Molecular Mechanisms of Translational Control

Eunice Kennedy Shriver National Institute Of Child Health & Human Development

Investigators

Linked publications, trials & patents

Abstract

We study the molecular mechanisms involved in assembly, function, and regulation of translation initiation complexes involved in protein synthesis, using yeast as a model system to exploit its powerful combination of genetics and biochemistry for dissecting complex cellular processes in vivo. The translation initiation pathway produces an 80S ribosome bound to mRNA with methionyl initiator tRNA (tRNAi) base-paired to the AUG start codon. The tRNAi is recruited to the 40S subunit in a ternary complex (TC) with GTP-bound eIF2 to produce the 43S preinitiation complex (PIC) in a reaction stimulated by eIFs 1, 1A, 3 and 5. The 43S PIC attaches to the 5' end of mRNA, facilitated by cap-binding complex eIF4F (comprised of eIF4E, eIF4G, and RNA helicase eIF4A) and PABP bound to the poly(A) tail, and scans the 5 untranslated region (UTR) for an AUG start codon in preferred sequence context. Scanning is promoted by eIFs 1 and 1A, which induce an open conformation of the 40S and binding of TC in a conformation suitable for scanning successive triplets entering the ribosomal P site (P-out), and by eIF4F and other RNA helicases, such as Ded1 and Dbp1, that remove secondary structure in the 5' UTR. AUG recognition leads to tighter binding of TC in the P-in state and evokes irreversible hydrolysis of the GTP bound to eIF2, dependent on the GTPase activating protein (GAP) eIF5, releasing eIF2-GDP from the PIC to leave tRNAi in the P site and allow joining of the 60S subunit to form the 80S initiation complex. The function of eIF2 is down-regulated by phosphorylation of its -subunit by protein kinase Gcn2 in response to amino acid starvation and other stresses that likely evoke stalling of ribosomes engaged in translation elongation. eIF2 interactions with mRNA control accurate start codon selection by the translation preinitiation complex Comparison of cryo-EM structures of 48S PICs in open, scanning-conducive or closed, arrested conformations revealed interactions unique to the closed complex between Arg residues R55 and R57 of domain 1 of the -subunit of eIF2 (eIF2-D1) with mRNA nucleotides 5 of the AUG codon, including the -3 residue of the Kozak context. We showed that substitutions of R55 and R57 reduce recognition of the poor-context AUG codon for SUI1 mRNA (encoding eIF1) and also UUG start codons in Sui- cells (the Ssu- phenotype). We further showed that the R55G-R57E Ssu- substitutions destabilize TC binding to 48S PICs reconstituted with mRNA with a UUG start codon in the in vitro reconstituted system, in the manner expected from specific destabilization of the closed complex at a near-cognate codon. Interestingly, residue R53 of eIF2-D1 interacts with rRNA residues exclusively in the open complex; and the R53E substitution was found to enhance initiation at UUG codons (the Sui- phenotype) and the poor-context SUI1 AUG, and also to confer the Gcd- phenotype, indicating slow recruitment of the TC to scanning 40S subunits engaged in reinitiation on GCN4 mRNA, in vivo. In the reconstituted system, R53E stabilized TC binding to UUG complexes while simultaneously reducing the on-rate of TC loading, all in the manner predicted for specific destabilization of the open complex and shift towards the closed state. We conclude that distinct interactions of eIF2-D1 with the rRNA or mRNA stabilize first the open, and then the closed, conformation of the PIC to regulate the accuracy and efficiency of start codon selection in vivo (Thakur et al., 2020). eIF4A and eIF4E interactions with distinct residues of the Ded1 N-terminus stimulate Ded1 function in translation initiation in vivo. Binding of eIF4F to the mRNA cap enhances recruitment of the 43S PIC to the 5' end and subsequent scanning of the 5UTR. Ded1 interacts with eIF4A and the eIF4G subunit of eIF4F; and eIF4A and eIF4G stimulate unwinding of a model RNA substrate by Ded1 in vitro. Previously, we showed that the Ded1 C-terminal domain (CTD) and its two interacting domains in eIF4G, RNA2 and RNA3, and the Ded1 N-terminal domain (NTD) that interacts with eIF4A, enhance Ded1 stimulation of 48S PIC assembly in the reconstituted system. Ded1 also interacts with eIF4E; but the binding sites for eIF4A and eIF4E in the Ded1-NTD were unknown. Substituting conserved residues 21-27 and 51-57 in the Ded1-NTD reduced Ded1 binding to eIF4A in vitro, impaired association between Ded1 and eIF4A in cell extracts, and reduced growth, bulk translation initiation, and translation of Ded1-hyperdependent reporter mRNAs with stem-loop (SL) insertions. Overexpressing eIF4A diminished the growth defects for each single substitution, but not for the 21-27/51-57 double substitution that is null for eIF4A binding, supporting the importance of Ded1-NTD/eIF4A interaction in cells. Substituting the residues 59-65 and 83-89 reduced Ded1-NTD binding to eIF4E in vitro and Ded1-eIF4E association in extracts, and reduced translation of Ded1-hyperdependent reporters. Combining all four NTD substitutions conferred a growth defect indistinguishable from NTD deletion, suggesting that eIF4A/eIF4E binding is the key function of the Ded1 NTD. Deleting the Ded1-CTD impairs growth only when combined with NTD substitutions, implying that the Ded1-CTD/eIF4G interaction is dispensable when Ded1 can interact with eIF4A/eIF4E. In the reconstituted system, Ded1 NTD substitutions that eliminate eIF4A binding reduce the rate of 48S PIC assembly on a Ded1-dependent mRNA harboring a 5' UTR SL, and increases the amount of Ded1 required for the half-maximal rate (K1/2). Disruption of Ded1-NTD/eIF4E interaction also increases the Ded1 K1/2 for 48S assembly. Thus, Ded1 NTD interactions with eIF4A and eIF4E stabilize a Ded1-eIF4E-eIF4G-eIF4A quaternary complex that enhances Ded1s ability to stimulate 48S PIC assembly on structured 5UTRs(Gulay et al.,2020). Reprogramming of mRNA translation by impaired ribosome recycling at stop codons favors efficiently translated mRNAs in yeast Formation of the 43S PIC is a rate-determining step of initiation. Ribosome recycling after termination produces free 40S subunits for reassembly of 43S PICs. Yeast mutants lacking orthologs of mammalian eIF2D (Tma64), and either MCT-1 (Tma20) or DENR (Tma22), are broadly impaired for 40S recycling; but it was unknown whether this defect alters the translational efficiencies (TEs) of mRNAs. It was also possible that Tma64/eIF2D can substitute for eIF2 in recruiting tRNAiMet during initiation. Consistent with impaired initiation, the tma64tma20 mutant exhibits reduced assembly of bulk polysomes. Ribosome profiling of this mutant reveals a marked reprogramming of translational efficiencies, wherein translation of the most efficiently translated (strong) mRNAs tends to be elevated, whereas translation of weak mRNAs generally declines. Profiling of the tma64 single mutant reveals none of the hallmarks of impaired 40S recycling nor changes in translation efficiencies, suggesting that the defects found in tma cells are associated with defective ribosome recycling versus loss of eIF2D function in Met-tRNAi recruitment. Remarkably similar translational re-programming was seen on reducing 43S PIC assembly by inducing phosphorylation of eIF2 or by decreasing total 40S subunit levels by depleting Rps26, without affecting ribosome recycling. Moreover, the tma mutation specifically impaired translation of mRNAs with cap-proximal SL structures that are expected to impede PIC attachment. Our findings suggest that strong mRNAs out-compete weak mRNAs in response to 43S PIC limitation achieved in various ways at the step of 43S PIC recruitment, in accordance with mathematical modeling of how translational efficiencies of different groups of mRNAs are altered by reduced ribosome abundance (Gaikwad et al., 2021).