The RNA recognition motif of eukaryotic translation initiation factor 3g (eIF3g) is required for resumption of scanning of posttermination ribosomes for reinitiation on GCN4 and together with eIF3i stimulates linear scanning

Abstract

Recent reports have begun unraveling the details of various roles of individual eukaryotic translation initiation factor 3 (eIF3) subunits in translation initiation. Here we describe functional characterization of two essential Saccharomyces cerevisiae eIF3 subunits, g/Tif35 and i/Tif34, previously suggested to be dispensable for formation of the 48S preinitiation complexes (PICs) in vitro. A triple-Ala substitution of conserved residues in the RRM of g/Tif35 (g/tif35-KLF) or a single-point mutation in the WD40 repeat 6 of i/Tif34 (i/tif34-Q258R) produces severe growth defects and decreases the rate of translation initiation in vivo without affecting the integrity of eIF3 and formation of the 43S PICs in vivo. Both mutations also diminish induction of GCN4 expression, which occurs upon starvation via reinitiation. Whereas g/tif35-KLF impedes resumption of scanning for downstream reinitiation by 40S ribosomes terminating at upstream open reading frame 1 (uORF1) in the GCN4 mRNA leader, i/tif34-Q258R prevents full GCN4 derepression by impairing the rate of scanning of posttermination 40S ribosomes moving downstream from uORF1. In addition, g/tif35-KLF reduces processivity of scanning through stable secondary structures, and g/Tif35 specifically interacts with Rps3 and Rps20 located near the ribosomal mRNA entry channel. Together these results implicate g/Tif35 and i/Tif34 in stimulation of linear scanning and, specifically in the case of g/Tif35, also in proper regulation of the GCN4 reinitiation mechanism.

FIG. 1.

The triple-Ala substitution of the highly conserved residues in RNPs of the g/Tif35 RRM in g/tif35-KLF impairs cell growth and the rate of general translation initiation. (A) The unpublished nuclear magnetic resonance solution structure of the human heIF3g RRM (K. Tsuda et al., unpublished; Protein Data Bank accession code 2CQ0) displays a canonical RRM fold with a four-stranded antiparallel β-sheet packed against two perpendicular α-helices. The highly conserved Arg242 in position 2 of RNP2 in β-sheet 1 and Phe282 and Phe284 in positions 3 and 5 of RNP1 in β-sheet 3 of human eIF3g RRM correspond to Lys194 and to Leu235 and Phe237, respectively, of yeast g/Tif35-RRM and are highlighted in red. (B) Amino acid sequence alignment of g/Tif35-RRM of Saccharomyces cerevisiae with that of other species. The amino acid sequence of S. cerevisiae g/Tif35 (accession number NP_010717) between residues 185 and 274 (the terminal residue) is aligned with its Schizosaccharomyces pombe (accession number CAA18400), Candida albicans (accession number Q59ZV5), Salmo salar (accession number ACI69727), Xenopus tropicalis (accession number Q28CY2), Bos taurus (accession number Q3ZC12), Mus musculus (accession number Q9Z1D1), and Homo sapiens (accession number O75821) homologs. The alignment was conducted with ClustalW (http://www.ch.embnet.org/software/ClustalW.html). Highly conserved sequences of RNP2 and RNP1 are shown in yellow and green, respectively, with the key residues that were subjected to site-directed mutagenesis highlighted in red. The sequence of g/tif35-KLF generated in this study is given at the bottom. (C) g/tif35-KLF severely impairs cell growth at elevated temperatures. The H464 (g/tif35Δ) strain was transformed with YCp22-g/Tif35-screen (top row) and YCp22-g/tif35-KLF (bottom row), and the resident YEp-TIF35-U (URA3) plasmid was evicted on medium containing 5-FOA. The resulting strains were then spotted in four serial 10-fold dilutions on SD medium and incubated at 30, 34, and 37°C for 2 days. The far right columns show results of Western analysis of WCEs from the very same strains grown at 34°C, using anti-g/Tif35 (g/Tif35 expression; lane 1) and anti-GCD11 (eIF2γ loading; lane 2) antibodies. (D) The g/tif35-KLF mutant reduces rates of translation initiation in vivo. Polysome profiles are shown for the strains in panel C cultured in YPD medium at 34°C and treated with cycloheximide just prior to harvesting. WCEs were separated by velocity sedimentation through a 5-to-45% sucrose gradient centrifugation at 39,000 rpm for 2.5 h. The gradients were collected and scanned at 254 nm to visualize the ribosomal species. Positions of 40S, 60S, and 80S species are indicated by arrows, and P/M ratios are given above the profiles.

FIG. 2.

The g/tif35-KLF mutant neither affects the integrity of eIF3 in the MFC nor impairs the RNA-binding activity of g/Tif35. (A) The g/tif35-KLF mutation does not reduce binding of g/TIF35 to b/Prt1, i/Tif34, or β-globin mRNA in vitro. Full-length WT g/Tif35 (lane 3) and mutant g/tif35-KLF (lane 4) fused to GST, and also GST alone (lane 2), were tested for binding to ³⁵S-labeled b/Prt1 and i/Tif34 and to ³²P-labeled β-globin mRNA. Lane 1 (In) contains 10% and 2.5% of input amounts of proteins and RNA, respectively, added to each reaction mixture. (B) The g/tif35-KLF mutation does not prevent g/Tif35 from associating with eIF3 in the MFC in vivo. WCEs were prepared from H421 (g/tif35Δ) bearing untagged g/Tif35 (lanes 1 to 4), H111 expressing 8×His-tagged g/Tif35 (lanes 5 to 8), and H112 expressing 8×His-tagged g/tif35-KLF (lanes 9 to 12). The WCEs were incubated with Ni²⁺-silica resin, and the bound proteins were eluted and subjected to Western blot analysis. Lanes 1, 5, and 9 contained 5% of the input WCEs (In); lanes 2, 6, and 10 contained 30% fractions eluted from the resin (E1); lanes 3, 7, and 11 contained 60% of the same fractions (E2); lanes 4, 8, and 12 contained 5% of the flowthrough (FT). The Western signals for a/Tif32, b/Prt1, eIF2, eIF5, and i/Tif34 in the E1 and E2 fractions for the WT g/TIF35 and mutant g/tif35-KLF strains were quantified, combined, normalized for the amounts of WT g/Tif35 in these fractions, and these data are plotted in the histogram on the right as percentages of the corresponding values calculated for the WT g/TIF35.

FIG. 3.

g/tif35-KLF reduces processivity of scanning and interferes with the reinitiation process by preventing posttermination retention of the 40S ribosome on GCN4 mRNA. (A) g/tif35-KLF imparts a strong Gcn⁻ phenotype, implicating g/Tif35 in regulation of translational control of GCN4 expression. Isogenic strain H464 (GCN2 g/TIF35; row 1) and H421 (gcn2Δ g/TIF35; row 2) transformed with an empty vector YCplac22 and strain H111 (GCN2 g/tif35Δ YCp22-g/TIF35-screen; row 3) and H112 (GCN2 g/tif35Δ YCp22-g/tif35-KLF; row 4) were spotted in four serial 10-fold dilutions on SD (left panel) or SD containing 30 mM 3-AT (right panel) and then incubated at 34°C for 3 or 6 days, respectively. (B, construct i) g/tif35-KLF reduces basal expression of GCN4-lacZ and prevents its full derepression upon starvation. H111 and H112 were transformed with p180 and grown in minimal medium for 6 h, and the β-galactosidase activities were measured in the WCEs and are expressed in units of nmol of o-nitrophenyl-β-d-galactopyranoside hydrolyzed per min per mg of protein. To induce GCN4-lacZ expression, strains grown in minimal medium for 2 h were treated with 10 mM 3-AT for 6 h. The mean values and standard deviations obtained from at least six independent measurements with three independent transformants, along with the activities the g/tif35-KLF mutant strain relative to the corresponding WT, are given in the table. White versus black squares indicate REI-permissive (uORF1) versus REI-nonpermissive (uORFs 2 to 4) uORFs. (Constructs ii to iv) The failure of g/tif35-KLF to derepress GCN4-lacZ is not caused by leaky scanning. H111 and H112 were transformed with p227 (ii), pM226 (iii), and pA80z (iv) and analyzed as for construct (i), except that they were not treated with 3-AT. Xs point to mutations eliminating the AUG start codons of uORFs 1 to 4. A paperclip symbol indicates an insertion of the nucleotide sequence. (Constructs v to viii) g/tif35-KLF blocks induction of GCN4 expression by reducing the amount of posttermination 40S ribosomes on uORF1, which resume scanning for reinitiation downstream. H111 and H112 were transformed with pG67 (v), pM199 (vi), p209 (vii), or p1014l (viii) and analyzed as described for constructs ii to iv. Scissors indicate deletions of the nucleotide sequence. (Constructs ix and x) g/tif35-KLF increases the translation-inhibitory effect of a stable stem-loop structure inserted in the 5′-UTR of uORF-less GCN4 mRNA. H111 and H112 were transformed with pWCB07 (ix) or pWCB06 (x) and analyzed as described for construct i. Both constructs contain the indicated sequences (with complementary bases underlined) inserted 21 nt 5′ of the GCN4 AUG codon. (C) β-Galactosidase activities obtained for constructs v to viii in panel B, with WT and g/tif35-KLF cells were plotted as a function of the intercistronic distance between uORF1 and the AUG start codon of GCN4-lacZ.

FIG. 4.

Overexpression of eIF1 in g/tif35-KLF diminishes GCN4 upregulation upon starvation even at 30°C. (A) Overexpression of eIF1 in g/tif35-KLF produces the Gcn⁻ phenotype at 30°C. Isogenic strains H111(g/TIF35) and H112 (g/tif35-KLF) were transformed with the following combinations of two vectors: empty vectors YEplac195 and YEplac181 (EV; row 3), pDSO22 and YEplac181 (hc eIF1A; row 4), YEpSUI1-U and YEplac181 (eIF1; row 5), YEp-SUI1 + TIF11 and YEplac181 (eIF1 + eIF1A; row 6), YEpTIF2(4A)-U and YEplac181 (hc eIF4A; row 7), YEpTIF4631(4G)-U and YEplac181 (eIF4G; row 8), YEpTIF2(4A)-L and YEpTIF4631(4G)-U (hc eIF4A + eIF4G; row 9). These strains, together with control strains H464 (GCN2; row 1) and H421 (gcn2Δ; row 2) transformed with YEplac195 and YEplac181, were spotted in four serial 10-fold dilutions on SD or SD medium containing 30 mM 3-AT and incubated at 30°C for 3 or 6 days, respectively. (B) High-copy-number expression of eIF1 in g/tif35-KLF severely blocks induction of GCN4-lacZ expression. Isogenic strains H111 (g/TIF35) and H112 (g/tif35-KLF) transformed either with empty vector YEplac112 (EV) or pCF82 (hc eIF1) were further transformed with p180 and analyzed as described for construct i in Fig. 3B. (C) Overexpression of eIF1 or eIF4A exacerbates the Gcn⁻ phenotype of a/tif32-Δ8. Isogenic strains YBS47 (a/TIF32) and YBS53 (a/tif32-Δ8) were transformed with the same combinations of two vectors as described for panel A, spotted together with the control strains (as for panel A) in four serial 10-fold dilutions on SD or SD containing 30 mM 3-AT, and incubated at 30°C for 2 to 8 days, as indicated.

FIG. 5.

The g/tif35-KLF mutation diminishes the requirement of uORF1 5′ enhancer sequences for efficient REI. (A) Schematic showing the predicted position of the 40S ribosome terminating at the stop codon of uORF1 from the GCN4 mRNA leader (based on data from reference 49). E, P, and A sites of the 40S ribosomes are aligned with the last two coding triplets and the TAA stop codon. The locations of the 5′ enhancer (labeled 5′-a/Tif32 enhancer to denote the interaction with the NTD of a/Tif32), linker, and buried parts of the sequences upstream of uORF1 are indicated at the top; the 3′ boundaries of the Δ40 deletion (identical to Δ160), Δ46 deletion, and the (C)₄GG multiple substitution are shown below the line depicting mRNA. (B, constructs i to v) The same experiment as described for construct i in Fig. 3B, except that H111 and H112 were transformed with p209, pBS64, pBS62, pVM11, and pBS63 (constructs i to v, respectively) and analyzed without 3-AT treatment. Activities relative to WT are given as percentages in boldface in the table to the right of the schematics.

FIG. 6.

g/Tif35 specifically interacts with Rps3 and Rps20 situated on the beak of the solvent-exposed side of the 40S subunit. Also shown in a revised model of the hypothetical location of eIF3 on the S. cerevisiae small ribosomal subunit. (A) g/Tif35 fused to GST (lane 3) or GST alone (lane 2) was tested for binding to ³⁵S-labeled Rps3, -20, and -2 essentially as described for Fig. 2A. (B and C) Revised hypothetical location of the S. cerevisiae eIF3 on the back side of the 40S subunit, based on the data presented in this study and data from reference . The cryo-electron microscopy reconstruction of the 40S subunit is shown from the solvent side, with RNA segments in yellow and proteins in green. Positions of Rps3 and Rps20, i/Tif34, the RRM of g/Tif35, and the extreme CTD of b/Prt1 are highlighted in bold. The mRNA entry and exit channels are designated by an asterisk and an X, respectively. The blue lines represent mRNA. Positions of Rps2, -3, and -9 were modified according to findings described in reference .

FIG. 7.

Genetic evidence that i/tif34-Q258R prevents induction of GCN4 expression by a combination of modestly increased leaky scanning of uORF1 and a severe reduction in the rate of scanning from uORF1. (A) i/tif34-Q258R imparts a strong Gcn⁻ phenotype, implicating i/Tif34 in regulation of translational control of GCN4. Isogenic strains H450 (GCN2 i/TIF34; row 1) and H420 (gcn2Δ i/TIF34; row 2) transformed with empty vector YCplac111 and strains H120 (GCN2 i/tif34Δ YCp111-i/TIF34; row 3) and H121 (GCN2 i/tif34Δ YCp111-i/tif34-Q258R; row 4) were spotted in four serial 10-fold dilutions on SD (left panel) or SD containing 30 mM 3-AT (right panel) and then incubated at 30°C for 6 or 8 days, respectively. (B, construct i) i/tif34-Q258R prevents full derepression of GCN4-lacZ expression upon starvation. Isogenic strains H120 (i/TIF34) and H121 (i/tif34-Q258R) were transformed with p180 and analyzed as described for construct i in Fig. 3B. (Constructs ii to iv) The i/tif34-Q258R mutation increases leaky scanning over the AUG start codon. H120 and H121 were transformed with p227 (ii), pM226 (iii), and pA80z (iv) and analyzed as described for constructs ii to iv in Fig. 3B. (Constructs v to vii) The i/tif34-Q258R mutation reduces the rate of scanning of posttermination 40S ribosomes from uORF1. H120 and H121 were transformed with pG67 (v), pM199 (vi), or p209 (vii) and analyzed as described for constructs ii to iv in Fig. 3B.

FIG. 8.

Simultaneous overexpression of eIFs 1 and 1A partially suppresses the Gcn⁻ phenotype of i/tif34-Q258R. (A) Isogenic strains H120 (i/TIF34) and H121 (i/tif34-Q258R) were transformed with the same combinations of two vectors as for Fig. 4A, spotted together with H450 (GCN2; row 1) and H420 (gcn2Δ; row 2) transformed with YEplac195 and YEplac181 in four serial 10-fold dilutions on SD or SD containing 10 mM 3-AT, and incubated at 30°C for 5 or 8 days. (B) High-copy-number expression of eIFs 1 and 1A increases the GCN4-lacZ activity in the i/tif34-Q258R cells treated with 3-AT. Isogenic strains H120 (i/TIF34) and H121 (i/tif34-Q258R) transformed either with empty vector YEplac112 (EV) or YEp-SUI1+TIF11-W (hc eIF1 + eIF1A) were further transformed with p180 and analyzed as described for construct i in Fig. 3B.