Proc.NatI.Acad.Sci.USAVol.82,pp.503-505,January1985GeneticsRepairofdef - PDF document

Proc.NatI.Acad.Sci.USAVol.82,pp.503-505,January1985GeneticsRepairofdefinedsinglebase-pairmismatchesinEscherichiacoli(bacteriophageX/heteroduplex/mutagenesis/geneconversion)CHRISTIANEDOHET*,ROBERTWAGNER*,ANDMIROSLAVRADMAN*t*InstitutJ.Monod,CentreNationaldelaRechercheScientifique,UniversitdParis7,Tour43,2placeJussieu,75251Paris,Cedex05,France;andtD~partementdeBiologieMoldculaire,UniversitdlibredeBruxelles,B1640Rhode-St-Gen~se,BelgiumCommunicatedbyMatthewMeselson,September27,1984ABSTRACTHeteroduplexeswithsinglebase-pairmis-matchesofknownsequencewerepreparedbyannealingsepa-ratedstrandsofbacteriophageXDNAandusedtotransfectEscherichiacoli.Eachoftheeightpossiblesinglebase-pairmismatcheswasconstructed.GeneticanalysisoftheprogenyphagesobtainedfromtransfectedbacteriaindicatesthattheE.colimismatchrepairsystemdoesnotrecognize(ordoesnotrepair)allsinglebase-pairmismatcheswithequalefficiency.Inparticular,theA-G,CAT,andC-Cmismatchesappeartobelessrepairedthananyothers.Themutatorcharacterofmis-matchrepair-deficientmutantssuggeststhatsuchunrepairedmismatchesshouldoccurinfrequentlyduringE.coliDNArep-lication.MismatchrepairisbelievedtocorrecterrorsarisingduringDNAreplicationbyremovingincorrect-i.e.,mispaired-basesfromnewlysynthesizedDNAstrands(1,2).Themis-matchrepairsysteminEscherichiacolidoesnotappeartoactinregionsofDNAwhereG-A-T-Csequencesarefullyadeninemethylated(3)anditappearstobethetransientun-dermethylationofnewlysynthesizedG-A-T-Csequencesintheregionimmediatelyfollowingthereplicationforkthatal-lowsmismatchrepairtooccuronlyonnewlysynthesizedstrands(1-3).Repairappearstoinvolvelocalizedexcisionandresynthesis(1,4,5).SomemutatormutantsofE.coli[mutU,mutL,mutS,mutU(uvrD,uvrE)]havebeenshowntobedeficientinmis-matchrepair(3,6-9).Themut-dependentincreaseoffor-wardmutationorreversionfrequenciesobservedforseveralmarkersdependsonboththemarkerandthemutmutation(8,10),suggestingthattheE.colimismatchrepairsystemmayrecognizeorrepaircertainmismatchesmoreefficientlythanothers.ItappearsthatsuchisthecaseforthemismatchrepairsystemofStreptococcuspneumoniaewheretwomis-matches,AKGandC-C,donotappeartoberepaired(11).TheresultsofexperimentsusingheteroduplexespreparedfromseparatedstrandsofbacteriophageXDNAtotransfectE.colicellssuggestthatatleastonemismatchmaynotberepairedinE.coliaswell,althoughthenatureandsequenceofthemismatchwerenotknown(3).TodeterminethespecificityoftheE.colimismatchrepairsystem,wehavemeasuredtheamountofrepairofmis-matchesinheteroduplexespreparedfromseparatedstrandsofunmethylatedbacteriophageXDNAcontainingse-quencedmutationsintheXcIgene.Mutantswerechosentoallowtheformation,inmutant/wild-typeheteroduplexes,ofeachoftheeightpossiblesinglebase-pairmismatches-ARC,A-G,A-A,T-G,T-C,T-T,C-C,andG-G.Theresultsindicatethatallmismatchesarenotrecognized(ornotrepaired)withequalefficiency.Inparticular,theA-G,COT,andCCmismatchesappeartobelessrepairedthananyothers.MATERIALSANDMETHODSXphageswithsequencedcImutationswereobtainedfromFranklinHutchinson(YaleUniversity)(Fig.1).Theoriginalphageswerec+(orcI)andalsocarriedthemarkersc1857ind-Oam29.ThephagesusedintheseexperimentsweremadeObyrecombination.cImutationswereoriginallyse-quenced,inthelaboratoryofFranklinHutchinson(12),bythemethodofMaxamandGilbert(20).DNAstrandswerepreparedandannealedasdescribed(1).Transfectionwasac-complishedbythemethodofMandelandHiga(13).Trans-fectionefficiencywasintherangeof5x10-8to5x10-6transfectedcellsperDNAheteroduplexatamultiplicityofabout0.1DNAmoleculepercell.TransfectionefficiencyvarieswithDNApreparationandbacterialstrainbutdoesnotappeartoaffecttheresults-i.e.,transfectionefficiencydoesnotappeartoberelatedtotheextentofmismatchre-pair.Transfectedcellswereplatedbeforelysis,givingrisetoinfectivecenters.TominimizeanyselectiveadvantageordisadvantageassociatedwiththecImutations,infectivecen-tersweregrownunderconditions(370C)inwhichallphagesusedintheseexperimentsgrowlyticallyduetothepresenceofthetemperature-sensitivec1857mutation.Individualin-fectivecenterswerepickedandthephageswerereplatedat320CtodeterminetheirgenotypeatcI.Infectivecenterswerescoredasclear(C)-i.e.,containingonlyphageswiththecImutation;turbid(C+)-i.e.,containingonlyphageswiththewild-typealleleofthecImutation;ormixed(C+/C)-i.e.,containingbothwild-typeandmutantphages.Theresultsofcontrolexperimentsinwhichinfectivecentersderivedfromcellsinfectedwithphageshavingonlythewild-typealleleofthecIgenewereanalyzedrevealthat-1%ofturbidinfectivecenterswillbescoredasmixedbecauseofforwardmutationofC+toC.Thissetsthelowerlimitofmixedinfectivecentersthatwillberecoveredintheseex-perimentssomewherebetween0.5%and1%.ReversionsofthecImutationsaremuchtooinfrequenttointerferewithscoring.Atleast350infectivecentersweretestedforeachheteroduplex.DifferentDNApreparationsofthesamehe-teroduplexgiveessentiallyidenticalresults.RESULTSANDDISCUSSIONHeteroduplexeswerepreparedfromseparatedstrandsoffourdifferentXcImutantsandtheirrespectivewild-typeal-lelestocreateeachoftheeightpossiblesinglebase-pairmis-matches(Table1).Themutationswerechosentofallwithinasmall(35-base-pair)regionofthecIgene(Fig.1).TheDNAwasisolatedfromphagesgrownindamE.coli(GM33-deficientinadeninemethylation;ref.17)sothatrepaironeitherstrandcouldbeobserved,allowinganystrandprefer-enceforrepairtobedetected.Transfectionsofdamandwild-typecellsgiveessentiallyidenticalresults(notshown),suggestingthat,inthissystem,methylationduringtransfectionisnotrapidenoughtoaffecttheresults.Asimilarresultwasobtainedbyothersutilizinghelperphage-mediatedtransfection(3).Theresultsoftransfectionswiththeeightdifferenthetero-duplexesarepresentedinTable1.Threetypesofinfective503Thepublicationcostsofthisarticleweredefrayedinpartbypagechargepayment.Thisarticlemustthereforebeherebymarked"advertisement"inaccordancewith18U.S.C.§1734solelytoindicatethisfact. Proc.NatLAcadSci.USA82(1985)Table1.GeneticanalysisofinfectivecentersderivedfromheteroduplexescontainingdefinedmismatchesTransfectedbacteriaHeteroduplexWildtype*mutLtMismatchMixed,%Clear,%Turbid,%TotalMixed,%Clear,%Turbid,%TotalpositionMismatch(C+/C)(C)(C+)number(C+/C)(C)(C+)number26c+1-G-5.639.355.148385.17.57.5643c-T-r26c-A-110.353.935.845877.316.66.1608c-C-r11c+-T-16.931.361.861183.06.110.8526c-T-r11c-A-112.257.929.969985.410.73.8596c+A-r45c+G-16.128.865.1101878.37.114.5434c-G-r45c-CI-39.436.624.086783.410.46.2356c+-C-r15c+-C-i80.06.013.949678.47.713.9504c-T-r15c-A-172.715.511.898081.213.75.0517c+GrThesourceof"c+"strands(bothIandr)forallheteroduplexeswasXc1857indI(sequence,ref.14).Xc1857indIphageswithadditionalcImutationswereusedtoprovide"c"strands.ThemutantsarespontaneoussinglebasesubstitutionsisolatedandsequencedinthelaboratoryofFranklinHutchinson(12).Mismatchpositionsareasfollows:position26,SP27,G-C-*A-Ttransition;position11,SP62,A-T-*T-Atransversion;position45,SP51,G-C-*C-Gtransversion;position15,SP57,G-C-*T-Atransversion.*E.coliC600(15).tE.coliES871(16).centersareobserved:(i)thoseyieldingonlyc+phages(pureturbid),(ii)thoseyieldingonlycIphages(pureclear),and(iii)thoseyieldingbothc+andcIphages(mixed).Pureinfec-tivecentersmayresultfrommismatchrepair,fromstrandloss(1),orfromcontaminatedstrandpreparations(transfec-tionswithself-annealedsinglestrandpreparationsrevealthatistrandpreparationsarefrequentlycontaminatedwithrstrands).Mixedinfectivecentersreflecteithertheabsenceofbothmismatchrepairandstrandloss,thechancesuperpo-sitionoftwotransfectedcellsinoneinfectivecenter,ortheuptakebyonecellofmorethanoneDNAmolecule.Earlierworkwithmultiplymismatchedheteroduplexesindicatesthatthechancesuperpositionoftransfectedcellsisarela-tivelyrareevent(1).TransfectingwithanequalmixtureofcIandc+homoduplexDNAmoleculesatamultiplicity10timeshigherthanthatusedforheteroduplexesproducednomorethan1%mixedinfectivecenters,mostofwhichpre-sumablyresultedfromforwardmutationsofc+tocI(seeMaterialsandMethods).Thus,mixedinfectivecentersde-riveprimarilyfromheteroduplexesthathaveescapedmis-matchrepair.Therelativefractionsofmixedandpureinfectivecentersaredeterminedby(i)theextentofmismatchrepair,(ii)theamountofstrandlossinheteroduplexesthathavenotbeenrepaired,and(iii)theamountofcontaminatinghomoduplexmolecules.AnupperlimitfortheamountofstrandlossandstrandcontaminationcanbeestimatedfromtheresultsofheteroduplextransfectionsinmutLbacteria,inwhichmis-matchrepairhasbeenshowntobeabsentoratleastmuchlessefficientthaninwild-typebacteria(3,5,7,9).Intheseexperimentspureinfectivecentersaccountforonly15-23%ofthetotalnumberofinfectivecenters,comparedtovaluesashighas94%whenthetransfectedcellsarewildtypewithrespecttomismatchrepair(Table1).Thefindingofmis-matchesthatdonotappeartoberepaired-i.e.,onesyield-ingalargefractionofmixedinfectivecentersinwild-typetransfections-indicatesthatthedifferencebetweenmutLandwild-typetransfectionsisnotduetoadifferenceintheextentofstrandloss.Thus,evenifitisassumedthatallpureinfectivecentersfromtransfectionsinmutLcellsarethere-sultofstrandlossorstrandcontamination,itdoesnotap-pearthatthesephenomenasubstantiallydiminishthefrac-tionofmixedinfectivecentersrecoveredunderconditionsinwhichmismatchrepaircanoccur.Theamountofstrandlossobservedintheseexperiments(nomorethan15-25%oftransfectingheteroduplexes)islessthanthatobservedinearlierexperiments(1)usinghelperphage-mediatedtrans-fection(60-70%).Wehavenoexplanationforthisdifferenceexcepttosuggestthatitmightreflectsomedifferenceinthereplication(symmetricvs.asymmetric)oftheheterodu-plexesinthetwotransfectionprocedures-i.e.,itmaybethatthepreinfectingXhelperphagesproduceXreplicationproteinssuchthattransfectingheteroduplexesinitiaterollingcirclereplicationmorerapidlythaninCaCl2-mediatedtrans-fections.ThedatapresentedinTable1indicatethatthemismatchesT*G,C*A,T'T,A-A,andG-Garerepairedatreasonablyhighefficiencyinwild-typebacteria-i.e.,mixedinfectivecenters.G-A,T-C,andCCmismatchesarepoorlyrepaired;G-AandT-Cmismatchesmaybecompletelyrefractoryto111526455ATGAGCACAAAAAAGAAACCATTAACACAAGAGCAGCTTGAGGACGCACG3TACTCGTGTTTTTTCTTTGGTAATTGTGTTCTCGTCGAACTCCTGCGTGCTTAATAGCFIG.1.DNAsequenceoftheregionoftheXcIgenecodingfortheamino-terminalendoftheXrepressor(14).Locationsandbasechangesofthemutationsusedinthisstudyareindicated.NumbersindicatethenumberofbasesfromthestartoftheCIPRMtranscript.504Genetics:DohetetaL Proc.NatLAcadSciUSA82(1985)505repair.Thus,itappearsthattheE.colimismatchrepairsys-temdoesnotrecognizeorrepairallmismatcheswithequalefficiency.Itisnotcleartowhatextenttheresultsareaffectedbynucleotidesequencecontext.However,thedataindicatethatneighboringnucleotidesequencecannotbethesolede-terminantofmismatchrepairefficiencysincethereexistsareciprocalpairofmismatches,whicharenecessarilyatthesamesiteintheDNA,inwhichonemismatch(GIG)iswellrepairedandtheother(C-C)isnot(Table1).ItmaybethatthespecificityoftheE.coliandS.pneumo-niaemismatchrepairsystemsisidentical,asitappearsthatinneithersystemareG-AandC-Cmismatcheswellrepaired(11).However,therearedifferingreportsoftheamountofrepairofCOTmismatchesinS.pneumoniae(11,18).Itispossiblethatmismatchesdifferwithrespecttotheamountoftimerequiredtorecognizeorrepairthem.ItmaybethatduringE.coliDNAreplicationmoretimeisavailableformismatchrepairthaninourexperimentalsystem,suchthatallmismatchesarerecognizedandrepaired.However,itseemsmostlikelythatthetimeavailableformismatchre-pairduringE.coliDNAreplication-i.e.,thetimebeforeadeninemethylationofnewlysynthesizedG-A-T-Cse-quences-isnolongerthanthetimeavailableformismatchrepairintheseexperiments-i.e.,thetimebetweenentryofheteroduplexDNAintothetransfectedcellsandinitiationofthefirstroundofXreplication.Thedatarevealnomismatch-dependentstrandpreferenceforrepair.However,theredoesappeartobeaconsistentbiasamongpureburstsinfavoroftstrandgenotypes(Table1).Thesourceofthisbiasisnotknown.Inthecaseoftrans-fectionsofmutLbacteria,itmaybeaccountedforbyIstrandcontamination,preferentialstrandloss,orresidualrepair.Intransfectionsofwild-typebacteriawithheteroduplexescon-tainingrepairablemismatches,inwhichpureburstsarepri-marilytheresultofmismatchrepair,thebiasmaybeduetopreferentialrepairofthenon-transcribedstrand-i.e.,itmaybethattranscriptionenzymescompetewithmismatchrepairenzymesforspaceonthetranscribedstrandsuchthatrepairismorecommononthenon-transcribedstrand(rstrandinthecIregion).ThefindingthattheE.colimismatchrepairsystemreadilyrecognizesandrepairsonlythreeofthesixtransversionmis-matches(A-A,TUT,andG-G)accountsforthefindingsthatthemutatoreffectsobservedinmutH,mutL,mutS,andmutUmutants,whicharedeficientinmismatchrepair,anddammutants,whichhaveundirectedmismatchrepair,aredueprimarilytoanincreaseintransitionandframeshiftmu-tations(10,19).[Wehavefoundthatasinglebaseadditionframeshiftmismatchcanberecognizedandrepairedonei-therstrandbytheE.colimismatchrepairsystem(21).]Giv-entheobservedpatternofmismatchrepairandthefactthatmutanddammutantsaremutators,itappearsthat,inE.coli,transversionmismatcheseitherarisemuchlessfre-quentlyduringDNAreplicationthantransitionmismatchesorarecorrectedbysomerepairsystem,perhapspolymer-ase-associated(e.g.,proofreading),whichactsbeforethemismatchrepairwearestudying.AsimilarsituationmayexistinS.pneumoniae,inwhichmismatchrepairhasalsobeenfoundtorepairthetransitionmismatchesbutonlysomeofthetransversionmismatches(11,18).Dr.F.Hutchinsongenerouslyprovidedthesequencedmutantsusedinthisstudy.R.W.istherecipientofaseniorfellowshipfromLigueNationaleFrangaisecontreleCancer.Thisworkwassupport-edbyLigueNationaleFrangaisecontreleCancerandagrantfromAssociationpourleDdveloppementdelaRecherchesurleCancer.1.Wagner,R.E.&Meselson,M.(1976)Proc.Natl.Acad.Sci.USA73,4135-4139.2.Radman,M.,Wagner,R.E.,Glickman,B.W.&Meselson,M.(1980)inProgressinEnvironmentalMutagenesis,ed.Ala-cevic,M.(Elsevier,Amsterdam),pp.121-130.3.Pukkila,P.J.,Peterson,J.,Herman,G.,Modrich,P.&Mesel-son,M.(1983)Genetics104,571-582.4.Wildenberg,J.&Meselson,M.(1975)Proc.Natl.Acad.Sci.USA72,2202-2206.5.Lu,A.L.,Clark,S.&Modrich,P.(1983)Proc.Natl.Acad.Sci.USA80,4639-4643.6.Nevers,P.&Spatz,H.(1975)Mol.Gen.Genet.139,233-243.7.Rydberg,B.(1978)Mutat.Res.52,11-24.8.Glickman,B.W.&Radman,M.(1980)Proc.Nat!.Acad.Sci.USA77,1063-1067.9.Radman,M.,Dohet,C.,Bourguignon,M.F.,Doubleday,0.P.&Lecomte,P.(1981)inChromosomeDamageandRe-pair,eds.Seeberg,E.&Kleppe,K.(Plenum,NewYork),pp.431-445.10.Cox,E.C.(1976)Annu.Rev.Genet.10,135-156.11.Claverys,J.P.,Mejean,V.,Gasc,A.M.&Sicard,M.(1983)Proc.Nat!.Acad.Sci.USA80,5956-5960.12.Skopek,T.R.&Hutchinson,F.(1982)J.Mol.Biol.159,19-33.13.Mandel,M.&Higa,A.(1970)J.Mol.Biol.53,159-162.14.Sauer,R.T.(1978)Nature(London)276,301-302.15.Appleyard,R.K.(1954)Genetics39,440-452.16.Vaccaro,K.K.&Siegel,E.C.(1977)Mutat.Res.42,443-446.17.Marinus,M.G.&Morris,N.R.(1975)Mutat.Res.28,15-26.18.Lacks,S.,Dunn,J.J.&Greenberg,B.(1982)Cell31,327-336.19.Glickman,B.W.(1979)Mutat.Res.61,153-162.20.Maxam,A.&Gilbert,W.(1977)Proc.Nat!.Acad.Sci.USA74,560-564.21.Wagner,R.,Dohet,C.,Jones,M.,Doutriaux,M.-P.,Hutchin-son,F.&Radman,M.(1984)ColdSpringHarborSymp.Quant.Biol.49,inpress.Genetics:DohetetaL