Malignant melanoma (MM) is a singular challenge to modern oncology as it is one of the few human cancers in which the age-adjusted incidence has increased throughout the last decade, and it is notoriously refractory to therapy. The genetics of MM in humans suggest two general classes of lesions, receptor tyrosine kinase (RTK) activation, and deletion of the Ink4a/Arf locus at 9p21. Mice with melanocyte specific overexpression of H-Ras (a signaling intermediate of several RTKs) and Ink4a/Arf loss have been previously shown to develop melanoma with high penetrance and short latency. The Ink4a/Arf locus, conserved in mouse and humans, encodes two distinct proteins, p16INK4a and p19ARF, both of which regulate critical tumor suppressor pathways, and each may play a substantive role in human melanoma. Mutational analysis of human tumors and melanoma- prone kindreds clearly establish p16INK4a as a suppressor of melanoma formation, while the role of p19ARF in human tumors is less clear. Previous work in the mouse, however, shows p19ARF, through its regulation of p53, to be a critical tumor suppressor gene in a wide range of tissues. I propose to define genetically the relative importance of these two proteins in murine melanoma using two novel but as of yet uncharacterized strains of knockout mice. Using standard gene targeting techniques coupled with CRE recombinase mediated excision of the remnant selectable marker (PGK-Neo), I have produced two independent knockout (KO) strains with specific deletion of either p16INK4a or p19ARF. The work described in this proposal involves the characterization in vitro and in vivo of the p16INK4a specific KO, with a particular emphasis on melanocyte biology and melanoma formation. To c o mplement the in vivo analysis of the effect of p16INK4a loss on tumorigenesis, I will cross the p16INK4a KO to a p53-deficient background to look for increased tumor formation and change in tumor spectrum. Melanocyte growth from both p16INK4a and p19ARF null mice will be characterized in vitro, with particular emphasis on their relative transformability. Finally, I will attempt to develop a well-defined murine model of melanoma by crossing the p16INK4a and p19ARF specific KO mice to H-RAS overexpressing mice, in the hope of establishing which of the proteins, or perhaps both, is the critical mediator of melanoma susceptibility in mice in vivo. Such a model could then serve as a foundation for future studies of melanoma progression and therapy, and the results of this work will have implications for the treatment of human MM.
