During the past five years, we have discovered several novel genetic alterations in colorectal tumors that are central to its pathogenesis. The most interesting of these affected PIK3CA, a lipid kinase mutated in more than 30% of the tumors. The mutations were clustered in two hot-spots and were altered at similar frequencies and positions in several other common tumor types. The PIK3CA gene is therefore one of the two most highly mutated oncogenes ever identified in human cancers. Our future efforts will be devoted to its further exploration and will incorporate four major components: 1. We will use targeted homologous recombination to disrupt the genes that appear to mediate the biologic effects of PIK3CA in colorectal cancer cells. By comparison of the properties of these cells to those in which mutant or wild-type forms of PIK3CA have been disrupted, we will define the role of each gene in each phenotype associated with PIK3CA mutation. 2. We will determine the changes in transcription associated with PIK3CA mutation using SAGE and microarray technologies. Genes identified through the expression analyses will themselves be subject to targeted disruption to test their physiologic roles in colorectal cancer cells. 3. We will develop small molecular inhibitors of the p110a enzyme encoded by PIK3CA, preferably specific for the mutant form but at minimum specific for the a isoform. Cells in which PIK3CA pathway genes have been disrupted will provide unique tests of specificity of such compounds. 4. We will determine the molecular structure of wild-type and mutant forms of the PIK3CA gene product, both alone and in complex with small molecule inhibitors. Such structures will be used to guide drug discovery as well as to provide insights into the mechanisms through which PIK3CA mutations affect enzymatic function.
to Public Health: These studies represent an integrated program of research that will illuminate key features of one of the most important pathways contributing to human neoplasia. It is likely to have substantial implications for the diagnosis and treatment of cancer patients.
Showing the most recent 10 out of 227 publications
