VARIOUS TRANSCRIPTS ARE GENERATED FROM THE VCSA1 GENE BY ALTERNATIVE SPLICING AND POLY(A) PROCESSING IN THE RAT SUBMANDIBULAR-GLAND

The members of the VCS (variable coding sequence) multigene family display extensive evolutionary divergence in the protein-coding region. The first described gene (VCSA1) was found to encode a major 0.7-kb mRNA (VCSA1*1T1) coding for a prohormone-like preproprotein, SMR1-VA1, in the submandibular gland (SMG) of Rattus norvegicus. We report here the cloning of four other VCSA1 cDNAs corresponding to mRNAs (VCSA1*1T2 to *1T5) expressed in the SMG, VCSA1*1T1 to *1T4 mRNAs share the three exons previously described and differ in their 3' untranslated regions (UTR). Their differences originate from the alternative utilization of four polyadenylation sites. Comparison of the tissue levels of VCSA1*1T1 and VCSA1*1T4 during post-natal development of the male rat SMG suggests that the poly(A) addition sites are both used at each stage. The fifth RNA transcript (VCSA1*1T5) contains only the first two exons. The nucleotide sequence of the cDNA reveals that VCSA1 has an additional exon (exon 4) which is spliced to exon 2 in VCSA1*1T5. In addition to VCSA1*1T1, at least VCSA1*1T4 and VCSA1*1T5 are actively translated in vivo, as revealed by their association to the polysomal fractions. The protein, P2-VA1, coded by VCSA1*1T5 is 68 amino acids in length and it is likely to be a glycosylated secretory protein. The putative mature P2-VA1 protein completely differs from the SMR1-VA1 pro-protein and very likely has a different function. VCSA1*1T1 is accumulated in the male rat SMG 200-1000-fold more than the other transcripts, Run-on experiments reveal that almost all transcription proceeds several hundred bp downstream from the poly(A) site corresponding to VCSA1*1T1. This suggests that the high levels of VCSA1*1T1 transcript are mainly due to post-transcriptional events.