The p53 tumor suppressor that plays a central role in the cellular response to genotoxic stress was suggested to be associated with the DNA repair machinery which mostly involves nucleotide excision repair (NER). In the present study we show for the first time that p53 is also directly involved in base excision repair (BER). These experiments were performed with p53 temperature-sensitive (ts) mutants that were previously studied in in vivo experimental models. We report here that p53 ts mutants can also acquire wild-type activity under in vitro conditions. Using ts mutants of murine and human origin, it was observed that cell extracts overexpressing p53 exhibited an augmented BER activity measured in an in vitro assay. Depletion of p53 from the nuclear extracts abolished this enhanced activity. Together, this suggests that p53 is involved in more than one DNA repair pathway.z 1999 Federation of European Biochemical Societies.

The p53 tumor suppressor gene plays an important role in both apoptosis and DNA repair pathways that are pivotal for genomic stability. Here we show that the treatment of cells with low doses of gamma-irradiation or cisplatin resulted in an immediate enhancement of p53-dependent DNA repair, measured by base excision repair (BER) activity. However, treatment of cells with high doses of DNA damaging agents resulted in a reduction in p53-dependent DNA repair and in the induction of p53-dependent apoptosis. Analysis of p53 upstream molecular events suggested that regulation of p53-associated DNA repair is ATM-dependent. Furthermore, we observed that while dephosphorylation of Ser376 at the C-terminus of the p53 protein was associated with enhancement in DNA repair, phosphorylation at the N-terminal Ser15 resulted in the reduction in DNA repair. The latter is also in correlation with an enhancement in the specific DNA binding activity and in the induction of apoptosis. Treatment of cells with a caspase inhibitor, prior to the damaging agent-blocked apoptosis, had no effect on the DNA repair pattern. Taken together, this suggests that the decision of cells to induce a p53-dependent DNA repair or apoptosis is most probably controlled by the level of genotoxic agent introduced to cells.

p53 is involved in several DNA repair pathways. Some of these require the speci®c transactivation of p53-dependent genes and others involve direct interactions between the p53 protein and DNA repair associated proteins. Previously, we have shown that p53 acts directly in Base Excision Repair (BER) when assayed under in vitro conditions. Our present data indicate that this involvement is independent of the transcriptional activity of the p53 molecule. We found that under both in vitro and in vivo conditions, a p53 transactivation-de®cient molecule, p53-22-23 was more ecient in BER activity than was wild type p53. However, mutations in the core domain or C-terminal alterations strongly reduced p53-mediated BER activity. These results are consistent with the hypothesis that the involvement of p53 in BER activity, a housekeeping DNA repair pathway, is a prompt and immediate one that does not involve the activation of p53 transactivation-dependent mechanisms, but rather concerns with the p53 protein itself. In an endogenous DNA damage status p53 is active in BER pathways as a protein and not as a transcription factor. Oncogene (2001) 20, 581 ± 589.