Several reactions in biological systems contribute to maintain the steady-state concentrations of superoxide anion (O ) and hydrogen peroxide (H₂O₂). The electron transfer chain of mitochondria is a well documented source of H₂O₂; however, the release of O from mitochondria into cytosol has not been unequivocally established. This study was aimed at validating mitochondria as sources of cytosolic O , elucidating the mechanisms underlying the release of O from mitochondria into cytosol, and assessing the role of outer membrane voltage-dependent anion channels (VDACs) in this process. Isolated rat heart mitochondria supplemented with complex I or II substrates generate an EPR signal ascribed to O . Inhibition of the signal in a concentration-dependent manner by both manganese-superoxide dismutase and cytochromec proteins that cannot cross the mitochondrial membrane supports the extramitochondrial location of the spin adduct. Basal rates of O release from mitochondria were estimated at ∼0.04 nmol/min/mg protein, a value increased ∼8-fold by the complex III inhibitor, antimycin A. These estimates, obtained by quantitative spin-trapping EPR, were confirmed by fluorescence techniques, mainly hydroethidine oxidation and horseradish peroxidase-basedp-hydroxyphylacetate dimerization. Inhibitors of VDAC, 4′-diisothiocyano-2,2′-disulfonic acid stilbene (DIDS), and dextran sulfate (in a voltage-dependent manner) inhibited O production from mitochondria by ∼55%, thus suggesting that a large portion of O exited mitochondria via these channels. These findings are discussed in terms of competitive decay pathways for O in the intermembrane space and cytosol as well as the implications of these processes for modulating cell signaling pathways in these compartments.

In this work, the topology of mitochondrial and H₂O₂ generation and their interplay with matrix GSH in isolated heart mitochondria were examined. We observed that complex I releases into the matrix (where it is converted to H₂O₂ by Mn-SOD) but not into the intermembrane space. No free radical generation was observed from complex II, but succinate treatment caused H₂O₂ generation from the matrix through a reverse electron flow to complex I. Complex III was found to release into the matrix and into the intermembrane space. Antimycin, which increases steady-state levels of (ubisemiquinone at the Qo site) in complex III, enhanced both H₂O₂ generation from the matrix and production from the intermembrane space. On the other hand, myxothiazol, which inhibits formation, completely inhibited antimycin induced toward the intermembrane space and inhibited H₂O₂ generation from the matrix by 70%. However, myxothiazol alone enhanced H₂O₂ production from complex III, suggesting that other components of complex III besides the can cause generation toward the matrix. As expected, mitochondrial GSH was found to modulate H₂O₂ production from the matrix but not generation from the intermembrane space. Low levels of GSH depletion (from 0—40%, depending on the rate of H₂O₂ production) had no effect on H₂O₂ diffusion from mitochondria. Once this GSH depletion threshold was reached, GSH loss corresponded to a linear increase in H₂O₂ production by mitochondria. The impact of 50% mitochondrial GSH depletion, as seen in certain pathological conditions in vivo, on H₂O₂ production by mitochondria depends on the metabolic state of mitochondria, which governs its rate of H₂O₂ production. The greater the rate of H₂O₂ generation the greater the effect 50% GSH depletion had on enhancing H₂O₂ production.