Tif5p (in yeast)
The translation initiation process is mediated by a series of partial reactions each of which requires the participation of a large number of protein factors collectively called eukaryotic (translation) initiation factors (eIFs). The process leads to the assembly of an 80S ribosome-bound initiator Met-tRNAi at the start AUG codon of an mRNA that is active in peptidyl transfer. According to the currently accepted view of translation initiation, the first step is the binding of the initiator Met-tRNAi to the heterotrimeric GTP-binding initiation factor eIF2 to form the Met-tRNAi•eIF2•GTP ternary complex. The ternary complex then binds to the 40S ribosomal subunit containing bound initiation factor eIF3, a process that requires two other initiation factors eIF1 and eIF1A. This interaction leads to the formation of a 43S preinitiation complex (40S•eIF3•eIF1•eIF1A•Met-tRNAi•eIF2•GTP) which is then recruited to the 5′-capped end of the mRNA via protein-protein interaction between the cap-bound initiation factor eIF4F and eIF3 bound to the 43S preinitiation complex. The resulting 43S complex then binds two RNA-helicases eIF4A and eIF4B and scans the mRNA in a 5′→ 3′ direction until it is positioned at the initiating codon (usually an AUG codon) by base-pairing interaction between the AUG codon and the anticodon of Met-tRNAi to form the 48S initiation complex (mRNA•40S•eIF3•eIF1•eIF1A•Met-tRNAieIF2•GTP). Concomitant with the formation of the 48S initiation complex, eIF1 is released from its binding site in the 48S complex. Subsequently, another initiation factor eIF5 interacts with eIF2 in the 48S initiation complex, and acting as a GTPase activating protein (GAP), activates the intrinsic GTPase activity of eIF2 to promote the hydrolysis of bound GTP. Hydrolysis of GTP leads to the immediate release of Pi and eventually of inactive eIF2•GDP (and other bound factors) during eIF5B-mediated joining of the 60 ribosomal subunit to the 48S ribosomal complex to form an 80S initiation complex (80S•Met-tRNAi•mRNA) that is competent to form the first peptide bond during the elongation phase of protein synthesis. (For a review, see Pestova et al. 2007; Hinnebusch et al. 2007; Sonenberg and Hinnebusch 2009; Lorsch and Dever 2010).
Isolation and Initial Characterization of eIF5
Initiation factor eIF5 was one among the seven distinct eIFs that were originally isolated in many laboratories from a variety of eukaryotic sources including rabbit reticulocyte lysates on the basis of its ability to stimulate in vitro translation of globin mRNA in a partially purified reconstituted system. The purified protein was shown to be a monomeric protein of apparent Mr = 150,000–170,000 (For a review, see Maitra et al. 1982 and references therein). Functional characterization of eIF5 in the translation initiation assay showed that eIF5 did not play a role in the in vitro formation of the 48S initiation complex. However, the protein was essential for the subsequent joining of the 60S ribosomal subunit to the 48S complex to form a functional 80S initiation complex. It was further shown that prior to or concomitant with the subunit joining reaction, eIF5 promoted the hydrolysis of GTP bound to eIF2 in the 48S initiation complex. Hydrolysis of GTP was shown to be a prerequisite for the joining of the 60S ribosomal subunit to the 48S complex to form the active 80S initiation complex (reviewed by Maitra et al. 1982). However, later, using a more direct and a specific assay that measured the ability of eIF5 to mediate the joining of the 60S ribosomal subunit to a preformed 48S initiation complex formed with an AUG trinucleotide in lieu of an entire mRNA, a new activity of about 50 kDa was isolated from calf liver extracts and rabbit reticulocyte lysates that migrated on SDS-polyacrylamide-gels with an apparent Mr of 58,000 (reviewed by Das and Maitra 2001). Subsequently, a similar activity was purified from the yeast Saccharomyees cerevisiae as a protein of about 45 kDa that migrated in SDS-polyacrylamide-gels as a protein of about 54,000–56,000 (Das and Maitra 2001).
This anomaly in molecular size of eIF5 was resolved when later work (Pestova et al. 2000) showed that the higher molecular weight polypeptide detected in earlier eIF5 preparations reported by others is another distinct initiation factor, designated eIF5B, that is required along with eIF5 for the joining of the 60S ribosomal subunit to the 48S initiation complex formed with a natural mRNA as the template. However, eIF5B has no role to play in the hydrolysis of GTP bound to eIF2 in the 48S initiation complex formed either with AUG or a natural mRNA as a template.
Biochemical characterization of the 50 kDa protein, designated as eIF5, showed that eIF5 promotes the hydrolysis of GTP only when the nucleotide is bound as a Met-tRNAi•eIF2•GTP ternary complex on the 40S ribosomal subunit. eIF5 fails to promote hydrolysis of either free GTP or GTP bound to eIF2 as a ternary complex. The reaction is unaffected by the presence of 60S ribosomal subunits, indicating that GTP hydrolysis reaction precedes the joining of the 60S ribosomal subunit to the 40S complex. Finally, following eIF5-promoted GTP hydrolysis, even though Pi is released almost instantaneously from the 40S ribosomes, eIF2•GDP complex formed remains initially bound to the ribosomal complex and can be released by prolonged incubation at 37 °C or by sucrose gradient centrifugation (Raychaudhuri and Maitra 1986).
Molecular Genetic Characterization eIF5
To show that the 50 kDa protein functions as a canonical translation initiation factor essential for translation of mRNAs in vivo, both the mammalian cDNA and the single copy essential yeast gene TIF5 encoding eIF5 were cloned and expressed in Escherichia coli (For a review see, Das and Maitra 2001). In each case, the purified recombinant mammalian protein (calculated molecular mass 48,926) or the yeast protein (calculated molecular mass 45,436) mimics eIF5 isolated from natural sources in molecular size, in specific activity, and in its ability to promote the hydrolysis of GTP bound to eIF2 in the 48S initiation complex. Additionally, to understand the function of eIF5 in vivo in yeast cells, a conditional mutant yeast strain in which a functional but a rapidly degradable form of eIF5 fusion protein was synthesized from the glucose repressible GAL promoter was constructed. Depletion of eIF5 from this mutant yeast strain resulted in inhibition of cell growth and the rate of in vivo protein synthesis. Analysis of the polysome profiles of eIF5-depleted cells showed greatly diminished polysomes with simultaneous increase in the pool of free 80S ribosomes and free 60S and 40S ribosomal subunits. Such a polysome/ribosome profile is characteristic of cells lacking an essential translation initiation factor. Furthermore, lysates of cells depleted of eIF5 were inactive in translation of both total yeast poly(A)+ and luciferase mRNAs in vitro. Addition of 45-kDa purified yeast eIF5 restored translation in these cell lysates. Similar observation was also made when a thermosensitive mutant yeast strain was used. These results show that the TIF5 gene product, a protein of 45,346 Da, is indeed an initiation factor required for initiation of protein synthesis. Additionally, mammalian eIF5 can functionally substitute for yeast eIF5 in maintaining yeast cell viability and growth, indicating that the mechanism of eIF5-promoted GTP hydrolysis is conserved from yeast to mammals.
eIF5 Is Not a GTPase Protein, but Rather Functions as a GTPase Activating Protein (GAP)
GTP-binding domains, G1-G4, are characteristic of members of the GTPase superfamily. An important feature of these proteins is that they not only bind GTP but also possesses an intrinsic GTPase activity, which is often stimulated by their interaction with an effector molecule that acts as a GTPase activating protein (GAP). Both mammalian and yeast eIF5 contain imperfect G1-G4 domains. The presence of imperfections in the G1-G4 domains of eIF5 explains why eIF5, by itself, does not bind or hydrolyze free GTP (Das and Maitra 2001). In contrast, the heterotrimeric eIF2 binds GTP and its γ subunit contains the conserved GTP-binding domains (G1-G4) (Hinnebusch et al. 2007; Lorsch and Dever 2010). Since interaction of eIF5 with Met-tRNA•eIF2•GTP bound in the 48S initiation complex is necessary for hydrolysis of GTP, the question arises whether eIF5 functions as a specific GAP to eIF2-GTPase (reviewed by Das and Maitra 2001; Hinnebusch et al. 2007; Pestova et al. 2007).
Binding studies show that eIF5 forms a complex with eIF2, the GTP-binding protein, by specifically interacting with the eIF2β subunit (Das et al. 1997; Asano et al. 1999). It was further demonstrated that the N-terminal lysine-rich region of eIF2β (designated K-boxes 1, 2, and 3) is involved in binding the C-terminal region of eIF5 containing a glutamic acid-rich bipartite motif (amino acids 345-347 and 384-386) (Asano et al.; 1999; Das and Maitra 2000). Mutational analysis showed that yeast eIF5 mutants E346A, E347A and E384A, E385A exhibited a marked decrease in binding to eIF2, and were also severely defective in eIF5-mediated hydrolysis of GTP bound to the 48S initiation complex (Das and Maitra 2000). These mutant eIF5 proteins were also defective in stimulating translation in eIF5-depleted yeast cell-free translation extracts, indicating that the eIF5•eIF2 interaction is essential for eIF5 function in vitro and in vivo.
eIF5 also binds to the γ subunit of eIF2. In fact, it has been demonstrated the amino terminal domain (NTD) of yeast eIF5 (residues 1–279) is as active as full-length eIF5 in activating the intrinsic GTPase activity of eIF2 (Hinnebusch et al. 2007). This observation led to the hypothesis that interaction between eIF2β and eIF5 does not activate the intrinsic GTPase eIF2γ, rather it may merely serve to recruit eIF5 to eIF2 in the 48S ribosomal complex. However, mutational studies in the γ and β subunits of yeast eIF5 are also consistent with an alternative hypothesis that eIF5-eIF2β interaction derepresses the GTPase activity of eIF2γ (reviewed by Pestova et al. 2007).
Further deletion analysis studies showed that an invariant arginine residue at position 15 at the N-terminal end of eIF5 is also essential for its GTP hydrolysis-promoting activity (Das et al. 2001; Paulin et al. 2001. This property of eIF5 is reminiscent of typical GAPs like RasGAPs and RhoGAPs that contain “arginine finger” motifs consisting of an invariant arginine residue at the N-terminus of their catalytic domains that is necessary for their GTPase-stimulating activity, in addition to motifs required for interacting with their respective G proteins. In typical GAPs, a secondary arginine residue stabilizes the finger loop carrying the primary arginine residue (reviewed by Das and Maitra 2001). In analogy with RasGAP and RhoGAP, it appears that either Lys33 or Lys55, or both in rat eIF5 constitute the “secondary” element required for eIF5 GAP function. (Das et al. 2001; Paulin et al. 2001).
However, it should be emphasized that while interactions between eIF5 and eIF2β and eIF5 and eIF2γ are necessary for activation of GTPase activity of eIF2γ, these interactions alone are not sufficient since binding of eIF2 as a Met-tRNAi•eIF2•GTP ternary complex to the 40S ribosomal subunit is essential for the GAP activity of eIF5 toward eIF2. Therefore, it has been postulated that the 40S ribosomal subunit also interacts with eIF2γ and stabilizes the switch regions of eIF2γ, which is required for the GTPase activity of eIF2 (Das and Maitra 2001; Das et al. 2001). In summary, during translation initiation, the γ subunit of eIF2 which contains the consensus GTP-binding domains G1-G4, is the presumed GTPase, while eIF5 and the 40S ribosome act as GAPs. It should be noted that as a GAP for eIF2, eIF5 increases the rate of GTP hydrolysis in the ribosomal preinitiation complex by over six orders of magnitude (Algire et al. 2005).
Role of eIF5 in AUG Start Codon Selection and the Link Between eIF5-Promoted GTP Hydrolysis and AUG Selection
In eukaryotes, translation is known to be initiated almost exclusively from AUG codons. This is in sharp contrast to prokaryotes where non-AUG codons such as UUG or GUG codons also serve as efficient sites for translation initiation from about 10% of mRNAs. To understand the mechanism by which the stringency of start site selection in eukaryotes is maintained, a series of elegant yeast genetic experiments was carried out by Donahue and his colleagues (Huang et al. 1997; Donahue 2000). It was shown that translation can be initiated from a UUG codon only in the presence of complementary suppressor mutations in several initiation factors – eIF1, α, β, and γ subunits of eIF2, and eIF5. The identification, in this screen of eIF2, a GTP-binding initiation factor, and eIF5, a protein known to be a GAP, indicated that a defect in GTP binding/hydrolysis may be the cause for the breakdown in the fidelity of translation start site selection. Indeed, subsequent biochemical assays (Huang et al. 1997) demonstrated that in each of the Sui mutants, either an elevated level of GTP hydrolysis by eIF2 or an aberrant dissociation of Met-tRNAi releases bound initiation factors from the 43S preinitiation complex prematurely and leaves the initiator Met-tRNA at the P site of the 40S subunit. This allows the translation machinery to initiate from a UUG codon. These data suggest that in eukaryotes, in order to ensure translation fidelity, hydrolysis of GTP must be suppressed during scanning of the mRNA by the 43S preinitiation complex. It can only occur when the 43S preinitiation complex has selected the initiation AUG codon and that the process of AUG selection and eIF5-promoted GTP hydrolysis is likely to be coupled. Since all the components required for GTP hydrolysis, namely eIF2, eIF5, and 40 ribosome, are present in the 43S preinitiation complex, there must exist a mechanism by which premature GTP hydrolysis and therefore aberrant initiation is prevented prior to AUG selection. The possibility exists that the coupling of eIF5-promoted GTP hydrolysis to the AUG selection process may be modulated by other initiation factors bound in the 43S preinitiation complex.
Biochemical support for the role of multiple initiation factors in AUG selection stemmed from a variety of protein-protein interaction studies. It was observed that in addition to its interaction with eIF2, eIF5 also interacts with eIF3, eIF4G, and eIF1 (reviewed by Hinnebusch et al. 2007; Pestova et al. 2007). While the N-terminal domain of eIF5 interacts with the G-domain of eIF2 γ, the C-terminal domain of eIF5 interacts with the K-boxes of eIF2β, as well as with eIF1, eIF3, and eIF4G. Additionally, work in the yeast S. cerevisiae has shown that eIF5 is a component of the multifactorial protein complex consisting of eIF2, eIF3, eIF1, and stoichiometric amount of Met-tRNAi. (Asano et al. 2000) that may be important for coordinating eIF5-promoted GTP hydrolysis to the AUG selection process. In fact, eIF5 was found to be associated with the 43S preinitiation complex bound to eIF4F at the 5′-cap structure of mRNA (Majumdar and Maitra 2005). It may, thus, be possible that the eIF5•eIF2 interaction, which is essential for GTP hydrolysis, is physically blocked in the 43S preinitiation complex prior to AUG selection via protein-protein interactions involving one or more bound initiation factors, and this block is released once AUG is selected by the ribosomal complex.
Subsequent in vitro biochemical studies on eIF5-mediated GTP hydrolysis demonstrated that addition of eIF5 to the 43S preinitiation complex containing bound ternary complex, as well as bound initiation factors eIF1, eIF1A, and eIF3 did not result in complete inhibition of hydrolysis of the GTP bound to the 43S preinitiation complex (Majumdar and Maitra 2005). However, near-complete abolition of GTP hydrolysis was observed when eIF4F protein (containing a bound 5′ mRNA cap structural analog) was added to the 43S preinitiation complex containing bound eIF1, eIF1A, and eIF3 (Majumdar and Maitra 2005). Similar results were obtained when the 43S preinitiation complex was positioned at the 5′-capped end of an mRNA via binding to eIF4F (Majumdar and Maitra 2005). Surprisingly, although both cap analog-bound eIF4F as well as eIF4F free of the cap analog-bound the 43S preinitiation complex with comparable efficiency, eIF4F could exert its effect on eIF5-promoted GTP hydrolysis only when it was bound to the cap analog. Taken together, these results indicate that the specific cap-bound conformation of the eIF4F protein was, perhaps, required to fully block eIF5-promoted GTP hydrolysis in the 43S preinitiation complex. Since, eIF5 was found to be stably associated with 43S preinitiation complex as well, it strongly favors the hypothesis that eIF5 was completely prevented from interacting with eIF2 in the 43S preinitiation complex in the presence of the other bound initiation factors, eIF1, eIF1A, eIF3, and eIF4F and as a result aberrant GTP hydrolysis (i.e., GTP hydrolysis prior to positioning of the 43S preinitiation complex at the AUG start codon) was prevented under these conditions.
eIF5, a single-subunit protein of about 49 kDa in mammals and 45 kDa in the yeast Saccharomyces cerevisiae, in conjunction with GTP and other translation initiation factors, plays an essential role in initiation of protein synthesis in eukaryotic cells. Following scanning of the 5′-UTR of an mRNA by 43S ribosomal preinitiation complex (40S•eIF3•mRNA•Met-tRNA•eIF2•GTP•eIF1•eIF1A) and positioning of the preinitiation complex at the AUG start codon of the mRNA to form the 48S initiation complex, eIF5 interacts with eIF2 bound as a Met-tRNAi•eIF2•GTP ternary complex in the ribosomal complex to promote the hydrolysis of bound GTP. Hydrolysis of GTP is a stringent prerequisite for the eventual release of eIF2•GDP and other bound initiation factors during eIF5B-mediated joining of the 60 ribosomal subunit to the 48S complex to form the functional 80S initiation complex. Extensive biochemical studies have shown that eIF5 physically interacts with eIF2 and acts a GTPase-activating protein (GAP) for eIF2, increasing the intrinsic GTPase activity of eIF2 by over six orders of magnitude. Additionally, recent genetic, biochemical, structural analysis of eIF5 and other initiation factors have provided compelling evidence that eIF5, by virtue of its interaction with other initiation factors, is also an essential component of the scanning 43S ribosomal complex and plays an important role in the stringency of the start codon selection. Specific single-site mutation of eIF5 has been shown to alter the ability of the scanning 43S preinitiation complex to utilize UUG codons as start sites in addition to AUG codons. Thus, in addition to its function as a constitutive GAP for eIF2, the initiative factor eIF5, in conjunction with other initiation factors in the ribosomal scanning complex, plays a direct role in start codon selection.
- Asano K, Krishnamoorthy T, Phan L, Pavitt GD, Hinnebusch AG. Conserved bipartite motifs in yeast eIF5 and eIF2B epsilon, GTPase-activating and GDP-GTP exchange factors in translation initiation, mediate binding to their common substrate eIF2. EMBO J. 1999;18:1673–88.PubMedPubMedCentralCrossRefGoogle Scholar
- Das S, Maitra U. Mutational analysis of mammalian translation initiation factor 5 (eIF5): role of interaction between the beta subunit of eIF2 and eIF5 in eIF5 function in vitro and in vivo. Mol Cell Biol. 2000;20:3941–50.Google Scholar
- Donahue TF. Genetic approaches to translation initiation in Saccharomyces cerevisiae. In: Soneberg N, JWB H, Matthews MB, editors. Translational control of gene expression. New York: Cold Spring Harbor Press; 2000. p. 487–502.Google Scholar
- Hinnebusch AG, Dever TE, Asano K. Mechanism of translation initiation in the yeast Saccharomyces cerevisiae. In: Mathews M, Sonneberg N, Hershey JWB, editors. Translational control in biology and medicine. New York: Cold Spring Harbor Laboratory Press; 2007. p. 225–68.Google Scholar
- Pestova TV, Lorsch JR, Helen CUT. The mechanism of translation initiation in eukaryotes. In: Mathews M, Sonenberg N, Hershey JWB, editors. Translational control in biology and medicine. New York: Cold Spring Harbor Laboratory Press; 2007. p. 87–128.Google Scholar