Oxygen dependence of binding of misonidazole to rodent and human tumors in vitro.

Misonidazole, a 2-nitroimidazole, has been shown to form metabolically induced adducts to cellular molecules at a very high rate in the absence of oxygen, and this rate decreases substantially as the oxygen concentration increases. Thus, it has considerable potential as a marker for hypoxic, radiation resistant cells in tumors. The dependence of the rate of adduct formation (binding) on oxygen concentration was studied for EMT6/Ed, Walker 256, and Dunning R3327-AT rodent tumors and for two human colon carcinomas, a human melanoma, and a human breast carcinoma. Fragments of these tumors were incubated with [14C]- or [3H]misonidazole in vitro at several oxygen concentrations and the quantity of misonidazole bound was determined from autoradiographs as a function of distance from the surfaces of the fragments. The Km of binding inhibition (oxygen concentration for half-maximal binding) for the tumors varied by a factor of 10. The range was centered on the range of values reported for the Km of cellular radioresistance (oxygen concentration for half-maximal radioresistance). The patterns of binding at depth within the tumor fragments indicated that gradients of cellular waste products and nutrients other than oxygen had minimal effects on binding. All tumors were capable of metabolizing oxygen to levels sufficiently low to yield the maximal binding rate, but the distance of penetration of oxygen varied, indicating a range of at least 4 in rates of oxygen consumption. The ratio of misonidazole bound by stromal tissue versus tumor cells ranged from 0.9 for a colon tumor to 0.3 for a breast tumor. These properties of misonidazole binding indicate that it should be a good marker for radiobiological hypoxia in tumors, providing adequate controls can be performed.

Misonidazole retention by normal tissues: a distinction between label on the ring and side chain.

The retention of misonidazole by mouse normal tissues and EMT6/Ed tumors was studied 24 hr after injection of ring labelled (2-14C) and side chain-labelled (3H) drug. Slightly more 3H than 14C was retained by tumors, but this was considered to be within experimental error of no difference. The ring label (14C) activity retained by tumors was 5-12 times greater than that retained by heart, kidney, brain, muscle and spleen. However, in the same animals, the side chain label (3H) was retained to an appreciably greater extent by the normal tissues, so that the ratio of activities retained in tumors and normal tissues was 3 to 4. This difference in discrimination between tumors and normal tissues implies that gamma-emitting or NMR-active analogues of misonidazole will detect hypoxia in tumors in situ more efficiently, if the active isotope is situated on the ring. The data also indicate that fragmentation products of the metabolism of misonidazole, which contain the side chain and exclude the 2-carbon, are responsible for 50-70% of the misonidazole retained by normal tissues when a side chain-label is used.

The labelling of EMT-6 tumours in BALB/C mice with 14C-misonidazole.

14C-misonidazole was injected IP into BALB/C mice bearing EMT-6 tumours. Tumour and normal tissue levels of radioactivity were determined by liquid scintillation procedures for times up to 72 h after drug administration. At times longer than 8 h, levels of 14C in tumour were 2-6 times higher than 14C levels measured in other normal tissues, with the exception of liver. The levels of 14C from misonidazole retained in liver tissue were comparable to those retained in the tumours. The clearance of 14C-misonidazole from the tumours and all normal tissues could be characterised by two distinct phases: a rapid phase (0.7 h half-life) followed by a much slower phase (approximately 50 h half-life). The distribution of retained 14C from labelled misonidazole within tumour-bearing mice was also measured by whole animal autoradiographic techniques. This procedure confirmed the biodistribution of 14C-misonidazole determined by liquid scintillation procedures. To enhance binding to hypoxic cells in vivo, 14C-misonidazole was administered in multiple doses over a three-hour period, simulating a prolonged exposure to the drug. The 14C from labelled misonidazole retained in tumour tissue was 4-15 times greater than that retained in normal tissues, liver tissue being again an exception. Myocardial ischaemia was induced by isoproterenol treatment and a two-fold increase of 14C from labelled misonidazole was found in the heart tissue of mice treated by the drug compared to controls. The increased binding of 14C-misonidazole to tumour cells after prolonged exposure to the drug and the increased binding of 14C-misonidazole to ischaemic heart tissue suggests that hypoxia is a factor in sensitiser adduct formation in vivo.