Gene expression profile in rat pancreatic islet and RINm5F cells.

To clarify tissue-specificity of pancreatic beta cells, comparison of mRNA expression in various conditions of the tissue of multiple organisms is important. Although the developed methodologies for mRNA monitoring such as microarray, rely on the growth of dbEST (database of expressed sequence tag), a large number of unknown genes in the genome, especially in the rat, have not been shown to be expressed. In this study, we have established the first database of ESTs from rat pancreatic islet and RINm5F cells. Two cDNA libraries were constructed using mRNAs from rat pancreatic islet and RINm5F cells to cover a wider spectrum of expressed genes. Over 40,000 clones were randomly selected from the two libraries and partially sequenced. The sequences obtained were subjected to BLAST database analyses. This large-scale sequencing generated 40,710 3'-ESTs. Clustering analysis and homology search of nucleotide and peptide databases using both 3'- and 5'-ESTs revealed 10,406 non-redundant transcripts representing 4078 known genes or homologs and 6328 unknown genes. To confirm actual expression, the unknown sequences were further subjected to dbEST search, resulting in the identification of 5432 significant matches to those from other sources. Interestingly, of the remaining sequences showing no match, 779 were found to be encoded by exon-intron organization in the corresponding genomic sequences, suggesting that these are newly found as actually expressed in this study. Since many genes are up- or down-regulated in differing conditions, applications of the expression profile should facilitate identification of the genes involved in cell-specific functions in normal and disease states.