Proteolysis and the biochemistry of life-or-death decisions

A Poptosis, the process by which cells commit suicide, is ubiquitous in both normal and diseased tissues, and its mechanism is conserved in animal species from humans at least to nematodes. Apoptotic cells undergo a stereotypical set of changes--shrinkage, cleavage of nuclear DNA, condensation of chromatin, fragmentation of the cytoplasm-that distinguish them from cells dying by other means. In embryogenesis, this process is required to form adult structures (as when digits become separated) and to delete immune cell clones specific for self-antigens, and in adult tissues it removes cells that have suffered damage to their genetic material. Thus, apoptosis helps to establish normal tissues and prevents cell proliferation that could otherwise lead to pathologic conditions such as autoimmunity or cancer. The recognition that a protein required for apoptosis in the worm, CED-3, is a homologue of the mammalian cysteine proteinase interleukin 1[3-converting enzyme (ICE; 1) set the stage for much of the current work on apoptosis. It is now clear that at least five proteinases in the ICE family are expressed in mammalian cells, and that the activity of some or all of these enzymes is required for apoptosis to occur (2). Inappropriate expression of intracellular proteinases including ICE has been shown to drive cells into the apoptotic program. Conversely, normal apoptosis can be prevented by supplying cells with proteinase inhibitors such as CrmA, a viral protein that acts specifically on proteinases of the ICE family (3). On the basis of results such as these, it seems clear that proteolysis is crucially involved in apoptosis, but it is much less clear why this is so. Apoptotic cells do not simply proteolyze themselves out of existence, and at least in the early stages of the process when the cell becomes irreversibly committed to dying, the proteinases induced are extremely limited in their specificity and the number of target proteins is small. Evidently, proteolysis of certain key molecules serves to flip a switch, initiating a series of steps that culminate in the death of the cell and its processing into apoptotic fragments that can be phagocytosed by neighboring cells. Valuable hints about the nature of this switch come from the work of Casciola-Rosen et al. (4) in this issue of the Journal of Experimental Medicine. In previous work (2, 5-10) these and other investigators identified several proteins that are targets ofapoptotic proteolysis, and therefore are candidates to be part of the switch. In the current study, they use a specific inhibitor ofapopain, a proteinase of the ICE faroily, and show by several criteria that apopain rather than ICE is responsible for the proteolysis they observe in apoptotic cells. This is consistent with the observation that in mice deficient for ICE, apoptosis still occurs in most cell types. Many of the known proteolytic targets are localized in the nucleus of normal cells: lamins; globular actin; the nuclear mitotic apparatus protein NuMA; the U1-70 kD protein, a component of the KNA splicing machinery; the enzyme poly(ADP-ribose) polymerase (PARP); and the catalytic subunit of the DNA-dependent protein kinase (DNA-PK~s). Of these, U1-70 kD, PARP and DNA-PKc~ are believed to be cleaved by apopain (5, 8, 10), and at least the latter two appear to be involved in regulating the cell cycle in response to DNA damage. This function suggests a model for the apoptotic switch in which damage to DNA activates apopain to cleave a set of nuclear proteins that control the progression of cells through the cell cycle or commit it to apoptosis.