p53 mutants exhibiting enhanced transcriptional activation and altered promoter selectivity are revealed using a sensitive, yeast-based functional assay

Changes in promoter specificity and binding affinity that may be associated with p53 mutations or post-translational modifications are useful in under standing p53 structure/ function relationships and categorizing tumor mutat ions. We have exploited variable expression of human p53 in yeast to identi fy mutants with novel phenotypes that would correspond to altered promoter selectivity and affinity. The p53 cDNA regions coding for the DNA binding a nd tetramerization domains were subjected to random PCR mutagenesis and wer e cloned directly by recombination in yeast into a vector with a GAL1 promo ter whose level of expression could be easily varied. p53 variants exhibiti ng higher than wild type levels of transactivation (supertrans) for the RGC responsive element were identified at low level of p53 protein expression. All the p53 mutants obtained with this screen were located in the DNA bind ing domain. Two out of 17 supertrans mutants have been found in tumors. Six mutations were in the L1 loop region between amino acids 115 and 124, The transactivation potential of a panel of supertrans p53 mutants on different promoters was evaluated using the p53 responsive elements, RGC, PIG3, p21 and bax. Although all mutants retained some activity with all promoters, we found different patterns of induction based on strength and promoter speci ficity. In particular none of the mutants was supertrans for the p21 respon sive element, Interestingly, further analysis in yeast showed that the tran sactivation function could be retained even in the presence of dominant-neg ative p53 tumor mutations that could inhibit wild type p53, Five mutants we re also characterized in human cells in terms of growth suppression and tra nsactivation of various promoters. These novel supertrans p53 mutants may b e useful in studies aimed at dissecting p53 downstream pathways, understand ing specific interactions between p53 and the DNA, and could replace wild t ype p53 in cancer gene therapy protocols. The approach may also prove usefu l in identifying p53 tumor mutations.