Abstract
We and others previously identified a region of hemizygous or homozygous deletion at chromosome band 8p23 in oral and oropharyngeal squamous cell carcinomas (OSCCs) and many other cancer types, suggesting the presence of a tumor‐suppressor gene (TSG) in this region. Recently, based on a single region of homozygous deletion in head and neck squamous cell carcinomas (HNSCC), a putative TSG, CUB and sushi multiple domains‐1 (CSMD1), has been identified. In the present study, we mapped three OSCC cell lines with previously described homozygous deletions at a high resolution onto a detailed physical map. Critically, this map covered a wider region than that used in previous studies, and in contrast to these studies, our results revealed multiple regions of homozygous deletion within a small interval on 8p23. To investigate this deletion pattern further, we generated a panel of 34 sequence tagged site (STS) markers spanning the region and tested these three cell lines and an additional 34 OSCC cell lines, identifying homozygous deletions in a further four. Combining the results from all seven deleted cell lines identified three non‐overlapping regions of homozygous deletion. This complex pattern could be consistent with the presence of multiple TSGs or one very large TSG in this region, and/or specific chromosomal instability. CSMD1 spans two of the three deleted regions and, therefore, would appear to be an excellent candidate for a TSG. However, deletion mapping with STSs corresponding to the exons of CSMD1 shows that some of the deletions do not interrupt its coding region, and in other cell lines the coding region is interrupted by two discontinuous homozygous deletions, suggesting the presence of redundant deletions. These results call into question whether the CSMD1 gene is the 8p23 TSG or whether this or any other genes at this locus are involved in the development of OSCC. © 2003 Wiley‐Liss, Inc.