Characterization of O6-methylguanine-DNA methyltransferase in transgenic mice introduced with the E. coli ada gene

225

Mutation Research, DNA Repair, 254 (1991) 225-230 © 1991 Elsevier Science Publishers B.V. 0921-8777/91/$03.50 ADONIS 0921877791000653 MUTDNA 06428

Characterization of O6-methylguanine-DNA methyltransferase in transgenic mice introduced with the E. coli ada gene Yoko Nakatsuru 1, Shoichi Matsukuma 2, Mutsuo Sekiguchi 3 and Takatoshi Ishikawa 1,4 Departments of t Experimental Pathology and 2 Cell Biology, Cancer Institute, Tokyo 170 (Japan), 3 Department of Biochemistry, Faculty of Medicine, Kyushu University, Fukuoka 812 (Japan) and 4 Department of Pathology, Faculty of Medicine, University of Tokyo, Tokyo 113 (Japan) (Received 10 July 1990) (Revision received 8 October 1990) (Accepted 9 October 1990)

Keywords: O6-Methylguanine-DNAmethyltransferase; Transgenic mouse; Enzyme induction; Methylnitrosourea resistance

Summary The characteristics of O6-methylguanine-DNA methyltransferase (Or-MTase) produced in transgenic mice, in which the introduced E. coli ada gene was expressed under the control of the metallothionein promoter, were investigated. Liver extracts from transgenic homozygotes showed approximately 3 times the control activity, a marked increase of up to about 8 times the non-transgenic control levels being observed 10 h after zinc treatment. Examination of the substrate specificity of the enzyme revealed that the activity in the transgenic mice is due to the introduced foreign gene. The enzyme possessed methylphosphotriester-DNA methyltransferase as well as O6-MTase, characteristic of the E. coli Ada protein. Comparison of differences in biological response between transgenic and non-transgenic mice after treatment with the alkylating carcinogen methylnitrosourea (MNU) at various doses revealed transgenic mice to be more capable of repairing Or-MTase activity, only showing signs of exhaustion at very high levels of exposure. In non-transgenic mice, on the other hand, the basal level of O6-MTase was low, and the activity was hardly detectable when the animals were treated with MNU.

Correspondence: Dr. T. Ishikawa, Department of Pathology, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku,Tokyo 113 (Japan). Abbreviations: Or-MTase, or-methylguartine-DNA methyltransferase; MNU, methylnitrosourea; Or-meG, Or-methylguanine; O4-meT, O4-methylthymine;Ch MT-I, Chinese hamster metaUothioneinI.

It is generally believed that DNA-repair mechanisms act to prevent initiation of cancer development. However, there is little direct evidence for this from in vivo animal models. While human xeroderma pigmentosum (XP) is probably the best example demonstrating a link between cancer proneness and DNA-repair deficiency (Cleaver, 1968), we do not have appropriate animal models.

226

We must create mutant animals with analogous DNA-repair defects by appropriate genetic manipulation. O6-Alkylguanine-DNA adduct is regarded as one of the most potent premutagenic lesions; it preferentially pairs with thymine rather than with cytosine, resulting in a GC to AT transition mutation (Saffhill et al., 1985). N-Nitroso compounds which yield relatively high levels of O6-alkyl guanine are potent inducers of tumors in many experimental animals (Lawley, 1984). The 0 6methylguanine-DNA adducts can be repaired by an enzyme known as O6-methylguanine-DNA methyltransferase (O6-MTase), which transfers a methyl group from the O6-methylguanine moieties of double-stranded DNA to the cysteine residues of the enzyme molecule O6-MTase itself (Lindahl, 1982). Compounds which directly damage DNA, such as nitrosoureas, are known to induce tumors preferentially in tissues which contain a low level of O6-MTase. There seems to be an inverse correlation between O6-MTase activity and tumor incidence induced by alkylating agents. The levels of O6-MTase activity vary greatly between species in rodents and man, the liver having the highest enzyme activity. Activity levels are generally several times higher in humans than in rodents (Gerson et al., 1986). We studied in detail age and strain dependence of this enzyme activity in various mouse strains (Nakatsuru et al., 1989). The enzyme activities in mice were generally lower than in rats and there were no significant interstrain differences between animals of the same age and sex. To examine whether increased levels of 0 6MTase activity can indeed decrease the susceptibility of animals to N-nitroso compounds for tumor induction, transgenic animals expressing O6-MTase genes at high levels would be useful. We have generated such animals by introducing the chimeric E. coli O6-MTase gene, ada, attached to the Chinese hamster metallothionein I (Ch MTI) gene promoter (Matsukuma et al., 1989). Recently another group also reported the production of transgenic mice expressing this chimeric ada gene with a P-enolpyruvate carboxykinase (GTP) gene promoter (Lim et al., 1990). The present report describes optimal conditions for enzyme induction in these transgenic mice and

provides evidence that the increased enzyme activity is indeed due to expression of the integrated bacterial ada gene. We further demonstrate that hepatic O6-MTase levels in the transgenic mice are highly preserved even after methylnitrosourea (MNU) treatment. Materials and methods

Animals Transgenic mice with integrated chimeric genes, composed of the E. coli ada coding sequence and Ch MT-I promoter, were generated as described previously (Matsukuma et al., 1989). The transgenic mice used in these studies were bred and maintained in our laboratory as a homozygous colony with respect to the integrated ada gene. Non-transgenic control mice (C3H/HeN) were purchased from Japan SLC laboratory (Hamamatsu-shi, Japan). They were used at the age of 7-10 weeks, housed in plastic cages at 23°C and fed on CE-2 diet (CLEA Japan, Inc., Tokyo, Japan) until sacrifice. Chemicals Poly[d(A,T)] was purchased from Pharmacia (Uppsala, Sweden) and calf thymus DNA from Sigma Chemical Co. (St. Louis, MO, U.S.A.). [3H]MNU (0.5-4.4 Ci/mmole) was purchased from New England Nuclear (Boston, MA, U.S.A.) and Amersham (Buckinghamshire, U.K.). MNU was the generous gift of Dr. M. Nakadate (National Institute of Hygienic Science, Tokyo). Purified Ada protein was prepared as described (Nakabeppu et al., 1985). Preparation of the methylated DNA substrate [3H]Methylated calf thymus DNA was prepared as described by Cathcart and Goldthwait (1981). Calf thymus DNA or poly[d(A, T)] was to give a concentration of 2 mg/ml dissolved in 20 mM ammediol buffer (pH 10) and [3H]MNU (1 mCi) was added. After incubation at 37 °C for 30 min, DNAs were precipitated with ethanol, dissolved in 50 mM Tris-HC1 (pH 7.5), 0.1 mM EDTA and dialyzed against the same buffer at 4 ° C overnight.

227 Preparation of liver extracts Mouse fiver samples (0.5 g) were homogenized in 6 volumes of buffer A (20 mM Tris-HC1, pH 8.5, 1 mM EDTA, 1 mM 2-mercaptoethanol, 5% (w/v) glycerol) at 4°C. Homogenates were centrifuged at 12,000 × g for 10 min. After centrifugation, resuspended homogenates were lysed by sonication (30 s, 3 times), recentrifuged at 22,000 × g for 30 min and then supernatants were precipitated with 80% (NH4)2SO4 at 4 ° C for 1 h. The precipitates were recovered by centrifugation at 22,000 × g for 20 min and dialyzed against buffer A. Protein concentrations were measured by the dye-binding assay (Bio Rad Laboratories, U.S.A.) using bovine serum albumin (BSA) as a standard. Measurement of 06-MTase activity The activity of O6-MTase in river extracts was measured as removal of the [3H]methyl adduct from the 0 6 position of guanine in [3H]methyl DNA (Nakatsuru et al., 1989). For the assay, 2.5 mg protein of river extract was combined with 97/~g [3H]methyl DNA (containing 10.5 pmole O6-methylguanine, O6-meG), in buffer M consisting of 66 mM Tris-HC1 (pH 8.3), 1.3 mM dithiothreitol and 0.3 mM EDTA in a total volume of 650 #1 and the mixture was incubated for 60 min at 37 ° C. The reaction was stopped by adding 0.25 ml of 1 N perchloric acid (PCA) after the addition of 0.1 ml of 3.6 mg/ml calf thymus DNA as a cartier. After standing at 4 ° C for 30 min, precipitates were collected by centrifugation at 12,000 rpm for 10 min and hydrolyzed with 0.5 ml 0.1 N PCA at 70°C for 30 min. After centrifugation, the supernatant was retained, and the precipitate subjected to the second cycle of acid hydrolysis. The combined supernatants were applied to HPLC analysis for 0 6meG. A Partisil 10 SCX column (4 × 200 mm; Whatman, U.K.) with 20 mM ammonium formate buffer, pH 4.0/7.5% methanol was run over 20 min as described (Nakatsuru et al., 1989). 06-MTase induction in transgenic mouse livers For the study of metal induction of the chimeric ada gene product, the transgenic mice were treated with ZnSO4. The animals (8 weeks old) were given a single i.p. injection of 30 mg/kg ZnSO 4, and

killed at various times after thre treatment; the rivers were then excised and the O6-MTase activities measured. Effect of M N U in 06-MTase activity in vivo Transgenic or non-transgenic mice (7 weeks old) were given a single i.p. injection of MNU (1, 3, 10, 30 and 100 mg/kg) 10 h after ZnSO 4 induction treatment. The animals (4 mice in each group) were killed 15 h later and O6-MTase activity in the liver was determined. Methyl acceptor activities of mouse liver extracts For fluorographic analysis, the liver extracts (42 /~g protein) were incubated with [3H]methyl D N A (38,000 disints/min) or [3Hlmethyl polyd(A,T)] (85,500 disints/min) at 37°C for 30 min in 60 /~l of buffer M. The reaction was terminated by heating at 90 °C for 5 min after adding 20/tl of 250 mM Tris-HC1 (pH 6.8), 20% 2-mercaptoethanol, 9.2% SDS, 40% glycerol. The samples were applied to 12.5% SDS/polyacrylamide gels and run at 100 mA for 4 h. After electrophoresis, the gels were fixed with 10% methanol, 10% acetic acid. To amplify tritium activity, the gels were soaked in ENLIGHTENING solution (New England Nuclear, U.S.A.) for 30 min. [3H]Methyl-accepted proteins were detected by fluorography. Results

Enhancement of 06-MTase activities in transgenic mouse liver by metal administration O6-MTase activities in female and male untreated transgenic mice were 2.3 and 2.7 times higher than levels in respective non-transgenic mice. After zinc administration, the levels of enzyme activity increased considerably, the highest activity being observed in females 10 h after the treatment (about 8 times higher than non-transgenic mouse liver values). The time course of change in repair activity for O6-meG in transgenic mouse livers after the metal treatment is shown in Table 1. In non-transgenic mice, endogenous 0 6MTase activity did not increase after zinc treatment, rather showing a tendency to decrease at the 10-h time point.

228

Fluorographic analyses of the methyl acceptor capacity of O6-MTase were made and a typical result is shown in Fig. 1. When extracts of liver tissues of transgenic mice were incubated with [3H]MNU-treated calf thymus D N A and then subjected to SDS/polyacrylamide gel electro° phoresis followed by fluorography, a main band corresponding to the bacterial Ada protein (39 kD) was detected. Levels of methyl acceptor capacity of transgenic mouse livers were clearly increased by metal administration. Bands corresponding to 22-kD and lower-MW proteins were also detected in the transgenic mouse samples, probably reflecting the existence of cleavage products of the Ada protein (Yoshikai et al., 1988). Neither intact nor cleaved Ada protein bands were detected in the non-transgenic samples. OO-MTase activity in transgenic mice E. coil Ada protein carries 2 distinct methyl-

transferase activities, one to transfer methyl groups from O6-methylguanine and O4-methylthymine (O4-meT) and the other to transfer methyl groups from phosphomethyltriester of methylated DNA, while most mammalian enzymes possess only the former activity (Nakabeppu et al., 1985; Koike et al., 1990). To confirm that the activity found in transgenic mouse liver is due to that expressed by the exogenous E. coli ada gene, the substrate specificity of the enzyme was examined. Calf

TABLE 1 ENHANCEMENT OF O6-MTase ACTIVITIES IN TRANSGENIC MOUSE LIVER BY METAL TREATMENT

Time after ZnSO4 treatment (h) Transgenicmice 0 4 6 8 10 20

Or-MTase activity (fmole/mg protein) Male

Female

430,498(464) 534,433(484) 523,457(490) 801,568 (685) 641,837(739) 680, 633 (657)

292, 560, 662, 526, 946, 844,

Non-transgenicmice 0 86,95,110 (97) 10 114,50, 78 (81) Figures in parentheses are the means.

548 (420) 481 (521) 693 (678) 689 (608) 1010 (978) 575 (710)

155, 95,164 (125) 75, 50, 83 (69)

I

Transgenic mice .~ ,,1

,...o~

~'7

~o/

1

39Kd .-* ~

2

3

4

J:.o

5

6

---

22 Kd "~

Fig. 1. Fluorographic resolution of comparative methyl acceptor activities of the liver extracts from transgenic and nontransgenic mice. Lane 1: E. coli Ada protein (5 #g); lane 2: transgenic mouse liver extract without metal treatment (430 fmole/mg protein, Or-MTase activity); lane 3: transgenic mouse, 6 h after ZnSO4 treatment (801 fmole/mg protein, Or-MTase activity); lane 4: transgenic mouse, 8 h after ZnSO4 treatment (837 fmole/mg protein, Or-MTase activity); lane 5: transgenic mouse, 6 h after CdCI2+ZnSO 4 treatment (865 fmole/mg protein, Or-MTase activity); lane 6: non-transgenic mouse without metal treatment (86 fmole/mg protein, 06MTase activity).

thymus D N A and poly[d(A,T)] were treated with [3H]MNU and used as substrates, the former containing both o r - m e G and methylphosphotriester, as major methyl donors for the methyltransferase reaction, while the latter contains only methylphosphotriester. The result shown in Fig. 2 clearly indicates that liver extract of transgenic mice is capable of transferring methyl groups from both methylated calf thymus D N A and methylated poly[d(A,T)]. Only a faint band was detected with the non-transgenic mouse sample when it was incubated with methylated DNA, but not with methylated poly[d(A,T)]. I n vivo 0 6 - M T a s e activities in transgenic mouse liver treated with M N U

To examine whether the Ada protein produced in transgenic mouse liver is competent for in vivo repair, the reduction of O6-MTase activity in

229 I

39Kd

"

2

3

4

5

6

activity were retained after treatment with relatively low concentrations of MNU (1, 3 and 10 mg/kg), and a substantial decrease in enzyme activity was attained only after treatment with considerably higher doses (30 and 100 mg/kg). In non-transgenic mice, the basal level of O6-MTase activity was low, and the activity was hardly detectable when 30 mg/kg or more of MNU was administered.

......

Discussion 22Kd-

Fig. 2. Methyl acceptor activities of liver extracts from different substrates. Liver extracts from 1 non-transgenic and 2 representative transgenic mice were incubated with [3H]MNUtreated calf thymus DNA (lanes 2, 4 and 6) or [3H]MNUtreated poly[d(A,T)] (lanes 1, 3 and 5). Lanes 1 and 2: transgenic liver extracts; lanes 3 and 4: non-transgenic liver extracts; lanes 5 and 6: E. coil Ada protein.

transgenic mouse liver treated with various doses of MNU was measured (Fig. 3). Mice pretreated with ZnSO 4 were injected with MNU intraperitoneally (1, 3, 10, 30 and 100 mg/kg), killed 15 h later and then the livers were excised immediately to measure O6-MTase activity. In trans~enic mice, considerably high levels of O6-MTase

i ls~ 1 b

~--_'tl

I

I

._c4~ •

~

10

:

100

MNU dose (mglkg)

Fig. 3. O~-MTase activity of liver extracts from transgenic and non-transgenic mice of both sexes after treatment with MNU at various doses. Transgenic and non-transgenic mice (4 mice in each group) pretreated with ZnSO4 (30 mg/kg, i.p.) were injected with MNU (1, 3, 10, 30 and 100 mg/kg, i.p.) and killed 15 h later at which time liver enzyme activities were measured. (A) transgenic male; (A) transgenic female; (I) non-transgenic male; (o) non-transgenic female.

Since O6-MTase activity in the mouse is relatively low compared to rats and humans, mice are particularly suitable host animals for the introduct i o n of foreign O6-MTase genes. Thus we have produced transgenic mice carrying the E. coli ada gene (Matsukuma et al., 1989). One transgenic mouse line which expressed both ada-specific mRNA and Ada protein could be propagated in homozygous state with respect to the integrated DNA and has been found to be highly reproductive for many generations. Liver extracts from transgenic homozygous mice have consistently demonstrated approximately 3 times the control activity without metal treatment. The level of enzyme activity can be increased by administration of metals, since a metal-responsive metallothionein promoter is attached to the ada. In the present detailed examination it was revealed that the enzyme activity markedly increased after treatment with zinc. The maximum activity, which was attained 10 h after treatment, reached 8 times the normal level, equivalent to that for man, thus indicating the model to be useful for simulating the human situation in showing the biological significance of this DNA-repair enzyme in in vivo carcinogenesis experiments. One assumption obviously requiting confirmation is that the demonstrated increase in enzyme activity in transgenic mice is due to expression of the integrated bacterial ada gene, but not to an enhancement of expression of the authentic mouse gene. Western blot analysis of liver extracts has revealed that a 39-kD protein, corresponding to the intact Ada protein, is produced in transgenic homozygotes (Matsukuma et al., 1989). To supply further evidence we determined, in the present study, the substrate specificity of the methyltrans-

230

ferase enzyme. It has been established that the E. coli Ada protein carries 2 distinct methyltrans-

ferase activities, one to transfer methyl groups from O6-meG and O4-meT and the other to transfer methyl groups from phosphomethyltriester, while the mammalian enzymes possess only the former activity. Fluorographic analyses revealed that the activity overproduced in transgenic mice could accept methyl groups from methylated poly[d(A,T)] as well as from methylated DNA, a clear indication that the activity has its origin in the E. coli. We also compared the in vivo biological responses between transgenic and non-transgenic mice after treatment with various doses of an alkylating carcinogen. After MNU treatment O6-MTase activity was preserved to a greater extent than in non-transgenic mice. This observation suggests that bacterial O6-MTase activity was effective for repairing O6-meG generated in the DNA of transgenic mice after exposure to relatively high levels of MNU. It can be inferred that O6-MTase can protect cellular DNA from continuous low-dose exposure to environmental alkylating carcinogens. Acknowledgements The technical assistance of Ms. Kaori Nose and Ms. Junko Sakurai is appreciated. Acknowledged also are Mr. Hiroaki Mitani and Mr. Hironori Murayama for animal care. This work was supported by grants-in-aid for cancer research from the Ministry of Education and Culture, the Ministry of Health and Welfare of Japan, the Uehara Memorial Foundation and the Smoking Research Foundation.

References

Cathcart, R., and D.A. Goldthwait (1981) Enzymatic excision of 3-methyladenine and 7-methylgnanine by a rat liver nuclear fraction, Biochemistry, 20, 273-280. Cleaver, J.E. (1968) Defective repair replication of DNA in xeroderma pigmentosum, Nature (London), 218, 652-656. Gerson, S.L., J.E. Trey, K. Miller and N. Berger (1986) Comparison of O6-alkylgnanine DNA alkyltransferase activity based on cellular DNA content in human, rat and mouse tissues, Carcinogenesis, 7, 745-749. Koike, G., H. Maki, H. Takeya, H. Hayakawa and M. Sekiguchi (1990) Purification, structure and biological properties of human O6-methylguanine-DNA methyltransferase, J. Biol. Chem., 265, 14754-14762. Lawley, P.D. (1984) Carcinogenesis by alkylating agents, in: C.E. Searle (Ed.), Chemical Carcinogens, 2nd edn., ACS Monograph, Vol. 182, American Chemical Society, Washington, DC, pp. 325-484. Lim, I.K., L.L. Dumenco, J. Yun, C. Donovan, B. Warman, N. Gorodetzkaya, T.E. Wagner, D.W. Clapp, R.W. Hanson and S.L. Gerson (1990) High level, regulated expression of the chimeric P-enolpyruvate carboxykinase (GTP)-bacterial O6-alkylguanine-DNA alkyltransferase (ada) gene in transgenic mice, Cancer Res., 50, 1701-1708. Lindahl, T. (1982) DNA repair enzymes, Annu. Rev. Biochem.. 51, 61-87. Matsukuma, S., Y. Nakatsuru, K. Nakagawa, T. Utakoji, H. Sugano, H. Kataoka, M. Sekiguchi and T. Ishikawa (1989) Enhanced O6-methylguanine-DNA methyltransferase activity in transgenic mice containing an integrated E. coli ada repair gene, Mutation Res., 218, 197-206. Nakabeppu, Y., H. Kondo, S. Kawabata, S. Iwanaga and M. Sekiguchi (1985) Purification and structure of the intact Ada regulatory protein of Escherichia coli K12, O6-methylguanine-DNA methyltransferase, J. Biol. Chem., 260, 7281-7288. Nakatsuru, Y., K. Aoki and T. Ishikawa (1989) Age and strain dependence of O6-methylguanine-DNA methyltransferase activity in mice, Mutation Res., 219, 51-56. Saffhill, R., G.P. Margison and P.J. O'Connor (1985) Mechanisms of carcinogenesis induced by alkylating agents, Biochim. Biophys. Acta, 823, 111-145. Yoshikai, T., Y. Nakabeppu and M. Sekiguchi (1988) Proteolytic cleavage of Ada protein that carries methyltransferase and transcriptional regulator activities, J. Biol. Chem., 263, 19174-19180.