[CANCERRESEARCH54,3001-3006,June1, 1994] Molecular Analysis of Ethyl Methanesulfonate-induced Mutations at the hprt Gene in the Ethyl Methanesulfonate-sensitive Chinese Hamster Cell Line EM-Cl! and Its Parental Line CHO9'

Christel W. Op het Veld, Matgorzata Z. Zdzienicka, Harry Vrieling, Paul H. M. Lohman, and Albert A. van Zeeland2 MGC-Department of Radiation Genetics and Chemical Muzagenesis, State University of Leiden, Wassenaarseweg 72, 2333 AL Leiden [C. W. 0. h. V., M. Z. Z., H. V., P. H. M. L., A. A. V. Z.J, and J. A. Cohen Institute, Interuniversizy Research Institute for Radiopathology and Radiation Protection, Leiden [M. Z. Z., H. V., A. A. v. Z.J, the Netherlands

ABSTRACF strong nucleophiles like the N-7 position of guanine, resulting in a relatively low 06/N-7-alkylguanine ratio. The Chinese hamster cell line EM-Cl! has been shown to be 5 times Molecular analysis of induced mutations, i.e. , the determination of more sensitive than its parental line CHO9, but not hypermutable, after mutational spectra, provides a valuable tool for the identification of treatment with ethyl methanesulfonate. Ethyl methanesulfonate-induced adducts that are involved in mutation induction. Moreover, the use of mutational spectra were determined at the hprt locus to investigate DNA repair deficient cell lines will generate additional information whether the same ndducts are responsible for mutation induction in both cell lines. The mutational spectra for EM-C!! and CHO9 show an im concerning the mutagenic potential of different types of DNA adducts. portent difference. GC-'AT transitions were found in both cell lines at Recently, a Chinese hamster ovary cell line, EM-Cl 1, which is very similar frequencies; however, the spectrum of CHO9 contains a class of sensitive to the cell killing effects of EMS, was isolated in our AT—'GCtransitions,which seems to be replaced by a group of deletions laboratory (2). EM-Cl 1 belongs to the same complementation group in EM-Cl!. Since the ethyl methanesulfonate-induced mutation frequency as EM9 (3) which shows a strongly enhanced cytotoxicity as well as for both lines is the same at equal exposure, it is hypothesized that the hypermutability after EMS treatment. Both cell lines are slow in the lesions leading to AT—'GCtransitions in CHO9 are responsible for the repair of DNA single strand breaks and the defect in these mutants is deletions in EM-Cl!. This phenomenon might be explained if the respon complemented by the human xrcc-l gene, isolated by Thompson et al. sible adduct(s) in CHO9 is bypassed resulting in replication errors, while (4). However, the precise biochemical defect remains to be deter blocking DNA synthesis in EM-Cl! causing the observed increase in cell mined (2, 3). A few differences have been observed between these death. In surviving EM-C!! cells, DNA strand exchanges might have two Chinese hamster ovary cell mutants. EM-Cl 1 is not sensitive to occurred at the position of stalled replication forks, leading to gross Uv irradiationandonlyslightlysensitivetoX-rays(2),whereasEM9 molecular changes. The adduct probably responsible for the AT—GC is sensitive to UV and X-rays (3). Thus EM-Cl 1 may have a more transitions in CHO9 and the deletions in EM-C!! is 3-ethyladenine. specific defect in the repair of DNA damage induced by alkylating agents and, therefore, appears to be a valuable tool for the determi INTRODUCFION nation of the nature of mutagenic lesions induced by monofunctional alkylating agents. Monofunctional alkylating agents such as EMS3 act directly on In this study we have used EM-Cl 1 and its parental line CHO9 to oxygen and nitrogen atoms in DNA bases and with oxygen moieties study the effects of EMS on the frequency and molecular nature of the of the phosphate backbone. A broad spectrum of DNA lesions is induced mutations in the hprt gene. formed, consisting of alkylated bases, phosphotriesters, and abasic sites due to spontaneous hydrolysis of unstable alkylation products or MATERIALS AND METHODS enzymatic hydrolysis by DNA glycosylases. The types of DNA le sions induced by an alkylating agent depend upon the reaction mech Cell Culture Conditions. EM-Cu Chinese hamsterovary cells and the anism and the reactivity of the alkylating agent, which can be ex parental line CHO9 were cultured in Ham's modified F-lO medium lacking pressed by the Swain-Scott constant. Alkylating agents with a low s hypoxanthine and thymidine and supplemented with 15% newborn calf serum (Gibco), 100 units/mi penicillin, and 0.1 mg/ml streptomycin (2). value, such as N-ethyl-N-nitrosourea (s 0.26), react with the nu EMS survival and mutation induction at the hprt locus were determined cleophilic centers in the DNA via a SN1 reaction, in which the rate according to the method of Zdzienicka and Simons (5). In brief, i0@cells were limiting step is the formation of the alkyl cation. These agents show treated in suspension for 1 h at 37°C in serum-free medium supplemented with a relatively low selectivity in their alkylation reaction. Since the 20 mi@t4-(2-hydroxyethyl)l-piperazethanesulfonic acid (pH 7.4) in a total number of oxygen atoms in DNA available for alkylation is larger volume of 10 ml. EMS (Eastman Co., Rochester, NY) was added directly to than the number of nitrogen atoms, compounds with a low s value the suspension or, in case of low exposures, EMS was diluted in phosphate give high levels of O-alkylation relatively to N-alkylations. These buffered saline just before use. agents further have a low chromosome breaking ability relative to Following treatment, cells were washed twice with phosphate buffered their capacity to induce gene mutations (1). EMS belongs to the saline, resuspended in medium with 15% newborn calf serum, and subse alkylating agents with a relatively high s value (s = 0.67) and follows quently cells were (a) seeded for survival (200 cells/94-mm dish; 5 dishes/ group) and (b) propagated for expression of induced mutants (3.5 X 10@'—2X a mixed SN1/SN2 type reaction. Therefore, EMS also reacts with 106 cells/l50-mm dish; 4 dishes/group); these dishes were subcultured after 4 days. Eight days after treatment cells were seeded for (a) cloning efficiency Received 12/22/93; accepted 3/28/94. (200 cells/94-mm dish; 5 dishes/group) and (b) selection of hprt-deficient Thecostsof publicationofthisarticleweredefrayedinpartby thepaymentofpage mutants (10@ cells/94-mm dish; 20 dishes/group) in medium containing 5 charges. This article must therefore be hereby marked advertisement in accordance with 18U.S.C.Section1734solelyto indicatethisfact. p@g/ml6TG.The mutant frequency was calculated by correcting the frequency I This work was supported by Grant KWF.90.04 from the Dutch Cancer Society (the of 6TG resistant clones with the corresponding cloning efficiency. Netherlands) and Grant EV5V-CF91-0012 from the Commission of the European Com Isolation of hprt Mutants for Moleculnr Analysis. To isolate hprt defi munities. dent mutants, 4—5 X i0@ C1-1O9 or EM-Cu cells were treated with 5 m@i 2 To whom requests for reprints should be addressed. 3 The abbreviations used are: EMS, ethyl methanesulfonate; 6TG, 6-thioguanine; EMS. After treatment, cells were washed twice, resuspended in complete cDNA, complementary DNA; PCR, polymerase chain reaction. medium, and subsequently divided in aliquots of 3.5 X l0@cells for CHO9 and 3001

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1994 American Association for Cancer Research. 3-ETHYI.ADENINE CAUSES AT-' OC TRANS@ON5

Table 1 Primers used for molecular analysis of hprt mutants Primer sequence Positiona cDNA synthesis vrl-16 5' GCAGAUCAAcTFGAATFCFCATC 3' 681 to 658

PCR cDNA vrll0-m13 5' CGACG1TGTAAAACGACGGCCAGT 675 to658 TCAACTTGAATFCFCATC 3b zee-P 5' GGCflCCFCCTCAGACCGCT 3' —50to —31

PCR genomic DNA ham @5C 5' AGcTFATGCFCFGA1TFGAAATCAGCFG 3' E2:—42to —15 ham 23 5' AUAAGATCVFACVFACCFGTCCATAATC 3' 12:17 to E2:96 ham 35C 5' GGAACFCGTCFATfCCGTGATITFA 3' E3:—34 to —10 ham 33 5' AAATACATACAAAACTAGGAUGCC 3' 13:58 to 34 ham 45C 5' GTGTATFCAAGAATATGCATGTAAATGATG 3' E4:—45to —16 ham 43 5' CAAGTGAGTGATFGAAAGCACAGTFAC 3' 14:80 to 54 ham 55C 5' AACATAT000TCAAATAUOTFCTAATAG 3' E5:—141to—112 ham 53 5' GGCVfACCTATAGTATACACACTAAGCFA 3' 15:68 to 42 ham 75C 5' GTFCFArFGTCITFCCCATATGTC 3' E7:—49to —26

Sequencing m13 5' GTAAAACGACGGCCAGTG 3' vrl-2 5' GCAAGCfTGCAACCI'TAACC 3' 490 to 471 zee-6 5' CFGATAAAATCFACAGTCAT 3' 302 to 283 zee-4 5' CCATGAGGAATAAACAC 3' 119 to 93

a Positions in the coding region are numbered according to the method of Jolly et aL (29) with the A of the ATG initiation codon being the first nucleotide. A designation such as 11:5 refers to the fifth base of intron 1. b Bases in italics, M13 sequence. C 5'-Biotinylated primer.

aliquots of 5 X i0@for EM-Cl 1. Each population was subcultured separately Mutational Spectrum of CHO9. hprt-deficient mutants were iso to ensure that all mutants obtained were independent. After 8 days of expres lated after an exposure to S m@i EMS. The spontaneous mutant sion time, each culture was harvested separately and 10@cells were seeded per frequency for CHO9 was 0.9 X i05, whereas after S mi@iEMS the plate (2 dishes/culture) in 6TG containing medium. After an additional growth mutant frequency was 22.8 X i0@, which indicates that most of the period of 10 days, only one 6TG resistant colony from each set of dishes was hprt mutants isolated for mutational analysis was induced by the EMS isolated. treatment. The RNA of the mutants was used to synthesize hprt cDNA In total, 25 independent hprt mutants from CHO9 and 41 hprt mutants from EM-Cu were isolated. which was amplified by PCR and subsequently used for direct Molecular Analysis of hprt Mutants. Total cytoplasmic RNA was iso sequencing. lated from 1—2x i07 cells of each mutant as described by Vrieling et a!. (6). Nearly all mutants isolated from CHO9 gave a PCR product (23 of The RNA was used to synthesize hprt cDNA as described by Menichini et aL 25 mutants). Sixteen of these mutants gave a full length PCR product, (7). About 10% of the cDNA product was used as template for PCR. Gel while seven mutants gave one or more smaller PCR products. In most purified PCR product was used in a second round of PCR with one of the of the mutants giving a full length PCR product the hprt-deficient primers being 5'-biotinylated. The biotinylated PCR product was bound to phenotype was the result of a single base pair substitution. Ten streptavidin coated beads (Dynal AS, Norway). Single stranded DNA to be GC—*ATand 4 AT—@GCtransitions were found (Table 2). In the used as substrate in the dideoxy sequencing reaction (Pharmacia Ti sequenc remaining two mutants with a full length cDNA PCR product, no ing kit) was prepared by denaturation of the bound DNA as described (7, 8). mutation was found in the cDNA sequence (C-EMS7 and C-EMS9), Sequencing was performed using a Pharmacia automatic sequencer. To analyze mutants missing one or more exons from the hprt cDNA, a crude possibly because of a mutation in the untranslated region of the gene cell lysate was used as source of genomic DNA for amplification of these affecting translation. exons using primers lying in the introns flanking the missing exon (7, 9). Seven mutants gave one or more smaller PCR products after Sequencing of the amplified hprt exon sequences was performed similar to amplification of the hprt cDNA, which could have arisen from intra sequencing of the amplified hprt cDNA (7). The primers that were used for PCR, cDNA synthesis, and sequencing are listed in Table 1. 100.0

RESULTS > EMS InducedCytotoxicityand Mutabilityat the hprtLocus. Cl) Cell killing by EMS was measured by the determination of the colony J 10.0 forming ability of the cells immediately after treatment. The data in Fig. 1 show that EM-Cl 1 is about 5 times more sensitive than the parental cell line CHO9 (Fig. 1), which is in the same range as previously observed by Zdzienicka et a!. (2). Mutation induction at 1.0 the hprt locus by EMS was measured as 6-thioguanine resistance. In both cell lines a linear increase in mutation induction with exposure is 0 10 20 30 40 seen (Fig. 2). Because of the high toxicity, mutation induction with EM-Cl 1 was measured only up to an EMS concentration of 6 m@i. EMS Concentration(mM) Mutability of the two cell lines was the same when compared at equal Fig. 1. Survival of CHO9 (•)andEM-Cl 1(A) cells after exposure to EMS. Eachpoint EMS concentration. is an independent measurement. 3002

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1994 American Association for Cancer Research. 3-E@FHYLADEN1NECAUSESAT-. GC TRANSITIONS

was missing in the cDNA. Therefore, it can be concluded that these .!@ 150 8 mutants have a mutation in one of the splice sites, which, subse quently, could be demonstrated for some mutants by sequencing of the . exon specific PCR product (see Table 2, splice mutations). 0 With two mutants (C-EMS15 and -21) no amplified cDNA product U, was obtained. The possibility that these mutants contain a large C deletion including the complete hprt gene was checked using genomic E DNA as template for PCR with primers for exons 2, 3, 4, 5, and 7—9. In both mutants all these exons were present in the genomic DNA, indicating the absence of a large deletion in these mutants. Possible I explanations for the lack of a PCR product are (a) a mutation affecting the stability of the messenger RNA or (b) a mutation in the promotor 0 region affecting hprt transcription. 0 10 20 30 40 MutationalSpectrumin EM-C!L hprtdeficientmutantswere also with this cell line isolated after an exposure to 5 mr@iEMS. The mutant frequency for EM-Cu at 5 mi@iEMS is 14.1 X l0@, which EMS concentration(mM) is between 7 and 23 times higher than the spontaneous mutant fre Fig. 2. Mutation induction at the hprt locus in parental CHO9 (0) and in EMS quency that varied from 0.61 to 2.1 X iO@. This indicates that most sensitive EM-Cl 1 (A) cells in response to EMS. Each point is an independent measure of the analyzed mutants are induced by the EMS treatment. ment. The mutational spectrum for EM-Cl 1 is shown in Table 3. Forty one mutants have been analyzed: 23 gave a full length PCR product; genic deletions or missplicing of the mRNA as a result of a point 7 mutants gave one or more smaller PCR products; whereas one mutation at one of the splice sites. Two of these mutants were missing mutant gave a larger PCR product. Ten mutants did not give any PCR exon 4, two mutants were missing exon 5, and three mutants were product when cDNA was used as template for the PCR reaction. missing either exons 2 and 3 or exon 3 alone. Genomic DNA from In 20 of the 23 mutants giving a full length PCR product, the hprt these mutants was subjected to PCR with primers lying in the introns deficient phenotype was due to a GC—÷ATtransition mutation. Mu flanking the exons missing from the cDNA in order to distinguish tant E-EMS18 carried an AT—'CG transversion. Two other mutants whether the mutant phenotype was the result of a small deletion in the (E-EMS2 and -3) carried a —1frame-shift mutation. One of these hprt coding sequence or due to missplicing. When genomic DNA was frame-shifts (E-EMS2) is at the boundary of exon 5/exon 6 in the used as a substrate, all mutants gave a PCR product of the exon that amplified cDNA and could have been caused by a mutation at the 5'

CHO9MutationPositiona Table 2 EMS induced mutants isolated from

sequence― acid changeTransitionsC-EMS2O110 Exon cDNA@' Splicec ChangeTarget 5' —.3'Amino

ThrC-EMS251 2 AT—'GCGTTTA T TCCTCIle—* ThrC-EMS8530 10 2 AT—@GCGTTTA T TCCTClIe—. GlyC-EMS18608e 7 AT—@GCGCCAG A CTTTGAsp—. 5crC-EMSI 8 8 AT—*GCTTTGA A TCATAAla—' ArgC-EMS1214811 18 2 GC—@ATCTCAT G GAGTGGly—. ThrC-EMS24151 3 GC—.ATGACTT 0 CCCGAMa—a GAGATArg-.stopC-EMS22152 3 GC-+ATTTGCC C GinC-EMS1209 3 GC—ATTGCCC 0 AGATGMg—. GluC-EMS17430 3 GC-*AT0 GGGCTGly—' AAACTGln—'stopC-EMS3508 6 GC—*ATCAATG C GAAGTArg—.stopC-EMS1O508 7 GC—+ATCCTCT C GAAGTArg—'stopC-EMS14508 7 GC—.ATCCTCT C GAAGTArg—.stopC-EMS23538 7 GC—'ATCCTCT C ArgSplice 8 GC—.ATTTGTT 0 GATTTGly—.

mutations C-EMS6 C-EMS13 2,3@ Splice SpliceC-EMS5E4:—iC-EMS162,3k 2,3kSplice AATGASpliceC-EMS19E4:—l 4 GC—@ATaacta g AATGASpliceC-EMS2E5:—l 4 GC—aATaacta g AATGTSpliceC-EMS4E5:—l 5 GC—'ATttcta g AATGTSpliceOtherC-EMS7Wild 5 GC—@ATttcta g

region.C-EMS9Wild type cDNA sequence. Possible mutation in untranslated region.C-EMS15No type cDNA sequence. Possible mutation in untranslated DNA.C-EMS21No cDNA was obtained, but exons 2, 3, 4, 5, and 7—9arepresent in genomic cDNA was obtained, but exons 2, 3, 4, 5, and 7—9arepresent in genomic DNA.

a See Table 1, Footnote a. b Exon where DNA alteration occurred or exon(s) which have been deleted from the hprt cDNA.

C (Putative) splice mutant. The indicated exon is deleted from the cDNA. d Exon sequences are in capital letters; intron sequences are in lower case letters. e cDNA molecules lacking exon 8 were also recovered from this mutant. 1Missing exon 3 or exon 2 + 3 from the amplified cDNA. 3003

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1994 American Association for Cancer Research. 3-ETHYLADENINECAUSES AT—'OCTRAN5@ONS

EM-CuMutant Table 3 EMSinduced mutantsisolatedfrom sequence acid Position― Exon cDNA@'SplicecChangeTarget 5' —*3―Amino change Transitions E-EMS5 208 3 GC—*AT TGAAG 0 GGGGC Gly-+ Arg E-EMS1I 208 3 GC—+AT TGAAG G GGGGC Gly-.Arg E-EMS41 208 3 GC—aAT TGAAG0 GGGGC Gly-Arg E-EMS48 208 3 GC—@AT TGAAG G GGGGC Gly-. Arg E-EMS42 209 3 GC—*AT GAAGG G GGGCT Gly-* Glu E-EMS45 325 4 GC—@AT ATGAT C AGTCA Gln—*stop E-EMS6 419 6 GC—.AT CACTG G TAAAA Gly-+Asp E-EMS47 419 6 GC—@AT CACTG G TAAAA Gly-Asp E-EMS1 464 6 GC—AT CA.ACC C CAAAA Pro-+ His E-EMS9 508 7 GC-+AT CCTCT C GAATG Arg—.stop E-EMS14 508 7 GC-+AT CCTCT C GAATG Arg-÷stop E-EMS37 508 7 GC—*AT CCTCT C GAAGT Arg—@stop E-EMS13 538 8 GC-'AT TTGTT G GATTT Gly-@Arg E-EMS39 538 8 GC—'AT TTGTT 0 GATTT Gly-@Arg E-EMS1O 544 8 GC-*AT GATTT G AAATT Glu—+Lys E-EMS16 599 8 GC-*AT CTTCA 0 GGATT Mg—+Lys E-EMS28 599 8 GC-*AT CTTCA 0 GGATT Arg—'Lys E-EMS7 601 8 GC-*AT TCAGG 0 ATTTG Asp—@Asn E-EMS36 601 8 GC—'AT TCAGG 0 ATTTG Asp-iAsn E-EMS38 617 9 GC-+AT TATTT 0 TGTCA Cys-@ Tyr

Transversions E-EMS18 573 8 AT-+CG CGAAG T GTTGG Tyr—@stop

Splice mutations E-EMS8 2,3e Splice E-EMS12 2,3e Splice E-EMS4 ES:—! 5 GC—.AT ttcta gAATGT Splice E-EMS27 8 Splice E-EMS34 At least two PCR products were obtained from cDNA. Splice E-EMS46 At least two PCR products were obtained from cDNA. Splice E-EMS49 At least two PCR products were obtained from cDNA. Splice

Deletions/insertions E-EMS15 Deletion exon 5 and exons 7—9fromgenomic DNA E-EMS17 Deletion exons 7—9from genomic DNA E-EMS19 Deletion exons 7—9from genomic DNA E-EMS22 Deletion exons 7—9fromgenomic DNA E-EMS25 Deletion exons 2, 4, and 7—9from genomic DNA E-EMS26 Deletion exons 7—9from genomic DNA E-EMS32 Deletion exons 2, 4, and 7—9fromgenomic DNA E-EMS33 Deletion exons 2, 4, and 7—9fromgenomic DNA E-EMS5O Deletion exons 2, 4, and 7—9from genomic DNA

Other E-EMS43 Duplication of exons 2 and 3 E-EMS2O Mutation affecting mRNA stability or hprt transcription E-EMS3 370 4 -T TAAAG(TT)GAGA5 Frame-shift E-EMS2 403 6 -G TTGA(GG)ACATA5 Frame-shift

a See Table 1, Footnote a. b Exon where DNA alteration occurred or exon(s) which have been deleted from the hprt cDNA. C (Putative) splice mutant. The indicated exon is deleted from the cDNA. d Exon sequences are in capital letters; intron sequences are lower case letters. e Missing exon 3 or exon 2 + 3 from the amplified cDNA. 1Only one of the nucleotides in parentheses has been deleted from the cDNA sequence, but which one cannot be determined because they are identical. splice site of intron 5. The second frame-shift mutation (E-EMS3) is exon. For one of these mutants (E-EMS2O) PCR products were at a iT site and could result from template slippage during DNA obtained for all exons checked, indicating that this mutant might carry replication. a mutation affecting either transcription or the stability of the mRNA. Seven mutants gave one or more smaller PCR products, which were The nine remaining mutants were missing some hprt exons from their due to splice mutations (see Table 3, splice mutants). One of these was DNA, indicating that these mutants carry deletions of (part of) the hprt missing exon 5 (E-EMS4), another was missing exon 8 (E-EMS27) gene. Mutants E-EM525, -32, -33, and -50 probably miss the com and two mutants were missing either exons 2 and 3 or exon 3 alone plete hprt gene, whereas mutants E-EMS15, -17, -19, -22, and -26 (E-EMS8 and -12) from the amplified cDNA. Mutants E-EMS34, -46, miss the 3'-terminal part of the gene only. and -49 also give more than one PCR product and therefore are Strand and Region Specificity. The spectra of both cell lines are considered to be splice mutants as well, although the molecular dominated by GC—@ATtransitions, which are most probably caused changes in the mutants have not been examined in detail. Mutant by 06-ethylguanine. In order to check whether the mutated nucleo E-EMS43 gave a larger PCR product due to a duplication of exons 2 tides in the coding sequence are randomly distributed over the two and 3. strands, our data were compared with a database containing over 300 Finally, there is a group of ten mutants that did not give a PCR independent mutations in Chinese hamster cells (10). In this database product when cDNA was used in the PCR reaction. The genomic 70% of the mutations at a GC base pair showed the mutated G to be DNA of these mutants was checked for the presence of each hprt present in the nontranscribed strand. In the CHO9 spectrum analyzed exon, by using primers in the intron sequences flanking that particular in 5 mutated guanines were present in the transcribed strand and 5 in 3004

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1994 American Association for Cancer Research. 3-ETHYLADENINE CAUSES AT—.GC TRANS@ONS the nontranscribed strand, whereas for EM-Cu 15 mutated guanines AT—@GCtransitions observed in CHO9 cells are absent in EM-Cl 1 were located in the nontranscribed strand and 5 in the transcribed and seem to be replaced by a similarly sized class of deletions (22%). strand. Thus, the observed distribution for both cell lines is probably Since the frequency of mutations at equal exposure to EMS is the not different from the distribution one would expect on basis of the same in both cell lines, we hypothesize that the DNA adduct which is amount of mutable guanines in each strand. causing the AT—*GCtransitions in CHO9, i.e., 3-ethyladenine, is also Some positions, e.g., 208 and 508, were mutated more than once: responsible for the deletions found in EM-Cu. An explanation for since the procedure for picking up mutants was devised in such way this phenomenon might be that DNA replication is able to perform that all are independent mutations, these positions might be hot spots translesion synthesis in CHO9, thereby causing replication errors for mutation. resulting in AT—*GCtransitions. The same DNA adduct in EM-Cl! might cause inhibition of DNA synthesis, causing increased cell DISCUSSION death. Evidence that 3-alkyladenine in DNA blocks replication has been shown in in vitro replication systems (26, 27). Among the EMS-induced Cytotoxicity and Mutability at the hprt Locus. surviving fraction of EM-C11 cells DNA strand exchanges might Despite the fact that EM-Cu is 5 times more sensitive to the cell occur at the position of stalled replication forks, leading to gross killing effects of EMS than CHO9, it is not hypermutable when molecular changes. This model is supported by the observation that compared at equal exposure. This indicates that DNA damage intro the frequency of EMS induced chromosomal aberrations in EM-Cl! duced by EMS leads to additional cell killing in EM-Cu, but that this is considerably higher than in CHO9 cells (2). damage does not lead to additional mutations. This is in contrast to Alternatively, base excision repair of 3-ethyladenine generates tran results obtained with another EMS-sensitive Chinese hamster cell line siently gaps in DNA which are visible as single strand breaks. EM (EM9) belonging to the same complementation group which has been Cli (as well as EM9) has been reported to be defective in repair of shown to be hypermutable with EMS relative to its parental line AA8 single strand breaks (2). These unrepaired breaks could be responsible (3). The difference between EM9 and EM-Cu in response to EMS, for the generation of deletion type of mutations. concerning mutability at the hprt locus, provides additional evidence The results presented here do not exclude the possibility that only for phenotypical heterogeneity among this class of DNA repair defi a subset of cells contributes to mutagenicity, because asynchronous cient Chinese hamster cell lines. This has already been suggested by cell populations were used. Assuming that DNA damage is converted Zdzienicka et al. (2) to explain other phenotypic differences between into mutations during DNA replication, the position in the cell cycle EM-Cu and EM9, such as the lower sensitivity to X-rays in EM-Cu in which cells are at the moment of treatment determines how much compared to EM9. These differences could be due to different muta time is available for the repair of induced lesions prior to mutation tions in the xrcc-! gene or to differences between the two parental fixation. For lesions that are slowly repaired or not repaired at all, like lines (CHO9 and AA8, respectively) which might carry additional 06-ethylguanine in Chinese hamster cells, the contribution to muta mutations influencing the phenotype of these mutants. tion induction will be similar in all cells. However, lesions, like MutationalAnalysisof InducedhprtMutants.Forcomparison 3-alkyladenine (28), that are rapidly repaired might contribute to of the nature of the EMS induced single base pair changes in CHO9 mutations only in cells that were just entering S phase at the time of and in EM-Cuu only the mutations in the coding sequence have been treatment. This could lead to an underestimation of the mutagenic considered. Both spectra are dominated by a large number of potential of such adducts. GC—@ATtransitions, in agreement with previous studies (11—15). The influence of the cell cycle on mutation induction is currently Many of the GC—@ATtransitions are probably caused by misincor under investigation in synchronized populations of Chinese hamster poration of thymine opposite 06-ethylguanine (16—18),which is a cells. These studies should help to determine the mechanism of persistent lesion in Chinese hamster cells that lack 06-alkylguanine formation of AT—*(JC@transitions after EMS treatment in particular DNA alkyltransferase activity (19). and the nature of mutagenic lesions in general. Furthermore, the In addition, the spectrum in CHO9 shows a class of AT—@GC biological role of the xrcc-1 gene remains to be unraveled. More transitions (16%), which is absent in the spectrum of EM-C1l. The insight into the function of its gene product will also contribute to a origin of these EMS induced AT—*GCtransitions, which have been better understanding of mechanisms leading to deletion type muta observed at low frequencies also by other investigators (12, 13), is not tions in EM-Cu. completely clear. These types of transitions induced by ENU (20, 21) are considered to be caused by 04-ethylthymine (22, 23). Since EMS REFERENCES treatment hardly induces this adduct in DNA (24), another lesion should be responsible for the EMS-induced AT—*GC transitions. 1. Vogel, E., and Natarajan, A. T. The relation between reaction kinetics and mutagenic action of monofunctional alkylating agents in higher eucaryotic systems: interspecies Indirect evidence that 3-alkyladenine might be the responsible muta comparisons. In: A. Holeander and F. J. de Serres (eds.), Chemical Mutagens, Vol. 7, genic lesion has recently been presented by Kiungland et a!. (25). pp. 295—336.New York: Plenum Publishing Corp., 1982. 2. Zdzienicka, M. Z., van der Schans, 0. P., Natarajan, A. T., Thompson, L H., Chinese hamster cells, transfected with the Escherichia coli 3-methyl Neuteboom, I., and Simons, J. W. I. M. A Chinese hamster ovary cell mutant adenine DNA glycosylase I gene, show a 2-fold reduction in hprt (EM-Cl 1) with sensitivity to simple alkylating agents and a high level of sister mutation frequency after methyl methanesulfonate treatment. 3-Alkyl chromatid exchanges. Mutagenesis, 7: 265—269,1992. 3. Thompson, L H., Brookman, K. W., Dillehay, L E., Carrano, A. V., Mazrimas, J. A., adenine might generate AT—@GCtransitions either directly or via the Mooney, C. L, and Minkler, J. L. A CHO-cell strain having hypersensitivity to formation of abasic sites during base excision repair of 3-alkyl mutagens, a defect in DNA strand-break repair, and an extraordinary baseline adenine. However, in case of a mechanism via abasic sites one would frquency of sister-chromatid exchange. Mutat. Rca., 95: 427—440, 1982. 4. Thompson, L H., Brookman, K. W., Jones, N. J., Allen, S. A., and Can, A. V. expect to find AT—@TAtransversions as well, because opposite to an Molecular cloning of the human XRCCI gene, which corrects defective DNA strand AP site preferentially adenine is incorporated (A-rule). Since no break repair and sister chromatid exchange. Mol. Cell. Biol., 10: 6160—6171, 1990. 5. Zdzienicka, M. Z., and Simons, J. W. I. M. Analysis of repair processes by the AT—@TAtransversions were found, we propose that EMS-induced determination of the induction of cell killing and mutations in two repair-deficient AT—+GCtransitions are generated by direct mispairing of 3-ethyl Chinese hamster ovary cell lines. Mutat. Res., 166: 59—69,1986. adenine with cytosine. 6. Vrieling, H., Niericker, M. J., Simons, J. W. I. M., and van Zeeland, A. A. Molecular analysis of mutations induced by N-ethyl-N-nitrosourea at the HPRT locus in mouse In the mutational spectrum of EM-Cu, the major class of muta lymphoma cells. Mutat. Res., 198: 99—106,1988. tions found is GC—@ATtransitions as observed in CHO9 cells. The 7. Menichini, P., Vrieling, H., and van Zeeland, A. A. Strand-specific mutation spectra 3005

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1994 American Association for Cancer Research. 3-ETHYLADENINE CAUSES AT-. GC TRANSITIONS

in repair-proficient and repair-deficient hamster cells. Mutat. Res., 251: 143—155, and the mutagenic properties of these bases. Mutat. Res., 233: 81—94,1990. 1991. 19. Dogliotti, E., Vitelli, A., Terlizzese, M., Di Muccio, A., Calcagnile, A., Safflotti, U., 8. Zhang, L-H., and Jenssen, D. Site specificity of N-methyl-N-nitrosourea-induced and Bignami, M. Induction kinetics of mutations at two genetic loci, DNA damage transition mutations in the hprt gene. Carcinogenesis (Lond.), 12: 1903—1909,1991. and repair in CHO cells after different exposure times to N-ethyl-N-nitrosourea. 9. Rossiter, B. J. F., Fuscoe, J. C., Munzy, D. M., Fox, M., and Caskey, C. T. The Carcinogenesis (Land.), 8: 25—31,1987. Chinese hamster HPRT gene: restriction map, sequence analysis, and multiplex PCR 20. Pastink, A., Vreeken, C., Nivard, M. J. M., Searles, L. L., and Vogel, E. W. Sequence deletion screen. Genomics, 9: 247—256,1991. analysis of N-ethyl-N-nitrosourea-induced vermillion mutations in Drosophila mela 10. Jansen, J. G., Vrieling, H., van Zeeland, A. A., and Mohn, G. R. The gene encoding nogaster. Genetics, 123: 123—129,1989. hypoxanthine-guanine phosphoribosyltransferase as target for mutational analysis: 21. Bronstein, S. M., Cochrane, J. E., Craft, T. R., Swenberg, J. A., and Skopek, T. R. PCR cloning and sequencing of the cDNA of the rat. Mutat. Res., 266: 105—116, Toxicity,mutagenicity,andmutationalspectraofN-ethyl-N-nitrosoureainhumancell 1992. lines with different DNA repair phenotypes. Cancer Res., 51: 5188—5197, 1991. I 1. Lebkowski, J. S., Miller, J. H., and Cabs, M. P. Determination of DNA sequence 22. Preston, B. D., Singer, B., and Loeb, L. A. Mutagenic potential of (Y'-methylthymine changes induced by ethyl methanesulfonate in human cells, using a shuttle vector in vivo determined by an enzymatic approach to site-specific mutagenesis. Proc. NatI. system. Mol. Cell. Biol., 6: 1838—1842,1986. Acad. Sci. USA, 83: 8501—8505, 1986. 12. Pastink, A., Heemskerk, E., Nivard, M. J. M., van Vliet, C. J., and Vogel, E. W. 23. Preston, B. D., Singer, B., and Loeb, L. A. Comparison of the relative mutagenicities Mutational specificity of ethyl methanesulfonate in excision-repair-proficient and of O-alkylthymines site-specifically incorporated into 4X174 DNA. J. Biol. Chem., -deficient strain of Drosophila melanogaster. Moi. Gen. Genet., 229: 213—218,1991. 262:13821—13827,1987. 13. Ashman, C. R. DNA base sequence changes in spontaneous and ethyl methanesul 24. Beranek, D. T. Distribution of methyl and ethyl adducts following alkylation with fonate-induced mutations of a chromosomally-integrated gene in Chinese hamster monofunctional alkylating agents. Mutat. Res., 231: 11—30,1990. ovary cells. Mutat. Res., 270: 115—124,1992. 25. Klungland, A., Fairbairn, L., Watson, A. J., Margison, G. P., and Seeberg, E. 14. Kunz, B. A., Gabriel, M., Kang, X., Kohalmi, S. E., and Terrick, K. A. DNA repair Expression of the E. coli 3-methyladenine DNA glycosylase I gene in mammalian modifies the site specific site and strand specificity of ethyl methanesulfonate mu cells reduces the toxic effects of methylating agents. EMBO J., 11: 4439—4444, 1992. tagenesis in yeast. Mutagenesis, 7: 461—469,1992. 26. Boiteux, S., Huisman, 0., and Laval, J. 3-Methyladenine residues in DNA induce the 15. Lee, G. S-F., Blonsky, K. S., Van On, D. L., Savage, E. A., Morgan, A. R., and von SOS function sfi4 in Escherichia coli. EMBO J., 3: 2569—2573, 1984. Borstel, R. C. Base alterations in yeast induced by alkylating agents with differing 27. Larson, K., Sahm, J., Shenkar, R., and Strauss, B. Methylation-induced blocks to in Swain-Scott substrate constants. J. Mol. Biol., 223: 617—626, 1992. vitro DNA replication. Mutat. Res., 150: 77—84,1985. 16. Loechler, E. L., Green, C. L., and Essigmann, J. M. In vivo mutagenesis by O@ 28. Habraken, Y., and Laval, F. Increased resistance of the Chinese hamster mutant irsl methylguanine built into a unique site in a viral genome. Proc. Natl. Acad. Sci. USA, cells to monofunctional alkylating agents by transfection of the E. coli or mammalian 81: 6271—6275,1984. N3-methyladenine-DNA-glycosylase genes. Mutat. Res., 293: 187—195,1993. 17. Williams, L. D., and Shaw, B. R. Protonated base pairs explain the ambiguous pairing 29. Jolly, D. J., Okayama, H., Berg, P., Esty, A. C., Filupa, D., Bohlen, P., Johnson, G. [email protected],84: 1779—1783,1987. G.,Shively,J.E., Hunkapilar,T.,andFiredmann,T.Isolationandcharacterization 18. Swann, P. F. Why do O'@-alkylguanine and O@-alkylthymine miscode? The relation of a full-length expressible cDNA for human hypoxanthine phosphoribosyl-trans ship between the structure of DNA containing O@-alkylguanineandO'-alkylthymine ferase. Proc. Nat!. Acad. Sci. USA, 80: 477—481,1983.

3006

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1994 American Association for Cancer Research. Molecular Analysis of Ethyl Methanesulfonate-induced Mutations at the hprt Gene in the Ethyl Methanesulfonate-sensitive Chinese Hamster Cell Line EM-C11 and Its Parental Line CHO9

Christel W. Op het Veld, Matgorzata Z. Zdzienicka, Harry Vrieling, et al.

Cancer Res 1994;54:3001-3006.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/54/11/3001

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/54/11/3001. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1994 American Association for Cancer Research.