While the complete etiology of epithelial-derived cancers is not yet known, several correlative genetic and environmental factors have been identified. One specific class of genetic events receiving increasing attention as both a marker and a contributing factor of oncogenesis involves microsatellite length mutations, manifested as microsatellite alteration (MA) and microsatellite instability (MSI) (Forgacs et al., 2001; Woerner et al., 2001). Microsatellite repeats are common and often polymorphic in mammalian genomes. We have developed and verified algorithms for predicting microsatellites, which are most likely to be subject to frequent mutation (Fondon et al., 1998). Somatic microsatellite length mutations are commonly observed in colon, endometrial, breast, and stomach cancers, and have been shown to be a prominent feature of some lung cancers (Girard, et al., 2001; Forgacs, et al., 2001; Wistuba et al., 2000). MSI has been shown to be a manifestation of defects in DNA mismatch repair genes, and is a useful indicator of a class of defects in DNA mismatch repair. We hypothesize that both somatic and germline microsatellite mutation may play an important etiological role in the development and progression of some cancers, and it is critical to have knowledge of their mutational frequency, complexity, and diversity among types of epithelial-derived cancers, as well as an understanding of how they vary in different normal genetic backgrounds. We have inspected the genome for gene-associated microsatellites, and have identified over 10,000 loci within transcribed regions of genes that are likely to experience expansion and/or contraction mutations. Many of these are in coding regions, with 8% of these predicted to result in frame-shifting variants when altered and some are in tumor suppressors (Wren et al., 2001).
In Aim 1 we will refine methods for predicting microsatellite mutation events associated with cancer progression, scan the recently completed human genome for predicted repeat polymorphisms in genes known or suspected to have a role in tumorigenesis, and select -150 of these loci for genotyping (Aim 2) in a panel of DNAs from lung, breast, ovarian and colorectal cancer cell lines, including many with matched B lymphocyte DNAs in addition to primary tumor and normal blood DNAs. We will also (Aim 3) investigate the effects of these mutations identified in Aim 2 on splicing by correlating them with the results of a quantitative global survey of splicing isoforms.
