lambda bar minigene-mediated inhibition of protein synthesis involves accumulation of peptidyl-tRNA and starvation for tRNA
1998; Springer Nature; Volume: 17; Issue: 13 Linguagem: Inglês
10.1093/emboj/17.13.3758
ISSN1460-2075
AutoresJavier Hernández‐Sánchez, J. Gerardo Valadez, Jesús Vega Herrera, Carlos Ontiveros, Gabriel Guarneros,
Tópico(s)Ubiquitin and proteasome pathways
ResumoArticle1 July 1998free access λ bar minigene-mediated inhibition of protein synthesis involves accumulation of peptidyl-tRNA and starvation for tRNA Javier Hernández-Sánchez Javier Hernández-Sánchez Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Search for more papers by this author Juan Gerardo Valadez Juan Gerardo Valadez Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Present address: Departamento de Biotecnología y Bioingeniería, CINVESTAV-IPN, Mexico City, Mexico Search for more papers by this author Jesús Vega Herrera Jesús Vega Herrera Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Search for more papers by this author Carlos Ontiveros Carlos Ontiveros Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Search for more papers by this author Gabriel Guarneros Corresponding Author Gabriel Guarneros Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Departamento de Genética Molecular, Centro de Investigación sobre Fijación de Nitrógeno, UNAM, Apartado Postal 565-A, Cuernavaca, Mexico Present address: Departamento de Genética y Biología Molecular, CINVESTAV-IPN, Mexico City, Mexico Search for more papers by this author Javier Hernández-Sánchez Javier Hernández-Sánchez Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Search for more papers by this author Juan Gerardo Valadez Juan Gerardo Valadez Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Present address: Departamento de Biotecnología y Bioingeniería, CINVESTAV-IPN, Mexico City, Mexico Search for more papers by this author Jesús Vega Herrera Jesús Vega Herrera Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Search for more papers by this author Carlos Ontiveros Carlos Ontiveros Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Search for more papers by this author Gabriel Guarneros Corresponding Author Gabriel Guarneros Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico Departamento de Genética Molecular, Centro de Investigación sobre Fijación de Nitrógeno, UNAM, Apartado Postal 565-A, Cuernavaca, Mexico Present address: Departamento de Genética y Biología Molecular, CINVESTAV-IPN, Mexico City, Mexico Search for more papers by this author Author Information Javier Hernández-Sánchez1, Juan Gerardo Valadez1,2, Jesús Vega Herrera1, Carlos Ontiveros1 and Gabriel Guarneros 1,3,4 1Departamento de Genética y Biología Molecular, Centro de Investigación y de Estudios Avanzados del IPN, Apartado Postal 14-740, Mexico City, 07000 Mexico 2Present address: Departamento de Biotecnología y Bioingeniería, CINVESTAV-IPN, Mexico City, Mexico 3Departamento de Genética Molecular, Centro de Investigación sobre Fijación de Nitrógeno, UNAM, Apartado Postal 565-A, Cuernavaca, Mexico 4Present address: Departamento de Genética y Biología Molecular, CINVESTAV-IPN, Mexico City, Mexico *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:3758-3765https://doi.org/10.1093/emboj/17.13.3758 Deceased October, 1996 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Expression of the bacteriophage λ two-codon, AUG AUA, barI minigene (bar+) leads to the arrest of protein synthesis in cells defective in peptidyl-tRNA hydrolase (Pth). It has been hypothesized that translation of the bar+ transcript provokes premature release and accumulation of peptidyl-tRNA (p-tRNA). Inhibition of protein synthesis would then result from either starvation of sequestered tRNA or from toxicity of accumulated p-tRNA. To test this hypothesis and to investigate the cause of arrest, we used a coupled in vitro transcription–translation system primed with DNA containing bar+ and the β-lactamase-encoding gene of the vector as a reporter. The results show that expression of bar+ minigene severely inhibits β-lactamase polypeptide synthesis by Pth-defective extracts and partially inhibits synthesis by wild-type extracts. Fractions enriched for Pth, or a homogeneous preparation of Pth, prevented and reversed bar+-mediated inhibition. A mutant minigene, barA702, which changes the second codon AUA (Ile) to AAA (Lys), was also toxic for Pth-defective cells. Expression of barA702 inhibited in vitro polypeptide synthesis by Pth-defective extracts and, as with bar+, exogenous Pth prevented inhibition. Addition of pure tRNALys prevented inhibition by barA702 but not by bar+. Expression of bar+ and barA702 led to release and accumulation of p-tRNAIle and p-tRNALys respectively but bar+ also induced accumulation of p-tRNALys. Finally, bar+ stimulated association of methionine with ribosomes probably as fMet-tRNAfMet and the accumulation of methionine and isoleucine in solution as peptidyl-tRNA (p-tRNA). These results indicate that minigene-mediated inhibition of protein synthesis involves premature release of p-tRNA, misincorporation of amino acyl-tRNA, accumulation of p-tRNAs and possibly sequestration of tRNAs. Introduction Peptidyl-tRNA hydrolase (Pth; EC 3.1.1.29), an enzyme essential for Escherichia coli viability, cleaves the ester linkage of peptidyl-tRNA (p-tRNA) to yield peptides and tRNA. It has been proposed that the natural substrates for Pth are p-tRNAs released from ribosomes (Atherly and Menninger, 1972; Menninger, 1976). A missense mutation in the gene encoding this enzyme,pth(ts), causes temperature-sensitive bacterial growth (Atherly and Menninger, 1972; García-Villegas et al., 1991). Upon shift to non-permissive temperatures, the mutant shows accumulation of p-tRNA, arrest of protein synthesis and cell death. It has been suggested that this arrest may result from toxicity of the p-tRNA and/or from starvation of the critical tRNA sequestered as p-tRNA (Atherly and Menninger, 1972). The latter possibility is supported by the observation that the thermosensitivity of the pth(ts) mutant is suppressed by overproduction of tRNALys (Hergué-Hamard et al., 1996), the tRNA which accumulates the fastest as p-tRNA upon temperature increase (Menninger, 1978). An excess of tRNALys would also favor accurate translation of the Lys codons, reducing the rate of misincorporation with a consequent decrease in the formation and dissociation of erroneous p-tRNA. Both pth(ts) and another mutation, pth(rap), prevent the vegetative growth of bacteriophage λ under conditions which allow abundant growth of the mutant bacteria (Guzmán and Guarneros, 1989; García-Villegas et al., 1991). Phage mutations, designated bar, that grow well on the pth mutant cells, map to three loci in the genome: barI, at the phage attachment site attP; barII within the ssb gene; and barIII near the immunity region (Guzmán and Guarneros, 1989). Plasmid constructs that express a λ DNA segment containing either the barI or barII loci cannot be maintained in pth mutants (Guzmán et al., 1990). Expression of wild-type bar sequences from the constructs causes a general arrest of protein synthesis and is toxic for these bacteria (Pérez-Morga and Guarneros, 1990). It has been suggested that wild-type bar expression kills the mutants because it leads to excessive dissociation of p-tRNA, which then accumulates and inhibits synthesis (Hernández et al., 1997). Subcloning of these sequences showed that a construct expressing only a 21 bp sequence of the barI region in a transcript context is not maintained in the pth(rap) mutant (Guzmán et al., 1990). Inspection of the 21 bp nucleotide sequence and 5′-nearby sequences in the λ bar+ region revealed the presence of a minigene: a DNA segment whose transcripts would contain a Shine–Dalgarno-like sequence appropriately spaced for translation from the sequence AUG AUA UAA. Furthermore, the λ barII region contains a Shine–Dalgarno-like sequence preceding the same two-codon ORF but terminated at UGA. We have proposed that toxicity involves translation of the bar+ transcripts because they associate with ribosomes and are released by run-off translation. It was also suggested that bar+ transcript–translation leads to 'drop-off' and, under limiting Pth activity, to accumulation of p-tRNA (Ontiveros et al., 1997). This drop-off could result either from misincorporation of aminoacyl-tRNA or from a natural propensity of bar+ messengers to pause at the stop codon by a defect in translation termination (for a review see Hernández et al., 1997). Under the so-called 'ribosome editor' hypothesis, amino acids inappropriately incorporated into the growing polypeptide chain lead to preferential dissociation of p-tRNA from the ribosomes (Menninger, 1977). It has been shown that dissociation of non-cognate p-tRNA happens more readily with tRNAs carrying short peptidyl chains than longer ones (Gast et al., 1987). On the other hand, to our knowledge, there is no published evidence supporting a defect in termination for faithful translation of messengers containing two or a few sense codons. In these experiments, we have used a cell-free transcription–translation system to investigate the mechanism of minigene-mediated inhibition of protein synthesis. From the results we conclude that ribosome drop-off of p-tRNA is a common event in minigene transcript–translation and that translation inhibition involves p-tRNA accumulation and aminoacyl-tRNA misincorporation. Results Expression of λ bar+ minigene inhibits protein synthesis in cell-free extracts The expression of bar+ from plasmids is lethal to pth(rap) mutant cells (Guzmán et al., 1990). Lethality may result from a translational defect because it correlates with inhibition of protein synthesis, whereas RNA synthesis remains unaffected (Pérez-Morga and Guarneros, 1990). To elucidate further the mechanism of inhibition, we used a DNA-dependent cell-free system for protein synthesis. We prepared S30 extracts from pth(rap) or wild-type cells and conducted assays primed by plasmid constructs containing bar+ and bla genes (Figure 1) as described in Materials and methods. The results showed that the construct containing bar+ did not express β-lactamase protein (Bla) in pth(rap) extracts and severely inhibited Bla synthesis in wild-type extracts (Figure 2, lanes 6 and 3). The construct carrying the bar101 mutation, a 1 bp substitution which changes the AUG start codon to AUA, is not toxic for Pth-defective cells (Guzmán et al., 1989). This construct directed Bla synthesis efficiently, as did the plasmid vector, both with wild-type and pth(rap) extracts (Figure 2, compare lanes 4 and 7 with lanes 2 and 5). As estimated by Northern blot assays, we found comparable concentrations of bla RNA and bar RNA, both in bar+ or bar101 reactions (data not shown). These in vitro results reproduce the basic features of in vivo protein-synthesis inhibition mediated by expression of bar+. Figure 1.Linear map of pFGbar constructs. bla is the gene encoding β-lactamase; oLpL, λ leftward operator-promoter fragment corresponding to co-ordinates 35 535–35 713 in the λ map (Daniels et al., 1983); bar is a BamHI (B)-HindIII (H) DNA fragment comprising nucleotides 27 976–27 480 in the λ map and including λ bar+ and tI transcription-terminator; ori replication origin of pBR322 (distances are not drawn to scale). The sequence below the map shows a segment of bar+-region transcript which includes bar+ minigene. The Shine–Dalgarno-like sequence is in bold, the two-codon ORF is underlined and the stop codon is in italics. The base substitutions bar101 and barA702 are indicated. Download figure Download PowerPoint Figure 2.λ bar+ minigene expression inhibits protein synthesis in vitro. In vitro reactions (total volume 50 μl) contained the indicated plasmids (Vector, pFG93; bar+, pFGbar+ carrying λ bar+ region; bar101, pFGbar101) and S30 extracts from either pth(rap) mutant, or wild-type cells as indicated. Reactions were stopped after 60 min at 37°C, and the products were analyzed by SDS–PAGE and autoradiography as described in Materials and methods. The migration position of β-lactamase protein in the gel is indicated. Download figure Download PowerPoint Pth prevents λ bar+ -mediated protein synthesis inhibition If bar+ -mediated protein-synthesis inhibition were caused by accumulation of p-tRNA as a consequence of defective Pth (Pérez-Morga and Guarneros, 1990; Ontiveros et al., 1997), the addition of exogenous Pth activity should prevent inhibition. To test this prediction as well as the possible involvement of another factor, we fractionated wild-type S30 extracts and assayed for in vitro bar+-mediated inhibition. Previous investigations have shown that Pth activity partitioned between the S150 supernatant and the ribosomal wash (RW) (Menninger et al., 1970; Brun et al., 1971). We prepared S150, RW fractions and 70S ribosomes from wild-type and Pth-defective cells as described in Materials and methods. Addition of S150 or RW fractions from wild-type cells to in vitro transcription–translation mixtures prepared from Pth-defective cell extracts, restored protein synthesis inhibited by bar+ expression (Figure 3A, lanes 7 and 8). Cellular fractions derived from a pth(rap) mutant were less effective than the respective fractions from wild-type cells in restoring Bla synthesis (Figure 3A, lanes 4 and 5). The amount of Pth antigen in the wild-type and pth(rap) fractions was comparable (Figure 3B), as shown by Western blot. Therefore, the difference in the restorative capacity could be accounted for by the wild-type and mutant Pth-specific activities. To show that Pth was the active component in protein synthesis restoration, we purified the Pth protein to homogeneity (Kössel et al., 1970). Pure Pth prevented protein synthesis inhibition by λ bar+ expression (Figure 3A, lane 9). We found a direct correlation, in the range of 2–40 ng/μl, between Pth concentration and its capacity to prevent bar+ inhibition (not shown). These results suggest that Pth-mediated restoration of protein synthesis is derived from its hydrolytic activity on free p-tRNA, which in turn implies that bar+-mediated inhibition of Bla synthesis is caused by the release and accumulation of uncleaved p-tRNA(s). Figure 3.Pth prevents λ bar-mediated inhibition of protein synthesis. (A) In vitro reaction mixtures contained the indicated plasmids, S30 extracts from pth(rap) mutant cells and 80 μg protein of cell fractions (RW, ribosomal washing; S150, 150 000 g supernatant; 70S, washed ribosomes) or 0.5 μg of purified Pth as indicated. After 1 h incubation at 37°C the products were processed and autoradiographed as described in Materials and methods. The migration position of β-lactamase protein is indicated. (B) Pth protein content in 70S, S150 and RW fractions estimated by immunoblot assay. Fractions were electrophoresed, blotted onto nitrocellulose membranes and Pth-antigen detected using a rabbit polyclonal serum against Pth as described in Materials and methods. Purified Pth was introduced as a control. Protein concentration was determined by densitometry from a standard curve with known concentrations of purified Pth in the same immunoblot. Download figure Download PowerPoint λ bar+ minigene transcript is translated From the nucleotide sequence of λ bar+ we would expect that abortive translation termination of its transcripts would yield fMet-Ile-tRNAIleAUA. To test this prediction, we monitored the incorporation of [35S]methionine into translation intermediates directed by the bar+ constructs with unlabeled isoleucine as the only additional amino acid. Therefore, the [35S]methionine-containing TCA-precipitable material would presumably be tRNA-bound intermediates and not polypeptides (see Materials and methods). Using sedimentation through sucrose gradients, we analyzed the distribution of [35S]methionine TCA-precipitable products. Expression of the bar+ construct promoted the association of [35S]methionine with ribosomes and its incorporation into the soluble fractions (Figure 4B). The mutant bar101 construct with the poor initiation codon AUA in the miniORF (Shinedling et al., 1987), failed to mediate detectable association of methionine with ribosomal fractions (Figure 4A). As the absence of cold isoleucine from the reaction mixtures did not affect the association of [35S]methionine to the ribosomal fraction (data not shown), the ribosome-associated methionine was probably fMet-tRNAfMet. Thus, the incorporation of [35S]methionine seems to depend on efficient translation initiation of the minigene mRNA. Other mRNAs encoded in the constructs, including that of the bla gene, did not affect incorporation even in the absence of inhibition by a bar101 mutation (Figure 4A). The loss of [35S]-methionine-containing material, by alkali treatment of the fractions, is consistent with the notion that it was bound to tRNA (Figure 4E). Addition of purified Pth to the in vitro reaction prevented the accumulation of [35S]methionine-containing TCA-precipitable material in the ribosomal fractions and reduced its concentration in the upper soluble fractions (Figure 4C). The addition of puromycin to the in vitro reactions released [35S]methionine from the ribosomal fraction (Figure 4D). Puromycin mimics an amino acyl-tRNA in the ribosomal A-site, and acts as an acceptor of the peptide or fMet-tRNA derivatives in the P-site releasing peptidyl-puromycin or fMetpuromycin (Nierhaus and Wittmann, 1980). Since fMet-tRNAfMet may be on the ribosomes, as argued above, the product of this reaction is probably fMet-puromycin. In the presence of [3H]isoleucine and unlabeled methionine as the only amino acids in the reaction, most of the TCA-insoluble 3H-label was incorporated into the soluble fraction (Figure 5A). Addition of Pth to these reactions significantly reduced the 3H-label incorporated in solution (Figure 5B). Based on these results we suggest that translation of the bar+ transcript yields fMet-IletRNAIleAUA which is released and accumulates in the soluble fraction. As the concentration of free tRNAIleAUAbecomes limiting it causes fMet-tRNAfMet to pause at the P-site on the ribosome. Figure 4.Distribution of [35S]methionine-containing products along the ribosomal profile of samples from in vitro bar+ primed reactions. In vitro reactions carried out as described in Materials and methods with [35S]methionine and cold isoleucine as the only amino acids added to the reaction, were directed with: (A) pFGbar101; (B) pFGbar+; (C) pFGbar+ in the presence of 10 ng/μl of Pth; (D) pFGbar+ further incubated with 1 mM puromycin at 37°C for 10 min; (E) pFGbar+. 50 μl samples from the in vitro coupled transcription–translation reactions were layered onto a 5–40% linear sucrose gradient containig 50 mM Tris–HCl, 3.5 mM Mg(Ac)2, 60 mM NH4Cl, 0.6 mM 2-mercaptoethanol and run as described in Materials and methods. Fractions (300 μl) were collected, and aliquots (10 μl) were drawn to measure absorbance at 260 nm and cold TCA-insoluble radioactivity. Gradient fractions from (E) were treated with 1 M NaOH for 10 min at 37°C before TCA precipitation. Absorbance at 260 nm (•); 35S-activity (○). Download figure Download PowerPoint Figure 5.Distribution of [3H]isoleucine-containing products along the ribosomal profile of samples from in vitro bar+ primed reactions. In vitro reactions carried out as described in Materials and methods with [3H]isoleucine and cold methionine as the only amino acids added to the reaction, were directed with: (A) pFGbar+; (B) pFGbar+ in the presence of 10 ng/μl of Pth. The gradients were done as described in Figure 4. Download figure Download PowerPoint tRNALys prevents protein-synthesis inhibition by the mutant barA702 If bar+-mediated protein-synthesis inhibition were caused by starvation of free tRNAIleAUA sequestered as p-tRNAIleAUA, adding excess tRNAIleAUA to the reaction mixtures should reverse the inhibition. However, this is technically a difficult experiment to do because the tRNAIleAUA is a very scarce isoacceptor, not easily available in pure form (Muramatsu et al., 1988). We therefore took advantage of a mutation in bar+, barA702, in which the AAA (Lys) codon, cognate to an abundant isoacceptor, substitutes the AUA at the second position in the bar+ ORF (J.G.Valadez et al., in preparation; Ikemura, 1985; Dong et al., 1996). Expression of barA702, like bar+, was toxic to pth(rap) cells (not shown) and prevented Bla synthesis in reactions prepared with pth(rap)-cell free extracts (Figure 6, lane 3). As predicted, the addition of a purified tRNALys preparation restored synthesis of the bla gene product (Figure 6, lanes 4 and 5). The addition of non-cognate tRNAs did not reverse the effects of barA702 expression (Figure 6, lanes 6 and 7), nor did the addition of tRNALys to mixtures primed by bar+ (AUA) minigene (not shown). Addition of tRNALys at various times up to 60 min after reaction initiation by the barA702 construct, restored Bla protein synthesis (data not presented). This result suggests that the accumulated p-tRNALys (Table I) is not toxic for translation initiation as proposed (Atherly, 1978) but it could be a competitive inhibitor of tRNALys. Inhibition by barA702 was also prevented by exogenous Pth to the same degree as Pth reversed the effect of bar+ regardless of whether Pth was added at the beginning of the reaction (compare Figure 6, lane 8 with Figure 3A, lane 9) or up to 30 min after initiation (not shown). This is consistent with the recycling of tRNAs from p-tRNAs generated during barA702 inhibition. Figure 6.tRNALys and Pth prevent barA702 mediated inhibition of protein synthesis. In vitro transcription–translation reactions prepared with Pth-defective S30 extracts as described in Materials and methods, were primed with pFGbarA702 DNA except lane 1 (primed with pFGbar101) and lane 2 (with pFGbar+). Where indicated, 5 or 10 μg of specific tRNA(s) or 0.5 μg of Pth were added to the reaction. The migration position of β-lactamase protein in the SDS–PAGE gel is indicated. Download figure Download PowerPoint Table 1. Accumulation of specific p-tRNA upon in vitro bar expression Plasmid Fraction % p-tRNA Ile Lys His None S 0.5 0.1 0.1 R 0.1 0.1 0.2 bar101 S 4.1 3.8 3.1 R 0.7 0.8 0.9 bar+ S 10.5 8.0 2.2 R 1.8 0.8 0.1 barA702 S 2.7 40.1 3.6 R 0.1 2.4 0.6 The percent of peptidyl-tRNA in solution (S) or in the ribosomal (R) fractions was estimated as the accepting activity for each of the indicated amino acids relative to the total accepting activity of the in vitro reactions as described in Materials and methods. Reactions were performed in the presence of all 20 amino acids. The values are averages of two independent determinations. bar expression causes p-tRNA accumulation In an attempt to test the assumption that unsuccessful translation of the bar+ transcript leads to accumulation of fMet-Ile-tRNAIleAUA, we assayed directly for the presence of p-tRNA upon in vitro minigene expression. Reactions were primed with plasmid constructs carrying either bar+ or barA702 minigenes in the presence of all 20 amino acids, and the p-tRNA was measured in ribosomal and soluble fractions. The data in Table I showed, relative to bar101 controls, that bar+ and barA702 expression induced accumulation of p-tRNAIle and p-tRNALys in solution, respectively, whereas no accumulation of control p-tRNAHis was observed in any case. Interestingly, bar+ expression, which does not encode lysine, accumulated p-tRNALys but barA702 did not accumulate p-tRNAIle. In both bar+ and barA702-inhibited reactions p-tRNAs were found in the soluble fractions. The significant values for accumulation of p-tRNA were above the average acceptor activity for each of the three amino acids assayed for the soluble and ribosomal fractions from control reactions without priming plasmid (Table I, lines 1 and 2). Discussion We have shown that expression of bacteriophage λ bar+ minigene in vitro leads to accumulation of p-tRNA and inhibition of protein synthesis. Presumably, p-tRNAs dissociate from ribosomes by a defect in translation termination and accumulate when Pth activity is limited. These data confirm and extend the notion, based on in vivo results, that translation of λ bar+ mRNA is essential for inhibition of protein synthesis. In fact, bar+ transcripts seem to act as true messengers because they associate with ribosomes and dissociate by run-off translation (Ontiveros et al., 1997). In vitro inhibition is stronger with pth-defective cell-free extracts than it is with wild-type extracts, presumably because they accumulate p-tRNA faster (Figure 2, compare lanes 6 and 3). The hydrolysis of p-tRNA suppresses inhibition since preparations enriched for Pth restored protein synthesis, and a direct correlation has been found between pure Pth concentration and its restoring capacity (Figure 3; data not shown). Free tRNA sequestration as uncleaved p-tRNA or direct toxicity by p-tRNA have been suggested as the possible causes of lethality in the thermosensitive mutant of E.coli pth. ts) at the restrictive temperature (Atherly and Menninger, 1972; Menninger et al., 1973; Menninger, 1976). Overproduction of tRNALys, the isoacceptor which accumulates the fastest as p-tRNA upon temperature upshift, overcomes lethality (Menninger, 1978; Hergué-Hamard et al., 1996), indicating that the growth limitation of the mutant could be due to tRNALys starvation or to a rise in the rate of misincorporation and a consequent increase in the formation and dissociation of erroneous p-tRNA. Improperly terminated translation of bar+ mRNA, AUG AUA UAA, is expected to yield fMet-Ile-tRNAIleAUA. Our in vitro results show that the accumulated material in the soluble fraction could include such p-tRNA: it contains methionine (Figure 4B); it is susceptible to alkaline degradation (Figure 4E); its accumulation is prevented by Pth (Figure 4C); it contains [3H]isoleucine (Figure 5); and its tRNA is specific for isoleucine (Table I). Mutant minigene barA702, which carries AAA (Lys) as a second codon, is toxic for pth-deficient cells and inhibits protein synthesis in the cell-free system (J.G.Valadez, unpublished results and Figure 6). Likewise, a construct carrying barA702 may have promoted the accumulation of [3H]-lysine-containing p-tRNALys because Pth treatment released the 3H-label and lysine accepting activity (Table I; data not shown). Neither bar+ nor barA702 mediate incorporation of 3H-labeled p-tRNA to the ribosomal fractions suggesting that little, if any, p-tRNA remains bound to ribosomes (Figure 5). Expression of bar+ constructs promoted association of formyl-[35S]Met-tRNAfMet to ribosomes (Figure 4B). As [35S]methionine was released by puromycin treatment (Figure 4D), it is likely that the fMet-tRNAfMet resided in the ribosomal P-site (Nierhaus and Wittmann, 1980). Such a situation would be expected by pausing of ribosomes at the AUA codons upon starvation for free tRNAIleAUA during bar+ mRNA translation. If translation termination proceeded normally, fMet-Ile-tRNAIleAUA would efficiently be cleaved into fMet-Ile and tRNAIleAUA. We do not know the extent to which this termination occurs, but substantial levels of defective termination are indicated by the accumulation of p-tRNAIle in the ribosome-free fraction of in vitro reactions (Table I). The nature of the p-tRNA depends upon the minigene expressed: bar+ provokes accumulation of p-tRNAIle, whereas barA702 leads to build up of p-tRNALys (Table I). This indicates that the last sense codon, before the stop in the minigene message, determines the specificity of the accumulated p-tRNA. However bar+, which does not contain lysine codons, also mediated accumulation of p-tRNALys (Table I). The 'ribosome editor' hypothesis proposes that inappropriate p-tRNAs, resulting from erroneous incorporation of aminoacyl-tRNA, preferentially dissociate from the ribosome (Menninger, 1977). Upon starvation for tRNAIleAUA, sequestered as p-tRNAIleAUA, Lys-tRNALys would compete for binding at AUA codons, misincorporate lysine, and dissociate as inappropriate fMet-Lys-tRNALys. This, in turn, may lead to starvation for tRNALys. The fact that barA702 did not promote misincorporation of isoleucine (Table I), might be explained by a lack of misreading of the AAA codon by the tRNAIleAUA. tRNALys, specific for lysine codons AAA and AAG, is a fairly abundant tRNA, (Ikemura, 1991; Dong et al., 1996), thus bar+-mediated inhibition is not only specific for scarce isoacceptors cognate to rare codons. A significant proportion of tRNALys accumulates as p-tRNA upon barA702 expression (40%, Table I) and an additional supply of tRNALys prevents inhibition caused by barA702 minigene (Figure 6, lanes 4 and 5). These features are
Referência(s)