Dipyridamole (DPM) enhanced the sensitivity of human ovarian carcinoma 2008 cells to etoposide (VP-16) producing a 5.5-fold reduction in 50% inhibitory concentration at a DPM concentration of 20 microM. This interaction was shown to be truly synergistic by isobologram and median effect analysis. DPM increased the steady-state VP-16 content of 2008 cells; a DPM concentration of 4 microM increased VP-16 content by 2-fold. DPM was 25 times less potent when cells were incubated in human plasma. In tissue culture medium 96% of the DPM was free, whereas in plasma only 15% was non-protein bound. DPM did not displace VP-16 from proteins under either condition. DPM did not increase the initial influx of VP-16 but did inhibit the initial efflux, reducing the efflux rate constant by 27%. DPM had no effect on the later stages of drug efflux, nor did it irreversibly bind VP-16 in the cell. The effect of DPM was evident within 1 min; once removed, the effect disappeared within 2 min. DPM is a potent nucleoside membrane transport inhibitor and can also inhibit cyclic AMP (cAMP) phosphodiesterase in platelets. Nitrobenzylthioinosine, another nucleoside transport inhibitor which competes for binding with DPM, did not enhance sensitivity to VP-16 or increase VP-16 cellular accumulation and did not block the effect of DPM. In 2008 cells, DPM did not increase cAMP; when cAMP was increased by incubation with dibutyryl cyclic 3':5'-AMP, there was no synergy with VP-16. The results indicate that enhanced sensitivity to VP-16 was not due to an effect of DPM on the protein binding of VP-16 or on cellular cAMP and suggest that it is not directly related to inhibition of nucleoside transport. This effect appears to be a newly identified mechanism of action for this agent.