Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. 2010 Dec; 186(4): 1187–1196.
PMCID: PMC2998303

The Role of eIF1 in Translation Initiation Codon Selection in Caenorhabditis elegans


The selection of a proper AUG start codon requires the base-pairing interactions between the codon on the mRNA and the anticodon of the initiator tRNA. This selection process occurs in a pre-initiation complex that includes multiple translation initiation factors and the small ribosomal subunit. To study how these initiation factors are involved in start codon recognition in multicellular organisms, we isolated mutants that allow the expression of a GFP reporter containing a non-AUG start codon. Here we describe the characterization of mutations in eif-1, which encodes the Caenorhabditis elegans translation initiation factor 1 (eIF1). Two mutations were identified, both of which are substitutions of amino acid residues that are identical in all eukaryotic eIF1 proteins. These residues are located in a structural region where the amino acid residues affected by the Saccharomyces cerevisiae eIF1 mutations are also localized. Both C. elegans mutations are dominant in conferring a non-AUG translation initiation phenotype and lead to growth arrest defects in homozygous animals. By assaying reporter constructs that have base changes at the AUG start codon, these mutants are found to allow expression from most reporters that carry single base changes within the AUG codon. This trend of non-AUG mediated initiation was also observed previously for C. elegans eIF2β mutants, indicating that these two factors play a similar role. These results support that eIF1 functions in ensuring the fidelity of AUG start codon recognition in a multicellular organism.

TRANSLATION initiation is thought to be one of the most complex cellular processes in eukaryotes. It involves at least 12 translation initiation factors (eIFs) comprising over 30 polypeptides (Pestova et al. 2007). These factors bring together an initiator methionyl tRNA (Met-tRNAi), the small ribosomal subunit, and a mRNA to form a 48S initiation complex. An important role performed by this complex is to select an AUG codon to initiate translation of the mRNA. Since the first AUG at the 5′ end of most mRNAs is selected as the start site, it is believed that the initiation complex scans for an AUG start codon as it moves from the 5′-capped end of the mRNA toward the 3′ end, as proposed in the ribosomal scanning model (Kozak 1978; Kozak 1989). The recognition of the AUG start codon is mediated by the anticodon of the Met-tRNAi, and the matching base-pairing interactions between the codon of the mRNA and the anticodon determine the site of initiation (Cigan et al. 1988). These base-pairing interactions are essential, but are likely not the only components required for accurately selecting the correct AUG start codon. Numerous initiation factors along with base-pairing interactions have been shown to aid in the AUG recognition process (Pestova et al. 2007).

Translation initiation factors involved in start codon selection fidelity were first identified through genetic studies performed in the yeast Saccharomyces cerevisiae. Mutant strains with a modified His4 gene that had an AUU instead of an AUG at the native start site were selected for the ability to survive on media lacking histidine (Donahue et al. 1988; Castilho-Valavicius et al. 1990). These mutants were found to be able to produce the His4 protein by using a downstream inframe UUG codon (the third codon within the His4 coding region) as the translation start site. Further analyses determined that non-AUG initiation occurred mostly from a UUG codon and not significantly from other codons (Huang et al. 1997). These mutants defined five genetic loci and were named sui1-sui5 (suppressor of initiation codon) on the basis of their ability to initiate translation at a non-AUG codon.

The sui1 suppressors were found to have missense mutations in eIF1. These missense mutations showed semidominant or codominant properties in non-AUG translation initiation while deletion of the eIF1 gene led to lethality in yeast (Yoon and Donahue 1992). eIF1 is a highly conserved protein with a size of approximately 12 kDa that plays a vital role in multiple translation initiation steps. eIF1 is incorporated into a multifactor complex that includes eIF1A, eIF3, and eIF5 and stimulates the recruiting of the ternary complex (consisting of eIF2 · GTP and the charged Met-tRNAi) to the small ribosomal subunit to form the 43S pre-initiation complex (Singh et al. 2004). eIF1 acts synergistically with eIF1A to promote continuous ribosomal scanning for AUG codons by stabilizing an open conformation that allows mRNA to pass through the complex (Maag et al. 2005; Cheung et al. 2007; Passmore et al. 2007). It also mediates the assembly of the ribosomal initiation complex at the AUG start codon (Pestova et al. 1998). eIF1 dissociates from the complex upon recognition of the AUG codon and this dissociation is necessary to trigger a series of conformational changes leading to the translation elongation phase (Algire et al. 2005). Consistent with these roles, sui1 mutations reduce the affinity of eIF1 for the ribosome and cause premature release of eIF1 at non-AUG codons (Cheung et al. 2007). Other sui mutations support the involvement of four additional genes in translation initiation fidelity in yeast. Mutations have been isolated in the heterotrimeric eIF2 as SUI2 (α-subunit) (Cigan et al. 1989), SUI3 (β-subunit) (Donahue et al. 1988), and SUI4 (γ-subunit) (Huang et al. 1997), and a mutation in eIF5 corresponds to the SUI5 mutant (Huang et al. 1997).

However, the genetic studies that identified these translation fidelity mutants were conducted only in yeast. It is not known if there are similar mechanisms regulating translation initiation fidelity in multicellular organisms. To address this question, we designed a genetic system to isolate C. elegans mutants that have reduced fidelity in AUG start codon selection (Zhang and Maduzia 2010). Mutants were selected on the basis of their ability to express a GFP reporter that contains a GUG codon in place of its native translation start site. Here we report the characterization of two mutants that have mutations in eIF1. Unlike yeast sui1 mutants, which preferred the UUG codon, these mutants are capable of using a subset of non-AUG codons for translation initiation. Our results are consistent with eIF1 playing a role in the fidelity of AUG codon selection, perhaps by discriminating base-pairing interactions between the codon and anticodon during start-site selection.


Growth and handling of worms and C. elegans strains:

Worms were grown on nematode growth medium (NGM) agar plates following standard procedures (Wood 1988). Relevant strains used are as follows:

Mutant screen:

Both eif-1(nb132) and eif-1(nb134) strains were isolated from a screen described previously (Zhang and Maduzia 2010). This screen mutagenized approximately 75,000 genomes with N-ethyl-N-nitrosourea (ENU) in a reporter strain carrying the integrated reporter transgene nbIs4.

Mapping of mutants and gene identification:

To map eif-1(nb132) to a chromosome, double heterozygous males [eif-1(nb132)/+; nbIs4/+] were crossed with strains MT464 and MT465, both of which have chromosomes marked with specific mutant genes. On the basis of segregation of these chromosomal markers from the GFP expression phenotype of eif-1(nb132) in the F2 generation, eif-1(nb132) was found to be linked to chromosome II. Several three-factor or multifactor crosses further localized the mutation to the left of bli-2. From strains carrying the double markers dpy-10unc-4, 12 non-Dpy-10 Unc-4 recombinants were isolated and all of them contained eif-1(nb132). From strains carrying the markers bli-2dpy-10, 10 Bli-2 non-Dpy-10 recombinants were isolated and none of them segregated eif-1(nb132). Data from both of these mapping results suggested that eif-1(nb132) was located to the left of bli-2. From worms with the genotype lin-31+dpy-10/+eif-1(nb132)+, 19 of 20 Lin-31 non-Dpy-10 recombinants segregated the eif-1(nb132) mutation, indicating that eif-1(nb132) was between Lin-31 and Dpy-10 (Figure 1A). SNP mapping with CB4856 and eif-1(nb132) bli-2 strains resulted in a total of 46 Bli-2 non-Eif-1 recombinants and each of these recombination events were found to occur to the right of SNP nbP102 (−2.50), consistent with the location of the eif-1(nb132) mutation being to the left of this marker. When SNP mapping with the triple mutant lin-31eif-1(nb132) unc-4, 15 recombinants with phenotype Lin-31 non-Eif-1 non-Unc-4 were found to occur to the left of SNP marker HW25014 (−3.01). Altogether, these mapping results consistently located eif-1(nb132) between −3.01 and −2.50 map units.

Figure 1.
Identification of mutations in eIF1. (A) Chromosomal location of eif-1(nb132). (B) GFP expression from extrachromosomal arrays coexpressing GUG–GFP reporter and the eif-1(nb132) (C65G) mutant eIF1. (C) eIF1 is an alternatively spliced open reading ...

Candidate gene T27F7.3 was sequenced using DNA lysate prepared from worms either heterozygous or homozygous for the eif-1(nb132) or eif-1(nb134) mutations. Worms were digested for 60 min at 65° with proteinase K (200 μg/ml) in 5 μl of lysis buffer (10 mm Tris–HCl, pH 8.3, 50 mm KCl, 2.5 mm MgCl2, 0.45% Nonidet P-40, 0.45% Tween 20, 0.01% gelatin). After heat inactivating the proteinase K at 95° for 15 min, 1 μl of the lysate was used as DNA template to amplify the candidate gene in a 25-μl PCR reaction using LongAmp Taq DNA polymerase (New England Biolabs, Ipswich, MA). The sequence obtained from the mutants was compared with the C. elegans wild-type reference sequence available at Wormbase (http://www.wormbase.org) using SeqMan from the LaserGene software package (DNASTAR, Inc.).

Construction of plasmids carrying wild-type and mutant eIF1 genes:

The T27F7.3 gene was amplified using Phusion high-fidelity DNA polymerase (New England Biolabs, NEB) with wild-type genomic DNA as a template under the following conditions: 98° for 30 sec followed by 35 cycles of 98° for 5 sec, 58° for 30 sec, and 72° for 3 min with a final incubation at 72° for 10 min. The primer pair (TTTCTGCAGACATCTCTCGCATC and TAGTTATGACGATGATGACTGGG) amplified a 4967-bp fragment that included a 1123-bp sequence 5′ to the first predicted exon of eIF1, a 2719-bp region containing all exons and introns and a 1125-bp sequence 3′ to the last exon. USER cloning sequence adaptors were attached to the above PCR product using Taq DNA polymerase following a short PCR at 94° for 30 sec followed by 5 cycles of 94° for 5 sec, 53° for 30 sec, and 65° for 4 min, and a final step at 65° for 10 min. The same primer pair was used except they also contained USER cloning sequence adaptors [GGAGACA(dU) and GGGAAAG(dU)] at their 5′ ends. The PCR product was inserted into the pNEB206A vector following the instructions in the USER Friendly cloning kit (NEB).

Constructs containing eIF1 mutants were generated from the above plasmid carrying the wild-type gene with primers containing the mutant nucleotides using the Phusion site-directed mutagenesis kit (NEB). The primers are (mutant nucleotide bases are boldface):


Generation of transgenic worms coexpressing eIF1 mutations and non-AUG reporters:

Extrachromosomal arrays were generated by injecting into the gonad with a mixture of 5 ng/μl of eIF1 plasmid, 12 ng/μl of non-AUG reporter, and 50 ng/μl of pRF4 following standard C. elegans transgenic procedures (Mello and Fire 1995). Non-AUG reporters were described previously (Zhang and Maduzia 2010).

Suppression of unc-62(t2012) by eif-1(nb132):

A strain with the genotype nbIs4;+/eif-1(nb132); +/eT1; unc-62(t2012) dpy-11(e224)/eT1 was constructed to obtain Dpy worms with the homozygous genotype unc-62(t2012) dpy-11(e224). Two types of Dpy worms were found: Dpy that expressed GFP due to the presence of eif-1(nb132)/+ and Dpy worms that did not and thus lacked the eif-1 mutation. These two types of Dpy worms were individually placed onto plates to examine if they produced progeny.

Determination of GFP expression:

Visual scoring of GFP levels was performed using a Leica MZFLIII microscope with the objective set at 5× zoom. Fluorescent images were captured with a Zeiss Axiovert 200M microscope using a 10× objective. The intensities of the fluorescent signals of entire images containing between 5 and 10 worms were analyzed and displayed in bar graphs as an average GFP intensity, which represents a mean gray value per pixel. GFP protein expressed from an identical number of worms for each mutant strain using Western blot analysis. Detailed procedures for determining GFP expression were described previously (Zhang and Maduzia 2010).


Missense mutations in a conserved structural interface of eIF1 lead to expression of a non-AUG reporter:

We have previously described a reporter system for isolating mutants that permits the use of non-AUG start codons in translation initiation (Zhang and Maduzia 2010). This reporter contains a GTG codon rather than an ATG codon at the start site of the GFP coding region. The GFP is not expressed from the reporter in wild-type worms due to a robust fidelity in translation initiation site selection, which allows translation to start only from the AUG codon. Mutants were isolated on the basis of their ability to allow GFP expression from this reporter in classic genetic screens. Here we describe the characterization of two mutants, eif-1(nb132) and eif-1(nb134).

To determine the identity of genes mutated in strains carrying the nb132 allele, we performed standard three-factor and SNP mappings to determine its chromosomal position. These mapping results consistently placed eif-1(nb132) between DNA polymorphisms HW25014 and nbP102 on chromosome II (Figure 1A). Close examination of the genes within this region focused our attention on T27F7.3, since it is predicted to encode the C. elegans initiation factor eIF1. The genomic DNA region of T27F7.3 was PCR amplified from mutant worms and sequenced. A single base mutation was found in eif-1(nb132) changing a T to a G. Since the eif-1(nb134) mutant strain had abnormalities similar to that of eif-1(nb132), we sequenced T27F7.3 in eif-1(nb134) worms and found a single base change, a G to an A, at a different position. To confirm identification of this gene, both of the identified mutations were engineered into T27F7.3 by site-directed mutagenesis and introduced into wild-type worms together with the GUG–GFP reporter. Transgenes containing either the eif-1(nb132) or eif-1(nb134) mutations, but not the wild-type gene, allowed GFP expression from this non-AUG reporter (Figure 1B).

T27F7.3 has a unique gene structure and its largest intron within this gene is predicted to contain an unrelated open reading frame that consists of the coding sequence of T27F7.4 (Figure 1C). T27F7.3 is predicted to encode the only eIF1 in the C. elegans genome. The splicing pattern of T27F7.3 was confirmed by our RT–PCR analysis (data not shown) and numerous EST clones in Wormbase. The T27F7.4 transcript is predicted to have all of the exons of T27F7.3, but the eIF1 coding frame is interrupted by an out-of-frame coding region that encodes a mannosyltransferase III involved in glycosylphosphatidylinositol anchor biosynthesis. The eif-1(nb132) and eif-1(nb134) mutations affect only the eIF1 coding region although they are likely present in the 3′-UTR of the mannosyltransferase mRNA (Figure 1C). Interestingly, we surveyed the presence of this special gene structure and found that it is conserved in all examined nematode genomes including C. briggsae, C. remanei, Pristionchus pacificus, and Brugia malayi. We examined the expression of these two genes by fusing GFP in frame with each of the respective coding regions. GFP expression was observed in a number of cells in transgenic animals carrying the eIF1::GFP fusion with higher expression levels appearing in the pharynx and intestine (Figure 1C), but was undetectable in worms carrying the mannosyltransferase III::GFP fusion (Figure 1D).

C. elegans eIF1 is a 109-amino acid protein that shares high-sequence homology with other eIF1s found in eukaryotes ranging from yeast to humans (Figure 2A). The eif-1(nb132) mutation substitutes an amino acid of similar size changing a cystine to a glycine at position 65 (C65G), and eif-1(nb134) leads to a G83R substitution that increases both the size of the side chain and the overall net charge of the protein (Figure 2A). Both affected residues are identical among all eukaryotic eIF1s examined. These mutations occur at residues not affected by previously identified yeast sui1 mutations. Strikingly, when the mutations are mapped to the human eIF1 crystal structure, both the C. elegans and yeast mutations localize to the same structural interface (Figure 2B). Thus, the C. elegans eIF1 mutations and sui1 mutations may affect a similar molecular process.

Figure 2.
Amino acid substitutions in eIF1 mutants. (A) Multiple sequence alignment of selected eukaryotic eIF1s. The amino acid changes in eif-1(nb132), eif-1(nb134), and yeast eIF1 mutations (sui1-1, sui1-4, sui1-17, and mof2-1) are shown and labeled according ...

Mutations in eIF1 are dominant for GFP expression of non-AUG reporters and recessive for growth defects:

Both eif-1(nb132) and eif-1(nb134) strains showed a high level of GFP signal expression from the GTG-containing GFP reporter in heterozygous (Figure 3, B and C) or homozygous (Figure 3, E and F) animals. Worms carrying the eif-1(nb132) allele had a higher level of GFP expression than that of eif-1(nb134) mutant animals, both as heterozygotes or homozygotes (Figure 3G). Western blot analyses with heterozygous animals also indicated that the GFP protein expressed is more abundant in mutant animals than in wild-type animals. Additionally, the GFP produced is the same size as the GFP synthesized from a reporter carrying a native ATG start codon (Figure 3H).

Figure 3.
GFP expression from a GFP reporter containing a GUG codon at the start site in eIF1 mutants. GFP images are shown for adult worms (A–C) and L1 larvae (D–F) with respective genotypes. (A) +/+, wild type. (B) eif-1(nb132)/+. ...

Homozygotes of either mutant also show growth arrest defects. eif-1(nb132) worms were able to grow through the four larval stages. However, they arrested as late L4 worms (9/9 L1 larvae examined) and could not develop into adults. The developing vulva and gonad, both characteristic of L4 development, were visible in these arrested worms but no oocytes were found in the gonad. In contrast, eif-1(nb134) homozygous worms arrested at the L1 larval stage (10/10 L1 larvae examined). A similar L1 larval arrest phenotype was observed when eIF1 activity was knocked down by RNA interference upon injecting eIF1 dsRNA (data not shown). This suggests that the growth arrest phenotype associated with these two mutations resembles the loss-of-function phenotype resulting from the RNAi knockdown of eIF1, and the eif-1(nb134) mutation appears to cause more severe detrimental defects.

Genetic suppression of an initiator AUG codon mutation in unc-62(t2012) by eif-1(nb132):

Since eif-1 mutants allowed GFP expression from transgenes containing a non-AUG reporter, we wondered whether they might also allow expression of endogenous mRNAs that contain a non-AUG codon at the translation start site. Previously, we used genetic suppression experiments to demonstrate that the eIF2β mutant iftb-1(nb101) was able to initiate translation from endogenous mutant unc-62(t2012) mRNAs containing an AUA codon in place of its AUG start codon (Zhang and Maduzia 2010). We performed similar genetic suppression experiments with eif-1(nb132). Mutation of the AUG start codon to AUA in unc-62(t2012) animals causes a maternal-effect lethal phenotype such that homozygous worms grow up normally but their progeny arrest as either embryos or early larvae (Van Auken et al. 2002). We found that the presence of the eif-1(nb132) mutation improves the viability of unc-62(t2012) worms. Out of nine unc-62(t2012) worms that are also heterozygous for the eif-1(nb132) mutation, two of them produced a total of four larvae that grew to either L4 larvae or sterile adults. In contrast, all 11 unc-62(t2012) worms with wild-type eIF1 in the background produced only dead embryos. Suppression of the maternal effect lethal phenotype of unc-62(t2012) by eif-1(nb132) indicates that, similar to iftb-1(nb101), the eif-1(nb132) mutation also allows translation of endogenous mRNAs containing non-AUG codons at the translation start site.

eIF1 containing the eif-1(nb132) or eif-1(nb134) mutations allows translation to start at a subset of non-AUG codons:

Since the eIF1 mutants were isolated using a non-AUG reporter that contains a GUG codon at the site of the native AUG start codon, we set out to determine if these mutants might also permit expression of GFP from reporters carrying other non-AUG codons using a transgenic assay we previously used to study dominant eIF2β mutants (Zhang and Maduzia 2010). In this assay, transgenic worms that coexpress a mutant eIF1 and a GFP reporter containing an altered start codon were examined for GFP expression. The GFP expression level was visually scored under a fluorescent microscope (Table 1). The scoring results generally correlated well with a separate analysis using fluorescent images to estimate GFP brightness on the basis of average intensity of fluorescent signals (Figure 4).

GFP expression from non-AUG reporters coexpressing eIF1 mutants
Figure 4.
Expression of GFP from non-AUG reporters coexpressing eif-1(nb132) or eif-1(nb134) mutant eIF1. (A–C) Fluorescent images for non-AUG reporters expressed with the eif-1(nb132) (C65G) transgene. (D) Expression for eif-1(nb134) (G83R) is summarized ...

We assayed reporters containing all possible single base changes in the AUG start codon as well as a reporter with two base changes. When the wild-type eIF1 transgene is present, little to no GFP expression is detected in transgenic worms containing these reporters (Table 1, column wt; Figure 4A). With either the eif-1(nb132) or eif-1(nb134) mutant transgene, some of the reporters show increased GFP expression. Reporters that have one base change at the first or the third base position within the AUG codon express GFP (Table 1, lines 1–3 and 7–9; Figure 4, A, C, and D). An exception was observed in multiple transgenic lines where eif-1(nb134) did not result in an increase in GFP expressed from a reporter containing AUC at the translation start site (Table 1, line 8; Figure 4D). Changes at the second base position to a C (and thus the codon ACG) allow GFP expression; however, changes to either an A or G (codons AAG or AGG) do not (Table 1, lines 4–6; Figure 4, B and D). Also, no expression was observed from the reporter that contained two simultaneous base changes (Table 1, line 10; Figure 4D). The relative GFP levels from most reporters co-injected with the eif-1(nb132) transgene appeared to be higher than the levels of those same reporters co-injected with the eif-1(nb134) transgene (Figure 4D).

C. elegans eIF1 carrying an equivalent yeast sui1 mutation allows translation initiation at a similar subset of non-AUG codons:

In contrast to what we observed with the eif-1 mutants described above, non-AUG initiation by yeast eIF1 sui1 mutants occurs at UUG but not with other non-AUG codons (Huang et al. 1997). To address whether eIF1 carrying these sui1 mutations are able to confer a similar phenotype in our reporter system, we constructed mutations in the C. elegans eIF1 gene (Figure 2A) corresponding to all known sui1 alleles (sui1-1, sui1-4, and sui1-17) isolated in the His4 suppressor screens (Yoon and Donahue 1992) and mof2-1, which was first isolated as a mutant with an altered frameshift efficiency but later found to have a reduction in start codon recognition fidelity as well (Cui et al. 1998). Since sui1 mutations are dominant or codominant in yeast (Yoon and Donahue 1992), it is possible that these mutations also behave dominantly in C. elegans, allowing the detection of their defects in our transgenic assay.

These mutants were initially assayed with non-AUG GFP reporters carrying GUG or UUG codons. The C. elegans eIF1 transgene carrying the sui1-4 mutation promoted GFP expression from both GUG and UUG reporters unlike transgenes carrying the wild-type eIF1 and the mutant eIF1 genes containing the sui1-1, sui1-17, or mof2-1 mutations (data not shown). These results indicate that only the sui1-4 mutant was able to confer non-AUG translation in the C. elegans reporter system, which is in contrast to that observed in yeast where all these mutations allowed translation to start at the UUG codon.

We further assayed the ability of the sui1-4 mutant to allow GFP expression from reporters containing other non-AUG start codons. GFP expression was observed from additional reporters that changed a single base either at the first or the third base position of the AUG codon (Table 1, lines 1–3 and 7–9; Figure 5, A, C, and D). An exception to this trend was seen in worms carrying a reporter containing the codon AUC. No GFP was expressed in these worms (Figure 5, C and D), similar to what was observed with this reporter in worms containing an eif-1(nb134) mutant transgene (Figure 4D). When the second base was changed, increased GFP expression was observed only when it was changed to a C but not when it was changed to a G or an A. Thus, eIF1 containing the sui1-4 mutation allows translation initiation to occur at a subset of non-AUG codons similar to what was observed for the eif-1 alleles (nb132 and nb134) in this C. elegans assay, as opposed to observations in yeast where non-AUG translation initiates only from the UUG codon.

Figure 5.
Expression of GFP from non-AUG reporters coexpressing sui1-4 (D84Y) mutant eIF1. Arrangement is similar to that in Figure 4 except that images from wild-type eIF1 transgene are not shown. (A) Changes at the first base position. (B) Changes at the second ...


In this study, we have isolated and characterized eIF1 mutants in the multicellular organism C. elegans. Several lines of evidence support that these eIF1 mutants have a defect in selecting the AUG start codon. First, the GUG reporter was constructed in such a way that the GFP mRNA has no in-frame AUG codon available for the synthesis of a functional GFP protein (Zhang and Maduzia 2010). Western blot analysis reveals that the size of GFP expressed from the GUG reporter is the same as that of GFP synthesized from the wild-type GFP reporter carrying a normal AUG start codon, consistent with translation initiating from similar positions on these mRNAs. We also showed that the eif-1(nb132) mutant is able to produce functional products from an endogenous mutant mRNA that contains an AUA codon at the translation start site in a genetic suppression experiment. Finally, in comprehensive assays of GFP reporters that have base changes only at the AUG start codon, these mutants allow GFP expression from some but not all reporters. These results are consistent with a notion that translation of these modified mRNAs likely initiates at the non-AUG codons due to a reduced fidelity in start codon recognition in eIF1 mutants, similar to yeast eIF1 sui1 mutants that allow translation initiation of a modified His4 selection marker gene from a non-AUG start codon (Yoon and Donahue 1992).

The ability of different non-AUG codons to initiate translation in these C. elegans eIF1 mutants is not random. GFP expression was observed only from a subset of reporters containing single base changes within the AUG either at the first or the third position of the start codon. In addition, changing the middle base position to a C (and thus resulting in the codon ACG) also gives considerable levels of expression. There is no detectable expression when two bases are changed. This phenomenon is not limited to mutants isolated from our C. elegans screen. The sui1-4 mutant, the only sui1 allele that permits non-AUG reporter expression when engineered into the C. elegans eIF1, follows a similar trend. These patterns of non-AUG codon usage are comparable with what we observed in C. elegans eIF2β mutants (Zhang and Maduzia 2010), but are different from that observed in yeast with the sui1 and other sui mutants, which show a much higher rate of translation initiation only at the UUG codon (Huang et al. 1997). As we proposed before (Zhang and Maduzia 2010), it is possible that the sequence context surrounding the non-AUG codons in the worm and yeast reporter constructs may contribute to this difference. Interestingly, analogous non-AUG usage is also observed in wild-type yeast (Kolitz et al. 2009) and mammalian cells (Peabody 1989) where sensitive assays allow the detection of low levels of protein expression from mRNAs with altered AUG start codons. This naturally occurring misrecognition indicates that discriminating two base-paired near-cognate codons from the perfect three-base-paired AUG codon is subject to mistakes. Mutations in translation initiation factors, such as eIF1 and eIF2β, further increase the levels of these mistakes.

The trend we observed in C. elegans mutants is best explained by a model where two base-pairing interactions between non-AUG codons and the anticodon of the Met-tRNAi are sufficient to trigger translation initiation, suggesting that wild-type eIF1 plays a role in monitoring proper base-pairing interactions when scanning for the AUG start site. It would be predicted that the Met-tRNAi, not a cognate tRNA matching an individual non-AUG codon, is used in translation initiation at these non-AUG start codons. This prediction is consistent with evidence in the literature. The translation initiation complex will bind only the Met-tRNAi as opposed to other tRNAs because Met-tRNAi has unique sequence and structural features that allow it to be loaded onto eIF2 of the ternary complex and enable it to fit into the P site of the ribosome (Pestova et al. 2007). In addition, it has been demonstrated that Met-tRNAi is indeed used to initiate translation of mRNAs at UUG codons in the yeast sui1 (Yoon and Donahue 1992) and SUI3 mutants (Donahue et al. 1988).

Localization of mutations on the three-dimensional protein structure suggests that a particular structural interface on eIF1 plays an important role in AUG start codon selection. Both C. elegans eif-1 and yeast sui1 mutations are missense mutations affecting amino acid residues that are highly conserved among all eukaryotic eIF1 proteins. Importantly, all affected residues cluster together on a narrow interface (Figure 3B) of the eIF1 crystal structure (Fletcher et al. 1999). Since both C. elegans and yeast mutations lead to initiation at non-AUG start codons, the colocalization of these affected residues indicates that this structural surface is critically involved in translation start codon selection. It is unclear how this interface interacts with other components in the initiation complex. This interface is known to be important in ribosome binding and the affinity of eIF1 for the small ribosomal subunit is severely reduced when either the sui1-1 or sui1-7 mutations are present (Cheung et al. 2007). However, these affected residues appear to mediate ribosomal binding indirectly as they are located slightly to the side of the ribosomal binding site of eIF1 (Lomakin et al. 2003). Additionally, there is evidence that eIF1 interacts with eIF5 (Reibarkh et al. 2008). It is quite possible that this eIF1 interface is involved directly or indirectly in the binding of these two factors.

Interestingly, we find that not all yeast sui1 mutations behave the same in the C. elegans assay as they do in yeast (Yoon and Donahue 1992). When engineered into the C. elegans eIF1 gene, only the sui1-4(D84Y) mutation, but not those of sui1-1(D84G), sui1-17(Q85P), and mof2-1(G108R), allowed detectable expression levels from the non-AUG-codon GFP reporter. Peculiarly, sui1-1(D84G) and sui1-4(D84Y) mutations affect the same amino acid residue, both resulting in a reduction of the net negative charge of eIF1. D84Y also increases the surface hydrophobicity, which may be more detrimental to specific molecular interactions. It is unclear why the other sui1 mutations did not show a similar phenotype in our assay. One possibility is that C. elegans eIF1 containing the sui1-1, sui1-17, or mof2-1 mutations cannot form a proper pre-initiation complex, thus making it impossible to assay for non-AUG translation initiation activity. It is also possible that these sui1 mutations simply do not allow translation initiation at non-AUG codons in the C. elegans reporter system. Either way, these results suggest that there are functional differences between yeast and C. elegans eIF1.


We gratefully acknowledge the encouragement and support of Donald Comb and New England Biolabs, Inc. We thank Tilde Carlow, Ana Egana, Brendan Galvin, Bill Jack, and Chris Noren for critical reading of the manuscript. Some C. elegans strains were obtained from the Caenorhabditis Genetic Center, which is funded by the National Center for Research Resources of the National Institutes of Health.


Available freely online through the author-supported open access option.


  • Algire, M. A., D. Maag and J. R. Lorsch, 2005. Pi release from eIF2, not GTP hydrolysis, is the step controlled by start-site selection during eukaryotic translation initiation. Mol. Cell 20 251–262. [PubMed]
  • Castilho-Valavicius, B., H. Yoon and T. F. Donahue, 1990. Genetic characterization of the Saccharomyces cerevisiae translational initiation suppressors sui1, sui2 and SUI3 and their effects on HIS4 expression. Genetics 124 483–495. [PMC free article] [PubMed]
  • Cheung, Y. N., D. Maag, S. F. Mitchell, C. A. Fekete, M. A. Algire et al., 2007. Dissociation of eIF1 from the 40S ribosomal subunit is a key step in start codon selection in vivo. Genes Dev. 21 1217–1230. [PMC free article] [PubMed]
  • Cigan, A. M., L. Feng and T. F. Donahue, 1988. tRNAi(met) functions in directing the scanning ribosome to the start site of translation. Science 242 93–97. [PubMed]
  • Cigan, A. M., E. K. Pabich, L. Feng and T. F. Donahue, 1989. Yeast translation initiation suppressor sui2 encodes the alpha subunit of eukaryotic initiation factor 2 and shares sequence identity with the human alpha subunit. Proc. Natl. Acad. Sci. USA 86 2784–2788. [PMC free article] [PubMed]
  • Cui, Y., J. D. Dinman, T. G. Kinzy and S. W. Peltz, 1998. The Mof2/Sui1 protein is a general monitor of translational accuracy. Mol. Cell. Biol. 18 1506–1516. [PMC free article] [PubMed]
  • Donahue, T. F., A. M. Cigan, E. K. Pabich and B. C. Valavicius, 1988. Mutations at a Zn(II) finger motif in the yeast eIF-2 beta gene alter ribosomal start-site selection during the scanning process. Cell 54 621–632. [PubMed]
  • Fletcher, C. M., T. V. Pestova, C. U. Hellen and G. Wagner, 1999. Structure and interactions of the translation initiation factor eIF1. EMBO J. 18 2631–2637. [PMC free article] [PubMed]
  • Huang, H. K., H. Yoon, E. M. Hannig and T. F. Donahue, 1997. GTP hydrolysis controls stringent selection of the AUG start codon during translation initiation in Saccharomyces cerevisiae. Genes Dev. 11 2396–2413. [PMC free article] [PubMed]
  • Kolitz, S. E., J. E. Takacs and J. R. Lorsch, 2009. Kinetic and thermodynamic analysis of the role of start codon/anticodon base pairing during eukaryotic translation initiation. RNA 15 138–152. [PMC free article] [PubMed]
  • Kozak, M., 1978. How do eucaryotic ribosomes select initiation regions in messenger RNA? Cell 15 1109–1123. [PubMed]
  • Kozak, M., 1989. The scanning model for translation: an update. J. Cell Biol. 108 229–241. [PMC free article] [PubMed]
  • Lomakin, I. B., V. G. Kolupaeva, A. Marintchev, G. Wagner and T. V. Pestova, 2003. Position of eukaryotic initiation factor eIF1 on the 40S ribosomal subunit determined by directed hydroxyl radical probing. Genes Dev. 17 2786–2797. [PMC free article] [PubMed]
  • Maag, D., C. A. Fekete, Z. Gryczynski and J. R. Lorsch, 2005. A conformational change in the eukaryotic translation preinitiation complex and release of eIF1 signal recognition of the start codon. Mol. Cell 17 265–275. [PubMed]
  • Mello, C., and A. Fire, 1995. DNA transformation. Methods Cell Biol. 48 451–482. [PubMed]
  • Passmore, L. A., T. M. Schmeing, D. Maag, D. J. Applefield, M. G. Acker et al., 2007. The eukaryotic translation initiation factors eIF1 and eIF1A induce an open conformation of the 40S ribosome. Mol. Cell 26 41–50. [PubMed]
  • Peabody, D. S., 1989. Translation initiation at non-AUG triplets in mammalian cells. J. Biol. Chem. 264 5031–5035. [PubMed]
  • Pestova, T. V., S. I. Borukhov and C. U. Hellen, 1998. Eukaryotic ribosomes require initiation factors 1 and 1A to locate initiation codons. Nature 394 854–859. [PubMed]
  • Pestova, T. V., J. R. Lorsch and C. U. T. Hellen, 2007. The mechanism of translation initiation in eukaryotes, pp. 87–128 in Translational Control in Biology and Medicine, edited by M. B. Mathews, N. Sonenberg and J. W. B. Hershey. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.
  • Reibarkh, M., Y. Yamamoto, C. R. Singh, F. del Rio, A. Fahmy et al., 2008. Eukaryotic initiation factor (eIF)1 carries two distinct eIF5-binding faces important for multifactor assembly and AUG selection. J. Biol. Chem. 283 1094–1103. [PubMed]
  • Singh, C. R., H. He, M. Ii, Y. Yamamoto and K. Asano, 2004. Efficient incorporation of eukaryotic initiation factor 1 into the multifactor complex is critical for formation of functional ribosomal preinitiation complexes in vivo. J. Biol. Chem. 279 31910–31920. [PubMed]
  • Van Auken, K., D. Weaver, B. Robertson, M. Sundaram, T. Saldi et al., 2002. Roles of the Homothorax/Meis/Prep homolog UNC-62 and the Exd/Pbx homologs CEH-20 and CEH-40 in C. elegans embryogenesis. Development 129 5255–5268. [PubMed]
  • Wood, W. B., 1988. The Nematode Caenorhabditis elegans. Cold Spring Harbor Laboratory.
  • Yoon, H. J., and T. F. Donahue, 1992. The suil suppressor locus in Saccharomyces cerevisiae encodes a translation factor that functions during tRNA(iMet) recognition of the start codon. Mol. Cell Biol. 12 248–260. [PMC free article] [PubMed]
  • Zhang, Y., and L. L. Maduzia, 2010. Mutations in Caenorhabditis elegans eIF2β permit translation initiation from non-AUG start codons. Genetics 185 141–152. [PMC free article] [PubMed]

Articles from Genetics are provided here courtesy of Genetics Society of America
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Gene
    Gene links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...