Role of Specific Quorum-Sensing Signals in the Regulation of Exopolysaccharide II Production within Sinorhizobium melilo...

http://dx.plos.org/10.1371/journal.pone.0042611 ( Publisher's URL )
MISSING IMAGE

Material Information

Title:
Role of Specific Quorum-Sensing Signals in the Regulation of Exopolysaccharide II Production within Sinorhizobium meliloti Spreading Colonies
Physical Description:
Serial
Creator:
Gao, Mengsheng
Coggin, Andrew
Yagnik, Kruti
Teplitski, Max
Publisher:
PLoS ONE
Publication Date:

Subjects

Genre:
serial   ( sobekcm )

Notes

Abstract:
Background Quorum sensing (QS) in Sinorhizobium meliloti involves at least half a dozen different N-acyl homoserine lactone (AHL) signals. These signals are produced by SinI, the sole AHL synthase in S. meliloti Rm8530. The sinI gene is regulated by two LuxR-type transcriptional regulators, SinR and ExpR. Mutations in sinI, sinR and expR abolish the production of exopolysaccharide II (EPS II). Methodology/Principal Findings This study investigated a new type of coordinated surface spreading of Rm8530 that can be categorized as swarming. Motility assays on semi-solid surfaces revealed that both flagella and EPS II are required for this type of motility. The production of EPS II depends on AHLs produced by SinI. Of these AHLs, only C16:1- and 3-oxo-C16:1-homoserine lactones (HSLs) stimulated swarming in an ExpR-dependent manner. These two AHLs induced the strongest response in the wggR reporter fusions. WggR is a positive regulator of the EPS II biosynthesis gene expression. The levels of the wggR activation correlated with the extent of swarming. Furthermore, swarming of S. meliloti required the presence of the high molecular weight (HMW) fraction of EPS II. Within swarming colonies, a recombinase-based RIVET reporter in the wggR gene was resolved in 30% of the cells, indicating an enhanced regulation of EPS II production in the subpopulation of cells, which was sufficient to support swarming of the entire colony. Conclusions/Significance Swarming behavior of S. meliloti Rm8530 on semi-solid surfaces is found to be dependent on the functional QS regulatory cascades. Even though multiple AHL signals are produced by the bacterium, only two AHLs species, C16:1- and 3-oxo-C16:1-HSLs, affected swarming by up-regulating the expression of wggR. While EPS II is produced by Rm8530 as high and low molecular weight fractions, only the HMW EPS II facilitated initial stages of swarming, thus, suggesting a function for this polymer.
Funding:
This work was supported by United States Department of Agriculture National Research Initiative 2007-35319-18158 (CRIS project # FLA-SWS-004631) in cooperation with The Current Research Information System # FLA-SWS-004984. Additional funding was provided by University of Florida Research Innovation Grant program. Research contribution of KY was funded by University of Florida Institute of Food and Agricultural Sciences Dean for Summer Research Internship Program. Publication of this article was funded in part by the University of Florida Open-Access Publishing Fund. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:

This item is licensed with the Creative Commons Attribution License. This license lets others distribute, remix, tweak, and build upon this work, even commercially, as long as they credit the author for the original creation.
System ID:
AA00012393:00001


This item is only available as the following downloads:


Full Text

PAGE 1

RoleofSpecificQuorum-SensingSignalsinthe RegulationofExopolysaccharideIIProductionwithinSinorhizobiummelilotiSpreadingColoniesMengshengGao ,AndrewCoggin¤,KrutiYagnik,MaxTeplitskiSoilandWaterScienceDepartment,UniversityofFlorida/InstituteofFoodandAgriculturalSciences,Gainesville,Florida,UnitedStatesofAmer icaAbstractBackground:Quorumsensing(QS)in Sinorhizobiummeliloti involvesatleasthalfadozendifferent N -acylhomoserine lactone(AHL)signals.ThesesignalsareproducedbySinI,thesoleAHLsynthasein S.meliloti Rm8530.The sinI geneis regulatedbytwoLuxR-typetranscriptionalregulators,SinRandExpR.Mutationsin sinI sinR and expR abolishthe productionofexopolysaccharideII(EPSII).Methodology/PrincipalFindings:ThisstudyinvestigatedanewtypeofcoordinatedsurfacespreadingofRm8530thatcan becategorizedasswarming.Motilityassaysonsemi-solidsurfacesrevealedthatbothflagellaandEPSIIarerequiredforthis typeofmotility.TheproductionofEPSIIdependsonAHLsproducedbySinI.OftheseAHLs,onlyC16:1-and3-oxo-C16:1homoserinelactones(HSLs)stimulatedswarminginanExpR-dependentmanner.ThesetwoAHLsinducedthestrongest responseinthe wggR reporterfusions.WggRisapositiveregulatoroftheEPSIIbiosynthesisgeneexpression.Thelevelsof the wggR activationcorrelatedwiththeextentofswarming.Furthermore,swarmingof S.meliloti requiredthepresenceof thehighmolecularweight(HMW)fractionofEPSII.Withinswarmingcolonies,arecombinase-basedRIVETreporterinthe wggR genewasresolvedin30%ofthecells,indicatinganenhancedregulationofEPSIIproductioninthesubpopulationof cells,whichwassufficienttosupportswarmingoftheentirecolony.Conclusions/Significance:Swarmingbehaviorof S.meliloti Rm8530onsemi-solidsurfacesisfoundtobedependentonthe functionalQSregulatorycascades.EventhoughmultipleAHLsignalsareproducedbythebacterium,onlytwoAHLs species,C16:1-and3-oxo-C16:1-HSLs,affectedswarmingbyup-regulatingtheexpressionof wggR .WhileEPSIIisproducedby Rm8530ashighandlowmolecularweightfractions,onlytheHMWEPSIIfacilitatedinitialstagesofswarming,thus, suggestingafunctionforthispolymer.Citation: GaoM,CogginA,YagnikK,TeplitskiM(2012)RoleofSpecificQuorum-SensingSignalsintheRegulationofExopolysaccharideIIProductionwithin Sinorhizobiummeliloti SpreadingColonies.PLoSONE7(8):e42611.doi:10.1371/journal.pone.0042611 Editor: GunnarF.Kaufmann,TheScrippsResearchInstituteandSorrentoTherapeutics,Inc.,UnitedStatesofAmerica Received February28,2012; Accepted July10,2012; Published August13,2012 Copyright: 2012Gaoetal.Thisisanopen-accessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense,whichpermitsunrestricted use,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited. Funding: ThisworkwassupportedbyUnitedStatesDepartmentofAgricultureNationalResearchInitiative2007-35319-18158(CRISproject # FLA-SWS-004631) incooperationwithTheCurrentResearchInformationSystem # FLA-SWS-004984.AdditionalfundingwasprovidedbyUniversityofFloridaResearchInnovation Grantprogram.ResearchcontributionofKYwasfundedbyUniversityofFloridaInstituteofFoodandAgriculturalSciencesDeanforSummerResearchI nternship Program.PublicationofthisarticlewasfundedinpartbytheUniversityofFloridaOpen-AccessPublishingFund.Thefundershadnoroleinstudydesi gn,data collectionandanalysis,decisiontopublish,orpreparationofthemanuscript. CompetingInterests: Theauthorshavedeclaredthatnocompetinginterestsexist. *E-mail:msgao@ufl.edu ¤Currentaddress:MillerSchoolofMedicine,UniversityofMiami,Miami,Florida,UnitedStatesofAmericaIntroductionS.meliloti isasoil a -proteobacterium,bestknownforitsabilityto establishnitrogen-fixingsymbioseswithplanthostsbelongingto thegenera Medicago Melilotus and Trigonella .Signalingand regulatoryeventsthattakeplaceduringtheearlystagesofthe symbiosesarestudiedandsomeofthesepathwaysarewelldefined [1–3].Lessstudiedarethebehaviorsofthisbacteriumoutsidethe hoststhatinfluencethesymbioses,suchasbacterialquorumsensingsignalingintherhizosphere[4],biofilmformation[5,6] andthemovementoftherhizobiumonsurfaces[7–9]. Bacteriausevarioustypesofmotilitytorelocatetheir populationsonsurfacesinsearchforamoresuitableenvironmentalniche[10].Typesofsurfacemotilityincludeswarming, sliding,gliding,andtwitching[11].Itisthoughtthatmotilityin rhizobiaiscriticalfortheestablishmentofthesymbiosisunder naturalconditions[8]becauseithelpsthebacteriatogainbetter accesstonutrients,expandintonewinchesandcolonizehosts. Swarmingmotilityisamulticellularbacterialmovementacross asurface.Itisdrivenbyrotatingflagellaandcoupledtothe productionofamucoidlayerthatfacilitatesthemovement [11,12].Thelatterservesassurfactantstoreducetensionbetween thesubstrateandthebacterialcellsattheswarmingfront[13]or aswettingagentstoextractwaterfromthesurroundings[11,12]. Surfactantsandwettingagentscanbecostlytosynthesize,but oncereleased,benefitothercellswithintherange,thusleadingto theircharacterizationas‘‘publicgoods’’[14].Thebenefits(aswell ascosts)andmechanismsofsuchcooperativebehaviorsarea subjectofresearch[15–17].Theproductionsofsomeofthose publicgoodsarecontrolledbyquorumsensing(QS)systems PLOSONE|www.plosone.org1August2012|Volume7|Issue8|e42611

PAGE 2

[18,19].BesidefunctionasQSsignals,AHLswithlong N -acyl chainsalsofunctionassurfactantsin Rhizobiumetli [20]. S.meliloti strainRm8530usestheAHLsynthaseSinItoproduce atleasesevenAHLmolecules.TheyareC12-HSL,C14-HSL,3oxo-C14-HSL,C16-HSL,3-oxo-C16:1-HSL,C16:1-HSL,C18-HSL [21,22].AtleasttwoLuxRtypetranscriptionalregulators,SinR andExpR,regulatetheexpressionof sinI [23–25].Inthepresence ofSinIAHLs,ExpRcontrolstheaccumulationofdozensof transcriptsincludingthoseencodedbytheEPSIIgenecluster [24,26,27]. EPSII,agalactoglucanpolymer,isoneofthetwosymbiotically importantexopolysaccharidesproducedby S.meliloti Rm8530 [23,28,29].EPSIIissecretedintwofractions,highandlow molecularweights.Alowmolecularweight(LMW)EPSIIfraction consistsof15–20disaccharidesubunits,itallowstherhizobial noduleinvasionin Medicagosativa [30],anditisalsocriticalforthe biofilmformationandautoaggregationunderlaboratoryconditions[6,31].Thefunctionforthehighmolecularweight(HMW) EPSIIfractionhasremainedelusive. TheEPSIIgeneclustercontains22genes.Itisorganizedinto wge (alsocalled expE ), wga ( expA ), wgd ( expD ), wggR ( expG )and wgcA ( expC )operons[23,32].WgcAiscriticalfortheinitiationofEPSII biosynthesis.Proteinsencodedby wge ( expE ), wga ( expA ),and wgd ( expD )operonsareresponsibleforthepolymerizationofEPSII [33].WggR,amemberofMarRfamilyofregulators,activates wga wgd wggR wgcA and wgeA operonsbyinteractingwiththe conservedpalindromemotifsinthetargetpromoterregions [34,35].Disruptionof wggR preventstheproductionofEPSII [33].ExpRstimulatestheexpressionsof wggR [24]andotherEPS IIgenesinthepresenceofSinIAHLandWggRprotein[27]. MucR,anotherregulatoryprotein,negativelyaffectstheEPSII synthesisbyrepressing wgaA wgdA ,and wggR genes[8,36,37]. Disruptionof mucR promotessynthesisoftheHMWEPSII fraction[30].Inaddition,thesynthesisofEPSIIin S.meliloti isalso regulatedbyphosphatestarvation[35,37]. Inthisstudy,wefirstdescribethecharacterizationofaflagellaandEPSII-dependentsurfaceswarmingbehaviorof S.meliloti Rm8530.WetheninvestgatedhowAHLsignalsproducedby Rm8530contributetotheregulationofthebacterialswarming. WefoundthatHMWEPSIIiscentralfortheintiationofswarm andthattheproductionofEPSIIiscontrolledbythespecificSinI AHLsthroughstimilatingtheexpressionofregulatorygene wggR .Results Astructuredsurfacespreadingbehaviorin S.meliloti Rm8530AsshowninFig.1, S.meliloti strainRm8530formedlarge mucoidcoloniesthatspreadslowlyoverthesurfaceofverysoft agar(0.4%)anddevelopeddistinctpatterns.Eventhough S.meliloti 8530bacteriawerepreviouslyshowntoslidonaharderagar medium(0.6%)[8,9]andspreadonregularagarmedium(1.5%) [24,33],thedistinctpatternsobservedinFig.1werenotseen underthoseconditions[8,9,33].Thestructuredsurfacespreading colonyofRm8530wasenclosedwithinanextracellularmucoid matrixandhadanentireedge.Spreadingwasmorepronounced whentheagarwasbasedon20-fold-dilutedLuria-Bertani(LB) medium. Tobettervisualizethepatternswithinthespreadingcolonyof Rm8530,thebacteriaweremarkedwithpDG71whichcontainsa constitutivelyexpressed gfp geneforgreenfluorescentprotein (GFP)[38].Thespreadingcolonyformationwasdocumentedover afive-dayperiodunderamicroscopeandseveraldistinctivestages intheformationofthespreadingcolonywereobserved(Fig.1). After10–14hoursofincubationofRm8530onthesoftagar,an unevendistributionofbacterialcellswithinthepre-swarming colonieswasobserved,resultinginmany‘‘terraces’’and‘‘valleys’’ orSwisscheese-likestructureinthecenterofthecolonies.Afew hourslater,coloniesappeared‘‘wet’’andbegantospread.The coloniesmovedat0.15–0.3mm/s(0.5–1mm/h)betweenday1(Istage)andday3(III-stage)aftertheinoculation(Fig.1).The colonieswereimmersedinandwereapparentlyguidedbythe almosttransparentmucoidmatrix.Thecolonyeventually developedafeather-likemorphologywithpoolsandchannelsof slime(Fig.1).Byspreadingoverthesurface,thebacteria multipliedtohighernumbers,presumablybygainingaccessto nutrients.Basedontheopticaldensity(OD600)measurements,in 3–4daysthetotalnumberofcellswithinthespreadingcoloniesof Rm8530wasestimatedtobe2–3timeshigherthanthosewithin coloniesofmutantsthatwereunabletospread(Fig.S1).However, growthratesofthesestrainsinshakecultureswerenearlyidentical.RoleofflagellainthesurfacespreadingBecausehyper-flagellationisoftenassociatedwithbacterial swarming[11,12],over70ofRm8530cellscollectedfromthe spreadingcolonieswereexaminedunderascanningelectron microscope(SEM)(Fig.2A)forthenumberofflagellaassociated withcells.Nohyper-flagellationwasobserved.Themajorityof cellshadtwotofourflagellapercell(Fig.2A),whichisconsistent withanearlierreportoftwotosixflagellaperswimming S.meliloti cell[39].CellscollectedfromcoloniesformedbyRm8530onhard agarwerenotassociatedwithflagella(Fig.2B).Thepresenceof flagellaonthecellsrecoveredfromthespreadingcolonies suggestedthatthistypeofspreadingisdistinctfromsliding,which isdefinedasapassiveexpansionoversemi-solidsurfaceswithina mucoidlayer[12]. Toconfirmthenatureofthistypeofsurfacespreading,we testednon-flagellatedmutantsMG320( fliP )andRm11601( flaA flaB )[40]fortheirabilitytoformstructuredspreadingcolonieson softagarsurfaces.BothmutantsproducedcopiousamountsofEPS II.However,theMG320mutantdidnotformdistinctfeather-like patternsanddidnotspreadasfastasthewildtype(Fig.2C). SimilarresultswereobservedforRm11601mutant(Fig.2D). Theseobservationsconfirmedthatflagellaareneededforthistype ofmotility.InRm910[41],amutantstrainlackingboth expR and fliP ,thesurfacespreadingwascompletelyabolished,indicating thatbothEPSIIandflagellaarerequiredforspreading(Fig.3C). BecausebothflagellaandEPSIIarerequiredforthistypeof motility,itcanbecharacterizedasswarming.EPSIisnotinvolvedinthesurfacespreadingBecausethe S.meliloti Rm8530swarmingcoloniesareenclosed withinamucoidmatrix,andthebacteriumsecretestwo exopolysaccharides,EPSIandEPSII,experimentswere conductedtodeterminewhetherornotbothEPSIandIIare involvedintheswarming.Swarmingphenotypesoftheisogenic EPSImutantRm11603( exoY )[40]andRm9030-2( wgaA )[24] wereanalyzed(Fig.3).The exoY geneencodesanenzymethatis involvedintheinitiationoftheassemblyofrepeatingunitsofEPS I[37,42].Thedisruptionof exoY didnotaffectswarming,whilea mutationin wgaA abolishedswarming.Theseresultsindicatethe EPSIhasnomajorfunctioninRm8530swarming.RoleofQSinRm8530swarmingThe sinI and expR mutantsof S.meliloti Rm8530havebeen previouslyshowntobeincapableofformingspreadingcolonieson softagar(0.3%),implyingthatQScontributestoRm8530 swarming[22].ThereareatleasttwopossibilitiestoaddresstheRegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org2August2012|Volume7|Issue8|e42611

PAGE 3

roleofQSintheswarming:AHLsmaydirectlyfacilitatespreading (asreportedfor R.etli )[20],ortheymaysetoffaQSregulatory cascadethatleadstotheexpressionofthegenesinvolvedinthe productionoftheEPSII. AsshowninFig.4A(andconsistentwithpreviousreports [22,27]),coloniesofthe sinI mutantMG32,the sinR mutant MG170,andthe expR mutantRm1021weredryanddidnot spread.Complementationofthe sinI mutantMG32withp sinI ,a vectorcarrying sinI genewith sinI promoter(downstreamandin thesamedirectionasthevector-borne lac promoterthatis functionalinrhizobia[43])fullyrestoredswarming(Fig.4A). Complementationofthe sinR mutantMG170withp sinR ,avector carrying sinR genewiththe sinR promoter,fullyrestoredswarming (Fig.4A).BecauseSinRisaknownpositivetranscriptional regulatorof sinI [44],wefurthertestedtheeffectof sinI (supplied in trans )onthebehaviorofMG170.Theintroductionofp sinI into MG170partiallyrestoredthecolonyspreadingphenotypeofthe sinR mutant(Fig.4A).Thispartialrestorationofswarmingin MG170byp sinI likelyreflectsthetranscriptionof sinI fromthe plasmid-borne lac promoter.Theseresultssuggestthatthemajor functionofSinRinswarmingisrestrictedtoitsroleincontrolling theexpressionof sinI .Complementationofthe expR mutant Rm1021withp expR ,avectorcarrying expR genewiththe expR promoter(placedinthesamedirectionanddownstreamfromthe vector-borne lac promoter),restoredswarming.However,the patternwasdistinctfromthatofthewildtype(Fig.4A).The introductionofpTH113(whichcarriesan8.5kbfragmentof S. meliloti chromosomeincluding sinRI )[45]didnotoverridethe swarmingdefectofthe expR mutationinRm1021,andthe Rm1021pTH113strainformeddrycolony(Fig.4A).These resultssuggestthatafunctionalExpRisresponsibleforthe perceptionoftheAHLsand/ortheregulationofthegenes involvedinswarmofRm8530. TofollowuponthehypothesisthatExpR-mediatedregulation wascentraltosurfacespreading,phenotypesofthegenes controlledbyExpRandinvolvedinEPSIIbiosynthesiswere tested.SinceWggRactivatestheexpressionofEPSIIgenesand thattheexpressionof wggR isstimulatedbyExpR[27],wetested thestrainRm9034[24],a S.meliloti 8530derivativecarryinga Figure1.StagesofS.melilotiRm8530colonyspreading. Aseriesofdark-field(toppanels)anditscorrespondingfluorescent(lowerpanels) imagesofa S.meliloti Rm8530pDG71(markedwithconstitutivelyexpressed gfp )swarmingcolonyspreadingovera0.4%agarsurface.Stagesof spreadingthatareshown:pre-spreading(stagep,3hoursafterincubation);formationofaSwisschees-likeappearanceandinitiationofspreading (stage1,20hoursafterincubation);formationoffeather-likepatternsandcontinuationofthecolonyspreading(stageII,III,40hand3daysafter inoculation,respectively);complexpatternsandlaterstagesofthecolonyspreading(stageIVandV,4dayand5daysafterincubationrespectively ). Micrographsweretakenusingadigitalcameraconnectedtoadissectingmicroscope.Ifasinglecolonycouldnotbecapturedasoneimage,images wereassembledinAdobePhotoshopCS,andedgesofassembledimageswereleftvisible. doi:10.1371/journal.pone.0042611.g001 Figure2.TheroleofflagellainthesurfacespreadingofS. meliloti8530.A. Electronmicrographsofaflagellated S.meliloti Rm8530cellfromatwo-dayoldspreadingcolonyformedonsoft0.4% agarandflagella-less S.meliloti 8530bacteriafromanon-spreading colonyformedonhard1.5%agar( B ). C. Appearanceofcoloniesformed byRm8530(wildtype),thenon-flagellatedmutantMG320( fliP )andthe EPSIIandnon-flagellatedmutantRmG910( expR,fliP )aftertwodaysof spreadingon0.4%agar. D. Appearanceofacolonyformedbya flagellinmutantRm11601( flaA,flaB )afterfourdaysofincubationunder similarconditions.CopiousamountsofmucoidEPSIIareseenonthe edgesofcoloniesformedbyMG320andRm11601. doi:10.1371/journal.pone.0042611.g002 Figure3.TheroleofEPSIandEPSIIinsurfacespreading. When S.meliloti8530 mutantsdefectiveinEPSI( exoY ,Rm11603)andEPSII ( wgaA ,Rm9030-2)biosynthesisweretestedon0.4%softagarsurfaces, onlyEPSII-deficientmutantwasunabletospread. doi:10.1371/journal.pone.0042611.g003 RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org3August2012|Volume7|Issue8|e42611

PAGE 4

mutationin wggR ,foritscolonymorphologyonsoftagar.As showninFig.4B,Rm9034formedadrycolonythatdidnot spread.TheintroductionofpKY wggR ,avectorcarrying wggR regulatedbybothitsnativepromoteranda lac promoterfromthe vector,restoredtheswarmingphenotype,butthesizeofthe colonywaslessthanthatofthewildtypeRm8530(Fig.4B).This suggeststhatthetimingandthelevelof wggR expressionarelikely important.ThepKY wggR plasmiddidnotrestoretheswarming defectofthe sinI mutant(Fig.4B).Theinabilityofoverexpressed expR or sinR and sinI torestoretheswarmphenotypeofthe wggR mutant(Fig.4B)suggeststhatthesegeneproductscontributelittle totheswarmphenotypewithoutWggR.TheinabilityofpKY wggR toover-ridethe sinI mutation(Fig.4B)isconsistentwiththe previouslydocumented[27,33]directinvolvementofExpR-AHL complexesintheregulationof wggR andsomeotherEPSIIgene expression.Timecourseof sinI and wggR expressionduringRm8530 swarmingTobegintounderstandthedynamicsoftheQSregulationin Rm8530swarming,activitiesofachromosomal gusA reporterand aplasmid-borne gfp reporter,eachfusedseparatelywith sinI and wggR ,weremeasuredseveraltimesduringbacterialgrowthonthe surfaceof0.4%agar.Inthewildtypebackground,theexpression ofthe sinI-gusA (MG301,Fig.5A)increasedafterfivehoursof growthwithinthecolonyonsoftagar(consistentwiththelate StageP,beforetheappearanceofpatternswithinthecolony, Fig.1)andthenkeptincreasingthroughoutalmosttheentire courseoftheswarming.Theexpressionofthesame sinI-gusA reporterinthe sinI backgroundremainedatlowlevels(MG302, Fig.5A).TheadditionofC16:1-HSL,oneofseveralSinIAHLs, addedintothesoftagarincreasedtheactivityofMG302tonearly wildtypelevels(Fig.5A).Thisconfirmsthat sinI isautoregulated withinswarmingcolonies.Thisconclusionwasfurthersupported bytestingtheactivityofpMG309(aplasmidcarryinga sinI-gfp fusion)inthewildtypestrainRm8530,inthe sinI mutantMG32, andinthe sinI mutantgrownonthesoftagarcontainingC16:1HSL.Theexpression-profilesof sinI-gfp weresimilartothoseof sinI-gusA fusions(fig.5Bcomparedto5A). Undersimilarconditions,theexpressionof wggR followeda similartimecourseas sinI asindicatedbyboththechromosomal wggR gusA reporters(MG305andMG306,Fig.6A)andbythe plasmid-borne wggR-gfp reporter(pMG310,Fig.6B).Theexpressionof wggR wasinducedbyC16:1-HSL(Fig.6).Theseresults indicatethattheexpressionof wggR dependsonSinIAHLwithin spreadingcolonies,thereforeprovidingimportantinformationon theroleoftheQSincontrollingEPSIIbiosynthesisanditsrolein swarming.Theseresultsarealsoinanagreementwithearlier transcriptionalstudies[24,33].SpecificSinIAHLsrestoreswarmingphenotypesofthe sinI andthe sinR mutants,butnotthe expR mutantand EPSIIdefectivemutantsSinIisknowntocatalyzethesynthesisofatleastsevendifferent AHLs[21,22].WetestedfourofSinIAHLsfortheirabilityto facilitateswarming.Theadditionof200nMofC16:1-and3-oxoC16:1HSLsseparatelyaddedintothesoftagargrowthmedium restoredswarmofthe sinI mutantMG32andthe sinR mutant MG170,butnotthe expR mutantRm1021(Fig.7,toprows).At thesameconcentration,neitherC14-nor3-oxo-C14-HSLaffected swarminginthe sinI ,the sinR andthe expR mutants(Fig.7,two bottomrows).ThisisconsistentwithearlierobservationsofC16:1andoxo-C16:1-HSLsrestoringsurfacespreadingofthe sinI mutant [22].TheabilityofAHLstorestoreswarminginthe sinR mutantis alsoconsistentwiththeabilityofplasmid-borne sinI topartially rescueswarminginthe sinR mutant(Fig.4A).Thisfurther supportsthehypothesisthatthemajorfunctionofSinRin S. meliloti Rm8530swarmingistostimulatethesynthesisofAHLsby controllingtheexpressionof sinI gene. TheadditionofotherAHLs,includingC8-HSL(at22mM), C12-HSL(18mM),andC16-HSL(3.5mM),separatelyaddedinto softagargrowthmediumdidnotrestoretheswarmingphenotypes ofthe sinI ,the sinR ,andthe expR mutants(datanotshown).These resultsindicatethattwospecificSinIAHLs(C16:1-andoxo-C16:1HSLs)areinvolvedinswarminginExpR-dependentfashion. ExpRhasaknownfunctionofcontrollingEPSIIproduction. StimulatedbyExpR(andinconcertwithit),WggRprotein interactswiththepromoterregionsoftheoperonsinvolvedinEPS IIbiosynthesisandsecretion,including wga and wgd [35].By testingtheswarmingphenotypesoftheEPSIImutantsinthe presenceorabsenceofAHLs,weinvastigated(1)whetherAHLs, assignals,affectswarmingindirectlyviatheExpR-WggRmediatedEPSIIsynthesisor(2)whetherAHLsfunctiondirectly assurfactantsorsurfacewettingagents.AsshowninFig.7, swarmingin wggR wgaA and wgdA mutantswasabolished.Neither theadditionof200nMofC16:1-,3-oxo-C16:1-,C14-,3-oxo-C14HSLs(Fig.7),northeadditionofshorterchainAHLs(C8-,C12HSLsat18–22mM)(datanotshown)restoredtheabilityofthese EPSIImutantstoswarmoverthesoftagarsurface.Theseindicate thatAHLsfunctionassignalsratherthansurfactantsinRm8530 swarmingmotility,andthattheirregulatoryeffectsonswarming requireExpR-andWggR-mediatedregulatorycascadesleadingto EPSIIbiosynthesis. Figure4.ContributionsofQuorumSensinggenestosurface spreading.A .ColonyspreadingphenotypesofRm8530(wildtype), MG32( sinI ),MG170( sinR )andRm1021( expR )(toprow);Complementationandepistasisexperimentswiththegenesofinterestsuppliedin trans:MG32p sinI ,MG170p sinR ,andRm1021pKB expR (secondrow); MG170p sinI andRm1021pTH113(carryingan8.5kbgenomic fragmentcontaining sinI and sinR genes)(thirdrow) B .Colony spreadingphenotypesofRm9034( wggR ),Rm9034pKY wggR ,Rm9034 p expRkm ,Rm9034pTH113,MG32pKY wggR .MG32,MG170,Rm9034are mutantsdirectlyderivedfromRm8530.Rm1021is expR -progenitorof S. meliloti 8530. doi:10.1371/journal.pone.0042611.g004 RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org4August2012|Volume7|Issue8|e42611

PAGE 5

Theeffectsof sinI sinR and expR on sinI expressionBecauseQSgenes sinI sinR and expR areneededforswarming tooccur,wetestedwhetherornottheycontributetoit independentlyorwhethertheyareallpartofoneregulatory hierarchy.Toaddressthisquestion,theactivityofpMG309(the plasmidcarryingthe sinI-gfp reporter)wastestedincolonies formedbythewildtypestrainRm8530anditsisogenic sinI mutantMG32, sinR mutantMG170andthe expR mutantRm1021 afteratwo-dayincubationon0.4%agar(Fig.8A).Thedisruption ofeither expR or sinI reducedtheexpressionof sinI by approximately3-fold,whilethedeletionof sinR hadthemost severeeffect.Thisisconsistentwiththeobservationsinliquid media[44].Theseobservationsalsomatchwiththephenotypesof thecorrespondingmutants: sinR mutantsareunabletoproduce SinIAHLs,while expR -defectivestrain S.meliloti 1021produces AHLs,althoughinlowerquantities[21,22].TheeffectofabroadrangeofAHLson sinI gene expressionBecausespecificC16:1-andoxo-C16:1-AHLsrestoredswarming ofthe sinI and sinR mutants(Fig.7),wetestedwhetherornotthisis duetothetwospecificAHLsstimulatedexpressionof sinI genein bacteiraonthesoftagar.FluorescenceofpMG309inthe sinI backgroundwasmeasuredafter2daysofincubationonagar Figure5.ExpressionofsinI.A. AverageGUSactivityofthechromosomal sinI-gusA reporterwasmeasuredinthewildtypebackground(MG301, filledcircles), sinI (MG302,filledtriangles),andMG302with150nMofC16:1-AHL(opentriangles). B. AverageGFPactivity(fluorescence/OD595)ofthe sinI-gfp fusionreporterpMG309inthewildtype(filleddiamonds),inMG32(filledsquares),andinMG32with150nMofC16:1-AHL(opensquares). FluorescenceoftheplasmidpVO190(whichcarriespromoterless gfp )inMG32isshownaslinewithastar.Averagesofthreetechnicalreplicationsare shown.Errorbarspresentstandarddeviation.Forbothassays,bacteriawerecollectedfromsoftagarsurfaces.Platescontainedeither150nMof C16:1-AHL(opensymbols)orsolventonly(methanol)(filledsymbols). doi:10.1371/journal.pone.0042611.g005 Figure6.ExpressionofthewggRgene.A. AverageGUSactivityofthe wggR-gusA merodiploidreporterinthewildtypebackground(MG305, filledcircles), sinI (MG306(filledtriangles)orinMG306with150nMofC16:1-HSL(opentriangles)inbacteriagrownonsurfacesofthesoftagar. Averagesofthreebiologicalreplicationswithinarepresentativeexperimentareshown,errorbarsarestandarddeviation. B. AverageGFPactivity [fluorescence/OD595]of wggR-gfp fusionreporterplasmidpMG310inRm8530(filleddiamonds),inMG32(filledsquares,shortdashes),andinMG32 with150nMofC16:1-HSL(opensquared).BackgroundfluorescenceofthepVO190vectorinMG32isshownasadashedlinewithstars. doi:10.1371/journal.pone.0042611.g006 RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org5August2012|Volume7|Issue8|e42611

PAGE 6

containingdifferentAHLs(Fig.8B).C14-,3-oxo-C14-andC16:1HSLsinduced sinI geneexpressionwhensuppliedat0.15–15nM andhigherconcentrations;3-oxo-C16:1wasactiveat0.075– 0.15nMandhigherconcentrations(Fig.8B).C8-(at22mM),C12(at18mM),andC16-HSL(at3.5mM)inducedthe sinI reporter activitybyapproximatelytwofold.Thus, sinI appearstorespond toabroadrangeofAHLswith3-oxo-C16:1-HSLbeingactiveat thelowestconcentrations.SpecificSinIAHLsstimulatetheexpressionof wggRToaddressthequestionofwhetherspecificSinIAHLsstimulate theexpressionofthe wggR gene,wefirstmeasuredandcompared theactivityoftheplasmidpMG310(carryingthe wggR-gfp reporter)inthewildtypeandthe sinI mutantgrownonsoftagar. Theactivityofthe wggR gfp reporterinthe sinI mutantwasmore than10timeslowerthanthatinthewildtype(Fig.9A).Next, responsesofthe wggR-gfp reportertodifferentAHLswere measuredinthe sinI mutant.AsshowninFig.9B,the wggR-gfp reporterwasunresponsivetoC14-HSLandonlyrespondedweakly to3-oxo-C14-HSLatthetwohighestconcentrations(750nMand 1500nM).Theactivityofthe wggR gfp reporterincreasedstrongly upontheadditionofC16:1-HSLand3-oxo-C16:1-HSL,anditdid soinadose-dependentfashion(Fig.9B).TheamountofC16:1HSLthatelicitedfullresponsivenessofthereporterwas approximately5-foldlowerthanthatof3-oxo-C16:1-HSL (Fig.9B).NeitherC14-,nor3-oxo-C14-HSLrestoredtheswarming phenotypeofthe sinI mutantstrain(Fig.9C).TheC16:1-HSLand oxo-C16:1-HSLstimulatedtheactivityof wggR-gfp reporterinthe sinI mutantcorrelatedwiththeextentandtheappearanceof swarming(Fig.9C).C8-(at22mM),C12-HSL(at18mM),and C16-HSL(at3.5mM)didnotinducethe wggR-gfp reporter.These datastronglysuggestthatspecificC16:1-andoxo-C16:1-AHL signalsstimulatestheexpressionof wggR geneinthe sinI mutantto enhancetheregulationofEPSIIproductionthatpromotedthe surfacemovement. TheseobservationsareconsistentwiththedatashowninFig.7 andfurthersuggestthateventhoughthe sinI promoteris responsivetoafairlybroadrangeofAHLs(Fig.8B),thesynthesis ofEPSIIistightlyregulatedbyWggR,andtheWggRregulation onEPSIIproductionistightlycontrolledbythetwospecificSinI AHLsC16:1-andoxo-C16:1-HSLthroughWggR. Figure7.SpecificAHLsinduceswarmingcolonyformation. Colonyspreadof S.meliloti Rm8530(wildtype),MG32( sinI ),MG170 ( sinR ),Rm1021( expR ),Rm9034( wggR ),Rm9030-2( wgaA ),andRm9032 ( wgdA )on0.4%agarcontainng1/20LBmedium(toprow).Coloniesof MG32,MG170,Rm1021, wggR wgaA wgdA onthemediumcontaining 200nMofsyntheticC16:1-HSL(secondrow),3-oxo-C16:1-HSL(thirdrow), C14-HSL(fourthrow)and3-oxo-C14-HSL(bottomrow).Inoculatedplates wereincubatedat30 u Cfor3daysbeforebeingphotographed. Genotypesofthestrainsareonthetopofthefigure;strainnamesare onthebottomofthefigure. doi:10.1371/journal.pone.0042611.g007 Figure8.ResponsesofthesinIreportertoAHLsignals.A. Activityofthe sinI-gfp reporterpMG309in S.meliloti 8530(wildtype),MG32( sinI ), Rm1021( expR )andMG170( sinR ).Adashedwhitelineindicatesaveragefluorescence/OD595(41717 6 2380units)fromRm8530andmutantscarrying pVO190vector. B. AverageactivityofpMG309inthe sinI mutantMG32grownon0.4%agarand20-fold-dilutedLBmediumplatescontainingC16:1-, 3-oxo-C16:1-,C14-and3-oxo-C14-HSLsattheindicatedconcentrations(orequivalentamountsofthemethanolsolvent).Bacteriawerecollectedfrom thespreadingcoloniesthatformedaftertwodaysofincubationonsoftagar.Averagesfromthreetechnicalreplicationswithinarepresentative experimentareshown,errorbarsarestandarddeviations.Theexperimentwasrepeatedtwicewithreproducibleresults. doi:10.1371/journal.pone.0042611.g008 RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org6August2012|Volume7|Issue8|e42611

PAGE 7

CrudeEPSIIanditsHMWfractionspartiallyrescue spreadingphenotypesofAHL-andEPSII-defective mutantsTotestwhetherEPSIIisresponsibleforfacilitatingthe bacterialspreadingoverthesoftsurfaces,EPSIIwascollected fromswarmingcoloniesformedbywildtypestrainRm8530, centrifuged,filter-sterilized,andthensize-fractioned.Crude extracellularmatrixharvestedfromswarmingcoloniesformed bywildtypepartiallyrestoredtheswarmingdefectphenotypesof theEPSIImutantsRm9034( wggR ),Rm9030-2( wgaA ),and Rm9032( wgdA )tothestageIofswarm(Fig.10,seealsoFig.1for themorphologyofanormalswarmingcolonyatstage1).The samephenotypeswereobservedfortheQSmutantsMG170 ( sinR ),Rm1021( expR )andMG32( sinI )suggestingthattheir contributionstoswarminginvolvecontrollingtheproductionof EPSII(Fig.9).CrudeEPSisolatedfromthe2–3dayoldcolonies ofthe wgaA mutantgrownonsoftagardidnotrestorethe swarmingofanymutantstestedinFig.10(datanotshown), indicatingthatothersurfacepolysaccharidesdonotplayarolein thistypeofswarming.Thisisexpected,becausenoneofEPSII mutantswasabletoswarmasshowninFig.10andtheEPSI mutantformednormalswarmingcolonies(Fig.3). BecauseEPSIIproducedbyRm8530isknowntoexistas LMWandHMWfractions,wesize-fractionatedthecollected extracellularmatrix(EPSII)totestwhichcomponenthadthe greatesteffectonswarming.AsshowninFig.10,theLMW fractionhadnoeffectonswarming.TheHMWfractionpartially restoredtheswarminginthemutantsdefectiveintheAHL synthesisandperception( sinI,sinRandexpR ),andinthemutants defectiveinEPSIIsynthesis( wggR,wgaA,wgdA ).Theextentof complementationbytheHMWfractionwasthesameaswiththe crudeEPSII.ThissuggeststhattheHMWfractionofEPSII producedby S.meliloti intheExpR-dependentmannercontributes totheearlystagesofswarming.Thisisthefirstreportedfunction forHMWEPSII.Expressionof wggR insubpopulationsofRm8530 swarmingcellsThemicroscopeimaging(Fig.1)showedpatternsofuneven distributionofpopulationinsideRm8530swarmingcoloniesand suggeststhatproductionoruptakeofAHLsignalsbyindividual cellsmayoccurdifferentlyunderthoseconditions.Therefore,we investigatedwhetherornot wggR genewasdifferentiallyexpressed amongindividualcellsundersimilarconditions.Forthispurpose, aresolvase-based invivo expressiontechnology(RIVET)[46]was usedtodetect wggR promoteractivityofindividualcellswithin coloniesonsoftagar. TheRIVETmethod[46]isbasedontheabilityofTnpR recombinasetocatalyzesite-specific‘‘resolution’’at res sequences. When tnpR isexpressedfromapromoterofinterest,activationof thispromoterdrivestheexpressionoftherecombinasegene. TnpRthenexcisesaselectabletetracyclineresistancegene.This resolutioneventcausesthelossofthetetracyclinemarkerand generatestetracyclinesensitiveprogeny.Therefore,theresolution Figure9.ResponsesofwggRtoAHLs.A. Averageactivity[fluorescence/OD595]ofthe wggR-gfp reporterpMG310in S.meliloti Rm8530(wild type),MG32( sinI )andMG32pVO190(vectorcontrol). B. Averageactivityof wggR-gfp (onpMG310)inMG32grownon0.4%agarand20-fold-diluted LBmediumplateswithdilutionsofAHLs(fromlefttoright)C16:1-,3-oxo-C16:1-,C14-,andoxo-C14-HSL.FluorescenceoftheMG32pMG310reporterin thenegativecontrol(solventonly)was7,995 6 646.Fortheassays,bacteriawerecollectedfromcoloniesafter3daysofincubationonsoftagarplates. Averageofthreebiologicalreplicaswithinarepresentativeexperimentareshown.Allessayswererepeatedatleasttwicewithreproducibleresult s. Errorbarsdenotestandarddeviations. C. FluorescentimagesofcoloniesformedbyMG32pMG310onsoftagarcontainingdifferentkindsand amountsofAHLs(asindicatedatthebottomoftheimage). doi:10.1371/journal.pone.0042611.g009 RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org7August2012|Volume7|Issue8|e42611

PAGE 8

ofRIVETreporterbacterialcells,calculatedaspercentageofthe tetracycline-sensitivecoloniesoverthetotal,areusedtoindirectly measuretheactivityofthepromoterthatdrivestheexpressionof tnpR [46]. S.meliloti MG102[47]isa wggR RIVETreporterof Rm8530strain.Itharborsachromosomalintegrated wggR-tnpR resolvasegenefusionanda res-tet-res cassetteinsertedinaneutral siteofRm8530chromosome[47].Similarly,MG103isa wggR RIVETreporterofthe sinI mutantstrain.MG103reporterwas constructedbyintroducingthe wggR-tnpR genefusionandthe restet-res cassetteintothechromosomeofthe sinI mutantusingthe protocolpreviouslydescribed[47] AverageresolutionofMG102was28 6 3%intriplicate swarmingcoloniesatitsthirddayofswarmingonsoftagar.The averageresolutionofMG103reporterwaslowinnon-swarming coloniesundersimilarconditions.A3-day-longgrowthofMG103 onsoftagarcontaining15and150nMofC16:1-HSLstimulated 10 6 5%and30 6 3%resolutionsoftheMG103reporter, respectively,comparedwith3.3 6 1.5%resolutionofthereporter grownonsoftagarwithoutC16:1-HSL. Theseresultsindicatethatapproximately1/3individualcellsin theswarmingcoloniesincreasedtheexpressionoftheir wggR gene inresponsetoC16:1-HSLsignalsandtherestofthemdidnot. Thus,itappearsthatSinIC16:1-HSLenhancedtheexpressionof the wggR insubpopulationstoupregulatetheirEPSIIproduction, andthiswassufficienttosupportapopulation-wideswarming. Howdidcellsinwhichtheexpressionof wggR remainedatlow levelscontributetoswarmofRm8530remainunknown.Spreadingofcoloniesinthepresenceof‘‘cheater’’ mutantsThediscoverythatonlyasubpopulationofthecellswithina colonycontributedtotheproductionoftheextracellularEPSIIin theQS-dependentmannerraisesimportantquestionsaboutthe population-wideeffectsofQS,resourceallocationswithinbacterial communities,andtheroleof‘‘cheaters’’[17].Fortheco-spreading assays,EPSIIandQSmutantswereused.AsshowninFig.11A, theaveragediameterofco-spreadingcoloniesformedbymixing thewildtypeandthe expR mutant(8:2)was15%lessthanthatof thewildtype.Diametersoftheco-spreadingcolonieswerefurther reducedwiththeincreasedproportionofthe expR mutant (Fig.11A).Similarresultswereobtainedwiththespreadingcoculturesofthe wggR and wgaA mutants(datanotshown).Mixed spreadingcoloniesformedbythe sinI mutantMG32andRm8530 wereidenticaltothoseofthewildtypecoloniesevenwhenthe mutantmadeupthemajorityofthepopulation(Fig.11B).These resultsarereminiscentofthoseobtainedin P.aeruginosa wherethe ‘‘publicgoods’’cheatersweremoredetrimentaltothecolonythan ‘‘signalcheaters’’[17].Discussion SwarmingbehaviorSotoetal.[7]firstobservedsurfaceswarmingin S.meliloti .Their G4WTstraindidnotswarmundertheconditionstested,buta fadD mutantdid.Ourresultsshowthat S.meliloti Rm8530strain canswarmonverysoftagar(0.4%).The fad mutantswarmingcells werehyperflagellated,andtheystoppedtheirpropagationin swarmingcolonies[7].Rm8530swarmingcellswerenot Figure10.ContributionsofEPSIItocolonyspreading. Surfacespreadingof S.meliloti Rm8530(wildtype),MG32( sinI ),MG170( sinR ),Rm1021 ( expR ),Rm9034( wggR ),Rm9030-2( wgaA ),andRm9032( wgdA )on0.4%agarand20-fold-dilutedLBmedium(toprow).CrudeEPS(harvestedfromthe spreadingcoloniesofRm8530)oritshighorlowmolecularweightfractions(20ml)werespottedontotheplatesurfaces,intowhich3.5mloftested bacteriawereadded.TheamountofsuppliedEPSIIrepresents 1/2–1/7oftheamountofEPSIIthatisproducedbya3-dayoldsingleswarming colony.Plateswereincubatedfor2–3daysat30 u Cbeforephotographing. doi:10.1371/journal.pone.0042611.g010 Figure11.Co-spreadingofS.meliloti8530andmutants. SwarmingcoloniesformedbythemixtureofRm8530(wildtype)and Rm1021( expR )(left),orbythemixtureofRm8530andMG32( sinI ) (right).Surfaceofsoftagarwereinoculatedwithmixedinoculacontain amutantand80%,50%,30%and10%ofRm8530.Asacontrol,pure culturesofRm8530andmutantswerespottedseparatelyonthesame plate.Photosweretakenaftertwodaysinoculation. doi:10.1371/journal.pone.0042611.g011 RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org8August2012|Volume7|Issue8|e42611

PAGE 9

hyperflagellated(Figure2A),andtheydidnotstoptheir propagationinswarmingcolonies(Figure.S1).Socialand cooperativebehaviorsareknowntooccurinswarmingcolonies inotherbacteria[14,15],andtheswarmingofRm8530was dependentoncontrolledsecretionofEPSII,consistentwith involvementofsocialorganizationinswarmingcolonies.Dual motilitysystems(A-motilityandS-motility)insoilbacterium Myxococcusxanthus werereported[48],andthosemotilitiesshow differentselectiveadvantagesonvarioussurfaces[48].Swarming ofRm8530studiedhereisoneofafewmotilityphenomena describedin S.meliloti sofar[7,9,27],andlikelytohelpthebacteria toadaptcomplexsurfaceenvironments.RegulationofswarmingbehaviorTheneedforExpR/SinQSsystemtoinitiateswarmingcolony in S.meliloti Rm8530seemstoberestrictedtogeneratingAHL signalsperceivingtheAHLsignals,andregulatingtheEPSII production(Fig.4andFig.7).TheabilityofSinIC16:1-HSLand oxo-C16:1-HSLtostimulateswarmingofthe sinI mutantandthe sinR mutantbutnotthe expR mutant,indicatesthatswarming colonyinitiationonthesoftsurfaceinvolvesthesespecificSinI AHLsactingassignalsmediatedbytheExpRreceptor.The inabilityofC16:1-HSLandoxo-C16:1-HSLtostimulateswarming ofthe wggR mutantindicatesthattheseAHLscontributelittleto swarmintheabsenceofWggR.Thus,itappearsthat sinI made C16:1-HSLand3-oxo-C16:1-HSLactivatetheExpRreceptorand thisdirectlyorindirectlyenhancesexpressionof wggR and contributestotheregulationoftheproductionofEPSII.This relationshipisconsistentwithearliertranscriptionalstudies [24,26,27,33].Inadditiontoenhancingtheexpressionof wggR C16:1-HSLwasshowntorestoretheexpressionofotherEPSII genesatthepresenceofExpR[26].Currentdatashowthat positiveregulationofEPSIIgenesbyExpRisdependenton WggR[27].Theseexplainwhyanoverexpressed wggR is incapabletostimulateswarminginthe sinI mutant(Figure4B). TheinabilityofC14–HSLandoxo-C14:1–HSLtostimulate swarmingofthe sinI ,the sinR ,the expR ,orthe wggR mutants (Figure7)suggeststhatthoseSinIAHLsnormallydonotactas signalsfortheinitiationofRm8530swarming.TheinabilityofC14andoxo-C14-AHLtostimulate wggR promoterinthepresenceof ExpR(Fig.9B)stronglysupportsourfindingthattheexpressionof wggR isspecificallystimulatedbytheC16:1-andoxo-C16:1-HSLs activatedExpR.Interestingly,earliergelshiftassaysshowedthat oxo-C14:1–HSLdidnotenhancetherelationshipbetweenExpR andthe wggR promoter[27].Theinabilityofoverexpressed sinI and sinR genes(Figure4B)andsyntheticAHLs(Figure7)to stimulateswarmingofEPSIImutantsindicatethatSinIAHLsdo notnormallyactassurfactantsand/orwettingagentsinRm8530 swarmingcells. Thelevelsof sinI expressionina sinIexpR doublemutantwere notsignificantlyaffectedbytheadditionofC16:1-HSLoroxoC16:1-HSLaddedintosoftagar(FigureS2B).Theseresultssupport ourconclusionthattheinitiationofswarmingdependsonthe interactionoftheseAHLswithExpR.Thelevelsof sinI expression inthe sinIexpR doublemutantweresignificantlyincreasedbythe additionofotherAHLs(oxo-C14-HSL,C12-HSLandC8-HSL) (Fig.S2A),raisingthequestionofwhetherornotthesesignal moleculesinteractwiththeSinRproteinorotherpredictedLuxRlikeproteinstoaffecttheexpressionof sinI .Theanswertothat questionremainsunknown. EPSIIissecretedintwomajorfractions:HMWandLMW.We haveshownthattheHMWfractionfacilitatedtheinitialstagesof swarmingandthatLMWfractionisnotcriticalforfacilitatingthe initialstagesofswarming(Figure10). Thisstudydemonstratesthatswarmingisasocialbehaviorthat canbeencouragedordiscouragedbychangesinQSsignaling inputandtheregulationingeneexpression.Whiletheinfluenceof QSonswarmingisstudiedintheaspectof wggR ,otherregulatory geneproductsmayalsocontributetothebehaviorthroughtheir effectsontheproductionofEPSIIandmotilitygenes.For example,MucR,aRosRhomolog,isapositiveregulatorofEPSI geneandanegativeregulatorofEPSIIgenesincluding wggR [35]. MucRmutantsproduceHMWEPSIIexclusively[30].ExpR/Sin QSsystemincreasesexpressionofthe wggR .WggRderepresses EPSIIproductionatthetranscriptionallevelfromMucR,while concurrentlyelevatingtheexpressionof wgeA ,resultinginthe synthesisoftheLMWfraction[33].TheroleofMucRin controllingswarmingofRm8530remainstobeinvestigated.MaterialsandMethods MediaandcultureconditionsForroutinepropagation,strainsof S.meliloti weregrownat30 u C inTYbroth[49]. E.coli wasculturedinLuria-Bertanibroth (FisherScientific,FairLawn,NewJersey07410,U.S.A.)at37 u C. Asneeded,mediaweresupplementedwithantibioticsatthe followingfinalconcentrations:streptomycin,250–500mg/ml; neomycin,100mg/ml;tetracycline,2.5–5mg/ml;gentamicin, 50mg/ml;kanamycin,25mg/ml.C16:1D9cis -(L)-homoserinelactone(referredtoas‘‘C16:1-HSL’’intext),3-oxo-C16:1D11cis -(L)homoserinelactone(referredtoas‘‘3-oxo-C16:1-HSL’’intext),3oxo-C14-andC14-homoserinelactoneswerefromCayman Chemical(EllsworthRoad,AnnArbor,MI,U.S.A.).OtherAHLs werefromSigma-Aldrich(St.Louis,MO,U.S.A.).StrainsandplasmidsStrainsandplasmidsusedinthisstudyarelistedinTable1. PrimersarelistedinTable2.Forcomplementationandepistasis studies,theintactORFsincludingpredictedregulatoryregions werePCRamplifiedfromgenomesof S.meliloti Rm8530or Rm1021withthefollowingprimers:forp sinI ,primersMG460and MG461;forpKB expR ,MT10andMT11;forpKY wggR ,DC3and DC10;forp sinR ,MG1866andMG18677.Theamplified fragmentswereinitiallyclonedintopCR2.1(Invitrogen,Carlsbad, CA,U.S.A.),andthensequenced.Onceconfirmed,fragments werereleasedfrompCR2.1withthefollowingrestriction endonucleases: EcoR I(forthe sinI ,the expR ,andthe sinR fragments), Spe I, Xba I(forthe wggR fragmnet)andthencloned intopBBRMCS1-basedvectors.[50].Plasmidp expR -kmwas constructedbyreleasingthe EcoR IfragmentfrompKB expR and insertingitinto EcoR IsiteofpBBR1-MCS2(Km).Finalconstructs wereconfirmedbydiagnosticrestrictiondigestsandsequencing. Validatedconstructsweremobilizedfrom E.coli intorhizobiaby tri-parentalconjugationasprevioulydescribed[51]. Plasmid-bornepromotergfp reporterfusionswereconstructed byPCRamplifyingagenomicregioncontainingdesired promoterswiththefollowingprimers:forpMG309( sinI-gfp ), MG43andMG44;forpMG310( wggR-gfp ),MG45andMG46. ThePCRfragmentswereclonedinto Kpn I/ Xho Isitesinfrontof thepromoterless gfp geneinthebroad-host-rangepVO190 plasmid(agiftfromDr.ValerieOke).Resultingplasmids,after sequencing,wereintroducedinto S.meliloti bytri-parental conjugationsasdescribed[51]. Tomakechromosomaltranscriptionalfusions,chromosomal integrativeplasmidscarrying S.meliloti DNAwereintroducedinto proper S.meliloti strainsasdescribed[22].PlasmidpVO3TnpR wasusedtomake S.melilotiwggR RIVETreporterstrainMG103; pVMG sinI 7wasusedtomake S.melilotisinI-gusA reporterstrainsRegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org9August2012|Volume7|Issue8|e42611

PAGE 10

MG301andMG302;pVO3wasusedtomake S.melilotiwggR-gusA reporterstrainsMG305andMG306. MG320mutant(Rm8530 fliP )wascreatedbytransducingthe fliP ::KmmutationfromRmG910[41]intoRm8530usingM12 aspreviouslydescribed[41].Themutantisolatewasbackcrossed onetimetoRm1021.SwimmingmotilitydefectsofMG320were confirmedbytheswimmingassayusinganestablishedprotocol [52].Constructionof S.meliloti MG170( D sinR )Tocreatethe sinR deletionmutantMG170,twoDNAregions flanking sinR genewerePCRamplifiedfromthegenomicDNAof S.meliloti Rm1021usingprimerpairsMG497andMG1751;and MG496andMG1752,respectively.ThetwoPCRfragmentswere digestedwith AvrII ,purified,ligatedtoeachother.Thiscreateda 1.9-kbfragmentof sinR deletionmutant(thedeletionwasfrom200 to670ntwithin sinR openreadingframe).Thefragmentof sinR mutantwasPCRamplifiedusingprimersMG497andMG496, clonedintopCR2.1(Invitrogen)andconfirmedbysequencing. Table1. Strainsandplasmidsusedinthisstudy.StrainorplasmidRelevantcharacteristic(s) Reference Plasmids pBBR1MCSBroad-host-rangecloningvectors [56] pDG71ConstitutiveP trp -Gfpmut3,Tc [38] pJQ200SK sacB suicidevector,Gm [50] pRK600Conjugaltransferhelperplasmid,Cm [57] pTH113pRK7813,hasanintact sinRandsinI ,Tc [45] pVMGpUC119derivative,promoterless gus withupstreamstopcodons,Nm[22] pVMG sinI7 pVMGderivative, sinI-gus ,Nm [47] pVO3pVMGderivative, wggR-gus ,Nm [47] pVO3TnpRpVO3derivative, wggR-tnpR-gus ,Nm [47] pVO190pBBRderivativecontainingapromoterlessgfpmut1,SpOke,Valerie pKB expR pBBR1MC-S5containinga1.3-kb Eco R1fragmentof expR ,GmThiswork pKY wggR pBBR1MC-S3containinga1.1-kb Spe IXba Ifragmentof wggR ,TcThiswork pMG307pJQ200SKcontaininga1.9-kb Spe IApa Ifragmetof sinR mutant, Avr II,GmThiswork pMG309pVO190containinga348-bp Kpn IXho Ifragmentof sinI promoterregionThiswork pMG310pVO190containinga764-bp Kpn IXho Ifragmentof wggR promoterregionThiswork p sinI pBBR1MC-S5containinga2.2-kb Eco RIfragmentof sinI ,GmThiswork p sinR pBBR1MC-S5containinga1.9-kb Eco RIfragmentof sinR ,GmThisworkS.melilotiRm1021SU47, expR 102::IS Rm 2011-1,Sm [58] Rm8530Rm1021, expR + Sm [24] Rm9030-2Rm8530, wgaA ( expA1 ):: lacZ -Gm,Sm,Gm[24] Rm9032Rm8530, wgdA ( expD1 ):: lacZ -Gm,Sm,Gm[24] Rm9034Rm8530, wggR ( expG )::lacZ-Gm,Sm,Gm [24] Rm11601Rm8530, flaAflaB ,Sm,Hy [40] Rm11603Rm8530, exoY ,Sm, [6] RmG910Rm1021, fliP ::kan,Sm,Km [41] MG32Rm8530, D sinI ,Sm [22] MG32rtrMG32,withintegrated resI-tet-resI cassette,Sm,Tc[47] MG75Rm1021, D sinI ,Sm [22] MG102Rm8530 wggR + wggR-tnpR res1-tet-resI ,Sm,Nm,Tc[47] MG103MG32rtr,withintegratedpVO3TnpR,Sm,Nm,Tc[47] MG170Rm8530, D sinR ,Sm Thiswork MG301Rm8530,withintegratedpVMGsinI7, sinI-gusA ,Sm,NmThiswork MG302MG32,withintegratedpVMGsinI7, sinI-gusA ,Sm,NmThiswork MG305Rm8530,withintegratedpVO3, wggR-gusA ,Sm,NmThiswork MG306MG32,withintegratedpVO3, wggR-gusA ,Sm,NmThiswork MG320Rm8530, fliP ::kan,Sm,Km Thiswork GUS:-glucuronidase;Sm,Sp,Km,Nm,Tc,Gm,Hy:resistanttospectinomycin,streptomycin,kanamycin,neomycin,gentamicin,hygromycinrespecti vely. doi:10.1371/journal.pone.0042611.t001 RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org10August2012|Volume7|Issue8|e42611

PAGE 11

Thefragmentofconfirmed sinR mutantwasexcisedasa1.9-kb Spe I/ Apa IfragmentandclonedintothesuicidevectorpJQ200SK [50],yieldingpMG307.pMG307wasmatedinto S.meliloti Rm8530withhelperplasmidpRK600[53].Gentamicinresistant transconjugantscontainingasinglecrossoverin sinR wasselected. Sucrosecounterselectionfordoublerecombinantswasperformed using5%sucroseaspreviouslydescribed[22,50].Boththe presenceofthe sinR deletionandtheabsenceofwildtype sinR in MG170mutantwasconfirmedbyPCRandsequencing.The defectintheAHLproductionofMG170wasconfirmedbya bioassay-coupledthin-layerchromatography(TLC)methodusing anestablishedprotocal[22].Swarmconditionsandthecollectionofextracellular polymersSwarmconditions.Log-phasecultures(OD600=0.4to0.8) ofbacteria,growninTYbroth(30 u C,225rpm)werecentrifuged andthepelletswerere-suspendedintheoriginalvolumeinsterile water.Typically,3.5mlofbacterialsuspensionswerespottedonto thesurfaceof1/20LBsolidifiedwith0.4%MolecularGenetics gradeagar(FisherScientific).Priortotheinoculation,plateswere cooledfor10minwithlidopenonasterileflowbench.AHLs wereaddedtoagarplates.Plateswereincubatedat30 u Cinan uprightposition,andtheappearanceofcolonieswasobserved daily.EPScollection.Extracellularmatrixwascollectedfromthe edgesofthespreadingcoloniesin S.meliloti 8530usingblunt-ended pipette( 100–150mlpercolony)andtransferredintoeppendorf tubes,spunattopspeedandthenthesupernatantwasspin-filtered usingcentrifugalnylonfilters(poresize0.22mm)toremove bacteria.Thebacteria-freepreparationwasfurtherfractionized intoHMWEPSandLMWEPSwithaMilliporeUltraFreeMC 30,000NMWLFilterUnits(MilliporeCorporation,Bedford,MA) followingthemanufacturer’sprotocol.Aliquotsofthematerial wereroutinelycheckedfortheabsenceofbacteriabyplatingonto TYagar.EPSfromtheEPSIImutantRm9030-2( wgaA )was collectedfromaplatewith40drycoloniesformedbyRm9030-2 bacteriaonsurfaceafter2daysofincubation.Thiswasdoneby floodingtheplatewith2mlofsterilewater,lettingtheplatestand for2minutes,andusingblunt-endedpipettetotransferaqueous suspensionscontainingEPSfromthemutant,andthenfollowing thesameprocessdescribedabovetopreparebacteria-freeEPS. 20mlofthepreparationwasusedforexperiments.b -Glucuronidase(GUS)assaysCellsgrownonthesoftagarwerecollectedfromthesurfacesat theindicatedtimesbyfloodingplateswith1.5mlofwaterfor 1minuteandthencarefullyaspiratingbacterialsuspensionsusing anestablishedprotocol[54].Afteropticaldensityofthesamples wasmeasuredat595nm(OD595),cellswerepermeabilizedwith lysozyme(200mgml2 1,37Cfor10min),andtheactivitywas assayedwithp-nitrophenylb o -glucuronide.GUSactivityis presentedinnanomolesperminuteperOD595unit 6 1,000as in[55].GFPmeasurementCellswerecollectedfromagarsurfacesusingthesamewayasin theGUSassayandsuspendedinwater.Quantitativegreen fluorescencewasmeasuredas[Fluorescence(1.0s)(Counts)/ OD595]in96-wellmicroliterplatesusingWallac1420,Multilabel counterwiththefiltersetforfluorescentmeasurement(485-nm bandpassexcitationfilteranda535-nmbarrierfilter)(PerkinElmer LifeandAnalyticalScience,WallacOy,P.O.Box10,FIN-20101 Turku,Finland).ImagingPicturesofspreadingcoloniesweretakenwithageldocimager orwithaCanonEOSRebelXSIcamera.Imageswereacquired fromOlympusMVx10dissectingscopeequippedwitha gfp filter) withMicroFirscamera(Optitronics,Goleta,CA,USA).Images werethentransferredintoAdobePhotoshopCS,brightnessand colorbalancewerenormalizedusingdefaultautomaticsettings. Multi-panelimageswereassembledinAdobePhotoshopCS. Table2. Primersusedforcloning.PrimerNameSequencePurpose DC3TTGGGGCCCTTGCTAATCAAAGGAConstructionofpKY wggR DC10ACGAATGCTACATGCATCConstructionofpKY wggR MG43ggggtACCGGGCCGGAAACGGAGGConstructionofpMG309 MG44ccgctcgAgTTTTTCGCTCCATGCGConstructionofpMG309 MG45ggggtaccACGACGGAGATCGCConstructionofpMG310 MG46ccgctcgAgTGGGAACGTACTTCCAAConstructionofpMG310 MG460GAAGAAATCGGGCTTTCCACCGAConstructionofp sinI MG461CGTCGCGAGCACATGATAGTAGAGConstructionofp sinI MG497ACGATCGTGCGCACGAATACGAConstructionofpMG307 MG1751TAGATTTCGGCGGcCTaGGCGCCGAAAGTConstructionofpMG307 MG496ACATCGGGCGATCGAGAACGGConstructionofpMG307 MG1752ATATCctAGgGAACGGTGCGTTTCTTConstructionofpMG307 MG1866ACGATCGTGCGCACGAATtCGAConstructionofp sinR MG1867TGCGACCGgaTCCGTTCACTATConstructionofp sinR MT10TTTGCGTTCTTCCCAAAAAACGCGGTAConstructionofpKB expR MT11AATGAAGCGCAATTTCAGATGCGACATConstructionofpKB expR Lowercaselettersinoligonucleotidesequencesindicatenucleotidesthatweremodifiedfromthepublishedsequencetocreateenzymesites. doi:10.1371/journal.pone.0042611.t002 RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org11August2012|Volume7|Issue8|e42611

PAGE 12

SupportingInformationFigureS1Bacterialgrowthwithinspreadingcolonies. A. Spreadingcolonyformedby S.meliloti Rm8530(wildtype) containsmorecellsthannon-swarmingcoloniesformedby sinI sinR and expR mutants(basedonOD600measurements),implying benefitsforcolonygrowth. B. Colonyappearancesofwildtype andmutantbacteriaonagarsurfacefromwhichcellswere harvested. (TIF)FigureS2EffectofAHLson sinI expressionina sinI expR doublemutant.A. ComparisonofC16-,C12-andC8HSLinducedGFPactivityof sinI-gfp (pMG309)inMG32( sinI ) andMG75( sinIexpR ). B. AverageGFPactivityofthe sinI -GFP promoterplasmid(pMG309)inMG75( sinIexpR )mutantwithor withoutAHLsaddedintosoftagar.Bacteriawerefromcolonies aftertwodaysincubationonsoftagar.Eachdatapointisan averageofthreetechnicalreplicationsfromarepresentative experiment.Errorbarsarestandarddeviations. (TIF)AcknowledgmentsWeappreciateK.BhanforconstructingplasmidpKB expR .WethankN. FujishigeforRmG910.WethankJ.E.GonzalezforRm11601and Rm11603.WespeciallythankV.Okeforsharingunpublishedplasmid pVO190andhelpfuladvice.AuthorContributionsConceivedanddesignedtheexperiments:MGMT.Performedthe experiments:MGACKYMT.Analyzedthedata:MGMT.Wrotethe paper:MGMT.References1.LongSR(2001)Genesandsignalsinthe rhizobium -legumesymbiosis.Plant Physiol125:69–72. 2.GibsonKE,KobayashiH,WalkerGC(2008)Moleculardeterminantsofa symbioticchronicinfection.AnnuRevGenet42:413–441. 3.TianCF,GarneroneAM,Mathieu-DemaziereC,Masson-BoivinC,BatutJ (2012)Plant-activatedbacterialreceptoradenylatecyclasesmodulateepidermal infectioninthe Sinorhizobiummeliloti Medicago symbiosis.ProcNatlAcadSciUSA 109:6751–6756. 4.MathesiusU,MuldersS,GaoM,TeplitskiM,Caetano-AnollesG,etal.(2003) Extensiveandspecificresponsesofaeukaryotetobacterialquorum-sensing signals.ProcNatlAcadSciUSA100:1444–1449. 5.FujishigeNA,KapadiaNN,DeHoffPL,HirschAM(2006)Investigationsof Rhizobium biofilmformation.FEMSMicrobiolEcol56:195–206. 6.RinaudiLV,GonzalezJE(2009)Thelow-molecular-weightfractionof exopolysaccharideIIfrom Sinorhizobiummeliloti isacrucialdeterminantof biofilmformation.JBacteriol191:7216–7224. 7.SotoMJ,Fernandez-PascualM,SanjuanJ,OlivaresJ(2002)A fadD mutantof Sinorhizobiummeliloti showsmulticellularswarmingmigrationandisimpairedin nodulationefficiencyonalfalfaroots.MolMicrobiol43:371–382. 8.BahlawaneC,McIntoshM,KrolE,BeckerA(2008) Sinorhizobiummeliloti regulatorMucRcouplesexopolysaccharidesynthesisandmotility.MolPlant MicrobeInteract21:1498–1509. 9.NogalesJ,LydiaBernabe u-Roda,VirginiaCue llar,andMar aJ..Soto(2012) ExpRisnotrequiredforswarmingbutpromotesslidingin Sinorhizobiummeliloti JBacteriolpublishedonlineaheadofprinton10February2012. 10.HarsheyRM(2003)Bacterialmotilityonasurface:manywaystoacommon goal.AnnuRevMicrobiol57:249–273. 11.VerstraetenN,BraekenK,DebkumariB,FauvartM,FransaerJ,etal.(2008) Livingonasurface:swarmingandbiofilmformation.TrendsMicrobiol16: 496–506. 12.KearnsDB(2010)Afieldguidetobacterialswarmingmotility.NatRev Microbiol8:634–644. 13.SuleS,CursinoL,ZhengD,HochHC,BurrTJ(2009)Surfacemotilityand associatedsurfactantproductionin Agrobacteriumvitis .LettApplMicrobiol49: 596–601. 14.CrespiBJ(2001)Theevolutionofsocialbehaviorinmicroorganisms.Trends EcolEvol16:178–183. 15.XavierJB,KimW,FosterKR(2011)Amolecularmechanismthatstabilizes cooperativesecretionsin Pseudomonasaeruginosa .MolMicrobiol79:166–179. 16.NadellCD,FosterKR,XavierJB(2010)Emergenceofspatialstructureincell groupsandtheevolutionofcooperation.PLoSComputBiol6:e1000716. 17.VenturiV,BertaniI,KerenyiA,NetoteaS,PongorS(2010)Co-swarmingand localcollapse:quorumsensingconveysresiliencetobacterialcommunitiesby localizingcheatermutantsin Pseudomonasaeruginosa .PLoSOne5:e9998. 18.OchsnerUA,ReiserJ(1995)Autoinducer-mediatedregulationofrhamnolipid biosurfactantsynthesisin Pseudomonasaeruginosa .ProcNatlAcadSciUSA92: 6424–6428. 19.LindumPW,AnthoniU,ChristophersenC,EberlL,MolinS,etal.(1998)NAcyl-L-homoserinelactoneautoinducerscontrolproductionofanextracellular lipopeptidebiosurfactantrequiredforswarmingmotilityof Serratialiquefaciens MG1.JBacteriol180:6384–6388. 20.DanielsR,ReynaertS,HoekstraH,VerrethC,JanssensJ,etal.(2006)Quorum signalmoleculesasbiosurfactantsaffectingswarmingin Rhizobiumetli .ProcNatl AcadSciUSA103:14965–14970. 21.MarketonMM,GronquistMR,EberhardA,GonzalezJE(2002)Characterizationofthe SinorhizobiummelilotisinR/sinI locusandtheproductionofnovelNacylhomoserinelactones.JBacteriol184:5686–5695. 22.GaoM,ChenH,EberhardA,GronquistMR,RobinsonJB,etal.(2005) sinI and expR -dependentquorumsensingin Sinorhizobiummeliloti .JBacteriol187: 7931–7944. 23.GlazebrookJ,WalkerGC(1989)Anovelexopolysaccharidecanfunctionin placeofthecalcofluor-bindingexopolysaccharideinnodulationofalfalfaby Rhizobiummeliloti .Cell56:661–672. 24.PellockBJ,TeplitskiM,BoinayRP,BauerWD,WalkerGC(2002)ALuxR homologcontrolsproductionofsymbioticallyactiveextracellularpolysaccharide IIby Sinorhizobiummeliloti .JBacteriol184:5067–5076.25.LlamasI,KeshavanN,GonzalezJE(2004)Useof Sinorhizobiummeliloti asan indicatorforspecificdetectionoflong-chain N -acylhomoserinelactones.Appl EnvironMicrobiol70:3715–3723. 26.MarketonMM,GlennSA,EberhardA,GonzalezJE(2003)Quorumsensing controlsexopolysaccharideproductionin Sinorhizobiummeliloti .JBacteriol185: 325–331. 27.McIntoshM,KrolE,BeckerA(2008)Competitiveandcooperativeeffectsin quorum-sensing-regulatedgalactoglucanbiosynthesisin Sinorhizobiummeliloti JBacteriol190:5308–5317. 28.ZhanHJ,LeverySB,LeeCC,LeighJA(1989)Asecondexopolysaccharideof Rhizobiummeliloti strainSU47thatcanfunctioninrootnoduleinvasion.Proc NatlAcadSciUSA86:3055–3059. 29.DownieJA(2010)Therolesofextracellularproteins,polysaccharidesandsignals intheinteractionsof rhizobia withlegumeroots.FEMSMicrobiolRev34:150– 170. 30.GonzalezJE,ReuhsBL,WalkerGC(1996)LowmolecularweightEPSIIof Rhizobiummeliloti allowsnoduleinvasionin Medicagosativa .ProcNatlAcad SciUSA93:8636–8641. 31.SorrocheFG,RinaudiLV,ZorreguietaA,GiordanoW(2010)EPSIIdependentautoaggregationof Sinorhizobiummeliloti planktoniccells.Curr Microbiol61:465–470. 32.BeckerA,RubergS,KusterH,RoxlauAA,KellerM,etal.(1997)The32kilobaseexpgeneclusterof Rhizobiummeliloti directingthebiosynthesisof galactoglucan:geneticorganizationandpropertiesoftheencodedgene products.JBacteriol179:1375–1384. 33.MuellerK,GonzalezJE(2011)Complexregulationofsymbioticfunctionsis coordinatedbyMucRandquorumsensingin Sinorhizobiummeliloti .JBacteriol 193:485–496. 34.BaumgarthB,BartelsFW,AnselmettiD,BeckerA,RosR(2005)Detailed studiesofthebindingmechanismofthe Sinorhizobiummeliloti transcriptional activatorExpGtoDNA.Microbiology151:259–268. 35.BahlawaneC,BaumgarthB,SerraniaJ,RubergS,BeckerA(2008)Fine-tuning ofgalactoglucanbiosynthesisin Sinorhizobiummeliloti bydifferentialWggR (ExpG)-,PhoB-,andMucR-dependentregulationoftwopromoters.JBacteriol 190:3456–3466. 36.KellerM,RoxlauA,WengWM,SchmidtM,QuandtJ,etal.(1995)Molecular analysisofthe RhizobiummelilotimucR generegulatingthebiosynthesisofthe exopolysaccharidessuccinoglycanandgalactoglucan.MolPlantMicrobe Interact8:267–277. 37.JanczarekM(2011)Environmentalsignalsandregulatorypathwaysthat influenceexopolysaccharideproductionin rhizobia .IntJMolSci12:7898–7933. 38.GageDJ(2002)AnalysisofinfectionthreaddevelopmentusingGfp-andDsRedexpressing Sinorhizobiummeliloti .JBacteriol184:7042–7046. 39.GotzR,SchmittR(1987) Rhizobiummeliloti swimsbyunidirectional,intermittent rotationofright-handedflagellarhelices.JBacteriol169:3146–3150. 40.GurichN,GonzalezJE(2009)Roleofquorumsensingin Sinorhizobiummeliloti Alfalfasymbiosis.JBacteriol191:4372–4382. 41.FinanTM,GoughC,TruchetG(1995)Similaritybetweenthe Rhizobiummeliloti fliPgeneandpathogenicity-associatedgenesfromanimalandplantpathogens. Gene152:65–67.RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org12August2012|Volume7|Issue8|e42611

PAGE 13

42.MullerP,KellerM,WengWM,QuandtJ,ArnoldW,etal.(1993)Genetic analysisofthe Rhizobiummeliloti exoYFQoperon:ExoYishomologoustosugar transferasesandExoQrepresentsatransmembraneprotein.MolPlantMicrobe Interact6:55–65. 43.KhanSR,GainesJ,RoopRM,2nd,FarrandSK(2008)Broad-host-range expressionvectorswithtightlyregulatedpromotersandtheirusetoexaminethe influenceofTraRandTraMexpressiononTiplasmidquorumsensing.Appl EnvironMicrobiol74:5053–5062. 44.McIntoshM,MeyerS,BeckerA(2009)Novel Sinorhizobiummeliloti quorum sensingpositiveandnegativeregulatoryfeedbackmechanismsrespondto phosphateavailability.MolMicrobiol74:1238–1256. 45.DriscollBT,FinanTM(1997)PropertiesofNAD( + )-andNADP( + )-dependent malicenzymesof Rhizobium ( Sinorhizobium ) meliloti anddifferentialexpressionof theirgenesinnitrogen-fixingbacteroids.Microbiology143:489–498. 46.CamilliA,MekalanosJJ(1995)Useofrecombinasegenefusionstoidentify Vibrio cholerae genesinducedduringinfection.MolMicrobiol18:671–683. 47.GaoM,TeplitskiM(2008)RIVET-atoolforinvivoanalysisofsymbiotically relevantgeneexpressionin Sinorhizobiummeliloti .MolPlantMicrobeInteract21: 162–170. 48.ShiW,ZusmanDR(1993)Thetwomotilitysystemsof Myxococcusxanthus show differentselectiveadvantagesonvarioussurfaces.ProcNatlAcadSciUSA90: 3378–3382. 49.BeringerJE(1974)Rfactortransferin Rhizobiumleguminosarum .JGenMicrobiol 84:188–198. 50.QuandtJ,HynesMF(1993)Versatilesuicidevectorswhichallowdirectselection forgenereplacementingram-negativebacteria.Gene127:15–21. 51.GlazebrookJ,WalkerGC(1991)Genetictechniquesin Rhizobiummeliloti MethodsEnzymol204:398–418. 52.GaoM,BarnettMJ,LongSR,TeplitskiM(2010)Roleofthe Sinorhizobium meliloti globalregulatorHfqingeneregulationandsymbiosis.MolPlantMicrobe Interact23:355–365. 53.OkeV,LongSR(1999)Bacterialgenesinducedwithinthenoduleduringthe Rhizobium -legumesymbiosis.MolMicrobiol32:837–849. 54.WangQ,FryeJG,McClellandM,HarsheyRM(2004)Geneexpression patternsduringswarmingin Salmonellatyphimurium :genesspecifictosurface growthandputativenewmotilityandpathogenicitygenes.MolMicrobiol52: 169–187. 55.BittnerAN,OkeV(2006)MultiplegroESLoperonsarenotkeytargetsof RpoH1andRpoH2in Sinorhizobiummeliloti .JBacteriol188:3507–3515. 56.KovachME,ElzerPH,HillDS,RobertsonGT,FarrisMA,etal.(1995)Four newderivativesofthebroad-host-rangecloningvectorpBBR1MCS,carrying differentantibiotic-resistancecassettes.Gene166:175–176. 57.FinanTM,KunkelB,DeVosGF,SignerER(1986)Secondsymbiotic megaplasmidin Rhizobiummeliloti carryingexopolysaccharideandthiamine synthesisgenes.JBacteriol167:66–72. 58.GalibertF,FinanTM,LongSR,PuhlerA,AbolaP,etal.(2001)Thecomposite genomeofthelegumesymbiont Sinorhizobiummeliloti .Science293:668–672.RegulationofSurfaceSpreadingin S.meliloti PLOSONE|www.plosone.org13August2012|Volume7|Issue8|e42611