The objective of this study was to investigate the protective effect of U50,488H, a selective kappa-opioid receptor agonist, in the ischemia/reperfusion (I/R) rat and to delineate the underlying mechanism. Rat heart I/R injury was induced by occluding the left anterior descending coronary artery for 45 min and restoring perfusion for 120 min. U50,488H or vehicle was intravenously injected before ischemia. Electrocardiogram, heart rate (HR), arterial blood pressure (ABP), left ventricular pressure (LVP), systolic function (+dp/dtmax), and diastolic function (-dp/dtmax) were monitored in the course of the experiment. Myocardial infarction size was evaluated. Plasma concentrations of cardiac troponin T (cTnT), creatine kinase (CK), and lactate dehydrogenase (LDH) were measured. Single rat ventricular myocyte was obtained by enzymatic dissociation method. The potassium currents (IK) of isolated ventricular myocytes were recorded with the whole-cell configuration of the patch-clamp technique. Compared with the sham control group, no significant change was found in HR, while ABP, LVP and +/-dp/dtmax were significantly reduced in the I/R group. Administration of U50,488H significantly lowered HR in both control and I/R groups. Compared with the vehicle-treated I/R group, administration of U50,488H had no significant effect on I/R-induced reduction in ABP, LVP, and +/-dp/dtmax. However, this treatment significantly reduced the myocardial infarction size, and markedly decreased the contents of plasma cTnT, CK and LDH. During ischemia and reperfusion, the incidence of ventricular arrhythmia in U50,488H-treated rats was significantly reduced. These effects were independent of the bradycardia induced by U50,488H, as the reducing infarct size and antiarrhythmic effect of U50,488H were still observed in animals in which heart rate was kept constant by electrical pacing. U50,488H and BRL-52537 still produced an antiarrhythmic effect when the rat heart was subjected to a shorter ischemic period of 10 min occlusion of coronary artery, which produced no infarction. IK of the myocytes were inhibited by U50,488H in a dose-dependent manner in normal and hypoxic rat ventricular myocytes. However, the effects of U50,488H on IK did not show any significant difference in normal and hypoxic myocytes. The above-described effects of U50,488H were totally blocked by nor-Binaltorphimine, a selective kappa-opioid receptor antagonist. The results suggest that kappa-opioid agonist U50,488H exerts its direct cardioprotective and antiarrhythmic effects against I/R via kappa-opioid receptor, which participates in the regulation of potassium channels in normal and hypoxic ventricular myocytes.