HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 278, Issue 20, 17672-17679, May 16, 2003

This Article Abstract Full Text (PDF) All Versions of this Article: 278/20/17672    most recent M212890200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stortchevoi, A. Articles by RajBhandary, U. L. Search for Related Content PubMed PubMed Citation Articles by Stortchevoi, A. Articles by RajBhandary, U. L. Social Bookmarking                      What's this?

Common Location of Determinants in Initiator Transfer RNAs for Initiator-Elongator Discrimination in Bacteria and in Eukaryotes^*

Alexei Stortchevoi, Umesh Varshney, and Uttam L. RajBhandary^§

From the Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139

Received for publication, December 18, 2002, and in revised form, March 10, 2003

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Initiator tRNAs are used exclusively for initiation of protein synthesis and not for elongation. We show that both Escherichia coli and eukaryotic initiator tRNAs have negative determinants, at the same positions, that block their activity in elongation. The primary negative determinant in E. coli initiator tRNA is the C1xA72 mismatch at the end of the acceptor stem. The primary negative determinant in eukaryotic initiator tRNAs is located in the TC stem, whereas a secondary negative determinant is the A1:U72 base pair at the end of the acceptor stem. Here we show that E. coli initiator tRNA also has a secondary negative determinant for elongation and that it is the U50·G64 wobble base pair, located at the same position in the TC stem as the primary negative determinant in eukaryotic initiator tRNAs. Mutation of the U50·G64 wobble base pair to C50:G64 or U50:A64 base pairs increases the in vivo amber suppressor activity of initiator tRNA mutants that have changes in the acceptor stem and in the anticodon sequence necessary for amber suppressor activity. Binding assays of the mutant aminoacyl-tRNAs carrying the C50 and A64 changes to the elongation factor EF-Tu·GTP show marginally higher affinity of the C50 and A64 mutant tRNAs and increased stability of the EF-Tu·GTP· aminoacyl-tRNA ternary complexes. Other results show a large effect of the amino acid attached to a tRNA, glutamine versus methionine, on the binding affinity toward EF-Tu·GTP and on the stability of the EF-Tu·GTP·aminoacyl-tRNA ternary complex.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A special methionine tRNA is used for the initiation of protein synthesis in all organisms studied (1-3). The initiator tRNA is used for the initiation of protein synthesis, whereas the elongator tRNA is used for the insertion of methionine into internal peptidic linkages. Because of their unique function, initiator tRNAs, tRNA^fMet in eubacteria and tRNA in eukaryotes, have several properties that are different from those of elongator tRNAs (3, 4). 1) Initiator tRNAs bind directly to the P site of the ribosome in a reaction facilitated by the initiation factors, IF2 in eubacteria and eIF2 in eukaryotes. In contrast, elongator tRNAs bind to the A site of the ribosome. 2) Initiator tRNAs are used exclusively for the initiation of protein synthesis. This is achieved by preventing the binding of initiator methionyl-tRNAs to the elongation factors (EF¹-Tu in eubacteria and eEF1 in eukaryotes), which carry aminoacyl-tRNAs to the A site on the ribosome (3). In eubacteria, such as Escherichia coli, the initiator Met-tRNA^fMet is formylated to formylmethionyl-tRNA^fMet (fMet-tRNA^fMet) by methionyl-tRNA formyltransferase (5, 6). The formyl group acts as a positive determinant for IF2 (7, 8). Concomitantly, it makes the fMet-tRNA^fMet an even worse substrate for EF-Tu (9), thereby ensuring sequestration of the initiator tRNA exclusively for initiation.

Initiator tRNAs also possess unique sequence and/or structural features that are absent in elongator tRNAs. One of the unique features of eubacterial initiator tRNA^fMet (Fig. 1) is the presence of a base pair mismatch between nucleotides 1 and 72 in the acceptor stem (C1xA72). Elongator tRNAs have Watson-Crick base pairs between these positions. It was shown earlier (10, 11) that this mismatch in the acceptor stem is a primary determinant for exclusion of tRNA^fMet from elongation. U1 and G72 mutations that generate a Watson-Crick base pair at this position allow the mutant tRNAs to bind to EF-Tu and act as elongators, whereas a U1G72 double mutation, which generates a U·G wobble base pair, does not (11, 12).

Eukaryotic initiator tRNAs also have highly conserved nucleotides at position 1 and 72. In contrast to the C1xA72 mismatch in the bacterial initiator tRNA, eukaryotic initiator tRNAs have a A1:U72 base pair. In vivo and in vitro work on eukaryotic initiator tRNAs has shown that the primary determinant for preventing activity of these tRNAs in elongation resides in the sequence/structure of the TC stem, with the A1:U72 base pair acting as a secondary determinant (13-17). For example, fungal and plant initiator tRNAs have a bulky 2'-O-phosphoribosyl modification on the ribose of nucleotide 64 in the TC stem (18-20). Removal of this bulky modification allows the initiator tRNA to bind to eEF1 and to act as an elongator (13-15). Vertebrate initiator tRNAs lack this bulky modification (21, 22). However, mutations of base pairs 50:64 and 51:63 allow the mutant tRNA to act as an elongator tRNA in vivo in mammalian cells suggesting that in the vertebrate initiator tRNAs it is the sequence-dependent perturbation of the TC stem that blocks the tRNA from acting as an elongator (17). Coupling of the 50:64 and 51:63 base pair mutations with A1:U72  G1:C72 mutation further increased the activity of the tRNA in elongation, suggesting that the rather weak A1:U72 base pair in the acceptor stem was a secondary determinant.

Another tRNA that does not bind to E. coli EF-Tu is the selenocysteinyl-tRNA^Sec, which binds to its own elongation factor encoded by the selB gene (23). In this tRNA also, the primary determinant, which prevents the tRNA from binding to EF-Tu, has been located in the TC stem (24).

Although the U1 and G72 mutants of E. coli tRNA^fMet are active in elongation, they are less active than the E. coli elongator methionine tRNA (tRNA^Met). This finding raised the question of whether there are other determinants in the E. coli tRNA^fMet which further limit its activity in elongation. In particular, in view of the results above with eukaryotic initiator tRNAs and tRNA^Sec, we have investigated whether the putative secondary determinant is located in the TC stem of the tRNA.

This paper describes studies on the role of the U50·G64 wobble base pair in the TC stem of E. coli tRNA^fMet. A wobble base pair at this site or in adjacent sites in the TC stem is found in most bacterial, mitochondrial, and chloroplast initiator tRNAs (25). (An update of this compilation can be found in www.uni-bayreuth.de/departments/biochemie/trna//.) We have mutated the U50·G64 wobble base pair to C50:G64 (C50 mutant) or to U50:A64 (A64 mutant) Watson-Crick base pairs and combined this to (i) mutations (U1, data not shown; G72 or G72G73) in the acceptor stem, which allow the tRNA to act in elongation; and (ii) mutations (U35A36) in the anticodon sequence, which allow the tRNAs to read the amber (UAG) termination codon. We show that mutations of the 50·64 wobble base pair to either C:G or U:A base pairs increases the activity in elongation of all of the mutant tRNAs in vivo. In addition, we have investigated whether this increase in elongation activity is due to increased affinity toward EF-Tu.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Mutant tRNA Genes-- The mutant tRNAs used in this work were all derived from tRNA. The tRNA genes were cloned into pRSVCATam1.2.5 vector (26) and expressed in E. coli B105, which lacks the tRNA species. For the sake of simplicity, the tRNA gene will be referred to as tRNA^fMet. The original DNA templates contained U35A36 and G72 or G72G73 mutations in the tRNA^fMet gene sequence. C50 and A64 mutations in TC stem of tRNA (Fig. 1) were introduced by QuickChange site-directed mutagenesis with the appropriate mutagenic primers, using Pfu DNA polymerase, according to Stratagene.

Purification of Enzymes and Proteins-- E. coli strain containing the pET24C(+) plasmid with the EF-Tu gene (27) was kindly provided by Dr. Linda Spremulli (University of North Carolina, Chapel Hill). Plasmid DNA was isolated from this strain and reintroduced into E. coli BL21DE3 strain for expression of His₆-tagged EF-Tu. EF-Tu was purified from a 1-liter culture of the BL21DE3 transformants grown in LB containing kanamycin (10 µg/ml) and induced for 3 h by the addition of 1 mM isopropyl--thiogalactoside. Wet bacterial pellet was suspended in EF-Tu lysis buffer (20 mM Tris-HCl, pH 8.0, 200 mM NaCl, 10 mM MgCl₂, 7 mM -mercaptoethanol, 50 µM GDP, 0.5 mM phenylmethylsulfonyl fluoride) at 1 g/15 ml, and the bacteria were lysed by freeze-thawing in the presence of 0.1 mg/ml of lysozyme. The lysate was treated with DNase I and clarified by centrifugation (12,000 × g) for 30 min at 4 °C. The supernatant was mixed with 2 ml of Talon-Sepharose CO²⁺ metal affinity resin (Clontech) and nutated for 30 min at 4 °C. The resin was allowed to settle and washed three times with 5 volumes of EF-Tu lysis buffer; the slurry was then transferred into a 10-ml disposable column, and the column was washed with 10 bed volumes of EF-Tu lysis buffer containing 10 mM imidazole, pH 7.5. EF-Tu was finally eluted with EF-Tu lysis buffer containing 30 mM imidazole and dialyzed three times each against 2 liters of EF-Tu storage buffer (20 mM Tris-HCl, pH 7.5, 10 mM MgCl₂, 5 mM EDTA, 0.2% sodium azide, 2 mM phenylmethylsulfonyl fluoride, 10 mM -mercaptoethanol, 10 µM GDP, and 10% glycerol). The dialyzed material was divided into 125-µl aliquots, quick-frozen in liquid nitrogen, and stored at 70 °C. Final EF-Tu concentration was 3 mg/ml.

Recombinant His₆-tagged glutaminyl-tRNA synthetase (GlnRS) was overproduced in E. coli M15 carrying the pQE60-QRS vector and purified from a 1-liter culture grown in LB containing 50 µg/ml carbenicillin using a Cobalt Talon column (5-ml bed volume), as described for EF-Tu except that the GlnRS lysis buffer contained 20 mM Tris-HCl, pH 7.5, 150 mM NaCl, and 8 mM -mercaptoethanol. The GlnRS was then dialyzed against GlnRS storage buffer (20 mM HEPES, pH 7.5, 100 mM KCl, 8 mM -mercaptoethanol), quick-frozen in 25-µl aliquots, and stored at 70 °C. Final concentration of GlnRS was 11 mg/ml. Aliquots were used only once after thawing, because of instability of the enzyme upon refreezing or upon storage at higher temperatures. Expression and purification of His₆-tagged methionyl-tRNA synthetase (MetRS) was as described earlier (28).

Regeneration of EF-Tu with GTP-- EF-Tu·GTP was regenerated from EF-Tu·GDP by incubation at 37 °C for 3 h in a solution containing 50 mM Tris-HCl, pH 7.4, 150 mM NH₄Cl, 10 mM MgCl₂, 3 mM phosphoenolpyruvate, 40 international units of pyruvate kinase, 200 µM GTP, and 5 mM dithiothreitol. EF-Tu·GDP concentration in the regeneration mixture was 2-4 µM. Regenerated EF-Tu·GTP was chilled on ice for 15 min and immediately used for K_D or k_off assay with aminoacyl-tRNAs labeled with [³H]Gln or [³⁵S]Met.

Purification of Wild Type and Mutant tRNA^fMet-- E. coli B105 transformants carrying the mutant tRNA^fMet genes were grown in 1 liter of 2YT medium containing 50 µg/ml ampicillin. Cells were suspended in a solution containing 10 mM Tris-HCl, pH 7.5, 10 mM MgCl₂, 10 mM EDTA at 1 g of wet pellet per 8 ml of buffer. The suspension was mixed vigorously for 10 min with an equal volume of water-saturated phenol, and the phases were separated by centrifugation. Nucleic acids in the aqueous layer were precipitated by addition of sodium acetate, pH 5.0, to 0.1 M followed by 2.5 volumes of 95% ethanol and storage for 2 h at 20 °C. The nucleic acid pellet was washed with 70% ethanol, air-dried, dissolved in ~10 ml of TE buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA), and applied to a DE52-DEAE-cellulose column (5-ml bed volume) equilibrated with 0.1 M Tris-HCl, pH 7.5. The column was washed with 20 bed volumes of washing buffer (0.1 M Tris-HCl, pH 7.5, 0.2 M NaCl) and eluted with 0.1 M Tris-HCl containing 1 M NaCl. Total tRNA from eluate was ethanol-precipitated, and the pellet was washed twice with ethanol and dissolved in TE buffer to a final concentration of 100-500 A₂₆₀/ml. Mutant or wild type tRNA^fMet was purified to homogeneity from the total tRNA by PAGE, as described (12).

Aminoacylation of tRNAs with [³H]Gln and [³⁵S]Met-- The reaction mixture (100 µl) contained 150 mM NH₄Cl, 10 mM MgCl₂, 10 µg/µl bovine serum albumin, 0.1 mM EDTA, 20 mM imidazole, pH 7.6, 2.5 mM ATP, 2.0 A₂₆₀ units of gel-purified tRNAs (wild type, G72, C50/G72, and A64/G72 mutant tRNA^fMet), 50 µM [³⁵S]Met (specific activity 12 mCi/µmol), and 5 µg of MetRS. Aminoacylation of the U35A36, U35A36/G72G73, U35A36/C50/G72G73, and the U35A36/A64/G72G73 mutant tRNAs with [³H]glutamine was carried out similarly except that the reaction volume was 300 µl and contained 5 mM dithiothreitol, 200 µM [³H]glutamine (specific activity, 3.35 mCi/µmol, ~7,370 dpm/pmol), and 55 µg of GlnRS instead of [³⁵S]methionine and MetRS. In both cases, aminoacylation was carried out at 37 °C for 15 min, and the reaction was stopped by cooling to 0 °C, adding 0.1 M sodium acetate, pH 5.0, and extracting with phenol equilibrated to pH 5.0 with sodium acetate. The aqueous layer was then extracted with phenol/chloroform (1:1, v/v) and finally chloroform, and the aminoacyl-tRNA in the aqueous layer was precipitated with ethanol and recovered by centrifugation. The pellet was washed twice with 70% ethanol at 4 °C, air-dried, and dissolved in 100 µl of ice-cold 5 mM sodium acetate, pH 5.0. The solution was centrifuged at 4 °C through Sephadex G-25 spin columns pre-equilibrated with 5 mM sodium acetate, pH 5.0, to separate aminoacyl-tRNA from ATP and free amino acids. Analysis of the extent of aminoacylation of tRNAs in vivo was carried out as described previously (26).

Assay for Suppression in Vivo-- E. coli CA274 transformants expressing the various mutant initiator tRNA genes were grown at 37 °C in 5 ml of LB medium containing 50 µg/ml ampicillin, until the A₆₀₀ reached 0.8-1.0. At this point, cultures were induced by adding 1 mM isopropyl--thiogalactoside and grown for an additional 2 h. After induction, the cultures were used for assay of -galactosidase activity (29) and for the preparation of bacterial extracts and isolation of total tRNA under acidic conditions (26, 30). Total tRNA was used for in vivo aminoacylation analysis, and bacterial extracts were used for assay of -lactamase (Calbiochem). Activities of -galactosidase in extracts were normalized to activities of -lactamase.

Assay for -Lactamase Activity-- Between 3 and 6 µg of protein in total extracts was taken in 0.9 ml of phosphate buffer Z (29), and the reaction was started by adding 100 µl of 0.5 mg/ml nitrocefin at room temperature. The reaction was terminated after 10 min by addition of 10% SDS (110 µl), and the absorbance was read at 486 nm.

Immunoblot Analysis of Bacterial Extracts with Anti--galactosidase and Anti--lactamase Antibodies-- Between 10 and 15 µg of protein in total extracts, normalized to -lactamase levels, were fractionated on an 8% polyacrylamide gel and transferred to an Immobilon-P membrane (Millipore) by electroblotting. The membrane was probed with polyclonal antibodies against -lactamase (5 Prime  3 Prime, Inc., Boulder, CO) and -galactosidase (ICN), using the guidelines provided by the suppliers. The working dilutions were 1:100,000 for anti--lactamase antibody and 1:20,000 for anti--galactosidase antibody. The immunoblot was probed with secondary horseradish peroxidase-conjugated goat anti-rabbit IgG (diluted 4,000-fold) and developed for 5 min with a 5-fold dilution of luminol/hydrogen peroxide mixture (1:1, Biolabs) with water, followed by a 15-s exposure to BioMax film (Eastman Kodak Co.).

Measurement of K_D for the EF-Tu·GTP·Aminoacyl-tRNA Ternary Complex-- We used a modification of the procedure described by LaRiviere et al. (31). Sequentially diluted samples of reconstituted EF-Tu·GTP (diluted from 1-2 µM to 1.95-3.9 nM in steps of 2-fold in buffer K: 50 mM Tris-HCl, pH 7.4, 10 mM MgCl₂, 150 mM NH₄Cl, 100 µM GTP, 0.5 mM phosphoenolpyruvate, 10 mM dithiothreitol, and 40 µg/ml pyruvate kinase) were mixed with an equal volume (25 µl) of ³H- or ³⁵S-labeled aminoacyl-tRNA (20-40 nM), and the mixture (50 µl) was incubated for 20 min on ice to allow equilibration between bound and free forms of aminoacyl-tRNA and EF-Tu·GTP. For each aminoacyl-tRNA, 12 such samples including no EF-Tu·GTP control were simultaneously treated with 5 µl of RNase A solution (Sigma, 0.2 mg/ml) for 30 s on ice. RNase treatment was stopped by addition of 5 µl of a 1 mg/ml solution of total E. coli tRNA (Sigma) followed by 25 µl of cold 20% trichloroacetic acid. The samples were filtered under vacuum through a sheet of Millipore membrane (HAWP, 0.45 µm) mounted on a Bio-Dot^TM apparatus (Bio-Rad). The membrane was washed 3 times with 200 µl each of 5% trichloroacetic acid, removed from the Bio-Dot^TM apparatus, immersed briefly (5 min) in ice-cold 95% ethanol, and air-dried. The membrane fragments corresponding to individual samples were cut out and placed into scintillation vials for counting of [³⁵S]Met or [³H]Gln radioactivity.

Measurements of Off Rates of the EF-Tu·GTP·Aminoacyl-tRNA Ternary Complexes-- The incubation mixture (100 µl) contained 4 µM EF-Tu·GTP and 40 nM aminoacyl-tRNA (³H- or ³⁵S-labeled) in buffer K. The mixture was incubated on ice for 20 min and then treated with 10 µl of RNase A (0.4 mg/ml, Sigma). Aliquots (10 µl) were taken out every 10 min (from 0 to 90 min) and immediately transferred to the wells, on a 96-well assay plate, containing 100 µl of pre-chilled 10% trichloroacetic acid with 2% casamino acids. The precipitated radioactivity was collected by filtration under vacuum as described above and counted.

Development of Experimental Data-- For calculation of K_D, EF-Tu concentrations were plotted as x axis (in nM), and the amounts of ternary complex (in nM) (calculated from counts/min of [³H]Gln or [³⁵S]Met in corresponding samples), as y axis. K_D values were calculated with the help of Kaleidagraph application program (Synergy software), from the curve fit based on Equation 1,

(Eq. 1)

For k_off measurements, time (in seconds) was plotted as x axis, and ln(tRNA/tRNA_input), calculated based on counts/min of the samples, was plotted as y axis. k_off values were then calculated using the "linear curve fit" option of Kaleidagraph, using Equation 2,

(Eq. 2)

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Mutant tRNAs Used-- Fig. 1 shows the mutant tRNAs used in this work. To examine the possible role of U50·G64 wobble base pair as a secondary determinant for restricting the activity of the E. coli initiator tRNA in elongation, this base pair was mutated to either a C50:G64 or U50:A64 base pair, and these mutations were combined with mutations in the acceptor stem and the anticodon sequence. The G72 mutation in the acceptor stem allows the E. coli initiator tRNA to act as an elongator (10), and the U35A36 mutation in the anticodon sequence allows the tRNA to read the amber termination codon UAG. As a result the U35A36/G72 mutant tRNA is an amber suppressor in E. coli. Because of the anticodon sequence change, the U35A36 and U35A36/G72 mutant tRNAs are now aminoacylated with glutamine instead of methionine (32, 33). However, the G72 mutation in the acceptor stem makes the U35A36/G72 mutant tRNA a poorer substrate for the E. coli glutaminyl-tRNA synthetase (GlnRS) than the U35A36 mutant tRNA (11, 34). Therefore, for some experiments, the G72 mutation was also combined with G73 mutation to make the mutant tRNAs better substrates for the E. coli GlnRS.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Cloverleaf structure of E. coli tRNAfMet and mutants used in this work. Arrows indicate the sites of mutations. The C50 or A64 mutations were introduced into tRNAs with a U35A36/G72 or U35A36/G72G73 mutant background (A) or a G72 mutant background (B). The first set of mutant tRNAs is aminoacylated with glutamine, and the second set is aminoacylated with methionine.

Activity of the Mutant tRNAs in Elongation in E. coli-- The activity of the mutant tRNAs in elongation in vivo was measured by their activities in suppression of an amber mutation in the -galactosidase gene. E. coli CA274 (HfrC lacZam trpEam) was transformed with the pRSV plasmid carrying the mutant tRNA genes, and cell extracts were used for measurement of -galactosidase activity. To correct for any possible fluctuations in plasmid copy number, the -galactosidase activities were normalized to the -lactamase activity encoded in the same pRSV plasmid. As shown before, the U35A36 mutant initiator tRNA is completely inactive as an elongator, whereas the U35A36/G72 and the U35A36/G72G73 mutants are active in elongation (Table I). Also, the activity in elongation of the U35A36/G72G73 series of mutants is higher than that of the corresponding mutants in the U35A36/G72 background. Introduction of C50 or A64 mutations in either the U35A36/G72 or the U35A36/G72G73 mutant backgrounds substantially increases -galactosidase activities. The same results were obtained in a U1/U35A36 mutant background (data not shown). These results suggest that the U50·G64 wobble base pair restricts the activity of the E. coli initiator tRNA in elongation. Interestingly, the A64 mutants with a weaker U50:A64 base pair consistently gave a higher activity in elongation (~5-fold) than the C50 mutants with the stronger C50:G64 base pair (~3-fold).

Immunoblot analyses on cell extracts confirm the above results and show that the increased activity of -galactosidase in cell extracts is due to increased synthesis of full-length -galactosidase protein. Total proteins in cell extracts were fractionated on an SDS-8% polyacrylamide gel, transferred to Immobilon membrane, and probed with anti--lactamase and anti--galactosidase polyclonal antibodies (Fig. 2). Although the levels of -lactamase, encoded in the same pRSV plasmid as the mutant tRNA genes, are, as expected, approximately the same (lanes 1-7), there is more of the full-length -galactosidase in some of the extracts compared with others, and the relative intensity of the bands corresponds to the relative levels of -galactosidase in cell extracts.

View larger version (92K): [in this window] [in a new window]   Fig. 2.   Immunoblot analysis, using rabbit anti--lactamase and anti--galactosidase polyclonal antibodies, of crude E. coli extracts transformed with mutant tRNAs shown in Fig. 1A. The truncated -galactosidase fragment is produced by chain termination at the site of the amber mutation. The intensity of the band corresponding to the full-length -galactosidase reflects the extent of suppression. As expected, this band and the band corresponding to -lactamase are absent in extracts from cells that do not carry the pRSVCAT vector with the mutant tRNA genes (lane 8).

It is extremely unlikely that the increased activity of the mutant tRNAs carrying the C50 and A64 changes is due to an effect on formylation of the tRNAs. First, biochemical (35) and structural studies (36) have shown that interactions of methionyl-tRNA formyltransferase (MTF) with the initiator tRNA are restricted to the acceptor stem and the D stem. Second, the U35A36/G72 and the U35A36/G72G73 mutant tRNAs are extremely poor substrates for MTF, and there is no detectable formylation of these tRNAs in E. coli (30, 37).

Extent of Aminoacylation of Mutant tRNAs in E. coli-- One possible explanation of the increased activity in elongation of tRNAs carrying the C50 or A64 mutations is increased aminoacylation of these mutant tRNAs by GlnRS. To investigate this possibility, total tRNA isolated under acidic conditions from E. coli CA274 transformants expressing the various mutant tRNAs was separated on an acid-urea polyacrylamide gel, and the uncharged and aminoacylated forms of the mutant tRNAs were detected by RNA blot hybridization using a deoxyribo-oligonucleotide probe complementary to sequences in the anticodon stem and loop common to all of the mutant tRNAs (Fig. 3). A probe for the endogenous tyrosine tRNA (tRNA^Tyr) was used as an internal control. The amounts of uncharged and aminoacylated forms of the mutant tRNAs (bands A and B of Fig. 3) were quantified using a PhosphorImager, and the pixels were used to calculate the percent of each mutant tRNA that is aminoacylated in the cell. The amount of the aminoacylated form of each of the mutant tRNAs was then normalized to the amount of total tRNA^Tyr present in each sample (bands C and D of Fig. 3). It can be seen (Table II) that within one set of mutants, the U35A36/G72 or the U35A36/G72G73, introduction of C50 or A64 mutations had little effect on the extent of aminoacylation of the tRNA and on the relative content of the mutant aminoacyl-tRNAs in the cell. As expected, the extent of aminoacylation of tRNAs carrying the U35A36/G72G73 mutation is higher than that of tRNAs carrying the U35A36/G72 mutation. These results suggest that the increased activity in elongation of the tRNAs carrying the C50 or the A64 mutations is due to their increased activity at a step following aminoacylation of the tRNA. This could be due either to an increased affinity of the mutant aminoacyl-tRNAs for EF-Tu or for the ribosomal A site.

View larger version (57K): [in this window] [in a new window]   Fig. 3.   Northern blot analysis of total RNA isolated under acidic conditions from E. coli transformed with plasmids carrying the various mutant tRNAfMet genes. The blot was probed with 32P-labeled oligonucleotides complementary to the anticodon stem-loop region of the U35A36 mutant tRNAfMet and wild type tyrosine tRNA (tRNATyr). Bands A and B correspond, respectively, to the uncharged and aminoacylated forms of the mutant tRNAs, and bands C and D correspond, respectively, to uncharged and aminoacylated forms of tRNATyr.

Binding Affinity of EF-Tu·GTP for the Mutant Aminoacyl-tRNA^fMet-- Two sets of mutant tRNAs were used for this work. One set contains the G72 and G73 mutations in the acceptor stem and the U35A36 mutation in the anticodon sequence (Fig. 1A). These mutant tRNAs were aminoacylated in vitro with [³H]glutamine using E. coli GlnRS. For maximal aminoacylation of the mutant tRNAs with E. coli GlnRS, the tRNAs with the G72G73 mutations were used rather than the ones with the G72 mutation alone (11). The other set of mutant tRNAs used contained just the G72 mutation in the acceptor stem (Fig. 1B). These tRNAs were aminoacylated in vitro with [³⁵S]methionine using E. coli MetRS. The G72G73 mutant tRNA could not be used because this tRNA is not aminoacylated by E. coli MetRS (38).

A ribonuclease protection assay was used to measure K, the equilibrium dissociation constant of the binding of EF-Tu·GTP to the various mutant aminoacyl-tRNAs. This assay relies on the finding (39) that free aminoacyl-tRNA is cleaved into short trichloroacetic acid-soluble fragments by RNase A, whereas the acceptor stem and the TC stem of the tRNA is protected from cleavage when the aminoacyl-tRNA is in the form of a ternary complex with EF-Tu·GTP. For the formation of the ternary complex, EF-Tu·GTP and the mutant aminoacyl-tRNAs, labeled with ³H- or ³⁵S-labeled amino acids, were mixed with various concentrations of His-tagged E. coli EF-Tu·GTP (serially diluted 2-fold from 1-2 µM down to 1.95-3.9 nM), and the mixture was left on ice for 20 min to allow complex formation to reach equilibrium. The amount of aminoacyl-tRNA in the ternary complex was determined by measuring the amount of ³H or ³⁵S radioactivity remaining insoluble in 10% trichloroacetic acid after a short treatment (30 s) with an excess of RNase A. The equilibrium binding curves were then used to obtain K values using a Kaleidagraph application program (see "Experimental Procedures") (40). Fig. 4, A and B, shows representative equilibrium binding curves for the two sets of mutant aminoacyl-tRNAs. Table III shows the mean K values based on five to seven separate experiments on different batches of EF-Tu and ³H- and ³⁵S-labeled aminoacyl-tRNAs. With both sets of mutant tRNAs, replacement of the U50·G64 wobble base pair by a C50:G64 or U50:A64 Watson-Crick base pair results in only a marginal increase in binding affinity. Perhaps the most striking result is that in contrast to the wild type E. coli [³⁵S]Met-tRNA^fMet, which binds extremely poorly to EF-Tu·GTP (Fig. 4B and Table III), the U35A36 mutant, which is aminoacylated with [³H]glutamine, binds quite well with a K of ~27 nM (Fig. 4A and Table III). Because the anticodon sequence of tRNAs is not involved in any of the interactions with EF-Tu·GTP, this result shows that the amino acid attached to the tRNA has a significant effect on its affinity for EF-Tu·GTP (9). This result provides further support to the conclusions of Uhlenbeck and co-workers (31) on the important role of the amino acid in binding of aminoacyl-tRNAs to EF-Tu·GTP, with glutamine being perhaps the most tightly binding amino acid.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Equilibrium binding curves of EF-Tu·GTP with the various mutant aminoacyl-tRNAs. A, tRNAs aminoacylated with glutamine; B, tRNAs aminoacylated with methionine. K values were determined using a Kaleidagraph application as described under "Experimental Procedures."

Stability of EF-Tu·GTP·Aminoacyl-tRNA Ternary Complexes-- In an extensive analysis of the binding of several aminoacyl-tRNAs to EF-Tu·GTP, Uhlenbeck and co-workers (31) have noted a strong correlation between the dissociation rate constants (k_off) of the EF-Tu·GTP·aminoacyl-tRNA ternary complexes and the K_D values. This suggests that k_on, the association rate constant, is the same for many of the aminoacyl-tRNAs. Therefore, the k_off values of the ternary complexes could be equally useful parameters in comparing the affinity of EF-Tu·GTP for a series of related mutant aminoacyl-tRNAs. Furthermore, although the measurement of K_D values is sensitive to errors in estimate of the amount of active EF-Tu·GTP in a preparation and the fraction of aminoacyl-tRNA that is active in binding to EF-Tu, measurement of k_off does not suffer from the limitations. Additionally, measurement of k_off values allows one to determine whether the extremely poor binding of wild type E. coli Met-tRNA^fMet to EF-Tu·GTP (Fig. 4B) is due to a very low association rate constant of binding or due to extreme instability of the ternary complex. We have, therefore, measured the k_off of EF-Tu·GTP complexes with the various mutant [³H]Gln- and [³⁵S]methionyl-tRNAs to obtain an independent and possibly a more accurate estimate of the binding affinities of the various aminoacyl-tRNAs toward EF-Tu·GTP.

Wild type and mutant aminoacyl-tRNAs (40 nM) were mixed with an excess of E. coli EF-Tu·GTP at a high concentration (4 µM) such that essentially all of the aminoacyl-tRNAs were in the form of a ternary complex. RNase A was added to the incubation mixture, and aliquots were taken every 10 min (except for Met-tRNA^fMet in which case it was every 5 s) for the measurement of residual trichloroacetic acid-precipitable radioactivity. Dissociation of the ternary complex renders the aminoacyl-tRNAs susceptible to RNase A cleavage. Therefore, the time-dependent loss of trichloroacetic acid-precipitable radioactivity can be used to measure k_off, the dissociation rate constant of the ternary complex (see "Experimental Procedures").

Fig. 5, A and B, shows representative examples of the dissociation rates of the amino acyl~tRNA·EF-Tu·GTP ternary complexes for the two sets of mutant tRNAs. Table IV lists the mean k_off values. With the U35A36 mutant tRNA that is aminoacylated with glutamine, introduction of the G72G73 mutation increases the stability of the ternary complex by about 3.5-fold. Introduction of additional mutations in the U50·G64 wobble base pair leads to further stabilization of the complex by an additional factor of 2.4-2.6-fold. With the wild type initiator tRNA that is aminoacylated with methionine, and that binds extremely poorly to EF-Tu·GTP, introduction of the G72 mutation leads to a large increase in stability of the ternary complex, by a factor of ~45-fold. Introduction of additional mutations in the U50·G64 wobble base pair increases further the stability of the ternary complex, although less than that for the corresponding mutant tRNAs that are aminoacylated with glutamine. As with the K values, comparison of the k_off values of the ternary complex formed with the wild type Met-tRNA^fMet and the U35A36 mutant Gln-tRNA^fMet highlights once more the striking effect of the amino acid attached to the tRNA on binding of the aminoacyl-tRNAs to EF-Tu·GTP.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Kinetics of dissociation of EF-Tu·GTP·aminoacyl-tRNA ternary complex. A, tRNA aminoacylated with glutamine; B, tRNA aminoacylated with methionine. Fraction of tRNA, trichloroacetic acid-precipitable radioactivity remaining after treatment for various times with RNase A divided by the trichloroacetic acid-precipitable radioactivity at time 0. Aliquots were taken out every 10 min except for the wild type (W.T.) tRNAfMet (B) in which case it was every 5 s.

The very rapid hydrolysis of wild type Met-tRNA^fMet by RNase A (Fig. 5B) led us to investigate whether this aminoacyl-tRNA forms a ternary complex at all with EF-Tu·GTP or whether the time-dependent hydrolysis simply reflects the rate of RNase A cleavage of unbound Met-tRNA^fMet. To distinguish among these possibilities, Met-tRNA^fMet was treated with RNase A in the presence or absence of EF-Tu·GTP. Results in Fig. 6 show that EF-Tu·GTP protects the Met-tRNA^fMet from RNase A cleavage. In the absence of EF-Tu·GTP, RNase A cleavage of Met-tRNA^fMet is extremely fast. Thus, the Met-tRNA^fMet does form a complex with EF-Tu·GTP; however, the complex is extremely unstable with a half-life of ~15 s. This extreme instability of the EF-Tu·GTP·Met-tRNA^fMet ternary complex explains our inability to detect the formation of such a complex in the equilibrium binding curves used to measure K_D values (Fig. 4B). The 30-s treatment with RNase A used to hydrolyze all of the unbound Met-tRNA^fMet would have hydrolyzed more than 75% of the Met-tRNA in the extremely unstable ternary complex.

View larger version (12K): [in this window] [in a new window]   Fig. 6.   Time-dependent cleavage by RNase A of Met-tRNAfMet, either free (EF-Tu·GTP) or in the form of a ternary complex (+EF-Tu·GTP). For the formation of the ternary complex, the Met-tRNAfMet was incubated with EF-Tu·GTP on ice for 20 min. The fraction of Met-tRNA, [35S]Met radioactivity remaining in trichloroacetic acid-precipitable form after treatment for various times with RNase A was divided by the trichloroacetic acid precipitable radioactivity at time 0.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The primary negative determinant blocking activity of E. coli initiator tRNA^fMet in elongation is the C1xA72 mismatch at the end of the acceptor stem. This is due to perturbation of the RNA helical structure caused by the C1xA72 mismatch (10). Here we have shown that the E. coli initiator tRNA has a secondary negative determinant, the 50·64 wobble base pair, that is located in the TC stem. Interestingly, eukaryotic initiator tRNAs also have negative determinants at these sites except that the primary negative determinant is located in the TC stem and the secondary negative determinant is the A1:U72 base pair, which is unique to and is highly conserved in eukaryotic initiator tRNAs (17). Plant and fungal initiator tRNAs have a bulky 2'-O-phosphoribosyl modification attached to the ribose of nucleotide 64. This bulky modification protrudes into the minor groove of the TC stem and most likely acts as a steric block for the binding of eEF1 (13-15, 41, 42). In vertebrate initiator tRNAs that lack the bulky modification, the 50:64 and possibly 51:63 base pairs act as negative determinants most likely by perturbing the sugar phosphate backbone in the TC stem (17). Thus, the negative determinants for elongation are located at the same sites in bacterial and in eukaryotic initiator tRNAs. Fig. 7 shows the location of these negative determinants in the initiator tRNAs in a ribbon diagram of the three-dimensional structure of tRNA.

View larger version (33K): [in this window] [in a new window]   Fig. 7.   Ribbon diagram of three-dimensional structure of tRNA highlighting the negative determinants for elongation in E. coli and in eukaryotic initiator tRNAs.

Does the secondary negative determinant in the TC stem of E. coli initiator tRNA affect elongation activity of the tRNA by modulating the binding affinity of the aminoacyl-tRNA to EF-Tu·GTP? Binding assays show that mutation of the U50·G64 wobble base pair to C50:G64 or U50:A64 increases only marginally the binding affinity of the aminoacyl-tRNAs to EF-Tu·GTP. The k_off values of the ternary complexes formed between EF-Tu·GTP and the mutant aminoacyl-tRNAs suggest a more significant increase, 2.4-2.6-fold, in stability of the ternary complexes formed with the C50 and A64 mutant aminoacyl-tRNAs (Table IV, top half). Whether these small improvements in binding affinity for EF-Tu·GTP and the stabilities of the EF-Tu·GTP·aminoacyl-tRNA ternary complexes are sufficient to account for the increased activities of these mutant tRNAs as amber suppressors in vivo (Table I) or whether the increased activity is also due to a better accommodation of the A64 and C50 mutant tRNAs into the ribosomal A site is not known. It is important to note, however, that 2-3-fold increases in vitro in EF-Tu·GTP binding affinity of the Su⁺7 tRNA aminoacylated with glutamine over that aminoacylated with tryptophan (43) results in selection in vivo of glutamine over tryptophan by a factor of 9 (44, 45).

A striking result of our work is the very large effect of the amino acid attached to the tRNA on EF-Tu binding affinity of wild type and mutant initiator tRNAs. The binding of the wild type tRNA aminoacylated with methionine is so unstable that it was not possible to obtain an accurate estimate of K (Fig. 4B and Table III). Binding assays carried out in 50 mM NH₄Cl instead of 150 mM NH₄Cl gave essentially the same result. In contrast, binding of the U35A36 mutant tRNA aminoacylated with glutamine was fairly strong with a K of 27 nM. Similarly, whereas the EF-Tu·GTP·Met-tRNA^fMet ternary complex had a dissociation rate constant of 453 × 10⁴ s¹ (Table IV) corresponding to a half-life of 15 s, the ternary complex formed with the U35A36 mutant Gln-tRNA^fMet had a dissociation rate constant of 6.21 × 10⁴ s¹ corresponding to a half-life of 17 min. Because the only difference between these two tRNAs is in the anticodon sequence and EF-Tu is known not to interact with the anticodon (9), the difference in binding affinities and stability of the ternary complexes is due to the different amino acids attached to the tRNA. Knowlton and Yarus (43) were the first to show that Su⁺7 tRNA aminoacylated with glutamine bound to EF-Tu·GTP with a K 2-3-fold lower than the same tRNA aminoacylated with tryptophan. More recently, in a thorough analysis, Uhlenbeck and co-workers (31, 46) have demonstrated a significant effect of the amino acid attached to a tRNA on its binding affinity for EF-Tu·GTP and have observed 60-150-fold differences in K_D values for the same tRNAs aminoacylated with glutamine, phenylalanine, valine, and alanine, with glutamine being the tightest binding amino acid. Our findings and those of Uhlenbeck and co-workers are also consistent with the known effects of amino acids attached to a tRNA on its interaction with other aminoacyl-tRNA or peptidyl-tRNA-binding proteins or enzymes such as MTF (11, 47-50), IF2 (51), eIF2 (52, 53), SelB (23), and peptidyl-tRNA hydrolase (54). Thus, proteins that bind to or utilize aminoacyl-tRNAs or formylaminoacyl-tRNAs as substrates are quite sensitive to the nature of the amino acid attached to the tRNA.

The binding (K, 27 nM) of the U35A36 mutant Gln-tRNA^fMet to EF-Tu·GTP is slightly stronger than that of the G72 mutant Met-tRNA^fMet (K, 34.8 nM) (Table III), yet the U35A36 mutant tRNA^fMet is essentially inactive as an elongator tRNA in vivo (Table I), whereas the G72 mutant is active as an elongator in vitro (10). The reason for it is that despite its relatively strong affinity for EF-Tu, the U35A36 mutant tRNA is a good substrate for MTF and therefore is formylated in vivo to fGln-tRNA. Because the amino group of glutamine is blocked in fGln-tRNA, this tRNA now binds very poorly to EF-Tu and cannot, in any case, participate in the elongation reaction on the ribosome because of the blocked amino group. In strains severely deficient in MTF, however, the U35A36 mutant initiator tRNA does act weakly as an elongator tRNA (55). Similarly, mutant tRNAs that are extremely poor substrates for MTF can act as elongators in vivo, albeit weakly, even though they contain a C1xA72 mismatch (56). These results highlight yet another important role of formylation of the initiator tRNA in ensuring sequestration of the tRNA for initiation.

The strong effect of methionine in modulating the binding affinity of Met-tRNA^fMet for EF-Tu highlights yet another important reason why methionine is the best amino acid for the initiation of protein synthesis in E. coli. Initiation requires formylation of the initiator Met-tRNA to fMet-tRNA by MTF. The formyl group then acts as a positive determinant for IF2. We and others (47-49) have shown that MTF prefers methionine over all the other amino acids tested. Similarly, IF2 also prefers formylmethionine over several other formyl amino acids tested (51). Our finding that EF-Tu binds extremely poorly to the wild type Met-tRNA^fMet but binds quite well to the U35A36 mutant Gln-tRNA^fMet provides yet another role for methionine in minimizing the affinity of the initiator Met-tRNA^fMet for EF-Tu. This allows MTF, which is present in extremely small amounts in E. coli (57), to compete for Met-tRNA^fMet with EF-Tu, which is the most abundant protein in E. coli (58). Finally, although fMet-tRNA^fMet is not a substrate for peptidyl-tRNA hydrolase (59), Varshney and co-workers (54) have shown that the U35A36 mutant fGln-tRNA^fMet is a substrate for peptidyl-tRNA hydrolase. Thus, methionine plays yet another important role in ensuring the availability of fMet-tRNA for initiation by preventing its hydrolysis by peptidyl-tRNA hydrolase.

	ACKNOWLEDGEMENTS

We thank Dr. Linda Spremulli for the strain for overproducing EF-Tu, Dr. Alexey Wolfson for advice on assays for EF-Tu binding, and Dr. Caroline Köhrer for help in the design of figures for electronic submission. We thank Annmarie McInnis for the continued enthusiasm and for patience and care in the preparation of this manuscript.

	FOOTNOTES

^* This work was supported by Grant R37GM17151 from the National Institutes of Health.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Present address: Dept. of Microbiology and Cell Biology, Indian Institute of Science, Bangalore 560012, India.

^§ To whom correspondence should be addressed: Dept. of Biology, Rm. 68-671, Massachusetts Institute of Technology, Cambridge, MA 02139. Tel.: 617-253-4702; Fax: 617-252-1556; E-mail: bhandary@mit.edu.

Published, JBC Papers in Press, March 13, 2003, DOI 10.1074/jbc.M212890200

	ABBREVIATIONS

The abbreviations used are: EF, elongation factors; GlnRS, glutaminyl-tRNA synthetase; MetRS, methionyl-tRNA synthetase; MTF, methionyl-tRNA formyltransferase.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				H. Roy and M. Ibba RNA-dependent lipid remodeling by bacterial multiple peptide resistance factors PNAS, March 25, 2008; 105(12): 4667 - 4672. [Abstract] [Full Text] [PDF]

				M. Ibba and D. Soll Aminoacyl-tRNAs: setting the limits of the genetic code Genes & Dev., April 1, 2004; 18(7): 731 - 738. [Full Text] [PDF]

				N. Hino, T. Suzuki, T. Yasukawa, K. Seio, K. Watanabe, and T. Ueda The pathogenic A4269G mutation in human mitochondrial tRNAIle alters the T-stem structure and decreases the binding affinity for elongation factor Tu Genes Cells, March 1, 2004; 9(3): 243 - 252. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 278/20/17672    most recent M212890200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stortchevoi, A. Articles by RajBhandary, U. L. Search for Related Content PubMed PubMed Citation Articles by Stortchevoi, A. Articles by RajBhandary, U. L. Social Bookmarking                      What's this?