We examined P-glycoprotein-mediated verapamil transport, using two drug-sensitive and multi-drug resistant cell-line couples, i.e. A2780, 2780AD and SW-1573, SW-1573/1R500. The interaction of 3H-labeled verapamil with cells was measured using a flow-through system. The verapamil-containing medium was pumped over the cells and monitored on-line for radioactivity. In the P-glycoprotein-expressing cells, verapamil accumulation was increased by vinblastine and some known multidrug resistant (MDR) modifiers. Subsequent removal of these modifiers caused release of verapamil into the medium against a verapamil concentration gradient. In this manner, we obtained evidence that verapamil is actively transported by the MDR-related P-glycoprotein. Using the flow-through system, we also exposed the cells to flowing culture medium containing daunorubicin, and measured the inhibition of daunorubicin efflux by verapamil. We found that, although the active efflux of daunorubicin was maximally blocked by verapamil short-term, longer-term active efflux of daunorubicin resumed. At a daunorubicin concentration in the flowing medium of 5 microM, increasing the verapamil concentration resulted in the same short-term effects, but in a significantly longer period of a maximal inhibition of daunorubicin efflux from the cells. At a daunorubicin concentration of 20 microM, increasing the verapamil concentration affected neither the short-term nor the long-term effects. These and other observations are in agreement with a model in which daunorubicin and verapamil are non-competing substrates for P-glycoprotein. In conclusion, we obtained evidence that verapamil is actively transported by the MDR-related P-glycoprotein and that verapamil and daunorubicin are non-competing substrates for P-glycoprotein. Consequently, the effectiveness of verapamil as an MDR antagonist may be compromised because it is extruded by P-glycoprotein.