1. Whole-cell patch-clamp method was applied to single smooth muscle cells freshly isolated from the rat inferior vena cava. 2. Depolarizing pulses, applied from a holding potential of -90 mV, activated both Na+ and Ca2+ channels. The fast Na+ current was inhibited by nanomolar concentrations of tetrodotoxin (TTX). The slow Ba2+ current (measured in 5 mM Ba2+ solution) was inhibited by Cd2+ and modulated by dihydropyridine derivatives. When the cells were held at a holding potential of -80 mV, racemic Bay K 8644 increased the Ba2+ current (ED50 = 10 nM) while racemic isradipine inhibited the current (IC50 = 21 nM). 3. The voltage-dependency of isradipine blockade was assessed by determining the steady-state availability of the Ca2+ channels. From the shift of the inactivation curve in the presence of isradipine, we calculated a dissociation constant of 1.11 nM for inactivated Ca2+ channels. Scatchard plots of the specific binding of (+)-[3H]-isradipine obtained in intact strips incubated in 5.6 mM or 135 mM K+ solutions confirmed the voltage-dependency of isradipine binding. 4. Specific binding of (+)-[3H]-isradipine was completely displaced by unlabelled (+/-)-isradipine, with an IC50 of 15.1 nM. This value is similar to the IC50 for inhibition of the Ba2+ current (21 nM) in cells maintained at a holding potential of -80 mV. 5. Bay K 8644 had no effects on the Ba2+ current kinetics during a depolarizing test pulse. The steady-state inactivation-activation curves of Ba2+ current were not significantly shifted along the voltage axis.6. The present data suggest the existence of two distinct dihydropyridine binding sites which can be bound preferentially by agonist or antagonist derivatives.