- Email: [email protected]
Bleomycin and X-ray-hypersensitive Chinese hamster ovary cell mutants: Genetic analysis and cross-resistance to neocarzinostatin
157
Mutation Research, 193 (1988) 157-165 DNA Repair Reports
Elsevier MTR 06271
Bleomycin and X-ray-hypersensitive Chinese hamster ovary cell mutants: Genetic analysis and cross-resistance to neocarzinostatin C r a i g N. R o b s o n a, A n d r e w H a l l b, A d r i a n L. H a r r i s a a n d I a n D. H i c k s o n
a
Departments of Clinical Oncology a and Haematology b, University of Newcastle-upon-Tyne, Royal Victoria Infirmary, Newcastle-upon-Tyne, NE1 4LP (Great Britain)
(Received24 July 1987) (Revisionreceived21 October1987) (Accepted5 November1987)
Keywords: Chinesehamsterovarymutants; Bleomycin;X-Rayhypersensitivity;Neocarzinostatin;Geneticanalysis;Adriamycin-sen-
sitive.
Summary We have previously reported the isolation of 3 mutants of Chinese hamster ovary cells which exhibit hypersensitivity to bleomycin. 2 mutants were isolated on the basis of bleomycin-sensitivity [designated BLM-1 and BLM-2, Robson et al., Cancer Res., 45 (1985) 5304-5309] and 1 as adriamycin-sensitive [ADR-1, Robson et al., Cancer Res., 47 (1987) 1560-1565]. Because bleomycin generates DNA-strand breaks via a free-radical mechanism, we have studied the survival response of these mutants to a range of drugs which also generate free radicals and consequently DNA-strand breaks. The mutants are all hypersensitive to phleomycin, which differs from bleomycin in being unable to intercalate due to a modified bithiazole moiety. However, BLM-2 cells alone are hypersensitive to pepleomycin, a semi-synthetic bleomycin analogue. In contrast, BLM-1 cells are more sensitive than BLM-2 to streptonigrin (which operates via a hydroquinone intermediate). ADR-1 cells show wild-type resistance to streptonigrin. The results obtained with neocarzinostatin, an antibiotic requiring thiol activation, are unusual in that both BLM-1 and BLM-2 are approximately 3-fold more resistant than parental cells. However, the steady-state intracellular level of the major non-protein thiol, glutathione, is not altered in BLM-1 or BLM-2 cells. ADR-1 cells show essentially wild-type resistance to neocarzinostatin. Analysis of cell hybrids shows that BLM-1 and BLM-2 cells are phenotypically recessive in combination with parental CHO-K1 cells and represent different genetic complementation groups not only from one another, but also from the bleomycin-sensitive mutant xrs-6, isolated on the basis of X-ray sensitivity by Jeggo and Kemp [Mutation Res., 112 (1983) 313-319]. These results indicate that at least 3 gene products are involved in cellular protection against bleomycin toxicity in mammalian cells.
Correspondence: Dr. I.D. Hickson, Department of Clinical Ontology, University of Neweastle-upon-Tyne, The Royal VictoriaInfirmary,Newcasfle-upon-TyneNE1 4LP (Great Britain).
The bleomycins are a family of glycopeptide antitumour antibiotics originally isolated from Streptomyces verticillus by Umezawa et al. (1966). They are though to exert their cytotoxicity prin-
0167-8817/88/$03.50 © 1988 ElsevierSciencePublishersB.V.(BiomedicalDivision)
158 cipally by degrading DNA in a reaction requiring Fe(II) and 0 2 a s cofactors (Lown and Sim, 1977). There is good evidence that the ultimate DNAdamaging agent is a form of activated, reduced oxygen produced as a consequence of oxidation of bleomycin-chelated Fe(II) to Fe(III) (Pratviel et al., 1986; Giloni et al., 1981). Bleomycin interacts in a specific manner with double-stranded DNA, causing the release of DNA bases, and the formation of single- and double-strand breaks (Giloni et al., 1981). Fe(III): bleomycin is readily recycled to Fe(II): bleomycin by sulphydryl reagents, including glutathione (GSH), a likely candidate for the in vivo cofactor (Povirk, 1983). Clinically used bleomycin is a mixture of bleomycins A 2 and B2, together with several other minor components, but the DNA-cleavage pattern, and the frequency of induced single- and double-strand DNA breaks, is indistinguishable from that seen with purified bleomycin A 2 (Kross et al., 1982). The drug is active against a specific range of tumours - lymphoid neoplasms, teratomas and squamous tumours, but there are tissuespecific toxicities associated with its use, e.g. pulmonary fibrosis and Raynaud's phenomenon (Sikik et al., 1985). In order to analyse the normal basal mechanisms of cellular resistance to bleomycin, we have generated two CHO cell lines (designated BLM-1 and BLM-2) that are hypersensitive to killing by bleomycin (Robson et al., 1985). We have compared their sensitivities to a range of drugs known to generate DNA-strand breaks via a free-radical mechanism, and observe significant variation between different drugs. Both mutants are hypersensitive to the bleomycin analogue, phleomycin, but are more resistant than parental cells to neocarzinostatin (NCS). Because of the role of thiol-containing compounds in the activation of NCS, and in the protection of cells against freeradical damage, we measured the intracellular level of GSH in BLM-1 and BLM-2 cells. However, GSH levels are unaltered in both of these mutants. We also show that another bleomycin-sensitive mutant, ADR-1, which was isolated on the basis of adriamycin sensitivity, exhibits mild cross-sensitivity to certain bleomycin analogues. Genetic analysis of BLM-1 and BLM-2 shows that they are both phenotypically recessive in hy-
brids with parental cells, and that they lie in different complementation groups from one another and from the X-ray- and bleomycin-sensitive CHO cell line, xrs-6, previously isolated by Jeggo and Kemp (1983). Thus, at least 3 gene products are necessary for normal cellular resistance to bleomycin and its analogues. Materials and methods Cell-culture conditions
The parental cell line used in this study was CHO-K1. Cells were routinely maintained in Ham's F10 medium (Northumbria Biologicals) supplemented with 5% foetal bovine serum, 5% newborn calf serum, glutamine (3 mM), and penicillin, streptomycin, nystatin (100 U/ml, 100/~g/ml and 50 U/ml, respectively). Cells were grown as monolayers in a humidified atmosphere at 37 °C under 5% CO 2. Each line was regularly tested to confirm that it was mycoplasma-free. Mutant isolation
The isolation of mutants BLM-1, BLM-2 and ADR-1, together with their survival responses to bleomycin and X-rays, and their cloning efficiencies and doubling times, has been described previously (Robson et al., 1985, 1987). Survival curves
Exponentially growing cells were removed with 0.02% EDTA in PBS (lacking Ca 2+ and Mg 2+) and seeded in 100-mm petri dishes to yield 500-10000 cells per dish. The petri dishes were incubated for 4 h to allow cells to adhere. The cells were then treated with a DNA-damaging agent, as outlined below. Drug, treatment
All drugs were dissolved in sterile distilled water and stored at - 2 0 ° C in aliquots. Cells were exposed to drug for 24 h, before being washed twice with PBS and returned to drug-free medium. Colonies were allowed to develop for 10-12days before being fixed with methanol: acetic acid (3 : 1), stained with crystal violet (400/~g/ml) and counted. Colonies containing more than approximately 50 cells were judged to be survivors.
159
Each point on a survival curve represents the average of at least 3 independent experiments. In all cases, the parental CHO-K1 strain was treated in an identical manner to act as a control. All survival curves were fitted to the data by eye. The D37 value represents the dose of DNAdamaging agent required to reduce survival of a population to 37% of the untreated control and represents the average dose required to kill a cell.
100
.>_.
GSH measurements Intracellular GSH levels were measured by the method of Griffith (1980).
:~ 10
O3
Complementation analysis Cell hybrids were generated by polyethylene glycol-induced cell fusion, as described by Davidson and Gerald (1976). The selectable markers Ecogpt or neo were introduced into the cells to be fused, as previously described (Robson and Hickson, 1986). Results
Cell lines used in this study The mutants BLM-1 and -2 (previously described in Robson et al., 1985) have D37 values for killing by bleomycin of 0.44 and 0.22 /tg/ml, respectively, compared to 3.2 /tg/ml for parental CHO-KI cells. Mutant ADR-1 was isolated as adriamycin-sensitive and is hypersensitive to a range of topoisomerase II inhibitors, including anthracyclines, acridines and epipodophyllotoxins. However, it shows cross-sensitivity to bleomycin (D37 of 1.5/tg/ml: Robson et al., 1987). Mutant xrs-6, isolated on the basis of X-ray sensitivity, is also hypersensitive to bleomycin (Jeggo and Kemp, 1983). Cross-sensitivity to bleomycin analogues Cell sensitivities to the bleomycin analogues, phleomycin and pepleomycin, were compared. Phleomycin was chosen since it produces a much lower double strand : single strand break ratio than bleomycin (Huang et al., 1981). Pepleomycin was reported to produce less pulmonary toxicity than bleomycin (Sikik et al., 1980), suggesting a difference in tissue toxicity, the basis of which is unknown.
0
1
2
3
4
Phleomycin (/~g/ml) Fig. 1. Survival of parental CHO-KI and mutant cell lines to phleomycin. ©, CHO-K1; D, BLM-1; e, BLM-2; II, ADR-1. Points represent the mean. Bars, S.E.
The mutants BLM-1, BLM-2 and ADR-1 are all hypersensitive to phleomycin (Fig. 1), and show the same relative ranking in their sensitivity to phleomycin as they do to bleomycin. While ADR-1 cells show a similar degree of sensitivity to phleomycin as to bleomycin, BLM-1 and BLM-2 are significantly less phleomycin sensitive. Only mutant BLM-2, which shows the most striking sensitivity to bleomycin, is significantly hypersensitive to pepleomycin (approximately 3fold, as judged by/)37 values; Fig. 2).
Cross-sensitivity to streptonigrin Streptonigrin is an antitumour antibiotic isolated from Streptomyces flocculus which associates with double-stranded DNA in the presence of Zn 2÷ (White, 1977). Once bound, it is subject to an NADPH-mediated reductase reaction, converting it to a hydroquinone. The latter generates hydrogen peroxide, and superoxide and hydroxyl radicals, leading to the formation of DNA-strand breaks (Cone et al., 1976).
160
100
100
.>
-~ 10
-~10
11 0
i
I
1
2
~
i
i
3
4
1
0
ADR-1 cells, although adriamycin- and bleomycin-sensitive, are no more sensitive to streptonigrin than are CHO-K1 cells (Fig. 3). Also, the ranking in sensitivity to streptonigrin of BLM-1 and BLM-2 cells is reversed, with BLM-1 showing greater sensitivity (3-fold compared to 2-fold for BLM-2). Cross resistance to neocarzinostatin The antitumour antibiotic NCS consists of an apoprotein and a non-protein chromophore (Dasgupta and Goldberg, 1986). The latter is responsible for the antibiotic action, binding reversibly to DNA in the absence of sulphydryl reagents (Povirk and Goldberg, 1984). In the presence of sulphydryl groups, it is activated to produce sequence specific single and double DNA-strand breaks, the release of bases and the formation of a covalent adduct (Koppen and Goldberg, 1985). The mutant ADR-1 shows the wild-type level of sensitivity to NCS, while BLM-1 and BLM-2
20
30
40
Streptonigrin (ng/ml)
Pepleomycin (/ag/ml) Fig. 2. Survival of parental CHO-KI and mutant cell lines to pepleomycin. O, CHO-K1; E3, BLM-1; @, BLM-2; ram,ADR-1. Points represent the mean. Bars, S.E.
t0
Fig. 3. Survival of parental CHO-KI and mutant fines to streptonigrin. O, CHO-K1; [3, BLM-1; @, BLM-2; II, ADR-1. Points represent the mean. Bars, S.E.
are more than 3-fold resistant to NCS, as judged values (Fig. 4).
b y 037
Intracellular G S H levels GSH is known to play a role in cellular protection against free radicals, and in the activation of NCS. Because the bleomycin-sensitive, NCS-resistant phenotype of BLM-1 and BLM-2 could be a consequence of a reduction in cellular thiol content, we measured the level of GSH in these mutants. The results (Table 1) shows that the steady-state GSH level in BLM-1 and BLM-2 does not differ significantly from that in parental cells. Complementation analysis Complementation analysis of mutants BLM-1, BLM-2 and xrs-6 was performed by the method described previously (Robson and Hickson, 1986), in which selectable markers are introduced by
161
100
100
.¢ ~: 10 oO
I
0
20
I
I
40
60
I
80
1 0
DNA transfection into the cell lines to be fused. Fig. 5 shows bleomycin survival curves for the hybrids generated by fusing BLM-1 or BLM-2 cells with the parental line. The hybrids BLM1/CHO-K1 and BLM-2/CHO-K1 show wild-type resistance to bleomycin, indicating that both mutants are phenotypically recessive. The analysis of self-cross hybrids of BLM-1, BLM-2 and xrs-6, and of the hybrids generated by fusing different combinations of these 3 mutants, is shown in Fig. 6. The hybrids BLM-2/BLM-2 TABLE 1
2
3
4
Bleomycin (,ug/ml)
Neoearzinostatin (ng/ml) Fig. 4. Survival of parental CHO-K1 and mutant cell lines to NCS. O, CHO-K1; n, BLM-1; o, BLM-2; II, ADR-1. Points represent the mean. Bars, S.E.
1
Fig. 5. Survival of CHO-Kl-containing hybrids, and of diploid controls, to bleomycin, zx, BLM-1; - , BLM-2; O, CHO-K1; n, CHO-K1/CHO-KI; e, BLM-1/CHO-KI; II, BLM-2/CHO-KI. Points represent the mean. Bars, S.E.
and xrs-6/xrs-6 show mutant levels of bleomycinsensitivity. However, the pairing BLM-1/BLM-1 shows a significantly enhanced resistance to bleomycin compared with that of BLM-1 cells alone, being only 2.5-fold more sensitive than parental cells compared to 7-fold for BLM-1 cells. The hybrids generated by fusing all combinations of the mutants BLM-1, BLM-2 and xrs-6 show near wild-type resistance to bleomycin (Fig. 6), indicating that the mutants lie in 3 different genetic complementation groups with respect to bleomycin-hypersensitivity.
INTRACELLULAR GSH LEVELS
Discussion Cell line
GSH level ( m o l e s / n a g soluble protein)
CHO-K1 BLM-1 BLM-2
20.2 19.1 20.3
Values represent the mean of 4 determinations.
The cross-sensitivity profiles and the complementation data show that the biochemical basis of basal resistance to bleomycin and its analogues is complex in mammalian cells.
162
100
. m
10 o~
0
I
I
I
I
1
2
3
4
Bleomycin (,ug/m II Fig. 6. Survival of hybrids of mutant cells, and of diploid controls, to bleomycin. A, xrs-6/xrs-6; v, BLM-1/BLM-1; m, BLM-2/BLM-2; O, BLM-1/BLM-2; Q, xrs-6/BLM-1; zx, xrs6/BLM-2; O, CHO-KI.
Only small amounts of bleomycin enter cells, suggesting that differences in uptake and subcellular localisation may affect toxicity (Roy and Horwitz, 1984). Alterations in membrane transport may account for some of the variations in toxicity between the bleomycin analogues studied here, but would not be expected to lead to X-ray sensitivity in the case of BLM-2, or NCS resistance in either BLM-1 or BLM-2. In addition, drug transport in ADR-1 cells is known to be comparable with that in wild-type cells (Robson et al., 1987). The mutant BLM-2 is hypersensitive to all of the bleomycin analogues tested. It is also the only mutant hypersensitive to X-rays. These results suggest that BLM-2 may be defective in DNAstrand break repair, and this has been confirmed (unpublished results). ADR-1 and BLM-1 show no apparent defect in DNA-strand break repair (Robson et al., 1987 and unpublished results).
The different cross-sensitivity patterns for bleomycin and its analogues shows the potential value of such mutants for demonstrating the differences in action between closely related drugs. Phleomycin differs from bleomycin in having a reduced bithiazole moiety. In vitro, when equalised for single-strand breakage, there are 8-fold fewer double-strand breaks generated by phleomycin than by bleomycin (Huang et al., 1981). Sequence specificity, however, remains the same (Kross et al., 1982). In spite of this marked difference in the ratio of single- to double-strand breaks, the 3 mutants show the same ranking in their sensitivity to phleomycin as they do to bleomycin. Since the bithiazole ring can intercalate in bleomycin (Povirk et al., 1979), but is unable to do so in the reduced form in phleomycin, it appears that this moiety is not the basis of the sequence specificity of DNA cleavage by the bleomycin analogues (Takeshita et al., 1981). The essentially wild-type resistance of BLM-1 and ADR-1 cells to pepleomycin, compared with hypersensitivity to phleomycin and bleomycin, suggests that these latter 2 drugs are killing the cells by a mechanism different from that of pepleomycin. Pepleomycin was developed as a semisynthetic, less toxic analogue of bleomycin and generally produces less pulmonary fibrosis (Ginsburg et al., 1984). However, it crosses the blood-brain barrier, and is more nephrotoxic and neurotoxic (Sikic et al., 1980). It has been shown that several membrane active pharmacological agents enhance pepleomycin toxicity, apparently due to increased intracellular calcium (Mizuno and Ishida, 1982a,b), suggesting the involvement of a non-DNA target. Similarly, modification and enhancement of the toxicity of bleomycin by dibucaine does not correlate with increased DNA damage (Berry et al., 1985). Also, treatment of CHO cells with 0.025% trypsin for 5 min is associated with an up to 50-fold increase in resistance to bleomycin given 1 h later, suggesting a membrane target may have been modified (Barranco et al., 1980). If bleomycin and its analogues kill cells by a DNA-dependent and a DNA-independent mechanism, these mutants may be able to elucidate the mechanisms. ADR-1 is hypersensitive to a range of topoisomerase II inhibitors, including the intercalators
163 m-AMSA, adriamycin and mitoxantrone. We previously showed that ADR-1 is also bleomycin-sensitive (Robson et al., 1987) and suggested that this may be due to intercalation of the bithiazole moiety. However, there is no cross-sensitivity to pepleomycin which has the same bithiazole moiety but lacks a charged terminal amine. This charge has been shown to be necessary for intercalation, by interacting with negatively charged phosphate groups (Lin and Gallman, 1981; Sakai et al., 1983). Since phleomycin does not appear to intercalate, but the mutant ADR-1 is still cross-sensitive, it suggests that this sensitivity is not related to intercalation per se, although it may be due to a type of DNA damage that involves topoisomerase II in repair. The cross-sensitivity patterns to streptonigrin show that BLM-2 is less sensitive to this free-radical generating antibiotic than is BLM-1. ADR-1 is not hypersensitive to streptonigrin, implying that DNA binding and free-radical generation do not necessarily produce hypersensitivity in this mutant. This suggest that streptonigrin is producing a different spectrum of lethal lesions from bleomycin, possibly related to differences in quinone reductase, essential for the activation of streptonigrin (Bachur et al., 1978). It is more likely that in BLM-1 there is a difference in free-radical detoxification, since this line is also hypersensitive to H202 (unpublished observation). NCS was studied because of its different sequence specificity (A-T residues) and different type of base damage, a 5'-terminal nucleoside aldehyde (Kappen and Goldberg, 1985). The cross resistance of two of the mutants to NCS is particularly surprising in the case of BLM-2, which has a defect in DNA-strand break repair (unpublished result). The NCS resistance of these mutants appears to result from a lower accumulation of DNA-strand breaks after NCS treatment compared to parental cells (unpublished observation). Thiols are essential for activation of the NCS chromophore, and NCS toxicity is greatly reduced in GSH-depleted cells (De Graf et al., 1985). Although the phenotype of BLM-1 and BLM-2 points to a possible alteration in cellular thiol content, the steady-state level of GSH is apparently unaltered in both of these mutants. Whether these mutants are defective in a GSH-de-
pendent enzyme, such as GSH reductase, leading to a failure of the cells to recover from oxidative stress, remains to be determined. In a recent study, 99.9% depletion of GSH left a residual DNA breakage 33% of that in cells with normal GSH levels. Some breaks were protein associated, and it was suggested that residual breaks may have been due to endonucleolytic activity or to topoisomerase II (Kapper et al., 1987). Therefore, other, as yet undefined, factors may be important in producing DNA strand breakage in the presence of NCS. The genetic analyses show that BLM-1, BLM-2 and xrs-6 cells represent different complementation groups. Thus, at least 3 gene products are required for efficient cellular protection against bleomycin-toxicity. Recently, Stamato et al. (1987) have isolated a bank of bleomycin-sensitive hamster mutants which, in certain cases, appear to have a profile of cross-sensitivities unlike those of BLM-1 and -2. These new mutants may, therefore, include more genetic classes. Although the hybrids xrs-6/xrs-6 and BLM2/BLM-2 show approximately diploid mutant levels of hypersensitivity to bleomycin, the pairing BLM-1/BLM-1 is significantly more resistant than BLM-1 cells alone. The reason for this is unknown. The increased cell volume to surfa.ce area ratio of the hybrids could lead to decreased drug uptake, while their longer doubling time could lead to a decreased sensitivity to a cell-cycle specific drug such as bleomycin. However, why this should particularly affect the BLM-1/BLM-1 hybrid is unclear. The mutants that we have isolated show major differences in their cell survival characteristics following treatment with various bleomycin analogues, and dearly possess different genetic alterations. They should prove to be suitable recipients for DNA transfection to isolate the human genes responsible for basal resistance to bleomycin. They may also be useful for a biochemical investigation of the mechanisms of cell killing by bleomycin, and for the screening of new analogues.
Acknowledgements We thank Dr. P. Jeggo for the kind use of the xrs-6 cell line, and the North of England Cancer Research Campaign for financial support.
164
References Bachur, N.R., S.L. Gordon and M.V. Gee (1978) A general mechanism for microsomal activation of quinone anticancer agents to free radicals, Cancer Res., 38, 1745-1750. Bachur, N.R., M.V. Gee and R.D. Friedman (1982) Nuclear catalyzed antibiotic free radical formation, Cancer Res., 42, 1078-1081. Barranco, S.C., W.E. Bolton and J.K. Novak (1980) Time-dependent changes in drug sensitivity expressed by mammalian cells after exposure to trypsin, J. Natl. Cancer Inst., 64, 913-916. Berry, D.E., R.E. Kilkuskie and S.M. Hecht (1985) DNA damage induced by bleomycin in the presence of dibucaine is not predictive of cell growth inhibition, Biochemistry, 24, 3214-3219. Cone, R., S.K. Hasan, J.W. Lown and A.R. Morgan (1976) The mechanism of degradation of DNA by streptonigrin, Can. J. Biochem., 56, 1042-1047. Dasgupta, D., and I.H. Goldberg (1986) Mode of reversible binding of neocarzinostatin chromophore to DNA: base sequence dependency of binding, Nucl. Acids Res., 14, 1089-1105. Davidson, R.L., and P.S. Gerald (1976) Improved techniques for the induction of mammalian cell hybridisation by polyethylene glycol, Somat. Cell Genet., 2, 165-176. DeGraff, W.G., A. Russo and J.B. Mitchell (1985) Glutathione depletion greatly reduces neocarzinostatin cytotoxicity in Chinese hamster V79 cells, J. Biol. Chem., 260, 8312-8315. Giloni, L., M. Takeshita, F. Johnson, C. Iden and A.P. Grollman (1981) Bleomycin-induced strand scission of DNA, J. Biol. Chem., 256, 8608-8615. Ginsburg, E., T.E. Gram and M.A. Trush (1984) A comparison of the pulmonary toxicity and chemotherapeutic activity of bleomycin-BAPP to bleomycin and pepleomycin, Cancer Chemother. Pharmacol., 12, 111-115. Griffith, O.W. (1980) Determination of glutathione and glutathione disulfide using glutathione reductase and 2vinylpyridine, Anal. Biochem., 106, 207-212. Huang, C.-H., C.K. Mirabelli, Y. Jan and S.T. Crooke (1981) Single-strand and double-strand deoxyribonucleic acid breaks produced by several bleomycin analogues, Biochemistry, 20, 233-237. Jeggo, P.A., and L.H. Kemp (1983) X-Ray sensitive mutants of a Chinese hamster ovary cell line: Isolation and cross-sensitivity to other DNA damaging agents, Mutation Res., 112, 313-319. Kappen, L.S., and I.H. Goldberg (1985) Activation of neocarzinostatin chromophore and formation of nascent DNA damage do not require molecular oxygen, Nucl. Acids Res., 13, 1637-1648. Kappen, L.S., T.E. EUenberger and I.H. Goldberg (1987) Mechanism and base specificity of DNA breakage in intact cells by neocarzinostatin, Biochemistry, 26, 384-390. Kilkuskie, R.E., T.L. Macdonald and S.M. Hecht (1984) Bleomycin may be activated for DNA cleavage by NADPH-cytochrome P-450 reductase, Biochemistry, 23, 6165-6171.
Kross, J., W.D. Henner, S.M. Hecht and W.A. Haseltine (1982) Specificity of deoxyribonucleic acid cleavage by bleomycin, phleomycin and tallysomycin, Biochemistry, 21, 4310-4318. Lin, S.Y., and A.P. Grollman (1981) Interactions of a fragment of bleomycin with deoxyribodinucleotides: nuclear magnetic resonance studies, Biochemistry, 20, 7589-7598. Lown, J.W., and S.K. Sim (1977) The mechanism of bleomycin-induced cleavage of DNA, Biochem. Biophys. Res. Commun., 77, 1150-1157. Mizuno, S., and A. Ishida (1982) Potentiation of bleomycin cytotoxicity by membrane-interacting drugs and increased calcium ions, Biochem. Biophys. Res. Commun., 107, 1021-1027. Murray, V., and R.F. Martin (1985) The sequence specificity of bleomycin-induced DNA damage in intact cells, J. Biol. Chem., 260, 10389-10391. Ozawa, S., et al. (1985) Mechanisms affecting pepleomycin sensitivity of Chinese hamster cell lines, J. Antibiotics, 38, 1257-1264. Povirk, L.F. (1979) Catalytic release of DNA bases by oxidation and reduction of an iron-bleomycin complex, Binchemistry, 18, 3989-3995. Povirk, L.F., and I.H. Goldberg (1984) Competition between anaerobic covalent linkage of neocarzinostatin chromophore to deoxyribose in DNA and oxygen-dependent strand breakage and base release, Biochemistry, 23, 6304-6311. Povirk, L.F., M. Hogan and N. Dattagupta (1979) Binding of bleomycin to DNA: intercalation of the bithiazole rings, Biochemistry, 18, 96-101. Pratviel, G., J. Bernadou and B. Meunier (1986) DNA breaks generated by the bleomycin-iron III complex in the presence of KHSO 5, a single oxygen donor, Biochem. Biophys. Res. Commun., 136, 1013-1020. Robson, C.N., and I.D. Hickson (1986) Genetic analysis of mitomycin C-sensitive mutants of a Chinese hamster ovary cell line, Mutation Res., 163, 201-208. Robson, C.N., A.L. Harris and I.D. Hickson (1985) Isolation and characterisation of Chinese hamster ovary cell lines sensitive to mitomycin C and bleomycin, Cancer Res., 45, 5304-5309. Robson, C.N., P.R. Hoban, A.L. Harris and I.D. Hickson (1987) Cross-sensitivity to topoisomerase II inhibitors in cytotoxic drug hypersensitive Chinese hamster ovary cell lines, Cancer Res., 47, 1560-1565. Roy, S.N., and S.B. Horwitz (1984) Characterization of the association of radiolabeled bleomycin A : with HeLa cells, Cancer Res., 44, 1541-1546. Sakai, T.T., J.M. Riordan and J.D. Glickson (1983) Bleomy¢in interactions with DNA, Studies on the role of the C-terminal cationic group of bleomycin A 2 in association with and degradation of DNA, Biochim. Biophys. Acta, 758, 176-180. Sikik, B.I., Z.H. Siddik and T.E. Gram (1980) Relative pulmonary toxicity and antitumor effects of two new bleomycin analogs, pepleomycin and tallysomycin Aa, Cancer treat. Rep., 64, 659-667.
165 Sikik, B.I., M. Rozensweig and S.K. Carter (Eds.) (1985) Bleomycin Chemotherapy, Academic Press, New York. Stamato, T.D., B. Peters, P. Patil, N. Denko, R. Weinstein and A. Giaccia (1987) Isolation and characterisation of bleomycin-sensitiveChinese hamster ovary cell lines, Cancer Res., 47, 1588-1592. Takeshita, M., L.S. Kappen, A.P. Grollman, M. Eisenberg and I.H. Goldberg (1981) Strand seission of deoxyribonucleic acid by neocarzinostatin, auromomycin and bleomycin:
studies on base release and nucleotide sequence specificity, Biochemistry, 20, 7599-7606. Umezawa, H., K. Maeda, T. Takeuchi and Y. Akami (1966) New antibiotics bleomycin A and B, J. Antibiot., Ser. A, 19, 200-209. White, J.R. (1977) Streptonigrin-transition metal complexes: binding to DNA and biological activity, Biochem. Biophys. Res. Commun., 77, 387-391.
Mutation Research, 193 (1988) 157-165 DNA Repair Reports
Elsevier MTR 06271
Bleomycin and X-ray-hypersensitive Chinese hamster ovary cell mutants: Genetic analysis and cross-resistance to neocarzinostatin C r a i g N. R o b s o n a, A n d r e w H a l l b, A d r i a n L. H a r r i s a a n d I a n D. H i c k s o n
a
Departments of Clinical Oncology a and Haematology b, University of Newcastle-upon-Tyne, Royal Victoria Infirmary, Newcastle-upon-Tyne, NE1 4LP (Great Britain)
(Received24 July 1987) (Revisionreceived21 October1987) (Accepted5 November1987)
Keywords: Chinesehamsterovarymutants; Bleomycin;X-Rayhypersensitivity;Neocarzinostatin;Geneticanalysis;Adriamycin-sen-
sitive.
Summary We have previously reported the isolation of 3 mutants of Chinese hamster ovary cells which exhibit hypersensitivity to bleomycin. 2 mutants were isolated on the basis of bleomycin-sensitivity [designated BLM-1 and BLM-2, Robson et al., Cancer Res., 45 (1985) 5304-5309] and 1 as adriamycin-sensitive [ADR-1, Robson et al., Cancer Res., 47 (1987) 1560-1565]. Because bleomycin generates DNA-strand breaks via a free-radical mechanism, we have studied the survival response of these mutants to a range of drugs which also generate free radicals and consequently DNA-strand breaks. The mutants are all hypersensitive to phleomycin, which differs from bleomycin in being unable to intercalate due to a modified bithiazole moiety. However, BLM-2 cells alone are hypersensitive to pepleomycin, a semi-synthetic bleomycin analogue. In contrast, BLM-1 cells are more sensitive than BLM-2 to streptonigrin (which operates via a hydroquinone intermediate). ADR-1 cells show wild-type resistance to streptonigrin. The results obtained with neocarzinostatin, an antibiotic requiring thiol activation, are unusual in that both BLM-1 and BLM-2 are approximately 3-fold more resistant than parental cells. However, the steady-state intracellular level of the major non-protein thiol, glutathione, is not altered in BLM-1 or BLM-2 cells. ADR-1 cells show essentially wild-type resistance to neocarzinostatin. Analysis of cell hybrids shows that BLM-1 and BLM-2 cells are phenotypically recessive in combination with parental CHO-K1 cells and represent different genetic complementation groups not only from one another, but also from the bleomycin-sensitive mutant xrs-6, isolated on the basis of X-ray sensitivity by Jeggo and Kemp [Mutation Res., 112 (1983) 313-319]. These results indicate that at least 3 gene products are involved in cellular protection against bleomycin toxicity in mammalian cells.
Correspondence: Dr. I.D. Hickson, Department of Clinical Ontology, University of Neweastle-upon-Tyne, The Royal VictoriaInfirmary,Newcasfle-upon-TyneNE1 4LP (Great Britain).
The bleomycins are a family of glycopeptide antitumour antibiotics originally isolated from Streptomyces verticillus by Umezawa et al. (1966). They are though to exert their cytotoxicity prin-
0167-8817/88/$03.50 © 1988 ElsevierSciencePublishersB.V.(BiomedicalDivision)
158 cipally by degrading DNA in a reaction requiring Fe(II) and 0 2 a s cofactors (Lown and Sim, 1977). There is good evidence that the ultimate DNAdamaging agent is a form of activated, reduced oxygen produced as a consequence of oxidation of bleomycin-chelated Fe(II) to Fe(III) (Pratviel et al., 1986; Giloni et al., 1981). Bleomycin interacts in a specific manner with double-stranded DNA, causing the release of DNA bases, and the formation of single- and double-strand breaks (Giloni et al., 1981). Fe(III): bleomycin is readily recycled to Fe(II): bleomycin by sulphydryl reagents, including glutathione (GSH), a likely candidate for the in vivo cofactor (Povirk, 1983). Clinically used bleomycin is a mixture of bleomycins A 2 and B2, together with several other minor components, but the DNA-cleavage pattern, and the frequency of induced single- and double-strand DNA breaks, is indistinguishable from that seen with purified bleomycin A 2 (Kross et al., 1982). The drug is active against a specific range of tumours - lymphoid neoplasms, teratomas and squamous tumours, but there are tissuespecific toxicities associated with its use, e.g. pulmonary fibrosis and Raynaud's phenomenon (Sikik et al., 1985). In order to analyse the normal basal mechanisms of cellular resistance to bleomycin, we have generated two CHO cell lines (designated BLM-1 and BLM-2) that are hypersensitive to killing by bleomycin (Robson et al., 1985). We have compared their sensitivities to a range of drugs known to generate DNA-strand breaks via a free-radical mechanism, and observe significant variation between different drugs. Both mutants are hypersensitive to the bleomycin analogue, phleomycin, but are more resistant than parental cells to neocarzinostatin (NCS). Because of the role of thiol-containing compounds in the activation of NCS, and in the protection of cells against freeradical damage, we measured the intracellular level of GSH in BLM-1 and BLM-2 cells. However, GSH levels are unaltered in both of these mutants. We also show that another bleomycin-sensitive mutant, ADR-1, which was isolated on the basis of adriamycin sensitivity, exhibits mild cross-sensitivity to certain bleomycin analogues. Genetic analysis of BLM-1 and BLM-2 shows that they are both phenotypically recessive in hy-
brids with parental cells, and that they lie in different complementation groups from one another and from the X-ray- and bleomycin-sensitive CHO cell line, xrs-6, previously isolated by Jeggo and Kemp (1983). Thus, at least 3 gene products are necessary for normal cellular resistance to bleomycin and its analogues. Materials and methods Cell-culture conditions
The parental cell line used in this study was CHO-K1. Cells were routinely maintained in Ham's F10 medium (Northumbria Biologicals) supplemented with 5% foetal bovine serum, 5% newborn calf serum, glutamine (3 mM), and penicillin, streptomycin, nystatin (100 U/ml, 100/~g/ml and 50 U/ml, respectively). Cells were grown as monolayers in a humidified atmosphere at 37 °C under 5% CO 2. Each line was regularly tested to confirm that it was mycoplasma-free. Mutant isolation
The isolation of mutants BLM-1, BLM-2 and ADR-1, together with their survival responses to bleomycin and X-rays, and their cloning efficiencies and doubling times, has been described previously (Robson et al., 1985, 1987). Survival curves
Exponentially growing cells were removed with 0.02% EDTA in PBS (lacking Ca 2+ and Mg 2+) and seeded in 100-mm petri dishes to yield 500-10000 cells per dish. The petri dishes were incubated for 4 h to allow cells to adhere. The cells were then treated with a DNA-damaging agent, as outlined below. Drug, treatment
All drugs were dissolved in sterile distilled water and stored at - 2 0 ° C in aliquots. Cells were exposed to drug for 24 h, before being washed twice with PBS and returned to drug-free medium. Colonies were allowed to develop for 10-12days before being fixed with methanol: acetic acid (3 : 1), stained with crystal violet (400/~g/ml) and counted. Colonies containing more than approximately 50 cells were judged to be survivors.
159
Each point on a survival curve represents the average of at least 3 independent experiments. In all cases, the parental CHO-K1 strain was treated in an identical manner to act as a control. All survival curves were fitted to the data by eye. The D37 value represents the dose of DNAdamaging agent required to reduce survival of a population to 37% of the untreated control and represents the average dose required to kill a cell.
100
.>_.
GSH measurements Intracellular GSH levels were measured by the method of Griffith (1980).
:~ 10
O3
Complementation analysis Cell hybrids were generated by polyethylene glycol-induced cell fusion, as described by Davidson and Gerald (1976). The selectable markers Ecogpt or neo were introduced into the cells to be fused, as previously described (Robson and Hickson, 1986). Results
Cell lines used in this study The mutants BLM-1 and -2 (previously described in Robson et al., 1985) have D37 values for killing by bleomycin of 0.44 and 0.22 /tg/ml, respectively, compared to 3.2 /tg/ml for parental CHO-KI cells. Mutant ADR-1 was isolated as adriamycin-sensitive and is hypersensitive to a range of topoisomerase II inhibitors, including anthracyclines, acridines and epipodophyllotoxins. However, it shows cross-sensitivity to bleomycin (D37 of 1.5/tg/ml: Robson et al., 1987). Mutant xrs-6, isolated on the basis of X-ray sensitivity, is also hypersensitive to bleomycin (Jeggo and Kemp, 1983). Cross-sensitivity to bleomycin analogues Cell sensitivities to the bleomycin analogues, phleomycin and pepleomycin, were compared. Phleomycin was chosen since it produces a much lower double strand : single strand break ratio than bleomycin (Huang et al., 1981). Pepleomycin was reported to produce less pulmonary toxicity than bleomycin (Sikik et al., 1980), suggesting a difference in tissue toxicity, the basis of which is unknown.
0
1
2
3
4
Phleomycin (/~g/ml) Fig. 1. Survival of parental CHO-KI and mutant cell lines to phleomycin. ©, CHO-K1; D, BLM-1; e, BLM-2; II, ADR-1. Points represent the mean. Bars, S.E.
The mutants BLM-1, BLM-2 and ADR-1 are all hypersensitive to phleomycin (Fig. 1), and show the same relative ranking in their sensitivity to phleomycin as they do to bleomycin. While ADR-1 cells show a similar degree of sensitivity to phleomycin as to bleomycin, BLM-1 and BLM-2 are significantly less phleomycin sensitive. Only mutant BLM-2, which shows the most striking sensitivity to bleomycin, is significantly hypersensitive to pepleomycin (approximately 3fold, as judged by/)37 values; Fig. 2).
Cross-sensitivity to streptonigrin Streptonigrin is an antitumour antibiotic isolated from Streptomyces flocculus which associates with double-stranded DNA in the presence of Zn 2÷ (White, 1977). Once bound, it is subject to an NADPH-mediated reductase reaction, converting it to a hydroquinone. The latter generates hydrogen peroxide, and superoxide and hydroxyl radicals, leading to the formation of DNA-strand breaks (Cone et al., 1976).
160
100
100
.>
-~ 10
-~10
11 0
i
I
1
2
~
i
i
3
4
1
0
ADR-1 cells, although adriamycin- and bleomycin-sensitive, are no more sensitive to streptonigrin than are CHO-K1 cells (Fig. 3). Also, the ranking in sensitivity to streptonigrin of BLM-1 and BLM-2 cells is reversed, with BLM-1 showing greater sensitivity (3-fold compared to 2-fold for BLM-2). Cross resistance to neocarzinostatin The antitumour antibiotic NCS consists of an apoprotein and a non-protein chromophore (Dasgupta and Goldberg, 1986). The latter is responsible for the antibiotic action, binding reversibly to DNA in the absence of sulphydryl reagents (Povirk and Goldberg, 1984). In the presence of sulphydryl groups, it is activated to produce sequence specific single and double DNA-strand breaks, the release of bases and the formation of a covalent adduct (Koppen and Goldberg, 1985). The mutant ADR-1 shows the wild-type level of sensitivity to NCS, while BLM-1 and BLM-2
20
30
40
Streptonigrin (ng/ml)
Pepleomycin (/ag/ml) Fig. 2. Survival of parental CHO-KI and mutant cell lines to pepleomycin. O, CHO-K1; E3, BLM-1; @, BLM-2; ram,ADR-1. Points represent the mean. Bars, S.E.
t0
Fig. 3. Survival of parental CHO-KI and mutant fines to streptonigrin. O, CHO-K1; [3, BLM-1; @, BLM-2; II, ADR-1. Points represent the mean. Bars, S.E.
are more than 3-fold resistant to NCS, as judged values (Fig. 4).
b y 037
Intracellular G S H levels GSH is known to play a role in cellular protection against free radicals, and in the activation of NCS. Because the bleomycin-sensitive, NCS-resistant phenotype of BLM-1 and BLM-2 could be a consequence of a reduction in cellular thiol content, we measured the level of GSH in these mutants. The results (Table 1) shows that the steady-state GSH level in BLM-1 and BLM-2 does not differ significantly from that in parental cells. Complementation analysis Complementation analysis of mutants BLM-1, BLM-2 and xrs-6 was performed by the method described previously (Robson and Hickson, 1986), in which selectable markers are introduced by
161
100
100
.¢ ~: 10 oO
I
0
20
I
I
40
60
I
80
1 0
DNA transfection into the cell lines to be fused. Fig. 5 shows bleomycin survival curves for the hybrids generated by fusing BLM-1 or BLM-2 cells with the parental line. The hybrids BLM1/CHO-K1 and BLM-2/CHO-K1 show wild-type resistance to bleomycin, indicating that both mutants are phenotypically recessive. The analysis of self-cross hybrids of BLM-1, BLM-2 and xrs-6, and of the hybrids generated by fusing different combinations of these 3 mutants, is shown in Fig. 6. The hybrids BLM-2/BLM-2 TABLE 1
2
3
4
Bleomycin (,ug/ml)
Neoearzinostatin (ng/ml) Fig. 4. Survival of parental CHO-K1 and mutant cell lines to NCS. O, CHO-K1; n, BLM-1; o, BLM-2; II, ADR-1. Points represent the mean. Bars, S.E.
1
Fig. 5. Survival of CHO-Kl-containing hybrids, and of diploid controls, to bleomycin, zx, BLM-1; - , BLM-2; O, CHO-K1; n, CHO-K1/CHO-KI; e, BLM-1/CHO-KI; II, BLM-2/CHO-KI. Points represent the mean. Bars, S.E.
and xrs-6/xrs-6 show mutant levels of bleomycinsensitivity. However, the pairing BLM-1/BLM-1 shows a significantly enhanced resistance to bleomycin compared with that of BLM-1 cells alone, being only 2.5-fold more sensitive than parental cells compared to 7-fold for BLM-1 cells. The hybrids generated by fusing all combinations of the mutants BLM-1, BLM-2 and xrs-6 show near wild-type resistance to bleomycin (Fig. 6), indicating that the mutants lie in 3 different genetic complementation groups with respect to bleomycin-hypersensitivity.
INTRACELLULAR GSH LEVELS
Discussion Cell line
GSH level ( m o l e s / n a g soluble protein)
CHO-K1 BLM-1 BLM-2
20.2 19.1 20.3
Values represent the mean of 4 determinations.
The cross-sensitivity profiles and the complementation data show that the biochemical basis of basal resistance to bleomycin and its analogues is complex in mammalian cells.
162
100
. m
10 o~
0
I
I
I
I
1
2
3
4
Bleomycin (,ug/m II Fig. 6. Survival of hybrids of mutant cells, and of diploid controls, to bleomycin. A, xrs-6/xrs-6; v, BLM-1/BLM-1; m, BLM-2/BLM-2; O, BLM-1/BLM-2; Q, xrs-6/BLM-1; zx, xrs6/BLM-2; O, CHO-KI.
Only small amounts of bleomycin enter cells, suggesting that differences in uptake and subcellular localisation may affect toxicity (Roy and Horwitz, 1984). Alterations in membrane transport may account for some of the variations in toxicity between the bleomycin analogues studied here, but would not be expected to lead to X-ray sensitivity in the case of BLM-2, or NCS resistance in either BLM-1 or BLM-2. In addition, drug transport in ADR-1 cells is known to be comparable with that in wild-type cells (Robson et al., 1987). The mutant BLM-2 is hypersensitive to all of the bleomycin analogues tested. It is also the only mutant hypersensitive to X-rays. These results suggest that BLM-2 may be defective in DNAstrand break repair, and this has been confirmed (unpublished results). ADR-1 and BLM-1 show no apparent defect in DNA-strand break repair (Robson et al., 1987 and unpublished results).
The different cross-sensitivity patterns for bleomycin and its analogues shows the potential value of such mutants for demonstrating the differences in action between closely related drugs. Phleomycin differs from bleomycin in having a reduced bithiazole moiety. In vitro, when equalised for single-strand breakage, there are 8-fold fewer double-strand breaks generated by phleomycin than by bleomycin (Huang et al., 1981). Sequence specificity, however, remains the same (Kross et al., 1982). In spite of this marked difference in the ratio of single- to double-strand breaks, the 3 mutants show the same ranking in their sensitivity to phleomycin as they do to bleomycin. Since the bithiazole ring can intercalate in bleomycin (Povirk et al., 1979), but is unable to do so in the reduced form in phleomycin, it appears that this moiety is not the basis of the sequence specificity of DNA cleavage by the bleomycin analogues (Takeshita et al., 1981). The essentially wild-type resistance of BLM-1 and ADR-1 cells to pepleomycin, compared with hypersensitivity to phleomycin and bleomycin, suggests that these latter 2 drugs are killing the cells by a mechanism different from that of pepleomycin. Pepleomycin was developed as a semisynthetic, less toxic analogue of bleomycin and generally produces less pulmonary fibrosis (Ginsburg et al., 1984). However, it crosses the blood-brain barrier, and is more nephrotoxic and neurotoxic (Sikic et al., 1980). It has been shown that several membrane active pharmacological agents enhance pepleomycin toxicity, apparently due to increased intracellular calcium (Mizuno and Ishida, 1982a,b), suggesting the involvement of a non-DNA target. Similarly, modification and enhancement of the toxicity of bleomycin by dibucaine does not correlate with increased DNA damage (Berry et al., 1985). Also, treatment of CHO cells with 0.025% trypsin for 5 min is associated with an up to 50-fold increase in resistance to bleomycin given 1 h later, suggesting a membrane target may have been modified (Barranco et al., 1980). If bleomycin and its analogues kill cells by a DNA-dependent and a DNA-independent mechanism, these mutants may be able to elucidate the mechanisms. ADR-1 is hypersensitive to a range of topoisomerase II inhibitors, including the intercalators
163 m-AMSA, adriamycin and mitoxantrone. We previously showed that ADR-1 is also bleomycin-sensitive (Robson et al., 1987) and suggested that this may be due to intercalation of the bithiazole moiety. However, there is no cross-sensitivity to pepleomycin which has the same bithiazole moiety but lacks a charged terminal amine. This charge has been shown to be necessary for intercalation, by interacting with negatively charged phosphate groups (Lin and Gallman, 1981; Sakai et al., 1983). Since phleomycin does not appear to intercalate, but the mutant ADR-1 is still cross-sensitive, it suggests that this sensitivity is not related to intercalation per se, although it may be due to a type of DNA damage that involves topoisomerase II in repair. The cross-sensitivity patterns to streptonigrin show that BLM-2 is less sensitive to this free-radical generating antibiotic than is BLM-1. ADR-1 is not hypersensitive to streptonigrin, implying that DNA binding and free-radical generation do not necessarily produce hypersensitivity in this mutant. This suggest that streptonigrin is producing a different spectrum of lethal lesions from bleomycin, possibly related to differences in quinone reductase, essential for the activation of streptonigrin (Bachur et al., 1978). It is more likely that in BLM-1 there is a difference in free-radical detoxification, since this line is also hypersensitive to H202 (unpublished observation). NCS was studied because of its different sequence specificity (A-T residues) and different type of base damage, a 5'-terminal nucleoside aldehyde (Kappen and Goldberg, 1985). The cross resistance of two of the mutants to NCS is particularly surprising in the case of BLM-2, which has a defect in DNA-strand break repair (unpublished result). The NCS resistance of these mutants appears to result from a lower accumulation of DNA-strand breaks after NCS treatment compared to parental cells (unpublished observation). Thiols are essential for activation of the NCS chromophore, and NCS toxicity is greatly reduced in GSH-depleted cells (De Graf et al., 1985). Although the phenotype of BLM-1 and BLM-2 points to a possible alteration in cellular thiol content, the steady-state level of GSH is apparently unaltered in both of these mutants. Whether these mutants are defective in a GSH-de-
pendent enzyme, such as GSH reductase, leading to a failure of the cells to recover from oxidative stress, remains to be determined. In a recent study, 99.9% depletion of GSH left a residual DNA breakage 33% of that in cells with normal GSH levels. Some breaks were protein associated, and it was suggested that residual breaks may have been due to endonucleolytic activity or to topoisomerase II (Kapper et al., 1987). Therefore, other, as yet undefined, factors may be important in producing DNA strand breakage in the presence of NCS. The genetic analyses show that BLM-1, BLM-2 and xrs-6 cells represent different complementation groups. Thus, at least 3 gene products are required for efficient cellular protection against bleomycin-toxicity. Recently, Stamato et al. (1987) have isolated a bank of bleomycin-sensitive hamster mutants which, in certain cases, appear to have a profile of cross-sensitivities unlike those of BLM-1 and -2. These new mutants may, therefore, include more genetic classes. Although the hybrids xrs-6/xrs-6 and BLM2/BLM-2 show approximately diploid mutant levels of hypersensitivity to bleomycin, the pairing BLM-1/BLM-1 is significantly more resistant than BLM-1 cells alone. The reason for this is unknown. The increased cell volume to surfa.ce area ratio of the hybrids could lead to decreased drug uptake, while their longer doubling time could lead to a decreased sensitivity to a cell-cycle specific drug such as bleomycin. However, why this should particularly affect the BLM-1/BLM-1 hybrid is unclear. The mutants that we have isolated show major differences in their cell survival characteristics following treatment with various bleomycin analogues, and dearly possess different genetic alterations. They should prove to be suitable recipients for DNA transfection to isolate the human genes responsible for basal resistance to bleomycin. They may also be useful for a biochemical investigation of the mechanisms of cell killing by bleomycin, and for the screening of new analogues.
Acknowledgements We thank Dr. P. Jeggo for the kind use of the xrs-6 cell line, and the North of England Cancer Research Campaign for financial support.
164
References Bachur, N.R., S.L. Gordon and M.V. Gee (1978) A general mechanism for microsomal activation of quinone anticancer agents to free radicals, Cancer Res., 38, 1745-1750. Bachur, N.R., M.V. Gee and R.D. Friedman (1982) Nuclear catalyzed antibiotic free radical formation, Cancer Res., 42, 1078-1081. Barranco, S.C., W.E. Bolton and J.K. Novak (1980) Time-dependent changes in drug sensitivity expressed by mammalian cells after exposure to trypsin, J. Natl. Cancer Inst., 64, 913-916. Berry, D.E., R.E. Kilkuskie and S.M. Hecht (1985) DNA damage induced by bleomycin in the presence of dibucaine is not predictive of cell growth inhibition, Biochemistry, 24, 3214-3219. Cone, R., S.K. Hasan, J.W. Lown and A.R. Morgan (1976) The mechanism of degradation of DNA by streptonigrin, Can. J. Biochem., 56, 1042-1047. Dasgupta, D., and I.H. Goldberg (1986) Mode of reversible binding of neocarzinostatin chromophore to DNA: base sequence dependency of binding, Nucl. Acids Res., 14, 1089-1105. Davidson, R.L., and P.S. Gerald (1976) Improved techniques for the induction of mammalian cell hybridisation by polyethylene glycol, Somat. Cell Genet., 2, 165-176. DeGraff, W.G., A. Russo and J.B. Mitchell (1985) Glutathione depletion greatly reduces neocarzinostatin cytotoxicity in Chinese hamster V79 cells, J. Biol. Chem., 260, 8312-8315. Giloni, L., M. Takeshita, F. Johnson, C. Iden and A.P. Grollman (1981) Bleomycin-induced strand scission of DNA, J. Biol. Chem., 256, 8608-8615. Ginsburg, E., T.E. Gram and M.A. Trush (1984) A comparison of the pulmonary toxicity and chemotherapeutic activity of bleomycin-BAPP to bleomycin and pepleomycin, Cancer Chemother. Pharmacol., 12, 111-115. Griffith, O.W. (1980) Determination of glutathione and glutathione disulfide using glutathione reductase and 2vinylpyridine, Anal. Biochem., 106, 207-212. Huang, C.-H., C.K. Mirabelli, Y. Jan and S.T. Crooke (1981) Single-strand and double-strand deoxyribonucleic acid breaks produced by several bleomycin analogues, Biochemistry, 20, 233-237. Jeggo, P.A., and L.H. Kemp (1983) X-Ray sensitive mutants of a Chinese hamster ovary cell line: Isolation and cross-sensitivity to other DNA damaging agents, Mutation Res., 112, 313-319. Kappen, L.S., and I.H. Goldberg (1985) Activation of neocarzinostatin chromophore and formation of nascent DNA damage do not require molecular oxygen, Nucl. Acids Res., 13, 1637-1648. Kappen, L.S., T.E. EUenberger and I.H. Goldberg (1987) Mechanism and base specificity of DNA breakage in intact cells by neocarzinostatin, Biochemistry, 26, 384-390. Kilkuskie, R.E., T.L. Macdonald and S.M. Hecht (1984) Bleomycin may be activated for DNA cleavage by NADPH-cytochrome P-450 reductase, Biochemistry, 23, 6165-6171.
Kross, J., W.D. Henner, S.M. Hecht and W.A. Haseltine (1982) Specificity of deoxyribonucleic acid cleavage by bleomycin, phleomycin and tallysomycin, Biochemistry, 21, 4310-4318. Lin, S.Y., and A.P. Grollman (1981) Interactions of a fragment of bleomycin with deoxyribodinucleotides: nuclear magnetic resonance studies, Biochemistry, 20, 7589-7598. Lown, J.W., and S.K. Sim (1977) The mechanism of bleomycin-induced cleavage of DNA, Biochem. Biophys. Res. Commun., 77, 1150-1157. Mizuno, S., and A. Ishida (1982) Potentiation of bleomycin cytotoxicity by membrane-interacting drugs and increased calcium ions, Biochem. Biophys. Res. Commun., 107, 1021-1027. Murray, V., and R.F. Martin (1985) The sequence specificity of bleomycin-induced DNA damage in intact cells, J. Biol. Chem., 260, 10389-10391. Ozawa, S., et al. (1985) Mechanisms affecting pepleomycin sensitivity of Chinese hamster cell lines, J. Antibiotics, 38, 1257-1264. Povirk, L.F. (1979) Catalytic release of DNA bases by oxidation and reduction of an iron-bleomycin complex, Binchemistry, 18, 3989-3995. Povirk, L.F., and I.H. Goldberg (1984) Competition between anaerobic covalent linkage of neocarzinostatin chromophore to deoxyribose in DNA and oxygen-dependent strand breakage and base release, Biochemistry, 23, 6304-6311. Povirk, L.F., M. Hogan and N. Dattagupta (1979) Binding of bleomycin to DNA: intercalation of the bithiazole rings, Biochemistry, 18, 96-101. Pratviel, G., J. Bernadou and B. Meunier (1986) DNA breaks generated by the bleomycin-iron III complex in the presence of KHSO 5, a single oxygen donor, Biochem. Biophys. Res. Commun., 136, 1013-1020. Robson, C.N., and I.D. Hickson (1986) Genetic analysis of mitomycin C-sensitive mutants of a Chinese hamster ovary cell line, Mutation Res., 163, 201-208. Robson, C.N., A.L. Harris and I.D. Hickson (1985) Isolation and characterisation of Chinese hamster ovary cell lines sensitive to mitomycin C and bleomycin, Cancer Res., 45, 5304-5309. Robson, C.N., P.R. Hoban, A.L. Harris and I.D. Hickson (1987) Cross-sensitivity to topoisomerase II inhibitors in cytotoxic drug hypersensitive Chinese hamster ovary cell lines, Cancer Res., 47, 1560-1565. Roy, S.N., and S.B. Horwitz (1984) Characterization of the association of radiolabeled bleomycin A : with HeLa cells, Cancer Res., 44, 1541-1546. Sakai, T.T., J.M. Riordan and J.D. Glickson (1983) Bleomy¢in interactions with DNA, Studies on the role of the C-terminal cationic group of bleomycin A 2 in association with and degradation of DNA, Biochim. Biophys. Acta, 758, 176-180. Sikik, B.I., Z.H. Siddik and T.E. Gram (1980) Relative pulmonary toxicity and antitumor effects of two new bleomycin analogs, pepleomycin and tallysomycin Aa, Cancer treat. Rep., 64, 659-667.
165 Sikik, B.I., M. Rozensweig and S.K. Carter (Eds.) (1985) Bleomycin Chemotherapy, Academic Press, New York. Stamato, T.D., B. Peters, P. Patil, N. Denko, R. Weinstein and A. Giaccia (1987) Isolation and characterisation of bleomycin-sensitiveChinese hamster ovary cell lines, Cancer Res., 47, 1588-1592. Takeshita, M., L.S. Kappen, A.P. Grollman, M. Eisenberg and I.H. Goldberg (1981) Strand seission of deoxyribonucleic acid by neocarzinostatin, auromomycin and bleomycin:
studies on base release and nucleotide sequence specificity, Biochemistry, 20, 7599-7606. Umezawa, H., K. Maeda, T. Takeuchi and Y. Akami (1966) New antibiotics bleomycin A and B, J. Antibiot., Ser. A, 19, 200-209. White, J.R. (1977) Streptonigrin-transition metal complexes: binding to DNA and biological activity, Biochem. Biophys. Res. Commun., 77, 387-391.