- Email: [email protected]
Assessment of micronucleus induction in SCCVII cells treated with bioreductive agents, WIN 59075 (SR 4233) and mitomycin C, under aerobic and hypoxic conditions
Genetic Toxicology
Mutation Research 342 (1995) 171-177
ELSEVIER
Assessment of micronucleus induction in SCCVII cells treated with bioreductive agents, WIN 59075 (SR 4233) and mitomycin C, under aerobic and hypoxic conditions Toru Shibata a,* Yuta Shibamoto a,h Keisuke Sasai a, Natsuo Oya Takehisa Takagi a, Rumi Murata a, Mitsuyuki Abe a 9
9
a 9
a Department of Radiology, Faculty of Medicine, Kyoto University, Kyoto606-Ol, Japan b Department of Oncology, Chest Disease Research Institute, Kyoto University, Kyoto606-01, Japan
Received 8 August 1994; revised 11 November 1994; accepted 5 December 1994
Abstract WIN 59075 (SR4233, tirapazamine) is a promising bioreductive antitumor agent preferentially more toxic to hypoxic cells and presently undergoing phase I clinical trials. In this investigation, we have examined the applicability of the cytokinesis-block micronucleus assay to assess the effects of bioreductive agents. SCCVII tumor cells were treated with WIN 59075 or mitomycin C at various concentrations under aerobic and hypoxic conditions. Significant induction of micronuclei in binucleate cells was demonstrated in a dose-dependent fashion and it appeared to be strongly correlated with the loss of clonogenicity in the colony assay. Both agents showed selectively higher toxicity to hypoxic cells than to aerobic cells and the ratios of the concentrations required to obtain the equivalent effects under aerobic and hypoxic conditions could be also estimated by this method as follows: the hypoxic toxicity ratios were 120-130 for WIN 59075 and 3.0-3.3 for mitomycin C. For several favorable characteristics, the cytokinesis-block micronucleus assay can provide an alternative, rapid, and reproducible means for evaluation of antitumor activities from chromosomal breakage caused by the bioreductive agents. Keywords: Cytokinesis-block micronucleus assay; WIN 59075; SR 4233; Tirapazamine; Hypoxia; Bioreductive drugs
I. Introduction The existence of hypoxic cells in solid tumors has been postulated to be one of the major causes of treatment failure by radiotherapy (Moulder and Rockwell, 1987). To overcome this problem,
* Corresponding author. Tel.: (81)75-751-3419; Fax: (81)75771-9749.
various therapeutic strategies (Hall and RoizinTowle, 1975; Brown and Yu, 1984; Rockwell et al., 1991) have been proposed and examined, including hypoxic cell radiosensitizers. Despite great expectations, advance in patient cure has not yet been achieved in the clinical trials (Dische, 1991). Further investigations are developing to search for potent agents with lower toxicity and higher sensitizing effects (Hill et al., 1986; Sasai et al., 1991; Shibata et al., 1994).
0165-1218/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0165-1218(94)00090-5
172
72 Shibata et al. /Mutation Research 342 (1995) 171-177
Recent studies have suggested that tumor hypoxia could be exploited by the use of the bioreductive agents which have the feature of preferentially killing the hypoxic cells (Kennedy et al., 1979). Brown and colleagues reported that WIN 59075 (SR 4233) showed highly selective toxicity toward hypoxic cells both in the in vitro and in vivo murine tumor system (Zeman et al., 1986; Brown and Lemmon, 1990; Brown, 1993). The hypoxic cytotoxicity of WIN 59075 is thought to result from DNA double strand breakage by oxidizing radical anions produced through bioreductive metabolism under hypoxic conditions (Baker et al., 1988). Recent studies using electron spin resonance have unequivocally identified the formation of a free radical under hypoxic conditions in vitro (Zeman et al., 1988). WIN 59075 is currently being tested in phase I clinical trials (Doherty et al., 1994) and novel analogues of WIN 59075 and the bioreductive compounds of other classes are being synthesized and tested. This is becoming one of the major topics in radiation oncology and cancer therapy (Adams and Stratford, 1994). The micronucleus (MN) assay has been widely used for the evaluation of mutagenicity of environmental agents in the field of mutation research (Heddle et al., 1983). Recently, the cytokinesis-block MN assay (Fenech and Morley, 1985) using cytochalasin-B (Cyt-B) has been applied as a rapid and alternative method to evaluate the cytotoxicity by various therapeutic radiation and chemicals (Wakata and Sasaki, 1987). This method has the advantage of the feasibility to distinguish cells after one mitosis from nondividing cells. We previously described the potential usefulness of this method for the predictive assay of tumor radiation sensitivity (Shibamoto et al., 1991; Shibamoto et al., 1994). For the evaluation of the efficacy of hypoxic cell radiosensitizers at low radiation doses, we showed that the cytokinesis-block MN assay was very useful (Shibata et al., 1994) and comparable data were obtained with the chromosomal aberration assay performed simultaneously with the same specimens (Oya et al., 1993). From these previous studies, this assay appears to demonstrate the extent of chromosomal breakage caused by various therapeutics.
However, no studies on the efficiency of WIN 59075 by the MN assay have been reported. If the MN assay provides reliable and reproducible results with WIN 59075 and bioreductive anticancer agents, it may have some promising applications not only to the evaluation of newly synthesized agents but also to the drug sensitivity prediction of human cancers, because this assay is already used in the clinic (Shibamoto et al., 1994). In this study, we examined the differential induction of micronuclei in SCCVII cells with WIN 59075 and mitomycin C under aerobic and hypoxic conditions. We also evaluated the doseeffect relationship in the MN assay in comparison with that in the colony formation assay and discussed the usefulness of the cytokinesis-block MN assay for the assessment of the bioreductive agents.
2. Materials and methods 2.1. Compounds and cell cultures
WIN59075 (SR 4233, tirapazamine), 3-amino1,2,4-benzotriazine-l,4 dioxide (CAS. 27314-97-2), was kindly provided by Sterling Winthrop Pharmaceuticals Research Division (Collegeville, PA, U.S.A.). Mitomycin C (MMC) was supplied by Kyowa Hakko Co. Ltd (Tokyo, Japan). Eagle's minimal essential medium supplemented with 12.5% fetal bovine serum was used throughout the present study. Both drugs were dissolved into saline immediately before the experiments and were diluted to appropriate concentrations with the medium. SCCVII tumor cells were prepared by trypsinization from exponentially growing cultures immediately before use (Shibata et al., 1994). 2.2. Drug exposure under hypoxic and aerobic conditions
According to the published methods (Shibamoto et al., 1992; Shibata et al., 1994), the cell suspensions (5 x 10 6 cells/ml) containing various concentrations of drugs in the plugged glass flasks were flushed with a gas mixture of
T. Shibata et al. /Mutation Research 342 (1995) 171-177
95% N 2 / 5 % C O 2 for 20 min at 4°C on iced water. After achievement of hypoxic conditions, the flasks were put into a water bath at 37°C to allow the cells to bioreductively metabolize the drugs. After 60 min, hypoxic treatment was stopped and the cells were reaerated. The drugs were removed by centrifugation and replaced by fresh medium. Then the cells were counted and diluted appropriately for both MN assay and colony assay. For treatment under aerobic conditions, the cell suspensions were kept in the incubator at 37°C and 5% CO 2.
2.3. The cytokinesis-block MN assay and the colony formation assay Experimental procedures were similar to those described previously (Shibata et al., 1994). The cells were plated into the dishes and Cyt-B was added at the concentration of 1.5 /xg/ml, which is the optimal concentration to inhibit cytokinesis but not karyokinesis (Shibamoto et al., 1991). After an appropriate incubation period, the cells were fixed with ethanol/acetic acid and stained with propidium iodide to score micronuclei in binucleated cells (BNC) under a fluorescence microscope. The number of micronuclei in BNC was determined from the observation of at least 400 BNC according to the published criteria (Shibamoto et al., 1992). The cell survival was determined by the standard colony formation method after 10 days of culture.
3. Results
3.1. Optimalization of the incubation time after the addition of Cyt-B The frequency of BNC changes with the incubation time. The BNC that have experienced the first nuclear divisions after treatment should be scored in the cytokinesis-block method. As a preliminary step, to determine the optimal incubation period, SCCVII cells were treated with WIN 59075 for 60 min and the number of main nuclei ( N ) in each cell was assessed after various durations (12, 24, 36, 48, and 60 h). The frequencies of
WIN 59075 (hypoxic) 0 . 0 0 5 m M / I hr
173
WIN 59075 laerobic) ImM/lhr
Control (aerobic)
100
~ 75 ~ 5o ~
25
0 12 24 36 48
12 24 36 48
12 24 36 48
Incubation time (hr)
Fig. 1. Changes of the frequency of BNC as a function of incubation time. The frequencies of the cells with I N (O), 2 N (e), and > 2 N (zx) of main nuclei are presented. The highest yields of B N C were obtained at 36 h in this experimental procedure.
1N, 2N, and > 2 N cells changed along the durations as shown in Fig. 1. Both for the cells treated with 5 /zM of WIN 59075 in hypoxia and for those with 1 mM in air, the highest yield of BNC was achieved at 36 h. As the drug concentrations increased, the frequency of BNC decreased and the peak time seemed to be delayed, but the optimal period was 36 h among the evaluable dose range by this method. Similar results were obtained with MMC treatment (data not shown).
3.2. M N induction and cytotoxicity by WIN 59075 and MMC The number of micronuclei in BNC was scored after treatment with various drug concentrations. The frequency of micronucleated BNC per total BNC was determined as %MNBNC. The baseline value of % M N B N C was 10.8 + 2.5% (mean + s.d.) in hypoxia and 9.5 + 3.8% in air without drugs, there being no significant difference. Under hypoxic conditions with WIN 59075, the induction of micronuclei was observed above the concentrations of 0.5-1 ~M. The % M N B N C increased steeply in a dose-dependent manner. On the other hand, no apparent induction of micronuclei was detectable below 100 /zM in air and great differences were notable in the concentrations necessary to obtain the equivalent effects in the two conditions. The differential effects were also observed in the colony assay. Dose-el-
174
T. Shibata et al. / Mutation Research 342 (1995) 171-177 o 0
I00
8O 7(1
I0 1
6O u Z
50-
~ 4o g
3o20100
,
J/
, , .• ,
.......
103
L
........
102
W I N
5 9 0 7 5
,
i o -I
........
i
i00
(mM)
102
10 3
10 4 ,
........
i
......
,
104 10 ~ MMC (mM)
....... I 0 -2
Fig. 2. %MNBNC as a function of the concentration of WIN59075 or MMC under aerobic ( o ) and hypoxic (o) conditions in the cytokinesis-block MN assay. Each dose-effect curve could be determined by the linear regression analysis according to the least-squares method, which indicated that %MNBNC increased exponentially for the logarithmic scale of the concentrations.
fect curves for M N induction and survival with M M C illustrated in Figs. 2 and 3 also exhibited a pattern similar to that with W I N 59075. To compare the efficiency of M N induction in hypoxia to that in air, we have determined the concentrations required to reach % M N B N C to 50% under aerobic or hypoxic conditions and the ratio of the concentrations (aerobic/hypoxic) was defined as the hypoxic toxicity ratio. The concentration required to give a surviving fraction of 0.1 was also calculated from each dose-survival curve, and then the hypoxic toxicity ratio was determined in a similar way. As shown in Table 1, the MN assay was confirmed to give ratios similar to those in the colony assay. There may also be another index for estimation, because the frequency of
lO 2
10 3
....... L 1o i
WIN 59075
........
i
100
,
10 4
10 3
10 2
MMC (mM)
(mM)
Fig. 3. Surviving fractions as a function of the concentration of WIN59075 or MMC under aerobic ( o ) and hypoxic (e) conditions in the colony assay. Dose-survival curves were fitted by the least-squares regression analysis.
BNC with multiple numbers of micronuclei increased corresponding to the increment in the concentrations of WIN 59075 under aerobic and hypoxic conditions as shown in Fig. 4. In Fig. 5, we plotted the present data based on the index MN frequency, which indicates the average number of micronuclei per single BNC. We have also obtained the ratio of the concentrations required to give a 0.5 MN frequency (Table 1). Although it is not yet clear which is the better index of chromosomal damage, the hypoxic toxicity ratios obtained from these two indices were similar.
4. Discussion Herein, we found that treatment by WIN 59075 and M M C under hypoxic and aerobic conditions could induce micronuclei in SCCVII cells in a
Table 1 The drug concentrations required to obtain the equivalent effects on each endpoint under aerobic and hypoxic conditions
MMC Colony assay MN assay WIN59075 Colony assay MN assay
Endpoints
Hypoxic a (/z M)
surviving fraction of 0.1 50% M N B N C 0.5 MN frequency
0.95 0.45 0.25
surviving fraction of 0.1 50% M N B N C 0.5 M N frequency
9.0 6.6 3.3
Aerobic a (/z M) 3.0 1.5 0.75 1250 790 430
Hypoxic toxicity ratio b (aerobic/hypoxic) 3.2 3.3 3.0 140 120 130
a These values were estimated from the dose-effect curves computed by the least-squares regression analysis. b The hypoxic toxicity ratio was obtained from the direct comparison of the differentials in the concentrations (aerobic/hypoxic).
T. Shibata et al. / Mutation Research 342 (1995) 171-177
control
WIN 5 9 0 7 5 / h y p o x i c 0.001mM 0.005mM
control
175
WIN 5 9 0 7 5 / a e r o b i c 0.5mM 1raM
100 80-
i
40200-
012345
012345
5
012345
012345
012345
012345
Micronuclei p e r single b i n u c l e a t e cell Fig. 4. Frequencies of BNC with multiple micronuclei at various concentrations of WIN59075 u n d e r aerobic and hypoxic conditions.
dose-dependent fashion. The MN assay was performed at slightly lower drug doses than the colony assay because it was not possible to obtain a sufficient number of BNC for determination of micronuclei at higher drug doses. We generally agree with other authors (Bonatti et al., 1983) that the MN assay should be performed at doses allowing the cell survival of more than 10% of the cells, and this seems to be a limitation of this assay. To clarify the quantitative relationship between MN induction and lethality, we calculated the increase in % M N B N C corresponding to D o value (the dose required to reduce surviving fraction to 37%) which means the average dose of 1 lethal event per cell. As shown in Table 2, the increase in of % M N B N C per lethal event for
each drug treatment in hypoxia and air ranged from 34 to 41%. These findings indicated that the induction of micronuclei by WIN 59075 and MMC in hypoxia and air was nearly be the same magnitude for lethal events. A report on cisplatin in CHO cells in the conventional MN assay (Bonatti et al., 1983) showed that MN represented only 10-20% of chromosomal damage related to lethal hits. This quantitative difference from our findings suggests that the mechanisms of cytotoxicity and MN induction are qualitatively different among chemotherapeutic agents even if the main cause of the cytotoxicity may be chromosomal damage. From the previous paper on radiosensitivity (Shibamoto et al., 1991), we estimated that the increase in % M N B N C per lethal event was about 36% in SCCVII cells in the cytokinesis-
Table 2 Comparison of the M N induction in the MN assay with loss of clonogenicity in the colony assay
W I N 59075 hypoxic aerobic MMC hypoxic aerobic
Do a (/zM)
%MNBNC/~M (%//xM)
b
% M N B N C / I e t h a l event (%)
4.9 + 0.6 5.2 _+ 1.0 x 102
7.5 + 0.8 6.4 _+ 1.0 x 10 2
37 + 9 34 + 11
3.3 + 0.5 x 10 -1 1.3 +_ 0.1
1.1 + 0.2 x 102 3.3 + 0.4 x 10
38 + 11 41 +_ 9
a Do value, the average dose per lethal hit, was determined for the reciprocal of the slope of the survival curve (37% dose slope). b Change of % M N B N C corresponding to increment of the concentration by 1 /xM.
foo
176
T. Shibata et al. / Mutation Research 342 (1995) 171-177
hypoxic
~
0.S
0
J 10 3
102
lO-t
WIN 5 9 0 7 5 (raM)
100
10 4 10 3 MMC (mM)
10-2
Fig. 5. M N frequency as a function of the concentration of WIN59075 or M M C u n d e r aerobic (O) and hypoxic (o) conditions in the cytokinesis-block MN assay. Curve fitting was done in a m a n n e r similar to that in Fig. 2.
block method. Bakker et al. (1993) also reported the increase of micronucleated cells per lethal event to be about 30 _+ 6% in rat rhabdomyosarcoma and ureteral cancer cell lines using the conventional MN assay. These previous findings with radiation were comparable to those in the present study. Previous mechanistic studies (Zeman et al., 1986; Baker et al., 1988) implicates that the bioreductive activation of WIN 59075 under hypoxic conditions causes the formation of toxic drug radicals by single electron transfer and that DNA double strand breaks are the most likely lethal lesions as in the case of radiation. Zeman and Brown (1991) also suggested that WIN 59075 behaves in a radiomimetic manner in most respects. Our present findings may be consistent with the hypothesis. In this study, the baseline values of %MNBNC in our SCCVII murine tumor cells (10.8% in hypoxia and 9.5% in air) were slightly higher than those of other cell lines (Shibamoto et al., 1991), but are almost compatible with the values obtained in our previous studies using SCCVII cells (Oya et al., 1993; Shibata et al., 1994). It has not been determined whether the spontaneous micronuclei in SCCVII cells derive from chromosome loss or chromosome breakage, but an analysis with Chinese hamster cells by the cytokinesisblock method (Wakata and Sasaki, 1987) indicated that the major source of baseline frequency of micronuclei are not the spontaneously occurring chromosomal structural aberrations but rather mitotic dysfunction. In a study using kine-
tochore-staining method (Weissenborn and Streffer, 1991), 40-52.7% of micronuclei in untreated human melanoma ceils were found to contain complete chromosomes as indicated by the kinetochore-positive staining but the proportion of micronuclei derived from whole chromosome decreased with increasing radiation dose. Also in human rectal carcinoma cells with high frequency of spontaneously formed micronuclei, at least 3050% of these micronuclei contained complete chromosomes. Further investigations are needed to clarify whether these findings can be applied to our SCCVII ceils. Recently, the MTT assay which is a colorimetric assay for cell growth has been introduced to evaluate the bioreductive drugs (Stratford and Stephens, 1989). The M T T assay has been reported to be a simple, rapid procedure and to give values of differential toxicity similar to those obtained by the clonogenic assay. While the MTT assay measures cellular metabolic activity, the MN assay detects more specific events based on DNA damage. Studies using human tumor cells in primary culture also seem to be possible (Shibamoto et al., 1994). Further investigation using this assay would lead to better understanding of cellular events caused by the bioreductive agents.
Acknowledgements This work was supported in part by the Grantin-Aid for Cancer Research (04151010, 05151011) from the Japanese Ministry of Education, Science, and Culture.
References Adams, G.E. and I.J. Stratford (1994) Bioreductive drugs for cancer therapy: the search for tumor specificity, Int. J. Radiat. Oncol. Biol. Phys., 29, 231-238. Baker, M.A., E.M. Zeman, V.K. Hirst and J.M. Brown (1988) Metabolism of SR 4233 by Chinese hamster ovary cells: the basis of selective hypoxic cytotoxicity, Cancer Res., 48, 5947-5952. Bakker, P.J.M., L.J. Tukker, J. Stap, C.H.N. V e e n h o f and J.A. Aten (1993) Micronuclei expression in tumours as a test for radiation sensitivity, Radiother. Oncol., 26, 69-72.
T. Shibata et al. / Mutation Research 342 (1995) 171-177
Bonatti, S., P.H.M. Lohman and F. Berends (1983) Induction of micronuclei in Chinese-hamster ovary cells treated with Pt co-ordination compounds, Mutation Res., 116, 149-154. Brown, J.M. and N.Y. Yu (1984) Radiosensitization of hypoxic cells in vivo by SR 2508 at low radiation doses: A preliminary report, Int. J. Radiat. Oncol. Biol. Phys., 10, 1207-1212. Brown, J.M. and M.J. Lemmon (1990) Potentiation by the hypoxic cytotoxin SR 4233 of cell killing produced by fractionated irradiation of mouse tumors, Cancer Res., 50, 7745-7749. Brown, J.M. (1993) SR 4233 (tirapazamine): a new anticancer drug exploiting hypoxia in solid tumours, Br. J. Cancer, 67, 1163-1170. Dische, S. (1991) A review of hypoxic cell radiosensitization, Int. J. Radiat. Oncol. Biol. Phys., 20, 147-152. Doherty, N., S.L. Hancock, S. Kaye, C.N. Coleman, L. Shulman, C. Marquez, C. Mariscal, R. Rampling, S. Senan and R.V. Roemeling (1994) Muscle cramping in Phase I clinical trials of Tirapazamine (SR 4233) with and without radiation, Int. J. Radiat. Oncol. Biol. Phys., 29, 379-382. Fenech, M. and A.A. Morley (1985) Measurement of micronuclei in lymphocytes, Mutation Res., 147, 29-36. Hall, E.J. and L. Roizin-Towle (1975) Hypoxic sensitizer: radiobiological studies at the cellular level, Radiology, 117, 453-457. Heddle, J.A., M. Hite, B. Kirkhart, K. Mavournin, J.T. MacGregor, G.W. Newell and M.F. Salamone (1983) The induction of micronuclei as a measure of genotoxicity, A report of the U.S. Environmental Protection Agency, Genetox Program, Mutation Res., 123, 61-118. Hill, R.P., S. Gulyas and G.F. Whitmore (1986) Studies of the in vivo and in vitro cytotoxicity of the drug RSU-1069, Br. J. Cancer, 53, 743-751. Kennedy, K.A., B.A. Teicher, S. Rockwell and A.C. Sartorelli (1979) The hypoxic tumor cell: a target for selective cancer chemotherapy, Biochem. Pharmacol., 29, 1-8. Moulder, J.E. and S. Rockwell (1987) Tumor hypoxia: its impact on cancer therapy, Cancer Metas. Rev., 5, 313-341. Oya, N., Y. Shibamoto, K. Sasai, T. Sugiyama and M. Abe (1993) In vivo radiosensitization efficacy of KU-2285 and etanidazole at clinically relevant low radiation doses, Int. J. Radiat. Oncol. Biol. Phys., 27, 1113-1119. Rockwell, S., M. Kelley, C.G. Irvin, C.S. Hughes, E. Porter, H. Yabuki and J. Fischer (1991) Modulation of tumor oxygenation and radiosensitivity by a perfluorooctylbromide emulsion, Radiother. Oncol., 22, 92-98.
177
Sasai, K., S. Nishimoto, K. Shimokawa, Y. Hisanaga, Y. Kitakabu, Y. Shibamoto, L. Zhou, J. Wang, M. Takahashi, T. Kagiya and M. Abe (1991) A fluorinated 2-nitroimidazole, KU-2285, as a new hypoxic cell radiosensitizer, Int. J. Radiat. Oncol. Biol. Phys., 20, 1249-1254. Shibamoto, Y., T. Shibata, S. Miyatake, Y. Oda, T. Manabe, G. Ohshio, K. Yagi, C. Streffer, M. Takahashi and M. Abe (1994) Assessment of the proliferative activity and radiosensitivity of human turnouts using the cytokinesis-block micronucleus assay, Br. J. Cancer, 70, 67-71. Shibamoto, Y., C. Streffer, C. Fuhrmann and V. Budach (1991) Tumor radiosensitivity prediction by the cytokinesis-block micronucleus assay, Radiation Res., 128, 293-300. Shibamoto, Y., C. Streffer, K. Sasai, N. Oya and M. Abe (1992) Radiosensitization efficacy of KU-2285, RP-170 and etanidazole at low radiation doses: assessment by in vitro cytokinesis-block micronucleus assay, Int. J. Radiat. Biol., 61, 473-478. Shibata, T., Y. Shibamoto, N. Oya, K. Sasai, R. Murata, T. Ishigaki and M. Abe (1994) Comparison of radiosensitizing effect of KU-2285 and SR-2508 at low drug concentrations and doses, Int. J. Radiat. Oncol. Biol. Phys., 29, 587-590. Stratford, I.J. and M.A. Stephens (1989) The differential hypoxic cytotoxicity of bioreductive agents determined in vitro by the MTT assay, Int. J. Radiat. Oncol. Biol. Phys., 16, 973-976. Wakata, A. and M.S. Sasaki (1987) Measurement of micronuclei by cytokinesis-block method in cultured Chinese hamster cells: comparison with types and rates of chromosome aberrations, Mutation Res., 190, 51-57. Weissenborn, U. and C. Streffer (1991) Micronuclei with kinetochores in human melanoma cells and rectal carcinomas, Int. J. Radiat. Biol., 59, 373-383. Zeman, E.M., J.M. Brown, M.J. Lemmon, V.K. Hirst and W.W. Lee (1986) SR-4233: a new bioreductive agent with high selective toxicity for hypoxic mammalian cells, Int. J. Radiat. Oncol. Biol. Phys., 12, 1239-1242. Zeman, E.M., V.K. Hirst, M.J. Lemmon and J.M. Brown (1988) Enhancement of radiation-induced tumor cell killing by the hypoxic cell toxin SR 4233, Radiother. Oncol., 12, 209-218. Zeman, E.M. and J.M. Brown (1991) Aerobic radiosensitization by SR4233 in rodent and human cells: mechanistic and therapeutic implications, Int. J. Radiat. Biol., 59, 117-131.
Mutation Research 342 (1995) 171-177
ELSEVIER
Assessment of micronucleus induction in SCCVII cells treated with bioreductive agents, WIN 59075 (SR 4233) and mitomycin C, under aerobic and hypoxic conditions Toru Shibata a,* Yuta Shibamoto a,h Keisuke Sasai a, Natsuo Oya Takehisa Takagi a, Rumi Murata a, Mitsuyuki Abe a 9
9
a 9
a Department of Radiology, Faculty of Medicine, Kyoto University, Kyoto606-Ol, Japan b Department of Oncology, Chest Disease Research Institute, Kyoto University, Kyoto606-01, Japan
Received 8 August 1994; revised 11 November 1994; accepted 5 December 1994
Abstract WIN 59075 (SR4233, tirapazamine) is a promising bioreductive antitumor agent preferentially more toxic to hypoxic cells and presently undergoing phase I clinical trials. In this investigation, we have examined the applicability of the cytokinesis-block micronucleus assay to assess the effects of bioreductive agents. SCCVII tumor cells were treated with WIN 59075 or mitomycin C at various concentrations under aerobic and hypoxic conditions. Significant induction of micronuclei in binucleate cells was demonstrated in a dose-dependent fashion and it appeared to be strongly correlated with the loss of clonogenicity in the colony assay. Both agents showed selectively higher toxicity to hypoxic cells than to aerobic cells and the ratios of the concentrations required to obtain the equivalent effects under aerobic and hypoxic conditions could be also estimated by this method as follows: the hypoxic toxicity ratios were 120-130 for WIN 59075 and 3.0-3.3 for mitomycin C. For several favorable characteristics, the cytokinesis-block micronucleus assay can provide an alternative, rapid, and reproducible means for evaluation of antitumor activities from chromosomal breakage caused by the bioreductive agents. Keywords: Cytokinesis-block micronucleus assay; WIN 59075; SR 4233; Tirapazamine; Hypoxia; Bioreductive drugs
I. Introduction The existence of hypoxic cells in solid tumors has been postulated to be one of the major causes of treatment failure by radiotherapy (Moulder and Rockwell, 1987). To overcome this problem,
* Corresponding author. Tel.: (81)75-751-3419; Fax: (81)75771-9749.
various therapeutic strategies (Hall and RoizinTowle, 1975; Brown and Yu, 1984; Rockwell et al., 1991) have been proposed and examined, including hypoxic cell radiosensitizers. Despite great expectations, advance in patient cure has not yet been achieved in the clinical trials (Dische, 1991). Further investigations are developing to search for potent agents with lower toxicity and higher sensitizing effects (Hill et al., 1986; Sasai et al., 1991; Shibata et al., 1994).
0165-1218/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0165-1218(94)00090-5
172
72 Shibata et al. /Mutation Research 342 (1995) 171-177
Recent studies have suggested that tumor hypoxia could be exploited by the use of the bioreductive agents which have the feature of preferentially killing the hypoxic cells (Kennedy et al., 1979). Brown and colleagues reported that WIN 59075 (SR 4233) showed highly selective toxicity toward hypoxic cells both in the in vitro and in vivo murine tumor system (Zeman et al., 1986; Brown and Lemmon, 1990; Brown, 1993). The hypoxic cytotoxicity of WIN 59075 is thought to result from DNA double strand breakage by oxidizing radical anions produced through bioreductive metabolism under hypoxic conditions (Baker et al., 1988). Recent studies using electron spin resonance have unequivocally identified the formation of a free radical under hypoxic conditions in vitro (Zeman et al., 1988). WIN 59075 is currently being tested in phase I clinical trials (Doherty et al., 1994) and novel analogues of WIN 59075 and the bioreductive compounds of other classes are being synthesized and tested. This is becoming one of the major topics in radiation oncology and cancer therapy (Adams and Stratford, 1994). The micronucleus (MN) assay has been widely used for the evaluation of mutagenicity of environmental agents in the field of mutation research (Heddle et al., 1983). Recently, the cytokinesis-block MN assay (Fenech and Morley, 1985) using cytochalasin-B (Cyt-B) has been applied as a rapid and alternative method to evaluate the cytotoxicity by various therapeutic radiation and chemicals (Wakata and Sasaki, 1987). This method has the advantage of the feasibility to distinguish cells after one mitosis from nondividing cells. We previously described the potential usefulness of this method for the predictive assay of tumor radiation sensitivity (Shibamoto et al., 1991; Shibamoto et al., 1994). For the evaluation of the efficacy of hypoxic cell radiosensitizers at low radiation doses, we showed that the cytokinesis-block MN assay was very useful (Shibata et al., 1994) and comparable data were obtained with the chromosomal aberration assay performed simultaneously with the same specimens (Oya et al., 1993). From these previous studies, this assay appears to demonstrate the extent of chromosomal breakage caused by various therapeutics.
However, no studies on the efficiency of WIN 59075 by the MN assay have been reported. If the MN assay provides reliable and reproducible results with WIN 59075 and bioreductive anticancer agents, it may have some promising applications not only to the evaluation of newly synthesized agents but also to the drug sensitivity prediction of human cancers, because this assay is already used in the clinic (Shibamoto et al., 1994). In this study, we examined the differential induction of micronuclei in SCCVII cells with WIN 59075 and mitomycin C under aerobic and hypoxic conditions. We also evaluated the doseeffect relationship in the MN assay in comparison with that in the colony formation assay and discussed the usefulness of the cytokinesis-block MN assay for the assessment of the bioreductive agents.
2. Materials and methods 2.1. Compounds and cell cultures
WIN59075 (SR 4233, tirapazamine), 3-amino1,2,4-benzotriazine-l,4 dioxide (CAS. 27314-97-2), was kindly provided by Sterling Winthrop Pharmaceuticals Research Division (Collegeville, PA, U.S.A.). Mitomycin C (MMC) was supplied by Kyowa Hakko Co. Ltd (Tokyo, Japan). Eagle's minimal essential medium supplemented with 12.5% fetal bovine serum was used throughout the present study. Both drugs were dissolved into saline immediately before the experiments and were diluted to appropriate concentrations with the medium. SCCVII tumor cells were prepared by trypsinization from exponentially growing cultures immediately before use (Shibata et al., 1994). 2.2. Drug exposure under hypoxic and aerobic conditions
According to the published methods (Shibamoto et al., 1992; Shibata et al., 1994), the cell suspensions (5 x 10 6 cells/ml) containing various concentrations of drugs in the plugged glass flasks were flushed with a gas mixture of
T. Shibata et al. /Mutation Research 342 (1995) 171-177
95% N 2 / 5 % C O 2 for 20 min at 4°C on iced water. After achievement of hypoxic conditions, the flasks were put into a water bath at 37°C to allow the cells to bioreductively metabolize the drugs. After 60 min, hypoxic treatment was stopped and the cells were reaerated. The drugs were removed by centrifugation and replaced by fresh medium. Then the cells were counted and diluted appropriately for both MN assay and colony assay. For treatment under aerobic conditions, the cell suspensions were kept in the incubator at 37°C and 5% CO 2.
2.3. The cytokinesis-block MN assay and the colony formation assay Experimental procedures were similar to those described previously (Shibata et al., 1994). The cells were plated into the dishes and Cyt-B was added at the concentration of 1.5 /xg/ml, which is the optimal concentration to inhibit cytokinesis but not karyokinesis (Shibamoto et al., 1991). After an appropriate incubation period, the cells were fixed with ethanol/acetic acid and stained with propidium iodide to score micronuclei in binucleated cells (BNC) under a fluorescence microscope. The number of micronuclei in BNC was determined from the observation of at least 400 BNC according to the published criteria (Shibamoto et al., 1992). The cell survival was determined by the standard colony formation method after 10 days of culture.
3. Results
3.1. Optimalization of the incubation time after the addition of Cyt-B The frequency of BNC changes with the incubation time. The BNC that have experienced the first nuclear divisions after treatment should be scored in the cytokinesis-block method. As a preliminary step, to determine the optimal incubation period, SCCVII cells were treated with WIN 59075 for 60 min and the number of main nuclei ( N ) in each cell was assessed after various durations (12, 24, 36, 48, and 60 h). The frequencies of
WIN 59075 (hypoxic) 0 . 0 0 5 m M / I hr
173
WIN 59075 laerobic) ImM/lhr
Control (aerobic)
100
~ 75 ~ 5o ~
25
0 12 24 36 48
12 24 36 48
12 24 36 48
Incubation time (hr)
Fig. 1. Changes of the frequency of BNC as a function of incubation time. The frequencies of the cells with I N (O), 2 N (e), and > 2 N (zx) of main nuclei are presented. The highest yields of B N C were obtained at 36 h in this experimental procedure.
1N, 2N, and > 2 N cells changed along the durations as shown in Fig. 1. Both for the cells treated with 5 /zM of WIN 59075 in hypoxia and for those with 1 mM in air, the highest yield of BNC was achieved at 36 h. As the drug concentrations increased, the frequency of BNC decreased and the peak time seemed to be delayed, but the optimal period was 36 h among the evaluable dose range by this method. Similar results were obtained with MMC treatment (data not shown).
3.2. M N induction and cytotoxicity by WIN 59075 and MMC The number of micronuclei in BNC was scored after treatment with various drug concentrations. The frequency of micronucleated BNC per total BNC was determined as %MNBNC. The baseline value of % M N B N C was 10.8 + 2.5% (mean + s.d.) in hypoxia and 9.5 + 3.8% in air without drugs, there being no significant difference. Under hypoxic conditions with WIN 59075, the induction of micronuclei was observed above the concentrations of 0.5-1 ~M. The % M N B N C increased steeply in a dose-dependent manner. On the other hand, no apparent induction of micronuclei was detectable below 100 /zM in air and great differences were notable in the concentrations necessary to obtain the equivalent effects in the two conditions. The differential effects were also observed in the colony assay. Dose-el-
174
T. Shibata et al. / Mutation Research 342 (1995) 171-177 o 0
I00
8O 7(1
I0 1
6O u Z
50-
~ 4o g
3o20100
,
J/
, , .• ,
.......
103
L
........
102
W I N
5 9 0 7 5
,
i o -I
........
i
i00
(mM)
102
10 3
10 4 ,
........
i
......
,
104 10 ~ MMC (mM)
....... I 0 -2
Fig. 2. %MNBNC as a function of the concentration of WIN59075 or MMC under aerobic ( o ) and hypoxic (o) conditions in the cytokinesis-block MN assay. Each dose-effect curve could be determined by the linear regression analysis according to the least-squares method, which indicated that %MNBNC increased exponentially for the logarithmic scale of the concentrations.
fect curves for M N induction and survival with M M C illustrated in Figs. 2 and 3 also exhibited a pattern similar to that with W I N 59075. To compare the efficiency of M N induction in hypoxia to that in air, we have determined the concentrations required to reach % M N B N C to 50% under aerobic or hypoxic conditions and the ratio of the concentrations (aerobic/hypoxic) was defined as the hypoxic toxicity ratio. The concentration required to give a surviving fraction of 0.1 was also calculated from each dose-survival curve, and then the hypoxic toxicity ratio was determined in a similar way. As shown in Table 1, the MN assay was confirmed to give ratios similar to those in the colony assay. There may also be another index for estimation, because the frequency of
lO 2
10 3
....... L 1o i
WIN 59075
........
i
100
,
10 4
10 3
10 2
MMC (mM)
(mM)
Fig. 3. Surviving fractions as a function of the concentration of WIN59075 or MMC under aerobic ( o ) and hypoxic (e) conditions in the colony assay. Dose-survival curves were fitted by the least-squares regression analysis.
BNC with multiple numbers of micronuclei increased corresponding to the increment in the concentrations of WIN 59075 under aerobic and hypoxic conditions as shown in Fig. 4. In Fig. 5, we plotted the present data based on the index MN frequency, which indicates the average number of micronuclei per single BNC. We have also obtained the ratio of the concentrations required to give a 0.5 MN frequency (Table 1). Although it is not yet clear which is the better index of chromosomal damage, the hypoxic toxicity ratios obtained from these two indices were similar.
4. Discussion Herein, we found that treatment by WIN 59075 and M M C under hypoxic and aerobic conditions could induce micronuclei in SCCVII cells in a
Table 1 The drug concentrations required to obtain the equivalent effects on each endpoint under aerobic and hypoxic conditions
MMC Colony assay MN assay WIN59075 Colony assay MN assay
Endpoints
Hypoxic a (/z M)
surviving fraction of 0.1 50% M N B N C 0.5 MN frequency
0.95 0.45 0.25
surviving fraction of 0.1 50% M N B N C 0.5 M N frequency
9.0 6.6 3.3
Aerobic a (/z M) 3.0 1.5 0.75 1250 790 430
Hypoxic toxicity ratio b (aerobic/hypoxic) 3.2 3.3 3.0 140 120 130
a These values were estimated from the dose-effect curves computed by the least-squares regression analysis. b The hypoxic toxicity ratio was obtained from the direct comparison of the differentials in the concentrations (aerobic/hypoxic).
T. Shibata et al. / Mutation Research 342 (1995) 171-177
control
WIN 5 9 0 7 5 / h y p o x i c 0.001mM 0.005mM
control
175
WIN 5 9 0 7 5 / a e r o b i c 0.5mM 1raM
100 80-
i
40200-
012345
012345
5
012345
012345
012345
012345
Micronuclei p e r single b i n u c l e a t e cell Fig. 4. Frequencies of BNC with multiple micronuclei at various concentrations of WIN59075 u n d e r aerobic and hypoxic conditions.
dose-dependent fashion. The MN assay was performed at slightly lower drug doses than the colony assay because it was not possible to obtain a sufficient number of BNC for determination of micronuclei at higher drug doses. We generally agree with other authors (Bonatti et al., 1983) that the MN assay should be performed at doses allowing the cell survival of more than 10% of the cells, and this seems to be a limitation of this assay. To clarify the quantitative relationship between MN induction and lethality, we calculated the increase in % M N B N C corresponding to D o value (the dose required to reduce surviving fraction to 37%) which means the average dose of 1 lethal event per cell. As shown in Table 2, the increase in of % M N B N C per lethal event for
each drug treatment in hypoxia and air ranged from 34 to 41%. These findings indicated that the induction of micronuclei by WIN 59075 and MMC in hypoxia and air was nearly be the same magnitude for lethal events. A report on cisplatin in CHO cells in the conventional MN assay (Bonatti et al., 1983) showed that MN represented only 10-20% of chromosomal damage related to lethal hits. This quantitative difference from our findings suggests that the mechanisms of cytotoxicity and MN induction are qualitatively different among chemotherapeutic agents even if the main cause of the cytotoxicity may be chromosomal damage. From the previous paper on radiosensitivity (Shibamoto et al., 1991), we estimated that the increase in % M N B N C per lethal event was about 36% in SCCVII cells in the cytokinesis-
Table 2 Comparison of the M N induction in the MN assay with loss of clonogenicity in the colony assay
W I N 59075 hypoxic aerobic MMC hypoxic aerobic
Do a (/zM)
%MNBNC/~M (%//xM)
b
% M N B N C / I e t h a l event (%)
4.9 + 0.6 5.2 _+ 1.0 x 102
7.5 + 0.8 6.4 _+ 1.0 x 10 2
37 + 9 34 + 11
3.3 + 0.5 x 10 -1 1.3 +_ 0.1
1.1 + 0.2 x 102 3.3 + 0.4 x 10
38 + 11 41 +_ 9
a Do value, the average dose per lethal hit, was determined for the reciprocal of the slope of the survival curve (37% dose slope). b Change of % M N B N C corresponding to increment of the concentration by 1 /xM.
foo
176
T. Shibata et al. / Mutation Research 342 (1995) 171-177
hypoxic
~
0.S
0
J 10 3
102
lO-t
WIN 5 9 0 7 5 (raM)
100
10 4 10 3 MMC (mM)
10-2
Fig. 5. M N frequency as a function of the concentration of WIN59075 or M M C u n d e r aerobic (O) and hypoxic (o) conditions in the cytokinesis-block MN assay. Curve fitting was done in a m a n n e r similar to that in Fig. 2.
block method. Bakker et al. (1993) also reported the increase of micronucleated cells per lethal event to be about 30 _+ 6% in rat rhabdomyosarcoma and ureteral cancer cell lines using the conventional MN assay. These previous findings with radiation were comparable to those in the present study. Previous mechanistic studies (Zeman et al., 1986; Baker et al., 1988) implicates that the bioreductive activation of WIN 59075 under hypoxic conditions causes the formation of toxic drug radicals by single electron transfer and that DNA double strand breaks are the most likely lethal lesions as in the case of radiation. Zeman and Brown (1991) also suggested that WIN 59075 behaves in a radiomimetic manner in most respects. Our present findings may be consistent with the hypothesis. In this study, the baseline values of %MNBNC in our SCCVII murine tumor cells (10.8% in hypoxia and 9.5% in air) were slightly higher than those of other cell lines (Shibamoto et al., 1991), but are almost compatible with the values obtained in our previous studies using SCCVII cells (Oya et al., 1993; Shibata et al., 1994). It has not been determined whether the spontaneous micronuclei in SCCVII cells derive from chromosome loss or chromosome breakage, but an analysis with Chinese hamster cells by the cytokinesisblock method (Wakata and Sasaki, 1987) indicated that the major source of baseline frequency of micronuclei are not the spontaneously occurring chromosomal structural aberrations but rather mitotic dysfunction. In a study using kine-
tochore-staining method (Weissenborn and Streffer, 1991), 40-52.7% of micronuclei in untreated human melanoma ceils were found to contain complete chromosomes as indicated by the kinetochore-positive staining but the proportion of micronuclei derived from whole chromosome decreased with increasing radiation dose. Also in human rectal carcinoma cells with high frequency of spontaneously formed micronuclei, at least 3050% of these micronuclei contained complete chromosomes. Further investigations are needed to clarify whether these findings can be applied to our SCCVII ceils. Recently, the MTT assay which is a colorimetric assay for cell growth has been introduced to evaluate the bioreductive drugs (Stratford and Stephens, 1989). The M T T assay has been reported to be a simple, rapid procedure and to give values of differential toxicity similar to those obtained by the clonogenic assay. While the MTT assay measures cellular metabolic activity, the MN assay detects more specific events based on DNA damage. Studies using human tumor cells in primary culture also seem to be possible (Shibamoto et al., 1994). Further investigation using this assay would lead to better understanding of cellular events caused by the bioreductive agents.
Acknowledgements This work was supported in part by the Grantin-Aid for Cancer Research (04151010, 05151011) from the Japanese Ministry of Education, Science, and Culture.
References Adams, G.E. and I.J. Stratford (1994) Bioreductive drugs for cancer therapy: the search for tumor specificity, Int. J. Radiat. Oncol. Biol. Phys., 29, 231-238. Baker, M.A., E.M. Zeman, V.K. Hirst and J.M. Brown (1988) Metabolism of SR 4233 by Chinese hamster ovary cells: the basis of selective hypoxic cytotoxicity, Cancer Res., 48, 5947-5952. Bakker, P.J.M., L.J. Tukker, J. Stap, C.H.N. V e e n h o f and J.A. Aten (1993) Micronuclei expression in tumours as a test for radiation sensitivity, Radiother. Oncol., 26, 69-72.
T. Shibata et al. / Mutation Research 342 (1995) 171-177
Bonatti, S., P.H.M. Lohman and F. Berends (1983) Induction of micronuclei in Chinese-hamster ovary cells treated with Pt co-ordination compounds, Mutation Res., 116, 149-154. Brown, J.M. and N.Y. Yu (1984) Radiosensitization of hypoxic cells in vivo by SR 2508 at low radiation doses: A preliminary report, Int. J. Radiat. Oncol. Biol. Phys., 10, 1207-1212. Brown, J.M. and M.J. Lemmon (1990) Potentiation by the hypoxic cytotoxin SR 4233 of cell killing produced by fractionated irradiation of mouse tumors, Cancer Res., 50, 7745-7749. Brown, J.M. (1993) SR 4233 (tirapazamine): a new anticancer drug exploiting hypoxia in solid tumours, Br. J. Cancer, 67, 1163-1170. Dische, S. (1991) A review of hypoxic cell radiosensitization, Int. J. Radiat. Oncol. Biol. Phys., 20, 147-152. Doherty, N., S.L. Hancock, S. Kaye, C.N. Coleman, L. Shulman, C. Marquez, C. Mariscal, R. Rampling, S. Senan and R.V. Roemeling (1994) Muscle cramping in Phase I clinical trials of Tirapazamine (SR 4233) with and without radiation, Int. J. Radiat. Oncol. Biol. Phys., 29, 379-382. Fenech, M. and A.A. Morley (1985) Measurement of micronuclei in lymphocytes, Mutation Res., 147, 29-36. Hall, E.J. and L. Roizin-Towle (1975) Hypoxic sensitizer: radiobiological studies at the cellular level, Radiology, 117, 453-457. Heddle, J.A., M. Hite, B. Kirkhart, K. Mavournin, J.T. MacGregor, G.W. Newell and M.F. Salamone (1983) The induction of micronuclei as a measure of genotoxicity, A report of the U.S. Environmental Protection Agency, Genetox Program, Mutation Res., 123, 61-118. Hill, R.P., S. Gulyas and G.F. Whitmore (1986) Studies of the in vivo and in vitro cytotoxicity of the drug RSU-1069, Br. J. Cancer, 53, 743-751. Kennedy, K.A., B.A. Teicher, S. Rockwell and A.C. Sartorelli (1979) The hypoxic tumor cell: a target for selective cancer chemotherapy, Biochem. Pharmacol., 29, 1-8. Moulder, J.E. and S. Rockwell (1987) Tumor hypoxia: its impact on cancer therapy, Cancer Metas. Rev., 5, 313-341. Oya, N., Y. Shibamoto, K. Sasai, T. Sugiyama and M. Abe (1993) In vivo radiosensitization efficacy of KU-2285 and etanidazole at clinically relevant low radiation doses, Int. J. Radiat. Oncol. Biol. Phys., 27, 1113-1119. Rockwell, S., M. Kelley, C.G. Irvin, C.S. Hughes, E. Porter, H. Yabuki and J. Fischer (1991) Modulation of tumor oxygenation and radiosensitivity by a perfluorooctylbromide emulsion, Radiother. Oncol., 22, 92-98.
177
Sasai, K., S. Nishimoto, K. Shimokawa, Y. Hisanaga, Y. Kitakabu, Y. Shibamoto, L. Zhou, J. Wang, M. Takahashi, T. Kagiya and M. Abe (1991) A fluorinated 2-nitroimidazole, KU-2285, as a new hypoxic cell radiosensitizer, Int. J. Radiat. Oncol. Biol. Phys., 20, 1249-1254. Shibamoto, Y., T. Shibata, S. Miyatake, Y. Oda, T. Manabe, G. Ohshio, K. Yagi, C. Streffer, M. Takahashi and M. Abe (1994) Assessment of the proliferative activity and radiosensitivity of human turnouts using the cytokinesis-block micronucleus assay, Br. J. Cancer, 70, 67-71. Shibamoto, Y., C. Streffer, C. Fuhrmann and V. Budach (1991) Tumor radiosensitivity prediction by the cytokinesis-block micronucleus assay, Radiation Res., 128, 293-300. Shibamoto, Y., C. Streffer, K. Sasai, N. Oya and M. Abe (1992) Radiosensitization efficacy of KU-2285, RP-170 and etanidazole at low radiation doses: assessment by in vitro cytokinesis-block micronucleus assay, Int. J. Radiat. Biol., 61, 473-478. Shibata, T., Y. Shibamoto, N. Oya, K. Sasai, R. Murata, T. Ishigaki and M. Abe (1994) Comparison of radiosensitizing effect of KU-2285 and SR-2508 at low drug concentrations and doses, Int. J. Radiat. Oncol. Biol. Phys., 29, 587-590. Stratford, I.J. and M.A. Stephens (1989) The differential hypoxic cytotoxicity of bioreductive agents determined in vitro by the MTT assay, Int. J. Radiat. Oncol. Biol. Phys., 16, 973-976. Wakata, A. and M.S. Sasaki (1987) Measurement of micronuclei by cytokinesis-block method in cultured Chinese hamster cells: comparison with types and rates of chromosome aberrations, Mutation Res., 190, 51-57. Weissenborn, U. and C. Streffer (1991) Micronuclei with kinetochores in human melanoma cells and rectal carcinomas, Int. J. Radiat. Biol., 59, 373-383. Zeman, E.M., J.M. Brown, M.J. Lemmon, V.K. Hirst and W.W. Lee (1986) SR-4233: a new bioreductive agent with high selective toxicity for hypoxic mammalian cells, Int. J. Radiat. Oncol. Biol. Phys., 12, 1239-1242. Zeman, E.M., V.K. Hirst, M.J. Lemmon and J.M. Brown (1988) Enhancement of radiation-induced tumor cell killing by the hypoxic cell toxin SR 4233, Radiother. Oncol., 12, 209-218. Zeman, E.M. and J.M. Brown (1991) Aerobic radiosensitization by SR4233 in rodent and human cells: mechanistic and therapeutic implications, Int. J. Radiat. Biol., 59, 117-131.