T HE CONCEPT THAT p53 is a tumor-suppressor gene was developed through two major facets. One facet entailed the laboratory experiments in which it was realized that wild-type p53 does not induce malignant transformation, but rather suppresses malignant ind~ ction,"~ whereas the other hinged mostly on the observation that more than 50% of human primary tumors exhibit mutations in the protein and that these mutated products are inactive in rendering cells growth arre~ t. 4.~

The early cDNA p53 constructs isolated were all derived from tumor cell origins. 8-" All these constructs that exhibited sequence variations were found to be very efficient in the induction of malignant transformation when cotransfected into normal primary cells with other oncogenes such as rus or Ela.""'Conversely, p53 cDNAs isolated from normal origins failed to induce any transformation in these assays.'Furthermore, the latter were found to mediate a direct inhibition of the induction of primary foci when added to a mutated p53 in a mixture with ElA. 3 This conclusion was further substantiated by additional experiments using a p53 temperature-sensitive (ts) mutant that at 32" C, when exhibiting the wild-type conformation, induced a cell growth arrest, whereas at 39" C, when exhibiting a mutant conformation, did not induce any growth arrest, but rather enhanced proliferation. I6 The observation that expression of wild-type p53 cDNA induced directly cell growth arrest was repeated in different cell types using a variety of mammalian expression vector^.'"'^ In all these experiments, growth arrest was mediated by wild-type p53 and was found in cell lines that either express endogenously a mutant p53 form'or are totally lacking the expression~ 5 3.'~ Analysis of changes in the cell cycle pattern have indicated that, in most cases, the arrest induced the accumulation of cells at the G, pha~ e.'~.'~ Inactivation of p53 in primary tumors and cell lines was shown to be mediated by several molecular mechanisms. The first p53 nonproducing cell lines reported were the Abelson murine leukemia virus (Ab-MuLV)-transformed pre-B~ ells*~.*~ and the Friend leukemic cell line^.^^,*^ In these examples, the p53 gene was inactivated by integration of viral particles into the p53 gene. In the case of the Ab-MulV cell line, L12, a Moloney viral particle integrated into the first p53 intron, thus rendering the gene inactive. 26 In these cells, a p53 Moloney fusion aberrant mRNA was expressed and no p53 protein was detected." In the case of Friend leukemic