175

Mutation Research, DNA Repair, 255 (1991) 175-182 © 1991 Elsevier Science Publishers B.V. 0921-8777/91/$03.50 ADONIS 092187779100092P MUTDNA 06455

The origin of O 6 - m e t h y l g u a n i n e - D N A methyltransferase in Chinese hamster ovary cells transfected with h u m a n D N A Keizo Tano a, Susumu Shiota a, Joanna S. Remack b, Thomas P. Brent b, Darell D. Bigner c and Sankar Mitra a a University of Tennessee-Graduate School of Biomedical Sciences and Biology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, b Department of Biochemical and Clinical Pharmacology, St. Jude Children's Research Hospital, Memphis, TN 38101 and c Department of Pathology, Duke University Medical Center, Durham, NC 27710 (U.S.A.) (Received 30 December 1990) (Revision received 19 March 1991) (Accepted 20 March 1991)

Keywords: O6-Methylguanine-DNA methyltransferase; CHO cell; M e r - / M e x - ; MGMT mRNA

Summary Transfection of Chinese hamster ovary (CHO) cells with human DNA has been shown in several laboratories to produce clones which stably express the DNA-repair protein, O6-methylguanine-DNA methyltransferase (MGMT), that is lacking in the parent cell lines (Mex- phenotype). We have investigated the genetic origin of the MGMT in a number of such MGMT-positive (Mex ÷) clones by using human MGMT eDNA and anti-human MGMT antibodies as probes. None of the five independently isolated Mex ÷ lines has human MGMT gene sequences. Immunoblot analysis confirmed the absence of the human protein in the extracts of these cells. The MGMT mRNA in the lines that express low levels of MGMT (0.6-1.4 × 10 4 molecules/cell) is of the same size (1.1 kb) as that present in hamster liver. One cell line, GC-1, with a much higher level of MGMT (4 x 10 4 molecules/cell) has two MGMT mRNAs, a major species of 1.3 kb and a minor species of 1.8 kb. It has also two MGMT polypeptides (32 and 28 kDa), both of which are larger than the 25 kDa MGMT present in hamster liver and other Mex ÷ transfectants. These results indicate that the MGMT in all Mex ÷ CHO cell clones is encoded by the endogenous gene. While spontaneous activation of the MGMT gene cannot be ruled out in the Mex ÷ cell clones, the intervention of human DNA sequences may be responsible for activation of the endogenous gene in the GC-1 line.

O6-Methylguanine-DNA methyltransferase (MGMT) (EC 2.1.1.63) is an unusual and ubiquitous repair protein responsible for the removal of

mutagenic and carcinogenic O6-alkylguanine produced in DNA by simple alkylating carcinogens such as N-alkylnitrosamines (Pegg and Singer,

Correspondence: Dr. Keizo Tano, University of Tennessee Graduate School of Biomedical Sciences and Biology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 (U.S.A.), Phone (615) 574-0963; Fax (615) 574-1274.

Abbreviations: CHO, Chinese hamster ovary; CNU, 2-chloroethyl-N-nitrosourea; MGMT, O6-methylguanine-DNA methyltransferase; MNU, N-methyl-N-nitrosourea, PAGE, polyacrylamide gel electrophoresis; PMSF, phenylmethanesulfonyl fluoride.

176

1984). The protein restores the original guanine in DNA by in situ dealkylation of the adduct in a second-order stoichiometric reaction in which the protein accepts the alkyl group at a unique cysteine residue resulting in its inactivation (Lindahl et al., 1988; Bhattacharyya et al., 1990; Foote et al., 1986). Various cultured cell lines display different characteristic levels of MGMT and some cell lines, called Mer- or Mex-, have undetectable MGMT (Day et al., 1980; Sklar and Strauss, 1981). The Mex- cells are much more sensitive to alkylating agents that produce adducts at the 0-6 position of guanine than the MGMTpositive Mex + cells (Day et al., 1987; Ding et al., 1985; Barrows et al., 1987). Two common Chinese hamster cell lines, namely Chinese hamster ovary (CHO) cells and V79 cells, have a Mex- phenotype (Foote et al., 1986). CHO cells were transfected with human DNA and transfectants were selected for resistance to 2-chloroethyl-N-nitrosourea (CNU), a DNAcross-linking agent (Ding et al., 1985; Barrows et al., 1987; Dolan et al., 1989; Dunn et al., 1991). The CNU-resistant cell clones were found to have significant levels of MGMT, consistent with the notion that the toxicity of CNU is due to cross-linking of DNA which results from the intermediate adduct O6-chloroethylguanine. Repair of this adduct by MGMT would prevent such cross-linking (Brent, 1985; Ludlum et al., 1986). Repeated attempts to clone the human MGMT gene from Mex + CHO cells following a protocol used for cloning human excision repair genes (Weber et al., 1988; Bootsma et al., 1988) have been unsuccessful (Dunn et al., 1991). This raised the question of whether the activity of MGMT in transfectants was encoded by the human gene or resulted from activation of the endogenous hamster gene. We have recently succeeded in cloning the human MGMT cDNA using an unrelated strategy (Tano et al., 1990). Monoclonal antibodies specific for the human MGMT have also been produced recently (Brent et al., 1990). The availability of the cDNA and antibody has now allowed us to address the question of the origin of MGMT in Mex + CHO cells. In this paper we show that the expression of MGMT in several transfected Mex + CHO cell lines is due to activation of the endogenous hamster gene. Interven-

tion of human DNA appeared to be necessary for the high-level MGMT expression in at least one of these clones. Materials and methods

Cell lines. Chinese hamster ovary (CHO) cell clone K1-BH 4 (O'Neil et al., 1977) was transfected by electroporation (Zerbib et al., 1985) with human (HeLa $3) cell DNA and pSV2 neo plasmid (Southern and Berg, 1982) for the isolation of GC-1 and GC-2 cell clones. These cells were cloned from transformants, originally selected for resistance to the antibiotic G418, after exposure to 2-chloroethyl-N-nitrosourea (CNU) as described previously (Dunn et al., 1991). The GC-13 clone was isolated in a secondary transfection procedure in which GC-1 cell DNA and pSV2 neo were introduced into fresh CHO cells by the CaPO 4 method (Lowy et al., 1978). CHOMex + (Ding et al., 1985) and BR1 and BRM1 (Barrows et al., 1987) cell lines were independently isolated by transfecting CHO cells with human DNA. Human leukemic lymphoblasts (CEM-CCRF line) were grown as previously described (Brent, 1985). Nucleic acid hybridization. 10 p~g of DNA were digested with BstEII, BamHI, or KpnI, separated by the electrophoresis on 0.8% agarose gel, and finally transferred to nylon membrane (Hybond-N, Amersham) for hybridization (Maniatis et al., 1982) with human MGMT cDNA (Tano et al., 1990). The cDNA probe was labeled with 32p-dCTP by the random hexanucleotide priming method (Feinberg and Vogelstein, 1983). Hybridization was carried out at 65°C for 20 h in 0.5 M sodium phosphate buffer (pH 7.2) containing 2 mM EDTA, 7% SDS and 1% nonfat dry milk. The filter was washed with 0.3 M NaC1/0.03 M sodium citrate, pH 7.0, 0.1% SDS at 65°C for 1 h, dried, and exposed to Kodak X-OMAT AR film. Poly(A)+RNA was extracted from proteinase K-digested cell lysates by oligo(dT)-cellulose chromatography according to the method of Badley et al. (1988). 2 /.~g of poly(A)+RNA was separated by electrophoresis on 1.2% agarose gel containing 2.2 M formaldehyde (Maniatis et al.,

177

1982) followed by capillary transfer to nylon membrane (Hybond-N, Amersham). Hybridization with 32p-labeled eDNA insert of pKT100 (Tano et al., 1990) or with human y-actin cDNA fragment (Gunning et al., 1983) was carried out as described above. The filters were washed with 0.15 M NaC1/0.015 M sodium citrate, pH 7.0, 0.1% SDS for 1 h at 65°C and exposed to X-ray film.

Analysis of methyltransferase by immunoblotting and fluorograph. 200-300 /zg of frozen cells or liver were sonicated in 2 vol. of 50 mM Tris-HCl (pH 7.5), 2 mM EDTA, 1 mM DTT, 0.02% NaN 3 and 0.1 M NaCI (buffer A) to which aprotinin and PMSF were added to 20 units/1 and 0.2 mM, respectively. The supernatant after ultracentrifugation (100000 X g, 45 min) (Fraction I) was either used as such for immunoblotting and fluorography or further purified, by affinity chromatography on 0.5 ml Sepharose 4B linked to single-stranded DNA (Brent, 1985) (Fraction II). The [3H-Me]-methylated MGMT was prepared by incubating these active MGMT preparations with [3H]MNU-treated calf-thymus DNA at 37°C for 30 min before analysis by S D S / P A G E in 0.75-mm thick slab gels (Laemmli, 1970). Following transfer of the protein to PVDF (ImmobilonP, Millipore) membranes by electroblotting (Matsudaira, 1987) the membranes were probed with monoclonal or polyclonal antibodies specific for MGMT (Brent et al., 1990; von Wronski et al., 1989) and were visualized by a silver-enhanced gold staining procedure (von Wronski et al., 1989). The same membranes were then treated with EN3HANCE spray (Dupont) before exposure to Kodak X-OMAT AR films for up to 2 weeks. Results

MGMT activity in transfected cells. Table 1 shows that all of the transfected cells except GC-1 have 6-13000 MGMT mol/cell compared to undetectable activity in the parent cells. GC-1 has a much higher level of MGMT expression (Dunn et al., 1991).

Southern blot analysis of hamster cell DNAs. Fig. 1A shows identical pattern of DNA fragments from GC-1, GC-2, GC-13, and Chinese hamster spleen that hybridized with the human MGMT eDNA. Only a few bands hybridized with the human cDNA probe for each of the enzymes. These profiles are totally different from that of human placental DNA (Fig. 1B). The absence of any additional MGMT-specific DNA band indicates the absence of human MGMT gene sequences in all the transfected lines. We have also confirmed the absence of human MGMT gene in other Mex+-CHO transfectant lines, namely C H O - M e x ÷ cells (Ding et al., 1985) and BR1 and BRM1 cells (Barrows et al., 1987) (data not shown).

Analysis of MGMT transcripts. The mature transcript of hamster MGMT is 1.1 kb, slightly larger than that of human MGMT. We determined the size of MGMT transcripts in hamster liver and transfectant CHO lines by Northern blot analysis (Fig. 2). The parent Mex- cells, similar to all human Mex- cell lines examined thus far (Tano et al., 1990; Fornace et al., 1990), do not have any detectable MGMT mRNA The stable MGMT messages of GC-2, GC-13, BR1 and BRM1 cells consisted of a single species of the same size as that of hamster liver. In contrast, the GC-1 cells have two species of MGMT mRNA of 1.8 and 1.3 kb, with the smaller message in larger abundance.

TABLE 1 MGMT ACTIVITY OF CHO TRANSFECTANT CELL LINES The MGMT activity was assayed as described earlier (Foote et al., 1983). Cell line

MGMT molecules/cell

AA8 and K1-BH4 (control) GC-1 GC-2 GC-13 CHO-Mex + BRM1 BR1

Undetectable ( < 200 molecules/cell) 40000 6000 11000 12000 13100 13900

178

A I

2

3

4

5

6

7

8

9

10

11

B

12

I

-

2

3

28.0

-

9.4-~

-

6 . 6 -N

-

4.4

-

2.3~

-

2.0-- N

Fig. 1. Southern blot analysis of Chinese hamster and human DNAs with human MGMT cDNA. (A) DNA samples from Chinese hamster spleen (lanes 1, 5 and 9), GC-1 cells (lanes 2, 6 and 10), GC-2 cells (lanes 3, 7 and 11) and GC-13 (lanes 4, 8 and 12) were digested with BstEII (lanes 1-4), BamHI (lanes 5-8), and KpnI (lanes 9-12~. (B) Human placental DNA was digested BstEII (lane 1), BamHl (lane 2), and Kpnl (lane 3). The position of markers correspond to HindlII-fragments of ,~ DNA.

T h e M G M T message in C H O - M e x + app e a r e d to be of the same size as that of the liver (data not shown).

Analysis of MGMT polypeptides by fluorography and immunoblotting. M o n o c l o n a l antibodies specific for the h u m a n M G M T were used to identify the genetic origin o f M G M T polypeptides in different Mex ÷ h a m s t e r cells. T h e fiuorograph of labeled M G M T shows that crude extracts o f h a m s t e r liver and GC-2 cells have a single radioactive polypeptide of a p p a r e n t molec-

ular mass of 28 kDa, about 3 k D a larger than the h u m a n M G M T (Fig. 3B). A l t h o u g h h u m a n M G M T is known to have a molecular mass of 21.7 kDa, it migrated with an a p p a r e n t size of 25 k D a in S D S / P A G E in these studies. T h e reasons for this discrepancy are not clear. It does not a p p e a r to be due to posttranslational modification of the protein (von Wronski et al., 1991), but might result from the high representation of low molecular weight amino acids as suggested by R y d b e r g et al. (1990). GC-13 also showed a single polypeptide b a n d of the same size as those from

179

hamster liver and GC-2 (data not shown). In contrast, GC-1 has at least two labeled polypeptides. The predominant species had an apparent size of 32-33 k D a and a minor species of 28-29 kDa (Fig. 3B). With partially purified preparations, multiple M G M T bands migrating in between these two bands have been observed and the amount of the 32-kDa band was correspondingly reduced. But because these were more commonly present in old samples, these were probably degradation products of the larger species. As predicted from the Southern blot data, none of the M G M T polypeptides in GC-1, GC-2, and GC-13 a p p e a r e d to be of human origin because none of them cross-reacted with a human MGMT-specific monoclonal antibody 4A.1 (Fig. 3A). Although nonspecific bands were observed for all cell extracts, a band in the same position

A

B

1

2

1

2

3

4

5

6

- 1.52

1.52

-1,28

1.28

-0.78

0.78

- 0.40

-0.28

kDa

97.4._ 66.2--

42.7-31.0-21.5--

14.4-12

54

1 2

54

Fig. 3. SDS/PAGE analysis of MGMT polypeptides. Panels A and B show immunoblot and fluorograph, respectively, of MGMT polypeptide in Fraction II of CEM cells (80 fmoles MGMT) (lane 1), GC-1 cells (230 fmoles) (lane 2), hamster liver (100 fmoles) (lane 3), and GC-2 cells (150 fmoles) (lane 4). Monoclonal antibody 4.A1 was used for immunoprobing (von Wronski et al., 1989). The fluorograph was exposed for 3 days.

was observed in both Western blot and fluorograph only for the human C E M cell extract. No cross-reacting band was observed in the positions of 28-kDa M G M T polypeptides in GC-1, GC-2 and hamster liver or 32-kDa polypeptide of GC-1 cell. The weak fluorograph band for the human cell extract was due to quenching of radioactivity in the antigen-antibody complex.

0.40

0,28

Fig. 2. Analysis of MGMT transcripts in hamster cells. Panels A and B show probing with MGMT eDNA Panel C (same membrane as used as Panel B) shows probing with human y-actin eDNA. Panel (A); 20/zg each of RNAs from Chinese hamster liver (lane 1) and GC-1 cells (lane 2). Panel (B); 2/zg each of RNAs from GC-1 cells (lane 1), GC-2 (lane 2), GC-13 (lane 3), BR1 (lane 4), BRM1 (lane 5) and HeLaS3 (lane 6) were used. An RNA ladder (BRL) was used as the marker.

Discussion

Several h u m a n genes involved in excision repair have been cloned by phenotypic rescue of repair-deficient C H O mutant cells transfected with human D N A (Weber et al., 1988; Bootsma et al., 1988). Following the first observation of Ding et al. (1985) that M e x - C H O cells can be made Mex + by transfection with human DNA, we and others have independently isolated similar Mex ÷ C H O cells (Ding et al., 1985; Barrows et al., 1987; Dolan et al., 1989; Dunn et al., 1991), However, attempts to clone the human M G M T gene from such cells, following the approach used

180 for excision repair genes, were unsuccessful (Dunn et al., 1991). Our success in cloning the human MGMT cDNA (Tano et al., 1990) has now allowed us to demonstrate the absence of the human MGMT gene in five independently isolated Mex ÷ CHO cell lines. Furthermore, MGMT polypeptides in at least GC-1 and GC-2 cells did not cross react with the human MGMTspecific antibodies. These data thus suggest that the MGMT expression in CHO cell transfectants is due to derepression of the endogenous hamster gene. Our preliminary experiments indicate that the human MGMT gene has a size larger than 150 kb (Tano et al., unpublished results), much larger than the estimated average size of DNA fragments used in transfection of CHO cells. Thus, in retrospect, the approach of phenotypic rescue was unlikely to have led to the cloning of the human MGMT gene. The GC-1 cell line has a unique feature in that its MGMT level is 5-10 times higher than that of all other Mex ÷ CHO cell lines tested. Our observation that MGMT mRNA in GC-1 consists of two species, both larger than the endogenous hamster mRNA, supports the possibility of a transcription fusion, at least in this cell line. The presence of two mRNA populations could reflect either alternate splicing of the transcript of one allele (Breitbart et al., 1987) or activation of both alleles of the MGMT gene in the diploid cell. It is interesting that the more abundant MGMT message of GC-1 cells has a size of 1.3 kb with a 1.8-kb message present in a smaller amount. Two GC-1 polypeptides of apparent molecular masses of 32 and 28 kDa have MGMT activity and the smaller polypeptide is of the same size as that present in hamster liver. Assuming similar stability of the two mRNAs, we may conclude that the 1.3-kb mRNA codes for the more abundant 32kDa MGMT protein, and the 1.8-kb mRNA in GC-1 codes for the smaller polypeptide. The 1.1kb MGMT mRNA in GC-2 cells is of similar size to that in hamster liver and apparently encodes the 28-kDa native protein. Taking into account all the observations described in this study, we propose the following possible model for the activation of MGMT in CHO cells. The completely repressed MGMT gene in CHO cells may be activated by insertion

of an exogenous DNA fragment within the regulatory sequence of the gene. Because electroporation leads to uptake of a small amount of exogenous DNA (Boggs et al., 1986) it is possible that some human DNA of unknown sequence derepressed the endogenous MGMT gene even though repetitive human DNA sequences could not be detected in these cells (Dunn et al., 1991). Depending on the precise location of insertion, the mature MGMT transcript may or may not contain a sequence of the transfecting DNA. In cells with low levels of MGMT, the promoter and the coding sequence of the transcript may remain unaltered. Thus, the level of transcription would be limited by the strength of the endogenous promoter and the MGMT level would not exceed that observed in hamster liver. While spontaneous activation of the MGMT gene in these cells cannot be ruled out, we have been unable to isolate such revertants by CNU-selection in the absence of prior transfection treatment (Dunn ct al., 1991). The much higher levels of MGMT observed in GC-1 cells could be the result of integration of an exogenous promoter sequence upstream of the hamster MGMT gene resulting in a fusion polypeptide. This model can be experimentally tested by cloning the cDNAs of GC-1 and GC-2 cells and by demonstrating the presence or absertce of human sequences in the mature transcripts. Furthermore, the cloning of the MGMT gene from GC-1 cells could lead to identification of the putative human promoter integrated in the hamster genome. It is even possible that this promoter sequence was derived from the human MGMT gene. In any case, these experiments may also elucidate the molecular mechanism of extinction of MGMT gene in Mex- cells.

Acknowledgments We thank Dr. R.S. Foote for a critical reading of the manuscript and helpful comments and Mr. W.C. Dunn for expert technical assistance. We also express our gratitude to Drs. E. Bresnick and L.B. Barrows for their gifts of cell lines. Ms. Daisy Sams provided expert secretarial assistance.

181 This work was s u p p o r t e d by the Office of H e a l t h a n d E n v i r o n m e n t a l Research, U.S. Dep a r t m e n t of Energy, u n d e r c o n t r a c t D E - A C 0 5 84OR21400 with the M a r t i n M a r i e t t a E n e r g y Systems, Inc., a n d by the Public H e a l t h Service grants C A 31721 (to S.M.) a n d C A 36888, C A 14799 a n d C A 21765 (to T.P.B.) a n d Public H e a l t h Service grants C A 11898 a n d NS 20023 a n d Bristol-Myers Squibb g r a n t 100-RIB (to D.D.B.).

References Badley, J.E., G.A Bishop, T. St. John and J.A Frelinger (1988) A simple, rapid method for the purification of polyA+ RNA, Bio. Tech., 6, 114-116. Barrows, L.B., A.H. Borchers and M.B. Paxton (1987) Transfectant CHO cells expressing O6-alkylguanine-DNA alkyltransferase display increased resistance to DNA damage other than O6-guanine alkylation, Carcinogenis, 8, 1853-1859. Bhattacharyya, D., R.S. Foote, A.M. Boulden and S. Mitra (1990) Physicochemical studies of human O6-methylguanine-DNA methyltransferase, Eur. J. Biochem., 193, 337-343. Boggs, S.S., R.G. Gregg, N. Borenstein and O. Smithies (1986) Efficient transformation and frequent single-site single copy insertion of DNA can be obtained in mouse erythroleukemia cells transformed by electroporation, Exp. Hematol., 14, 988-994. Bootsma, D., A. Westerveld and J.H.J. Hoeijmakers (1988) DNA repair in human cells: from genetic complementation to isolation of genes, Cancer Sure., 7, 303-315. Breithart, R.E., A. Andreadis and B. Nadal-Ginard (1987) Alternative splicing: A ubiquitous mechanism for the generation of multiple protein isoforms from single genes, Annu. Rev. Biochem., 50, 467-494. Brent, T.P. (1985) Isolation and purification of O6-alkylguanine-DNA alkyltransferase from human leukemic cells, Prevention of chloroethylnitrosourea-induced crosslinks by purified enzyme, Pharmacoi Ther., 31, 121-140. Brent, T.P., M.A. von Wronski, C.T. Pegram and D.D. Bigner (1990) Immunoaffinity purification of human O6-alkylguanine-DNA alkyltransferase using newly developed monoclonal antibodies, Cancer Res., 50, 58-61. Day, R.S., C.H.J. Ziolkowski, D.A. Scudiero, S.A. Meyer, A.S. Lubiniecki, A.J. Girardi, S.M. Galloway and G.D. Bynum (1980) Defective repair of alkylated DNA by human tumour and SV40-transformed human cell strains, Nature (London), 288, 724-727. Day III, R.S., M.A. Babich, D.B. Yarosh and D.A. Scudiero (1987) The role of O6-methylguanine in human cell killing, sister chromatid exchange induction and mutagenesis: A review, J Cell Sci., Suppl. 6, 333-353. Ding, R., K. Ghosh, A. Eastman and E. Bresnick (1985) DNA-mediated transfer and expression of a human DNA

repair gene that demethylates O6-methylguanine, Mol. Cell. Biol., 5, 3293-3296. Dolan, M.E., L. Norbeck, C. Clyde, N.K. Hora, L.C. Erickson and A.E. Pegg (1989) Expression of mammalian O6-alkylguanine-DNAalkyltransferase in a cell line sensitive to alkylating agents, Carcinogenesis, 10, 1613-1619. Dunn, W.C., K. Tano, G.J. Horesovsky, R.J. Preston and S. Mitra (1991) The role of O6-alkylguanine in cell killing and mutagenesis in Chinese hamster ovary cells, Carcinogenesis, 12, 83-89. Feinberg, A.P., and B. Vogelstein (1983) A technique for radiolabeling DNA restriction fragments to high specific activity, Anal. Biochem., 132, 6-13. Foote, R.S., B.C. Pal and S. Mitra (1983) Quantitation of O6-methylguanine-DNA methyltransferase in HeLa cells, Mutation Res., 119, 221-228. Foote, R.S., W.C. Dunn and S. Mitra (1986) Regulation of repair of O6-methylguanine in mammalian cells, in: N.K. Notani and P.S. Chauhan (Eds.), Environmental Mutagenesis and Carcinogenesis, Bhaba Atomic Research Center, Bombay, pp. 99-115. Fornace, A J., M.A. Papathanasiov, M.C. Hollander and D.B. Yarosh (1990) Expression of the O6-methylguanine-DNA methyltransferase gene MGMT in Mer + and Mer- human tumor cells, Cancer Res., 50, 7908-7911. Gunning, P., P. Ponte, H. Okayama, J. Engel, H. Blau and L. Kedes (1983) Isolation and characterization of full-length cDNA clones for human a-, /3-, and y-actin mRNA: Skeletal but not cytoplasmic actins have an amino-terminal cysteine that is subsequently removed, Mol. Cell. Biol., 3, 787-795. Laemmli, W.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature (London), 227, 680-685. Lindahl, T., B. Sedgewick, M. Sekiguchi and Y. Nakabeppu (1988) Regulation and expression of the adaptive response to alkylating agents, Annu. Rev. Biochem., 57, 133-157. Lowy, D.R., E. Rands and E.M. Scolnick (1978) Helper-independent transformation by unintegrated Harvey Sarcoma Virus DNA, J. Virol., 26, 291-298. Ludlum, D.B., J.R. Mehta and W.P. Tong (1986) Prevention of' 1-(-3-deoxycytidyl),2-(1-deoxyguanosinyl)ethane crosslink formation in DNA by rat liver O6-alkylguanine-DNA alkyltransferase, Cancer Res., 46, 3353-3357. Maniatis, T., E.F. Fritsch and J. Sambrook (1982) Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, New York. Matsudaira, P. (1987) Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes, J Biol. Chem., 262, 10035-10038. O'Neill, J.P., P.A. Brimer, R. Machanoff, G.P. Hirsch and A.W. Hsie (1977) A quantitative assay of mutation induction at the hypoxanthine-guanine phosphoribosyl transferase locus in Chinese hamster ovary cells (CHO/HGPRT system), Mutation Res., 45, 91-101. Pegg, A., and B. Singer (1984) Is O6-alkylguanine necessary for initiation of carcinogenesis by alkylating agents? Cancer Invest., 2, 223-236.

182 Rydberg, B., N. Spurr and P. Karran (1990) cDNA cloning and chromosomal assignment of the human O6-methyl guanine-DNA methyltransferase, J. Biol. Chem., 265, 9563-9569. Sklar, R., and B. Strauss (1981) Removal of O6-methyl guanine from DNA of normal and xeroderma pigmentosum-derived lymphoblastoid lines, Nature (London), 289, 417-420. Southern, P.J., and P. Berg (1982) Transformation of mammalian cells to antibiotic resistance with a bacterial gene under control of the SV40 early region promotor, J. Mol. Appl. Genet., 1,327-341. Tano, K., S. Shiota, J. Collier, R.S. Foote and S. Mitra (1990) Isolation and structural characterization of a cDNA clone encoding the human DNA repair protein for O6-al kylguanine, Proc. Natl. Acad. Sci. (U.S.A.), 87, 686-690. von Wronski, M.A, T.P. Brent and D.D. Bigner (1989) Ex-

pression of O6-alkylguanine-DNA alkyltransferase in mer + and m e t - human cell extracts probed with specific monoclonal antibodies, Cancer Commun., 1,323-327. yon Wronski, M.A., S. Shiota, K. Tano, S. Mitra, D.D. Bigner and T.P. Brent (1991) Structural and immunological comparison of indigenous human O6-methylguanine-DNA methyltransferase with that encoded by a cloned cDNA, J. Biol. Chem. 266, 1064-1070. Weber, C.A., E.P. Sazafar, S.A. Stewart and L.H. Thompson (1988) Molecular cloning and biological characterization of a human gene, ERCC2, which corrects the nucleotide excision repair defect in CHO UV5 cells, Mol. Cell. Biol., 8, 1137-1146. Zerbib, D., F. Amalric and J. Teissie (1985) Electric field mediated transfection: Isolation and characterization of a TK + subclone, Biochem. Biophys. Res. Commun., 129, 611-618.