Angiotensin converting enzyme (ACE) inhibitors enhance contractile function of myocardium "stunned" by a brief episode of coronary artery occlusion, yet their mechanism(s) of action remain unresolved. In addition to possible hemodynamic effects, ACE inhibitors may stimulate the synthesis of cardioprotective prostaglandins. Furthermore, the beneficial effects of ACE inhibitors that contain a sulfhydryl group may be due in part to the ability of thiol compounds to act as nonspecific antioxidants or direct scavengers of cytotoxic oxygen-derived free radicals. To investigate this question we compared the effects of (1) the sulfhydryl-containing ACE inhibitor zofenopril, (2) the sulfhydryl-containing stereoisomer of captopril (SQ 14,534) with essentially no ACE inhibitor properties, (3) the nonsulfhydryl-containing ACE inhibitor enalaprilat, and (4) solvent alone, given at the time of reperfusion, on recovery of contractile function after 15 minutes of coronary occlusion in the anesthetized open-chest dog. Segment shortening in control animals remained depressed or "stunned" after reperfusion, recovering to only -5 ± 12% of baseline preocclusion values at 3 hours after reperfusion. In contrast, all three treatment agents attenuated postischemic dysfunction: segment shortening was restored to 33 ± 12%, 54 ± 6%, and 83 ± 5% of baseline values at 3 hours after reflow in dogs treated with SQ 14,534 (p < 0.05), zofenopril (p < 0.01), and enalaprilat (p < 0.01), respectively (all vs control value). These improvements in segment shortening did not appear to be the result of altered oxygen supply or demand after reperfusion, inasmuch as no significant differences in systemic hemodynamic parameters or myocardial blood flow were observed among the groups. In the second phase of the study, we found that the improved contractile function associated with enalaprilat treatment could largely be reversed by infusion of the potent cyclooxygenase inhibitor indomethacin: segment shortening was reduced from 69 ± 12% at 2 hours after treatment/reperfusion to 38 ± 12% at 2 hours after indomethacin infusion (p < 0.01 vs 2 hours after reperfusion). Infusion of indomethacin had no effect, however, on the improved contractile function associated with zofenopril treatment. We therefore conclude that sulfhydryl- versus nonsulfhydryl-containing agents enhance contractile function of stunned myocardium by different mechanisms of action: enalaprilat attenuates postischemic dysfunction at least in part by a prostaglandin-mediated mechanism, whereas the salutary effects of zofenopril and SQ 14,534 may be due in part to the antioxidant properties of the sulfhydryl moiety.