- Email: [email protected]
The ribosome-associated inhibitor A reduces translation errors
BBRC Biochemical and Biophysical Research Communications 320 (2004) 354–358 www.elsevier.com/locate/ybbrc
The ribosome-associated inhibitor A reduces translation errors Dmitry E. Agafonov and Alexander S. Spirin* Institute of Protein Research, Russian Academy of Sciences, 142290 Pushchino, Moscow Region, Russia Received 17 April 2004 Available online 11 June 2004
Abstract Recently we have reported about a novel stress response protein (pY or RaiA) associated with Escherichia coli ribosomes that inhibits translation at the aminoacyl-tRNA binding stage. Here we show that leucine misincorporation during in vitro poly(U) translation is inhibited by this protein much stronger than the incorporation of phenylalanine. The miscoding counteraction by RaiA is especially strong at the concentrations of magnesium ions close to those observed in vivo and diminishes at higher magnesium concentrations. The results obtained suggest that the anti-miscoding activity of RaiA could be the main function of the protein, rather than the inhibition of translation. The role of the protein in adaptation of cells to environmental stress is discussed. Ó 2004 Published by Elsevier Inc. Keywords: Ribosome; Cell-free translation; Ribosome-associated inhibitor; Miscoding; Stress response
Some years ago a novel protein pY was discovered in the bacterial ribosome [1,2] and identified as the product of the gene yfia in the Escherichia coli chromosome [3]. The protein was shown to have its binding site at the small ribosomal subunit, settle at the ribosome subunit interface, and stabilize the ribosome against dissociation [3]. The protein appeared in the ribosomal fraction of the E. coli cells grown to stationary phase [4]. The accumulation of the protein in the ribosomal fraction was also detected in the E. coli cells subjected to temperature downshift [5]. Our experiments with cell-free translation systems demonstrated that the protein inhibited translation at the elongation stage by impeding the binding of aminoacyl-tRNA to the ribosomal A site; the protein was named ribosome-associated inhibitor A or RaiA [5]. The three-dimensional structure of RaiA [6,7] and its homolog from Haemophilus influenzae (HI0257, 64% sequence identity with E. coli RaiA) [8] was solved by NMR study in solution. Both structures are similar (comparison is given in [6,7]) and possess b-a-b-b-b-a topology resembling the a-b-b-b-a topology of the double-stranded RNA-binding domain. When com*
Corresponding author. Fax: +7-095-924-0493. E-mail addresses: [email protected], [email protected]. ru, (A.S. Spirin). 0006-291X/$ - see front matter Ó 2004 Published by Elsevier Inc. doi:10.1016/j.bbrc.2004.05.171
pared with dsRBD proteins, little sequence homology and different arrangement of charged amino acid residues were mentioned, thus leading to the conclusion that RaiA and its homologs have another mode of interaction with RNA [6–8]. The NMR titration and gel shift experiments show that RaiA does not bind to model fragment of 16S ribosomal RNA, which comprises a double-stranded region and a UUCG tetraloop [7] indicating the absence of dsRBD-like affinity to RNA. RaiA belongs to a protein family with unknown function and is present in a number of bacterial species [9]. The homolog of RaiA was also found in chloroplasts of Spinacia oleracea as a component of the 30S ribosomal subunit [10] and was reported to be incorporated into E. coli ribosomes in vivo [11]. Expression of the RaiA homolog from cyanobacteria Synechococcus PCC 7002 only under dark conditions [12] taken together with appearance of RaiA in ribosomes in response to either excessive cell density [4,5] or temperature downshift [5] suggests that proteins from this family can be involved in general adaptation of the translation machinery to the environmental stress. In this study the cell-free translation experiments have demonstrated that RaiA is capable of strongly reducing mistranslation. The anti-miscoding activity is suggested as the main function of RaiA in vivo.
D.E. Agafonov, A.S. Spirin / Biochemical and Biophysical Research Communications 320 (2004) 354–358
355
Table 1 Effect of RaiA on miscoding level in poly(U)-directed cell-free translation systems at different magnesium concentrations Mg2þ concentration (mM)
RaiA to ribosomes ratio
Phe incorporated (pmol)
Leu incorporated (pmol)
Leu/Phe (%)
8
No RaiA 1:1
19.18 1.2 7.12 0.6
Not detected Not detected
Not calculated Not calculated
9
No RaiA 1:1
19.58 1.3 13.32 0.9
0.36 0.02 0.05 0.004
1.84 0.38
10
No RaiA 1:1 2:1
22.32 1.4 23.81 1.7 21.25 1.5
1.37 0.08 0.5 0.03 0.3 0.02
6.14 2.1 1.41
11
No RaiA 1:1
21.54 1.6 23.17 1.6
1.95 0.13 0.98 0.057
9.05 4.23
14
No RaiA 2:1
20.77 1.5 21.37 1.7
7.88 0.63 5.9 0.41
37.94 27.61
The amount of incorporated amino acid after 60 min of incubation was calculated from specific activities of the commercial radioactive amino acids.
Materials and methods Poly(U)-directed translation in E. coli RNA-free S100 extract [13] was carried out at 30 °C for 1 h. The final concentration of the cell extract in the reaction mixtures corresponded to 1 mg of total protein of the extract per ml. The reaction mixtures contained 1.5 mM ATP, 0.2 mM GTP, 0.6 mg/ml E. coli total tRNA, 5 mM phosphoenolpyruvate, 0.025 mg/ml pyruvate kinase, 0.2 mg/ml poly(U), and 0.22 lM ribosomal subunits each in 50 mM Hepes–KOH buffer, pH 7.5, with 115 mM NH4 Cl, 0.1 mM EDTA, and 1 mM DTT. The concentration of MgCl2 varied from 8 to 14 mM in different experiments. Other components were 9.5 lM [14 C]phenylalanine (Amersham–Pharmacia Biotech, 504 mCi/mmol) and 9.5 lM leucine when the phenylalanine incorporation was tested, or 9.5 lM [3 H]leucine (Amersham–Pharmacia Biotech, 161 Ci/mmol) and 9.5 lM phenylalanine when the leucine incorporation was tested. All results (except Table 1) are expressed as net incorporation of amino acid (incorporation in the presence, minus that in the absence, of added poly(U)) in counts per min. Purified protein RaiA (purity of the protein was evaluated by SDS–PAGE and mass spectrometric analysis as described in [3]) was added to the translation mixture to concentrations of 0.22 lM (equimolar amount relative to ribosomes) or 0.44 lM (twofold excess relative to ribosomes), and the incubation was carried out in parallel with that in the absence of RaiA. The aliquots of 5 ll were taken during incubation to measure the radioactive amino acid incorporation into hot trichloroacetic acid-insoluble product. Presented data were from at least three independent experiments.
the poly(U) template [14,15]. Fig. 1 demonstrates that the addition of RaiA to the poly(U)-directed cell-free translation system at 8 mM Mg2þ reduced the incorporation of phenylalanine into polypeptide, as it was previously reported [5]. The inhibition reached 60–70% at the equimolar ratio of RaiA to ribosomes. It was found impossible, however, to judge about the extent of inhibition of mistranslation because of very low original level of leucine incorporation (less than 0.05 pmol) at this Mg2þ concentration both in the absence and in the presence of RaiA. The fact of high accuracy of the poly(U)-directed cell-free translation system at sub-optimal Mg2þ concentrations was mentioned earlier [16].
Results and discussion In order to check the possible effect of RaiA on mistranslation, the cell-free translation experiments were performed as described previously [5] but at several different Mg2þ concentrations in the presence of phenylalanine and leucine (at their equal concentrations) as substrates for polypeptide synthesis. Such an experimental setup allows one to discriminate between codoncognate incorporation of phenylalanine (encoded by UUU or UUC codons) and incorporation of leucine caused by false ribosomal recognition of the near codoncognate leucine isoacceptor tRNAs (UUA or UUG) on
Fig. 1. Effect of RaiA on cell-free translation of poly(U) at 8 mM Mg2þ . Incorporation of [14 C]phenylalanine into newly synthesized polypeptide was measured by radioactivity of hot 10% trichloroacetic acid-insoluble precipitate. The presence of RaiA in the reaction mixture is indicated by open symbols. Filled squares, the absence of RaiA. Open circles, RaiA-to-ribosomes ratio is 0.3:1. Open triangles, RaiAto-ribosomes ratio is 1:1.
356
D.E. Agafonov, A.S. Spirin / Biochemical and Biophysical Research Communications 320 (2004) 354–358
Miscoding in cell-free translation systems is known to be stimulated by high Mg2þ concentrations, as well as by diamines and polyamines [14–16]. In order to obtain a detectable level of miscoding, the Mg2þ concentration in the poly(U)-directed cell-free translation system was increased (Fig. 2). In our experiments the level of leucine incorporation at 10 mM Mg2þ was estimated to be approximately 6–7%. Fig. 2A demonstrates that at 10 mM Mg2þ the presence of even twofold molar excess of RaiA over ribosomes only slightly, if any, decreased phenylalanine incorporation rate. On the other hand, the presence of RaiA led to significant inhibition of leucine
Fig. 2. Effect of RaiA on cell-free translation of poly(U) at 10 mM Mg2þ . Molar ratio of RaiA to ribosomes was 2:1. Incorporation of radioactive amino acids into newly synthesized polypeptide was measured by radioactivity of hot 10% trichloroacetic acid-insoluble precipitate. The presence of RaiA in the reaction mixture is indicated by open symbols. (A) Incorporation of [14 C]phenylalanine in the presence of unlabeled leucine. (B) Incorporation of [3 H]leucine in the presence of unlabeled phenylalanine.
incorporation (Fig. 2B). Thus, the presence of RaiA in the reaction mixture made translation much more accurate. Further increase of Mg2þ concentration from 10 to 14 mM stimulated the poly(U)-directed leucine incorporation 5–6 times, although the polyphenylalanine synthesis remained almost the same (Fig. 3). The presence of RaiA at 14 mM Mg2þ had negligible effect, if any, on the phenylalanine incorporation (Fig. 3A) and caused only 30% decrease of the leucine incorporation (Fig. 3B). The cell-free translation data obtained at different Mg2þ concentrations in the 8–14 mM range in the presence and absence of RaiA are summarized in Table 1. The results demonstrate that the yield of
Fig. 3. Effect of RaiA on cell-free translation of poly(U) at 14 mM Mg2þ . Experimental conditions and symbols are as in Fig. 2. (A) Incorporation of [14 C]phenylalanine in the presence of unlabeled leucine. (B) Incorporation of [3 H]leucine in the presence of unlabeled phenylalanine.
D.E. Agafonov, A.S. Spirin / Biochemical and Biophysical Research Communications 320 (2004) 354–358
Fig. 4. Improvement of translation fidelity by RaiA at different magnesium concentrations. The ratio between RaiA and ribosomes was 1:1. The difference in translation accuracy was calculated as Leu/Phe ratio in the absence of RaiA divided by Leu/Phe ratio in the presence of RaiA. Initial values are given in Table 1.
translation and the translation accuracy are strictly governed by the magnesium ion concentration and the amount of added RaiA. Fig. 4 demonstrates the improvement of translation fidelity by RaiA (Leu/Phe ratio in the absence of RaiA divided by Leu/Phe ratio in the presence of RaiA; initial values are given in Table 1), which was plotted as a function of magnesium ion concentration. It is clearly seen that despite of inhibitory effect on overall protein production observed in the presence of RaiA (Fig. 1, Table 1), the protein dramatically reduces translation errors. The miscoding counteraction is especially pronounced upon Mg2þ decrease, and the extrapolation of the data shows that RaiA could make translation much more accurate at 5–7 mM Mg2þ (Fig. 4) which is generally considered to be the in vivo concentration of magnesium. On the other hand, the miscoding inhibition effect of RaiA at low magnesium concentration became negligible at higher magnesium. It seems that heightened unspecific affinity of both cognate and non-cognate aminoacyltRNA to the ribosome remains unaffected by RaiA (Fig. 3). In contrast, the diminished affinity of aminoacyl-tRNA to the ribosome at low magnesium concentration makes the translation system very sensitive to RaiA (Fig. 1). The presented data taken together with the A site inhibition effect of RaiA [5] lead us to the suggestion that the mechanism of RaiA function is a decrease of aminoacyl-tRNA affinity to the ribosomal A site. This effect results in an improvement of discrimination between cognate and non-cognate
357
binding of tRNA to the ribosome and thus raises the fidelity of translation. The accuracy of protein biosynthesis is one of the main demands to translation machinery [17]. The incorrect recognition of non-cognate aminoacyl-tRNA by the ribosome results in amino acid substitution and can dramatically damage protein structure and function. It is well documented that the ribosome components, both proteins and RNA, play an important role in correct selection of aminoacyl-tRNA from the milieu [18–21]. Here we reported for the first time that a novel ribosome-associated protein appearing in response to environmental stresses such as excessive cell culture density [4] or temperature downshift [5] can counteract miscoding. At the same time, the stress conditions mentioned are known to promote miscoding both in vivo and in vitro. Thus, it is well documented that lowered temperature is among physical factors that directly stimulate miscoding [14,15,22]. The excessive cell culture density at stationary phase of growth results in amino acid starvation of the cells thus leading to miscoding in vivo [23,24]. Hence, it can be presumed that the main function of RaiA is the counteraction to the miscoding effects under stress conditions. On the other hand, the protein biosynthesis rate declines upon temperature downshift [25] and at stationary phase [26] that, together with the reported absence of RaiA in polysomes [4], correlates well with the inhibitory effect of RaiA on translation [5]. These facts do not contradict the idea about the miscoding inhibitory function of RaiA. Thus, the absence of the protein in polysomes may reflect just the dynamics in the competition between aminoacyl-tRNA and RaiA for ribosomal A site: as far as translating ribosomes are “frozen” on mRNA the RaiA binding site could be blocked by tRNA. The inhibition of protein biosynthesis upon stress conditions could be also mediated by means of a number of different proteins involved in the response [27]. It seems very likely that RaiA could posses the binary biological role in stress-dependent translation rearrangement: the protein reduces the rate of protein biosynthesis adjusting it to changed environment but at the same time improves translation fidelity. The in vivo experiments that could provide additional proofs for this suggestion are now in progress.
Acknowledgments We thank V.A. Kolb for valuable discussions and critical reading of the manuscript, and A.B. Chetverin, K.S. Vassilenko, V.A. Shirokov, and A. Kommer for helpful suggestions and comments. This work was supported by Grants # 99-04-48087 and # 00-15-97903 from the Russian Foundation for Basic Research and by the Program on Physico-Chemical Biology of the Russian Academy of Sciences. The Alexander von Humboldt Foundation supported presentation of the data at RNA 2003 Conference by D.E. Agafonov.
358
D.E. Agafonov, A.S. Spirin / Biochemical and Biophysical Research Communications 320 (2004) 354–358
References [1] M.M. Yusupov, A.S. Spirin, Hot tritium bombardment technique for ribosome surface topography, Methods Enzymol. 164 (1988) 426–439. [2] D.E. Agafonov, V.A. Kolb, A.S. Spirin, Proteins on ribosome surface: measurements of protein exposure by hot tritium bombardment technique, Proc. Natl. Acad. Sci. USA 94 (1997) 12892– 12897. [3] D.E. Agafonov, V.A. Kolb, I.V. Nazimov, A.S. Spirin, A protein residing at the subunit interface of bacterial ribosome, Proc. Natl. Acad. Sci. USA 96 (1999) 12345–12349. [4] Y. Maki, H. Yoshida, A. Wada, Two proteins, YfiA and YhbH, associated with resting ribosomes in stationary phase Escherichia coli, Genes Cells 5 (2000) 965–974. [5] D.E. Agafonov, V.A. Kolb, A.S. Spirin, Ribosome-associated protein that inhibits translation at the aminoacyl-tRNA binding stage, EMBO Rep. 2 (2001) 399–402. [6] K. Ye, A. Serganov, W. Hu, M. Garber, D.J. Patel, Ribosomeassociated factor Y adopts a fold resembling a double-stranded RNA binding domain scaffold, Eur. J. Biochem. 269 (2002) 5182– 5191. [7] A. Rak, A. Kalinin, D. Shcherbakov, P. Bayer, Solution structure of the ribosome-associated cold shock response protein Yfia of Escherichia coli, Biochem. Biophys. Res. Commun. 299 (2002) 710–714. [8] L. Parsons, E. Eisenstein, J. Orban, Solution structure of HI0257, a bacterial ribosome binding protein, Biochemistry 40 (2001) 10979–10986. [9] D.E. Agafonov, V.A. Kolb, A.S. Spirin, A novel stress-response protein that binds at the ribosomal subunit interface and arrests translation, Cold Spring Harbor Symp. Quant. Biol. 66 (2001) 509–514. [10] C.H. Johnson, V. Kruft, A.R. Subramanian, Identification of a plastid-specific ribosomal protein in the 30 S subunit of chloroplast ribosomes and isolation of the cDNA clone encoding its cytoplasmic precursor, J. Biol. Chem. 265 (1990) 12790–12795. [11] M.G. Bubunenko, A.R. Subramanian, Recognition of novel and divergent higher plant chloroplast ribosomal proteins by Escherichia coli ribosome during in vivo assembly, J. Biol. Chem. 269 (1994) 18223–18231. [12] X. Tan, M. Varughese, W.R. Widger, A light-repressed transcript found in Synechococcus PCC 7002 is similar to a chloroplast-
[13]
[14] [15]
[16]
[17] [18] [19]
[20] [21] [22]
[23]
[24]
[25]
[26]
[27]
specific small subunit ribosomal protein and to a transcription modulator protein associated with sigma 54, J. Biol. Chem. 269 (1994) 20905–20912. L.M. Gold, M. Schweiger, Synthesis of bacteriophage-specific enzymes directed by DNA in vitro, Methods Enzymol. 20 (1971) 537–542. W. Szer, S. Ochoa, Complexing ability and coding properties of synthetic polynucleotides, J. Mol. Biol. 8 (1964) 823–834. S.M. Friedman, I.B. Weinstein, Lack of fidelity in the translation of synthetic polyribonucleotides, Proc. Natl. Acad. Sci. USA 52 (1964) 988–996. L.P. Gavrilova, I.N. Perminova, A.S. Spirin, Elongation factor Tu can reduce translation errors in poly(U)-directed cell-free systems, J. Mol. Biol. 149 (1981) 69–78. C.G. Kurland, Translational accuracy in vitro, Cell 28 (1982) 201– 202. L. Gorini, Ribosomal discrimination of tRNAs, Nat. New Biol. 234 (1971) 261–264. L. Gorini, Streptomycin and misreading of the genetic code, in: M. Nomura, A. Tissieres, P. Langyel (Eds.), Ribosomes, CSHL Press, Cold Spring Harbor, New York, 1974, pp. 791–803. H.F. Noller, Ribosomal RNA and translation, Annu. Rev. Biochem. 60 (1991) 191–227. R. Green, H.F. Noller, Ribosomes and translation, Annu. Rev. Biochem. 66 (1997) 679–716. J.L. Manley, R.F. Gesteland, Suppression of amber mutants in vitro induced by low temperature, J. Mol. Biol. 125 (1978) 433– 447. P.H. O’Farrell, The suppression of defective translation by ppGpp and its role in the stringent response, Cell 14 (1978) 545– 557. J. Parker, J.W. Pollard, J.D. Friesen, C.P. Stanners, Stuttering: high-level mistranslation in animal and bacterial cells, Proc. Natl. Acad. Sci. USA 75 (1978) 1091–1095. H.K. Das, A. Goldstein, Limited capacity for protein synthesis at zero degrees centigrade in Escherichia coli, J. Mol. Biol. 31 (1968) 209–226. R. Kolter, D.A. Siegele, A. Tormo, The stationary phase of the bacterial life cycle, Annu. Rev. Microbiol. 47 (1993) 855– 874. C.O. Gualerzi, A.M. Giuliodori, C.L. Pon, Transcriptional and post-transcriptional control of cold-shock genes, J. Mol. Biol. 331 (2003) 527–539.
The ribosome-associated inhibitor A reduces translation errors Dmitry E. Agafonov and Alexander S. Spirin* Institute of Protein Research, Russian Academy of Sciences, 142290 Pushchino, Moscow Region, Russia Received 17 April 2004 Available online 11 June 2004
Abstract Recently we have reported about a novel stress response protein (pY or RaiA) associated with Escherichia coli ribosomes that inhibits translation at the aminoacyl-tRNA binding stage. Here we show that leucine misincorporation during in vitro poly(U) translation is inhibited by this protein much stronger than the incorporation of phenylalanine. The miscoding counteraction by RaiA is especially strong at the concentrations of magnesium ions close to those observed in vivo and diminishes at higher magnesium concentrations. The results obtained suggest that the anti-miscoding activity of RaiA could be the main function of the protein, rather than the inhibition of translation. The role of the protein in adaptation of cells to environmental stress is discussed. Ó 2004 Published by Elsevier Inc. Keywords: Ribosome; Cell-free translation; Ribosome-associated inhibitor; Miscoding; Stress response
Some years ago a novel protein pY was discovered in the bacterial ribosome [1,2] and identified as the product of the gene yfia in the Escherichia coli chromosome [3]. The protein was shown to have its binding site at the small ribosomal subunit, settle at the ribosome subunit interface, and stabilize the ribosome against dissociation [3]. The protein appeared in the ribosomal fraction of the E. coli cells grown to stationary phase [4]. The accumulation of the protein in the ribosomal fraction was also detected in the E. coli cells subjected to temperature downshift [5]. Our experiments with cell-free translation systems demonstrated that the protein inhibited translation at the elongation stage by impeding the binding of aminoacyl-tRNA to the ribosomal A site; the protein was named ribosome-associated inhibitor A or RaiA [5]. The three-dimensional structure of RaiA [6,7] and its homolog from Haemophilus influenzae (HI0257, 64% sequence identity with E. coli RaiA) [8] was solved by NMR study in solution. Both structures are similar (comparison is given in [6,7]) and possess b-a-b-b-b-a topology resembling the a-b-b-b-a topology of the double-stranded RNA-binding domain. When com*
Corresponding author. Fax: +7-095-924-0493. E-mail addresses: [email protected], [email protected]. ru, (A.S. Spirin). 0006-291X/$ - see front matter Ó 2004 Published by Elsevier Inc. doi:10.1016/j.bbrc.2004.05.171
pared with dsRBD proteins, little sequence homology and different arrangement of charged amino acid residues were mentioned, thus leading to the conclusion that RaiA and its homologs have another mode of interaction with RNA [6–8]. The NMR titration and gel shift experiments show that RaiA does not bind to model fragment of 16S ribosomal RNA, which comprises a double-stranded region and a UUCG tetraloop [7] indicating the absence of dsRBD-like affinity to RNA. RaiA belongs to a protein family with unknown function and is present in a number of bacterial species [9]. The homolog of RaiA was also found in chloroplasts of Spinacia oleracea as a component of the 30S ribosomal subunit [10] and was reported to be incorporated into E. coli ribosomes in vivo [11]. Expression of the RaiA homolog from cyanobacteria Synechococcus PCC 7002 only under dark conditions [12] taken together with appearance of RaiA in ribosomes in response to either excessive cell density [4,5] or temperature downshift [5] suggests that proteins from this family can be involved in general adaptation of the translation machinery to the environmental stress. In this study the cell-free translation experiments have demonstrated that RaiA is capable of strongly reducing mistranslation. The anti-miscoding activity is suggested as the main function of RaiA in vivo.
D.E. Agafonov, A.S. Spirin / Biochemical and Biophysical Research Communications 320 (2004) 354–358
355
Table 1 Effect of RaiA on miscoding level in poly(U)-directed cell-free translation systems at different magnesium concentrations Mg2þ concentration (mM)
RaiA to ribosomes ratio
Phe incorporated (pmol)
Leu incorporated (pmol)
Leu/Phe (%)
8
No RaiA 1:1
19.18 1.2 7.12 0.6
Not detected Not detected
Not calculated Not calculated
9
No RaiA 1:1
19.58 1.3 13.32 0.9
0.36 0.02 0.05 0.004
1.84 0.38
10
No RaiA 1:1 2:1
22.32 1.4 23.81 1.7 21.25 1.5
1.37 0.08 0.5 0.03 0.3 0.02
6.14 2.1 1.41
11
No RaiA 1:1
21.54 1.6 23.17 1.6
1.95 0.13 0.98 0.057
9.05 4.23
14
No RaiA 2:1
20.77 1.5 21.37 1.7
7.88 0.63 5.9 0.41
37.94 27.61
The amount of incorporated amino acid after 60 min of incubation was calculated from specific activities of the commercial radioactive amino acids.
Materials and methods Poly(U)-directed translation in E. coli RNA-free S100 extract [13] was carried out at 30 °C for 1 h. The final concentration of the cell extract in the reaction mixtures corresponded to 1 mg of total protein of the extract per ml. The reaction mixtures contained 1.5 mM ATP, 0.2 mM GTP, 0.6 mg/ml E. coli total tRNA, 5 mM phosphoenolpyruvate, 0.025 mg/ml pyruvate kinase, 0.2 mg/ml poly(U), and 0.22 lM ribosomal subunits each in 50 mM Hepes–KOH buffer, pH 7.5, with 115 mM NH4 Cl, 0.1 mM EDTA, and 1 mM DTT. The concentration of MgCl2 varied from 8 to 14 mM in different experiments. Other components were 9.5 lM [14 C]phenylalanine (Amersham–Pharmacia Biotech, 504 mCi/mmol) and 9.5 lM leucine when the phenylalanine incorporation was tested, or 9.5 lM [3 H]leucine (Amersham–Pharmacia Biotech, 161 Ci/mmol) and 9.5 lM phenylalanine when the leucine incorporation was tested. All results (except Table 1) are expressed as net incorporation of amino acid (incorporation in the presence, minus that in the absence, of added poly(U)) in counts per min. Purified protein RaiA (purity of the protein was evaluated by SDS–PAGE and mass spectrometric analysis as described in [3]) was added to the translation mixture to concentrations of 0.22 lM (equimolar amount relative to ribosomes) or 0.44 lM (twofold excess relative to ribosomes), and the incubation was carried out in parallel with that in the absence of RaiA. The aliquots of 5 ll were taken during incubation to measure the radioactive amino acid incorporation into hot trichloroacetic acid-insoluble product. Presented data were from at least three independent experiments.
the poly(U) template [14,15]. Fig. 1 demonstrates that the addition of RaiA to the poly(U)-directed cell-free translation system at 8 mM Mg2þ reduced the incorporation of phenylalanine into polypeptide, as it was previously reported [5]. The inhibition reached 60–70% at the equimolar ratio of RaiA to ribosomes. It was found impossible, however, to judge about the extent of inhibition of mistranslation because of very low original level of leucine incorporation (less than 0.05 pmol) at this Mg2þ concentration both in the absence and in the presence of RaiA. The fact of high accuracy of the poly(U)-directed cell-free translation system at sub-optimal Mg2þ concentrations was mentioned earlier [16].
Results and discussion In order to check the possible effect of RaiA on mistranslation, the cell-free translation experiments were performed as described previously [5] but at several different Mg2þ concentrations in the presence of phenylalanine and leucine (at their equal concentrations) as substrates for polypeptide synthesis. Such an experimental setup allows one to discriminate between codoncognate incorporation of phenylalanine (encoded by UUU or UUC codons) and incorporation of leucine caused by false ribosomal recognition of the near codoncognate leucine isoacceptor tRNAs (UUA or UUG) on
Fig. 1. Effect of RaiA on cell-free translation of poly(U) at 8 mM Mg2þ . Incorporation of [14 C]phenylalanine into newly synthesized polypeptide was measured by radioactivity of hot 10% trichloroacetic acid-insoluble precipitate. The presence of RaiA in the reaction mixture is indicated by open symbols. Filled squares, the absence of RaiA. Open circles, RaiA-to-ribosomes ratio is 0.3:1. Open triangles, RaiAto-ribosomes ratio is 1:1.
356
D.E. Agafonov, A.S. Spirin / Biochemical and Biophysical Research Communications 320 (2004) 354–358
Miscoding in cell-free translation systems is known to be stimulated by high Mg2þ concentrations, as well as by diamines and polyamines [14–16]. In order to obtain a detectable level of miscoding, the Mg2þ concentration in the poly(U)-directed cell-free translation system was increased (Fig. 2). In our experiments the level of leucine incorporation at 10 mM Mg2þ was estimated to be approximately 6–7%. Fig. 2A demonstrates that at 10 mM Mg2þ the presence of even twofold molar excess of RaiA over ribosomes only slightly, if any, decreased phenylalanine incorporation rate. On the other hand, the presence of RaiA led to significant inhibition of leucine
Fig. 2. Effect of RaiA on cell-free translation of poly(U) at 10 mM Mg2þ . Molar ratio of RaiA to ribosomes was 2:1. Incorporation of radioactive amino acids into newly synthesized polypeptide was measured by radioactivity of hot 10% trichloroacetic acid-insoluble precipitate. The presence of RaiA in the reaction mixture is indicated by open symbols. (A) Incorporation of [14 C]phenylalanine in the presence of unlabeled leucine. (B) Incorporation of [3 H]leucine in the presence of unlabeled phenylalanine.
incorporation (Fig. 2B). Thus, the presence of RaiA in the reaction mixture made translation much more accurate. Further increase of Mg2þ concentration from 10 to 14 mM stimulated the poly(U)-directed leucine incorporation 5–6 times, although the polyphenylalanine synthesis remained almost the same (Fig. 3). The presence of RaiA at 14 mM Mg2þ had negligible effect, if any, on the phenylalanine incorporation (Fig. 3A) and caused only 30% decrease of the leucine incorporation (Fig. 3B). The cell-free translation data obtained at different Mg2þ concentrations in the 8–14 mM range in the presence and absence of RaiA are summarized in Table 1. The results demonstrate that the yield of
Fig. 3. Effect of RaiA on cell-free translation of poly(U) at 14 mM Mg2þ . Experimental conditions and symbols are as in Fig. 2. (A) Incorporation of [14 C]phenylalanine in the presence of unlabeled leucine. (B) Incorporation of [3 H]leucine in the presence of unlabeled phenylalanine.
D.E. Agafonov, A.S. Spirin / Biochemical and Biophysical Research Communications 320 (2004) 354–358
Fig. 4. Improvement of translation fidelity by RaiA at different magnesium concentrations. The ratio between RaiA and ribosomes was 1:1. The difference in translation accuracy was calculated as Leu/Phe ratio in the absence of RaiA divided by Leu/Phe ratio in the presence of RaiA. Initial values are given in Table 1.
translation and the translation accuracy are strictly governed by the magnesium ion concentration and the amount of added RaiA. Fig. 4 demonstrates the improvement of translation fidelity by RaiA (Leu/Phe ratio in the absence of RaiA divided by Leu/Phe ratio in the presence of RaiA; initial values are given in Table 1), which was plotted as a function of magnesium ion concentration. It is clearly seen that despite of inhibitory effect on overall protein production observed in the presence of RaiA (Fig. 1, Table 1), the protein dramatically reduces translation errors. The miscoding counteraction is especially pronounced upon Mg2þ decrease, and the extrapolation of the data shows that RaiA could make translation much more accurate at 5–7 mM Mg2þ (Fig. 4) which is generally considered to be the in vivo concentration of magnesium. On the other hand, the miscoding inhibition effect of RaiA at low magnesium concentration became negligible at higher magnesium. It seems that heightened unspecific affinity of both cognate and non-cognate aminoacyltRNA to the ribosome remains unaffected by RaiA (Fig. 3). In contrast, the diminished affinity of aminoacyl-tRNA to the ribosome at low magnesium concentration makes the translation system very sensitive to RaiA (Fig. 1). The presented data taken together with the A site inhibition effect of RaiA [5] lead us to the suggestion that the mechanism of RaiA function is a decrease of aminoacyl-tRNA affinity to the ribosomal A site. This effect results in an improvement of discrimination between cognate and non-cognate
357
binding of tRNA to the ribosome and thus raises the fidelity of translation. The accuracy of protein biosynthesis is one of the main demands to translation machinery [17]. The incorrect recognition of non-cognate aminoacyl-tRNA by the ribosome results in amino acid substitution and can dramatically damage protein structure and function. It is well documented that the ribosome components, both proteins and RNA, play an important role in correct selection of aminoacyl-tRNA from the milieu [18–21]. Here we reported for the first time that a novel ribosome-associated protein appearing in response to environmental stresses such as excessive cell culture density [4] or temperature downshift [5] can counteract miscoding. At the same time, the stress conditions mentioned are known to promote miscoding both in vivo and in vitro. Thus, it is well documented that lowered temperature is among physical factors that directly stimulate miscoding [14,15,22]. The excessive cell culture density at stationary phase of growth results in amino acid starvation of the cells thus leading to miscoding in vivo [23,24]. Hence, it can be presumed that the main function of RaiA is the counteraction to the miscoding effects under stress conditions. On the other hand, the protein biosynthesis rate declines upon temperature downshift [25] and at stationary phase [26] that, together with the reported absence of RaiA in polysomes [4], correlates well with the inhibitory effect of RaiA on translation [5]. These facts do not contradict the idea about the miscoding inhibitory function of RaiA. Thus, the absence of the protein in polysomes may reflect just the dynamics in the competition between aminoacyl-tRNA and RaiA for ribosomal A site: as far as translating ribosomes are “frozen” on mRNA the RaiA binding site could be blocked by tRNA. The inhibition of protein biosynthesis upon stress conditions could be also mediated by means of a number of different proteins involved in the response [27]. It seems very likely that RaiA could posses the binary biological role in stress-dependent translation rearrangement: the protein reduces the rate of protein biosynthesis adjusting it to changed environment but at the same time improves translation fidelity. The in vivo experiments that could provide additional proofs for this suggestion are now in progress.
Acknowledgments We thank V.A. Kolb for valuable discussions and critical reading of the manuscript, and A.B. Chetverin, K.S. Vassilenko, V.A. Shirokov, and A. Kommer for helpful suggestions and comments. This work was supported by Grants # 99-04-48087 and # 00-15-97903 from the Russian Foundation for Basic Research and by the Program on Physico-Chemical Biology of the Russian Academy of Sciences. The Alexander von Humboldt Foundation supported presentation of the data at RNA 2003 Conference by D.E. Agafonov.
358
D.E. Agafonov, A.S. Spirin / Biochemical and Biophysical Research Communications 320 (2004) 354–358
References [1] M.M. Yusupov, A.S. Spirin, Hot tritium bombardment technique for ribosome surface topography, Methods Enzymol. 164 (1988) 426–439. [2] D.E. Agafonov, V.A. Kolb, A.S. Spirin, Proteins on ribosome surface: measurements of protein exposure by hot tritium bombardment technique, Proc. Natl. Acad. Sci. USA 94 (1997) 12892– 12897. [3] D.E. Agafonov, V.A. Kolb, I.V. Nazimov, A.S. Spirin, A protein residing at the subunit interface of bacterial ribosome, Proc. Natl. Acad. Sci. USA 96 (1999) 12345–12349. [4] Y. Maki, H. Yoshida, A. Wada, Two proteins, YfiA and YhbH, associated with resting ribosomes in stationary phase Escherichia coli, Genes Cells 5 (2000) 965–974. [5] D.E. Agafonov, V.A. Kolb, A.S. Spirin, Ribosome-associated protein that inhibits translation at the aminoacyl-tRNA binding stage, EMBO Rep. 2 (2001) 399–402. [6] K. Ye, A. Serganov, W. Hu, M. Garber, D.J. Patel, Ribosomeassociated factor Y adopts a fold resembling a double-stranded RNA binding domain scaffold, Eur. J. Biochem. 269 (2002) 5182– 5191. [7] A. Rak, A. Kalinin, D. Shcherbakov, P. Bayer, Solution structure of the ribosome-associated cold shock response protein Yfia of Escherichia coli, Biochem. Biophys. Res. Commun. 299 (2002) 710–714. [8] L. Parsons, E. Eisenstein, J. Orban, Solution structure of HI0257, a bacterial ribosome binding protein, Biochemistry 40 (2001) 10979–10986. [9] D.E. Agafonov, V.A. Kolb, A.S. Spirin, A novel stress-response protein that binds at the ribosomal subunit interface and arrests translation, Cold Spring Harbor Symp. Quant. Biol. 66 (2001) 509–514. [10] C.H. Johnson, V. Kruft, A.R. Subramanian, Identification of a plastid-specific ribosomal protein in the 30 S subunit of chloroplast ribosomes and isolation of the cDNA clone encoding its cytoplasmic precursor, J. Biol. Chem. 265 (1990) 12790–12795. [11] M.G. Bubunenko, A.R. Subramanian, Recognition of novel and divergent higher plant chloroplast ribosomal proteins by Escherichia coli ribosome during in vivo assembly, J. Biol. Chem. 269 (1994) 18223–18231. [12] X. Tan, M. Varughese, W.R. Widger, A light-repressed transcript found in Synechococcus PCC 7002 is similar to a chloroplast-
[13]
[14] [15]
[16]
[17] [18] [19]
[20] [21] [22]
[23]
[24]
[25]
[26]
[27]
specific small subunit ribosomal protein and to a transcription modulator protein associated with sigma 54, J. Biol. Chem. 269 (1994) 20905–20912. L.M. Gold, M. Schweiger, Synthesis of bacteriophage-specific enzymes directed by DNA in vitro, Methods Enzymol. 20 (1971) 537–542. W. Szer, S. Ochoa, Complexing ability and coding properties of synthetic polynucleotides, J. Mol. Biol. 8 (1964) 823–834. S.M. Friedman, I.B. Weinstein, Lack of fidelity in the translation of synthetic polyribonucleotides, Proc. Natl. Acad. Sci. USA 52 (1964) 988–996. L.P. Gavrilova, I.N. Perminova, A.S. Spirin, Elongation factor Tu can reduce translation errors in poly(U)-directed cell-free systems, J. Mol. Biol. 149 (1981) 69–78. C.G. Kurland, Translational accuracy in vitro, Cell 28 (1982) 201– 202. L. Gorini, Ribosomal discrimination of tRNAs, Nat. New Biol. 234 (1971) 261–264. L. Gorini, Streptomycin and misreading of the genetic code, in: M. Nomura, A. Tissieres, P. Langyel (Eds.), Ribosomes, CSHL Press, Cold Spring Harbor, New York, 1974, pp. 791–803. H.F. Noller, Ribosomal RNA and translation, Annu. Rev. Biochem. 60 (1991) 191–227. R. Green, H.F. Noller, Ribosomes and translation, Annu. Rev. Biochem. 66 (1997) 679–716. J.L. Manley, R.F. Gesteland, Suppression of amber mutants in vitro induced by low temperature, J. Mol. Biol. 125 (1978) 433– 447. P.H. O’Farrell, The suppression of defective translation by ppGpp and its role in the stringent response, Cell 14 (1978) 545– 557. J. Parker, J.W. Pollard, J.D. Friesen, C.P. Stanners, Stuttering: high-level mistranslation in animal and bacterial cells, Proc. Natl. Acad. Sci. USA 75 (1978) 1091–1095. H.K. Das, A. Goldstein, Limited capacity for protein synthesis at zero degrees centigrade in Escherichia coli, J. Mol. Biol. 31 (1968) 209–226. R. Kolter, D.A. Siegele, A. Tormo, The stationary phase of the bacterial life cycle, Annu. Rev. Microbiol. 47 (1993) 855– 874. C.O. Gualerzi, A.M. Giuliodori, C.L. Pon, Transcriptional and post-transcriptional control of cold-shock genes, J. Mol. Biol. 331 (2003) 527–539.