Apoptosis, or programmed cell death (PCD), has been studied by developmental biologists over decades, primarily when dealing with morphogenesis and differentiation. Tumor biologists might have been expected to recognize its significance earlier, in view of the fact that 2 of the hallmarks of a tumor are its loss of morphological tissue structure and frequently its de-differentiation. When, however, oncogenes were found which prevented apoptosis (Bcl-2) and tumor-suppressor genes ( p 5 3 ) which promoted apoptosis, and when some of them were shown to have effects on the cell cycle, apoptosis became a familiar topic among molecular-tumor biologists, the more so when it could be shown that chemotherapeutic agents (e.g., adriamycin) triggered apoptosis.
Most of the discussions at this meeting were devoted, not to the downstream events or the process of apoptosis itself, but rather to the regulation of apoptosis induction, i.e., the level at which interrelationships with the neoplastic process are being sought. Since 2 earlier meetings in this series dealt with tumor-suppressing genes (Burger and Croce, 1990) and with thep53 gene and its products (Levine and Burger, 1992) and a sizeable literature onp53 and apoptosis has accumulated, we divided the meeting into a "pS3" day and a "mostly non-pS3" day.
The "mostly non-p53"day
In the introductory discussion it was generally agreed that the criteria used to define apoptosis should not be the peculiar DNA degradation of apoptosis, but the characteristic nuclear and total cellular shrinkage. Subsequently Dr. Frederick 0. Cope reported the isolation of a p49 protein which is expressed both by breast-tumor cells and by intestinal epithelial cells that are resistant to apoptosis. This protein can inhibit apoptosis in numerous cell types. It was also noted that intestinal epithelial cells express early kinase (EAK-1) when apoptosis is induced. This is blocked by p49. p49 also blockspS.3 and CKII induction by certain chemotherapeutic agents. The integration of these elements was discussed in the context of a cell-cycle-apoptosis oscillatory regulatory model.
In a thought-provoking survey, Dr. Martin C. Raff suggested that all of our cells (except perhaps blastomeres) are capable of undergoing apoptosis and constitutively express all of the proteins required to execute the death program. The program operates by default when cells fail to receive survival signals from other cells. The program can run and Bcl-2 can protect cells in the absence of a nucleus, mitochondria1 respiration or cdc-2-kinase activity, and in the virtual absence of oxygen: in conditions in which reactive oxygen species are unlikely to be generated.
As seen in mammalian cells, over-expression of wild-type humanpS3 can inhibit cell growth, and is lethal in the fission yeast Schizosaccharomyces pornbe. Dr. James R. Bischoff described a novel human calcium-binding protein that can suppress the lethal effectts) ofp.53 in S. pornbe and cause delay of apoptosis in IL3-dependent pre-B cells. This observation suggests that fission yeast and mammalian cells die by a common mechanism. He also reported that a colleague at his institution, Dr. Maria J. Fernandez-Sarabia, found a rus-related protein (R-ras, p23) which binds specifically to Bcl-2. The latter turned out to bind to the nucleotide-free form of R-ras, p23, suggesting that Bcl-2 may play a role in regulating the activation of R-ras.
Dr. Douglas R. Green presented a signal-transduction cascade leading to apoptosis, in which ligation of the CD95/Fas molecule activates the hydrolysis of sphingomyelin to ceramide which in turn activates Ras. The resultant apoptosis is dependent on this Ras activation. H e further described activation-induced apoptosis in T-cell hybridomas, where expression and interaction of Fas and Fas-ligand results in cell death, even in a single cell.
The expression and interaction of members of the Bcl-2 family play a certain role in determining the apoptotic threshold of a cell. Dr. Craig B. Thompson pointed out that, in lymphocytes, Bcl-2 and Bcl-xL work coordinately to regulate the resistance of the cell to apoptosis. Bcl-2 sets the steady-state apoptosis threshold of the cell, while Bcl-xL expression is responsive to extracellular stimuli, allowing for dynamic variation in the apoptosis threshold during an immune response.
Dr. Peter J.A. Davies presented evidence suggesting that the induction of tissue transglutaminase is a frequent occurrence in cells undergoing apoptotic cell death. In myeloid leukemia cells, both the induction of the transglutaminase and the expression of apoptosis depend on ligand activation of retinoid X receptors (RXR). This retinoid-induced accumulation of tissue transglutaminase in apoptotic cells may facilitate the formation of apoptotic bodies by transglutaminase-mediated activation of phospholipase C or by enzyme-mediated promotion of irreversible protein cross-linking reactions.
As a frame for the molecular studies on apoptosis, Dr. Rolf Schulte-Hermann investigated the relevance of active cell death for cancer development in vivo, using rat liver as a model. The rates of birth and death of cells increase steadily from normal to pre-neoplastic to adenoma and carcinoma cells. Tumor promoters inhibit apoptosis in pre-neoplastic cell foci, thereby favoring their preferential growth. Anti-promoting regimens, such as promoter withdrawal or fasting, preferentially enhance apoptosis in (pre)-neoplastic lesions, and can lead to extinction of initiated clones.
~
This meeting was sponsored by Hoffmann-LaRoche Inc., Nutley, NJ, USA. The Tumor Biology Programme of UICC (Dr. R. Brentani) and the participants express their gratitude for the unselfish support and low-key presence of the sponsor.