ELUCIDATION OF THE p53-DEPENDENT PROGRAMMED CELL DEATH PATHWAY

The tumor suppressor gene, p53, is mutated in over 50% of human tumors. The normal function of p53 protein is to respond to a number of different stimuli, including DNA damage, cellular transformation (activated oncogenes) and hypoxia (insufficient oxygen concentrations) by inducing either cell growth arrest or programmed cell death (apoptosis). It is now clear that the primary tumor suppressor function of p53 relies on its ability to induce apoptosis during the initiation and progression of a tumor. Currently, the downstream events mediating p53-dependent apoptosis are largely unknown. p53 is a transcription factor that activates the expression of particular genes, and down-regulates, or represses, the expression of others.

Multiple lines of data from this and other laboratories implicate the transcriptional repression function of p53 as necessary for the ability of this protein to induce apoptosis. We have successfully isolated the first p53-repressed gene, and have shown that repression of transcription of this gene by p53 is necessary for the induction of apoptosis. The goal of future work is to continue the identification and characterization of p53-repressed genes, and to elucidate the mechanism of transcriptional repression by this tumor suppressor protein. A natural corollary of this goal is to further assess the importance of this activity to the ability of p53 to induce programmed cell death or apoptosis.

Employing an inducible-p53 cell line system that distinguishes between p53mediated growth arrest and apoptosis, we have found that this tumor suppressor protein can repress the expression of specific genes. Importantly, transcriptional repression by p53 occurs during p53-mediated apoptosis, but not when apoptosis is inhibited by the proto-oncogene, bcl-2, and growth arrest ensues. Therefore, we have identified transcriptional repression of gene expression by p53 as an essential component of the downstream pathway of p53-dependent apoptosis. The identification and characterization of genes whose expression is repressed by p53 is likely to be a fundamental step toward understanding how this protein induces apoptosis, and, logically, how this protein suppresses tumor growth.

Using a combination of differential display, RNAse protection, and micro-chip oligo-nucleotide array technologies, we have identified over fifty known and novel genes whose expression is down-regulated following p53 induction. One of these genes encodes the microtubule-associated protein, Map4. Map4 is a ubiquitously-expressed protein that polymerizes and stabilizes microtubules. Interestingly, many of the genes identified by micro-array analysis as p53-repressed genes encode components of the microtubule network. These include alpha tubulin, beta tubulin, gamma tubulin, and the microtubule-associated proteins, Map4 and stathmin. In collaboration with Dr. George, we have shown that the micro-tubule-destabilizing protein stathmin (also called oncoprotein 18) is repressed only in cells with wild type p53 following ultraviolet and gamma irradiation, as well as during hypoxia. These stimuli are known to activate p53 protein as a transcriptional repressor. The accumulated data indicate that p53 induction during apoptosis interferes with the integrity and normal functions of microtubules, which are essential components of cell division and protein trafficking. Current efforts are focused on an analysis of tubulin gene repression by p53, and on the contribution of altered microtubule dynamics to the process of apoptosis.

Map4 promoter studies. A primary goal of this work will be to elucidate the mechanism of transcriptional repression by p53. Future efforts will focus on more narrowly delineating the p53 binding site in the Map4 promoter, using gel shift assays and mutagenesis analysis. Our data indicate that p53 must cooperate with at least one other DNA-binding protein in order to bind and repress the Map4 promoter. Therefore, these studies will be extended to determine which protein partners are necessary for DNA binding to this element by p53.

The p53-mSin3a-HDAC complex. Many transcriptional repressors use an evolutionarily conserved mechanism for accomplishing repression of gene expression. Specifically, transcriptional repressors like the retinoblastoma protein, pRb, and the Myc-binding protein, Mad, repress transcription by bringing to promoters enzymes called histone deacetylases (HDACs). HDACs function to repress transcription by deacetylating histones, and enhancing their affinity for DNA. This serves to "tighten" up local chromatin, and reduce accessibility to the transcriptional apparatus. We have found that specific chemical inhibitors of HDACs selectively inhibit the ability of p53 to repress the transcription of genes like Map4 and stathmin. Further, in collaboration with Dr. Evans, we have shown that p53 exists in vivo in a complex with HDACs; this interaction is mediated by the evolutionarily conserved corepressor, mSin3a. Significantly, the domains of p53 necessary for repression map exactly to the two mSin3a interaction domains on p53. Future studies will address the mechanisms whereby the p53-mSin3a interaction is controlled in the cell.

We have obtained evidence for the existence of a novel class of genes that are transcriptionally induced only during p53dependent apoptosis, and not during p53-dependent growth arrest. As such, this class of genes represents what may be integral mediators of programmed cell death. Taking advantage of our matched cell line system that distinguishes between p53-mediated growth arrest and apoptosis, we are utilizing a number of molecular approaches to begin to identify and characterize such genes and their protein products. It is believed that characterization of such genes, and elucidation of their role in programmed cell death, will be paramount to our understanding of the downstream pathway of p53-dependent apoptosis.

The publications listed below represent work carried out prior to M.M. joining Fox Chase.

DEL SAL, G., MURPHY, M., RUARO, E.M., LAZAREVIC, D., LEVINE, A.J., SCHNEIDER, C. Cyclin D1 and p21/waf1 are both involved in p53 growth suppression. .Oncogene 12:177-185, 1996.

MURPHY, M., HINMAN, A.H., LEVINE, A.J. Wild type p53 negatively regulates the expression of a microtubule-associated protein. Genes Dev. 10:2971-2980, 1996.

MURPHY, M., LEVINE, A.J. The role of p53 in apoptosis. In Apoptosis Regulatory Genes, edited by C.S. Potten, C. Booth, J.W. Wilson. Thompson Science, pp. 1-31, 1998.

Zhang, C., Yang, J.-M., White, E., Murphy, M., Levine, A.J., Hait, W.N. The role of MAP4 expression in the sensitivity to paclitaxel and resistance to vinca alkaloids in p53 mutant cells. Oncogene 16:1617-1624, 1998.

^a   D.L. George: University of Pennsylvania School of Medicine, Philadelphia PA 19104

^b   R. Evans: The Salk Institute for Biological Studies, La Jolla CA 92138

Illustrations or unpublished data in these reports should not be used without permission of the author.

Fox Chase Cancer Center

Scientific Report 1998