Localization of a promoter in the putative internal ribosome entry site of the Saccharomyces cerevisiae TIF4631 gene

Abstract

The 5′-region of the TIF4631 gene of Saccharomyces cerevisiae (encoding the translation initiation factor eIF4G1) was reported earlier to harbor a very active internal ribosome entry site (IRES) allowing for internal initiation of translation of TIF4631 mRNA. Here, we report the presence of a promoter in the region −112 to −36 relative to the translation initiation codon of the TIF4631 gene. This promoter stimulates transcription from a start site at position −36 and generates an mRNA that is actively translated in vitro and able to sustain growth of yeast cells in vivo as the only source of eIF4G. The data show that the IRES activity reported earlier is due to this promoter. On the contrary, the presumed IRES represents a strongly inhibitory element for translation in vitro.

INTRODUCTION

Regulation of translation at the level of initiation plays an important role in the control of gene expression. In eukaryotes, initiation of translation of most mRNAs presumably occurs via the cap-dependent pathway (for reviews, see Pain 1996; Hershey and Merrick 2000; Dever 2002). This pathway involves recognition of the cap structure m⁷GpppN (where N is any nucleotide and m is a methyl group) at the 5′-end of mRNA by the cap-binding protein eIF4E (eukaryotic initiation factor 4E) bound to the anchor protein eIF4G (for reviews, see Sachs et al. 1997; Gingras et al. 1999) and recruitment of eIF4A and eIF4B and ATP-hydrolysis-dependent melting of RNA secondary structure in the 5′-untranslated region (UTR) by the DEAD-box helicase eIF4A. This creates a binding site for the 40S ribosomal subunit, which carries the ternary complex eIF2–GTP–Met-tRNAi and the initiation factors eIF1, eIF1A, eIF3, and eIF5 (43S initiation complex). After binding to mRNA near the cap structure, the 43S complex scans the mRNA in the 3′ direction in search of the initiator AUG codon (Kozak 1989). AUG recognition by base-pairing with the anticodon of Met-tRNAi (Cigan et al. 1988) triggers the hydrolysis of GTP bound to eIF2, release of initiation factors from the 40S ribosomal subunit, and binding of a 60S ribosomal subunit to form a 80S initiation complex competent to translate the open reading frame (ORF) of the mRNA.

An alternative initiation pathway by which a number of viral and some eukaryotic cellular mRNAs are translated is internal initiation mediated by an internal ribosome entry site (IRES; for reviews, see Jackson and Kaminski 1995; Belsham and Jackson 2000; Carter et al. 2000; Jackson 2000; Hellen and Sarnow 2001). The initiation factor requirement for internal initiation depends on IRES type and structure. Translation initiation on most of the viral IRESs does not require either cap-binding factor eIF4E or intact eIF4G (Belsham and Sonenberg 1996; Belsham and Jackson 2000; Jackson 2000) except for the IRES element of hepatitis A virus, which is known to require intact initiation factor eIF4G (Borman and Kean 1997). In contrast, recruitment of 40S ribosomes to hepatitis C virus and swine fever virus IRES does not require the initiation factors of the eIF4 group (Pestova et al. 1998), and cricket paralysis virus IRES directs initiation without requirement for any of the canonical initiation factors (Wilson et al. 2000).

The yeast Saccharomyces cerevisiae is a powerful system to study the mechanism and regulation of translation initiation; however, rather little is known about internal initiation in this system (Altmann et al. 1990; Iizuka et al. 1994; Paz et al. 1999). Reported examples of internal initiation in yeast include reporter mRNAs carrying the poliovirus 5′-UTR (Altmann et al. 1990), a 148-nucleotide IRES element located in the 5′-UTR of the coat protein gene of crucifer-infecting tobamovirus (crTMV; Dorokhov et al. 2002); the cricket paralysis viral RNA (Thompson et al. 2001); the mRNAs coding for the transcription factors TFIID, HAP4, and YAP1 (Iizuka et al. 1994); the mRNA coding for URE2 (Komar et al. 2003); and the mRNA encoding translation initiation factor eIF4G1. The latter was reported to be very potent (Zhou et al. 2001).

Yeast eIF4G is represented by two proteins that are 50% identical at the amino acid level, namely, eIF4G1 (encoded by the gene TIF4631) and eIF4G2 (encoded by the gene TIF4633; Goyer et al. 1993). Internal initiation of translation of mRNA encoding eIF4G seems plausible because synthesis of eIF4G protein may have to occur under conditions in which eIF4G is limiting in cells. Like in most cases of internal initiation, the presence of an IRES in eIF4G1 mRNA was demonstrated by placing the 5′-UTR in a dicistronic reporter construct between two ORFs. The IRES then supported second ORF translation (Zhou et al. 2001).

We attempted to study internal initiation in yeast extracts with dicistronic mRNAs and chose the 5′-UTR of TIF4631 mRNA for our experiments. To our surprise, we found a promoter in this region of the TIF4631 gene.

RESULTS

The 5′-UTR of the TIF4631 gene inhibits translation

We selected the 5′-UTR of the TIF4631 mRNA (encoding translation initiation factor eIF4G1) for in vitro studies of internal initiation of translation because this nucleotide sequence had been reported to contain one of the most potent internal ribosome entry sites (IRES) described for yeast (Zhou et al. 2001). We inserted this sequence into DNA constructs containing one or two reporter genes (Fig. 1A ▶), obtained the corresponding mono- or dicistronic mRNAs by in vitro transcription and tested them in a yeast in vitro translation system (strain CWO4; Fig. 1B ▶; Table 1 ▶). This system supports both cap-dependent and internal initiation as shown with a reporter construct carrying the IRES of yeast URE2 mRNA (Komar et al. 2003). Capped dicistronic mRNA (RNA 1) yielded strong first ORF (Renilla luciferase) and lower second ORF (Photinus luciferase) expression. Second ORF expression from this mRNA was in the order of 10% of what was obtained when this ORF was expressed from a monocistronic mRNA (RNA 3). The rather high expression from the second ORF may be due to reinitiation and/or leaky scanning because it was partially sensitive to cap analog, although less sensitive than first ORF expression. Reinitiation appears more plausible than leaky scanning (scanning through 32 AUG codons before reaching the Photinus initiation codon) because the intergenic region is very short (13 nt). Insertion of 505 nt of the sequence preceding the start codon of the TIF4631 ORF (referred to here as the TIF4631 5′-UTR) significantly inhibited second ORF translation in dicistronic mRNA (RNA 2) while having little effect on first ORF translation. This indicates that this sequence interferes with reinitiation and/or leaky scanning and does not support internal initiation. Inhibition of reinitiation and scanning by the TIF4631 5′-UTR is expected because this nucleotide sequence contains several AUG codons and short ORFs (Goyer et al. 1993).

FIGURE 1.

In vitro translation. Reporter RNAs encoding Renilla and/or Photinus luciferase with or without the 5′-UTR of TIF4631 mRNA were compared for translation in yeast extracts (Materials and Methods). (A) Schematic representation of in vitro transcribed RNAs encoding Renilla and/or Photinus luciferase. Thick lines represent the 5′-UTR of TIF4631 mRNA from position −508 to −3 relative to the translation initiation codon. pA indicates the presence of a 30-residue-long poly(A) tail at the 3′-end of the RNA. (B) In vitro translation. The concentrations of dicistronic RNAs (1 and 2) and monocistronic RNAs (3 to 6) were 33 ng/μL and 13 ng/μL, respectively. The presence of 1 mM of cap analog m⁷G(5′)ppp(5′)G in the translation reaction is indicated. (RLU) Relative luminescence units.

TABLE 1.

Yeast strains used in this study

Strain	Genotype	Reference
CWO4	MATα ade2-1 his3-11,15 leu2-3,112 trp1-1 ura3 canR	Banroques et al. 1986
CBY19	MATa ade2 his3 trp1 ura3 leu2 canR TIF4631::LEU2 tif4632::ura3 <ycp50-TIF4631 URA3>	Berset et al. 1998
CBY1.1	MATa ade2 his3 trp1 ura3 leu2 canR TIF4631::LEU2 tif4632::ura3	Berset et al. 1998
334	MATa ade2 his3 trp1 ura3 leu2 canR cdc33::LEU2 <pCDC33, CEN, TRP1>	Altmann et al. 1989
EGY48	MATα his3 trp1 ura3 LexA op(X6)-LEU2	Zhou et al. 2001

To test the TIF4631 5′-UTR directly for its effect on translation initiation, we measured Photinus luciferase expression from capped and uncapped monocistronic mRNA carrying the TIF4631 5′-UTR (RNA 4 and 6) and compared it with control mRNAs (RNAs 3 and 5). The TIF4631 5′-UTR strongly repressed translation (200–600-fold reduction). We conclude from these data that the TIF4631 5′-UTR does not carry an IRES with activity in our in vitro translation system and that this sequence acts as an inhibitor of translation initiation.

The 5′-UTR of the TIF4631 gene contains a promoter

To understand the discrepancy between our in vitro and the in vivo translation experiments reported earlier (Zhou et al. 2001), we analyzed Renilla and Photinus luciferase expression in vivo in our strains CWO4 and 334 (Table 1 ▶) from the vectors pGal-R.P, pGal-R.4G(−508/−3), pGal-R.4G(−250/−3).P (Table 2 ▶) encoding dicistronic mRNAs (Fig. 2 ▶, top) and compared the expression levels with those previously described by Zhou et al. (2001) with strain EGY48. Using these vectors we were able to reproduce their results: upon induction of transcription of the dicistronic mRNAs with galactose in all three strains (Fig. 2 ▶, bottom, +) the low second ORF expression from construct 1 was stimulated 100–1000-fold when the TIF4631 5′-UTR (from position −508 to −3 in construct 2 or from position −250 to −3 relative to the translation initiation codon in construct 3) was inserted between the two ORFs. Surprisingly, second ORF expression was about the same in all strains analyzed when the transcription of the dicistronic mRNA was not induced (Fig. 2 ▶, bottom, −). Under these conditions, dicistronic mRNA levels should be at least 20-fold reduced as shown by the reduction of first ORF translation. This control (which had not been mentioned by Zhou et al. 2001) indicates the presence of a promoter in the TIF4631 5′-UTR between positions −250 and −3, which leads to the synthesis of a monocistronic mRNA encoding Photinus luciferase.

TABLE 2.

Plasmids used in this study

Plasmid	Comments	Reference
In vivo experiments
pGal-R.P (=pMyr-RP)	Dicistronic construct under GAL1/10 promoter	Zhou et al. 2001
pGal-R.4G(−508/−3).P (=pMyr-p150/RP)	Dicistronic construct under GAL1/10 promoter	Zhou et al. 2001
pGal-R.4G(−250/−3).P (=pMyr-p150/RP(250–508))	Dicistronic construct under GAL1/10 promoter	Zhou et al. 2001
pRS313	Yeast shuttle vector, HIS3 as selectable marker gene	Sikorski and Hieter 1989
pRS313+230.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
pRS313+130.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
pRS313+22.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
pRS313-112.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
pRS313-170.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
pRS313-320.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
pRS313-370.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
pRS313-470.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
pRS313-520.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
pRS313-530.4G	Yeast shuttle vector, HIS3 as selectable marker gene	This work
In vitro experiments
pSP6R.P	Used for in vitro transcription	This work
pSP6R.4G(−505/−3).P	Used for in vitro transcription	This work

FIGURE 2.

In vivo expression of dicistronic reporter constructs. DNA constructs encoding dicistronic mRNAs without the 5′-UTR (pGal-R.P) or with the TIF4631 5′-UTR (−508 to −3; pGal-R.4G(−508/−3).P) or (−250 to −3; pGal-R.4G(−250/−3).P) were transformed into yeast strains CWO4, 334 and EGY48. Transformants were tested for expression of Renilla and Photinus luciferase (Materials and Methods) by measuring 1 μL of a 1/100 dilution of extract. Induction of expression with galactose is indicated by +/−. Measured luciferase activities were normalized to values corresponding to 1 μg of protein.

Identification of a monocistronic mRNA encoding Photinus luciferase

To identify the putative monocistronic mRNA transcribed from the dicistronic construct pGal-R.4G(−508/−3), we carried out RT-PCR on poly(A)⁺ RNA isolated from CWO4 cells transformed either with the plasmid pGal-R.P or pGal-R.4G(−508/−3) (Fig. 3A ▶). Reverse transcription was primed with the oligonucleotide P′, PCR with the reverse primer P′ and the forward primer P (to detect transcripts encoding the second ORF, P-fragment) or forward primer R (to detect transcripts encoding both ORFs, R-fragment). Control experiments omitting reverse transcriptase did not give any signals indicating the absence of contaminating DNA (Fig. 3B ▶, lanes 2,4,6,8,10,12,14,16). PCR reactions with the plasmids pGal-R.P (minus TIF4631 5′-UTR; Fig. 3B ▶, lanes 17,18) and pGal-R.4G(−508/−3) (plus TIF4631 5′-UTR; Fig. 3B ▶, lanes 19,20) were performed to optimize PCR conditions. Induction with galactose of transcription from the GAL1/10 promoter on plasmid pGal-R.P (lanes 13–16) led to the synthesis of dicistronic mRNA, increasing its concentration strongly above the level found in the presence of glucose (Fig. 3B ▶, cf. lanes 9 and 11 with lanes 13 and 15). Induction of transcription on plasmid pGal-R.4G(−508/−3) generated a dicistronic mRNA with a longer intercistronic region (Fig. 3B ▶, lanes 5,7). In the presence of glucose, we could not detect the dicistronic mRNA (Fig. 3B ▶, lane 1) but a monocistronic mRNA containing the Photinus luciferase ORF (Fig. 3B ▶, lane 3). This result confirms our hypothesis that a promoter in the 5′-UTR of TIF4631 generates a monocistronic mRNA. We assume that transcription from this promoter may inhibit residual transcription from the GAL1/10 promoter under repressing conditions through steric hinderance and thus lower the amount of the full-length dicistronic mRNA, rendering it undetectable (Fig. 3B ▶, lane 1).

FIGURE 3.

Detection of monocistronic Photinus luciferase mRNA. (A) Schematic representation of dicistronic mRNAs. The reverse primer P′ (hybridizing to nucleotides 282–301 of the Photinus luciferase ORF) was used for cDNA synthesis. The forward primer P (hybridizing to nucleotides 1–21 of the Photinus luciferase ORF) or the forward primer R (hybridizing to nucleotides 903–917 of the Renilla luciferase ORF) were used with the reverse primer P′ for PCR, resulting in the synthesis of a P-fragment or R-fragment. (B) Wild-type CWO4 yeast cells were transformed with the plasmids pGalR.P or pGalR.4G(−508/−3).P, and expression was induced with galactose for 3 h at 30°C (Materials and Methods). Total RNA was extracted, poly(A)⁺ RNA was isolated and DNase-treated, and reverse transcription and PCR were performed as detailed in Materials and Methods. For PCR reactions (15 μL), 1.2 μL of cDNA preparation was amplified. Two pairs of primers, shown in A, were used for amplification of each cDNA. The PCR was run for 30 cycles. From each reaction, 4 μL was loaded on a 2% agarose gel and stained with ethidium bromide.

Localization of the promoter in the TIF4631 5′-UTR

To localize the promoter, we deleted parts of the TIF4631 5′-UTR by exonuclease III digestion (Materials and Methods) and tested the truncated constructs by inserting them in the promoterless yeast shuttle vector pRS313 and testing for complementation of a yeast strain (CBY19, Table 1 ▶) lacking both chromosomal copies of TIF4631 and TIF4632. This strain carries the wild-type TIF4631 gene on a URA3 plasmid and is therefore viable. By plating transformants on 5-FOA (5-fluoro-orotic acid) plates (selection for loss of URA3 plasmid), cells carrying a truncated TIF4631 gene but still able to express eIF4G1 could be selected (Fig. 4A ▶). On plates lacking 5-FOA, all transformants grew because of the presence of the wild-type TIF4631 gene on the plasmid ≤ycp50-TIF4631; URA3> (Fig. 4A ▶, left plate). On plates containing 5-FOA, transformants carrying plasmids with deletions in the TIF4631 5′-UTR from position −530 (full length) to −112 (relative to the translation initiation codon) were able to grow, whereas transformants carrying plasmids with longer deletions (+22 to +230) were not able to grow (Fig. 4A ▶, right plate). Transformants that grew on plates containing 5-FOA expressed full-length eIF4G1 as shown by Western blotting (Fig. 4B ▶): the TIF4631 gene with only 112 nt upstream of the start AUG codon codes for the synthesis of eIF4G1 of the same size as the one encoded by the TIF4631 gene with 530 nt upstream of the AUG codon or by wild-type chromosomal TIF4631. These data demonstrate the presence of a promoter in the region −1 to −112 of the TIF4631 gene.

FIGURE 4.

Localization of a promoter in the TIF4631 gene. (A) pRS313 plasmids containing the TIF4631 gene with different 5′-UTRs (numbers indicate the nucleotide position relative to the AUG start codon of the ORF) were transformed into yeast strain CBY19. Transformed cells were grown on minimal medium plates with or without 0.7% 5-FOA at 30°C. (B) Strain CWO4 (wild-type) and CBY19 transformants containing pRS313–530.4G or pRS313–112.4G were grown (after selection on 5-FOA plates) as described (Materials and Methods). Pellets corresponding to 1 mL of cells (samples 1, 4, 7), 2 mL of cells (samples 2, 5, 8), or 5 mL of cells (samples 3, 6, 9) were fractionated by SDS polyacrylamide gel electrophoresis and analyzed by Western blotting with anti-eIF4G1 antibody (Materials and Methods). The arrow on the left points to eIF4G1; arrows on the right indicate the position of molecular mass markers (in kilodaltons).

Determination of transcription start sites

To determine the transcription initiation start site(s) we analyzed the length of the TIF4631 mRNA 5′-UTR by RT-PCR and RACE. We isolated poly(A)⁺ RNA from CWO4 (wild-type), CBY19/pRS313–112.4G and CBY19/pRS313–530.4G (data for this strain are not shown). For RT-PCR, cDNA was synthesized from poly(A)⁺ RNA using a primer hybridizing to positions 476–497 in the ORF of the TIF4631 mRNA and was then amplified by PCR with a collection of primers hybridizing to different sites in the 5′-UTR of the TIF4631 gene. PCR control experiments using the TIF4631 gene as template showed that every primer pair produced the expected DNA fragment (Fig. 5A ▶, DNA controls). TIF4631 mRNA derived from wild-type cells as well as from cells carrying the shortest functional TIF4631 gene (deletion to position −112) gave rise to PCR products ending in the region of nucleotide positions −59 to −85. We conclude from these data that the longest form of mRNA produced by the promoter present in the TIF4631 5′-UTR (full length or deletion −112) carries an untranslated region of ~70–90 nt.

FIGURE 5.

Identification of transcription start sites by RT-PCR and RACE. (A) RT-PCR (Materials and Methods) was done with poly(A)⁺ RNA from CWO4 cells (wild-type) and CBY19 cells containing the plasmid pRS313–112.4G after selection on 5-FOA. The reverse primer MA150 hybridized to nucleotide positions 497–476 in the TIF4631 ORF, and the forward primers hybridized to the 5′-UTR of the TIF4631 gene at the positions indicated at the top of the figure. DNA controls were done with the plasmid pRS313–530.4G. (B) Putative promoter elements. The sequence of the TIF4631 gene from position −112 to position +27 is shown (position 1 represents the A of the ATG start codon). Putative promoter elements are underlined. (C) Primer extension experiments: 500,000 cpm of ³²P-labeled primer MV01 complementary to nucleotides 81–97 of the eIF4G ORF was hybridized to 20 μg of total RNA isolated from strain CW04 grown to an OD₆₀₀ of 0.62 (lane 1) or strain CBY1.1 (which contains a TIF4631 deletion) grown to an OD₆₀₀ of 0.56 (lane 2) and extended with reverse transcriptase.

To map the transcription initiation site(s) more precisely, we used RACE experiments on poly(A)⁺ RNA from wild-type CWO4 cells. Sequence analysis of the RACE products was used to determine the ends of the cDNAs produced in the reverse transcriptase reaction. The analysis of 10 RACE products revealed two ending at nucleotide position −75, one ending at −54, five ending at −36, one ending at −35, and one ending at −31. In these experiments we cannot distinguish between true transcription start sites and incomplete cDNA synthesis caused by premature reverse transcription termination or mRNA degradation. However, we conclude that the longest transcripts we can detect start at nucleotide position −75 and the most abundant at −36 (Fig. 5B ▶).

We visualized the products of primer extension reactions by performing reverse transcriptase reactions using a 5′-(³²P)-labeled oligonucleotide that hybridized close to the start AUG of TIF4631 mRNA. Two main signals were obtained when using total RNA from strain CWO4 that corresponds to transcription start sites at positions −38 and −2 (Fig. 5C ▶, lane 1). Also, two longer products were detected in this reaction at positions around −390 and −580 (data not shown). We don’t know if those products were obtained because of DNA traces in our RNA preparations or if they correspond to real transcription start sites. The first explanation seems unlikely as control experiments with total RNA prepared from a yeast strain carrying a TIF4631 deletion (strain CBY1.1; Table 1 ▶) did not give any signal in our reverse transcription reactions (Fig. 5C ▶, lane 2). In any case, our primer extension experiments confirm the existence of a main transcription start site located around position −38, which we also detected in our RACE experiments.

Because a stretch corresponding to nucleotides −580 to −3 of the 5′-UTR was found to be inhibitory for translation (Fig. 1 ▶), we next determined the translational activity of Photinus luciferase mRNAs with shorter 5′-UTR sequences derived from the TIF4631 gene. We produced capped Photinus luciferase mRNAs with the 5′-UTRs extending to positions −75 and −36 and tested them for in vitro translation in yeast extracts. The activities of these mRNAs were on the order of 1000-fold higher than for mRNA containing a 5′-UTR extending to position −508 (Fig. 6 ▶). These results indicate that TIF4631 mRNAs with 5′-UTRs of 36 or 75 nt are highly efficient for translation.

FIGURE 6.

In vitro translation of capped Photinus luciferase mRNAs. Reporter RNAs encoding Photinus luciferase with different TIF4631 5′-UTR sequences were translated in yeast extracts (Materials and Methods). The concentration of RNA was 13 ng/μL. (RLU) Relative luminescence units.

DISCUSSION

In this work, we identified a promoter in the region −112 to −36 (relative to the AUG initiator codon) of the TIF4631 gene encoding yeast eIF4G1. The identification of the promoter-bearing region is based on the complementation experiments described in Figure 4 ▶. These data show that a DNA construct retaining sequences up to −112 is still functional, whereas a DNA construct starting at position +22 is not functional in vivo. DNA constructs with truncated TIF4631 genes starting at nucleotide positions +22, +130, and +230 cannot complement, although they encode mRNAs with an AUG codon at position 256 (amino acid 86). Initiation of translation at this AUG codon leads to the synthesis of N-terminally truncated eIF4G, which was shown to be functional in vivo (Tarun et al. 1997). This shows that the promoter driving the expression of the TIF4631 gene on the plasmid used in our complementation experiments is not plasmid-borne because such a promoter should lead to the synthesis of mRNAs encoding truncated but functional eIF4G1 protein.

By using Northern analysis, we previously identified a single TIF4631 mRNA of ~3800 nt (Goyer et al. 1993). In the course of this work, we decided to perform RT-PCR and RACE to identify the 5′-end of TIF4631 mRNAs because small differences in mRNA length cannot be determined as accurately by Northern analysis. Because the majority of TIF4631 mRNAs that we identified in wild-type yeast cells using RACE have their 5′-end at position −36 (Fig. 5B ▶), primer extension experiments show two main mRNA bands at position −38 and −2 (Fig. 5C ▶), and a truncated TIF4631 gene extending to position −112 encodes full-length eIF4G1 (Fig. 4B ▶), we conclude that the promoter elements must be located in the nucleotide sequence between −36 and −112. This region contains two binding sites for transcription factor Bas2 (from −109 to −104 and from −45 to −39) and one binding site for transcription factors Gcr1 and Gcn4 (located in the region from −88 to −103; Fig. 5B ▶; http://cgsigma.cshl.org/jian/). Perhaps, the Bas2 element located at position −104 to −109 is sufficient to drive transcription without further enhancer elements. Further experiments will be required to precisely map the DNA elements required for efficient transcription of TIF4631 mRNA.

Previously, longer 5′-UTRs for TIF4631 mRNAs detected by primer extension experiments have been reported (Goyer et al. 1993). We have also detected potentially longer transcripts in our primer extension experiments, but, as mentioned above, we do not know if they correspond to real mRNA 5′-ends or if they are artifacts. One possible error source is the presence of residual DNA in mRNA preparations. We observed longer DNA bands in our RT-PCR experiments when DNase treatment was insufficient (data not shown) and therefore always ran an RT-PCR control experiment without reverse transcriptase to make sure that all DNA bands detected originated from RNA (as shown in Fig. 3B ▶). From our experiments, we cannot exclude that additional longer forms of TIF4631 mRNA are produced and perhaps translated by internal initiation. However, we state that longer forms of TIF4631 mRNA are not essential for yeast growth, although they may have a function under certain physiological conditions. From our in vitro data, we also conclude that there is no strong IRES sequence in the 5′-UTR of TIF4631.

We have not found any specific phenotype for yeast strains carrying TIF4631 genes with truncated 5′-UTRs. The generation time at 30°C in YPD was similar (100 min) for the strain CBY19 transformed with pRS313–112.4G, pRS313–170.4G, pRS313–320.4G, pRS313–370.4G, pRS313– 470.4G, pRS313–520.4G, or pRS313–530.4G after selection on 5-FOA. Furthermore, we did not detect significant differences between these strains in recovery after starvation, sporulation (diploid strains), or growth at lower (22°C) or higher (37°C) temperatures. We therefore conclude that the promoter in the region −112 to −36 is sufficient for growth under all these conditions by promoting the synthesis of mRNA with short 5′-UTR that is translated in a cap-dependent fashion.

Our findings are reminiscent of the situation in higher eukaryotes, where a strong promoter was found in the putative IRES of the gene encoding eIF4G1. This promoter also lies in a nucleotide sequence earlier believed to harbor an IRES (Han and Zhang 2001). As in our case, no evidence for an IRES could be found when a dicistronic mRNA containing this eIF4G1 sequence was translated in vitro or in vivo.

Finally, these results should remind us that second ORF translation from dicistronic mRNAs is insufficient to claim internal initiation (Kozak 2001) and that several additional control experiments have to be carried out, among them the determination of potential promoter activity in the putative IRES. Along this line, the presence of cryptic promoters in the genes TFIID, HAP4, and YAP1 encoding yeast transcription factors may be responsible for the expression of these proteins from dicistronic DNA constructs when encoded by the second ORF (Hecht et al. 2002).

MATERIALS AND METHODS

Yeast strains

The yeast strains used in this study are shown in Table 1 ▶.

Plasmids

The plasmids used in this study are shown in Table 2 ▶.

SP6R.P

The Photinus luciferase ORF was amplified by PCR on luciferase T7 DNA from Promega (plasmid provided with the kit TNT T7 Coupled Reticulocyte Lysate System) with a forward primer introducing a BamHI restriction site followed by an NcoI site (at the ATG start codon of the luciferase ORF) and the reverse primer introducing a SacI site. The resulting DNA fragment was inserted in BamHI/SacI-cut SP64 Poly(A) plasmid (Promega), producing plasmid SP6P. The plasmid SP6R.P was created introducing the Renilla luciferase ORF as an HindIII/BamH1 fragment into a HindIII/BamHI-cut plasmid SP6P.

SP6R.4G(−508/−3).P

The 5′-untranslated region of the TIF4631 gene from position −508 to −3 was amplified by PCR on genomic DNA of S. cerevisiae. The forward primer inserted a BamHI site at the 5′-end and the reverse primer an NcoI site (at the ATG start codon of the TIF4631 ORF) at the 3′-end. This sequence was then inserted into a BamH1/NcoI-cut plasmid SP6R.P.

pGal-R.P, pGal-R.4G(−508/−3).P, pGal-R.4G(−250/−3).P

These constructs were kindly provided by W. Zhou, who referred to them as pMyr-RP (pGal-R.P), pMyr-p150/RP (pGal-R.4G(−508/−3).P), and pMyr-p150/RP (250–508) (pGal-R.4G(−250/−3).P) (Zhou et al. 2001).

Deletion of nucleotide sequences in the 5′-region of the TIF4631 gene

The double-stranded nested deletion kit (Pharmacia) was used to produce deletions in the 5′-region of the TIF4631 gene on the plasmid pRS313.4G (vector pSR313 with inserted 3.5-kb BamHI/EcoRI fragment carrying the 5′-region [−530/−1] and the TIF4631 ORF with a His6x tag). This vector was digested with BamHI and SacI and the linerarized vector was incubated with exonuclease III at 28°C for different times to digest DNA from the 3′-end. S1 nuclease was then used to digest DNA single strands. Finally, the Klenow fragment of Pol I was used to produce blunt ends, and DNA was recircularized by ligation and transformed into Escherichia coli XL2blue. Plasmids from transformants were analyzed for the length of the 5′-region of the TIF4631 gene by restriction enzyme digestion and sequencing. Plasmids containing inserts starting at positions −520, −470, −370, −320, −170, −112, +22, +130, and +230 relative to the ATG start codon (called pRS313–530.4G, pRS313–520.4G, pRS313–470.4G, pRS313–370.4G, pRS313–320.4G, pRS313–170.4G, pRS313–112.4G, pRS313+22.4G, pRS313+130.4G, and pRS313+230.4G) were further analyzed.

Primers

The oligonucleotide primers used in this study are shown in Table 3 ▶.

TABLE 3.

Oligonucleotide primers used in this study

Primer	Sequence
MV01	5′-GTTGCTGAGATTCCTGC-3′
MV07	5′-CCATGTGTCCACCTCTGA-3′
MA150	5′-GGTTGCGGAGACACAGTAGATC-3′
P′	5′-CAACTCCGATAAATAACGCG-3′
R	5′-GTTCGTTGAGCGAGT-3′
P	5′-ATGGAAGACGCCAAAAACATA-3′
−483/−460	5′-GTAAGGCTTTTTTCAATATCTCTG-3′
−409/−391	5′-TCATGATGGCAGACTTCCA-3′
−314/−294	5′-CTTTTTCACCGTATTTTTGTG-3′
−251/−228	5′-CTCGTACTGTTTCACTGACAAAAG-3′
−136/−111	5′-CCAATCTTGATATTGTGATAATTTAC-3′
−112/−87	5′-ACTTAATTATGATTCTTCCTCTTCCC-3′
−85/−59	5′-CAATTTCTTAAAGCTTCTTACTTTAC-3′
−60/−37	5′-ACTCCTTCTTGCTCATAAATAAGC-3′
−36/−11	5′-AAGGTAAGAGGACAACTGTAATTACC-3′
T7-75/Nco1	5′-TAATACGACTCACTATAGGAAAGCTTCTTA CTTTACTCCTTCTTG-3′
T7-36	5′-TAATACGACTCACTATAGGAAGGTAAGA GGACAACTGTAATTACCTATTACAATAAT GGAAGACGCCAAAAAC-3′
T7-36/Nco1	5′-TAATACGACTCACTATAGGAAGGTAAGA GGACAACTGTAATTACC-3′
T7-508/Nco1	5′-TAATACGACTCACTATAGGAATCATTTTTTT GAAAATTACATTAATAA-3′
Sp6pA/rev	5′-TATGACATGATTACGAATTCGGTT-3′

In vitro transcription

Plasmids SP6R.P and SP6 R.4G(−508/−3).P were linearized with BsrBI. DNA for transcription of monocistronic RNA (constructs 3–6; Fig. 1A ▶) was obtained by PCR on plasmid SP6 R.4G(−508/−3).P using the reverse primer SP6pA/rev and the forward primers T7–75/Nco1, T7–36, T7–36/Nco1, and T7–508/Nco1, which introduced a T7 promoter. DNA was purified by phenol/chloroform extraction and ethanol precipitation (2.5 volumes) in the presence of 2.5 M ammonium acetate.

In vitro transcription reactions contained 1 mM ATP, UTP, and CTP; 0.2 mM GTP; 1 mM m⁷G(5′)ppp(5′)G (New England Biolabs); and (per microliter of reaction mixture) 40–50 ng of DNA, 0.8 U of RNasin, 1 U of RNA polymerase (SP6 or T7), and 6000 cpm of ³⁵S-UTP (to quantify the RNA by measurement of UTP incorporation). After 2 h of incubation at 37°C, the RNA was phenol/chloroform-extracted and precipitated with ethanol (2.5 volumes) in the presence of 2.5 M ammonium acetate. Pellets were resuspended in DEPC-treated water.

In vitro translation

Yeast extracts were prepared as described (Altmann and Trachsel 1997). Briefly, cells were collected in the logarithmic growth phase and treated with β-mercaptoethanol, followed by treatment with zymolyase and regeneration in rich medium containing 1 M Sorbitol. Spheroblasts were homogenized with a dounce homogenizer. After isolation of the S100 fraction by centrifugation, the extract was passed through a Sephadex G25 column. The OD₂₆₀ peak of the void fraction was collected and stored in liquid nitrogen.

Extracts were treated for 10 min at room temperature with micrococcal nuclease (Amersham Pharmacia Biotech). Translation reactions (15 μL) contained 28 mM HEPES/KOH (pH 7.4), 180 mM KAc, 3 mM MgAc₂, 1 mM ATP, 0.4 mM GTP, 12 mM creatine phosphate, 50 μg/mL creatine phosphokinase, 1.2 mM DTT, and 50 μM amino acids. The RNA concentrations used for translation are indicated in the figure legends. Translation mixes were incubated for 1 h at 20°C.

Luciferase assay

Photinus luciferase assays (50 μL) were done in Eppendorf tubes (saturated with BSA) in 20 mM Tris-phosphate (pH 7.8), 1 mM MgAc₂, 2.7 mM MgSO₄, 0.1 mM EDTA, 0.5 mM ATP, 34 mM DTT, 0.47 mM D-Luciferine (Sigma), and 0.27 mM Coenzyme A (Sigma). Measurements were done in a luminometer TD20/20 (Turner Design) by integrating the signal for 10 sec.

Dual luciferase assays were done using the Promega kit (dual luciferase reporter assay system) as recommended by the manufacturer using the luminometer TD20/20.

In vivo expression of dicistronic constructs

Wild-type CWO4, strain 334, and EGY48 (Invitrogen) were transformed with the plasmids pGal-R.P, pGal-R.4G(−508/−3).P, and pGal-R.4G(−250/−3).P using the lithium acetate method (Ito et al. 1983). Transformed cells were grown in minimal medium with 2% glucose at 30°C to exponential phase (OD₆₀₀ ~ 0.6), washed with minimal medium, divided into two equal parts, and either incubated with 2% galactose and 1% raffinose (induction) or with 2% glucose (control). After 3 h of induction at 30°C, the cells were collected and resuspended in two volumes of 1× passive lysis buffer (Promega) containing 1 mg/mL porcine gelatine (PLB-GP). Cells were lysed by adding one volume of glass beads (45–65 μm) and vortexing four times for 30 sec. Extracts were centrifuged at 16,000g for 3 min at 4°C. Lysates (diluted 1/100 in PLB-GP buffer) were measured using the dual luciferase reporter assay system (Promega). Protein concentrations were determined with the Bio-Rad protein assay.

Complementation experiments

Plasmids pRS313 (control), pRS313+230.4G, pRS313+130.4G, pRS313+22.4G, pRS313–112.4G, pRS313–170.4G, pRS313–320.4G, pRS313–370.4G, pRS313–470.4G, pRS313–520.4G, and pRS313–530.4G were transformed into yeast strain CBY19 (Berset et al. 1998) using the lithium acetate method (Ito et al. 1983). The haploid CBY19 strain has both chromosomal gene copies (TIF4631 and TIF4632) disrupted and carries the wild-type TIF4631 gene under its own promoter on a URA3 plasmid (ycp50-TIF4631, URA3). The cells were analyzed for their ability to survive on 0.7% 5-FOA plates (loss of ycp50-TIF4631).

SDS polyacrylamide gel electrophoresis and Western blot analysis

Yeast cells were grown to exponential phase and harvested by centrifugation. The cell pellet from 1 mL of culture was resuspended in 250 mM Tris-HCl (pH 6.8), 10 mM DTT, 10% SDS, 0.1% β-mercaptoethanol, 50% glycerol, and 0.5% bromophenol blue. Cells were lysed by heating to 100°C in this buffer for 1 min. Equal amounts of protein were loaded on 10% SDS polyacrylamide gels (Anderson et al. 1973; Berset et al. 1998) and blotted onto nitrocellulose for 45 min at 60 V in a Mini Trans Blot Cell (Bio-Rad). Blots were saturated with 2.5% BSA in TBS (10 mM Tris-HCl at pH 7.5, 150 mM NaCl) for 1 h at room temperature and incubated overnight with rat polyclonal antibodies against eIF4G1 (amino acids 542–883; 1:500 dilutions in TBS containing 0.5% BSA). After washing with TBS, blots were decorated for 1 h with peroxidase-conjugated rabbit anti-rat Ig (Dako) and stained with 0.018% chloronaphthol and 0.006% H₂O₂ in TBS. Equal protein loading was verified by Coomassie blue staining.

Isolation of total RNA from yeast cells

Yeast cells were harvested and resuspended in 10 mM Tris-HCl (pH 7.5), 2 mM EDTA, 150 mM LiCl (6 mL of buffer per gram of cells). Glass beads (45–60 μm, 10 g/g of cells), LiDS to 1% final concentration, and phenol/chloroform (10 mL/g of cells) were added. The mixture was vortexed three times for 30 sec and centrifuged at 5000g for 5 min at 4°C. The supernatant was re-extracted twice with phenol/chloroform and precipitated with 2.5 volumes ethanol containing 100 mM potassium acetate. RNA was resuspended in DEPC-treated water.

Isolation poly(A)⁺ RNA

Oligo(dT)₂₅ dynabeads (Dynal) were used as recommended by the manufacturer: 300 μg of total RNA was incubated with 200 μL of dynabeads and poly(A)⁺ RNA eluted in 8 μL of DEPC-treated water. Poly(A)⁺ RNA was treated with RNase-free DNase (1 U/μL in 10 mM Tris-HCl at pH 8, 10 mM MgSO₄, 1 mM CaCl₂). DNase was inactivated by addition of 2 mM EGTA.

RT-PCR

Reverse transcription and PCR were carried out in a one-tube reaction in 20 mM Tris-HCl (pH 8.5), 50 mM KAc, 2.5 mM MgAc₂, 10 mM DTT, 0.4 mM dNTPs, and 1 mM primers. Reverse transcription reactions (15 μL) contained 1.2 μL of poly(A)⁺ RNA and 3 U of AMV reverse transcriptase and were incubated for 45 min at 48°C.

For the PCR reaction, 0.1 U/μL of Taq DNA polymerase was added. Some cDNA preparations were diluted before PCR (see figure legends). Controls (minus AMV reverse transcriptase) never gave any signal.

RACE

The 5′ RACE kit (Invitrogen) was used with poly(A)⁺ RNA as recommended by the manufacturer, except that AMV reverse transcriptase was used. The reverse transcription primer was MA150. PCR amplification of dC-tailed cDNA was done with primers MV07 and Abridge Anchor Primer. For the final amplification, MV07 and AUAP primers were used.

PCR product were purified through QIAGEN qiaquick columns as recommended and inserted into pGEM-T vector (Promega). Clones were analyzed by PCR, restriction digestion, and DNA sequencing.

Abbreviations

• eIF, eukaryotic initiation factor

• SDS, sodium dodecyl sulfate

• 5-FOA, 5-fluoro-orotic acid

• IRES, internal ribosome entry site

• Ac, acetate

• RACE, rapid amplification of cDNA ends

• PCR, polymerase chain reaction

• RT, reverse transcription

• RLU, relative luminescence units

• ORF, open reading frame

• nt, nucleotide

• DTT, dithiothreitol

• EGTA, ethylene glycol-bis-(2-aminoethyl)-N,N,N′,N′-tetraacetic acid

• EDTA, ethylenedioxy-diethylene-dinitrilo-tetraacetic acid

• BSA

• bovine serum albumin

• DEPC, diethylpyrocarbonate