(CANCER RESEARCH 48, 5947-5952, November 1, 1988] Metabolism of SR 4233 by Chinese Hamster Ovary Cells: Basis of Selective Hypoxie Cytotoxicity1 Margaret A. Baker,2 Elaine M. Zeman, V. Kate Hirst, and J. Martin Brown Department of Therapeutic Radiology, Division of Therapeutic Radiology, Stanford University, Stanford, California 94305 ABSTRACT The metabolism of SR 4233 (3-amino-l,2,4-bentotriazine-l,4-dioxide), recently reported as highly toxic to hypoxic cells in vitro, was studied by using suspensions of Chinese hamster ovary cells. The rates of formation of two known reduction products, the 1-oxide and the unoxygenated 3- aminobenzotriazine, were measured in aerobic and hypoxic cell suspen sions for drug treatments producing both hypoxic and aerobic cytotoxic- ity. Formation of the 1-oxide and a small amount of the 3-aminobenzo- triazine occurred preferentially in hypoxic suspensions. These metabo lites were relatively nontoxic to either aerobic or hypoxic cells, implying another mechanism of toxicity. The activation of SR 4233 by single electron transfer, hypothetically forming a toxic drug radical, was ex plored. Aerobic stimulation of oxygen consumption in respiration-inhib ited cells and malondialdehyde release from aerobic cells in the presence of SR 4233 indicated the formation of active oxygen species during drug activation. Increased malondialdehyde release in hypoxic cells and its attenuation by the hydrogen donor, dimethylthiourea, implied the pres ence of an oxidizing radical. Unlike the nitroimidazole, misonidazole, hypoxic metabolism of SR 4233 did not deplete intracellular glutathione or result in increased binding of drug metabolites to cellular macromol- ecules. These results are consistent with macromolecular damage caused by an oxygen sensitive, nonbinding, drug-free radical intermediate with oxidizing properties as the mechanism of selective hypoxic toxicity of SR 4233. INTRODUCTION SR 4233 is the lead compound in a series of benzotriazine- yV-oxides we are evaluating as potential chemotherapeutic and radiosensitizing agents. This class of compounds has not yet been tested for cancer chemotherapeutic use, although some have been used as antimalarials (1, 2). The basis for the pro posed activity of SR 4233 is its high selective toxicity for hypoxic cells (3) and its ability, when combined with the vaso- dilating agent hydra hi/inc. to produce extensive killing of tumor cells in vivo (4). The differential toxicity of SR 4233, 50- to 200-fold more for hypoxic than for aerobic cells, depending on the cell line, is higher than for most other bioreductively acti vated drugs, such as the nitroimidazoles and quinone antitumor antibiotics (5-7). The one bioreductively activated compound for which a differential cytotoxicity similar to that for SR 4233 has been reported (8) is a bifunctional compound, RSU 1069, which has both a nitro group and an a/iridine ring. The selective hypoxic toxicity of these compounds makes them potentially useful as adjuncts to radiation therapy for the treatment of solid tumors with a significant hypoxic fraction, or in combi nation with vasodilating agents which selectively increase tumor hypoxia (4). Reduction products of SR 4233 and the sequence of their formation can be predicted from the chemical reduction steps Received 2/8/88; revised 7/18/88; accepted 7/27/88. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This work was supported by USPHS Grant CA 15201 from the National Cancer Institute, Department of Health and Human Services. 2To whom requests for reprints should be addressed, at Department of Radiation Oncology, Division of Biochemical and Radiation Oncology, Univer sity of Pennsylvania Medical School, Philadelphia, PA 19104. for the parent compound (Fig. 1). First is the loss of the 4- position oxide requiring two electron transfer to form either a hydroxide ion or, depending on the extent of protonation, a water molecule as the leaving group, thus producing the mono- N-oxide, SR 4317. The second site for reduction is the 1- position oxide, also requiring two electrons to produce a hy- droxyl radical or water and the triazine, SR 4330 (Fig. 1). Sequential production of SR 4317 and SR 4330 by radiolytic generation of reducing equivalents in an anaerobic system con taining formate, or in an anaerobic solution of xanthine oxidase plus xanthine has been reported (9). In the presence of air, reduction of SR 4233 was not measurable. The differential toxicity of SR 4233 is greatest for CHO3 cells and for murine tumor cell lines when tested in vitro (3). We therefore used a CHO cell line to examine the differential metabolism of SR 4233 in aerobic and hypoxic cells. We also measured the toxicity of the metabolites themselves to hypoxic and aerobic cells. Possible mechanisms of hypoxic and aerobic toxicity such as oxidative damage, glutathione depletion, and adduci formation were also tested. In addition, the formation of active oxygen species in aerobic cells was examined. MATERIALS AND METHODS Cells and Drugs. SR 4233 (3-amino-l,2,4-benzotriazine-l,4-dioxide) and its reduction products, SR 4317 (3-amino-l,2,4-benzotriazine-l- A'-oxide) and SR 4330 (3-amino-l,2,4-benzotriazine), were synthesized at SRI International, Menlo Park, CA, by Dr. William W. Lee under contract from the National Cancer Institute. Chinese hamster ovary (CHO-To) cells (from Dr. A. M. Rauth, Ontario Cancer Institute, Toronto, Ontario, Canada) were grown in suspension in a-MEM me dium (Grand Island Biologicals Laboratories, Santa Clara, CA) con taining 10% fetal bovine serum (Flow Laboratories, Inc., Inglewood, CA), 126 mg/ml penicillin, and 146 mg/ml streptomycin. MISO and [2-'4C]MISO were obtained from the National Cancer Institute. The [3-14C]SR 4233 was synthesized and generously supplied by Dr. Keith R. Laderoute (Ontario Cancer Institute). Silicon oil (Dow Corning 550 fluid) was purchased from K. R. Anderson Co., Inc., Santa Clara, CA, and mineral oil was purchased from E. R. Squibb and Sons, Princeton, NJ. DMTU and 2-vinylpyridine were purchased from Aldrich Chemical Co., Milwaukee, WI. All other reagents were purchased from Sigma Chemical Co., St. Louis, MO. Benzotriazine Metabolism by CHO Cells. Cells were harvested in exponential phase of growth by centrifugation for 5 min at 300 x g and resuspended at 5 x IO6cells/ml in BBS. BBS contained per liter: 7.53 g NaCl, 2.20 g NaHCO3, 0.40 g KCi, 0.46 g KH2PO4, 0.14 g CaCI2, 0.10 g MgCl2 6H2O, 0.10 g MgSO4-7H2O, 0.9 g glucose (5 mM), and 0.01 g phenol red. Use of BBS decreased the HPLC background caused by medium and serum components and allowed the cell metabolism experiments to be done with the same gas mixtures as were used in the cytotoxicity assays (containing 5% CO2), while maintaining a constant pHof7.4. The cell suspension was added in equal amounts (15 ml) to each of several 25-ml double side-armed flasks (Wheaton, Millville, NJ) and 3The abbreviations used are: CHO, Chinese hamster ovary; MDA, malondi aldehyde; MISO, misonidazole; DMTU, dimethylthiourea; TBARS, thiobarbi- turic acid reactive substances; a-MEM, a-minimal essential medium; BBS, bicar bonate-buffered balanced salts solution; HPLC, high pressure liquid chromatog- raphy; GSH, reduced glutathione; GSSG, oxidized glutathione; TBA, thiobarbituric acid. 5947 on June 6, 2015. © 1988 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from