RNA-based mutation screening in hereditary nonpolyposis colorectal cancer

Abstract

Hereditary nonpolyposis colorectal cancer (HNPCC) is a cancer syndrome inherited in an autosomal dominant fashion. Four susceptibility genes are known, which code for DNA mismatch repair enzymes. The purpose of this study was to identify the HNPCC gene defects in a cohort of Australian HNPCC families and to evaluate the use of RNA-based screening methods. Six mutations were identified, four in the hMLH1 gene and two in hMSH2, by using a combination of DNA-based and RNA-based methods. One of the hMLH1 defects was a missense mutation, and the other five mutations would be expected to result in a shortened protein. These included a rare type of mRNA splicing mutation in hMLH1 exon 17. By use of reverse-transcriptase (RT) PCR, defective transcripts were detectable for three of the hMLH1 mutations but not for the fourth one, which was predicted to cause skipping of exon 15. Furthermore, many more alternative transcripts for the hMLH1 gene were found than previously described, and these were more abundant in the RNA samples prepared from whole blood than from lymphoblastoid cell lines. This confounded RNA-based screening for HNPCC mutations, because it was difficult to determine which aberrant RT-PCR fragment was the real hereditary defect. One of the splice-site mutations reported here causes skipping of exons 9 and 10, which also occurs as an alternative transcript. When the protein-truncation test was used, the results were indistinguishable between the patients in this family and controls. Other aberrant transcripts were also observed that varied in size between individuals but were unrelated to the hereditary defects. This study has important implications for the design of reliable diagnostic tests for HNPCC gene defects.