Human cells that lack O1-alkylguanine DNA alkyltransferase (AT) activity can remove O1-butylguanine (O1-wBuG) produced in cellular DNA by exposure to N-n-butyl-N-nitrosourea as determined by radioimmunoassay of enzyme digests of DNA. Fibroblasts from xeroderma pigmentosum (XP) complementation groups A and G that show <5 % unscheduled DNA synthesis following exposure to UVC failed to remove O1-nBuG. Hence it appears that O1- alkylguanine is repaired in cells that lack AT by a process that is defective in XP cells, presumably nucleotide excision repair. Neither V79 nor V79/79 Chinese hamster cell lines have AT activity and both are able to remove O1-nBuG from DNA. However, only V79/79 is able to remove O1MeG, suggesting some substrate specificity of the excision repair process. Comparison of relative levels of O1-alkylation by N-mcthyl-, N-ethyl-, N-propyl- and N-n-butyl-nitrosourea indicate that approximately equal levels of O1-alkylation are produced by equitoxic doses of these agents.

Alkylation of DNA at the O1 atom of guanine by monofunctional alkylating agents produces an adduct that introduces miscoding errors in transcription and replication of DNA (reviewed by Saffhill et al. 1985). The mutagenic and carcinogenic potential of this adduct is reduced in most cells by O1-alkylguanine DNA alkyltransferase (AT), a protein that transfers the alkyl residue from the O1-alkylguanine to a cysteine in its polypeptide chain, a process that inactivates the catalytic function (Harris et al. 1983). Cells that lack AT are sensitive to the cytotoxic and mutagenic action of mono- and some bi-functional alkylating agents whose primary attack is at O1-guanine. Included in the latter class are the chloroethyl nitrosoureas, which form DNA crosslinks if the initial O1-adduct is not repaired (Erickson et al. 1980; Scudiero et al. 1984a).

Human cells possess a nucleotide excision repair pathway that can excise bulky DNA adducts that are assumed to be similar to the archetypal lesion, the ultraviolet light (u.v.)-induced pyrimidine dimer, in causing topological distortions of the DNA double helix. In this paper we show that O1-n-butylguanine (O1- n BUG) can be repaired in cell lines deficient in AT, probably by nucleotide excision repair since xeroderma pigmentosum cells, defective in excision repair of pyrimidine dimers, are also defective in repair of the butyl adduct. We show also that a similar process appears to act on O1-methylguanine (O1MeG) in Chinese hamster V79 cells, and present some preliminary data concerning the expression of this activity in clones derived from cell fusion experiments.

Cell lines

Human fibroblast cell lines, MRC-5 (Jacobs et al. 1970) and its simian virus 40 (SV40)- transformed derivative MRC-5 SV2 (Hoschtscha & Holliday, 1983) were kindly donated by Dr S. Fairweather. Xeroderma pigmentosum fibroblasts XP25RO (De Weerd-Kastelein et al. 1972) and the SV40-transformed line XP12RO(SV) (Teo et al. 1983), which carries the same XP group A mutation, were provided by Dr C. Arlett. These cells were maintained in Eagle’s minimal essential medium (MEM) supplemented with 15 % foetal calf serum.

The origins of the cloned Chinese hamster lines V79A-2 and its u.v.-, X-ray- and alkylationsensitive derivative V79/79-1 have been described previously (Durrant & Boyle, 1982). Both lines were maintained in MEM plus 10% foetal calf serum. All cell lines were routinely screened for mycoplasma infection (Chen, 1977) and penicillin (100 IUml − 1) and streptomycin (50 μ gml − 1) were added to experimental cultures but not to stock cultures.

Alkylation of calf thymus DNA in vitro

Calf thymus DNA was dissolved in 50mM-potassium phosphate buffer, pH 7 · 4, at 1 mg ml − 1 and alkylated by incubation at 37 ° C for 2h with appropriate quantities of specific N-alkyl-N- nitrosoureas.

Radioimmunoassay of O1-nBuG and O1MeG

The formation and persistence of O1-nBuG and O1MeG in DNA were quantified in radioimmunoassay (RIA) systems developed with monoclonal antibodies specific for either of these adducts (Saffhill et al. 1982; Wild et al. 1983) and validated by comparison with standard radiochromatographic estimations (Wild et al. 1983; Saffhill, 1984).

O1-alkylguanine DNA alkyltransferase activity

Pellets of up to 2 × 101 cells harvested by trypsinization were washed with PBS and resuspended in 1 · 4 ml buffer containing 50mM-Tris-HCl, pH 7 · 8, 1 mM-EDTA and 3mM-dithiothreitol. The suspensions were exposed to three 10-s pulses at 10 gm peak-to-peak in an MSE sonicator and 87 gg phenylmethylsulphonyl fluoride (PMSF) was added in 10 μ l of ethanol. Sonicates were centrifuged at 4 ° C for 10 min at 20 000 revs min − 1 in a Coolspin microcentrifuge. The supernatants were assayed for AT activity either by HPLC analysis (Cooper et al. 1982) in early experiments, or by a more rapid and sensitive assay that measures the transfer of radioactively labelled methyl groups from substrate DNA to the AT protein (Margison et al. 1985). Substrates were prepared (Margison et al. 1985) using N-[1H]methyl-A’-nitrosourea ([1H]MNU) of 1 · 6 or 5 · 4 Ci mmol − 1 (HPLC assays) or 5 · 4 or 30 Ci mmol − 1 (protein assays).

Dot-blot and Southern analysis

High molecular weight DNA was extracted by a method similar to that of Blin & Stafford (1976). Optical density ratios of 260/280 nm were close to 2-0 and molecular weights in excess of 100 kilobases (kb) were demonstrated by electrophoresis in 0 · 3 % agarose.

The high molecular weight DNA was nick-translated by Escherichia coli DNA polymerase I using [α -1P]dCTP (Rigby et al. 1977). Labelled DNA was separated from excess α -1P by centrifugation through a column of Sephadex G50/80. By this method specific activities of >1 × 101 cts min − 1μ g − 1 DNA were obtained.

Further fractions of high molecular weight DNA were digested with EcoRI and the fragments were separated by electrophoresis in 1 % agarose, then transferred from the gel to a nitrocellulose filter by capillary action (Southern, 1975). Filters were baked in vacuo for 2h at 80 ° C and prehybridized in BLOTTO (Johnson et al. 1984). They were then probed with nick-translated MRC-5 total DNA at a stringency of 2 × SSC (SSC is 0 · 15M-NaCl, 0 · 015 M-sodium citrate, pH7-0) at 65 ° C overnight and processed for autoradiography using AR X-ray film in a cassette with an intensifying screen.

For dot blots, cells were harvested and dilutions of 2 × 101, 7 × 101, 2 × 101, 7 × 101, 2 × 101 and 7 × 101 cells per ml were prepared and 50((I of each dilution was applied to wells of a Bio-rad dot blot manifold. The cells were lysed by contact with 3MM papers soaked in 0 · 5M-NaOH, 1 · 5 M- NaCl. The filter was neutralized and baked in a vacuum oven for 2h at 80 ° C, prehybridized in BLOTTO, then probed with nick-translated DNA as above.

Alkylation schedule and cell survival assay

Exponentially growing monolayers of cells were treated in T30 flasks with.N-n -butyl-.V- nitrosourea (BNU) or MNU for 30min at 37 ° C as previously described (Boyle et al. 1986b). Survival of colony-forming ability of human cells was measured in the presence of gamma irradiated feeder cells (Boyle et al. 1986a).

With human cells

Alkyltransferase activity

Typical protein dependence curves for AT activity in cell sonicates are shown in Fig. 1, from which specific activities in terms of fmol methyl transferred per mg protein were calculated as 200 (MRC-5), 2 (MRC-5 SV2), 625 (XP25RO) and 0 (XP12RO (SV)). Values less than 10fmol mg − 1 protein are within the range of background determinations. Thus the two SV40-transformed lines are devoid of AT activity, consistent with the findings of others (Day et al. 1980; Teo et al. 1983).

Fig. 1.

AT activity in sonicates of human cells. The protein dependence of AT activity was measured by the protein assay (Materials and Methods) using sonicates of MRC-5 (O), MRC-5(SV2) (•), XP25RO (▫) and XP12RO(SV) (▪). The incubation time was 30 min.

Fig. 1.

AT activity in sonicates of human cells. The protein dependence of AT activity was measured by the protein assay (Materials and Methods) using sonicates of MRC-5 (O), MRC-5(SV2) (•), XP25RO (▫) and XP12RO(SV) (▪). The incubation time was 30 min.

Formation and persistence of O1-nBuG

The formation of O1-MBUG (O1-n- butylguanine) was measured by RIA immediately after treatment of cells with 4mM-BNU for 30 min at 37 ° C. A value of 13 · 0 ± 2-Oftmol adduct per mol guanine was obtained in six experiments, indicating that the rate of alkylation was similar in all cell lines.

Fig. 2 shows the persistence of the adduct during the 24 h following alkylation. The curves indicate that half the adduct was removed from MRC-5 and MRC-5 SV2 in 17 h and 22 h, respectively, whereas the XP lines removed only 5 % in 24 h. Thus normal human cell lines are able to remove O1-MBUG efficiently whereas XP cells are deficient in this ability.

Fig. 2.

Persistence of O1-MBUG in DNA of human cells treated with BNU. Cells were exposed to 4mM-BNU for 30 min, then harvested immediately or incubated in fresh medium for the indicated times. Cellular DNA was extracted and processed as described in Materials and Methods and the O1-HBUG measured by RIA. Points represent individual determinations expressed as a percentage of the amount of adduct induced (13-0 ± 2 · 0 μ mol per mol guanine; n = 6). Symbols as for Fig. 1.

Fig. 2.

Persistence of O1-MBUG in DNA of human cells treated with BNU. Cells were exposed to 4mM-BNU for 30 min, then harvested immediately or incubated in fresh medium for the indicated times. Cellular DNA was extracted and processed as described in Materials and Methods and the O1-HBUG measured by RIA. Points represent individual determinations expressed as a percentage of the amount of adduct induced (13-0 ± 2 · 0 μ mol per mol guanine; n = 6). Symbols as for Fig. 1.

Evidence that AT can repair O1-nBuG in vivo

To determine whether AT can remove O1-MBUG in vivo we treated MRC-5 cells with doses of MNU or BNU that resulted in 10% survival of colony-forming ability. Immediately and at intervals after treatment cells were harvested and residual AT activity in sonicates was measured. On the basis of the assumption that transfer of butyl residues would inactivate AT stoichiometrically as in the case of methyl transfer, a decline in AT in cells treated with BNU would imply that AT can accept butyl residues, although this would give no information concerning the origin of the residues, whether from O1-MBUG or from phosphotriesters.

Table 1 shows that butylation causes an immediate loss of 40 % AT activity from which the cells recover during subsequent incubation. The greater loss of activity seen with MNU may reflect the fact that at equitoxic doses the alkylation of the O1 atom of guanine by MNU is 50% greater than by BNU (see Discussion). Alternatively, AT might act more slowly on O1-MBUG, the abundance of which is reduced by excision repair. Further experiments are in progress to establish the relative stoichiometry of inactivation of AT by these agents and the kinetics of resynthesis.

Table 1.

Changes in AT activity in MRC-5 cells exposed to equitoxic doses of MNU and BNU

Changes in AT activity in MRC-5 cells exposed to equitoxic doses of MNU and BNU
Changes in AT activity in MRC-5 cells exposed to equitoxic doses of MNU and BNU

Relative sensitivity of human cells to killing by monofunctional alkylnitrosoureas

Dose—response curves for survival of colony-forming ability of each of the four human cell lines were determined with N-methyl-, N-ethyl-, N-propyl- and N-n- butyl-N-nitrosoureas. Fig. 3 shows the Dio (dose at which 10% of cells survive to form colonies) values from those curves plotted as a function of alkyl chain length. The extreme sensitivity of SV40-transformed cells to methylation rapidly disappears when higher alkylating species are used. However, an approximately twofold increase in sensitivity due to the presence of the XP mutation is seen with all alkylating agents.

Fig. 3.

Effect of alkyl chain length on sensitivity of human cells to cell killing by alkyl nitrosoureas. Survival of colony-forming ability was measured following exposure of each of the cell lines to MNU, ENU, PNU or BNU for 30min. Survival curves were constructed from the results of duplicate experiments, some of which have been reported previously (Boyle et al. 1986a). D10 values derived from these curves are plotted as a function of the carbon chain length of the alkyl adducts. Symbols as for Fig. 1.

Fig. 3.

Effect of alkyl chain length on sensitivity of human cells to cell killing by alkyl nitrosoureas. Survival of colony-forming ability was measured following exposure of each of the cell lines to MNU, ENU, PNU or BNU for 30min. Survival curves were constructed from the results of duplicate experiments, some of which have been reported previously (Boyle et al. 1986a). D10 values derived from these curves are plotted as a function of the carbon chain length of the alkyl adducts. Symbols as for Fig. 1.

To determine the relative yields of O1-alkylguanine from different nitrosoureas under standard conditions, calf thymus DNA was alkylated in 50 mM-phosphate and adduct yields were measured by radiochromatography or RIA (Table 2).

Table 2.

Relative in vitro yields of O1-alkylguanine

Relative in vitro yields of O1-alkylguanine
Relative in vitro yields of O1-alkylguanine

With Chinese hamster cells

Repair of O1MeG and O1-nBuG in V79, V79/79 and hybrid cells

Previously (Wild et al. 1983) we showed that V79/79 cells can repair O1MeG whereas the parent line V79 cannot. However, both cell lines are capable of repairing O1-MBUG at about equal rates (Fig. 4). These results are summarized in Table 3, which also shows the half-times for removal of O1MeG, 7-MeG and 3-MeA obtained by radio-chromatography of acid hydrolysed DNA extracted from MNU-treated cells. The differential removal of O1MeG by the two cell lines is clearly not reflected in the removal of the other adducts, implying that removal of O1MeG in V79/79 is mediated by a specific pathway and is not the result, for example, of enhanced nonspecific nuclease attack.

Table 3.

Repair of alkyl purines and AT activity in Chinese hamster cell lines

Repair of alkyl purines and AT activity in Chinese hamster cell lines
Repair of alkyl purines and AT activity in Chinese hamster cell lines
Fig. 4.

Persistence of O1-alkylguanine in Chinese hamster cells. Chinese hamster V79 (○, •) and V79/79 (▫) cells were exposed to MNU or BNU and the relevant O1 adduct in the DNA of cells harvested immediately or at intervals after alkylation was determined by RIA. A is reproduced from Wild et al. (1983) by kind permission of IRL Press. A. 2mM-MNU for 30min at 37°C; B, 4mM-BNU for 30min at 37°C.

Fig. 4.

Persistence of O1-alkylguanine in Chinese hamster cells. Chinese hamster V79 (○, •) and V79/79 (▫) cells were exposed to MNU or BNU and the relevant O1 adduct in the DNA of cells harvested immediately or at intervals after alkylation was determined by RIA. A is reproduced from Wild et al. (1983) by kind permission of IRL Press. A. 2mM-MNU for 30min at 37°C; B, 4mM-BNU for 30min at 37°C.

The repair capacity was also determined by radiochromatography of three hybrid clones from fusion of V79A-2 HPRT X V79/79-1 APRT (Table 3). Two of the lines were able to repair O1MeG whilst the other, which had a reduced chromosome complement, behaved like the V79 parent and lacked repair of this adduct.

AT activity in Chinese hamster cell lines

Table 3 also shows the AT activity of the cell lines. All the assays performed on sonicates of Chinese hamster cells gave values that were indistinguishable from background even when relatively large amounts of protein were used.

The repair of O1MeG in hamster cells lacking AT activity is not unique to V79/79 cells

In order to investigate genetic factors controlling the repair of O1-alkylguanine a series of fusions was performed between MRC-5 and an HPRT- derivative of V79A-2 with selection in HAT medium plus lX10-5M-ouabain. The majority of colonies isolated were true hybrids containing human and hamster chromosomes. However, human DNA could not be detected in the colonies listed in Table 3 when tested by dot-blot and Southern analysis using nick-translated MRC-5 DNA as the probe (data not shown). Clones MV34 and MV4.4 contained 32 and 33 chromosomes, respectively, whereas clones MV33, MV7 and MV9 contained either 22 or 24 chromosomes, similar to that of the V79A-2 parent.

None of these clones had AT activity above background, but clones MV34, MV33 and MV4.4 removed 40% or more O1MeG in 24 h following alkylation with 2mM-MNU (Table 4).

Table 4.

Persistence of O1MeG and AT activity in Chinese hamster cell lines after exposure to polyethylene glycol

Persistence of O1MeG and AT activity in Chinese hamster cell lines after exposure to polyethylene glycol
Persistence of O1MeG and AT activity in Chinese hamster cell lines after exposure to polyethylene glycol
Table 5.

Calculation of O1-alkylguanine molecules per cell per lethal hit

Calculation of O1-alkylguanine molecules per cell per lethal hit
Calculation of O1-alkylguanine molecules per cell per lethal hit

Day et al. (1980) have demonstrated host cell reactivation of methylated viral DNA in human cells designated mer+ (methyl repair), but not in others designated mer−, which lack AT activity. The finding that some mer−, cells showed more repair replication of cellular DNA than did wild-type cells following exposure to N-methyl- N′-nitro-N-nitrosoguanidine (MNNG), but not after dimethylsulphate (DMS) or u.v., provided the first indication that O1-alkylguanine might be repaired by a specific excision repair process (Scudiero et al. 19846). We have now shown that AT-deficient normal human cells, but not XP cells, are able to remove chemically stable O1-MBUG from their DNA, an observation that strongly suggests that this adduct is repairable by nucleotide excision repair. This interpretation is supported by preliminary evidence of O1-n BudG in the culture medium supernatants of excision-proficient cells (Saffhill & Boyle, unpublished data). Support for excision repair of O1-nBuG also comes from experiments in bacteria where uvr and rec mutations have been shown to influence the frequency of site-specific transition mutations induced in phage 0X174 by this adduct (Chambers et al. 1985). The possibility that cells of XP complementation group A have a special deficiency in this respect is ruled out by the finding that group G cells, which are like group A cells in being strongly deficient in nucleotide excision of pyrimidine dimers, are also strongly defective in O1-nBuG repair (Boyle et al. 1986a).

In Chinese hamster cells an alternative pathway to AT also exists, which we tentatively assume to be excision repair. Neither V79 nor V79/79 has detectable AT activity, but both lines are able to repair O1-wBuG, a situation analogous to MRC-5 SV2 cells. The analogy can be strengthened semi-quantitatively. A dose of 1J m−,2 produces about 54000 dimers per cell and 4mM-BNU produces a roughly comparable yield of 26000 O1-nBuG per cell. At this level of damage, which is nonsaturating for excision of dimers, the ratio of excision (strictly, incision) of dimers in normal untransformed human versus hamster cells was 5·3 (Collins et al. 1982). Human cells transformed by SV40 showed a reduced rate of excision of dimers (Squires et al. 1982) so that it is likely that the ratio of excision of dimers in MRC-5 SV2 versus V79 is nearer 1 as is the ratio for excision of O1-MBUG (Figs 1, 2). V79/79 cells appear somewhat exceptional compared to other Chinese hamster cell lines in that they can also remove O1MeG by excision repair. As yet we have no data concerning excision of O1MeG in human cells.

The ability of N19/19 cells to repair O1MeG was inherited in two hybrid cell lines. We cannot identify the origins of individual chromosomes in these intraspecies hybrids, but because they contain nearly the sum of the chromosome numbers of the parent cells, whilst another hybrid has a reduced chromosome number and lacks O1 repair, our tentative conclusion is that the repair activity is dominantly expressed in hybrids, suggesting that the lack of repair in V79 is not due to a trans-acting repressor.

Repair of O1MeG was also observed in another group of clones derived from V79 and MRC-5 cells that had been subjected to a polyethylene glycol fusion protocol followed by selection in HAT plus ouabain. No human chromosomes could be distinguished in these cell lines and no human DNA was detected when dot-blot and Southern blot analysis were performed under conditions used to detect single copy genes. Following growth in HAT-free medium for several passages after isolation, all were found to be resistant to 6-thioguanine. Hence it would appear that these lines represent transient HAT-resistant phenocopies of V79 that survived the initial selection. An alternative explanation, that they represent hybrids that rapidly lost all human chromosomes, is possible but unlikely, since they were isolated at a frequency similar to that of true hybrids obtained from the same fusion experiment.

We favour the idea that the appearance of excision activity in these clones may be the result of hypomethylation of deoxy cytidine. An analogous situation appears to cause the switching on of HPRT in fusions involving a partner thought to possess a deletion in HPRT because of its resistance to reverse mutagenesis (Watson et al. 1972; Bakay et al. 1973). In one series of experiments the fusogen was omitted but switching on of the repressed HPRT still occurred when the cells were incubated at high density under fusion conditions (Shin et al. 1973). In retrospect, it seems likely that these conditions were favourable for hypomethylation of HPRT, a known cause of activation of this gene (Migeon et al. 1985).

Some alkyl adducts, e.g. 3-methylpurines, are cytotoxic because they block DNA synthesis. Others, e.g. 7-MeG and methyl phosphotriesters, appear to have little effect on cell survival. The question whether O1-alkylguanine is cytotoxic as well as promutagenic is contentious. The fact that mer+ cells with AT activity are more resistant to MNU than are mer cells, which lack AT activity, has been considered cogent evidence that O1MeG is cytotoxic. Furthermore, transfection of the E. coli AT gene into V79 cells increased their resistance to alkylating agents (Margison & Brennand, 1987). On the other hand, selective inactivation of AT by incubation of cells with O1MeG either had no effect (Karran & Williams, 1985) or only slightly increased sensitivity to monofunctional alkylating agents (Dolan et al. 1985), but increased sensitivity to bifunctional alkylating agents to a greater extent (Dolan et al. 1985). The argument centres upon the uncertainty that it is O1MeG rather than some other lesion produced with a similar dose response that causes lethality. The lack of AT activity in SV40-transformed fibroblasts means that MRC-5 cells are hypersensitive to the effects of MNU, but they show normal sensitivity to higher alkyl nitrosoureas. Against the background of excision deficiency of XP, SV40 transformation also adds an increment of sensitivity towards ENU. However, the excision defect in XP25RO gives an approximately twofold increase in sensitivity towards all the nitrosoureas tested.

According to target theory, the D37 dose, at which 37 % of cells survive to form colonies, will on average produce one lethal hit per cell. Knowing the extent of alkylation produced in vivo by 4mM-BNU and the relative extents of O1-alkylation produced in vitro under standard conditions by BNU, PNU, ENU and MNU (Table 2) we can gain an approximate estimate of the numbers of O1-alkyl residues per lethal hit per cell (Table 5). This shows that the differential sensitivity to MNU associated with SV40 transformation is correlated with a difference of about 15 000 O1MeG molecules per human cell (Harris et al. 1983). In a similar manner the excision defect of XP25RO is associated with a difference of 5000–7000 O1-alkylguanine molecules, independent of alkyl chain length. It is of interest to note that in the AT-proficient, excision-proficient MRC-5 cells D37 values for all nitrosoureas are associated with 11 800 ± 3300 O1-alkylguanine molecules per cell, i.e. an almost constant value despite MNU being 3-6-fold more reactive than BNU at constant dosage and producing proportionately less O1-alkylation (O1/N1 ratio, about 0 · 1) compared to other nitrosoureas (O1/N1 ratios = 0 · 6 – 0 · 7). Therefore, this observation tends to support the idea that O atom, rather than N atom, alkylation is predominant in the cytotoxicity of alkylnitrosoureas. A further inference is that the sparsity of O1 lesions (on average 1 per 300 kilobase-pairs) makes it unlikely that cytotoxicity results from the overlapping of single strand breaks produced during excision of these lesions.

Bacterial and mammalian AT activities accept a range of alkyl groups including some chloroethyl derivatives. In vitro, transfer of methyl is more efficient than that of higher alkyl groups, including butyl (Morimoto et al. 1985). We do not know the relative rates of repair in vivo of methyl and butyl adducts; however, our present results show that in human cells O1-nBuG is accepted by AT in a reaction that is rapid compared with that of excision repair. The present rough estimates (Table 5) show that D37 doses of MNU produce 15 100 O1MeG and 9750 O1-nBuG molecules per cell. Since human cells contain 10000-20000 AT molecules per cell, further work is necessary to determine to what extent the greater loss of AT in MNU-treated cells (Table 1) is due to stoichiometric inactivation of all AT molecules or due to a faster rate of reaction.

We thank Josie Hopkins, Ann Hallam, Gail McGown, Menna Davies, Janice Smith and Brian Keenan for skilled technical help in different phases of this work, and the Cancer Research Campaign for financial support.

Bakay
,
B.
,
Graf
,
M.
,
Carey
,
S.
,
Nissinen
,
E.
&
Nyhan
,
W. L.
(
1973
).
Re-expression of HPRT activity following cell fusion with polyethylene glycol
.
Biochem. Genet.
16
,
227
237
.
Beranek
,
D. T.
,
Weiss
,
C. C.
&
Swenson
,
D. H.
(
1980
).
A comprehensive quantitative analysis of methylated and ethylated DNA using high pressure liquid chromatography
.
Carcinogenesis
1
,
595
606
.
Blin
,
N.
&
Stafford
,
D. W.
(
1976
).
Isolation of high molecular weight DNA
.
Nucl. Acids Res.
3
,
2303
2308
.
Boyle
,
J. M.
,
Margison
,
G. P.
&
Saffhill
,
R.
(
1986a
).
Evidence for the excision repair of O6-nButyl deoxyguanosine in human cells
.
Carcinogenesis
7
,
1987
1990
.
Boyle
,
J. M.
,
Saffhill
,
R.
,
Margison
,
G. P.
&
Fox
,
M.
(
1986b
).
A comparison of cell survival, mutation and persistence of putative promutagenic lesions in Chinese hamster cells exposed to BNU or MNU
.
Carcinogenesis
7
,
1981
1985
.
Chambers
,
R. W.
,
Sledziewska-Goiska
,
E.
,
Hirani-Hojatti
,
S.
&
Borowy-Borowski
,
H.
(
1985
).
UvrA and recA mutations inhibit a site-specific transition produced by a single O6- methylguanine in gene G of bacteriophage 0X174
.
Proc. natn. Acad. Sci. U.SA.
82
,
7137
7177
.
Chen
,
T. R.
(
1977
).
In situ dectection of mycoplasma contamination by fluorescent Hoechst 33258 stain
.
Expl Cell Res.
104
,
255
—264.
Collins
,
A.
,
Jones
,
C.
&
Waldren
,
C.
(
1982
).
A survey of DNA repair incision activities after ultraviolet irradiation of a range of human, hamster and human-hamster hybrid cell lines
.
J. Cell Sci.
56
,
423
440
.
Cooper
,
D. P.
,
O’Connor
,
P. J.
&
Margison
,
G. P.
(
1982
).
Effect of acute doses of 2-acetylaminofluorine on the capacity of rat liver to repair methylated purines in DNA in vivo and in vitro
.
Cancer Res.
42
,
4203
4209
.
Day
,
R. S.
,
Ziolkowski
,
C. H. J.
,
Scudiero
,
D. A.
,
Meyer
,
S. A.
,
Lubiniecki
,
A. S.
,
Girardi
,
A. J.
,
Galloway
,
S. M.
&
Bynum
,
G. D.
(
1980
).
Defective repair of alkylated DNA by human tumor and SV40 transformed human cell strains
.
Nature, Lond.
288
,
724
727
.
De Weerd-Kastelein
,
E. A.
,
Keuzer
,
W.
&
Bootsma
,
D.
(
1972
).
Genetic heterogeneity of xeroderma pigmentosum demonstrated by somatic cell hybridisation
.
Nature, new Biol.
238
,
80
83
.
Dolan
,
M. E.
,
Corsico
,
C. D.
&
Pegg
,
A. E.
(
1985
).
Exposure of HeLa cells to O(6)-alkyl- guanines increases sensitivity to the cytotoxic effects of alkylating agents
.
Biochem. biophys. Res. Commun.
132
,
178
185
.
Durrant
,
L. G.
&
Boyle
,
J. M.
(
1982
).
Potentiation of cell killing by inhibitors of poly (ADP- ribose) polymerase in four rodent cell lines exposed to N-methyl-N-nitrosourea or UV light
.
Chem.-Biol. Interact.
38
,
325
338
.
Erickson
,
L. C.
,
Bradley
,
M. O.
,
Ducore
,
J. M.
,
Ewig
,
R. A. G.
&
Kohn
,
K. W.
(
1980
).
DNA crosslinking and cytotoxicity in normal and transformed human cells treated with antitumor nitrosoureas
.
Proc. natn. Acad. Sci. U.SA.
77
,
467
471
.
Harris
,
A. L.
,
Karran
,
P.
&
Lindahl
,
T.
(
1983
).
O6-methylguanine-DNA methyltransferase of human lymphoid cells. Structural and kinetic properties and absence in repair deficient cells
.
Cancer Res.
43
,
3247
3252
.
Hoschtscha
,
L. I.
&
Holliday
,
R.
(
1983
).
Limited and unlimited growth of SV40 transformed cells from human diploid MRC-5 fibroblasts. J’
.
Cell Sci.
63
,
77
99
.
Jacobs
,
J. P.
,
Jones
,
C. M.
&
Baille
,
J. P.
(
1970
).
Characterisation of a human diploid cell designated MRC-5
.
Nature, Lond.
227
,
168
170
.
Johnson
,
D. A.
,
Gautsch
,
J. W.
,
Sportsman
,
J. R.
&
Elder
,
J. H.
(
1984
).
Improved technique utilizing nonfat dry milk for analysis of proteins and nucleic acids transferred to nitrocellulose
.
Gene Analyt. Tech.
1
,
3
8
.
Karran
,
P.
&
Williams
,
S. A.
(
1985
).
The cytotoxic and mutagenic effects of alkylating agents on human lymphoid cells are caused by different DNA lesions
.
Carcinogenesis
6
,
789
792
.
Margison
,
G. P.
&
Brennand
,
J.
(
1987
).
Functional expression of the Escherichia colialkyltransferase gene in mammalian cells
.
J. Cell Sci. Suppl.
6
,
83
96
.
Margison
,
G. P.
,
Cooper
,
D. P.
&
Brennand
,
J.
(
1985
).
Cloning of the E. coliO6-methylguanine and methyl-phosphotriester methyltransferase gene using a functional DNA repair assay
.
Nucl. Acids Res.
13
,
1939
1952
.
Migeon
,
B. R.
,
Johnson
,
G. G.
,
Wolf
,
S. F.
,
Axelman
,
J.
&
Schmidt
,
M.
(
1985
).
Hyperexpression of HPRT induced by 5-azacytidine in mouse-human hybrid reactivants
.
Am. J. hum. Genet.
37
,
608
611
.
Morimoto
,
K.
,
Dolan
,
M. E.
,
Scicchitano
,
D.
&
Pegg
,
A. E.
(
1985
).
Repair of O6-propyl- guanine and O6-butylguanine in DNA by O6-alkylguanine-DNA-alkyltransferase from rat liver and E
.
coli. Carcinogenesis
6
,
1027
1031
.
Rigby
,
P. W. J.
,
Dieckmann
,
M.
,
Rhodes
,
C.
&
Berg
,
P.
(
1977
).
Labeling deoxyribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase I
.
J. molec. Biol.
113
,
237
251
.
SAFFHILL
,
R.
(
1984
).
In vitro reaction of N-n-butyl-N-nitrosourea and n-butylmethane sulphonate with guanine and thymine bases of DNA
.
Carcinogenesis
5
,
621
625
.
Saffhill
,
R.
,
Margison
,
G. P.
&
O’Connor
,
P. J.
(
1985
).
Mechanisms of carcinogenesis induced by alkylating agents
.
Biochim. biophys. Acta
823
,
111
145
.
Saffhill
,
R.
,
Strickland
,
P. T.
&
Boyle
,
J. M.
(
1982
).
Sensitive radioimmunoassays for O6-nbutyl deoxyguanosine, O2-nbutyl thymidine and O4-nbutyl thymidine
.
Carcinogenesis
3
,
547
552
.
Scudiero
,
D. A.
,
Meyer
,
S. A.
,
Clatterbuck
,
B. E.
,
Mattern
,
M. R.
,
Ziolkowski
,
C. H. J.
&
Day
,
R. S.
(
1984b
).
Relationship of DNA repair phenotypes of human fibroblasts and tumor strains to killing by N-methyl-N’-nitro-N-nitrosoguanidine
.
Cancer Res.
44
,
961
969
.
Scudiero
,
D. A.
,
Meyer
,
S. A.
,
Clatterbuck
,
B. E.
,
Mattern
,
M. R.
,
Ziolkowski
,
C. H. J.
&
Day
,
R. S.
(
1984a
).
Sensitivity of human cell strains having different abilities to repair O6- methylguanine in DNA to inactivation by alkylating agents including chloroethyl nitrosoureas
.
Cancer Res.
44
,
2467
2474
.
Shin
,
S.
,
Caneva
,
R.
,
Schildkraut
,
C. L.
,
Klinger
,
H. P.
&
Siniscalo
,
M.
(
1973
).
Cells with phosphoribosyltransferase activity recovered from mouse cells resistant to 8-azaguanine
.
Nature, Lond.
241
,
194
196
.
Southern
,
E.
(
1975
).
Detection of specific sequences among DNA fragments by gel electrophoresis. J’
.
molec. Biol.
98
,
503
517
.
Squires
,
S.
,
Johnson
,
R. T.
&
Collins
,
A. R. S.
(
1982
).
Initial rates of DNA incision in UV- irradiated human cells. Differences between normal, xeroderma pigmentosum and tumour cells
.
Mutat. Res.
95
,
389
404
.
Teo
,
I. A.
,
Lehmann
,
A. R.
,
Muller
,
R.
&
Rajewsky
,
M. F.
(
1983
).
Similar rate of elimination of O6-ethylguanine elimination from DNA in normal human fibroblast and xeroderma pigmentosum cell strains not transformed by SV40
.
Carcinogenesis
4
,
1075
—1077.
Watson
,
B.
,
Gormley
,
I. P.
,
Gardiner
,
S. E.
,
Evans
,
H. J.
&
Harris
,
H.
(
1972
).
Reappearance of murine hypoxanthine guanine phosphoribosyl transferase activity in mouse A9 cells after attempted hybridisation with human cell lines
.
Expl Cell Res.
75
,
401
409
.
Wild
,
C. P.
,
Smart
,
G. M.
,
Saffhill
,
R.
&
Boyle
,
J. M.
(
1983
).
Radioimmunoassay of O6- methyl deoxyguanosine in DNA of cells alkylated in vitro
.
Carcinogenesis
4
,
1605
1609
.