Genome-wide analysis of translational efficiency reveals distinct but overlapping functions of yeast DEAD-box RNA helicases Ded1 and eIF4A

Abstract

DEAD-box RNA helicases eIF4A and Ded1 are believed to promote translation initiation by resolving mRNA secondary structures that impede ribosome attachment at the mRNA 5 ′ end or subsequent scanning of the 5 ′ UTR, but whether they perform unique or overlapping functions in vivo is poorly understood. We compared the effects of mutations in Ded1 or eIF4A on global translational efficiencies (TEs) in budding yeast Saccharomyces cerevisiae by ribosome footprint profiling. Despite similar reductions in bulk translation, inactivation of a cold-sensitive Ded1 mutant substantially reduced the TEs of >600 mRNAs, whereas inactivation of a temperature-sensitive eIF4A variant encoded by tif1-A79V (in a strain lacking the ortholog TIF2) yielded <40 similarly impaired mRNAs. The broader requirement for Ded1 did not reflect more pervasive secondary structures at low temperature, as inactivation of temperature-sensitive and cold-sensitive ded1 mutants gave highly correlated results. Interestingly, Ded1-dependent mRNAs exhibit greater than average 5 ′ UTR length and propensity for secondary structure, implicating Ded1 in scanning through structured 5 ′ UTRs. Reporter assays confirmed that cap-distal stem-loop insertions increase dependence on Ded1 but not eIF4A for efficient translation. While only a small fraction of mRNAs shows a heightened requirement for eIF4A, dependence on eIF4A is correlated with requirements for Ded1 and 5′ UTR features characteristic of Ded1-dependent mRNAs. Our findings suggest that Ded1 is critically required to promote scanning through secondary structures within 5 ′ UTRs, and while eIF4A cooperates with Ded1 in this function, it also promotes a step of initiation common to virtually all yeast mRNAs.

Mechanism and Regulation of Protein Synthesis in Saccharomyces cerevisiae

“…Secondary structure near the 59 cap of the mRNA could prevent ribosome association with the mRNA, whereas secondary structure further down the 59 UTR could prevent ribosome scanning. Several studies in yeast have demonstrated that insertion of stem-loop structures in the 59 UTR interferes with translation (Baim and Sherman 1988;Cigan et al 1988b;Abastado et al 1991;Vega Laso et al 1993;Berthelot et al 2004;Sen et al 2015). As expected, more stable stem-loop structures are more deleterious than weaker stem loops; however, the impact of cap-proximal vs. more distal secondary structure varies in the different published reports.…”

“…The mRNAs most affected were not ones with longer 59 UTRs, suggesting that eIF4G's role in 43S PIC recruitment to mRNAs is more critical than its role in promoting scanning on long or structured 59 UTRs (Park et al 2011). Second, ribosome profiling of an eIF4A ts mutant found that the translational efficiency of most mRNAs were similarly affected by loss of eIF4A (Sen et al 2015). As reporter mRNAs with 59 UTRs of differing lengths and secondary structures were all affected by 30-50% in the eIF4A mutant, it was concluded that eIF4A is globally important for optimal initiation on all mRNAs (Sen et al 2015).…”

“…Second, ribosome profiling of an eIF4A ts mutant found that the translational efficiency of most mRNAs were similarly affected by loss of eIF4A (Sen et al 2015). As reporter mRNAs with 59 UTRs of differing lengths and secondary structures were all affected by 30-50% in the eIF4A mutant, it was concluded that eIF4A is globally important for optimal initiation on all mRNAs (Sen et al 2015).…”

“…Inactivation of a ded1 cold-sensitive mutant impaired global translation rates, and greater than average reductions in translational efficiency were observed for only $600 genes (Sen et al 2015). The Ded1 hyperdependent mRNAs had longer than average 59 UTR lengths and greater propensity for secondary structure, thereby implicating Ded1 as critical for scanning through structured 59 UTRs (Sen et al 2015).…”

Understanding biases in ribosome profiling experiments reveals signatures of translation dynamics in yeast

“…The exact mechanism of this inhibition, however, is not completely understood, with recent studies suggesting that CHX binds to a ribosome's E-site along with a deacylated tRNA to block further translocation [19,20]. The majority of the rapidly growing body of ribosome profiling experiments in yeast have followed this original CHX-pretreatment protocol [13,[21][22][23][24][25][26][27][28][29][30]. Several groups have applied a variety of conceptually similar computational methods to the data produced by these experiments to infer the average speed with which each codon identity is translated.…”