......p» ... 1... v.1.uuumuvwmh-I View ... . 0: . Al. 1 CV 1 -. ... We... 3..., .. fl - ...: . . .....MWI ...? . . 1.4%? .m. . . . .. ...-MW. ,. .mmwwdu...wn . . . .... : . . . .... .... Mu. .. ... .. .. , .... . L . .1 l. . u .. I - \.n1.\..ul.nh.~11o ¢ Ln; lokUthmwwf‘u‘ .ner In? a . mwummfi . .. ...... .. Mm... 1 V.. J... ..1. ... .T‘ v Mg ...-1». .IWv - .umul. .4 3.... 1.. ... .. .. 1.. .WJINLWWMWWPHUfi fimnmmlwmns .. .. .1... ...... $4..”va ...... .ua.fl...-...»“«..zll. . .. . .. ... .. . hug“! hrlflhnw M.» . . .V€$.~ ;W 49!! K . . . -wwxw..NHA§ - .. .IFL..." I \v t. 1...... ... .3... at... v. I “IN .0! . 1. ...uk! _ .....er . ... .. . r ; ... .. . .. . 1. . .. ... . . 1 . ..r. . . . no . .u . .... z . . u .. . Us . ....urf ...-.-» - vanmytpat ..n. ...m. :1 .-. .u Wmflnwxw . ...... fl...» ...vmuHKWvar . . . , .. .. -..»...WOL . .. $5.41... $4...- . ... %®.«.. I ... . . . .. n..tfi.u- . .....rnxmmfluk ...vl .v.hl- .Mqu “h“. VIN -01.... ..K- I ..I. t. V..-WWW”! v In. .% \hvvrhubt \ l. I .flu.l. IMQI n ibflgr .Ifl. .0 1H. . ,nniiamfl..fl. {...}. . . . ...... u l. up- . (:de i. .. ..me . \1 uh? “‘7 rt “ti-VF. . .v ....cV. . . IL. '1... {llmmv ckrt ... mg Armhwkl M... - .. M41. .... t)? . . I {..WUB. +0.1. 1... g... . . . \lr‘“-. pk... r1). .. 1.“. ... HHMWWMWVW (.... ... rpmmmmn. 4%.»... I. v raging. :70"! . . . In. 1!- .1 .H. lllll .. . .. | .... . - ‘11: 1 a. .. l . (Iain. “NW4“..M... ARI. )l4.l\l.\ . ..l blurhb n.01fl.nfilv.rxknn.~fflnu1uafi¢h . I ‘4 wwlflflkhflillhflrl‘flut‘lwlv‘ . (Wmnnmudhql\§ nub . ... . .. . .. i... .L. 0‘2“. “I‘. "an aw.‘q.l¥t a MW...” MmmnmuWWMbumwfldu ..fllwW‘Uf‘. 4. -.- I... ... 0“:me 44......ch ...“...WwflwmmfiWMIWnuwwwn..- F- o v m unflqnfin- . . “train“ {If x 5 - n. .3.“ 2% -i.)lif;h\h.vf..fl.t~wfl.\ .5. > I H. "A.” .l ’ .1. .(it’l‘hll ll . M. ....H 46m@ Ln. .... lubflhxhuxruuxwg. I :nJuL) .. . . -. ...o).l........m.Mn.l.....€WIl J.WI.IU.W..I. N—N=O /’ H [CH3—N=N+]+ OH— CH3++ N2 DAUX CH3-DNA 25 Figure 3. Sites of alkylation of DNA. Cancer Invest., 1, 223—231. From Pegg (1984), BASE W ADENINE: (NI GUANINE= I 3 m2 CYTOSINE= 1% HO 1' *2 PHOSPHATE: _o-§§-o_ OH Figure 3 POSITIONS ALKYLATED 1-, 3-,7- 317-, 06. 3301’— 3-, 03,01 27 Table 1. Relative proportions of alkylation products in DNAa Percent of Total Alkylation DMNb DEN MMS EMS DMS Alkylation MNU ENU (i=0 . 83) (2‘0 . 67) (i=0 . 86) site (30.42)C (_s_=o.26) DMH Adenine N—1 O07 003 102 107 1 N-3 8.0 4.0 11. 4.2 16. N"? 105 004 109 109 20 Cytosine d 0 0.1 3. ND 0.3 ND Guanine N-B 008 006 007 003 10 N57 680 120 830 580 790 0 7.5 8. 0.3 2. 0.2 Thymine N23 0.3 0.8 ND ND ND 04 0.1 7. ND ND ND 0 0.7 2. ND ND ND Alkyl- phosphates 12. 53. 1. 12. 1 aAdapted from Pegg (1983), In: Reviews in Biochemical Toxicology, (Hodgson, Bend, Philpot, eds.), Elsevier Biomedical, pp. 83—133. bDMN, dimethylnitrosamine; MNU, N-methyl-N-nitrosourea; DMH, 1,2- dimethylhydrazine; DEN, diethylnitrosamine; ENU, N-ethyl—N-nitrosourea; MMS, methylmethanesulfonate; EMS, ethylmethanesulfonate; DMS, dimethyl- sulfate. CExamples of_3 values from Lawley (1974), Mutation Res.,_g3, 283. dND, Not detected. [—12 28 proportion of alkylation products formed in DNA depends on the alkylating agent. An inverse correlation exists between the Swain—Scott constant, s, for a particular alkylating agent and the proportion of oxygen adducts formed. Agents that have lower E values form a higher percentage of oxygen adducts. MNNG has a low Swain—Scott constant of 0.42, and 6 6 the ratio of O -methylguanine (O —MeG) to 7—methylguanine (N7—MeG) in DNA is approximately 0.1. The s value for DMS is 0.86 and the 06-MeG to N7-MeG ratio is approximately Ali-fn’ :‘0 IR" 0.003 (Lawley, 1974). In addition, methylating agents have higher g values than the corresponding ethylating agents. 2. 06-MeG Involved in the Induction of Tumors in Laboratory Animals Two decades ago it was believed that 7—alkylguanine was the adduct responsible for the production of liver tumors in rats after exposure to DMN (Magee and Farber, 1962). This conclusion was reached because the adduct 7-alkylguanine was persistent and present in the greatest amount. The same conclusion was reached using the alkylating agents DMS and MNU (Singer 1975). This hypothesis began to be discredited about ten years ago when researchers found that amount or presence of 7—alkylguanine did not correlate with the sites of tumor production. Ethyl methanesulfonate (EMS) was much more carcinogenic in rats than methyl methanesulfonate (MMS) even though 7—ethylguanine was present in smaller amounts than 7-methylguanine (Swann and Magee, 1971). The first 29 indication that methylating agents could react at sites other than nitrogen was first detected by Loveless (1969) and later Lawley and Thatcher (1970) found 06-methylguanine to be present in MNNG-treated DNA, but not DMS-treated DNA. Because experimental findings showed that MNNG is a powerful carcinogen whereas DMS is a weak carcinogen, it was postulated that organ specific carcinogenesis correlated 6-alkylguanine. 6 with the presence of 0 Experimental studies relating O —alkylguanine to carcinogenesis have been provided by measurements of abundance of this lesion in various organs in experimental 6-MeG in the animals. Alkylating agents that produce little 0 DNA of a particular organ are poor inducers of tumors in that organ. Goth and Rajewsky (1974) demonstrated that in young rats exposed to N—ethyl-N—nitrosourea, O6-ethylguanine persisted in the target organ (brain). However, in the nontarget organ (liver), this lesion was repaired. A similar finding by Margison and Kleihues (1975) showed the preferential accumulation of 06-methylguanine in rat brain DNA following exposure to N-methyl-N—nitrosourea. Frei gt il° (1978) measured the extent of alkylation of various DNA sites after treatment of mice with radiolabelled ENU and MNU and measured the extent of induction of thymic lymphomas. At doses which yielded an equal carcinogenic response, the only lesion which showed an equal extent of alkylation was 6 0 -alkylguanine. This correlation between the persistence of O6-alkylguanine and carcinogenesis has also been 30 confirmed by other studies dealing with other alkylating agents as well as other target tissues (Pegg, 1977; Montesano, 1981). The findings that O6 —alkylguanine correlates well with carcinogenesis is supported by the work of Loveless (1969) and also Gerchman and Ludlum (1982) who found that this altered base can mispair during replication and thereby lead to mutations, whereas the abundant lesion, 7-methylguanine does not cause misincorporation (see Figure 4). Although the causal connection between persistence of 06-alkylguanine in target organs and the subsequent induction of tumors is supported by the above mentioned experiments, there are data that contradict this hypothesis. A controversy exists to date as to whether or not o6-alkylguanine is necessary for initiation of carcinogenesis by alkylating agents (Pegg, 1984; Singer, 1984). Several experiments have indicated that 6 0 -alkylguanine persists in the brain of young hamsters, gerbils, Xenopus laevus, and many strains of mice; even though the brain is not the target organ (Buechler gt gt., 1977; Rogers and Pegg, 1977; Hodgson gt gt., 1980; Kleihues gt 31., 1980; Nicoll gt gt., 1975). Another example is the work of Likhachev and coworkers (1983) where the persistence of 06-alkylguanine was highest in the nontarget organ, the brain, after exposure to MNU or ENU. However, this was not a well conducted experiment since 06-MeG was not measured in the target organ. mm at w .J . 1, .. aw... . 1. var .l I 31 Figure 4. 06—MeG mispairing. Upper diagram, cytosinezguanine; center, cytosine:06-methylguanine; lower, thyminezos-methylguanine. R, represents the position of the deoxyribose moeity of the sugar—phosphate backbone. From Margison and O'Connor (1979), In: Chemical Carcinogens and DNA, Grover, P.L. (ed.), CRC Press, Boca Raton, FL, pp. 112—149. ‘0 Guad- Figure 4 33 Other factors must be considered in interpreting the O6-a1kylguanine experiments. These include cell proliferation in an organ, the role of other adducts (e.g., 2 2 O and O4-thymine and O -cytosine) in carcinogenesis and the 6—alkylguanine in the variability of the persistence of 0 different cell types of a particular organ. The role of cell division is important in interpreting the experiments that implicate 06 -alky1guanine in carcinogenesis. Methylated bases can lead to tumors only if DNA synthesis occurs while the lesion is present. For example, Kleihues and coworkers (1973) found that after exposure to MNU, all the DNA of the cells in the brain is methylated. However, only the glial cells which divide form tumors. Furthermore, in adult rats, methylating agents do not form tumors after a single dose, but do produce tumors in neonatal rats where the liver cells are dividing or in cases of partial hepatectomy (Pegg, 1977; Pegg, 1983; Craddock, 1971). Singer's laboratory (1976) demonstrated that ENU reacts readily with pyrimidines in DNA and that the O-alkylpyrimidines are capable of mispairing. Singer and coworkers (1981) showed that in young rats treated with ENU, the target organ (brain) contained O-ethylpyrimidines and these lesions are more persistent than O6—ethylguanine. Swenberg and colleagues (1984) demonstrated when one is considering repair in the organ as a whole, repair in the cell types that make up that organ must be considered. They separated hepatocytes from nonparenchymal cells and found 34 that repair of DMH-induced damage in the nonparenchymal cells was poor. The tumors which developed were of nonparenchymal cell origin. Thus, if repair is measured in the whole organ, the result will indicate repair is occurring even though one of the cell types is not repairing at all. These experiments taken as a whole indicate that 06-methylguanine is necessary to initiate tumors but other factors determine whether or not the neoplastic process occurs (Pegg, 1984; Singer, 1984). 3. Is 06—MeG the Mutagenic Lesion in Mammalian Cells in Culture ? Warren gt gt. (1979) and Goth-Goldstein (1980) found that after treatment with radiolabelled alkylating agents, Chinese hamster lung (V79) and Chinese hamster ovary (CHO) cell lines were able to excise 3—alkyladenine and 7-alkylguanine residues, but ggt 06-alkylguanine residues. Goth-Goldstein (1980) suggested that since O6—alkylguanine can mispair during DNA replication, a deficiency in the removal of this lesion predicts that the induced mutation frequency will be higher in cells that cannot remove this lesion than in cells that can remove it. However, she only measured the number of adducts remaining after nearly 20 hours and the mutations experiments to test her speculation were not carried out. 35 Since V79 cells do not excise 06-MeG from DNA, they are ideal for experiments designed to establish the role of 06-MeG in the mutagenicity and cytotoxicity of methylating agents. Newbold and coworkers (1980) demonstrated that the mutagenicity of methylating agents in V79 cells is 6-MeG adduct. The major lesions 7-MeG, 6 associated with the O 3-methy1adenine (3-MeA) and O -MeG were measured and also the induced mutation frequency after exposure to the weak carcinogens DMS and MMS and the potent carcinogen MNU. The mutagenicity, but not the cytotoxicity, of each agent reflected its carcinogenicity and differences in mutation frequency were paralleled by differences in the amount of 06-MeG. A problem with this study is that only some of the types of DNA lesions were quantitated. Another type of lesion might also correlate with mutations. Newbold‘s work is supported by Suter gt gt. (1980). In V79 cells the levels of DNA alkylation were measured for the adducts: 7-MeG, 3—MeA and 06-MeG. Also measured was the induced mutation frequency after exposure to MNU or DMS. At 6-alky1ation approximately equal numbers of equal levels of O mutants were induced. The survival data did not correlate with 06-MeG, but from their data, it appears that a correlation with 3-MeA could be made. Heflich gt gt. (1982) exposed Chinese hamster ovary cells to tritiated ENU or EMS and measured the induction of TG resistant cells and quantitated the DNA adducts formed. Statistical analysis indicated that mutation induction was 36 highly correlated with alkylation of guanine at the 06 position. Sister chromatid exchanges (SCE) were highly correlated with the lesions 3—methyladenine and 3—methylguanine. Work from the same laboratory by Morris gt gt. (1982) reported that the biological responses of cell killing and SCE are highly correlated. Their interpretation from the two papers is that the group of adducts resulting in mutations at the HPRT locus is not the same set of lesions responsible for inducing cell killing. In contrast, Fox and Brennard (1980) have shown evidence for the involvement of lesions other than 06-alkylguanine in mammalian cell mutagenesis. Using a reverse mutation assay (that is, reverting TG resistant cells from HAT medium [hypoxanthine, aminopterin, thymine] sensitivity to HAT medium resistance) and measuring levels of alkylation, they found that EMS, a weak carcinogen, and MNU, a potent carcinogen, both were effective mutagens in the reverse direction. Since EMS produces little of the 06-alkylation product and MNU produces more of the product (per total 6 amount bound), the amount of O -alkylation does not correlate with the mutational response. The dose response of methylating agents at the HPRT locus in Chinese hamster cells (Couch and Hsie, 1978; Couch _t _t., 1978; Thielman _t _t., 1979) was examined and no deviation from linearity was reported. Also the previously mentioned studies by Newbold t l. (1980) and Suter gt gt. (1980) suggest that the increase in mutation frequency as a 37 function of dose is linear. A decreased response or threshold at low dose of a mutagenic agent may be connected with changes in the mechanism of repair of the lesions at different dose levels. A threshold implies that the repair is error-free at low doses and that this repair process is saturated at higher doses. However, the above investigators did not pay specific attention to the mutational response at low doses. They measured the response at higher doses and extrapolated to lower doses. Jenssen and Ramel (1980) investigated the dose response curves for the induction of TG resistant mutants by low doses of the monofunctional alkylating agents EMS, ENU, MMS and MNU in V79 cells. They found that at low doses, the ethylating agents give a linear response, whereas the methylating agents MMS and MNU demonstrate a threshold in the induced mutation response to 6-thioguanine resistance. The threshold suggests a repair mechanism at low doses for methylating agents. This implies that at low doses for ethylating agents the repair mechanism is not as efficient as for methylating agents and/or another repair system is involved in the repair of ethyl adducts. 4. Responses of Human Cells to Alkylating Agents The persistence of the lesion 06—MeG has also been studied in human cells. Goth-Goldstein (1977) treated SV40-transformed XP fibroblasts (defective in UV excision repair, complementation group A) and normal SV40-transformed 38 3'2 fibroblasts with MNU and ENU in order to measure the '3 alkylation products O6—alkylguanine and 7—alkylguanine. She found that normal cells remove 06—alkylguanine, whereas in XP cells the O6-alkylguanine is removed to a lesser extent. These results were confirmed by Altamirano-Dimas gt gt. (1979) who showed that lymphoblastoid cells from a xeroderma patient (complementation group C) are unable to remove 06 —MeG residues in DNA after treatment of the cells with MNNG. However, XP cells are able to remove 3—methyl adenine adducts. They also found that UV excision defective mutants 5; of E. ggtt, uvr A and uvr B, are able to excise MNNG-induced 06-methylguanine adducts, indicating that the repair mechanism for these adducts is not the same as the mechanism for UV induced lesions. Therefore it was surprising that XP cells were also unable to remove 06—MeG. Sklar and Strauss (1981) examined the kinetics of removal of the methyl group from 06—MeG in normal and XP—derived lymphoblastoid lines. The removal of the methyl group from 06-MeG occurred rapidly with a half life of less than one hour. They distinguished between two cell types: Mex+ (methyl excision proficient), cells that can remove 06—MeG and Mex— (methyl excision deficient), cells which are unable to remove the adduct after a low dose of MNNG. The mex+ and mex— categories are similar to the mer+ and mer— categories of Day _t _t. (1979; 1980) who reported strains which differ in their ability to reactivate MNNG—treated adenoviruses. Mer_ cells can support the growth of 39 adenovirus 5 normally, however, when the virus is treated with MNNG, the mer- cells cannot support its growth. The assay consists of inactivating the adenovirus with MNNG and subsequently assaying for plaque-forming ability, using monolayers of human fibroblasts. Day gt gt. ( 1980) reported that mer_ strains: 1) show a greater sensitivity to MNNG induced cytotoxicity, 2) are deficient in removing 6-MeG from their DNA after MNNG treatment, and 3) have an 0 increased susceptibility to SCE after MNNG treatment. Medcalf and Lawley (1981) showed that at non-toxic doses of MNU, normal human fibroblasts exhibit rapid removal of 06-methylguanine (tl/2 = 1 hour), but at higher doses, removal is slow (tl/2 > 24 hours). They predicted a non-linear dose response for mutation induction in human fibroblasts. A marked non—linear dose response was reported by Penman gt gt. (1979) for induction of mutations in human lymphoblasts by the methylating agent N-methyl-N—nitrosourethane. However, these authors did not prove that the threshold was due to the more efficient removal of 06-MeG at low doses. Such a threshold in the induced mutational response of normal human fibroblasts to MNNG was also observed by Jacobs and DeMars (1978). In contrast to Goth-Goldstein (1977) and Bodell gt gt. (1979), who observed that following treatment with ENU, SV40 transformed XP fibroblasts removed O6—alkylguanine from DNA at a slower rate than SV40-transformed normal human fibroblasts, Teo t l. (1983) working with nontransformed 40 fibroblasts, found similar rates of elimination of the O6 -ethylguanine adduct in two normal and two XP cell lines. The difference between the experimental groups may be due to the SV40 transformation process. It has been reported by Day gt gt. (1980), Sklar and Strauss (1981) and Shiloh gt al. (1983) that transformation of human cells by SV40 or Epstein-Barr virus may be accompanied by a loss of the 6 ability of the transformed cells to remove O -methylguanine from their DNA. D. Role of Methyltransferase in the DNA Repair of Alkylation Damage . l. Methyltransferase in Bacterial Cells At the same time that observations dealing with tumor and mutation induction were being published, studies on the alkylation repair mechanism in t. ggtt bacteria were being performed. These eventually yielded the following insights into the properties of the alkylation repair mechanism in g. ggtt. Bacterial cells are capable of repairing 06—MeG by a demethylating mechanism which involves a suicide enzyme. The model emerged that the 06—methylguanine lesion was repaired by the direct transfer of methyl groups from 06-methylguanine to protein cysteine residues on a protein acceptor molecule (Olsson and Lindahl, 1980). Karran and coworkers (1979) showed that when DNA which contained the radioactively labelled 06-methylguanine (methyl group 41 labelled) substrate was incubated with extracts from MNNG- treated g. ggtt, the radioactivity was removed from the exogenous DNA. Extracts from uninduced cells (not treated with MNNG) did not remove the radioactive lesions from DNA. However, no free radioactive 06—methylguanine could be found. Olsson and Lindahl (1980) concluded that following repair, the radioactive material was probably transferred to a protein. This conclusion was based on the discovery of radioactive S-methylcysteine. Further evidence for the model that the 06-methylguanine lesion is repaired by the direct transfer of the methyl groups to protein cysteine residues was obtained by Foote and coworkers (1980). By using a substrate radioactively labelled in the purine ring rather than the methyl group, they were able to demonstrate that the model described above regenerated guanine from the mutagenic lesion. Lindahl gt gt., (1982) and Robins and Cairns (1979) showed that the repair of 06-methy1guanine was limited by the amount of acceptor protein in the reaction, suggesting that repair protein was being consumed in the reaction. Purification of §° ggtt 06-methylguanine-DNA methyltransferase to homogeneity by Demple and coworkers (1982) allowed these workers to demonstrate conclusively that the molecule that removes the methyl group becomes inactivated in this process. 42 2. The Adaptive Response in E. coli An important aspect of this repair mechanism is that the 6—MeG is limited by the amount of methyl removed from 0 amount of acceptor protein initially present (suicide enzyme). However, in g. ggtt the demethylating mechanism for alkylation damage is inducible. This inducibility is termed the adaptive response. This response in E. ggtt was first demonstrated by Samson and Cairns (1977). They added a subtoxic dose of MNNG to exponentially growing cultures of g. ggtt. At various times thereafter, samples from the culture were treated with a higher dose of MNNG and the resulting number of mutants and extent of killing were measured. The bacteria became increasingly difficult both to kill and to mutate if they were pretreated with a subtoxic dose of MNNG before the larger dose of MNNG. The resistance to MNNG mutagenesis was prevented by the addition of chloramphenicol, suggesting that protein synthesis is required for attaining resistance to MNNG. Cairns and coworkers (1981) also showed that the adaptive response was specific for alkylating agents. Resistance to mutation induction was highly efficient, but exhibited saturable kinetics, that is, adapted bacteria could be challenged with low doses of a specific methylating agent without significant induction of mutations, but at higher doses of the alkylating agent, the frequency of mutants induced was the same as in unadapted cells. A good 43 correlation was found between the presence of the DNA lesion O6 -methylguanine and the frequency of mutations. Adaptation to killing and to induction of mutations have been found to occur via different pathways since mutants of g. ggtt_have been isolated which possess the adaptive response to mutation induction, but not to killing and vice versa (Evensen and Seeberg, 1982). The adaptive response in t. ggtt consists in part in the induction of methyltransferase which transfers the methyl group from 06-MeG in DNA to a protein acceptor. In addition to the MT, the adaptive response involves the induction of a DNA glycosylase which catalyzes the removal of N-methylated purines such as 3—methy1adenine (Karran t 1., 1982b; Evensen and Seeberg, 1982). 3. Methyltransferase in Mammalian Cells is not Fully Characterized The loss of 06-methylguanine from DNA has been demonstrated with cell extracts from rodent liver, kidney, and other mammalian tissues, and a variety of human tissues and cells (Pegg _t _t., 1983; Bogden gt gt., 1981; Lemaitre gt gt., 1982; Pegg and Wiest, 1983; Pegg, 1978; Pegg, 1983b; Pegg gt gt., 1982; Waldstein gt gt., 1982a; Waldstein gt 1., 1982b; Waldstein t 1., 1982c). The removal of O6 —methy1guanine in mammalian cells is similar to that carried out by the E. coli protein, however, the mammalian system has not been fully characterized. Human and rodent 44 cells contain an activity with a molecular weight of 20,000 6 which demethylates O -methylguanine (Bogden gt gt., 1981; Mehta _t _t., 1981). The reaction involves the transfer of the methyl group from 06-methylguanine to a cysteine residue of a protein acceptor molecule. For both the g. ggtt and liver protein, direct evidence for this mechanism is provided by the demonstration that stoichiometric loss of O6 —methylguanine, generation of S—methylcysteine and production of guanine in DNA takes place (Lindahl, 1982; Pegg gt gt., 1982; Pegg gt gt., 1983; Demple gt gt., 1982; Foote gt gt., 1980; Mitra gt gt., 1982). Demple and coworkers (1982) have purified the bacterial protein to homogeneity and have shown that the methyl acceptor sites reside in the same protein as that which catalyzes the transfer. This may also be true for the mammalian protein (Foote gt gt., 1980). The mammalian enzyme has not been purified to homogeneity as of this writing. As in the g. ggtt suicide enzyme reaction, the protein acceptor becomes inactivated. This is indicated by the stoichiometry between 6 S-alkylcysteine production and O —alkylguanine removal by mammalian cell extracts (Pegg gt gt., 1983; Bogden gt gt., 1981; Lemaitre t 1., 1982; Pegg and Wiest, 1983; Pegg, 1983b; Waldstein t 1., 1982c; Mehta, 1981; Craddock, 1982). 45 4. In Vitro Properties of MT Are in Good Agreement with in Vivo Studies The tg vitro (studies with cell free extracts) properties of methyltransferase are in good agreement with the persistence of O6 -MeG in DNA tg tttg. The initial rapid rates of removal tg tttg are similar to those seen tg vitro with substrates alkylated to the same extent (Scicchitano and Pegg, 1982). Also tg XiXEI the repair of 06-methylguanine at high doses becomes saturated which agrees with the tg vitro characteristics of methyltransferase. Since the acceptor site is not readily regenerated, the removal reaction proceeds rapidly until all the acceptor sites have been consumed. The number of lesions that can be repaired rapidly is determined by the initial number of protein molecules (Lindahl, 1982; Cairns gt gt., 1981; Schendel, 1981). This capacity to repair is both species and organ specific. Table 2 shows the content of methyltransferase in different cells and tissues. Rat liver has a high amount of methyltransferase and can remove 06—methylguanine tg gttg, whereas rat brain has a low amount of MT and is unable to remove this lesion. The content of methyltransferase in cells from different tissues correlates well with the carcinogenesis data from methylating agents described in a previous section. The fact that hamster hepatocytes have a lower amount of methyltransferase than rat hepatocytes correlates well with 46 Table 2. Content of MT in different cells and tissuesa Source Molecules of MT per Cell Human liver 750,000 brain 30,000 lymphocytes l4,000-140,000 HeLa cells 100,000 Rat liver hepatocytes 60,000 liver nonparenchymal cells 12,000 brain 1,500 kidney 12,000 Hamster liver 28,000 aAdapted from Pegg (1984), Cancer Investigation t, 223. 47 the observation that single doses of DMN will cause tumors in hamsters, but not in rats (Montesano, 1981; O'Connor gt _t., 1982). Rat hepatocytes in culture also have a low rate of DNA synthesis and are found to be more resistant than rat brain or kidney cells to MNU (Pegg, 1977; Kleihues gt gt., 1982; Lindahl, 1982). tg XiXQ studies have also indicated that methyltransferase is inducible to a small extent. The activity of methyltransferase was increased when rats were chronically exposed to DMN (Montesano t 1., 1980; Pegg gt .flkfimmfii gt., 1981; Pegg and Perry, 1981b). Similar increases were produced by the alkylating agents DMH and diethylnitrosamine (DEN) (Swenberg _t _t., 1982; Pegg and Perry, 1981b). A moderate increase in methyltransferase (7—fold) has been seen in rats following partial hepatectomy or treatment with hepatotoxins (Pegg _t _t., 1981; Pegg and Perry, 1981b). However, in adapted E. coli a 100—fold increase can be seen. 5. The Adaptive Response in Mammalian Cells An inducible process for 06—methylguanine-DNA methyltransferase has not been definitely observed in mammalian cells, although some suggestive observations have been published. Samson and Schwartz (1980) demonstrated that chronic treatment of CHO cells and an SV40—transformed human skin fibroblast line, (using non—toxic levels of MNNG) made the cells resistant to alkylation—induced cell killing and 48 sister chromatid exchanges. CHO cells were found to have increased resistance to MNNG—induced mutation after adaptation (Samson and Schwartz, 1983; Schwartz and Samson, 6-MeG from DNA 1983). However, CHO cells failed to remove O (Goth—Goldstein, 1980) and lacked either constitutive or inducible MT activity (Foote and Mitra, 1984) and therefore the mechanism for this response in CHO cells may differ from that in g. ggtt. In Chinese hamster V79 cells, Kaina (1982) reported a moderate resistance to cell killing, a reduction 1 in sensitivity to mutation, and a reduction in the induction :q of chromosome aberrations after a single pretreatment of the cells with low doses of MNNG or MNU before challenge with a higher dose. Durrant and coworkers (1981) reported an increased survival in Chinese hamster V79 cells following a challenge dose of MNU after previous subtoxic doses of MNU. However, a reduction in induced mutation frequency was not seen. Rat epithelial cell lines exposed to subtoxic doses of MNNG showed a reduction in dose dependence of cell killing when challenged with increasing doses of MNU, as demonstrated by Montesano _t gt. (1982). Recently, rat hepatoma cells pretreated with alkylating agents showed increased resistance to alkylation—induced cell killing and mutagenesis and an increased repair capacity for 06—MeG (Laval and Laval, 1984). In contrast, Karran gt gt. (1982) found that the adaptive response to the cytotoxic effects of MNU is absent in normal human fibroblasts. This may reflect the fact that human cells constitutively have high levels of 49 methyltransferase (see Table 2). In addition to the evidence for an adaptive response in human cells (Samson and Schwartz, 1980), Waldstein gt gt. (1982) found that treatment of mammalian cells with multiple low doses of MNNG resulted in the increase of methyltransferase. However, these results should be viewed with caution since other laboratories have not been able to reproduce these results with the same cell lines (Yarosh gt gt., 1984; Myrnes t 1., 1982; Foote and Mitra, 1984). E. Studies Dealing with the Repair of Alkylated Bases Other than 06—MeG Other enzymes which are involved in the repair of alkylation damage are glycosylases which are able to correct the lesions 3—methyladenine and 7—methylguanine. Glycosylases split the bond between the N-9 position of a purine and the deoxyribose, yielding an apurinic site in DNA. The apurinic sites are then repaired by an excision repair system. 3-Alky1adenine is repaired in DNA by a glycosylase. This has been demonstrated in extracts from human lymphoblasts (Brent, 1979; Singer and Brent, 1981) and placenta (Gallager and Brent, 1982) and in extracts from rodent liver and other tissues (Margison and Pegg, 1981; Cathcart and Goldthwait, 1981; Craddock _t _t., 1982; Pegg, 1982). Whether the glycosylase is specific for 3—alkyladenine has not been clear, since the enzyme has not been purified to homogeneity . -«~.LJ'~L'::'I’.#' v 50 from mammalian extracts. In bacteria 3—methyladenine-DNA glycosylase is important in cell survival. This has been demonstrated through the use of tgg mutants which are deficient in the 3—methyladenine-DNA glycosylase. The tgg mutants are more sensitive to the killing effect induced by alkylating agents (Mamet-Bratley _t _t., 1982; Lindahl, 1982). Enzymatic removal of 7-alkylguanine from DNA has been demonstrated using extracts from: 1) human lymphoblasts (Singer and Brent, 1981), 2) rat liver (Margison and Pegg, 1981; Cathcart and Goldthwait, 1981), and 3) hamster liver (Cathcart and Goldthwait, 1981). In these studies the removal rate was enhanced compared to the rate of spontaneous hydrolysis. 7—methylguanine-DNA glycosylase has not been fully characterized and its substrate specificity is not certain. The removal of 7—methylguanine from DNA is probably not important since in bacteria this adduct does not appear to contribute to mutagenesis and does not affect survival. Besides 06-a1kylguanine, the alkylation products 02— and 2 o4-a1kylthymine and O —alkylcytosine are also possible miscoding lesions (Singer t 1., 1978). The removal of these adducts was studied by Bodell _t gt., (1979) in human fibroblasts after exposure to ENU. The human fibroblasts used were an XP line and a normal line. They found that both lines had similar ability to remove the alkylpyrimidine products. However, the XP line did not remove 51 06-alkylguanine as well as the normal line and this confirmed the work of Goth—Goldstein (1977). III. MATERIALS AND METHODS A. Chemicals Chemical mutagens were obtained from the following sources: MNNG, Pfaltz and Bauer (Flushing, NY); ENU, MNU and 4-NQO, NCI Carcinogenesis Program. All were stored as solids at -200 C. All chemical mutagens except 4—NQO were dissolved in 100% DMSO (Burdick and Jackson, Muskegon, MI) prior to treatment of the cell cultures. 4-NQO was dissolved in 100% ethanol (Burdick and Jackson, Muskegon, MI). For mutagenicity experiments, 6-thioguanine was obtained from Sigma Chemical Co. (St. Louis, MO) and a 80 mM solution was prepared by dissolving 0.164 g in 2 m1 of 1 N NaOH and 98 m1 of distilled water. HEPES was also obtained from the Sigma Chemical Co. (St. Louis, MO). In experiments to deplete MT by exposure to exogenous 06-MeG, 06-MeG was a gift from Dr. A.E. Pegg, Pennsylvania State University (Hershey, PA) and was synthesized by the 6-MeG was method of Balsiger and Montgomery (1960). 0 dissolved by sonication into serum free Eagle's medium. Formaldehyde was obtained from Mallinckcrodt (Detroit, MI). Tritiated thymidine (Amersham, Arlington Heights, IL) and aphidicolin (Sigma, St. Louis, M0) were used in cell 52 53 synchrony experiments. A stock solution of aphidicolin was prepared by dissolving in 100% DMSO and storing at —200 C. Crystal violet and methylene blue used for colony staining were obtained from Mallinckcrodt (Detroit, MI). Bisbenzimide # 33258 Fluorochrome, used for determining mycoplasma contamination of cells, was obtained from Calbiochem-Behring (San Diego, CA). B. Cell Lines The characteristics and sources of the cell lines used in this study are listed in Table 3. Fibroblasts from foreskin material of normal newborns were initiated in our laboratory as previously described (McCormick and Maher, 1981). The cell lines obtained from the two XP patients, XP12BE and XP12R0, are very deficient in nucleotide excision repair of UV—induced damage (complementation group A) (Robbins gt gt., 1974; de Weerd-Kastelein gt gt., 1972). The non-transformed cell lines were between passage 6 and 16 when used (We use the term "passage" to refer to the number of population doublings the cells have undergone following the first subculture). 54 .4: .qovmom .opSPHpqu nooch ponsmmuwgmm .wz .gwuswo .nouwomom Hwofiuo: mom opzpwpqu .HE .uflonpmn .hpwmpo>fidb madam wnhmzm .hpwmpo>flqb ouwpm qwmflnowsw .nsogqu ow< .92 .mHHH>xoom .qowpomaaoo mmspfiso maze swowpms< .coshommnmnp mapfi>10¢>mw .mnogpzw hp copmflquHM Hg .m.m opaaae om ell sweeps gnaw ma 4smm egg .m.e oufiaam Feuvm e-- s33.9 :me ma wmmm msH opnafle smuom as on ammonn :me mo wemmzu mzH mpnqflp mm-w_ gs mg ammonp anew ma simmso mzH opflaHm msuwe he m_ swmogp qflxm mm mmmmso Mmpswfifins .m assessafia: msxmm mu- smaonp anew mm >mommimx 009a mpficnm mmuom us e ammonp :me mx mmmimx mzH msmpHeHHas _muom gs mi smmonp anew flagpou shoes emsH manage mguom mspme s303 :Hsm asapoa _ioozu omsmno .o.o oufinfim fir sh mv hmmown dwxw Hasnoc m¢m> nooaa magnum _N ha ma KnEOE swam awaken onquo ml: opflnflm mbuOm opmqooq :meosom Hasnon cram mu- manage mmumg opmaomq aflxmmpom Hwenon moqm mu- manage me10m spaces: anxmopom Hugues swam opdpaso :H vahocofiOHmwm nonov mammanonnfim wo magnum ocoo pmfifimmzm qmmmeflq mnflqoao mo om< monsom HmOquHo mcwuw:Mfimmm emflespm mmnHH Hamo mo mogumwpopownmzo .m manna 55 C. Culture Medium and Conditions Cells were routinely cultured in a humidified atmosphere (5% CO2 in air) in Eagle's medium or in modified Ham's F10 medium lacking hypoxanthine (Grand Island Biological Co., Grand Island, NY), with additional NaHCO3 to bring the level to 2.2g/liter to provide additional buffering capacity. The medium was supplemented with 10 or 15% (vol/vol) fetal bovine serum (Sterile Systems, Logan, UT / KC Biologicals, Lenexa, KA / GIBCO, Grand Island, NY / Biocell, Carson, CA / Flow Laboratories, McLean, VA), and the antibiotics penicillin and streptomycin. This medium was referred to as complete culture medium. For mutant selection, growth medium was supplemented with 40uM 6-thioguanine (TG). D. Storage and Recovery of Cells from Liquid Nitrogen Cells not immediately needed for experiments were stored in liquid nitrogen for future use. Cells were detached from the culture vessel with trypsin (GIBCO, Grand Island, NY) and resuspended in freezing medium. This medium consisted of Ham's F10 or Eagle's medium, supplemented with 20% fetal bovine serum, antibiotics and 10% DMSO (anhydrous; stored over molecular sieve). The cell suspension in freezing medium was transferred to a freezing vial and placed at -800 C in an insulated styrofoam container for 2 hours. This 56 procedure allowed the cells to undergo freezing at approximately one degree C per minute. Vials were then placed in a liquid nitrogen freezer where the cells can be stored indefinitely. In order to obtain the greatest recovery and viability of cells during the thawing process, the frozen cells were warmed in a 370 C water bath as quickly as possible and then pipetted directly into a culture flask which contained at least 25 ml of complete culture medium. 25 ml of media was required for thawing to dilute the concentration of DMSO present in the freezing medium. The cells were allowed to attach for 6—18 hours and then fresh complete culture medium was added. E. Testing for Mycoplasma Cells were allowed to attach to coverslips in 35 mm diameter dishes under sterile conditions for at least 6 hours. The coverslips with attached cells were then rinsed in phosphate buffered saline (PBS) and fixed in 25% PBS/75% methanol solution for 15 minutes. The fixative was then o\0 removed and a solution of 0.001 (w/v) of Hoechst dye 33258 in absolute ethanol was added and the cells were stained for 10 minutes before being removed from the dye and air dryed. The coverslips were mounted on glass slides with 50% buffered glycerol, pH 5.5 and the cells were examined for mycoplasma contamination with a fluorescence microscope. 57 F. Cell and Colony Counts After using 0.25% trypsin (Grand Island Biological Company, Grand Island, NY) to detach cells from the tissue culture vessel (Corning Glass Works, Corning, NY), the cells were suspended in medium and then stored on ice for up to one hour before plating. Counting the cells was performed by using a hemacytometer or using an electronic counter (Coulter Count, Hialeah, FL). Serial dilutions were carried out to obtain cells at various concentrations. Colonies were visually scored at 14 days after seeding for cytotoxicity and induced mutation measurements. The colonies were stained before counting and only colonies of 50 cells or greater were counted. The plates containing the colonies were rinsed in saline and stained with 1% crystal violet (in 70% ethanol) or stained in (0.2 %) methylene blue after being fixed in methanol. G. Cytotoxicity and Cloning Efficiency Determinations Sensitivities of cell lines to various chemical mutagens were determined as described by McCormick and Maher (1981) and Maher gt gt. (1979). Cytotoxicity or survival was determined from the loss of colony-forming ability by an tg gttg assay in which the cells were treated at cloning densities. Cells were plated into 60-mm or 100—mm diameter dishes at several different densities (50-1000 cells per 58 dish; 4-6 dishes per density) to obtain at least one set in which the number of colonies ranged from 20-80 after chemical treatment. Eighty colonies per dish are easily countable for such cells as XP12ROSV which develop dense, small diameter colonies. Cell density is important, since cells plated at more than 10,000 cells per 21-cm2 show an increased cloning ability compared to those plated at less than 10,000. Therefore, cells plated at high densities tend to give an artificially increased survival. Feeder layers of X—irradiated cells were not used. The protocol for determining the cytotoxic effects of chemical carcinogens is diagrammed in Figure 5. Cells in exponential growth were trypsinized and plated at appropriate densities for determining survival 6-12 hours before treatment with MNNG, ENU or 4—NQO. Complete culture medium was replaced with serum free medium, buffered with 15mM Hepes, pH 7.4. MNNG was freshly dissolved in anhydrous dimethylsulfoxide and immediately delivered into the serum free medium in the dishes by micropipette. ENU was also dissolved in DMSO. 4-NQO was dissolved in 95% ethanol. The final concentration of dimethylsulfoxide did not exceed 1% and the final concentration of ethanol did not exceed 0.45%. After 1 hour incubation, the medium containing the chemical mutagen was removed and replaced with complete culture medium. Cells were fed one week after treatment and stained at 2 weeks. 59 Figure 5. Protocol for determining the cytotoxic and mutagenic effects of chemical carcinogens 60 m mtzmwm mezozmH mzszmmpma woz_nm=on, :pzoma zoap<5=aoa a>msm mmov >pawzmn zen: >HHqumma>msw ozazono p< >HHUonpop>u mnnmo mp £30 230 2.00 £30 1 - :2. 6933— 60:03:50 9.203.. 230 @ beamed .Eoncgo a: 33:20 a «a £3 9:395 uzzazscoaxu 32m 69 After the 15 minute pulse, cells were removed from the dishes with trypsin, and transferred onto glass fiber filters (Whatman, GF/C). The filters with cells were then washed twice with ice cold PBS and ice cold trichloroacetic acid (10%) and then washed once with ice cold ethanol (95%). The filter was dried under a sunlamp for 5-10 minutes. The radioactivity was measured by placing the glass fiber filter into a toluene-based scintillant and reading the counts per minute with a Beckman liquid-scintillation counter. IV. RESULTS A. Sensitivity of Cells to the Killing Effect of MNNG As mentioned in the literature review section, several groups of investigators had reported that fibroblasts derived from patients with familial polyposis (FP) or Gardner's syndrome (GS) are abnormally sensitive to a variety of DNA-damaging agents. These results suggest that cells from such patients might be deficient in DNA repair. If so, these cells might also be more sensitive than normal cells to the mutagenic effect of these same agents. Before beginning a study comparing the frequency of mutants induced in normal, FP and GS cells by carcinogens, a comparison of their sensitivity to cell killing was carried out. N-methyl—N-nitro—N-nitrosoguanidine (MNNG) was selected as the carcinogen of interest because N—nitroso compounds have been implicated in colon cancer. Figure 7A compares the survival curves of an FP cell line and four GS cell lines with that of normal fibroblasts, SL68 (neonatal) and GM0011 (fetal), exposed to MNNG at cloning densities. The data were obtained from a series of cytotoxicity studies, but SL68 cells were always included in each experiment for comparison. Therefore, the survival 70 71 Figure 7. Comparison of cytotoxicity (A) and mutagenicity (B) induced by MNNG in human fibroblasts. Cells in exponential growth were treated with MNNG for l h at a 5 cells per 150—mm dish for density of not more than 8 x 10 mutagenicity determination. The background mutation frequencies per 106 clonable cells were: 3 and 8 for normal cells (SL68); l3 and 0 for normal cells (GM0011); 0, 7 and 11 for FP cells (GM2355); 0 and 0 for GS cells (GM3948); 45 for GS cells (2938); and 20 and 53 for GS cells (GM3314). The mutagenicity data have been corrected for these backgrounds. The cloning efficiencies of these cell lines are listed in Table 3. Bars representing the standard error of the mean are included for the cytotoxicity data. Symbols without error bars represent individual determinations of survival or determinations for which the symbol drawn is greater than the error bar. Lines were fitted by eye. 72 MNNG(qu1hr) 4 6 8 IO ICDC3fflF1:QE:, I I l I77 I I I I _I ‘éqj 0 SL68, g C 60 . 9. AGM2355 ; g “ Bf - GM3948 55 S? 41C)§%$ Ji!L\. 53(3A43y3l‘4 (f) o I I: [I 2938 52 g g? n 2974 Lu 2 , \o OGMO0I I fr) 3 _g‘. .\ Lu 0 20 .‘ 0- I e A A '0 I I I l I I I q I I [40)- I I I I I I I I I I 0 " B on |2C3“- _J m d j I 0 Lu Ioo— , }_ U z: _. a 5 . (I) g 80'- (7) Z _ [iii] 3 O SL68 0 o 60— x A GM2355 8 - . z GM33I4 5’33 40 go a :1 2938 a CL OGMOOII Z - 20 o :3 /‘ C) IQJ'I'IrI/.1 1 L J I I It 0 2 4 6 8 IO Figure 7 MNNG (uM x1 hr) 73 data for SL68 cells exposed to particular doses of MNNG in many different experiments were combined and averaged. The error bars shown in Figure 7A give the standard error of the mean for SL68 cells. Since our normal strain, SL68, is derived from neonatal fibroblasts and the FP and GS cell lines are from patients that range in age from 11 to 45 years old, it was important to investigate the survival data for age matched normal cells. Also, survival experiments for other neonatal cells were carried out to find the range of cytotoxicity induced by MNNG in normal fibroblasts. The MNNG survival data obtained with a series of neonatal foreskin-derived cells and skin fibroblasts from a 16-year-old (CRL1220) and a 45—year-old person (VE45) were similar to that shown for SL68 cells (Table 4). The data in Figure 7A show that the FP cells and three GS cell lines had normal sensitivity to MNNG cytotoxicity, but GM3314 and GM0011 cells were extremely sensitive. GM3314 fibroblasts are from a patient with Gardner‘s syndrome. GM0011 cells are from an apparently normal fetus (8 weeks old). To determine whether the sensitivity to MNNG of the fetal line GM0011 was a general characteristic of fetal skin fibroblasts, we measured the response of three other fetal fibroblast lines (WI-38, 57372 and GM1380) to the cytotoxicity of MNNG. The results are presented in Table 5. All three cell lines showed a normal response, similar 74 Table 4. Cytotoxicity of MNNG in normal human fibroblasts % SURVIVAL MNNG (uM)a SL68 VE45 CRL1220 0 100 100 100 3.5 55 50 52 5 33 22 29 7 12 7 9 aCells were exposed for 1 hour. 75 Table 5. Cytotoxicity of MNNG in normal neonatal and fetal cell lines 9 SURVIVAL O normal neonatal normal fetal MNNG (uM) SL68 WI—38 GM1380 57372 GM0011 0 100 100 100 100 100 0.5 93 71 80 82 55 l 85 65 78 71 33 5 44 16 27 32 2 76 to that of the normal neonatal cell line SL68. GM0011 cells were recently reported by Middlestadt gt gtt (1984) to exhibit increasing sensitivity to MNNG with increasing passage number in culture. The GM0011 cells shown in Figure 7A were treated at passage 8. They showed slightly higher survival at passage 6 (Da = 1.5uM) and slightly lower survival at passage 14 (Dg = 0.5uM) (data not shown ). However, the difference was not significant enough to warrant further investigation. B. Sensitivity of Cells to the Mutagenic Effect of MNNG Figure 7B gives the mutagenicity results obtained with an FP cell line, four GS cell lines and the normal fibroblast cell line SL68. Each mutation experiment was accompanied by a cytotoxicity experiment and these corresponding data are included in Figure 7A along with the results of experiments comparing survival only. GM3314 and GM0011, the two cell lines that were abnormally sensitive to the cytotoxic effect of MNNG, showed a dose—dependent increase in TG resistant cells at very low doses, i.e., < 1.5uM. All the other cell lines showed little or no increase in mutant frequency at these levels of MNNG. Only after treatment with MNNG doses of 2uM or greater did they exhibit an increase in mutant frequency. Note that the increase was not linear. No significant increase in mutant frequency occurred at an MNNG dose of 2uM. Between 3uM and 77 4uM, the increase was linear, but then the frequency increased rapidly. GM3948 cells, taken from a Gardner‘s syndrome patient, appeared to be slightly more resistant to the mutagenic effect of MNNG than were the rest of the cell lines with normal resistance. However, much more data would be needed to prove that this difference was statistically significant. C. Comparing Human Cell Lines for Their Levels of 06-Methylguanine—DNA Methyltransferase Activity It is known that human cells contain 06—methylguanine-DNA methyltransferase (Grafstrom gt al., 1984). As explained in the literature review, this repair protein functions not as an enzyme, but as a methyl protein acceptor molecule. Since human cells have a finite number of such proteins, the supply may well be exhausted by very large doses of methylating agents. If this protein acceptor molecule caused rapid removal of the majority of the potentially mutagenic lesions induced by MNNG at low doses, but was unable to do so in cells exposed to higher doses (Medcalf and Lawley, 1981), this would explain the observed "threshold" on the mutation curve for the majority of cell lines shown in Figure 78. Such a hypothesis predicts that GM3314 cells and GM0011 cells are deficient in this repair mechanism, i.e., have a much lower level of MT activity than the other cell lines tested. 78 The level of MT activity (ability to remove methyl 6 position of guanine) in the cell lines groups from the 0 shown in Figure 7B and in two additional skin fibroblast cell lines prepared from foreskins of normal neonates (SL66 and SL70) was determined by measuring the decrease of 6 radioactively labelled O —methylguanine from a DNA substrate using procedures previously described (Pegg gt gt., 1983; Pegg and Balog, 1979). For such studies, approximately 107 cells of each cell were prepared, harvested in the appropriate buffer (see section III — Materials and Methods), stored at -800C and transported to the laboratory of Dr. Anthony Pegg where the analysis was conducted. The analyses shown in Table 6 were conducted by myself under the direction of Dr. Pegg during my stay at Pennsylvania State University, Hershey, PA. Later analyses were carried out by Dr. Pegg and his associate, Dr. Eileen Dolan. For comparative purposes, three additional cell lines were also examined. These included two SV40 virus-transformed cell strains (XP12ROSV and GM637) known to differ in ability to support survival of MNNG—treated adenovirus and classified respectively as methyl repair deficient (mer_) and methyl repair proficient (mer+) (Day gt 'gt., 1980). The third cell line, designated XP12BE, is a non-transformed XP cell line, previously shown to be extremely deficient in excision repair of DNA damage induced by ultraviolet radiation (Robbins gt gt., 1974) or by carcinogens that form multi—ringed DNA adducts (Heflich gt 79 gt., 1980; Yang _t _t., 1980), but to exhibit normal sensitivity to the killing action of MNNG (Simon gt gt., 1981). Table 6 shows the results of the MT activity assay for each of these cell lines. Each value listed is the mean of three determinations. The GM3948 cells, which exhibited the lowest frequency of MNNG-induced mutations (Figure 7B) showed one of the highest levels of MT activity. GM3314 and GM0011, the two sensitive strains, showed much lower or virtually non-detectable levels of MT, as did the mer- XP12ROSV cell line. Although the data in Table 6 suggest that the level of MT activity in GM0011 declined with increased passage number, one cannot be certain of this because the levels observed were so low. (A value of 8 fmol methyl removed per mg protein is close to or below the detection limit of the assay.) Of the 8 cell lines with measureable levels of MT activity, the GM637 cells, exhibited the lowest values. The FP cell line (GM2355) and three GS cell lines (GM3948, GM3314 and GM2938) showed levels of MT activity similar to that of the three normal foreskin-derived cell lines. 80 Table 6. Comparison of the levels of 06-methylguanine—DNA methyltransferase in cell extracts of various human fibroblast cell lines Cell line 06-methylguanine-DNA methyltransferase activity (fmol 06-MeG removed per mg protein) Individual experiments Mean XPIZBE '"1'935753; “““““ _EII GM3948 148, 267 208 SL70 202 SL66 152 GM2355 126, 159 142 SL68 112, 135 123 2938 103, 108 105 GM637 121, 85, 83 96 GM0011 p. 6b 16 GM0011 p. 13 <10 GM3314 l0, 6 8 XP12ROSV 12, 3 7 aEach experiment used different preparations of cells and the MT activity of the cell extracts was measured at 2 or 3 protein concentrations and the average value is given in this column. The next column gives the mean value from the individual experiments. bThe cells were supplied as passage 3 and allowed to replicate 8—fold. 81 4 D. Ability of MT to Remove Methyl from 0 —Methylthymine It has recently been shown (McCarthy gt gt., 1983; McCarthy t 1., 1984; Ahmmed and Laval, 1984) that the methyltransferase of E. coli can remove methyl groups from the O4 6 position of thymine, as well as from the 0 position of guanine. Because 04-methylthymine has been shown to miscode tg ytttg (Loveless ,1969), it was important to see if human MT could remove this lesion. Therefore, XP12BE cells, the cell line with the highest level of MT activity (see Table 6), were grown to large numbers, prepared and shipped to the laboratory of Dr. Pegg to be tested for the ability of the extracts to remove methyl 4 position of thymine. groups from the O The assay consists of testing the ability of human cell extracts to remove methyl from the 04 position of thymine using a radioactively labelled poly(dT)°poly(dA) substrate. At the same time, the cell extract was tested for the removal of 3-methylthymine, 06—methylguanine and 7—methylguanine. The results are shown in Table 7. The XP12BE cell extract had normal activity against 06-methylguanine but was completely inactive against 04-methylthymine. 82 Table 7. Specificity of human 06 transferase -methylguanine-DNA methyl- Methylated bases remaining in substrate (pmol) Extracta 04-methyl- 3-methy1- 06-methyl— 7-methyl— added thymine thymine guanine guanine None 0.36 0.51 0.54 5.6 1.35 units 0.39 0.56 0.04 >4.3 of alkyl- transferase from human fibroblasts XP12BE aThe extract was incubated with the appropriate DNA substrate (either methylated poly(dT)'poly(dA) for analysis of methylated thymines or methylated calf thymus DNA for methylated guanines) and the effect on speciEic methylated bases was determined by HPLC. One unit of O -methylguanine- DNA methyltransferase activity correspongs to the ability to remove 1 pmol of methyl groups from O —methylguanine in DNA. 83 E. Comparing the MNNG Sensitivitygof These Other Three Cell Lines Once the level of MT activity in cell lines XP12ROSV, GM637, and XP12BE was known, these same cell lines were compared to the others for their response to the cytotoxic and mutagenic effect of MNNG. The results are shown in Figure 8. The data for the various strains shown in Figure 7 have been included in Figure 8 for comparison. As predicted from their level of MT activity shown in Table 6, XP12BE cells showed a normal response to MNNG; GM637 cells were more sensitive than normal; and XP12ROSV cells were extremely sensitive to the cytotoxic and mutagenic action of MNNG. As was seen for GM3948 cells, the response of the XP12BE cells, which had a high level of MT activity, to the mutagenic effect of MNNG appeared to be somewhat lower than that of normal cells. For comparative purposes, Table 8 presents the data from two representative experiments for each cell line except GM3314. The latter cell line had background mutant frequencies that ranged from 20—500 mutants per 106 clonable cells. In only two experiments did we find the background frequencies to be low: 20 and 53. It will be seen that a dose-dependent increase in the frequency of TG resistant cells was obtained in GM0011 and XP12ROSV cells at "threshold doses", doses that did not cause a significant increase in normal cells or in GS cell lines that were shown 84 Figure 8. Comparison of cytotoxicity (A) and mutagenicity (B) induced by MNNG in XP12BE cells and two SV40-transformed cell lines. Data from Figure 6 are shown for comparative purposes. The background mutation frequencies per 106 clonable cells were: 2 for XPlZBE cells (CRL1223); 2 and 18 for GM637 cells; and 0, l and 0 for XP12ROSV cells. The cloning efficiencies of these cell lines are listed in Table 3. Lines shown were fitted by eye. PERCENT SURVIVAL CLONING ABILITY INDUCED TG RESISTANT CELLS Figure 8 PER I06 CLONABLE CELLS 85 MNNG (ILMx1hr) O 2 4 6 8 IO I I I I i N C) @o 0SL68 AGM2355 IGM3948 EIGM33I4 :1 2938 * U 2974 AXPIZBE OGM637 OXPIZROSV O OGMOOII A. A I O I I I I I40 I20 I00 0 SL68 AGM2355 I GM3948 EGM33I4 :1 2938 AXPIZBE OGM637 OXPIZROSV OGMOOII L I I O 2 4 MNNG (quIhr) 6 8 | O 86 Table 8. Representative mutagenicity data of MNNG in various cell lines Target MNNG Original Cells Thioguanine Cloning Mutant: cells concen- surviving selec ed resistant efficiency per 10 tration fraction (x 10 ) colonies at clonable (uM x 1 h) selection cells SL68 0 1.00 1.19 5 0.50 8 1 0.86 1.19 10 0.65 13 expt. 2 0.95 1.19 10 0.59 14 24 4 0.51 1.19 35 0.56 52 6 0.26 1.19 87 0.51 143 expt. 0 1.00 1.32 2 0.53 3 21 4 0.56 1.32 16 0.42 29 6 0.27 1.32 26 0.31 64 8 0.11 1.29 28 0.22 100 GM3948 0 1.00 1.76 0 0.53 0 4.5 0.46 1.76 18 0.38 27 expt. 5.5 0.35 1.76 12 0.36 19 54 6.5 0.16 1.76 21 0.28 43 0 1.00 1.72 0 0.18 0 expt. 1.6 0.85 1.76 1 0.17 3 22 2.5 0.76 1.69 1 0.19 3 3.6 0.62 1.63 2 0.14 9 GM2355 0 1.00 1.32 7 0.47 11 3 0.68 1.32 14 0.49 22 expt. 4.5 0.48 1.32 22 0.40 42 19 6.0 0.32 1.32 46 0.38 92 0 1.00 1.32 5 0.54 7 expt. 3 0.71 1.32 13 0.40 25 18 4.5 0.34 1.32 23 0.38 46 6.0 0.36 1.32 17 0.32 40 87 Table 8. (continued) Target MNNG Original Cells Thioguanine Cloning Mutantg cells concen- surviving selec ed resistant efficiency per 10 tration fraction (x 10 ) colonies at clonable (uM x 1 h) selection cells GM637 0 1.00 1.76 2 0.28 4 1.0 0.55 1.76 15 0.31 27 expt. 2.0 0.44 1.76 39 0.25 89 38 3.5 0.31 1.76 59 0.26 130 0 1.00 1.41 18 0.32 40 expt 0.5 0.69 1.41 16 0.28 41 49 1.5 0.59 1.41 25 0.28 62 3.0 0.50 1.39 29 0.28 75 XP12ROSV 0 1.00 2.05 1 0.38 1 0.25 0.42 2 11 8 0.42 9 expt. 0.40 0.34 2 06 11 0.40 13 51 0.60 0.19 2 11 19 0.36 25 0 1.00 1.74 0 0.52 0 expt. 0.1 0-78 1.74 3 0.53 3 37 0.2 0.53 1.74 6 0.48 7 0.3 0.37 1.69 11 0.47 14 GM0011 0 1.00 0.68 2 0.23 13 expt. 0.65 0.31 1.34 10 0.25 30 yy' 1.00 0 26 66 23 0.28 49 0 1.00 1 74 0 0.40 0 expt. 0.3 0.84 1 76 23 0.45 29 62 0.6 0.84 1 74 33 0.46 41 1 0 0.63 1 94 30 0.36 44 GM3314 0 1.00 1.74 6 0.065 53 0 125 0.77 1.76 7 0.068 58 expt. 0.25 0.40 1.76 8 0.050 90 32 0.50 0.24 1.76 4 0.040 57 88 to have normal levels of MT activity. These SV40-transformed cells differ from non—transformed fibroblasts in that they are less tightly attached to the surface of the culture vessel. Therefore, in determining the mutation frequencies with these cell lines, as well as with the others, the frequencies were calculated not only from the observed number of mutant colonies as shown in Table 8, but also using the Poisson distribution function. That is, the chance of a TG resistant colony per dish was calculated from the number of dishes containing no colonies. One determines the probability of a mutational event occurring per experimental culture dish (x) by measuring the fraction of culture dishes containing no clones, P(0), and applying the formula P(0) = exp (-x). Use of this method eliminates the possibility that the frequencies observed were artificially increased by the formation of satellite colonies. No evidence of such increase was found. The results were the same using either method, indicating that formation of satellites was not a problem. 89 F. Sensitivity of Another SV40 Transformed Cell Line to the Cytotoxic Action of MNNG Day and coworkers (1980) reported that the SV40 transformed cell line XP12RO that they used called XPT703 was intermediate in sensitivity to MNNG. Our results (Table 9) indicate that XP12ROSV was extremely sensitive to MNNG. The assay of Day gt gt. (1980) was inactivation of adenovirus but this assay usually gives similar results to a survival curve. Therefore, we examined the line from Dr. Day and measured the cytotoxic response of XP12ROSV and XPT703 to MNNG under the same experimental conditions. The results showed that indeed XPT703 was not as sensitive as XP12ROSV. G. Measuring Effect of MT Depletion on Mutations and Cell Survival 1. Inhibition of MT by Formaldehyde Pegg and coworkers (1983) have found that in mammalian cell extracts, methyltransferase is specific for the removal of the lesion 06-MeG. If 06 —MeG is the lesion responsible for cytotoxicity and mutagenicity in human cells, then if one could inhibit the action of the DNA repair protein or decrease the number of molecules and then challenge such cells with MNNG, it should be possible to detect a decreased 90 Table 9. The cytotoxicity of MNNG in two viral transformed XP cell lines % SURVIVAL MNNG dose(uM)a XP12ROSV40 XP12T703 experiment 64 0 100 100 0.25 30 48 0.50 35 81 2.0 17 65 3.0 8 50 experiment 59 0 100 100 0.25 36 65 0.40 17 54 0.60 11 46 2.00 pr 40 4.00 NP 31 aCells were exposed for 1 hour. bNot performed. 91 survival and an increased mutation frequency. Use of such procedures should give an indication of the biological role of the 06-MeG lesion in DNA. Preliminary observations by Grafstrom _t _t. (1983) indicated that MT activity in human cells could be inhibited by exposing cells to formaldehyde. Therefore SL68 cells plated at cloning densities were exposed to concentrations of formaldehyde of 0, 25, 50, 75 and 100 uM in serum-free medium for 5 hours to determine the cytotoxicity of formaldehyde in preparation for its use in subsequent survival and mutation experiments with the alkylating agents MNNG and ENU. THe results are shown in Table 10. To determine the effect of low doses of formaldehyde on these two parameters, cells were plated at cloning densities to determine cell survival in l00—mm dishes and at 0.4 x 106 cells per 150-mm dishes to determine induced mutation frequencies. The cells were allowed to attach for 6 hours before addition of 60uM formaldehyde (80% survival) and either 1.75 mM ENU or 0.5uM MNNG for 1 hour. This solution was then removed and formaldehyde (60uM) in serum-free media was added for an additional four hours before complete medium was added. Table 11 shows the results. No increase in mutation frequency was seen. However, the level of MT activity in the cells pre-treated with formaldehyde was not measured in this experiment. Therefore the negative result may reflect the fact that the formaldehyde concentration used did not inhibit MT significantly. Further 92 Table 10. tg situ cytotoxicity of formaldehyde (HCHO) in normal cells (SL68) HCHO % SURVIVAL (uM) a 0 100 25 95 50 87 75 76 100 41 aCells exposed for 5 hours. 93 Table 11. Effect of formaldehyde (HCHO) on the cytotoxic and mutagenic response of SL68 cells to alkylating agents Dose Dose Dose Percent Mutants cho ENU MNNG SURVIVAL (x 106) 0 0 0 100 0 60uM 0 0 72 0 0 0.75mM 0 93 69 0 1.75mM 0 58 96 60uM 1.75mM 0 27a 102 0 0 0.5uM 77 0 0 0 5.0uM 50 54 60uM 0 0.5uM 80b 0 aThis corresponds to 38% of its own control. bThis corresponds to 110% of its own control. 94 experimentation using formaldehyde to reduce the MT activity in human cells was not pursued because sometime later Dr. Peter Karran reported (personal communication) that addition of the free base 06 —methylguanine to the medium caused a decrease in the level of MT activity in cells. Therefore, experiments to reduce MT activity by exposure of cells to exogenous 06-MeG were carried out. This was the preferred method since an increase in mutations with formaldehyde plus an alkylating agent could be due to the fact that formaldehyde causes strand breaks and this could be a synergistic effect (Grafstrom gt gt., 1984). 2. Depletion of MT by Exposure to Exogenous 06—MeG Experiments were conducted in normal human cells (SL68) to determine whether or not the level of methyltransferase activity could be reduced by exposure to exogenous 06-MeG in the culture medium. As shown in Table 12, when exogenous 06-methylguanine was present in the 6 cells culture medium (0.4 or 0.8 mM) of SL68 cells (8 x 10 per 150-mm dish) for 15 or 24 hours, there was a substantial loss of MT activity in extracts prepared from the cells (30—70 x 106). The average reduction in MT activity was 65%. Increasing the dose of exogenous 06-MeG from 0.4 mM to 0.8 mM did not decrease the activity substantially. It should be noted that cells with depleted levels of MT 6 activity could regenerate MT after the removal of the O -MeG medium (Table 13). In MT depleted cells, the MT activity 95 Table 12. Effect of exposure to exogenous 06-MeG for 15 h or 24 h on the level of MT activity in human cell extracts Control population 06-MeG-exposed population Experi- MT activity Dgse MT activity Activity ment (fmol removed 0 -MeG (fmol removed) remaining number per mg protein)a (mM) per mg protein) (% of control) I 272 0.4 108 40% II 210 0.4 80 38% III 480 0.4b 200 42% b 0.8 150 31% IV 317 0.4 66 21% aEach value is the mean of 3 determinations. b15 h exposure. 96 Table 13. Regeneration of MT activity in cell ex racts after depletion by exposure to exogenous O -MeG MT activity Experimental (fmol removed Percent conditions per mg protein) of control Ng exposure to 317 100% O -MeG 2% h of exogenous 66 21% O —MeG (0.4mM) 24 h after remgval 258 80% of exogenous O -MeG 48 h after remgval 298 94% of exogenous 0 -MeG 97 regenerated to 81% of its initial value in 24 hours and to 94% at 48 hours. The ability of the free base, 06-MeG, to act as a substrate for MT provides an opportunity to specifically manipulate MT levels. This permits one to determine the effect of MT on the cytotoxic and/or mutagenic effect of alkylating agents. Cells were plated at a density of 0.5-1.0 x 106 cells per l00-mm dish for determining cell survival and induced mutations. The cells were allowed to attach for 6 hours before addition of the 06-MeG medium. 3: 6-MeG medium was removed and " Fifteen or 24 hours later, the O the MT-depleted populations and their corresponding non-depleted control populations were challenged with various doses of MNNG and assayed for cytotoxicity and mutagenicity. As shown in Figure 9, at every dose of MNNG the frequency of TG resistant cells was higher in the MT-depleted cells (triangles) than in the corresponding non-depleted populations (circles). Data from a representative mutagenicity experiment are shown in Table 14. The non-depleted populations showed no increase in mutant frequency at doses of MNNG below 2.5uM, a linear increase at intermediate doses of MNNG, but then the frequency increased rapidly. In contrast, in the depleted populations, a dose of 2uM MNNG was sufficient to induce mutations and the frequencies induced by 3.5uM MNNG were similar to those induced by 4.5uM in the populations with normal levels of MT activity. From the insert in Figure 9, 98 Figure 9. Mutagenicity induced by MNNG in populations of human fibroblasts with reduced levels of MT activity or with normal levels. Data derived using cells with decreased MT activity caused by 15 or 24 h exposure to 0.4mM exogenous 06-MeG are shown as triangles, that from cells with normal levels of MT are shown as circles. Cells were treated with 6 cells per 150-mm MNNG for 1 h at a density of ~1 x 10 diameter dish and selected for TG resistance (40uM) after an expression period of 7 d. The induced frequencies have been corrected for the background frequencies in the populations not treated with MNNG and for the cloning efficiencies of the cells at time of selection. Data from individual experiments are distinguished by the shading of the symbols used, e.g., Exp. II, 0, A ; Exp. IV, (D, A. 99 _ _ _ 6 I/ I1 I /A O/ _ A A 00/ I 4 I A O/ A on I IIWI/I A $2 A/MJTW All.» .. _ _ _ O O O O O O 6 2 8 4 mujwo “1.542040 @0_ mm& mu_n_m_o ._.Z<._.m_mwm 0H omoaoz. MNNG (11M x 1h) Figure 9 100 .omSI o msoqmmoxm 284.0 m o» vmmomxc mHHoow now 00.0 mm.o No.0 F 00* Aaoquoc mo gem V mm.o 4m.o 4w.o F 00? mas o m. mHHoc wgoao P pom amass nowpomamm mo mswp pm homewowmmm mgwqoao wm>nomno mowqoaoo uqmpmwmmp ocwgmsmOHse A low NV Aaoppdoo mo bwpooamm pqmonomv mHHoo mHHmo pomnmp proe Hw>w>ssm mHHmo pomnwp mo hoqmwowmmm mneqoao Assv wzzz omoa Aqflmposm ms pom co>oSon Hosmv maamo wowed» aw sesssees as memsaeoseee seas: esseess as 822: go seeossemsess use assessosoese see so assesses as to soseefiees to posses .41 eHese 101 it can be seen that at high doses of MNNG (SuM), the difference in the mutagenic response of the two populations '(non—depleted and depleted) was less pronounced, as expected, if the number of potentially mutagenic lesions induced by high doses of MNNG far exceeds the repair capacity of either population. Each mutation experiment was accompanied by a replating cytotoxicity experiment and these corresponding data are shown in Table 15. The data indicate that populations with decreased levels of MT activity, caused by 15 or 24 hours of 6—MeG, were not significantly more exposure to exogenous O sensitive to the killing action of MNNG than non—depleted populations. H. Effect of the Cell Cycle on MNNG Induced Cytotoxicity and Mutagenicity 1. Use of Aphidicolin to Synchronize Human Fibroblasts Recent studies by Grossman gt gt. (1985) in this laboratory demonstrated that it is possible to synchronize human cells using a Gl/S block. The block is aphidicolin, which is known to inhibit DNA polymerase alpha (Pedrali-Noy _t _t., 1980). In preparation for those studies, experiments were carried out designed to determine the length of the S-phase in synchronized human cells following release from a Gl/S block. The cells had been synchronized by being grown to confluence and starved for mitogens (G0 102 mm mm 00F mm or cow mm mm COP oo— 00F Aeeseaeee 92V >4 mm 00— mm mm 009 mm mm 00— so OOF Awopoammvlgoqv macapmasmom women» mo Hm>w>nsm pgmonmm mm.o m>.o 4m.o Aeeeeeeesnazv mm.o om.o mm.o Avopoammvnqogv mQOHpMHsmom pompwp mo homewowmmo mgwqoao A23 @222 omom R—m Rmv &wm Aaoppqoo mo &v quGHMSmp sessseos a: 822: to assessoeoeso mg» as sesssees as to soseefieee so posses >H HHH HH popes: pace prmmxm .mF mange 103 state), then plated at a lower density and incubated for 24 hours in medium containing aphidicolin (Grossman gt gt., 1985). The onset and length of the S phase was determined by measuring the extent of incorporation of radiolabelled thymidine during 15 minute pulses every hour for 16 hours. The incorporation data in Figure 10 shows that upon release from the aphidicolin block the S phase starts immediately and is 8 hours in length. 2. Effect of Cell Cycle on Mutagenesis by MNNG The effect of the cell cycle on the frequency of mutations induced by MNNG in normal SL68 cells is shown in Figure 11. Cells were treated with MNNG, 1.5 to 2.5 hours into the onset of the S phase (solid circles) at a density of l x 106 cells per 150—mm dish. The dashed line represent data (taken from Figure 7) for SL68 cells that were in exponential (asynchronous) growth when treated with MNNG. When synchronous cells were treated with MNNG (3, 4 and 5uM) during the S phase, the induced mutation frequency was approximately 3 times higher than if asynchronous cells were treated with MNNG. 3. Effect of Cell Cycle on the Cytotoxicity of MNNG The effect of cell cycle on the survival of SL68 cells after treatment with MNNG is shown in Figure 12. The dashed line represents data (taken from Figure 7) for SL68 cells 104 3 Figure 10. Incorporation of H—TdR during S phase in cells synchronized by release from aphidicolin. At the times post-release, cells were pulse labeled for 15 min with 3H-TdR and analyzed for incorporation of label into acid insoluble material as described in Materials and Methods, section III. 105 O m _ — _ _ _ _— _ 0 0 0 0 0 m 8 6 4 2 6.9 x .89 meets no 292404482. 1 234567 I C)O I HOURS Figure 10 106 Figure 11. Mutagenicity of MNNG in SL68 cells at S and with time for repair. The background mutation frequencies per 106 clonable cells were: 4 and 12 for experiments in which cells were treated at S; 0 and 57 for experiments in which cells were treated 24 h to S. The asynchronous curve has been taken from Figure 6. The mutagenicity data have been corrected for these backgrounds. 107 v III// l /./O /// I S S ..n /. ///p/A 0 0 ...I ....l m /. // l as 4., I x 2 A I 0 o u .. _ p _ _ F _ _ _ _ _ _ _ p _ O O O O O O O O M 2 O 8 6 4 2 3.8 ”Tm/420.5 so. mme mjmo 4259mm”. me 8082. 2 4 6 8 MNNG (qulh) 0 Figure 11 108 Figure 12. Cytotoxicity of SL68 cells treated at S 109 MNNG (ILMxlh) 4 2 H _ 4 d _ _ _ — — Tl \\\\ IJ \ I \ I \\ ll IA-v. ‘\\\ I1 \ II A-'\\\. ll \ I\ I :4 _ _ _ _ _ _ _ _ O O O 6 4 2 m >._._I__m< 02.2040 .._<>_>m3m HZMUGMQ Figure 12 110 that were asynchronous when treated with MNNG. When synchronous cells (0.2 x 106 cells per l00—mm dish, 24 hr. aphidicolin) were treated with MNNG (3, 4 and 5uM) at 1.5—2.5 hours into the onset of the S phase (solid circles) and then replated at cloning densities, the survival was found to be no different from that of asynchronous cells. I. Effect of Holding Cells at the G1/8 Border Post-treatment on the Mutagenicity and Cytotoxicity of MNNG Liquid holding or recovery experiments are designed to measure the repair of DNA lesions that occur when the cells are held in a non-dividing state. Otherwise, if the cells are dividing, the lesion can become fixed in the S phase (Konze-Thomas gt gt., 1982; Yang gt gt., 1982). With the inhibitor of DNA synthesis, aphidicolin, it is possible to hold cells in a non-replicating state. If cells in aphidicolin are treated with MNNG and held at the Gl/S border to allow lesions to be repaired before fixation, then the mutation frequency should decrease because of repair of the 06-MeG lesion. 1. Measuring the Cytotoxic Effect of Incubation of Cells in Aphidicolin Before attempting to determine whether holding cells in aphidicolin affected their response to MNNG, experiment were initiated to determine if aphidicolin itself was cytotoxic. 111 Cells (0.5 x 106 per 100-mm dish) were synchronized with aphidicolin and held in aphidicolin for an additional 0, 8, 16 or 24 hours before they were seeded at cloning densities. The results in Figure 13 clearly demonstrate that holding cells in aphidicolin medium decreases the cloning efficiency of the cells at 0 uM MNNG. The cloning efficiencies were 0.80 (100%), 0.76 (84%), 0.63 (79%) and 0.48 (61%) for cells held in aphidicolin medium for 0, 8, 16 and 24 hours. 2. Measuring the Effect of Holding Cells in Aphidicolin on Their Response to MNNG To determine whether cells have the ability to remove potentially cytotoxic or mutagenic lesions induced by MNNG, when given time to repair between treatment and entry into S, cells were released from confluence and plated at 1.5 x 106 cells per 150-mm dish (mutagenicity) and 0.6 x 106 cells per 100-mm dish (cytotoxicity) in medium containing aphidicolin. The cells were then held in aphidicolin medium for 0, 8, 16 or 24 hours after treatment with MNNG (0, 3, 4 or 5 uM). Cells were held in aphidicolin medium for 0 and 24 hours for the mutagenicity determinations and 0, 8, l6 and 24 hours for the survival determinations. Figure 13 demonstrates that the survival decreases when holding cells in aphidicolin for 16 or 24 hours after treatment with MNNG. The survival data for an 8 hour holding period in aphidicolin medium was not significantly different than the data for 0 hours. These results indicate that holding cells 112 Figure 13. The cytotoxic effect of holding human cells (SL68) at the Gl/S border in aphidicolin medium and the effect of holding on their cytotoxic response to MNNG. Cells were held in aphidicolin for 0 (Q), 8 (. ), 16 (A ), or 24 ( I. ) hours after treatment with several concentrations of MNNG. The cloning efficiency for the control was 80%. 113 MNNG (11M xlh) 4 6 r I 2 I >._._I__m< 022040 I_<>_>mDm HzmommE 10 Figure 13 114 in aphidicolin medium for longer than 16 hours is toxic to the cells and also that holding in aphidicolin does not cause any significant change in survival. The mutagenicity data are illustrated in Figure 11. For 0 hours holding in aphidicolin medium, the data are represented by the solid circles (0 hours to S) and holding for 24 hours is represented by the the open circles (24 hours to S). The induced mutation frequency decreased with holding in aphidicolin medium. This preliminary result indicates that the potentially mutagenic adduct is removed when time is allowed for repair. V. DISCUSSION 6 A. 0 -MeG Is a Cytotoxic and a Mutagenic Adduct Simple alkylating agents, such as N-methyl—N'—nitro—N-nitrosoguanidine (MNNG), can react with all nucleophilic sites in DNA (Singer and Kusmierek, 1982). The lesions 06—methylguanine, 02-methylthymine, 04-methylthymine and 02-methylcytosine are capable of miscoding tg ttttg and have been proposed as potentially mutagenic (Singer and Kusmierek, 1982). Studies with CHO cells (Heflich gt gt., 1982) and V79 cells (Newbold gt gt., 1980) show that o6 -a1kylation of guanine in DNA is strongly correlated with mutation induction, but not cell killing. However, in the present comparative study using a series of human fibroblast cell lines, the data suggest that 06—MeG is a potentially cytotoxic and mutagenic adduct. 1.Correlation Between MT Activity and Resistance of Human Cells to the Mutagenic and Cytotoxic Effects of MNNG The data in Tables 6 and 7 and Figure 8 indicate that the ability of human cells to remove methyl groups from the 06 position of guanine is highly correlated with resistance 115 116 to both the cytotoxic and mutagenic effect of MNNG, 6-methylguanine is a potentially cytotoxic suggesting that O lesion as well as a potentially mutagenic lesion. The three cell lines virtually devoid of MT activity were extremely sensitive to the cytotoxic effect of MNNG and exhibited mutation induction at doses which are non-mutagenic for cells with measurable levels of MT. A cell line with an intermediate level of MT activity showed intermediate sensitivity to the cytotoxicity and mutagenicity of MNNG. The five cell lines with normal levels of MT showed a distinct shoulder on their survival curves and a corresponding threshold on their mutation induction curves. However, Waldstein gt gt. (1982c) reported that cells that were deficient and proficient in the reactivation of MNNG-treated adenovirus (mer‘ and mer+) had the same levels of MT activity. However, these investigators used only a single, relatively low amount of protein in their assay on crude cell extracts. They determined MT activity from the number of radioactive counts transferred to an acid precipitable fraction. It is difficult to detect consistent differences between mer+ and mer- cells under these experimental conditions. Furthermore, studies with Chinese hamster cell lines with unlimited lifespans, i.e., CHO cells (Heflich gt gt., 1982) and V79 cells (Newbold gt gt., 1980) 6 showed that O -a1kylation of guanine in DNA was strongly correlated with mutation induction, but not with 117 cytotoxicity. Further work will be needed to see if these differences reflect the species difference in the target cells. 6 2. MT Is Spgcific for the Removal of O -MeG Besides 06-MeG, another miscoding adduct formed in DNA by methylating agents is 04—methylthymine. Recently, three laboratories have reported preliminary data indicating that gt ggtt contains a MT protein which acts on 04-methylthymine (McCarthy gt gt., 1983; McCarthy gt gt., 1984; Ahmmed and Laval, 1984). Therefore, in this study it was important to determine whether MT in human cells was specific for the removal of 06-MeG. The experiments in this study indicate that the cell extract with the highest activity of MT (XPlZBE) was active 6 position of guanine but did not on methyl groups at the 0 remove methyl groups from 04-methylthymine, and this indicates that human cell methyltransferase differs from that of E. coli (McCarthy gt gt., 1983, McCarthy gt gt., 6-MeG is 1984, Ahmmed and Laval, 1984). This argues that O the principal mutagenic and cytotoxic lesion induced in human cells by MNNG. We did not test the XP12BE cell extract for ability to remove methyl groups from phosphate triesters, i.e., lesions repaired by the MT protein of g. ggtt (McCarthy gt gt., 1983). Recently it has been shown by Dolan gt gt. (1984) that partially purified preparations of O6-alky1guanine-DNA 118 alkyltransferase from rat liver cannot repair 04-methylthymine in a methylated poly(dT)‘poly(dA) substrate. In contrast, Becker and Montesano (1985) found that rat liver MT repairs 06—MeG as well as 04—methylthymine residues in DNA. These workers used double-stranded poly [d(A-T)°d(A-T)] which was alkylated with tritiated MNU as their substrate. Recently, Richardson gt gt. (1985) investigated the accumulation and removal of 04—methylthymine and O4—ethylthymine in liver DNA from rats exposed to 1,2-dimethylhydrazine or diethylnitrosoamine using a sensitive radioimmune assay to detect these adducts. Their study showed that rat liver efficiently removes 04-methylthymine tg vivo. The contradiction between Dolan t l. (1984) and these investigators might be due to repair of 04-methylthymine in rat liver by enzymes other than MT and/or the inability of rat liver MT to act on the poly(dA)‘poly(dT) substrate. However, recently Pegg has used a DNA substrate and there still was no removal of methyl from 04—methylthymine (personal communication). In addition the persistence of O4-ethylthymine (Muller and Rajewsky, 1983; Swenberg _t gt., 1984) and of alkylphosphate triesters (Margison and O'Connor, 1979) in rat tissues under conditions when O6—alkylguanine is repaired rapidly, suggests that mammalian cell methyltransferases are specific for O6—alky1guanine. 119 3. MT Is in Limited Supply and Can, Therefore, Become Depleted by Low Doses The fact that there was a shoulder on the survival curve and a threshold on the mutation curve for the cells with high levels of MT, but none on the curves of the three cell lines that lack the repair system is further evidence that 06-methylguanine might be a principal cytotoxic and mutagenic lesion. Such a threshold was also found for normal human fibroblasts by Jacobs and Demars (1978). This result is expected if at low doses of MNNG, MT—proficient 6 position cells rapidly remove the methyl groups from the O of guanine before these can exert their potentially harmful effect, but at higher doses, the number of 06—MeG lesions far exceeds the number of repair molecules, so that killing and mutation induction occurs. Our biochemical data support this hypothesis. From the MT activity observed, we estimate that in the human cell fibroblasts that were investigated in this study, the number of molecules of MT per cell, range from less than 1200 for a MT deficient line (XP12ROSV) to 39,000 for a MT proficient cell line (XP12BE). The MT activity we observed was comparable to the work of Harris gt gt. (1983) and Grafstrom gt gt. (1984), who observed similar activities in other human cell lines. In lymphoid cell lines, Harris gt gt. (1983) found that MT proficient cell lines contained as much as 25,000 molecules of MT per cell. Similarly, Grafstrom gt gt. (1984), reported that epithelial cells and 120 fibroblasts contained respectively 26,000 and 29,000 molecules of MT per cell. In keeping with our data, Medcalf and Lawley (1981) showed that at non—toxic or only slightly toxic doses of N—methyl—N-nitrosourea, normal human fibroblasts exhibit 6-methylguanine (half life of l h), but at rapid removal of 0 doses 3— to 4-times higher, removal is extremely slow (half-life > 24 h). A similar rapid rate of removal at low doses of MNNG by a human lymphoblastoid cell line was reported by Sklar and Strauss (1981). 4. The Frequency of MNNG—Induced Mutants Is Higher in Cells Which Are Specifically Depleted of MT Activity The data in this study suggest that 06—methylguanine is the principal cytotoxic and mutagenic lesion induced in human cells by MNNG. However, it is always possible that other DNA repair systems are also operating on the MNNG-induced lesions and these systems influenced the results. Therefore, the MT activity in normal cells was specifically decreased by exposure to exogenous 06—MeG and the effect on cytotoxicity and mutagenicity of MNNG was determined. The results indicated that at low doses, the frequency of TG resistant cells was significantly higher in MT-depleted cells than in the corresponding non-depleted 6 populations. This is the result expected if 0 —MeG is the principal mutagenic lesion. The fact that at high doses of 121 MNNG (5uM), the differences in the mutagenic response of the two populations (non-depleted and depleted) was less 6---MeG is the pronounced also supports the hypothesis that O principal mutagenic lesion. This is because at a dose of 5uM, even the cells with normal levels of MT activity give evidence that the number of potentially mutagenic lesions induced far exceeds the repair capacity. Our results in Table 12 indicating that MT activity of cultured human cells can be greatly reduced by exposure to exogenous 06-MeG substrate in the medium, agree with Dolan gt gt. (1985) who demonstrated that when HeLa cells are exposed to exogenous 06-MeG for 8-24 hours, an 80% reduction in MT activity is observed, indicating that MT activity of cultured human cells can be greatly reduced by exposure to exogenous 06—MeG substrate in the medium. They found, as we did, that increasing the exogenous 06-MeG concentration above 0.2mM or increasing the exposure time beyond four hours did not further deplete the MT activity in the cells. Our work indicates that the regeneration in normal human diploid fibroblasts is 24 hours or less. Dolan gt gt. (1985) also found that the loss of MT activity brought about by exogenous 06-MeG is reversible on removing the substrate, but at least 48 hours are required for complete restoration of the activity in HeLa cells. These data are in keeping with those of Yarosh and coworkers (1984) who found that when human tumor cell lines are treated with MNNG, MT is depleted, but regeneration of the activity is complete in 48 122 hours. In contrast, Karran and Williams (1985) recently showed that the restoration of MT activity in human lymphoid cells is complete in 4 hours. These differences in regeneration times are probably due to the different cell lines used. Dolan _t _t. (1985) also observed that the loss of MT activity is not prevented by the addition of inhibitors of nucleic acid or protein synthesis, suggesting 6 that protein synthesis or the incorporation of O —MeG into nucleic acids is not required for the loss of activity. 5. Survival Is Not Affected When MT Activity Is Depleted Studies comparing the survival of human cell lines constitutively deficient in MT activity with those 6 containing normal levels indicate that O -methylguanine is involved in the cytotoxicity of MNNG (Domoradzki et 1., 1984; Scudiero _t _t., 1984; Harris gt gt., 1983). Yet, our experiments indicate that normal populations with depleted—MT activity were not significantly more sensitive to the killing action of MNNG than non-depleted populations. A possible explanation for this lack of increased sensitivity is that, in contrast to mutation induction by MNNG which is essentially complete within a short period of time following exposure (Figure 11) cell death (loss of ability to form a colony) reflects a process that is not completed until much later and regeneration of MT protein in these human fibroblasts occurred rapidly enough to remove the potentially cytotoxic lesions before their effect was 123 made permanent ("fixed"). Evidence in support of this hypothesis was obtained by measuring the regeneration time of MT activity in depleted populations. Regeneration was found to be rapid and occurred within 24 hours. Therefore, the lesions were probably removed before survival could be affected more than in the control populations. 6 6. Aphidicolin Synchrony to Study Repair of O -MeG Aphidicolin, a tetracyclic diterpenoid mycotoxin, specifically inhibits DNA polymerase alpha but not polymerase beta or gamma. The inhibition of polymerase alpha ifl,Xi££2 correlates positively with the inhibition of DNA replication 12.21X2 (Sugino and Nakayama, 1980). An additional role for polymerase alpha cannot be excluded (Miller and Chinault, 1982; Ciarrocchi gt gt., 1982; Dresler gt gt., 1982). However, there is controversy concerning the effect of aphidicolin on DNA repair synthesis (Seki gt gt., 1982; Giulotto _t gt., 1981; Pedrali-Noy and Spadari, 1980). It is possible to synchronize cells using aphidicolin because polymerase alpha, involved in DNA replication, is blocked. We adapted this method to synchronize diploid human fibroblasts and measured the frequency of MNNG—induced mutants (resistance to TG) at the S phase of the cell cycle when mutations are "fixed" and at the Gl/S border when sufficient time has been allowed for DNA repair. In theory, a synchronous population of normal fibroblasts (normal levels of MT) treated with MNNG, just as the gene for TG 124 resistance is to be replicated, should have virtually no time to repair and the frequency of mutations induced per dose should approximate that observed in a MT deficient strain. When SL68 cells (MT proficient) were treated with MNNG, at the S phase, the frequency of MNNG—induced mutations increased per dose and approximated the frequency seen in a MT deficient strain. Actually, the induced mutation frequency for the MT deficient strain may be two-fold higher than the theoretical yield because mutagenic lesions may well persist in these cells through successive rounds of DNA replication and continue to generate mutations. When SL68 cells were treated at the Gl/S border and allowed 24 hours to repair their DNA before entering the S phase, the frequency of MNNG—induced mutations per dose decreased to a level below that normally seen with SL68 cells. The mutagenicity results indicate that in cells capable of removing lesions, the time available for repair between treatment and the S phase is the important factor in determining the ultimate mutagenicity of MNNG treatment in human cells. In contrast to the mutation data in which DNA synthesis appears to be the critical event which converts DNA damage into mutations, our cytotoxicity data indicate that no single cell cycle-related event, such as DNA synthesis, is critical in translating DNA damage into cell death. The cytotoxicity data in cells treated at S or at the Gl/S 125 border and allowed 24 hours to repair before being allowed to begin cycling are similar. Yet during that time, the potentially mutagenic lesions, presumably 06-MeG, were 6—MeG is also a principal removed by the cells. If 0 potentially cytotoxic lesion, these lesions were removed during that period in aphidicolin. The fact that the survival of the cells held at the Gl/S border was not higher than that of cells not held in aphidicolin is consistent with previous findings of Konze-Thomas gt gt. (1982) and Yang _t gt. (1982). It has been suggested by Konze—Thomas gt gt. (1982) that the amount of time available for repair before cell death is regulated by the cells' need for critical cellular proteins and their respective mRNAs, so that if the DNA template for transcription of these mRNAs is still blocked by lesions at the time the cell has need for them, reproductive death (i.e., inability to form a colony) is the result. Although the kinds of proteins being synthesized by cells that are being held at the Gl/S border undoubtedly differ from those being synthesized by cells in early S, the overall average need for transcription and translation by the two populations may well be comparable. If so, this could explain why the lethal effect of MNNG was comparable in cells treated at the onset of S phase or at the Gl/S border followed by time post-treatment before cell cycling. When synchronous cells are treated with MNNG, a 300% increase occurs in the induced mutation frequency when 126 compared to MNNG treated asynchronous cells. In contrast, the difference in induced mutation frequency for UV treated synchronous and asynchronous cells (Konze-Thomas gt gt.,1982; Maher 22.2l-r 1979) is not as dramatic, only 33%. The large difference between MNNG treated synchronous and asynchronous populations can be attributed to several factors. In the asynchronous population only a small fraction of the cells are progressing through the S phase 6-MeG is occurring. Therefore, the and rapid repair of 0 induced mutation frequency is low. In the synchronous population, the induced mutation frequency is near the theoretical yield probably because the mutations are rapidly put in since the cells are treated when the HPRT gene is being replicated and direct miscoding during replication is very likely taking place. Also in the synchronous population, MNNG doesn't interfere as much with DNA replication as UV does (Grossman, Maher and McCormick; unpublished studies). Therefore, in the MNNG-treated synchronous cells, replication proceeds on the template still containing the 06-MeG lesions and miscoding can take place. All of the above mentioned factors work to increase the frequency of TG resistant mutants. B. Sensitivity to MNNG Is Not a Common Feature of FP and GS Persons with familial polyposis coli and Gardner's syndrome develop multiple polyps of the lower digestive 127 tract which are predisposed to malignancy. A number of investigators have reported that fibroblasts derived from skin biopsies of such FF and GS patients are abnormal in their response to alkylating agents (Hori gt gt., 1980; Miyaki gtht., 1982; Paterson gt gt., 1981; Barfknecht and Little, 1982). Because methylating agents have been implicated in colon carcinogenesis (IARC, 1974), we determined whether fibroblasts from FP and GS patients were abnormally sensitive to mutations induced by such carcinogens. A positive result would support the hypothesis that mutations occurring in the colon epithelial cells of these patients at an abnormally high frequency are a contributing factor in the disease. The results of this study indicate that the predisposition to colon cancer of FP and GS patients is not necessarily correlated with an increased sensitivity of their fibroblasts to mutations induced by methylating carcinogens. The study also indicates that lack of MT is not an essential feature of PP and GS. 1. Lack of MT Is Not an Essential Feature of PP and GS Abnormal increased sensitivity to the mutagenic effects of carcinogens has been shown to be characteristic of skin fibroblasts from XP patients, both classic (Robbins gt gt., 1974; Maher _£._l°r 1977; Yang gt gt., 1980; Arlett and Harcourt, 1983 ) and variant (Maher gt gt., 1976; Myrh gt gt., 1979; Patton gt gt., 1984) and more recently of 128 fibroblasts from persons with hereditary cutaneous malignant melanoma (Howell gt gt., 1984). However, the present study shows that this is not necessarily the case for FP or G8 patients. Skin fibroblasts from only one of the individuals predisposed to colon cancer (GM3314) showed a low level of MT activity and hypersensitivity to MNNG. The cells of an affected daughter of this individual (GM3948) had a high level of MT activity and showed a normal response to killing by MNNG and a somewhat lower than normal response to mutation induction. Middlestadt _t _t. (1985) recently demonstrated that cells from five affected individuals in a GS family each have a different level of MT activity, ranging from 10% to 100% of normal. This argues that the lack of methyltransferase in fibroblasts is not an essential feature of Gardner's syndrome or familial polyposis coli. Similarly, the finding that fibroblasts from an apparently normal fetus (GM0011) are deficient in MT and hypersensitive to MNNG suggests that these characteristics are not restricted to cells from persons with a genetically inherited predisposition to cancer. 2. Comparison of Our Cytotoxicity Data with That of Others Our survival curve data for normal cells agree with that of Paterson _t _t. (1981) who reported that exposure for 1 hour to 7.3uM MNNG in serum-free medium lowered the colony-forming ability of normal cells to 10% of the 129 untreated control (D10)° In our experiment Dlg = 8uM. However, in contrast to our results, they reported a D10 for GM3948 as 2.8uM and for GM3314 as 2.1uM. Our data indicate a D10 of 8uM for GM3948 and extrapolation of our data for GM3314 gives a 010 of 0.6uM. Our results with FP cell line GM2355 agree with the MNNG-induced cytotoxicity data recently reported by Barfknecht and Little (1982). However, they reported that GS strain 2938 was more sensitive than normal cells to MNNG. The difference in results might be accounted for by the low cloning efficiencies in their experiments (l-9%). 3. Possible Mechanism for Hereditary Colon Cancer The underlying mechanism responsible for the increased susceptibility to colon cancer is unknown. However, Lipkin _t _t. (1983) has provided insight into this process. The cellular proliferative zone in the mucosa of the large intestine of humans is located in approximately the basal (lower) three—fourths of the colonic crypts. Epithelial cells migrate toward the gut lumen and are extruded from the mucosal surface between the crypts. In humans, about 10% to 20% of all cells undergo DNA synthesis in the lower three-fourths of the crypt column. In the upper portion of the crypt, cell proliferation decreases, more cells differentiate and proliferation stops as cells near the surface (Lipkin gt gt., 1970). In individuals with a hereditary predisposition to colon cancer, such as FP, an 130 early abnormal characteristic of colonic epithelial cells is cell replication in the upper as well as the lower portions of the crypts. The adenomatous polyps apparently form as a direct result of this abnormal cell replication pattern. Cell cycle studies in rat epithelial cells (Tong gt_gt., 1980) indicate that the frequency of mutations induced by MNNG is higher in cells treated in early S phase than in early G1 or in a non-proliferating state. This is supported by the cell cycle studies described in this dissertation. This S-phase sensitivity was also found in human cells exposed to UV radiation (Konze-Thomas gt gt., 1982) or a reactive metabolite of benzo(a)pyrene (Maher _t _t., 1976 ). These data suggest that the colonic epithelial cells of FP patients should be at a greater risk of being mutated by endogenous or exogenous mutagens than the comparable epithelial cells from normal individuals, even if their rate of DNA repair is normal. These mutant cells can expand to form new clones and undergo additional mutagen—induced events. This could account for the increased sensitivity of FR patients to colon cancer, but does not rule out many other explanations. APPEND IX APPENDIX CLONING THE GENE FOR METHYLTRANSFERASE INTRODUCTION A direct approach to determine whether 06—MeG is the DNA adduct responsible for both the mutagenic and cytotoxic response to MNNG would be to transfer the MT repair gene into a methyltransferase deficient cell line and to determine what effect this has on survival and induced mutations after treatment with MNNG. Although it is known that the response of human cells to mutagens is determined by the proficiency of DNA repair systems, there exists little detailed knowledge of the nature of these systems or the genes that control them. The characterization and isolation of DNA repair genes by recombinant DNA technology is a new approach to the elucidation of repair systems. Rubin t 1., (1983) and Westerveld _t _t., (1984) identified a human DNA repair gene following DNA mediated gene transfer of DNA from human cells proficient in DNA repair into Chinese hamster ovary mutant cells that, like XP cells, are sensitive to a variety of DNA damaging agents and are defective in excision repair. These 131 132 investigators co-transfected the repair deficient (sensitive to mitomycin-C) recipient cells (Chinese hamster ovary cells) with human HeLa cell DNA and the plasmid pSV2-GPT containing the dominant ecogpt gene from the bacterium E; ggtt. After calcium phosphate mediated gene transfer, cells were initially selected for the expression of the ecogpt gene (selectable marker) in growth medium containing mycophenolic acid. Selection of the ecogpt gene provides a means of pre-selecting the fraction of transfection competent recipient cells. The cells were then selected for repair proficiency in medium containing mitomycin-C. Secondary transfections yielded transformants that were proficient in repairing DNA damage, contained the ecogpt gene and contained human DNA sequences. In a similar study, Westerveld and coworkers (1984) identified a human DNA repair gene that complemented the repair defect in a Chinese hamster ovary mutant cell line. To demonstrate the successful integration of human DNA, both studies made use of the fact that the human genome contains a family of highly repetitive sequences (Alu sequences). They were able to distinguish human DNA against a background of CHO DNA by the hybridization technique of Southern (1975). Westerveld and coworkers (1984) were also able to identify the DNA repair gene. They constructed a cosmid recombinant library in g. ggtt from the DNA of the repair proficient transformant and screened the resulting library by colony filter hybridization using a ecogpt probe which 133 was the selectable marker in their system. We reasoned that using an approach similar to the experiments described above, it should be possible to transfer the MT gene from 8 MT repair proficient cell to a MT repair deficient cell. In the DNA mediated gene transfer method, the transformants would be co-selected for resistance to an alkylating agent, such as MNNG, and for resistance to neomycin because the selectable marker gene neo would be utilized in a co—transfection. The cell line XP12ROSV was chosen as the recipient of DNA from a MT-proficient cell line (rat liver cells) because it is deficient in MT, because it has an infinite lifespan, and because it is extremely sensitive to MNNG. A two species system (human, rodent) such as the one used by Rubin ‘gt gt, (1983) was employed, so that after transfection, it would be possible to distinguish the DNA sequences from the transfected DNA from those of the recipient cells by Southern blot analysis of the DNA (Southern, 1975). Figure A-l illustrates the protocol to be used. Rat liver DNA and pSVZ-neo DNA (a selectable marker) are co-transfected into XP12ROSV cells by a calcium phosphate (CaPO4) precipitation procedure (Graham and van der Erb, 1973). The cells are then selected for resistance to MNNG and geneticin (which selects for expression of the neo gene) and the resistant colonies further analyzed for MT activity, for the presence of rat DNA sequences, for continued resistance to MNNG and for the frequency of MNNG induced 134 XP12ROSV CaP04 rat liver DNA (MT proficient) MT deficient ppt. + immortal "< pSV2-neo (selectable MNNG sensitive marker gene) SELECT for MNNG resistance (MNNGR) SELECT for neomycin resistance (neoR) grow cells that are MNNGR and neoR Isolate DNA Assay Check if Probe for fromR R for MT cells rat DNA MNNG neo contain sequences cells. rat MT or against human Use for a 2nd (induced) DNA background. round of human MT. transfection into XP12ROSV cells. Figure A-1. Scheme of MT Cloning Experiment 135 mutations. To isolate the gene, a series of secondary transformations need to be performed in order to isolate cells that are free of most nonessential transforming DNA sequences. MATERIALS AND METHODS Cells and Culture Media XP12ROSV cells were used and the source of these cells is given in Section II along with the culture conditions. Isolation of Rat Liver DNA Nuclei from rat liver cells were isolated initially and then DNA was isolated from the nuclei. Rat liver was homogenized in a Waring blender using a lysing solution which consisted of 2% Titron X—l00, 2mM CaClZ, 2mM M9804, 30 mM KCl, 100 mM NaCl and 50 mM Tris-HCl, pH 7.4. The tissue was homogenized until it was microscopically observed that the nuclei were free of the cell debris. The sample wa centrifuged for 5 minutes (30 xg). The pellet was re-suspended in 50 mM Tris-HCl, pH 7.4, 10 mM EDTA, 100 mM NaCl. Protease K was added to a final concentration of 100 ug/ml and sodium dodecyl sulfate to a final concentration of 0.5%. The sample was incubated at 500C for 6-12 hours. The DNA was extracted with Kirby phenol-chloroform solution until the interface was clear. TlRNase was added at 25 136 units/ml and incubated at 370C for 30 minutes. DNA extraction was repeated in Kirby phenol-chloroform. The DNA was ethanol precipitated, pelleted and re-suspended in buffer consisting of 10 ml EDTA, 100 mM Tris-HCl, and 50 mM NaCl. Isolation of Plasmid pSV2-neo DNA Plasmid DNA was isolated by an alkaline extraction procedure according to the method of Birnboim and Doly (1979). DNA Transfection XP12ROSV monolayers were transfected with rat liver and pSVZ-neo plasmid DNA as a co-precipitate by the method of Sutherland and Bennett (1984). Cells were plated into 150-mm diameter tissue culture dishes at a density of 0.8 x 106 cells per dish. Twenty-four hours later, the culture medium was removed and the cells were washed twice with serum-free medium. The cells were overlayed with 5 m1 of polyethyleneglycol (PEG) solution prepared by adding 20 m1 of serum-free medium to 10 g of melted PEG 6000 (Sigma, St. Louis, MO), which was cooled to 37°C before use. After 2 minutes, the PEG solution was removed and the dishes were washed twice with serum free medium. Growth medium was added (30 ml), plus 0.3 ml of a DNA CaPO4 co-precipitate. This co-precipitate was made by mixing plasmid DNA (13 ug/ml) and rat liver DNA (256 ug/ml) in a total of 27 ml 137 with 0.5 M CaCl2 (27 m1) and 2 x Hepes buffered saline (54 ml; 280 mM NaCl; 10 mM KCl; 1.4 mM NazHPO4; 40 mM Hepes, pH 7.05) at a ratio of 1:1:2. The medium was removed 18 to 24 h later and the cultures were washed twice with serum free medium and refed with fresh growth medium. Seven days following transfection, the cells were harvested and seeded in 100-mm diameter dishes and selected for resistance to MNNG and neomycin. Selection for Resistance to Geneticin For selection with Geneticin (Grand Island Biological Co., Grand Island, NY), the cells were seeded at 4 x 105 cells per l00-mm diameter dish. After allowing 12 to 18 hours for cell attachment, geneticin, dissolved in Hepes pH 7.4, was added to a final concentration of 200 ug/ml of active compound. The cells were fed with growth medium plus geneticin after one week. At two weeks, the transfection frequency was determined by staining several dishes in 1% crystal violet and counting the number of colonies. Selection for Resistance to MNNG Cells were seeded at 4 x 105 cells per 100-mm dish and treated with 2 uM MNNG for one hour. If selecting for MNNG resistant and neo-resistant colonies, then following MNNG treatment, the cells were exposed tg situ to geneticin. 138 RESULTS Initially, two DNA-CaPO4 co—precipitation techniques were compared in order to determine which technique would give a higher transfection frequency. Normal cells (SL68) and also MT deficient cells (XP12ROSV) were transfected with different concentrations of the plasmid pSV2-neo by the suspension method of Chu and Sharp (1981) or the monolayer method of Sutherland and Bennett (1984). The results in Table A-1 indicate that for XP12ROSV, the monolayer method 6 cells gave the highest yield of transfectants; 318 per 10 selected per 10 ug of pSV2-neo. This frequency is comparable to the one obtained by Spivak _t _t. (1984). Co-transfection of XP12ROSV (MT deficient) cells with pSV2-neo and rat liver DNA (MT proficient) by the technique described in the Materials and Methods section of this appendix, resulted in 474 neo resistant colonies per 0.4 x 6 10 cells per dish. Selection in geneticin and MNNG resulted in 53 neo resistant, MNNG resistant clones per 0.4 x 106 cells per dish. A total of 250 dishes were used because the frequency of transformants was expected to be low based on the results of such a transfection protocol by Rubin gt gt. (1983). The high frequency we obtained could be explained if the XP12ROSV cells were not as sensitive to MNNG as was expected. Extrapolation from the survival data shown in 139 Table A-1. Comparison of two DNA-CaPO co-precipitation techniques in SL68 cells (normal) and IP12ROSV (MT deficient) cells8 Cells Method No. of ug No. of total cells pSV2—neo cells clones selected transfected XP12ROSV suspensionb 1 x 10 10 1 x 10 38 XP12ROSV suspension 1 x 106 5 1 x 106 34 XP12ROSV monolayerC 1 x 106 10 1 x 106 318 XP12ROSV monolayer 1 x 106 5 1 x 106 206 . SL68 suspension 1 x 106 10 1 x 106 7 SL68 suspension 1 x 106 5 1 x 106 1 SL68 monolayer 1 x 106 10 1 x 106 4 SL68 monolayer 1 x 106 5 1 x 106 5 aTransfection with pSV2-neo. bChu and Sharp (1981). cDescribed in Methods and Materials section of this appendix. 140 Figure 8 of the Results section indicates that at a dose of 2uM of MNNG, only 0.01% of the cells should survive. However, only doses of up to 0.7uM had been tested previously and the latter dose gave 10% survival. Therefore, the survival curve of XP12ROSV was determined at higher doses of MNNG. Figure A-2 illustrates the survival curve for XP12ROSV at higher doses of MNNG. Instead of an exponential decline, the survival curve showed two components, which is indicative of a mixed population of cells. These results suggested that in the transfection protocol, we had selected for MNNG resistant cells but the latter had not necessarily attained resistance by acquiring the rat MT gene. There was a resistant sub-population of cells in the XP12ROSV cell line even before transfection. Before trying to repeat the cloning experiment, it was necessary to isolate a pure population of MNNG sensitive cells such that a dose of 2uM would yield a 0.01% survival. Therefore, the XP12ROSV cells were seeded into two 96 well microtiter dishes at a density of 1 cell or less per well and allowed to form colonies. Per microtiter dish, 30 clones were obtained. These clones were replicated and their cells were then tested for resistance to MNNG (2uM) by exposing the developing clones at the 200-1000 cell stage. Several of the clones that appeared to be extra sensitive to MNNG were chosen for further study of their survival after MNNG treatment. The tg situ cytotoxicity results of such a 141 Figure A-2. tg situ cytotoxicity of MNNG in the human transformed cell line XP12ROSV and in its derivatives. The cloning efficiencies were 77% for XP12ROSV; 48% for clone B-9; 40% for clone B—21 and 40% for clone A—22. 142 MNNG (IIMth) o I 2 3 4 I l I T IOO 60 4o PERCENT SURVIVAL CLONING ABILITY '0 : : 6 Z 0 E 4 " O I " 2 .. _ O l : : : ‘ A : 0.6 — -_- 04 '_'_ o XPI2ROSV _ ° _ I XP12ROSV Clone B-9 _ O 2 O XP12ROSV Clone B-Zl " A XP12ROSV Clone A-22“ 0.1 l l l I II 143 study are shown in Figure A-2. Clone XP12ROSV, A—22, was found to be the most sensitive. DISCUSSION Use of the clonal line of XP12ROSV, A-22, should make the cloning experiment outlined above feasible. By the method of Rubin _t _t. (1983), the rat MT gene would be isolated but this gene might serve as a probe for identifying the human MT gene. An alternative and more recent method of isolating the human MT gene would be to use the human cosmid library of Lau and Kan (1983, 1984). Their system utilizes a cosmid vector that can shuttle cloned sequences between bacterial and mammalian cells. They constructed a complete human cosmid library. DNA from the total library (containing thymidine kinase) was transfected into mouse L cells (thymidine kinase deficient) by the calcium phosphate method. Transformants (thymidine kinase proficient) were then selected for resistance to HAT medium. As the cosmid vector contains the cohesive ends of the bacteriophage lambda, they were able to retrieve human DNA sequences. Total DNA from the transformants was packaged ifl.li££2 with lysogenic bacterial extracts and used to infect E. ggtt. The sequence of interest (thymidine kinase gene) was then found in one of the resulting cosmids. 144 Human cells that lack the MT with cosmid DNA carrying the neo cosmid DNA would be selected for geneticin. Cells that have also be selected for their ability to gene could be transfected gene. Cells that integrate their ability to grow in integrated a MT gene would grow in MNNG. Cells from MNNG and neo resistant colonies could then be isolated, propagated and the MT gene rescued from the chromosome by the technique of lambda packaging (Hohn, 1979). Such studies are to be carried out in Carcinogeneis laboratory. the near future in the LIST OF REFERENCES LI ST OF REFERENCES Ahmmed, Z. and Laval, J. (1984), Enzymatic repair of OZalkylated thymidine residues in DNA: involvement of a 02-methylthymidine-DNA methyltransferase and an O -methylthymine DNA glycosylase, Biochem. Biophys. Res. Commun., tgg, l-8. Akamatsu, N., Miyaki, M., Suzuki, K., Ono, T. and Sasaki, M.S. (1983), Mechanism of increased susceptibility to 4-nitroquinoline—l-oxide in cultured skin fibroblasts from patients with familial polyposis coli, Mutation Res., tgg, 173-180. Altamirano-Dimas, M., Sklar, R. and Strauss, B. (1979), Selectivity of the excision of alkylation products in a xeroderma pigmentosum-derived lymphoblastoid line, Mutation Res., pg, 197-206. Arlett, C.F. (1977), Mutagenicity testing with V79 Chinese hamster cells, In: Handbook of Mutagenicity Test Procedures, Kilbey, Legator, Nichols, Ramel (eds.), Elsevier Scientific Pub., New York, NY, pp. 175-192. Arlett, C.F. and Harcourt, S.A. (1983), Variation in response to mutagens amongst normal and repair-defective human cells, In: Induced Mutagenesis Molecular Mechanisms and Their Implications for Environmental Protection, Lawrence, C.W. (ed.), Plenum Press,New York, pp. 249-270. Balsiger, R.W. and Montgomery, J.A. (1960), Synthesis of potential anticancer agents. XXV. Preparation of 6-alkoxy— 2-aminopurines, J. Org. Chem., gg, 1573-1575. Barfknecht, T.R. and Little, J.B. (1982), Abnormal sensitivity of skin fibroblasts from familial polyposis patients to DNA alkylating agents, Cancer Res., 3, 1249-1254. Bauer, K.H. (1928), Mutationstheorie der Geschwulst-Entstehung , Julius Springer, Berlin. B cker, R.A. and Montesano, R. (1985), Repair of O -methyldeoxythymidine residues in DNA by mammalian liver extracts, Carcinogenesis, g, 313-317. Birnboim, H.C. and Doly, J. (1979), A rapid alkaline extraction procedure for screening recombinant plasmid DNA, Nucleic Acids Res., 1, 1513-1523. 145 146 Bodell, W.J., Singer, B., Thomas, G.H., and Cleaver, J.E. (1979), Evidence for removal at different rates of O-ethyl pyrimidines and ethylphosphotriesters in two human fibroblast cell lines, Nucleic Acids Res., g, 2819—2829. Bogden, J.M., Eastman, A. and Bres ick, E. (1981), A system in mouse liver for the repair of O -methylguanine lesions in methylated DNA, Nucleic Acids Res., 2, 3089. Boveri, T. (1929), The Origin of Malignant Tumors, Fischer Verlag, Jena (reprinted by Williams and Wflkins, Baltimore). Bradford, M.M. (1976), A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem., 1t, Braun, A.C. (1974), The Biology of Cancer, Addison-Wesley, London. Brennand, J., Saffhill, R. and Fox, M. (1982), The effects of methylated thymidines upon cgltures of V79 cells and the mechanism of incorporation of O -methylthymidine into their DNA, Carcinogenesis, 3, 219—222. Brent, T.P. (1979), Partial purification and characterization of a human 3-methyladenine-DNA glycosylase, Biochemistry, tg, 911. Bgecheler, J. and Kleihues, P. (1977), Excision of O —methylguanine from DNA of various mouse tissues following a single injection of N—methyl—N—nitrosourea, Chem. Biol. Interact., tg, 325—333. Burdette, W.J. (1955), The significance of mutation in relation to the origin of tumours. A review, Cancer Res., 15, 201. Burke, P.G., Lutzner, M.A., Clarke, D.D. and Robbins, J.H. (1971), Ultraviolet—stimulated thymidine incorporation in xeroderma pigmentosum lymphocytes, J. Lab. Clin. Med., 11, 759-767. Cairns, J., Robins, P., Sedgwick, B. and Talmud, P. (1981), The inducible repair of alkylated DNA, Prog. Nucleic Acid RES. M010 BiOlo, _2_6_’ 237-2440 147 Cathcart, R. and Goldthwait, D.A. (1981), Enzymatic excision of 3-methyladenine and 7-methylguanine by rat liver nuclear fraction, Biochemistry, 3g, 273. Chu, E.H.Y. and Powell, 8.8. (1976), Selective systems in somatic cell genetics, In: Advances in Human Genetics, H. Harrie and K. Hirschorn (eds.), Vol. 7, Plenum Pub. Corp., New York, NY, pp. 189-258. Chu, G. and Sharp, P.A. (1981), SV40 DNA transfection of cells in suspension: analysis of the efficiency of transcription and translation of T-antigen, Gene, tt, 197-202. Ciarrocchi, G., Jose, J.G. and Linn, S. (1979), Further characterization of a cell—free system for measuring replicative and repair DNA synthesis with cultured human fibroblasts and evidence for the involvement of DNA E4 polymerase alpha in DNA, Nucleic Acids Res., 1, 1205—1219. Cleaver, J.E. (1969a), Defective repair replication of DNA in xeroderma pigmentosum, Nature, 218, 652—656. Cleaver, J.E. (1969b), Xeroderma pigmentosum: A human disease in which an initial stage of DNA repair is defective, Proc. Natl. Acad. Sci. U.S.A., gt 428-435. Cleaver, J.E. (1974), Repair processes for photochemical damage in mammalian cells., Adv. Radiat. Biol., t, 1-75. Cleaver, J.E. and Bootsma, D. (1975), Xeroderma pigmentosum: Biochemical and genetic characteristics, Ann. Rev. Genet., 9, 19-380 Corsaro, C.M. and Migeon, B.R. (1977), Comparison of contact—mediated communication in normal and transformed human cells in culture, Proc. Natl. Acad. Sci. U.S.A., 1t, 4476-4480. Couch, D.B. and Hsie, A.W. (1978), Mutagenicity and cytotoxicity of congeners of two classes of nitroso compounds in Chinese hamster ovary cells, Mutation Res., 21, 209-216. Couch, D.B., Forbes, N.L. and Hsie, A.W. (1978), Comparative mutagenicity of alkylsulfate and alkanesulfonate derivatives in Chinese hamster ovary cells, Mutation Res., g1, 217-224. Craddock, V.M. (1971), Liver carcinomas induced in rats by single administration of dimethylnitrosamine after partial hepatectomy, J. Natl. Cancer Inst., 31, 889—907. 148 Craddock, V.M. and Henderson, A.R. (1982), The activity of 3—methyladenine DNA glycosylase in animal tissues, Carcinogenesis, t, 747. Craddock, V.M., Henderson, A.R. and Gash, 86 (1982), Nature of the constitutive and induced mammalian O -methylguanine DNA repair enzyme, Biochem. Biophys. Res. Commun., 107, 546. Day, R.S. and Ziolkowski, C.H.J. (1979), Human brain tumor cell strains with deficient host—cell reactivation of N-methyl—N'-nitro-N—nitrosoguanidine— damaged adenovirus 5, Nature, 212, 797-799. Day, R.S., Ziolkowski, C.H.J., Scudiero, D.A., Meyer, S.A., Lubiniecki, A.S., Girardi, A.J., Galloway, S.M. and Bynum, G.D. (1980), Defective repair of alkylated DNA by human tumour and SV40 transformed human cellure, Egg, 724-727. Demple, B., Jacobsson, A., Olsson, M., Robins, P. and Lindahl, T. (1982), Repair of alkylated DNA in Escherichia coli, Physical properties of O -methylguanine-DNA methyl transferase, J. Biol. Chem., ggz, 13776. Dickson, E. and Robertson, H.D. (1976), Potential regulatory roles for RNA in cellular development, Cancer Res., gg, 3387. Dodson, L.A., Foote, R.S., Mitra, S. and Masker, W.E. (1982), Mutagenesig of bacteriophage T7 in Vitro by incorporation of O —methylguanine during DNA synthesis, Proc. Natl. Acad. Sci., 12, 7660. Dolan, M.E., Scicchitano, D., Singer, B. and Pegg, A.E. (1984), Comparison of repair of methylated pyrimidines in poly(dT) by extracts from rat liver and Escherichia coli, Biochem. Biophys. Res. Commun., ttt, 324-336. Domoradzki, J., Pegg, A.E., Dolan, M.E., Maher, V.M. and M Cormick, J.J. (1984), Correlation between 0 -methylguanine-DNA methyltransferase activity and resistance of human cells to the cytotoxic and mutagenic effect of N—methyl—N'-nitro-N-nitrosoguanidine, Carcinogenesis, g, 1641-1647. Drake, J.W. and Baltz, R.H. (1976), The biochemistry of mutagenesis, Ann. Rev. Biochem., 35, 11—37. Dresler, S.L., Roberts, J.D. and Lieberman, N.W. (1982), Characterization of deoxyribonucleic acid repair synthesis in permeable human fibroblasts, Biochem. gt, 2557-2564. 149 Durrant, L.G., Margison, G.P. and Boyle, J.M. (1981), Pretreatment of Chinese hamster V79 cells with MNU increases survival without affecting DNA repair or mutagenicity, Carcinogenesis, t, 55—60. Edenberg, H.J. (1976), Inhibition of DNA replication by ultraviolet light, Biophys. J. tg, 849-860. Evensen, G. and Seeberg, E. (1982), Adaptation to alkylation resistance involves the induction of a DNA glycosylase, Foot , R.S., Mitra, S. and Pal, B.C. (1980), Demethylation of O -methylguanine in a synthetic DNA polymer by an inducible activity in Escherichia coli, Biochem. Biophys. Res. Commun., 21, 654-659. Fgote,R and Mitra, S. (1984), Lack of induction of O-methylguanine- DNA methyltransferase in mammalian cells treated with N-methy1-N'-nitro-N——nitrosoguanidine, Carcinogenesis, g, 277-281. Fox, M. and Brennand, J. (1980), Evkdence for the involvement of lesions other than 0 -alkylguanine in mammalian cell mutagenesis, Carcinogenesis, t, 795-799. Frei, J.V., Swenson, D.H., Warren, W. and Lawley, P. (1978), Alkylation of deoxyribonucleic acid in vivo in various organs of C57BL mice by the carcinogens N-methyl—N—nitrosourea, N-ethyl-N-nitrosourea and ethyl methanesulphonate in relation to induction of thymic lymphoma. Some applications of high—pressure liquid chromatography, Biochem. J., tZt, 1031-1044. Friedberg, E.C., Anderson, C., Bonura, T., Cone, R. and Simmons, R. (1978), Base excision repair, In: DNA Repair Mechanisms, P.C. Hanawalt, E.C. Friedberg and C.F. Fox, (eds.), Academic Press, New York, NY. Fujiwara, Y., Tatsumi, M. and Sasaki, M.S. (1977), Cross—link repair in human cells and its possible defect in Fanconi's anemia cells, J. Mol. Biol. 113, 635-649. Gallagher, P.E. and Brent, T.P. (1982), 3-Methyladenine—DNA glycosylase from human placenta, Biochemistry, tt, 6404. Gardner, E. J. and Richard, R. C. (1953), Multiple cutaneous and subcutaneous lesions occurring simultaneously with hereditary polyposis and osteomatosis, Am. J. Hum. Genet., t, 139-148. 150 Ggrchman, L.L. and Ludlum, D.B. (1982), The Properties of O -methylguanine in templates for RNA polymerase, Biochim. Biophys. Acta., 308, 310-316. German, J. (1972), Genes which increase chromosomal instability in somatic cells and predispose to cancer, Prog. Giulotto, E. and Mondello, C. (1981), Aphidicolin does not inhibit the repair synthesis of mitotic chromosomes, Biochem. Biophys. Res. Commun., 22, 1287-1294. thh, R. and Rajewsky, M.F. (1974), Persistence of O -ethylguanine in rat brain DNA: Correlation with nervous system-specific carcinogenesis by ethylnitrosourea, Proc. Natl. Acad. Sci. USA, 1t, 639-643. Goth, R. and Rajewsky, M.F. (1974), Molecular and cellular mechanisms associated with pulse-carcinogenesis in the rat nervous system by ethylnitrosourea: Ethylation of nucleic acids and elimination rates of ethylated bases from the DNA of different tissues, Z. Krebsforsch., gt, 37-64. Goth-Goldstein, R. (1977), Repair of DNA damaged by alkylating carcinogens is defective in xeroderma pigmentosum-derived fibroblasts, Nature, 267, 81-82. Goth-Goldstein, R. (198%), Inability of Chinese hamster ovary cells to excise O -alkylguanine, Cancer Res., 32, 2623-2624. Grafstrom, R.C., Pegg, A.E., Trump, B.F. and Harris, C.C. (1984), Alkylguanine-DNA alkyltransferase activity in normal human tissues and cells, Cancer Res., gt, 2855-2857. Grafstrom, R.C., Pegg, .E. and Harris, C.C. (1983) Formaldehyde inhibits 0 -methyl excision repair in human bronchial cells. AACR. abstracts, 71. Graham, F.L. and van der Eb, A.J. (1973), A new technique for the assay of infectivity of human adenovirus 5 DNA, Grossman, A., Maher, V.M. and McCormick, J.J. (1985), The frequency of mutants in human fibroblasts UV—irradiated at various times during S-phase suggests that genes for thioguanine and diptheria toxin resistance are replicated early, Mutat. Res. (in press). Grossman, L. (1978), Nucleotide excision repair, In: DNA Repair Mechanisms, P.C. Hanawalt, E.C. Friedberg and C.F. Fox (eds.), Academic Press, New York, NY. 151 Gupta, R.S. and Siminovitch, L. (1978), Diphtheria toxin resistant mutants of CHO cells affected in protein synthesis: A novel phenotype, Somat. Cell Genet., 1, 553-571. Hanawalt, P.C., Friedberg, E.C. and Fox, C.F. (1978), (eds.), DNA Rgpair Mechanisms, Academic Press, New York, NY. Hanawalt, P.C., Cooper, P.K., Ganesan, A.K. and Smith, C.A. (1979), DNA repair in bacteria and mammalian cells, Ann. Rev. Biochem., gt, 783-836. H rris, A.L., Karran, P. and Lindahl, T. (1983), O -Methylguanine-DNA methyltransferase of human lymphoid cells: structural and kinetic properties and absence in repair-deficient cells, Cancer Res., gt, 3247-3252. Harris, G., Lawley, P.D. and Olsen, I. (1981), Mode of action of methylating carcinogens: comparative studies of murine and human cells, Carcinogenesis, t, 403. Heflich, R.H., Hazard, R.M., Lommel, L., Scribner, J.D., Maher, V.M. and McCormick, J.J. (1980), A comparison of the DNA binding, cytotoxicity, and repair synthesis induced by human fibroblasts by reactive derivatives of aromatic amide carcinogens, Chem. Biol. Interactions, tt, 43-56. Heflich, R.H., Beranek, D.T., Kodell, R.L. and Morris, S.M. (1982), Induction of mutations and sister-chromatid exchanges in Chinese hamster ovary cells by alkylating agents. Relationship to specific DNA adducts, Mutat. Res., tgg, 147-161. Heim, S. (1983), tg Vitro growth characterization of skin fibroblasts from patients with adenomatosis of the colon and rectum and their relatives, Clin. Genet., tt, 41—48. Higgins, N.P., Kato, K., and Strauss, B., (1976), A model for replication repair in mammalian cells, J. Mol. Biol., 101, 417-425. Hodgson, R., Swann, P., Clothier, R. and B 115, M. (1980), The persistence in Xenopus laevis DNA of O -methylguanine produced by exposure to N-methyl-N-nitrosourea, Eur. J. Cancer, tg, 481-486. Hohn, B. (1979), tg vitro packaging of lambda and cosmid DNA, Methods Enzymol., gt, 299. Hori, T., Murato, M. and Utsunomija, J. (1980), Chromosome abberations induced by N-methyl-N'-nitro-N-nitrosoguanidine in cultured skin fibroblasts from patients with adenomatosis coli, Gann, 1t, 628-633. 152 Howell, J.N., Greene, M.H., Corner, R.C., Maher, V.M. and McCormick, J.J. (1984), Fibroblasts from patients with hereditary cutaneous malignant melanoma are abnormally sensitive to the mutagenic effect of simulated sunlight and 4-nitroquinoline-l-oxide, Proc. Natl. Acad. Sci. U.S.A., gt, 1179-1183. IARC (1974), Monographs on the Evaluation of Carcinogenic Risk of Chemicals to Man, Vol. 4, International Agency for Research on Cancer, Lyon, pp. 145—151. Jacobs, L. and DeMars, R. (1978), Quantification of chemical mutagenesis in diploid human fibroblasts: induction of azaguanine-resistant mutants by MNNG, Mutat. Res., gt, 29-53 0 Jenssen, D. and Ramel, C. (1980), Relationship between I chemical damage of DNA and mutations in mammalian cells, ‘ Mutation Res., 1t, 339-347. Kaina, B. (1982), Enhanced survival and reduced mutation and aberration frequencies induced in V79 Chinese hamster cells pre-exposed to low levels of methylating agents, Mutation Res., gt, 195-211. Kaplan, S.H. (1959), Some implications of indirect induction mechanisms in carcinogenesis: A review, Cancer Res., tt, 791. Karran, P., Lindahl, T. and Griffin, B. (1979), Adaptive rgsponse to alkylating agents involves alteration in-situ of O -methylguanine residues in DNA, Nature, 280, 76-77. Karran, P., Arlett, C.F. and Broughton, B.C. (1982a), An adaptive response to the cytotoxic effects of N-methyl-N-nitrosourea is apparently absent in normal human fibroblasts, Biochimie, gt, 717-721. Karran, P., Hjelmgren, T. and Lindahl, T. (1982b), Induction of a DNA glycosylase for N-methylated purines is part of the adaptive response to alkylating agents, Nature, 296, 770-773. Karran, P. and Williams, S.A. (1985), The cytotoxic and mutagenic effects of alkylating agents on human lymphoid cells are caused by different DNA lesions, Carcinogenesis, g, 789-792. Kleihues, P., Magee, P.N., Austoker, J., Cox, D. and Mathias, A.P. (1973), Reaction of N-methyl-N—nitrosourea with DNA of neuronal and glial cells in vivo, FEBS Lett., 'tt, 105-108. 153 Kleihues, P., Bamborschke, S. and Doerjer, G. (1980), Persistence of alkylated DNA bases in the Mongolian gerbil (Meriones unguiculatus) following a single dose of methylnitrosourea, Carcinogenesis, t, 111-113. Kleihues, P., Patzschke, K. and Doerjer, G. (1982), DNA modification and repair in the experimental induction of nervous system tumors by chemical carcingens, NY Acad. Sci., tgt, 290-303. Kleijer, W.J., de Weerd-Kastelein, E.A., Sluyter, M.L., Keijzer, W., de Wit, J. and Bootsma, D. (1973), UV-induced DNA repair synthesis in cells of patients with different forms of xeroderma pigmentosum and of heterozygotes, Mutat. Res., 3g, 417-428. Konze-Thomas, B., Hazard, R.M., Maher, V.M. and McCormick, J.J. (1982), Extent of excision repair before DNA synthesis determines the mutagenic but not the lethal effect of UV radiation, Mutat. Res., gt, 421—434. Kopelovich, L. (1977), Phenotypic markers in human skin fibroblasts as possible diagnostic indices of hereditary adenomatosis of the colon and rectum, Cancer, tg, 2534-2541. Kopelovich, L., Conlon, S. and Pollack, R. (1977), Defective organization of actin in cultured skin fibroblasts, Proc. Natl. Acad. SCio U080A0’ E, 3019-3g220 Kopelovich, L., Bias, N.E. and Helson, L. (1979), Tumor promoter alone induces neoplastic transformation of fibroblasts from humans genetically predisposed to cancer, Nature, tgt, 619-621. Kraemer, K.H. (1977), Progressive degenerative diseases associated with defective DNA repair: Xeroderma pigmentosum and ataxia telangiectasia, In: Cellular Senescence and Somatic Cell Genetics: DNA Repair Processes, W.W. Nichols and D.G. Murphy (eds.), Symposia Specialists, Miami, FL, pp. 37-71. Lau, Y.-F. and Kan, Y.W. (1983) Versatile cosmid vectors for the isolation, expression, and rescue of gene sequences: studies with the human-globin gene cluster. Proc. Natl. Acad. Sci. USA, gt, 5225-5229. Lau, Y.-F. and Kan, Y.W. (1984) Direct isolation of the functional human thymidine kinase gene with a cosmid shuttle vector. Proc. Natl. Acac. Sci. USA, gt, 414-418. 154 Laval, F. and Laval, J. (1984), Adaptive response in mammalian cells: Cross reactivity of different pretreatments on cytotoxicity as contrasted to mutagenicity, Proc. Natl. Acad. Sci. U.S.A., gt, 1062-1066. Lawley, P.D. and Thatcher, C.J. (1970), Methylation of deoxyribonucleic acid in cultured mammalian cells by N—methyl-N'-nitro-N-nitrosoguanidine. The influence of cellular thiol concentration on the extent of methylation and the 6-oxygen atom of guanine as a site of methylation, Biochem. J., ttg, 693-707. Lawley, P.D. (1974), Some chemical aspects of dose-response relationships in alkylation mutagenesis, Mutation Res., gt, 283-295. Lawley, P.D. and Martin, C.N. (1975), Molecular mechanisms in alkylation mutagenesis: induced reversion of bacteriophage TrII AP72 by ethyl methanesulphonate in relation to extent and mode of ethylation of purines in bacteriophage DNA, Biochem. J., ttg, 85. Lawley, P.D. (1976), Carcinogenesis by alkylating agents, In: Chemical Carcinogens, C.E. Searle (ed.), American Chemical Society Monograph, ttt, American Chemical Society, Washington, DC, pp. 83-244. Lehmann, A.R. (1974), Postreplication repair of DNA in mammalian cells, Life Sci., tg, 2005-2016. Lehmann, A.R. and Karran, P. (1981), DNA repair, In: Int. Rev. Cytol., 1g, G.H. Bourne and J.F. Danielli (eds.), Academic Press, New York, NY, pp. 101-146. Lemaitre, M., Renard, A. and Verly, W.G. (19%2), A common chromatin factor involved in the repair of O -methylguanine and O -ethylguanine lesions in DNA, FEBS Lett., 144, 242. Lewi , J.G. and Swenberg, J.A. (1980), Differential repair of O -methylguanine in DNA of rat hepatocytes and nonparenchymal cells, Nature (London), 228, 185-187. Likhachev, A.J., Ivanov, M.N., Bresil, H. Planche-Martel, G., Montesano, R. and Margison, G.P. (1983), Carcinogenicity of single doses of N-nitroso-N—methylurea and N-nitroso-N-ethylurea in Syrian golden hamsters and the persistence of alkylated purines in the DNA of various tissues, Cancer Res., tt, 829-833. Lindahl, T. (1979), DNA glycosylases, endonucleases for apurinic/apyrimidinic sites, and base excision repair, Prog. Nucleic Acid Res. Mol. Biol., gg, 135-192. 155 Lindahl, T. (1982), DNA repair enzymes, Ann. Rev. Biochem., gt, 61. Lindahl, T., Demple, B. and R bins, P. (1982), Suicide inactivation of the g. coli 0 -methylguanine-DNA methyltransferase, EMBO, t, 1359-1368. Lipkin, M., Bell, B., Stalder, G. and Troncale, F. (1970), The development of abnormalities of growth in colonic epithelial cells of man. In: Carcinoma of the Colon and Antecedent Epithelium, H. Burdette, ed., Charles C. Thomas, Springfield, IL, pp. 213-221. Lipkin, M., Friedman, E., Higgins, P. and Augenlicht, L. (1983), Rodent and human models for the analysis of intestinal tumor development, In: Milman, H.A. and Sell, S. (eds.), Application of Biological Markers to Carcinogen Testing, Plenum Press, New York, pp. 61-89. Little, J.B., Nove, J. and Weichselbaum, R.R. (1980), Abnormal sensitivity of diploid skin fibroblasts from a family with Gardner's syndrome to the lethal effects of X-irradiation, ultraviolet light and mitomycin-C, Mutation Res., 1t, 241—250. Loeb, L.A., Springgate, C.F., and Battula, N. (1974), Errors in DNA replication as a basis of malignant changes, Cancer Res., tt, 2311. Loveless, A. (1969), Possible relevance of O6 alkylation of deoxyguanosine to the mutagenicity and carcinogenicity of nitrosamines and nitrosamides, Nature (London), ggt, 206-208. McCann, J., Choi, B., Yamasaki, E. and Ames, B.N. (1975), The detection of carcinogens as mutagens in the Salmonella microsome test: assay of 300 chemicals, Proc. Natl. Acad. Sci. U.S.A., 1g, 5135-5139. McCarthy, J.G., Edington, B.V. and Schendel, P.E. (1983), Inducible repair of phosphotriesters in Escherichia coli, Proc. Natl. Acad. Sci. U.S.A., gt, 7380-7384. McCarthy, T.V., Karran, P. and Lindahl, T. (1984), Inducible repair of O-alkylated DNA pyrimidines in Escherichia coli, EMBO J., 2, 545-5590 McCormick, J.J. and Maher, V.M. (1981), Measurement of colony-forming ability and mutagenesis in diploid human cells, In: Friedberg, E.C. and Hanawalt, P.C. (eds.), DNA Repair, A Laboratory Manual of Research Procedures, Vol. 2, Marcel Dekker, New York, pp. 501-521. 156 McKusick, V.A. (1962), Genetic factors in intestinal polyposis, J. Am. Med. Assoc., 182, 271-277. Magee, P.N. and Farber, E. (1962), Toxic liver injury and carcinogenesis. Methylation of rat-liver nucleic acids by dimethylnitrosamine in vivo, Biochem. J., gt, 114-124. Maher, V.M., Curren, C.D., Ouellette, L.M. and McCormick, J.J. (1976), Effect of DNA repair on the frequency of mutations induced in human cells by ultraviolet irradiation and by chemical carcinogens, In: Fundamentals in Cancer Prevention, P.N. Magee gt gt., (eds.), Univ. Park Press, Baltimore, MD, pp. 363-382. Maher, V.M. and McCormick, J.J. (1976), Effect of DNA repair on the cytotoxicity and mutagenicity of UV irradiation and of chemical carcinogens in normal and xeroderma pigmentosum — cels, In: Biology of Radiation Carcinogenesis, Yuhas, J.M., I Tennant, R.W., and Regan, J.D. (eds.), Raven Press, New York, NY, p. 129. Maher, V.M., McCormick, J.J., Grover, P.L. and Sims, P. (1977), Effect of DNA repair on the cytotoxicity and mutagenicity of polycyclic hydrocarbon derivatives in normal and xeroderma pigmentosum human fibroblasts, Mutation Res., .tt, 117-138. Maher, V.M., Dorney, D.J., Mendrala, A.L., Konze—Thomas, B. and McCormick, J.J. (1979), DNA excision repair processes in human cells can eliminate the cytotoxic and mutagenic consequences of ultraviolet irradiation, Mutat. Res., _g, 311-323. Mamet-Bratley, M.D. and Karska-Wysocki, B. (1982), Role of 3-methyladenine-DNA glycosylase in host—cell reactivation of methylated T7 bacteriophage, Biochim. Biophys. Acta, 698, 29. Margison, G.P. and Kleihues, P. (1975), Chemical carcinogenesis in the nervous system. Preferential accumulation of O -methylguanine in rat brain DNA during repetitive administration of N-methyl-N-nitrosourea, Biochem. J., ttg, 521. Margison, G.P. and O'Connor, P.J. (1979), Nucleic acid modification by N-nitroso compounds, In: Chemical Carcinogens and DNA, Grover, P.L. (ed.), CRC Press, Boca Raton, FL, pp. 112-149. Margison, G.P. and Pegg, A.E. (1981), Enzymatic release of 7-methylguanine from methylated DNA by rodent liver extracts, Proc. Natl. Acad. Sci., 1g, 861. 157 Mgdcalf, A.S.C. and Lawley, P.D. (1981), Time course of O -methylguanine removal from DNA of N-methyl-N-nitroso- urea-treated human fibroblasts, Nature, 289, 796-799. Medcalf, A.S.C. and Wade, M.H. (1983), Comparison of mutagenicity of N-methyl-N-nitrosourea and N-ethyl-N-nitrosourea in human diploid fibroblasts, Carcinogenesis, t, 115-118. Mehta, J.R., Ludlum6 D.B., Renard, A. and Verly, W.G. (1981), Repair of O -ethylguanine in DNA by a chromatin fraction from rat liver: transfer of the ethyl group to an acceptor protein, Proc. Natl. Acad. Sci. U.S.A., 1g, 6766-6770. Middlestadt, M.V., Norton, G. and Paterson, M.C. (1984), Absence of O -methylguanine-DNA methyltransferase activity in a nontransformed human fetal fibroblast line, Abstract, 15th Annual Meeting, Environmental Mutagen Society, p. 114. Middlestadt, M.V., McWilli ms, R., Mirzayans, R., Norton, G. and Paterson, M. (1985), O -Methylguanine repair in human fibroblasts, J. Cell. Biochem., abstracts, p. 47. Miller, J.A. (1970), Carcinogenesis by Chemicals: An Overview, G.H.A. Clowes Memorial Lecture, Cancer Res., tt, Miller, E.C. and Miller, J.A. (1971), The mutagenicity of chemical carcinogens: correlations, problems, and interpretations, In: Chemical Mutagens- Principles and Methods for Their Detection, A. Hollaender (ed.), Vol. I, Plenum Press, New York, NY, pp. 83-119. Miller, E.C. and Miller, J.A. (1976), The metabolism of chemical carcinogens to reactive electrophiles and their possible mechanisms of action in carcinogenesis, In: Chemical Carcinogens, C.E. Searle (ed.), American Chemical Society Monograph, No. ttt, American Chemical Society, Washington, DC, pp. 737-762. Miller, M.R. and Chinault, D.N. (1982), Evidence that DNA polymerases alpha and beta participate differentially in DNA repair synthesis induced by different agents, J. Biol. Chem., ggt, 46-49. Mgtra, S., Pal, B.C. and Foote, R.S. (1982), O -methylguanine-DNA methyl transferase in wild-type and ada mutants of Escherichia coli, J. Bacteriology, 152, 534. 158 Miyaki, M., Akamatsu, N. and Hirono, U. (1980), Characteristics of cultured fibroblasts of skin from patients with adenomatosis coli and Peutz-Jeghers' syndrome, In: Gann Monograph on Cancer Research, T. Nagayo and W. Mori, (eds.), Tokyo, Japan Scientific Societies Press, pp. 115-125. Miyaki, M., Akamatsu, N., Ono, T., Tonomura, A. and Utsunomiya, J. (1982), Morphologic transformation and chromosomal changes induced by chemical carcinogens in skin fibroblasts from patients with familial adenomatosis coli, J. Natl. Cancer Inst., t, 563-571. Moehring, J.M. and Moehring, T.J. (1979), Characterization of the diphtheria toxin resistance system in Chinese hamster ovary cells, Somat. Cell Genet., g, 453-468. Montesano, R. and Bartsch, H. (1976), Mutagenic and carcinogenic N-nitroso compounds: possible environmental hazards, Mutation Res., tg, 179. Montesano, R., Bartsch, H. and Tomatis, L. (1976), (ed.), Screening tests in chemical carcinogenesis, IARC Scientific Publications, 12, 661, International Agency for Research on Cancer (IARC), Lyon. Montesano, R., Bresil, G., Planche-Martel, G., Margison, G.P. and Pegg, A.E. (1980), Effect of chronic tregtment of rats with dimethylnitrosamine on the removal of 0 -methyl- guanine from DNA, Cancer Res., tt, 452. Montesano, R. (1981), Alkylation of DNA and tissue specificity in nitrosamine carcinogenesis, J. Supramol. Struct. Cell Biochem., t1, 259-273. Montesano, R., Bresil, H., Drevon, C. and Piccoli, C. (1982), DNA repair in mammalian cells exposed to multiple doses of alkylating agents, Biochimie, 8-9, 591. Morris, S.M., Heflich, R.H., Beranek, D.T. and Kodell, R.L. (1982), Alkylation-induced sister-chromatid exchanges correlate with reduced cell survival, not mutations, Mutation Res., ttg, 163-168. Mgller, R. and Rajewsky, M.F. (1983),4Enzymatic removal of O -ethylguanine versus stability of O -ethylthymine in the DNA of rat tissues exposed to the carcinogen ethylnitrosourea: possible interference of guanine-O6 alkylation with 5-cytosine methylation in the DNA of replicating target cells, Z. Naturforsch., tg, 1023—1029. 159 Myhr, B.C., Turnbull, D. and DiPaolo, J.A. (1979), Ultraviolet mutagenesis of normal and xeroderma pigmentosum variant human fibroblasts, Mutation Res., gg, 341-353. Myrnes, B., Giercksky, K.E. and Krokan, H. (1982), Repair of O -methylguanine residues in DNA takes place by a similar mechanism in extracts from HeLa cells, human liver, and rat liver, J. Cell. Biochem., gt, 381-392. Nagao, M., Sugimura, T. and Matsushima, T. (1978), Environmental mutagens and carcinogens, Ann. Rev. Genet. 12, 117. Newbold, R.F., Warren, W., Medcalf, A.S.C. and Amos, J. (1980), Mutagenicity of carcinogenic methylating agents is associated with a specific DNA modification, Nature, 283, 596-599. Nicoll, J.W., Swann, P.E. and Pegg, A.E. (1975), Effect of dimethylnitrosamine on persistence of methylated guanines in rat liver and kidney DNA, Nature (London), 254, 261—262. O'Connor, P.J., Cooper, D.P., Smith, R.A. and Margison, G.P. ( 982), Species differences in the inducibility of hepatic O -alkylguanine repair in rodents, Biochimie, gt, 769-773. Olsson, M. and Lindahl, T. (1980), Repair of alkylated DNA 1 Escherichia coli: methyl group transfer from O -methylguanine to a protein cysteine residue, J. Biol. Chem., ggg, 10569-10571. Pappenheimer, A.M. (1977), Diphtheria toxin, Ann. Rev. Paterson, M.C. and Smith, B.P. (1979), Ataxia-telangiectasia: an inherited human disorder involving hypersensitivity to ionizing radiation and related DNA-damaging chemicals, Ann. Rev. Genet. tt, 291-318. Paterson, M.C., Smith, B.P., Krush, A.J. and McKeen, E.A. (1981), In vitro hypersensitivity to N-methyl-N'-nitro-N-nitrosoguanidine in a Gardner syndrome family, Radiat. Res., gt, 483. Patton, J.D., Rowan, L.A., Mendrala, A.M., Howell, J.N., Maher, V.M. and McCormick, J.J. (1984), Xeroderma pigmentosum fibroblasts including cells from XP variants are abnormally sensitive to the mutagenic and cytotoxic action of broad spectrum simulated sunlight, Photochem. Photobiol. tg, 37-42. 160 Pedrali-Noy, G. and Spadari, S. (1979), Effect of aphidicolin on Viral and human DNA polymerases, Biochem. Biophys. Res. Commun. gg, 1194-2002. Pedrali-Noy, G. and Spadari, S. (1980), Aphidicolin allows a rapid and simple evaluation of DNA-repair synthesis in damaged human cells, Mutation Res., tt, 389-394. Pedrali-Noy, G. Spadari, S., Miller-Faures, A., Miller, A.D.A., Kruppa, J. and Koch, G. (1980), Synchronization of HeLa cell cultures by inhibition of DNA polymerase with aphidicolin, Nucleic Acids Res. g, 377-387. Pegg, A.E. (1977), Formation and metabolism of alkylated nucleosides: Possible role in carcinogenesis by nitroso compounds and alkylating agents, Adv. Cancer Res., gg, 195-269. Pegg, A.E. (1978), Enzymatic removal of 06-methylguanine I from DNA by mammalian cell extracts, Biochem. Biophys. Res. Commun., gt, 166. Pegg, A.E. an Balog, B. (1979), Formation and subsequent excision of O -ethylguanine from DNA of rat liver following administration of diethylnitrosamine, Cancer Res., tg, 5003-5010. Pegg, A.E. and Perry, W. (1981a), Alkylation of nucleic acids and metabolism of small doses of dimethylnitrosamine in the rat, Cancer Res., tt, 3128-3132. Pegg, A.E. and Perry6 W. (1981b), Stimulation of transfer of methyl groups from O -methylguanine in DNA to protein by rat liver extracts in response to hepatotoxins, Carcinogenesis, 2, 1195. Pegg., A.E., Perry, W. and Bennett, R.A. (1981), Effect of partial hepatectomy on removal of O -methylguanine from alkylated DNA by rat liver extracts, Biochem. J., 197, 195. Pegg, A.E. (1982), Repair of alkylated DNA by cell extracts from various organs and species, In: Organ and Species Specificity in Chemical Carcinogenesis, R.J. Langenbach, S. Nesnow and J.M. Rice (eds.), New York, Plenum Press, p. 545. Pegg, A.E., Roberfoid, M., von Bahr, C., Foote, R.S., Mitra, S., Bresil, H., Likha hev, A. and Montesano, R. (1982), Removal from DNA of O -methylguanine by human liver fractions, Proc. Natl. Acad. Sci., tg, 5162-5165. P 99, A.E. and Wiest, L. (1983), Regulation of O —methylguanine-DNA transmethylase activity in rat liver and kidney, Cancer Res., tt, 972-975. 161 Pegg, A.E., Wiest, L., Foote, R.S., Mitr , S. and Perry, W. (1983), Purification and properties of O -methylguanine-DNA transmethylase from rat liver, J. Biol. Chem., ggg, 2327-2333. Pegg, A.E. (1983a), Alkylation and subsequent repair of DNA after exposure to dimethylnitrosamine and related carcinogens, In: Reviews in Biochemical Toxicology, Hodgson, B., Bend, J.R. and Philpot, R.M. (eds.), Elsevier, New York, NY, pp. 83-133. Pegg, A.E. (1983b), Repair of 06-methylguanine in DNA by mammalian tissues, In: Biochemical Basis of Chemical Carcinogenesis, M. Kramer (ed.), Raven Press, New York. Pegg, A.E. (1984), Methylation of the 06 position of guanine in DNA is the most likely initiating event in carcinogenesis by methylating agents, Cancer Invest., g, 223-231. Pegg, A.E., Scicchitano, D. and Dolan, M.E. (1984), Comparison of the rates of rgpair of 06-alkylguanine in DNA by rat liver and bacterial O -a1kylguanine-DNA alkyltransferase, Cancer Res., tt, 3806-3811. Penman, B.W., Hoppe IV, H. and Thilly, W.G. (1979), Concentration—dependent mutation by alkylating agents in human lymphoblasts and salmonella typhimurium: N-methyl-N-nitrosourethane and beta-propiolactone, JNCI, t, 903-907. Pfeffer, L.M., Lipkin, M., Stutman, P. and Kopelovich, L. (1976), Growth abnormalities in cultured human skin fibroblasts, J. Cell Physiol., gt, 29-38. Pfeffer, L.M. and Kopelovich, L. (1977), Differential genetic susceptibility of cultured human skin fibroblasts to transformation by Kirsten murine sarcoma virus, Cell, tt, 313-320. Richardson, E.C., Dyroff, M.C., Boucheron, J.A. and S enberg, J.A. (1985), Differential repair of O -alkylthymidine following exposure to methylating and ethylating hepatocarcinogens, Carcinogenesis, g, 625-629. Robbins, J.H., Kraemer, K.H., Lutzner, M.A., Festoff, B.W. and Coon, H.G. (1974), Xeroderma pigmentosum: an inherited disease with sun sensitivity, multiple cutaneous neoplasms and abnormal DNA repair, Ann. Intern. Med., gt, 221-248. Roberts, J.J. (1978), The repair of DNA modified by cytotoxic, mutagenic, and carcinogenic chemicals, Adv. Radiation Biol., t, 211-436. 162 Robins, P. and Cairns, J. (1979), Quantitation of the adaptive response to alkylating agents, Nature, 280, 74-76. R gers, K.J. and Pegg, A.E. (1977), Formation of O -methylguanine by alkylation of rat liver, colon, and kidney DNA following administration of 1,2-dimethylhydrazine, Cancer Res., tt, 4082-4084. Rubin, J.S., Joyner, A.L., Bernstein, A. and Whitmore, G.F. (1983), Molecular identification of a human DNA repair gene following DNA-mediated gene transfer, Nature, 306, 206-208. Samson, L. and Cairns, J. (1977), A new pathway for repair in Escherichia coli, Nature, 267, 281-283. Samson, L. and Schwartz, J.L. (1980), Evidence for an adaptive DNA repair pathway in CH0 and human skin fibroblast cell lines, Nature, 287, 861-863. . Samson, L. and Schwartz, J.L. (1983), The induction of resistance to alkylation damage in mammalian cells, In: Induced Mutagenesis, C. Lawrence (ed.), Plenum, New York, NY, pp. 291-309. Schendel, P. (1981), Inducible repair systems and their implications for toxicology, In: CRC Crit. Rev. Toxicol., L. Goldberg (ed.), CRC Press, Boca Raton, Florida, p. 3311. Schwartz, J.L. and Samson, L. (1983), Mutation induction in Chineses hamster ovary cells after chronic pretreatment with MNNG, Mutation Res., 119, 393-397. Sgicchitano, D. and Pegg, A. . (1982), Kinetics of repair of O -methylguanine in DNA by O -methylguanine-DNA methyltransferase in vitro and tg vivo, Biochem. Biophys. Res. Commun., 109, 995-997. Scudiero, D.A., Meyer, S.A., Clatterbuck, B.R., Mattern, M.R., Ziolkowski, G.H. and Day, R.S. (1984), Sensitivity of hgman cell strains having different abilities to repair 0 -methylguanine in DNA to inactivation by alkylating agents including chloroethylnitrosourea, Cancer Res., tt, 2467-2474. Seki, S., Ohashi, M., Ogura, H. and Oda, T. (1982), Possible involvement of DNA polymerases alpha and beta in bleomycin-induced unscheduled DNA synthesis in permeable HeLa cells, Biochem. Biophys. Res. Commun., ttt, 1502-1508. Setlow, R.B. (1978), Repair deficient human disorders and cancer, Nature 271, 713-717. 163 S iloh, Y., Tabor, E. and Becker, Y. (1983), Repair of O -methylguanine in human fibroblast cell strains: Deficient repair capacity of lymphoblastoid cell lines does not reflect a genetic polymorphism, Mutation Res., ttg, 47—58. Singer, B. (1975), Chemical effects of nucleic acid alkylation and their relation to mutagenesis and carcinogenesis, Prog. Nucleic Acid Res. Mol. Biol., tg, 219-284; tg, 330-332. Singer, B. (1976), All oxygens in nucleic acids react with carcinogenic ethylating agents, Nature (London), 264, 333-339. Singer, B., Fraenkel-Conrat, H. and Kusmierek, J. (1978), Preparation and temp ate activities of polynucleotides containing 0 - and O -alkyluridine, Proc. Natl. Acad. Sci. U.S.A., tg, 1722-1726. Singer, B. and Brent, T.P. (1981), Human lymphoblasts contain DNA glycosylase acttvity excising N-3 and N-7 methyl and ethyl purines but not 0 -a1kylguanines or 1-alkyladenines, Proc. Natl. Acad. Sci. U.S.A., tg, 856. Singer, B., Spengler, S. and Bodell, W.J. (1981), Tissue-dependent enzyme-mediated repair or removal of O-ethyl pyrimidines and ethyl purines in carcinogen-treated rats, Carcinogenesis, 3, 1069-1073. Singer, B. and Kusmierek, J.T. (1982), Chemical mutagenesis, Ann. Rev. Biochem., gg, 655-693. Singer, B. (1984), Alkylation of the 06 position of guanine is only one of many chemical events that may initiate carcinogenesis, Cancer Invest., g, 233-238. Sirover, M.A. and Loeb, L.A. (1974), Erroneous base-pairing induced by a chemical carcinogen during DNA synthesis, Nature 252, 414. S lar, R. and Strauss, B. (1981), Removal of O -methylguanine from DNA of normal and xeroderma pigmentosum derived lymphoblastoid lines, Nature, 2 9, 417-420. Snow, E.T., Foote, R.S and Mitra, S. (1983), Replication and demethylation of O -methylguanine in DNA, Prog. Nucleic Southern, E. (1975), Detection of specific sequences among DNA fragments separated by gel electrophoresis, J. Mol. Biol., gg, 503. 164 Spivak, G., Ganesan, A.K. and Hanawalt, P.C. (1984), Enhanced transformation of human cells by UV-irradiated pSV2 plasmids, Mol. Cell Biol., t, 1169-1171. Strauss, B.S. (1976), Repair of DNA adducts produced by alkylation, In: Agigg, Carcinogenesis and Radiation Biology, K.C. Smith (ed.), Plenum, New York and London, pp. 287—314. Strauss, B.S. (1977), Molecular biology of the response of cells to radiation and to radiomimetic chemicals, Cancer, tt, 471-480. Stoltz, D.R., Poirier, L.A., Irving, C.C., Stich, H.F., Weisberger, J.H. and Grice, H.C. (1974), Evaluation of short—term tests for carcinogenicity, Toxicol. Appl. Pharmacol., gt, 157-180. Sugino, A. and Nakayoma, K. (1980), DNA polymeraseek mutants from a Drosophilia melangaster cell line, Proc. Natl. Acad. Sci. U.S.A., tt, 7049-7053. Suter, W., Brennand, J., McMillan, S. and Fox, M. (1980), Relative mutagenicity of antineoplastic drugs and other alkylating agents in V79 Chinese hamster cells, independence of cytotoxic and mutagenic responses, Mutation Res., tt, 171—181. Sutherland, B.M. (1978), Enzymatic photoreactivation of DNA, In: DNA Repair Mechanisms, P.C. Hanawalt, E.C. Friedberg and C.F. Fox (eds.), Academic Press, New York. Sutherland, B.M. and Bennett, P.V. (1984), Transformation of human cells by DNA transfection, Cancer Res. tt, 2769-2772. Sutton Jr., R.L. (1938), Early epideimal neoplasia: description and interpretation, Arch. Derm. Syph., tt, 737. Swann, P.E. and Magee, P.N. (1971), Nitrosamine—induced carcinogenesis. The alkylation of N-7 of guanine of nucleic acids of the rat by diethylnitrosamine, N—ethyl-N-nitrosourea and ethyl methanesulfonate, Biochem. J., tgg, 841-847. Swenberg, J.A., Bedell, M.A., Billings, K.C., Umbenhauer, D6R° and Pegg, A.E. (1982), Cell specific differences in O -alkylguanine DNA repair activity during continuous carcinogen exposure, Proc. Natl. Acad. Sci. U.S.A., tg, 5499. 165 Swenberg, J.A., Dryoff, M.C., Bedell, M.A., Popp, J.A., Huh, N , Kirstein, U. and Rajewsky, M. . (1984), O -ethyldeoxythymidine, but not 0 -ethyldeoxyguanosine, accumulates in hepatocyte DNA of rats exposed continuously to diethylnitrosamine, Proc. Natl. Acad. Sci. USA, gt, 1692-1695. Szybalski, W. (1964), Chemical reactivity of chromosomal DNA as related to mutagenicity: Studies with human cell lines, Cold Spring Harbor Symp. Quant. Biol., gt, 151-160. Temin, H.M. (1974), On the origin of genes for neoplasia, G.H.A. Clowes Memorial Lecture, Cancer Res. tt, 2835. Teo, I.A., Lehmann, A.R., Muller, R. and Rajewsky, M.F. (1983), Similar rate of O -ethylguanine elimination from DNA in normal human fibroblast and xeroderma pigmentosum cell strains not transformed by SV40, Carcinogenesis, t, 1075-1077. Thielmann, H.W., Schroder, C.H., O'Neill, J.P., Brimer, P.A. and Hsie, A.W. (1979), Relationship between DNA alkylation and specific-locus mutation induction by N—methyl- and N-ethyl-N-nitrosourea in cultured Chinese hamster ovary cells (CHO/HGPRT system), Chem. Biol. Interact., gg, 233-243. Van Ness, B.G., Howard, J.B. and Bodley, J.W. (1980), ADP-ribosylation of elongation factor 2 by diphtheria toxin: NMR spectra and proposed structures of ribosyl-diphthamide and its hydrolysis products, J. Biol. Chem., ggg, 10710-10716. Waldstein, E.A., Cao, E.-H., Bender, A. and Setlow, R.S. (19828), Abilities of extracts of human lymphocytes to remove O -methylguanine from DNA, Mutation Res., tg, 405. Waldstein, E.A, Cao, E.-H., Miller, E.P., Cronkite, E.P. and Setlow, R.B. (1982b), Extracts of chronic lymphocytic leukemia lymphgcytes have a high level of DNA repair activity for O -methylguanine, Proc. Natl. Acad. Sci., tt, 4786. Waldstein, E.A, Cao, E.-H and Setlow, R.B. (l982c), Adaptive resynthesis of O -methylguanine-accepting protein can explain the differences between mammalian cells proficient and deficient in methyl excision repair, Proc. Natl. Acad. Sci. USA, tt, 5117. Waldstein, E.A, Cao, E -H. and Setlow, R.B. (l982d), Adaptive increase of O -methylguanine-acceptor protein in HeLa cells following N-methyl-N‘-nitro-N-nitrosoguanidine treatment, Nucleic Acids Res., tt, 4595-4604. 166 Warren, W., Crathorn, A.R. and Shooter, K.V. (1979), The stability of methylated purines and of methylphosphotriesters in the DNA of V79 cells after treatment with N-methyl-N—nitrosourea, Biochim. Biophys. Acta, ggt, 82-88. de Weerd-Kastelein, E.A., Keijzer, W. and Bootsma, D. (1972), Genetic heterogeneity of xeroderma pigmentosum demonstrated by somatic cell hybridization, Nature, ggt, 80-83. Weisberger, J.H. and Williams, G.M. (1975), Metabolism of Chemical Carcinogens, In: Cancer, F.F. Becker (ed.), t, Plenum Press, New York, NY, pp. 185-234. Westerveld, A., Hoeijmakers, J.H.J., van Duin, M., de Wit, J., Odijk, H., Pastink, A., Wood, R.D. and Bootsma, D. (1984), Molecular cloning of a human DNA repair gene, Nature, ttt, 425-429. Yang, L.L., Maher, V.M. and McCormick, J.J. (1980), E ror-free excision of the cytotoxic, mutagenic N -deoxyguanosine DNA adduct formed in human fibroblasts by (+/-)-]8, 8d-dihydroxy, 9X, l0d-epoxy- 7,8,9,l0- tetrahydrobenzo[a]pyrene, Proc. Natl. Acad. Sci. U.S.A., tt, 5933—5937. Yang, L.L., Maher, V.M. and McCormick, J.J. (1982), Relationship between excision repair and.the cytotoxic and mutagenic effect of the ‘anti' 7,8-diol-9,l0—epoxide of benzo[a]pyrene in human cells, Mutat. Res. tt, 435-447. Yarosh, D.B., Rice, M., Day III, R.S., Foote, R.S. and Mitra, S. (1984), O -methylguanine-DNA methyltransferase in human cells, Mutation Res., 131, 27-36. IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII