multimodular enzyme involved in secondary metabolite production in bacteria and fungi [12] . Other hypocrealean fungi are known to produce NRPS or PKS products with activity against insects , such as destruxins ( Metarhizium robertsii ) [13] , efrapeptins ( Tolypocladium spp . ) [14] , and ergot alkaloids ( clavicipitalean endophytes of grasses including Claviceps and Epichloë spp . ) [15] . Many of these compounds also have pharmaceutical applications and/or roles in antibiosis , pathogenesis , and competitive interactions between organisms [13] , [14] . The genome sequence of T . inflatum thus provides an opportunity to characterize the secondary metabolite arsenal of an insect-pathogenic fungus with potential for both elucidation of the biosynthetic cluster of the immunosuppressant drug cyclosporin and discovery of novel gene clusters and metabolites with applications in medicine and agriculture . Hypocrealean fungi display considerable flexibility of lifestyles . They include plant-pathogens , plant-saprobes , plant-endophytes , mycoparasites , and pathogens of insects , spiders , rotifers , and nematodes . Transitions between different lifestyles have occurred multiple times in the evolutionary history of the order [16] , [17] . T . inflatum is a pathogen of beetle larvae [18] , but is also able to live saprotrophically in soil during the asexual phase of its lifecycle ( Figure 1 ) . It is one of the few insect pathogenic ( entomopathogenic ) fungi sequenced to date , although Hypocreales contains three families ( Clavicipitaceae , Cordycipitaceae , and Ophiocordycipitaceae ) that are particularly rich in entomopathogenic species . The genomes of M . robertsii and M . acridum ( Clavicipitaceae ) have provided insights into expansions of gene families , especially those for secreted proteins , with roles in insect pathogenesis [19] . Other studies have shown changes in the profiles of carbohydrate active enzymes ( CAZymes ) , cytochrome P450s , and proteases in insect pathogens when compared to closely related plant pathogens [20] . The genome of Cordyceps militaris ( Cordycipitaceae ) , a common pathogen of moth pupae used in traditional Chinese Medicine , revealed aspects of the mating systems of entomopathogenic fungi [21] . These taxa belong to three separate families of Hypocreales that represent parallel diversifications of entomopathogenic fungi [16] . Here we present the results from whole genome sequencing and RNA-Seq analyses of T . inflatum ( Ophiocordycipitaceae ) , which represents the first draft genome from the third entomopathogenic family within Hypocreales . Through phylogenomic and comparative genomic analyses we demonstrate that the NRPS responsible for cyclosporin biosynthesis ( simA ) exhibits complex patterns of homology with other NRPSs and that the cyclosporin gene cluster is unique to Tolypocladium with no evidence for horizontal transfer of a complete cluster from other fungi or bacteria . RNA-Seq analyses in a cyclosporin-inducing medium clearly delineate a secondary metabolite cluster responsible for cyclosporin biosynthesis . RNA-Seq analyses in media simulating insect pathogenesis indicate that several genes within the cluster , including the homolog of the cyclosporin binding protein cyclophilin , are also upregulated in response to insect hemolymph , supporting a role for both cyclosporin and the cyclophilin gene in pathogenesis of insects .
A karyotype study of the sequenced strain T . inflatum NRRL 8044 ( ATCC 34921 ) indicated that T . inflatum has 6 chromosomes ranging in size from 3 . 8 to 6 . 6 Mb and a mini supernumerary chromosome of 1 Mb with a total genome size of approximately 30 . 45 Mb [22] . The total size of the assembly ( 30 . 348 Mb ) closely matched this estimate and contained 194 contigs in 101 scaffolds with an N50 of 1 . 5 Mb and an Nmax of 3 . 56 Mb . The MAKER 2 . 0 annotation pipeline [23] predicted 9 , 998 protein coding genes ( loci tagged as TINF ) with greater than 90% having support from either protein ( Fusarium graminearum , Nectria haematococca , Trichoderma reesei , Tr . virens , and M . robertsii ) or EST data ( C . militaris , Beauveria bassiana , M . robertsii , Tr . reesei , and assembled T . inflatum RNA-Seq reads ) . Analysis using the Core Eukaryotic Genes Mapping Approach ( CEGMA ) pipeline [24] estimated that the annotations represent >98% of coding regions based on completeness of a conserved set of eukaryotic proteins . The average gene length ( 1 . 67 kb ) , exon length ( 570 bp ) and intron length ( 77 . 5 bp ) were similar to estimates from other Ascomycota ( Table 1 ) [25] . However , T . inflatum has a higher average GC content ( 58% ) and a more compact genome with higher gene density ( 329 genes/Mb ) than closely related filamentous ascomycetes ( Table 1 ) . Only the MAT1-2 mating type locus was detected in the sequenced strain , indicating that T . inflatum is likely heterothallic ( Figure S1 ) . The estimated proportion of repeat sequence ( 1 . 24% ) , which agrees well with previous experimental estimates ( 1% ) [26] , is relatively low compared to other filamentous ascomycetes ( Table 1 ) . In total , T . inflatum contained a slightly larger number of retrotransposons compared to DNA transposons ( Table S1 ) . Retrotransposons were dominated by two classes , LINE elements and the Gypsy family of LTRs , while DNA transposons were mostly comprised of the hAT family ( Table S1 ) . Several novel repeat elements , including the CPA ( cyclosporin production associated ) element [26] and the first characterized fungal hAT transposon ( Restless ) [27] , were previously identified in the sequenced strain ( NRRL 8044 ) . The CPA element , which shows greatest similarity to a RecQ DNA helicase in M . robertsii [28] , was named based on the observation that multiple copies were found only in cyclosporin producing strains of T . inflatum while only a single copy was present in other strains of T . inflatum and related Tolypocladium and Beauveria species . We identified 12 copies of the CPA element in T . inflatum NRRL 8044 that were found dispersed across eleven scaffolds , but none of them associated with the scaffold containing the cyclosporin biosynthetic cluster . A single partial copy was also found in C . militaris and Tr . virens and multiple copies were present in Tr . reesei , Metarhizium spp . , and especially N . haematococca , in which the transposon has undergone expansion to 27 copies ( Table S1 ) . We conclude that presence and expansion of CPA elements is not unique to T . inflatum and that it is not associated with the evolution or production of cyclosporin . Similarly , copies of Restless were also found in other hypocrealean taxa and were particularly expanded in F . oxysporum , which harbored over double ( 55 ) the number of elements in T . inflatum ( 26 ) ( Table S1 ) . Restless generates partial deletion copies of the transposon and several deletion variants ( ΔRst 1-6 ) have been found in T . inflatum , Neurospora crassa , and Penicillium chrysogenum [29] . With the exception of Tr . atroviride , all hypocrealean taxa contained either an intact or a deletion variant of Restless , indicating the transposon was present in the ancestor of Hypocreales . However , of the genomes analyzed , deletion variants were most abundant and diverse in T . inflatum ( Table S1 ) , suggesting it has been particularly active in T . inflatum . T . inflatum is a member of the class Sordariomycetes and is related to the widely studied filamentous ascomycete N . crassa , which together with the wilt pathogens Verticillium dahliae and Verticillium albo-atrum , served as an outgroup to the order Hypocreales in our phylogenomic analyses ( Figure 2A , node 1 ) . Orthologous clustering of proteins using MCL [30] identified a total of 36 , 532 orthologous clusters of proteins across the 14 taxa analyzed ( N . crassa , V . dahliae , V . albo-atrum , N . haematococca , F . graminearum , F . oxysporum , F . verticillioides , Tr . virens , Tr . atroviride , Tr . reesei , C . militaris , M . robertsii , M . acridum , and T . inflatum ) . Using the phylogenomic pipeline Hal [31] , these 36 , 532 clusters were filtered to identify 2 , 769 clusters containing only single-copy orthologous proteins . A concatenated alignment was built from this subset of clusters and used to construct a maximum likelihood phylogeny . The inferred phylogeny recovered a topology consistent with the four major families of Hypocreales ( Figure 2A ) . The earliest diverging group , Nectriaceae , comprises primarily plant pathogenic species including the wheat head blight fungus F . graminearum and related species F . oxysporum , F . verticillioides , and N . haematococca ( Figure 2A , node 3 , green ) . Sequenced representatives of Hypocreaceae include members of the genus Trichoderma ( Tr . atroviride , Tr . reesei , Tr . virens ) . Although the ability of Trichoderma spp . to grow on and digest plant-based compounds is well-documented , recent comparative genomic studies support an evolutionary history characterized by mycoparasitism [32] ( Figure 2A , node 6 , blue ) . Hypocreaceae forms a sister group ( node 5 ) to the primarily insect pathogenic Cordycipitaceae , which includes the moth pathogen and traditional Chinese medicinal fungus C . militaris . The remaining two families , Ophiocordycipitaceae , of which T . inflatum is the first sequenced representative , and Clavicipitaceae ( Figure 2A , node 7 , red ) , which includes the two insect pathogenic biocontrol species M . robertsii and M . acridum , comprise , with the exception of the clavicipitaceous endophytes , primarily insect pathogens . This topology is consistent with standard multigene ( e . g . , five to seven loci ) phylogenetic analyses that have sampled an order of magnitude more species of Hypocreales and ancestral character state reconstructions from previously published multigene datasets [16] , [17] , which support a major transition within the order from plant hosts/substrates in early diverging lineages ( Nectriaceae ) to primarily insect ( Clavicipitaceae , Cordycipitaceae , Ophiocordycipitaceae ) or fungal ( Hypocreaceae ) hosts ( Figure 2A , node 2 ) . However , the placement of Cordycipitaceae has been controversial . The removal of fast evolving sites [33] in these genome-scale analyses provides stronger bootstrap support for the placement of C . militaris ( Cordycipitaceae ) as monophyletic with Trichoderma ( Hypocreaceae ) and not with the other insect pathogens of Metarhizium ( Clavicipitaceae ) and Tolypocladium ( Ophiocordycipitaceae ) . These results provide additional support for polyphyletic origins and parallel diversifications of insect pathogenic fungi in three separate families in Hypocreales [34] . Out of the total 36 , 532 orthologous clusters , those shared by one or more descendants of each node were mapped to the phylogeny to produce a phylogenetic profile of orthologous clusters ( Figure 2A ) . A total of 7 , 964 clusters containing 112 , 539 proteins from both within Hypocreales and from outgroup taxa ( N . crassa , V . dahliae and V . albo atrum ) , while 1 , 746 clusters containing 13 , 658 proteins mapped uniquely to the node representing the origin of Hypocreales ( Figure 2A , node 1 ) . Within Hypocreales , plant associated species in the Nectriaceae had the largest number of unique clusters although these genomes also contained a larger number of protein coding genes per genome ( Table 1 ) . T . inflatum had 1 , 675 species-unique clusters containing 1 , 709 proteins ( out of a total of 9998 ) . T . inflatum shared a larger number of clusters with fungal pathogens in Hypocreaceae ( 1750 ) than with other insect pathogens in Clavicipitaceae and Cordycipitaceae ( 190 ) or with plant pathogens in Nectriaceae ( 109 ) ( Figure 2B ) . All hypocrealean taxa , including T . inflatum , shared a similar profile of Gene Ontology ( GO ) Slim categories ( Figures 3A , S2 ) . However , GO Slim profiles of genes found in orthologous clusters unique to each of the insect pathogens C . militaris ( Cordycipitaceae ) , M . robertsii and M . acridum ( Clavicipitaceae ) , and T . inflatum ( Ophiocordycipitaceae ) showed lineage specific differences ( Figure 3B ) . Metarhizium spp . ( Clavicipitaceae ) had a larger proportion of species-unique genes associated with GO molecular functions of protein binding ( 22–26% vs 13–16% ) , oxidoreductase activity ( 20% vs 8–11% ) , and peptidase activity ( 7–9% vs 2–3% ) relative to either T . inflatum or C . militaris ( Figures 3B , S2B ) . These results are consistent with expansions of both proteases ( peptidase activity ) and P450s ( oxidative activity ) in Metarhizium spp . , Tr . virens , and all plant pathogenic species in Nectriaceae , but not in the other insect pathogens T . inflatum or C . militaris ( Table S2 ) . In fact , M . acridum and particularly M . robertsii , which is known to live in association with the plant rhizosphere [35] , showed overall profiles of CAZymes , proteases , and P450s more similar to plant pathogens than to the other insect pathogens ( Table S2 ) . T . inflatum contained a larger proportion of species-unique genes associated with the GO molecular function of transporter activity ( 7% vs 2% ) than other insect pathogens , while C . militaris had a larger proportion of species-unique genes associated with transferase activity ( 39% vs 13–22% ) ( Figures 3B , S2B ) . For GO biological process categories , T . inflatum also contained a larger proportion of species-unique genes related to transport ( 11% vs 3–4% ) but a smaller proportion of genes involved in DNA-dependent transcription ( 7% vs 18–23% ) relative to other insect pathogens ( Figure S2D ) . A large proportion of species-unique genes in all insect pathogens ( 31–74% ) were associated with membranes , particularly endomembrane systems ( Figure S2F ) , consistent with the importance of secreted proteins in these fungi [19] . These differences in gene content and ontology between the three insect pathogenic lineages corroborate phylogenetic evidence for parallel and qualitatively distinct evolutionary radiations of insect pathogens in Hypocreales that reflect adaptations to distinct ecologies . While T . inflatum is best known as the original source of CsA [36] , it is also known to produce other bioactive secondary metabolites including insecticidal compounds such as efrapeptins [37] and tolypin [38] , diketopiperazines [39] , and the carboxysterol antibiotic ergokonin-C [39] . In addition to well-known core enzymes involved in producing fungal secondary metabolites ( NRPSs , PKSs , prenyltransferases , and terpene cyclases ( TC ) ) , a large number of modifying enzymes such as racemases , methyltransferases , acetyl transferases , prenyltransferases , cytochrome P450 monooxygenases ( P450s ) and oxidoreductases are often required for synthesis of the final bioactive products . In fungi , these are often found clustered with the core enzymes to form secondary metabolite biosynthetic gene clusters [40] , which some hypothesize may facilitate or be driven by horizontal transfer [41] . Others suggest clustering may minimize the number of coordinated interactions between regulatory elements [42] . We identified a total of 14 NRPSs , 20 PKSs , 4 Hybrid PKS/NRPSs , 11 putative NRPS-like enzymes , 5 putative PKS-like enzymes , and one dimethylallyl-tryptophan synthase ( DMATS ) in the T . inflatum genome ( Table S3 ) , indicating that T . inflatum has a large potential for secondary metabolite production . The majority of these core enzymes fell within one of the 36 secondary metabolite clusters identified by SMURF [43] or an additional 2 clusters ( 38 total ) identified by antiSMASH [44] . In addition to the NPRS in the cyclosporin cluster , phylogenomic analyses of T . inflatum NRPS adenylation domains ( Figure S3 , Table S4 ) identified homologs of a number of functionally characterized NRPSs from fungi , including three peptaibol synthetases ( TINF05969 , TINF07827 , TINF07876 ) , both intracellular ( ChNPS2 - TINF08996 ) and extracellular ( ChNPS6 - TINF01764 and TINF06175 ) siderophore synthetases , and a homolog of conserved NRPS-like proteins involved in morphological development ( ChNPS10 TINF09755 ) ( Figure S3 , Table S4 ) . We also identified an NRPS ( TINF02556 ) whose A-domains group with those of the ergot alkaloid synthetases ( cpps1-4 ) from the grass endophyte Claviceps purpurea ( Figures S3 , S4A ) . The ergot alkaloid cluster in C . purpurea contains two trimodular NRPSs ( cpps1 , cpps4 ) , two monomodular NRPSs ( cpps2 , cpps3 ) , and a DMATS [45] , [46] . The alkaloid secondary metabolite clusters recently discovered in the insect pathogens M . robertsii and M . acridum [19] contain homologs of the two monomodular NRPSs ( cpps2 , cpps3 ) , the DMAT synthase , and the majority of modifying enzymes found on the 5′ end of the C . purpurea cluster . In contrast , the antiSMASH predicted cluster in T . inflatum lacks homologs of these monomodular NRPSs and other ergot alkaloid biosynthetic genes but contains a four modular NRPS ( TINF02556 ) with A-domains that show closest similarity to the trimodular NRPSs ( cpps1 and cpps4 ) from C . purpurea , as well as other genes involved in secondary metabolism ( Figure S4B ) . We also identified an additional cluster in Metarhizium spp . which contains a 7 modular NRPS ( MAA_06559 , MAC_08899 ) with A-domains that also show homology to cpps1 and cpps4 , but which also lacks other genes from the ergot alkaloid cluster ( Figure S4B ) . T . inflatum does contain one DMAT synthase . However , it is located on a different scaffold in a distinct secondary metabolite cluster predicted by antiSMASH to be involved in terpene biosynthesis ( Figure S4C ) . While further chemical data is needed , we conclude it is unlikely that the T . inflatum cluster produces an ergot alkaloid compound similar to those produced by C . purpurea . Cyclosporins are cyclic depsipeptides belonging to a class of cyclic undecapeptides [10] , [47] , and T . inflatum is known to produce 25 different analogs of cyclosporin ( cyclosporins A-I and K-Z ) [48] , [49] . CsA is composed of 11 substrate molecules produced by an NRPS encoded by the single 45 . 8 kb simA locus , and like the products of many NRPSs , CsA contains several non-proteinogenic substrates including 2-aminobutyric acid , D-alanine , and ( 4R ) -4-[ ( E ) -2-butenyl]-4-methyl-threonine ( Bmt ) [10] . The simA gene displays a modular structure typical of NRPSs , consisting of 11 modules comprised of three core catalytic domains: adenylation ( A ) , which binds and activates the substrate , thiolation ( T ) , which attaches substrates to the NRPS , and condensation ( C ) , which forms a peptide bond between adjacent substrates . Each module activates one of the eleven substrates and additional methylation ( M ) domains are present which methylate substrates 2 ( Leu ) , 3 , ( Leu ) , 4 ( Val ) , 5 ( Bmt ) , 8 ( Leu ) and 10 ( Leu ) . Various other fungi within Hypocreales ( Acremonium , Chaunopycnis , Fusarium , Isaria , Nectria , Neocosmospora , Trichoderma and Verticillium ) have been reported to synthesize a common profile of cyclosporins A-D and E-F [49] , [50] . Only a few fungi outside of Hypocreales ( Aspergillus terreus ) [51] have been reported to produce cyclosporin A , while others ( Leptostroma , Cylindrotrichum , Stachybotrys ) produce a single and often novel cyclosporin-related compound [52] , [53] . A previous phylogenomic study of NRPSs from 37 complete fungal genomes found that simA grouped sister to a clade of bacterial NRPSs but found no complete ( 11 modular ) homolog of simA in either bacteria or other fungi [54] . Similarly , we employed BLAST searches of the NCBI database and HMMER searches across closely related hypocrealean fungi and found no complete homologs of simA ( Figures 4 , S3 , S5 ) . The phylogenetic tree constructed from A-domain sequences of the top 50 BLAST hits to simA from the NCBI nr database ( Figure 4 , S5 ) showed that no bacterial sequences group within the simA clade . This suggests that simA likely evolved by duplication of modules within fungi rather than through recent horizontal transfer from bacteria . Individual adenylation domains from several other fungal NRPSs group within the simA clade with 100% bootstrap support ( Figures 4B , S5 ) . These include NRPSs synthesizing several other known fungal cyclic depsipeptides such as enniatin synthetase ( esyn1 ) ( Fusarium equiseti ) [55] , beauvericin ( bbBeas ) and bassianolide ( bbBsls ) synthetases ( Beauveria bassiana ) , the NRPS responsible for biosynthesis of the antifungal compound aureobasidin A ( aba1 ) ( Aureobasidium pullulans ) , and two modules of destruxin synthetase ( dtxS1 ) ( M . robertsii ) ( Figures 4B , S5 ) . These compounds share similar functions , having either anti-insect ( beauvericin , bassianolide , destruxin , and cyclosporin A ) [13] , [56]–[58] and/or antifungal properties ( aureobasidin A [59] and cyclosporin A [60] ) . Fungal NRPSs containing adenylation ( A ) domains found in the simA clade , share a complex history of evolution through duplication and fusion of modules [61] . For example , the A-domain of module 1 of the NRPSs synthesizing enniatin ( esyn1 ) , beauvericin ( bbBeas ) , and bassianolide ( bbBsls ) synthetases all code for an identical non-amino acid substrate , D-2-hydroxyisovaleric acid ( Hiv ) and group together phylogenetically in a clade distinct from the simA clade ( Figures 4A , C , 5 , S5 [enniatin module 1 clade] ) . In contrast , the A-domain from module 2 of enniatin and the C-terminal modules of all of these genes fall within the simA clade ( Figures 4A , B , 5 , S5 [enniatin module 2 clade] ) , suggesting fusion of modules . Other fungal NRPSs from Magnaporthe grisea ( XP_369222 . 2 ) and Aspergillus species ( XP_001394009 . 2 , XP 682495 . 1 , XP 001267445 . 1 ) display a similar pattern . Others , such as NPS1 and NPS3 from Cochliobolus heterostrophus and destruxin synthetase from M . robertsii , contain two A-domains grouping in the simA clade , but remaining A-domains group with the Epichloë festucae NRPS perA , outside both the simA and the enniatin module 1 clades ( Figures 5 , S3 ) . The C-terminal modules of these same NRPSs , containing A-domains in the simA clade , show evidence of duplication of NRPS modules followed by divergence of A-domain substrate specificities . Enniatins , for example , are known to vary in the N-Me-amino acid incorporated by the second A-domain [62] . The C-terminal modules ( 2–8 ) of aureobasidin A synthetase ( aba1 ) provide the clearest example of extensive module duplication and divergence within a single species as all A-domains group as a single monophyletic group with 100% bootstrap support and many share over 95% sequence similarity ( Figure 4B ) [63] . While the sequence of duplications within simA is more complex , A-domains of simA that group together with greater than 50% bootstrap support encode for identical substrate amino acid specificities ( [A2 , A3 , A8 , A10; bs = 78% ) for Leu and ( A4 and A9; bs = 98% ) for Val] ) , suggesting these domains represent more recent duplications that have not diverged in specificity encoding regions ( Figures 4B , 5 ) . The average pairwise dN/dS ratio across all A-domains was low ( ώ = 0 . 182 ) , indicating that most sites within A domains are under purifying selection . However , the branch-site REL method of the HYPHY package [64] detected significant evidence of episodic positive selection ( p<0 . 0001 ) on the branch separating the clade coding for Leu ( A2 , A3 , A8 , A10 ) from other A-domains . These results are consistent with a process in which duplication followed by lineage specific changes at a few amino-acid positions contributed to the evolution of the species-unique cyclosporin metabolite . The two computational methods used for defining secondary metabolite clusters , SMURF and antiSMASH , identified slightly different boundaries to the simA cluster ( Figures 6 , 7 ) . The secondary metabolite gene cluster surrounding simA predicted by antiSMASH contains 22 genes and spans over 112 kb , while SMURF delineated a slightly smaller cluster of 10 genes spanning approximately 93 kb ( Figures 6 , 7 ) . In order to utilize transcriptional data to define the cyclosporin metabolite cluster , an RNA-Seq time course experiment in a cyclosporin-inducing medium ( SM medium ) [65] was conducted . Three biological replicates each were grown in the inducing ( SM ) and a rich control medium , Sabouraud Dextrose Broth ( SDB ) , and were sampled at two-day intervals for a total of six time points ( days 2 , 4 , 6 , 8 , 10 , and 12 ) . Total RNA isolated from these samples was prepared for RNA-Seq and remaining mycelia and culture filtrate was harvested and analyzed by LS-MS in order to correlate gene expression patterns with production of cyclosporin metabolites . Expression profiling under cyclosporin-inducing conditions ( SM medium ) clearly identified a cluster of genes surrounding simA that were significantly ( q-value<0 . 01 ) upregulated ( Table S5 ) during production of cyclosporins as detected by an HPLC peak ( Figures 6 , S6 ) that included analogs with nominal molecular masses of 1188 , 1202 ( Cyclosporin A ) , 1216 , and 1218 Da ( data not shown ) . Upregulation of genes within the cluster became highly significant at time point 3 ( day 6 ) which corresponded with the first detection of large quantities of cyclosporin in the culture filtrate by LC-MS and were consistently and strongly upregulated at time points 4 , 5 , and 6 that also showed an HPLC peak for CsA ( Figure 6 , S6 , Table S5 ) . The 5′ boundary of this cluster corresponded to the SMURF computational prediction beginning at simA ( TINF00159 ) , while the 3′ edge of this cluster corresponded to the 3′ boundary of the antiSMASH cluster prediction ( TINF07874 ) ( Figures 6 , 7 , Table S5 ) . LC-MS profiles for the cyclosporin-containing fraction were generated consistently from culture filtrate extracts obtained for each time point . Under the HPLC protocol used , analogs of cyclosporins eluted predominantly between 32–40 min , with the peak maximum for cyclosporin A ( the major product , molecular mass 1202 Da ) at 38 min . The overall production of cyclosporins peaked at time point 4 ( day 8 ) in SM medium ( Figure 6 , Figure S6 ) . Beginning at time point 4 , an increase in the number of overlapping peaks for closely-eluting analogs of cyclosporin A is consistent with depletion of specific amino acids in the culture media leading to relaxed substrate specificity of the cyclosporin NRPS A-domains and production of distinct cyclosporin analogs ( Figure 6 , Figure S6 ) . However , the existence of more than one isoform with the same molecular mass prevents the rigorous assignment of these metabolites ( molecular masses 1188 , 1202 , 1216 , and 1218 Da ) by mass spectrometry alone . This combination of computational and experimental approaches provided a more robust method for characterization of secondary metabolite clusters and demonstrates the utility of transcriptional data in confirming cluster boundaries . Although NRPSs such as simA and esyn1 contain A-domains with shared ancestry ( Figure 4 , S3 , S5 ) , the metabolite clusters containing these core metabolite genes do not share other homologous genes ( Figure 7 ) . Components of secondary metabolite clusters other than the core backbone enzymes function in synthesis of precursors , mediation of intermediate steps , transport and delivery , and modifications of the final metabolite [66] . While further functional studies ( e . g . gene knockouts ) are needed , the simA cluster contains genes which likely function in both synthesizing substrates for the NRPS and modification or activation of the cyclosporin product . The unusual non-proteinogenic amino acid substrate D-alanine must be supplied by an independent alanine racemase [67] as simA itself does not contain racemase activity . Similarly , it was shown that one of the unusual amino acid substrates of cyclosporin , ( 4R ) -4-[ ( E ) -2-butenyl]-4-methyl-threonine ( Bmt ) , is synthesized by a polyketide biosynthetic mechanism [68] . As a cyclosporin mutant ( Cyb56 ) was shown to accumulate Bmt [69] , this substrate is likely synthesized by T . inflatum . The discovery of both a D-alanine racemase ( TINF00247 ) and a PKS gene ( TINF00267 ) within the cluster strongly suggests their involvement in production of these two unusual substrate molecules ( Figure 7 ) . The cluster also contains an aminotransferase ( TINF00351 ) , an enzyme involved in the synthesis of branched chain amino acids such as the Leu and Val residues found in cyclosporin . Several genes in the cluster belong to gene families commonly found in fungal secondary metabolite clusters , including a cytochrome P450 ( TINF00470 ) and a dehydrogenase ( TINF00195 ) . Two transcription factors , a C2H2 zinc-finger transcription factor ( TINF00183 ) on the 5′-edge of the cluster and a putative basic leucine zipper ( bZIP ) transcription factor ( TINF0394 ) on the 3′-end of the cluster , are candidates for a cluster-specific transcriptional regulator ( Figure 7 ) . Adjacent to the alanine racemase ( TINF0247 ) is a gene ( TINF00586 ) belonging to the cyclophilin family of peptidylprolyl isomerases ( IPR002130 ) ( Figure 7 ) . The first isolated cyclophilin , human cyclophilin A ( hCypA ) , was identified almost thirty years ago as the cellular target of cyclosporin [3] . Binding to hCypA is a prerequisite for the immunosuppressive activity of CsA as it causes CsA to undergo a conformational change to an entirely trans peptide conformation , which puts the calcineurin-binding motif of CsA in the reverse orientation compared to the crystal structure of CsA bound to a tetrapeptide substrate [70] , [71] and primes CsA to create a better fit to its cellular target calcineurin . Cyclophilins have since been identified in nearly all kingdoms of life , including animals , plants , insects , fungi , protists , and bacteria , and they are classified based on their cellular location , domain organization and function [72] , [73] . While different cyclophilins vary in their binding affinity for CsA , all exhibit petidyl isomerase activity that facilitates conformational changes from cis to trans at peptide bonds preceding prolines ( peptidyl-prolyl bonds ) , and thus may function as general molecular chaperones in protein folding [74] . They are also implicated in diverse cellular processes including cell signaling , cell cycle control , intracellular transport , stress response , and virulence in both plant [75] and animal pathogens [76] . Using an HMM model to the conserved cyclophilin-like domain ( CLD ) of cyclophilins ( Pfam: PF00160 . 16 ) , we identified ten proteins , including TINF00586 , containing the conserved CLD domain ( PF00160 . 16 ) in the T . inflatum genome ( Figure 8 A , B ) . The simA cluster cyclophilin ( TINF00586 ) has highest and equally scoring BLAST hits to two S . cerevisiae proteins , Cpr1 ( YDR155C ) ( e−41 ) , the yeast homolog of hCypA , and the yeast mitochondrial cyclophilin Cpr3 ( YML078W ) ( e-41 ) , and it contains an N-terminal signal peptide with a localization signal to mitochondria ( Figure 8B ) [77] . In a phylogeny including the conserved CLD domains of major cyclophilins from other fungi , animals , bacteria , and protists , the T . inflatum cyclophilins group in diverse locations in the phylogeny , suggesting that T . inflatum , like other eukaryotes , contains a full suite of cyclophilins ( Figures 8A , S7 ) . The simA cluster cyclophilin is similar in domain structure to hCypA and other cyclophilin A homologs . It groups phylogenetically with greater than 70% bootstrap support in a clade with another T . inflatum cyclophilin gene ( TINF04375 ) and a number of fungal cyclophilins with roles in morphological development and pathogenesis in either plant ( BCP1 ( Botrytis cinerea ) [78] , CpCYP1 ( Cryphonectria parasiticus ) [75] , and CYP1 ( Magnaporthe grisea ) [79] ) or animal ( Cpa1 and Cpa2 ( Cryptococcus neoformans ) systems [80] ( Figures 8A , C [Fungal CypA Clade] , S7 ) . Several putative cyclophilins previously cloned from T . inflatum , including two that are hypothesized to be alternately spliced products of a single gene targeted to the cytosol and mitochondria , respectively [81] , and another gene coding for an approximately 19 . 5 kDa protein ( cptA ) [82] , are nearly identical in sequence and all group closely with TINF04375 ( Figure 8C , S7 ) . Alternative splicing of this single gene is consistent with the finding that other cyclophilin genes in this clade CypA ( N . crassa ) [83] and BCP1 ( B . cinerea ) [78] ) , also produce alternately spliced mitochondrial and cytosolic isoforms ( Figure 8C , S7 ) . The simA cluster cyclophilin ( TINF00586 ) is distinct in sequence from TINF04375 and groups at the base of this clade with the CpCYP1 of C . parasiticus ( Figure 8C , S7 ) . While the direct mechanism of toxicity of CsA in insects remains unknown , histopathological changes consistent with Mitochondrial Pore Transition Permeability ( MPT ) , such as swollen , electron-dense , and occasionally lysed mitochondria , have been observed in several insect species treated with CsA [84] . We hypothesize that the simA cluster cyclophilin may be involved in targeting CsA to the insect mitochondria . Other possible functions include a role in folding of the cyclosporin peptide during export , creation of a pre-activated CsA-CYPA cocktail prior to delivery to the host , protection of CsA from proteolysis by endopeptidases [85] , binding to detoxifying proteins in hemolymph [86] , or auto protection for T . inflatum against CsA toxicity . Toxins and other secondary metabolites are suspected to function in insect pathogenesis but their expression patterns and modes of action remain poorly characterized . Previous studies suggest that many secondary metabolites are expressed at very low levels under most experimental conditions [87] , and their expression is elicited only in response to specific stimuli . Like many insect pathogenic fungi , T . inflatum exhibits a complex lifecycle encompassing a saprobic growth phase in soil and a pathogenic growth phase on and within the insect host ( Figure 1 ) . The pathogenic phase initiates with an infection phase that involves growth on , and penetration of , the insect cuticle . This infection phase is followed by a colonization phase , which initially involves a yeast-like ( hyphal body ) growth phase within insect hemolymph , and ultimately switches to a filamentous growth form that colonizes the insect to form an endosclerotium . Previous studies have shown that cyclosporin has immunosuppressive functions in insects [58] , as well as in humans , suggesting a role for and expression of cyclosporin inside the insect . In order to evaluate the expression of the simA cluster in relation to insect pathogenesis , RNA-Seq was carried out on fungal cultures grown on ( 1 ) minimal medium supplemented with insect cuticle ( infection stage ) , and ( 2 ) Grace's insect medium supplemented with insect hemolymph ( colonization stage ) . Each of these treatment samples ( cuticle and hemolymph ) were compared to a control grown on SDB . Transcriptional responses in these media were compared qualitatively to responses in the cyclosporin-inducing ( SM ) medium . In the strongly inducing SM medium , nearly all genes within the cluster were upregulated with extremely high significance ( q-value<0 . 0005 ) ( Figure 6 , Table S5 ) , but their relative expression levels varied widely ( Figures 9A , S6 , Table S5 ) . The most highly expressed gene in the cluster was the cyclophilin gene ( TINF00586 ) . However , this gene also had relatively high levels of constitutive expression in the control SDB medium and underwent only a 3 . 18× log2 fold increase in expression during time point 5 ( Figure 6 , Table S5 ) . In contrast , most other genes within the cluster , with the exception of TINF00536 and TINF00426 , had very low levels of constitutive expression in control SDB medium , but were more highly upregulated in SM medium ( Figures 6 , 9A , S6 , Table S5 ) . For example , the PKS ( TINF00267 ) , the cytochrome P450 ( TINF00470 ) , the D-alanine racemase ( TINF00247 ) , the dehydrogenase ( TINF00195 ) , a hypothetical protein ( TINF00377 ) , and the aminotransferase ( TINF00351 ) had over a 9× log2 fold increase in expression in SM medium compared to the control ( Figure 6 , Table S5 ) . Several other genes , including simA ( TINF00159 ) , a cytochrome b-2-like protein ( TINF0174 ) , two hypothetical proteins ( TINF00141 and TINF07874 ) , and the bZIP transcription factor ( TINF00394 ) experienced between 5–9× log2 fold increases in expression ( Figure 6 , Table S5 ) . The C2H2 transcription factor on the 5′ end of the cluster ( TINF00183 ) experienced less than 1× log2 fold increases in expression , suggesting that the bZIP transcription factor ( TINF001374 ) is more likely the cluster-specific transcriptional regulator ( Figure 6 , Table S5 ) . While the simA NRPS itself was not significantly upregulated in media simulating stages of insect pathogenesis ( cuticle and hemolymph media ) , several cluster genes showed significant upregulation ( q-value<0 . 05 ) ( Figure 9B , Table S6 ) . In cuticle medium , the PKS gene ( TINF00267 ) and the cytochrome P450 ( TINF00470 ) were significantly upregulated ( Figure 9B , Table S6 ) . In hemolymph medium , a greater number of cluster genes including the PKS gene ( TINF00267 ) , the bZIP transcription factor ( TINF00394 ) , the cyclophilin homolog ( TINF00586 ) , and a hypothetical protein containing a thioester/thiol ester dehydrase-isomerase domain ( TINF00377 ) were significantly upregulated ( Figure 9B , Table S6 ) . The P450 ( TINF00470 ) was the most highly upregulated gene in the cuticle medium ( 3 . 25× log2 fold ) , while the bZIP transcription factor ( TINF00394 ) was the most highly upregulated gene in hemolymph medium ( 3 . 58× log2 fold ) ( Table S6 ) . Importantly , the cyclophilin gene ( TINF00586 ) showed significant upregulation ( 1 . 82× log2 fold ) only in hemolymph media and had the highest relative expression level of all cluster genes in hemolymph media ( Figure 9B , Table S6 ) . However , culture media conditions can only simulate conditions of insect pathogenesis , and differential expression cannot be solely attributed to the added insect components as differences in the composition and pH of the basal media used may also influence expression patterns . RNA samples were also harvested 24 hours after transfer from a rich SDB medium to hemolymph medium . Given that full induction of cyclosporin production took nearly 6 days in a strongly inducing medium , it is possible that the weaker response observed in media containing insect components reflects an early stage of cyclosporin induction . The larger number of significantly upregulated genes and the strong upregulation of the bZIP transcription factor ( TINF00394 ) only in hemolymph media , however , suggests a possible role for cyclosporin and the cyclophilin homolog during the colonization phase inside an insect host ( Figure 9B , Table S6 ) . The disjunct distribution of cyclosporin biosynthesis across fungal taxa poses interesting questions about the origins and evolution of the cyclosporin biosynthetic cluster . In order to search for homologs of cyclosporin cluster genes in other fungi , the top 25 BLAST hits to the antiSMASH predicted simA cluster genes plus 10 flanking genes on either side in the NCBI nr database were aligned and phylogenies constructed using maximum likelihood ( Figure S8 ) . Pairwise BLASTP searches among the set of the fourteen hypocrealean fungi analyzed for phylogenomic analyses were also performed to identify reciprocal best-pair hits between these genomes and these best-pair BLASTP hits were considered as orthologs . These analyses revealed that only one gene ( TINF00195 ) between the C2H2 transcription factor ( TINF00183 ) , adjacent to the 5′ end of the RNA-Seq defined cluster ( Figure 10 , red line ) , and the 3′-end of the RNA-Seq defined cluster ( Figure 10 , blue line , at TINF07874 ) had hits above e−05 to bacterial genes ( Figure S8 ) . Similarly , only a few genes ( TINF00557 , TINF00586 , TINF00426 , TINF00174 , TINF00267 , TIN00470 , and TINF00394 ) had orthologs in other sequenced hypocrealean taxa ( Figures 10 , S8 ) . In contrast , most genes flanking this region on both sides , as well as the 8 genes at the 5′-end of the antiSMASH predicted cluster ( TINF00177 to TINF00557 ) contained single-copy orthologs in nearly all other hypocrealean taxa that showed relatively conserved synteny with those in T . inflatum ( Figures 10 , S8 ) . The 8 genes on the 5′ end of the antiSMASH predicted cluster that had homologs in other Hypocreales were excluded from the cluster by both SMURF and the RNA-Seq predicted cluster ( Figures 6 , 7 ) . Additionally , the few homologs of genes within the RNA-Seq defined cluster in other hypocrealean taxa were scattered elsewhere in these genomes and not located between these conserved flanking regions ( Figure 10 , S8 ) . In most species , no additional genes were found between the 5′ flank ( TINF00183 ) and the 3′ flank ( TINF07874 ) of the RNA-Seq defined cluster and the intervening region between these boundaries was less than 5 kb ( Figure 10 ) . F . oxysporum contained a single additional gene in this region , while C . militaris underwent an inversion that added additional genes , none of which were orthologs of the simA cluster genes ( Figure 10 ) . These results suggest one of two hypotheses regarding the origin of the nearly 100 kb simA cluster region in T . inflatum that is missing in other Hypocreales: 1 ) the cluster has been horizontally transferred from another fungal species into this site in T . inflatum , or 2 ) this region has evolved by recruitment of genes from other regions of the T . inflatum genome . Horizontal transfer of complete secondary metabolite clusters between heterologous bacterial species by transposition has been demonstrated [88] , and evidence exists for horizontal transfer of large secondary metabolite clusters among fungi [89] , [90] . However , BLAST searches did not detect a homologous cluster in other fungal taxa . To further test for horizontal transfer , we utilized the customized pipeline ( CRAP ) [89] to scan for syntenic homologs of simA cluster genes in 195 sequenced fungal genomes . We did not detect a cluster containing both the NRPS and PKS genes and a majority of accessory genes from the simA cluster . While these results do not preclude the existence of a nearly complete cluster in a yet to be sequenced fungus , available evidence suggests that the cluster more likely evolved by a process of recruitment of genes into the cluster from elsewhere within the T . inflatum genome . The fact that simA and esyn1 synthetases clearly share related A-domains while the clusters containing these NRPSs lack other shared genes leads us to hypothesize that these clusters have evolved by recruitment of distinct modifying enzymes into regions surrounding these core NRPSs through rearrangement and transposition . Transposons have been found adjacent to other secondary metabolite gene clusters in fungi , such as the gliotoxin cluster in Aspergillus fumigatus [91] . While their role in shaping the evolution of secondary metabolite clusters remains speculative , they represent a potential mechanism for gene recruitment . A gypsy LTR retrotransposon , related to F . oxysporum SKIPPY [92] , was found on the 3′ end of the simA cluster ( Figure 10 ) , and several lines of evidence suggest that the region surrounding the simA cluster may be prone to rearrangement in other taxa . Both Tr . atroviride and C . militaris show evidence of an inversion or transposition in this region of the genome . In Tr . atroviride , this is simply an inversion which did not add additional genes . In C . militaris , an additional 0 . 5 Mb of sequence distinct from the T . inflatum sequence is found between the two rearranged flanking regions , but this additional sequence does not contain any homologs of simA cluster genes nor is it predicted by SMURF or antiSMASH to contain other secondary metabolite clusters ( Figure 10 ) . Although the simA clade itself was shown previously to group sister to a clade of bacterial NRPSs [54] , no bacterial homolog of simA were found within the simA clade . Thus , we conclude that simA evolved through duplication and divergence of fungal NRPS modules within T . inflatum rather than by recent horizontal transfer from bacteria ( Figures 4 , S5 , S8 ) . A number of related NRPSs that share homologous A-domains with simA , including enniatin , beauvericin , bassianolide , and aureobasidin A synthetases , evolved through a similar process of module duplication , but also fusion of distantly related NRPS modules ( Figures 4 , 5 , S5 ) . Interestingly , all of these metabolites possess insecticidal or fungicidal properties . Other genes within the secondary metabolite gene clusters containing these NRPSs , however , do not show homology with those in the simA cluster and are not syntenic with the simA cluster . Regions syntenic with the simA cluster in other hypocrealean fungi instead lack the nearly 100 kb of the simA cluster present in T . inflatum . However , searches for orthologs of the simA cluster genes in other fungi using BLAST searches of the NCBI nr database and the CRAP pipeline found no clusters containing more than a few genes showing similarity to those in the simA cluster . While horizontal transfer from a yet un-sampled fungus cannot be ruled out , we suggest that the simA cluster instead had a lineage specific origin , having evolved through recruitment of genes from other locations in the T . inflatum genome . The discovery of a homolog of hCypA , the cellular target of cyclosporin in mammalian systems , within the simA cluster of T . inflatum is novel . This is the first report of a cyclophilin gene located within a secondary metabolite gene cluster , although several genes shown to be dependent on the activity of calcineurin , the target of CsA , are suspected of being located within a secondary metabolite biosynthetic cluster in B . cinerea [78] . The up-regulation of the cluster cyclophilin in hemolymph and high expression levels under both inducing conditions and in insect hemolymph medium suggests a role for this gene in mediating the activity of cyclosporin in vivo . Elucidation of the simA gene cluster through a combination of computational and transcriptional approaches opens the door for functional studies and chemical analyses to address mechanisms of action in nature and potential novel applications of these compounds in pathogenicity and medicine .
Tolypocladium inflatum NRRL 8044 , the original strain from which cyclosporin was isolated , was obtained from NRRL . Cultures were grown for 2 . 5 weeks on cornmeal agar to induce sporulation . For DNA extractions , conidia from agar plates were used to inoculate potato dextrose broth cultures , which were grown for 3 days before harvesting . Lyophilized mycelia were ground in liquid nitrogen and genomic DNA was isolated using the Qiagen genomic tip 500 following the manufacturer-supplied protocol for isolation of genomic DNA from plants and filamentous fungi ( Qiagen ) . A 350 bp insert Illumina library was prepared by shearing two samples of approximately 5 µg of genomic DNA in a Biorupter XL sonicator for 20 min with a cycle of 30 sec on and 30 sec off . These samples were pooled and prepared following the Illumina protocol for paired-end sequencing . Illumina sequencing was performed on the Illumina GAII machine at the Center for Genome Research and Biocomputing ( CGRB ) at Oregon State University . For 454 sequencing , a shotgun library and a 3-kb paired end 454 library were each prepared and sequenced on a full plate using titanium chemistry at the Duke IGSP Sequencing Core Facility at Duke University . 454 reads were assembled using the Newbler Assembler version 2 . 3 ( 454 Life Sciences ) using a combined shotgun and 3 kb paired end library assembly . Out of a total of 4 , 260 , 863 input 454 reads ( 1 , 516 , 236 single end shotgun reads , 1 , 067 , 664 mate pair reads with both pairs , and 1 , 676 , 963 singleton mate pair reads ) , 4 , 077 , 744 ( 95 . 70% ) were assembled into the Newbler Assembly . Illumina reads were first trimmed to 50 bp , and sequences containing adaptors , N's , or greater than 2 bp with a quality score below 20 were filtered out of the dataset . A total of 37 , 643 , 435 quality filtered reads , containing 33 , 586 , 130 paired end reads , were submitted to SOAP Gapcloser [93] to fill gaps in the Newbler scaffolds . Finally , a mapping assembly was performed in MIRA to map the quality filtered Illumina reads to the SOAP Gapcloser assembly to correct for homopolymer base pair errors . Genome annotations were created in MAKER 2 . 00 [23] using three ab initio gene prediction models: an AUGUSTUS model trained for F . graminearum , a GeneMark model trained for T . inflatum via self-training , and a SNAP model trained for F . graminearum . Protein datasets from F . graminearum , Nectria haematococca , Tr . reesei , Tr . virens , and M . robertsii were submitted to MAKER as protein evidence . ESTs from C . militaris , B . bassiana , M . robertsii , and Tr . reesei downloaded from GenBank were included as EST evidence . Illumina PE RNA-Seq reads from T . inflatum PDB grown cultures ( see RNA isolation and sequencing ) were trimmed to 70 bp and filtered to remove sequences containing adaptors , N's , or greater than 2 bp with a quality score below 20 , and assembled into transcripts using OASES with a coverage cutoff of 3 [94] . Assembled transcripts were input into MAKER as EST evidence . Transfer RNAs ( tRNAs ) were identified with tRNA scan-SE [95] . Repeat elements were identified with Repeat Masker ( Smit , AFA , Hubley , R & Green , P . RepeatMasker Open-3 . 2 . 8 1996–2010 http://www . repeatmasker . org ) using the fungal transposon species library ( database version 20090604 ) as input and crossmatch version . 990329 . The number of CPA ( NCBI Accession AM990997 . 1 ) and Restless ( NCBI Accession Z69893 . 1 ) elements was identified in hypocrealean taxa by supplying these sequences as a library to Repeat Masker using the same settings . Proteins for the set of hypocrealean taxa utilized for functional annotations , as well as three Sordariomycete outgroups ( N . crassa , V . dahliae , and V . albo-atrum ) , were clustered using MCL [30] at inflation parameter 3 within the Hal pipeline [31] and then filtered to identify single-copy orthologs . Single copy orthologs were aligned separately using MUSCLE [96] and then concatenated . The concatenated alignment was used to infer a phylogeny using RAxML with the best fit model of amino acid substitution for each gene partition estimated by ProtTest [97] and branch support estimated from 1000 bootstrap replicates . In a second set of analyses , amino acid rate categories were estimated across eight categories using PAML [98] and the effect of amino acid positions with high rates of substitution was determined by repeating the same RAxML analyses without the 6th , 7th , and 8th rate categories . The ultrametric tree used for CAFÉ analyses was computed in r8s [99] from the phylogeny inferred by the first RAxML analysis using previously estimated divergence dates for Hypocreales [16] ( Table S2 ) . The number of clusters and proteins within clusters were mapped to nodes using a custom perl script . A functional annotation pipeline was developed for the hypocrealean fungi F . graminearum , F . oxysporum , F . verticillioides , N . haematococca , Tr . reesei , Tr . virens , Tr . atroviride , M . robertsii , M . acridum , C . militaris , and T . inflatum . Functional annotations were classified with InterproScan ( http://www . ebi . ac . uk/interpro ) using hmmpfam , PatternScan , ProfileScan and blastprodom databases and converted to GO with the Interpro2GO mapping ( version 11/05/2011 ) . Orthologous groups of proteins for these eleven fungi as well as the model fungi S . cerevisiae , Schizosaccharomyces pombe , and N . crassa were determined with Inparanoid [100] , [101] . GO enrichments were performed by transferring additional GO annotations associated with proteins from model fungi to uncharacterized proteins within the same orthologous cluster . Protein localization signals excluding the plastid location were identified using TargetP [102] and Predotar [103] while transmembrane proteins were characterized using TMHMM [104] . Carbohydrate active enzymes ( CAZymes ) were determined by BLAST searches against the CAZymes database with a cutoff of <e−30 . Proteases were identified by BLASTP searches of the MEROPS [105] database ( http://merops . sanger . ac . uk/ ) with default settings ( cutoff of e−05 ) , and P450 enzymes were identified as BLAST hits to entries in the Nelson P450 [106] database below a cutoff of <e−20 . The program CAFÉ , using the default settings , was used to analyze gene families ( CAZymes , P450s , and proteases ) expanded or contracted in taxa within Hypocreales . NRPS and PKS secondary metabolite genes and their domain structures were characterized using three methods: 1 ) HMMER searches using models build for Adenylation ( A ) , Thiolation ( T ) , and Condensation ( C ) domains of NRPSs [54] , 2 ) SMURF [43] , and 3 ) antiSMASH [44] . Adenylation ( A ) domains from NRPSs and ketosynthase ( KS ) domains from PKSs were extracted from the same eleven hypocrealean taxa included in the functional annotation pipeline and aligned together with known NRPSs from fungi and several outgroup A-domains ( Acetyl CoA Synthetases , AcylCoA ligases , Ochratoxin , and CPS1 ) using MAFFT [107] . The alignment was examined and manually edited to remove regions of ambiguous alignment and a maximum likelihood phylogeny constructed in RAxML [108] using 1000 bootstrap replicates and the best fit model identified by ProtTest [97] ( REVF plus gamma for A domains ) . The simA NRPS and the PKS gene found within the cyclosporin cluster were also subjected to BLAST against the NCBI nr database and A-domains from the top 50 hits were extracted , aligned , and analyzed as described above to identify any putative bacterial homologs of simA . The identified secondary metabolite genes were placed into secondary metabolite gene clusters using SMURF and antiSMASH . Protein sequences from the T . inflatum cyclosporin cluster plus ten flanking genes were used to search the genomes of related hypocrealean fungi using TBLASTN and BLASTP searches . The TBLASTN searches identified genes flanking the cyclosporin cluster that localized to the same contig or scaffold in other species . Genome-wide best-pair BLASTP searches were used to refine this search and identify those genes in the simA cluster with best pairwise BLASTP hits in other genomes and these were considered putative orthologs . The genomic sequence was also aligned with MAUVE at the DNA level to confirm synteny relationships ( data not shown ) [109] . To identify potential homologs in fungal species other than hypocrealean taxa and to address phylogenetic evidence for horizontal transfer , all T . inflatum cyclosporin cluster genes plus ten flanking genes of the antiSMASH predicted cluster were subjected to BLASTP against the nr database and the top 25 hits were extracted and aligned with MAFFT [107] . These alignments were filtered to remove regions of poor alignment using Gblocks ( with relaxed setting ) , and ProtTest [97] was used to identify best-fit models for each gene alignment . Maximum likelihood phylogenies were constructed using RAxML [108] with the corresponding best-fit protein model and 100 bootstrap replicates . The simA cluster genes were also run through a customized pipeline ( CRAP ) [89] to search 195 sequenced fungal genomes and 1150 bacterial genomes for syntenic homologs of the simA cluster . For initial coverage of the transcriptome , conidia from 2 1/2 week-old cultures grown on cornmeal agar were adjusted to a concentration of 1×107 spores/mL and 1 mL was used to inoculate a 100 mL liquid culture of potato dextrose broth . Mycelia were harvested after three days , flash frozen and ground in liquid nitrogen . RNA was extracted using the Qiagen RNAeasy kit . cDNA was prepared using the Mint cDNA Kit ( Evrogen ) , sheared by nebulization for 6 min at 34 psi , prepared using the Illumina protocol for PE sequencing , and sequenced in one lane of a paired end 80 bp run on the Illumina GAII . For the RNA-Seq time course experiment in cyclosporin inducing ( SM medium ) , two 200 mL flasks containing 100 mL of YM medium ( 4 g yeast extract , 20 g malt extract in 1L ddH20 ) were each inoculated with approximately 1 mL of 1×107 conidia/mL . After two days growth in YM medium , 10 mL of the YM culture was transferred to 125 mL of SM production media or Sabouraud Dextrose Broth ( SDB ) in 200 mL flasks . All cultures were grown at 21°C . Three flasks ( 3 biological replicates ) for each treatment ( SM or SDB ) were harvested every two days for a total of six time points ( days 2 , 4 , 6 , 8 , 10 , 12 ) . Tissue was flash frozen and ground in liquid nitrogen and total RNA extracted in TRIzol® . Remaining tissue and culture filtrate was frozen at −20°C until chemical extraction for HPLC analyses . The RNA samples were prepared using the Illumina TruSeq RNA sequencing kit , randomized , and sequenced across three lanes of a 50 bp SE run on the Illumina HiSeq2000 . For RNA-Seq expression studies in insect media , cultures were grown under three conditions: 1 ) a rich medium of Sabouraud dextrose broth ( SDB ) , 2 ) a minimal salts medium [110] supplemented with 10% ( w/v ) black vine weevil ( Otiorhynchus sulcatus , Coleoptera ) insect cuticle cleaned from soft tissue with sodium tetraborate [111] to simulate the infection stage , and 3 ) Grace's insect medium ( Gibco , unsupplemented ) with 10% ( v/v ) filter sterilized black vine weevil hemolymph added to simulate the colonization stage . The minimal salts medium utilized for cuticle media contained 0 . 02% KH2PO4 , 0 . 01% MgSO4 , 0 . 2 p . p . m . FeSO4 , 1 . 0 p . p . m . ZnSO4 , 0 . 02 p . p . m . NaMoO4 , 0 . 02 p . p . m . CuSO4 , 0 . 02 p . p . m . MnCl2 adjusted to pH 6 . 5 These cultures were grown in a 2-stage fermentation that has proven a reproducible method for eliciting expression of proteins in response to specific elicitors in insect pathogenic fungi [112] . Conidia from cultures grown on cornmeal agar for 2 . 5 weeks were used to inoculate a 100 mL culture of rich media ( SDB ) at a final concentration of approximately 1×106 conidia/mL . These cultures were grown on a shaking incubator for 48 hours , washed in sterile water , and approximately 500 mg wet weight of mycelia transferred to three replicates of 2 mL cultures for each media condition . Mycelia were grown for an additional 24 hours before harvesting . All cultures were grown at 21°C . RNA was extracted with TriZol® according the manufacturers protocol ( Invitrogen ) , polyA RNA isolated using the Ambion PolyA Purist kit , and cDNA prepared using the Superscript III kit and random hexamer primers ( Invitrogen ) . A 450 bp insert library was prepared for each biological replicate according to the Illumina PE protocol and all samples were multiplexed in each of three lanes of a SE 40 bp run . Barcodes were trimmed from 40 bp Illumina reads to a length of 36 bp for the insect pathogenesis experiment and the first and last nine bases were trimmed from the 50 bp reads based on quality score profiles to a length of 40 bp for the cyclosporin inducing experiment . Reads were mapped to gene models using GENE-Counter [113] . Differential expression was analyzed in GENE-Counter , which utilizes a negative binomial model and the NBP-Seq R package to model differential gene expression [114] . For the time course experiment , the three biological replicates in SM media were compared with the three biological replicates in the SDB media at each time point separately . For the insect assays , pairwise comparisons of SDB vs cuticle medium and SDB vs hemolymph medium were performed in GENE-Counter . Q-values and fold changes ( log2 transformed ) were calculated using the normalized expression values from NBP-Seq . Relative expression levels were calculated as Reads Per Kilobase of transcript per Million mapped reads ( RPKM ) [115] . Culture filtrates from three biological replicates at each time point were pooled for chemical extraction . Each culture filtrate was applied to a glass column containing Diaion® HP20 resin ( 20 g , Supelco ) , which had been sonicated in MeOH ( to de-gas ) and then pre-washed with H2O ( 200 mL ) . The column loaded with sample was then eluted sequentially with H2O ( 200 mL , to desalt the sample ) , MeOH ( 100 mL ) and acetone ( 100 mL ) . The latter two organic solvent eluents were combined and concentrated to provide an organic extract from each culture filtrate . In each case , the organic extract was applied to a C18 reversed-phase solid phase extraction ( RP18 SPE ) cartridge ( 10 g ) , which had been primed in 100% methanol , and then equilibrated in 70% methanol in water . The SPE cartridge was then eluted sequentially with 70% and 100% methanol in H2O before being washed with dichloromethane . The cyclosporin-containing SPE fractions ( 100% methanol , determined by direct injection MS ) from each SM culture filtrate extract were selected for comparative HPLC ( used to establish protocols for peak collection ) and also LC-MS profiling , alongside the corresponding control SDB medium extracts . HPLC of each sample ( 50 µg per 5 injection ) was performed using a linear gradient from 60–100% methanol in H2O over 40 min followed by isocratic 100% methanol for 20 min ( column: Synergi Hydro-RP , 4 . 6×250 mm , 0 . 6 mL/min ) . LC-MS of 5 µg-containing aliquots was performed under identical HPLC solvent conditions using a Synergi Hydro-RP , 2×100 mm column with a flow rate of 0 . 2 mL/min . HPLC was performed on a Shimadzu HPLC system comprising a SIL-20AC autosampler , dual LC-20AD solvent pumps and a SPD-M20A UV/VIS photodiode array detector . LC-ESI ( + ) MS data were obtained using an AB SCIEX QTrap 3200 mass spectrometer interfaced with a Shimadzu Prominence HPLC system . HPLC-grade solvents were used for all chemical extraction and fractionation . The Whole Genome Shotgun projects have been deposited at DDBJ/EMBL/GenBank under the accession number AOHE00000000 . | Tolypocladium inflatum , the fungus from which the immunosuppressant drug cyclosporin was isolated , is a prolific producer of secondary metabolites with potential applications in medicine and agriculture . We have sequenced the first draft reference genome of T . inflatum , which also represents the first genome of a novel family of insect pathogenic fungi , Ophiocordycipitaceae . We present comparative genomic and evolutionary analyses of the cyclosporin nonribosomal peptide synthetase ( simA ) , which highlight the lineage specific nature of cyclosporin's origin and the homology of cyclosporin adenylation ( A ) domains with other fungal NRPSs whose products show anti-insect activity . RNA-Seq data profiles the expression patterns of the cyclosporin gene cluster in an inducing medium and in response to media simulating distinct stages of insect pathogenesis . Sequencing of the T . inflatum genome has uncovered the metabolite gene cluster responsible for cyclosporin biosynthesis and characterized complex patterns of its evolution . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"biochemistry",
"genomics",
"drugs",
"and",
"devices",
"chemical",
"biology",
"chemistry",
"microbiology",
"biology",
"evolutionary",
"biology"
] | 2013 | The Genome of Tolypocladium inflatum: Evolution, Organization, and Expression of the Cyclosporin Biosynthetic Gene Cluster |
Trachoma is widely considered a disease of poverty . Although there are many epidemiological studies linking trachoma to factors normally associated with poverty , formal quantitative data linking trachoma to household economic poverty within endemic communities is very limited . Two hundred people with trachomatous trichiasis were recruited through community-based screening in Amhara Region , Ethiopia . These were individually matched by age and gender to 200 controls without trichiasis , selected randomly from the same sub-village as the case . Household economic poverty was measured through ( a ) A broad set of asset-based wealth indicators and relative household economic poverty determined by principal component analysis ( PCA , ( b ) Self-rated wealth , and ( c ) Peer-rated wealth . Activity participation data were collected using a modified ‘Stylised Activity List’ developed for the World Bank’s Living Standards Measurement Survey . Trichiasis cases were more likely to belong to poorer households by all measures: asset-based analysis ( OR = 2 . 79; 95%CI: 2 . 06–3 . 78; p<0 . 0001 ) , self-rated wealth ( OR , 4 . 41 , 95%CI , 2 . 75–7 . 07; p<0 . 0001 ) and peer-rated wealth ( OR , 8 . 22 , 95% CI , 4 . 59–14 . 72; p<0 . 0001 ) . Cases had less access to latrines ( 57% v 76 . 5% , p = <0 . 0001 ) and higher person-to-room density ( 4 . 0 v 3 . 31; P = 0 . 0204 ) than the controls . Compared to controls , cases were significantly less likely to participate in economically productive activities regardless of visual impairment and other health problems , more likely to report difficulty in performing activities and more likely to receive assistance in performing productive activities . This study demonstrated a strong association between trachomatous trichiasis and relative poverty , suggesting a bidirectional causative relationship possibly may exist between poverty and trachoma . Implementation of the full SAFE strategy in the context of general improvements might lead to a virtuous cycle of improving health and wealth . Trachoma is a good proxy of inequality within communities and it could be used to target and evaluate interventions for health and poverty alleviation .
Trachoma is leading infectious cause of blindness worldwide [1] . Trachomatous trichiasis ( TT ) is the late stage consequence of repeated conjunctival Chlamydia trachomatis infection in which eyelashes turn towards the eye , causing pain and eventually irreversible blinding corneal opacification ( CO ) . About 229 million people live in trachoma endemic areas , and approximately 7 . 3 million have untreated TT [2 , 3] . More than 2 . 4 million people are visually impaired from trachoma worldwide , among which between 439 , 000 and 1 . 2 million are estimated to be irreversibly blind [2 , 4] . The WHO recommends the SAFE Strategy for trachoma control [5] . This involves Surgery for trichiasis , Antibiotics for infection , Facial cleanliness and Environmental improvements to suppress chlamydial infection and transmission . Trachoma has long been considered a disease of poverty [6] . It is believed that the decline in trachoma observed in Europe , North America and elsewhere over the last century , in the absence of specific control measures , was largely attributable to general improvements in socio-economic status [7 , 8] . Trachoma remains prevalent in developing and marginalised communities , particularly in Africa , where crowded living conditions are common and access to clean water , sanitation and health care are often limited [6 , 8 , 9] . However , not all people living in such settings acquire active or scarring trachoma . It is possible that , within apparently homogeneous communities , the individuals who are most vulnerable to developing the blinding complications of trachoma are the poorest members of the poorest communities , although this has not been adequately investigated [10] . Moreover , the disability that TT causes may lead to reduced productivity , unemployment and loss of income , putting additional financial pressure on an already strained household [11–13] . The effect of trachoma on income may begin prior to the visual impairment , with the pain and the photophobia from trichiasis limiting function [13 , 14] . Of note , blindness has generally been associated with lower socio-economic status [15–17] . In low and middle income countries ( LMICs ) resources are often shared within households . Therefore , relative wealth or poverty in LMICs needs to be measured at household level , as the economic impact of a medical condition or intervention potentially affects the whole family [18] . In low-income settings estimating income can be difficult , as many people are self-employed and incomes are subject to significant short-term fluctuations [18 , 19] . In addition , people may earn from sources that they do not wish to disclose . Consumption expenditure data are considered more reliable than income data [16 , 19] . However , this method is subject to recall bias and requires detailed questionnaires , which are time consuming and costly to administer [19] . An alternative approach is to use a range of asset and housing characteristics as proxy indicators for household wealth and socio-economic status [19 , 20] . A key advantage of this approach is that it measures the long-term financial status of a household , and is less vulnerable to short-term fluctuations than income and consumption expenditure [19 , 20] . On the other hand , asset score only measure relative poverty , which may preclude regional or international comparability . There is surprisingly little direct data that formally quantifies the relationship between trachoma and economic poverty , and none that specifically focuses on the scarring sequelae . The aim of this study was to investigate in detail the relationship between poverty and trachomatous trichiasis through an asset-based analysis , self-rated and peer-rated wealth measures , and participation in productive activities .
This study was reviewed and approved by the Food , Medicine and Healthcare Administration and Control Authority of Ethiopia , the National Health Research Ethics Review Committee of the Ethiopian Ministry of Science and Technology , Amhara Regional Health Bureau Research Ethics Review Board Committee , the London School of Hygiene and Tropical Medicine ( LSHTM ) Ethics Committee , and Emory University Institutional Review Board . Written informed consent in Amharic was obtained prior to enrolment from participants . If the participant was unable to read and write , the information sheet and consent form were read to them and their consent recorded by thumbprint . This case-control study was nested within a clinical trial of two alternative surgical treatments for trichiasis . From the 1000 trichiasis cases recruited into the trial , every fifth consecutive case was also enrolled into this economic poverty study and matched to a non-trichiasis control . This approach was chosen for logistical and methodological reasons , in order to identify and collect data from controls within the shortest possible time period following case recruitment . Cases were defined as individuals with one or more eyelashes touching the eyeball or with evidence of epilation in either or both eyes in association with tarsal conjunctival scarring . People with trichiasis of other causes , recurrent trichiasis and those under 18 years were excluded . Trichiasis cases were identified mainly through community-based screening . Trichiasis screeners and counsellors ( Eye Ambassadors ) visited every household in their target village , identified and referred trichiasis cases to health facilities where surgical services were provided . Some individuals self-presented or were referred by local health workers . Recruitment was mainly from three districts of West Gojam Zone , Amhara Region , Ethiopia between February and May 2014 . This area has one of the highest burdens of trachoma worldwide [21] . Controls were individuals without clinical evidence or a history of trichiasis ( including surgery and epilation ) , and who came from households without a family member with trichiasis or a history of trichiasis , as we wanted to measure household level relative poverty , which requires comparison of trichiasis case households with households without trichiasis cases . One control was individually matched to each trichiasis case by location , sex and age ( +/- two years ) . The research team visited the sub-village ( 30–50 households ) of the trichiasis case requiring a matched control . A list of all potentially eligible people living in the sub-village of was compiled with the help of the sub-village administrator . One person was randomly selected from this list using a lottery method , given details of the study and invited to participate if eligible . If a selected individual refused or was ineligible , another was randomly selected from the list . When eligible controls were not identified within the sub-village of the case , recruitment was done in the nearest neighbouring sub-village , using the same procedures . Data on detailed demographic characteristics were collected . Household economic poverty was measured through ( a ) Asset based wealth indicators , ( b ) Self-rated wealth , and ( c ) Peer-rated wealth . Activity participation data was collected using a modified ‘Stylised Activity List’ developed for the World Bank’s Living Standards Measurement Survey [22] . Visual acuity of both cases and controls were measured and cases underwent detailed trachoma examination . To detect a difference in asset-based principal component analysis ( PCA ) similar to that found in the Cataract Impact Study ( mean and standard deviation of asset based PCA score in cataract cases and their controls 0 . 6 and 2 . 0; and 0 . 3 and 2 . 6 , respectively ) with an alpha of 0 . 05 and 95% power , at least 346 ( 173 in each group ) participants were required [16] . We recruited 200 trichiasis cases and 200 age , sex and location matched non-trichiasis controls . Data were double-entered into Access ( Microsoft ) and transferred to Stata 11 ( StataCorp ) for analysis . Conditional logistic regression was used to compare basic characteristics of matched cases and controls .
Cases and controls were well matched in terms of location , gender and age and had similar levels of literacy , household size and household occupation ( Table 1 ) . Compared to the controls , the trichiasis cases were less likely to be married , more likely to be either unemployed or work as daily labourers , less likely to have a family member with formal education and more likely to have experienced a health problem during the last month . As expected , cases were more likely to be visually impaired than the controls ( 37 . 0% v 3 . 0% , respectively; OR = 69 . 0; 95%CI 9 . 58–496 . 82; p<0 . 0001 ) The asset variables used in the PCA are described in Table 2 and their summary statistics are shown in S1 Table . The PCA was based on a combination of 28 asset values . The other 32 measured assets were excluded as they were present in less than 5% or more than 95% of the participants’ households . Households were generally poor . About 67% had a latrine , among which 65% were of the “non-improved” pit latrine type without a concrete slab . About half ( 54% ) had their cattle dwelling within the main house . Ownership of durable assets such as mobile phones and radio was low ( <30% ) . Only 17% of the households had access to electricity . About 12% of the households had taken a government loan . Overall , cases had fewer household and agricultural assets than controls and were more likely to have a government loan ( Table 2 ) . There was no difference in the ownership of the house they were living in ( 92 . 0% vs 94% , p = 0 . 22 ) , or access to electricity ( 18·5% v 16·5% , p = 0·40 ) . Case households had fewer rooms ( 1 . 22 vs 1 . 55 , p<0·0001 ) , and had a higher density of persons per room than the controls: 4 . 0 , 95%CI 3 . 6–4 . 4 vs 3 . 3 , 95%CI 3 . 0–3 . 6 respectively ( P = 0 . 020 ) . The overall asset index accounts for 21% of the total variance ( S1 Table ) . Among the three subset asset indices , the agricultural asset indicators had the highest factor scores and accounted for the highest weights in measuring wealth in this population . In contrast , the housing characteristics and utilities index , except for the number of metal roof sheets , had generally lower factor scores and contributed lower weights in estimating wealth than the other two subset indices . Among all indices , number of oxen and cows owned ( 0 . 324 ) , the number of metal roof sheets ( 0 . 320 ) and amount of land owned in hectares ( 0 . 319 ) had the highest weights in estimating wealth . In contrast , access to electricity ( -0 . 096 ) having cattle dwelling within the main house ( -0 . 024 ) and having a government loan ( -0 . 038 ) had negative weights . Fig 1 illustrates the distribution of the subset and overall asset indices , in order to determine whether clumping or truncation were present in this data . Overall , there was evidence of truncation and clumping when the three subset indices ( Fig 1A to 1C ) are used separately . However , the distribution of the overall combined factor scores was much smoother; and clumping and truncation were not observed ( Fig 1D ) . There was a strong association between being a trichiasis case and asset based household economic poverty: OR = 2 . 79; 95%CI , 2 . 06–3 . 78; p<0 . 0001 ( Table 3 ) . This relationship persisted after adjusting for marital status , and highest family education ( OR = 2 . 78; 95%CI , 2 . 00–3 . 87; p<0 . 0001 ) . For stratified analyses we combined “richest” and “rich” with “middle” because of small numbers , to create a “middle & above” category with three levels of socio-economic status measure to facilitate data modelling . Compared to the controls , trichiasis cases were more likely to be from the poorest ( OR = 2 . 65; 95%CI , 2 . 05–3 . 42; p<0 . 0001 ) households than from the middle & above households ( Table 4 ) . In the stratified analysis , the association between asset based household economic poverty and trichiasis persisted regardless of age , gender , marital status , and in people with normal visual acuity after adjusting for the matching variables and family education ( Table 4 ) . On both the self-rated and peer-rated scores , the households of trichiasis cases were rated poorer than controls ( Table 3 ) . This association persisted in both self-rated ( OR = 3 . 99; 95%CI , 2 . 43–6 . 54; p<0 . 0001 ) and peer-rated ( OR = 9 . 10; 95%CI , 4 . 79–17 . 27; p<0 . 0001 ) wealth measures after adjusting for marital status and highest family education . Compared to the controls , the trichiasis case households were more likely to be rated as poorest and poor rather than middle or affluent by themselves ( OR = 3 . 74; 95%CI , 2 . 55–5 . 49; p<0 . 0001 ) and their peers ( OR = 10 . 57; 95%CI , 6 . 42–17 . 41; p<0 . 0001 ) compared to the other households in their villages ( Table 4 ) . Using the 0 to 100 scale ( poorest to richest ) , the mean self-rated scores for cases and controls were 34 . 1 v 49 . 1 ( p<0 . 0001 ) and for peer-rated scores they were 27 . 5 v 50 . 3 ( p<0 . 0001 ) . The association of lower self-rated and peer-rated wealth with trichiasis persisted regardless of age , gender , marital status , and in people with normal visual acuity after adjusting for the matching variables and family education ( Table 4 ) . The asset based socio-economic classification of households was found to be robust and produced similar ranking of households when the overall index was compared with the different subset indexes; the Spearman rank correlation coefficient ranged between 0 . 88 and 0 . 94 . A Spearman rank correlation coefficient between asset index and self-rated wealth index , asset index and peer-rated wealth index , and self and peer-rated wealth indexes were 0 . 58 , 0 . 70 and 0 . 63 , respectively . Trichiasis cases were significantly less likely to participate in household , outdoor , agricultural and leisure activities , even after controlling for the presence of other health problems during the preceding month , ( Table 5 ) . However , the trichiasis cases were slightly more likely to participate in daily labouring and self-employment activities such as selling goods . These associations persisted in multivariable analysis after controlling for self reported health problems during the preceding month , except for leisure activities . In stratified analyses by vision , trichiasis cases with normal vision were significantly less likely to participate in processing of agricultural products and in productive outdoor activities such as fetching wood and travelling compared to controls with normal vision ( Table 5 ) . After adjusting for the matching variables and self reported health problems , trichiasis cases were significantly more likely to report difficulty in performing all productive and leisure activities than the controls: >66% of the cases reported difficulty in all productive activities in contrast to <5% of controls ( Table 6 ) . Similarly , trichiasis cases were significantly more likely to report receiving assistance in doing all productive activities compared to controls . In contrast to other activities , higher proportions of trichiasis cases received assistance particularly in agricultural activities such as farming , animal husbandry and processing agricultural products ( Table 6 ) . In a univariable analysis ( Table 7 ) , being a household head with trichiasis had a strong association with economic poverty ( OR = 3 . 29; 95%CI , 1 . 89–5 . 75; p<0 . 0001 ) while visual impairment had a borderline association ( OR = 1 . 71; 95%CI , 0 . 98–2 . 97; p = 0 . 058 ) . Not having a marriage partner ( OR = 9 . 41; 95%CI , 4 . 16–21 . 31; p<0 . 0001 ) , no family member with formal education ( OR = 4 . 95; 95%CI , 1 . 73–14 . 16; p = 0 . 0028 ) and a main family job of daily labouring ( OR = 19 . 64; 95%CI , 2 . 32–166 . 49; p = 0 . 0063 ) as opposed to farming were independently associated with economic poverty ( Table 7 ) . Families in which there were more people of a productive age were less likely to be poor than their counterparts ( OR = 0 . 32; 95%CI , 0 . 16–0 . 60; p = 0 . 0005 ) ( Table 7 ) . In a multivariable analyses , participating in animal husbandry ( OR = 0 . 05; 95%CI , 0 . 02–0 . 12; p<0 . 0001 ) and agricultural product processing ( OR = 0 . 50; 95%CI , 0 . 27–0 . 91; p = 0·024 ) activities were independently associated with wealthier households while house cleaning ( OR = 2 . 05; 95%CI , 1 . 03–4 . 08; p = 0 . 042 ) and self employment ( OR = 2 . 77; 95%CI , 1 . 25–6 . 18; p = 0 . 012 ) activities were associated with poorer households .
The age distribution , gender profile and literacy status of the trichiasis cases in this study were comparable with those reported in our earlier studies in Ethiopia as well as other studies of trichiasis patients elsewhere in Sub-Saharan Africa [31 , 38–40] . This suggests that the results are probably generalizable for this region of Ethiopia at least . The households of trichiasis cases were significantly less well off than controls in terms of ownership of almost all asset indicators measured . Consistent with the literature , trichiasis cases had significantly smaller and more crowded households [6 , 41] . Cases had less latrine access and more kept their cattle within the house , which is consistent with observations that active trachoma is associated with poor sanitation access [41–43] . These differences reflect a gap in the implementation of the “E” component of the SAFE strategy , which needs on-going emphasis in this region . We have found clear evidence from each measure that even within trachoma-endemic communities individuals and households affected by trichiasis are significantly economically poorer than those that are not . Within endemic communities some individuals or families appear to be more severely affected by the disease and develop sight-threatening complications . This raises the important question of whether the association between poverty and trichiasis arises from a general state of impoverishment or whether there are a number of critical factors that primarily drive the relationship that might be amenable to focused intervention . The data we present here suggest that the relationship between poverty and trachoma could possibly be bidirectional . Poverty may contribute to trachoma . This study provides evidence that even within superficially homogeneous endemic communities relative poverty plays a major part in the vulnerability of families to scarring disease . Firstly , trichiasis cases were more likely than the controls to come from households where the main family job is daily labouring and from families with no or lower formal education . Both of these factors have a major influence on income and health awareness , which in turn increase the vulnerability of the family to trachoma . Consistent with this , studies from Malawi , Tanzania and Ethiopia identified that children from lower socio-economic households had a higher prevalence of active trachoma than their counterparts indicating an association between poverty and active trachoma [10 , 44 , 45] . Secondly , previously described risk factor associations for active trachoma such as crowding and poor access to latrine , characterised the households of the trichiasis cases in this study . Such conditions are believed to promote the transmission of Chlamydia trachomatis within endemic communities , sustaining higher prevalence levels . Poorer households and communities may be less likely to have either the resources or the awareness to access treatment and sustain a sufficiently hygienic environment to control trachoma [8 , 17 , 46 , 47] . Households with higher income were more likely to have a latrine than their counterparts in a study conducted in the same area [48] . Trachoma may also contribute to poverty . Poor health frequently results in loss of productivity through disability and diversion of resources [11] . Trichiasis and its associated visual impairment probably lead to a loss of income , exacerbating pre-existing poverty in a “vicious cycle” [12 , 13] . Previously healthy and productive adults can be rendered dependent on others , unable to work or fully care for themselves due to pain , photophobia or visual impairment [13] . We found clear evidence of reduced activity and participation among trichiasis cases . Trichiasis cases were less likely than the controls to participate in productive household activities , outdoor activities ( shopping/marketing , fetching wood and water ) and agricultural activities ( farming , animal husbandry and processing agricultural products ) . The stratified analysis found trichiasis cases with normal vision are less likely to participate in outdoor and agricultural activities than controls . This is consistent with a study of Tanzanian women with trichiasis without visual impairment , who had a degree of functional limitation which was comparable to those with visual impairment [14] . We found evidence that households with fewer economically productive adults and where the family head had trichiasis tended to be poorer . Conversely , households where trichiasis cases participated in agricultural activities were better off . Even where the trichiasis cases were undertaking specific activities , they reported much more difficulty and greater need for assistance than the controls . Similarly in another study , trichiasis cases reported difficulty in performing day-to-day farming activities [49] . These observations all point towards households with someone with trichiasis being under greater financial strains through reduced income contribution and greater needs and dependence of the person with trichiasis . The burden of disability caused by trachoma has been estimated between 171 , 000 and 1 . 3 million DALYs , with economic losses of 5–8 billion USD/year [4 , 12 , 13] . The economic loss from trichiasis alone due to lost productivity was estimated to be 3 billion USD/year [12 , 13] . This study comprehensively assesses the relationship between trachoma and economic poverty using four different measures , with a robust process to select suitable community controls . The asset index quantifies the long-term economic welfare of trachoma affected communities , which is important as trachoma and its sequelae are probably related to long-term SES [19 , 20] . The asset index has the practical advantage that it is much less affected by recall or measurement bias during data collection [19] . Most of the housing characteristics , utilities and durable assets were collected through direct observation minimising miss-measurement . Broad ranges of asset data were collected increasing the power of the study in the following ways . Clumping and truncation , potential problems that can arise with PCA of asset data and compromise its suitability for defining socio-economic strata , did not occur when all asset indices were combined into a single index . This indicates that the data from this study is sufficient to measure economic status and effectively infer inequality between different socio-economic strata and that in this region assessment of economic status by asset measurement requires a wider pool of parameters , particularly including agricultural assets . Encouragingly , the asset based poverty measure was moderately and strongly correlated with the self-rated and peer-rated wealth measures . Poverty is a complex multidimensional problem with many causes and manifestations . Therefore there are many ways in which poverty can be measured . Here we only examined the economic aspect using relative measures such as low asset ownership . We use the first principal component ( PC1 ) to measure socio-economic status . However , there is no clear description of the number of principal components to use and often the factor scores derived from the other principal components are difficult to interpret [27] . Despite the comparability of the amount of variance explained by PC1 with other studies , there is uncertainty whether the first component alone sufficiently explains all the pertinent variation . Asset scores are usually developed to be locally relevant , to allow ranking of people within the same community with respect to poverty . Unfortunately , socio-economic classifications based on asset ownership quintiles measure relative poverty within a given context and face the limitation of lacking international comparability . Therefore , between region or country comparison of SES should be done with caution [28] . We did not collect consumption or expenditure data , and so were not able to assess absolute poverty levels . Although a community based screening method was used to identify trichiasis cases , it is possible that some cases might have been missed , which could potentially introduce non-response bias . Similarly , it is possible that some potential controls were not listed by the sub-village administrators . Self and peer-rated wealth are subjective measures , which might have suffered from the tendency to favour ranking households in the middle of the distribution . The activity participation data relied on the participant’s recall ability on what s/he had done in the last week . Finally , our results suggest that a bidirectional relationship may possibly exist between trachoma and poverty . However , the authors recognise that inference about causality is speculative as it is not possible to draw firm conclusions from a cross-sectional observational study such as this . In this study we found a clear association between trichiasis and household economic poverty by all three economic measures . Trichiasis cases were more likely to have economically poor households and less likely to participate in productive activities regardless of visual impairment , more likely to report difficulty in performing productive activities and more likely to need assistance in performing activities than controls . These suggest that the causative relationships between poverty and trachoma may possibly involve bidirectional interaction: poor households are more affected by trachoma and the scarring sequelae of trachoma and trichiasis reduces productivity even prior to the development of visual impairment , which might lead to additional poverty . These data are anticipated to be useful in advocacy and to support programme leaders and funders to secure resources to promote trachoma prevention linked to socio-economic development in trachoma-endemic communities . Implementation of the full SAFE strategy in the context of general improvements might lead to a virtuous cycle of improving health and wealth . Trachoma is a good proxy of inequality within communities and it could be used to target and evaluate interventions for health and poverty alleviation . Measuring the effect of trichiasis surgery on household economic poverty through longitudinal studies would provide an indication of the relative contribution of trichiasis to poverty , as improved health potentially leads to improved productivity and income . | Trachoma has long been considered a disease of poverty . However , there is surprisingly little direct data that formally quantifies the relationship between trachoma and economic poverty , and none that specifically focuses on trichiasis . We compared 200 people with trachomatous trichiasis ( TT ) to 200 people ( controls ) without the condition , who were matched on age and sex , living in the same community , in Amhara Region , Ethiopia . We measured household relative poverty using three measures: household assets , self-rated wealth and peer-rated wealth . We also measured activity participation . We found TT case households were poorer by all relative economic measures . We found cases less likely to participate in economically productive activities regardless of visual impairment and other health problems , more likely to report difficulty and need assistance performing activities . The results suggest that the causative relationship between poverty and trachoma could possibly be bidirectional: poor households are more affected by trachoma and trichiasis reduces productivity even prior to development of visual impairment , which may exacerbate poverty . Implementation of the SAFE strategy in the context of general socioeconomic improvements might lead to a virtuous cycle of improving health and wealth . Trachoma could be used as proxy of inequality and to target and evaluate interventions for health and poverty alleviation . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Trachoma and Relative Poverty: A Case-Control Study |
Methods for improving microbial strains for metabolite production remain the subject of constant research . Traditionally , metabolic tuning has been mostly limited to knockouts or overexpression of pathway genes and regulators . In this paper , we establish a new method to control metabolism by inducing optimally tuned time-oscillations in the levels of selected clusters of enzymes , as an alternative strategy to increase the production of a desired metabolite . Using an established kinetic model of the central carbon metabolism of Escherichia coli , we formulate this concept as a dynamic optimization problem over an extended , but finite time horizon . Total production of a metabolite of interest ( in this case , phosphoenolpyruvate , PEP ) is established as the objective function and time-varying concentrations of the cellular enzymes are used as decision variables . We observe that by varying , in an optimal fashion , levels of key enzymes in time , PEP production increases significantly compared to the unoptimized system . We demonstrate that oscillations can improve metabolic output in experimentally feasible synthetic circuits .
A central goal of synthetic biology is to create new tools and strategies to improve production of metabolites , chemicals and proteins from microbial sources . A particular focus of this field has been the advancement of genetic constructs to exquisitely control gene expression in cells . Traditionally , researchers have modified microbial production strains in a variety of ways that include gene knockouts , gene overexpression , and heterologous pathway expression [1]–[3] . With expanding availability of genome-wide datasets and large metabolic models , emphasis has shifted from single-gene manipulations to genome-wide alterations to improve microbial production processes [4]–[7] . Many of these techniques allow entire clusters of genes to be manipulated simultaneously and are designed to fine tune metabolic regulators for optimal production of a desired product . In this study , we propose a novel method to improve metabolic production of a desired product that relies on time-periodic oscillations of cellular enzymes . Natural oscillations have been observed in many biological systems [8]–[10] . Early studies have established that oscillations within cellular circuitry can have profound impact on the behavior of a culture [11] . In E . coli , oscillations have been studied both experimentally and computationally . Experimentally , glycolytic oscillations have traditionally been generated in response to periodic control of the feed source or external stresses [12]–[16] . Likewise , Chassagnole et al . demonstrated that oscillations observed experimentally could be described using a kinetic model of central carbon metabolism [13] , [17] . Additionally , several theoretical studies have shown that oscillations in enzyme levels could increase intracellular concentrations of metabolites in simplified biological circuits [18] , [19] . We consider oscillating enzymes within the context of E . coli metabolism and suggest ways that these ideas could be implemented experimentally . Computationally , our principal contribution is the use of dynamic optimization to tune these oscillatory responses in a way that maximizes production of a desired metabolite . This contrasts the exclusive use of parameter sensitivity to make control decisions , as used in earlier works [18] , [19] . In this work , we explore optimized time-periodic oscillations of a subset of enzymes within a metabolic pathway as a strategy to increase metabolite production . Our hypothesis is based on well-established results from the operation of chemical reactors . This literature shows that the amount of product generated over time can be increased by operating reactors in a non-steady state , time-periodic regime [20] , [21] . As a prototype system , we use a modified version of a previously published kinetic model of E . coli central carbon metabolism ( Figure 1 ) [17] . Within this system , we explore the use of oscillations in enzyme levels to increase intracellular levels of a key metabolite , phosphoenolpyruvate ( PEP ) . PEP is an important metabolite both in cellular physiology and a key precursor for industrially-important compounds [22] . Specifically , PEP levels control the flow of glucose into the cell and allosterically regulate enzymes within central carbon metabolism [23] , [24] . In addition to its cellular functions , PEP is a limiting precursor for microbial production of aromatic amino acids [25] , which are important building blocks for products in the chemical , pharmaceutical , and food industries [22] , [26] . In light of these facts , many metabolic engineering strategies have been employed to improve PEP availability for aromatic amino acid production [23] , [27]–[29] . These studies provide valuable insights into the changes in levels of metabolites , like PEP , that can be achieved through genetic modification . In this study , we expand an established model of E . coli central carbon metabolism [17] , [30] , [31] by adding two gluconeogenesis reactions from another experimentally-validated model [32] and incorporate enzyme levels into the model using methods described previously [33] , [34] . Addition of the phosphoenolpyruvate synthase ( PPS ) and fructose-1 , 6-bisphosphatase ( FBP ) reactions is motivated by previous research results that suggest they are important controllers of PEP levels [25] , [35] . Following the incorporation of these reactions , we used sensitivity analysis to identify several key enzymes that impact production of PEP . The levels of these enzymes were then assumed to vary in a periodic fashion , and were modeled as cosine waves whose amplitude , period , and phase properties were optimized to maximize metabolite production ( Methods , Optimizations ) . Finally , we explored the oscillatory properties of an experimentally feasible small synthetic circuit to increase PEP production .
A key aspect of this study was to identify appropriate enzyme candidates that could be periodically varied in time to increase PEP production . Enzyme levels are incorporated into our model as an additional term to the reaction rate of each enzyme ( Methods , Kinetic Model ) . We assumed that the enzyme levels in the original model were at steady state and arbitrarily set their values to one . As a result , changes in enzyme levels are defined as deviations from this nominal value . To identify enzymes that are important for PEP production , we conducted a sensitivity analysis of the system using step tests . These consisted of monitoring the evolution of PEP levels in time after increasing each enzyme level to be 50% greater than its steady state value ( 1 to 1 . 5 ) . Based on this analysis ( Figure 2 ) , nine enzymes appeared to affect PEP levels: phosphofructokinase ( PFK ) , glyceraldehyde phosphate dehydrogenase ( GAPDH ) , pyruvate kinase ( PK ) , phosphoenolpyruvate carboxylase ( PEPC ) , ribose-phosphate pyrophosphokinase ( RPPK ) , serine synthesis ( SER ) , synthesis 1 ( SYN1 ) , 2-Dehydro-3-deoxyphosphoheptonate aldolase ( DAHPS ) , and glucose-6-phosphate dehydrogenase ( G6PDH ) . Sensitivity analysis indicated that increasing the levels of PFK or GAPDH led to increased PEP levels . This result is somewhat expected , since these two enzymes represent critical branch points for simultaneously controlling flux down the main glycolysis pathway and flux returning from the pentose phosphate pathway . Individually increasing levels of the remaining seven enzymes resulted in decreased levels of PEP . Four of these enzymes ( PEPC , SYN1 , DAHPS , and PK ) catalyze reactions that directly consume PEP at the PEP node ( Figure 1 ) . Two of these enzymes , SER and RPPK , catalyze reactions that are located upstream of the PEP node and direct flux out of the model . Finally , G6PDH is one of the main factors that determines whether the metabolic flux follows glycolysis or the pentose phosphate pathway [17] . The latter three enzymes likely reduce PEP production by playing a more general role in directing metabolic flux out of the boundaries of the system . Practical implementation of enzyme oscillations ( further discussed below ) could be achieved through heterologous expression of an enzyme from a plasmid source . In this scenario , it is typical that enzyme concentrations would reach much higher levels than natural expression levels in the cell . To represent the high amount of expression that can be obtained from inducing strong promoters [36] , we allowed enzyme levels to reach 20 times their nominal levels in our simulations . With this assumption in place , we first explored independent oscillations of each of the nine PEP-influencing enzymes . As shown in Figure 3 , a wide range of increases in total PEP levels ( 1%–28 . 3% relative to the case of no oscillation ) is observed across the nine enzymes identified as sensitive . While individual enzyme oscillations show smaller improvements than simulated knockouts or over expressions ( Table S1 ) , these results showed the potential of oscillating intracellular enzyme levels to positively affect PEP levels compared to the unoptimized case ( Figure 3 ) . After examining the benefits of individual oscillations relative to the base case of no oscillation , we reasoned that the oscillating multiple metabolic enzymes simultaneously could further push PEP gains . We next analyzed the effect of combined oscillations in our system . We grouped all nine sensitive enzymes together into a “PEP-influencing cluster” and focused on optimizing collective expression of this cluster as a unit to mimic natural systems , where large groups of enzymes are co-regulated to produce a specific phenotype [37] . This PEP-influencing cluster represents the theoretical maximum number of enzymes that we hypothesized would significantly influence PEP levels . We modeled oscillations of all the enzymes in the PEP-influencing cluster using simple cosine forcing functions ( Methods , Dynamic Optimization ) . Then , we optimized properties of the waves ( i . e . amplitude , frequency and phase ) for maximum PEP production . Levels of each enzyme were independently optimized to maximize PEP levels given constraints on metabolite levels derived from experiments ( Methods , Dynamic Optimization ) [38]–[40] . We observed that oscillations of this nine enzyme cluster ( Figure 4A ) , caused PEP levels to oscillate ( Figure 4B ) . These PEP oscillations resulted in a 2 . 2-fold increase in total PEP levels over an 8 hour time horizon relative to the non-oscillating unoptimized system , suggesting that regulating multiple enzymes in a periodic fashion needed to be further explored as a metabolic optimization strategy . We also compared the oscillatory strategy to a time invariant optimization of the levels of the nine enzymes ( Figure 4B , Table S2 ) . The time invariant optimization calculates the PEP gain possible by fine tuning enzyme levels at constant levels . We were encouraged to see that oscillating enzymes produced more PEP than the unoptimized case ( Figure 4B ) . It is important to point out that we constrained PEP levels in our simulations to a 10-fold range ( 1 mM–10 mM ) similar to the total variability of PEP concentrations reported in the literature [40] , [41] . We have also verified that describing the variation of enzyme levels in terms of square waves in the optimization calculation leads to a similar increase in the concentration of the PEP ( data not shown ) . Our optimization revealed that PEP is able to reach higher levels by causing a key enzyme that removes PEP from the system ( PEPC ) to simultaneously be at low levels while enzymes that help to produce PEP ( GAPDH and PFK ) are at high intracellular levels . To quantify the relationship between changes in enzyme levels and changes in PEP levels , we calculated correlation coefficients for the time series data corresponding to enzyme levels in the PEP-influencing cluster ( Figure 4C ) . Correlation coefficients indicate how closely changes in one variable ( a given enzyme level ) correlate to changes in another variable ( PEP levels ) . PFK , GAPDH , PEPC , G6PDH , RPPK and DAHPS form a highly-correlated , synchronized group of enzymes that is primarily responsible for the changes in PEP levels . PK and SER form a secondary enzyme group that also tunes PEP levels ( Figure 4C ) . Up to this point we had considered the theoretical maximum number of enzymes that could be in our oscillatory circuit to qualitatively evaluate the potential of synthesizing multi-enzyme clusters to improve PEP production . However , given the experimental convenience of manipulating a smaller number of genes , we tested the impact of constructing an smaller oscillatory circuit . To select the enzymes in this circuit , we analyzed all combinations ( of enzymes within the nine enzyme cluster ) of two-enzyme clusters to understand which combinations positively impacted PEP the most ( data not shown ) . We did not consider all 32 enzymes for this analysis given the weak influence of most enzymes on PEP levels . We confirmed that individual enzymes that resulted in the largest PEP gains when oscillated independently ( i . e . GAPDH , PEPC , or PFK ) , also produced the largest PEP gains when oscillated with a second enzyme . In particular we noted the largest PEP gain from oscillating GAPDH and PFK simultaneously . To this cluster we added a third enzyme , RPPK ( which is essential for cellular viability ) , to test for additional PEP gains from oscillations that would be difficult to obtain through traditional methods . We optimized periodic expression of RPPK , GAPDH , and PFK by simulating these genes in a recently described light-inducible system [42] ( Figure 5A ) . In this circuit , the bacterial two-component system , YF1 ( histidine kinase repressed by light ) /FixJ ( response regulator ) , represses the expression of transcripts from the FixK2 promoter . A second repressor protein , lambda phage cl , is expressed from the FixK2 promoter which represses the lambda phage promoter pR . A reversible physical input ( i . e . light ) simultaneously represses production of genes controlled by FixK2 , and cause pR promoter genes to be expressed since the promoter is no longer being repressed . Importantly , in the presence of light , GAPDH and PFK genes are expressed and RPPK is suppressed simultaneously . The optimized circuit ( Figure 5B ) showed a significant increase in PEP levels ( 1 . 86-fold increase in total PEP concentration ) relative to the levels observed in the unoptimized case with no oscillations ( Figure 5C ) . Although oscillating the cluster produced less PEP than time-invariant optimization of the three enzyme levels , we were encouraged by the fact that the oscillating three-enzyme circuit produced far more PEP than the unoptimized case and 85% as much PEP as the oscillating nine-enzyme circuit ( Figure 5C , Table S3 ) . This data validated the potential of selecting influential enzyme clusters and of periodically oscillating them as a way to increase targeted production of a metabolite of interest .
We have shown that tuned periodic oscillations of selected enzyme levels in a metabolic pathway can have a positive effect on metabolite production . These findings agree with observations made in chemical reactors that a higher cumulative yield of product can be reached by operating the reactor in a periodic fashion [20] , [21] , [43] . In this study , we report an experimentally viable three enzyme oscillating cluster that can lead to a 1 . 86-fold increase in PEP production compared to the original ( unoptimized ) system . The motivation of this work was to evaluate the tradeoff between drastic alterations in gene expression and more moderate metabolic changes ( i . e . oscillations ) that can lessen metabolic burden and tune essential genes . A key question was how the levels of PEP increase obtained by periodic enzyme oscillations compare to traditional strategies of genetic deletions and enzyme overexpression . We expected that constitutive overexpression of enzymes over time ( where levels are always at a maximum ) would lead to higher PEP levels than the periodic oscillation cases ( where those maximum enzyme levels are only periodically achieved ) . On average an increase of 32% in PEP levels was observed by deleting individual genes that were negatively correlated with PEP production , relative to the oscillation of these same individual genes . A similar trend was discovered when individual enzymes that positively correlated with PEP production were constitutively overexpressed relative to when they were independently oscillated . To gauge the accuracy of our projections we compared the results of our simulated knockouts and overexpression to experimental data on these modifications [2] , [29] , [44]–[46] . This comparison shows that , while qualitatively correct , our model is significantly underestimating the metabolite concentration increases gained by mutant strains ( i . e . our model projects a PEPC ( ppc ) mutant to have a 1 . 67-fold increase in PEP concentration , but experimental data shows that the knockout produces a 3-fold increase [40] ) . We suspect that , likewise , our model underestimates the gains in PEP levels obtained from oscillatory simulations . We believe that oscillatory strategies could prove valuable for several reasons . First , oscillations provide an additional way to manipulate expression of essential genes . Second , this approach can reduce the metabolic burden in cells that is observed as a result of constitutive overexpression of multiple proteins . Although our model cannot capture these effects , it has been well established that consistently overexpressed proteins can become a large metabolic burden on the cell representing as much as 15–40% of the total cellular protein produced by recombinant cells [47]–[49] . Third , oscillations could be valuable in situations where there are growth tradeoffs in producing the final product . For example , if the final product is toxic to cell growth [50] or causes cells to form spores [51] , then enzyme oscillations could allow cells some recovery time and prolong their viability . Finally , oscillatory strategies could also be valuable if recombinant enzymes are oscillated synchronously with natural periodic rhythms found in many cells types [52]–[54] . We also envision using oscillatory strategies to tune global regulatory genes resulting in the simultaneous coordination of many genes and pathways . For these reasons , a dynamic strategy can provide a complementary approach to current methods depending on the particular metabolic optimization problem . The method outlined here is generally applicable to any organism that is genetically pliable and for which a kinetic model can be constructed . Oscillating enzyme levels inside cells requires ( 1 ) a method to induce periodic changes in vivo and ( 2 ) the ability to create and manipulate regulatory clusters . Since the creation of the Goodwin oscillator [55] in the 1960s , researchers have been creating more robust and sophisticated synthetic oscillators [56] . The Repressilator , is another created by Elowitz and Leibler , is a good example of a synthetic oscillator , where each of the three genes inhibit transcription of its successor and cause sustained oscillations to form [57] . These simple oscillators have led to the development of a fast , robust tunable synthetic oscillator inside living cells [56] . In addition to these oscillators , light inducible systems have shown the potential to modulate gene expression in a highly controllable fashion [58] , [59] . While many synthetic oscillatory systems have been rigorously tested computationally and validated experimentally , their incorporation into larger regulatory circuitry has been less explored . The number of innovations in the engineering of synthetic circuits to control gene expression in vivo is expected to continue to rise . Manipulation of regulatory clusters , whether naturally occurring or rationally designed , has already proven to be an effective method to improve metabolite production [4] , [27] . Changes in system and flux profiles can be achieved by altering global regulatory systems , including methods such as knocking out transcriptional regulators [60] , tuning promoters [5] , and altering post transcriptional regulatory systems [4] . Some of these methods have already been applied to increase carbon flow through the PEP node . For instance , by manipulating the Carbon Storage Regulator system using overexpression and knockouts , intracellular PEP levels can be increased 2 to 3-fold [27] , [61] . Oscillating components of regulatory circuits , like parts of the Carbon Storage Regulator system , provides the potential additional advantage of bypassing the negative impact of multiple gene deletions and/or gene overexpression on cell growth that has been widely reported in the literature . Oscillating enzymes levels could be a useful strategy for improving production of metabolites in conjunction with traditional methods . Oscillations would be ideal when controlling the levels of genes that hinder or completely impair cellular growth , including many genes in central carbon metabolism [62] . These oscillatory clusters can tune and coordinate expression of multiple cellular enzymes . It is important to note that oscillations of individual proteins could be further customized by tuning the promoters , altering ribosome binding sites , and using different protein degradation tags to change the rate of degradation for each protein . Exploiting customized regulatory “parts” to creatively control gene expression has become common in the construction of synthetic genetic circuits [63] . While initial synthetic oscillatory circuits may take the simpler form similar to the RPPK-GAPDH-PFK circuit , our future research will also consider larger circuits that tune multiple genes directly from the chromosome .
The kinetic model used in this study is an adaptation of the model created by Chassagnole et al . [17] ( Figure 1 ) . The metabolite concentrations are modeled by dynamic mass balance equations , resulting in a set of ordinary differential equations of the form: ( 1 ) where S is the n x m dimensional stoichiometric matrix , r is an m-dimensional vector of reaction rates , C is an n-dimensional vector of metabolite concentrations , P is a k-dimensional parameter vector , and e is an m-dimensional vector of enzyme levels . The second term is a dilution factor that accounts for biomass generation demands on that metabolite ( μ = specific growth rate ) . One of the drawbacks of the original model [17] is that the accumulation of co-metabolites ( ATP , NAD , etc . ) is expressed as explicit time dependent functions rather than as mass balance equations . Since we ran simulations over a significantly longer time frame than originally modeled , we removed the time-dependent descriptions of the co-metabolites and replaced them with constant values following the approach taken in other studies [32] , [34] . Enzyme levels were added to the originally published model [17] by introducing an additional term , , into the reaction rate ( renz ) equations: ( 2 ) Where rmax is the maximum reaction rate , ê is the steady state enzyme concentration ( arbitrarily set to equal one ) and e is the current enzyme concentration , such that enzyme levels ( ) are observed as deviations from their steady state values similar to previous studies [33] , [34] . The model contains mass balance equations for 18 metabolites in the glycolysis , gluconeogenesis , and pentose phosphate pathways . It contains 32 reactions , including two gluconeogenic reactions catalyzed by phosphoenolpyruvate synthase ( PPS ) and fructose-1 , 6-bisphosphatase ( FBP ) , which were taken from a previous study [32] . The parameters for these equations were taken directly from the aforementioned work of Usuda et al . [32] with the exception of the rmax values which were recalculated using the top-down approach previously described by Rizzi et al . [64] . A table of all parameters , equations and modified mass balance equations is provided in the supplementary materials ( Text S1 ) . A sensitivity analysis of enzyme levels on PEP levels was carried out by performing a battery of step tests on the dynamic model . Specifically , with the model initially at its nominal steady state , the concentration of each enzyme was increased to 1 . 5 times its steady state value; the model was subsequently simulated for a sufficiently long period of time for a new steady state to be reached . The subset of enzymes that led to more than a 2% change in PEP levels became the targets of the optimization . For the optimization calculations , it was assumed that the levels of each key enzyme vary in time following a simple sinusoidal function , ( 3 ) Where A is the amplitude , t is time , ω is the frequency , φ is the phase , and h is a constant bias ( h = 1 ) . gPROMs [65] was used to determine the optimal values of these parameters . The optimization was formulated as:Subject to: Process model equations ( 4 ) Where Cmetabolite represents the concentration of metabolites in millimoles/liter , ω in radians/second , φ in radians . Constraints on the amplitudes of the enzyme level waves allow enzymes to have a maximum of 20 times their steady state concentration ( Lower bound: 1/20th ) . We assumed that enzymes that improve PEP production ( GAPDH and PFK ) would be added to the oscillatory circuit on a plasmid without modifying the genomic copies of the gene . To model this assumption , the levels of GAPDH and PFK were not allowed to pass below their original steady state levels of 1 . 0 . The constraints were within the fold changes in enzyme levels described in experimental studies [39] , [66] . Constraints for metabolite concentrations ( i . e . Cpyr , Cpep , Crib5p , Ce4p , and Cg6p ) were set using the fold change observed in experimental studies as constraints [38] , [40] . These studies were chosen because they consider large scale E . coli responses to perturbations and metabolic profile changes that could be achieved by manipulating artificial regulatory systems . As a control to compare the performance of the dynamic optimizations , time invariant optimizations were also run by replacing the time dependent enzyme level descriptions ( Equation 3 ) with a constant term for the enzyme level ( h ) . These constant levels were then used as the optimization variables to maximize PEP production . In these optimizations , all enzymes were allowed to vary between 0 and 20 except for essential enzymes which were allowed to vary between 0 . 25 and 20 . With these exceptions all other constraints and conditions controlling the dynamic optimization were applied to the static optimizations . Correlation coefficients for the time series corresponding to enzyme and metabolite levels were calculated in MATLAB based on the Pearson product moment correlation coefficient formula: ( 5 ) Where x is a matrix of the time-dependent enzyme levels of enzyme i and y is a matrix of the time-dependent enzyme levels of enzyme j . The resulting matrix was then colored using a grayscale to indicate the highest and lowest correlation values . A correlation coefficient close to one indicates a strong correlation . | Until recently , engineering of microbial strains has been heavily focused on removing or overexpressing individual genes . Although valuable , these efforts do not take into account the potential benefits of tuning enzymes and , thus , do not exploit the full diversity of available synthetic tools to regulate gene expression . Tuning enzyme levels is a key factor in gene expression because optimal levels for an enzyme may change over time in living systems . In this study , we use a kinetic model of Escherichia coli to explore how to increase metabolite levels by oscillating sets of enzymes over time . We discover that oscillating relevant clusters of enzymes can increase metabolite levels . When combined with recent experimental studies that demonstrate the ability to build synthetic oscillators and regulators inside living cells , suggesting that oscillating enzymes levels could be of practical relevance to metabolic engineering . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biotechnology",
"bioengineering",
"biological",
"systems",
"engineering",
"computer",
"and",
"information",
"sciences",
"network",
"analysis",
"engineering",
"and",
"technology",
"synthetic",
"biology",
"biology",
"and",
"life",
"sciences",
"metabolic",
"networks",
"regu... | 2014 | Optimizing Metabolite Production Using Periodic Oscillations |
Although genetic and non-genetic studies in mouse and human implicate the CD40 pathway in rheumatoid arthritis ( RA ) , there are no approved drugs that inhibit CD40 signaling for clinical care in RA or any other disease . Here , we sought to understand the biological consequences of a CD40 risk variant in RA discovered by a previous genome-wide association study ( GWAS ) and to perform a high-throughput drug screen for modulators of CD40 signaling based on human genetic findings . First , we fine-map the CD40 risk locus in 7 , 222 seropositive RA patients and 15 , 870 controls , together with deep sequencing of CD40 coding exons in 500 RA cases and 650 controls , to identify a single SNP that explains the entire signal of association ( rs4810485 , P = 1 . 4×10−9 ) . Second , we demonstrate that subjects homozygous for the RA risk allele have ∼33% more CD40 on the surface of primary human CD19+ B lymphocytes than subjects homozygous for the non-risk allele ( P = 10−9 ) , a finding corroborated by expression quantitative trait loci ( eQTL ) analysis in peripheral blood mononuclear cells from 1 , 469 healthy control individuals . Third , we use retroviral shRNA infection to perturb the amount of CD40 on the surface of a human B lymphocyte cell line ( BL2 ) and observe a direct correlation between amount of CD40 protein and phosphorylation of RelA ( p65 ) , a subunit of the NF-κB transcription factor . Finally , we develop a high-throughput NF-κB luciferase reporter assay in BL2 cells activated with trimerized CD40 ligand ( tCD40L ) and conduct an HTS of 1 , 982 chemical compounds and FDA–approved drugs . After a series of counter-screens and testing in primary human CD19+ B cells , we identify 2 novel chemical inhibitors not previously implicated in inflammation or CD40-mediated NF-κB signaling . Our study demonstrates proof-of-concept that human genetics can be used to guide the development of phenotype-based , high-throughput small-molecule screens to identify potential novel therapies in complex traits such as RA .
Rheumatoid arthritis ( RA ) is a common autoimmune disease for which there is no known cure . A diverse number of biological pathways are altered in patients with RA , which impinge on a wide-variety of cell types , tissue types and organ systems – innate immune cells ( e . g . , neutrophils , dendritic cells , mast cells , platelets ) , adaptive immune cells ( e . g . , B and T cells ) , bone , cartilage , synovial fibroblasts , vascular cells , brain , muscle , and fat [1] . Accordingly , the task of sorting through which biological pathways cause disease , as compared to those pathways that are simply a consequence of disease , is a daunting challenge . Without knowing the critical causal pathways , it is very difficult to develop novel therapeutics to treat or cure RA . There are fundamental principles of human genetics that make it a promising strategy to identify critical biological pathways and novel therapeutic targets in complex traits such as RA [2] . Since risk alleles are randomly assigned at meiosis , are independent of non-genetic confounding , and are unmodified by the disease itself , human genetics can help distinguish between cause and consequence . Moreover , risk alleles indicate if a pathway is up or down regulated in disease – a critical first step in drug development . Risk alleles help calibrate the amount of target modulation that is tolerable in humans , as gain-of-function and loss-of-function mutations in the same gene can be assessed for clinical phenotypes in carriers of these mutations . Consistent with these concepts , known drug targets that are safe and effective in humans appear on the list of genes identified by genome-wide association studies ( GWAS ) of common diseases [3] , which suggests that other GWAS hits represent targets worthy of further investigation [4] . However , there are important challenges in translating SNP associations from human genetics ( and GWAS in particular ) to novel therapeutics . First , the causal gene must be identified within the risk locus , as there are often multiple genes in the region of linkage disequilibrium . Compounding this challenge , most GWAS hits are to non-coding variants that cannot pinpoint specific genes . Second , the risk allele must be experimentally validated as gain- or loss-of-function in a relevant human tissue , in order to guide whether a drug should inhibit or activate ( respectively ) the target of interest . Third , the biology of the risk allele should be recapitulated in an assay system suitable for a high-throughput screen ( HTS ) . And fourth , the HTS should demonstrate performance characteristics that make it robust for screening large chemical libraries . The CD40-CD40L pathway represents a good example of a pathway for which human genetics may help guide drug development . The pathway is upregulated in multiple diseases [5]–[7] , including autoimmune diseases such as RA [8]–[15] . GWAS identified a common variant in the CD40 locus that increases risk of RA , which suggests that CD40 upregulation is a cause rather than a consequence of chronic inflammation [16] . Loss-of-function mutations in both CD40 and CD40L result in immunodeficiency , but only in the homozygous state , indicating that 50% inhibition of CD40-CD40L signaling ( as observed in heterozygous mutation carriers ) should be safely tolerated in humans [17] . Despite these findings , there are currently no approved drugs that inhibit CD40-CD40L signaling , and there are no drugs in clinical trials ( www . clinicaltrials . gov ) . Others have developed small molecules that disrupt CD40-CD40L binding , but these compounds have not been tested in humans [18] . Antibodies to CD40L were effective in treating inflammatory diseases , but resulted in thrombotic events due to the presence of CD40L on platelets [19]–[28] . Human genetics suggests that inhibiting intracellular signaling of CD40-mediated signaling will also likely be effective in humans , without adverse events related to thrombosis . Here , we demonstrate a strategy that uses findings from GWAS to guide the development of a drug screen for the identification of small molecules inhibiting the CD40-CD40L intracellular signaling pathway . We hypothesize that such molecules might in turn be safe and effective in treating inflammation observed in RA patients . We set out ( 1 ) to investigate the biology of the CD40 risk allele , by fine-mapping the CD40 locus and analyzing the function of the risk allele in primary B cells from healthy donors; ( 2 ) to recapitulate the biology of the CD40 risk allele in an assay system suitable for a high-throughput screen ( HTS ) of small molecule drugs; and ( 3 ) to conduct a pilot HTS to search for known and novel inhibitors of CD40-mediated signaling in human B cells .
We first performed comprehensive genotyping at the CD40 risk locus to fine-map the causal allele . Additional details can be found in Text S1 . We used a dataset in which RA case-control samples were genotyped at high density across the CD40 locus with the Illumina Immunochip platform ( Eyre et al . Nat Genet . in press ) . In total , we analyzed 492 SNPs in 7 , 222 seropositive RA patients and 15 , 870 controls ( Table S1 ) . As shown in Figure 1A , our analysis revealed that the strongest signal of association was shared by 2 “equivalent” non-coding SNPs in linkage disequilibrium ( LD ) at r2 = 0 . 98 , rs6032662 and rs4810485 ( P = 1 . 4×10−9 , OR = 1 . 17 per copy of the risk allele ) . After conditional analysis , no additional signal of association remained , indicating that one of these 2 SNPs , or one of the other six equivalent non-coding SNPs in LD at r2>0 . 80 , is the underlying causal allele at this locus ( Figure S1A , Table S2A ) . We refer to rs4810485 as the index SNP; the major G allele is the RA risk allele and the minor T allele is the non-risk allele . A previous study showed that a perfect proxy of the CD40 RA risk allele increases CD40 protein on the surface of B lymphocyte cells from 23 healthy individuals [29] . To confirm and extend these findings , we developed a flow cytometry assay to measure CD40 on the surface of primary human CD19+ B cells . We demonstrated high reproducibility of the assay on blood samples drawn from the same individual >3 months apart ( r2 = 0 . 76 , Figure S2 ) . We measured CD40 protein levels from 90 healthy control subjects . We performed high-density SNP genotyping across the CD40 locus , using the same genotyping array as in our case-control study of RA risk . Strikingly , the strongest signal of association across the CD40 locus was at the RA risk allele ( Figure 1B; P = 3×10−9 , Table S2 ) . After conditional analysis , no additional SNP was significant ( Figure S1B ) . Healthy control subjects homozygous for RA risk allele have ∼33% more CD40 on the surface of primary human CD19+ B lymphocytes than subjects homozygous for the non-risk allele ( Figure 1C ) . The RA risk allele ( the G allele of rs4810485 ) explains 31% of variation observed in CD40 protein level in these healthy control subjects , and was the strongest signal among the genome-wide set of SNPs tested for association with CD40 protein levels ( Figure S1C ) . To complement this finding , we examined CD40 gene expression in peripheral blood mononuclear cells of 1 , 469 unrelated individuals [30] . As shown in Figure 1D , we found that the RA risk allele was an expression quantitative trait locus ( eQTL ) on CD40 gene expression ( P = 8 . 2×10−13 ) . Similar findings have been reported for the RA risk allele in other immune cell types [31] , [32] . Taken together , our data demonstrate unequivocally that the RA risk allele ( the G allele of rs4810485 ) is a gain-of-function mutation that leads to increased level of CD40 on the surface of primary human CD19+ B cells ( and possibly other immune lineages within PBMC's ) . We next sought to determine the biological consequences of having more CD40 on the surface of B cells , in order to determine the most appropriate assay for a drug screen . Engagement of CD40 by its trimerized ligand ( tCD40L ) leads to phosphorylation of p65 , a subunit of NF-κB ( Figure 2A ) . To determine the effect of p65 phosphorylation in a human B cell line ( BL2 ) with varying levels of CD40 protein , we derived clones in which CD40 mRNA was knocked-down with shRNA [33] . In two independent cell lines , we observed CD40 protein levels at 40% and 55% compared to the BL2 parent line , respectively ( Figure 2B ) . We used the parent BL2 line and two BL2/shRNA lines to activate the CD40 signaling pathway with tCD40L . In both BL2/shRNA cell lines , we observed a reproducible decrease in phosphorylation of p65 at Ser536 at 15 and 30 minutes following tCD40L activation . The levels of NF-κB p65 phosphorylation , as measured by Western blot , correlated with the levels of CD40 protein across all three B cell lines ( Figure 2C ) . That is , more CD40 on the surface of B cells ( as is the case for carriers of the RA risk allele ) has increased activation of the classical NF-κB pathway ( as measured by phosphorylation of NF-κB p65 ) . A Western blot is not suitable for a high-throughput screen ( HTS ) . Based on our functional analysis , we developed a luciferase reporter assay that can be used in an HTS to identify inhibitors of CD40 signaling pathway . For this assay , we generated a BL2 cell line ( BL2-NFκB-Luc ) that was transfected with a luciferase reporter construct driven by a basal promoter element ( TATA box ) joined to tandem repeats of the NF-κB response element . To optimize conditions for an HTS , we performed a series of experiments with BL2-NFκB-Luc cells . First , we performed a titration of tCD40L , and found approximately 80% activation at 64 ng/ml tCD40L ( Figure 2D ) . Second , we determined the optimal time course following 64 ng/ml tCD40L activation , and found maximum activation ( 5 . 6-fold induction ) at 4 . 5 hours . Third , we performed a titration of a known inhibitor of the classical NF-κB signaling pathway , IKK inhibitor VII ( Milipore ) ( Figure 2E ) . To confirm that the decrease of NF-κB activity by this inhibitor is not due to cytotoxicity , we used an anti-PARP antibody ( 116-kDa poly-ADP-ribose nuclear polymerase ) to demonstrate by Western blot that the decrease in NF-κB phosphorylation following IKK inhibition was not simply due to cell death ( Figure S3 ) . And fourth , we determined the fold-increase and fold-inhibition of luciferase activity following tCD40L activation and IKK inhibition , respectively . We observed robust performance of our assay under a specific set of conditions ( Figure 2F ) , with a Z'-factor≈0 . 8 [34] ( where a Z'-factor of >0 . 50 is considered appropriate for a small-molecule screen [35] ) . We optimized our luciferase assay in a 384-well format . We conducted a pilot screen of 2 , 240 chemical compounds ( of which 1 , 982 are in PubChem ) , each assayed in duplicate experiments . The chemical compounds comprise bioactive compounds ( including FDA-approved drugs ) , commercially available drug-like molecules , targeted collections ( e . g . , biased for kinases ) , stereochemically-diverse compounds , and purified natural products . Following normalization of luciferase activity to correct for variability across plates , we determined fold-change in luciferase activity relative to our positive ( IKK inhibitor VII ) and neutral ( 0 . 5% DMSO ) controls . The Z'-factors for this pilot screen ranged from 0 . 63–0 . 85 ( average of 0 . 79 ) . We observed strong correlation between the two experimental replicates ( r2 = 0 . 94; Figure 3A ) . We identified 81 compounds ( 4 . 1% of all compounds ) with >2SD decrease in luciferase activity relative to DMSO controls ( which we refer to as preliminary “hit” compounds; Table S3 ) . Even at this liberal threshold , there are more compounds that decrease luciferase activity than would be expected by chance alone ( P = 0 . 006 ) . As further evidence that many of these hit compounds represent true positive findings , we observed enrichment of known anti-inflammatory agents using text-mining of PubChem annotations [36] , [37] ( Figure S5A–S5C ) and enrichment of shared chemical structure pertaining to corticosteroids and their analogs ( Figure S6 and Table S5 ) , which are known NF-κB inhibitors [38] . To confirm inhibition of CD40-mediated NF-κB signaling , we conducted a series of counter-screens using our primary assay ( tCD40L-activated BL2-NFκB-Luc cells ) and another B cell line , Ramos RA-1 , transfected with the same NF-κB luciferase reporter construct ( Ramos-NFκB-Luc ) . In addition to activation with tCD40L , we activated BL2-NFκB-Luc cells with LPS ( which binds to TLR4 and signals through the classical NF-κB pathway ) and Ramos-NFκB-Luc cells with TNF-alpha ( which also signals through the classical NF-κB pathway ) . We measured luciferase activity following exposure with each of the 73 compounds across 8 different concentrations ( stock concentrations were not available for 8 compounds ) . For 20 of the 73 compounds ( ∼1% of all compounds tested ) , we observed consistent , dose-dependent inhibition across all four assays ( BL2-NFκB-Luc cells activated with tCD40 and LPS; Ramos-NFκB-Luc activated with tCD40L and TNF-alpha ) , without evidence of cellular toxicity ( Table S6 ) . To ensure that inhibition was due to effect of the chemical compound and not an experimental artifact , we re-ordered 14 of the compounds that were available commercially , confirmed chemical structure and purity using high performance liquid chromatography and mass spectrometry , and re-tested these compounds using the same BL2- and Ramos-NFκB-Luc assays . We confirmed that several compounds known to inhibit inflammation ( e . g . , indoprofen; PubChem Compound ID [CID] 3718 ) [39] or NF-κB signaling ( e . g . , 3-[ ( 4-methoxyphenoxy ) methyl]benzohydrazide; CID 843208 ) [40] are potent inhibitors in our assays . We also identified corticosteroids ( e . g . , 4-hydroxy-estradiol; CID 5282360 ) and inhibitors of inflammatory arthritis in murine models of RA ( tranilast; CID 5282230 ) [41] , [42] . Representative examples are shown in Figure 3B and Figure S7 . Equally importantly , however , we identified 2 chemical compounds not previously implicated in inflammation , NF-κB signaling or inflammatory arthritis ( Figure 3C and Figure 4 ) . For both , the relative IC50 was <20 µM ( Table 1 ) , with >50% decrease in luciferase activity ( Table 2 ) . Finally , we tested 2 “known” and 2 “novel” compounds for their ability to inhibit tCD40L-mediated NF-κB signaling in primary CD19+ B cells from healthy control subjects . We used flow cytometry to measure CD86 cell-surface protein levels , as CD86 expression is up-regulated upon activation of B cells with CD40L [43] . We observed dose-dependent decrease in CD86 expression with all 4 compounds ( Figure 4 ) . Thus , we confirmed that the 2 known inhibitors and the 2 novel compounds identified in our HTS inhibit the NF-κB signaling pathway in primary CD19+ B cells from healthy donors .
Our study exploits fundamental principles of human genetics to guide a small molecule drug screen in RA . We show that upregulation of the CD40-CD40L signaling pathway is a cause rather than a consequence of disease , as the RA risk allele increases levels of CD40 on the surface of B lymphocyte cells and increases CD40-mediated NF-kB signaling . Based on our genetic findings , a prediction is that drugs that attenuate CD40-mediated NF-kB signaling will either protect from RA or treat symptoms in patients with active disease . From our screen , we identify two compounds that support this hypothesis: tranilast , which reduces inflammation in a mouse model of RA [41] , and a corticosteroid , which is a potent anti-inflammatory drug used to treat RA [44] , [45] . Equally important , we discover 2 novel small-molecules that inhibit CD40 signaling through the classical NF-κB pathway in primary CD19+ B cells . There are few examples where GWAS was used to guide drug discovery . One example is PCSK9 , where a loss-of-function variant is associated with lower levels of LDL cholesterol and protection from cardiovascular disease [46]–[49] . However , the original finding implicating PCSK9 and LDL cholesterol came not from GWAS , but from sequencing in families with autosomal dominant high LDL levels and an increased incidence of coronary heart disease [50] . In 2012 , a randomized control trial was published that a monoclonal antibody to PCSK9 significantly reduced LDL cholesterol levels in healthy volunteers and in subjects with hypercholesterolemia [51] , [52] . Another example is BCL11A and persistence of fetal hemoglobin in sickle cell anemia . In 2008 , a GWAS found an association with a common , non-coding variant of the hemoglobin silencing factor gene , BCL11A , and HbF expression [53] , [54] . Based on these data , together with data from animal models [55] , repressors of BCL11A are under development for the treatment of sickle cell disease [56] . Our study illustrates another example , as we demonstrate that genetic findings can be instrumental in developing optimal high-throughput drug screens . In RA , many biological pathways have been implicated . Consequently , identifying relevant pathways is critical for the development of molecules that will be effective in treating the disease . Our strategy successfully links an RA risk allele to a biological process suitable for an HTS . First , we show unequivocally that the RA risk allele leads to increased levels of CD40 protein on the surface of CD19+ B cells , thereby establishing a causal link between increased CD40 protein levels and risk of RA . Second , we establish a direct relationship between amount of CD40 on the surface of B cells and an intracellular biological pathway , NF-kB signaling . In doing so , we recapitulate the effect of the CD40 risk allele in an assay system suitable for an HTS . There are important limitations of our study . First , the chemical library tested in our study is small relative to libraries in academic centers and industry ( which often contain hundreds of thousands to millions of compounds ) [57] , [58] . Second , our screen did not identify inhibitors specific to CD40 signaling . Whether a more selective CD40 inhibitor would be a better therapeutic than a more general inhibitor requires additional studies . That our HTS identified two “known” drugs that inhibit inflammation reinforces that our general strategy is successful . Third , we have not yet tested our compounds in animal models of RA . However , one of our known compounds , tranilast , has been shown by others to inhibit collagen-induced arthritis in the mouse [41] . Fourth , we do not yet know the target of our “novel” small molecule inhibitors of CD40-mediated NF-κB signaling . One of these compounds ( CID 7309015 ) has been annotated in PubChem as an inhibitor of retinoic acid-related orphan receptor ( ROR ) gamma , a transcription factor that has a central role in the differentiation of CD4+ Th17 cells . However , this PubChem annotation has not yet been linked to a PubMed manuscript . The other compound has not been confirmed as active in any PubChem assay , and therefore represents a novel tool compound for further study . In conclusion , we demonstrate a strategy to translate GWAS findings into HTS to identify novel small-molecule inhibitors of the CD40 signaling pathway . Given the wealth of GWAS data that has accumulated in recent years , human genetics represents a promising approach to develop safe and effective therapies to treat complex human diseases such as RA .
Six case-control collections were included for genotyping using the Illumina Immunochip platform ( Table S1 ) , as part of the RACI consortium [59]; the GWAS datasets have been previously described , and include 4 collections that did not overlap with the Immunochip dataset [60] . All six Immunochip datasets were clustered together using the Illumina Genome Studio algorithm . Initial data filtering steps in GenomeStudio included: removal of samples with call rate<90% and removal of SNPs with poor clustering quality metrics ( call frequency<0 . 98 , cluster separation<0 . 4 ) . Further quality control was performed in the six individual population datasets separately . First , samples with call rate <99% were excluded . Second , SNPs with call rate <99% in either the RA cases or controls were excluded . To address population stratification , we selected a set of common SNPs ( MAF>5% ) , pruned to remove SNPs in LD . We calculated pairwise identity-by-state ( IBS ) statistics using PLINK [61] , and removed one individual from each pair of individuals who were 2nd degree or closer relatives . Principal components analysis ( PCA ) was subsequently performed using EIGENSTRAT [62] . After exclusion of individuals of non-European ancestry , as determined by clustering with CEU HapMap ( phase II ) , a second PCA was performed to further remove outliers . Cases with anti-CCP negative or missing anti-CCP status data were removed , leaving 7 , 222 CCP+ cases and 15 , 870 controls for association analysis . To avoid duplicate samples , we used IBS estimates to remove related samples between the Immunochip and GWAS collections . Specifically , we selected a set of genotyped SNPs with missing-genotype rate<0 . 5% , MAF >5% and Hardy-Weinberg equilibrium ( HWE ) P>10−3 that were shared across all 10 collections . When related samples were identified ( siblings or duplicates ) , the sample from the GWAS data was removed ( to preferentially keep genotyped data rather than imputed data in the subsequent association analyses ) , bringing the total sample size to 9 , 785 seropositive RA cases and 33 , 742 controls ( Table S1 ) . Finally , we computed a chi-square test to assess the difference in missingness between cases and controls and removed SNPs with a Pmissing<10−2 , together with SNPs in departure form Hardy-Weinberg equilibrium ( PHWE>5 . 7×10−7 ) . To test for association with risk of RA , we used PLINK to conduct logistic regression analysis of the six Immunochip RA case-control status , including 10 eigenvectors as covariates . We conducted an inverse-variance weighted meta-analysis to combine the results across the 6 collections , for the 156 , 520 SNPs across the genome with results in one or more collections , including 492 SNPs across the CD40 locus . We also computed Cochran's Q statistic and I2 statistic to assess heterogeneity across collections . Meta-analysis and heterogeneity statistics computation was adapted from the MANTEL program . Sequence data at the CD40 locus was generated as part of a larger experiment to perform pooled sequencing of 25 RA risk genes ( Supplementary Material ) . Using Syzygy ( a pooled variant caller ) [63] , we estimated the allele frequencies in the overall sample set ( 500 cases and 650 controls of European ancestry geographically and genetically matched ) . We observed a strong correlation between genotype frequencies available from our GWAS data and frequencies estimated using the method in Syzygy indicating accurate experimental recovery of the pool composition . For the CD40 region , we had no coverage of exon 1 , but near complete coverage of the remaining eight exons ( Figure S4 ) . After stringent quality control ( Supplementary Material ) , we observed 4 coding variants at the CD40 locus: two coding-synonymous SNPs and two missense SNPs . None of the SNPs was associated with RA either in a single-SNP analysis or in a gene-burden test ( Table S2B ) . CD40 protein levels on the surface of unstimulated CD19+ B cells were measured in healthy control subjects from the PhenoGenetic Project of Brigham and Women's Hospital , a living biobank of 1 , 739 subjects free of chronic inflammatory and infectious diseases recruited from the general population of Boston , MA . Subjects used in this experiment were randomly selected from the biobank . Fresh PBMCs were isolated from 10 ml blood with 5 ml Ficoll-Hypaque ( GE; Cat#07908 ) . PBMCs were washed once in 0 . 1%BSA/PBS and blocked with FcR blocking reagent ( Milteyi Biotec; Cat . #120-000-442 ) . After red blood cells were lysed in 10 ml human red blood cell lysis buffer , 0 . 25×106 isolated PBMCs were double-labeled with an anti-CD19-FITC ( eBioscience; Cat . #11-0199-73 ) and anti-hCD40/TNFRSF5-PE ( R&D; Cat . #FAB6321P ) . The CD40 levels were measured by FACS analysis with PE-GeoMFI on CD19+ cells . As a negative control , an anti-IgG2B-PE ( R&D; Cat . #IC0041P ) was used; in addition , we used frozen BL2 cells to normalize for day-to-day variation . In total , 97 subjects had both CD40 protein levels measured and genotyping generated using the Immunochip beadset at Yale University . The same initial data-filtering steps described above were performed . Following QC , 90 samples with a call rates >99% were included in the analysis . After HapMap phase III PCA , no sample was removed based on ethnicity . A second PCA was performed to compute eigenvectors to include in the association analysis . We conducted a linear regression analysis to test for CD40 protein level-SNP association ( PLINK ) . Ten eigenvectors were included as covariates in the linear model . Details of the eQTL analysis have been previously described [30] . In short , we assessed the effect of rs4810485 on CD40 in whole peripheral blood in a collection of 1 , 469 samples ( 1 , 240 samples run on the Illumina HT12v3 platform , 229 samples run on the Illumina H8v2 platform ) . We used a Spearman rank correlation and meta-analysis using a weighted Z-method to calculate statistical significance of the rs4810485 G alleles CD40 gene expression levels . BL2 cells were purchased from DSMZ ( Germany; Cat . # ACC 625 ) . Both BL2/shRNA cell line and BL2-NFκB-Luc cell line were derived from BL2 cells , as described below . All cells were cultured in RPMI 1640 medium ( Life Technologies , Inc ) supplemented with 10% FBS . For CD40 activation , cells were incubated at 37°C for 15 minutes with 64 ng/ml trimerized CD40 ligand ( tCD40L ) . Two independent BL2 cell lines were generated , in which CD40 was knocked-down using a HuSH shRNA Plasmid , pGFP-V-RS ( Origene; Cat . # TR30007 ) . A double-stranded DNA oligo containing a hairpin structure with sequence specific to human CD40 gene ( sequence below ) [33] was cloned into the pGFP-V-RS , according to manufacture specifications ( Origene ) . 5′GATCGGCGAATTCCTAGACACCTGTTTCAAGAGAACAGGTGTCTAGGAATTCGCTTTTTTGAAGCT3′ The plasmid ( 20 µg ) was linearized by ScaI and cotransfected into BL2 cells with a puromycin selection vector by electroporation at 300 V and 950 µF . Transfected cells were cultured in regular medium for 2 hours before they were serially diluted and plated into 96-well plates . For selection , cells were grown in 0 . 3 µg/ml puromycin . Single colonies were picked two weeks later and CD40 levels were measured by Western blot and FACS analysis . For the immunoblot detection , 0 . 25×106 BL2 cells were activated either with or without tCD40L . Whole-cell lysate was prepared with 10 µl RIPA buffer and then subjected to electrophoresis on a 7 . 5% SDS-PAGE gel under reducing conditions . Proteins were electro-blotted onto a nitrocellulose membrane . The membranes were detected by antibodies for CD40 ( Santa Cruz; cat#: sc-13128 ) , NF-κB p65 ( Cell signaling; cat#: 4767 ) and phospho-NF-κB p65 ( Ser536 ) ( Cell signaling; cat#: 3033 ) . A cignal lenti luciferase reporter construct driven by a basal promoter element ( TATA box ) joined to tandem repeats of the NF-κB response element was infected into BL2 and Ramos RA-1 cells , according to manufacture specifications ( Qiagen; Cat . # CLS-013L ) . We call these two lines as BL2-NFκB-luc and Ramos-NFκB-luc , respectively . Single colonies were selected on 96 well plates with 0 . 3 µg/ml puromycin . Positive clones were further screened by luciferase assay with Steady-Glo assay system ( Promega ) after 4 hr activation by 64 ng/ml tCD40L . Since the expression of luciferase gene is controlled by the activity of NF-κB , we were able to measure the activity of NF-κB following activation with tCD40L by measuring the activity of luciferase . For LPS activation in BL2 cells , a BL2-NFκB-luc line was transfected with episomal DNA of TLR4 ( pUNO1-hTLR4a , InvivoGen ) for high levels of TLR4 expression . We call this line as BL2-TLR4-NFκB-luc . The high-throughput screening assay was optimized in 384-well format in collaboration with the Broad Institute Probe Development Center ( BIPDeC ) . Briefly , 10 µL BL2-NFκB-luc cells at 25K cells/well were plated into each well on a 384-well plate ( Perkin Elmer; Cat . # 6008230 ) using a Multidrop Combi dispenser ( Thermo ) . For each compound , DMSO or positive control IKK inhibitor VII , 25 nL was transferred using a pin tool ( Cybio ) . The final concentration for each compound was 9 . 4 µM; DMSO 0 . 25%; and IKK inhibitor VII 50 µM . After cells were incubated at 37°C for 1 hour , 10 uL ( 192 ng/mL ) tCD40L was added to make a final concentration of 64 ng/ml . Cells were incubated at 37°C for 4 . 5 hrs before 5 uL 1X Steady-Glo luciferase substrate was added . Luciferase activity was read after 5 min using LJL analyst plate reader ( LJL BioSystems ) . Each compound was tested in duplicate . A complete list of the compounds can be found in Table S3 . We refer to this set of compounds as our “2K screening” set . Seventy-three of 81 “hit” compounds from the primary screen were advanced into a series of counter-screens . The 73 compounds were selected because stock concentrations of each compound were available for dose-titration experiments . Each compound was tested across a range of concentrations from 50 µM to 0 . 39 µM ( 2-fold decrease between doses ) . In addition to screening BL2-NFκB-luc cells activated with tCD40L , we also screened BL2-TLR4-NFκB-luc cells activated with 16 ng/ml LPS ( Sigma ) . We screened an additional B cell line transfected with the same luciferase reporter contruct , Ramos-NFκB-luc cells , and activated with 64 ng/ml tCD40L and 64 ng/ml TNFα ( eBioscience ) . Cell viability of both lines was evaluated by adding 5 ul 0 . 5X CellTiter-glo . CD86 protein levels were measured in human primary CD19+ B cells from the PhenoGenetic Project purified by MACs ( Milteyi Biotec; Cat# 130-091-151 ) . Purified human primary CD19+ B cells ( 1×106 ) were pre-incubated with 10 ng/ml IL4 and different concentrations of drugs in each well of a 24-well plate at 37°C for one hour . Cells were activated with and without 64 ng/ml tCD40L . After 48 hours , cells were stained with anti-CD19-FITC/anti-CD86-PE ( Biolengd; Cat#305405 ) . CD86 expression was measured by PE GeoMFI on CD19+ gated B cells . The raw signals of each 384-well microtiter wells were normalized using the “Neutral Controls minus Inhibitors” method in Genedata Assay Analyzer ( v7 . 0 . 3 ) . The median raw signal of the intraplate neutral-control wells was set to a normalized activity value of 0 . The median raw signal of the intraplate positive-control wells was set to a normalized absolute activity value of 100 . The plate pattern correction algorithm “Assay Median” in Genedata ( v7 . 0 . 3 ) was applied to the normalized plate data . We used DMSO neutral controls to define 95% confidence intervals ( CI ) of our 2K screening compounds . We defined “hits” as those compounds outside of 95% CI in both dimensions of the replicate experiments . This led to 86 compounds that inhibited luciferase activity , consistent with our positive IKK inhibitor control . We defined a compound as promiscuous if it satisfies both of the following two rules: ( 1 ) the number of assays in which it has been tested is larger than 50 , and ( 2 ) the ratio between the number of hits and the number of assays in which it has been screened is larger than 0 . 25 . Based on these criteria , we found 40 promiscuous compounds , 5 of which had >2SD inhibition of luciferase activity , yielding 81 compounds that inhibited luciferase activity . A list of all compounds tested , including the annotation of 40 promiscuous compounds , can be found in Tables S3 and S4 . To calculate relative IC50 , the nls function in the R package of stats ( 2 . 14 version ) was used to fit the four-parameter logistic non-linear regression model as follows . This method represents the nonlinear ( weighted ) least-squares estimates of the parameters of a nonlinear model , with the equation ( xn is concentration and yn is luciferase intensity ) : . It estimates the four parameters L , h , α , and β . The starting point of L is the maximum value at smallest concentration; the starting point of h is the minimum difference between the minimum value at the largest concentration and the average value at smallest concentration . The starting point of α and β are 4 and 2 , respectively . The R function smooth . spline was used to smooth the estimated points in the curves . In some cases where a singular gradient happens , the parameters are not estimable; when a plateau was not observed , the curve was smoothed by loess . smooth function in R and the percentage was calculated by the observed values . The 95% confidence interval ( CI ) was computed based on the observed data and transferred into percentage . To calculate percent maximum inhibition for each compound , we determined the difference between the average activity at the lowest drug concentration and the average activity at the highest drug concentration . The percent maximum inhibition was then obtained by dividing the difference by the average activity at the lowest drug concentration . | A current challenge in human genetics is to follow-up “hits” from genome-wide association studies ( GWAS ) to guide drug discovery for complex traits . Previously , we identified a common variant in the CD40 locus as associated with risk of rheumatoid arthritis ( RA ) . Here , we fine-map the CD40 signal of association through a combination of dense genotyping and exonic sequencing in large patient collections . Further , we demonstrate that the RA risk allele is a gain-of-function allele that increases the amount of CD40 on the surface of primary human B lymphocyte cells from healthy control individuals . Based on these observations , we develop a high-throughput assay to recapitulate the biology of the RA risk allele in a system suitable for a small molecule drug screen . After a series of primary screens and counter screens , we identify small molecules that inhibit CD40-mediated NF-kB signaling in human B cells . While this is only the first step towards a more comprehensive effort to identify CD40-specific inhibitors that may be used to treat RA , our study demonstrates a successful strategy to progress from a GWAS to a drug screen for complex traits such as RA . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"medicine",
"rheumatoid",
"arthritis",
"immune",
"cells",
"genetics",
"of",
"the",
"immune",
"system",
"b",
"cells",
"clinical",
"immunology",
"drug",
"research",
"and",
"development",
"drugs",
"and",
"devices",
"genetics",
"... | 2013 | Human Genetics in Rheumatoid Arthritis Guides a High-Throughput Drug Screen of the CD40 Signaling Pathway |
Trypanosoma cruzi , the etiological agent of Chagas' disease , presents nutritional requirements for several metabolites . It requires heme for the biosynthesis of several heme-proteins involved in essential metabolic pathways like mitochondrial cytochromes and respiratory complexes , as well as enzymes involved in the biosynthesis of sterols and unsaturated fatty acids . However , this parasite lacks a complete route for its synthesis . In view of these facts , T . cruzi has to incorporate heme from the environment during its life cycle . In other words , their hosts must supply the heme for heme-protein synthesis . Although the acquisition of heme is a fundamental issue for the parasite’s replication and survival , how this cofactor is imported and distributed is poorly understood . In this work , we used different fluorescent heme analogs to explore heme uptake along the different life-cycle stages of T . cruzi , showing that this parasite imports it during its replicative stages: the epimastigote in the insect vector and the intracellular amastigote in the mammalian host . Also , we identified and characterized a T . cruzi protein ( TcHTE ) with 55% of sequence similarity to LHR1 ( protein involved in L . amazonensis heme transport ) , which is located in the flagellar pocket , where the transport of nutrients proceeds in trypanosomatids . We postulate TcHTE as a protein involved in improving the efficiency of the heme uptake or trafficking in T . cruzi .
Trypanosoma cruzi is the etiological agent of Chagas' disease or American trypanosomiasis . It is estimated that about 6 to 7 million people are infected worldwide , mostly in Latin America and southern states of USA , where Chagas’ disease is considered endemic . It is also becoming relevant in non-endemic regions due to migrations and the absence of control in blood banks and organ transplantation ( http://www . who . int/mediacentre/factsheets/fs340/en/ ) [1 , 2] . T . cruzi has a complex life cycle alternating between two hosts and displaying at least four developmental stages . In the mammalian host , it is present as intracellular forms , mainly amastigotes ( replicative stage ) or transiently as intracellular epimastigotes , and as bloodstream trypomastigotes ( non replicative , infective stage ) . In the insect vector , it is present as epimastigotes ( replicative stage ) and metacyclic trypomastigotes ( non-replicative , infective stage ) . During the different life-cycle stages , the parasite faces different environments and has to adapt its metabolism to the nutritional availability through the different hosts [3 , 4] . T . cruzi presents nutritional requirements for several nutrients and cofactors , and heme is included in this list . As other trypanosomatids responsible for human diseases ( T . brucei and Leishmania spp . ) , it lacks the enzymes for heme biosynthesis [5 , 6] . However these trypanosomatids present several heme-proteins involved in essential metabolic pathways like the biosynthesis of ergosterol , unsaturated fatty acids and mitochondrial cytochromes in the respiratory chain [7] . In this context , it is reasonable to hypothesize that T . cruzi must acquire heme from the extracellular environments along its life cycle . Several mechanisms have been proposed for heme transport in different organisms . In gram-positive and gram-negative bacteria , direct heme uptake , bipartite heme receptors or hemophore-heme uptake systems were described [8 , 9] . In eukaryotic organisms some proteins were found playing a role as heme transporters , but none of them showed to be homologous to those of prokaryotes . The complete characterization of heme transport and distribution in these organisms is a long way from being completely understood yet [10 , 11] . As mentioned before , only a small number of proteins were identified and characterized as heme transporters . It is worth mentioning CeHRG-1 and CeHRG-4 , both having a highly conserved function in modulating heme homeostasis in Caenorhabditis elegans [11–13] . CeHRG-4 was suggested to be involved in heme uptake by C . elegans intestinal cells and CeHRG-1 seems to mediate intracellular compartment heme delivery . Additionally , proteins involved in heme uptake have been identified in Leishmania spp . These proteins are the ABC transporter , ABCG5 , responsible for intracellular heme traffic derived from the hemoglobin breakdown [14] and LHR1 postulated as a plasma membrane heme transporter [15] . Besides , it was shown that T . cruzi epimastigotes’ proliferation depends on heme in a dose dependent manner and that the heme uptake might occur via a specific porphyrin transporter , possibly a member of the ABC-transporter family [16 , 17] . In the present work , we show for the first time that heme transport takes place in the replicative life-cycle stages: the epimastigotes occurring in the midgut of the insect vector and the intracellular amastigotes occurring in mammalian host-cells . Notably , heme uptake does not occur in the infective non-replicative stage trypomastigotes . In addition , we characterize a T . cruzi protein ( TcHTE ) , which promotes heme uptake when expressed in yeasts and improves the heme uptake/trafficking in T . cruzi epimastigotes . Remarkably , TcHTE is located in the flagellar pocket region , where it is postulated that nutrient transport takes place in trypanosomatids .
Dulbecco's Modified Eagle Medium ( DMEM ) was obtained from Life Technologies . Fetal Bovine Serum ( FBS ) was purchased from Natocor SA ( Córdoba , Argentina ) . δ-Aminolevulonic acid ( δ-ALA ) , Hemin ( Fe ( III ) protoporphyrin IX ) , Zn ( II ) mesoporphyrin IX ( ZnMP ) , Zn ( II ) protoporphyrin IX ( ZnPP ) , Sn ( IV ) mesoporphyrin IX ( SnMP ) and Ga ( III ) protoporphyrin IX ( GaPP ) were obtained from Frontier Scientific ( Logan , UT , USA ) . Hemin was dissolved in 50% ( v/v ) EtOH , 0 . 01 M NaOH at a final concentration of 1 mM , filtered using a 22 μm syringe disposable filter and it was stored at -80°C . ZnMP , ZnPP , SnMP and GaPP were dissolved in DMSO at a final concentration of 10–20 mM and were stored at -80°C . The working solutions were prepared dissolving the stock solution in 0 . 01 M NaOH at a final concentration of 1 mM . Escherichia coli strains used for all cloning procedures were grown at 37°C in Luria–Bertani medium supplemented with ampicillin ( 100 μg/mL ) or kanamycin ( 50 μg/mL ) as necessary . The yeast Saccharomyces cerevisiae cells ( hem1Δ ) were grown either in a rich medium ( YP , 1% yeast extract , 2% peptone ) , or in a synthetic complete medium ( SC ) lacking the appropriate nutrients for plasmid selection , supplemented with hemin or δ-ALA when it was necessary . In both cases , 2% Glucose ( Glc ) or 3% Glycerol–2% Ethanol ( Gly–EtOH ) were used as carbon sources . S . cerevisiae hem1Δ , MATα , ura3-52 , leu2-3 , 112 , trp1-1 , his3-11 , ade6 , can1-100 ( oc ) , heme1::HIS3 , used for heterologous complementation assays , was generously provided by Dennis R . Winge , University of Utah . The yeast cells were transformed using the lithium acetate method [18] . The Vero cell line ( ATCC CCL-81 , already available in our laboratory ) was routinely maintained in DMEM medium supplemented with 0 . 15% ( w/v ) NaHCO3 , 100 U/mL penicillin , 100 μg/mL streptomycin and 10% heat inactivated Fetal Bovine Serum ( DMEM 10% FBS ) at 37°C in a humid atmosphere containing 5% CO2 . During the T . cruzi infections , Vero cells were incubated in DMEM medium supplemented with 0 . 15% ( w/v ) NaHCO3 , 100 U/mL penicillin , 100 μg/mL streptomycin and 2% heat inactivated Fetal Bovine Serum ( DMEM 2% FBS ) at 37°C in a humid atmosphere containing 5% CO2 . The epimastigotes of T . cruzi ( Dm28c strain ) were maintained in mid-log phase by successive dilutions through Liver Infusion Tryptose ( LIT ) medium supplemented with 10% heat inactivated FBS ( LIT 10% FBS ) and 20 μM hemin [19] , at 28°C . Metacyclic trypomastigotes were obtained by spontaneous differentiation of Dm28c epimastigotes at 28°C . Cell-derived trypomastigotes were obtained after the infection of the monolayered Vero cells with metacyclic trypomastigotes . After two rounds of infections , the cell-derived trypomastigotes were used to perform analogs uptake assays or to infect Vero cells to obtain amastigotes . Intracellular amastigotes obtained 48–72 hours after the infection of the monolayered Vero cells were used for analogs uptake assay . The purity of all the obtained forms as well as their viability was evaluated by microscopic observation . The orthologous sequences of LHR1 in other trypanosomatids were obtained from TriTrypDB ( http://tritrypdb . org/tritrypdb/ ) [24] using LmjF . 24 . 2230 from Leishmania major as seed . Amino acids multiple sequence alignments were carried out using the version 2 . 0 of Clustal W and Clustal X software ( http://www . ebi . ac . uk/Tools/msa/clustalw2 , [25] ) . Transmembrane-domains were predicted with TMHMM software , accessible from ExPASy Bionformatics Resource Portal ( http://www . cbs . dtu . dk/services/TMHMM-2 . 0 , [26] ) . The epimastigotes were fixed with 500 μL of formaldehyde 3 . 7% ( w/v ) in PBS . Then they were washed twice with PBS , suspended in 40 μL of PBS , settled on polylysine-coated coverslips and mounted with VectaShield reagent . Also the trypomastigotes incubated with HAs were settled on polylysine-coated coverslips and mounted with VectaShield reagent . Additionally , the infected and non-infected Vero cells treated with HAs were mounted with VectaShield reagent . All the images were acquired with a confocal Nikon Eclipse TE-2000-E2 microscope containing the following fluorescent filter cubes: 99748-C55027 and 99749-C87049 , and using Nikon EZ-C1 software . The emission of the samples displaying red fluorescence ( treated with HAs or treated with secondary antibody conjugated with Alexa Fluor 555 ) was registered at 605/75nm . The emission of the samples containing GFP fusion proteins was registered at 515/30 nm . The emission of the samples treated with DAPI was registered at 450/30 nm . All the images were processed using the ImageJ software [21 , 22] . The heme content was quantified by the alkaline pyridine method [23] . 50–250 x 106 cells of T . cruzi epimastigotes were collected and washed twice with PBS . Samples were suspended in 500 μL with PBS and then combined in a 1 mL cuvette with 500 μL of a stock solution containing 0 . 2 N NaOH and 40% ( v/v ) pyridine . The oxidized and reduced ( by addition of sodium dithionite ) spectra were recorded between 450 nm and 700 nm . The 557 nm peak was identified in the differential -reduced minus oxidized- spectrum . The heme concentration was estimated using the molar extinction coefficient 23 . 98 mM−1 cm−1 . The experiments were performed by triplicate , and the data is presented as the mean ± SD ( standard deviation ) . All the assays were independently reproduced at least 2–4 times . Statistically significant differences between groups ( p values less than 0 . 05 ) were assessed using Mann Whitney U test ( intracellular heme content ) or Kruskal-Wallis one-way analysis of variance by ranks followed by Dunn's multiple comparison post-test ( HAs cytotoxicity effect on Vero cells ) , as appropriate ( GraphPad , version 3 . 0 ) .
As previously reported , the 10 minutes treatment with heme analogs ( SnPP , PdMP and ZnMP ) prevented heme incorporation by epimastigotes and competed with heme for its transport [17] . Therefore , it is rational to use this type of compounds to trace the heme transport in other life-cycle stages than epimastigotes . In this work we used the derivatives of protoporphyrin IX ZnPP ( Zn ( II ) protoporphyrin IX ) and GaPP ( Ga ( III ) protoporphyrin IX ) and the derivatives of mesoporphyrin IX ZnMP ( Zn ( II ) mesoporphyrin IX ) and SnMP ( Sn ( IV ) mesoporphyrin IX ) to trace heme transport along the different T . cruzi life-cycle stages . All these heme analogs ( HAs ) are fluorescent and mimic heme in structure ( S1 Fig ) . First , we confirmed HAs transport in the epimastigote stage , where exponential growth phase epimastigotes were incubated with the different HAs ( 100 μM in PBS ) for 5 minutes and HAs incorporation was analyzed by confocal microscopy ( Fig 1A ) . The confocal images revealed that ZnPP and ZnMP were clearly imported and maintained by the parasite while samples treated with GaPP and SnMP did not show the expected signal . To confirm this observation a fraction of treated epimastigotes were lysed and the soluble fraction ( total cell-free extracts ) was analyzed by direct FI measurements ( S1 Text and S2 Fig ) . Parasites treated with ZnPP , ZnMP and less extent with GaPP displayed a fluorescent signal , while those treated with SnMP did not show any ( Figs 1A and S2A ) . The viability of the treated epimastigotes was also evaluated , therefore aliquots of treated parasites were taken , washed and incubated in LIT 10% FSB plus hemin , and the cell growth was followed up for four days . We did not detect irreversible toxicity as consequence of the 5 minutes treatment with these HAs ( S2B Fig ) . As mentioned above , the fluorescent signal of the epimastigotes incubated with GaPP was weaker than those of ZnPP and ZnMP ( S2 Fig ) and was not detected by confocal microscopy ( Fig 1A ) . This fact could be due to three possibilities: 1 ) the analog could have been degraded , 2 ) the analog could have been exported back to the extracellular medium or 3 ) the GaPP transport was less efficient than the other HAs transport . To determine the possible GaPP degradation by any enzymatic activity ( like heme oxygenase activity ) , we evaluated the stability of this compound in presence of total cell-free extract of epimastigotes . The analog specific absorbance was followed up during 5 hours and no significant change on the acquired spectra was observed , ruling out the degradation of GaPP by intracellular components . Alternatively , it was possible to postulate that the observed low FI signal could be the result of an inefficient import process or a rapid export response , determining a stationary state with low intracellular concentrations of GaPP . On the other hand , it remains to be elucidated why SnMP is not accumulated into the cell at detectable level as other HAs . To explore if T . cruzi is able to import heme in other life-cycle stages , we evaluated the transport of these compounds by trypomastigotes using similar assays to those used for epimastigotes . In this case , cell-derived trypomastigotes were incubated 30 minutes with 100 μM of HAs and then analyzed by confocal microscopy . Surprisingly , a fluorescent signal was not observed in samples treated with any of the used HAs ( Fig 1B ) , suggesting that trypomastigotes did not import heme , at least under these experimental conditions . We were also interested in evaluating the heme transport in the intracellular T . cruzi amastigotes , the replicative life-cycle stage during the mammalian host infection . To accomplish this , we first evaluated the toxicity and transport of the HAs into the mammalian host cells . The Vero cells were treated with 100 μM of the analogs ( ZnPP , GaPP , ZnMP and SnMP ) and after 2 hours their viability and the HAs uptake were evaluated . No toxicity was detected in these experimental conditions ( S3 Fig ) , and 30 minutes of incubation were enough to allow the compounds to get inside and become available for the amastigotes ( Fig 2A ) . After that , non-infected and infected cells ( 2–3 days post infection ) were treated with 100 μM analogs for 30 minutes and analyzed by confocal microscopy ( Fig 2A and 2B ) . All HAs–including SnMP- produced a homogeneously distributed fluorescent signal inside the non-infected cells , showing that they were imported and maintained in the intracellular environment . However , when infected Vero cells were incubated with HAs , a completely different fluorescent pattern was observed ( Fig 2B ) . After the treatment with ZnPP , ZnMP and GaPP , they showed a punctuated fluorescent signal overlapping with the location of intracellular amastigotes . Interestingly , infected cells treated with SnMP showed the same fluorescent pattern as non-infected cells . These results strongly suggest that the HAs are being taken up by the cells and , except for SnMP , efficiently imported by T . cruzi amastigotes . Remarkably , intracellular amastigotes could discriminate between the assayed HAs , as previously observed for epimastigotes . After searching in T . cruzi databases , we identified a gene ( AYLP01000162 from Dm28c genomic sequence and TcCLB . 504037 . 10 and TcCLB . 511071 . 190 from CL Brener genomic sequence ) that encodes for a protein with a 55% sequence similarity to LHR1 , a heme transporter identified in L . amazonensis [15] and 38% with CeHRG-4 ( C . elegans protein [12] ) . The T . cruzi protein sequence was scanned in silico for transmembrane spanners in order to predict its secondary structure and our results indicated the presence of four probable transmembrane domains ( TMD ) as it was also predicted for LHR1 and CeHRG-4 . Besides their similarities–in sequence and predicted topology- two essential tyrosine residues present in LHR1 ( Y18 and Y129 [32] ) are conserved in the T . cruzi protein , which correspond to Y16 and Y127 following the T . cruzi protein numbering . However , another relevant tyrosine ( Y80 in LHR1 ) is substituted by a phenylalanine , F78 , in the T . cruzi protein . Nevertheless , none of these residues are conserved in the C . elegans protein CeHRG-4 ( S4 Fig ) . In the same way , other residues or motifs relevant for CeHRG-4 activity were not conserved in the T . cruzi protein . This evidence pointed out that the T . cruzi putative protein could play a role in T . cruzi heme transport therefore we decided to clone its gene for further characterization . The complete ORF was amplified , sequenced and cloned upstream of a sequence encoding for a C-terminal His-tag or a His-GFP-tag . To find the possible participation of this T . cruzi putative protein in heme transport we tested its activity in hem1Δ yeasts ( deficient in heme synthesis , [33] ) . Both constructs were cloned into a yeast expression vector , which were used to transform hem1Δ cells . The transformed yeasts were exposed to a selective medium supplemented with different hemin concentrations . The yeast cells expressing the T . cruzi gene ( both versions ) were able to grow at a lower hemin concentration compared to the control ( Fig 3 ) , indicating that , similarly to LHR1 [15 , 32] , the T . cruzi protein promoted heme transport by hem1Δ yeast cells ( Fig 3A ) and also allowed the recovery of the respiratory growth when a non-fermentable carbon source such as glycerol-ethanol was used ( Fig 3B ) . The presence of the T . cruzi protein ( 6His-GFP version ) in total cell extracts was verified by western blot analysis ( Fig 3C ) . Taking into account that heme uptake has been improved by the expression of the T . cruzi gene , we named its product TcHTE for T . cruzi Heme Transport Enhancer protein . The yeast cells expressing a functional TcHTE . 6His-GFP were used to analyze the protein location by confocal microscopy ( Fig 3D ) . The images confirmed that TcHTE . 6His-GFP is present in the plasma membrane , as it was expected for proteins involved in transport processes . Considering that the addition of 15–30 μM hemin is an essential requirement to maintain epimastigotes of T . cruzi in all axenic media used up to now , first we evaluated the parasite’s growth when the medium was supplemented with different concentrations of hemin . We did not observe differences in its growth ( S5 Fig ) or in the intracellular heme content when the epimastigotes were maintained with 5 or 20 μM hemin ( 3 . 572 ± 0 . 012 and 3 . 667 ± 0 . 129 nmol heme/109 parasites respectively , p>0 . 05 ) . After that , to unveil the role of TcHTE in T . cruzi , we analyzed the epimastigotes’ growth and heme content in parasites over-expressing the TcHTE gene . In this case , Dm28c . pLEW13 epimastigotes were transfected with the TcHTE6HIS-GFP versions of the gene cloned in the inducible T . cruzi expression vector pTcINDEX-GW or with the empty vector as control . After selection , the parasites containing the TcHTE . 6HIS-GFP gene or the empty vector were used to test their growth under different conditions of heme availability . First , these transfected parasites were kept for 72 hours in the absence of hemin , and then the cultures were either maintained without the addition ( 0 μM hemin ) or supplemented with 5 and 20 μM hemin . In all cases 0 . 5 μg/mL of tetracycline was added to induce the gene expression and the growth was monitored . We did not observe significant differences when these parasites were maintained in a medium supplemented with 5 μM or without hemin , suggesting that the induction of TcHTE . 6HIS-GFP gene expression did not alter the parasite growth in these conditions ( Fig 4A and 4B ) . Surprisingly , when parasite cultures were supplemented with 20 μM hemin , the growth was moderately impaired in epimastigotes expressing the TcHTE . 6HIS-GFP gene compared to the control as it is shown in Fig 4B . Also the intracellular heme content was quantified in epimastigotes maintained in 20 μM hemin , and it significantly increased from 3 . 1 ± 0 . 2 in the control to 5 . 3 ± 0 . 5 nmol heme/109 parasites in the epimastigotes expressing the TcHTE . 6HIS-GFP gene ( p<0 . 05 ) . Finally , we evaluated the location of the TcHTE . 6His-GFP protein . In that case , transfected parasites expressing TcHTE . 6HIS-GFP were analyzed by confocal microscopy . Most of GFP fluorescent signals were punctually located in a region corresponding to the flagellar pocket close to the flagellum emerging site , where it is postulated that metabolite transport processes take place in the T . cruzi ( Fig 5A ) . The TcHTE . 6His-GFP location was verified by the flagellar complex isolation assay where the GFP signal was observed next to the flagellar complex ( Fig 5B ) .
Trypanosoma cruzi requires a supply of heme like other trypanosomatids , which are auxotrophic for this cofactor . In view of these facts , it is postulated that T . cruzi must import it from the different hosts along its different life-cycle stages [7] . Since heme is a toxic molecule , it is accepted that these organisms must possess an essential and controlled pathway for its uptake and distribution [6 , 34] . Recently , several proteins were described as heme transporters or involved in heme transport in trypanosomatids [14 , 15] and other organisms with heme auxotrophy like C . elegans [35] . Among them , CeHRG-4 and CeHRG-1 proteins from C . elegans [12 , 13] and LHR1 from L . amazonensis , were described as proteins involved in heme transport or playing a role as a heme transporter . Also , LHR1 activity was relevant for the infectivity of L . amazonensis [32 , 36] . In addition , the activity of an ABC transporter–ABCG5- was reported as a heme transporter involved in the intracellular heme distribution in L . donovani [14] . Moreover , it was demonstrated that heme is relevant for T . cruzi epimastigote proliferation and that its uptake was inhibited by the treatment with ABC transporters’ inhibitors , indicating that heme transport should be dependent on an active transporter [16 , 17] . Nevertheless , up to date , the identity of the T . cruzi heme transporter was not unraveled . In this work we used different fluorescent heme analogs ( ZnPP , GaPP , ZnMP and SnMP ) to study heme transport along the different T . cruzi life-cycle stages . Only ZnPP , ZnMP , and to a lesser extent GaPP ( only detected by direct fluorescent measurements ) were accumulated inside the epimastigotes . The weak fluorescent signals detected in parasites incubated with GaPP could be a consequence of its slower uptake , its faster export , its degradation , or any combination of them . Since we were not able to detect GaPP degradation in the presence of total cell-free extracts we exclude this possibility . This reinforces the idea that the reduced intracellular concentration of GaPP might be a consequence of an unfavorable relationship between influx and efflux when compared to ZnMP and ZnPP . On the other hand , SnMP was not accumulated at detectable levels in epimastigotes . These results suggest that the HAs were selectively taken up by the T . cruzi epimastigotes . In spite of its relevance , no information was found , up to now , about heme transport in other life-cycle stages besides epimastigotes . Our results reveal that only replicative stages ( epimastigotes and amastigotes ) , but not infective stages ( trypomastigotes ) , were able to uptake heme . Moreover , based on our observations , we postulate that T . cruzi amastigotes are able to selectively withdraw most of the HAs ( ZnPP , ZnMP and GaPP ) available from their environment , the host cell cytosol . Apparently , SnMP did not accumulate in intracellular amastigotes as it was observed in epimastigotes . These results reinforce the hypothesis that T . cruzi presents an exclusive heme transporter that might discriminate between structurally related compounds . Remarkably , the heme uptake regulation during the life cycle together with the proposed selectivity of the heme transporter points that this process is critical , being very specific and tightly regulated . Recently , LHR1 was identified as a protein relevant for heme transport in L . amazonensis [32] . Therefore , we described a T . cruzi protein -herein named TcHTE- that shows sequence similarity to LHR1 and also with the C . elegans protein CeHRG-4 [12] . In spite of their sequence similarity and the conserved predicted topology , TcHTE , as well as LHR1 and CeHRG-4 , does not show any recognized domains or motifs that allow its classification as an active or ABC transporter [37 , 38] , as it was expected based on previous reported results [17] . Nevertheless , the hypothetical T . cruzi protein was characterized . TcHTE activity was first evaluated in the yeast hem1Δ ( HEM1 gene encodes for the 5-aminolevulenic acid synthase , the first enzyme in heme biosynthesis ) as it was previously used to assess the heme transport activity of CeHRG-4 and LHR1 proteins [13 , 15] . Since yeast exhibit a low affinity transport activity , hem1Δ cells can grow in a medium supplemented with high concentrations of hemin . The expression of the TcHTE gene enhanced heme transport and/or distribution in these yeast cells allowing their growth at lower hemin concentrations and also the recovery of the respiratory growth . Moreover , TcHTE was located on the yeast's plasma membrane as it was expected for a transporter protein , the same was observed with LHR1 [14 , 35] and Ce-HRG4 [13] . Nevertheless the results presented here do not allow us to discriminate if TcHTE is itself the heme transporter or if it is an essential protein that promotes/enhances the heme transport ( or trafficking ) activity . Finally , we explored the role of TcHTE in T . cruzi epimastigotes . We did not observe significant differences when epimastigotes overexpressing the TcHTE gene were grown in a medium supplemented with low hemin concentrations . However , when these epimastigotes were exposed to 20 μM hemin–the standard hemin concentration used in axenic medium- the presence of extra TcHTE turned out to be a disadvantage to maintain the optimal growth and an increment in the intracellular heme was observed . In the later case , the extra TcHTE could push on the heme uptake rendering more intracellular heme than the parasite is able to metabolize causing a mild negative effect on growth rates . These observations strongly suggest that TcHTE displays a role enhancing heme transport or trafficking . Additionally , epimastigotes transfected with a plasmid containing TcHTE . 6HIS-GFP were analyzed by confocal microscopy . The GFP fluorescent signal was located on the flagellar pocket region , similarly to the reported hemoglobin receptor of Leishmania donovani [39] . Surprisingly , TcHTE was not homogenously distributed along the plasma membrane or in internal vesicles as it was observed for LHR1 on L . amazonensis promastigotes [15] . These observations might indicate differences in heme transport between these trypanosomatids , which could be mediated by different transporters and/or involving different mechanisms . Although many transporters have been biochemically characterized in T . cruzi , their location remains unknown [4 , 40 , 41]; however it is established that the trypanosomatids’ transport proceeds through the flagellar pocket [39 , 42] . Two different mechanisms ( or pathways ) were proposed for heme acquisition by T . cruzi epimastigotes . One of them , probably via the endocytosis , is responsible for the hemoglobin import , and the other , a faster pathway ( few seconds ) , allows the transport of free heme ( followed by the fluorescent heme analog PdMP ) depending on an active transporter [16] . In both cases the fluorescent signal ( from the heme analog PdMP or hemoglobin-rhodamine ) was initially observed in the anterior part of the parasite ( on the flagellar pocket region ) , which corresponds with the cellular location for TcHTE protein . Although the TcHTE protein itself does not exhibit any distinctive signature allowing its classification as an active transporter , the results presented here allowed us to postulate that TcHTE plays a relevant role in promoting/enhancing the heme transport in order to regulate this activity . However the identity of the heme transporter still remains unknown . In summary , the results presented here show , for the first time , that T . cruzi transports heme along its replicative life-cycle stages , the epimastigote in the midgut of the insect vector and the intracellular amastigote in the mammalian host , but not in the infective form , the trypomastigote stage . Our findings revealed the existence of a protein transporter that might discriminate between structurally related compounds . Besides , we presented the first characterization of a T . cruzi protein , named TcHTE , which is conserved in trypanosomatids . We postulate that TcHTE plays a relevant role in T . cruzi heme transport/trafficking enhancement regulating this activity , without excluding it as part of or associated with the transporter complex . Considering the critical role of heme as an essential cofactor in T . cruzi , and the fact that its uptake inhibition results deleterious for the parasite , new approaches will be required for the complete identification of the molecular partners involved in its transport and distribution . Their identification and characterization might result in new molecular targets to inhibit parasite proliferation . | Trypanosoma cruzi is a protozoan parasite that causes Chagas’ disease , the most common parasitic disease in the Americas , but due to migration movements it has become relevant in non-endemic countries . T . cruzi , as other trypanosomatids responsible for human diseases ( T . brucei and Leishmania spp . ) , lacks the enzymes for heme biosynthesis . However , it presents several heme-proteins involved in essential metabolic pathways , like biosynthesis of ergosterol and unsaturated fatty acids and mitochondrial cytochromes in its respiratory chain . In this work , we used different fluorescent compounds , which are structurally similar to heme ( heme analogs ) , to explore the heme uptake in T . cruzi . The treatment of the different forms of T . cruzi with these compounds shows that this parasite is able to import heme during the replicative stages: epimastigote in the insect vector and intracellular amastigote in the mammalian host . We also characterized a T . cruzi protein , TcHTE , located in the flagellar pocket , where the transport of nutrients takes place in trypanosomatids . We propose TcHTE as a protein that improves the efficiency of the heme uptake . Considering the critical role of heme as an essential cofactor in T . cruzi , the identification of proteins involved in heme transport and distribution might result in the discovery of new molecular targets that may inhibit parasite proliferation . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2016 | The Trypanosoma cruzi Protein TcHTE Is Critical for Heme Uptake |
In 2015 , Brazil was faced with the cocirculation of three arboviruses of major public health importance . The emergence of Zika virus ( ZIKV ) presents new challenges to both clinicians and public health authorities . Overlapping clinical features between diseases caused by ZIKV , Dengue ( DENV ) and Chikungunya ( CHIKV ) and the lack of validated serological assays for ZIKV make accurate diagnosis difficult . The outpatient service for acute febrile illnesses in Fiocruz initiated a syndromic clinical observational study in 2007 to capture unusual presentations of DENV infections . In January 2015 , an increase of cases with exanthematic disease was observed . Trained physicians evaluated the patients using a detailed case report form that included clinical assessment and laboratory investigations . The laboratory diagnostic algorithm included assays for detection of ZIKV , CHIKV and DENV . 364 suspected cases of Zika virus disease were identified based on clinical criteria between January and July 2015 . Of these , 262 ( 71 . 9% ) were tested and 119 ( 45 . 4% ) were confirmed by the detection of ZIKV RNA . All of the samples with sequence information available clustered within the Asian genotype . This is the first report of a ZIKV outbreak in the state of Rio de Janeiro , based on a large number of suspected ( n = 364 ) and laboratory confirmed cases ( n = 119 ) . We were able to demonstrate that ZIKV was circulating in Rio de Janeiro as early as January 2015 . The peak of the outbreak was documented in May/June 2015 . More than half of the patients reported headache , arthralgia , myalgia , non-purulent conjunctivitis , and lower back pain , consistent with the case definition of suspected ZIKV disease issued by the Pan American Health Organization ( PAHO ) . However , fever , when present , was low-intensity and short-termed . In our opinion , pruritus , the second most common clinical sign presented by the confirmed cases , should be added to the PAHO case definition , while fever could be given less emphasis . The emergence of ZIKV as a new pathogen for Brazil in 2015 underscores the need for clinical vigilance and strong epidemiological and laboratory surveillance .
Since 2015 , Brazil is faced with the challenge of three co-circulating arboviruses of major public health importance . For the last 30 years , dengue virus ( DENV ) infection has been the main mosquito-transmitted threat in the country , which has suffered several epidemics caused by all four serotypes , [1 , 2] fostered by the widespread presence of its main vector , Aedes aegypti , in densely populated urban areas . [3–5] In spite of all efforts , a sustained reduction of the Ae aegypti population has not been achieved . The emergence of Chikungunya virus ( CHIKV ) and Zika virus ( ZIKV ) in Brazil [6 , 7] poses new challenges to clinicians and public health authorities due to overlapping clinical features and the fact that validated serological assays for ZIKV that can reliably distinguish between acute disease and past exposure are currently not available . Most importantly , ZIKV infections are suspected to be associated with congenital abnormalities [8] and with neurological complications such as Guillan-Barré syndrome ( GBS ) [9] , while CHIKV infections have been associated with persisting arthralgia [10] . Both ZIKV as well as DENV are members of the family Flaviviridae , whereas CHIKV is an alphavirus of the Togaviridae family . ZIKV was first described in humans in Africa in 1952 [11] and has recently caused epidemics in the Pacific region . [12–14] In May 2015 , ZIKV was identified in Northeast Brazil associated with an outbreak of acute exanthematous disease in the state of Bahia , [6 , 15] followed by several other locations , including the state of Rio de Janeiro . [6 , 15–17] ZIKV was considered to cause a benign infection , leading to a self-limited disease consisting of rash , fever ( often of low intensity and short-termed or even absent ) , and mild arthralgia and therefore did not receive much scientific attention until recently [14] . However , new findings suggest an association of recent ZIKV disease with GBS amongst adults[18] and with congenital abnormalities ( e . g . microcephaly and ocular lesions in neonates ) born from women reporting a ZIKV-like disease during pregnancy . [19–21] Here we describe the clinical , epidemiological , and virological features of patients with acute exanthematous disease at a tertiary reference centre in Rio de Janeiro during an outbreak of ZIVK in the first half of 2015 . We analyze the temporal and spatial occurrence of suspected ZIKV patients over the time of the outbreak and compare the profile of clinical signs and symptoms in PCR-confirmed ZIKV patients with the current PAHO case definition for ZIKV disease .
The study was conducted at the Laboratorio de Pesquisa Clínica de Doenças Febris Agudas of the Instituto Nacional de Infectologia Evandro Chagas ( INI ) , which is part of the regional Dengue-Research Center of Fundação Oswaldo Cruz ( FIOCRUZ ) , a reference center of the dengue network in Rio de Janeiro , Brazil . Patients with acute febrile disease seen at this outpatient clinic are either referred from other health units in Rio de Janeiro , or are spontaneously seeking care as they live in the nearby district of ‘Manguinhos’—an impoverished neighborhood close to the FIOCRUZ research center with a human development index of 0 . 726 , a population of 36 , 160 persons in an area of 261 square kilometres , and an estimated number of 10 , 816 households . Since January 2015 physicians at INI/FIOCRUZ observed an increase of cases characterized by rash , with either absent or low-grade and short-termed fever , and sometimes associated with arthralgia and/or conjunctivitis . The disease was clinically distinct from DENV which led to a systematic syndromic investigation utilizing a specific laboratory algorithm , which included diagnostics assays for detection of DENV , CHIKV , and ZIKV ( after the reports of ZIKV transmission in the Northeast of Brazil[22] ) . The investigations were performed as part of the ongoing study on “Detection of unusual clinical presentations of dengue” , which was reviewed and approved by the local Ethics Committee ( CAAE 0026 . 0 . 009 . 000–07 ) . Written informed consent was obtained from all patients or their legal representatives . For the purpose of the analysis presented here we concentrate on the first half of the year 2015 . After the identification of ZIKV infection in a patient with no travel history outside Rio de Janeiro in May 2015[16] , additional prospective surveillance at the outpatients' clinic was initiated in order to specifically identify patients with suspected ZIKV disease . Patients with acute onset of generalized macular or papular rash were considered suspected cases of ZIKV . We interviewed patients using a standard case report form ( CRF ) to collect information about demographic features , clinical signs and symptoms , and the severity of the disease ( S1 File ) . Blood samples were collected during the acute phase ( i . e . , within 7 days after the onset of symptoms ) and during the convalescent phase . Samples were stored at -70°C and analyzed in the Flavivirus Diagnostic and Reference Laboratory of Fiocruz . Acute phase serum samples were tested by qRT-PCR for ZIKV [23] and RT-PCR for DENV RNA [24] . CHIKV qRT-PCR testing was performed for a random sample of 25% of patients as no patient tested positive and there was no on-going transmission of CHIKV in Rio at the time of this stud1y . Cases were classified as ZIKV infection if ZIKV RNA was detected in the serum . Phylogenetic analysis of nucleic acid sequences derived from 10 random ZIKV positive samples ( out of 119 ) was performed using 327 base pairs of the envelope protein ( GeneBank accession number KT381874 ) . The tree was inferred using the maximum likelihood algorithm based on the Tamura 3-parameter model as implemented in MEGA 6 . The numbers shown to the left of the nodes represent bootstrap support values > 70 ( 1 , 000 replicates ) . The tree was rooted with Spondoweni virus . The home addresses of the cases were georeferenced using Google maps . Statistical analyses and maps were performed using software R version 3 . 2 . 2 . libraries RCurl , RJSONIO , ggmap and ggplot2 . [25]
Between 1st of January and 31st of July 2015 , 364 suspected cases of ZIKV disease were identified based on the presence of acute onset rash with or without fever . Of these , 262 ( 71 . 9% ) were tested and 119 ( 45 . 4% ) were confirmed by the detection of ZIKV RNA through qRT-PCR assay . All of the samples with sequence information available clustered within the Asian genotype ( Fig 1 ) . RT-PCR assays using consensus primers for nucleic acid of other arboviruses , including DENV and CHIKV , were negative in these patients . No DENV RNA was detected in any of the 143 acute phase serum samples that tested negative for ZIKV . Suspected ZIKV cases started to appear in January 2015 , peaking in May ( Fig 2 ) . Their geographical distribution is shown in Fig 3 . In spite of the clustering in the surrounding neighborhood ‘Manguinhos’ , cases residing in other neighborhoods in the Metropolitan area of Rio de Janeiro were also included ( Fig 3 ) . Baseline characteristics of confirmed ( tested ZIKV positive ) and unconfirmed ( tested ZIKV negative ) patients are presented in Table 1 . The median age of these patients was 37 years ( range 9 to 60 ) . A majority of the cases were females ( 158/262–60 . 3% ) , and among these 73% were aged 15–49 years ( 116/158–73 . 4% ) . The same pattern was seen with regard to the confirmed cases only ( 60 . 5% female [72/119] and 76 . 4% among those in the age group of 15–49 years [55/72] ) . No travel histories were recorded for the confirmed cases , thus infections were acquired locally . Only 38% of the patients recalled exposure to mosquito bites . Arterial hypertension was the most frequent comorbidity recorded , followed by diabetes . Geographic clustering of cases ( defined as more than one case in the household , neighborhood or work ) occurred in half of all confirmed cases ( 62/119–51 . 1% ) . The most commonly reported symptoms in the first four days of the disease were rash , itching , prostration , headache and arthralgia ( with or without associated oedema ) ( Table 2 ) . Although fever was not observed at presentation in the majority of patients ( 64% ) , 43 patients ( 36% ) reported a history of fever not lasting longer than one day , or the occurrence of a single fever peak on the first day of disease . The median duration of rash was 5 . 5 days ( range 3 to 7 ) , and of arthralgia 9 days ( range 2 to 21 ) . Mild hemorrhagic symptoms ( petechiae , minor mucosal bleeding ) were reported in 21% of the confirmed and 11% of the unconfirmed cases , while no severe bleeding was observed . One patient was hospitalized with neurological symptoms consisting of peripheral nerve impairment , and is described in more detail elsewhere ( Brasil et al , in press ) . No deaths or other severe complications were associated with the ZIKV disease in this series . Four women with confirmed ZIKV were pregnant . One had a miscarriage at 10 weeks of pregnancy , three weeks after the acute disease episode . The remaining three ( two with ZIKV infection during the 18th week and one during the 35th week ) delivered normal babies without any clinical evidence of infection or congenital abnormalities . All patients had full blood counts performed . The median leucocyte count of confirmed ZIKV cases was 4 , 590 cells/mm3 ( 2 , 240 to 11 , 570 cells/mm3 ) , the median platelet count was 201 , 000 cells/mm3 ( 102 , 000 to 463 , 000 cells/mm3 ) and the median haematocrit was 41 . 2% ( 33 . 2 to 50 . 3% ) .
This is the first report of a ZIKV outbreak in the state of Rio de Janeiro , based on a large number of suspected ( n = 364 ) and laboratory confirmed cases ( n = 119 ) . We were able to demonstrate that ZIKV was circulating in Rio de Janeiro as early as January 2015 . The outbreak began in the first half of 2015 with a transmission peak in May 2015 . The clinical picture of the disease was characterized by the sudden increase of exanthematous disease with absent or short-termed and low-grade fever . The early documentation as well as the detailed clinical characterization was possible through a prospective syndromic surveillance study , which included a systematic collection of blood samples for the detection of unusual forms of dengue . In fact , 11% of the confirmed cases date back to the months before ZIKV transmission was reported in Northeast Brazil , in May ( Fig 2 ) . The detection of ZIKV RNA in the serum of acutely ill patients and the absence of nucleic acid of other arboviruses provide convincing evidence that the outbreak was caused by ZIKV . Although DENV is common in Rio de Janeiro , none of the 252 patients for whom acute-phase specimens were available , tested positive for DENV , which is surprising and possibly highlights explosive transmission dynamics of ZIKV . Patients were also tested for other pathogens at the attending physician’s discretion , but no other causes were found ( i . e . CHIKV , rubella , cytomegalovirus , EBV , toxoplasmosis ) . Based on this study , there is no evidence that ZIKV was circulating in Rio de Janeiro before January 2015 . However , a larger and more representative sample would be necessary to make a more definitive statement about ZIKV circulation in Rio de Janeiro before 2015 . The peak of cases was observed in May ( Fig 2 ) , after the rainy season—which is in line with temporal patterns observed in previous outbreaks in Yap Island and French Polynesia . [13 , 14] In May 2015 , atypically high precipitation was recorded in Rio de Janeiro [26] , which may have affected the vector distribution and abundance . It is noteworthy , that the first DENV epidemic in Rio de Janeiro in 1986 started in the same geographical area from where the majority of ZIKV cases reported here originated [27] . This is likely to be associated with the high human population density , abundant Ae aegypti populations , the precarious socioeconomic status , and lack of infrastructure . The observation that ZIKV cases clustered within households is interesting and needs to be verified in other settings . If this is a consistent feature , it could be consistent with a high vectorial capacity or hint towards other routes of transmission . The possibility of transmission via mucosal contact is supported by evidence of sexual transmission [28 , 29] and by the detection of ZIKV RNA in saliva and urine . [30 , 31] Further studies need to be carried out to address this question in more detail , with important implications on epidemiology and disease control of ZIKV . The age distribution with a bias towards adults is consistent with the overall age profile of the clinic . Female patients were overrepresented with regard to males , which might be due to differences in health care–seeking behavior . An epidemic of microcephaly in newborns in Northeast Brazil has been the focus of attention by public health authorities in Brazil and worldwide . There is emerging evidence that these congenital abnormalities may be associated with exposure to ZIKV during pregnancy . [32] Among the ZIKV-positive pregnant women followed in our study , one out of four had a miscarriage in the 10th week of pregnancy . However , no further investigations were performed in this case . In a recently published cohort of pregnant women with ZIKV infection , we have described two miscarriages amongst 72 confirmed cases [21] , although the real frequency of this and other complications is yet to be determined . More recently , the state of Rio de Janeiro has reported a tenfold increase in microcephaly cases in 2015 compared to previous years [32] and , due to the long-lasting medical and economic consequences , it is of high importance that multidisciplinary studies are conducted to determine the incidence and the phenotypic variability of congenital abnormalities related to ZIKV infection during pregnancy . The clinical signs and symptoms of ZIKV observed in Rio de Janeiro have many similarities with those described in previous reports and case series . [14] The similarity and frequency of clinical manifestations between confirmed and unconfirmed cases supports the notion that the majority of suspect cases were suffering from ZIKV disease , which could be due to current limits of detecting low quantities of RNA within a potentially short viremic period and the lack of a reliable serological test to ascertain recent exposure and distinguish it from DENV infections . As in DENV and CHIKV infections , rash was a common feature in ZIKV . However , pruritus/itching was a prominent characteristic of the maculo-papular rash in the majority of confirmed cases ( 79% ) from the onset of symptoms , which can potentially help in distinguishing ZIKV from other arboviral infections . Enlarged lymph nodes , especially in the cervical and retro-auricular regions , were also found frequently , which led clinicians to consider rubella as an important differential diagnosis . More than half of the patients reported headache , arthralgia , myalgia , non-purulent conjunctivitis , and lower back pain , consistent with the case definition of suspected ZIKV issued by PAHO . [8] However , fever was absent in most cases . In our opinion , pruritus , the second most common clinical sign presented by the confirmed cases , should be added to the PAHO case definition . DENV , which has been circulating in Brazil for many years , was considered an unlikely diagnosis in many patients because fever was only observed in a minority ( 36% ) . When present , fever was usually short-termed and the temperature relatively low . From a clinical management perspective , it is important to differentiate ZIKV both from CHIKV and especially from DENV due to the need to monitor cases for possible evolution of severe and life-threatening clinical outcomes . Standard laboratory values ( e . g . hematology or biochemistry ) did not show a distinct pattern in the confirmed ZIKV patients and are unlikely to be helpful . The leucocyte count in ZIKV patients was in many cases moderately decreased as in other viral infections ( < 5 , 000 cells mm3 ) and both platelet counts and haematocrit were within the normal range . In this study , only one patient was hospitalized with febrile illness and a neurological manifestation developed afterwards ( manuscript submitted ) . An increase in cases with Gullain-Barré syndrome ( GBS ) was reported during the epidemic In French Polynesia in 2013/14 [12] , as well as in the Northeast of Brazil during the current outbreak . [6] It remains to be seen if ZIKV is causally related to higher rates of neurological complications . Our data suggests that in a previously naïve adult population , ZIKV causes a self-limited and mostly benign disease characterized by a pruritic maculopapular rash and absent or low-grade , short-termed fever . ZIKV-like disease was not a notifiable condition in Rio de Janeiro until October 22th , 2015 . This partially explains why our data differs from the official figures , [22] as the syndromic surveillance study that was implemented in 2007 ( long before the ZIKV outbreak started ) is more likely to detect changes in the clinical patterns of dengue compared to the routine health system . Similar to the dynamic in the Northeast of Brazil , a considerable increase in the number of microcephaly in neonates was registered in Rio de Janeiro during 2015 . As of January 2016 , 122 microcephaly cases were documented—compared to the average of 10 cases per year registered in the previous years . [32] This increase could potentially be attributed to the first wave of ZIKV transmission observed between January and July 2015 in Rio de Janeiro . In Pernambuco , in the Northeast region of Brazil , 1306 cases of microcephaly were reported as of the second epidemiological week of 2016 [32] which are attributed to the ZIKV outbreak in early spring 2015 . At this point , health services must be alerted to the potential for an even larger epidemic during the summer of 2015–2016 spreading to additional locations and affecting the susceptible proportion of the population that was not exposed during the last transmission season . The emergence of ZIKV as a new pathogen for Brazil in 2015 underscores the ease with which pathogens travel between continents and the need for clinical vigilance and strong epidemiological and laboratory surveillance systems . The phylogenetic analysis somehow is in line with the speculative hypothesis that ZIKV was possibly introduced to Rio de Janeiro during the VI World Sprint Championship canoe race in August 2014 , which included teams from four Pacific countries ( French Polynesia , New Caledonia , Cook Islands , and Easter Island ) where the virus circulated during 2014 . [33] In 2016 , Rio de Janeiro will be hosting the Summer Olympics and Paralympics games , which will attract a high number of national and international visitors . The threat caused by ZIKV has far reaching implications on tourism and industry . [34] Reliable and sensitive surveillance of arboviral disease that includes a system for the detection of emerging pathogens is of paramount importance to manage the complex challenges ahead . Our findings have demonstrated that ZIKV was circulating in Rio de Janeiro at least five months before its detection was announced by the health authorities , which must be taken into consideration for future design and implementation of effective syndromic surveillance systems . | Zika virus ( ZIKV ) has been identified in 2015 in Brazil for the first time , causing outbreaks of an illness characterized by skin rash and absent or low grade and short-termed fever . It is difficult to distinguish ZIKV from Dengue ( DENV ) or ( CHIKV ) based on the acute clinical presentation . The virus is closely related to DENV , and therefore antibody tests also have problems distinguishing between the two viruses due to cross-reactivity . Recent findings suggest that in a minority of ZIKV cases neurological disease can develop , and that babies born from mothers reporting a ZIKV-like illness during pregnancy may suffer from congenital abnormalities , in many cases a small head or brain . Here we report about an outbreak of ZIKV disease in Rio de Janeiro in the first half of the year 2015 , which reached its peak in May/June 2015 . This is the first published description of a ZIKV outbreak in Latin America . It is interesting to note that confirmed cases appeared as early as January 2015 . Cases were confirmed based on the detection of the viral genome in the blood of the patients . The clinical characterization of the confirmed cases and unconfirmed cases proved to be very similar . Itching or itching rash has been suggested to be added to the case definition issued by the Pan American Health Organization ( PAHO ) . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"microcephaly",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"chikungunya",
"infection",
"pathogens",
"geographical",
"locations",
"microbiology",
"tropical",
"diseases",
"alphaviruses",
"viruses",
"developmental",
"b... | 2016 | Zika Virus Outbreak in Rio de Janeiro, Brazil: Clinical Characterization, Epidemiological and Virological Aspects |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.