strandisdisplacedfromtheD-loopstructureandannealswitha3homologousstran - PDF document

strandisdisplacedfromtheD-loopstructureandannealswitha3homologousstrandthatisavailableduetoresectionofthesecondendoftheDSB.Thus,ageneconversionwithoutlossofsequenceinformationistheÞnalresultofthereaction(Figure1).ThemechanismsofDSBrepairarecurrentlythefocusofinterest,asthedevelopmentofzincÞnger(Carroll,2011)andTALEnucleases(BogdanoveandVoytas,2011)enablestheinductionofDSBsalmostanywhereingenomes.Inplants,inductionofDSBscanbeusedtomutategenesbyNHEJetal.,2010;PuchtaandHohn,2010;Zhangetal.2010)forgenetargeting(Puchtaetal.,1996;Shuklaetal.2009;Townsendetal.,2009;Fauseretal.,2012)ortoremoveunwantedsequencesfromthegenomebyNHEJorSSA(SiebertandPuchta,2002;Petolinoetal.,2010).Whereasinyeastandanimals,therolesofmanyfactorsinvolvedinHRhavebeencharacterisedindetailfordifferentrecombinationreactions,suchquestionshavebeensparselyaddressedinplants.Sometimeago,wewereabletoestablishreporterlinesthatenabledustodiscriminatebetweentheSSAandSDSApathways(Figure1)(Oreletal.2003).Theassaysystemdependsontherestorationofa-glucuronidasegeneafterinductionofDSBsbytherarecuttinghomingendonucleaseI-I.Afterpropa-gationofrecombinedplantmaterialwewereabletodemonstratebySouthernblotsandPCRwithsubsequentsequenceanalysisthatrestorationofthemarkerinthedifferentassaylineswasindeedduetohomologousrecombination.(Oreletal.,2003).Usingtherespectivemarkers,thefrequenciesofbothpathwayscouldbecom-pared,showingthattheSSApathwayappearstobeapproximatelyÞvetimesmoreefÞcientthanthegeneconversionpathway.Recently,weappliedtheselinestoelucidatetheroleofrecombinationintermediateprocessingfactorsRAD5A,MUS81andRECQ4AinSSAandSDSAetal.,2010).Interestingly,theassaysystemwasalsousedduringthediscoveryoftheinvolvementofsmallRNAsinDSBrepairandforthecharacterisationoffactorsinvolvedintherespectivesmallRNAprocessingpathwayetal.,2012).Here,weapplythetransgeniclinestodeÞnetheroleofanumberoffactorsthatareputativelyinvolvedinDNAprocessingintheSSAandSDSAmechanisms.WetestedfactorsinvolvedinbothDSBendprocessing(COM1,MRE11)andintheprocessofstrandexchange(RAD51,RAD51C,XRCC3andRAD54).Moreover,wealsotestedtwofactorsthat,accordingtorecentresults,seemtobeinvolvedinthestabilisationofcertainrecombinationintermediates(FANCMandSMC6B).SetupoftheassaysystemTheefÞciencyofsomaticHRpathwayscanbeaddressedbyusingtransgenicplantsharbouringdifferentconÞgurationsofanon-functionalmarkergene(inourcase,dase)thatcanonlyberestoredbytherespectivepathway(Figure1).AccordingtothemodelfortheSDSAmechanism,theorientationofthedonorsequenceinrelationtothebreakhasnoinßuenceongeneconversion,andthebreakshouldliewithinahomologousregionthatcanberepairedbytheuseofatemplatethatishomologoustobothendsofthebreak.ThetworecombinationreporterlinesDU.GUS-8andIU.GUS-8differonlybytheorientationofthepartofthe T P b a r T I restriction site I-I-I expression P P RBLB SceI restriction site RB U T U U TI-I expressionTPDGU.US-1(b)(c)(d) Figure1.Modelsandsubstratesforsingle-strandannealing(SSA)andsynthesis-dependentstrand-annealing(SDSA).ModelsoftheSSAandtheSDSApathwaysofrecombinationaredepictedin(c)(SSA)and(d)(SDSA).Thehomologousrecombination(HR)eventsaredeterminedbyusingthereporterlinesDGU.US-1(forSSA,a)andDU.GUS-8andIU.GUS-8(forSDSA,b).InDGU.US-1theI-IrestrictionsiteisßankedbytwopartsofaGUSgeneharbouringanoverlapof557bp.The5fragmentwasfusedwiththe35Spromoterofthecaulißowermosaicvirus(brown)andthe3-GUSfragmenttoanopalinesynthase(NOS)terminatorofAgrobacteriumtumefaciens(green).Furthermore,theDGU.US-1constructharboursaresistancegeneforphosphinothricinresistance()ßankedbya35Spromoteranda35Sterminator(brown)(b).IntheDU.GUS-8andIU.GUS-8linea1087-bpGUSfragmentisinsertedindirect(DU.GUS)orininverted(IU.GUS)orientationnexttotherightborder.Nexttotheleftborderisanon-functionalGUSgenewithalinkersequenceharbouringanI-Irecognitionsite.DU.GUS-8andIU.GUS-8alsocontainthehygromycin-resistancegene(hygromycinBphosphotransferase)fusedwithaNOSpromoter(green)andaNOSterminator.ThewayinwhichthemarkergenescanberestoredafterI-IinducedHRbytherespectivepathwayisdepicted.P,promoter;T,NadineRothetal.2012TheAuthorsThePlantJournal2012BlackwellPublishingLtd,ThePlantJournal,(2012),,781Ð790 -glucuronidasemarkerthatshouldbeusedasatemplateforrepairafterinductionofaDSBattheI-Isite(Oreletal.,2003).Therefore,weexpectedtoÞndsimilarrecom-binationefÞcienciesinthesamegeneticbackgroundwithbothlines.However,thetransgeneconstructinDGU.US-1isconstructedinsuchawaythatafterbreakinductionbyI,restorationofthemarkerisonlypossiblebyanneal-ingthetwodirectrepeats.TheI-IexpressionlineusedinthisstudycontainedanartiÞcialI-Iopenreadingframe(ORF)optimisedforplantexpressionfusedtoadouble35STheArabidopsisplantsusedfortheSSAandSDSArecombinationassaysmustbeheterozygousforboththereporterconstructandtheI-Iexpressingconstructinahomozygousmutantortherespectivewild-typeback-ground.Dependingonwhethertherespectivemutantisfertileorsterile,differentapproachestoachievethisstatusmustbetaken.Inthecaseoffertilemutants(),thereporterlinesandanI-Iexpressinglinearecrossedwiththerespectivemutantlinesindependently.IntheFgeneration,plantsthatarehomozygousfortheIexpressionconstructorthereporterconstructintherespectivemutantorthecorrespondingwild-typeback-groundareidentiÞedbyPCRandpropagated.AsaÞnalstep,thereportersubstratesarecrossedwiththeI-Iexpress-ingline,eitherinthemutantorthecorrespondingwild-typebackground.Inthenextgeneration,allseedsarehete-rozygousforboththeI-Iexpressingconstructandthereportersystem.Theseseedsarethensownout,andrecombinationfrequenciesaredeterminedbycountingbluesectorsafterhistochemicalstaining(Figure2).Ifthemutantissterile(),plantsthatarehomozygousforthedifferenttransgenesbuthemizygousforthemutatedgenesmustbeproduced.Aftercrossingthehemizygousmutantsintherespectivehomozygousmarkerbackgrounds,therequiredgenotypesmustbeidentiÞedbyPCRbeforetherecombinationfrequenciesaredetermined(Figure2). I-I Reporter line+/+Gene PCR based genotyping and propagationF3 I-I I-I I-I I-I I-I Reporter line+/+Gene+/– F2PCR based genotyping and propagationF3 I-I I-I I-I PCR based genotyping and propagation I-I Comparison of HR frequencies (a)(b) Figure2.Crossingschemeforfertilemutantlinesandsterilemutantlines.ToestablishthereporterassaysÞrstthereporterlines(DGU.US,DU.GUSorIU.GUS)aswellasanI-Iexpressionlineareindependentlycrossedwiththerespectivehomozygousmutant(a)orthesegregatingmutantline(incaseofsterilityofthemutant,b).Inthesecondgenerationaftercrossing,allhplantsareidentiÞedbyPCR-basedgenotypingandpropagated.Inalaststep,thereportersubstrateandtheI-Iexpressingconstructarebroughttogetherbycrossingtherespectiveplants,eitherinthemutantorinthewild-typebackground.ThismethodologyresultsinplantsthatareheterozygousforboththereporterandtheI-Iexpressingconstructinahomozygousmutantorwild-typebackground(a).Toassesshowthelossofageneofinterestinßuenceshomologousrecombination(HR),thenumberofrepaireventsaftertheinductionofadouble-strandbreakarecomparedbetweenthemutantandthecorrespondingwiplants.Incaseofsterility,mutantlinesmustbepropagatedinaheterozygousmutantbackground,andtheresultingassayplantsmustbegenotypedbyPCRtoidentifyhomozygoussiblings(b).Recombinationfactorsandpathwaysofhomologousdouble-strandbreakrepair2012TheAuthorsThePlantJournal2012BlackwellPublishingLtd,ThePlantJournal,(2012),,781Ð790 Inallcases,mutantrecombinationfrequencieswerenormalisedtotherespectivewildtypes.Inallcases,atleastthreeindependentrecombinationassayswereanalysed.Inthefollowingsections,theresultsofanalysesofthedifferentfactorswillbepresentedinanorderthatrelatestotheirroleintheconsecutivestepsofDSBrepair.COM1andMRE11aredispensableforefcienthomologousrecombinationbytheSDSAandtheSSApathwayaftersite-specicDSBinductionInDSBprocessinginmeiosis,MRE11andCOM1/SAE2areessentialtoprocessanintermediateofthecleavagereactionconsistingofSPO11covalentlylinkedtothe5terminiofDNA.InArabidopsis,theinsertionmutantissterileduetomeioticdefects,indicatingaconservedfunction.Moreover,COM1isinvolvedincrosslinkrepairinArabi-dopsis(Uanschouetal.,2007).Similarly,plantscarryinganullalleleofMRE11aresterile(Puizinaetal.,2004)andsensitivetoDNA-damagingagents(BundockandHooykaas,AsshowninFigure3(a),inthecaseofCOM1wedetectedonlyaveryminorreductioninthemeanrecombinationfrequencyforallthreedifferentassaylinesthatwereindeednotstatisticallysigniÞcant.InthecaseofMRE11(Figure3b)theoutcomewasdifferent,astheefÞciencyofrecombina-tionwasenhancedintheSSAlinebyapproximatelyhalfatastatisticallysigniÞcantlevel(=0.036).ThemeanwasalsoapproximatelyathirdhigherforbothSDSAlines(althoughduetohighervariationinthecaseofIU.GUS,theresultwasonlystatisticallysigniÞcantforDU.GUS).Thus,recombina-tionapparentlyproceedswithhigherefÞciencyinthemutantthaninthewildtype,independentofthemecha-nism.TheseresultsaremostlikelyduetothefactthatMRE11isalsoinvolvedinacertainclassofNHEJeventsinArabidopsis(Heacocketal.,2004).AsNHEJiscompetingwiththetwoHRpathwaysanalysedinthisstudy,inthemutantmorebreaksarechannelledintoHR.Never-theless,ourresultsindicatethatMRE11playsnoroleinanyofthetwopathwaystested.Thus,neitherMRE11norCOM1arerequiredforSSAandSDSAinsomaticcells,atleastifDSBsareinducedbyasequence-speciÞcendonuclease.RAD51,RAD51C,XRCC3andRAD54areimportantforhomologousrecombinationbytheSDSApathwaybutnotfortheSSApathwayIthasbeendemonstratedforyeastthatproteinsinvolvedinstrandexchangearerequiredforgeneconversionandcrossovers.RAD51ÞlamentstabilityandD-loopformationarecontrolledbytheRAD51paralogues(Liuetal.,2011)andtheSWI2/SNF2ATPaseRAD54(CeballosandHeyer,2011).Inouranalysis,weconcentratedoureffortsonRAD54RAD51anditstwoparaloguesXRCC3andRAD51C.Currentknowledgeabouttheinvolvementofthesefactorsindiffer-entrecombinationmechanismsinArabidopsisisquiterudimentary.ForRAD51(Lietal.,2004)anditsparalogueXRCC3(BleuyardandWhite,2004),ithasbeenreportedthat 40100120 06080100120 WTWTWTcom1com1 WT*0406080100120140160 2060100120 406080100120140160180200Rel. recombination events [%]WT(a)(b) Figure3.Recombinationfrequenciesofmutantplants.Therelativerecombinationfrequenciesofthe(a)andthe(b)mutantplantsarepresentedinrelationtothecorrespondingwild-type(WT)controlplants(100%).ThehomologousrecombinationfrequenciesweredeterminedusingthereporterlinesDGU.US-1(SSA),IU.GUS-8(SDSA)andDU.GUS-8(SDSA)asdepictedinFigure1.Therelativerecombinationiscalculatedasthemeanvalueofthreeindependentexperiments.ErrorbarsindicatetheSD.Asterisksindicate-valuesfromtwo-tailedpaired-tests.(***0.001;**=0.001Ð0.01;*=0.01Ð0.05;&#x-365;&#x.500;0.05,noasterisk=notsigniÞcant).NadineRothetal.2012TheAuthorsThePlantJournal2012BlackwellPublishingLtd,ThePlantJournal,(2012),,781Ð790 bothmutantsaresterileandsensitivetoDNA-damagingagents,althoughnodataaboutsomaticHRhavebeenpublished.Usingrecombinationtrapsthatcouldnotdis-criminatebetweendifferentrecombinationpathways,ithasbeenreportedpreviouslythatthelossoftheparalogueRAD51C(Abeetal.,2005;Bleuyardetal.,2005)andtheATPaseRAD54(Osakabeetal.,2006;Shakedetal.,2006)resultsinadefectinHRandsensitivityagainstcross-linkingTotesttheinvolvementofthesefactorsinSSAandSDSA,weappliedourassaysystem.InthecaseofRAD51,wewereonlyabletocombinethemutantbackgroundwiththerecombinationassaylinesDGU.US-1andIU.GUS-8,butnotDU.GUS-8,asthelocationoftheDU.GUS-8transgeneandRAD51aretoocloseonchromosomeVtocombinebothbycrossing.OurresultsclearlydemonstratethatlossofRAD51hasastronginßuenceonSDSA,resultinginareductionofthewild-typerecombinationefÞciencybyalmost80%.Incontrast,thereisaminimalbutstatisticallyinsigniÞcantreductioninthecaseofSSA(Figure4a).ThisÞndingdemonstratesthatthestrandexchangeproteinisessentialfortheinitiationofD-loopsbutnotforasimpleannealingreaction.Notsurprisingly,almostthesameresultisobservedforbothparaloguestested,RAD51CandXRCC3.InthecaseofXRCC3,nodecreaseinefÞciencycanbedetectedinSSA.InthecaseofRAD51C,ourexperimentsshowareductionofapproximatelyaquarter,whichisbarelystatisticallysigniÞcant(=0.04).However,inbothmutant 406080100120 406080100120 406080120Rel. recombination events [%] 0406080100120 406080100120 406080120 406080100120 406080100120 WTrad51CWTrad51Crad51C 406080100120Rel. recombination events [%]WT*********** 406080100120Rel. recombination events [%]DGU.US(b)(c)(d) Figure4.Recombinationfrequenciesofmutantplants.Therepresentativediagramsshowtherelativerecombinationfrequenciesofthe(c)and(d)mutantplantspresentedinrelationtothecorrespondingwild-type(WT)controlplants(100%).ThehomologousrecombinationfrequenciesweredeterminedusingthereporterlinesDGU.US-1(SSA),IU.GUS-8(SDSA)andDU.GUS-8(SDSA),asdepictedinFigure1.Therelativerecombinationiscalculatedasthemeanvalueofthreeindependentexperiments.ErrorbarsindicatetheSD.Asterisksindicate-valuesfromtwo-tailedpaired-tests.(***0.001;**=0.001Ð0.01;*=0.01Ð0.05;&#x-365;&#x.500;0.05,noasterisk=notsigniÞcant).Recombinationfactorsandpathwaysofhomologousdouble-strandbreakrepair2012TheAuthorsThePlantJournal2012BlackwellPublishingLtd,ThePlantJournal,(2012),,781Ð790 backgrounds,theefÞciencyofSDSAwithbothrecombina-tiontrapsisseverelyreduced(Figure4b,c).Finally,theabsenceoftheSWI2/SNF2ATPaseRAD54hasnoinßuenceonrecombinationfrequencyfortheSSApathway,butrecombinationisreducedtoaquarterincomparisonwiththewildtypeintheIU.GUS-8andintheDU.GUS-8assaylines(Figure4d).Thus,RAD54,aswellasRAD51anditsparalogues,areextremelyimportantforSDSAinTheroleofSMC6BandFANCMintheSDSAandtheSSApathwaysFurthermore,wewereinterestedinelucidatingtheroleoftwofactorsthatwereonlyrecentlyshowntobeinvolvedinHRinArabidopsis:FANCMandSMC6B.Thehumanhered-itarydiseaseFanconianaemia(FA)leadstoseveresymp-toms,includingdevelopmentaldefectsandbreakdownofthehaematopoieticsystem.FAiscausedbysinglemuta-tionsinthegenes,oneofwhichencodesfortheDNAtranslocaseFANCM(KnollandPuchta,2011).Recently,wewereabletodemonstratethatAtFANCMactsduringmeiosisasananti-recombinasetosuppressectopicrecombination-dependentchromosomeinteractionsandisinvolvedinthesuppressionofinterference-insensitivecrossovers(Knolletal.,2012).IthasbeenpostulatedthatyeastFANCMbindstorecombinationintermediateslikeD-loopstructuresandthusmightcontroltheefÞciencyofthereaction(Prakashetal.,2009).Interestingly,AtFANCMsuppressesspontane-oussomaticHRviaaRECQhelicase(AtRECQ4A)-indepen-dentpathwaybutontheothersideisrequiredforDSB-inducedHR(Knolletal.,2012).ItwasthereforeimportanttotesthowtheabsenceofFANCMwouldinßu-enceSSAandSDSA.Ourresults(Figure5a)indicatethatgeneconversionefÞciencybySDSAisreducedforbothrecombinationtrapstolessthanhalfofthewildtype.Sur-prisingly,SSAefÞciencyisreduced,buttoalesserextent,toapproximatelytwo-thirdsofthewildtype.TheSMC5/6complex,togetherwithcohesin,isinvolvedinDSBrepairbysisterchromatidrecombinationduringthephaseinyeastsandmammals(DePiccolietal.,2009).WhereasinArabidopsisthemutationofthesingleSMC5homologueisnon-viable,mutationofoneortheotherSMC6homologue(SMC6AandSMC6B)resultsinfertileplants,althoughthedoublemutantisalsonon-viable.WewereabletodemonstratethatSMC6AandSMC6BarebothrequiredforefÞcientrepairofDNAdamageviaHRinsomaticcells(Watanabeetal.,2009).However,usinganassaythatcouldnotdiscriminatebetweendifferenttypesofintra-andintermolecularrecombinationmechanisms,itwasnotpossibletodeÞnetheroleoftheSMC6homologuesinHRingreaterdetail.Forthecurrentstudy,weusedthesamealleleofSMC6Bthathasbeenusedpreviously(Watanabeetal.,2009)todeÞnetheroleoftheproteininSSAandSDSA.Interestingly,incontrasttoourpreviousresultsthatshowedadrasticdecreaseinHRwithandwithoutinductionofrandomDSBsbybleomycinwiththelineDGU.US1etal.,2009),ourrecentanalysiswithsite-speciÞcinductionoftheDSBbetweentheoverlapsshowedareductionofonlyapproximatelyaÞfth(Figure5b).BothSDSAlinesshowedsimilarresults,althoughthereduction 60100120 4060120 WTWTWTfancmfancm 406080100120 406080120 WTWTWT Figure5.Recombinationfrequenciesofmutantplants.(a)and(b)depicttherelativerecombinationfrequenciesofthe(a)andthe(b)mutantplants,whicharepresentedinrelationtothecorrespondingwild-type(WT)controlplants(100%).ThehomologousrecombinationfrequenciesweredeterminedusingthereporterlinesDGU.US-1(SSA),IU.GUS-8(SDSA)andDU.GUS-8(SDSA),asdepictedinFigure1.Therelativerecombinationiscalculatedasthemeanvalueofthreeindependentexperiments.ErrorbarsindicatetheSD.Asterisksindicate-valuesfromtwo-tailedpaired-tests.(***0.001;**=0.001Ð0.01;*=0.01Ð0.05;&#x-365;&#x.500;0.05,noasterisk=notsigniÞcant).NadineRothetal.2012TheAuthorsThePlantJournal2012BlackwellPublishingLtd,ThePlantJournal,(2012),,781Ð790 wasonlystatisticallysigniÞcantinthecaseofDU.GUS-8.ThisÞndingindicatesthatunderthesespeciÞcconditions,themajorityofrecombinationreactionsindeedoccurintra-molecularlyandthatsisterchromatidsareminimallyThespecicityoftheassaysystemsforintrachromatidTodetectspeciÞcrecombinationpathways,transgenicplantlineswereusedinwhichareportergeneisrestoredafterinductionofasite-speciÞcDSBwiththemeganucleaseI.OnemuststatethatboththesetupofthetransgeneaswellastheinductionofaDSBatauniquepositionarepre-requisitesthatthemarkergeneisrestoredÐatleastintheoverwhelmingmajorityofcasesÐbytherespectivemech-anism,whichisnicelyexempliÞedbytheDGU.US-1line.Indeed,onecanenvisagedifferenttypesofmarkerrestora-tionmechanismsincaseswhereDSBsareinducedbygenotoxinsatrandompositionswithinthetransgenicmar-kersequence.InadditiontoSSA,break-inducedreplication(BIR)usingthesisterchromatidasatemplateorunequalsisterchromatidexchangeviaSDSAcanleadtotheresto-rationofthemarker(Watanabeetal.,2009).OnlythefactthatthebreakisexclusivelyinduceddirectlybetweentheoverlapschannelsthereactionintotheSSApathway.ThisphenomenonisclearlydemonstratedbythebehaviouroftheGU.USconstruct(DGU.US-1line)intheabsenceofSMC6B.IncasesinwhichDSBsareinducedrandomlybybleomycin,therecombinationefÞciencyisdrasticallyreducedbyapproximatelyhalfanorderofmagnitudeetal.,2009),indicatingthatatleastincellsintheSorGphasethesisterchromatidplaysamajorroleinHR.Incontrast,onlyamilddefectwasdetectedafterI-mediatedDSBinduction(areductionofapproximatelyaÞfth),demonstratingthatundertheseconditionsintra-molecularSSA,butnotintermolecularsisterchromatidrecombination,preferentiallyoccurs.Inaddition,theSDSAmarkerlinesshowonlyamildreductioninHRefÞciency.ThisÞndingalsoindicatesthatinthiscasehomologyclosebyonthesamebutnotonthesisterchromatidisthepre-ferredmatrixforrepairingtheDSB.OnehastokeepinmindthatbesidetherepaireventsthatleadtotherestorationofthemarkermostDSBsarestillrepairedbyNHEJunderourexperimentalconditions.AlthoughduetotheexperimentalsetupNHEJrepairshouldhardlyresultinarestorationofthemarkergenewecannotexclude,thatasmallfractionoffunctionalGUSORFsarerestoredbyimpreciseNHEJrepairinourassays.IncaseoftheDGU.US-1lineNHEJwouldhavetoresultinadeletionofexactlytheduplicatedregionof557bp.IncaseoftheDU/IU.GUSlinesthe31-bplinkersequenceharbouringtheI-restrictionsitewouldeitherhavetobeeliminatedcom-pletelyorpartiallybyimpreciseNHEJ,resultinginafunctionalORFofthemarker.TheinductionofDSBsmostprobablydoesnottakeplaceatalltimesandinallcellsinourassay.Therefore,aminorfractionoftheHReventsthatleadtotherestorationofthemarkermightnotbeinducedbyanI-ImediatedDSBbutbyanaturallyoccurringDSBorbyotherkindsofDNAdamagelikestalledreplicationforks.AsI-Iexpressionleadstotheenhancementofrecombinationfrequencybyonetotwoordersofmagnitude(Oreletal.,2003)theseeventsrepresentonlyatinyfractionofalleventsandthusshouldnotinßuencetheoutcomeofouranalysis.TheroleofDSBendprocessingOurÞndingsthatneitherCOM1norMRE11arerequiredforSSAorSDSAsuggestthatneitherproteinisneededaftertheproductionofprocessibleDSBends,whichisthecaseiftheyaregeneratedbyhomingnucleases.However,onehastokeepinmindthatmostDSBsthatarisenaturallyduringthelifetimeofanorganismmighthavedifferentproperties.X-rayirradiationmightresultinbreakswithcomplexendsthatcannotbedirectlyprocessedbypolymerasesorligases.Moreover,inthecaseofcovalentlinkageofproteinstoDNA,theDNAmustbesetfreebeforeitcanbeprocessedbytherepairmachinery.Interestingly,ourresultsalsoindicatethat,intheabsenceofMRE11,moreDSBsarerepairedbySSAaswellasSDSA.WedonotfavourthehypothesisthatMRE11isadirectnegativeregulatorofbothpathwaysforthefollowingreasons:asacomplex,MRE11andRAD50formafunctionalnuclease(Daoudal-Cotterelletal.,2002),andthereareindi-cationsthatthecomplexisalsoinvolvedinNHEJinArabid-opsis(Puizinaetal.,2004).Thus,adefectinNHEJmightbecompensatedbyrepairingmorebreaksbytheuseofhomologoussequences.Indeed,usingaconventionalrecombinationtrapthatcouldnotdiscriminatebetweenpathways,ithasbeenreportedpreviouslythatHRisalsoenhancedintheabsenceofRAD50(Gherbietal.,2001).TheimportanceofmultiplefactorsforprocessingrecombinationintermediatesinSDSAWeshowedthattherecombinaseRAD51anditsparaloguesRAD51CandXRCC3areofgreatimportanceforSDSAbutnotforSSA,whichcanbeeasilyexplained,asRAD51Þla-mentformationisrequiredforinvasionofasinglestrandintoadouble-strandedregion.ThisphenomenonsupportsourÞndingthattheArabidopsisdoublemutant,whichisdeÞcientinRAD51Þlamentformation,alsohasatremendousdefectinHR(Seeligeretal.,2012).Weproposethat,similartoyeast,theparaloguesareneededforthestabilisationofrecombinationintermediatesassociatedwiththeRAD51Þlament(Liuetal.,2011).Asinmammals,threemoreparaloguesexistinadditiontoRAD51CandXRCC3,namelyRAD51B,RAD51DandXRCC2(Bleuyardetal.,2005;etal.,2005).TheproteinsarefoundinatleasttwoRecombinationfactorsandpathwaysofhomologousdouble-strandbreakrepair2012TheAuthorsThePlantJournal2012BlackwellPublishingLtd,ThePlantJournal,(2012),,781Ð790 distinctcomplexesinhumancells.OnecomplexcontainsRAD51B,RAD51C,RAD51DandXRCC2,whereastheothercomplexconsistsofRAD51CwithXRCC3(Massonetal.2001).AswedetectedamildbutsigniÞcantreductionintheSSAefÞciencyinthecaseofRAD51CdeÞciency,itwillbeinterestingtotestwhetherthelossofanyoftheotherthreeparaloguesalsohassomeminorinßuenceonSSAinaddi-tiontoastrongdefectinSDSA.ItisinterestingtocomparethecurrentresultswithdataweobtainedusingthesameassaysystemtodeÞnetheroleoffactorsinvolvedintheprocessingofDNArecombinationintermediates(Mannussetal.,2010),namelytheSWI2/SNF2ATPaseRAD5A(Chenetal.,2008),theendonucleaseMUS81etal.,2006)andthehelicaseRECQ4A(Hartungetal.,2007).ForIU.GUS,therecombinationfrequencieswerereducedbyhalfinthemutantbackgroundcomparedwiththewild-typecontrol.Incontrast,nosignif-icantdifferencewasobservedbetweenthemutantsandthewild-typecontrolswiththeDGU.USreportersystem.Thus,bothSWI2/SNF2ATPasesRAD5AandRAD54areinvolvedingeneconversion,butnotintheSSApathwayofHRinsomaticplantcells.BecauselossofRAD54reducedSDSAmoredramatically,itsroleseemstobemoreprominent.Incontrast,aslightreductionoftheHRefÞciencyintheSSApathwaywasobservedinboththeandthemutants(Mannussetal.,2010).Thereductionwaslessthanathird,arguingthatotherfactorsmighthaveplayedmoreprominentrolesorthattheywereabletosubstituteforthenucleaseorforthehelicase.Interestingly,amuchstrongereffectwasobservedwiththeSDSAsubstrate,inwhichtheHRefÞciencywasreducedtolessthanhalfforbothsinglemutants.ThisobservationisreminiscentofFANCMinthisstudy.Indeed,FANCMandRECQ4A-amemberoftheRTRcomplex(Hartungetal.,2008)-bothareinvolvedinsuppressingreplication-associatedHRintwoindependentpathwaysinArabidopsis(Knolletal.,2012).Therefore,oneistemptedtospeculatethatbothhelicasesmightbeinvolvedintheprocessingofasubsetofslightlydifferentrecombinationintermediatesthatmightarisewithintheSDSApathway.Additionally,RECQ4AandMUS81mightbeinvolvedintheprocessingofdifferentclassesofrecombinationintermediates,asthedoublemutantisnon-viablebutcanberescuedbyknockingoutRAD51C,which,asdemonstratedinthisstudy,isinvolvedinSDSA(Hartungetal.,2006;Mannussetal.,2010).Thus,FANCM,RECQ4AandMUS81couldallbeinvolvedinalternativemeansofprocessingSDSA-dependentrecombi-nationintermediatesthatariseafterstrandexchange.Withthecurrentwork,wewereabletoidentifyagroupoffactorsthatplayamajorroleinSDSA.TheseÞndingsareaclearindicationthatthepathwayisquitecomplexandrequiresalargernumberofproteinsthatareinvolvedintheprocessingofdifferenttypesofintermediatesduringdiffer-entstepsofthereaction.Formally,wecannotexcludethefactthatthe-glucuronidasegeneinIU.GUS-8andDU.GUS-8canalsoberestoredbytheclassicaldoublestrandbreakrepair(DSBR)mechanism(Szostaketal.,1983).However,thismechanismplaysnosigniÞcantroleinDSBrepairinsomaticplantcellsastheinitiationofrecombinationhomologytooneendofthebreakissufÞcient(Puchta,1998).TheDSBRmodelpostulatesthathomologyonbothendsofthebreakisrequired.RecentÞndingsinyeastalsoindicatethattherepairbytheDSBRmechanismisonlyaminorpathwayofhomologousDSBrepairinmitoticcellsetal.,2010).Single-strandannealingseemstobemuchsimpler,andtherequiredfunctionsmightbepresentredundantlyinplantcells.ThismightalsobethereasonwhywecouldnotuntilnowidentifyafactorwhoselosswouldseverelyinßuenceSSA.EXPERIMENTALPROCEDURESTheT-DNAinsertionlinesofCOM1(At3g52115),MRE11(At5g54260),XRCC3FANCM(At1g35530)and(At5g61460)wereallobtainedfromtheSALKcollection(Alonsoetal.,2003).Themutantallelescom1-2etal.,2007),(BleuyardandWhite,2004),etal.,2005),etal.,2006),fancm-1etal.,2012)andetal.,2009)werepreviouslydescribed.TheGABIT-DNAinsertionlineGA-BI_134A01()of(At5g20850)wasdescribedbefore(Lietal.,2004)andprovidedbyBerndReis.TheallelehasitsT-DNAinsertioninexon17andissterile,astrongindicationofaloss-of-functionallele.TheI-IexpressionlinewasproducedbycloningtheartiÞcialIORFoptimisedforplantexpression(Puchtaetal.,1993)fusedtoadouble35SpromoterandanoctopineterminatorintheplasmidPZP221(Hajdukiewiczetal.,1994).Theresultingvectorwastrans-ferredintoAgrobacteriumtumefaciens,andtheresultingstrainswereusedtotransformArabidopsisutilisingtheßoraldipmethod(CloughandBent,1998).Bysegregationanalysis,wecoulddeÞnedifferentI-IexpressinglinesthatharbourtheI-Iexpressioncassetteateachsinglegenomiclocus.InthefollowingSouthernblotanalysis,weselectedforlinesthatcarryonlyonecopyoftheIconstructatadeÞnedlocus.Fortherecombinationassaysinthisstudy,weusedtheI-IexpressinglineI-I-8.TheI-linecontainstheI-IcassetteonchromosomeIattheposition4125836withadeletionof1182bpandaninsertionof47bpattheleftborderoftheT-DNAand152bpattherightborder.Thereporterlineswerepreviouslydescribed(Oreletal.,2003).Additionally,wecharacterisedtheIU.GUS-8,DU.GUS-8andDGU.US-1linesbySouthernblotanalysisandgainedevidencethattheseeachcarryasinglecopyoftherecombinationtrapatonelocus.Furthermore,wedeterminedtheintegrationsiteofthereportercassettesviasiteÞnderPCR(Tanetal.,2005).IntheDGU.US-1line,thereportercassetteisinsertedonchromosomeIIIattheposition7386798withadeletionof10bpandaninsertionof9bpattherightborderoftheT-DNAand4bpattheleftborderoftheT-DNA.TheDU.GUS-8lineharboursthereportercassetteonchromosomeVattheposition7081288withaduplicationof6bpatthe5and3sideoftheT-DNAandaninsertionof25bpattherightborderoftheT-DNAand25bpattheleftborderoftheT-DNA.ThereportercassetteintheIU.GUSlineisinsertedonchromosomeIattheposition27372237withadeletionof40bpandaninsertionof6bpattherightsideoftheNadineRothetal.2012TheAuthorsThePlantJournal2012BlackwellPublishingLtd,ThePlantJournal,(2012),,781Ð790 T-DNAborderandaninsertionof10bpattheleftborder.Furthermore,fortherecombinationassays,therespectivewild-typeplantswerealwaysproducedsimultaneouslyandtakenfromthesamecrossingandsegregationasthemutantofinterestwiththesamemarkerlinebackgrounds.AllT-DNAmutantlines,theIexpressionlineandtheGUSreporterlinesareintheColumbia-0background.PlanthandlingandgrowthconditionsArabidopsisseedsweresterilisedin6%sodiumhypochloritefor5minandrinsedseveraltimeswithsterilewater.Plantsweregrowninchambersat22C(CU-36L4,PercivalScientiÞc,http://www.percival-scientiÞc.com/)underwhitelight(16-hlight/8-hdark)andadarkphaseat20C.Forassays,sterilisedseedswerespreadontosolidgerminationmedia(GM)-agar[4.9gLMurashigeandSkoogmicro-andmacro-elements,includingvitaminsandMESbuffer(Duchefa,http://www.duchefa.com/),10gLsucrose,pH5.7,andplant-agar(Duchefa)].Forpropagationandcrossings,plantsweregrowninagreen-houseunderconstant22Cwithalightphaseof16handadarkphaseof8h.AnalysisofrecombinationGenerally,recombinationfrequenciesweremeasuredintheFgenerationaftercrossingplantsharbouringtherespectivereporterconstructandtheI-Iexpressingconstructinahomozygousmutantorthesegregatedwild-typebackground.WeÞrstcrossedbothreporterlinesandanI-Iexpressionlineindependentlyfromoneanotherwiththerespectivemutantlines.Forcrossing,weremovedsepals,petalsandstamenstoaccessthegynoecium.Pollinationwasperformedbytappingmaturestamensfromthefatherplanttothestigma.SeedsfromcrosseswerepropagatedthroughtheFandFgenerations.IntheFplantshomozygousforthetransgenesandtherespectivemutantorthecorrespondingwild-typebackgroundwereidentiÞedbyPCRwithprimersfortherespectiveloci.AsaÞnalstep,thereportersubstrateswerecrossedwiththeI-Iexpressinglineeitherinthemutantorthecorrespondingwild-typebackground.TheseedsofthesecrossingswerethensownoutonPetridishescontainingsolidGMmediumsupplementedwiththeantibioticsphosphinotricin(PPT)andhygromycinforthedifferentreporterlinestoexcludeplantsthatareself-fertilisedfromthemotherline(whichwasalwaystheI-Iexpressingline).TheDGU.US-1linecarriesaPPTresistancemarker,andtheIU.GUS-8andDU.GUS-8linescarryahygromycinresistancemarker.After2weeks,plantletswerehisto-chemicallystainedinanX-Glucstainingsolutionasdescribedetal.,1994)for2daysat37C.Destainingofleafpigmentswith70%ethanolovernightat60Cfacilitatedthefollowinganalysisofrecombinationeventsbycountingbluesectorsunderabinocular.Forsterilemutants,weÞrstcrossedheterozygousmutantsintothereporterandI-Iexpressinglines.Then,genotypedFprogeny,heterozygousforthemutationandhomozygousforeitherthereporterortheI-Itransgene,werecrossedtoobtainasegregatingpopulationwiththedesiredgenotypeinthenextgeneration.ForanalysisoftherecombinationefÞciency,130seedsperlinewerespreadontoGMmediasupplementedwiththeantibioticsasmentionedabove.After2weeksofgrowth,therootsoftheplantletswereremovedandusedforDNAisolationandsubsequentPCRanalysis.Eachsingleseedlingwasplacedintoasinglewellofa24-wellplateandmarkedindividually.StainingwasperformedbyÞllingeachwellwith1.5mlofstainingsolution.After2days,theplantletsweredestainedwith70%ethanol.CorrelationofthePCRresultsobtainedwiththerootsremovedbeforestainingenabledustoidentifyplantshomozygousfortherespectivemutantorwild-typebackground.Plantletswithidenticalgeneticback-groundswerethenpooled,andthenumberofrecombinationeventswasdeterminedperseedlingusingabinocular.Foreveryline,between35and45plantsperassaywereanalysed.Theresultswereobtainedfromatleastthreeindependentexper-iments.Errorbarsindicatethestandarddeviationbetweenthesethreeexperiments.Thevaluesofthemutantlinesarepresentedinrelationtothefrequencyofthecorrespondingwild-typecontrolplants(100%).-valueswerecalculatedfromtwo-tailed,paired-testsbetweenthewildtypeandthecorrespondingmutantline.SigniÞcantdifferences,deÞnedashaving0.001(***),=0.001Ð0.01(**),=0.01Ð0.05(*)and&#x-365;&#x.400;0.05(noasterisk=notsigniÞcant)areindicatedbyasterisksabovethecorrespondingbars.TheauthorswishtothankMarenScheidleandMandyMeierfortechnicalassistanceandManfredFockeforreadingthemanuscript.TheworkwasfundedbytheERCAdvancedGrantÔCOMRECÕandtheDFGgrantPu137-8.Abe,K.,Osakabe,K.,Nakayama,S.,Endo,M.,Tagiri,A.,Todoriki,S.,Ichikawa,H.andToki,S.(2005)ArabidopsisRAD51Cgeneisimportantforhomologousrecombinationinmeiosisandmitosis.PlantPhysiol.139,896Ð908.Alonso,J.M.,Stepanova,A.N.,Leisse,T.J.etal.(2003)Genome-wideinser-tionalmutagenesisofArabidopsisthaliana.,653Ð657.Bleuyard,J.Y.andWhite,C.I.(2004)TheArabidopsishomologueofXrcc3playsanessentialroleinmeiosis.EMBOJ.,439Ð449.Bleuyard,J.Y.,Gallego,M.E.,Savigny,F.andWhite,C.I.(2005)DifferingrequirementsfortheArabidopsisRad51paralogsinmeiosisandDNAPlantJ.,533Ð545.Bogdanove,A.J.andVoytas,D.F.(2011)TALeffectors:customizableproteinsforDNAtargeting.,1843Ð1846.Bundock,P.andHooykaas,P.(2002)Severedevelopmentaldefects,hyper-sensitivitytoDNA-damagingagents,andlengthenedtelomeresinArabi-dopsisMRE11mutants.PlantCell,2451Ð2462.Bzymek,M.,Thayer,N.H.,Oh,S.D.,Kleckner,N.andHunter,N.(2010)DoubleHollidayjunctionsareintermediatesofDNAbreakrepair.,937ÐCarroll,D.(2011)Genomeengineeringwithzinc-Þngernucleases.,773Ð782.Ceballos,S.J.andHeyer,W.D.(2011)FunctionsoftheSnf2/Swi2familyRad54motorproteininhomologousrecombination.Biochim.Biophys.Acta,509Ð523.Charbonnel,C.,Gallego,M.E.andWhite,C.I.(2010)Xrcc1-dependentandKu-dependentDNAdouble-strandbreakrepairkineticsinArabidopsisplants.PlantJ.,280Ð290.Charbonnel,C.,Allain,E.,Gallego,M.E.andWhite,C.I.(2011)KineticanalysisofDNAdouble-strandbreakrepairpathwaysinArabidopsis.DNARepair,611Ð619.Chen,I.P.,Mannuss,A.,Orel,N.,Heitzeberg,F.andPuchta,H.(2008)AhomologofScRAD5isinvolvedinDNArepairandhomologousrecombi-nationinArabidopsis.PlantPhysiol.,1786Ð1796.Clough,S.J.andBent,A.F.(1998)Floraldip:asimpliÞedmethodforAgro-bacterium-mediatedtransformationofArabidopsisthaliana.PlantJ.Daoudal-Cotterell,S.,Gallego,M.E.andWhite,C.I.(2002)TheplantRad50-Mre11proteincomplex.FEBSLett.,164Ð166.DePiccoli,G.,Torres-Rosell,J.andAragon,L.(2009)Theunnamedcomplex:whatdoweknowaboutSmc5-Smc6?ChromosomeRes.,251Ð263.Fauser,F.,Roth,N.,Pacher,M.,Ilg,G.,Sanchez-Fernandez,R.,Biesgen,C.andPuchta,H.(2012)Inplantagenetargeting.Proc.NatlAcad.Sci.USA,7535Ð7540.Gherbi,H.,Gallego,M.E.,Jalut,N.,Lucht,J.M.,Hohn,B.andWhite,C.I.HomologousrecombinationinplantaisstimulatedintheabsenceofEMBORep.,287Ð291.Recombinationfactorsandpathwaysofhomologousdouble-strandbreakrepair2012TheAuthorsThePlantJournal2012BlackwellPublishingLtd,ThePlantJournal,(2012),,781Ð790 Hajdukiewicz,P.,Svab,Z.andMaliga,P.(1994)Thesmall,versatilepPZPfamilyofAgrobacteriumbinaryvectorsforplanttransformation.PlantMol.,989Ð994.Hartung,F.,Suer,S.,Bergmann,T.andPuchta,H.(2006)TheroleofAtMUS81inDNArepairanditsgeneticinteractionwiththehelicaseAtRecQ4A.cleicAcidsRes.,4438Ð4448.Hartung,F.,Suer,S.andPuchta,H.(2007)TwocloselyrelatedRecQhelicaseshaveantagonisticrolesinhomologousrecombinationandDNArepairinArabidopsisthaliana.Proc.NatlAcad.Sci.USA,18836Ð18841.Hartung,F.,Suer,S.,Knoll,A.,Wurz-Wildersinn,R.andPuchta,H.Topoisomerase3alphaandRMI1suppresssomaticcrossoversandareessentialforresolutionofmeioticrecombinationintermediatesinArabid-opsisthaliana.PLoSGenet.,e1000285.Heacock,M.,Spangler,E.,Riha,K.,Puizina,J.andShippen,D.E.MolecularanalysisoftelomerefusionsinArabidopsis:multiplepathwaysforchromosomeend-joining.EMBOJ.,2304Ð2313.Knoll,A.andPuchta,H.(2011)TheroleofDNAhelicasesandtheirinteractionpartnersingenomestabilityandmeioticrecombinationinplants.J.Exp.,1565Ð1579.Knoll,A.,Higgins,J.D.,Seeliger,K.,Reha,S.J.,Dangel,N.J.,Bauknecht,M.,Schropfer,S.,Franklin,F.C.andPuchta,H.(2012)TheFanconiAnemiaOrthologFANCMEnsuresOrderedHomologousRecombinationinBothSomaticandMeioticCellsinArabidopsis.PlantCell,1448Ð1464.Li,W.,Chen,C.,Markmann-Mulisch,U.,Timofejeva,L.,Schmelzer,E.,Ma,H.andReiss,B.(2004)TheArabidopsisAtRAD51geneisdispensableforvegetativedevelopmentbutrequiredformeiosis.Proc.NatlAcad.Sci.,10596Ð10601.Liu,J.,Renault,L.,Veaute,X.,Fabre,F.,Stahlberg,H.andHeyer,W.D.Rad51paraloguesRad55-Rad57balancetheantirecombinaseSrs2inRad51Þlamentformation.,245Ð248.Mannuss,A.,Dukowic-Schulze,S.,Suer,S.,Hartung,F.,Pacher,M.andPuchta,H.(2010)RAD5A,RECQ4A,andMUS81havespeciÞcfunctionsinhomologousrecombinationanddeÞnedifferentpathwaysofDNArepairinArabidopsisthaliana.PlantCell,3318Ð3330.Masson,J.Y.,Tarsounas,M.C.,Stasiak,A.Z.,Stasiak,A.,Shah,R.,McIlwraith,M.J.,Benson,F.E.andWest,S.C.(2001)IdentiÞcationandpuriÞcationoftwodistinctcomplexescontainingtheÞveRAD51paralogs.GenesDev.Orel,N.,Kyryk,A.andPuchta,H.(2003)Differentpathwaysofhomologousrecombinationareusedfortherepairofdouble-strandbreakswithintan-demlyarrangedsequencesintheplantgenome.PlantJ.,604Ð612.Osakabe,K.,Abe,K.,Yamanouchi,H.etal.(2005)ArabidopsisRad51Bisimportantfordouble-strandDNAbreaksrepairinsomaticcells.PlantMol.,819Ð833.Osakabe,K.,Abe,K.,Yoshioka,T.,Osakabe,Y.,Todoriki,S.,Ichikawa,H.,Hohn,B.andToki,S.(2006)IsolationandcharacterizationoftheRAD54genefromArabidopsisthaliana.PlantJ.,827Ð842.Osakabe,K.,Osakabe,Y.andToki,S.(2010)Site-directedmutagenesisinArabidopsisusingcustom-designedzincÞngernucleases.Proc.NatlAcad.Sci.USA,12034Ð12039.Paques,F.andHaber,J.E.(1999)Multiplepathwaysofrecombinationinducedbydouble-strandbreaksinSaccharomycescerevisiae.Microbiol.Mol.Biol.,349Ð404.Petolino,J.F.,Worden,A.,Curlee,K.,Connell,J.,StrangeMoynahan,T.L.,Larsen,C.andRussell,S.(2010)ZincÞngernuclease-mediatedtransgenePlantMol.Biol.,617Ð628.Prakash,R.,Satory,D.,Dray,E.etal.(2009)YeastMph1helicasedissociatesRad51-madeD-loops:implicationsforcrossovercontrolinmitoticrecom-GenesDev.,67Ð79.Puchta,H.(1998)Repairofgenomicdouble-strandbreaksinsomaticplantcellsbyone-sidedinvasionofhomologoussequences.PlantJ.,331Ð340.Puchta,H.(2005)Therepairofdouble-strandbreaksinplants:mechanismsandconsequencesforgenomeevolution.J.Exp.Bot.,1Ð14.Puchta,H.andHohn,B.(2010)Breakingnews:plantsmutaterightontarget.Proc.NatlAcad.Sci.USA,11657Ð11658.Puchta,H.,Dujon,B.andHohn,B.(1993)HomologousrecombinationinplantcellsisenhancedbyinvivoinductionofdoublestrandbreaksintoDNAbyasite-speciÞcendonuclease.NucleicAcidsRes.,5034Ð5040.Puchta,H.,Dujon,B.andHohn,B.(1996)Twodifferentbutrelatedmecha-nismsareusedinplantsfortherepairofgenomicdouble-strandbreaksbyhomologousrecombination.Proc.NatlAcad.Sci.USA,5055Ð5060.Puizina,J.,Siroky,J.,Mokros,P.,Schweizer,D.andRiha,K.(2004)Mre11deÞciencyinArabidopsisisassociatedwithchromosomalinstabilityinsomaticcellsandSpo11-dependentgenomefragmentationduringmeio-PlantCell,1968Ð1978.Seeliger,K.,Dukowic-Schulze,S.,Wurz-Wildersinn,R.,Pacher,M.andPuc-hta,H.(2012)BRCA2isamediatorofRAD51-andDMC1-facilitatedhomologousrecombinationinArabidopsisthaliana.NewPhytol.,364ÐShaked,H.,Avivi-Ragolsky,N.andLevy,A.A.(2006)InvolvementoftheArabidopsisSWI2/SNF2chromatinremodelinggenefamilyinDNAdam-ageresponseandrecombination.,985Ð994.Shukla,V.K.,Doyon,Y.,Miller,J.C.etal.(2009)PrecisegenomemodiÞcationinthecropspeciesZeamaysusingzinc-Þngernucleases.,437ÐSiebert,R.andPuchta,H.(2002)EfÞcientrepairofgenomicdouble-strandbreaksbyhomologousrecombinationbetweendirectlyrepeatedse-quencesintheplantgenome.PlantCell,1121Ð1131.Swoboda,P.,Gal,S.,Hohn,B.andPuchta,H.(1994)Intrachromosomalhomologousrecombinationinwholeplants.EMBOJ.,484Ð489.Szostak,J.W.,Orr-Weaver,T.L.,Rothstein,R.J.andStahl,F.W.(1983)Thedouble-strand-breakrepairmodelforrecombination.,25Ð35.Tan,G.,Gao,Y.,Shi,M.,Zhang,X.,He,S.,Chen,Z.andAn,C.(2005)Site-Finding-PCR:asimpleandefÞcientPCRmethodforchromosomewalking.NucleicAcidsRes.,e122.Townsend,J.A.,Wright,D.A.,Winfrey,R.J.,Fu,F.,Maeder,M.L.,Joung,J.K.andVoytas,D.F.(2009)High-frequencymodiÞcationofplantgenesusingengineeredzinc-Þngernucleases.,442Ð445.Uanschou,C.,Siwiec,T.,Pedrosa-Harand,A.,Kerzendorfer,C.,Sanchez-Moran,E.,Novatchkova,M.,Akimcheva,S.,Woglar,A.,Klein,F.andSchlogelhofer,P.(2007)AnovelplantgeneessentialformeiosisisrelatedtothehumanCtIPandtheyeastCOM1/SAE2gene.EMBOJ.,5061Ð5070.Watanabe,K.,Pacher,M.,Dukowic,S.,Schubert,V.,Puchta,H.andSchubert,(2009)Thestructuralmaintenanceofchromosomes5/6complexpro-motessisterchromatidalignmentandhomologousrecombinationafterDNAdamageinArabidopsisthaliana.PlantCell,2688Ð2699.Wei,W.,Ba,Z.,Gao,M.,Wu,Y.,Ma,Y.,Amiard,S.,White,C.I.,RendtlewDanielsen,J.M.,Yang,Y.G.andQi,Y.(2012)AroleforsmallRNAsinDNAdouble-strandbreakrepair.,101Ð112.Zhang,F.,Maeder,M.L.,Unger-Wallace,E.etal.(2010)Highfrequencytar-getedmutagenesisinArabidopsisthalianausingzincÞngernucleases.Proc.NatlAcad.Sci.USA,12028Ð12033.NadineRothetal.2012TheAuthorsThePlantJournal2012BlackwellPublishingLtd,ThePlantJournal,(2012),,781Ð790