In a number of experimental systems, the early stage of the apoptotic process, i.e., the stage that precedes nuclear disintegration, is characterized by the breakdown of the inner mitochondrial transmembrane potential (delta psi(m)). This delta psi(m) disruption may involve mitochondrial permeability transition (PT). Here, we address the question of whether PT transition would suffice to cause apoptosis or, rather, it would constitute a secondary event not causally involved in the apoptotic cascade. Protoporphyrin IX (PPIX), a ligand of the mitochondrial benzodiazepin receptor that is well known for its PT-triggering capacity, induces delta psi(m) disruption, enhanced generation of superoxide anions, as well as signs of nuclear apoptosis in thymocytes and in T cells. The sequence of events triggered by PPIX mimics that observed in natural apoptosis. The PT inhibitory compound bongkrekic acid, a specific ligand of the mitochondrial adenine nucleotide translocator, is an efficient inhibitor of protoporphyrin IX-induced delta psi(m) disruption. Bongkrekic acid prevents all PPIX-induced phenomena, including superoxide anion generation, chromatinolysis, and oligonucleosomal DNA fragmentation. In contrast, inhibitors of mRNA or protein synthesis fail to suppress PPIX-induced delta psi(m) disruption and apoptosis. Transfection-enforced hyperexpression of the apoptosis-inhibitory proto-oncogene bcl-2 also inhibits PPIX-induced delta psi(m) disruption, hyperproduction of reactive oxygen species, and nuclear DNA loss. The delta psi(m)-stabilizing effect of Bcl-2 is observed both in cells and in isolated mitochondria. In conclusion, these data are compatible with the hypotheses that mitochondrial PT is self-sufficient to trigger apoptosis and that Bcl-2 may directly regulate PT.