Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (156K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (37)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Kozak, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kozak, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 4853-4859  

Primer extension analysis of eukaryotic ribosome-mRNA complexes
Introduction
Materials And Methods
   Plasmids
   Synthesis of mRNAs
   Primer extension assay of initiation complexes
Results And Discussion
   Primer extension reaction conditions
   Ribosome binding reaction conditions
   Application of the toeprinting assay to test the fidelity of initiation
   A test of the processivity of scanning
   Trapping of 40S subunits upstream from the AUG codon
References

Primer extension analysis of eukaryotic ribosome-mRNA complexes

Primer extension analysis of eukaryotic ribosome-mRNA complexes

Marilyn Kozak*

Department of Biochemistry, University of Medicine and Dentistry of New Jersey, 675 Hoes Lane, Piscataway,NJ 08854-5635, USA

Received July 21, 1998; Revised and Accepted September 10, 1998

ABSTRACT

Conditions are described for using a primer extension inhibition (toeprinting) assay to study the initiation step of protein synthesis in rabbit reticulocyte lysates. These studies revealed that chloramphenicol acetyltransferase mRNA, which is widely used as a reporter, forms unusually labile initiation complexes. This and other unexpected problems were solved by adjustments in pH and temperature during the reverse transcriptase step. Complications that may occur during the ribosome binding step were also examined, including the possibility of rapid mRNA degradation. The suitability of inhibitors commonly used to block the elongation phase of translation was studied. The refined toeprinting assay was used to confirm context-dependent selection of the AUG start codon. Absence of the m7G cap did not subvert the process wherein initiation is restricted to the first AUG codon. The fidelity of initiation was impaired, however, when NaF was introduced during the ribosome binding step. In a preliminary assessment of the processivity of scanning, no dissociation of 40S ribosomal subunits was detected as the distance from the cap to the AUG codon was increased to nearly 300 bases. With an mRNA that contains a pseudoknot upstream from the AUG codon, the toeprinting assay revealed 40S ribosomal subunits trapped behind the base paired structure. Thus the assay is usable for mapping some intermediates as well as for detecting conventional 80S initiation complexes.

INTRODUCTION

The initiation of translation in eukaryotes is known to occur via a scanning process, but the mechanism of scanning and the function of numerous protein factors await definition. Although some important aspects of the initiation mechanism have been elucidated by working with a subset of purified components (1-6), it is not clear that all of the protein factors required for efficient assembly of functional initiation complexes have been identified. One way to judge progress toward that goal is to compare the properties of complexes assembled from purified components with complexes that form in a highly active, unfractionated translation system. The processivity of scanning, context effects on AUG codon recognition and the ability to discriminate between two close AUG codons are some of the properties of interest. With an unfractionated translation system, the interaction of ribosomes with mRNA can be limited to the initiation step by including antibiotics that inhibit elongation. The resulting complexes are usually monitored by sucrose gradient centrifugation.

Some years ago, a much simpler assay was developed for monitoring initiation complexes in prokaryotic translation systems (7). In this `toeprinting' assay, a 32P-labeled oligonucleotide is annealed to the mRNA downstream from the presumed site of initiation and reverse transcriptase is used to extend the radioactive primer up to the position of the bound ribosome. The assay works well in prokaryotes (8-10), where initiation complexes can be assembled from purified components; but the use of unfractionated translation systems in eukaryotes complicates the primer extension reaction. Earlier, I obtained reproducible toeprinting results with a reticulocyte translation system by using column chromatography to purify initiation complexes before the reverse transcriptase step (11,12). Sucrose gradient purification of ribosome-mRNA complexes prior to toeprinting was employed in another study (13). Both of these solutions are cumbersome, however, when there are many mRNAs to be analyzed. Quantitative comparisons among mRNAs are also made difficult by the variable dilution that occurs during chromatographic purification of initiation complexes.

Here I describe conditions that allow toeprinting directly after the ribosome binding step in a reticulocyte translation system. I describe some unexpected complications and some solutions. Chloramphenicol acetyltransferase (CAT) mRNA, which is often employed in translation experiments, forms unusually labile initiation complexes. Recognition of this and other potential problems is important as the toeprinting assay gains wider use (14-16).

The improved reaction conditions are used herein to demonstrate context-dependent selection of the AUG start codon and perturbation of that reaction by NaF. An assessment of the ability of 40S ribosomal subunits to scan over long distances is also presented. The scanning model predicts that the introduction of a base paired structure upstream from the AUG codon should block scanning and the toeprinting assay provides a way to test that prediction.

MATERIALS AND METHODS

Plasmids

Riboprobe vector pSP64 (Pharmacia Biotech Inc) was previously modified by inserting a CAT cartridge preceded by a promoter for T7 RNA polymerase (17). As described previously (12), a cassette mutagenesis technique, targeted to a HindIII site adjacent to the promoter, was used to vary the 5[prime]-untranslated region (5[prime]-UTR) of the encoded mRNA. The particular sequences used in this study are depicted in Figure 1. The structures of all plasmids were confirmed by appropriate dideoxy chain termination sequencing reactions using Sequenase 2 (US Biochemical Corp.).


Figure 1. Sequences of mRNAs used for toeprinting assays. For the mRNA designated T7-AUGcat (line 1), the sequence of the 5[prime]-UTR and first 14 codons is shown. The rest of the protein coding sequence is given in Alton and Vapnek (18). For the mRNAs depicted in lines 2 and 3, the last 13 nt (GAUCCGAGAUUUU) correspond to nt 26-38 in the 5[prime]-UTR of T7-AUGcat mRNA. Downstream from this sequence the three mRNAs are identical. The AUG codon highlighted in T7-AUGpre mRNA (line 2) thus precedes the normal start site for translation of CAT. In T7-aug(AC41)AUG mRNA, two AUG codons (marked #1 and #2) precede AUGcat. The m7G cap appended to the 5[prime]-end of all transcripts is not shown.


Synthesis of mRNAs

CsCl-purified plasmid DNA, linearized by digestion with AvaI, was used as the template for transcription by T7 RNA polymerase (Gibco BRL). The reactions were carried out as described previously (12). mRNAs were purified by extraction with phenol, concentrated by ethanol precipitation, dissolved in water and stored at -70°C. All mRNAs were capped unless it is explicitly stated otherwise.

Primer extension assay of initiation complexes

The deoxyoligonucleotide CTCAAAATGTTCTTTACGATGCC, which would serve to prime the final reverse transcriptase step, was labeled at the 5[prime]-end by incubation with T4 polynucleotide kinase and [[gamma]-32P]ATP (3000 Ci/mmol; New England Nuclear). This primer is complementary to CAT codons 16-23. In step 1 of the toeprinting reaction, the 32P-labeled primer was pre-annealed to the mRNA by heating for 1 min at 65°C followed by incubation at 37°C for 8 min in 40 mM Tris-HCl (pH 7.5) and 0.2 mM EDTA. The primer-mRNA complexes were than held on wet ice for ~15 min while the reticulocyte reaction mixtures were assembled.

The ribosome binding reactions (step 2) used micrococcal nuclease-treated rabbit reticulocyte lysate obtained from Promega Corp. or Boehringer Mannheim Corp. The source of lysate is stipulated in the legend to each figure. Unless otherwise indicated, the reaction mixtures containing 45% reticulocyte lysate were supplemented with 90 µg/ml cycloheximide, 200 µM sparsomycin, 2 mM Mg(CH3COO)2 and 100 mM KCl (for Promega) or 100 mM potassium acetate (for Boehringer Mannheim). The potassium salt used in each case follows the supplier's recommendation.

In the standard assay, 25 µl aliquots of this mixture containing 2 µl of mRNA/primer (0.1-0.25 µg of mRNA) from step 1 were incubated in 6 × 50 mm glass tubes at 25°C for 6 min and then diluted with 20 vol of cold buffer containing 50 mM Tris-HCl (pH 7.5), 40 mM KCl, 6 mM MgCl2, 5 mM DTT, 110 µg/ml cycloheximide and 575 µM of each of four dNTPs. Primer extension (step 3) was initiated by adding 2 U/µl Superscript II reverse transcriptase (Gibco BRL) and incubating at 25°C for 10 min. Reactions were terminated by extracting with phenol. These standard conditions were arrived at after testing alternative conditions as indicated in various figure legends.

Primer extension products were mixed with formamide (40%) and EDTA (8 mM) and heated at 95°C for 3 min before layering onto 8% polyacrylamide sequencing gels. For reference, RNA sequence ladders were generated by primer extension with dideoxynucleotides using avian myeloblastosis virus (AMV) reverse transcriptase (Gibco BRL). Autoradiograms of the dried gels were obtained by exposure of X-omat AR film (Kodak) with an intensifying screen at -70°C for 1-8 h.

RESULTS AND DISCUSSION

The simplest mRNA constructed for these studies has a 58 nt 5[prime]-UTR preceding the AUG codon that initiates translation of CAT (Fig. 1, line 1). A second mRNA, with an AUG codon introduced upstream, was designed to test the properties of initiation complexes formed in a different environment. The upstream, in-frame start site (Fig. 1, line 2) is designated AUGpre because it initiates translation of an N-terminally extended `preCAT' protein. The experiments that follow show that, although T7-AUGpre mRNA retains the AUGcat codon downstream, under normal conditions initiation is limited to the 5[prime] proximal AUGpre codon, as expected from the scanning mechanism of initiation (19).

These two mRNAs were used in a series of experiments to develop reaction conditions for reliable toeprinting in vitro using a rabbit reticulocyte translation system. A primer complementary to CAT codons 16-23 was used with both mRNAs. When reverse transcriptase extends this primer without interruption to the 5[prime]-end of the mRNA, the product is 127 (with T7-AUGcat mRNA) or 173 nt (with T7-AUGpre mRNA) long. When primer extension is interrupted by an 80S ribosome bound at the AUG codon, the foreshortened products are 50 (AUGcat) or 113 nt long. This reflects the fact that an 80S ribosome, centered on the AUG codon, protects 15-16 nt 3[prime] of that codon (20).

The toeprinting reaction is carried out in three steps. In the first step, a 32P-labeled deoxyoligonucleotide is annealed to the mRNA. This preannealing of the primer produces a much stronger toeprint than is obtained when ribosomes are allowed to bind the mRNA before addition of the primer. Preannealing of the primer would not be expected to work well with translation systems in which RNase H activity is high, but RNase H is reported to be less of a problem with rabbit reticulocyte lysates than with some other systems (21,22).


Figure 2. Comparison of avian and murine reverse transcriptases in toeprinting reactions. A standard ribosome binding reaction at 25°C using Promega's reticulocyte lysate was carried out with T7-AUGcat mRNA. In preparation for primer extension, the ribosome binding reactions were diluted 20- (lanes 5 and 6) or 5-fold (lane 7). Primer extension was carried out at 25°C for 10 min with either AMV reverse transcriptase (2.5 U/100 µl, lane 5) or with Superscript II (200 U/100 µl, lanes 6 and 7). In the T lane of the sequencing ladder, a black dot marks the position of the first AUG codon. The ribosome-mediated primer extension pause site labeled AUGcat maps 15-16 nt downstream from this AUG. The primer is extended all the way to the 5[prime]-end in the fraction of mRNA molecules not engaged by ribosomes.


In the second step of the protocol, ribosomes are allowed to engage the mRNA. This is followed by dilution into appropriate buffer (Materials and Methods) and addition of reverse transcriptase, which extends the primer to the edge of the bound ribosome. The experiments below evaluate various reaction conditions for steps 2 and 3. These experiments are followed by some examples in which the refined toeprinting assay is used to analyze aspects of the initiation process.

Primer extension reaction conditions

In various toeprinting experiments, the primer extension step has been carried out using either AMV reverse transcriptase (7,13,16) or Superscript II (11,14,15). The latter enzyme derives from Moloney murine leukemia virus. With initiation complexes formed in a rabbit reticulocyte translation system, I tested both enzymes and found that Superscript II gave the stronger toeprint, as can be seen by comparing lanes 5 and 6 in Figure 2. With Superscript II, 20-fold dilution of the ribosome binding reaction mixture produced a strong, compact toeprint (Fig. 2, lane 6). The toeprint was more diffuse when the ribosome binding reaction mixture was diluted only 5-fold before the primer extension step (Fig. 2, lane 7). Based on these results, Superscript II was used in all subsequent experiments, with 20-fold dilution of the ribosome binding mixture. The reverse transcriptase reaction appears to be complete after 10 min. Longer incubation times do not increase the yield of primer extension products but may, under some conditions (below), decrease the yield.

Toeprinting of ribosome-mRNA complexes has sometimes been carried out at 37°C (11,12) or higher temperatures (14) in the hope of minimizing pauses caused by secondary structure. To inactivate an inhibitor (perhaps RNase H or a phosphatase) in extracts from some sources, some investigators have subjected ribosome-mRNA complexes to a temperature of 50°C or higher (14,15) prior to primer extension. The experiment in Figure 3 shows that elevated temperatures can produce erratic results.


Figure 3. Comparison of the stability of initiation complexes located at different AUG codons. Incubation of T7-AUGpre (lanes 1-3) or T7-AUGcat mRNA (lanes 4-6) in Promega's reticulocyte lysate was followed by primer extension using Superscript II for 15 min at 30 (lanes 1 and 4), 37 (lanes 2 and 5) or 42°C (lanes 3 and 6). As in Figure 2, the bands in the autoradiogram labeled AUGpre and AUGcat reflect ribosomes bound at the first AUG codon in each mRNA. An additional band (marked Echo) appears in lanes 4-6 under conditions that destabilize initiation complexes at AUGcat.

The most surprising finding is that the stability of ribosome-mRNA complexes during toeprinting varies depending on the sequence around the AUG codon. For example, ribosomes bound at AUGpre were not demonstrably destabilized when the temperature for the reverse transcriptase step was raised from 30 to 42°C (Fig. 3, lanes 1-3), but toeprinting of ribosomes at AUGcat was strongly affected by the temperature used for primer extension (Fig. 3, lanes 4-6). Analysis of an aliquot of these reactions by glycerol gradient centrifugation revealed that ribosomes dissociate from T7-AUGcat mRNA at 42°C, which explains the diminished toeprint. Unexpectedly, this is paralleled by the appearance of a new band (labeled Echo) upstream from AUGcat (Fig. 3, lane 6). Note that, at both AUGpre and AUGcat, the flanking sequence includes A in position -3 and G in position +4. Therefore, the temperature-induced dissociation of ribosomes at the latter site is not related to conventional context effects. (To confirm this, I toeprinted AUGpre after changing A-3 to U-3; the strength of the new toeprint was identical at 30 and 42°C.) The results of other experiments suggested that the stem-loop structure downstream from AUGcat (Fig. 1, line 1) contributed to the instability of initiation complexes at that site. I did not attempt further to define the destabilizing sequences, however, because my main goal was to define reaction conditions under which the toeprinting assay would give reliable results irrespective of the sequence of the mRNA.


Figure 4. Toeprinting of initiation complexes using different inhibitors to block the elongation phase of translation. The ribosome binding step was carried out with T7-AUGpre mRNA in Promega's reticulocyte lysate at 30°C using the following antibiotic inhibitors: lane 1, sparsomycin (200 µM) and cycloheximide (90 µg/ml); lane 2, cycloheximide alone (900 µg/ml); lane 3, anisomycin (0.4 mM); lane 4, anisomycin (1.6 mM). For the primer extension step that follows ribosome binding, the standard reaction mixture which includes cycloheximide (Materials and Methods) was used for all samples. The band labeled AUGpre is the toeprint caused by 80S ribosomes.


Loss of the toeprint at AUGcat and generation of the artifactual `echo' were prevented when the temperature was lowered to 25°C for the reverse transcriptase step. Toeprinting results were improved further by adjusting the pH. Although the supplier recommends pH 8.3 for reactions with Superscript II, this pH promotes extraneous pauses during primer extension, caused apparently by reticulocyte-derived proteins that stabilize base paired structures in the mRNA. These artifactual pauses were reduced or eliminated when reverse transcription was carried out at pH 7.5. Based on these results, incubation with Superscript II at 25°C for 10 min at pH 7.5 was adopted as our standard protocol for the primer extension step.

Ribosome binding reaction conditions

Successful toeprinting requires 80S ribosomes to be held at the AUG initiator codon. This can be accomplished by a combination of sparsomycin and cycloheximide (Fig. 4, lane 1) or by a higher concentration of cycloheximide alone (lane 2). Although anisomycin is a potent inhibitor of elongation in conventional translation assays (23), the more sensitive toeprinting assay shows that ribosomes are not completely immobilized by anisomycin even at high concentrations (Fig. 4, lanes 3 and 4). It might be relevant that, whereas anisomycin inhibits aminoacyl-tRNA binding in the A-site (24), sparsomycin and cycloheximide promote inert binding of aminoacyl-tRNA in the P-site (25,26). Based on the clean result in lane 1 of Figure 4, a combination of sparsomycin and cycloheximide was used for all subsequent experiments in this study.

Judging from the time course and the extent of ribosome binding at AUGcat, as shown in Figure 5, there is little difference in the efficiency of toeprinting using a reticulocyte lysate from Promega (lanes 2-4) versus Boehringer Mannheim (lanes 6-8). The toeprint at AUGcat is not seen in control reactions (lanes 1 and 5) in which the ribosome binding mixtures were assembled at 4°C and diluted directly into primer extension buffer. This confirms that ribosome binding, which occurs normally during step 2, does not continue during the subsequent primer extension step.


Figure 5. Comparison of ribosome binding and mRNA stability using different sources of reticulocyte lysate. For the ribosome binding step of the toeprinting assay, T7-AUGcat mRNA was incubated at 30°C with reticulocyte lysate from Promega (lanes 1-4) or Boehringer Mannheim (lanes 5-8). The duration of incubation is indicated above each lane. Lanes 1 and 5 show control reactions in which mRNA was added to the reticulocyte lysate at 4°C and the mixture was diluted directly into buffer for the primer extension step. The amount of full-length mRNA in these control lanes is taken as 100%. The other lanes show diminished recovery of full-length mRNA, which was quantified (as cited in the text) by cutting and counting the band labeled 5[prime]-end.

The toeprinting assay enables one to monitor not only the formation of initiation complexes but also the stability of the input mRNA. Relative to the yield of full-length mRNA recovered from a control reaction which had been kept at 4°C (Fig. 5, lane 1), the amount of full-length mRNA detectable after 9 min incubation at 30°C in Promega's translation system was only 14% (Fig. 5, lane 4). This contrasts with 50% residual full-length mRNA after 9 min incubation at 30°C in Boehringer Mannheim's system (Fig. 5, lane 8). The rapid loss of full-length primer extension product has been seen with every batch of lysate obtained from Promega during the past 14 months. Inclusion of RNasin (RNase inhibitor from Promega, used at 1 U/µl) did not lessen the apparent degradation of mRNAs which could be slowed somewhat, however, by incubating at a lower temperature. For this reason, the ribosome binding step was carried out at 25°C in subsequent experiments. Given the ease with which Superscript II penetrates base paired structures (below), the reduction in full-length mRNA in Figure 5 is not likely to result from a conformational rearrangement in the mRNA that prevents primer extension. Other possible explanations are not ruled out, however.

If the loss of full-length primer extension product indeed reflects degradation of the mRNA, the rapid degradation might compromise experiments, such as tests for internal initiation of translation, that depend on the mRNA remaining intact. The facility with which uncapped mRNAs are translated in Promega's lysate (Fig. 6B, lane 3) adds to the concern that this system might sometimes produce artifactual results by generating and translating broken transcripts.ABC


Figure 6. Context-dependent recognition of AUG codons. An mRNA that contains two upstream AUG codons separated by 41 nt (Fig. 1, line 3) was incubated at 25°C in reticulocyte lysate from Boehringer Mannheim (A) or Promega (B and C). Except in lane 3 (A and B), all mRNAs were capped. An mRNA with a weak (W) context flanking AUG#1 (UCAaugG) was tested in lane 1 (A and B). This was changed to a strong (S) context (ACCaugG) in the mRNAs tested in lanes 2 and 3. Using capped mRNA that has a strong context flanking AUG#1, the effect of including NaF during the ribosome binding step in Promega's reticulocyte lysate is shown in (C). NaF was tested at concentrations of 12.5 (lane 2), 25 (lane 3) and 50 mM (lane 4).

Application of the toeprinting assay to test the fidelity of initiation

An earlier version of the toeprinting assay was used to investigate effects of flanking sequences on recognition of the AUG codon (12). To be sure of those results, the refined reaction conditions developed in the present study were used to retest some effects of context. The scanning model (19) predicts that, when the first AUG codon resides in a favorable context (ACCaugG; 27), 40S ribosomal subunits should stop scanning and 80S initiation complexes should assemble uniquely at that site. That prediction is upheld by the results in Figure 6A and B, lanes 2. In contrast, a matched mRNA in which the context around the first AUG codon was weakened (UCAaugG) allowed initiation at both the first and second AUG codons (Fig. 6A and B, lanes 1). Therefore, with these capped mRNAs, proper sensitivity to context is seen with translation systems from both Promega and Boehringer Mannheim.


Figure 7. A test of the processivity of scanning by 40S ribosomal subunits. The ribosome binding step using Promega's reticulocyte lysate was carried out with mRNAs in which the distance between AUG#1 and AUG#2 was varied from 11 to 251 nt. This was accomplished by repetition (or omission) of the AC-rich 30 nt insert depicted in line 3 of Figure 1. If 40S subunits dissociate in the course of scanning, the toeprint at AUG#2 should diminish in intensity as the distance to be scanned increases (lanes 1-5). No diminishment is seen at AUG#2, however. The incidental decline in the 5[prime]-end extension product (top-most band in each lane) and in the toeprint at AUG#1 (middle band) probably reflects incomplete extension by reverse transcriptase on the longer mRNAs, perhaps because the repetitious insert causes depletion of some dNTPs. This artifact would not affect the quantitation of initiation events at AUG#2 which, unlike AUG#1, is the same distance from the primer in all five mRNAs.

Using Promega's translation system and an mRNA in which the first AUG is in a strong context, the toeprint at AUG#1 was nearly as strong with uncapped as with capped mRNA (Fig. 6B, lanes 2 and 3). When the same uncapped mRNA was tested in Boehringer Mannheim's system, however, the formation of initiation complexes was barely detectable (Fig. 6A, lane 3). Thus Boehringer Mannheim's system better recapitulates in vitro the cap-dependent quality of translation in vivo (28-30). It is noteworthy that, when the uncapped mRNA was translated in Promega's system, initiation was still restricted to the first AUG codon (Fig. 6B, lane 3). This contradicts speculation that, in the absence of facilitation by the m7G cap, ribosomes have a general ability to initiate translation at internal sites (31). Our experiments consistently have shown that eukaryotic ribosomes are restricted to engaging mRNAs at the 5[prime]-end, even in the absence of a cap (32). (Note that the results in Fig. 6B do not rule out concern about artifactual initiation at internal AUG codons due to mRNA breakage in this system. That still might occur with other, more highly structured mRNAs that produce stable degradative intermediates.)

The usual restriction of initiation to the first AUG codon, when it is in a strong context, is abrogated by some chemical inhibitors, such as NaF. When 50 mM NaF was included during the ribosome binding step, for example, the toeprinting assay revealed ribosomes at AUG#1, AUG#2 and AUG#3 (Fig. 6C, lane 4). In the presence of NaF, Met-tRNA associated with 40S ribosomal subunits has been shown to be abnormally susceptible to deacylation (33). The implied abnormality in binding of Met-tRNA to 40S subunits might impair the AUG codon-anticodon interaction which underlies the stop-scanning step (34). Rapid joining of the 60S ribosomal subunit might be required to fix a paused 40S subunit at the first AUG codon, in which case the inhibition of 60S subunit joining by NaF (35) might also contribute to the leaky scanning seen in Figure 6C. The impaired fidelity of initiation caused by NaF mimics, albeit less dramatically, the perturbation caused by edeine (36).


Figure 8. A pseudoknot within the 5[prime]-UTR blocks access to the AUG codon. Ribosome binding was carried out with an mRNA that contains a pseudoknotted structure (PK, lanes 2, 5 and 6) or with a control mRNA that lacks this structure (lanes 1, 3 and 4). Incubation was for 8 min using reticulocyte lysate from Promega (lanes 1 and 2) or Boehringer Mannheim (lanes 3-6). The standard 25°C temperature was used with the Boehringer Mannheim lysate, but with the Promega lysate the temperature was lowered to 20°C in order to minimize breathing (opening) of the base paired structure. Cap analog (0.8 mM m7GDP) was included in the reactions shown in lanes 4 and 6. In lanes 1-6, ribosome binding was followed by extension with Superscript II, as usual. In lanes 7 and 8, the ribosome binding step was omitted. The primer was annealed to the pseudoknot-containing mRNA and extended using either Superscript II (lane 7) or AMV reverse transcriptase (lane 8). The primer extension pause sites marked PK 3[prime] edge (lane 8) and PK 5[prime] edge (lanes 2, 5 and 6) were mapped by running a second, longer gel which included a sequence ladder.



Figure 9. Glycerol gradient analysis of initiation complexes. Capped, 32P-labeled mRNAs were incubated in Boehringer Mannheim's translation system under the conditions used for toeprinting in Figure 8. The insert shows the structure of the pseudoknot-containing mRNA ([cir]) used here and in Figure 8. (The parenthetical 60 nt are an unstructured spacer sequence. Downstream from the BamHI site, the sequence was the same as that shown for T7-AUGpre in Fig. 1.) The control mRNA (l) used here and in Figure 8 contains point mutations that disrupt the pseudoknot. Another control mRNA in which the pseudoknot was moved close to the cap was unable to engage 40S or 80S ribosomes (data not shown). The 10-30% glycerol gradients shown here were centrifuged at 39 000 r.p.m. for 4 h at 4°C.

A test of the processivity of scanning

In Figure 7, the toeprinting assay was used to assess the ability of 40S ribosomal subunits to scan over long distances. In this experiment, the distance from the m7G cap to the first AUG codon was held constant while the distance between the first and second AUG codons (or the distance from the cap to the second AUG) was increased. This was accomplished by repeating a particular sequence (as explained in the legend to Fig. 7) so that the effect of distance could be assessed without continuously varying the sequence. In all five mRNAs tested in Figure 7, the first AUG codon lies in the same unfavorable context (UCAaugG), therefore, leaky scanning should allow the same percentage of 40S ribosomal subunits to bypass AUG#1 in all five mRNAs. The question was how many of these 40S subunits would reach AUG#2 as the intervening sequence was lengthened from 11 (Fig. 7, lane 1) to 251 nt (Fig. 7, lane 5). Judging from the constancy of the toeprint at AUG#2 in lanes 1-5, there was little if any dissociation of 40S ribosomal subunits in the course of scanning this sequence.

The toeprint at AUG#2 can be taken as a measure of processivity only if ribosomes indeed reach AUG#2 by scanning. An alternative explanation might be that lengthening the upstream sequence had no deleterious effect because ribosomes enter directly at AUG#2. But that possibility is ruled out by the control mRNA (Fig. 6B, lane 2) in which initiation at AUG#2 was precluded simply by changing three bases preceding AUG#1. This proves that access to AUG#2 depends on leaky scanning at AUG#1.

This first attempt to assess the processivity of 40S ribosomal subunit-factor complexes on mRNA suggests that distance per se, in the absence of secondary structure, is not an impediment during the scanning phase of initiation.

Trapping of 40S subunits upstream from the AUG codon

The scanning model predicts that a base paired structure inserted within the 5[prime]-UTR should block the advance of 40S ribosomal subunits (19). This was tested by introducing a pseudoknot upstream from AUGcat. With this mRNA, initiation complexes were studied using both the toeprinting assay (Fig. 8) and conventional glycerol gradient centrifugation (Fig. 9). The insert in Figure 9 shows the sequence of the pseudoknot, which was copied from the mRNA of infectious bronchitis virus. In the viral mRNA, this structure has been shown to slow the movement of 80S ribosomes under physiological conditions (37). Although scanning 40S ribosomal subunits have some ability to disrupt base paired structures (38), this ability is generally less than that of 80S elongating ribosomes.

Indeed, toeprinting assays revealed that a pseudoknot at the midpoint of the 5[prime]-UTR prevented ribosomes from reaching the AUG start codon (Fig. 8, lanes 2 and 5). Whereas the unstructured control mRNA shows the usual primer extension pause at AUGcat (Fig. 8, lanes 1 and 3), this was supplanted in lanes 2 and 5 by an upstream pause that mapped to the 5[prime] edge of the pseudoknot. Inasmuch as Superscript II is not stopped by the pseudoknot on protein-free mRNA (Fig. 8, lane 7), the upstream pause in lanes 2 and 5 reflects the binding of some component in the reticulocyte lysate. Were it simply proteins that bind to the base paired structure, primer extension would probably terminate at the 3[prime]-edge of the stabilized structure. But the pause in lanes 2 and 5 maps to the 5[prime]-edge of the pseudoknot, the point where a scanning 40S subunit would be expected to stop. The inhibition of the upstream pause by m7GDP (Fig. 8, lane 6) is consistent with attributing it to 40S or 80S ribosomes.

Although the toeprinting assay cannot distinguish between 40S and 80S ribosomes, analysis by glycerol gradient centrifugation (Fig. 9) showed clearly that the pseudoknot-containing mRNA binds only the expected 40S ribosomal subunit. This result confirms an earlier study in which a ribonuclease protection assay was used to map the approximate location of 40S subunits stalled behind a hairpin structure (39). By demonstrating the existence of an intermediate predicted by the scanning model, these experiments add to the growing evidence in support of the model.

REFERENCES

1. Chaudhuri,J., Si,K. and Maitra,U. (1997) J. Biol. Chem., 272, 7883-7891. MEDLINE Abstract

2. Das,S., Maiti,T., Das,K. and Maitra,U. (1997) J. Biol. Chem., 272, 31712-31718. MEDLINE Abstract

3. Farruggio,D., Chaudhuri,J., Maitra,U. and RajBhandary,U.L. (1996)Mol. Cell. Biol., 16, 4248-4256. MEDLINE Abstract

4. Clemens,M.J., Pain,V.M., Wong,S.-T. and Henshaw,E.C. (1982) Nature, 296, 93-95. MEDLINE Abstract

5. Price,N. and Proud,C. (1994) Biochimie, 76, 748-760. MEDLINE Abstract

6. Huang,H., Yoon,H., Hannig,E.M. and Donahue,T.F. (1997) Genes Dev., 11, 2396-2413. MEDLINE Abstract

7. Hartz,D., McPheeters,D.S., Traut,R. and Gold,L. (1988) Methods Enzymol., 164, 419-425. MEDLINE Abstract

8. Hartz,D., McPheeters,D.S. and Gold,L. (1989) Genes Dev., 3, 1899-1912. MEDLINE Abstract

9. Rinquist,S., Cunningham,P., Weitzmann,C., Formenoy,L., Pleij,C., Ofengand,J. and Gold,L. (1993) J. Mol. Biol., 234, 14-27.

10. Spedding,G., Gluick,T.C. and Draper,D.E. (1993) J. Mol. Biol., 229, 609-622. MEDLINE Abstract

11. Kozak,M. (1995) Proc. Natl Acad. Sci. USA, 92, 2662-2666(figures reprinted on p. 7134). MEDLINE Abstract

12. Kozak,M. (1997) EMBO J., 16, 2482-2492. MEDLINE Abstract

13. Anthony,D.D. and Merrick,W.C. (1992) J. Biol. Chem., 267, 1554-1562. MEDLINE Abstract

14. Cao,J. and Geballe,A.P. (1996) Mol. Cell. Biol., 16, 603-608. MEDLINE Abstract

15. Wang,Z. and Sachs,M.S. (1997) Mol. Cell. Biol., 17, 4904-4913. MEDLINE Abstract

16. Pestova,T.V., Hellen,C.U.T. and Shatsky,I.N. (1996) Mol. Cell. Biol., 16, 6859-6869. MEDLINE Abstract

17. Kozak,M. (1994) J. Mol. Biol., 235, 95-110. MEDLINE Abstract

18. Alton,N.K. and Vapnek,D. (1979) Nature, 282, 864-869. MEDLINE Abstract

19. Kozak,M. (1989) J. Cell Biol., 108, 229-241. MEDLINE Abstract

20. Kozak,M. (1977) Nature, 269, 390-394.

21. Minshull,J. and Hunt,T. (1986) Nucleic Acids Res., 14, 6433-6451. MEDLINE Abstract

22. Boiziau,C., Kurfurst,R., Cazenave,C., Roig,V., Thuong,N.T. and Toulmé,J.J. (1991) Nucleic Acids Res., 19, 1113-1119. MEDLINE Abstract

23. Edens,B., Thompson,H.A. and Moldave,K. (1975) Biochemistry, 14, 54-61. MEDLINE Abstract

24. Carrasco,L., Barbacid,M. and Vazquez,D. (1973) Biochim. Biophys. Acta, 312, 368-376. MEDLINE Abstract

25. Oen,H., Pellegrini,M. and Cantor,C.R. (1974) FEBS Lett., 45, 218-222. MEDLINE Abstract

26. Monro,R.E., Celma,M.L. and Vazquez,D. (1969) Nature, 222, 356-358. MEDLINE Abstract

27. Kozak,M. (1991) J. Biol. Chem., 266, 19867-19870. MEDLINE Abstract

28. Horikami,S.M., De Ferra,F. and Moyer,S.A. (1984) Virology, 138, 1-15. MEDLINE Abstract

29. Fuerst,T.R. and Moss,B. (1989) J. Mol. Biol., 206, 333-348. MEDLINE Abstract

30. Lo,H.-J., Huang,H.-K. and Donahue,T.F. (1998) Mol. Cell. Biol., 18, 665-675. MEDLINE Abstract

31. Svitkin,Y.V., Ovchinnikov,L.P., Dreyfuss,G. and Sonenberg,N. (1996) EMBO J., 15, 7147-7155. MEDLINE Abstract

32. Kozak,M. (1979) Nature, 280, 82-85.

33. O'Rourke,J.C. and Godchaux,W. (1975) J. Biol. Chem., 250, 3443-3450. MEDLINE Abstract

34. Cigan,A.M., Feng,L. and Donahue,T.F. (1988) Science, 242, 93-97. MEDLINE Abstract

35. Hoerz,W. and McCarty,K.S. (1971) Biochim. Biophys. Acta, 228, 526-535. MEDLINE Abstract

36. Kozak,M. and Shatkin,A.J. (1978) J. Biol. Chem., 253, 6568-6577. MEDLINE Abstract

37. Somogyi,P., Jenner,A.J., Brierley,I. and Inglis,S.C. (1993) Mol. Cell. Biol., 13, 6931-6940. MEDLINE Abstract

38. Kozak,M. (1986) Proc. Natl Acad. Sci. USA, 83, 2850-2854. MEDLINE Abstract

39. Kozak,M. (1989) Mol. Cell. Biol., 9, 5134-5142. MEDLINE Abstract

*Tel: +1 732 235 5355; Fax: +1 732 235 5356

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 17 Oct 1998
Copyright©Oxford University Press, 1998.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
S. Gao, Y. Zhao, L. Kong, P. Toselli, I.-N. Chou, P. Stone, and W. Li
Cloning and Characterization of the Rat Lysyl Oxidase Gene Promoter: IDENTIFICATION OF CORE PROMOTER ELEMENTS AND FUNCTIONAL NUCLEAR FACTOR I-BINDING SITES
J. Biol. Chem., August 31, 2007; 282(35): 25322 - 25337.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
K. Y. Song, C. K. Hwang, C. S. Kim, H. S. Choi, P.-Y. Law, L.-N. Wei, and H. H. Loh
Translational repression of mouse mu opioid receptor expression via leaky scanning
Nucleic Acids Res., March 12, 2007; 35(5): 1501 - 1513.
[Abstract] [Full Text] [PDF]

Home page
 Genes Dev.Home page
I. Ventoso, M. A. Sanz, S. Molina, J. J. Berlanga, L. Carrasco, and M. Esteban
Translational resistance of late alphavirus mRNA to eIF2{alpha} phosphorylation: a strategy to overcome the antiviral effect of protein kinase PKR
Genes & Dev., January 1, 2006; 20(1): 87 - 100.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
P. J. Talmud, J. Palmen, W. Putt, L. Lins, and S. E. Humphries
Determination of the Functionality of Common APOA5 Polymorphisms
J. Biol. Chem., August 5, 2005; 280(31): 28215 - 28220.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
P. S. Gould and A. J. Easton
Coupled Translation of the Respiratory Syncytial Virus M2 Open Reading Frames Requires Upstream Sequences
J. Biol. Chem., June 10, 2005; 280(23): 21972 - 21980.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
S. de Breyne, R. S. Monney, and J. Curran
Proteolytic Processing and Translation Initiation: TWO INDEPENDENT MECHANISMS FOR THE EXPRESSION OF THE SENDAI VIRUS Y PROTEINS
J. Biol. Chem., April 16, 2004; 279(16): 16571 - 16580.
[Abstract] [Full Text] [PDF]

Home page
 RNAHome page
E. P. PLANT, K. L. M. JACOBS, J. W. HARGER, A. MESKAUSKAS, J. L. JACOBS, J. L. BAXTER, A. N. PETROV, and J. D. DINMAN
The 9-A solution: How mRNA pseudoknots promote efficient programmed -1 ribosomal frameshifting
RNA, February 1, 2003; 9(2): 168 - 174.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
J. Y. Park, S. H. Park, J. E. Choi, S. Y. Lee, H.-S. Jeon, S. I. Cha, C. H. Kim, J.-H. Park, S. Kam, R. W. Park, et al.
Polymorphisms of the DNA Repair Gene Xeroderma Pigmentosum Group A and Risk of Primary Lung Cancer
Cancer Epidemiol. Biomarkers Prev., October 1, 2002; 11(10): 993 - 997.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
M. Kos, S. Denger, G. Reid, and F. Gannon
Upstream Open Reading Frames Regulate the Translation of the Multiple mRNA Variants of the Estrogen Receptor alpha
J. Biol. Chem., September 27, 2002; 277(40): 37131 - 37138.
[Abstract] [Full Text] [PDF]

Home page
 J. Virol.Home page
U. Chatterji, A. de Parseval, and J. H. Elder
Feline Immunodeficiency Virus OrfA Is Distinct from Other Lentivirus Transactivators
J. Virol., August 28, 2002; 76(19): 9624 - 9634.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
M. Kozak
Constraints on reinitiation of translation in mammals
Nucleic Acids Res., December 15, 2001; 29(24): 5226 - 5232.
[Abstract] [Full Text] [PDF]

Home page
 Plant Physiol.Home page
L. Wang and S. R. Wessler
Role of mRNA Secondary Structure in Translational Repression of the Maize Transcriptional Activator Lc
Plant Physiology, March 1, 2001; 125(3): 1380 - 1387.
[Abstract] [Full Text]

Home page
 J. Biol. Chem.Home page
Z. Wang, A. Gaba, and M. S. Sachs
A Highly Conserved Mechanism of Regulated Ribosome Stalling Mediated by Fungal Arginine Attenuator Peptides That Appears Independent of the Charging Status of Arginyl-tRNAs
J. Biol. Chem., December 31, 1999; 274(53): 37565 - 37574.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (156K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (37)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Kozak, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kozak, M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?