grownkkk755.com

ARTICLENaturalvariationatXND1impactsroothydraulicsandtrade-offforstressresponsesinArabidopsisNingTang1,ZaighamShahzad1,4,FabienLonjon2,OlivierLoudet3,FabienneVailleau2&ChristopheMaurel1Soilwateruptakebyrootsisakeycomponentofplantperformanceandadaptationtoadverseenvironments.
Here,weuseagenome-wideassociationanalysistoidentifytheXYLEMNACDOMAIN1(XND1)transcriptionfactorasanegativeregulatorofArabidopsisroothydraulicconductivity(Lpr).
ThedistinctfunctionalitiesofaseriesofnaturalXND1variantsandasinglenucleotidepolymorphismthatdeterminesXND1translationefciencydemonstratethesignicanceofXND1naturalvariationatspecies-widelevel.
Phenotypingofxnd1mutantsandnaturalXND1variantsshowthatXND1modulatesLprthroughactiononxylemformationandpotentialindirecteffectsonaquaporinfunctionandthatitdiminishesdroughtstresstolerance.
XND1alsomediatestheinhibitionofxylemformationbythebacterialelicitoragellinandcounteractsplantinfectionbytherootpathogenRalstoniasolanacearum.
Thus,geneticvariationatXND1,andxylemdifferentiationcontributetoresolvingthemajortrade-offbetweenabioticandbioticstressresistanceinArabidopsis.
DOI:10.
1038/s41467-018-06430-8OPEN1BPMP,CNRS,INRA,MontpellierSupAgro,UniversitédeMontpellier,34060Montpellier,France.
2LIPM,UniversitédeToulouse,INRA,CNRS,31326Castanet-Tolosan,France.
3InstitutJean-PierreBourgin,INRA,AgroParisTech,CNRS,UniversitéParis-Saclay,78000Versailles,France.
4Presentaddress:InstituteofMolecular,CellandSystemsBiology,CollegeofMedical,VeterinaryandLifeSciences,UniversityofGlasgow,BowerBuilding,GlasgowG128QQ,UK.
CorrespondenceandrequestsformaterialsshouldbeaddressedtoC.
M.
(email:christophe.
maurel@cnrs.
fr)NATURECOMMUNICATIONS|(2018)9:3884|DOI:10.
1038/s41467-018-06430-8|www.
nature.
com/naturecommunications11234567890():,;Thegrowthperformanceandsurvivalofterrestrialplants,whetherinfavorableoradverseenvironments,cruciallydependonaproperuptakeandmanagementofwater.
Mostplantspeciesforagethesoilforwaterthroughcontinuousgrowthanddevelopmentofrootsintoaramiedarchitecture.
Theintrinsicwatertransportpropertiesofroottissues(i.
e.
,theirroothydraulicconductivity,Lpr)arealsoimportantforefcientuptakeandtransferofwatertowardstheshoots.
Lprshowsahighenvironmentalplasticity,withtypicalregulationsdependingontheavailabilityofwater,mineralnutrientsoroxygeninthesoil1.
Thesamestimuliactonrootgrowthanddevelopment,therebyalteringrootsystemarchitecture(RSA)2.
Overall,growthandwatertransportpropertiesofroots,whichcombineintotheso-calledroothydraulicarchitecture,determinetheplant'scapacitytocapturesoilwaterunderchangingorheterogeneoussoilcon-ditions.
Plantsalsodisplayremarkableintraspecicgeneticvar-iationsinRSAandhydraulics3–7,withpossibleimpactsonabioticstressresponses.
Thus,theultimatequestionistounderstandhowcombinedgeneticandphysiologicaladjustmentsofRSAandrootwaterpermeabilitycontributetoplantadaptationtospecichabitatsorclimaticscenarios.
Rootwatertransportpersereliesonseveralfundamentalprocesses,oftenpresentedassequential.
Radialwaterow,fromthesoiltothevasculatureinthestele,ismediatedthroughcellwalls(apoplasticpath)orfromcell-to-cell.
Thelatterpathcom-binestranscellular(acrosscellmembranesandaquaporins)andsymplastic(acrossplasmodesmata)transport.
Wateristhenaxi-allytransportedtotheaerialpartsthroughxylemvessels.
Thecontextsinwhichrootxylemcanbehydraulicallylimitingarestilldebated8.
BasedonPoiseuille'slawoflaminarow,itwascal-culatedthatunderwatersufcientconditionsthistissueissup-posedlynotlimitingwithrespecttorootstructuresmediatingradialtransport9.
Thismaynotbetrueinroottips,wherebyvesselsarenotfullydifferentiated.
Xylemcavitationunderdroughtcanalsodramaticallyreduceplanthydrauliccon-ductanceandconferhighplantvulnerability10.
Conversely,droughtimpactsxylemdifferentiationfurthersupportingacru-cialroleofvasculartransportintheseconditions5,8.
However,thisviewmaynotapplytocertainspeciesorundermildwaterstresssinceintraspecicvariationofxylemsizeinmajorcropssuchasricewasnotassociatedtoanygrowthadvantage,especiallyunderwaterdecit11.
Thus,thelinksthatxylemvesseldifferentiationestablishesbetweenrootgrowthanddevelopment,hydraulics,andstressresponsesarenotfullyestablished8.
Incontrast,detailedstudieshaverevealedhowpositioningofxylemaxescontributestoearlyvascularpatternformationandhowsub-sequentdifferentiationofxylemtrachearyelementsoccursthroughcellclearancebyprogrammedcelldeathanddepositionoflignininsecondarycellwalls12.
ThesedevelopmentalprocessesarecontrolledbyregulatorynetworksinvolvingNAC(NAM,ATAF1,2,andCUC2)13andMYB(myeloblastosis)-typetran-scriptionfactors14.
Theformerproteinshaveplayedakeyevo-lutionaryroleinwater-conductingstructuresfrommosstovascularplants15.
Whileroothydraulicsispronetonebiophysicalanalyses,geneticdissectionofthistraithasbeensomewhatlaggingdueinparttotechnicaldifcultiesindeningproperhydraulicpheno-types.
Reversegeneticanalysesofcandidategeneshaveuncoveredthelimitingroleofaquaporins1orendodermalbarriers16,17.
Incontrast,othercomponentswhichdeterminerootanatomyorenvironmentalsignalingandwhichcanpotentiallyinterferewithroothydraulicpropertieshavebeenpoorlyexplored18.
Quanti-tativegeneticsapproachesbasedonintraspecicvariationsofroothydraulics3,4couldhelpuncoversuchmolecularcompo-nents.
Inlinewiththeseideas,quantitativetraitlocus(QTL)analysisofLprinabiparentalrecombinantpopulation(Bur-0*Col-0)ofArabidopsisledtothecloningofhydraulicconductivityofroot1,aRaf-likeMAPKKKgenethatactsasanegativeregulatorofLpr19.
Here,weperformagenome-wideassociationanalysisasanotherapproachtoidentifygenescon-trollingroothydraulicsinArabidopsis.
WeidentifyXYLEMNACDOMAIN1(XND1)asakeynegativeregulatorofArabidopsisroothydraulicsatthespecies-widelevel.
OurstudyalsorevealshowgeneticvariationatXND1maycontributetothetrade-offbetweenabioticstresstoleranceandbioticdefenseinArabidopsis.
ResultsAGWAstudyuncoverstwonovelgenescontrollingLpr.
Asetof143ArabidopsisaccessionsfromtheRegMappanel20wasphenotypedforroothydraulics(SupplementaryData1).
AfourfoldvariationofLprwasobservedamongaccessions(Sup-plementaryFigure1),withacoefcientofvariationof0.
245andabroad-senseheritabilityh2=0.
36.
Conditionalgenomewideassociation(GWA)mappingusing250ksinglenucleotidepoly-morphisms(SNP)data20,andanacceleratedmixed-modelalgo-rithmmethodwithfourmarkersascofactors21revealedtwoSNPsthatweresignicantlyassociatedwithLprvariation(Bonferronimultipletestingcorrectionatα=0.
05)andcontributedto18.
3%and7.
3%ofthegeneticvariance,respectively(Fig.
1a).
Onewaslocatedonchromosome(Chr)1(position13,612,169),whiletheotherwasonChr5(position25,787,448).
Considering20-kbgenomicregionssurroundingthesetwoassociationSNPs(Fig.
1b,c),weidentiedeightcandidategenesforLprvariation.
NoneofthesegenesshowedSNPsthatwereinstronglinkagedis-equilibrium(LD)withthecorrespondingGWAstudies(GWAS)peakSNP(SupplementaryFigure2).
Theywerethereforefurtherevaluatedusingatotalof15Col-0-derivedT-DNAinsertionlines.
AsfortheChr1region,twoallelicinsertionlinesforAt1g36240showed,withrespecttoCol-0,anincreaseinLprby11or17%,whilemutantlinesforthreeneighboringgenesdidnotshowanysignicantLprphenotype(SupplementaryTable1).
WhenconsideringthefourgeneslocatedintheChr5region,weobservedsignicantlyalteredLprspecicallyforthreeT-DNAinsertionlinesofAt5g64530(SupplementaryTable1;Fig.
1d).
ThedatarevealthepowerofGWAmappingforidentifyinggeneticdeterminantsofspecictraitslikeroothydraulics.
At1g36240encodesaputativeribosomalproteinwhichwillbeinvestigatedinanotherwork.
At5g64530encodesXylemNACDomain1(XND1),aNACtranscriptionfactorsthatantagonizesxylemdifferentiation,bynegativelyregulatingsecondarycellwallsynthesisandprogrammedcelldeath22.
Itsputativefunctioninaxialwatertransportledustoexamineincloserdetailsitscon-tributiontoroothydraulics,whichhasnotbeendescribedpreviously.
XND1negativelyregulatesroothydraulics.
Quantitativereversetranscriptionpolymerasechainreaction(qRT-PCR)analysesrevealedcontrastingXND1mRNAabundanceinthethreexnd1allelicmutants(Fig.
1d).
xnd1–3andxnd1–4,whichbothexhibitaT-DNAinsertionintheXND1promoterregionappearedasknock-downandactivationlines,respectively.
ConsistentwithaT-DNAinsertioninthesecondintron,xnd1–5canratherbeconsideredasaknock-outallele.
Whenconsideringtherootdryweight(DW),primaryandtotalrootlength,orlateralrootdensityinplantsgrowninhydroponics,allthreexnd1genotypeshadarootarchitecturesimilartothatofCol-0(SupplementaryFigure3).
Incontrast,andbyreferencetoCol-0,Lprwasincreasedinhypofunctionalmutants(xnd1–3,xnd1–5)byupto29.
7±7.
6%anddecreasedby18.
6±4.
7%inthexnd1–4over-expressionmutant.
TheseresultsindicatethatXND1actsasatruenegativeregulatorofroothydraulics.
Consistentwiththis,ARTICLENATURECOMMUNICATIONS|DOI:10.
1038/s41467-018-06430-82NATURECOMMUNICATIONS|(2018)9:3884|DOI:10.
1038/s41467-018-06430-8|www.
nature.
com/naturecommunicationsoverexpressionofXND1underthecontrolofaCaMV35Spro-moter(SupplementaryFigure4A,B)resultedinadramatic(39.
5±7.
7%)reductioninLpr(Fig.
1e).
XND1isexpressedinrootxylem,preferentiallyinassociationwithdifferentiatingtrachearyelements22,23.
TransgenicexpressionofaGFP–XND1fusionproteinunderthecontrolofxylemspecicpromoterXCP2Pro(SupplementaryFigure4C,D)reducedLprby25.
3±4.
3%(Fig.
1f),implyingthatXND1actsinvasculartissuestoexerthydrauliceffects.
AllelicdiversityatXND1validatesGWAanalyses.
TheLprassociatedSNPidentiedatposition25,787,448/Chr5duringGWAanalysisislocated7.
9kbapartfromtheXND1codingregion,whereasSNPsthatarecloserdidnotshowanysignicantassociation.
Thishintstopossibleallelicheterogeneityandmul-tiplemutationsinXND1thatwouldcontributetoLprvariationandarelinkedtotheassociatedSNP.
Newlyreleasedgenomicsequences24wereusedtoevaluateincloserdetailsnaturalvar-iationatXND1among112accessions,ofwhich85belongtotheinitiallyinvestigatedpanelwhiletheremainingcorrespondstoaccessionsforwhichphenotypicdataweregeneratedlater(Sup-plementaryData2).
Consideringagenomicregionfrom2kbupstreamto0.
3kbdownstreamoftheXND1codingregion,weusedageneralizedlinearmodeltotesttheassociationwithLprat27polymorphicsites(minorallelefrequency(MAF)>0.
05),includingSNPsandINDELs(Fig.
2a,b).
OnesingleSNPlocatedatposition25,795,349inthe5′-UTRofXND1surpassedthesignicancethresholdandwasthereafternamedSNPUTR.
SeveralSNPsalsopointedtopossibleassociationsinthepromoterregion,consistentwithputativeallelicheterogeneity.
Further,weusedtheeightpolymorphismsshowingthelowestPvalues(P0.
05)intheindicatedXND1genomicregionwasinvestigatedinasetof112accessions.
Thex-axisshowsthenucleotidepositionofeachvariant,withemptyandlledcirclesindicatingINDELsandSNPs,respectively.
They-axisshowsthe–log10(P)fortheassociationtests,withthesignicancethresholdatα=0.
05indicatedwithadashedline.
bTheeightpolymorphismsselectedforfurtheranalysisareprojectedontoaschematicrepresentationofXND1genestructure.
Forposition1962,+andrepresentaninsertionanddeletion,respectively.
Theboxesrepresentexons,withsolidandemptyboxesshowingtranslatedanduntranslatedregions,respectively.
TheSNPatChr5-P25,795,349thatsurpassedthesignicancethresholdinaislocatedinthe5'-UTRofXND1andindicatedasSNPUTR.
cTheeightselectednucleotidepolymorphisms(withtheirdistancefromtranslationstartsiteshownonthetop)denesixhaplogroups(H1–H6).
RepresentativeaccessionsandmeanLpr±SEwithineachhaplogroup(n,accessionsnumber)areshown.
One-wayANOVA(Fisher'sLSD,P0.
05(equivalenttoaminorallelecount(MAC)≥8)wereconsidered.
Forconditionalanalysis,foursuccessivestepswereperformedfollowinganinitialcalculationwithAMM.
Ateachstep,oneamongtheninemosthighlyassociatedSNPsidentiedinthepreviouscalculationwasarbitrarilyselectedasacofactor.
Inpractice,fourSNPs(Chr1-P13,612,169;Chr5-P25,787,448;Chr5-P21,846,701,andChr4-P17,518,747)werestepwiseincludedascofactors.
Tocor-rectformultipletesting,aBonferronicorrectionwithanominalsignicancethreshold(α)of0.
05wasapplied,correspondingtoanuncorrectedPvalueof2.
61*107.
TheproportionofgeneticvarianceexplainedbySNP(13,612,169/Chr1)andSNP(25,787,448/Chr5)wasdeterminedusingcoefcientsofdeter-minationfromsimplelinearregressions.
ForXND1-basedlocalassociationana-lysis,weextractedfromSalkArabidopsis1001Genomesdatabase(http://signal.
salk.
edu/atg1001/index.
php)genomicsequencesofXND1(encompassingaregionfrom2kbupstreamto300bpdownstreamofXND1codingsequence)from112accessions.
Ofthese,85belongtotheinitiallyinvestigatedpanelforGWAmapping(SupplementaryData2).
Associationanalysisbetweenpolymorphicsites(includ-ingINDELsandSNPswithMAF>0.
05)andLprwereperformedwithTASSELversion5usingageneralizedlinearmodel45.
ThesignicancethresholdwassetataPvalueof0.
05permarkernumber.
HaplotypeswereclassiedbasedoneightpolymorphicsitesshowingthelowestPvaluesaccordingtoXND1-basedlocalassociationanalysis.
Thehaplogroupscontainingatleastveaccessionswereusedforfurthercomparativeanalysis.
Geneticcomplementationofxnd1.
A7088bpgenomicregionharboringXND1wasampliedfromCol-0,Bur-0,Ty-0,andFei-0genomicDNAusingaiProofHigh-FidelityPCRKit(Bio-Rad)andclonedintoapGreen0179vector46.
AQuikChangeSite-DirectedMutagenesisKit(Agilentstratagene)wasusedtomutatetheSNPUTR(Chr5_P25,795,349)oftheclonedCol-0orBur-0XND1fragments.
AllprimersequencesusedarelistedinSupplementaryTable2.
Theconstructswereconrmedbysequencingandtransferredintoxnd1–5mutantplantsusingtheoraldipmethod47.
Foreachconstruct,threetoveindependenthomozygoustransgeniclineswereselectedinT3generationon30mgL1hygromycinB(Sigma),checkedforXND1expression(seebelow)andphenotypedforLpr.
Quantitativegeneexpression.
XND1andPIPmRNAabundancewaschar-acterizedintransgeniclinesand/ornaturalaccessionsusingqRT-PCR.
TotalRNAwasextractedfromArabidopsisrootsandreverse-transcribedusingaSVTotalRNAIsolationSystem(Promega)andM-MLVreversetranscriptase(Promega),respectively.
PCRwasperformedonanoptical384-wellplatewithaLightCycler480system(Roche)usingSYBRGreenIMaster(Roche)orSYBRPremixExTaq(TaKaRa),accordingtothemanufacturer'sinstructions.
TIP41-likeprotein(At4g34270),PP2A3(At1g13320),andSANDfamilyprotein(At2g28390)wereselectedasreferencegenes48,basedontheirexpressionstabilityamongaccessionsevaluatedusingaNormFindersoftware49.
AllprimersequencesusedarelistedinSupplementaryTable2.
Relativeexpressionlevelsweredeterminedusingthe2(ΔΔC(T))method50,andcalibratedwithrespecttotranscriptabundanceinthewild-typecontrol,unlessotherwisestated.
CharacterizationofRSA.
Rootsofhydroponicallygrown22-day-oldplantswereharvestedandpreservedin20%ethanolsolution.
Thewholerootsystemswereimmersedinwaterandpositionedonasquarepetridish(24cm*24cm),soastoavoidrootoverlapping,andthenimagedusinganEpsonV850Proscannerat600dpi.
Primaryandtotalrootlengths,andlateralrootdensitywereanalyzedwithanOPTIMASsoftware(version6.
1).
Thelateralrootdensitywasdeterminedontheprimaryroot,ina12cmregionstartingfromthetip.
RootDWwasdeterminedafterdesiccationat80°Cforatleast24h.
Roothistologicalanalyses.
Rootsof21-to23-day-oldplantswerecutin2-cm-longsegmentsstartingfrom0.
3cmofthetip.
Rootsegmentswereembeddedin4%low-meltingpointagarose(Euromedex,1670-B)andcross-sectioned(~100μmthickness)usingaMicro-CutH1200Vibratome(Bio-Rad)accordingtothemanufacturer'sinstruction.
XylemmorphologywasobservedunderaBH-2BrighteldMicroscope(Olympus)andquantiedusingImageJ.
Xylemvesselswereidentiedfromtheirthickenedcellwall.
Xylemsizeandabundancewereassessedfromtheareaofallvesselsandvesselnumber,respectively.
Fortestingtheeffectofg22onxylemformation,19-day-oldplantswereexposedfor4daystoahydroponicsolutioncontaining0.
25Mg22or0.
025%DMSOasmocktreat-ment,andxylemmorphologywasanalyzedasdescribedabove.
ExpressionofuorescentXND1fusionproteins.
A1465bpregionharboringXND15′-UTRandcodingsequence,withoutstopcodonandinitswild-typeorSNPUTR-mutatedBur-0form,wasampliedfromabovementionedXND1-pGreen0179clones.
ThefragmentwasclonedusingtheGatewayTechnology(Invitrogen),downstreamofa35SCaMVpromoterandinfusionwithGFPinapGWB505vector51.
TheprimersusedarelistedinSupplementaryTable2.
Theconstructswereconrmedbysequencingandtransferredintoxnd1–5mutantplants.
Foreachconstruct,ninetotenindependenttransgeniclineswereselectedinT2generationon30mgL1hygromycinB(Sigma),culturedinhydoponicsandcheckedforXND1-GFPtranscriptabundanceandGFPuorescenceintensityinroots.
TheprimersusedforqRT-PCRarelistedinSupplementaryTable2.
ForquanticationoftheGFPuorescenceintensity,rootsof21-to23-day-oldplantswereobservedusingauorescencemicroscope(ZeissAxioObserver7)andmeangrayvaluesin400mroottipswerequantiedusinganImageJprogram(NIH,USA).
DatawerenormalizedtothecorrespondingvalueofthelineC9-3,whichpossessedthelowesttranscriptabundanceanduorescenceintensity.
Droughtstresstreatments.
Col-0plants,xnd1T-DNAinsertionandcom-plementationlinesweregrownintrayslledwithpeatsoil(NeuhausHuminSubstratN2,Klasmann-Deilmann).
Fourtosixtrayswereusedperexperiment.
Eachtraywasdividedinfourquarters,eachcontainingsixplantsofaspecicgenotype(quarter-splitmanner).
Thedimensions(length*width*depth)ofthetrayswere18cm*13cm*5.
5cm.
Plantsweremaintainedinagrowthchamberat20°Cand65%relativehumidity,withcyclesof8hoflight(250molphotonsm2s1)and16hofnight,andsufcientwatering.
After22days,plantsweresubjectedtodroughtbywithholdingwaterfor24daysandre-irrigatedfor5days.
Gravimetricsoilwatercontentwasaround30%attheendofthewaterdecitperiod.
Thecontroltreatmentwasconductedinthesameconditions,butwithcontinuouswatering.
Shootfreshweight(FW)wasdeterminedimmediatelyafterharvestwhereasshootDWwasmeasuredafterfurtherdesiccationforatleast4daysat60°C.
ShootwatercontentwascalculatedastheFW-to-DWdifference.
Alldatawerenormalizedtothecorre-spondingmeanvalueofCol-0plantsinthesametray.
Bacterialinoculations.
SeedsofCol-0plantsandxnd1T-DNAinsertionlinesweresurfacesterilizedfor20minwitha12%sodiumhypochloritesolution,washedvetimeswithsterilewaterandsownonaMSsolidmedium.
After8daysat20°Cinagrowthchamber,plantletsweretransferredtoJiffypots(JiffyFrance,Lyon,France)andgrownfor3weeksundershortdaysconditionsat22°Cand70%ARTICLENATURECOMMUNICATIONS|DOI:10.
1038/s41467-018-06430-810NATURECOMMUNICATIONS|(2018)9:3884|DOI:10.
1038/s41467-018-06430-8|www.
nature.
com/naturecommunicationsrelativehumiditywith9hoflight(250μmolm2s1).
Exposedrootsoftheplantswereimmersedfor20mininasuspensioncontaining108bacteria/mLofR.
sola-nacearumGMI1000strain52.
Inoculatedplantswerethentransferredtoanewtrayonarmsurfaceofpottingsoil,andincubatedinagrowthchamberat75%relativehumiditywithcyclesof12hoflight(100μmolm2s1)at27°Cand12hofnightat26°C.
Plantpositionwasrandomizedpriortoinoculation.
Symptomappearancewasscoreddailyandindependentlyforeachplant,usingamacroscopicscaledescribingtheobservedwilting:0,nowilting;1,25%ofleaveswilted;2,50%;3,75%;4,completewilting.
Forsubsequentanalysis,thedataweretransformedintoabinaryindex:0,<50%leaveswilted;1,≥50%wiltedleavesinordertoconstructsurvivalcurves.
WethenappliedtheKaplan–Meiersurvivalanalysis53withtheGehan–Breslow–WilcoxonmethodtocomputethePvalueandtestthenullhypothesisofidenticalsurvivalexperienceofthetestedmutant.
APvaluelowerthan0.
05wasconsideredtobesignicant.
Thesurvivalcurvesrepresentapoolofthreetechnicalreplicates,eachwith24–32plants,and3independentbiologicalreplicatescorrespondingto240plantsforCol-0andforeachofthexnd1T-DNAinsertionlines.
Forbacterialinternalgrowthmeasurements,aR.
solanacearumGMI1000derivativestraincarryingagentamycineresistancecassette54wasused.
Rosettesofthreetosixpairsofplantswereharvestedat3and4dayspost-inoculation,sterilizedin70%ethanolandrinsedthreetimesinsterilewater.
Therosetteswerethenweighted,grindedandre-suspendedinsterilewater.
BacterialconcentrationsweredeterminedbyplatingdilutionsonBmedium.
Fourbiologicalreplicatesweredone.
Comparisonofinplantabacterialmultiplicationinxnd1genotypeswiththatinCol-0wasperformedthroughaMann–Whitneytest.
AllstatisticalanalyseswereperformedwithaPrismversion5.
0software(GraphPadSoftware,SanDiego,CA,USA).
Statisticalanalyses.
Unlessotherwiseindicated,statisticalsignicanceofthedatawasassessedusingeitheraStudent'sttest(*P<0.
05)orone-wayANOVA(low-ercaseletters:P<0.
05).
Student'sttestswereperformedusingEXCELwhereasaSTATISTICAsoftwarewasusedforANOVAandmultiplecomparisontests(Fisher'sleastsignicantdifference).
DataavailabilityThedatasupportingthendingsofthestudyareavailableasSupplementarydataorfromthecorrespondingauthoronreasonablerequest.
Received:22January2018Accepted:4September2018References1.
Maurel,C.
etal.
Aquaporinsinplants.
Physiol.
Rev.
95,1321–1358(2015).
2.
Shahzad,Z.
&Amtmann,A.
Foodforthought:hownutrientsregulaterootsystemarchitecture.
Curr.
Opin.
Plant.
Biol.
39,80–87(2017).
3.
Sutka,M.
etal.
NaturalvariationofroothydraulicsinArabidopsisgrowninnormalandsaltstressconditions.
PlantPhysiol.
155,1264–1276(2011).
4.
Adachi,S.
etal.
Characterizationofaricevarietywithhighhydraulicconductanceandidenticationofthechromosomeregionresponsibleusingchromosomesegmentsubstitutionlines.
Ann.
Bot.
106,803–811(2010).
5.
Lynch,J.
P.
,Chimungu,J.
G.
&Brown,K.
M.
Rootanatomicalphenesassociatedwithwateracquisitionfromdryingsoil:targetsforcropimprovement.
J.
Exp.
Bot.
65,6155–6166(2014).
6.
Uga,Y.
etal.
ControlofrootsystemarchitecturebyDEEPERROOTING1increasesriceyieldunderdroughtconditions.
Nat.
Genet.
45,1097–1102(2013).
7.
Uga,Y.
,Okuno,K.
&Yano,M.
QTLunderlyingnaturalvariationinsteleandxylemstructuresofriceroot.
Breed.
Sci.
58,7–14(2008).
8.
Vadez,V.
Roothydraulics:theforgottensideofrootsindroughtadaptation.
FieldCropsRes.
165,15–24(2014).
9.
Steudle,E.
&Peterson,C.
A.
HowdoeswatergetthroughrootsJ.
Exp.
Bot.
49,775–788(1998).
10.
Choat,B.
etal.
Globalconvergenceinthevulnerabilityofforeststodrought.
Nature491,752–755(2012).
11.
Yambao,E.
B.
,Ingram,K.
T.
&Real,J.
G.
RootxyleminuenceonthewaterrelationsanddroughtresistanceofriceJ.
Exp.
Bot.
43,925–932(1992).
12.
DeRybel,B.
,Mahonen,A.
P.
,Helariutta,Y.
&Weijers,D.
Plantvasculardevelopment:fromearlyspecicationtodifferentiation.
Nat.
Rev.
Mol.
CellBiol.
17,30–40(2016).
13.
Olsen,A.
N.
,Ernst,H.
A.
,Leggio,L.
L.
&Skriver,K.
NACtranscriptionfactors:structurallydistinct,functionallydiverse.
TrendsPlant.
Sci.
10,79–87(2005).
14.
Heo,J.
O.
,Blob,B.
&Helariutta,Y.
Differentiationofconductivecells:amatteroflifeanddeath.
Curr.
Opin.
PlantBiol.
35,23–29(2017).
15.
Xu,B.
etal.
ContributionofNACtranscriptionfactorstoplantadaptationtoland.
Science343,1505–1508(2014).
16.
Pster,A.
etal.
Areceptor-likekinasemutantwithabsentendodermaldiffusionbarrierdisplaysselectivenutrienthomeostasisdefects.
eLife3,e03115(2014).
17.
Ranathunge,K.
&Schreiber,L.
WaterandsolutepermeabilitiesofArabidopsisrootsinrelationtotheamountandcompositionofaliphaticsuberin.
J.
Exp.
Bot.
62,1961–1974(2011).
18.
Lefebvre,V.
etal.
ESKIMO1disruptioninArabidopsisaltersvasculartissueandimpairswatertransport.
PLoSOne6,e16645(2011).
19.
Shahzad,Z.
etal.
Apotassium-dependentoxygensensingpathwayregulatesplantroothydraulics.
Cell167,87–98(2016).
20.
Horton,M.
W.
etal.
Genome-widepatternsofgeneticvariationinworldwideArabidopsisthalianaaccessionsfromtheRegMappanel.
Nat.
Genet.
44,212–216(2012).
21.
Seren,U.
etal.
GWAPP:awebapplicationforgenome-wideassociationmappinginArabidopsis.
PlantCell24,4793–4805(2012).
22.
Zhao,C.
,Avci,U.
,Grant,E.
H.
,Haigler,C.
H.
&Beers,E.
P.
XND1,amemberoftheNACdomainfamilyinArabidopsisthaliana,negativelyregulateslignocellulosesynthesisandprogrammedcelldeathinxylem.
PlantJ.
53,425–436(2008).
23.
Zhao,C.
,Craig,J.
C.
,Petzold,H.
E.
,Dickerman,A.
W.
&Beers,E.
P.
ThexylemandphloemtranscriptomesfromsecondarytissuesoftheArabidopsisroot-hypocotyl.
PlantPhysiol.
138,803–818(2005).
24.
The1001GenomesConsortium.
1135genomesrevealtheglobalpatternofpolymorphisminArabidopsisthaliana.
Cell166,481–491(2016).
25.
Bazakos,C.
,Hanemian,M.
,Trontin,C.
,Jimenez-Gomez,J.
M.
&Loudet,O.
Newstrategiesandtoolsinquantitativegenetics:howtogofromthephenotypetothegenotype.
Annu.
Rev.
PantBiol.
68,435–455(2017).
26.
Verslues,P.
E.
,Lasky,J.
R.
,Juenger,T.
E.
,Liu,T.
W.
&Kumar,M.
N.
Genome-wideassociationmappingcombinedwithreversegeneticsidentiesneweffectorsoflowwaterpotential-inducedprolineaccumulationinArabidopsis.
PlantPhysiol.
164,144–159(2014).
27.
Kerdaffrec,E.
etal.
MultipleallelesatasinglelocuscontrolseeddormancyinSwedishArabidopsis.
eLife5,e22502(2016).
28.
Grimm,D.
G.
etal.
easyGWAS:Acloud-basedplatformforcomparingtheresultsofgenome-wideassociationstudies.
PlantCell29,5–19(2017).
29.
Zhao,C.
etal.
XYLEMNACDOMAIN1,anangiospermNACtranscriptionfactor,inhibitsxylemdifferentiationthroughconservedmotifsthatinteractwithRETINOBLASTOMA-RELATED.
NewPhytol.
216,76–89(2017).
30.
Hinnebusch,A.
G.
,Ivanov,I.
P.
&Sonenberg,N.
Translationalcontrolby5′-untranslatedregionsofeukaryoticmRNAs.
Science352,1413–1416(2016).
31.
Merchante,C.
,Stepanova,A.
N.
&Alonso,J.
M.
Translationregulationinplants:aninterestingpast,anexcitingpresentandapromisingfuture.
PlantJ.
90,628–653(2017).
32.
Grant,E.
H.
,Fujino,T.
,Beers,E.
P.
&Brunner,A.
M.
CharacterizationofNACdomaintranscriptionfactorsimplicatedincontrolofvascularcelldifferentiationinArabidopsisandPopulus.
Planta232,337–352(2010).
33.
Tixier,A.
etal.
Arabidopsisthalianaasamodelspeciesforxylemhydraulics:doessizematterJ.
Exp.
Bot.
64,2295–2305(2013).
34.
Richards,R.
A.
&Passioura,J.
B.
Abreedingprogramtoreducethediameterofthemajorxylemvesselintheseminalrootsofwheatanditseffectongrainyieldinrain-fedenvironments.
Aust.
J.
Agric.
Res.
40,943–950(1989).
35.
Wortemann,R.
etal.
GenotypicvariabilityandphenotypicplasticityofcavitationresistanceinFagussylvaticaL.
acrossEurope.
TreePhysiol.
31,1175–1182(2011).
36.
Plavcova,L.
&Hacke,U.
G.
Phenotypicanddevelopmentalplasticityofxyleminhybridpoplarsaplingssubjectedtoexperimentaldrought,nitrogenfertilization,andshading.
J.
Exp.
Bot.
63,6481–6491(2012).
37.
Atkinson,C.
J.
&Taylor,J.
M.
EffectsofelevatedCO2onstemgrowth,vesselareaandhydraulicconductivityofoakandcherryseedlings.
NewPhytol.
133,617–626(1996).
38.
Kondo,Y.
etal.
PlantGSK3proteinsregulatexylemcelldifferentiationdownstreamofTDIF-TDRsignalling.
Nat.
Commun.
5,3504(2014).
39.
Kang,S.
etal.
TheArabidopsistranscriptionfactorBRASSINOSTEROIDINSENSITIVE1-ETHYLMETHANESULFONATE-SUPPRESSOR1isadirectsubstrateofMITOGEN-ACTIVATEDPROTEINKINASE6andregulatesimmunity.
PlantPhysiol.
167,1076–1086(2015).
40.
Hayward,A.
C.
Biologyandepidemiologyofbacterialwiltcausedbypseudomonassolanacearum.
Annu.
Rev.
Phytopathol.
29,65–87(1991).
41.
Xin,X.
F.
etal.
Bacteriaestablishanaqueouslivingspaceinplantscrucialforvirulence.
Nature539,524–529(2016).
42.
Nuruzzaman,M.
,Sharoni,A.
M.
&Kikuchi,S.
RolesofNACtranscriptionfactorsintheregulationofbioticandabioticstressresponsesinplants.
Front.
Microbiol.
4,248(2013).
43.
Smakowska,E.
,Kong,J.
,Busch,W.
&Belkhadir,Y.
Organ-specicregulationofgrowth-defensetradeoffsbyplants.
Curr.
Opin.
PlantBiol.
29,129–137(2016).
NATURECOMMUNICATIONS|DOI:10.
1038/s41467-018-06430-8ARTICLENATURECOMMUNICATIONS|(2018)9:3884|DOI:10.
1038/s41467-018-06430-8|www.
nature.
com/naturecommunications1144.
Funk,V.
,Kositsup,B.
,Zhao,C.
&Beers,E.
P.
TheArabidopsisxylempeptidaseXCP1isatrachearyelementvacuolarproteinthatmaybeapapainortholog.
PlantPhysiol.
128,84–94(2002).
45.
Bradbury,P.
J.
etal.
TASSEL:softwareforassociationmappingofcomplextraitsindiversesamples.
Bioinformatics23,2633–2635(2007).
46.
Hellens,R.
P.
,Edwards,E.
A.
,Leyland,N.
R.
,Bean,S.
&Mullineaux,P.
M.
pGreen:aversatileandexiblebinaryTivectorforAgrobacterium-mediatedplanttransformation.
PlantMol.
Biol.
42,819–832(2000).
47.
Clough,S.
J.
&Bent,A.
F.
Floraldip:asimpliedmethodforAgrobacterium-mediatedtransformationofArabidopsisthaliana.
PlantJ.
16,735–743(1998).
48.
Czechowski,T.
,Stitt,M.
,Altmann,T.
,Udvardi,M.
K.
&Scheible,W.
R.
Genome-wideidenticationandtestingofsuperiorreferencegenesfortranscriptnormalizationinArabidopsis.
PlantPhysiol.
139,5–17(2005).
49.
Andersen,C.
L.
,Jensen,J.
L.
&Orntoft,T.
F.
Normalizationofreal-timequantitativereversetranscription-PCRdata:amodel-basedvarianceestimationapproachtoidentifygenessuitedfornormalization,appliedtobladderandcoloncancerdatasets.
CancerRes.
64,5245–5250(2004).
50.
Livak,K.
J.
&Schmittgen,T.
D.
Analysisofrelativegeneexpressiondatausingreal-timequantitativePCRandthe2(-DeltaDeltaC(T))Method.
Methods25,402–408(2001).
51.
Nakagawa,T.
etal.
ImprovedGatewaybinaryvectors:high-performancevectorsforcreationoffusionconstructsintransgenicanalysisofplants.
Biosci.
Biotechnol.
Biochem.
71,2095–2100(2007).
52.
Salanoubat,M.
etal.
GenomesequenceoftheplantpathogenRalstoniasolanacearum.
Nature415,497–502(2002).
53.
Bland,J.
M.
&Altman,D.
G.
Survivalprobabilities(theKaplan–Meiermethod).
Br.
Med.
J.
317,1572(1998).
54.
Guidot,A.
etal.
MultihostexperimentalevolutionofthepathogenRalstoniasolanacearumunveilsgenesinvolvedinadaptationtoplants.
Mol.
Biol.
Evol.
31,2913–2928(2014).
AcknowledgmentsThisworkwassupportedbytheAgenceNationaledelaRecherche(ANR-11-BSV6-018).
WethankDrEricP.
Beers(VirginiaPolytechnicInstituteandStateUniversity,USA)forprovidingtheXND1-ectopicexpressionmaterials,Dr.
TsuyoshiNakagawa(ShimaneUniversity,Japan)forprovidingtheGFP-containingvectorpGWB505,Dr.
CarineAlconandDr.
GenevièveConéjérofromthePHYVplatformfortheirhelpinhistologicalexperiments,andXavierDumontfortechnicalassistanceonplanttransformations.
F.
L.
wasfundedbyagrantfromtheFrenchMinistryofNationalEducationandResearch.
TheIJPBandLIPMbenetfromthesupportoftheLabExSaclayPlantSciences-SPS(ANR-10-LABX-0040-SPS)andLabExTULIP(ANR-10-LABX-41),respectively.
AuthorcontributionsN.
T.
andC.
M.
conceivedanddesignedtheexperimentswithinputfromZ.
S.
,O.
L.
,andF.
V.
N.
T.
andZ.
S.
performedGWASandphenotypescreeningofT-DNAinsertionlines.
F.
L.
andF.
V.
performedtheRalstoniainfectionassays.
N.
T.
conductedallotherexperiments.
N.
T.
andC.
M.
analyzedthedataandwrotethepaperwhichwascheckedbyallauthors.
AdditionalinformationSupplementaryInformationaccompaniesthispaperathttps://doi.
org/10.
1038/s41467-018-06430-8.
Competinginterests:Theauthorsdeclarenocompetinginterests.
Reprintsandpermissioninformationisavailableonlineathttp://npg.
nature.
com/reprintsandpermissions/Publisher'snote:SpringerNatureremainsneutralwithregardtojurisdictionalclaimsinpublishedmapsandinstitutionalafliations.
OpenAccessThisarticleislicensedunderaCreativeCommonsAttribution4.
0InternationalLicense,whichpermitsuse,sharing,adaptation,distributionandreproductioninanymediumorformat,aslongasyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCommonslicense,andindicateifchangesweremade.
Theimagesorotherthirdpartymaterialinthisarticleareincludedinthearticle'sCreativeCommonslicense,unlessindicatedotherwiseinacreditlinetothematerial.
Ifmaterialisnotincludedinthearticle'sCreativeCommonslicenseandyourintendeduseisnotpermittedbystatutoryregulationorexceedsthepermitteduse,youwillneedtoobtainpermissiondirectlyfromthecopyrightholder.
Toviewacopyofthislicense,visithttp://creativecommons.
org/licenses/by/4.
0/.
TheAuthor(s)2018ARTICLENATURECOMMUNICATIONS|DOI:10.
1038/s41467-018-06430-812NATURECOMMUNICATIONS|(2018)9:3884|DOI:10.
1038/s41467-018-06430-8|www.
nature.
com/naturecommunications