Hydrogen peroxide (H2O2) serves as a precursor for highly reactive oxygen intermediates. However, the respiratory function of myocytes is relatively resistant to exogenously administered H2O2. In this study, we examined whether or not the reduction of cellular defense increases the toxicity of H2O2. Rat heart myocytes were isolated by collagenase digestion. Respiratory rates of myocytes, suspended in a medium containing sucrose, 3-N-morpholino-propanesulfonic acid, EGTA and bovine serum albumin, were determined polarographically in the presence of pyruvate and malate with or without 2,4-dinitrophenol (DNP). Mitochondrial membrane potentials were measured by using [3H]triphenylmethylphosphonium+. Cellular defense was attenuated by i) inhibiting the catalase activity by 3-amino-1,2,4-triazole (AT), ii) reducing the glutathione concentration by diethyl maleate (DEM) or ethacrinic acid (EA), and iii) permeabilizing the sarcolemmal membrane by saponin. The dose-response relationship between H2O2 (0.1-5 mM) and mitochondrial membrane potential was not greatly affected by these experimental conditions. Myocyte respiration was inhibited by 5 mM H2O2, particularly that measured in the presence of DNP (48% of control). DEM treatment did not significantly affect the respiratory inhibition by H2O2, whereas the degree of inhibition was somewhat greater following EA or AT treatment. By contrast, the sensitivity of cellular respiration to H2O2 was potentiated approximately two orders of magnitude by the permeabilization of sarcolemmal membrane; thus, 100 microM H2O2 inhibited both DNP-stimulated and unstimulated respiration to 17% and 35% of control, respectively. The results indicate that factors existing in the sarcolemma and/or in the cytosol, which become ineffective and/or are diluted, respectively, following permeabilization with saponin, are important cellular defense mechanisms in alleviating the toxic effect of exogenous H2O2 on the respiration of mitochondria in situ in myocytes.