The long-term goal of this project is to determine the specificity of repair of mismatched bases in vitro and in vivo and to elucidate the mechanism(s) by which base mismatches are generated. To this end we have developed a forward mutation system, and isolated and characterized mutations arising in wild type and mismatch repair defective strains. In the mismatch repair defective strains, the majority of the mutations are AT to GC transitions. Surprisingly, these transition mutations are clustered in three hotspots near the only two 5'-GATC-3' sequences in our target gene.
Specific aim 1 of the current proposal seeks to determine the effect on hotspot formation when one or the other 5'-GATC-3' site is deleted. The effect on mutagenesis of creating a new 5'-GATC-3' site will also be assessed.
Specific aim 2 seeks to determine the mechanism by which mutations are generated in the mismatch repair deficient strains. We will seek second-site suppressors which either increase or decrease spontaneous mutation in a mismatch repair defective strain. We will also test the effect of alleles of genes known to be involved in DNA replication and/or repair on mutation frequency. These studies will allow us to identify the process by which hotspots are generated.
Specific aim 3 seeks to elucidate two related aspects of mismatch repair. Is the efficiency of mismatch repair correlated with distance of the mismatch from the nearest 5'- GATC-3'? Since we have characterized many different kinds of mutations in our system, we can answer this question using a variety of transition, transversion and frameshift mutations. In fact, only in our system can this question be answered. The second part of Specific aim 3 seeks to resolve the question of what number of mismatched bases the repair system can recognize. We have a defined set of deletion mutations which can resolve this issue. The fourth Specific aim will allow determination of mutation spectrum in a strain overproducing the dam methylase. In theory, the spectrum should be identical to that in mismatch repair defective strains. We will also use a strain deleted for the dam and mutH genes to rule in or out the possibility that the spectrum we observe is due to binding of the products of these genes to 5'-GATC-3' sites. The unique mutation system we have developed will give the greatest chance of success to answer these questions.
