Low ATP levels interfered with the apoptotic execution at two central steps: (1) the release of mitochondrial cytochrome c to the cytosol, an event directly leading to caspase activation in apoptosis, was delayed by 1 h, and (2) execution caspases were not activated. In ATP depleted cells, the extent of STS- or anti-CD95-induced cytochrome c translocation to the cytosol was similar to that found in cells with high ATP levels. Nevertheless, caspases were not activated in ATP depleted cells. This suggests that one ATP-dependent step is at the level of the caspase-activating apoptosome-complex between Apaf-1, pro-caspase-9, and cytochrome c. The absence of caspase activity in ATP depleted cells was not due to an early inactivation of the apoptosome components, because caspases in cytosols isolated from such cells could still be activated in a cell-free system. Therefore, the lack of adequate ATP concentrations seems to be directly responsible for the block of the apoptotic execution at this step. However, at least in a cell-free system, dATP promoted the apoptosome activation more efficiently than ATP.
By examining in more detail the events upstream and downstream to the mitochondrial control of apoptosis it was also demonstrated that the release of cytochrome c is due to a non-selective translocation mechanism. In Jurkat cells, this was true for apoptosis, induced by a variety of substances, and for the above necrosis model. At least one other mitochondrial intermembrane protein, namely the adenylate kinase, is released in parallel to cytochrome c, suggesting a non-selective permeabilization of the outer mitochondrial membrane during cell death. Caspases, once activated, have been shown to permeabilize the outer mitochondrial membrane. However, such a secondary proteolytic degradation was not the reason for the non-selective release in our system, since the caspase-inhibitor zVADfmk did neither alter the kinetics nor the extent of the STS-induced translocation of both proteins.
The endogenous mediator nitric oxide (NO) was able to block the execution of apoptosis, and converted death to necrosis. Like other inhibitors of mitochondrial ATP-synthesis, NO delayed the release of cytochrome c and inhibited caspase activation. This was solely due to ATP depletion by the inhibition of mitochondrial oxidative phosphorylation. Graded repletion of ATP by glucose supplementation restored activation of caspase-3, -7, and -8, and the formation of apoptotic features, including the exposure of phosphatidylserine on the cell surface. This result may be relevant for the understanding of pathologic situations with enhanced NO production and localized high rates of necrotic cell death.
The anti-apoptotic protein Bcl-2 was able to protect from STS-induced necrosis in ATP depleted cells. Overexpression of Bcl-2 blocked the release of cytochrome c in cells with high or low ATP levels. As shown before, in cells with high ATP levels, Bcl-2 also prevented the loss of the mitochondrial membrane potential. In ATP depleted cells, however, Bcl-2 did not inhibit an initial STS-induced drop of the mitochondrial membrane potential. Therefore, the mechanisms whereby Bcl-2 prevents cell death and favors retention of cytochrome c in the mitochondria neither require the maintenance of the mitochondrial membrane potential?nor of normal ATP levels. Thus, Bcl-2 can reduce the overall cell death regardless whether it occurs by apoptosis or necrosis, which further implies the existence of common signaling events in these two modes of cell death.
Overall, this study shows that decisions on the execution of cell death may be taken downstream to the activation of a given death pathway. While some of the events in cell death signaling may be common to several death pathways (e.g. mitochondrial dysfuction), others (e.g. caspase activation) may be more specific for the execution of apoptosis.