Mouse models for colorectal cancer (CRC) have been generated with germline mutations based on familial syndromes. It is unclear if these would be good representations of the most frequent forms of CRC, sporadic cases. Though APC mutations have been found in most tumors, there is a significant number in which this gene is unaltered. We must more clearly understand the pathways and cellular changes involved in carcinogenesis in order to develop novel clinical modalities. As such, we hypothesize that APC-independent pathways must be involved in CRC development. We propose to develop a new model for sporadic CRC using adenoviral enemas to deliver Cre recombinase to a mouse with a conditional mutation in the cancer predisposition gene Mlh1. We will use a novel approach to simultaneously inactivate Mlh1 and express GFP only in those cells exposed to Cre. Thus, we can document the natural history of the cellular events that lead to adenoma formation by following the fate of individual foci of precancerous lesions. We will characterize the genomic and gene expression changes in cells at different stages of cancer progression by array comparative genomic hybridization (aCGH) and Genechip analyses, looking for APC-independent pathways.
