The repair of DNA methylation damage in Saccharomyces cerevisiae

Curr Genet(1996)30:461—468 Springer-Verlag1996 ORIGINAL PAPER

Wei Xiao·Barbara L.Chow·Lane Rathgeber

The repair of DNA methylation damage in Saccharomyces cerevisiae Received:13June/24July1996

Abstract The major genotoxicity of methyl meth-anesulfonate(MMS)is due to the production of a lethal 3-methyladenine(3MeA)lesion.An alkylation-speci?c base-excision repair pathway in yeast is initiated by a Mag13MeA DNA glycosylase that removes the damaged base,followed by an Apn1apurinic/ apyrimidinic endonuclease that cleaves the DNA strand at the abasic site for subsequent repair.MMS is also regarded as a radiomimetic agent,since a number of DNA radiation-repair mutants are also sensitive to MMS.To understand how these radiation-repair genes are involved in DNA methylation repair,we performed an epistatic analysis by combining yeast mag1and apn1 mutations with mutations involved in each of the RAD3,RAD6and RAD52groups.We found that cells carrying rad6,rad18,rad50and rad52single mutations are far more sensitive to killing by MMS than the mag1 mutant,that double mutants were much more sensitive than either of the corresponding single mutants,and that the e?ects of the double mutants were either addi-tive or synergistic,suggesting that post-replication and recombination-repair pathways recognize either the same lesions as M AG1and A PN1,or else some di?er-ent lesions produced by MMS treatment.Lesions han-dled by recombination and post replication repair are not simply3MeA,since over-expression of the MAG1 gene does not o?set the loss of these pathways.Based on the above analyses,we discuss possible mechanisms for the repair of methylation damage by various path-ways.

Key words DNA repair·Methylation damage·Epistasis analysis·Saccharomyces cerevisiae

W.Xiao()·B.L.Chow·L.Rathgeber

Department of Microbiology,University of Saskatchewan, Saskatoon,SK,S7N5E5,Canada

Communicated by B.Kunz Introduction

Simple methylating agents,such as methyl methanesul-

fonate(MMS)and dimethyl sulfate,methylate DNA in

vivo and in vitro mainly at7-deoxyguanine and3-

deoxyadenine.The resulting3-methyladenine(3MeA)

blocks DNA synthesis and is thus considered a lethal lesion.In addition,S,1-type methylating agents,such as N-methyl-N -nitro-N-nitrosoguanidine(MNNG),

also form a signi?cant portion of O-alkyl lesions,such

as O -methylguanine(O MeG)and O -methyl-

thymine(O MeT),which are considered to be more

mutagenic and carcinogenic(reviewed in Beranek

(1990).Ubiquitous alkylation-speci?c repair pathways

include an O MeG DNA-repair methyltransferase

(MTase)that removes the methyl group from O MeG

and/or O MeT,and a3MeA DNA glycosylase

that removes3MeA lesions by cleaving the base-

sugar N-glycosyl bond.The resulting abasic site is subject to cleavage by an apurinic/apyrimidinic (AP)endonuclease,followed by a base-excision repair process,which most likely requires deoxyribophos-phodiesterase,polymerase,and ligase in order to re-store DNA structure(Barnes et al.1993;Friedberg et al.1995).

In the yeast Saccharomyces cerevisiae,three genes

coding for the Mag13MeA DNA glycosylase(Chen

et al.1989,1990;Berdal et al.1990),the Apn1AP

endonuclease(Popo?et al.1990;Ramotar et al.1991)

and the Mgt1O MeG MTase(Xiao et al.1991;Xiao

and Samson1992)have been isolated and the deletion mutants characterized.Mutations in these genes result-ed in an enhanced sensitivity to alkylation-induced killing and/or mutagenesis but not to radiation dam-age.In addition,apn1(Ramotar et al.1991)and mgt1 (Xiao and Samson1992,1993)mutants also showed an increased spontaneous mutation rate,suggesting an endogenous source of DNA alkylation.Epistasis analy-sis of combined double mutants con?rmed two inde-pendent DNA-alkylation repair pathways initiated by

the DNA-repair MTase and the3MeA glycosylase (Xiao and Samson1993).

However,the above-mentioned DNA-alkylation re-pair pathways do not appear to be the only cellular mechanisms for the processing of DNA-methylation damage.Additional mutants have been isolated that are sensitive to MMS,but not to UV and X-rays (Prakash and Prakash1977).Furthermore,mutants whose genes are involved in DNA radiation-repair pathways exhibit an enhanced sensitivity to simple alkylating agents.These mutants include rad1,rad2, rad4,rad10and rad14in the RAD3group,rad50,rad51, rad52,rad54and rad57in the RAD52group,and rad5, rad6,rad8,rad18and rev3in the RAD6group(Fried-berg1988).RAD1,RAD2,RAD4,RAD10and RAD14 are absolutely required for nucleotide-excision repair. Proteins in the RAD52group are involved in mitotic and/or meiotic recombination and recombination re-pair.The RAD6group genes most likely constitute a post replication-repair and mutagenesis pathway (Haynes and Kunz1981;Prakash et al.1993;Friedberg et al.1995).The deduced Rad18amino-acid sequence predicts a nucleotide-binding protein(Chanet et al. 1988;Jones et al.1988).Rad18functions primarily with Rad5in error-free postreplication repair(Johnson et al. 1992),whereas RE?3encodes a non-essential,repair-speci?c DNA polymerase(Morrison et.al.1989)which, together with Rev7,forms DNA polymerase that bypasses a thymine-thymine dimer(Nelson et al.1996). RAD6encodes a ubiquitin-conjugating enzyme (Jentsch et al.1987)that functions in both post replica-tion repair and mutagenesis(Prakash et al.1993).In vitro experiments(Bailly et al.1994)demonstrated that Rad6and Rad18form a complex which is most likely targeted to single-stranded DNA via the Rad18nucleo-tide-binding activity.

Due to their genotoxicity to a wide spectrum of radiation-repair mutants,methylating agents such as MMS have been referred to as radiomimetic agents. However,it is not clear to date which type of lesion(s) induced by MMS is the target of various repair path-ways.For example,DNA treated with MMS in vitro does not produce an appreciable fraction of strand breaks.Why are RAD52-pathway mutants so sensitive to MMS treatment?We are interested in how these radiation-repair pathways are involved in the repair of DNA-methylation damage.Previous studies with radi-ation repair mutants have been focused on either the genotoxicity of ethylating agents(Cooper and Waters 1987)or induced mutagenesis by methylating agents (Prakash1974).In the present study,we have attem-pted to address two questions:(1)the relative sensitivity of radiation-repair mutants to a given methylating agent;and(2)the mechanism by which these repair pathways participate in the protection against methyla-tion damage.We found that DNA recombination-and post replication-repair pathways contribute signi?-cantly to resistance to killing by DNA-methylating agents,and discuss possible roles these two pathways play in the protection against methylation damage. Materials and methods

Media,growth conditions and genetic methods.Yeast cells were cul-tured at30°C either in a rich YPD medium or in a synthetic dextrose medium supplemented with amino acids and bases as described(Sherman et al.1983);5-?uoroorotic acid(5-FOA)me-dium was also made as described(Boeke et al.1987)to select for Ura\mutants.Yeast transformations were performed by the LiAc method(Ito et al.1983).For targeted integration and gene disrup-tion,plasmid DNA was digested with restriction enzymes and pre-cipitated by cold ethanol prior to transformation.

Plasmids.Prior to yeast transformation,the rad4 ::oRA3cassette was released by digesting pDG38(Gietz and Prakash1988)with Xba I;the rad6 ::?Eo2cassette was released by digesting pDG315 (Kang et al.1992)with Bam HI#Hin dIII;the rad18 ::1RP1cas-sette was released by digesting prad18 ::TRP1(a gift from Dr.A. Morrison)with Eco RI#Bst YI;the rad50 ::hisG-oRA3-hisG cas-sette was released by digesting pNKY83(Alani et al.1989)with Eco RI#Bgl II;the rad52 ::?Eo2cassette was released by digest-ing pBR HS::LEU2(Dornfeld and Livingston1991)with Sna BI-Dra I;and the rev3 ::?Eo2cassette was released by digesting pAM56(a gift from Dr.A.Morrison)with Xba I.Plasmid YEp13A contains the entire MAG1gene in a multicopy plasmid YEp13 (Chen et al.1989).Plasmid pWX1210was made by inserting the 2.26-kb Eco RI fragment from pUC4.0(Chen et al.1989)into the Eco RI site of plasmid YEp123(YEp,oRA3,P "& -1!7! ,Xiao and Fontanie1995)in an orientation such that the MAG1gene is under the control of the ADH1promoter.

?east strains.The S.cerevisiae haploid strains employed are listed in Table1.An isogenic background for all the strains was main-tained by targeted gene disruption in a common parental strain, DBY747,originally obtained from D.Botstein(Stanford University, USA).Strain WXY9216was selected from JC8901on a5-FOA plate and the Ura\phenotype was con?rmed,followed by Southern hybridization to con?rm the mag1::hisG structure.Other rad and rev mutants were made by disrupting these genes in the wild-type, mag1or apn1mutant as described.All the transformants were?rst identi?ed by Southern hybridization for the disruption of targeted genes and then used for phenotypic analyses.

Cell killing and gradient plate assay.Two types of quantitative killing experiments were performed.For a liquid killing experiment, overnight yeast cultures were used to inoculate fresh YPD at a 10-fold dilution.Cells were allowed to grow until about2;10 cells/ml.MMS was added to the culture at a?nal concentration as speci?ed and aliquots were taken at given intervals.Cells from each sample were collected via centrifugation,washed,diluted,and plated on YPD.The colonies were counted after a3-day incubation. Untreated cells were also plated and scored as100%survival.For a gradient plate assay,an MMS gradient was formed by pouring a bottom layer of YPD#MMS medium in a tilted square Petri dish.A series of plates were made,each containing di?erent concen-trations of MMS as indicated in the bottom agar.After agar solidi?-cation,the Petri dish was returned?at and a top layer of YPD medium was added.Cell cultures were printed onto each plate across the gradient using a microscope slide and the plates were incubated for2—3days as speci?ed.Although results from the gradient plate assay may vary by factors such as drug concentration, preparation of plates,cell density and time of incubation,we found that it is more reproducible than the liquid killing and can accomo-date a large number of samples.Each experiment was repeated at

462

Table1Saccharomyces

cerevisiae strains Strain Genotype Source/References

DBY747a his3- 1leu2-3,112trp1-289ura3-52 D.Botstein

JC8901DBY747with mag1::hisG-oRA3-hisG Chen et al.(1990)

WXY9105DBY747with apn1 ::HIS3Xiao and Samson(1993)

WXY9107DBY747with mag1::hisG-oRA3-hisG apn1 ::HIS3Xiao and Samson(1993)

WXY9216DBY747with mag1::hisG This study

WXY9394DBY747with rad4 ::hisG-oRA3-hisG This study

WXY9395DBY747with mag1::hisG and rad4::hisG-oRA3-hisG This study

WXY9376DBY747with rad6 ::?Eo2This study

WXY9377DBY747with mag1::hisG-oRA3-hisG and rad6::?Eo2This study

WXY9378DBY747with apn1 ::HIS3and rad6 ::?Eo2This study

WXY9326DBY747with rad18 ::1RP1This study

WXY9327DBY747with mag1::hisG-oRA3-hisG and rad18 ::1RP1This study

WXY9221DBY747with rad50 ::hisG-oRA3-hisG This study

WXY9323DBY747with mag1::hisG and rad50 ::hisG-oRA3-hisG This study

WXY9324DBY747with apn1 ::HIS3and rad50 ::hisG-oRA3-hisGr This study

WXY9387DBY747with rad52 ::?Eo2This study

WXY9382DBY747with rev3 ::?Eo2This study

WXY9383DBY747with mag1::hisG-oRA3-hisG and rev3 ::?Eo2This study

WXY9384DBY747with apn1 ::HIS3and rev3 ::?Eo2This study

least three times and comparisons were always restricted to the same

set of plates.

Results

Experimental design

It has been well established that treatment of cells with

S,2-type alkylating agents,such as MMS,produces

3MeA as a predominant lethal lesion(Boiteux et al.

1984).The major adduct7-methylguanine is considered

non-toxic and non-mutagenic,except that it may be

depurinated at a higher rate than deoxyguanine(Be-

ranek1990).It is also known that the yeast MAG1gene

encodes a major,if not the only,3MeA DNA

glycosylase(Chen et al.1989).The mag1mutant fails to

remove3MeA from DNA in vitro and renders cells

hypersensitive to MMS in vivo(Chen et al.1989,1990).

Furthermore,the APN1gene encodes the only type-II

AP endonuclease(Popo?et al.1990),which acts in the

same pathway as(and most likely downstream from)

3MeA glycosylase(Xiao and Samson1993).Our goal

in the present study was to examine if other DNA-

repair pathways are involved in the repair of DNA-

methylation damage,and if so,what is the possible

mechanism of such repair.We made double mutants

combining mag1or apn1with mutations in other

DNA-repair epistasis groups and asked whether the

alkylation-induced killing e?ect in these double mu-

tants is additive,synergistic,or remains unchanged

compared with their respective single mutants.If the

double mutant is not more sensitive than either one of

the single mutants,the two genes are likely to be

involved in the same repair pathway.This e?ect is

illustrated in Fig.1by a gradient plate assay,where the

Fig.2MMS-induced killing with mag1and rad4strains by a gradi-ent plate assay.(?),DBY747(wt);(?),WXY9216(mag1);(?), WXY9394(rad4)and(?),WXY9395(mag1rad4).The plates were incubated at30°C for44h

Wang et al.1994).RAD1,RAD2,RAD4,RAD10and RAD14are required for nucleotide-excision repair but not for transcription.Only the rad4mutant was exam-ined in the present study.As shown in Fig.2,the rad4 mutant itself exhibits only a slightly increased sensitiv-ity to MMS on a gradient plate assay.Nevertheless,the mag1rad4double mutant was more sensitive than either of the single mutants,and the e?ect seems to be additive.

Alkylation repair and RAD6-group genes

The RAD6group is heterogeneous and probably con-sists of two major pathways for post replication repair and mutagenesis,respectively.Current observations are consistent with a model that suggests Rad18and Rad5are involved in post replication repair,while Rev1,Rev3and Rev7are involved in mutagenesis,and Rad6plays a critical role in both pathways(Prakash et al.1993;Friedberg et al.1995).

The rad6mutant exhibits markedly increased killing by MMS(Kupiec and Simchen1984).Indeed,the rad6 mutant is more sensitive to MMS than mag1and apn1in a gradient plate assay(Fig.3).Nevertheless, while the MMS sensitivity for rad6and apn1is prob-ably additive(Fig.3B),for rad6and mag1the e?ect is clearly synergistic.At the lowest MMS concentration we used(0.001%),the mag1rad6double mutant did not grow at all,whereas on the0.002%MMS plate,both rad6and mag1single mutants grew to full size (Fig.3A).

The rad18single mutant exhibited an increased MMS sensitivity comparable to that of the rad6mu-tant.The mag1rad18double mutant was very sensitive to MMS(Fig.4).A10min treatment with0.2%MMS resulted in no survivors in a total of5;10 mag1 rad18cells((0.00002%);mag1and rad18mutations are thus synergistic for MMS-induced killing as judged by quantitative analysis.The expected survival for an additive e?ect in the mag1rad18mutant(50%;

0.00035%)is at least10-fold higher than that observed ((0.00002%)after a10-min treatment with0.2% MMS.We noticed a high frequency of MMS-resistant revertants from both rad6 and rad18 strains.This was evident both in respect to colonies growing on the high concentration of MMS gradient plates,as well as the non-linear survival response to MMS in liquid killing experiments(Fig.4,rad18).The MMS-resistant colonies were true revertants,which also acquired res-istance to UV(data not shown).Thus,these cells share the same phenotype with srs2/radH mutants isolated in Fig.3A,B MMS-induced killing with mag1,apn1and rad6strains by a gradient plate assay.A mag1and rad6.B apn1and rad6.(?), DBY747(wt);(?),JC8901(mag1);(?),WXY9105(apn1);(?), WXY9376(rad6);(?),WX9377(mag1rad6),and(?),WXY9378 (apn1rad6).The plates were incubated at30°C for72h

464

Fig.40.2%MMS-induced liquid killing with mag1and rad18 strains.(?),DBY747(wt);(?),WXY9216(mag1);(?),WXY9326 (rad18)and(?),WXY9327(mag1rad18).The results are the average of two independent experiments.The dotted line(for mag1rad18) indicates the minimum killing e?ect within our detection limit(i.e., no survivors in5;10 cells)

a similar manner(Lawrence and Christensen1979; Aboussekhra et al.1989;Schiestl et al.1990). Contrary to the rad6and rad18mutants,the rev3 single mutant showed only a slightly increased MMS sensitivity.The mag1re v3and apn1rev3double mu-tants were only slightly more sensitive to MMS than the mag1and apn1single mutants,respectively(Fig.5). These results suggest a simple additive e?ect on MMS-induced killing between the rev3mutation and DNA-alkylation repair mutations.

Alkylation repair and RAD52-group genes

Genes in the RAD52group are involved in DNA-re-combination repair.Their mutants are more sensitive to ionizing radiation and double-strand breaks than to UV radiation.To our surprise,the rad50mutant was even more sensitive to MMS than the most sensitive DNA-alkylation repair mutant mag1(Fig.6),as was the rad52mutant(data not shown).Within40min, only0.0003%of the rad50cells survived0.2%MMS treatment,compared with'50%apn1survival and 0.03%mag1survival under the same conditions.The mag1rad50double mutant was extremely sensitive to MMS:a10-min incubation with0.2%MMS resulted in only0.00004%survival and a20-min MMS treat-ment resulted in no survivors in a4;10 cell popula-tion(Fig.6A).The apn1mutant was less sensitive to MMS than the mag1mutant;nevertheless,the apn1 rad50double mutant was more sensitive than the rad50 single mutant(Fig.6B).The killing e?ect for the mag1 rad50double mutant is most likely synergistic.If the e?ect was additive,the expected survival after a10-min treatment would be0.1%,which is10 -fold higher than that observed(0.00004%).In contrast,the killing e?ect for apn1and rad50appears to be additive,as shown in Fig.6C.

MMS sensitivity in MAG1-over expressing cells

The synergistic e?ect of some double mutants suggest that the two genes/pathways share a common lesion substrate.Since the major lethal lesion produced by MMS is3MeA(Boiteux et al.1984),and the Mag1 3MeA DNA glycosylase removes the damaged base,it was thought that if the Mag1protein was over-produced,it might over ride the MMS sensitivity due to defects in other DNA-repair pathways.YEp13A is a YEp-based multicopy plasmid containing the MAG1 gene;cell extracts from DBY747/YEp13A transform-ants have about10-fold more3MeA DNA glycosylase activity than extracts of DBY747(Chen et al.1989). Fig.5A,B MMS-induced killing with mag1,apn1and rev3strains by a gradient plate assay.A mag1and rev3.B apn1and rev3.(?), DBY747(wt);(?),JC8901(mag1);(?),WXY9105(apn1);(?), WXY9382(rev3);(?),WX9383(mag1re v3),and(?),WXY9384 (apn1re v3).The plates were incubated at30°C for44h

465

Fig.6A—C0.2%MMS-induced liquid killing with mag1,apn1and rad50strains.A mag1and rad50.B apn1and rad50.(?),DBY747 (wt);(?),WXY9216(mag1);(?),WXY9105(apn1);(?),WXY9221 (rad50);(?),WX9323(mag1rad50),and(?),WXY9324(apn1rad50). The results are the means of three independent experiments.C ex-pected and observed surviving fractions of the apn1rad50double mutant treated with0.2%MMS.The surviving fraction of each single mutant was multiplied to obtain the expected surviving frac-tion of the double mutant when the e?ect is additive(?).It was compared with the observed surviving fraction of the corresponding double mutant(?).Data are taken from Fig.6B

pWX1210is a multicopy plasmid containing the

MAG1gene under the control of a strong ADH1pro-moter.Furthermore,the upstream repressing sequence

(Xiao et al.1993)has been removed from the MAG1

promoter in the P "& -MAG1construct.DBY747cells harboring pWX1210have'100-fold MAG1mRNA

than untransformed cells,as judged by Northern hy-

bridization(data not shown).Both YEp13A and

pWX1210fully complemented MMS-sensitivity in

a mag1mutant,but the transformants were not more resistant to MMS than the wild-type cells.In contrast, pWX1210transformants of apn1,rad6,rad18and rad52 mutants,and YEp13A transformants of the rad50mu-tant showed the same MMS-sensitivity as their respect-ive single mutants,suggesting that over expression of the MAG1gene does not suppress the repair defects in other pathways.

Discussion

In this study,we generated a large number of single and double null mutants in an otherwise isogenic strain and performed epistasis analyses by comparing the level of MMS sensitivity between double mutants and their corresponding single mutants.The results lead us to the following interpretations.

We showed that di?erent yeast repair genes/path-ways contribute di?erently to the protection of yeast cells against killing by MMS.We have observed the following order of sensitivity to MMS:rad6, rad18'rad50,rad52

The additive e?ect observed for the rad4mag1 double mutant suggests that the nucleotide-excision and MAG1base-excision repair pathways may target di?erent lesions induced by MMS.It is thus conceiv-able that the RAD3pathway may recognize some MMS-induced lesions which are less abundant,but cause more distortion,than3MeA.

Our results also suggest that the RE?3-dependent mutagenesis pathway probably provides a tolerance mechanism by a trans lesion bypass of the replication-blocking3MeA lesion,albeit at the cost of mutagenesis. This is consistent with a recent study of Rev3-Rev7in the trans lesion bypass(Nelson et al.1996),as well as the role of SOS mutagenesis in bacteria,where RecA and UmuCD may assist DNA polymerase III in by-passing a replication-blocking lesion at the cost of possible mutation(Friedberg et al.1995).It also ex-plains why MMS is lethal,but not mutagenic,to Salmonella in the Ames test unless plasmid pKM101is also present(Walker1983).

A comparison of killing by MMS between double mutants and their respective single mutants suggests that post replication-and recombination-repair path-ways are distinct from,yet may share some common lesions with,the MAG1base-excision repair pathway. Our results exclude the possibility that these two path-ways simply act downstream from the base excision since:(1)rad6,rad18,rad50and rad52mutants are even more sensitive to MMS than the most-sensitive mutant (mag1)in the base-excision repair pathway;(2)double mutants containing rad6,rad18,rad50or rad52with

466

mag1or apn1are more sensitive to MMS than either of the corresponding single mutants,suggesting an addi-tive or synergistic e?ect;and(3)the RAD6and RAD52 pathways may provide DNA damage-tolerance mecha-nisms rather than actual removal of the adduct (Prakash et al.1993;Friedberg et al.1995).Further-more,if the RAD6and RAD52pathways were to act downstream from the base excision,double mutants with mag1and apn1would be as sensitive to MMS as mag1and apn1single mutants,or less sensitive than the corresponding rad6,rad18,rad50and rad52single mu-tants.

Our?nding that some post replication-and recombi-nation-repair pathway mutants are more sensitive to MMS than the mag1mutant is rather surprising given that Mag1DNA glycosylase is the only known enzy-matic activity in yeast cells to remove3MeA(Chen et al.1989,1990;Berdal et al.1990).Furthermore,the synergistic e?ects of a mag1mutation with rad6,rad18, rad50(and most likely rad52)indicate that they target some common lesion(s),whereas results from MAG1 over expression argue against the hypothesis that post replication-repair(RAD6and RAD18)and recombina-tion-repair(RAD50and RAD52)pathways simply rec-ognize a3MeA lesion.To reconcile this apparent con-?ict,we propose that these two pathways may recog-nize a3MeA-induced block of the replication fork having single-stranded DNA ends instead of3MeA adducts or DNA-strand breaks per se.Hence RAD6, RAD52pathways and Mag1may compete for two distinct forms of a common lesion,3MeA.We further propose that a3MeA DNA polymerase complex is inaccessible to Mag1DNA glycosylase and is thus more lethal than a free3MeA,which is a mere pre-lethal lesion.This is reminiscent of a model proposed for the processing of UV-induced lesions in yeast (Friedberg et al.1995).Indeed we have frequently ob-served MMS-tolerant revertants from rad6and rad18 deletion mutant strains,which are believed to be srs2 mutants that channel UV lesions to the RAD52path-way(Schiestl et al.1990).

Acknowledgements The authors thank many laboratories and indi-viduals for providing plasmids.Special editorial assistance from Dr.B.Kunz is appreciated.This work was supported by a Medical Research Council of Canada operating grant MT-12633to W.X. W.X.is a Research Scientist of the National Cancer Institute of Canada.L.R.was supported by a B.Sc.(Med)program from the College of Medicine,University of Saskatchewan. References

Aboussekhra A,Chanet R,Zgaga Z,Cassier-Chauvat C,Heude M, Fabre F(1989)RADH,a gene of Saccharomyces cerevisiae encod-ing a putative DNA helicase involved in DNA repair.Character-istics of radH mutants and sequence of the gene.Nucleic Acids Res17:7211—7219

Alani ES,Subbian S,Kleckner S(1989)The yeast RAD50gene encodes a predicted153-kD protein containing a purine nucleo-

tide-binding domain and two large heptad-repeat regions.

Genetics122:47—57

Bailly V,Lamb J Sung P,Prakash S,Prakash L(1994)Speci?c complex formation between yeast RAD6and RAD18proteins:

a potential mechanism for targeting RAD6ubiquitin-conjugat-

ing activity to DNA damage sites.Genes Dev8:811—820 Barnes DE,Lindahl T,Sedgwick B(1993)DNA repair.Curr Opin Cell Biol5:424—433

Beranek DT(1990)Distribution of methyl and ethyl adducts follow-ing alkylation with monofunctional alkylating agents.Mutat Res 231:11—30

Berdal KG,Mjoras M,Bjelland S,Seeberg E(1990)Cloning and expression in Escherichia coli of a gene for an alkylbase DNA glycosylase from Saccharomyces cerevisiae;a homologue to the bacterial alkA gene.EMBO J9:4563—4568

Boiteux S,Huisman O,Laval J(1984)3-methyladenine residues in DNA induce the SOS function s?A in Escherichia coli.EMBO J3:2569—2573

Boeke JD,Trueheart J,Natsoulis G,Fink GR(1987)5-Fluoroorotic acid as a selective agent in yeast molecular genetics.Methods Enzymol154:164—175

Chanet R,Magana-Schwencke N,Fabre F(1988)Potential DNA-binding domains in the RAD18gene product of Saccharomyces cerevisiae.Gene74:543—547

Chen J,Der?er B,Maskati A,Samson L(1989)Cloning a eukaryotic DNA glycosylase repair gene by the suppression of a DNA repair defect in Escherichia coli.Proc Natl Acad Sci USA 86:7961—7965

Chen J,Der?er B,Samson L(1990)Saccharomyces cerevisiae 3-methyladenine DNA glycosylase has homology to the AlkA glycosylase of E.coli and is induced in response to DNA alkyla-tion damage.EMBO J9:4569—4575

Cooper AJ,Waters R(1987)A complex pattern of sensitivity to simple monofunctional alkylating agents exists amongst the rad mutants of Saccharomyces cerevisiae.Mol Gen Genet 209:142—148

Dornfeld KJ,Livingston DM(1991)E?ects of controlled RAD52 expression on repair and recombination in Saccharomyces cerevisiae.Mol Cell Biol11:2013—2017

Feaver WJ,Svejstrup JQ,Bardwell L,Bardwell AJ,Buratowski S, Gulyas KD,Donahue TF,Friedberg EC,Kornberg RD(1993) Dual roles of a multiprotein complex from S.cerevisiae in tran-scription and DNA repair.Cell75:1379—1387

Friedberg EC(1988)Deoxyribonucleic acid repair in the yeast Saccharomyces cerevisiae.Microbiol Rev52:70—102 Friedberg EC,Walker G,Siede W(1995)DNA repair and mutagen-esis.ASM Press,Washington,DC

Gietz RD,Prakash S(1988)Cloning and nucleotide sequence analy-sis of the Saccharomyces cerevisiae RAD4gene required for excision repair of UV-damaged DNA.Gene74:535—541 Guzder SN,Qiu H,Sommers CH,Sung P,Prakash L,Prakash S(1994a)DNA repair gene RAD3of S.cerevisiae is essential for transcription by RNA polymerase II.Nature367:91—94 Guzder SN,Sung P,Bailly V,Prakash L,Prakash S(1994b)RAD25 is a DNA helicase required for DNA repair and RNA polymerase II transcription.Nature369:578—581

Haynes RH,Kunz BA(1981)DNA repair and mutagenesis in yeast.

In:Strathern J,Jones EW,Broach JR(eds)The molecular biol-ogy of the yeast Saccharomyces cerevisiae:life cycle and inherit-ance.Cold Spring Harbor Laboratory,Cold Spring Harbor, New York,pp371—414

Ito H,Fukuda Y,Murata K,Kimura A(1983)Transformation of intact yeast cells treated with alkali cations.J Bacteriol 153:163—168

Jentsch S,McGrath JP,Varshavsky V(1987)The yeast DNA repair gene RAD6encodes a ubiquitin-conjugating enzyme.Nature 329:131—134

Johnson RE,Henderson ST,Petes TD,Prakash S,Bankmann M, Prakash L(1992)Saccharomyces cerevisiae RAD5-encoded DNA repair protein contains DNA helicase and zinc-binding

467

sequence motifs and a?ects the stability of simple repetitive sequences in the genome.Mol Cell Biol12:3807—3818 Jones,JS,Weber S,Prakash L(1988)The Saccharomyces cerevisiae RAD18gene encodes a protein that contains potential zinc?nger domains for nucleic acid binding and a putative nucleotide bind-ing sequence.Nucleic Acids Res16:7119—7131

Kang X,Yadao F,Gietz RD,Kunz BA(1992)Elimination of the yeast RAD6ubiquitin conjugase enhances base-pair transitions and G·C P T·A transversions as well as transposition of the Ty element:implications for the control of spontaneous mutation.

Genetics130:285—294

Kupiec M,Simchen G(1984)Cloning and integrative deletion of the RAD6gene of Saccharomyces cerevisiae.Curr Genet8:559—566 Lawrence CW,Christensen R(1979)Metabolic suppressors of trimethoprim and ultraviolet light sensitivities of Saccharomyces cerevisiae rad6mutants.J Bacteriol139:866—876

Morrison A,Christensen RB,Alley J,Beck AK,Bernstine EG, Lemontt JF,Lawrence CW(1989)RE?3,a Saccharomyces cerevisiae gene whose function is required for induced mutagen-esis,is predicted to encode a nonessential DNA polymerase.

J Bacteriol171:5659—5667

Nelson JR,Lawrence CW,Hinkle DC(1996)Thymine-thymine dimer bypass by yeast DNA polymerase .Science272:1646—1649 Popo?,SC,Spira AS,Johnson AW,Demple B(1990)The yeast structural gene(APN1)for the major apurinic endonuclease: homology to E.coli endonuclease IV.Proc Natl Acad Sci USA 87:4193—4197

Prakash L(1974)Lack of chemically induced mutation in repair-de?cient mutants of yeast.Genetics78:1101—1118.

Prakash L,Prakash S(1977)Isolation and characterization of MMS-sensitive mutants of Saccharomyces cerevisiae.Genetics 86:33—55

Prakash S,Sung P,Prakash L(1993)DNA repair genes and proteins of Saccharomyces cerevisiae.Annu Rev Genet27:33—70 Ramotar D,Popo?SC,Gralla EB,Demple B(1991)Cellular role of yeast APN1apurinic endonuclease/3 -diesterase:repair of oxida-

tive and alkylation DNA damage and control of spontaneous mutation.Mol Cell Biol9:4537—4544

Schiestl RH,Prakash S,Prakash L(1990)The SRS2suppressor of rad6mutations of Saccharomyces cerevisiae acts by channeling DNA lesions into the RAD52DNA repair pathway.Genetics 124:817—831

Sherman F,Fink GR,Hicks J(1983)Methods in yeast genetics.Cold Spring Harbor Laboratory,Cold Spring Harbor,NY

Walker G(1983)Molecular principles underlying the Ames Salmonella/microsome test:elements and design of short-term mutagenesis tests.In:Kolber AR,Wong TK,Grant LD,DeWos-kin RS,Hughes TJ(eds)In vitro toxicity testing of environmental agents.Plenum,New York,pp15—39

Wang Z,Svejstrup JQ,Feaver WJ,Wu X,Kornberg RD,Friedberg EC(1994)Transcription factor b(TFIIH)is required during nucleotide-excision repair in yeast.Nature368:74—76 Xiao W,Fontanie T(1995)Expression of the human MGM1O -methylguanine DNA methyltransferase gene in a yeast al-kylation-sensitive mutant:its e?ects on both exogenous and endogenous DNA alkylation damage.Mutat Res336:133—142 Xiao W,Samson L(1992)The Saccharomyces cerevisiae MG11 DNA repair methyltransferase gene:its promoter and entire coding sequence,regulation and in vivo biological functions.

Nucleic Acids Res20:3599—3606

Xiao W,Samson L(1993)In vivo evidence for endogenous DNA alkylation damage as a source of spontaneous mutation in eu-karyotic cells.Proc Natl Acad Sci USA90:2117—2121

Xiao W,Der?er B,Chen J,Samson L(1991)Primary sequence and biological functions of a Saccharomyces cerevisiae O -methyl-guanine/O -methylthymine DNA-repair methyltransferase gene.

EMBO J10:2179—2186

Xiao W,Singh KK,Chen B,Samson L(1993)A common element involved in transcriptional regulation of two DNA alkylation repair genes(MAG and MG11)of Saccharomyces cerevisiae.Mol Cell Biol13:7213—7221

.

468