Recent studies have demonstrated that inherited mutations in the human homologues of the Saccharomyces cerevisiae mismatch repair genes MSH2 and MLH1 are responsible for approximately 90% of hereditary nonpolyposis colon cancer (HNPCC). These results indicate that inherited defects in mismatch repair can cause cancer and analysis of sporadic tumors has provided evidence that a significant proportion of sporadic tumors may also be defective for mismatch repair. These results have defined a new, major pathway for cancer susceptibility. The goal of this proposal is to develop model systems for the analysis of mismatch repair and its relationship to cancer predisposition. To achieve these goals, the following lines of investigation will be performed. (1) Mice which are heterozygous for msh2 mutations will be constructed to develop an animal model system for HNPCC. The mouse MSH2 gene has been cloned, an exon encoding a conserved region of the MSH2 protein has been sequenced and this information has been used to initiate construction of mice containing mutations that disrupt this region of the gene. The complete sequence of the mouse MSH2 cDNA and the intron/exon junctions of the mouse genomic locus will be determined. This will facilitate the construction of additional mouse mutations and the analysis of msh2 mutations that inactivate the second copy of the MSH2 gene in tumors. The mouse system will also be used to study the types of other mutations that accumulate during tumor progression. (2) The human MSH2 and MLH1 genes are highly homologous to the corresponding S. cerevisiae genes which will allow the S. cerevisiae genes to be used to study the mutations found in humans. Mutations found in human tumors and HNPCC kindreds will be made in the corresponding S. cerevisiae genes by site specific mutagenesis. The resulting mutants will be characterized to determine the genetic properties of the different mutations found in humans. The properties of these mutations will be correlated with the properties of the HNPCC kindreds and sporadic tumors in which the mutations were found. (3) S. cerevisiae strains in which the normal S. cerevisiae mismatch repair proteins have been functionally substituted for by human proteins or chimeric S. cerevisiae/human proteins will be developed. This system will be used to study the effect of mutations found in humans on mismatch repair. (4) Genetic and biochemical methods will be used to identify genes encoding proteins that interact with the human MSH2 and MLH1 proteins. These genes will be analyzed to determine if they encode mismatch repair proteins and if HNPCC kindreds and tumors have mutations in these genes. The results of these studies should allow us to elucidate the repair pathways that reduce the endogenous mutation rates in cells, understand better how defects in these repair pathways lead to the development of cancer and define additional cancer susceptibility genes.
Showing the most recent 10 out of 143 publications
