Published in Cancer Detection and Prevention 2000; 24(Supplement 1).

Identification of gastric and colon cancer-associated genes using SEREX approach

A Line MSc 1, A Stengrevics MD 2, G Li PhD 3, Z Slucka 1, E Jankevics PhD 1, RC Rees PhD 3

1 Biomedical Research and Study Centre, University of Latvia, Riga, Latvia, 2 Latvian Oncology Center, Riga, Latvia and, 3 Dept Life Sciences, Nottingham Trent University, Nottingham, UK,

AIMS: Screening of tumour-derived cDNA expression libraries with cancer patient serum (SEREX) has proven to be a powerful method for identification of cancer-associated genes. In this study we applied SEREX strategy to gastric and colon carcinoma. METHODS: Two cDNA expression libraries were constructed from specimens of gastric and colon adenocarcinoma and screened with autologous patient sera. Serum-reactive clones were isolated and sequenced. DNA sequences were analyzed by homology search through GenBank/EMBL and UniGene databases. Relative mRNA expression levels in 15 tumour and matched normal tissues were examined by semi-quantitative RT-PCR and were correlated with clinico-pathologic features. RESULTS: Twenty-three distinct serum-reactive cDNA clones were isolated from gastric and colon cancer-derived libraries, 12 of them represent genes of unknown function. Three genes have a known association with human cancer – growth factor epithelin, putative oncogene TACC1 and RHAMM. Sequence analysis revealed several previously unknown splice variants of 3 genes and indicates that all of the novel genes are expressed in multiple normal tissues. Comparison of expression levels showed that epithelin, NAP1L1 and two novel genes are overexpressed but one putative transcription factor is downregulated in tumour versus normal mucosa. RT-PCR data also confirmed differential expression of the predicted splice variants. CONCLUSIONS: Twenty-three potential antigens were found in gastric and colon carcinoma. The genes were identified and searched for mutations or splice variants. Semi-quantitative RT-PCR data shows that several of them are differentially expressed in cancer versus normal tissue. Putative cancer-associated splice variant of one of the ubiquitously expressed genes was identified.

KEY WORDS: Tumour antigens, Quantitative RT-PCR, Overexpression, Alternative splicing.

For more information, contact

Paper presented at the International Symposium on Impact of Biotechnology on Cancer Diagnostic & Prognostic Indicators; Geneva, Switzerland; October 28 - 31, 2000; in the section on genetic risk assessment.