Activation of the tumor suppressor protein p53 leads to cell cycle arrest and/or apoptosis. The p53 protein therefore provides a critical brake on tumor development. The p53 gene is mutated in over half of all human cancers and the p53 pathway is thought to be impaired in the majority of the remainder. Primary tumors derived from the neural crest, including malignant melanoma and neuroblastoma, exhibit a low rate of p53 mutation. Frequently cells derived from these tumors require expression of the embryological oncogene Pax3 for survival. Pax3 is a transcription factor essential for embryological development. Genetic and pharmacological studies in mice have shown that a major role of Pax3 in the developing neural crest is to keep p53 down-regulated until a specific developmental program is completed. We hypothesised that Pax3 expression may lead to p53 functional inhibition in neural crest derived tumors. We have shown that in the malignant melanoma cell line UACC-62, down-regulation of Pax3 using specific siRNA leads to a 91% decrease in cell viability. This is associated with increased PARP cleavage, a marker of apoptosis. This reduction in cell viability in response to Pax3 down-regulation is completely rescued by the addition of p53 specific siRNA. In a model system, Pax3 inhibits the induction of the p53 responsive pro-apoptotic gene Bax by over 50% (p<0.01). In the same system Pax3 leads to a fall in p53 protein level by up to 89% relative to controls; this is associated with a decrease in p53 half life from 8 hours to 2 hours. To define the structural elements of Pax3 required for p53 inhibition we have created serial truncations of the Pax3 protein. We have also inserted point mutations in Pax3 at sites known to cause Pax3 related diseases in humans and mice. Any mutation that we have made that affects the ability of Pax3 to function as a transcription factor abolishes its repression of p53. Our data identifies an essential role for fully functional Pax3 in the suppression of wild-type p53 function in neural crest derived tumors, including malignant melanoma. Future work will focus on the mechanisms underlying the functional interaction between these two proteins and the potential for therapeutic intervention in this pathway.

[Proc Amer Assoc Cancer Res, Volume 47, 2006]