Comparative genomics reveals diversity among xanthomonads infecting tomato and pepper

MISSING IMAGE

Material Information

Title:
Comparative genomics reveals diversity among xanthomonads infecting tomato and pepper
Series Title:
BMC Genomics
Physical Description:
Mixed Material
Language:
English
Creator:
Pontis, Neha
Krasileva, Ksenia
Chow, Virginia
Almeida, Nalvo F.
Patil, Prabhu B.
Ryan, Robert P.
Sharlach, Molly
Behlau, Franklin
Dow, J. Max
Momol, M. T.
White, Frank F.
Preston, James F.
Vinatzer, Boris A.
Koebnik, Ralf
Setubal, João C.
Norman, David J.
Staskawicz, Brian J.
Jones, Jeffrey B.
Publisher:
BioMed Central
Publication Date:

Notes

Abstract:
Background: Bacterial spot of tomato and pepper is caused by four Xanthomonas species and is a major plant disease in warm humid climates. The four species are distinct from each other based on physiological and molecular characteristics. The genome sequence of strain 85-10, a member of one of the species, Xanthomonas euvesicatoria (Xcv) has been previously reported. To determine the relationship of the four species at the genome level and to investigate the molecular basis of their virulence and differing host ranges, draft genomic sequences of members of the other three species were determined and compared to strain 85-10. Results: We sequenced the genomes of X. vesicatoria (Xv) strain 1111 (ATCC 35937), X. perforans (Xp) strain 91-118 and X. gardneri (Xg) strain 101 (ATCC 19865). The genomes were compared with each other and with the previously sequenced Xcv strain 85-10. In addition, the molecular features were predicted that may be required for pathogenicity including the type III secretion apparatus, type III effectors, other secretion systems, quorum sensing systems, adhesins, extracellular polysaccharide, and lipopolysaccharide determinants. Several novel type III effectors from Xg strain 101 and Xv strain 1111 genomes were computationally identified and their translocation was validated using a reporter gene assay. A homolog to Ax21, the elicitor of XA21-mediated resistance in rice, and a functional Ax21 sulfation system were identified in Xcv. Genes encoding proteins with functions mediated by type II and type IV secretion systems have also been compared, including enzymes involved in cell wall deconstruction, as contributors to pathogenicity. Conclusions: Comparative genomic analyses revealed considerable diversity among bacterial spot pathogens, providing new insights into differences and similarities that may explain the diverse nature of these strains. Genes specific to pepper pathogens, such as the O-antigen of the lipopolysaccharide cluster, and genes unique to individual strains, such as novel type III effectors and bacteriocin genes, have been identified providing new clues for our understanding of pathogen virulence, aggressiveness, and host preference. These analyses will aid in efforts towards breeding for broad and durable resistance in economically important tomato and pepper cultivars.
General Note:
Publication of this article was funded in part by the University of Florida Open-Access publishing Fund. In addition, requestors receiving funding through the UFOAP project are expected to submit a post-review, final draft of the article to UF's institutional repository, IR@UF, (www.uflib.ufl.edu/ufir) at the time of funding. The Institutional Repository at the University of Florida (IR@UF) is the digital archive for the intellectual output of the University of Florida community, with research, news, outreach and educational materials

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
Resource Identifier:
doi - 10.1186-1471-2164-12-146
System ID:
AA00008951:00001


This item is only available as the following downloads:


Full Text

PAGE 1

RESEARCHARTICLE OpenAccessComparativegenomicsrevealsdiversityamong xanthomonadsinfectingtomatoandpepperNehaPotnis1,KseniaKrasileva2,VirginiaChow3,NalvoFAlmeida4,PrabhuBPatil5,RobertPRyan6, MollySharlach2,FranklinBehlau1,7,JMaxDow6,MTMomol1,FrankFWhite8,JamesFPreston3,BorisAVinatzer9, RalfKoebnik10,JooCSetubal11,DavidJNorman12,BrianJStaskawicz2,JeffreyBJones1*AbstractBackground: Bacterialspotoftomatoandpepperiscausedbyfour Xanthomonas speciesandisamajorplant diseaseinwarmhumidclimates.Thefourspeciesaredistinctfromeachotherbasedonphysiologicaland molecularcharacteristics.Thegenomesequenceofstrain85-10,amemberofoneofthespecies, Xanthomonas euvesicatoria ( Xcv )hasbeenpreviouslyreported.Todeterminetherelationshipofthefourspeciesatthegenome levelandtoinvestigatethemolecularbasisoftheirvirulenceanddifferinghostranges,draftgenomicsequences ofmembersoftheotherthreespeciesweredeterminedandcomparedtostrain85-10. Results: Wesequencedthegenomesof X.vesicatoria ( Xv)strain1111(ATCC35937), X.perforans ( Xp )strain91-118 and X.gardneri ( Xg )strain101(ATCC19865).Thegenomeswerecomparedwitheachotherandwiththe previouslysequenced Xcv strain85-10.Inaddition,themolecularfeatureswerepredictedthatmayberequiredfor pathogenicityincludingthetypeIIIsecretionapparatus,typeIIIeffectors,othersecretionsystems,quorumsensing systems,adhesins,extracellularpolysaccharide,andlipopolysaccharidedeterminants.SeveralnoveltypeIIIeffectors from Xg strain101and Xv strain1111genomeswerecomputationallyidentifiedandtheirtranslocationwas validatedusingareportergeneassay.AhomologtoAx21,theelicitorofXA21-mediatedresistanceinrice,anda functionalAx21sulfationsystemwereidentifiedin Xcv .Genesencodingproteinswithfunctionsmediatedbytype IIandtypeIVsecretionsystemshavealsobeencompared,includingenzymesinvolvedincellwalldeconstruction, ascontributorstopathogenicity. Conclusions: Comparativegenomicanalysesrevealedconsiderablediversityamongbacterialspotpathogens, providingnewinsightsintodifferencesandsimilaritiesthatmayexplainthediversenatureofthesestrains.Genes specifictopepperpathogens,suchastheO-antigenofthelipopolysaccharidecluster,andgenesuniqueto individualstrains,suchasnoveltypeIIIeffectorsandbacteriocingenes,havebeenidentifiedprovidingnewclues forourunderstandingofpathogenvirulence,aggressiveness,andhostpreference.Theseanalyseswillaidinefforts towardsbreedingforbroadanddurableresistanceineconomicallyimportanttomatoandpeppercultivars.BackgroundBacterialspotdiseaseoftomatoandpepperpresentsa seriousagriculturalproblemworldwide,leadingtosignificantcroplossesespeciallyinregionswithwarmand humidclimate.Thediseaseischaracterizedbynecrotic lesionsonleaves,sepalsandfruits,reducingyieldand fruitquality[1].Thediseaseiscausedbyarelatively diversesetofbacterialstrainswithinthegenus Xanthomonas;strainnomenclatureandclassificationfor thestrainsthatinfectpepperandtomatohavegone throughconsiderabletaxonomicrevisioninrecentyears. Currently,thepathogensareclassifiedintofourdistinct pathogengroups(A,B,C,andD)withinthegenus Xanthomonas .StrainsbelongingtogroupsA,BandD infectbothtomatoandpepper.GroupCstrainsare pathogeniconlyontomato[2,3].Thesephenotypically andgenotypicallydistinctstrainshavedifferentgeographicdistributions.StrainsofgroupAandBare foundworldwide.Cstrainshavebeenincreasinglyfound intheU.S.,Mexico,Brazil,Koreaandregionsbordering *Correspondence:jbjones@ufl.edu Contributedequally1DepartmentofPlantPathology,UniversityofFlorida,Gainesville,FL,USA FulllistofauthorinformationisavailableattheendofthearticlePotnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 2011Potnisetal;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommons AttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,andreproductionin anymedium,providedtheoriginalworkisproperlycited.

PAGE 2

theIndianOcean,andDgroupstrainsarefoundinthe formerYugoslavia,Canada,CostaRica,U.S,Braziland regionsoftheIndianOcean[4-8].Threeofthefour groupsexceptforDwereoriginallydescribedasasingle pathovarwithin Xanthomonascampestris andreferred toas X.campestris pv. vesicatoria .TheDgroupconsistedofastrainisolate dfromtomatothathadbeen designated Pseudomonasgardneri formanyyears[9] althoughDeLeyprovidedevidenceforplacementinthe genus Xanthomonas [10].Subsequentlyallfourgroups wereclassifiedasseparate speciesonthebasisofphysiologicalandmolecularcharacteristicsasfollows: Xanthomonaseuvesicatoria (groupA), Xanthomonas vesicatoria (groupB), Xanthomonasperforans (groupC), and Xanthomonasgardneri (groupD)[11]. Basedon16SrRNAanalysis, X.euvesicatoria strain 85-10(Agroup)and X.perforans (Cgroup)together formamonophyleticgroup,whereas X.vesicatoria (B group)and X.gardneri (Dgroup)clustertogetherwith X.campestris pv. campestris ( Xcc ) Xcc strain33913[11]. Recently,aphylogenetictreewasconstructedbasedon MLST(multi-locussequencetyping)dataforA,B,C andDgroupstrainsandotherxanthomonads[12].The MLSTapproachrevealedthat X.euvesicatoria and X.perforans formagroupalongwith X.citri strain306 ( Xac ). X.gardneri ismostcloselyrelatedto X.campestris pv. campestris strainswhile X.vesicatoria formsa distinctclade[12].Thisdiversityamongthefourgroups makesthe Xanthomonas -tomato/peppersysteman excellentexampletostudypathogenco-evolution,as distinctspecieshaveconvergedonacommonhost. Whileintegratedmanagementapproachesforcontrol ofbacterialspotdiseaseareavailable,thedevelopment ofhostresistanceismoreeconomicalandenvironmentallybenignforthecontrol ofthedisease[13,14].Host resistancemayalsoberequiredtoreplacethelossof someintegratedmanagementtools.Useofcopperand streptomycinspraysovertheyears,forexample,hasled tothedevelopmentofresistantstrains[5].Atthesame time,geneticresistancehasbeenlostduetoraceshifts inpathogenpopulations[15-17].Designingnewand possiblydurableresistancerequiresknowledgeofpathogenicityfactorspossessedbythefourgroups. Manycandidatepathogenicityfactorshavebeenidentifiedinstrainsof Xanthomonas .Anumberofvirulence factorsareemployedbyxanthomonadstogainentry intoleaforfruittissue,andgainaccesstonutrients, whilesimultaneouslyoverc omingorsuppressingplant defenses.Differentsecretionsystemsandtheireffectors havebeenshowntocontributetothevirulenceofplant pathogens.ThetypeIIIsecretionsystem(T3SS) encodedbythe hrp (HypersensitiveResponseand Pathogenicity)genecluster[18,19]andtypeIIIsecreted effectorshavebeenwidelystudiedfortheirrolein hypersensitivityandpathogenicity.Effectorscommon betweenstrainsarebelievedtoberesponsibleforconservedvirulencefunctionandavoidanceofhostdefense. Differencesineffectorsuiteshaveevolvedinclosely relatedstrainsofplantpathogensandstrain-specific effectorsmayhelptoescaperecognitionbyhost-specific defenses[20-25].Importantinsightsintopathogenicity mechanismsof X.euvesicatoria strain 85-10 (hereafter, Xcv )havebeenobtainedwithitsgenomesequence[26]. Herewereportdraftgenomesequencesoftypestrains oftheotherthreebacterialspotpathogenspecies: X.vesicatoria strain1111(Xv 1111)(ATCC35937), X.perforans strain91-118( Xp 91-118),and X.gardneri strain101( Xg 101)(ATCC19865).Wehaveannotated andanalyzedpredictedpathogenicityfactorsinthedraft genomes.Additionally,wehaveinvestigateddifferentiationbetweenxanthomonadsthatmightexplaindifferencesindiseasephenotypesandinhostrange.ResultsandDiscussionDraftgenomesequencesof Xv strain1111, Xp strain91118and Xg strain101wereobtainedbycombining Roche-454(pyrosequencing)andIlluminaGA2(Solexa) sequencingdataInitially,wesequenced Xv strain1111(ATCC35937) (hereafter Xv ), Xp strain91-118(hereafter Xp )and Xg strain101(ATCC19865)(hereafter Xg )by454 pyrosequencing[27]. Denovo assemblyusingNewbler assemblerresultedin4181,2360and4540contigs, respectively,for Xv Xp and Xg ,withapproximately10foldcoverageforeachstrain(Additionalfile1:Table S1).Manypathogenicitygenes,includingtypeIIIeffectors,existedintheformoffragmentsgiventherelatively lowcoverageofthe454-basedassembly.Morecomplete assemblieswereobtainedusingIlluminasequencing [28]. Denovo assembliesofaround100-foldcoverage wereconstructedfromtheIlluminadataaloneorcombinedwithpre-assembled454longreadsusingCLC GenomicWorkbench[29].Combined454andIllumina sequencingproducedamuchbetterassemblythan eithertechnologyalone(Table1).Therefore,combined assemblieswerechosenforallsubsequentanalyses.The averagecontigsizeinthecombined454andIllumina assemblieswasaround18kbfor Xv and Xp ,and10kb for Xg .TheN50(minimumnumberofcontigsneeded tocover50%oftheassembly)valueswere37and40for Xv and Xp ,respectively,and83for Xg indicatingthat finalassembliesconsistofafewlargecontigsallowing reasonablyaccuratewholegenomecomparisons. Thethreestrainswerededucedtocontainplasmidsas evidencedbythepresenceofgenesthatareknownto beinvolvedinplasmidmaintenance(e.g. parB/F genes). Wehaveusedadjacencytosuchgenestoinferoccurrenceofcertainothergenesonplasmids.Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page2of23

PAGE 3

Relationshipsofthestrainstootherxanthomonadsusing wholegenomecomparisons16SrRNAanalysisandMLST-basedphylogeneticanalysisshowedthediversityamongthefourbacterialspot species.Wecarriedoutphylogeneticanalysisbasedon orthologousprotein-codinggenesfromdraftgenomes andreferencexanthomonads(Figure1).Wholegenome comparisonswereperformedusingtheMUMiindex [30]toassesspairwisedistancebetweenthedraftgenomesandavailablereference Xanthomonas genomesas showninthephylogenetictreeandthedistancematrix (Additionalfile2:Fig.S2).Anotherprogram,dnadiff, basedonnucmer[31]showedtheextentofhomologies amongthesharedregionsofthegenomesbypairwise comparisons(Additionalfile3:TableS3).Allofthe methodsyieldedconsistentresults:wewereableto ascertainthatamongthethreenewlysequencedstrains inrelationshiptotheprevi ouslysequencedstrains, Xp and Xcv formtheclosestpair,whichisinturnclosest to Xac .Next, Xg isclosestto Xcc ,with Xv forminga cladewith Xg andthe Xcc speciesgroup(Figure1, AdditionalfilesS2andS3).Fourxanthomonadsshowvariationintheorganizationof thetypeIIIsecretiongeneclustersAnnotationoftherespectivetypeIIIsecretiongeneclusters,or hrp genesshowedthat Xp hasanalmostidentical andsyntenic hrp clustertothatof Xcv (Figure2).The mostnotabledifferenceisthat hpaG and hpaF encode thefusionproteinXopAEin Xp ,whiletheyarepresent asseparategenesin Xcv .Adjacenthypotheticalprotein XCV0410(126aminoacidprotein)isabsentfrom Xp Xv and Xg showgreatersimilaritytothecore hrp cluster genesof Xcc thantothatof Xcv Xv and Xg contain hrpW associatedwiththe hrp clusterasin Xcc .Additionally, xopD in Xv and Xg isnotassociatedwiththe hrp clusterasin Xcc (referredtoas psv in Xcc). PsvA shows 74%and84%sequenceidentitytotherespectivehomologsfrom Xv and Xg XopA ( hpa1 )fromXcv seemstobe absentfrom Xv and Xg .Interestingly,wefoundanovel candidateeffectorgene(named xopZ2 )upstreamof hrpW in Xv and Xg (Seebelow,Additionalfile4:Fig.S4). Table1Generalsequencingandcombined(454andsolexa) denovo assemblyfeaturesofdraftgenomesof Xv Xp and XgXanthomonasvesicatoria ( Xv ) Xanthomonasperforans ( Xp ) Xanthomonasgardneri (Xg) Numberofcontigs296291552 N50* 37 40 83 Meancontiglength 18,686 18,082 10,014 Longestcontig 153,834 133,836 88,536 Totallengthofcontigs5,531,090 5,262,127 5,528,125*N50-numberofcontigsthatcover50%ofthegenomeassembly. X p Xcv Xac 0.009 0.015 Xo o 0.012 0.040 Xv 0.025 0.044 Xg 0.011 0.040 Xcc 0.023 0.047 Sm 0.182 0.259 0.02 Figure1 Maximumlikelihoodtreebasedonorthologous genesfromxanthomonadsand Stenotrophomonas Concatenatedaminoacidsequencesoftheorthologousgenesfrom fourbacterialspotpathogenstrainsalongwithothersequenced xanthomonadswereconsideredintheanalysis. Stenotrophomonas maltophilia ( Sm )wasusedasanoutgroup.Theevolutionaryhistory wasinferredusingtheMaximumlikelihoodmethod.Thetreeis drawntoscale,withbranchlengthscorrespondingtothe evolutionarydistances.Theevolutionarydistanceswerecomputed usingtheMaximumCompositeLikelihoodmethodandareinthe unitsofthenumberofbasesubstitutionspersite. Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page3of23

PAGE 4

Finally,the hrp-associatedeffector xopF1 isconserved andintactinallfourtomatoandpepperpathogens.Areportergeneassayconfirmstranslocationofnovel typeIIIeffectorsWeidentifiedandannotatedT3SSeffectorsfromthe threenewlysequencedxanthomonads(SeeMethods). Severalcandidateeffectors,whichhadnotyetbeen experimentallyconfirmedinxanthomonads,andcandidateeffectorswithplausibletranslocationmotifswere identified(Tables2,3,and4).Corroborativeevidence forT3SS-mediatedtranslocationofthecandidateeffectorswasassessedbyconstructingfusiongeneswiththe C-terminalendofAvrBs2codingsequence(avrBs262574aa)inarace6strainof X.euvesicatoria .Translocationwasmeasuredinpeppercv.ECW20R,containingtheresistancegene Bs2 (Additionalfile4:Fig.S4). Genes xopAO,xopG,xopAM,andXGA_0724 (belongingtothe avrBs1 classofeffectors),ofwhich homologswerepreviouslyfoundin Pseudomonas species,weredemonstratedtodirectAvrBs2-specific hypersensitivereactionsinECW20R(Tables3, Table4,Additionalfile4:Fig.S4).Anothercandidate effectorgene xopZ2 ,associatedwiththehrp clustersin Xv and Xg (Figure2),wasalsofunctionalinthe AvrBs2-basedassay.Thus,weidentifiedfiveeffectors ( xopAO xopG xopAM xopZ2 ,XGA_0724)thathave notbeenpreviouslyrecognizedin Xanthomonas and showedtheirfunctionality.Coreeffectorsamongfourxanthomonadsgiveinsight intoinfectionstrategiesofthepathogenComparingthedraftgenomesequencesofthethree xanthomonadswiththatof Xcv allowedustoidentify thecoreeffectorsconservedinallfourstrainsaswellas strain-specificeffectors(Tables2,3,and4). Figure2 ComparisonoftypeIIIsecretionsystemcluster,itsassociatedtypeIIIeffectorgenesand helpergenesofthreedraft genomeswithalreadysequencedxanthomonads.TypeIIIsecretiongeneclustersinfivestrainsareshown.Boxesofthesamecolour indicateorthologousgenes.Genesofspecialinterestdiscussedinthepaperarelabeled. Xp hasnearidentical hrp clusteras Xcv ; Xv and Xg containmosaic hrp clusterwithorganizationandgenecontentsimilarto Xcc ,butassociatedeffectorsaresimilarto Xcv alongwithnovel effectorgeneassociatedwiththecluster. Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page4of23

PAGE 5

Table3TypeIIIeffectorsspecifictoeachspeciesEffector LocustagsEffector homolog Pfamdomains/biochemicalmotifs Comments/Reference Effectorsspecificto Xv XopJ2XVE_4840(partial); XVE_3769(partial) AvrBsTC55-familycysteineproteaseorSer/Thracetyltransferase [40] XopAG XVE_2415AvrGf1 [39] XopAI XVE_4756XAC3230 [25] Effectorsspecificto Xg classavrBs1XGA_0724 AvrA(84% identity) Thisstudy AvrHah1 (Fragmentedin assembly) XGA_4845/XGA_3187AvrBs3Transcriptionalactivator,nuclearlocalizationAvrBs3presentinfew euvesicatoriastrains[41]. XopAO XGA_1250AvrRpm1 (61%identity) Thisstudy XopAQ XGA_2091Rip6/rip11 Noknowndomains [46] XopASXGA_0764/0765HopAS1 Noknowndomains Thisstudy Effectorsspecificto Xp XopC2 XPE_3585Rsp1239 Haloaciddehalogenase-likehydrolase [24] XopJ4 XPE_1427AvrXv4SUMOprotease(experimental);YopJ-likeserine threonineacetyltransferasedomain(predicted) [38,105] XopAF XPE_4185AvrXv3 Transcriptionalactivatordomain [37] XopAE XPE_2919HpaF/G LRRprotein [24] Effectorsspecificto Xcv AvrBs1 XCVd0104AvrBs1 [26] XopC1 XCV2435 XopCPhosphoribosyltransferasedomainandhaloacid dehalogenase-likehydrolase [105] XopJ1 XCV2156 XopJC55-familycysteineproteaseorSer/Thracetyltransferase [105] XopJ3 XCV0471AvrRxvC55-familycysteineproteaseorSer/Thracetyltransferase [26] XopO XCV1055 Unknown [26] XopAA XCV3785 ECFEarlychlorosisfactor,proteasome/cyclosomerepeat [26] XopAI XCV4428AvrRxo1 [26] Table2CoreeffectorspresentinallfourtomatoandpepperxanthomonadsEffector class XcvXvXpXg PfamdomainsReferences AvrBs2XCV0052XVE_4395XPE_2126XGA_3805Glycerophosphoryldiesterphosphodiesterase[104] XopDXCV0437XVE_2372XPE_2945XGA_3151C48-familySUMOcysteineprotease(Ulp1 proteasefamily);EARmotif [105] XopF1XCV0414XVE_3220XPE_2922XGA_2763-[105] XopKXCV3215XVE_0354XPE_1077XGA_3563-[106] XopLXCV3220XVE_0359XPE_1073XGA_0320LRRprotein[107] XopNXCV2944XVE_0564XPE_1640XGA_0350ARM/HEATrepeat[108] XopQXCV4438-XPE_0810XGA_0949InosineuridinenucleosideN-ribohydrolase[105] XopRXCV0285XVE_0593XPE_1215, XPE_3295 XGA_1761-[106] XopXXCV0572XVE_3610 XVE_3609(partial) XPE_1488 XPE_1553 XGA_3272(secondcopy withframeshift) -[109] XopZ1XCV2059+(*)XPE_2869+(*)-[106] XopADXCV4315/ 4314/4313 XVE_4177XPE_4269XGA_0755SKWPrepeatprotein[110]*XvandXgcontaineffector xopZ2 belongingtothesamefamily xopZ .Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page5of23

PAGE 6

Atleast11effectorgenesformacoresetofcommon effectorsforxanthomonadsinfectingtomatoandpepper (Table2).Ofthese11,eighteffectorgenes( avrBs2, xopK,xopL,xopN,xopQ,xopR,xopX and xopZ )were foundtobeconservedinallsequencedxanthomonads includingthethreedraftgenomespresentedherewith theexceptionsof X.albilineans and X.campestris pv. armoraciae .Thesegenesmightbenecessaryformaintainingpathogenicityofthesexanthomonadsinawide rangeofhostplants.XopNhasbeenreportedtosuppressPAMP(pathogen-associatedmolecularpattern)triggeredimmunitybyinteractingwithtomatoTARK1 andTFT1[32]. XopF1 isconservedintomatoandpepperxanthomonads.Althoughahomologof xopF1 is foundin Xcc ,therespectivegeneistruncated[34]. Hence, xopF1 isapotentialpathogenicitydeterminant intomato.A xopF1 deletionmutantof Xcv didnot showanydifferenceinvirulencewhencomparedtowild type Xcv onthesusceptiblecultivarofpepperECW, suggestingXopF1isnotthelonefactorforpathogenicityof Xcv onpepper[33].Anothereffectorgene, xopD isassociatedwiththe hrp geneclusterin Xcv and Xp However, xopD appearstohavetranslocatedtoanother locationinthegenomeincaseof Xg,Xv and Xcc strains.XopDisannotatedas Psvvirulenceprotein in Xcc genome[34]andhasbeenshowntobeachimeric proteinsharingaCterminuswithXopDfrom Xcv [35]. Although xopD homologsfrom Xv and Xg aresyntenic withthe psv genein Xcc Xv and Xg haveintactfulllengthcopiesof xopD asin Xcv,indicatingthatthe x opD couldbeanothereffectorexclusivetothetomato pathogensandapossiblepathogenicitydeterminantin tomato. XopD hasbeenshowntoenhancepathogen survivalintomatoleavesbydelayingsymptomdevelopment[36].Twotandemcopiesof xopX arefoundin Xg However,onegenein Xg appearstobeinactiveduetoa frameshiftmutation.In Xp ,thetwocopiesofxopX ar e found indifferentlocationsinthegenomewithneighboringgenes,includingchaperonegene groEL ,whichis alsoduplicated.Orthologsof xopZ arealsofoundinall fourxanthomonads,with82%identityfor Xcv and Xp and35%identityfor Xg and Xv.Apartfromlow sequenceidentityin Xv and Xg ,gene-specificrearrangementsappeartohaveoccurredwithineachortholog. Weproposethattheoveralllowaminoacidrelatedness (pairwisesequenceidentitiesbelow50%)ofthiseffector Table4EffectorsspecifictoparticulargroupsofspeciesEffector class LocustagsPfamdomainsComments/References Effectorscommontoallpepperpathogens Xv Xcv and Xg XopE2XCV2280,XVE_1190, XGA_2887 Putativetransglutaminase[114] XopGXCV1298,XVE_4501, XGA_4777 M27familypeptidaseclostridiumtoxin Thisstudy Effectorscommonto Xv Xg butabsentfrom Xp and Xcv XopAMXVE_4676,XGA_3942 Thisstudy HrpWXVE_3222,XGA_2761 Pectatelyase HrpWassociatedwithhrpcluster,Maynotbe T3SE[111] AvrXccA1XVE_5046,XGA_0679LbHdomaincontaininghexapeptiderepeats(X-[STAV]-X[LIV]-[GAED]-X)-acyltransferaseenzymeactivity MaynotbeT3SE[112] XopZ2XGA_2762,XVE_3221 Notknown Thisstudy;Associatedwithhrpcluster. Effectorscommonto Xg and Xcv butabsentfrom Xp and Xv XopBXGA_4392,XCV0581 [113] Effectorscommonto Xp and Xcv butabsentfrom Xg and Xv XopE1XPE_1224,XCV0294 Putativetransglutaminase [114] XopF2XPE_1639,XCV2942 [105] XopIXPE_3711,XCV0806 F-boxdomain [115] XopPXPE_3586,XPE_4695 (Partial),XCV1236 [105] XopVXPE_4158,XCV0657 [106] XopAKXPE_4569,XCV3786 NotconfirmedtobeeffectorinXanthomonas; Homologofeffectorin Pseudomonas XopAPXPE_1567,XCV3138 LipaseclassIII 45%identitytohomologin Xp;Homologofrip38 fromR.solanacearumRS1000[46] Effectorspresentin Xv and Xp butabsentfrom Xg and Xcv XopARXVE_3216,XPE_2975 [46] Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page6of23

PAGE 7

in Xv and Xg warrantsassigningtheproteinstoanew familywithinthe xopZ class,namedxopZ2 ,whilethe orthologsfrom Xcv and Xp belongtofamilyof xopZ1 as originallydescribedin Xoo andassupportedbypairwise sequenceidentitiesofatleast60%(seeabove,Figure2, Table4).Effectorsuniqueto Xp mightberesponsiblefor restrictinggrowthonpepperXp ispathogeniconlyontomato.Theavirulencegene, avrXv3 ,presentin Xp ,waspreviouslyshowntoelicitan hypersensitiveresponse(HR)inpeppercv.ECW[37]. An avrXv3 knockoutmutantof Xp isnotvirulentin peppercv.ECWindicatingthatotherfactorsareassociatedwithhostspecificity.Comparingeffectorrepertoiresofthepepperpathogens Xg Xcv ,and Xv with Xp mayprovidecluestothefactorsthatareresponsiblefor reducedvirulence(Table4).Besides avrXv3 ,theonly effectorspresentin Xp andabsentorinactivein Xg Xv and Xcv are xopC2 xopAE and xopJ4 ( avrXv4 )(Table 3).Thegene avrXv4 isabsentfromothersequenced xanthomonadsandshowsgene-for-geneinteraction withthe Xv4 resistancegenefromthewildtomatorelative Solanumpennellii butdoesnotcontributeto restrictedgrowthof Xp onpepper[38].Theeffector xopC2 isahomologoftheeffector rsp1239 from Ralstoniasolanacearum GMI1000and xopAE encodesanLRR proteinwithhomologytothe R.solanacearum effector PopC.Bothgenes, xopC2 and xopAE ,aretruncatedin Xcv .Therefore,thesetwoeffectorsmaytriggerimmunityinpepper.Interestingly, Xp containsaparalogof xopP .Thetwocopiesarefoundnexttoeachotherin thegenomeandshare75%identityattheaminoacid level.Thesecondcopyisnexttothecandidateeffector xopC2 ,whichisuniqueto Xp amongtomatoandpepperpathogens.Effectors xopC2 and xopP maybothact torestrictgrowthinpepper.Moreover,thereareatleast twoeffectors, xopE2 and xopG ,presentinthepepper pathogens Xcv Xv and Xg butabsentfrom Xp .These effectorsmaybeessentialpathogenicityfactorsin pepper.Species-specificeffectorsXv possessestwouniqueeffectorgenes, xopAG ( avrGf1 ) and xopAI (Table3).Aphylogeneticanalysisof xopAG showedthat xopAG from Xv iscloselyrelatedto xopAG from X.citriAw,whichhasbeenshowntoberesponsible forcausinganHRongrapefruit[39].XopAIisachimeric protein,whichcontainsaconservedmyristoylationmotif atitsNterminus,likeXopJ1.Thiseffectorclassalso includesthehomolog XAC3230 from Xac aswellas XAUB_26830andXAUC_23780from X.fuscans subsp. aurantifolii strainsBandC,respectively[25].Thepresenceoftransposonsandphageelementsinclose proximityhelpstoexplaintheevolutionofthisnovel effectorin Xac byterminalreassortment[35]. Xv also containseffectorgene avrBsT ,whichisresponsiblefor thehypersensitiverespons eonpepper.Lossoftheplasmidcontaining avrBsT in Xcv strain75-3allowsthe straintocausediseaseonpepper[40]. Xg containsatleasttwoeffectors, avrHah1 (an avrBs3 -likeeffectorgene)and xopBas does Xcv ,and sharesequenceidentityof82%and86%respectivelyto thecorrespondingeffectorsof Xcv .However,AvrHah1 appearstospecifyadifferentphenotypewhencompared toavrBs3from Xcv AvrHah1 wasshowntoberesponsibleforincreasedwatersoakingonpepperECW-50Rand 60R,whereas Xcv strainscarrying avrBs3 showaphenotypethatconsistsofsmallraisedflecklesionsonpepper [41].Anothereffectorgene, xopB ,hasaPIPboxatthe 5 endin Xcv ,whereasthehomologin Xg doesnotcontainaPIPbox.Neighboringgenesto xopB inthe respectivestrainsarecompletelydifferentbetweengenomes,suggestinglackofsyntenybetweenthetwospeciesinthisregion(Table4). XopB from Xg is92% identicalattheaminoacidleveltothehomologin Xcv Deletionmutantsof xopB from Xcv didnotshowany differenceinvirulence,indicatingitdoesnotcontribute significantlytovirulence[42].However, xopB maycontributetovirulencein Xg .Wealsoidentifiedeighteffectorgenesthatareuniqueto Xcv (Table3).Withthe exceptionof xopAA (earlychlorosisfactor),allofthese genesbelongtoregionsoflowGCcontentcomparedto averagegenomeGCcontent(64.75%): avrBs1 (42%), xopC1 (48%), xopJ1 ( xopJ )(57%), xopJ3 ( avrRxv )( 52%), xopO (52%), xopAJ ( avrRxo1 ) (51%).Feweffectorsaresharedamongphylogeneticallyrelated groupstrainsAlthough Xp and Xcv ,and Xv and Xg formdistinctphylogeneticgroups(Figure1),relativelyfeweffectorsare sharedbetweenthesespecies.For Xp and Xcv,they shareatleastsixeffectorsxopE1,xopF2,xopP,xopV, xopAK,xopAP ,whichareabsentfromtheothertwo genomes(Table4). Xv and Xg appeartobemostclosely relatedtostrainsof X.campestris pv. campestris ,and thisrelationshipisreflectedinthesuiteofeffector genes.Infact, Xg and Xv sharefoureffectorgeneswith Xcc ,namely, xopAM,avrXccA1,hrpW and xopZ2 ,with thecaveatthat hrpW and avrXccA1 maynotfunctionas intracellulareffectors(Table4).Furthermore,thegenomicregionscontainingthesegenesaresyntenicin Xg Xv and Xcc.X.gardneri showsevidenceofeffectoracquisitionby horizontalgenetransferEffectorhomologsof avrA,hopAS1 and avrRpm1 from P.syringae pv. tomato T1and P.syringae pv. syringaePotnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page7of23

PAGE 8

B728aarefoundin Xg with79%,41%and61%identity attheaminoacidlevel,respectively(Table3,Additional file4:Fig.S4).Other X.gardneri strainsalsocontain theseeffectorsbasedonPCRscreening(datanot shown).Thesethreeeffectors,XGA_0724(belongingto avrBs1 class),XGA_0764/XGA_0765( xopAS)and XGA_1250( xopAO ),areuniqueto X.gardneri .TheC terminalregionofXGA_0724shows53%identityto avrBs1 from Xcv .Henceaccordingtothe Xanthomonas effectornomenclature[24],XGA_0724from Xg was placedundertheclass avrBs1 .XGA_0764/XGA_0765 andXGA_1250havenotyetbeenreportedtobefound inxanthomonadsandwer eassignedtonewclasses xopAS and xopAO. X.gardneri strainshavebeenfound tobeassociatedwithtomatoandhavealoweroptimum temperaturefordiseasedevelopmentsimilartothatof pathovarsof Pseudomonassyringae [43].Ahighscore byAlien_hunteranalysis[44],alongwithverylowGC content(45%forXGA_0724and48%forXGA_01250, 59%forXGA_0764/XGA_0765)andtheproximityof mobilegeneticelementsprovidesevidenceforhorizontalgenetransfer(Additionalfile5:TableS5).Effector xopAS appearstobeseparatedintotwoORFs XGA_0764andXGA_0765byinternalstopcodon.The functionalityofeffector xopAS needstobeconfirmedby inplanta reportergeneassay. AvrA of P.syringae pv. tomato PT23wasshowntocontributetovirulenceon tomatoplants[45].AcquisitionofXGA_0724by Xg mighthaveconferredincreasedvirulenceontomato. AvrRpm1 from P.syringae pv. syringae possessesamyristoylationmotif,whichisabsentfromhomologsin Xg Thismodificationin Xg mighthavebeenacquiredto escapehostrecognition.Anothercandidateeffector gene, xopAQ,in Xg isfound68bpsdownstreamofa perfectPIPbox.Thegeneshows65%identityatthe aminoacidlevelto rip6/11 ,anoveleffectorfrom R.solanacearum RS1000[46].AllfourxanthomonadscontainAx21codinggenebut only Xcv containsafunctionalsulfationgeneThe ax21 (activatorofXA21-mediatedimmunity)gene isconservedamong Xanthomonas speciesandispredictedtoencodeatypeI-secretedproteinthatmay serveasaquorumsensingsignalingmolecule[47].A 17-aminoacidsulfatedpeptidefromtheN-terminal regionof Xanthomonasoryzae pv. oryzae ( Xoo )Ax21 (axYS22)wasshowntobindandactivatetheXA21 receptorkinasefromrice,demonstratingthatAx21isa conservedPAMPthatcanactivateplantimmunesignaling[48].The ax21 geneispresentin Xcv (93%identity with Xoo PXO99protein), Xp (94%), Xv (91%),and Xg (88%).TheaxYS22peptideis100%conservedin Xcv, Xp and Xv ,whilein Xg thereisachangefromleucineto isoleucineatresidue20;thisisunlikelytoalterthe activityofthepeptide,sin cechangingthisresidueto alaninehadnoeffectonrecognitionbyXA21[48]. RecognitionofaxYS22bytheXA21receptorrequires sulfationoftyrosine22,whichrequirestheputativesulfotransferaseRaxST.Incontrastto ax21 ,the raxST geneismorevariableinthesegenomes,whichisconsistentwithareportofsequencedifferencesinthisgene among Xoo strains[49].Furthermore,in Xp ,thereisa single-nucleotideinsertionatposition65,causingaframeshiftmutation.The Xv and Xg genomesdonotcontain raxST ;therefore,the ax21 geneproductsmaybe nonfunctionalinthesestrains.Thesefindingshave implicationsforthefurtherstudyoftheroleofAx21in quorumsensingandvirulence,aswellasfortheusefulnessoftheXA21receptortoconferresistanceto xanthomonadsincropplants.TwotypeIIsecretionsystemsareconservedinallfour Xanthomonas genomesMostcell-walldegradingenz ymes,suchascellulases, polygalacturonases,xylanases,andproteases,are secretedbyatypeIIsecretionsystem(T2SS).TheXps T2SS,presentinallxanthomonads,hasbeenstudiedfor itscontributiontovirulencein Xcc and Xoo [50,51]. AnotherT2SScluster,knownastheXcssystem,is foundonlyincertainspeciesof Xanthomonas,e.g.Xcc, Xac ,and Xcv .TheXpssystemsecretesxylanasesand proteasesandisundercontrolof hrpG and hrpX [52], indicatingdifferentialregulation.BothXpsandXcssystemsarepresentinallthreedraftgenomes.Xanthomonadspossessdiverserepertoiresofcell-wall degradingenzymes,whicharepresentindiverse genomicarrangementpatternsEachspeciesof Xanthomonas hasitsowncollectionof genesencodingendoxylanases,endoglucanases,andpectatelyaseswhichcontribut etocellwalldeconstruction duringpathogenesis.Weh avecomparedtheserepertoiresfromthethreedraftgenomesandotherxanthomonadsasdetailedinTable5.Thegenesaredesignated fordifferentfamiliesofglycosylhydrolases(GH)and polysaccharidelyases(PL)thatincludetheenzymesthat cleaveglycosidicbondsinthestructuralpolysaccharides ofplantcellwalls. Genesencodingsecretedendoxylanasesregulatedby thexps geneshavebeendescribedfortheircontributionstovirulence,includingXCV0965[52]encoding GH30endoxyalanase.TheGH30familycatalysesthe cleavageofmethylglucuronoxylansinthecellwallsof monocotsanddicotsata b -1,4-xylosidicbondpenultimatetoonelinkingthexyloseresiduethatissubstituted byan a -1,2-linked4-O-methylglucuronateresidue [53,54].Suchanenzymesecretedby Erwiniachrysanthemi generatesoligosaccharidesthatarenotPotnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page8of23

PAGE 9

assimilatedforgrowth,suggestingafunctioninwhichit contributestocellwalldeconstructionforaccesstopectatesforgrowthsubstrate[53].Itisinterestingtonote theorthologousgenesencodingGH30enzymesare absentin Xg and Xv,withatruncated xyn30 genein Xac .Onthebasisofsequencehomology, xyn30 genes mayalsocontributetovirulencein Xoo,XccandXp ThemorecommonGH10endoxylanases,whichoccur inseveralbacterialandfungalphyla,havebeenimplicated inthevirulenceofplantpathogenicbacteriaandfungi [55,56].In Xoo ,deletionofthegeneencodingaGH10 xyn10B resultedindiminishedvirulence[57].All sequenced Xanthomonas genomescontaineithertwoor threecopiesof xyn10 genes,allofwhicharewithinagene clusterthatmaycompriseasingleoperon(Figure3).The GH10endoxylanasesarethebeststudiedofallofthexylanases,andstructure/functionrelationshipsmaybeinferred onthebasisofgenesequence.Theactionofthese enzymesonglucuronoxylansgeneratesxylotriose,xylobiose,andsmallamountsofxylosethatgenerallyserveas substratesforgrowth.Alsogeneratedismethylglucuronoxylotriose,thatisformedtotheextentthatxylose residuesinthe b -1,4xylanbackbonearesubstitutedwith a -1,2-linked4-O-methylglucuronateresidues[58]. Anadjacentgeneclusterinanoppositeorientation contains agu67 geneencodingaGH67 a -glucuronidase thatservestocatayzetheremovalof4-O-methylglucuronatefromthereducingterminusofmethylglucuronoxylotriose.Thisactivityprovidesasynergisticfunctionto theoverallxylanolyticproce sstogeneratexylotriose, whichisconvertedtoxylosebyxylanasesandxylosidasesforcompletemetabolism[59].Thecoregulationof operonsencodingXynBandAgu67enzymesoccursasa logicalconditiontocoordin ateexpressionofgenesthat encodetheseandadditionalenzymesthatcollectively processglucuronxylansandglucuronoarabinoxylansfor completemetabolism.Theaccessoryenzymesandtransportersnecessaryforthefunctionoftheseenzymesare embeddedwithintheseoperonsinGrampositivebacteria[60-62]andsharesimilaritiesnotedherewith Xanthomonas spp..Theseincludethegenesencoding twoglycohydrolases,a b -xylosidaseandan a -L-arabinofuranosidase.Alsoincludedinthisclusteraregenes encodingenzymesforintracellularmetabolismof Table5RepertoireofcellwalldegradingenzymesinxanthomonadsGenenameFamilyEnzymaticfunction XpXacXcvXvXgXcc strain33913 Xoo strain KACC Xylanases xyn10A GH10Endob-1,4-xylanase EC:3.2.1.8 201442544360233711724118 4429 xyn10B 201642524358-4428 xyn10C 202042494355233303414115 aguA GH67 a -glucuronidase EC:3.2.1.139 431842274333471224734102 4419 xyn51A GH51 b-D-Arabino-furanosidase EC:3.2.1.55 0180128613351029/103023031191 1317 xyn5A GH5Endob-1,4-xylanase EC:3.2.1.8 46820933/34partial0965--0857 3618 Glucanases cel8A GH8Endo-1,4b-Dglucanase1965351636410432cel9A GH9 234525222704132705882387 Pectatelyases pel1A PL1 Pectatelyase EC:4.2.2.2 384135623687193340240645 0821 pel1B 156329863132351208932815 pel1C -23732569-pel3A PL3 Pectatelyase EC:4.2.2.2 -2922-322227611219 pel4A PL4Rhamno-galacturonanlyase EC:4.2.2.197535053632259245313377/78/791078 pel9A PL9 Pectatelyase EC:4.2.2.2 -2278192718532265 pel10A PL10Pectatelyase EC:4.2.2.2 -406951240122 -Differentcellwalldegradingenzymes,suchasxylanase,pectatelyase,glucanases,werecomparedfortheirrepertoiresamongalreadysequencedxanthomonads includingourdraftgenomes.Genesidentifiedbytheirlocustagsintherespectivegenomes.Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page9of23

PAGE 10

glucuronateandxylose,includingglucuronateisomerase;xyluloseisomerase;D-mannonatedehydratase;and D-mannonateoxidoreductase.Genesencodingmannitol dehydrogenaseandthehexur onatetransporter,aswell astheTonB-dependentreceptorandLacItranscriptionalregulator,flankthesetwooperons. Thearrangementandcontentofxylanolyticenzymes differentiate Xanthomonas speciesintothreegroups (Figure3).Here,weproposeacommonnomenclaturefor xylanases,thegenesforwhichhavebeenannotatedinthe sequencedgenomes.Membersofthefirstgroupare Xac, Xcv and Xp inwhichallthreegenesencodingGH10 endoxylanases( xyn10A xyn10B and xyn10C )arepresent, andwithadditionalgenesfurtherdownstreaminthiscluster.Membersofthesecondgroupare Xcc Xv and Xg in whichgenesencodingtwoofthethreeendoxylanasesare present( xyn10A and xyn10C )andwhereoneormoreof thethedownstreamgenesareabsent. Xoo strainsrepresentathirdgroupinwhichadifferentsetoftwoendoxylanaseencodinggenesarepresent( xyn10A and xyn10B ) andwherethe b -galactosidaseandgluconolactonasegenes flanking xyn10C areabsent.Itisnoteworthythattheorganizationofgenesintheclusterencodingthe a -glucuronidaseisconservedacross Xanthomonas species.GenesinvolvedinseveralTypeIVsecretionsystemsare presentingenomesandplasmidsLike Xcv ,thetomatopathogens, Xg Xv and Xp ,also appeartocontainmorethanonecopyofatypeIV secretionsystem(T4SS)cluster(Figure4A,B).Two T4SSclusters(VirandDot/Icmtype)arepresentin Xcv ,andgenesbelongingtobothofthesesystemsare foundonplasmids[26].TheDot/Icmtypesystemis absentfrom Xv, Xp and Xg In Xv and Xp ,genesforoneT4SSareonaplasmid andthesecondoneonthechromosomewhilein Xg twoT4SSgeneclustersareonaplasmidandoneison thechromosome.ThetwoT4SSclustersonplasmidsof Xg donotshowanysimilaritytothegenesforT4SSin Xac Xcv, Xcc and Xoo .OfthetwoT4SSclustersin Xg Figure3 Xylanaseclusterorganization .Threetypesofclusterorganizationscanbefoundwithinxanthomonads.A)Foundin Xac,Xcv and Xp containingthreeendoxylanasegenes xyn10A,xyn10B and xyn10C ;B)Foundin Xcc,Xv and Xg containingtwoendoxylanases xyn10A and xyn10C ; andC)Foundin Xoo containing xyn10A and xyn10B withinendoxylanaseoperon. Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page10of23

PAGE 11

oneisalsofoundin Xv and Xp .Thisclusterappearsto beexclusivetothesethreetomatopathogens(Figure 4A).Thegenesbelongingt othisclustershowlow(3045%)identitytotheT4SSclustersfrom Ralstonia,Burkholderia,Bradyrhizobium ,and Stenotrophomonasmaltophilia .Theotherclusterfrom Xg ,whichisabsent from Xv and Xp ,showsveryhighidentity(98%)and syntenytotheT4SSclusterof Burkholderiamultivorans andaround89%identitytoaT4SSclusterof Acidovoraxavenae subsp. citrulli (Figure4B). ApartfromtheplasmidborneT4SSgenes, Xcv also containsaportionofatypeIVsystemclusteronthe chromosomeandconsistsof VirB6 VirB8,VirB9,VirD4 genes.Thischromosomalclusterisflankedbyatransposonelement(IS1477)thatmightindicateitshorizontal genetransfer. Xp Xg and Xv genomescontainacomplete chromosomalT4SSclustershowinghighidentitytothe T4SSchromosomalclustersfrom Xcc (Figure4C).TypeVsecretedadhesinsfunctioninsynergismduring pathogenesisDifferentadhesinshavebeenshowntofunctionatdifferentstagesoftheinfectionprocessstartingwith attachment,entry,latersurvivalinsidehosttissueand colonizationbypromotingvirulence[63,64].FhaB hemagglutinin,importantfor leafattachment,survival insideplanttissueandbiofilmformation,ispresentin allfourtomatopathogens.In Xcv fhaB isdividedinto twoseparateopenreadingframes,XCV1860and XCV1861,withthetwo-partnersecretiondomainsbeing presentinXCV1860.Sequencealignmentindicatesthat fhaB ispossiblyinactivatedin Xcv bytheinternalstop codonthatseparatesXCV1860fromXCV1861.Inthe caseof Xoo PXO99A,the Xanthomonas adhesin-like proteinsXadAandXadBpromotevirulencebyenhancingcolonizationoftheleafsurfaceandleafentry throughhydathode[64].Asin Xcv and Xac, Xp encodes Figure4 TypeIVsecretionsystem .A)SchematicrepresentationoftypeIVsecretionsystemclustercommonto Xp, Xv and Xg (Plasmidborne); B)TypeIVclusteruniqueto Xg (plasmidborne);C)ChromosomaltypeIVclusterorganizationin Xcv Xv Xp and Xg. Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page11of23

PAGE 12

twocopiesof xadA ,while Xv and Xg possessasingle orthologof xadA asdoes Xcc .YapHandthetypeIV pilusproteinPilQwereshowntobeinvolvedinvirulencein Xoo duringlaterstagesofgrowthandmigration inxylemvessels.In Xcv, Xc ,and Xoo KACC,twocopies of yapH arepresent.Therearetwo pilQ orthologsin Xcv andonlyoneinothersequencedxanthomonads. Nexttothe fhaB and fhaC adhesingenes, hms operonis presentinthegenomesofxanthomonads,thehomologs ofwhichare pga operongenesin E.coli involvedinbiofilmformation[65].TypeVIsecretionsystemispresentin Xcv Xv and XpTypeVIsecretionsystem(T6SS)hasbeenshown recentlytocontributetohostpathogeninteractionsduringpathogenesisin Vibriocholerae Burkholderiapseudomallei and Pseudomonasaeruginosa.Hcp (Haemolysin-coregulatedprotein)andVgr(valine-glycinerepeats)proteinsareexportedbytheT6SS[66]. T6SSclusterscanbeassignedtothreedifferenttypesin xanthomonads(Table6). Xcv and Xp possesstwotypes ofT6SSs(type1and3);whereas Xv containsonlyasingletypeofT6SS,type3.Asin Xcc,thereisnoT6SS clusterin Xg (Table6,Additionalfile6:TableS6).LPSlocusdisplaysremarkablevariationinsequenceand numberofcodinggenesandshowshostspecific variationThelipopolysaccharide(LPS)biosynthesisclusterhas beenstudiedindetailin Xcc [67],whichcomprises threeregions;region1from wxcA to wxcE involvedin biosynthesisofwatersolubleLPSantigen;region2 ( gmd,rmd)codingforLPScoregenes;andregion3 from wxcK to wxcO codingforenzymesformodification ofnucleotidesugarsandsugartranslocationsystems. ThisLPSbiosynthesislocusispositionedbetweenhighly conservedhousekeepinggenes,namelycystathionine gammalyase( metB )andelectrontransportflavoprotein ( etfA),asreportedinotherxanthomonads[68].Comparisonofthisclusterfromdraftgenomestothealready sequencedxanthomonadsrevealedhighvariabilityinthe numberofgenesandtheirsequences. Xv and Xg have anidenticaltypeofLPSgeneclusterof17.7kbencoding14openreadingframes(Figure5A)whichissimilar inorganizationandsequenc eidentitytotheLPSlocus from Xcc strains.Interestingly, Xg and Xv alsocontain twoglycosyltransferasesinvolvedinsynthesisofxylosylatedpolyrhamnanasseenin Xcc [69],incontrastto glycosyltransferases( wbdA1,wbdA2 )involvedinsynthesisofpolymannanin Xcv [26].Thissuggeststhatbasic structureofO-antigenin Xg and Xv issimilarto Xcc Thethreetomato/pepperpathogens Xcv Xv and Xg haveretainedanancestraltypeofLPSgenecluster (Figure5Aand5B).Ontheotherhand, Xp hasacquired anovelLPSgeneclusterduringthecourseofevolution andiscompletelydifferentinsequenceandnumberof genesthatareencoded.In Xp ,thisLPSlocusis17.3kb longandencodes12ORFs,allofwhichareabsentin thecorrespondinggenomicregionof Xcv, Xv or Xg AlsothefirstfiveORFsflankingthe metB sideofthe LPSlocusin Xp (Figure5A,ORFscoloredinred) showedverylowornoidentitytoregion1oftheLPS Table6TypeVIsecretionclustersindifferentxanthomonadsStrain T6SS#1 T6SS#2 T6SS#3 Phosphorylation-typeregulators:Kinase/Phosphatase/Forkhead Kinase/Phosphatase/Forkhead AraC-typeregulators: AraC Xvv NCPPB702 YES / / Xcm NCPPB4381 YES / / Xaub / / YES Xauc / / YES Xac / / XAC4116-XAC4148 Xv / / YES Xp YES / YES Xcv XCV2120-XCV2143 / XCV4206-XCV4244 Xoo KACC10331 XOO3034-XOO3052 XOO3466-XOO3517 / Xoo MAFF311018 XOO2886-XOO2906 XOO3286-XOO3319 / Xoo PXO99A XOO0245-XOO0270 XOO2029-XOO2060 / Xoc BLS256 XOC2523-XOC2545 XOC1309-XOC1370 / Xg /// Xcc ATCC33913 / / / Xcc 8004 / / / Xca 756C / / / Xalb /// Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page12of23

PAGE 13

locusintheotherxanthomonads.However,theseORFs stillbelongtothesamePfamfamilies[70]thatare usuallypresentinthisregion,forexample,ABCtransportersandglycosyltransferases.Thesecondhalfofthe LPSclusterflanking etfA sideencodessixORFs,which arehomologsoftheLPSclustergenesfrom Xac Xcm and Xoo .Phylogeneticinsightbasedonconserved metB and etfA genesthatflanktheLPSlocussuggestthatthe ancestorofallthe Xanthomonas pathogensofpepper andtomatostudiedinthispaperhadthesameLPSgene cluster,howeverputativehorizontalgenetransferevents atthislocushaveledtotheacquisitionofanovelLPS geneclusterin Xp (Figure5B).Alien_hunteranalysisalso supportsthisacquisitionwithahighscoreshowingthis regiontobelongtoananomalousregion(Additionalfile 5:TableS5).Thiseventmighthaveplayedamajorrole inchangingthespecificityof Xp towardstomatoandits dominanceoveritsrelative(s)asreportedpreviously[71], similartovariantepidemicstrainof Vibriocholerae reportedtobeamajorreasonforitsemergenceandcholeraoutbreakduringthe1990 sintheIndiansubcontinent[72].Identityintermsofsequencesandgene organizationamongpepperpathogensandabsenceof thosegenesfrom X.perforans andanovelLPSclusterin thetomatopathogen X.perforans suggestaroleofthis clusterinhostspecificvariation.AnalysisofDSFcell-cellsignalingsystemRpfC/RpfGaretwo-componentsignalingfactorsandare involvedinDSF(diffusiblesign alfactor)cell-cellsignaling[73-76],knowntoco-ordinatevirulenceandbiofilm geneexpression.Thegenomesof Xv Xp ,and Xg carryan rpf ( r egulationof p athogenicity f actors)genecluster (Table7)thatisfoundinallxanthomonadsandwhich encodescomponentsgoverningthesynthesisandperceptionofthesignalmoleculeDSF[74,75].TheRpfofthe DSFsystemregulatesthesynthesisofvirulencefactors andbiofilmformationandisrequiredforthefullvirulenceof Xcc Xac, Xoc,and Xoo [77-81].RpfFisresponsibleforthesynthesisofDSF,whereas,RpfCandRpfGare implicatedinDSFperceptionandsignaltransduction [73-76].RpfCisacomplexsensorkinase,whereasRpfG isaresponseregulatorwithaCheY-likereceiverdomain thatisattachedtoanHD-GYPdomain.HD-GYP domainsactindegradationofthesecondmessengercyclicdi-GMP[82].Inadditiontogenesencodingtheseproducts, Xg and Xp have rpfH ,whichencodesamembrane proteinrelatedtothesensoryinputdomainRpfCbut whosefunctionisunknown. Xv contains rpfH butwith aninternalstopcodon,whereasfunctional rpfH ispresentin Xcv and Xcc,andtotallyabsentin Xac and Xoo .Cyclicdi-GMPsignalingCyclicdi-GMPisasecondmessengerknowntoregulatearangeoffunctionsindiversebacteria,including thevirulenceofanimalandplantpathogens[83-85]. Figure5 TheStructureandphylogenyoftheLPScluster .A)SchematiccomparisonofLPSgeneclustersdescribedinthepresentstudy. Genesconservedindifferentstrainsaregivenidenticalcolor.Genesspecifictoindividualstrainsaregivenuniquecolor. Hpopro indicatesan ORFencodingforahypotheticalprotein.Theredcolor-codedgenesin Xp genesareabsentinanyofthesequencedxanthomonads.B) Phylogenetictreebasedonconserved metB and etfA genesthatflankthevariableLPSlocus.Strainsabbreviationsareasinthemaintext.Arrow indicatesthehorizontalgenetransfereventinthelineagethatgaveriseto Xp Table7Acomparisonof rpf clusterfrom rpfB to rpfG foundacrossarangeof Xanthomonas genomesGeneName Xcc 8004 XooXcvXvXpXg rpfB XC_2331XOO2868XCV1921293405302948 rpfF XC_2332XOO2869XCV1920293205282950 rpfC XC_2333XOO2870XCV1919293005262952 rpfH XC_2334AbsentXCV19182928/2926*05242954 rpfG XC_2335XOO2871XCV1917292405222956 Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page13of23

PAGE 14

Thecellularlevelofcyclicdi-GMPiscontrolledbya balancebetweensynthesisbyGGDEFdomaindiguanylatecyclasesanddegradationbyHD-GYPorEAL domainphosphodiesterases.GGDEF,EALandHDGYPdomainsarelargelyfoundincombinationwith othersignalingdomains,suggestingthattheiractivities incyclicdi-GMPturnovercanbemodulatedbyenvironmentalcues.Anumberofproteinsinvolvedincyclicdi-GMPsignalinghavebeenimplicatedinvirulence of Xcc [86,87].Thegenomeof Xcv encodes3proteins withanHD-GYPdomainand33proteinswith GGDEFand/orEALdomains.Asinother Xanthomonas spp.,theHD-GYPdomainproteinsarecompletely conservedin Xcv, Xv Xg and Xp .Thereisalsoalmost completeconservationofGGDEF/EALdomainproteinsbetween Xcv andthreedraftgenomes,although Xv hasnoorthologofXCV1982(Additionalfile7: TableS7).Inaddition,theEALdomainprotein (XCVd0150)encodedonaplasmidin Xcv isabsentin theotherstrains.Copperresistance( cop )genesarepresentin Xv and copperhomeostasis( coh )genesarepresentinall strainsAmongthe Xcv Xv, Xp and Xg strainssequenced, Xv istheonlyoneresistanttocopperandtheonlystrain harboringasetofplasmidbornegenes,namely copL copA,copB,copM,copG,copC,copD ,and copF thatare alsopresentincopperresistantstrainsof Xac (unpublisheddata/Behlau,F.personalcommunication)and S maltophilia [88].Genes copA and copB havebeenpreviouslyannotatedascopperresistancerelatedgenes formanydifferentxanthomonadgenomesincluding Xoo Xoc Xcv Xac and Xcc .Homologsofthesegenes arealsopresentin Xv,Xg and Xp andarelocatedon thechromosome.Additionally,upstreamof copA on thechromosomeofallstrains,thereisanORFthat shareshomologywithplasmid copL .Incontrastto whathasbeenpublished,chromosomal copA and copB arenotresponsibleforcopperresistancebutlikelyfor copperhomeostasisand/ortolerance.Whilestrains harboringtheplasmid-borne cop genes,likein Xv,are resistanttocopperandcangrowonMGYagar(manitol-glutamateyeastagar)amendedwithupto400mg L-1ofcoppersulfatepentahydrate,strainsthathave onlythechromosomal cop genesasfor Xcv, Xp and Xg ,aresensitivetocopperandcanonlygrowon mediaamendedupto75mgL-1ofcopper.Nucleotide sequenceofplasmid cop genesin Xv are98%similar totheonesfoundin Xac and Stenotrophomonas whereaschromosomal copLAB from Xv is83%identicaltohomologORFsin Xcv Xg and Xp .When copL copA and copB genesfrom Xv locatedontheplasmid arecomparedtothehomologsonthechromosomeof thesamestrain,theidentityofnucleotidesequencesis 27,73,and65%,respectively.Toavoidfurtherconfusionormisinterpretation, wesuggestthatthenomenclatureofthechromosomal copL copA and copB genes inxanthomonadsshouldbechangedto cohL cohA and cohB ,respectively,referringtocopperhomeostasis genes.Newnomenclaturehasbeenadoptedinthe annotationofthedraftgenomes.Genesuniqueto X.perforans ascomparedtopepper pathogensgivecluestoitspredominanceover Xcv inthe fieldandhostspecificityThirteengeneclusterswerefoundtobespecifictothe tomatopathogen Xp whencomparedtotheotherthree strains(Additionalfile8:TableS8).Apartoftheclustersaresyntenictothegeno micregionsspecifictothe threepepperpathogens,suggestingthereplacementof thesegenomicregionsfrompepperpathogensincorrespondtotheseregionin Xp .Thesereplacedregionsin Xp mightprovidepotentialcandidatesforhostrange determinants.Mostnotab leamongtheseregionswas theLPSclustergenes(Seeabove).Othersuchregions includetheavirulencegenes avrXv3 and avrXv4 ,aTIRlikedomaincontainingprotein,oxidoreductases,and bacteriocin-likeproteinsthatwerenotfoundinany othersequencedxanthomonads.Importanceofbacteriocin-likegenesin Xp hasalreadybeenstudiedforitspredominanceinthefieldoverT1strains[89,90]. Alien_hunteranalysisshowedthatthebacteriocinBCNAregionbelongstoananomalousregionindicating possiblehorizontalgenetransferofthisregion(Additionalfile5:TableS5).Pepperpathogenicity/aggressivenessfactorsincreased in planta growthof XpComparisonofproteomesof Xv,Xg,Xcv against Xp showed68genesexclusive topepperpathogenswhich mightbecandidatevirulencefactorsonpepper(Additionalfile9:TableS9).Theseinclude16geneswith knownfunction,35codingformobilegeneticelements, and17geneswithunknownfunction/hypotheticalproteins.Outofthe16geneswithknownfunction, xopG wasconfirmedtobeatypeIIIeffectorusingthe avrBs2 reportergeneassayand6genesbelongtotheLPSbiosynthesisgenecluster.These16genesweresearched againstalreadysequencedgenomesof Xac Xcc and Xoo The wxcO gene,whichcodesforO-antigen,hasbeen identifiedtobeavirulencefactorinthe X.fuscans beanpathosystembysubtract ivehybridization[91]. Threegenes,XCV1298,XCV1839and wxcO ,wereinitiallyselectedfortheverificat ionoftheircontributionto virulenceinpepper.Indivi dualgenesalongwiththeir promoterregionswereclonedintopLAFR3andconjugatedindividuallyandincombinationinto X.perforansPotnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page14of23

PAGE 15

ME24(91-118 avrXv3),whichnolongerelicitsanHR inpepper.However, inplanta growthofME24ismore similartothatofanavirulentstrainthanthevirulent pepperstrainTED3race6.ME24transconjugantscarrying wxcO and XCV1839 incombinationshowed increased inplanta growthandalsocomparatively increasednumberoflesionsonpeppercv.ECWwhen comparedtoME24revealingthatthesetwogenesplay infactaroleinpepperpathogenicity(Figure6).Genesspecificto Xg ascomparedtoothertomato/ pepperpathogensmayexplainitsaggressivenatureon tomatoandpepperComparisonofgenesfrom Xg against Xcv Xp and Xv genes showedthepresenceof625genesspecificto Xg (Additional file10:TableS10).TheseincludefourtypeIIIeffectors ( avrBs1 member, xopAO avrHah1 xopAQ ),twenty-one genesbelongingtotheuniquetypeIVsecretionsystemclusterandassociatedgenes.Thesegenescanbespeculatedto contributetotheaggressivenatureof Xg strainsontomato andpepper. Xg alsocontainsauniquebetaxylosidasenot presentinanyotherxanthomonads.TypeIIsecretedbeta xylosidasehasbeenstudiedforitsroleinplantcellwall digestion.Moreover, Xg containsXGA_3730codingfora hemolysin-typecalcium-bindingrepeatcontainingprotein,a homologofwhichisfoundin Xylella strainswith55% sequenceidentity.In Xylella,thisgeneisannotatedasa memberofafamilyofporeformingtoxins/RTXtoxins.Its homologisalsofoundinotherplantpathogens(i.e. P.syringae pv. syringae B728aand R.solanacearum GMI1000). ThisproteinhasbeendescribedasatypeIeffectorin X.fastidiosa straintemecula(PD1506)[92].RTXtoxinfamily members,especiallyofthehemolysintype,havebeenshown tobevirulencefactorsinavarietyofcelltypesineukaryotes [93,94].Finally,ageneXGA_0603codingforlanthionine synthetase(lantibioticbiosynthesis)isfoundamongthese Xg specificgenes,ahomologofwhichisfoundin Xvm NCPPB702.LanLenzymesinpathogenicbacteriacontribute tovirulencebymodifyingthehostsignalingpathways,in mostcasesbyinactivatingMAPKs[95].Genescommontoalltomatopathogensbutabsentfrom othersequencedxanthomonadsInordertoseewhatdefinesthetomatopathogens,we comparedthefoursequencedgenomes( Xv Xp Xg and Xcv)toothersequencedxanthomonads.Wefound sevengenesthatwereconservedinallfourtomato pathogensandabsentfrommostofothersequenced xanthomonadswiththeexceptionof Xcm,Xvv Xaub and Xauc ,whichpossesshomologsforsixoutofthese sevengenes(Table8).Onlythehypotheticalprotein XCV2641seemstobespecifictothefourtomato pathogens.Thisgeneshowsonly35%sequenceidentitytoagenefrom Xvv and Xcm .Ahomologofthe hypotheticalprotein,XCV4416wasfoundin Xau ,but isabsentfromallothersequencedxanthomonads. Geneshomologousin Xcm and Xvv includetwotransposasegenesbothbelongingtothetransposase17 superfamily(XCV0615,XCV0623),XCV0041(putative penicillinamidasefragment),XCV0111(lignostilbenealpha,betadioxygenase),XCV0112(uncharacterized proteinconservedinbacteria)(Table8).Interestingly, XCV0111encodesaproteinknowntobeinvolvedin phenylpropanoiddegradation.Phenylpropanoidsare wellknownplantsecondarymetabolitesinducedduringdefenseresponseuponpathogenattack[96].It appearsthatthefourtomatopathogensalongwith Xvv and Xcm haveacquiredthisfunctiontodisarmthe basalplantdefense.Theevolutionofpathogenicityclusterscorrespondsto theMLST-basedphylogenyThecorrelationbetweentreetopologyusingMLSTand phylogenybasedonthesequencesofpathogenicityclustersandthe avrBs2 effectorgene,whichisfoundinall xanthomonads,wastested.BasedonMLST, Xp and Xcv grouptogetheralongwith Xac while Xg ismoreclosely relatedto Xcc. Xv formsadifferentcladeandismore closelyrelatedtothe Xcc group.Ascanbeseenin Figure7,phylogenybasedonMLSTiscongruentwith phylogenybasedonthepathogenicityclusters( gum hrp cluster)andbasedonthe avrBs2 effector,suggesting thatoveralltheseclusterswereverticallyinheritedfrom themostrecentcommonancestorofthesestrains.ConclusionsTheinteractionof Xanthomonas strainswithtomatoand pepperrepresentsamodelsystemforstudyingplantpathogenco-evolutionbecauseofthediversitypresent Figure6 Pepperspecificitygenesincreasing inplanta growth of Xp Inplanta growthofPM1transconjugants(combined2 [XCV1839+ wxcO ];combined3[XCV1839+ wxcO + xopG ]);PM1and peppervirulentstrainpepperrace6representedinlog(CFU/cm2of leaftissue)at0,2,4,and6dayspostinoculation. Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page15of23

PAGE 16

amongthestrainscausingbacterialspot.Althoughthe four Xanthomonas speciesinfectthesamehost,tomato, andcauseverysimilardisease,theyaregeneticallydiverse pathogens.Thecomparativegenomicanalysishasprovidedinsightsintotheevolutionofthesestrains.Whole genomecomparisonsrevealedthat Xg and Xv aremore closelyrelatedto Xcc than Xcv and Xp .Afewpathogenicityclusters,suchas hrp, xcs and xps of Xg and Xv ,were similarintermsofgeneticorganizationandsequence identityto Xcc (Figure8).However,afewpathogenicity clustersofthefourstrainsbelongingtofourphylogenetic groupsshoweddifferentevolutionaryorigins.Whilethe Table8GenespresentinallfourtomatoandpepperpathogensbutabsentfromallothersequencedxanthomonadsLocustagfor Xcv 85-10 PossiblefunctionHomologpresentinanyothergeneraGC content XCV0623Transposase17superfamilyHypoprotein-COGbelongingto transposase,inactivederivatives In Stenotrophomonas Acidovorax Xanthomonascampestris pv. musacearumNCPPB4381 0.59 XCV2641Hypotheticalprotein X.c .musacearumand X.c .vasculorum (identity37,31%respectively) 0.65 XCV4416Hypotheticalprotein Pectobacteriumcarotovorum X.fuscans pv.aurantifolii 0.57 XCV0615Transposase17superfamilyHypotheticalproteinCOG1943 (transposase,inactivatedderivates) Acidovorax X.c .musacearumand X.c vasculorum 0.62 XCV0112COG4704uncharacterizedproteinconservedinbacteria Stenotrophomonas X.c .musacearumand X. c .vasculorum 0.65 XCV0111putativelignostilbene-alpha,beta-dioxygenase-phenylpropanoid compounddegradation Stenotrophomonas Ralstonia X.c musacearumand X.c .vasculorum 0.66 XCV0041putativepenicillinamidase(fragment) Ralstonia X.c .musacearumand X.c vasculorum 0.64 Figure7 CorrelationbetweenphylogeniesbasedonMulti-LocusSequenceTyping(MLST)coregenomeandpathogenicityclusters : Concatenatedaminoacidsequencesofthesixgenes fusA gapA gltA gyrB lacF lepA fromfourbacterialspotpathogenstrainsalongwithother sequencedxanthomonadsareconsideredintheanalysis.TheevolutionaryhistorywasinferredusingtheNeighbor-Joiningmethod.The percentageofreplicatetreesinwhichtheassociatedtaxaclusteredtogetherinthebootstraptest(1000replicates)isshownnexttothe branches.Thetreeisdrawntoscale,withbranchlengthsinthesameunitsasthoseoftheevolutionarydistancesusedtoinferthephylogenetic tree.TheevolutionarydistanceswerecomputedusingtheMaximumCompositeLikelihoodmethodandareintheunitsofthenumberofbase substitutionspersite.PhylogeneticanalyseswereconductedinMEGA4. Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page16of23

PAGE 17

pepperpathogens Xcv, Xv and Xg possesssimilarLPS biosynthesiscluster,partoftheLPSclusterfrom Xp is similartotheonefrom Xac (Figure8). Xv containsfew effectors,including xopAG ( avrGf1 )andxopAI thelatter ofwhichwaspreviouslyfoundtobeuniquetocitrus pathogens Xac Xaub and Xauc [25]. Xg hasanumberof effectorshomologousto P syringae typeIIIeffectorssuggestingprobablehorizontaltransferoftheseeffectors. Xg containsauniqueT4SSalongwiththeonethatisexclusiveto Xp Xv and Xg Xp hastwoT6SSs,asfoundin Xcv Xv hasonlyoneT6SSwhichissimilartothatof Xac. Xg hasnoT6SSasseenfor Xcc (Figure8).While Xg and Xv showcloserelationshipto Xcc basedonwhole genomecomparisons,fewpathogenicityclustersmentionedaboveseemtobeconservedamongtomato/pepperxanthomonads. TypeIIIeffectorshavebeeninvestigatedfortheircontributiontopathogenicityandhost-rangespecificity.In additiontohomologsoftheknowneffectors,weidentifiednoveleffectorsinthedraftgenomes.Bycomparing effectorrepertoiresoftomatopathogens,twopossible candidatepathogenicitydeterminants, xopF1 and xopD wereidentified,ofwhich xopD isresponsiblefordelayingsymptomdevelopment,andinturn,isimportantfor pathogensurvival.Uniquegenespresentin Xg include thenoveleffectors xopAO xopAQ,xopAS andan avrBs1 memberaswellasafewothervirulencefactors,which havebeencharacterizedinotherplantpathogensand whichcouldexplaintheaggressivenatureof Xg onpepper.EachspeciescontainsatleastthreeuniquetypeIII effectors,whichcouldexplainhostpreferencesamong thestrainsandtheiraggressivenessontomato/pepper. ComparisonoftheLPSclustersbetweenthefourspeciesrevealedsignificantvariation. Xp hasacquireda novelLPSclusterduringevolution,whichmightbe responsibleforitspredominanceanditslimitedhost range.Asseenfromthe inplanta growthassayof Xp avrXv3 mutantcarryingtheLPSO-antigenfrom Xc v theLPSclusterfrompepperpathogenscanbeacontributortotheincreased inplanta growthof Xp avrXv3 mutantonpepper,butisnottheabsolutevirulence determinant.UseoftheXA21receptorsimilartothe Figure8 Adiagrammaticrepresentationofrelationshipamongbacterialspotxanthomonads, Xac and Xcc withrespecttopresenceor absenceofpathogenicityclusters .Similarcolorshadeindicateshighidentityandsimilarclusterorganization.Lowersequenceidentities comparedtothereferenceareindicatedbyfadedgrayshades.Referencestrainisindicatedbyasterisknexttothesymbol.Theabsenceof certainpartofclusterisindicatedbywhite.InthecaseofLPScluster, Xv and Xac containnovelclusterregionsintheCterminalregionwhichis indicatedbyadifferentcolor. Xac and Xcv containaplasmidbornetypeIVcluster.AlthoughitdiffersfromtypeIVApresentinotherbacterial spotxanthomonads, Xac and Xcv clusterismentionedhereundertypeIVAwithdifferentcolors.Ablankspaceindicatescompleteabsenceof geneclusterinthatparticularspecies.Amoredetailedrepresentationofindividualclusterscanbefoundinfigures2through5. Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page17of23

PAGE 18

Xoo -ricesystemin Xcv -tomato/peppercouldbeoneof thewaystoconferresistancetoxanthomonadsdueto presenceofasimilarAX21peptideandafunctionalrax systemin Xcv.Commonanduniquegenesencoding enzymesinvolvedincellwalldeconstructionarecandidatesforfurtherstudytodefinehostpreferenceand virulence. Inconclusion,comparisonofdraftgenomesobtained bynextgenerationsequencinghasallowedanin-depth studyofdiversegroupsofbacterialspotpathogensat thegenomiclevel.Thisanalysiswillserveasabasisto inferevolutionofnewvirulentstrainsandovercoming existinghostresistance.Theknowledgeofpotential virulenceorpathogenicityfactorsisexpectedtoaidin devisingeffectivecontrolstrategiesandbreedingfor durableresistanceintomatoandpeppercultivars.MethodsGenomesequencingXv, Xp and Xg weresequencedby454-pyrosequencing [27]atcoreDNAsequencingfacility,ICBR,University ofFlorida. Xanthomonas isolatesweregrownovernight innutrientbroth.Genomi cDNAwasisolatedusing CTAB-NaClextractionmethod[97]andresuspendedin TEbuffer(10mMTrispH8,1mMEDTApH8). LibrariesoffragmentedgenomicsDNAweresequenced on454-GenomeSequencer,FLXinstrumentatInterdisciplinaryCenterforBiotechnologyResearch(ICBR)at UF. Denovo assemblieswereconstructedusing454 NewblerAssembler[27].Thethreedraftgenomeswere obtainedwitharound10coverage. ForIlluminasequencing,the Xanthomonas strains werepurifiedfromsingle-colonyandgrownovernight inliquidcultures.GenomicDNAwasisolatedbyphenol extractionandprecipitatedt wicewithisopropanol,and finallydissolvedinTEbuffer.DNAwasthenpurifiedby cesiumchloridedensitygradientcentrifugationandprecipitatedwith95%ethanol,thendissolvedinTEbuffer. LibrariesoffragmentedgenomicDNAwithadaptersfor paired-endsequencingwerepreparedaccordingtothe protocolprovidedbyIllumina,Inc.withminormodifications.Thelibrariesweresequencedonthe2GGenomeAnalyzeratCenterofGenomeResearch& BiocomputingatOregonStateUniversityandpost-processedusingastandardIlluminapipeline[28].We obtainedapproximately8-10million60-bpreadsfor eachgenome,providingroughly95predictedcoverage.AssemblyandannotationDenovo assemblywasgeneratedonNewblerassembler (version2.3;454LifeScience,Branford,CT)using454sequencingreadsforeachgenome.CLCworkbench[29] wasusedinthenextstepforcombining454-basedcontigswithilluminareads,w herein,454basedcontigs wereusedaslongreadstofillingapsgeneratedduring combined denovo assembly.Thesecombinedassemblies ofeachgenomewereuploadedonIMG-JGI(JointGenomeInstitute,WalnutCreek,California)serverforgene calling.ThegenepredictionwascarriedoutusingGeneMark.Pfam,InterPro,COGsassignmentswerecarried outforidentifiedgenes.Pathogenicityclustersdescribed inthepaperweremanuallyannotated.WholegenomecomparisonsWealigneddraftgenomesagainstreference Xanthomonas genomesusingnucmer[31]ofMUMmerprogram (version3.20)anddnadiffwasusedtocalculatepercentageofalignedsequences.WehavealsocomparedgenomesusingtheMUMindex[30]tomeasuredistances betweentwogenomes.Themaximaluniqueexact matchesindex(MUMi)distancecalculationwasperformedusingtheMummerprogram(version3.20). Mummerwasrunonconcatenatedcontigsorreplicons (achievedbyinsertingastringof20symbols N betweencontigorrepliconsequences)ofeachgenome. ThedistancecalculationsperformedusingtheMUMi scriptarebasedonthenumberofmaximalunique matchesofagivenminimallengthsharedbytwogenomesbeingcompared.MUMivaluesvaryfrom0for identicalgenomesto1forverydistantgenomes[30].PhylogeneticanalysisMLSTsequences( fusA gapA gltA gyrB lacF lepA )for allthegenomeswereobtainedinconcatenatedform fromPAMDBwebsitehttp://pamdb.org.Genesand theircorrespondingaminoacidsequencesspanning gum hrp clusterweredownloadedfromNCBIgenbank sequencesofsequencedgenomes.Aminoacidsequences ofproteinsoftheseclustersfor Xcv and Xcc wereused asquerytosearchforhomologyagainstdraftgenomes of Xp,Xv and Xg .Theaminoacidsequenceswerethen concatenatedforeachpathogenicityclusterandthen alignedusingCLUSTALWignoringgaps.Neighbourjoiningtreeswereconstructedwithboostrapvaluefor 1000replicatesusingMEGA4[98].Codonpositions includedwere1st+2nd+3rd+Noncoding.Allpositions containinggapsandmissingdatawereeliminatedfrom thedataset(Completedeletionoption).Therewerea totalof2723positionsinthefinaldataset.PhylogenyreconstructionSpeciestree .Weusedasupermatrixapproachasinpreviouswork[25].Proteinsequencesofsix Xanthomonas genomes(ingroups)andthe S.maltophilia R551-3genome(outgroup)wereclust eredin5,096familiesusing OrthoMCL[99].Wethenselectedfamilieswithoneand onlyonerepresentativefro meachoftheingroupgenomesandatmostoneoutgroupprotein,resultinginPotnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page18of23

PAGE 19

2,282families.Theirsequenceswerealignedusing MUSCLE[100]andtheresultingalignmentswereconcatenated.Non-informativecolumnswereremoved usingGblocks[101],resultingin792,079positions. RAxML[102]withthePROTGAMMAWAGFmodel wasusedtobuildthefinaltree.Predictionofeffectorrepertoires,cloningofcandidate effectorsandconfirmationusingavrBs2reportergene assayAdatabasewascreatedcollectingalltheknownplant andanimalpathogeneffectors.Usingalltheseknown effectorsasquery,tblastnanalysiswasperformed againstallcontigsofthedraftgenomesof Xv Xg and Xp withe-valueof10-5[103].Pfamdomainswere searchedforpossibledomainsfoundinknowneffectors inpredictedsetofORFsofdraftgenomesequences. Candidateeffectorswereclassifiedaccordingtothe nomenclatureandclassificationschemeforeffectorsin xanthomonadsrecently[24].Candidateeffectorsshowing<45%identityataminoacidleveltotheknown effectorswereconfirmedfortheirtranslocationusing avrBs2 reportergeneassay. N-terminal100aminoacidregionalongwith upstream500bpssequenceofcandidategeneswere PCRamplifiedusingprimerswithBglIIrestrictionsites atthe5 ends.FollowingdigestionwithBglII,PCR ampliconswereligatedwithBglII-digestedpBS(BglII:: avrBs262-574::HA )(courtesyofDr.MaryBethMudgett, Stanforduniversity),andlatertransformedinto E.coli DH5 a .In-framefusionswereconfirmedbyDNA sequencingusingF20andR24primers.BamHI-KpnI fragmentscontainingthecandidategenefusedto avrBs2 wasthenclonedintopUFR034.Resultingplasmidswere thenintroducedinto Xcv pepperrace6(TED3 containingmutationin avrBs2 )bytri-parentalmating.The resulting Xcv strainswereinoculatedon Bs2 peppercv. ECW20Randkeptat28Cingrowthroom.After24 hours,strongHRwasindicatingsuccessfultranslocation ofcandidateeffectorfusions.Cloningofpepperspecificitygenesin XpThethreegenesmentionedabovewereclonedindividuallyandincombinationinpLAFR3vectorandconjugatedin Xp 91-118 avrXv3 mutantPM1.ThePM1 transconjugantswiththethreeindividualgenesand combinedonesalongwithvirulentpepperrace6strain wereinfiltratedat105CFU/mlconcentrationinpepper cv.ECWandleavesweresampledatevery48hours afterinoculation.Thesampleswereplatedonnutrient agar,incubatedat27CandCFU/mlcountswereenumerated.Experimentwascarriedoutintriplicateand repeatedthreetimes.DatabasesubmissionThedraftgenomesequencesof Xanthomonasvesicatoria ATCC35937( Xv )havebeendepositedatDDBJ/EMBL/ GenBankunderaccessionnumberAEQV00000000.The draftgenomesequencesof Xanthomonasperforans 91118( Xp )havebeendepositedatDDBJ/EMBL/GenBank underaccessionnumberAEQW00000000.Thedraft genomesequencesof Xanthomonasgardneri ATCC 19865( Xg )havebeendepositedatDDBJ/EMBL/GenBankunderaccessionnumberAEQX00000000.Theversiondescribedinthispaperisthefirstversion, AEQV01000000,AEQW01000000,AEQX01000000.All threedraftgenomeswillbe releaseduponmanuscript acceptance.AdditionalmaterialAdditionalfile1:TableS1:Generalfeaturesofthesequencingdata andofthe denovo assembliesofdraftgenomesof Xv Xp and Xg usingindividualsequencingmethods Additionalfile2:FigureS2:1a)PhylogenetictreebasedonMUMi indices;1b)DistancematrixbasedonMUMiindices .MUMiprogram wasusedtocalculatepairwisedistancesbetweendraftgenomesand reference Xanthomonas genomes. Additionalfile3:TableS3:Wholegenomecomparisonsusing MUMmerdnadiffprogram .%coverageofthealignedcontigsand% identitiesoftherespectivecontigsagainstreferencegenomeshasbeen shownforeachdraftgenome. Additionalfile4:FigureS4:AvrBs2-basedHRassayconfirms translocationofnoveleffectors .Hypersensitiveresponsereaction indicatingpresenceoftranslocationsignalwasrecorded24hrsafter inoculationonpeppercv.ECW20Rwithcandidateeffectors xopZ2 (a), avrBs1 (b) ,xopG (d) ,xopAM (e) ,xopAO (f)conjugatedinrace6strain alongwithcontrolrace6strain(c).Allthestrainsshowedwater-soaking onpeppercv.ECWafter48hrsafterinoculation Additionalfile5:TableS5:Evidenceofthehorizontalgenetransfer usingAlien_hunteranalysis Additionalfile6:TableS6:Genes/contigsrepresentingT6SSindraft genomesascomparedto Xcv Additionalfile7:TableS7:Domainarchitectureanddistributionof proteinswithHD-GYP,GGDEFand/orEALdomainsencodedby genomesofdifferent Xanthomonas strains Additionalfile8:TableS8:Genesuniqueto Xp ,groupedinclusters Additionalfile9:TableS9:Genescommontoallpepperpathogens butabsentfrom Xp Additionalfile10:TableS10:Genesuniqueto Xg .Genesofspecial interestarehighlightedinredandyellow Listofabbreviations Xcv : Xanthomonaseuvesicatoria strain85-10;formerly, Xanthomonas campestris pv. vesicatoria strain85-10; Xv : Xanthomonasvesicatoria strain 1111(ATCC35937); Xp : Xanthomonasperforans strain91-118; Xg : Xanthomonasgardneri strain101(ATCC19865); Xoo : Xanthomonasoryzae pv. Oryzae;Xcc : Xanthomonascampestris pv. Campestris;Xcm : Xanthomonas campestris pv. musacearum NCPPB4381; Xvv : Xanthomonas vasicola pv. vasculorum NCPPB702; Xac : Xanthomonas citri subsp.citristrain306;formerly, Xanthomonasaxonopodis pv. citri strain306; Xaub : Xanthomonasfuscans subsp. aurantifolii Bstrain; Xauc : Xanthomonasfuscans subsp. aurantifoliiC strain; Xoc:Xanthomonasoryzaepv.Oryzicola;Xca:XanthomonascampestrisPotnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page19of23

PAGE 20

pv. Armoracie;Xalb:Xanthomonasalbilineans;Sm:Stenotrophomonas maltophilia Acknowledgements ThisresearchwassupportedbyUSDA-NIFAspecialgrantT-STAR(J.B.Jones, J.F.PrestonandM.T.Momol,USDA2006-34135-17569).WethankIMG-JGI forprovidinggenomeannotationpipelineandinterfaceforthecomparative genomicanalyses.SpecialthankstoKostantinosMavromatisinhelping preparationofGenbankfiles. Authordetails1DepartmentofPlantPathology,UniversityofFlorida,Gainesville,FL,USA.2DepartmentofPlant&MicrobialBiology,UniversityofCalifornia,Berkeley, Berkeley,CA,USA.3DepartmentofMicrobiologyandCellScience,University ofFlorida,Gainesville,FL,USA.4FaculdadedeComputao,Universidade FederaldeMatoGrossodoSul,CampoGrande,MS,Brazil.5Instituteof MicrobialTechnology(CSIR),Sector39A,Chandigarh160036,India.6BIOMERITResearchCentre,BiosciencesInstitute,UniversityCollegeCork, Ireland.7Fundecitrus-FundodeDefesadaCitricultura,Av.AdhemarPereira deBarros,201,14807-040Araraquara,SP.Brazil.8DepartmentofPlant Pathology,KansasStateUniversity,Manhattan,KS,USA.9DepartmentofPlant Pathology,PhysiologyandWeedSciences,VirginiaTech,Blacksburg,VA, USA.10LaboratoireGnomeetDveloppementdesPlantes,IRD-CNRSUniversit-dePerpignan,CentreIRD,911Av.Agropolis,BP64501,34394 Montpellier,France.11VirginiaBioinformaticsInstitute,VirginiaPolytechnic InstituteandStateUniversity,Blacksburg,VA,USA.12InstituteofFoodand AgriculturalSciences,Mid-FloridaResearch&EducationCenter,Universityof Florida,Apopka,FL,USA. Authors contributions JBJconceivedtheproject.JBJandBJSoversawgenomicsequencing.JBJ providedthestrains.KKdidgenomeassembly.NP,RPR,RK,VC,PBP,FB annotatedpathogenicityclustersinthegenomes.DJNandBJShelped createaneffectordatabase.NPcarriedouteffectoranalysisandconfirmed themexperimentally.NP,BAV,RK,FFWandJBJinterpretedeffectoranalysis. NP,RPR,RK,VC,JFP,PBP,JMD,MS,TMdidanalysesofdifferent pathogenicityclustersofthethreegenomesandhelpedwriting correspondingsectionsinthemanuscript.NPoversawtheexperimental validationsofthepathogenicityclusters.NFAcreatedorthologfamilies,and didphylogeneticanalysisbasedonallorthologusfamiliesofthethreedraft genomesandthereferencegenomes.JCSdidMUMianalysisand constructedphylogenetictreebasedonthatanalysis.NPgeneratedthe GenBankfiles.DJN,BJS,FFW,RK,JBJ,BAV,JFP,andJCShelpedwithdata analyses.NPandJBJwrotethefinalmanuscript.Allauthorsapprovedthe finalmanuscript. Received:1December2010Accepted:11March2011 Published:11March2011 References1.PohroneznyK,VolinRB: Theeffectofbacterialspotonyieldandquality offreshmarkettomatoes. HortScience 1983, 18 :69-70. 2.JonesJB,StallRE,BouzarH: Diversityamongxanthomonadspathogenic onpepperandtomato. AnnuRevPhytopathol 1998, 36 :41-58. 3.JonesJB,BouzarH,StallRE,AlmiraEC,RobertsPD,BowenBW,SudberryJ, StricklerPM,ChunJ: Systematicanalysisofxanthomonads( Xanthomonas spp.)associatedwithpepperandtomatolesions. IntJSystEvolMicrobiol 2000, 50 :1211-1219. 4.BouzarH,JonesJB,SomodiGC,StallRE,DaouzliN,LambeRC,Felix GastelumR,TrinidadCorreaR: Diversityof Xanthomonascampestris pv. vesicatoria intomatoandpepperfieldsofMexico. CanJPlantPathol 1996, 18 :75-77. 5.BouzarH,JonesJB,StallRE,LouwsFJ,SchneiderM,RademakerJLW,de BruijnFJ,JacksonLE: Multiphasicanalysisofxanthomonadscausing bacterialspotdiseaseontomatoandpepperintheCaribbeanand CentralAmerica:Evidenceforcommonlineageswithinandbetween countries. Phytopathology 1999, 89 :328-335. 6.KimSH,OlsonTN,PefferND,NikolaevaEV,ParkS,KangS: Firstreportof bacterialspotoftomatocausedby Xanthomonasgardneri in Pennsylvania. PlantDisease 2010, 94 :638. 7.HamzaAA,Robene-SoustradeI,JouenE,GagnevinL,LefeuvreP,ChiroleuF, PruvostO: Geneticandpathologicaldiversityamong Xanthomonas strainsresponsibleforbacterialspotontomatoandpepperinthe southwestIndianOceanregion. PlantDisease 2010, 94 :993-999. 8.MyungIS,MoonSY,JeongIH,LeeYK,LeeYH,RaDS: Bacterialspotof tomatocausedby Xanthomonasperforans ,anewdiseaseinKorea. Plant Disease 2009, 93 :1349. 9.SuticD: Bakteriozecrvenogpatlidzana(Tomatobacteriosis). PosebnaIzd InstZashtBiljaBeograd 1957, 6 :1-65,(specialedition).Beograd:Instituteof PlantProtein.(EnglishsummaryRevApplMycol1957,36:734-735.). 10.DeLeyJ: Modernmolecularmethodsinbacterialtaxonomy:evaluation, application,prospects. Proceedingsofthe4thInternationalConferenceon PlantPathogenicBacteria,Angers 1978,347-357. 11.JonesJB,LacyGH,BouzarH,StallRE,SchaadNW: Reclassificationofthe xanthomonadsassociatedwithbacterialspotdiseaseoftomatoand pepper. SystApplMicrobiol 2004, 27 :755-762. 12.AlmeidaNF,YanS,CaiR,ClarkeCR,MorrisCE,SchaadNW,SchuenzelEL, LacyGH,SunX,JonesJB,CastilloJA,BullCT,LemanS,GuttmanDS, SetubalJC,VinatzerBA: PAMDB,Amultilocussequencetypingand analysisdatabaseandwebsiteforplant-associatedmicrobes. Phytopathology 2010, 100 :208-215. 13.ObradovicA,JonesJB,MomolMT,BaloghB,OlsonSM: Managementof tomato bacterial spotinthefieldbyfoliarapplicationsof bacteriophagesandSARinducers. PlantDisease 2004, 88 :736-740. 14.LouwsFJ,WilsonM,CampbellHL,CuppelsDA,JonesJB,ShoemakerPB, SahinF,MillerSA: Fieldcontrolofbacterialspotandbacterialspeckof tomatousingaplantactivator. PlantDisease 2001, 85 :481-488. 15.KearneyB,StaskawiczBJ: Widespreaddistributionandfitnesscontribution of Xanthomonascampestris avirulencegene avrBs2 Nature 1990, 346 :385-386. 16.GassmannW,DahlbeckD,CjesnokovaO,MinsavageGV,JonesJB, StaskawiczBJ: Molecularevolutionofvirulenceinnaturalfieldstrains of Xanthomonascampestris pv. vesicatoria JBacteriol 2000, 182:7053-7059. 17.StallRE,JonesJB,MinsavageGV: Durabilityofresistanceintomatoand peppertoxanthomonadscausingbacterialspot. AnnuRevPhytopathol 2009, 47 :265-84. 18.BonasU,SchulteR,FenselauS,MinsavageGV,StaskawiczBJ,StallRE: Isolationofageneclusterfrom Xanthomonascampestris pv. vesicatoria thatdeterminespathogenicityandhypersensitiveresponseonpepper andtomato. MolPlantMicrobeInteract 1991, 4 :81-88. 19.KimJG,ParkBK,YooCH,JeonE,OhJ,HwangI: Characterizationofthe Xanthomonasaxonopodis pv.glycinesHrpPathogenicityIsland. JBacteriol 2003, 185 :3155-3166. 20.NimuraK,MelottoM,HeS: Suppressionofhostdefenseincompatible plantPseudomonassyringae interactions. CurrOpinionPlantBiol 2005, 8 :361-368. 21.GrantSR,FisherEJ,ChangJH,MoleBM,DanglJL: Subterfugeand manipulation:TypeIIIeffectorproteinsofphytopathogenicbacteria. AnnuRevMicrobiol 2006, 60 :425-449. 22.SarkarS,GordonJ,MartinG,GuttmanD: Comparativegenomicsofhostspecificvirulencein Pseudomonassyringae Genetics 2006, 174 :1041-1056. 23.RohmerL,GuttmanDS,DanglJL: Diverseevolutionarymechanismsshape thetypeIIIeffectorvirulencefactorrepertoireintheplantpathogen Pseudomonassyringae Genetics 2004, 167 :1341-1360. 24.WhiteFF,PotnisN,JonesJB,KoebnikR: TheTypeIIIeffectorsof Xanthomonas. MolPlantPathol 2009, 10 :749-766. 25. Moreira LM,AlmeidaNFJr,PotnisN,DigiampietriLA,AdiSS,BortolossiJC, daSilvaAC,daSilvaAM,deMoraesFE,deOliveiraJC,deSouzaRF, FacincaniAP,FerrazAL,FerroMI,FurlanLR,GimenezDF,JonesJB, KitajimaEW,LaiaML,LeiteRPJr,NishiyamaMY,NetoJR,NocitiLA, NormanDJ,OstroskiEH,PereiraHAJr,StaskawiczBJ,TezzaRI,FerroJA, VinatzerBA,SetubalJC: Novelinsightsintothegenomicbasisofcitrus cankerbasedonthegenomesequencesoftwostrainsof Xanthomonas fuscans subsp. aurantifolii BMCGenomics 2010, 11 :238. 26.ThiemeF,KoebnikR,BekelT,BergerC,BochJ,BttnerD,CaldanaC, GaigalatL,GoesmannA,KayS,KirchnerO,LanzC,LinkeB,McHardyAC, MeyerF,MittenhuberG,NiesDH,Niesbach-KlsgenU,PatschkowskiT, RckertC,RuppO,SchneikerS,SchusterSC,VorhlterFJ,WeberE, PhlerA,BonasU,BartelsD,KaiserO: Insightsintogenomeplasticityand pathogenicityoftheplantpathogenicbacterium XanthomonasPotnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page20of23

PAGE 21

campestris pv. vesicatoria revealedbythecompletegenomesequence. JBacteriol 2005, 187 :7254-7266. 27.MarguliesM,EgholmM,AltmanW,AttiyaS,BaderJ,BemdenL,BerkaJ, BravermanM,ChenY,ChenZ,DewellS,DuL,FierroJ,GomesX,BegleyR, RothbergJ: Genomesequencinginopenmicrofabricatedhighdensity picoliterreactions. Nature 2005, 437 :376-380. 28.BentleyDR: Wholegenomeresequencing. CurrOpinionGenetDev 2006, 16 :545-552. 29.CLCGenomicsWorkbench: Whitepaperon denovo assemblyinCLC NGSCell3.0beta. 2010[http://www.clcbio.com]. 30.DelongerM,KarouiME,PetitM: AgenomicdistancebasedonMUM indicatesdiscontinuitybetweenmostbacterialspeciesandgenera. JBacteriol 2009, 191 :91-99. 31.KurtzS,PhillippyA,DelcherAL,SmootM,ShumwayM,AntonescuC, SalzbergSL: Versatileandopensoftwareforcomparinglargegenomes. GenomeBiol 2004, 5 :R12. 32.KimJG,LiX,RodenJA,TaylorKW,AakreCD,SuB,LalondeS,KirikA, ChenY,BaranageG,McLaneH,MartinGB,MudgettMB: Xanthomonas T3S effectorXopNsuppressesPAMP-triggeredimmunityandinteractswitha tomatoatypicalreceptor-likekinaseandTFT1. PlantCell 2009, 21 :1305-1323. 33.BttnerD,NoelL,StutmannJ,BonasU: Characterizationofthe nonconserved hpaB hrpF regioninthe hrp pathogenicityislandfrom Xanthomonascampestris pv. vesicatoria MolPlantMicrobeInteract 2007, 20 :1063-1074. 34.SilvaACR,FerroJA,ReinachFC,FarahCS,FurlanLR,QuaggioR,MonteiroVitorelloCB,VanSluysMA,AlmeidaNF,AlvesLMC,AmaralAM, BertoliniMC,CamargoLE,CamarotteG,CannavanF,CardozoJ, ChambergoF,CiapinaLP,CicarelliRM,CoutinhoLL,Cursino-SantosJR,ElDorryH,FariaJB,FerreiraAJ,FerreiraRC,FerroMI,FormighieriEF, FrancoMC,GreggioCC,GruberA,KatsuyamaAM,KishiLT,LeiteRP, LemosEG,LemosMV,LocaliEC,MachadoMA,MadeiraAM,MartinezRossiNM,MartinsEC,MeidanisJ,MenckCF,MiyakiCY,MoonDH, MoreiraLM,NovoMT,OkuraVK,OliveiraMC,OliveiraVR,PereiraHA, RossiA,SenaJA,SilvaC,deSouzaRF,SpinolaLA,TakitaMA,TamuraRE, TeixeiraEC,TezzaRI,TrindadedosSantosM,TruffiD,TsaiSM,WhiteFF, SetubalJC,KitajimaJP: Comparisonofthegenomesoftwo Xanthomonas pathogenswithdifferinghostspecificities. Nature 2002, 417 :459-463. 35.StavrinidesJ,MaW,GuttmanD: Terminalreassortmentdrivesthe quantumevolutionoftypeIIIeffectorsinbacterialpathogens. PLoS Pathogens 2006, 2 :e104. 36.KimJG,TaylorKW,HotsonA,KeeganM,SchmelzEA,MudgettMB: XopD SUMOproteaseaffectshosttranscription,promotespathogengrowth, anddelayssymptomdevelopmentin Xanthomonas -infectedtomato leaves. PlantCell 2008, 20 :1915-1929. 37.Astua-MongeG,MinsavageGV,StallRE,DavisMJ,BonasU,JonesJB: ResistanceoftomatoandpeppertoT3strainsof Xanthomona campestris pv. vesicatoria is specified byaplant-inducibleavirulence gene. MolPlantMicrobeInteract 2000, 13 :911-921. 38.Astua-MongeG,MinsavageGV,StallRE,VallejosCE,DavisMJ,JonesJB: Xv4 Avrxv4 :Anewgene-for-geneinteractionidentifiedbetween Xanthomonascampetris pv. vesicatoria raceT3andthewildtomato relative Lycopersiconpennellii MolPlantMicrobeInteract 2000, 13 :1346-1355. 39.RybakM,MinsavageGV,StallRE,JonesJB: Identificationof Xanthomonas citri ssp. citri hostspecificitygenesinaheterologousexpressionhost. MolPlantPathol 2009, 10 :249-262. 40.MinsavageGV,DahlbeckD,WhalenMC,KearnyB,BonasU,StaskawiczBJ, StallRE: Gene-for-generelationshipsspecifyingdiseaseresistancein Xanthomonascampestris pv. vesicatoria -pepperinteractions. MolPlant MicrobeInteract 1990, 3 :41-47. 41.SchornackS,MinsavageGV,StallRE,JonesJB,LahayeT: Characterizationof AvrHah1,anovelAvrBs3-likeeffectorfrom Xanthomonasgardneri with virulenceandavirulenceactivity. NewPhytol 2008, 179 :546-556. 42.NoelL,ThiemeF,NennstielD,BonasU: C-DNA-AFLPanalysisunravelsa genome-widehrpG-regulonintheplantpathogen Xanthomonas campestris pv. vesicatoria MolMicrobiol 2001, 41 :1271-1281. 43.AraujoER,PereiraRC,MoitaAW,FerreiraMASV,Caf-FihoAC,QuezadoDuvalAM: Effectoftemperatureonpathogenicitycomponentsof tomatobacterialspotandcompetitionbetween Xanthomonasperforans and X.garnderi IIIInternationalsymposiumontomatodiseases 2010. 44.VernikosGS,ParkhillJ: Interpolatedvariableordermotifsforidentification ofhorizontallyacquiredDNA:revisitingthe Salmonella pathogenicity islands. Bioinformatics 2006, 22 :2196-203. 45.LorangJM,ShenH,KobayashiD,CookseyD,KeenNT: AvrAandavrEin Pseudomonassyrinage pv. tomato PT23playaroleinvirulenceon tomatoplants. MolPlantMicrobeInteract 1994, 7 :508-515. 46.MukaiharaT,TamuraN,IwabuchiM: Genome-wide identification oflarge repertoireof Ralstoniasolanacearum typeIIIeffectorproteinsbyanew functionalscreen. MolPlantMicrobeInteract 2010, 23 :251-262. 47.LeeS.-W,JeongK.-S,HanS.-W,LeeS.-E,PheeB.-K,HahnT.-R,RonaldP: The Xanthomonasoryzae pv.oryzaePhoPQtwo-componentsystemis requiredforAvrXA21activity, hrpG expression,andvirulence. JBacteriol 2008, 190 :2183-2197. 48.LeeS.-W,HanS.-W,SririyanumM,ParkC.-J,SeoY.-S,RonaldPC: AtypeIsecreted,sulfatedpeptidetriggersXA21-mediatedinnateimmunity. Science 2009, 326 :850-853. 49.daSilvaFG,ShenY,DardickC,BurdmanS,YadavRC,deLeonAL, RonaldPC: BacterialgenesinvolvedintypeIsecretionandsulfationare requiredtoelicitthericeXa21-mediatedinnateimmuneresponse. Mol PlantMicrobeInteract 2004, 17 :593-601. 50.JhaG,RajeshwariR,SontiR: Bacterialtypetwosecretionsystemsecreted proteins:double-edgedswordsforplantpathogens. MolPlantMicrobe Interact 2005, 18 :891-898. 51.WangL,RongW,HeC: Two Xanthomonas extracellular polygalacturonases,PghAxcandPghBxc,areregulatedbytypeIII secretionregulatorsHrpXandHrpGandarerequiredforvirulence. Mol PlantMicrobeInteract 2008, 21 :555-563. 52.SzczesnyR,JordanM,SchrammC,SchulzS,CogezV,BonasU,BttnerD: FunctionalcharacterizationoftheXcsandXpstypeIIsecretionsystems fromtheplantpathogenicbacterium Xanthomonascampestris pv. vesicatoria NewPhytol 2010, 187 :983-1002. 53.HurlbertJC,PrestonJF: Functionalcharacterizationofanovelxylanase fromcornstrainsof Erwiniachrysanthemi JBacteriol 2001, 183 :2093-2100. 54.StJohnFJ,RiceJD,PrestonJF: CharacterizationofXynCfrom Bacillus subtilis subspecies subtilis strain168andAnalysisofItsRolein DepolymerizationofGlucuronoxylan. JBacteriol 2006, 24 :8617-8626. 55.SunQ,HuJ,HuangG,GeC,FangR,HeC: Type-IIsecretionpathway structuralgene xpsE,xylanase-andcellulasesecretionandvirulencein Xanthomonasoryzae pv. oryzae PlantPathol 2005, 54 :15-21. 56.GoesaertH,GebruersK,BrijsK,CourtinCM,DelcourJA: XIP-type endoxylanaseinhibitorsindifferentcereals. JC e realSci 2003, 38:317-324. 57.RajeshwariR,JhaG,SontiR: Roleofaninplanta-expressedxylanaseof Xanthomonasoryzae pv. oryzae inpromotingvirulenceonrice. MolPlant MicrobeInteract 2005, 18 :830-837. 58.BielyP,VrsanskaMM,TenkanenM,KluepfelD: Endo-beta-1,4-xylanase families:differencesincatalyticproperties. JBiotechnol 1997, 57 :151-166. 59.PrestonJF,HurlbertJC,RiceJD,RagunathanA,StJohnFJ: Microbial StrategiesfortheDepolymerizationofGlucuronoxylan:Leadstothe BiotechnologicalApplicationsofEndoxylanases. In ApplicationofEnzymes toLignocellulosics.VolumeCh12. Editedby:MansfieldSD,SaddlerJN.ACS SymposiumSeriesNo.855;2003:191-210. 60.ShulamiS,GalO,SonensheinAL,ShohamY: Theglucuronicacidutilizationgeneclusterfrom Bacillusstearothermophilus T-6. JBacteriol 1999, 181 :3695-3704. 61.ShulamiS,ZaideG,ZolotnitskyG,LangutY,FeldG,SonensheinAL, ShohamY: Atwo-componentsystemregulatestheexpressionofanABC transporterforxylo-oligosaccharidesin Geobacillusstearothermophilus ApplEnvironMicrobiol 2007, 73 :874-84. 62.ChowV,NongG,PrestonJF: Structure,functionandregulationofthe aldouronate-utilizationgeneclusterfrom Paenibacillus sp.JDR-2. JBacteriol 2007, 189 :8863-8870. 63.ElTahirY,KuuselaP,SkurnikM: Functionalmappingofthe Yersinia enterocolitica adhesinYadA.IdentificationofeightNSVAIG-Smotifson theamino-terminalhalfoftheproteininvolvedincollagenbinding. Mol Microbiol 2000, 37 :192-206. 64.DasA,RangarajN,SontiR: Multipleadhesion-likefunctionsof Xanthomonasoryzae pv. oryzae areinvolvedinpromotingleaf attachment,entryandvirulenceonrice. MolPlantMicrobeInteract 2009, 22 :73-85.Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page21of23

PAGE 22

65.WangX,PrestonJF,RomeoT: The pgaABCD locusof Escherichiacoli promotesthesynthesisofapolysaccharideadhesinrequiredforbiofilm formation. JBacteriol 2004, 186 :2724-2734. 66.BoyerF,FichantG,BerthodJ,VandenbrouckY,AttreeI: Dissectingthe bacterialtypeVIsecretionsystembygenomewideinsilicoanalysis: whatcanbelearnedfromavailablemicrobialgenomicresources? BMC Genomics 2009, 10 :104. 67.VorhlterFJ,NiehausK,PhlerA: Lipopolysaccharidebiosynthesisin Xanthomonascampestris pv. campestris :aclusterof15genesisinvolved inthebiosynthesisoftheLPSO-antigenandtheLPScore. MolGenet Genomics 2001, 266 :79-95. 68.PatilPB,BogdanoveAJ,SontiRV: Theroleofhorizontaltransferinthe evolutionofahighlyvariablelipopolysaccharidebiosynthesislocusin xanthomonadsthatinfectrice,citrusandcrucifers. BMCEvolBiol 2007, 7 :243. 69.MolinaroA,SilipoA,LanzettaR,NewmannM,DowM,ParrilliM: Structural elucidationoftheO-chainofthelipopolysaccharidefrom Xanthomonas campestris strain8004. CarbohydrRes 2003, 338 :277-281. 70.FinnRD,MistryJ,TateJ,CoggillP,HegerA,PollingtonJE,GavinOL, GunesekaranP,CericG,ForslundK,HolmL,SonnhammerEL,EddySR, BatemanA: ThePfamproteinfamiliesdatabase. NucleicAcidResearch Databaseissue 2010, 38 :D211-222. 71.JonesJB,BouzarH,SomodiGC,StallRE,PerneznyK,El-MorsyG,ScottJW: EvidenceforthePreemptiveNatureofTomatoRace3of Xanthomonas campestris pv. vesicatoria inFlorida. Phytopathology 1998, 88 :33-38. 72.MooiFR,BikEM: Theevolutionofepidemic Vibriocholerae strains. Trends Microbiol 1997, 5 :161-165. 73.SlaterH,Alvarez-MoralesA,BarberCE,DanielsMJ,DowJM: AtwocomponentsysteminvolvinganHD-GYPdomainproteinlinkscell-cell signallingtopathogenicitygeneexpressionin Xanthomonascampestris MolMicrobiol 2000, 38 :986-1003. 74.HeYW,ZhangLH: Quorumsensingandvirulenceregulationin Xanthomonascampestris FEMSMicrobiolRev 2008, 32 :842-857. 75.DowM: Diversificationofthefunctionofcell-to-cellsignalingin regulationofvirulencewithinplantpathogenicxanthomonads. SciSignal 2008, 1 :23. 76.RyanRP,McCarthyY,AndradeM,FarahCS,ArmitageJP,DowJM: Cell-cell signal dependent dynamicinteractionsbetweenHD-GYPandGGDEF domainproteinsmediatevirulencein Xanthomonascampestris ProcNatl AcadSciUSA 2010, 107 :5989-5994. 77.BarberCE,TangJL,FengJX,PanMQ,WilsonTJG,SlaterH,DowJM, WilliamsP,andDanielsMJ: Anovelregulatorysystemrequiredfor pathogenicityof Xanthomonascampestris ismediatedbyasmall diffusiblesignalmolecule. MolMicrobiol 1997, 24 :555-566. 78.DowJM,CrossmanL,FindlayK,HeYQ,FengJX,TangJL: Biofilmdispersal in Xanthomonascampestris iscontrolledbycell-cellsignalingandis requiredforfullvirulencetoplants. ProcNatlAcadSciUSA 2003, 100 :10995-11000. 79.ChatterjeeS,SontiRV: rpfF mutantsof Xanthomonasoryzae pv. oryzae are deficientforvirulenceandgrowthunderlowironconditions. MolPlant MicrobeInteract 2002, 15 :463-471. 80.SicilianoF,TorresP,SendnL,BermejoC,FilipponeP,VelliceG,RamalloJ, CastagnaroA: Analysisofthemolecularbasisof Xanthomonas axonopodis pv. citri pathogenesisin Citruslimon ElectronJBiotechnol 2006, 9. 81.WangL,MakinoS,SubedeeA,BogdanoveAJ: Novelcandidatevirulence factorsinricepathogen Xanthomonasoryzae pv. oryzicola asrevealed bymutationalanalysis. ApplEnvironMicrobiol 2007, 73 :8023-8027. 82.RyanRP,FouhyY,LuceyJF,CrossmanLC,SpiroS,HeYW,ZhangLH, HeebS,CamaraM,WilliamsP,DowJM: Cell-cellsignalingin Xanthomonas campestris involvesanHD-GYPdomainproteinthatfunctionsincyclic di-GMPturnover. ProcNatlAcadSciUSA 2006, 103 :6712-6717. 83.RomlingU,GomelskyM,GalperinMY: C-di-GMP:thedawningofanovel bacterialsignallingsystem. MolMicrobiol 2005, 57 :629-639. 84.JenalU,MaloneJ: MechanismsofCyclic-di-GMPSignalinginBacteria. AnnuRevGenet 2006, 40 :385-407. 85.HenggeR: Principlesofc-di-GMPsignallinginbacteria. NatureRev Microbiol 2009, 7 :263-273. 86.RyanRP,FouhyY,LuceyJF,JiangBL,HeYQ,FengJX,TangJL,DowJM: Cyclicdi-GMPsignallinginthevirulenceandenvironmentaladaptation of Xanthomonas campestris Mol Microbiol 2007, 63 :429-442. 87.HeYW,BoonC,ZhouL,ZhangLH: Co-regulationof Xanthomonas campestris virulencebyquorumsensingandanoveltwo-component regulatorysystemRavS/RavR. MolMicrobiol 2009, 71 :1464-1476. 88.CrossmanVC,GouldJM,DowGS,VernikosA,OkazakiM,SebaihiaD, SaundersC,ArrowsmithT,CarverN,PetersE,AdlemA,KerhornouA, LordL,MurphyK,SeegerR,SquaresS,RutterMA,QuailMA,RajandreamD, HarrisC,ChurcherSD,BentleyJ,ParkhillNR,AvisonMB: Thecomplete genome,comparativeandfunctionalanalysisof Stenotrophomonas maltophilia revealsanorganismheavilyshieldedbydrugresistance determinants. GenomeBiol 2008, 9 :R74. 89.HertAP,RobertsPD,MomolMT,MinsavageGV,Tudor-NelsonSM,JonesJB: Relativeimportanceofbacteriocin-likegenesinantagonismof Xanthomonasperforans tomatorace3to Xanthomonaseuvesicatoria tomatorace1strains. ApplEnvironMicrobiol 2005, 71 :3581-3588. 90.Tudor-NelsonSM,MinsavageGV,StallRE,JonesJB: Bacteriocin-like substancesfromtomatorace3strainsof Xanthomonascampestris pv. vesicatoria Phytopathology 2003, 93 :1415-1421. 91.AlaviSM,SanjariS,DurandF,BrinC,ManceauC,PoussierS: Assessmentof thegeneticdiversityof Xanthomonasaxonopodis pv. phaseoli and Xanthomonasfuscans subsp.fuscansasabasistoidentifyputative pathogenicitygenesandatypeIIIsecretionsystemoftheSPI-1family bymultiplesuppressionsubtractivehybridizations. ApplEnvironMicrobiol 2008, 74 :3295-3301. 92.ReddyJD,ReddySL,HopkinsDL,GabrielDW: TolCisrequiredfor pathogenicityof Xylellafastidiosa in Vitisvinifera grapevines. MolPlant MicrobeInteract 2007, 20 :403-410. 93.LallyET,HillRB,KiebaLR,KorstoffJ: TheinteractionbetweenRTXtoxins andtargetcells. TrendsMicrobiol 1999, 7 :356-361. 94.LinhartovaI,BumbaL,MasinJ,BaslerM,OsickaR,KamanovaJ, ProchazkovaK,AdkinI,Hejnova-HolubovaJ,SadilkovaL,MorovaJ,SeboP: RTX-toxins:ahighlydiversefamilysecretedbyacommonmechanism. FEMSMicrobiolRev 2010, 34 :1076-1112. 95.GotoY,LiB,ClaesenJ,ShiY,BibbMJ,vanderDonkWA: Discoveryof uniquelanthioninesynthetasesrevealsnewmechanisticand evolutionaryinsights. PLoSBiol 2010, 8 :e1000339. 96.DixonRA,AchnineL,KotaP,LiuC,ReddyMS,WangL: The phenylpropanoidpathwayandplantdefence-agenomicsperspective. Mol Plant Pathol 2002, 3 :371-390. 97.AusubelFM,BrentR,KingstonRE,MooreDD,SeidmanJG,SmithJA, StruhlK:In Currentprotocolsinmolecularbiology.Volume1. NewYork,N.Y.: JohnWileyandSons;1994:2.4.1-2.4.2. 98.TamuraK,DudleyJ,NeiM,KumarS: MEGA4:MolecularEvolutionary GeneticsAnalysis(MEGA)softwareversion4.0. MolBiolEvol 2007, 24 :1596-1599. 99.LiL,StoeckertCJJr,RoosDS: OrthoMCL:identificationoforthologgroups foreukaryoticgenomes. GenomeRes 2003, 13 :2178-2189. 100.EdgarRC: MUSCLE:multiplesequencealignmentwithhighaccuracyand highthroughput. NucleicAcidsRes 2004, 32 :1792-1797. 101.CastresanaJ: Selectionofconservedblocksfrommultiplealignmentsfor theiruseinphylogeneticanalysis. MolBiolEvol 2000, 17 :540-552. 102.StamatakisA: RAxML-VI-HPC:maximumlikelihood-basedphylogenetic analyseswiththousandsoftaxaandmixedmodels. Bioinformatics 2006, 22 :2688-2690. 103.AltschulSF,MadeenTL,SchafferAA,ZhangJ,ZhangZ,MillerW, LipmanDJ: GappedBLASTandPSI-BLAST:anewgenerationofprotein databasesearchprograms. NucleicAcidsRes 1997, 25 :3389-3402. 104.KearneyB,StaskawiczBJ: Widespreaddistributionandfitnesscontribution of Xanthomonascampestris avirulencegene avrBs2 Nature 1990, 346 :385-386. 105.RodenJA,BeltB,RossJB,TachibanaT,VargasJ,MudgettMB: Agenetic screentoisolatetypeIIIeffectorstranslocatedintopeppercellsduring Xanthomonas infection. ProcNatlAcadSciUSA 2004, 101 :16624-16629. 106.FurutaniA,TakaokaM,SanadaH,NoguchiY,OkuT,TsunoK,OchiaiH, TsugeS: IdentificationofnoveltypeIIIsecretioneffectorsin Xanthomonasoryzae pv. oryzae MolPlantMicrobeInteract 2009, 22 :96-106. 107.JiangW,JiangBL,XuRQ,HuangJD,WeiHY,JiangGF,CenWJ,LiuJ, GeYY,LiGH,SuLL,HangXHTangDJ,LuGT,FengJX,HeYQ,TangJL: IdentificationofsixtypeIIIeffectorgeneswithPIPboxin Xanthomonas campestris pv. campestris andfiveofthemcontributeindividuallytofull pathogenicity. MolPlantMicrobeInteract 2009, 22 :1401-1411.Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page22of23

PAGE 23

108.KimJG,LiX,RodenJA,TaylorKW,AakreCD,SuB,LalondeS,KirikA, ChenY,BaranageG,McLaneH,MartinGB,MudgettMB: Xanthomonas T3S effectorXopNsuppressesPAMP-triggeredimmunityandinteractswitha tomatoatypicalreceptor-likekinaseandTFT1. PlantCell 2009, 21 :1305-1323. 109.MetzM,DahlbeckD,MoralesCQ,AlSadyB,ClarkET,StaskawiczBJ: The conserved Xanthomonascampestris pv. vesicatoria effectorproteinXopX isavirulencefactorandsuppresseshostdefensein Nicotiana benthamiana PlantJ 2005, 41 :801-814. 110.GuidotA,PriorP,SchoenfeldJ,CarrereS,GeninS,BoucherC: Genomic structureandphylogenyoftheplantpathogen Ralstoniasolanacearum fromgenedistributionanalysis. JBacteriol 2007, 189 :377-387. 111.ParkDS,HyunJW,ParkYJ,KimJS,KangHW,HahnJH,GoSJ: Sensitiveand specificdetectionof Xanthomonasaxonopodis pv. citri byPCRusing pathovarspecificprimersbasedonHrpWgenesequences. MicrobiolRes 2006, 161 :145-149. 112.XuRQ,LiXZ,WeiHY,JiangB,LiK,HeYQ,FengJX,TangJL: Regulationof eight avr genesby hrpG and hrpX in Xanthomonascampestris pv. campestris andtheirroleinpathogenicity. ProgressinNaturalScience 2006, 16 :1288-1294. 113.NoelL,ThiemeF,NennstielD,BonasU: C-DNA-AFLPanalysisunravelsa genome-wide hrpG -regulonintheplantpathogen Xanthomonas campestris pv. vesicatoria MolMicrobiol 2001, 41 :1271-1281. 114.ThiemeF,SzczesnyR,UrbanA,KirchnerO,HauseG,BonasU: NewtypeIII effectorsfrom Xanthomonascampestris pv. vesicatoria triggerplant reactionsdependentonaconservedN-myristoylationmotif. MolPlant MicrobeInteract 2007, 20 :1250-1261. 115.ThiemeF: GenombasierteIdentifizierungneuerpotentieller Virulenzfaktorenvon Xanthomonascampestris pv. vesicatoria Thesis Mathematisch-Naturwissenschaftlich-TechnischeFakulttderMartin-Luther Universitt,Halle-Wittenberg;2008.doi:10.1186/1471-2164-12-146 Citethisarticleas: Potnis etal .: Comparativegenomicsrevealsdiversity amongxanthomonadsinfectingtomatoandpepper. BMCGenomics 2011 12 :146. Submit your next manuscript to BioMed Central and take full advantage of: Convenient online submission Thorough peer review No space constraints or color gure charges Immediate publication on acceptance Inclusion in PubMed, CAS, Scopus and Google Scholar Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Potnis etal BMCGenomics 2011, 12 :146 http://www.biomedcentral.com/1471-2164/12/146 Page23of23

PAGE 24

Additional file 1 Table S1: General features of the sequencing data and of the de novo assemblies of draft genomes of Xv Xp and Xg using individual sequencing methods. Xanthomonas vesicatoria ( Xv ) Xanthomonas perforans ( Xp ) Xanthomonas gardneri (Xg) D e novo assembly based on 454 data Number of reads 397,923 476,897 408,422 Number of bases 39,726,904 45,692,099 38,828,836 Mean length of reads 99.84 95.81 95.07 Max length of reads 351 240 239 Min of length of reads 35 35 35 Number of assembled cont igs 4,181 2,360 4,540 Mean length of contigs 1,235 2,036 1,060 N50 701 317 767 Length of the longest contig 17,234 38,931 32,542 Total length of contigs 5,167,430 4,805,047 4,816,042 Calculated genome coverage 7.78 X 8.95 X 7.61 X De novo assembly based on Solexa data Number of reads 10,567,052 7,470,680 8,463,426 Number of bases 608,552,220 448,240,800 507,805,560 Max length of reads 60 60 60 Min of length of reads 60 60 60 Number of assembled reads 10,142,537 7,150,918 8,184,012 Number of assembled contigs 568 492 722 Mean length of contigs 9,275 9,866 7,261

PAGE 25

N50 84 68 112 Length of the longest contig 69,349 78,549 82,425 Total length of contigs 5,268,663 4,854,240 5,242,611 Calculated genome coverage 115X 88X 94X

PAGE 26

Additional file 2 Figure S2: 1a) Phylogenetic tree based on MUMi indices; 1b) Distance matrix based on MUMi indices. MUMi program was used to calculate pairwise distances between draft genomes and reference Xanthomonas genomes.

PAGE 27

0.01 Xcv Xp Xoo Xcc Xg Xv Xac 0 0.11 0.10 0.10 0.11 0.36 0.07 6.06e-03 0.38 0.37 0.38 0.21

PAGE 28

! Xac Xcc Xoo Xcv Xv Xp Xg Xac 0 0.77 0.67 0.41 0.76 0.41 0.75 Xcc 0.77 0 0.81 0.77 0.77 0.77 0.75 Xoo 0.67 0.81 0 0.67 0.80 0.67 0.79 Xcv 0.41 0.77 0.67 0 0.76 0.19 0.76 Xv 0.76 0.77 0.80 0.76 0 0.76 0.76 Xp 0.41 0.77 0.67 0.19 0.76 0 0.76 Xg 0.75 0.75 0.79 0.76 0.76 0.76 0

PAGE 29

Additional file 3 Table S3: Whole genome comparisons using MUMmer dnadiff program. % coverage of the aligned contigs and % identities of the respective contigs against reference genomes has been shown for each draft genome. Genome comparison % of contigs of draft genome aligned % of average identity for the aligned sequences Xp Xcv 85.57 98.1 Xac 85.91 93.8 Xcc 74.23 87.36 Xoo MAFF 77.32 90.5 Xg Xcv 78.44 88.57 Xac 79.71 88.05 Xcc 83.33 88.83 Xoo MAFF 72.83 87.9 Xv Xcv 83.11 87.86 Xcc 76.35 87.37 Xac 80.07 87.90 Xoo MAFF 69.26 87.68

PAGE 31

Additional file 5 Table S5: Evidence of the horizontal gene transfer using alien hunter analysis. Gene/ Gene cluster Locus tag Score for Alien Hunter (Threshold = 12.496) % GC tRNA/transposase/mobile genetic elements in the vicinity Evidence of HGT avrB s1 XGA_0724 32.031 45 Transposase Good xopAO XGA_1250 13.735 48 Predicted to be located on plasmid Good xopAS XGA_0764/XGA_0765 19.844 59 Transposase Good xopG XGA_4501 XVE_4777 XCV1298 21.272 50 ISxac2 transposase in Xcv Good xopAQ X GA_2091 Does not belong to anomalous region 51 Could not be predicted Weak xopZ2 XGA_2762 XVE_3221 Does not belong to anomalous region 69 IS30 transposase 3000 bps apart Weak LPS cluster XPE_3787 to XPE_3795 Belongs to anomalous region 50 No Good Bacteriocin cluster XPE_0786 to XPE_0790 Belongs to anomalous region 50 tranposase Good

PAGE 32

Additional file 6 Table S6: Genes/contigs representing T6SS in draft genomes as compared to Xcv T6SS subtype #1 T6SS subtype #3 Xp XCV homologs Xp XCV homologs Xv XCV homologs Contig 33 XCV2120 XCV2127(N) Contig 120 XCV4244 XCV4236(N) Contig 233 XCV42 44 XCV4216 Contig 287 XCV2127(i) Contig 287 XCV4236(i) Contig 288 XCV2127 (i) Contig 288 XCV4236 (i) Contig 291 XCV2127 (i) Contig 291 XCV4236 (i) Contig 238 XCV2127 (i) Contig 238 XCV4236 (i) Contig 254 XCV2127 (i) Contig 254 XCV4236 (i) Contig 90 XCV2127(C) XCV2137(N) Contig 44 XCV4236(C) XCV4216 Contig 240 XCV2137(C) XCV2144 no homolog XCV 4215 no homolog XCV 4215 Contig 116 XCV 4214 XCV4209(N) Contig 183 XCV 4214 XCV4212(N) Contig 133 XCV4209(C) XCV4206(N) Contig 148 XCV4213(N) XCV4206(N) Contig 233 XCV4206(i) Contig 195 XCV4206(C) Contig 175 XCV4206(C)

PAGE 33

Additional file 7 Table S7. Domain architecture and distribution of proteins with HD GYP, GGDEF and/or EAL domains encoded by genomes of different Xanthomonas strains Domain architecture of proteins Xcc8004 Xcc33913 XooK ACC 10331 XooP XO99 A XooM AFF 311018 Xcv Xv Xp X g XC0249 XCC0239 absent absent absent XCV0266 0573 1195 1780 XC0362 XCC0350 XOO4220 ORF03945 XOO3989 XCV0363 3039 4492 3433 X C0420 XCC0407 XOO0111 ORF03778 XOO0067 XCV0450 2497 2956 1448 XC0496 XCC0484 XOO0520 ORF02944 XOO0483 XCV0529 1437 3029 2716 XC0613 XCC3546 absent absent absent XCV0645 2531 4167 4044 1 XC0637 XCC3523 XOO4021 ORF04155 XOO3792 XCV0667 2531 3570 5056 XC0641 XCC3519 XOO4016 b ORF04147 XOO3786 XCV0672 0665 3574 4680 XC0675 XCC3486 XOO3988 ORF04275 XOO3759 XCV0699 1226 4018 4301 XC0831 XCC3333 XOO0933 a ORF02615 a XOO0855 a XCV3588 1507 1922 2867 XC1036 XCC3120 absent absent absent absent absent absent absent XC1383 XCC2731 XOO1440 ORF02019 XOO1324 XCV3048 5029 2 1275 0529 XC1411 XCC2703 XOO1467 ORF04753 XOO1368 XCV3023 2771 0357 5134 XC1476 XCC2641 XOO1551 ORF04826 XOO1439 XCV2973 3314 1829 1596 XC1582 XCC2563 XOO3246 ORF01278 XOO3074 XCV2860 1735 0013 2269 XC1755 XCC2361 XOO02798 ORF00476 XOO2640 XCV2672 0992 0406 1397 XC1766 XCC2350 XOO2787 ORF00466 XOO2627 XCV2660 0980 0394 2619 XC1803 XCC2313 absent absent absent XCV2623 0942 3324 5028 XC1824 XCC2291 XOO2725 ORF00403 XOO2570 XCV2595 3188 3346 3744 XC1841 XCC2274 XOO2708 ORF00384 XOO2553 XCV2579 3205 1807 2022 XC2161 XCC2023 absent absent absent XCV2102 1554 2913 3058 XC2226 XCC1959 XOO2561 ORF01021 XOO2420 XCV2041 4978 3 3128 1884 XC2228 XCC1958 XOO2563 ORF01019 XOO2422 XCV2039 4980 3130 1886 XC2274 XCC1913 XOO2614 ORF00967 XOO2473 XCV1985 4032 1307 1461 XC2275 XCC1912 XOO2615 ORF0096 5 XOO2474 XCV1983 4030 1306 1460 XC2276 XCC1911 XOO2616 ORF00964 XOO2475 XCV1982 absent 1305 1459 XC2324 XCC1865 XOO2860 ORF00058 XOO2715 XCV1929 2941 0537 0024

PAGE 34

XC2335 XCC1854 XOO2871 ORF00070 XOO2725 XCV1917 2924 0522 2956 XC2459 XCC1777 XOO2354 ORF00678 XOO2236 XCV1825 5096 0307 3401 XC2715 XCC1520 absent absent absent XCV1610 3722 1158 2000 XC2793 XCC1445 XOO2330 a OR F00647 XOO2206 a XCV1546 0243 3986 1105 XC2795 XCC1443 absent ORF00649 XOO2208 XCV1544 0241 3988 1961 XC2866 XCC1372 absent absent absent XCV1477 171 4823 4622 XC2946 XCC1294 XOO1879 ORF01741 XOO1775 XCV1398 2082 4745 1494 XC3163 XCC1086 absent absent absent XCV1210 1545 221 0886 XC3800 XCC3729 absent absent absent absent absent absent absent XC3829 XCC3759 absent absent absent absent absent absent absent XC3962 XCC3877 XOO4445 a ORF03877 a XOO4185 a XCV4051 0685 2596 0663 XC4313 XCC4224 absent ORF04275 absent XCV4470 4882 4444 4813 absent absent XOO2331 ORF00647 absent absent absent absent absent absent absent absent absent absent XCV2971 3316 1827 1598 absent absent absent absent absent XCVd0150 c absent absent absent 1 Domains were predicted using the prosite research tool from the Expert Protein analysis system ( http://ca.expasy.org/prosite ). This analysis revealed all proteins with the conserved HD domain; the subset of these protein s with the HD GYP domain was identified by checking for the presence of the GYP motif and by BLASTP alignments. The proteins also contain a variety of input domains including cyclic nucleotide binding domain (CNMP_BI), PAS, PAC, and uncharacterized domains or undefined regions are marked by a /. Complete domain d escriptions and functions of other domains are given at http://ca.expasy.org/prosite a Proteins with N terminal truncations; b Gene has an internal stop codon. c Plasmid borne.

PAGE 35

Additional file 8 Table S8: Genes unique to Xp grouped in clusters. Locus tag in Xp / Gene OID Distribution of flanking genes Function Cluster 1 LPS cluster genes XPE_3787 to XPE_3795 Present in Xcv Lipopolysaccharide biosynthesis cluster Cluster 2 Chemotaxis protein histidine kinase inactivated by transposase carrying 3 genes (in yellow) along with it in Xp XPE_4460 In all 4 chemotaxis protein histidine kinase XPE_4461 In all 4 transposase XPE_4462 Fe S oxidoreductase XPE_4463 XPE_4464 X PE_4465 In all 4 chemotaxis protein histidine kinase Cluster 3 Carrying unique genes in Xp not present in any plant pathogens XPE_1809 Transposase XPE_1810 TIR like domain, cyclic nucleic acid binding domain XPE_1811 Hypothetical protein XPE_1 812 Hypothetical protein XPE_1813 Hypothetical protein Cluster 4 avrXv4 and phage genes in the neighbourhood Cluster 5 XopC from Xcv is replaced by other unique genes in Xp XPE_3067 present in 306 hypothetical protein XPE_3068 present in 306 hypothetical protein XPE_3069 XAC2120 XPE_3070 mdmC from Xac306 Predicted O methyltransferase XPE_3071 not called in gene calling in 306 Hypothetical protein Cluster 6 Carrying bacteriocin genes XPE_0786 to XPE_0790 Cluster 7 fl anked by phage integrase XPE_2401 Predicted transcription regulator containing HTH domain XPE_2402 Uncharacterized protein conserved in bacteria XPE_2403 present in Xv XPE_2404 XPE_3894 plasmid mobilization system relaxase XPE_3895 XCV1122 XPE_3896 52% hypothetical predicted ATPase

PAGE 36

protein [Legionella pneumophila str. Corby] XPE_3897 hypo protein from Legionella XPE_3898 present in Xv, Xg XPE_3899 in Xv XCV1116 XPE_3900 Xcc B100_3109 exonuclease VII XPE_3901 present in Xg Cluster 8 Upstream flanking genes are conserved in order in all xanthomonads; while downstream are transposase genes in Xcv. XPE_3601 present in all xanthomonads XPE_3602 XPE_3603 XPE_3604 XPE_3605 XPE_3606 XPE_3607 XPE_3608 XPE_3609 cluster 9 Flanking genes conserved in Xcv XPE_3366 XAUB_37550 95% hypothetical protein XPE_3367 XCV0352 XPE_3368 XCV0353 XPE_3369 no hit to any plant pathogen hypothetical protein XPE_3370 no hit to any plant pathogen hypothetic al protein XPE_3371 no hit to any plant pathogen hypothetical protein XPE_3372 no hit to any plant pathogen hypothetical protein XPE_3373 no hit to any plant pathogen Activator of Hsp90 ATPase homolog 1 like protein. XPE_3374 no hit to any other plant pathogen hypothetical protein Cluster 10 Upstream and downstream flanking genes present in Xcv XPE_1376 no hit to any other plant pathogen XPE_1377 No hit to any other plant pathogen XPE_1378 no hit to any other plant pathogen XPE_1379 no hit to any other plant pathogen XPE_1380 EF hand calcium binding protein

PAGE 37

Cluster 11 Upstream, downstream flanking genes also present in Xcv XPE_0135 XAC3183 Hypothetical protein XPE_0136 XAC3182 Hypothetical protein Cluster 12 Flanking genes p resent in Xcv. In place of following unique genes, Xcv contains ISXac3 transposase. XPE_0734 signal peptide, transmemb helices XPE_0735 hypothetical protein XPE_0736 Xccb100_0356, vasculorum and musacearum hypothetical protein Cluster 13 XPE_218 3 In Xoo Type I site specific restriction modification system, R (restriction) subunit and related helicases XPE_2185 Xoo Type I restriction modification system methyltransferase subunit XPE_2187 Xoo Uncharacterized conserved protein XPE_2190 Xoo Unchar acterized conserved protein XPE_2192 Xoo Type I restriction modification system methyltransferase subunit XPE_2194 Xoo Type I site specific restriction modification system, R (restriction) subunit and related helicases XPE_2195 Xoo, Xoc hypothetical pro tein

PAGE 38

Additional file 9 Table S9: Genes common to all pepper pathogens but absent from Xp Locus tag in Xcv85 10 Gene symbol Product name Evidence to be involved in pathogenicity/ virulence Genes with the known functions XCV2278 Pectate lyase precursor XC V3713 wxcL Glycosyltransferase XCV3715 wxcN Putative membrane protein involved in synthesis of cell surface polysaccharide XCV3716 wxcO Putative carbohydrate translocase Alavi, SM et. al ., 2008 in X. fuscans bean pathosystem and this study. XCV3718 gmd GDP mannose 4,6 dehydratase (EC: 4.2.1.47) XCV3720 wxcB Putative protein kinase XCV3722 wzm O antigen ABC transporter permease XCV4257 rpmB LSU ribosomal protein L28P XCV1298 Type III effector (homolog of hopH1 from Pseudomonas syringae ) XC V1839 Hypothetical protein This study XCVc0007 kfrA KfrA protein XCV0510 hsdS1 Type I site specific deoxyribonuclease (specificity subunit) XCV0513 hsdM1 Type I site specific deoxyribonuclease (modification subunit) XCV2820 Putative type IV pilus assembly protein PilV XCV3312 Transcriptional regulator, AraC family XCV2191 Putative DoxD like family membrane protein

PAGE 39

Genes coding for mobile genetic elements XCVb0012 Putative ISxac3 transposase (fragment) XCVb0018 tnpR Tn5045 resolvase XCVc0040 Site specific recombinase/resolvase family protein XCVd0025 ISxac3 transposase (fragment) XCVd0071 Phage integrase family protein XCVd0097 tnpA Tn5044 transposase XCVd0109 tnpR Tn5045 resolvase XCVd0115 Tn5044 traposase XCV0355 ISxac3 transposase XCV0619 Transposase XCV0706 ISxac3 transposase XCV1118 ISxac3 transposase XCV1553 Phage related integrase XCV1698 ISxac3 transposase XCV1843 ISxac3 transposase XCV1848 Putative integrase/recombinase XCV2158 ISxac 3 transposase XCV2217 Phage related integrase XCV2261 Phage related integrase XCV2263 ISxac3 transposase (fragment) XCV2273 Tn5044 transposase XCV2295 Putative ISxac3 transposase (fragment) XCV2439 Tn5044 trasposase XCV2453 Filamentous phage Cf1c protein XCV2461 Filamentous phage phiLf related protein XCV2474 Filamentous phage Cf1c protein XCV2477 ISXac3 transposase XCV2484 Phage related integrase XCV2615 Integrase XCV2690 ISxac3 transposase XCV2712 Putative transpo sase (fragment) XCV2867 ISxac3 trasposase XCV3384 ISxac3 trasposase

PAGE 40

XCV3397 ISxac3 trasposase XCV3410 ISxac3 trasposase Genes with function unknown XCVd0055 XCV0648 XCV1188 XCV1189 XCV1187 XCV1303 XCV1596 XCV 1937 XCV2455 XCV2857 XCV2958 XCV3162 XCV3326 XCV3986 XCV4135 XCV4262 XCV4421