Cloning and sequencing of the cDNA for the 4.2 kb mRNA of eIF-2alpha reveal the presence of 4 consensus polyadenylation sites. In human tissues only two sites are recognized in vivo. Secondary structure analysis of the region containing the two silent sites show an area of highly predicted secondary structure emcompassing these two sites. Northern blot analysis of multiple mouse tissues showed the presence of both the 1.6 and 4.2 kb mRNAs, establishing conservation of the 4.2 mRNA 3'UTR. In these tissues the 1.6 kb mRNA was highly expressed, but the 4.2 kb mRNA was very weakly expressed. Interestingly one of the two silent polyadenylation sites was strongly expressed in mouse testis tissue, which showed the presence of 3 mRNAs for eIF-2alpha. Subsequently, human testis tissue was also found to expressed three mRNAs for eIF-2alpha. The expression of this third polyadenylation site, which is unrecognized in other tissues, may provide evidence that secondary structure plays a role in recognition of cleavage and polyadenylation. We have demonstrated that there is a difference between the stabilities in vivo for the 1.6 and 4.2 kb mRNAs of eIF-2alpha. We are in the process of developing in vitro assays to ascertain the importance of sequence and/or secondary structure of the 3'UTR to this difference. Deletion mutants have been prepared, which delete areas of predicted secondary structure, to aid in the mapping of the areas of sequence/structure responsible for the increased stability of the 4.2 kb mRNA. We have also demonstrated a significant difference in the translational efficiency of the two mRNAs. The 1.6 kb transcript is translated 3X the 4.2 kb transcript. Deletion mutations show that there is a region in the middle of the 3'UTR which when deleted will mostly restore translational activity. This region is a targeted region of predicted secondary structure. It remains to be determined if it is sequence or structure dependent mechanisms which controls the translational rate of these two transcripts.
