GLUCOSE-DEPENDENT TURNOVER OF THE MESSENGER-RNAS ENCODING SUCCINATE-DEHYDROGENASE PEPTIDES IN SACCHAROMYCES-CEREVISIAE - SEQUENCE ELEMENTS IN THE 5'-UNTRANSLATED REGION OF THE IP MESSENGER-RNA PLAY A DOMINANT ROLE

We have demonstrated previously that glucose repression of mitochondri al biogenesis in Saccharomyces cerevisiae involves the control of the turnover of mRNAs for the iron protein (Ip) and flavoprotein (Fp) subu nits of succinate dehydrogenase (SDH). Their half-lives are > 60 min i n the presence of a nonfermentable carbon source (YPG medium) and < 5 min in glucose (YPD medium). This is a rare example in yeast in which the half-life of an mRNA can be controlled by manipulating external co nditions. In our current studies, a series of Ip transcripts with inte rnal deletions as well as chimeric transcripts with heterologous seque nces (internally or at the ends) have been examined, and we establishe d that the 5'-untranslated region (5' UTR) of the Ip mRNA contains a m ajor determinant controlling its differential turnover in YPG and YPD. Furthermore, the 5' exonuclease encoded by the XRN1 gene is required for the rapid degradation of the Ip and Fp mRNAs upon the addition of glucose. In the presence of cycloheximide the nucleolytic degradation of the Ip mRNA can be slowed down by stalled ribosomes to allow the id entification of intermediates. Such intermediates have lost their 5' e nds but still retain their 3' UTRs. If protein synthesis is inhibited at an early initiation step by the use of a pl tl mutation (affecting the initiation factor eIF3), the Ip and Fp mRNAs are very rapidly degr aded even in YPG. Significantly, the arrest of translation by the intr oduction of a stable hairpin loop just upstream of the initiation codo n does not alter the differential stability of the transcript in YPG a nd YPD. These observations suggest that a signaling pathway exists in which the external carbon source can control the turnover of mRNAs of specific mitochondrial proteins. Factors must be present that control either the activity or more likely the access of a nuclease to the sel ect mRNAs. As a result, we propose that a competition between initiati on of translation and nuclease action at the 5' end of the transcript determines the half-life of the Ip mRNA.