Prooxidants open both the mitochondrial permeability transition pore and a low-conductance channel in the inner mitochondrial membrane.
Mitochondria can be induced by a variety of agents/conditions to undergo a permeability transition (MPT), which nonselectively increases the permeability of the inner membrane (i.m.) to small (<1500 Da) solutes. Prooxidants are generally considered to trigger the MPT, but some investigators suggest instead that prooxidants open a Ca(2+)-selective channel in the inner mitochondrial membrane and that the opening of this channel, when coupled with Ca(2+) cycling mediated by the Ca(2+) uniporter, leads ultimately to the observed increase in mitochondrial permeability [see, e.g., Schlegel et al. (1992) Biochem. J. 285, 65]. S. A. Novgorodov and T. I. Gudz [J. Bioenerg. Biomembr. (1996) 28, 139] propose that the i.m. contains a pore that, upon exposure to prooxidants, can open to two states, one of which conducts only H(+) and one of which is the classic MPT pore. Given the current interest in increased mitochondrial permeability as a factor in apoptotic cell death, it is important to determine whether i.m. permeability is regulated in one or multiple ways and, in the latter event, to characterize each regulatory mechanism in detail. This study examined the effects of the prooxidants diamide and t-butylhydroperoxide (t-BuOOH) on the permeability of isolated rat liver mitochondria. Under the experimental conditions used, t-BuOOH induced mitochondrial swelling only in the presence of exogenous Ca(2+) (>2 microM), whereas diamide was effective in its absence. In the absence of exogenous inorganic phosphate (P(i)), (1) both prooxidants caused a collapse of the membrane potential (DeltaPsi) that preceded the onset of mitochondrial swelling; (2) cyclosporin A eliminated the swelling induced by diamide and dramatically slowed that elicited by t-BuOOH, without altering prooxidant-induced depolarization; (3) collapse of DeltaPsi was associated with Ca(2+) efflux but not with efflux of glutathione; (4) neither Ca(2+) efflux nor DeltaPsi collapse was sensitive to ruthenium red; (5) collapse of DeltaPsi was accompanied by an increase in matrix pH; no stimulation of respiration was observed; (6) Sr(2+) was able to substitute for Ca(2+) in supporting t-BuOOH-induced i.m. depolarization, but not swelling; (7) in addition to being insensitive to CsA, the collapse of DeltaPsi was also resistant to trifluoperazine, spermine, and Mg(2+), all of which block the MPT; and (8) DeltaPsi was restored (and its collapse was inhibited) upon addition of dithiothreitol, ADP, ATP or EGTA. We suggest that these results indicate that prooxidants open two channels in the i.m.: the classic MPT and a low-conductance channel with clearly distinct properties. Opening of the low-conductance channel requires sulfhydryl group oxidation and the presence of a divalent cation; both Ca(2+) and Sr(2+) are effective. The channel permits the passage of cations, including Ca(2+), but not of protons. It is insensitive to inhibitors of the classic MPT.