Endogenous opioids have nitric oxide (NO)-dependent cardiovascular actions. In the light of biological evidence of accumulation of endogenous opioids in cholestasis and also existence of NO-dependent bradycardia in cholestatic subjects, this study was carried out to evaluate the role of endogenous opioids in the generation of bradycardia in a rat model of cholestasis. Male Sprague-Dawley rats were used to induce cholestasis by surgical ligation of the bile duct, with sham-operated animals serving as a control. The animals were divided into six groups which received naltrexone [20 mg/kg/day, subcutaneously (s.c.)], N(G)-L-nitro-arginine methyl ester (L-NAME, 3 mg/kg/day, s.c.), aminoguanidine (200 mg/kg/day, s.c.), L-arginine (200 mg/kg/day, s.c.), naltrexone + L-NAME (20 and 3 mg/kg/day, s.c) or saline. One week after the operation, a lead II electrocardiogram (ECG) was recorded and the spontaneously beating atria of the animals were then isolated and the chronotropic responses to epinephrine evaluated. The plasma L-nitro-tyrosine level and alanine amino transferase and alkaline phosphatase activities were also measured. The heart rate of cholestatic animals was significantly lower than that of control rats in vivo and this bradycardia was corrected with daily adminstration of naltrexone or L-NAME. The basal spontaneous beating rate of atria in cholestatic animals was not significantly different from that of sham-operated animals in vitro. Cholestasis induced a significant decrease in the chronotropic effect of epinephrine. This effect was corrected by daily injection of naltrexone or L-NAME, or concurrent administration of naltrexone + L-NAME, and was not corrected by aminoguanidine. L-arginine had an equivalent effect to L-NAME and increased the chronotropic effect of epinephrine in cholestatic rats but not in control animals. Bile duct ligation increased the plasma activity of liver enzymes as well as the level of L-nitro-tyrosine. L-arginine and naltrexone treatment significantly decreased the elevation of liver enzymes in bile duct-ligated rats. Pretreatment of cholestatic animals with naltrexone or L-NAME decreased the plasma L-nitro-tyrosine level. The results suggest that either prevention of NO overproduction or protection against liver damage is responsible for recovery of bradycardia after naltrexone administration.