In,. 1. Rodrarion Oncology Biol Phys.. Vol. 8. pp. 693496 Printed m the US A. All rights reserved.

036&3016/82/03069344$03.00/0 Copyright 0 1982 Pergamon Press Ltd.

??Session V-Mechanisms of Cytotoxicity and Chemosensitization

MIS~ONIDAZOLE: INTER-RELATED FACTORS AFFECTING CYTOTOXICITY

CAMERON J. KOCH, PH.D.

Radiobiology, Cross Cancer Institute, I 1560 University Avenue, Edmonton, Alberta, Canada, T6G 122

and

ROBERT L. HOWELL, MS.C.

Immunology and Bacteriology, University of Western Ontario, London, Ontario, Canada

A detailed investigation was undertaken of reported modifiers of the toxicity towards hypoxic cells of misonidazole. The modifiers tested were medium type, cell type, cell density, concentration of misonidazole, addition of serum, addition of sulfhydryl, addition of oxygen and pH. The latter 5 modifiers were found to be most important and were studied in many of the possible combinations. Serum has its greatest protective effect at low concentrations (e.g. 0.5 mM) of misonidaz.ole. In the absence of serum, the (log,,) survival curve for misonidazole toxicity can be described mathematically as a function of time by a shoulder (Do) inversely related to misonidazole concentration, and a slope (l/D,,) related directly to log,, misonidazole concentration. Sulfhydryl’s (cysteamine) protective effect dominates at high concentrations of misonidazole. The protective action of SH can change to potentiative in the absence of serum or at high pH. Addition of oxygen results in overall protection but no relative changes in the effects of the other modifiers. Drugs like ascorbate and disulfide may only potentiate toxicity to the level found in the absence of serum.

Nitroimidazole, Cytotoxicity, Sulfhydryl, Oxygen, Misonidazole.

INTRODUCTION

Since Sutherland’s original demonstration of specific cytotoxicity toward hypoxic cells by metronidazole,8 there has been a dramatic increase in the amount of research in this area. Initial interest centered on the potentiation of the toxicity in view of the possibility that nitroaromatic drugs could be used as chemotherapeutic agents specific for hypoxic cells. However, results from clinical trials suggested that the primary limitation in the clinical use of hypoxic cell radiosensitizers like misonidazole was ner- vous system toxicity.” Thus, much recent work has been done first, to try to understand the biochemical basis of the toxicity of the nitromidazoles and secondly to identify means to reduce this toxicity. Since there is still no appropriate in vitro model for neurotoxicity,’ studies have focused on hypoxic cell cytotoxicity.

To.date, agents like ascorbate3.5 and certain disulf- hide? have been shown to increase cytotoxicity whereas even very low concentrations of oxygen have been shown to decrease cytotoxicity.S.9,‘o Unfortunately, serum, sulf- hydryls and pH changes have been shown both to increase

or decrease cytotoxicity, depending presumably on the specific techniques or cell lines used. One is always skeptical of such inverse results, but in our own laborato- ry, over the course of two years, we noted a continued decline in the ability of a single serum batch to protect against the cytotoxic effects of 5.0 mM misonidazole.

These results prompted a detailed study of the relation- ships between factors known to modify nitroaromatic cytotoxicity. The factors initially considered were medium type, cell density, cell type, concentration of misonidazole, presence or absence of serum, presence or absence of sulfhydryls, oxygen and other drugs and pH. In preliminary experiments we found little variation in cytotoxicity with various medium or cell types, and with cell density so long as the cell density was low enough to prevent nutrient depletion and pH change’ during the experiment (i.e. for up to 20 hours, less than 250 K cells/ml). The present paper details our findings on the effects of misonidazole concentration (5.0 vs 0.5 mM), serum (f 13%), sulfhydryl (+2.5 mM cysteamine), oxy- gen (+2000 ppm partial pressure) and pH (6.5 - 7.9).

Reprint requests to: C.J Koch, Ph.D. Acknowledgements-Work supported by the National Cancer Institute of Canada and Ithe Alberta Heritage Trust Fund- Applied Cancer Research. The authors wish to thank Gail Page

for typing the manuscript. Dr. C.E. Smithen kindly provided the misonidazole.

Accepted for publication 5 November I98 I.

693

694 Radiation Oncology 0 Biology 0 Physics March-April 1982, Volume 8, No. 3 and No. 4

METHODS AND MATERIALS

The apparatus and methods used for these experiments have been detailed in other reports.4.5 Briefly, glass petri dishes (50 mm) were inoculated with approximately 250K V79 Chinese hamster fibroblasts, and incubated over night at 37“C in a humidified atmosphere of 95% air, 5% CO,. Three mls of drug containing medium (Eagle’s basal medium with 20 mM Hepes (pH 7.3 except as noted)) were added to the plates after aspiration of the original medium and the plates were sealed in aluminum chambers. The oxygen concentration in the gas phase of the chambers was changed to the desired level by means of a series of gas changes accomplished with a manifold and vacuum pump.4,5 After the medium in the dishes had equilibrated with the gas phase (-3 hours) the chambers were placed in a water bath at 37°C. At appropriate intervals, the chambers were removed from the water bath and opened. Cells were detached from the plates by 0.05% trypsin after aspirating the drug containing medium and viability was assessed by the fraction of cells forming colonies after 1 week incubation in fresh medium at 37OC.

lowing relationship: the time of the ‘shoulder’ region (analogous to Do for radiation survival curves) varies inversely as the concentration, whereas the final slope (analogous to Do) varies approximately as the log,,, of the concentration. This relationship fails in the presence of serum as the concentration decreases, depending on the serum batch.

The effect of sulfhydryl (2.5 mM Cysteamine (MEA)) varies substantially depending upon whether serum is present (Fig. 2). In the absence of serum cysteamine potentiates the toxicity of 5 mM misonidazole, whereas in the presence of serum some protection is seen.

Similar effects of sulfhydryl are seen at higher oxygen concentrations (Fig. 3). When the oxygen content of the gas phase of the chambers is 2000 ppm, substantial protection of the overall toxicity of 5 mM misonidazole is seen.5,9,‘0 Although serum alone has essentially no effect on toxicity, its presence or absence again determines whether MEA decreases or increases toxicity respec- tively. Note that at 2000 ppm there is no toxicity of the hypoxia alone on control cells whereas substantial killing of control cells occurs under conditions of extreme hypoxia (Fig. 1) for the lengthy incubation times used.

RESULTS AND DISCUSSION

Although serum did not prove to be a consistent protector against the toxicity of 5 mM misonidazole’,’ it afforded almost complete protection against 0.5 mM misonidazole (Fig. 1). In the absence of serum, the concentration dependence of cytotoxicity obeys the fol-

The effect of pH is complex in the sense that the simultaneous addition of sulfhydryl and/or serum changes the overall response markedly (Fig. 4). First, as the pH shifts away from the optimum value of 7.3, the effect of serum is less pronounced. At low pH, sulfhydryl protects against the cytotoxicity whereas at high pH sulfhydryl potentiates considerably the cytotoxic effects of misonidazole.

0 4 8 12 16

TIME AT 37” C (Hours)

Fig. I. Effect of serum and misonidazole concentration. Sur- vival of cells (V79 Chinese hamster fibroblasts) after exposure to extreme hypoxia (< 10 ppm 0,) in the presence or absence of misonidazole (MISO) and/or serum (SER). Standard errors of geometric mean survival after several experiments are shown when greater than size of point. Number of points and experi- ments listed for each curve. 0 = No MIS0 (Control) + SER, (76 pts/ 12 expts.); A = 0.5 mM MIS0 + SER, (15 pts/4 expts.); A = 0.5 mM MIS0 - SER. (39 pts/4 expts.); W = 5.0 mM MIS0 + SER, (61 pts/l expts.); 0 = 5.0 mM MIS0 - SER, (42 pts/7 expts.).

Our results confirm the interesting effect on cytotoxic-

0 2 4 6

TIME AT 37” C (Hours)

Fig. 2. Effect of sulfhydryl. Survival of cells after exposure to extreme hypoxia in the presence of misonidazole with and without cysteamine (MEA) or serum (SER). W = 5 mM MIS0 + SER (63/8); 0 = 5 mM MIS0 - SER (38/5); A = 5 mM MIS0 + SER + 2.5 mM MEA (21/3); A = 5 mM MIS0 - SER + 2.5 mM MEA (10/2). Results with 0.5 mM MIS0 showed essentially no effect of cysteamine and were not statisti- cally different from the 0.5 mM MIS0 curves in Figure 1.

MIS cytotoxicity 0 C. J. KOCH 695

TIME Al- 37 O C (Hours) TIME AT 37” C (Hours) Fig. 3. Effect of low partial pressure of oxygen. When the atmosphere inside the chamber contains 2000 ppm 0, (esti- mated oxygen partial pressure ~600 ppm at cell monolayer) there is overall protection against cytotoxicity but the relative effects of serum and cysteamine are the same as for extreme hypoxia. No toxicity is seen for controls at this oxygen tension. 0 0 = No MIS0 (controls) + SER + MEA (l8/3); ??= 5 mM MIS0 + SER(l7/2); 0 = 5mM MIS0 - SER(31/4);A’- 5 mM MIS0 + SER + 2.5 mM MEA (19/2); A = 5 mM MIS0 - SER + 2.5 mM MEA (25/3). Results with 0.5 mM MIS0 show similar protection by oxygen and reduced effect of MEA.

ity of pH and SH first described by Palcic et ~1.~ In synergestic effects of other drugs are greatly diminished, addition these experiments show that effects on the and in our experiments with several serum batches, we cytotoxicity of other drugs can be modified substantially found practically no toxicity at all. Thus even under by as yet unidentified components of serum. In general we conditions of extreme hypoxia it is going to be difficult to have found that if the serum batch used causes substantial model toxic effects of radiosensitizers at physiological protection against cytotoxicity, then SH will lead to concentrations. In order to obtain inter-lab consistency, it further protection whereas agents like ascorbate and is suggested that experiments be done both in the presence disulfhides will enhance the cytotoxicity but only back to or absence of serum, and with more precise measurement the level seen in the absence of serum. In contrast, if the and control of oxygen levels and pH. In recent experi- serum batch has no effect or even potentiates the toxici- ments we have found that high pH potentiates the toxicity ty,’ ascotbate and disulfhides have little or no effect and of disulfhides so a possible explanation for the results of

sulfhydryl may potentiate the cytotoxicity. Most impor- Figure 4 is that at high pH in the presence of an oxidizing

tant, as the misonidazole concentration decreases, and the agent (MISO), sulfhydryl is more rapidly converted to oxygen concentration increases (both changes are in disulfhide, which is responsible for the additional toxici-

keeping with what might be expected in vim) most of the ty.

I2 I I I I I I 4 0 2 4 6

Fig. 4. Effect of pH. The only major effect of pH is a dramatic increase of toxicity at high pH in the presence of cysteamine. All curves 5 mM MISO, extreme hypoxia. Points were done in duplicate but represent only I experiment. More recent experi- ments under slightly different conditions confirm the overall results however. + + = pH 7.9 + SER respectively; ??0 = pH 7.9 -t 2.5 mM MEA + SER respectively; ??0 = pH 6.5 + SER respectively; A A = pH 6.5 + 2.5 mM MEA 2 SER respec- tively. The pH dependence of MIS0 toxicity has not yet been tested at low MIS0 concentration or 2000 ppm 02.

REFERENCES Clarke, C., Dawson, K.B., Sheldon, P.W., Chaplin, D.J., Stratford, I.J., Adam:<, G.E.: Quantitative cytochemical method for assessing the neurotoxicity of misonidazole. In Radiation Sensitizers, L.W. Brady (Ed.). New York, Mas- son. 1980, pp. 245-249. Howell, R.L., Koch, C.J.: Modification by sulfhydtyls disulfides and ascorbate of radiosensitizing and toxic prop- erties of misonidazole with hypoxic cells (Abs. B-l-6). 6th Int. Congress Radiat. Res. Tokyo Japan, 1980, pp. I I I. Josephy, P.D., Palcic, B., Skarsgard, L.D.: Ascorbate enhanced cytotoxicity of misonidazole. Nature 271: 370- 371, 1978.

5.

6.

and radiation-sensitizing agents: II Radiosensitivity of hypoxic or aerobic Chinese hamster fibroblasts in the presence of cysteamine and misonidazole: Implications for the ‘Oxygen Effect’. Radiat. Res. 87: 265-283, I98 1. Koch, C.J., Howell, R.L., Biaglow, J.E.: Ascorbate anion potentiates cytotoxicity of nitro-aromatic compounds under hypoxic and anoxic conditions. Br. J. Cancer 39: 321-329, 1979. Palcic, B., Skov, K.A., Skarsgard, L.D.: Effect of reducing agents on misonidazole cytotoxicity. In Radiation Sensitiz- ers, L.W. Brady (Ed.). New York, Masson. 1980, pp. 438-440.

Koch, C.J., Howell, R.L.: Combined radiation-protective 7. Stratford, I.J., Gray, P.: Some factors affecting the specific

696 Radiation Oncology 0 Biology ??Physics

toxicity of misonidazole towards hypoxic mammalian ceils. Br. J. Cancer 37 (Suppl II): 129-l 32, 1978.

8. Sutherland, R.M.: Selective chemotherapy of non-cycling cells in an in vifro tumor model. Cancer Res. 34: 3501- 3504, 1974.

9. Taylor, Y.C., Rauth, A.M.: Sulphydryls, ascorbate and oxygen as modifiers of the toxicity and metabolism of misonidazole in vifro. Br. J. Cancer 41: 8922900, 1980.

IO. Taylor, Y.C., Rauth, A.M.: Oxygen tension, cellular respi-

March-April 1982, Volume 8, No. 3 and NO. 4

ration and redox state as variables influencing the cytotox- icity of the radiosensitizer misonidazole. Radiat. Res. (In press) I98 1.

I I. Urtasun, R.C., Chapman, J.D., Feldstein, M.L., Band, R.P., Rabin, H.E., Wilson, A.E., Marynowski, B., Starrev- eld, A., Shnitka, T.: Peripheral neuropathy related to misonidazole. Incidence and Pathology. Br. J. Cancer 37 (Suppl III): 271-275, 1978.