Reconstitution of peptide bond formation with Escherichia coli 23S ribosomal RNA domains

Science

Volumn 281, Issue 5377, 1998, Pages 666-669

  Nitta, Itaru ^a   Kamada, Yoshie ^a   Noda, Hiroe ^a   Ueda, Takuya ^a   Watanabe, Kimitsuna ^a  

^a UNIVERSITY OF TOKYO   (Japan)

Author keywords

[No Author keywords available]

Indexed keywords

NEOMYCIN; PEPTIDE; RIBOSOME PROTEIN; RIBOSOME RNA; SPARSOMYCIN;

ARTICLE; BACTERIUM MUTANT; CHEMICAL BOND; CONTROLLED STUDY; ESCHERICHIA COLI; NONHUMAN; PRIORITY JOURNAL;

ESCHERICHIA COLI;

EID: 0032584644     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.281.5377.666     Document Type: Article

References (42)

1. H. F. Noller et al., Nucleic Acids Res. 9, 6167 (1981).
2. H. A. Raué, J. Klootwijk, W. Musters, Prog. Biophys. Mol. Biol. 51, 77 (1988).
3. C. R. Woese, The Genetic Code: The Molecular Basis for Genetic Expression (Harper & Row, New York, 1967), pp. 179-195; F. H. C. Crick, J. Mol. Biol. 38, 367 (1968).
4. C. R. Woese, The Genetic Code: The Molecular Basis for Genetic Expression (Harper & Row, New York, 1967), pp. 179-195; F. H. C. Crick, J. Mol. Biol. 38, 367 (1968).
5. H. F. Noller, Annu. Rev. Biochem. 60, 191 (1991); R. Green and H. F. Noller, ibid. 66, 191 (1997).
6. H. F. Noller, Annu. Rev. Biochem. 60, 191 (1991); R. Green and H. F. Noller, ibid. 66, 191 (1997).
7. H. Schulze and K. H. Nierhaus, EMBO J. 1, 609 (1982).
8. F. J. Franceschi and K. H. Nierhaus, J. Biol. Chem. 265, 16676 (1990).
9. H. F. Noller, V. Hoffarth, L. Zimniak, Science 256, 1416 (1992).
10. I. Nitta, T. Ueda, K. Watanabe, RNA 4, 257 (1998).
11. 4Cl. Before the assay for peptide bond formation, the transcripts in the self-folding buffer were heated at 65°C for 10 min and then cooled to 37°C over a period of 90 min for the complete self-folding of 23S rRNA or of its individual domains, or at 37°C for 20 min to reconstitute the active complex from several domains.
12. 2].
13. 3/ methanol) as well as by HPLC with an octadecylsilane column using authentic materials synthesized chemically. In addition, the spot corresponding to AcPhe-Phe was extracted from the TLC plate [J. G. Seidman, B. G. Barrell, W. H. McClain, J. Mol. Biol. 99, 733 (1975)] and acid hydrolysis of AcPhe-Phe was carried out in 6N HCl for various time periods (1 to 12 hours) at 105°C [ W. H. McClain, C. Guthrie, B. G. Barrell, J. Mol. Biol. 81, 157 (1973)]. After the reaction of 1 hour, not only Phe but AcPhe could be detected on the TLC plate, the latter of which disappeared quickly in the longer reaction periods. After 12 hours of reaction, AcPhe-Phe was completely converted to Phe. This means that AcPhe-Phe is degraded into Phe and acetic acid, partly via AcPhe. We then examined what happened if only one of the two input tRNAs was used. When only AcPhe-tRNA was incubated with domain V, AcPhe was solely recovered by the above procedure, whereas Phe and a very small amount of Phe-Phe were detected in the case of using only Phe-tRNA, suggesting that the peptide transfer can occur between two Phe-tRNAs, but not between two AcPhe-tRNAs. For further confirmation of the products, we carried out the following experiments. When peptide bond formation was performed with domain V using radiolabeled AcPhe-tRNA and unlabeled PhetRNA, we found only spots corresponding to AcPhe-Phe and AcPhe. In contrast, two spots for AcPhe-Phe and Phe were observed in the cases of unlabeled AcPhe-tRNA and labeled Phe-tRNA, which are the same results as already reported (8). This also serves to confirm that the product is really AcPhe-Phe.
14. 3/ methanol) as well as by HPLC with an octadecylsilane column using authentic materials synthesized chemically. In addition, the spot corresponding to AcPhe-Phe was extracted from the TLC plate [J. G. Seidman, B. G. Barrell, W. H. McClain, J. Mol. Biol. 99, 733 (1975)] and acid hydrolysis of AcPhe-Phe was carried out in 6N HCl for various time periods (1 to 12 hours) at 105°C [ W. H. McClain, C. Guthrie, B. G. Barrell, J. Mol. Biol. 81, 157 (1973)]. After the reaction of 1 hour, not only Phe but AcPhe could be detected on the TLC plate, the latter of which disappeared quickly in the longer reaction periods. After 12 hours of reaction, AcPhe-Phe was completely converted to Phe. This means that AcPhe-Phe is degraded into Phe and acetic acid, partly via AcPhe. We then examined what happened if only one of the two input tRNAs was used. When only AcPhe-tRNA was incubated with domain V, AcPhe was solely recovered by the above procedure, whereas Phe and a very small amount of Phe-Phe were detected in the case of using only Phe-tRNA, suggesting that the peptide transfer can occur between two Phe-tRNAs, but not between two AcPhe-tRNAs. For further confirmation of the products, we carried out the following experiments. When peptide bond formation was performed with domain V using radiolabeled AcPhe-tRNA and unlabeled PhetRNA, we found only spots corresponding to AcPhe-Phe and AcPhe. In contrast, two spots for AcPhe-Phe and Phe were observed in the cases of unlabeled AcPhe-tRNA and labeled Phe-tRNA, which are the same results as already reported (8). This also serves to confirm that the product is really AcPhe-Phe.
15. H. A. Raué, W. Musters, C. A. Rutgers, J. V. Riet, R. J. Planta, rRNA: from Structure to Function, W. E. Hill et al., Eds. (American Society for Microbiology, Washington, DC, 1990), pp. 217-235.
16. Additional experimental results indicated that the reaction was completely suppressed by denaturing at 95°C any of the 23S rRNA domains as well as by digestion of the domains with either pyrimidine-specific ribonuclease A (RNase A) or guanine-specific RNase T1 (28). Furthermore, RNAs unrelated to rRNA, such as polyuridylate and polyadenylate, were found to never enhance AcPhe-Phe synthesis at high concentrations on either a mole or weight basis, whereas 16S rRNA did so at the same level as that of the domains other than V (28). These findings give weight to the proposition that tRNAs are unspecifically gathered by 23S rRNA domains other than domain V.
17. Effective activation in the presence of domain VI was further supported by the following facts. First, the amount of AcPhe-Phe formed by domain V together with domain I, II, III, or IV was similar to that formed by each of these domains without domain V (compare Fig. 1, C and D), whereas the amount formed by domain V with VI reached about the same level as that formed by the total domain complex when the concentration of domain Vl was 4 μM or more. Second, at relatively low concentrations of additional domain VI, the incorporation of phenylalanine increased linearly with increments in the concentration of domain VI, but leveled off at around 5 μM, indicating that domain V was the central module for peptide bond formation and VI was the activator. Finally, the activity with 1 μM of domain V and 2 μM of domain VI was twice that with 3 μM of domain V only (Fig. 1, C and D).
18. D. Moazed and H. F. Noller, Cell 57, 585 (1989); R. R. Samaha, R. Green, H. F. Noller, Nature 377, 309 (1995); C. M. T. Spahn, J. Remme, M. A. Schäfer, K. H. Nierhaus, J. Biol. Chem. 271, 32849 (1996); C. M. T. Spahn, M. A. Schäfer, A. A. Krayevsky, K. H. Nierhaus, ibid., p. 32857; B. T. Porse, H. P. Thi-Ngoc, R. A. Garrett, J. Mol. Biol. 264, 472 (1996); R. Green, R. R. Samaha, H. F. Noller, ibid. 266, 40 (1997).
19. D. Moazed and H. F. Noller, Cell 57, 585 (1989); R. R. Samaha, R. Green, H. F. Noller, Nature 377, 309 (1995); C. M. T. Spahn, J. Remme, M. A. Schäfer, K. H. Nierhaus, J. Biol. Chem. 271, 32849 (1996); C. M. T. Spahn, M. A. Schäfer, A. A. Krayevsky, K. H. Nierhaus, ibid., p. 32857; B. T. Porse, H. P. Thi-Ngoc, R. A. Garrett, J. Mol. Biol. 264, 472 (1996); R. Green, R. R. Samaha, H. F. Noller, ibid. 266, 40 (1997).
20. D. Moazed and H. F. Noller, Cell 57, 585 (1989); R. R. Samaha, R. Green, H. F. Noller, Nature 377, 309 (1995); C. M. T. Spahn, J. Remme, M. A. Schäfer, K. H. Nierhaus, J. Biol. Chem. 271, 32849 (1996); C. M. T. Spahn, M. A. Schäfer, A. A. Krayevsky, K. H. Nierhaus, ibid., p. 32857; B. T. Porse, H. P. Thi-Ngoc, R. A. Garrett, J. Mol. Biol. 264, 472 (1996); R. Green, R. R. Samaha, H. F. Noller, ibid. 266, 40 (1997).
21. D. Moazed and H. F. Noller, Cell 57, 585 (1989); R. R. Samaha, R. Green, H. F. Noller, Nature 377, 309 (1995); C. M. T. Spahn, J. Remme, M. A. Schäfer, K. H. Nierhaus, J. Biol. Chem. 271, 32849 (1996); C. M. T. Spahn, M. A. Schäfer, A. A. Krayevsky, K. H. Nierhaus, ibid., p. 32857; B. T. Porse, H. P. Thi-Ngoc, R. A. Garrett, J. Mol. Biol. 264, 472 (1996); R. Green, R. R. Samaha, H. F. Noller, ibid. 266, 40 (1997).
22. D. Moazed and H. F. Noller, Cell 57, 585 (1989); R. R. Samaha, R. Green, H. F. Noller, Nature 377, 309 (1995); C. M. T. Spahn, J. Remme, M. A. Schäfer, K. H. Nierhaus, J. Biol. Chem. 271, 32849 (1996); C. M. T. Spahn, M. A. Schäfer, A. A. Krayevsky, K. H. Nierhaus, ibid., p. 32857; B. T. Porse, H. P. Thi-Ngoc, R. A. Garrett, J. Mol. Biol. 264, 472 (1996); R. Green, R. R. Samaha, H. F. Noller, ibid. 266, 40 (1997).
23. D. Moazed and H. F. Noller, Cell 57, 585 (1989); R. R. Samaha, R. Green, H. F. Noller, Nature 377, 309 (1995); C. M. T. Spahn, J. Remme, M. A. Schäfer, K. H. Nierhaus, J. Biol. Chem. 271, 32849 (1996); C. M. T. Spahn, M. A. Schäfer, A. A. Krayevsky, K. H. Nierhaus, ibid., p. 32857; B. T. Porse, H. P. Thi-Ngoc, R. A. Garrett, J. Mol. Biol. 264, 472 (1996); R. Green, R. R. Samaha, H. F. Noller, ibid. 266, 40 (1997).
24. In our experiment, we used native tRNA instead of a tRNA fragment. According to footprinting data, intact tRNA has multiple contact sites with rRNA [M. Dabrowski, C. M. T. Spahn, K. H. Nierhaus, EMBO J. 14, 4872 (1995)], so local mutation of the 23S rRNA did not seem to be critical with regard to tRNA-rRNA interaction, and high domain V mutant concentrations resulted in an increment of local concentrations of tRNAs due to unspecific tRNA-rRNA interactions. In support of this notion, the AcPhe-Phe synthesis reaction was completely eliminated by digesting the domain V wild type or its mutants with either RNase A or RNase T1 (28).
25. D. Vazquez, Inhibitors of Protein Biosynthesis (Springer-Verlag, Berlin, 1979); E. Cundliffe, Antibiotic Inhibitors of Ribosome Function, E. F. Gale, E. Cundliffe, P. E. Reynolds, M. H. Richmond, M. J. Waring, Eds. (Wiley, London, 1981), pp. 402-545; G. T. Tan, A. DeBlasio, A. S. Mankin, J. Mol. Biol. 261, 222 (1996).
26. D. Vazquez, Inhibitors of Protein Biosynthesis (Springer-Verlag, Berlin, 1979); E. Cundliffe, Antibiotic Inhibitors of Ribosome Function, E. F. Gale, E. Cundliffe, P. E. Reynolds, M. H. Richmond, M. J. Waring, Eds. (Wiley, London, 1981), pp. 402-545; G. T. Tan, A. DeBlasio, A. S. Mankin, J. Mol. Biol. 261, 222 (1996).
27. D. Vazquez, Inhibitors of Protein Biosynthesis (Springer-Verlag, Berlin, 1979); E. Cundliffe, Antibiotic Inhibitors of Ribosome Function, E. F. Gale, E. Cundliffe, P. E. Reynolds, M. H. Richmond, M. J. Waring, Eds. (Wiley, London, 1981), pp. 402-545; G. T. Tan, A. DeBlasio, A. S. Mankin, J. Mol. Biol. 261, 222 (1996).
28. T. P. Hausner, U. Geigenmüller, K. H. Nierhaus, J. Biol. Chem. 263, 13103 (1988).
29. The lowest concentration of sparsomycin needed to almost suppress peptide bond formation with the naked rRNA. 3 mM, was much higher than the concentrations required in the in vitro system containing intact ribosomes, which are on the order of less than 100 μM. A likely explanation is that the binding of antibiotics affecting ribosomes probably requires some of the ribosomal proteins. This likelihood is supported by the result obtained previously (8). In the conventional puromycin reaction with intact ribosomes which follows the formation of a single peptide bond uncoupled from the numerous other processes of translation [R. R. Traut and R. E. Monro, J. Mol. Biol. 10, 63 (1964)], 0.5 mM was a sufficient concentration of puromycin for use as an aminoacyltRNA analog in peptide transfer; however, even higher concentrations of the antibiotic had no effect in the previous system with naked Z3S rRNA.
30. Although E. coli SS rRNA is essential for reconstituting the active 50S subunit, 5S rRNA had no effect on peptide bond formation with the naked 23S rRNA or its domains (28). The most likely explanation is that 5S participates not in peptide transfer but in association of the subunit or translocation, or both, probably cooperating with ribosomal proteins. By analogy with known interactions of 5S and 18S rRNA in eukaryotes, it has been proposed that 5S rRNA is involved in an interaction with 16S rRNA in the ribosome function [A. A. Azad, Nucleic Acids Res. 7, 1913 (1979)].
31. It has been recently reported that modifications of E. coli 23S rRNA are essential for peptide bond formation between N-acetylmethionine-(3′)ACCAAC(5′) and puromycin [R. Green and H. F. Noller, RNA 2, 1011 (1996)]; by contrast, unmodified 23S rRNA can work in the system with AcPhe-tRNA and Phe-tRNA, but our unpublished data revealed that the system was abolished by replacement of Phe-tRNA with puromycin. These findings imply that the role of modified nucleotides is not peptide transfer itself but interaction with the antibiotic.
32. W. Stiege, C. Glotz, R. Brimacombe, Nucleic Acids Res. 11, 1687 (1983); D. Moazed and H. F. Noller, Biochimie 69, 879 (1987).
33. W. Stiege, C. Glotz, R. Brimacombe, Nucleic Acids Res. 11, 1687 (1983); D. Moazed and H. F. Noller, Biochimie 69, 879 (1987).
34. I. G. Wool, A. Glück, Y. Endo, Trends Biochem. Sci. 17, 266 (1992).
35. J. A. Doudna and T. R. Cech, RNA 1, 36 (1995).
36. G. van der Horst, A. Christian, T. Inoue, Proc. Natl. Acad. Sci. U.S.A. 88, 184 (1991).
37. R. H. Jacobson, X. J. Zhang, R. F. DuBose, B. W. Matthews, Nature 369, 761 (1994).
38. K. Shiba and P. Schimmel, Proc. Natl. Acad. Sci. U.S.A. 89, 1880 (1992).
39. I. Nitta and K. Watanabe, data not shown.
40. Using an in vitro mutagenesis kit (Bio-Rad), which follows the conventional Kunkel's method, plasmids carrying 235 rRNA genes mutated in the first region were prepared from pEC23SN. On the other hand, using PCR, 23S rRNA genes in the second region possessing mutations were amplified from plasmids kindly given by C. M. T. Spahn, and were embedded into pUC119. All the mutant genes were confirmed by dideoxy sequencing. Domain V genes mutated in both regions were obtained from these plasmids by PCR amplification, with the forward primer carrying the class III T7 promoter fused with the 5′ end of domain V, and were in vitro transcribed with T7 RNA polymerase.
41. Sparsomycin was a gift from K. Igarashi. Neomycin sulfate, a mixture of 85% neomycin B and 15% neomycin C, was obtained from Sigma. Stock solutions of 100 mM antibiotics were prepared so as to have a pH of 7.5 by adjustment with KOH just before use. Before the reaction, the domain V transcript was treated with the antibiotic at 0°C for 15 min. The slight activities observed with 10 μM of domains other than V were unaffected by either sparsomycine or neomycin (28).
42. We thank K. H. Nierhaus and C. M. T. Spahn for providing plasmids containing mutant 23S rRNA genes and critical reading of the manuscript, K. Igarashi for sparsomycin, and H. F. Noller and W. H. McClain for advice on recovering the products from TLC plates and their analysis. This work was supported by a Grant-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Science, Sports and Culture (Japan) and by the Japan Society for the Promotion of Science under the "Research for the Future" program.