A mouse mutant lacking the p53 proline-rich domain is deficient in cell cycle arrest, but competent for apoptosis and tumor suppression

3636

The p53 tumor suppressor gene maintains the integrity of the genome by initiating the pathways for growth arrest, apoptosis, or modulating certain types of DNA repair in response to various stresses that contribute to tumorigenicity. Protein-protein interactions at a proline-rich region, which lies between the transactivation and DNA binding domains, are proposed to contribute to how p53 integrates signals to execute different programs. Transfection studies suggest that deleting the proline-rich region of p53 preserves cell cycle arrest but prevents apoptosis, however such studies have been inconsistent. We used homologous recombination to generate a mouse mutant lacking the p53 proline-rich domain to gain a more accurate understanding of the p53 pathway in vivo. The mouse proline-rich deletion (ΔP) exhibits different responses to stress than indicated by transfection studies. In ΔP MEFs, the ΔP protein was less abundant than wild-type (wt) p53 before and after stress, most likely because of increased sensitivity to MDM2-mediated degradation. ΔP was able to transactivate the target genes p21, MDM2, Noxa and Puma, although the resulting transcripts were less abundant than for wtp53. ΔP MEFs did not undergo cell cycle arrest, likely because of insufficient p21 levels. By contrast, ΔP can induce apoptosis in both E1A infected MEFs and thymocytes after DNA damage. To further investigate the effects of ΔP in a relevant in vivo setting, we tested the ability of ΔP to rescue MDM2 or MDMX deficiency in mice, since the early embryonic lethality in MDM2 and MDMX deficient mice have been attributed to induction of apoptosis or cell cycle arrest, respectively. The ΔP allele was unable to rescue MDM2 deficiency but was able to rescue MDMX deficiency. We next evaluated the capacity of the ΔP allele to suppress tumorigenicity. We compared tumor formation by E1A and Ras expressing p53 wt, p53 null, and ΔP MEFs. Cells expressing wild type p53 failed to form tumors while ΔP and p53 null MEFs both formed tumors of similar size, indicating that ΔP p53 is a poor suppressor of oncogene-induced tumors in a xenograft model. We are currently studying the onset and spectrum of tumors in p53 ΔP/ΔP mice. Our data reveals the importance of using in vivo models to evaluate structure-function relationships in the p53 pathway. This model will be important for future analyses of the impact of p53-mediated cell cycle arrest and apoptotic programs to tumor suppression in different tissues.