Summary In this report, we investigate several examples of hypoxia-induced drug resistance and compare them with P-glycoprotein associated multidrug resistance (MDR). EMT6/Ro cells exposed to drugs in air immediately after hypoxic treatment developed resistance to adriamycin, 5-fluorouracil, and actinomycin D. However, these cells did not develop resistance to colchicine, vincristine or cisplatin. When the cells were returned to a normal oxygen environment, they lost resistance. There was no correlation between the content of adriamycin and the development of adriamycin resistance induced by hypoxia. There was no difference between the efflux of adriamycin from aerobic cells and that from hypoxia-treated cells. The mRNA for P-glycoprotein was not detected in the hypoxia-treated cells. These results suggest that hypoxia-induced drug resistance is different from P-glycoprotein associated multidrug resistance.As a tumour grows, heterogeneities of cellular microenvironments occur, such as the development of oxygen gradients in the tumour as a consequence of deficient vascularisation, and cause hypoxic cells that may be resistant to radiotherapy (Sutherland, 1988). Several studies using monolayer cultures (Smith et al., 1980;Teicher et al., 1981Teicher et al., , 1985 and the multicell spheroid system (Sutherland et al., 1979) Conditions of hypoxic exposure Before being gassed with N2, the cells were supplied with 5 ml of fresh complete medium and allowed to equilibrate in a humidified 37°C incubator. Cells undergoing hypoxic stress were isolated in specially designed hypoxic chambers at room temperature (Sutherland et al., 1982). The chambers were repeatedly evacuated and filled every 15 min for 2.25 h with the appropriate gas mixtures certified to contain less than 10 ppm 02. The sealed chambers were then removed to a warm room (37'C) at a time point, t, referred to hereafter as '0' hours of hypoxia.Preparation of drugs and conditions of drug exposure All drugs used were obtained from Sigma Chemical Company. Stock solutions of adriamycin (ADR), vincristine, and actinomycin D (ACTD) were prepared with phosphatebuffered saline (PBS). Solutions of 5-fluorouracil (5-FU), colchicine, and cisplatin were made before each experiment. The solvents used were PBS for ADR, ACTD and cisplatin and distilled water for 5-FU. Absolute ethanol was used as a solvent for colchicine in order to obtain a sufficiently high concentration for the experiments. The final concentration of alcohol in the medium was 1%. As a control for the alcohol solvent, we established that 1% of absolute ethanol in cultures for 2 h had no effect on plating efficiency. Drug treatment was started under aerobic conditions at 37'C in the incubator after culture dishes were removed from the chambers at the zero time. Exposure times for ADR, ACTD and cisplatin were 1 h. A 2 h exposure time was used for 5-FU, colchicine and vincristine to cause significant cell killing. Exposure times were limited to 1-2 h to avoid the effect of release from a ORPs-induced state,...