We study the regulation of amino acid and vitamin biosynthetic genes in budding yeast as a means of dissecting mechanisms of translational and transcriptional control of gene expression. Transcription of these genes is coordinately induced by transcriptional activator Gcn4 in response to amino acid limitation. Gcn4 expression is coupled to amino acid levels through a translational control mechanism involving four short upstream open reading frames (uORFs) in GCN4 mRNA, which repress Gcn4 synthesis in nutrient-replete cells. Paradoxically, GCN4 translation is derepressed in starvation or stress conditions where general protein synthesis is reduced by decreased assembly of the eIF2-GTP-Met-tRNAi ternary complex (TC), which delivers Met-tRNAi to the small (40S) ribosomal subunit for assembly of the 43S preinitiation complex (PIC). TC assembly is reduced by phosphorylation of the alpha subunit of eIF2 (eIF2a-P) by the protein kinase Gcn2, conserved in all eukaryotes, converting eIF2 from substrate to inhibitor of its guanine nucleotide exchange factor (GEF), eIF2B. Hence, GCN4 translation is a sensitive in vivo reporter of impaired TC loading on 40S subunits, a pivotal step of the initiation pathway. We previously exploited this feature to dissect the functions of eIF2B subunits in GDP-GTP exchange, and in mediating the inhibitory effect of eIFa-P on GEF activity. We also implicated particular domains/residues of factors, eIF1, -1A, and 3, and also residues of 18S rRNA located near the P decoding site of the 40S subunit, in stimulating TC loading on 40S subunits in vivo. Recently, we discovered a 10-amino acid repeat in the C-terminal tail of eIF1A that is critical for both TC recruitment and accurate initiation at AUG start codons. Mutations in these repeats, dubbed scanning-enhancers SE1 and SE2, derepress GCN4 translation and reduce TC loading on 40S subunits in vitro, and simultaneously increase initiation at UUG start codons in vivo (the Sui- phenotype). Importantly, all of these defects are suppressed by a mutation in an N-terminal segment of eIF1A, dubbed the scanning inhibitor (SI). These and other findings suggest that the eIF1A SEs (i) promote TC binding to 40S in a conformation conducive to scanning, dubbed POUT;and (ii) block an alternative conformation incompatible with scanning but required for AUG recognition, PIN, with Met-tRNAi fully accommodated in the P site. The SI element antagonizes the SEs to destabilize POUT, promote PIN, and enable selection of AUG codons. Thus, eIF1A regulates AUG selection by controlling distinct modes of Met-tRNAi binding to the 40S P site. We identified a segment in the largest (a) subunit of eIF3 that is required for attachment of the 43S PIC to mRNA, efficient scanning of long structured mRNA leaders, and efficient start codon recognition. Mutations in this segment impairing these functions also disrupt eIF3a binding to both the N-terminal segment of eIF3b and eIF3j, which bind to one another and also influence scanning and AUG recognition in vivo. Interestingly, the adjacent, extreme C-terminal region of eIF3a binds to the 40S proteins Rps2 and Rps3, which together with other findings, locates the eIF3 a-b-j module near the 40S mRNA entry channel. This location could enable the a-b-j module to regulate 43S attachment to mRNA, efficiency of scanning, and the transition between scanning-conducive and initiation-competent conformations of the PIC. Previously, we and others showed that start codon recognition depends on dissociation of eIF1 from the 40S, enabling completion of GTP hydrolysis by eIF2 in the TC and stabilizing the closed, scanning-incompatible conformation of the PIC. Thus, mutations that decrease eIF1 affinity for the 40S, dubbed type I Sui- mutations, evoke premature dissociation of eIF1 at non-AUG codons and confer the increased ratio of UUG:AUG initiation characteristic of Sui- mutants. Recently, we uncovered a novel class of Sui- mutations in eIF1 (type II) with the opposite effect of decreasing eIF1 dissociation from the PIC, but only when AUG occupies the P site. Thus, they appear to elevate the UUG:AUG ratio by eliminating the selective advantage of AUG over non-AUG codons, thus implicating eIF1 in distinguishing AUG from non-AUG triplets in the P site. We also obtained evidence that eIF5, the GTPase activating protein for eIF2, stimulates eIF1 dissociation to promote start codon recognition. As the eIF5 N-terminal domain and eIF1 share structural similarity, eIF5 might compete with eIF1 for a common binding site through molecular mimicry. This finding implies a second function for eIF5, beyond GAP activity, in start codon recognition. eIF2B is the guanine nucleotide exchange factor (GEF) for eIF2, which stimulates formation of the TC in a manner inhibited by phosphorylated eIF2 (eIF2(aP)). While eIF2B contains five subunits, the e/Gcd6 subunit is sufficient for GEF activity in vitro. The d/Gcd2 and b/Gcd7 subunits function with a/Gcn3 in the eIF2B regulatory subcomplex that mediates tight, inhibitory binding of eIF2(aP)-GDP, but the essential functions of Gcd2 and Gcd7 are not well understood. We showed recently that the depletion of wild-type Gcd7 from cells, three lethal Gcd7 amino acid substitutions, and a synthetically lethal combination of substitutions in Gcd7 and eIF2a all impair eIF2 binding to eIF2B without reducing Gcd6 abundance in the native eIF2BeIF2 holocomplex. Additionally, we found that non-lethal Gcd7 mutations that impair eIF2B function display extensive allele-specific interactions with mutations in the S1 domain of eIF2a(harboring the phosphorylation site), which binds to eIF2B directly. Consistent with this, we observed that Gcd7 can overcome the toxicity of eIF2(aP) and rescue native eIF2B function when overexpressed in cells together with Gcd2 or eIF2B subunit g/Gcd1. In aggregate, these findings provide compelling evidence that b/Gcd7 is crucial for binding unphosphorylated substrate by eIF2B in vivo beyond its role in mediating inhibition of eIF2B by eIF2(aP). Snf1 is the ortholog of mammalian AMP-activated kinase, and is responsible for activation of glucose-repressed genes at low glucose levels in budding yeast. We recently demonstrated that Snf1 promotes formation of phosphorylated eIF2a, by stimulating the kinase activity of Gcn2 during histidine starvation of glucose-grown cells. Thus, eliminating Snf1 or mutating its activation loop lowers Gcn2 kinase activity, reducing autophosphorylation of Thr-882 in the Gcn2 activation loop, and decreases eIF2(aP) levels in starved cells. Consistent with these findings, eliminating Reg1, a negative regulator of Snf1 activity, provokes Snf1-dependent hyperphosphorylation of both Thr-882 and eIF2a. Interestingly, Snf1 also promotes eIF2phosphorylation in the non-preferred carbon source galactose, but this occurs by inhibition of protein phosphatase (PP) 1a, known as Glc7, and the PP2A-like enzyme Sit4, rather than by activation of Gcn2 function. Both Glc7 and Sit4 were found to physically interact with eIF2 in cell extracts, supporting their direct roles as eIF2a phosphatases. Our results show that Snf1 modulates the level of eIF2a phosphorylation by different mechanisms depending on the kind of nutrient deprivation that exists in cells. When combined with our previously published findings, it can now be stated that two major nutrient-signaling kinases that respond to the availability of different nutrients, Snf1 (carbon) and Tor (nitrogen), make critical contributions to setting the level of eIF2(aP), a key regulator of translation initiation in nutrient-limited cells.
Showing the most recent 10 out of 64 publications
