We evaluated the ability of alpha-phenyl-tert-butyl nitrone (PBN) to trap free radicals and to protect the rat myocardium during ischemia and reperfusion. Isolated bicarbonate buffer-perfused hearts (n = 8) were subjected to 20 min global ischemia (37 degrees C) followed by reperfusion with 0.4 to 4.0 mM PBN. Coronary effluent containing the PBN adduct was extracted in toluene. Electron spin resonance analysis of the toluene extract revealed a PBN-hydroxyl adduct. To verify this assignment, a Fenton system was used to generate an authentic PBN-hydroxyl adduct (n = 8), which yielded the same ESR spectra as the reperfusion-derived adduct. The structure of the adduct formed in the Fenton system was confirmed by gas chromatography-mass spectrometry. The ESR parameters of the PBN-hydroxyl adduct were exquisitely sensitive to solvent polarity during extraction of the adduct. Extraction of an authentic PBN-hydroxyl adduct into chloroform, chloroform:methanol, and toluene closely matched the ESR parameters obtained during reperfusion of ischemic myocardium in other animal models. To determine whether PBN could confer any protective effect during ischemia or reperfusion, hearts (n = 8/group) were subjected to 35 min global ischemia at 37 degrees C with the St. Thomas' II cardioplegic solution followed by 30 min reperfusion. Percent recovery (mean +/- SEM) of developed pressure, rate pressure product, and leakage of lactate dehydrogenase during reperfusion in control hearts were 58 +/- 3%, 48 +/- 4% and 3.2 +/- 0.5 IU/15 min/g wet wt. PBN at a concentration of 0.4 mM or 4.0 mM when present either during ischemia alone or reperfusion alone did not exert any effect upon recovery of developed pressure, rate pressure product or post-ischemic enzyme leakage. We conclude that PBN fails to improve contractile recovery and reduce enzyme leakage during reperfusion of myocardium subjected to global ischemia.