Although the formation of oxygen-derived free radicals (or reactive oxygen species; ROS) and the release of endogenous opioid peptides (EOP) have been independently reported to be the major arrhythmogenic factors in ischemic hearts, possible relations between these two factors have seldom been investigated. Thus, we studied whether the ROS and EOP were related in the progression of ischemia-induced arrhythmias. Isolated rat hearts perfused in the Langendorff mode were treated with dynorphin A1-13 (kappa EOP receptor agonist), and/or allopurinol (xanthine oxidase inhibitor), before the onset of ischemia induced by ligating the left coronary arteries. Ischemic period lasted for 30 min, during which cardiac rhythms were recorded. At the end of ischemia, hearts were analyzed for the glutathione and ascorbate levels. Allopurinol (100 nmoles/heart) was effective in reducing the severity of arrhythmia (arrhythmia score: Mean +/- SEM 3.00 +/- 0.80 for allopurinol, 5.75 +/- 0.41 for placebo, p < 0.01), while dynorphin (10 micrograms/heart) potentiated the arrhythmia (6.71 +/- 0.52, p < 0.05 vs. placebo). Coadministration of allopurinol and dynorphin was capable of reducing arrhythmia (5.57 +/- 0.65) when compared with the administration of dynorphin alone (6.71 +/- 0.52, p < 0.05). Tissue oxidative stress was evaluated by the concentrations of glutathione (GSH) and ascorbate. Allopurinol did not significantly elevate tissue GSH concentrations (1.46 +/- 0.05 mumoles/g wet wt) in ischemic hearts, while dynorphin alone significantly decreased the GSH concentrations (0.96 +/- 0.08, p < 0.05) when compared with the placebo (1.32 +/- 0.03). The dynorphin-induced GSH decrease cannot be reversed by coadministration with allopurinol (0.90 +/- 0.104). Allopurinol significantly elevated tissue ascorbate levels (0.16 +/- 0.01) when compared with placebo (0.10 +/- 0.01, p < 0.05). Interestingly, dynorphin alone also elevated the tissue ascorbate concentrations (0.16 +/- 0.02). Coadministration of allopurinol and dynorphin further spiked the ascorbate levels (0.28 +/- 0.05, p < 0.01). In conclusion, the results suggested that ischemia-induced arrhythmia mechanisms might involve the formation of superoxide and other ROS, which were probably generated from the release of EOP (or EOP/EOP receptor interactions). Superoxide, the formation of which can be inhibited by allopurinol that exerted antiarrhythmic effect, was probably scavenged by ascorbate in myocardial ischemia. The ROS resulting from EOP/EOP receptor interactions were probably scavenged by glutathione system. Elevated ascorbate levels in dynorphin-treated hearts might result from the compensatory synthesis induced by decreased glutathione levels.