A comparative genomics screen identifies a Sinorhizobium meliloti 1021 sodM-like gene strongly expressed within host pla...

MISSING IMAGE

Material Information

Title:
A comparative genomics screen identifies a Sinorhizobium meliloti 1021 sodM-like gene strongly expressed within host plant nodules
Physical Description:
Mixed Material
Language:
English
Creator:
Queiroux, Clothilde
Washburn, Brain K.
Davis, Olivia M.
Stewart, Jamie
Brewer, Tess E.
Lyons, Michael R.
Jones, Kathryn M.
Publisher:
BioMed Central (BMC Microbiology)
Publication Date:

Notes

Abstract:
Background: We have used the genomic data in the Integrated Microbial Genomes system of the Department of Energy’s Joint Genome Institute to make predictions about rhizobial open reading frames that play a role in nodulation of host plants. The genomic data was screened by searching for ORFs conserved in α-proteobacterial rhizobia, but not conserved in closely-related non-nitrogen-fixing α-proteobacteria. Results: Using this approach, we identified many genes known to be involved in nodulation or nitrogen fixation, as well as several new candidate genes. We knocked out selected new genes and assayed for the presence of nodulation phenotypes and/or nodule-specific expression. One of these genes, SMc00911, is strongly expressed by bacterial cells within host plant nodules, but is expressed minimally by free-living bacterial cells. A strain carrying an insertion mutation in SMc00911 is not defective in the symbiosis with host plants, but in contrast to expectations, this mutant strain is able to out-compete the S. meliloti 1021 wild type strain for nodule occupancy in coinoculation experiments. The SMc00911 ORF is predicted to encode a “SodM-like” (superoxide dismutase-like) protein containing a rhodanese sulfurtransferase domain at the N-terminus and a chromate-resistance superfamily domain at the C-terminus. Several other ORFs (SMb20360, SMc01562, SMc01266, SMc03964, and the SMc01424-22 operon) identified in the screen are expressed at a moderate level by bacteria within nodules, but not by free-living bacteria. Conclusions: Based on the analysis of ORFs identified in this study, we conclude that this comparative genomics approach can identify rhizobial genes involved in the nitrogen-fixing symbiosis with host plants, although none of the newly identified genes were found to be essential for this process. Keywords: Rhizobia, Sinorhizobium meliloti, Alfalfa, Symbiosis, Nitrogen fixation, Bacteria, Legume, Genomics, α-proteobacteria
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 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.
System ID:
AA00013923:00001

Full Text

PAGE 1

RESEARCHARTICLEOpenAccessAcomparativegenomicsscreenidentifiesa Sinorhizobiummeliloti 1021 sodM -likegene stronglyexpressedwithinhostplantnodulesClothildeQueiroux1,BrianKWashburn1,OliviaMDavis1,2 †,JamieStewart1 †,TessEBrewer1,MichaelRLyons1,3and KathrynMJones1*AbstractBackground: WehaveusedthegenomicdataintheIntegratedMicrobialGenomessystemoftheDepartmentof Energy ’ sJointGenomeInstitutetomakepredictionsaboutrhizobialopenreadingframesthatplayarolein nodulationofhostplants.ThegenomicdatawasscreenedbysearchingforORFsconservedin -proteobacterial rhizobia,butnotconservedinclosely-relatednon-nitrogen-fixing -proteobacteria. Results: Usingthisapproach,weidentifiedmanygenesknowntobeinvolvedinnodulationornitrogenfixation,as wellasseveralnewcandidategenes.Weknockedoutselectednewgenesandassayedforthepresenceof nodulationphenotypesand/ornodule-specificexpression.Oneofthesegenes,SMc00911,isstronglyexpressedby bacterialcellswithinhostplantnodules,butisexpressedminimallybyfree-livingbacterialcells.Astraincarryingan insertionmutationinSMc00911isnotdefectiveinthesymbiosiswithhostplants,butincontrasttoexpectations, thismutantstrainisabletoout-competethe S.meliloti 1021wildtypestrainfornoduleoccupancyincoinoculationexperiments.TheSMc00911ORFispredictedtoencodea “ SodM-like ” (superoxidedismutase-like) proteincontainingarhodanesesulfurtransferasedomainattheN-terminusandachromate-resistancesuperfamily domainattheC-terminus.SeveralotherORFs(SMb20360,SMc01562,SMc01266,SMc03964,andtheSMc01424-22 operon)identifiedinthescreenareexpressedatamoderatelevelbybacteriawithinnodules,butnotbyfree-living bacteria. Conclusions: BasedontheanalysisofORFsidentifiedinthisstudy,weconcludethatthiscomparativegenomics approachcanidentifyrhizobialgenesinvolvedinthenitrogen-fixingsymbiosiswithhostplants,althoughnoneof thenewlyidentifiedgeneswerefoundtobeessentialforthisprocess. Keywords: Rhizobia, Sinorhizobiummeliloti ,Alfalfa,Symbiosis,Nitrogenfixation,Bacteria,Legume,Genomics, -proteobacteriaBackgroundSinorhizobiummeliloti 1021isasoilbacteriumthat establishesanitrogen-fixingsymbiosiswiththehost plants Medicagosativa (alfalfa)and Medicagotruncatula (reviewedin[1,2]).Theseplantsarenotonlyagriculturallyimportant,butarealsokeymodelorganismsfor studyingthesymbioticinteractionbetweenrhizobial bacteriaandtheirplanthosts.Thegoalsofthisstudyare toincreaseourunderstandingofthisprocessandprovide practicalinsightsthatmayleadtotheproductionofmore efficientsymbioticstrainsofrhizobia.Increasingtheefficiencyofsymbioticnitrogenfixationisimportantinthat itreducestheneedforindustrialproductionofnitrogen fertilizers,whichisextremelycostlyintermsofpetroleum andnaturalgas.In2007,theUSapplied13milliontons ofindustrially-producednitrogenfertilizertocrops[3]. Fertilizerscontinuetobeusedtoincreaseyieldsoflegumecrops[3],demonstratingthatthereisconsiderable roomforimprovementinthesesymbioticassociations. *Correspondence: kmjones@bio.fsu.edu†Equalcontributors1DepartmentofBiologicalScience,FloridaStateUniversity,BiologyUnitI, 230A,89ChieftainWay,Tallahassee,FL32306-4370,USA Fulllistofauthorinformationisavailableattheendofthearticle 2012Queirouxetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedtheoriginalworkisproperlycited.Queiroux etal.BMCMicrobiology 2012, 12 :74 http://www.biomedcentral.com/1471-2180/12/74

PAGE 2

S.meliloti fixesnitrogeninrootnodulesformedby thehostplant,convertingdinitrogengastoammonia. Thedevelopmentofthesenodulesrequiresthatseveral signalsbeexchangedbetweentheplantandtherhizobial bacteria.Flavonoidcompoundsproducedbyhostplants signal S.meliloti toproducelipochitooligosaccharides calledNodfactors(NFs)[4].NFactivatesmultiple responsesinhostplants,includingtightcurlingofroot hairsthattrapsbacterialcellswithinthecurl,andcell divisionsintherootcortex,whichestablishthenodule primordium[5,6].Thebacteriainvadeandcolonizethe rootsthroughstructurescalledinfectionthreads,which originatefrommicrocoloniesofbacteriatrappedinthe curledroothaircells[1,7].Newinfectionthreadsinitiate ateachcelllayer,eventuallydeliveringthebacteriato theinnerplantcortex[7].There,therhizobialbacteria areendocytosedbyrootcorticalcellswithinindividual compartmentsofhost-cellmembraneorigin[2,8]. Withinthesecompartments,signalsprovidedbythe plantandthelow-oxygenenvironmentinducethebacteriatodifferentiateintoaformcalleda “ bacteroid ” ,and tobeginexpressingnitrogenase,thenitrogen-fixingenzyme,andotherfactorsthatarerequiredforthesymbiosis[9,10]. Rhizobialfixationofdinitrogenrequiresnotonlythe expressionofnitrogenase(encodedbythegenes nifK and nifD [11]),butalsotheassemblyofcofactorsand largeinputsofenergyandreductant[12].Nitrogenfixationalsorequiresanitrogenasereductase,encodedby nifH [11];iron-molybdenumcofactorbiosynthesisproteins,encodedby nifB nifE and nifE ;andelectrontransferflavoproteinsandferredoxins( fixA,fixB,fixC,fixX ) [13-16].Bacteroidsalsoincreasetheirrespirationrate, increasingtheexpressionofthe fixNOQP cytochromec oxidaseoperons[17-20]. Manyoftheproteinsrequiredfornitrogenfixationare tightlyregulatedbyoxygen-sensingsystemsandareproducedbyrhizobialbacteriaonlywhentheyencountera low-oxygenenvironment[21].Nitrogenaseandsomeof theotherfactorsinvolvedinnitrogenfixationareextremelyoxygen-sensitive[22],thustheirexpression underinappropriateconditionswouldbeineffective. Evenundermicroaerobicconditions,mostrhizobialbacteriaarenotcapableofnitrogenfixationinthefreelivingstate[23].Thereasonsforthisarenotcompletely understood,thoughitisknownthatlegumesofthe invertedrepeat-lackingclade(IRLC),suchasalfalfaand M.truncatula, whichformindeterminate-typenodules, imposeaspecificdifferentiationprogramontheintracellularbacteria,mostlikelythroughtheactivityof plant-producedbioactivepeptides[9,24].Bacteroidsalso receivenutrientsfromthehostplant,suchasthecarbon sourcemalate[25-27].Multiplebacterialcellularprocessesanddifferentiationprogramscontributetothe successofthesymbiosiswithhostplants,andoneofour goalsistousecomparativegenomicstopredictpreviouslyuncharacterized S.meliloti openreadingframes (ORFs)thatmaybeinvolvedintheseprocesses,totest thesepredictions,andunderstandthemechanisms involved.Inotherbacterialspecies,comparativegenomicsofbacterialstrainshasbeenusefulinfindingnew genesthatareinvolvedinmetabolicpathwaysandin identifyingvirulencefactorsthatdistinguishpathogenic strainsfromcommensalstrains(examplesinclude: [28,29]).Inthisstudy,acomparisonofORFSfromnitrogen-fixing,plant-hostnodulatingrhizobiawithcloselyrelatednon-nitrogen-fixingbacteriahasidentifiedORFs thatareexpressedby Sinorhizobiummeliloti withinhost plantnodules.MethodsGenomecomparisonsSearcheswereconductedattheDepartmentofEnergy JointGenomeInstitute ’ sIntegratedMicrobialGenomes website,http://img.jgi.doe.gov/cgi-bin/pub/main.cgi.All ofthegenomestobecomparedwereselectedfromthe genomedisplayunderthe “ FindGenomes ” tab(see Table1forcomparedgenomes).Theselectedgenomes weresaved.The “ Phylogeneticprofiler ” forsinglegenes wasusedtofindgenesin Sinorhizobium/Ensifermeliloti withhomologsinthegenomestobeintersectedand withouthomologsinthegenomestobesubtracted(see Table1).Thesearcheswereconductedat20 – 80%identityandthecompletedataoutputislistedinAdditional file1:TableS1.BacterialstrainsandgrowthconditionsS.meliloti 1021strainsweregrownat30Cineither LBMC(LuriaBertani[Miller]mediumsupplemented with2.5mMMgSO4and2.5mMCaCl2),or1/10LB7%sucrosemedium,with1mMMgSO4and0.25mM CaCl2,orM9salts-10%sucrosemedium,supplemented with1 g/mLbiotin[40].Bacterialplatescontained 1.5%BactoAgar.Selectionsagainststrainscarryingthe sacB geneintheplasmidpK19mobsacwereperformed inM9supplementedwith10%w/vsucroseor1/10LB7%sucrose[41].Appropriateantibioticswereusedat thefollowingconcentrationsfor S.meliloti strains: streptomycin500or1000 g/mL;neomycin200 g/mL. E.coli strainsweregrownat37CinLBmedium[40], withappropriateantibioticsusedatthefollowing concentrations:kanamycin50 g/mL;chloramphenicol 10 g/mL.Constructionof S.meliloti mutantstrainsMutantstrainsof S.meliloti 1021withdisruptionsin ORFsdescribedinTable2wereconstructedbyamplifyinginternalORFfragmentsusingPhusionpolymeraseQueiroux etal.BMCMicrobiology 2012, 12 :74 Page2of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 3

(NewEnglandBiolabs,Ipswich,MA,USA)andcloning intotheplasmidpJH104,whichcarriesaneomycin/ kanamycinresistancemarker(JeanneHarris,Univ.Vermont,personalcommunication)[42].Insertionofthe pJH104plasmidalsocreatestranscriptionalfusionsto the uidA -glucuronidase(GUS)gene.Non-disrupting GUSinsertionsofsomeORFs(describedinTable2) wereconstructedbyamplifyingtheentireORForoperonandcloningtheproductintopJH104,andconjugatinginto S.meliloti .Deletionmutantstrainswere constructedbyamplifyingfragmentsflankingtheORF tobedeletedandcloningthefragmentsintothe sacB gene-containingsuicidevectorpK19mobsac[41].(Some fragmentswereinitiallyclonedintopCR-BluntII-TOPO usingtheZero-TOPO-Bluntcloningkit[Invitrogen,San Diego,CA,USA].)MutantstrainsarelistedinTable2. Primers(EurofinsMWGOperon,Huntsville,AL,USA) andrestrictionenzymes(NewEnglandBiolabs,Ipswich, MA,USA)usedforamplificationandcloningofdisruption,non-disruptinginsertion,ordeletionfragmentsare listedinAdditionalfile2:TableS2.Plasmidsweremobilizedinto S.meliloti bytriparentalconjugationas describedpreviously[43]. S.meliloti exconjugantswere selectedonLBMCmediumcontaining200 g/mLneomycinand1000 g/mLstreptomycin.Unmarkeddeletionstrainswereselectedforlossofthe sacB gene carriedbythepK19mobsacvectorbyplatingneomycinresistantexconjugantstoeitherM9salts – 10%sucrose mediumor1/10LB-7%sucrosemedium.Strainsconstructedbyphage M12transductionofplasmidinsertionsinto S.meliloti 1021aredenotedintheTablesas “ Xsd ” .Transductionsusingphage M12wereperformed accordingtopublishedprotocols[44].Foreachmutant produced,atleasttwostrainswereisolated.Forsomeof themutants,includingthosewhichcarryanunmarked ORFdeletion,multipleindependentisolateswere obtainedbyselectingexconjugantsfrommultipleindependentconjugations.Formostofthemutantscarrying aninsertionofthepJH104plasmid,theindependentisolatesweretheoriginalisolateandstrainsconstructedby transductionoftheneomycin-resistancemarkerinto wildtype S.meliloti 1021viaphage M12[44].PlantnodulationassaysThehostplant Medicagosativa (alfalfa)cv.Iroquois waspreparedforinoculationwith S.meliloti asinLeigh etal. (1985)withmodifications:seedsweresterilized for5minutesin50%bleach,rinsedinsterilewater,and germinatedfor3dayson1%w/vplantcellculturetestedagar/water(Sigma,St.Louis,MO,USA)[45]. Seedlingswerethenmovedtoindividual100mmx 15mmJensen ’ smediumplates[46],andinoculated with100 LofOD600=0.05 S.meliloti oftheappropriatestrain.PlantsweregrowninaPercivalAR-36L Table1GenomeORFscomparedwith S.meliloti 1021GenomeSubtractedorintersectedLifestyle Agrobacteriumtumefaciens C58(Cereon)[ 30 31 ]subtractedplantpathogen Agrobacteriumtumefaciens C58(Dupont)[ 30 31 ]subtractedplantpathogen Bartonellabacilliformis KC583subtractedmammalianpathogen Bartonellahenselae Houston-1subtractedmammalianpathogen Bartonellaquintana Toulousesubtractedmammalianpathogen Bartonellatribocorum CIP105476subtractedmammalianpathogen Brucellaabortus bv19-941subtractedmammalianpathogen Brucellacanis ATCC23365subtractedmammalianpathogen Brucellamelitensis 16Msubtractedmammalianpathogen Brucellamelitensis bvAbortus2308subtractedmammalianpathogen Brucellaovis ATCC25840subtractedmammalianpathogen Brucellasuis ATCC23445subtractedmammalianpathogen Brucellasuis 1330subtractedmammalianpathogen Caulobactercrescentus CB15[ 32 ]subtractedfree-living Caulobacter sp.K31[ 33 34 ]subtractedfree-living Bradyrhizobiumjaponicum USDA110[ 35 ]intersectednitrogen-fixingplantsymbiont Mesorhizobiumloti MAFF303099[ 36 ]intersectednitrogen-fixingplantsymbiont Rhizobiumetli CFN42[ 37 ]intersectednitrogen-fixingplantsymbiont Rhizobiumleguminosarum bv.viciae3841[ 38 ]intersectednitrogen-fixingplantsymbiont Sinorhizobiummedicae WSM419[ 39 ]intersectednitrogen-fixingplantsymbiont Queiroux etal.BMCMicrobiology 2012, 12 :74 Page3of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 4

Table2 S.meliloti 1021-derivedmutantstrainsORFPredictedfunctionLength (aminoacids) TypeofmutationStrainname SMc01562hypotheticalprotein96deletion SMc01562.6 SMc01562.25 SMc01562.100 SMc01562hypotheticalprotein96non-disruptinginsertionof pJH104GUSmarker A104U.original A104U.Xsd1 A104U.Xsd6 A104U.Xsd25 A104U.Xs100 SMc01986hypotheticalprotein119deletion SMc01986.1 SMc01986.6 SMc01986.25 SMc01986.100 SMc01986hypotheticalprotein119non-disruptinginsertionof pJH104GUSmarker C104.1A.Xsd1 C104.1A.original C104.2B.Xsd100 SMc00135hypotheticalprotein243deletion SMc00135.B1 SMc00135.B17 SMc00135hypotheticalprotein243non-disruptinginsertionof pJH104GUSmarker B104.3A B104.4B B104.2C SMc01422hypotheticalprotein (probableoperonwith SMc01423,SMc01424) 128deletion(SMc01422,SMc01423, SMc01424alldeletedinthisstrain) SMc01422-24.D21 SMc01422-24.D29 SMc01423probablenitrilehydratase subunit 219deletionsameasabove SMc01424probablenitrilehydratase subunit 213deletionsameasabove SMc01424-01422hypotheticalprotein(probable operonwithSMc01423,SMc01422) 213non-disruptinginsertionof pJH104GUSmarker D104.2A D104.3B D104.1C SMa0044hypotheticalprotein89deletion SMa0044.c1 SMa0044.c6 SMa0044.c10 SMa0044non-disruptinginsertionof pJH104GUSmarker 89SMa0044.104.1A SMa0044.104.1B SMa0044.104.4C SMb20431hypoth.arylmalonate decarboxylase 261ORF-disruptinginsertionof pJH104GUSmarker SMb20431.original SMb20431.Xsd1 SMb20360hypotheticalprotein243ORF-disruptinginsertionof pJH104GUSmarker SMb20360.original SMb20360.Xsd1 Queiroux etal.BMCMicrobiology 2012, 12 :74 Page4of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 5

incubator(Perry,IA,USA)at21C,with60 – 70%relativehumidity,and100 – 175 molm 2s 1light.Plants weremeasuredat5weeksand6.5weeksofgrowth. t-tests(unpaired,two-tailed)wereperformedinMicrosoftExcelandinGraphPad(http://www.graphpad.com/ quickcalcs/ttest1.cfm?Format=C). Nodulationcompetitionassayswereperformedin thesamewayastheplantassaysdescribedabove, exceptthatstrainstobetestedincompetitionagainst oneanotherwerepreparedasamixed1:1inoculum immediatelybeforeinoculation.Bacteriawereharvested fromnodulesafter5or6.5weeksofgrowthbyexcising thenodulesfromroots,surfacesterilizingin20%bleach for5min.,washinginsterile,distilledwater,andcrushingthenodulesin1.5mLtubeswithamicro-pestle (Kimble-Chase,Vineland,NJ),inLB+0.3Mglucose [45].Dilutionsofthematerialfromcrushednodules wereplatedonLBMC+500 g/mLstreptomycin.ColonieswerepatchedfromtheseplatestoLBMC+500 g/ mLstreptomycinand200 g/mLneomycintodetermine thefractionofbacteriathatcarrytheneomycinresistancemarkerintheinsertionplasmidpJH104.Detectionof -glucuronidaseactivityandimagingofroot nodules -glucuronidaseexpressionbybacteriawithinnodules wasdetectedbyexcisingnodules,surfacesterilizing with20%bleachfor5min.,rinsinginsterilewater,and staininginX-glucbuffer(1mM5-bromo-4-chloro-3indolyl-beta-D-glucuronicacid,cyclohexylammonium salt;0.02%SDS;50mMNa-phosphate,pH7)[47]for theamountoftimeindicatedinTable3.Wholenodules wereimagedonanAZ100Multi-ZoomMicroscope equippedwithaDS-Fi1,5Megapixelcolorcamera (NikonInstrumentsU.S.,Melville,NY). -glucuronidase expressionbybacteriaonLBMCplateswasdetectedby streakingbacteriatoplatesthathadbeenspreadwith 40 LofX-glucsolution(100mM5-bromo-4-chloro-3indolyl-beta-D-glucuronicacid,cyclohexylammoniumsalt solutionindimethylformamide).ResultsComparisonsof Sinorhizobiummeliloti openreading frameswiththoseofotherrhizobiaandwithnonnitrogenfixing -proteobacteriaRhizobialfunctionsrequiredforsymbioticnitrogenfixationwithlegumeplantshavetypicallybeendiscovered throughtheclassicalbacterialgenetictechniqueof transposonmutagenesis,followedbyscreeningmutants forlossofsymbioticfunction.Wehaveusedanalternativecomparativegenomicsstrategytosearchforrhizobialgenesinvolvedinsymbiosis.Inthisapproach, searchesoftheJointGenomeInstitute,IntegratedMicrobialGenomes(JGIIMG)system[48]wereperformedto findORFsthat S.meliloti 1021shareswiththesymbiotic nitrogen-fixing -proteobacteria( -rhizobia) S.medicae WSM419, Rhizobiumetli CFN42, Rhizobiumleguminosarum bv.viciae, Mesorhizobiumloti MAFF303099,and Bradyrhizobiumjaponicum USDA110.Anovelaspectof thisstrategyisthatthesesearcheswererestrictedbyprior eliminationofall S.meliloti ORFsthatarepresentinany of15non-nitrogen-fixing,non-symbiotic -proteobacteria Table2 S.meliloti 1021-derivedmutantstrains (Continued)SMc03964hypotheticalprotein300ORF-disruptinginsertionof pJH104GUSmarker SMc03964.original SMc03964.Xsd6 SMc00911hypotheticalprotein275ORF-disruptinginsertionof pJH104GUSmarker SMc00911.original SMc00911.Xsd1 SMc00911.original2 SMa1334hypotheticalprotein398ORF-disruptinginsertionof pJH104GUSmarker(mayhavea polareffecton3 genes Sma1332,-1331,-1329) SMa1334.original SMa1334.Xsd1 SMc01266hypotheticalprotein438ORF-disruptinginsertionofpJH104 GUSmarker(mayhaveapolar effecton3 geneSmc01265) SMc01266.original SMc01266.Xsd1 greA transcriptionelongationfactor158ORF-disruptinginsertionof pJH104GUSmarker greA.12.4.1a expA1(wgaA) EPSIIbiosynthesisenzyme490ORF-disruptinginsertionofTn5-Nm in expA „ symbioticallyproficient, competitorassaystrain expA125::Tn5.Xsd1 Queiroux etal.BMCMicrobiology 2012, 12 :74 Page5of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 6

(specieslistedinTable1).(SeeMaterialsandMethodsfor searchprocedure.)Thegenomesusedintheanalysiswere chosenbasedontherhizobialgenomesavailableinthe JGIIMGdatabasewhentheanalysiswasinitiallyperformed.Thesearcheswereconductedatmultipleidentity levels(20% – 80%),andtheoutputdatafromallthe searchesispresentedinAdditionalfile1:TableS1.The genomesubtractionseliminatedgenescommonto proteobacteriawithnon-symbioticlifestyles.Forexample, asearchconductedat50%identity,intersectingthe S. meliloti ORFswithhomologsinthe5 -rhizobiaspecies yields1281genes.However,whenthesearchforhomologsisconductedwithsubtractionoftheORFsfromthe 15non-rhizobialspecies,thesearchyieldis58genes (Additionalfile3:TableS3). Theresultofthesearcheswasalistof139ORFscommontothe -rhizobia(listedinAdditionalfile3:Table S3),butnotfoundinthenon-nitrogen-fixing,nonsymbiotic -proteobacteria.Amongthese139ORFs were11genesknowntobeinvolvedinnitrogenfixation (Table4andAdditionalfile3:TableS3),including: nifH nifD nifK nifB nifE nifN fixA fixB ,and fixC (see Table3Expressionof -glucuronidase(GUS)fusionsORFstrain%ofnodules withGUS expression Strength ofnodule GUSexpression StainingtimePatternofnodule GUSexpression Free-living GUS expression N/A S.meliloti 1021wildtype (negativecontrol) 0/39=0% variablenone SMc00911 SMc00911.original18/20=90%++++1.5 … 3.75hrwholenodule+ SMc00911.Xsd118/18=100%++++1.5 … 3.75hrwholenodulen.d. SMc00911.original2n.d.n.d.N/AN/A+ SMb20360 SMb20360.original8/13=62%++3 … 5hrinvasionzone-fixationzone SMb20360.Xsd113/16=81%++3 … 5hrinvasionzone-fixationzone SMc00135 B104.3A6/8=75%+2 … 3hrinvasionzone-interzone+ B104.4B8/8=100%+2 … 3hrinvasionzone-interzone++ B104.2C6/8=75%++2 … 3hrinvasionzone-interzone++ SMc01562 A104U.original7/8=88%+4 … 6hrinterzone A104U.Xsd13/7=43%+/ 4 … 6hrinterzone-fixationzonen.d. A104U.Xsd68/8=100%+4 … 6hrinterzone-fixationzonen.d. A104U.Xsd253/8=38%+/ 4 … 6hrinterzone-fixationzonen.d. A104U.Xs1004/9=44%+4 … 6hrfixationzonen.d. SMc01266 SMc01266.original13/18=72%+3hrinvasionzone-fixationzone+/ SMc01266.Xsd113/18=72%++3hrinvasionzone SMc03964 SMc03964.original8/15=53%++3 … 5hrinterzone+/ SMc03964.Xsd69/19=47%++3 … 5hrinterzone-fixationzone SMc01424-22 D104.2A0/8=0% 4 … 6hrN/A+/ D104.3B7/8=88%++4 … 6hrinvasionzone-interzone+/ D104.1C6/8=75%+4 … 6hrinvasionzone-fixationzone+/ SMa0044 SMa0044.104.1A4/8=50%+/ 6 … 7hrinvasionzone-interzone+++ SMa0044.104.1B4/8=50%+/ 6 … 7hrinterzone+++ SMa0044.104.4C4/8%50%+/ 6 … 7hrinterzone+++ SMb20431 SMb20431.original10/16=63%+5 … 12hrinvasionzone-fixationzone SMb20431.Xsd111/15=73%+5 … 12hrinterzone SMc01986 C104.1A.Xsd10/6=0% 24hrN/An.d. C104.1A.originaln.d.n.d.24hrn.d.+/ C104.2B.Xsd1002/18=11%+/ 24hrfixationzonen.d. SMa1334 SMa1334.original0/11=0% 5 … 24hrN/A SMa1334.Xsd10/13=0% 5 … 24hrN/A Queiroux etal.BMCMicrobiology 2012, 12 :74 Page6of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 7

Introduction)and8knowntobeinvolvedinNodfactor production,including nodA nodB nodC nodJ and nodI [5],thus13.7%(19/139)oftheORFsselectedbythis comparativegemonicsapproacharealreadyknownto beimportantforsymbioticfunction. Therewerealso44hypotheticalproteins/proteinsof unknownfunctionamongthe139ORFsdetectedinthe comparativegenomicscreen.Thepredictedfunctionsof theremainingORFsincludedtransposases transcriptionalregulators,transportproteins,andadenylate/guanylatecyclases(Table4).Theseareclassesofgenesthat mayparticipateinmanyofthefunctionsthatdistinguish -rhizobiafromtheirnon-symbiotic -proteobacterial relatives,suchassignalingtothehostplant,reprogrammingtheirmetabolismfornitrogenfixation,andimportingspecificnutrientsanddifferentiationsignalsfromthe plant[9,10,49].Also,atypicaladenylatecyclaseshave beennotedbeforeintherhizobia[50].Constructionandsymbiosisassaysofmutantsin conservedgenesThirteenofthe139conservedORFswerechosenfor furtherstudybecausetheyareofundeterminedfunction in S.meliloti andhavenoclosehomologsinthe S.meliloti genomethatmightbeexpectedtoprovideredundantfunction.SixofthelongerORFs,including SMc00911,weredisruptedbycloningasmallinternal ORFfragmentintotheplasmidpJH104,conjugatingthe plasmidinto S.meliloti 1021,andselectingforsinglecrossoverinsertion/disruptionmutants.(Additionalfile 2:TableS2listsprimersequencesanddisruptionfragmentsizesandpositions.)Forthe6remainingORFs,3 thatareunder750bplong(SMc01562,SMc01986 andSMc00135)and3thatareallinasingleoperon (SMc01424,SMc01423,andSMc01422),deletionwas judgedtobeabetterstrategythandisruption. SMc01424,SMc01423,andSMc01422werealldeleted asasinglesegmentfromthestartcodonofSMc01424 tothestopcodonofSMc01422.Theendpointsofthe individualdeletionsofSMc01562,SMc01986,and SMc00135weredictatedbythepositionofthemost suitablePCRprimers.(Additionalfile2:TableS2lists primersequencesanddeletionsizesandpositions.)Eitherthedisruptionorthedeletionstrategyisexpected toresultinastrainthatdoesnotproduceafull-length versionoftheproteinencodedbythatORF.These ORFsandtheinsertionand/ordeletionmutantstrains ofeacharelistedanddescribedinTable2.Theresultingmutantstrainswerethentestedforsymbioticproficiencyonthehostplantalfalfa. Fortheinitialphenotypicanalysis,theabilityofthe mutantstosuccessfullyprovidetheplantswithfixednitrogenwasdetermined.Alfalfaplantswereinoculated withthebacterialmutantsandafter5weeksofgrowth, theshootlengthattainedonnitrogen-freemediumwas comparedwithplantsinoculatedwiththe S.meliloti 1021wildtypeasthepositivecontrolanduninoculatedplantsasthenegativecontrol.Figure1shows theshootlengthofalfalfaplantsinoculatedwithwild type S.meliloti 1021orwithdisruptionmutantstrains oftheORFsSMb20360,SMb20431,SMc00911, SMa1344,SMc01266,andSMc03964.Alfalfaplants inoculatedwiththesestrainsattainasimilaraverage shootlengthasthatofthewildtype,demonstratingthat allofthesestrainsareabletoformasuccessfulsymbiosiswiththishostplant.Figure2presentsthesametype ofassayasFigure1fordeletionmutantsintheORFs SMc01562,SMc01986,SMc01424-22,SMc00135,and SMa0044.AdditionaldataontheplantassaysinFigures1 and2ispresentedinTable5.Thenumberofplants inoculatedwitheachstrain,theaveragenumberofmature,pinknodulesperplantandtheaveragenumberof whitepseudonodulesperplantareshown.Allofthese mutantstrainsareabletomountasuccessfulsymbiosis withthehostplantalfalfa.SMc00911isthemoststronglyexpressedinthenodule oftheconservedORFSTodetermineifthe13ORFsanalyzedinthisstudy mightplayaroleinsymbiosis,despitethefactthatthey arenotstrictlyrequiredforsymbiosis,theexpression patternofeachoftheseORFswasdeterminedbothfor bacteriawithinthenoduleandinthefree-livingstate. TheSMc00911ORFisverystronglyexpressedbybacteriawithinthenodule(Figure3B – F),butitexpressed ataverylowlevelbyfree-livingbacteriaonLBMC plates(Figure3G).ThenodulesshowninFigure3are expressing -glucuronidase(GUS)fromapJH104plasmidinsertioninSmc00911.Thenodulesshownwere stainedfor3.75hr.Thereisstrongstainingthroughout thenodule,withslightlyweakerstainingattheinvasion zonenearthedistalendofthenodule.ThenoduleexpressionoftheSMc00911::GUSfusionismuchstronger Table4Functiondistributionofthe139ORFsfrom genomesearches(SeeAdditionalfile3:TableS3for completegenelist)FunctionNumberofORFs Nitrogenfixation11 Nodfactorproduction/modification8 Transposase10 Predictedtranscriptionalregulator8 Predictedtransportprotein14 Predictedadenylate/guanylatecyclase7 Otherpredictedfunction37 Hypotheticalprotein44 Queiroux etal.BMCMicrobiology 2012, 12 :74 Page7of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 8

thantheexpressionofanyoftheotherfusionstested (seeFigure4andTable3).Incontrast,SMc00911is expressedataverylowlevelbyfree-living S.meliloti carryingtheSMc00911::GUSfusiongrownonLBMC plates(Figure3GandTable3).Forcomparison, Figure3Galsoshowsthata greA ::GUSfusionstrainof S. meliloti constructedwiththesamereporterinsertion plasmid,pJH104,isstronglyexpressedundertheseconditions.Table3summarizestheexpressiondataforall oftheGUSfusionstrains. TwooftheotherORFstested,SMb20360and Smc00135,arealsostronglyexpressedinnodules (Figure4B – E,Table3),andanothersix,SMc01562, SMc01266,SMc03964andthethreeORFsinthe SMc01424-22operonaremoderatelyexpressed (Figure4F – M,Table3).Ofthese,onlySMc00135is expressedatapproximatelythesamelevelbybacteria withinthenoduleandbyfree-livingbacteria(Additional file4andAdditionalfile5showimagesofthefree-living expressionofGUSfusionsofalltheORFstested).However,noneoftheotherORFsthatareexpressedinthe noduleareexpressedasstronglyasSMc00911(Figure3 andFigure4).TwooftheORFs,SMa0044and SMb20431,areexpressedataverylowlevelinthenodule,andnonoduleexpressionwasdetectedfor SMc01986andSMa1334(Figure4).Sma0044hasanunusualexpressionpatterninthatitisexpressedstrongly byfree-livingbacteria(Additionalfile5A),butits Figure2 Plantshootlengthincm,5weeksafterinoculationwithdeletionmutantstrains(summarizedinTable 3 ). ForeachoftheORF deletions,theplantphenotypeofatleasttwoisolates/andortransductantsofeachstrainareshown.Meanvaluesaregivenabovegraphbars. Errorbarsrepresentstandarderrorofthemean.Asterisksindicatesampleswithmeanheightssignificantlydifferentfromthewildtype.The numberofplantstestedandthenumberofnodules/plantfortheseassaysarepresentedinTable4. Figure1 Plantshootlengthincm,5weeksafterinoculationwithinsertionmutantstrains(mutantstraininformationissummarizedin Table 3 ). Foreachofthe6ORFdisruptions,theplantphenotypeoftheoriginalisolateandthatofaphage M12transductantofthatstrainare shown.Meanvaluesaregivenabovegraphbars.Errorbarsrepresentstandarderrorofthemean.Asterisksindicatesampleswithmeanheights significantlydifferentfromthewildtype.Thenumberofplantstestedandthenumberofnodules/plantfortheseassaysarepresentedinTable4. Queiroux etal.BMCMicrobiology 2012, 12 :74 Page8of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 9

expressionappearstobemuchreducedinthenodule (Figure4N – O). BecauseofthestrongexpressionofSMc00911bybacteriainthenodule,theSMc00911mutantstrainswere chosenforfurtherstudyincompetitionexperiments (seebelow).AninsertionmutantofSMc00911out-competesthe S. meliloti 1021wildtypefornoduleoccupancyMany S.meliloti mutantstrainsthatareabletoforma successfulsymbiosiswhensinglyinoculatedonhost plantsaredeficientintheabilitytosuccessfullycompete fornoduleoccupancyagainstthewildtypestrainina mixedinfection[42,51].Competitivenodulationexperimentsarelikelytobeabetterapproximationofthesituationthatrhizobialbacteriaencounterinthesoil,where theymaybecompetingagainstseveraldifferentrhizobial strainsforhostplantinvasionandnoduleoccupancy. TheSMc00911insertionmutantstrainswerechosenfor competitionanalysisbecausethisORFisstrongly expressedinthenoduleandthesestrainsmightbe expectedtobeatacompetitivedisadvantageinthe Table5MeannodulenumberORFStrainnameNumber ofalfalfa plantstested Meannumber pinknodules/ plantstd.error Meannumberwhite pseudonodules/ plantstd.error N/A S.meliloti 1021wildtype, dataset1(seeFigure 1 ) 911.91.03.2+1.2 SMb20360 SMb20360.original817.42.54.51.2 SMb20360.Xsd11014.71.74.41.4 SMb20431 SMb20431.original1112.81.63.00.6 SMb20431.Xsd11113.31.93.80.8 SMc00911 SMc00911.original1114.32.53.30.8 SMc00911.Xsd11115.31.83.21.1 SMa1334 SMa1334.original1015.72.15.70.9 SMa1334.Xsd11116.41.13.61.7 SMc01266 SMc01266.original1114.42.44.20.5 SMc01266.Xsd11117.81.64.61.2 SMc03964 SMc03964.original1116.31.64.20.5 SMc03964.Xsd61015.22.34.00.9 N/A uninoculated,dataset1 (seeFigure 1 ) 500 N/A S.meliloti 1021wildtype, dataset2(seeFigure 2 ) 17912.50.53.20.3 SMc01562 SMc01562.62414.11.32.20.4 SMc01562.252511.61.22.50.5 SMc01562.1002411.80.92.00.6 SMc01986 SMc01986.12618.01.84.50.8 SMc01986.62615.32.14.40.8 SMc01986.252517.22.36.81.1 SMc01986.1002516.81.86.71.0 SMc01424-22 SMc01422-24.D2111013.10.73.70.4 SMc01422-24.D2910911.10.63.60.3 SMc00135 SMc00135.B18114.00.72.80.3 SMc00135.B177613.50.93.30.4 SMa0044 SMa0044.c12411.81.34.20.6 SMa0044.c62512.61.23.00.8 SMa0044.c102413.51.22.00.5 N/A uninoculated,dataset2 (seeFigure 2 ) 8200.10.1 Queiroux etal.BMCMicrobiology 2012, 12 :74 Page9of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 10

absenceofthefull-lengthSMc00911protein.However, incontrasttoexpectations,theSMc00911insertionmutantstrainsstronglyout-competethe S.meliloti 1021 wildtypestrainfornoduleoccupancyinamixed1:1infection(Table6).Ofthenodulestestedfromplants inoculatedwitha1:1mixtureof1021wildtypeandan SMc00911insertionmutant,allofthenoduleswere colonizedbyeithertheSMc00911insertionmutant aloneorbyamixtureofthemutantandthewildtype (Table6).Lessthan22%ofthemixed-inoculumnodules werecolonizedby1021wildtypealone.Also,allofthe mixednodulescontainedalargerproportionof SMc00911insertionmutantbacteriathan1021wildtype bacteria(Table6).Therecoveredbacteriafromoneof the8nodulesthathadbeeninoculatedwiththe SMc00911.Xsd1strainaloneincludedasmallnumberof neomycin-sensitivecolonies(Table6,line3).Thissuggeststhatthegenedisruptionplasmidinsertedinthe SMc00911ORFislostbybacteriainthenoduleatavery lowrate.Takentogether,thesecompetitionresults suggestthatdisruptionoftheSMc00911ORFactually confersacompetitiveadvantageto S.meliloti inthe symbiosiswithhostplants.TheSMc00911ORFispredictedtoencodea275aminoacidproteinwitha rhodanese-likesulfurtransferasedomainfromamino acids7 – 100andachromate-resistanceproteindomain fromaminoacids122 – 256[52].TheSMc00911mutants carrythepJH104-GUS-expression/disruptionplasmid insertedatnucleotideposition597outof828total nucleotides,whichwouldresultintheproductionofa truncatedproteincontainingonlyaminoacids1 – 199, basedonthe S.meliloti 1021genomesequence[53,54]. ThustheSMc00911insertionmutantsarepredictedto produceaproteinthatcontainsthewholerhodaneselikesulfurtransferasedomain,butonlyaportionofthe chromate-resistanceproteindomain. IncontrasttotheSMc00911insertionmutants,deletion mutantsofSMc01562(whichisexpressedinthenodule, butatamuchlowerlevelthanSMc00911(Figure4))are abletocompeteaseffectivelyas S.meliloti 1021wildtype Figure3 Expressionof -glucuronidase(GUS)-encodingreportergene uidA insertedwithinSMc00911. S.meliloti withinalfalfaroot nodules( B … F )expressGUSinsertedinSMc00911throughoutthenodule.Panel A showsanalfalfanoduleinvadedbywildtype S.meliloti 1021 thatdoesnotexpressGUS(subjectedtothesamestainingprocedureas B … F ).(RootsinB,C,andDwereinoculatedwithstrainSMc00911.Xsd1. RootsinEandFwereinoculatedwithstrainSMc00911.original.)Noduleswerestainedfor3.75hrafter5weeksofgrowthpost-inoculation.Scale barscorrespondto0.1mm.Panel G showsSMc00911-controlledGUSexpressionin S.meliloti grownonsolidLBMCmedium.Wildtype S.meliloti 1021isshownasanegativecontrolforGUSexpressionandastraincarryingthesameGUSinsertionplasmidinthe greA geneisshownasa positivecontrolforGUSexpressioninfree-livingcells.StrainSMc00911.originalanda M12transductantofthisstrainweretestedonplants. Queiroux etal.BMCMicrobiology 2012, 12 :74 Page10of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 11

Figure4 Expressionof -glucuronidase(GUS)-encodinggene uidA expressedunderthecontrolofthepromoterelementsofthe followingORFs:SMb20360(BandC);SMc00135(DandE);SMc01562(FandG);SMc01266(HandI);SMc03964(JandK);SMc01424-22 (LandM);SMa0044(NandO);SMb20431(PandQ);SMc01986(RandS);SMa1334(TandU). SMb20360andSMc00135arestrongly expressedinthenodules.(SeeTable3forpercentageofnoduleswithGUSexpressionandstainingtimes.)SMc01562,SMc01266,SMc03964and theSMc01424-22operonareexpressedatamoderatelevelinthenodules.TheremainingORFsareexpressedataverylowlevelinthenodule (ornotatall). S.meliloti 1021wildtypeisshowninPanel A asanegativecontrolforGUSexpression.Scalebarscorrespondto0.1mm. Queiroux etal.BMCMicrobiology 2012, 12 :74 Page11of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 12

againstacompetitorassaystraincarryinganeomycinresistancemarker(datanotshown),suggestingthatthe lossofthisproteinconfersneitherasymbioticdisadvantagenoranadvantageto S.meliloti 1021.DiscussionSmc00911,aconservedrhizobialORFexpressedstrongly inthenoduleOurcomparativegenomicsscreenhasidentifiedan S.meliloti 1021ORF(SMc00911)thatisstrongly expressedwithinhostplantnodules,butisexpressedin thefree-livingstateataverylowlevel.Surprisingly,disruptionofthisORFconfersacompetitiveadvantagefor noduleoccupancyon S.meliloti 1021.Smc00911ispredictedtoencodea275aminoacidproteinwithoverall similaritytoSodM-like(superoxidedismutase-like)proteins[55,56].Thereare57 “ SodM-likeproteins ” with >40%identitytoSMc00911intheNCBIdatabase[56]. SMc00911containstwodistinct,conserveddomains:a 94aminoaciddomain(aminoacids7 – 100)similarto theGlpEsufurtransferase/rhodanesehomologydomain (cd01444),anda135aminoacid(aminoacids122 – 256) chromate-resistance-expo rtedproteindomain(pfam09828) [52].TheSMc0911mutantstrainsconstructedinthis studyarepredictedtoproduceaproteinconsistingof thefirst199aminoacidsofthefull-lengthproteinplus fouraminoacidsencodedbythemultiplecloningsiteof pJH104,beforeencounteringastopcodon(Melanie Barnett,StanfordUniversitypersonalcommunication) [53,54].Thistruncatedproteinproductwouldinclude theentirerhodanese-homologydomainandapproximatelyhalfofthechromate-resistanceproteindomain. Onepossibilityisthatthecompetitiveadvantagethatthe SMc00911-insertionmutantstrainshaveagainstthe1021 wildtypestrainisduetotheexpressionofthistruncated protein,ratherthansimplyaloss-of-functionofthefulllengthprotein.EventhoughSMc00911isannotatedasa “ SodM-like ” proteinintheNCBIdatabase[53,54,56], thereareonlytwoshortsegmentsofsimilarity(8amino acids[38%identity]and11aminoacids[36%identity]) withaproteinconfirmedtobeaSodMfrom Xanthomonascampestris pv. campestris (accessionno.p53654) [57].Thus,sincetheN-terminalsimilarityofSMc00911 totheGlpEsufurtransferase/rhodanesehomologydomainandtheC-terminalsimilaritytothechromateresistanceproteindomainarebothgreaterthanthesimilarityofthisproteintoSodM, “ SodM-like ” maynotbe themost-appropriateannotationforthisORF.Thereare two sod ORFsinthe S.meliloti 1021genome, sodB (SMc00043)(SMc02597)andabacteriocuprein-family sodC (SMc02597)[2,53,54].An S.meliloti 1021 sodB loss-of-functionmutantformsafunctionalsymbiosis withhostplants[58],whilethesymbioticphenotypeofa sodC mutanthasnotbeenreported.Expressionofother hizobialconservedORFSAlthoughtheyarenotrequiredfordevelopmentofa functionalsymbiosisby S.meliloti 1021,theORFs SMb20360andSMc00135arealsostronglyexpressedin nodules,whileSMc01562,SMc01266,SMc03964and theSMc01424-22operonaremoderatelyexpressed (Figure4;Table3).However,theexpressionof SMc00135isnotspecifictothenodule(Figure4and Additionalfile5).SMb20360ispredictedtoencodea proteinoftheClp-proteasesuperfamily(COG0740), withspecificsimilaritytoClpP[52].Polarlocalizationof theClpXPproteasecomplexwithin S.meliloti cellshas beenfoundtobeimportantfor S.meliloti bacteroiddifferentiation[59],anditispossiblethatClpPproteases playaroleinthebacteroiddifferentiationprocess.Interestingly,inanotherstudy,asignature-taggedmutantin SMb20360wasfoundtobehighlycompetitiveforsurvival,inthefree-livingstate,incompetitionexperiments undersalt-anddetergent-stressedconditions[60]. SMc01562ispredictedtoencodeamemberofthe GYD-domaincontainingproteinsuperfamily(COG4274) [52].Nofunctionhasbeenreportedforthisproteinfamily[56].SMc01266ispredictedtoencodeamemberof theVonWillebrandfactortypeA(vFWA)superfamily (cl00057),howeverproteinscontainingavFWAdomain Table6SMc00911-disruptionstrainsout-compete S.meliloti 1021wildtypefornoduleoccupancyInoculumNumberof nodules tested* Numberofnodules containingno neomycin-resistant bacteria Numberofnodules containingonly neomycin-resistant bacteria Numberofnodules containingamixture ofneomycin-resistant andsensitivebacteria Averagepercent ofneomycin-resistant bacteriainmixed nodules S.meliloti 1021wildtype (neomycin-sensitive) 84=100%0=0%0=0%N/A SMc00911.original (neomycin-resistant) 160=0%16=100%0=0%N/A SMc00911.Xsd1 (neomycin-resistant) 160=0%15=93.8%1=6.3%95.2%0.00% SMc00911.original:1021 „ mixed1:1327=21.9%18=56.3%7=21.9%67.4%14.2% SMc00911.Xsd1:1021 „ mixed1:1312=6.5%21=67.7%8=25.8%76.7%9.8%*1 – 2nodules/plantwereanalyzed.Queiroux etal.BMCMicrobiology 2012, 12 :74 Page12of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 13

participateinawidevarietyoffunctions[61].Expression ofSMc01266haspreviouslybeenshowntoincreasein bacteroids[62](referenceSupplementalDataset3),and duringphosphatestress[63].Smc03964ispredictedto possessatwin-arginineexportsignal[64],andtoencode amemberofthemetallophosphatasesuperfamily (cl13995),agroupofphosphataseswithdiversefunctions [52].ORFsSMc01424,SMc01423,andSMc01422appear tobepartofasingleoperonandtheyencode,respectively,apredictednitrilehydratasealphasubunitprotein, anitrilehydratasebetasubunitprotein,andanitrile hydrataseactivatorprotein[53,54].Nitrilehydratases functioninthedegradationofxenobioticcompounds, buttheyarealsoinvolvedintryptophanmetabolism,specificallyintheconversionof3-indoleacetonitriletoindole-3-acetamide,whichisaprecursoroftheplant hormoneauxin[65,66].SMa0044hasanunusualexpressionpatterninthatitisexpressedataverylowlevelinapproximatelyhalfofthenodulestested(Table3;Figure4), butisexpressedquitestronglybyfree-living S.meliloti on LBMCmedium(Additionalfile5).SMa0044ispredicted toencodeamemberoftheDUF2277superfamily,which ishasnoknownfunction[52].ConclusionsThegoalofthisstudywastoidentify S.meliloti 1021 ORFsinvolvedinhostplantnodulationandnitrogenfixation.Thecomparativegenomicsmethodweemployed wasabletorediscover19ORFsthathavepreviously beenshowntobeimportantfornodulationand/ornitrogenfixation.Theearlierstudiesthatidentifiedthese genes,inmostcases,employedtheclassicalbacterial genetictechniquesoftransposonmutagenesis,followed bystrainisolationandphenotypicscreening[11,67][68]. Ourstudyidentified9additional S.meliloti ORFs(out ofthe13weanalyzed)thatwehaveshownareexpressed primarilyinhostplantnodules.Howevernoneofthese newlyidentifiedORFswererequiredfordevelopmentof afunctionalsymbiosisundertheconditionswetested. Ourresultssuggestthattheaccumulatedtransposon screensforessential S.meliloti nodulation/nitrogenfixationgenesmaybenearingsaturation.However,the comparativegenomicsmethoddescribedabovemightbe veryeffectiveforidentifyingfactorsinvolvedintheproductionofaphenotypecommontoagroupofbacterial speciesthathavenotyetbeenstudiedbyclassicaltransposonmutagenesisscreens.AdditionalfilesAdditionalfile1: TableS1. JointGenomeInstitute,Integrated MicrobialGenomesPhylogeneticProfilesearchdataonsinglegenes. Additionalfile2: TableS2. Primersusedtoamplify S.meliloti 1021 fragmentsforconstructionofinsertionmutantsanddeletionmutants. Additionalfile3: TableS3. Genelistof139ORFscompiledfromsearch datainAdditionalfile1:TableS1. Additionalfile4: Free-livingexpressionof -glucuronidase(GUS) underthecontrolofthepromotersofthefollowingORFs: A) clockwisefromlowerleft „ SMc01266; greA (positivecontrolforGUS expression); S.meliloti 1021wildtype(negativecontrolforGUS expression);SMb20431;SMa1334.(ThecroppedplatewedgesinpanelA areallfromthesameplate.)B)clockwisefromlowerright „ SMc01986; SMc01562;SMc03964; greA ; S.meliloti 1021;asecondstreakofSMc03964. C)(clockwisefromleft) greA ; S.meliloti 1021;SMb20360(twoseparate strains).Specificstrainnamesareshowninthephotolabels.Thegrowth mediumisLBMC,withstreptomycin500ug/mL. Additionalfile5: Free-livingexpressionof -glucuronidase(GUS) underthecontrolofthepromotersofthefollowingORFs: A) SMa0044.MultipleisolatesoftheSMa0044::GUSfusionsareshownin comparisonwith greA (positivecontrolforGUSexpression)and S.meliloti 1021wildtype(negativecontrolforGUSexpression).B)SMc00135. MultipleisolatesoftheSMc00135::GUSfusionsareshownincomparison with greA and S.meliloti 1021wildtype.C)theSMc01424-01422operon. MultipleisolatesoftheSMc01424-01422:GUSfusionsareshownin comparisonwith greA and S.meliloti 1021wildtype.Thegrowthmedium isLBMC,withstreptomycin500ug/mL.GUSexpressionstrainsthatwere testedfornoduleexpressionaredenotedwithanasteriskandare describedinTables3and4. Competinginterests Theauthorsdeclarethattheyhavenocompetinginterests. Authors ’ contributions KMJconceivedofthestudy,performedthegenomecomparisons,designed experiments,constructedbacterialmutantstrains,performedexperiments, interpretedresultsanddraftedthemanuscript.CQdesignedexperiments, constructedbacterialmutantstrains,performedexperiments,interpreted resultsandhelpeddraftthemanuscript.BKWconstructedbacterialmutant strains,performedexperiments,andhelpeddraftthemanuscript.OMD,JS, TEB,andMRLconstructedbacterialmutantstrainsandperformed experiments.Allauthorsreadandapprovedthefinalmanuscript. Acknowledgments TheauthorswishtothankSharonLong,MelanieBarnett,andJeanneHarris forplasmidpJH104;GrahamWalkerforplasmidpK19mobsac;andMichikoE. Taga,PennyJ.BeuningandGeorgeW.Batesforcriticalreadingofthe manuscript. Thisworkwasfundedbystart-upfundsprovidedtoKMJbyFloridaState University. Authordetails1DepartmentofBiologicalScience,FloridaStateUniversity,BiologyUnitI, 230A,89ChieftainWay,Tallahassee,FL32306-4370,USA.2Presentaddress: DepartmentofPsychology,BrooklynCollege,Brooklyn,NY,USA.3Present address:ArthrexOrthopaedicProducts,Jacksonville,FL,USA. Received:24January2012Accepted:4May2012 Published:15May2012 References1.JonesKM,KobayashiH,DaviesBW,TagaME,WalkerGC: Howrhizobial symbiontsinvadeplants:the Sinorhizobium-Medicago model. NatRev Microbiol 2007, 5 (8):619 … 633. 2.GibsonKE,KobayashiH,WalkerGC: Moleculardeterminantsofasymbiotic chronicinfection. AnnuRevGenet 2008, 42: 413 … 441. 3.HuangW: DataSets:U.S.FertilizerUseandPrice .In EditedbyService UER:usda.gov;2008. 4.PetersNK,FrostJW,LongSR: Aplantflavone,luteolin,inducesexpression of Rhizobiummeliloti nodulationgenes. Science 1986, 233: 977 … 980. 5.GageDJ: Infectionandinvasionofrootsbysymbiotic,nitrogen-fixing rhizobiaduringnodulationoftemperatelegumes. MicrobiolMolBiolRev 2004, 68 (2):280 … 300. 6.OldroydGE,DownieJA: Nuclearcalciumchangesatthecoreofsymbiosis signalling. CurrOpinPlantBiol 2006, 9 (4):351 … 357.Queiroux etal.BMCMicrobiology 2012, 12 :74 Page13of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 14

7.TimmersAC,AuriacMC,TruchetG: Refinedanalysisofearlysymbiotic stepsoftheRhizobium-Medicagointeractioninrelationshipwith microtubularcytoskeletonrearrangements. Development 1999, 126 (16):3617 … 3628. 8.CatalanoCM,CzymmekKJ,GannJG,SherrierDJ: Medicagotruncatula syntaxinSYP132definesthesymbiosomemembraneandinfection dropletmembraneinrootnodules. Planta 2007, 255 (3):541 … 550. 9.MergaertP,UchiumiT,AlunniB,EvannoG,CheronA,CatriceO,Mausset AE,Barloy-HublerF,GalibertF,KondorosiA, etal : Eukaryoticcontrolon bacterialcellcycleanddifferentiationintheRhizobium-legume symbiosis. ProcNatlAcadSciUSA 2006, 103 (13):5230 … 5235. 10.PrellJ,PooleP: Metabolicchangesofrhizobiainlegumenodules. Trends Microbiol 2006, 14 (4):161 … 168. 11.RuvkunGB,SundaresanV,AusubelFM: DirectedtransposonTn5 mutagenesisandcomplementationanalysisof Rhizobiummeliloti symbioticnitrogenfixationgenes. Cell 1982, 29 (2):551 … 559. 12.PooleP,AllawayD: Carbonandnitrogenmetabolismin Rhizobium AdvMicrobPhysiol 2000, 43: 117 … 163. 13.HirschAM,SmithCA: Effectsof Rhizobiummelilotinif and fix mutantson alfalfarootnoduledevelopment. JBacteriol 1987, 169: 1137 … 1146. 14.Masson-BoivinC,GiraudE,PerretX,BatutJ: Establishingnitrogen-fixing symbiosiswithlegumes:howmanyrhizobiumrecipes? TrendsMicrobiol 2009, 17 (10):458 … 466. 15.MeadeHM,LongSR,RuvkunGB,BrownSE,AusubelFM: Physicaland geneticcharacterizationofsymbioticandauxotrophicmutantsof Rhizobiummeliloti inducedbytransposonTn5mutagenesis. JBacteriol 1982, 149 (1):114 … 122. 16.EarlCD,RonsonCW,AusubelFM: Geneticandstructuralanalysisofthe RhizobiummelilotifixA,fixB,fixC ,and fixX genes.JBacteriol 1987, 169 (3):1127 … 1136. 17.PreisigO,AnthamattenD,HenneckeH: Genesforamicroaerobically inducedoxidasecomplexin Bradyrhizobiumjaponicum areessentialfor anitrogen-fixingendosymbiosis. ProcNatlAcadSciUSA 1993, 90 (8):3309 … 3313. 18.ArunothayananH,NomuraM,HamaguchiR,ItakuraM,MinamisawaK, TajimaS: Coppermetallochaperonesarerequiredfortheassemblyof bacteroidcytochromecoxidasewhichisfunctioningfornitrogen fixationinsoybeannodules. PlantCellPhysiol 2010, 51 (7):1242 … 1246. 19.RubioLM,LuddenPW: Biosynthesisoftheiron-molybdenumcofactorof nitrogenase. AnnuRevMicrobiol 2008, 62: 93 … 111. 20.BeckerA,BergesH,KrolE,BruandC,RubergS,CapelaD,LauberE,Meilhoc E,AmpeF,deBruijnFJ, etal : Globalchangesingeneexpressionin Sinorhizobiummeliloti 1021undermicrooxicandsymbioticconditions. MolPlantMicrobeInteract 2004, 17 (3):292 … 303. 21.DavidM,DaveranML,BatutJ,DedieuA,DomergueO,GhaiJ,HertigC, BoistardP,KahnD: Cascaderegulationof nif geneexpressionin Rhizobiummeliloti Cell 1988, 54 (5):671 … 683. 22.WongPP,BurrisRH: Natureofoxygeninhibitionofnitrogenasefrom Azotobactervinelandii ProcNatlAcadSciUSA 1972, 69 (3):672 … 675. 23.DreyfusBL,ElmerichC,DommerguesYR: Free-living Rhizobium strainable togrowonn(2)asthesolenitrogensource. ApplEnvironMicrobiol 1983, 45 (2):711 … 713. 24.VandeVeldeW,ZehirovG,SzatmariA,DebreczenyM,IshiharaH,KeveiZ, FarkasA,MikulassK,NagyA,TiriczH, etal : Plantpeptidesgovern terminaldifferentiationofbacteriainsymbiosis. Science 2010, 327 (5969):1122 … 1126. 25.DriscollBT,FinanTM: NAD+-dependentmalicenzymeof Rhizobium meliloti isrequiredforsymbioticnitrogenfixation. MolMicro 1993, 7 (6):865 … 873. 26.DriscollBT,FinanTM: PropertiesofNAD(+)-andNADP(+)-dependent malicenzymesofRhizobium(Sinorhizobium)melilotianddifferential expressionoftheirgenesinnitrogen-fixingbacteroids. Microbiology 1997, 143 (Pt2):489 … 498. 27.RogersA,AinsworthEA,LeakeyAD: Willelevatedcarbondioxide concentrationamplifythebenefitsofnitrogenfixationinlegumes? PlantPhysiol 2009, 151 (3):1009 … 1016. 28.RaskoDA,RosovitzMJ,MyersGS,MongodinEF,FrickeWF,GajerP,Crabtree J,SebaihiaM,ThomsonNR,ChaudhuriR, etal : Thepangenomestructure ofEscherichiacoli:comparativegenomicanalysisofE.colicommensal andpathogenicisolates. JBacteriol 2008, 190 (20):6881 … 6893. 29.deCrecy-LagardV,ElYacoubiB,delaGarzaRD,NoirielA,HansonAD: Comparativegenomicsofbacterialandplantfolatesynthesisand salvage:predictionsandvalidations. BMCGenomics 2007, 8: 245. 30.GoodnerB,HinkleG,GattungS,MillerN,BlanchardM,QurolloB,Goldman BS,CaoY,AskenaziM,HallingC, etal : Genomesequenceoftheplant pathogenandbiotechnologyagentAgrobacteriumtumefaciensC58. Science 2001, 294 (5550):2323 … 2328. 31.WoodDW,SetubalJC,KaulR,MonksDE,KitajimaJP,OkuraVK,ZhouY, ChenL,WoodGE,AlmeidaNFJr, etal : Thegenomeofthenatural geneticengineer Agrobacteriumtumefaciens C58. Science 2001, 294 (5550):2317 … 2323. 32.NiermanWC,FeldblyumTV,LaubMT,PaulsenIT,NelsonKE,EisenJA, HeidelbergJF,AlleyMR,OhtaN,MaddockJR, etal : Completegenome sequenceof Caulobactercrescentus ProcNatlAcadSciUSA 2001, 98 (7):4136 … 4141. 33.MannistoMK,TiirolaMA,Salkinoja-SalonenMS,KulomaaMS,PuhakkaJA: Diversityofchlorophenol-degradingbacteriaisolatedfrom contaminatedborealgroundwater. ArchMicrobiol 1999, 171 (3):189 … 197. 34. GenomeProject:Caulobactersp.K31 [http://img.jgi.doe.gov/cgi-bin/pub/ main.cgi?section=TaxonDetail&page=taxonDetail&taxon_oid=641522612]. 35.KanekoT,NakamuraY,SatoS,MinamisawaK,UchiumiT,SasamotoS, WatanabeA,IdesawaK,IriguchiM,KawashimaK,etal : Completegenomic sequenceofnitrogen-fixingsymbioticbacterium Bradyrhizobium japonicum USDA110. DNARes 2002, 9 (6):189 … 197. 36.KanekoT,NakamuraY,SatoS,AsamizuE,KatoT,SasamotoS,WatanabeA, IdesawaK,IshikawaA,KawashimaK, etal : Completegenomestructureof thenitrogen-fixingsymbioticbacterium Mesorhizobiumloti (supplement). DNARes 2000, 7 (6):381 … 406. 37.GonzalezV,SantamariaRI,BustosP,Hernandez-GonzalezI,Medrano-SotoA, Moreno-HagelsiebG,JangaSC,RamirezMA,Jimenez-JacintoV, Collado-VidesJ, etal : Thepartitioned Rhizobiumetli genome:geneticand metabolicredundancyinseveninteractingreplicons. ProcNatlAcadSciU SA 2006, 103 (10):3834 … 3839. 38.YoungJP,CrossmanLC,JohnstonAW,ThomsonNR,GhazouiZF,HullKH, WexlerM,CursonAR,ToddJD,PoolePS, etal : Thegenomeof Rhizobium leguminosarum hasrecognizablecoreandaccessorycomponents. GenomeBiol 2006, 7 (4):R34. 39.ReeveWG,ChainP,O ’ HaraG,ArdleyJ,NandesenaK,BrauL,TiwariRP, MalfattiS,KissH,LapidusA, etal : Completegenomesequenceofthe Medicago microsymbiont Ensifer ( Sinorhizobium ) medicae strainWSM419. StandardsinGenomicSciences 2010, 2: 77 … 86. 40.SambrookJ,FritschEF,ManiatisT: Molecularcloning:alaboratorymanual 2ndedition.ColdSpringHarbor,N.Y:ColdSpringHarborLaboratoryPress; 1982. 41.SchferA,TauchA,JgerW,KalinowskiJ,ThierbachG,PhlerA: Small mobilizablemulti-purposecloningvectorsderivedfromthe Escherichia coli plasmidspK18andpK19:selectionofdefineddeletionsinthe chromosomeof Corynebacteriumglutamicum Gene 1994, 145: 69 … 73. 42.FergusonGP,DattaA,CarlsonRW,WalkerGC: Importanceofunusually modifiedlipidAin Sinorhizobium stressresistanceandlegumesymbiosis. MolMicrobiol 2005, 56 (1):68 … 80. 43.GlazebrookJ,WalkerGC: GenetictechniquesinRhizobiummeliloti. MethodsEnzymol 1991, 204: 398 … 418. 44.FinanTM,HartweigE,LeMieuxK,BergmanK,WalkerGC,SignerER: General transductioninRhizobiummeliloti. JBacteriol 1984, 159 (1):120 … 124. 45.LeighJA,SignerER,WalkerGC: Exopolysaccharide-deficientmutantsof Rhizobiummeliloti thatformineffectivenodules. ProcNatlAcadSciUSA 1985, 82: 6231 … 6235. 46.VincentJM: AManualforthePracticalStudyoftheRoot-NoduleBacteria Oxford:Blackwell;1970. 47.JeffersonRA,KavanaghTA,BevanMW: GUSfusions:beta-glucuronidaseas asensitiveandversatilegenefusionmarkerinhigherplants. EMBOJ 1987, 6 (13):3901 … 3907. 48.MarkowitzVM,IvanovaNN,SzetoE,PalaniappanK,ChuK,DaleviD,Chen IM,GrechkinY,DubchakI,AndersonI, etal : IMG/M:adatamanagement andanalysissystemformetagenomes. NucleicAcidsRes 2008, 36 (Databaseissue):D534 … D538. 49.PrellJ,WhiteJP,BourdesA,BunnewellS,BongaertsRJ,PoolePS: Legumes regulateRhizobiumbacteroiddevelopmentandpersistencebythe supplyofbranched-chainaminoacids. ProcNatlAcadSciUSA 2009, 106 (30):12477 … 12482.Queiroux etal.BMCMicrobiology 2012, 12 :74 Page14of15 http://www.biomedcentral.com/1471-2180/12/74

PAGE 15

50.Tellez-SosaJ,SoberonN,Vega-SeguraA,Torres-MarquezME,CevallosMA: The RhizobiumetlicyaC product:characterizationofanoveladenylate cyclaseclass. JBacteriol 2002, 184 (13):3560 … 3568. 51.CampbellGR,SharypovaLA,ScheidleH,JonesKM,NiehausK,BeckerA, WalkerGC: Strikingcomplexityoflipopolysaccharidedefectsinacollection of Sinorhizobiummeliloti mutants. JBacteriol 2003, 185 (13):3853 … 3862. 52.Marchler-BauerA,LuS,AndersonJB,ChitsazF,DerbyshireMK,DeWeeseScottC,FongJH,GeerLY,GeerRC,GonzalesNR, etal : CDD:aConserved DomainDatabaseforthefunctionalannotationofproteins. NucleicAcids Res 2011, 39 (Databaseissue):D225 … D229. 53.GalibertF,FinanTM,LongSR,PhlerA,AbolaP,AmpeF,Barloy-HublerF, BarnettMJ,BeckerA,BoistardP, etal : Thecompositegenomeofthe legumesymbiont Sinorhizobiummeliloti Science 2001, 293 (5530):668 … 672. 54.BeckerA,BarnettMJ,CapelaD,DondrupM,KampPB,KrolE,LinkeB, RubergS,RunteK,SchroederBK, etal : Aportalforrhizobialgenomes: RhizoGATEintegratesa Sinorhizobiummeliloti genomeannotation updatewithpostgenomedata. JBiotechnol 2009, 140 (1 … 2):45 … 50. 55.Barloy-HublerF,CheronA,HellegouarchA,GalibertF: Smc01944,a secretedperoxidaseinducedbyoxidativestressesinSinorhizobium meliloti1021. Microbiology 2004, 150 (Pt3):657 … 664. 56.AltschulSF,MaddenTL,SchafferAA,ZhangJ,ZhangZ,MillerW,LipmanDJ: GappedBLASTandPSI-BLAST:anewgenerationofproteindatabase searchprograms. NucleicAcidsRes 1997, 25 (17):3389 … 3402. 57.WattSA,TellstromV,PatschkowskiT,NiehausK: Identificationofthe bacterialsuperoxidedismutase(SodM)asplant-inducibleelicitorofan oxidativeburstreactionintobaccocellsuspensioncultures. JBiotechnol 2006, 126 (1):78 … 86. 58.DaviesBW,WalkerGC: DisruptionofsitACompromisesSinorhizobium melilotiforManganeseUptakeRequiredforProtectionAgainst OxidativeStress. JBacteriol 2006,189 (5):2101 … 2109. 59.KobayashiH,DeNiscoNJ,ChienP,SimmonsLA,WalkerGC: Sinorhizobium melilotiCpdR1iscriticalforco-ordinatingcellcycleprogressionandthe symbioticchronicinfection. MolMicrobiol 2009, 73 (4):586 … 600. 60.PobigayloN,WetterD,SzymczakS,SchillerU,KurtzS,MeyerF,Nattkemper TW,BeckerA: Constructionofalargesignature-taggedmini-Tn5 transposonlibraryanditsapplicationtomutagenesisof Sinorhizobium meliloti ApplEnvironMicrobiol 2006, 72 (6):4329 … 4337. 61.ColombattiA,BonaldoP,DolianaR: TypeAmodules:interactingdomains foundinseveralnon-fibrillarcollagensandinotherextracellularmatrix proteins. Matrix 1993, 13 (4):297 … 306. 62.BarnettMJ,TomanCJ,FisherRF,LongSR: Adual-genomeSymbiosisChip forcoordinatestudyofsignalexchangeanddevelopmentina prokaryote-hostinteraction. ProcNatlAcadSciUSA 2004, 101 (47):16636 … 16641.Epub12004Nov16612. 63.KrolE,BeckerA: Globaltranscriptionalanalysisofthephosphate starvationresponseinSinorhizobiummelilotistrains1021and2011. Mol GenetGenomics 2004, 272 (1):1 … 17.Epub2004Jun2023. 64.MeloniS,ReyL,SidlerS,ImperialJ,Ruiz-ArguesoT,PalaciosJM: The twin-argininetranslocation(Tat)systemisessentialfor Rhizobium-legumesymbiosis. MolMicrobiol 2003, 48 (5):1195 … 1207. 65.KobayashiM,SuzukiT,FujitaT,MasudaM,ShimizuS: Occurrenceof enzymesinvolvedinbiosynthesisofindole-3-aceticacidfrom indole-3-acetonitrileinplant-associatedbacteria,Agrobacteriumand Rhizobium. ProcNatlAcadSciUSA 1995, 92 (3):714 … 718. 66.SpaepenS,VanderleydenJ,RemansR: Indole-3-aceticacidinmicrobialand microorganism-plantsignaling. FEMSMicrobiolRev 2007, 31 (4):425 … 448. 67.BuikemaWJ,LongSR,BrownSE,vandenBosRC,EarlC,AusubelFM: PhysicalandgeneticcharacterizationofRhizobiummelilotisymbiotic mutants. JMolApplGenet 1983, 2 (3):249 … 260. 68.EgelhoffTT,LongSR: Rhizobiummeliloti nodulationgenes:identification of nodDABC geneproducts,purificationofnodAprotein,andexpression of nodAin Rhizobiummeliloti JBacteriol 1985, 164 (2):591 … 599.doi:10.1186/1471-2180-12-74 Citethisarticleas: Queiroux etal. : Acomparativegenomicsscreen identifiesa Sinorhizobiummeliloti 1021 sodM -likegenestrongly expressedwithinhostplantnodules. BMCMicrobiology 2012 12 :74. 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 Queiroux etal.BMCMicrobiology 2012, 12 :74 Page15of15 http://www.biomedcentral.com/1471-2180/12/74


xml version 1.0 encoding utf-8 standalone no
mets ID sort-mets_mets OBJID sword-mets LABEL DSpace SWORD Item PROFILE METS SIP Profile xmlns http:www.loc.govMETS
xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
xsi:schemaLocation http:www.loc.govstandardsmetsmets.xsd
metsHdr CREATEDATE 2012-10-02T16:10:33
agent ROLE CUSTODIAN TYPE ORGANIZATION
name BioMed Central
dmdSec sword-mets-dmd-1 GROUPID sword-mets-dmd-1_group-1
mdWrap SWAP Metadata MDTYPE OTHER OTHERMDTYPE EPDCX MIMETYPE textxml
xmlData
epdcx:descriptionSet xmlns:epdcx http:purl.orgeprintepdcx2006-11-16 xmlns:MIOJAVI
http:purl.orgeprintepdcxxsd2006-11-16epdcx.xsd
epdcx:description epdcx:resourceId sword-mets-epdcx-1
epdcx:statement epdcx:propertyURI http:purl.orgdcelements1.1type epdcx:valueURI http:purl.orgeprintentityTypeScholarlyWork
http:purl.orgdcelements1.1title
epdcx:valueString A comparative genomics screen identifies a Sinorhizobium meliloti 1021 sodM-like gene strongly expressed within host plant nodules
http:purl.orgdctermsabstract
AbstractBackgroundWe have used the genomic data in the Integrated Microbial Genomes system of the Department of Energy’s Joint Genome Institute to make predictions about rhizobial open reading frames that play a role in nodulation of host plants. The genomic data was screened by searching for ORFs conserved in α-proteobacterial rhizobia, but not conserved in closely-related non-nitrogen-fixing α-proteobacteria.ResultsUsing this approach, we identified many genes known to be involved in nodulation or nitrogen fixation, as well as several new candidate genes. We knocked out selected new genes and assayed for the presence of nodulation phenotypes and/or nodule-specific expression. One of these genes, SMc00911, is strongly expressed by bacterial cells within host plant nodules, but is expressed minimally by free-living bacterial cells. A strain carrying an insertion mutation in SMc00911 is not defective in the symbiosis with host plants, but in contrast to expectations, this mutant strain is able to out-compete the S. meliloti 1021 wild type strain for nodule occupancy in co-inoculation experiments. The SMc00911 ORF is predicted to encode a “SodM-like” (superoxide dismutase-like) protein containing a rhodanese sulfurtransferase domain at the N-terminus and a chromate-resistance superfamily domain at the C-terminus. Several other ORFs (SMb20360, SMc01562, SMc01266, SMc03964, and the SMc01424-22 operon) identified in the screen are expressed at a moderate level by bacteria within nodules, but not by free-living bacteria.ConclusionsBased on the analysis of ORFs identified in this study, we conclude that this comparative genomics approach can identify rhizobial genes involved in the nitrogen-fixing symbiosis with host plants, although none of the newly identified genes were found to be essential for this process.
http:purl.orgdcelements1.1creator
Queiroux, Clothilde
Washburn, Brian K
Davis, Olivia M
Stewart, Jamie
Brewer, Tess E
Lyons, Michael R
Jones, Kathryn M
http:purl.orgeprinttermsisExpressedAs epdcx:valueRef sword-mets-expr-1
http:purl.orgeprintentityTypeExpression
http:purl.orgdcelements1.1language epdcx:vesURI http:purl.orgdctermsRFC3066
en
http:purl.orgeprinttermsType
http:purl.orgeprinttypeJournalArticle
http:purl.orgdctermsavailable
epdcx:sesURI http:purl.orgdctermsW3CDTF 2012-05-15
http:purl.orgdcelements1.1publisher
BioMed Central Ltd
http:purl.orgeprinttermsstatus http:purl.orgeprinttermsStatus
http:purl.orgeprintstatusPeerReviewed
http:purl.orgeprinttermscopyrightHolder
Clothilde Queiroux et al.; licensee BioMed Central Ltd.
http:purl.orgdctermslicense
http://creativecommons.org/licenses/by/2.0
http:purl.orgdctermsaccessRights http:purl.orgeprinttermsAccessRights
http:purl.orgeprintaccessRightsOpenAccess
http:purl.orgeprinttermsbibliographicCitation
BMC Microbiology. 2012 May 15;12(1):74
http:purl.orgdcelements1.1identifier
http:purl.orgdctermsURI http://dx.doi.org/10.1186/1471-2180-12-74
fileSec
fileGrp sword-mets-fgrp-1 USE CONTENT
file sword-mets-fgid-0 sword-mets-file-1
FLocat LOCTYPE URL xlink:href 1471-2180-12-74.xml
sword-mets-fgid-1 sword-mets-file-2 applicationpdf
1471-2180-12-74.pdf
sword-mets-fgid-3 sword-mets-file-3 applicationvnd.ms-excel
1471-2180-12-74-S3.XLS
sword-mets-fgid-4 sword-mets-file-4
1471-2180-12-74-S2.XLS
sword-mets-fgid-5 sword-mets-file-5 imagejpeg
1471-2180-12-74-S4.JPEG
sword-mets-fgid-6 sword-mets-file-6
1471-2180-12-74-S5.JPEG
sword-mets-fgid-7 sword-mets-file-7
1471-2180-12-74-S1.XLS
structMap sword-mets-struct-1 structure LOGICAL
div sword-mets-div-1 DMDID Object
sword-mets-div-2 File
fptr FILEID
sword-mets-div-3
sword-mets-div-4
sword-mets-div-5
sword-mets-div-6
sword-mets-div-7
sword-mets-div-8


!DOCTYPE art SYSTEM 'http:www.biomedcentral.comxmlarticle.dtd'
ui 1471-2180-12-74ji 1471-2180fm dochead Research articlebibl title p A comparative genomics screen identifies a it Sinorhizobium meliloti 1021 sodM-like gene strongly expressed within host plant nodulesaug au id A1 snm Queirouxfnm Clothildeinsr iid I1 email cqueiroux@bio.fsu.eduA2 Washburnmi KBrianwashburn@bio.fsu.eduA3 ce yes DavisMOliviaI2 odavis@gc.cuny.eduA4 StewartJamiejls08d@fsu.eduA5 BrewerETesstb08c@my.fsu.eduA6 LyonsRMichaelI3 mrl06d@fsu.eduA7 ca JonesMKathrynkmjones@bio.fsu.eduinsg ins Department of Biological Science, Florida State University, Biology Unit I, 230A, 89 Chieftain Way, Tallahassee, FL, 32306-4370, USAPresent address: Department of Psychology, Brooklyn College, Brooklyn, NY, USAPresent address: Arthrex Orthopaedic Products, Jacksonville, FL, USAsource BMC Microbiologyissn 1471-2180pubdate 2012volume 12issue 1fpage 74url http://www.biomedcentral.com/1471-2180/12/74xrefbib pubidlist pubid idtype doi 10.1186/1471-2180-12-74pmpid 22587634history rec date day 24month 1year 2012acc 452012pub 1552012cpyrt 2012collab Queiroux et al.; licensee BioMed Central Ltd.note This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.kwdg kwd RhizobiaSinorhizobium melilotiAlfalfaSymbiosisNitrogen fixationBacteriaLegumeGenomicsα-proteobacteriaabs sec st AbstractBackgroundWe have used the genomic data in the Integrated Microbial Genomes system of the Department of Energy’s Joint Genome Institute to make predictions about rhizobial open reading frames that play a role in nodulation of host plants. The genomic data was screened by searching for ORFs conserved in α-proteobacterial rhizobia, but not conserved in closely-related non-nitrogen-fixing α-proteobacteria.ResultsUsing this approach, we identified many genes known to be involved in nodulation or nitrogen fixation, as well as several new candidate genes. We knocked out selected new genes and assayed for the presence of nodulation phenotypes and/or nodule-specific expression. One of these genes, SMc00911, is strongly expressed by bacterial cells within host plant nodules, but is expressed minimally by free-living bacterial cells. A strain carrying an insertion mutation in SMc00911 is not defective in the symbiosis with host plants, but in contrast to expectations, this mutant strain is able to out-compete the S. meliloti 1021 wild type strain for nodule occupancy in co-inoculation experiments. The SMc00911 ORF is predicted to encode a “SodM-like” (superoxide dismutase-like) protein containing a rhodanese sulfurtransferase domain at the N-terminus and a chromate-resistance superfamily domain at the C-terminus. Several other ORFs (SMb20360, SMc01562, SMc01266, SMc03964, and the SMc01424-22 operon) identified in the screen are expressed at a moderate level by bacteria within nodules, but not by free-living bacteria.ConclusionsBased on the analysis of ORFs identified in this study, we conclude that this comparative genomics approach can identify rhizobial genes involved in the nitrogen-fixing symbiosis with host plants, although none of the newly identified genes were found to be essential for this process.bdy BackgroundSinorhizobium meliloti 1021 is a soil bacterium that establishes a nitrogen-fixing symbiosis with the host plants Medicago sativa (alfalfa) and Medicago truncatula (reviewed in abbrgrp abbr bid B1 1B2 2). These plants are not only agriculturally important, but are also key model organisms for studying the symbiotic interaction between rhizobial bacteria and their plant hosts. The goals of this study are to increase our understanding of this process and provide practical insights that may lead to the production of more efficient symbiotic strains of rhizobia. Increasing the efficiency of symbiotic nitrogen fixation is important in that it reduces the need for industrial production of nitrogen fertilizers, which is extremely costly in terms of petroleum and natural gas. In 2007, the US applied 13 million tons of industrially-produced nitrogen fertilizer to crops B3 3. Fertilizers continue to be used to increase yields of legume crops 3, demonstrating that there is considerable room for improvement in these symbiotic associations.S. meliloti fixes nitrogen in root nodules formed by the host plant, converting dinitrogen gas to ammonia. The development of these nodules requires that several signals be exchanged between the plant and the rhizobial bacteria. Flavonoid compounds produced by host plants signal S. meliloti to produce lipochitooligosaccharides called Nod factors (NFs) B4 4. NF activates multiple responses in host plants, including tight curling of root hairs that traps bacterial cells within the curl, and cell divisions in the root cortex, which establish the nodule primordium B5 5B6 6. The bacteria invade and colonize the roots through structures called infection threads, which originate from microcolonies of bacteria trapped in the curled root hair cells 1B7 7. New infection threads initiate at each cell layer, eventually delivering the bacteria to the inner plant cortex 7. There, the rhizobial bacteria are endocytosed by root cortical cells within individual compartments of host-cell membrane origin 2B8 8. Within these compartments, signals provided by the plant and the low-oxygen environment induce the bacteria to differentiate into a form called a “bacteroid”, and to begin expressing nitrogenase, the nitrogen-fixing enzyme, and other factors that are required for the symbiosis B9 9B10 10.Rhizobial fixation of dinitrogen requires not only the expression of nitrogenase (encoded by the genes nifK and nifDB11 11), but also the assembly of cofactors and large inputs of energy and reductant B12 12. Nitrogen fixation also requires a nitrogenase reductase, encoded by nifH11; iron-molybdenum cofactor biosynthesis proteins, encoded by nifBnifE and nifE; and electron transfer flavoproteins and ferredoxins (fixA, fixB, fixC, fixX) B13 13B14 14B15 15B16 16. Bacteroids also increase their respiration rate, increasing the expression of the fixNOQP cytochrome c oxidase operons B17 17B18 18B19 19B20 20.Many of the proteins required for nitrogen fixation are tightly regulated by oxygen-sensing systems and are produced by rhizobial bacteria only when they encounter a low-oxygen environment B21 21. Nitrogenase and some of the other factors involved in nitrogen fixation are extremely oxygen-sensitive B22 22, thus their expression under inappropriate conditions would be ineffective. Even under microaerobic conditions, most rhizobial bacteria are not capable of nitrogen fixation in the free-living state B23 23. The reasons for this are not completely understood, though it is known that legumes of the inverted repeat-lacking clade (IRLC), such as alfalfa and M. truncatula, which form indeterminate-type nodules, impose a specific differentiation program on the intracellular bacteria, most likely through the activity of plant-produced bioactive peptides 9B24 24. Bacteroids also receive nutrients from the host plant, such as the carbon source malate B25 25B26 26B27 27. Multiple bacterial cellular processes and differentiation programs contribute to the success of the symbiosis with host plants, and one of our goals is to use comparative genomics to predict previously uncharacterized S. meliloti open reading frames (ORFs) that may be involved in these processes, to test these predictions, and understand the mechanisms involved. In other bacterial species, comparative genomics of bacterial strains has been useful in finding new genes that are involved in metabolic pathways and in identifying virulence factors that distinguish pathogenic strains from commensal strains (examples include: B28 28B29 29). In this study, a comparison of ORFS from nitrogen-fixing, plant-host nodulating rhizobia with closely-related non-nitrogen-fixing bacteria has identified ORFs that are expressed by Sinorhizobium meliloti within host plant nodules.MethodsGenome comparisonsSearches were conducted at the Department of Energy Joint Genome Institute’s Integrated Microbial Genomes website, http://img.jgi.doe.gov/cgi-bin/pub/main.cgi. All of the genomes to be compared were selected from the genome display under the “Find Genomes” tab (see Table tblr tid T1 1 for compared genomes). The selected genomes were saved. The “Phylogenetic profiler” for single genes was used to find genes in Sinorhizobium/Ensifer meliloti with homologs in the genomes to be intersected and without homologs in the genomes to be subtracted (see Table 1). The searches were conducted at 20–80% identity and the complete data output is listed in Additional file supplr sid S1 1: Table S1.suppl Additional file 1 text b Table S1. Joint Genome Institute, Integrated Microbial Genomes Phylogenetic Profile search data on single genes.file name 1471-2180-12-74-S1.xls
Click here for file
table Table 1caption Genome ORFs compared withS. meliloti1021tgroup align left cols 3 colspec colname c1 colnum 1 colwidth 1* center c2 2 c3 thead valign top row rowsep entry GenomeSubtracted or intersectedLifestyletbody Agrobacterium tumefaciens C58 (Cereon) B30 30B31 31subtractedplant pathogenAgrobacterium tumefaciens C58 (Dupont) 3031subtractedplant pathogenBartonella bacilliformis KC583subtractedmammalian pathogenBartonella henselae Houston-1subtractedmammalian pathogenBartonella quintana Toulousesubtractedmammalian pathogenBartonella tribocorum CIP 105476subtractedmammalian pathogenBrucella abortus bv 1 9-941subtractedmammalian pathogenBrucella canis ATCC 23365subtractedmammalian pathogenBrucella melitensis 16 Msubtractedmammalian pathogenBrucella melitensis bv Abortus 2308subtractedmammalian pathogenBrucella ovis ATCC 25840subtractedmammalian pathogenBrucella suis ATCC 23445subtractedmammalian pathogenBrucella suis 1330subtractedmammalian pathogenCaulobacter crescentus CB15 B32 32subtractedfree-livingCaulobacter sp. K31 B33 33B34 34subtractedfree-livingBradyrhizobium japonicum USDA 110 B35 35intersectednitrogen-fixing plant symbiontMesorhizobium loti MAFF303099 B36 36intersectednitrogen-fixing plant symbiontRhizobium etli CFN 42 B37 37intersectednitrogen-fixing plant symbiontRhizobium leguminosarum bv. viciae 3841 B38 38intersectednitrogen-fixing plant symbiontSinorhizobium medicae WSM419 B39 39intersectednitrogen-fixing plant symbiontBacterial strains and growth conditionsS. meliloti 1021 strains were grown at 30°C in either LBMC (Luria Bertani [Miller] medium supplemented with 2.5 mM MgSOsub 4 and 2.5 mM CaCl2), or 1/10 LB-7% sucrose medium, with 1 mM MgSO4 and 0.25 mM CaCl2, or M9 salts-10% sucrose medium, supplemented with 1 μg/mL biotin B40 40. Bacterial plates contained 1.5% BactoAgar. Selections against strains carrying the sacB gene in the plasmid pK19mobsac were performed in M9 supplemented with 10% w/v sucrose or 1/10 LB-7% sucrose B41 41. Appropriate antibiotics were used at the following concentrations for S. meliloti strains: streptomycin 500 or 1000 μg/mL; neomycin 200 μg/mL. E. coli strains were grown at 37°C in LB medium 40, with appropriate antibiotics used at the following concentrations: kanamycin 50 μg/mL; chloramphenicol 10 μg/mL.Construction of S. meliloti mutant strainsMutant strains of S. meliloti 1021 with disruptions in ORFs described in Table T2 2 were constructed by amplifying internal ORF fragments using Phusion polymerase (New England Biolabs, Ipswich, MA, USA) and cloning into the plasmid pJH104, which carries a neomycin/kanamycin resistance marker (Jeanne Harris, Univ. Vermont, personal communication) B42 42. Insertion of the pJH104 plasmid also creates transcriptional fusions to the uidA β-glucuronidase (GUS) gene. Non-disrupting GUS insertions of some ORFs (described in Table 2) were constructed by amplifying the entire ORF or operon and cloning the product into pJH104, and conjugating into S. meliloti. Deletion mutant strains were constructed by amplifying fragments flanking the ORF to be deleted and cloning the fragments into the sacB gene-containing suicide vector pK19mobsac 41. (Some fragments were initially cloned into pCR-Blunt II-TOPO using the Zero-TOPO-Blunt cloning kit [Invitrogen, San Diego, CA, USA].) Mutant strains are listed in Table 2. Primers (Eurofins MWG Operon, Huntsville, AL, USA) and restriction enzymes (New England Biolabs, Ipswich, MA, USA) used for amplification and cloning of disruption, non-disrupting insertion, or deletion fragments are listed in Additional file S2 2: Table S2. Plasmids were mobilized into S. meliloti by triparental conjugation as described previously B43 43. S. meliloti exconjugants were selected on LBMC medium containing 200 μg/mL neomycin and 1000 μg/mL streptomycin. Unmarked deletion strains were selected for loss of the sacB gene carried by the pK19mobsac vector by plating neomycin-resistant exconjugants to either M9 salts--10% sucrose medium or 1/10 LB-7% sucrose medium. Strains constructed by phage ϕM12 transduction of plasmid insertions into S. meliloti 1021 are denoted in the Tables as “Xsd”. Transductions using phage ϕM12 were performed according to published protocols B44 44. For each mutant produced, at least two strains were isolated. For some of the mutants, including those which carry an unmarked ORF deletion, multiple independent isolates were obtained by selecting exconjugants from multiple independent conjugations. For most of the mutants carrying an insertion of the pJH104 plasmid, the independent isolates were the original isolate and strains constructed by transduction of the neomycin-resistance marker into wild type S. meliloti 1021 via phage ϕM12 44. Additional file 2 Table S2. Primers used to amplify S. meliloti 1021 fragments for construction of insertion mutants and deletion mutants.1471-2180-12-74-S2.xls
Click here for file
Table 2S. meliloti1021-derived mutant strains5 c4 4 c5 ORFPredicted functionLength (amino acids)Type of mutationStrain nameSMc01562hypothetical protein96deletionΔSMc01562.6ΔSMc01562.25ΔSMc01562.100SMc01562hypothetical protein96non-disrupting insertion of pJH104 GUS markerA104U.originalA104U.Xsd1A104U.Xsd6A104U.Xsd25A104U.Xs100SMc01986hypothetical protein119deletionΔSMc01986.1ΔSMc01986.6ΔSMc01986.25ΔSMc01986.100SMc01986hypothetical protein119non-disrupting insertion of pJH104 GUS markerC104.1A.Xsd1C104.1A.originalC104.2B.Xsd100SMc00135hypothetical protein243deletionΔSMc00135.B1ΔSMc00135.B17SMc00135hypothetical protein243non-disrupting insertion of pJH104 GUS markerB104.3AB104.4BB104.2 CSMc01422hypothetical protein (probable operon with SMc01423,SMc01424)128deletion (SMc01422, SMc01423, SMc01424 all deleted in this strain)ΔSMc01422-24.D21 ΔSMc01422-24.D29SMc01423probable nitrile hydratase subunit β219deletionsame as aboveSMc01424probable nitrile hydratase subunit α213deletionsame as aboveSMc01424-01422hypothetical protein (probable operon with SMc01423,SMc01422)213non-disrupting insertion of pJH104 GUS markerD104.2AD104.3BD104.1 CSMa0044hypothetical protein89deletionΔSMa0044.c1ΔSMa0044.c6ΔSMa0044.c10SMa0044non-disrupting insertion of pJH104 GUS marker89SMa0044.104.1ASMa0044.104.1BSMa0044.104.4 CSMb20431hypoth. arylmalonate decarboxylase261ORF-disrupting insertion of pJH104 GUS markerSMb20431.originalSMb20431.Xsd1SMb20360hypothetical protein243ORF-disrupting insertion of pJH104 GUS markerSMb20360.originalSMb20360.Xsd1SMc03964hypothetical protein300ORF-disrupting insertion of pJH104 GUS markerSMc03964.originalSMc03964.Xsd6SMc00911hypothetical protein275ORF-disrupting insertion of pJH104 GUS markerSMc00911.originalSMc00911.Xsd1SMc00911.original2SMa1334hypothetical protein398ORF-disrupting insertion of pJH104 GUS marker (may have a polar effect on 3′ genes Sma1332,-1331,-1329)SMa1334.originalSMa1334.Xsd1SMc01266hypothetical protein438ORF-disrupting insertion of pJH104 GUS marker (may have a polar effect on 3′ gene Smc01265)SMc01266.originalSMc01266.Xsd1greAtranscription elongation factor158ORF-disrupting insertion of pJH104 GUS markergreA.12.4.1aexpA1 (wgaA)EPSII biosynthesis enzyme490ORF-disrupting insertion of Tn5-Nm in expA—symbiotically proficient, competitor assay strainexpA125::Tn5.Xsd1Plant nodulation assaysThe host plant Medicago sativa (alfalfa) cv. Iroquois was prepared for inoculation with S. meliloti as in Leigh et al. (1985) with modifications: seeds were sterilized for 5 minutes in 50% bleach, rinsed in sterile water, and germinated for 3 days on 1% w/v plant cell culture-tested agar/water (Sigma, St. Louis, MO, USA) B45 45. Seedlings were then moved to individual 100 mm x 15 mm Jensen’s medium plates B46 46, and inoculated with 100 μL of OD600 = 0.05 S. meliloti of the appropriate strain. Plants were grown in a Percival AR-36 L incubator (Perry, IA, USA) at 21°C, with 60–70% relative humidity, and 100–175 μmol msup −2 s−1 light. Plants were measured at 5 weeks and 6.5 weeks of growth. t-tests (unpaired, two-tailed) were performed in Microsoft Excel and in GraphPad (http://www.graphpad.com/quickcalcs/ttest1.cfm?Format=C).Nodulation competition assays were performed in the same way as the plant assays described above, except that strains to be tested in competition against one another were prepared as a mixed 1:1 inoculum immediately before inoculation. Bacteria were harvested from nodules after 5 or 6.5 weeks of growth by excising the nodules from roots, surface sterilizing in 20% bleach for 5 min., washing in sterile, distilled water, and crushing the nodules in 1.5 mL tubes with a micro-pestle (Kimble-Chase, Vineland, NJ), in LB + 0.3 M glucose 45. Dilutions of the material from crushed nodules were plated on LBMC + 500 μg/mL streptomycin. Colonies were patched from these plates to LBMC + 500 μg/mL streptomycin and 200 μg/mL neomycin to determine the fraction of bacteria that carry the neomycin-resistance marker in the insertion plasmid pJH104.Detection of β-glucuronidase activity and imaging of root nodulesβ-glucuronidase expression by bacteria within nodules was detected by excising nodules, surface sterilizing with 20% bleach for 5 min., rinsing in sterile water, and staining in X-gluc buffer (1 mM 5-bromo-4-chloro-3-indolyl-beta-D-glucuronic acid, cyclohexylammonium salt; 0.02% SDS; 50 mM Na-phosphate, pH 7) B47 47 for the amount of time indicated in Table T3 3. Whole nodules were imaged on an AZ100 Multi-Zoom Microscope equipped with a DS-Fi1, 5 Megapixel color camera (Nikon Instruments U.S., Melville, NY). β-glucuronidase expression by bacteria on LBMC plates was detected by streaking bacteria to plates that had been spread with 40 μL of X-gluc solution (100 mM 5-bromo-4-chloro-3-indolyl-beta-D-glucuronic acid, cyclohexylammonium salt solution in dimethylformamide). Table 3Expression of β-glucuronidase (GUS) fusions7 c6 6 c7 ORFstrain% of nodules with GUS expressionStrength of nodule GUS expressionStaining timePattern of nodule GUS expressionFree-living GUS expressionN/AS. meliloti 1021 wild type (negative control)0/39 = 0%−variablenone−SMc00911SMc00911.original18/20 = 90%++++1.5–3.75 hrwhole nodule+SMc00911.Xsd118/18 = 100%++++1.5–3.75 hrwhole nodulen.d.SMc00911.original2n.d.n.d.N/AN/A+SMb20360SMb20360.original8/13 = 62%++3–5 hrinvasion zone-fixation zone−SMb20360.Xsd113/16 = 81%++3–5 hrinvasion zone-fixation zone−SMc00135B104.3A6/8 = 75%+2–3 hrinvasion zone-interzone+B104.4B8/8 = 100%+2–3 hrinvasion zone-interzone++B104.2 C6/8 = 75%++2–3 hrinvasion zone-interzone++SMc01562A104U.original7/8 = 88%+4–6 hrinterzone−A104U.Xsd13/7 = 43%+/−4–6 hrinterzone-fixation zonen.d.A104U.Xsd68/8 = 100%+4–6 hrinterzone-fixation zonen.d.A104U.Xsd253/8 = 38%+/−4–6 hrinterzone-fixation zonen.d.A104U.Xs1004/9 = 44%+4–6 hrfixation zonen.d.SMc01266SMc01266.original13/18 = 72%+3 hrinvasion zone-fixation zone+/−SMc01266.Xsd113/18 = 72%++3 hrinvasion zone−SMc03964SMc03964.original8/15 = 53%++3–5 hrinterzone+/−SMc03964.Xsd69/19 = 47%++3–5 hrinterzone-fixation zone−SMc01424-22D104.2A0/8 = 0%−4–6 hrN/A+/−D104.3B7/8 = 88%++4–6 hrinvasion zone-interzone+/−D104.1 C6/8 = 75%+4–6 hrinvasion zone-fixation zone+/−SMa0044SMa0044.104.1A4/8 = 50%+/−6–7 hrinvasion zone-interzone+++SMa0044.104.1B4/8 = 50%+/−6–7 hrinterzone+++SMa0044.104.4 C4/8% 50%+/−6–7 hrinterzone+++SMb20431SMb20431.original10/16 = 63%+5–12 hrinvasion zone-fixation zone−SMb20431.Xsd111/15 = 73%+5–12 hrinterzone−SMc01986C104.1A.Xsd10/6 = 0%−24 hrN/An.d.C104.1A.originaln.d.n.d.24 hrn.d.+/−C104.2B.Xsd1002/18 = 11%+/−24 hrfixation zonen.d.SMa1334SMa1334.original0/11 = 0%−5–24 hrN/A−SMa1334.Xsd10/13 = 0%−5–24 hrN/A−ResultsComparisons of Sinorhizobium meliloti open reading frames with those of other rhizobia and with non-nitrogen fixing α-proteobacteriaRhizobial functions required for symbiotic nitrogen fixation with legume plants have typically been discovered through the classical bacterial genetic technique of transposon mutagenesis, followed by screening mutants for loss of symbiotic function. We have used an alternative comparative genomics strategy to search for rhizobial genes involved in symbiosis. In this approach, searches of the Joint Genome Institute, Integrated Microbial Genomes (JGI IMG) system B48 48 were performed to find ORFs that S. meliloti 1021 shares with the symbiotic nitrogen-fixing α-proteobacteria (α-rhizobia) S. medicae WSM419, Rhizobium etli CFN 42, Rhizobium leguminosarum bv. viciae, Mesorhizobium loti MAFF303099, and Bradyrhizobium japonicum USDA110. A novel aspect of this strategy is that these searches were restricted by prior elimination of all S. meliloti ORFs that are present in any of 15 non-nitrogen-fixing, non-symbiotic α-proteobacteria (species listed in Table 1). (See Materials and Methods for search procedure.) The genomes used in the analysis were chosen based on the rhizobial genomes available in the JGI IMG database when the analysis was initially performed. The searches were conducted at multiple identity levels (20%–80%), and the output data from all the searches is presented in Additional file 1: Table S1. The genome subtractions eliminated genes common to α-proteobacteria with non-symbiotic lifestyles. For example, a search conducted at 50% identity, intersecting the S. meliloti ORFs with homologs in the 5 α-rhizobia species yields 1281 genes. However, when the search for homologs is conducted with subtraction of the ORFs from the 15 non-rhizobial species, the search yield is 58 genes ( Additional file S3 3: Table S3).Additional file 3 Table S3. Gene list of 139 ORFs compiled from search data in Additional file 1: Table S1.1471-2180-12-74-S3.xls
Click here for file
The result of the searches was a list of 139 ORFs common to the α-rhizobia (listed in Additional file 3: Table S3), but not found in the non-nitrogen-fixing, non-symbiotic α-proteobacteria. Among these 139 ORFs were 11 genes known to be involved in nitrogen fixation (Table T4 4 and Additional file 3: Table S3), including: nifHnifDnifKnifBnifEnifNfixAfixB, and fixC (see Introduction) and 8 known to be involved in Nod factor production, including nodAnodBnodCnodJ and nodI5, thus 13.7% (19/139) of the ORFs selected by this comparative gemonics approach are already known to be important for symbiotic function. Table 4Function distribution of the 139 ORFs from genome searches (SeeAdditional file3: Table S3for complete gene list)FunctionNumber of ORFsNitrogen fixation11Nod factor production/modification8Transposase10Predicted transcriptional regulator8Predicted transport protein14Predicted adenylate/guanylate cyclase7Other predicted function37Hypothetical protein44There were also 44 hypothetical proteins/proteins of unknown function among the 139 ORFs detected in the comparative genomic screen. The predicted functions of the remaining ORFs included transposases, transcriptional regulators, transport proteins, and adenylate/guanylate cyclases (Table 4). These are classes of genes that may participate in many of the functions that distinguish α-rhizobia from their non-symbiotic α-proteobacterial relatives, such as signaling to the host plant, reprogramming their metabolism for nitrogen fixation, and importing specific nutrients and differentiation signals from the plant 910B49 49. Also, atypical adenylate cyclases have been noted before in the rhizobia B50 50.Construction and symbiosis assays of mutants in conserved genesThirteen of the 139 conserved ORFs were chosen for further study because they are of undetermined function in S. meliloti and have no close homologs in the S. meliloti genome that might be expected to provide redundant function. Six of the longer ORFs, including SMc00911, were disrupted by cloning a small internal ORF fragment into the plasmid pJH104, conjugating the plasmid into S. meliloti 1021, and selecting for single-crossover insertion/disruption mutants. ( Additional file 2: Table S2 lists primer sequences and disruption fragment sizes and positions.) For the 6 remaining ORFs, 3 that are under 750 bp long (SMc01562, SMc01986 and SMc00135) and 3 that are all in a single operon (SMc01424, SMc01423, and SMc01422), deletion was judged to be a better strategy than disruption. SMc01424, SMc01423, and SMc01422 were all deleted as a single segment from the start codon of SMc01424 to the stop codon of SMc01422. The endpoints of the individual deletions of SMc01562, SMc01986, and SMc00135 were dictated by the position of the most suitable PCR primers. ( Additional file 2: Table S2 lists primer sequences and deletion sizes and positions.) Either the disruption or the deletion strategy is expected to result in a strain that does not produce a full-length version of the protein encoded by that ORF. These ORFs and the insertion and/or deletion mutant strains of each are listed and described in Table 2. The resulting mutant strains were then tested for symbiotic proficiency on the host plant alfalfa.For the initial phenotypic analysis, the ability of the mutants to successfully provide the plants with fixed nitrogen was determined. Alfalfa plants were inoculated with the bacterial mutants and after 5 weeks of growth, the shoot length attained on nitrogen-free medium was compared with plants inoculated with the S. meliloti 1021 wild type as the positive control and uninoculated plants as the negative control. Figure figr fid F1 1 shows the shoot length of alfalfa plants inoculated with wild type S. meliloti 1021 or with disruption mutant strains of the ORFs SMb20360, SMb20431, SMc00911, SMa1344, SMc01266, and SMc03964. Alfalfa plants inoculated with these strains attain a similar average shoot length as that of the wild type, demonstrating that all of these strains are able to form a successful symbiosis with this host plant. Figure F2 2 presents the same type of assay as Figure 1 for deletion mutants in the ORFs SMc01562, SMc01986, SMc01424-22, SMc00135, and SMa0044. Additional data on the plant assays in Figures 1 and 2 is presented in Table T5 5. The number of plants inoculated with each strain, the average number of mature, pink nodules per plant and the average number of white pseudonodules per plant are shown. All of these mutant strains are able to mount a successful symbiosis with the host plant alfalfa.fig Figure 1 Plant shoot length in cm, 5 weeks after inoculation with insertion mutant strains (mutant strain information is summarized in Table3
Plant shoot length in cm, 5 weeks after inoculation with insertion mutant strains (mutant strain information is summarized in Table 3). For each of the 6 ORF disruptions, the plant phenotype of the original isolate and that of a phage ϕM12 transductant of that strain are shown. Mean values are given above graph bars. Error bars represent standard error of the mean. Asterisks indicate samples with mean heights significantly different from the wild type. The number of plants tested and the number of nodules/plant for these assays are presented in Table 4.
graphic 1471-2180-12-74-1 Figure 2 Plant shoot length in cm, 5 weeks after inoculation with deletion mutant strains (summarized in Table3)
Plant shoot length in cm, 5 weeks after inoculation with deletion mutant strains (summarized in Table 3). For each of the ORF deletions, the plant phenotype of at least two isolates/and or transductants of each strain are shown. Mean values are given above graph bars. Error bars represent standard error of the mean. Asterisks indicate samples with mean heights significantly different from the wild type. The number of plants tested and the number of nodules/plant for these assays are presented in Table 4.
1471-2180-12-74-2 Table 5Mean nodule numberORFStrain nameNumber of alfalfa plants testedMean number pink nodules/ plant ± std. errorMean number white pseudonodules/plant ± std. errorN/AS. meliloti 1021 wild type, data set 1 (see Figure 1)911.9 ± 1.03.2 + 1.2SMb20360SMb20360.original817.4 ± 2.54.5 ± 1.2SMb20360.Xsd11014.7 ± 1.74.4 ± 1.4SMb20431SMb20431.original1112.8 ± 1.63.0 ± 0.6SMb20431.Xsd11113.3 ± 1.93.8 ± 0.8SMc00911SMc00911.original1114.3 ± 2.53.3 ± 0.8SMc00911.Xsd11115.3 ± 1.83.2 ± 1.1SMa1334SMa1334.original1015.7 ± 2.15.7 ± 0.9SMa1334.Xsd11116.4 ± 1.13.6 ± 1.7SMc01266SMc01266.original1114.4 ± 2.44.2 ± 0.5SMc01266.Xsd11117.8 ± 1.64.6 ± 1.2SMc03964SMc03964.original1116.3 ± 1.64.2 ± 0.5SMc03964.Xsd61015.2 ± 2.34.0 ± 0.9N/Auninoculated, data set 1 (see Figure 1)500N/AS. meliloti 1021 wild type, data set 2 (see Figure 2)17912.5 ± 0.53.2 ± 0.3SMc01562ΔSMc01562.62414.1 ± 1.32.2 ± 0.4ΔSMc01562.252511.6 ± 1.22.5 ± 0.5ΔSMc01562.1002411.8 ± 0.92.0 ± 0.6morerows SMc01986ΔSMc01986.12618.0 ± 1.84.5 ± 0.8ΔSMc01986.62615.3 ± 2.14.4 ± 0.8ΔSMc01986.252517.2 ± 2.36.8 ± 1.1ΔSMc01986.1002516.8 ± 1.86.7 ± 1.0SMc01424-22ΔSMc01422-24.D2111013.1 ± 0.73.7 ± 0.4ΔSMc01422-24.D2910911.1 ± 0.63.6 ± 0.3SMc00135ΔSMc00135.B18114.0 ± 0.72.8 ± 0.3ΔSMc00135.B177613.5 ± 0.93.3 ± 0.4SMa0044ΔSMa0044.c12411.8 ± 1.34.2 ± 0.6ΔSMa0044.c62512.6 ± 1.23.0 ± 0.8ΔSMa0044.c102413.5 ± 1.22.0 ± 0.5N/Auninoculated, data set 2 (see Figure 2)8200.1 ± 0.1SMc00911 is the most strongly expressed in the nodule of the conserved ORFSTo determine if the 13 ORFs analyzed in this study might play a role in symbiosis, despite the fact that they are not strictly required for symbiosis, the expression pattern of each of these ORFs was determined both for bacteria within the nodule and in the free-living state. The SMc00911 ORF is very strongly expressed by bacteria within the nodule (Figure F3 3B–F), but it expressed at a very low level by free-living bacteria on LBMC plates (Figure 3G). The nodules shown in Figure 3 are expressing β-glucuronidase (GUS) from a pJH104 plasmid insertion in Smc00911. The nodules shown were stained for 3.75 hr. There is strong staining throughout the nodule, with slightly weaker staining at the invasion zone near the distal end of the nodule. The nodule expression of the SMc00911::GUS fusion is much stronger than the expression of any of the other fusions tested (see Figure F4 4 and Table 3). In contrast, SMc00911 is expressed at a very low level by free-living S. meliloti carrying the SMc00911::GUS fusion grown on LBMC plates (Figure 3G and Table 3). For comparison, Figure 3G also shows that a greA::GUS fusion strain of S. meliloti constructed with the same reporter insertion plasmid, pJH104, is strongly expressed under these conditions. Table 3 summarizes the expression data for all of the GUS fusion strains.Figure 3 Expression of β-glucuronidase (GUS)-encoding reporter geneuidAinserted within SMc00911
Expression of β-glucuronidase (GUS)-encoding reporter gene uidA inserted within SMc00911.S. meliloti within alfalfa root nodules (B–F) express GUS inserted in SMc00911 throughout the nodule. Panel A shows an alfalfa nodule invaded by wild type S. meliloti 1021 that does not express GUS (subjected to the same staining procedure as B–F). (Roots in B, C, and D were inoculated with strain SMc00911. Xsd1. Roots in E and F were inoculated with strain SMc00911.original.) Nodules were stained for 3.75 hr after 5 weeks of growth post-inoculation. Scale bars correspond to 0.1 mm. Panel G shows SMc00911-controlled GUS expression in S. meliloti grown on solid LBMC medium. Wild type S. meliloti 1021 is shown as a negative control for GUS expression and a strain carrying the same GUS insertion plasmid in the greA gene is shown as a positive control for GUS expression in free-living cells. Strain SMc00911.original and a ϕM12 transductant of this strain were tested on plants.
1471-2180-12-74-3 Figure 4 Expression of β-glucuronidase (GUS)-encoding gene uidAexpressed under the control of the promoter elements of the following ORFs: SMb20360 (B and C); SMc00135 (D and E); SMc01562 (F and G); SMc01266 (H and I); SMc03964 (J and K); SMc01424-22 (L and M); SMa0044 (N and O); SMb20431 (P and Q); SMc01986 (R and S); SMa1334 (T and U)
Expression of β-glucuronidase (GUS)-encoding gene uidA expressed under the control of the promoter elements of the following ORFs: SMb20360 (B and C); SMc00135 (D and E); SMc01562 (F and G); SMc01266 (H and I); SMc03964 (J and K); SMc01424-22 (L and M); SMa0044 (N and O); SMb20431 (P and Q); SMc01986 (R and S); SMa1334 (T and U). SMb20360 and SMc00135 are strongly expressed in the nodules. (See Table 3 for percentage of nodules with GUS expression and staining times.) SMc01562, SMc01266, SMc03964 and the SMc01424-22 operon are expressed at a moderate level in the nodules. The remaining ORFs are expressed at a very low level in the nodule (or not at all). S. meliloti 1021 wild type is shown in Panel A as a negative control for GUS expression. Scale bars correspond to 0.1 mm.
1471-2180-12-74-4 Two of the other ORFs tested, SMb20360 and Smc00135, are also strongly expressed in nodules (Figure 4B–E, Table 3), and another six, SMc01562, SMc01266, SMc03964 and the three ORFs in the SMc01424-22 operon are moderately expressed (Figure 4F–M, Table 3). Of these, only SMc00135 is expressed at approximately the same level by bacteria within the nodule and by free-living bacteria ( Additional file S4 4 and Additional file S5 5 show images of the free-living expression of GUS fusions of all the ORFs tested). However, none of the other ORFs that are expressed in the nodule are expressed as strongly as SMc00911 (Figure 3 and Figure 4). Two of the ORFs, SMa0044 and SMb20431, are expressed at a very low level in the nodule, and no nodule expression was detected for SMc01986 and SMa1334 (Figure 4). Sma0044 has an unusual expression pattern in that it is expressed strongly by free-living bacteria (Additional file 5A), but its expression appears to be much reduced in the nodule (Figure 4N–O).Additional file 4Free-living expression of β-glucuronidase (GUS) under the control of the promoters of the following ORFs: A) clockwise from lower left—SMc01266; greA (positive control for GUS expression); S. meliloti 1021 wild type (negative control for GUS expression); SMb20431; SMa1334. (The cropped plate wedges in panel A are all from the same plate.) B) clockwise from lower right—SMc01986; SMc01562; SMc03964; greA; S. meliloti 1021; a second streak of SMc03964. C) (clockwise from left) greA; S. meliloti 1021; SMb20360 (two separate strains). Specific strain names are shown in the photo labels. The growth medium is LBMC, with streptomycin 500 ug/mL.1471-2180-12-74-S4.jpeg
Click here for file
Additional file 5 Free-living expression of β-glucuronidase (GUS) under the control of the promoters of the following ORFs: A) SMa0044. Multiple isolates of the SMa0044::GUS fusions are shown in comparison with greA (positive control for GUS expression) and S. meliloti 1021 wild type (negative control for GUS expression). B) SMc00135. Multiple isolates of the SMc00135::GUS fusions are shown in comparison with greA and S. meliloti 1021 wild type. C) the SMc01424-01422 operon. Multiple isolates of the SMc01424-01422: GUS fusions are shown in comparison with greA and S. meliloti 1021 wild type. The growth medium is LBMC, with streptomycin 500 ug/mL. GUS expression strains that were tested for nodule expression are denoted with an asterisk and are described in Tables 3 and 4.1471-2180-12-74-S5.jpeg
Click here for file
Because of the strong expression of SMc00911 by bacteria in the nodule, the SMc00911 mutant strains were chosen for further study in competition experiments (see below).An insertion mutant of SMc00911 out-competes the S. meliloti 1021 wild type for nodule occupancyMany S. meliloti mutant strains that are able to form a successful symbiosis when singly inoculated on host plants are deficient in the ability to successfully compete for nodule occupancy against the wild type strain in a mixed infection 42B51 51. Competitive nodulation experiments are likely to be a better approximation of the situation that rhizobial bacteria encounter in the soil, where they may be competing against several different rhizobial strains for host plant invasion and nodule occupancy. The SMc00911 insertion mutant strains were chosen for competition analysis because this ORF is strongly expressed in the nodule and these strains might be expected to be at a competitive disadvantage in the absence of the full-length SMc00911 protein. However, in contrast to expectations, the SMc00911 insertion mutant strains strongly out-compete the S. meliloti 1021 wild type strain for nodule occupancy in a mixed 1:1 infection (Table T6 6). Of the nodules tested from plants inoculated with a 1:1 mixture of 1021 wild type and an SMc00911 insertion mutant, all of the nodules were colonized by either the SMc00911 insertion mutant alone or by a mixture of the mutant and the wild type (Table 6). Less than 22% of the mixed-inoculum nodules were colonized by 1021 wild type alone. Also, all of the mixed nodules contained a larger proportion of SMc00911 insertion mutant bacteria than 1021 wild type bacteria (Table 6). The recovered bacteria from one of the 8 nodules that had been inoculated with the SMc00911.Xsd1 strain alone included a small number of neomycin-sensitive colonies (Table 6, line 3). This suggests that the gene disruption plasmid inserted in the SMc00911 ORF is lost by bacteria in the nodule at a very low rate. Taken together, these competition results suggest that disruption of the SMc00911 ORF actually confers a competitive advantage to S. meliloti in the symbiosis with host plants. The SMc00911 ORF is predicted to encode a 275 amino acid protein with a rhodanese-like sulfurtransferase domain from amino acids 7–100 and a chromate-resistance protein domain from amino acids 122–256 B52 52. The SMc00911 mutants carry the pJH104-GUS-expression/disruption plasmid inserted at nucleotide position 597 out of 828 total nucleotides, which would result in the production of a truncated protein containing only amino acids 1–199, based on the S. meliloti 1021 genome sequence B53 53B54 54. Thus the SMc00911 insertion mutants are predicted to produce a protein that contains the whole rhodanese-like sulfurtransferase domain, but only a portion of the chromate-resistance protein domain. Table 6SMc00911-disruption strains out-competeS. meliloti1021 wild type for nodule occupancyInoculumNumber of nodules tested*Number of nodules containing no neomycin-resistant bacteriaNumber of nodules containing only neomycin-resistant bacteriaNumber of nodules containing a mixture of neomycin-resistant and sensitive bacteriaAverage percent of neomycin-resistant bacteria in mixed nodulestfoot 1–2 nodules/plant were analyzed.S. meliloti 1021 wild type (neomycin-sensitive)84 = 100%0 = 0%0 = 0%N/ASMc00911.original (neomycin-resistant)160 = 0%16 = 100%0 = 0%N/ASMc00911.Xsd1 (neomycin-resistant)160 = 0%15 = 93.8%1 = 6.3%95.2% ± 0.00%SMc00911.original:1021—mixed 1:1327 = 21.9%18 = 56.3%7 = 21.9%67.4% ± 14.2%SMc00911.Xsd1:1021—mixed 1:1312 = 6.5%21 = 67.7%8 = 25.8%76.7% ± 9.8%In contrast to the SMc00911 insertion mutants, deletion mutants of SMc01562 (which is expressed in the nodule, but at a much lower level than SMc00911 (Figure 4)) are able to compete as effectively as S. meliloti 1021 wild type against a competitor assay strain carrying a neomycin-resistance marker (data not shown), suggesting that the loss of this protein confers neither a symbiotic disadvantage nor an advantage to S. meliloti 1021.DiscussionSmc00911, a conserved rhizobial ORF expressed strongly in the noduleOur comparative genomics screen has identified an S. meliloti 1021 ORF (SMc00911) that is strongly expressed within host plant nodules, but is expressed in the free-living state at a very low level. Surprisingly, disruption of this ORF confers a competitive advantage for nodule occupancy on S. meliloti 1021. Smc00911 is predicted to encode a 275 amino acid protein with overall similarity to SodM-like (superoxide dismutase-like) proteins B55 55B56 56. There are 57 “SodM-like proteins” with >40% identity to SMc00911 in the NCBI database 56. SMc00911 contains two distinct, conserved domains: a 94 amino acid domain (amino acids 7–100) similar to the GlpE sufurtransferase/rhodanese homology domain (cd01444), and a 135 amino acid (amino acids 122–256) chromate-resistance-exported protein domain (pfam09828) 52. The SMc0911 mutant strains constructed in this study are predicted to produce a protein consisting of the first 199 amino acids of the full-length protein plus four amino acids encoded by the multiple cloning site of pJH104, before encountering a stop codon (Melanie Barnett, Stanford University personal communication) 5354. This truncated protein product would include the entire rhodanese-homology domain and approximately half of the chromate-resistance protein domain. One possibility is that the competitive advantage that the SMc00911-insertion mutant strains have against the 1021 wild type strain is due to the expression of this truncated protein, rather than simply a loss-of-function of the full-length protein. Even though SMc00911 is annotated as a “SodM-like” protein in the NCBI database 535456, there are only two short segments of similarity (8 amino acids [38% identity] and 11 amino acids [36% identity]) with a protein confirmed to be a SodM from Xanthomonas campestris pv. campestris (accession no. p53654) B57 57. Thus, since the N-terminal similarity of SMc00911 to the GlpE sufurtransferase/rhodanese homology domain and the C-terminal similarity to the chromate-resistance protein domain are both greater than the similarity of this protein to SodM, “SodM-like” may not be the most-appropriate annotation for this ORF. There are two sod ORFs in the S. meliloti 1021 genome, sodB (SMc00043) (SMc02597) and a bacteriocuprein-family sodC (SMc02597) 25354. An S. meliloti 1021 sodB loss-of-function mutant forms a functional symbiosis with host plants B58 58, while the symbiotic phenotype of a sodC mutant has not been reported.Expression of other αhizobial conserved ORFSAlthough they are not required for development of a functional symbiosis by S. meliloti 1021, the ORFs SMb20360 and SMc00135 are also strongly expressed in nodules, while SMc01562, SMc01266, SMc03964 and the SMc01424-22 operon are moderately expressed (Figure 4; Table 3). However, the expression of SMc00135 is not specific to the nodule (Figure 4 and Additional file 5). SMb20360 is predicted to encode a protein of the Clp-protease superfamily (COG0740), with specific similarity to ClpP 52. Polar localization of the ClpXP protease complex within S. meliloti cells has been found to be important for S. meliloti bacteroid differentiation B59 59, and it is possible that ClpP proteases play a role in the bacteroid differentiation process. Interestingly, in another study, a signature-tagged mutant in SMb20360 was found to be highly competitive for survival, in the free-living state, in competition experiments under salt- and detergent-stressed conditions B60 60. SMc01562 is predicted to encode a member of the GYD-domain containing protein superfamily (COG4274) 52. No function has been reported for this protein family 56. SMc01266 is predicted to encode a member of the Von Willebrand factor type A (vFWA) superfamily (cl00057), however proteins containing a vFWA domain participate in a wide variety of functions B61 61. Expression of SMc01266 has previously been shown to increase in bacteroids B62 62 (reference Supplemental Dataset 3), and during phosphate stress B63 63. Smc03964 is predicted to possess a twin-arginine export signal B64 64, and to encode a member of the metallophosphatase superfamily (cl13995), a group of phosphatases with diverse functions 52. ORFs SMc01424, SMc01423, and SMc01422 appear to be part of a single operon and they encode, respectively, a predicted nitrile hydratase alpha subunit protein, a nitrile hydratase beta subunit protein, and a nitrile hydratase activator protein 5354. Nitrile hydratases function in the degradation of xenobiotic compounds, but they are also involved in tryptophan metabolism, specifically in the conversion of 3-indoleacetonitrile to indole-3-acetamide, which is a precursor of the plant hormone auxin B65 65B66 66. SMa0044 has an unusual expression pattern in that it is expressed at a very low level in approximately half of the nodules tested (Table 3; Figure 4), but is expressed quite strongly by free-living S. meliloti on LBMC medium ( Additional file 5). SMa0044 is predicted to encode a member of the DUF2277 superfamily, which is has no known function 52.ConclusionsThe goal of this study was to identify S. meliloti 1021 ORFs involved in host plant nodulation and nitrogen fixation. The comparative genomics method we employed was able to rediscover 19 ORFs that have previously been shown to be important for nodulation and/or nitrogen fixation. The earlier studies that identified these genes, in most cases, employed the classical bacterial genetic techniques of transposon mutagenesis, followed by strain isolation and phenotypic screening 11B67 67B68 68. Our study identified 9 additional S. meliloti ORFs (out of the 13 we analyzed) that we have shown are expressed primarily in host plant nodules. However none of these newly identified ORFs were required for development of a functional symbiosis under the conditions we tested. Our results suggest that the accumulated transposon screens for essential S. meliloti nodulation/nitrogen fixation genes may be nearing saturation. However, the comparative genomics method described above might be very effective for identifying factors involved in the production of a phenotype common to a group of bacterial species that have not yet been studied by classical transposon mutagenesis screens.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsKMJ conceived of the study, performed the genome comparisons, designed experiments, constructed bacterial mutant strains, performed experiments, interpreted results and drafted the manuscript. CQ designed experiments, constructed bacterial mutant strains, performed experiments, interpreted results and helped draft the manuscript. BKW constructed bacterial mutant strains, performed experiments, and helped draft the manuscript. OMD, JS, TEB, and MRL constructed bacterial mutant strains and performed experiments. All authors read and approved the final manuscript.bm ack AcknowledgmentsThe authors wish to thank Sharon Long, Melanie Barnett, and Jeanne Harris for plasmid pJH104; Graham Walker for plasmid pK19mobsac; and Michiko E. Taga, Penny J. Beuning and George W. Bates for critical reading of the manuscript.This work was funded by start-up funds provided to KMJ by Florida State University.refgrp How rhizobial symbionts invade plants: the Sinorhizobium-Medicago modelJonesKMKobayashiHDaviesBWTagaMEWalkerGCNat Rev Microbiol200758619lpage 63310.1038/nrmicro1705pmcid 2766523link fulltext 17632573Molecular determinants of a symbiotic chronic infectionGibsonKEKobayashiHWalkerGCAnnu Rev Genet20084241344110.1146/annurev.genet.42.110807.091427277058718983260Huang W: Data Sets: U.S. Fertilizer Use and Price. In. Edited by Service UER: usda.gov; 2008A plant flavone, luteolin, induces expression of Rhizobium meliloti nodulation genesPetersNKFrostJWLongSRScience198623397798010.1126/science.37385203738520Infection and invasion of roots by symbiotic, nitrogen-fixing rhizobia during nodulation of temperate legumesGageDJMicrobiol Mol Biol Rev200468228030010.1128/MMBR.68.2.280-300.200441992315187185Nuclear calcium changes at the core of symbiosis signallingOldroydGEDownieJACurr Opin Plant Biol20069435135710.1016/j.pbi.2006.05.00316713329Refined analysis of early symbiotic steps of the Rhizobium-Medicago interaction in relationship with microtubular cytoskeleton rearrangementsTimmersACAuriacMCTruchetGDevelopment1999126163617362810409507Medicago truncatula syntaxin SYP132 defines the symbiosome membrane and infection droplet membrane in root nodulesCatalanoCMCzymmekKJGannJGSherrierDJPlanta20072553541550Eukaryotic control on bacterial cell cycle and differentiation in the Rhizobium-legume symbiosisMergaertPUchiumiTAlunniBEvannoGCheronACatriceOMaussetAEBarloy-HublerFGalibertFKondorosiAetal Proc Natl Acad Sci U S A2006103135230523510.1073/pnas.0600912103145882316547129Metabolic changes of rhizobia in legume nodulesPrellJPoolePTrends Microbiol200614416116810.1016/j.tim.2006.02.00516520035Directed transposon Tn5 mutagenesis and complementation analysis of Rhizobium meliloti symbiotic nitrogen fixation genesRuvkunGBSundaresanVAusubelFMCell198229255155910.1016/0092-8674(82)90171-46288262Carbon and nitrogen metabolism in RhizobiumPoolePAllawayDAdv Microb Physiol20004311716310907556Effects of Rhizobium meliloti nif and fix mutants on alfalfa root nodule developmentHirschAMSmithCAJ Bacteriol1987169113711462119113818542Establishing nitrogen-fixing symbiosis with legumes: how many rhizobium recipes?Masson-BoivinCGiraudEPerretXBatutJTrends Microbiol2009171045846610.1016/j.tim.2009.07.00419766492Physical and genetic characterization of symbiotic and auxotrophic mutants of Rhizobium meliloti induced by transposon Tn5 mutagenesisMeadeHMLongSRRuvkunGBBrownSEAusubelFMJ Bacteriol198214911141222165986274841Genetic and structural analysis of the Rhizobium meliloti fixA, fixB, fixC, and fixX genesEarlCDRonsonCWAusubelFMJ Bacteriol19871693112711362119103029021Genes for a microaerobically induced oxidase complex in Bradyrhizobium japonicum are essential for a nitrogen-fixing endosymbiosisPreisigOAnthamattenDHenneckeHProc Natl Acad Sci U S A19939083309331310.1073/pnas.90.8.3309462898386371Copper metallochaperones are required for the assembly of bacteroid cytochrome c oxidase which is functioning for nitrogen fixation in soybean nodulesArunothayananHNomuraMHamaguchiRItakuraMMinamisawaKTajimaSPlant Cell Physiol20105171242124610.1093/pcp/pcq07920519277Biosynthesis of the iron-molybdenum cofactor of nitrogenaseRubioLMLuddenPWAnnu Rev Microbiol2008629311110.1146/annurev.micro.62.081307.16273718429691Global changes in gene expression in Sinorhizobium meliloti 1021 under microoxic and symbiotic conditionsBeckerABergesHKrolEBruandCRubergSCapelaDLauberEMeilhocEAmpeFde BruijnFJMol Plant Microbe Interact200417329230310.1094/MPMI.2004.17.3.29215000396Cascade regulation of nif gene expression in Rhizobium melilotiDavidMDaveranMLBatutJDedieuADomergueOGhaiJHertigCBoistardPKahnDCell198854567168310.1016/S0092-8674(88)80012-62842062Nature of oxygen inhibition of nitrogenase from Azotobacter vinelandiiWongPPBurrisRHProc Natl Acad Sci U S A197269367267510.1073/pnas.69.3.6724265324501581Free-living Rhizobium strain able to grow on n(2) as the sole nitrogen sourceDreyfusBLElmerichCDommerguesYRAppl Environ Microbiol198345271171324235016346220Plant peptides govern terminal differentiation of bacteria in symbiosisVan de VeldeWZehirovGSzatmariADebreczenyMIshiharaHKeveiZFarkasAMikulassKNagyATiriczHScience201032759691122112610.1126/science.118405720185722NAD+-dependent malic enzyme of Rhizobium meliloti is required for symbiotic nitrogen fixationDriscollBTFinanTMMol Micro19937686587310.1111/j.1365-2958.1993.tb01177.xProperties of NAD(+)- and NADP(+)-dependent malic enzymes of Rhizobium (Sinorhizobium) meliloti and differential expression of their genes in nitrogen-fixing bacteroidsDriscollBTFinanTMMicrobiology1997143Pt 24894989043124Will elevated carbon dioxide concentration amplify the benefits of nitrogen fixation in legumes?RogersAAinsworthEALeakeyADPlant Physiol200915131009101610.1104/pp.109.144113277310119755541The pangenome structure of Escherichia coli: comparative genomic analysis of E. coli commensal and pathogenic isolatesRaskoDARosovitzMJMyersGSMongodinEFFrickeWFGajerPCrabtreeJSebaihiaMThomsonNRChaudhuriRJ Bacteriol2008190206881689310.1128/JB.00619-08256622118676672Comparative genomics of bacterial and plant folate synthesis and salvage: predictions and validationsde Crecy-LagardVEl YacoubiBde la GarzaRDNoirielAHansonADBMC Genomics2007824510.1186/1471-2164-8-245197107317645794Genome sequence of the plant pathogen and biotechnology agent Agrobacterium tumefaciens C58GoodnerBHinkleGGattungSMillerNBlanchardMQurolloBGoldmanBSCaoYAskenaziMHallingCScience200129455502323232810.1126/science.106680311743194The genome of the natural genetic engineer Agrobacterium tumefaciens C58WoodDWSetubalJCKaulRMonksDEKitajimaJPOkuraVKZhouYChenLWoodGEAlmeidaNFsuf JrScience200129455502317232310.1126/science.106680411743193Complete genome sequence of Caulobacter crescentusNiermanWCFeldblyumTVLaubMTPaulsenITNelsonKEEisenJAHeidelbergJFAlleyMROhtaNMaddockJRProc Natl Acad Sci U S A20019874136414110.1073/pnas.0610292983119211259647Diversity of chlorophenol-degrading bacteria isolated from contaminated boreal groundwaterMannistoMKTiirolaMASalkinoja-SalonenMSKulomaaMSPuhakkaJAArch Microbiol1999171318919710.1007/s00203005069810201097Genome Project: Caulobacter sp. K31 [http://img.jgi.doe.gov/cgi-bin/pub/main.cgi?section = TaxonDetail&page = taxonDetail&taxon_oid = 641522612]Complete genomic sequence of nitrogen-fixing symbiotic bacterium Bradyrhizobium japonicum USDA110KanekoTNakamuraYSatoSMinamisawaKUchiumiTSasamotoSWatanabeAIdesawaKIriguchiMKawashimaKDNA Res20029618919710.1093/dnares/9.6.18912597275Complete genome structure of the nitrogen-fixing symbiotic bacterium Mesorhizobium loti (supplement)KanekoTNakamuraYSatoSAsamizuEKatoTSasamotoSWatanabeAIdesawaKIshikawaAKawashimaKDNA Res20007638140610.1093/dnares/7.6.38111214974The partitioned Rhizobium etli genome: genetic and metabolic redundancy in seven interacting repliconsGonzalezVSantamariaRIBustosPHernandez-GonzalezIMedrano-SotoAMoreno-HagelsiebGJangaSCRamirezMAJimenez-JacintoVCollado-VidesJProc Natl Acad Sci U S A2006103103834383910.1073/pnas.0508502103138349116505379The genome of Rhizobium leguminosarum has recognizable core and accessory componentsYoungJPCrossmanLCJohnstonAWThomsonNRGhazouiZFHullKHWexlerMCursonARToddJDPoolePSGenome Biol200674R3410.1186/gb-2006-7-4-r34155799016640791Complete genome sequence of the Medicago microsymbiont Ensifer (Sinorhizobium) medicae strain WSM419ReeveWGChainPO’HaraGArdleyJNandesenaKBrauLTiwariRPMalfattiSKissHLapidusAStandards in Genomic Sciences20102778610.4056/sigs.43526303525921304680SambrookJFritschEFManiatisTMolecular cloning: a laboratory manualpublisher Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N. Yedition 21982Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicumSchäferATauchAJägerWKalinowskiJThierbachGPühlerAGene1994145697310.1016/0378-1119(94)90324-78045426Importance of unusually modified lipid A in Sinorhizobium stress resistance and legume symbiosisFergusonGPDattaACarlsonRWWalkerGCMol Microbiol2005561688010.1111/j.1365-2958.2005.04536.x15773979Genetic techniques in Rhizobium melilotiGlazebrookJWalkerGCMethods Enzymol19912043984181658566General transduction in Rhizobium melilotiFinanTMHartweigELeMieuxKBergmanKWalkerGCSignerERJ Bacteriol198415911201242156016330024Exopolysaccharide-deficient mutants of Rhizobium meliloti that form ineffective nodulesLeighJASignerERWalkerGCProc Natl Acad Sci USA1985826231623510.1073/pnas.82.18.62313910263862129VincentJMA Manual for the Practical Study of the Root-Nodule BacteriaBlackwell, Oxford1970GUS fusions: beta-glucuronidase as a sensitive and versatile gene fusion marker in higher plantsJeffersonRAKavanaghTABevanMWEMBO J1987613390139075538673327686IMG/M: a data management and analysis system for metagenomesMarkowitzVMIvanovaNNSzetoEPalaniappanKChuKDaleviDChenIMGrechkinYDubchakIAndersonINucleic Acids Res200836Database issueD534D538223895017932063Legumes regulate Rhizobium bacteroid development and persistence by the supply of branched-chain amino acidsPrellJWhiteJPBourdesABunnewellSBongaertsRJPoolePSProc Natl Acad Sci U S A200910630124771248210.1073/pnas.0903653106271834019597156The Rhizobium etli cyaC product: characterization of a novel adenylate cyclase classTellez-SosaJSoberonNVega-SeguraATorres-MarquezMECevallosMAJ Bacteriol2002184133560356810.1128/JB.184.13.3560-3568.200213515112057950Striking complexity of lipopolysaccharide defects in a collection of Sinorhizobium meliloti mutantsCampbellGRSharypovaLAScheidleHJonesKMNiehausKBeckerAWalkerGCJ Bacteriol2003185133853386210.1128/JB.185.13.3853-3862.200316159412813079CDD: a Conserved Domain Database for the functional annotation of proteinsMarchler-BauerALuSAndersonJBChitsazFDerbyshireMKDeWeese-ScottCFongJHGeerLYGeerRCGonzalesNRNucleic Acids Res201139Database issueD225D229301373721109532The composite genome of the legume symbiont Sinorhizobium melilotiGalibertFFinanTMLongSRPühlerAAbolaPAmpeFBarloy-HublerFBarnettMJBeckerABoistardPScience2001293553066867210.1126/science.106096611474104A portal for rhizobial genomes: RhizoGATE integrates a Sinorhizobium meliloti genome annotation update with postgenome dataBeckerABarnettMJCapelaDDondrupMKampPBKrolELinkeBRubergSRunteKSchroederBKJ Biotechnol20091401–24550265659519103235Smc01944, a secreted peroxidase induced by oxidative stresses in Sinorhizobium meliloti 1021Barloy-HublerFCheronAHellegouarchAGalibertFMicrobiology2004150Pt 365766414993315Gapped BLAST and PSI-BLAST: a new generation of protein database search programsAltschulSFMaddenTLSchafferAAZhangJZhangZMillerWLipmanDJNucleic Acids Res199725173389340210.1093/nar/25.17.33891469179254694Identification of the bacterial superoxide dismutase (SodM) as plant-inducible elicitor of an oxidative burst reaction in tobacco cell suspension culturesWattSATellstromVPatschkowskiTNiehausKJ Biotechnol20061261788610.1016/j.jbiotec.2006.02.02216603270Disruption of sitA Compromises Sinorhizobium meliloti for Manganese Uptake Required for Protection Against Oxidative StressDaviesBWWalkerGCJ Bacteriol2006189521012109185575317172335Sinorhizobium meliloti CpdR1 is critical for co-ordinating cell cycle progression and the symbiotic chronic infectionKobayashiHDe NiscoNJChienPSimmonsLAWalkerGCMol Microbiol200973458660010.1111/j.1365-2958.2009.06794.x275602419602145Construction of a large signature-tagged mini-Tn5 transposon library and its application to mutagenesis of Sinorhizobium melilotiPobigayloNWetterDSzymczakSSchillerUKurtzSMeyerFNattkemperTWBeckerAAppl Environ Microbiol20067264329433710.1128/AEM.03072-05148959816751548Type A modules: interacting domains found in several non-fibrillar collagens and in other extracellular matrix proteinsColombattiABonaldoPDolianaRMatrix19931342973068412987A dual-genome Symbiosis Chip for coordinate study of signal exchange and development in a prokaryote-host interactionBarnettMJTomanCJFisherRFLongSRProc Natl Acad Sci U S A2004101471663616641Epub 12004 Nov 1661210.1073/pnas.040726910152792215542588Global transcriptional analysis of the phosphate starvation response in Sinorhizobium meliloti strains 1021 and 2011KrolEBeckerAMol Genet Genomics20042721117Epub 2004 Jun 202315221452The twin-arginine translocation (Tat) system is essential for Rhizobium-legume symbiosisMeloniSReyLSidlerSImperialJRuiz-ArguesoTPalaciosJMMol Microbiol20034851195120710.1046/j.1365-2958.2003.03510.x12787349Occurrence of enzymes involved in biosynthesis of indole-3-acetic acid from indole-3-acetonitrile in plant-associated bacteria, Agrobacterium and RhizobiumKobayashiMSuzukiTFujitaTMasudaMShimizuSProc Natl Acad Sci U S A199592371471810.1073/pnas.92.3.7144269011607511Indole-3-acetic acid in microbial and microorganism-plant signalingSpaepenSVanderleydenJRemansRFEMS Microbiol Rev200731442544810.1111/j.1574-6976.2007.00072.x17509086Physical and genetic characterization of Rhizobium meliloti symbiotic mutantsBuikemaWJLongSRBrownSEvan den BosRCEarlCAusubelFMJ Mol Appl Genet1983232492606363587Rhizobium meliloti nodulation genes: identification of nodDABC gene products, purification of nodA protein, and expression of nodA in Rhizobium melilotiEgelhoffTTLongSRJ Bacteriol198516425915992142932997121