This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xu, W.
Right arrow Articles by Murgola, E. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xu, W.
Right arrow Articles by Murgola, E. J.

 Previous Article  |  Next Article 

Journal of Bacteriology, February 2002, p. 1200-1203, Vol. 184, No. 4
0021-9193/01/$04.00+0     DOI: 10.1128/jb.184.4.1200-1203.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Mutations in the GTPase Center of Escherichia coli 23S rRNA Indicate Release Factor 2-Interactive Sites

Wenbing Xu,,{dagger} Frances T. Pagel, and Emanuel J. Murgola*

Department of Molecular Genetics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030

Received 15 June 2000/ Accepted 14 November 2001


arrow
ABSTRACT
 
Mutations in the GTPase center of Escherichia coli 23S rRNA were characterized in vivo as UGA-specific nonsense suppressors. Some site-directed mutations did not exhibit suppressor activity and were interspersed among suppressor mutations. Our results demonstrate the involvement of the two adjacent loops of this conserved rRNA structure in UGA-dependent translation termination and, taken with previous in vitro analyses and with consideration of the crystal structure of the GTPase center RNA, indicate that nucleotides 1067, 1093, 1094, and 1095 are sites of interaction with release factor 2.


arrow
TEXT
 
Although it is well established that rRNAs play an active role in every aspect of translation, it is still not clear which sites or structures function specifically in the termination stage, participating, for example, in interactions with peptide chain release factors, recognition of the termination codons, transmission of the termination signal to the hydrolytic center, or hydrolysis of peptidyl-tRNA.

We previously described two mutations that cause readthrough specifically at UGA codons, one at bulge nucleotide 1054 in 16S rRNA (13), the other at loop nucleotide 1093 in 23S rRNA (9). In vitro, both are defective in UGA-dependent hydrolysis of ribosome-bound peptidyl-tRNA (1) and are defective in the binding of release factor 2 (RF2)(2), which functions specifically at the UGA termination codon. The 23S rRNA site, nucleotide (nt) 1093, is a virtually universally conserved nucleotide in the highly conserved structure referred to historically as the GTPase center (GC), although other names have been used. Participation of this region in elongation is well documented (18), and it has also been implicated in the initiation and termination stages of polypeptide synthesis (14).

To elucidate the functional interactions of the GC RNA during peptide chain termination, we searched for more potentially termination-defective mutations in that region of 23S rRNA, based on the demonstrated (1, 2) usefulness of screening for codon-specific nonsense suppressors, that is, suppressors that work at one or two but not all three of the termination codons and that do not suppress related missense mutations.

We subjected a GC-containing segment (SacI to I-CeuI) of an rrnB operon cloned in plasmid pNO2680 (12) to random PCR mutagenesis and screened for codon-specific suppressors on glucose minimal medium. In a preliminary report (12), the initial screen revealed only suppressors of UGA, located only in the GC, and all were UGA specific. Here we report the further characterization of those mutants, of others found in similar screens, and of mutations made by site-directed mutagenesis.

After the randomly mutagenized fragments were subcloned, the plasmid population was used to transform two Trp auxotrophs, trpA(UGA15) and trpA(UGA211). Each contained a pcnB mutation, which lowers the copy number of ColE1-type plasmids, to allow rRNA mutations that, when highly expressed, might cause the cells to grow more slowly or not at all to be obtained. The validity of that concern was verified (Table 1). The ampicillin-resistant transformants were screened for Trp+ colonies whose prototrophy depended on a plasmid-associated suppressor. The SacI to I-CeuI fragment from each plasmid-associated suppressor was subcloned into a wild-type plasmid (pNO2680) to verify the location of the new suppressors in the mutagenized segment and for characterization of the suppression and growth properties. DNA sequence analysis revealed that all of those suppressor mutations (Table 1, footnote e) were located within a short region of the mutagenized segment, the highly conserved GC region of 23S rRNA (Fig. 1), suggesting that specific nucleotides in this structure are particularly important for termination.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Nonsense suppression and temperature-dependent growth of GTPase center mutants



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 1. Detail of GC RNA. The apparent RF2-interactive sites (white letters on solid black circles) were designated as such on the basis of (i) in vivo analyses of UGA suppressor mutations (this study); (ii) in vitro analyses of the G1093A mutation (1, 2); (iii) in vitro studies (16) of the binding to mutant GC RNA of thiostrepton, which inhibits RF activity (4); and (iv) considerations from the crystal structure of the GTPase center RNA complexed with protein L11 (5, 6, 19).

The mutations identified in the screens occurred at five nucleotide positions in 23S rRNA, nt 1067, 1093, 1094, 1095/6, and 1097; though we obtained all four types of mutations at 1093, only one type was found at the other four positions (Table 1). Therefore we constructed the other mutations at those positions as well as at 1098 for the analysis of the closing G·A base pair of the 1093 to 1098 hexaloop (22).

Mutant plasmids were tested for their ability to suppress UGA and UAG mutations at codon positions 15, 115, 211, and 243 in trpA as well as UAA at 15, 115, and 211. In the PcnB- strains, all of the suppressor-plus mutations were UGA specific, failing to suppress UAA or UAG (or any of the missense mutations). The UGA results are presented in Table 1. With the trpA mutations employed, there was no bias against suppression of the other two termination codons: the UAG and UAA mutations tested were suppressible by amber and/or ocher suppressor tRNAs, as well as by several rRNA nonsense suppressors.

The mutations that suppressed UGA at trpA position 211 are particularly informative, since we also tested several missense mutations at 211, mutations that differ from UGA by one nucleotide, including UGG. The fact that they suppressed neither the two other termination codons at the same position nor missense mutant codons differing from UGA by a single base (AGA, CGA, UGU, UGC, and UGG) suggests that an effect of the rRNA mutations on general translational accuracy is not the basis for the UGA readthrough. It could be suggested that the mutations lead to selective enhancement of misreading by tRNAs whose decoding specificity differs from UGA by one base but not by tRNAs similarly related to UAA and UAG. However, the failure of mutants to suppress UGG211 while suppressing UGA211 is particularly significant in this regard, since, in Escherichia coli, the tRNAs that read UGA-related codons ending in A also read the corresponding codon ending in G.

To examine the effects of the suppressor mutations on cell growth, we introduced the suppressor plasmids into the PcnB+ strain AL1 (3). Transformants were selected on Luria-Burrous (LB) agar plates containing ampicillin (100 µg/ml). In some cases, stable transformants were obtained only in the presence of plasmid pcI857 (9), which contains a temperature-sensitive bacteriophage lambda repressor gene that can regulate the rrnB promoter in pNO2680. The results, presented in Table 1, show that all of the suppressors exhibited temperature-dependent growth inhibition, most strikingly on supplemented minimal medium.

Quantification of the percentages of mutant rRNAs in different fractions of ribosomes indicated clearly that the failure of the suppressor-negative mutants to exhibit mutant phenotypes was not simply because their mutant RNAs were not incorporated sufficiently into the ribosomes (data not shown).

Deletion of any one of the nucleotides in the highly conserved hexaloop (nt 1093 to 1098) resulted in very strong suppression and growth inhibition, suggesting that decreasing the size of the loop interferes with its functions or interactions. While the base substitutions at 1093, 1094, and 1095 also resulted in mutant phenotypes, substitutions at 1096, 1097, and 1098 did not. The mutant phenotypes of one exception, A1098C, are consistent with the deletion results and are attributable to formation of a strong base pair (G-C) with G1093, in place of the G·A base pair, resulting in a smaller than normal loop (22). We conclude that, in addition to the identity of nt 1093, 1094, and 1095 (as well as the identity of nt 1067 in the adjacent loop), the size of the hexaloop is important for the function of that structure in translation termination. Furthermore, we provide evidence here that the "U-turn" structure found at nt 1094 to 1096 in the hexaloop (5, 19) is critical for normal termination. Indeed, the base substitution mutations at nt 1094 were comparable to the deletion mutations, exhibiting stronger mutant phenotypes than did the base substitutions at the other nucleotide positions (Table 1). Finally, the transversions at and deletion of nt 1094 caused stronger mutant phenotypes than the transition substitution. These differences suggest a critical functional role of nt 1094 as the first nucleotide in the U-turn structure.

Deletions such as {Delta}(1079-1082), at sites implicated in the higher-order structure of the GC (6, 10), resulted in severe temperature-conditional lethality but no suppression at the permissive temperature, indicating that the suppression observed with other mutations was due to alteration of specific interactions for termination rather than simply disruption of higher-order structures in the region. Conversely, the two UGA-suppressing base substitutions at nt 1067 and 1095 gave no evidence of higher-order structural changes within the GC RNA (16).

Nonsense suppression caused by deletion of nt A1067 or A1095 is consistent with the evidence that those two nucleotides are directly involved in the binding to ribosomes of thiostrepton, a peptide antibiotic that inhibits release factor-mediated termination in an in vitro termination assay system (4, 16). It has been shown by others that nt A1095 and A1067 are necessary for the binding of thiostrepton to ribosomes and that, specifically, base changes at nt 1067 and 1095 result in decreased binding of thiostrepton (16).

We constructed those mutants and tested them for growth defects and readthrough (Table 2). For the three base substitutions at each position, the order of increasing non-wild-type phenotypes (that is, UGA suppression and high-temperature growth inhibition) corresponded to the order of decreasing thiostrepton binding observed by Rosendahl and Douthwaite (16). These results demonstrate that nt 1067 and 1095 are necessary for normal peptide chain termination in vivo and suggest that RF2 may bind to the GC RNA in a manner similar to thiostrepton, which inhibits RF2- (and also RF1-) dependent termination. nt 1093 has been implicated in RF2 binding by the results of in vitro termination assays with one of the mutations at that position, A1093 (1, 2). These results, together with the severity of the phenotypes resulting from base substitutions at nt 1094, support the suggestion that the four identified nucleotides, nt 1067, 1093, 1094, and 1095, are RF2-interactive sites (Fig. 1).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Correlation between UGA suppression and thiostrepton affinity for mutations at nt 1067 and 1095 of E. coli 23S rRNA

Crystallographic analyses of the isolated L11-GC RNA complexes (5, 19) showed that the 1093-1098 and 1065-1073 loops are situated very close to each other, as proposed in earlier studies by Garrett, Douthwaite, and colleagues (7, 15, 16). Each of two of the termination-interactive nucleotides, 1067 and 1095, is the middle nucleotide of a U-turn, which causes them to bulge out, be exposed to solvent, and to come into close proximity to each other (5, 19). Those nucleotides, as well as 1093 and 1094, are entirely uncovered by L11, except for a possible occasional interaction with the N-terminal portion of L11 (19). It has been pointed out (5, 6) that the high conservation of the exposed bases at 1067 and 1095 is difficult to rationalize simply in terms of the structure of the GC RNA, suggesting their interaction with other translational molecules.

Considering those observations along with the results presented here of our functional studies in vivo, as well as the in vitro assays of the ribosomes from one of the mutants (1, 2), it seems plausible that both loops interact with part of RF2. It is not clear whether nt 1093 and 1094 interact directly with RF2 or influence the interaction of RF2 with nt 1067 and 1095 by altering the U-turn and hence the precise configurations of the bulged 1067 and 1095 nucleotides. However, it has been noted (5, 6) that from a structural standpoint, it is difficult to understand why A at 1093 does not substitute for G with impunity.

The suppression caused by the mutations in the GC is UGA preferential, indicating that RF1 does not interact with the two loops of the GC. Despite some structural similarity to RF2, RF1 must bind to the ribosome and function differently from RF2. Limitation in vivo of the amount of L11 protein available for assembly into ribosomes, as well as complete knockout of the chromosomal L11 gene, results in UAG-specific readthrough (11, 21; N. Van Dyke, W. Xu, and E. J. Murgola, submitted for publication), consistent with in vitro experiments with L11-deficient ribosomes that demonstrated the requirement for L11 in UAG-dependent termination (17). The fact that L11 inhibits somewhat RF2-dependent termination at UGA (11, 17, 21; N. Van Dyke et al., submitted) is consistent with the strong hydroxyl radical cleavages, generated from L11 amino acid residue 19, that were observed with nt 1066, 1067, 1094, and 1095 of the GC RNA (8). Furthermore, as mentioned above, thiostrepton inhibits both RF1-and RF2-dependent termination (4). The binding of thiostrepton to the GC RNA is stimulated by the NH2-terminal domain of L11 (20), and it has been proposed that the antibiotic also binds to ribosomal protein L11 (19). Therefore, it may be that thiostrepton interferes with UAG termination by way of an effect on RF1 through the amino terminus of L11, while affecting UGA termination by competing with RF2 for the RF2-interactive nt 1067, 1093, 1094, and 1095.

In summary, based on our results and other considerations presented here, in particular the fact that not only the 23S mutation G1093A but also the 16S mutation C1054A have been shown in vitro to be defective in the binding of RF2 to the ribosome (1, 2), we conclude that when RF2 is bound to the ribosome in the A site, it is in contact with both nt 1054 of 16S rRNA and the 1067 and 1095 loops of 23S rRNA. This conclusion is further supported by reflections on the recent 5.5-Å crystal structure of the 70S ribosome, in which the authors point out the proximity and likely contact of protein L11 and the loops of the GC RNA with the elbow region of an A-site tRNA and of the bulged base C1054 of 16S rRNA with the apex of the A-site tRNA anticodon loop (23). The relevance of those observations here is that, to the extent that RF2 may "mimic" an A-site tRNA, it too can be in contact with both C1054 and the tips of the loops of the GC of 23S rRNA. In any case, it appears that the GC, as part of the RF2 ribosome-binding site, plays a role in the transmission of a signal from the termination codon to the "hydrolytic center."


arrow
ACKNOWLEDGMENTS
 
We are grateful to W. J. Pagel for incisive editorial comments, to Alexey Arkov, Natalya Chernyaeva Van Dyke, and Klas Hedenstierna for helpful discussions, and to Madhu Kumar for technical assistance.

This work was supported by a grant to E.J.M. from the National Institute of General Medical Sciences (GM21499). The automated DNA sequencing facility at the M. D. Anderson Cancer Center was supported by grant CA16672 from the National Cancer Institute.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Molecular Genetics (Box 11), M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030. Phone: (713) 792-8939. Fax: (713)-794-4295. E-mail: mannyj{at}mdanderson.org. Back

{dagger} Present address: P.O. Box 382009, Cambridge, MA 02238. Back


arrow
REFERENCES
 
    1
  1. Arkov, A. L., D. V. Freistroffer, M. Ehrenberg, and E. J. Murgola. 1998. Mutations in RNAs of both ribosomal subunits cause defects in translation termination. EMBO J. 17:1507-1514.[CrossRef][Medline]
    2. 2
  2. Arkov, A. L., D. V. Freistroffer, M. Yu. Pavlov, M. Ehrenberg, and E. J. Murgola. 2000. Mutations in conserved regions of ribosomal RNAs decrease the productive association of peptide-chain release factors with the ribosome during translation termination. Biochimie 82:671-682.[Medline]
    3. 3
  3. Arkov, A. L., A. Mankin, and E. J. Murgola. 1998. An rRNA fragment and its antisense can alter decoding of genetic information. J. Bacteriol. 180:2744-2748.[Abstract/Free Full Text]
    4. 4
  4. Brot, N., W. P. Tate, C. T. Caskey, and H. Weissbach. 1974. The requirement for ribosomal proteins L7 and L12 in peptide chain termination. Proc. Natl. Acad. Sci. USA 71:89-92.[Abstract/Free Full Text]
    5. 5
  5. Conn, G. L., D. E. Draper, E. E. Lattman, and A. G. Gittis. 1999. Crystal structure of a conserved ribosomal protein-RNA complex. Science 284:1171-1174.[Abstract/Free Full Text]
    6. 6
  6. Draper, D. E., G. L. Conn, A. G. Gittis, D. GuhaThakurta, E. E. Lattman, and L. Reynaldo. 2000. RNA tertiary structure and protein recognition in an L11-RNA complex, p. 105-114. In R. A. Garrett, S. R. Douthwaite, A. Liljas, A. T. Matheson, P. B. Moore, and H. F. Noller (ed.), The ribosome: structure, function, antibiotics, and cellular interactions. ASM Press, Washington, D.C.
    7. 7
  7. Egebjerg, J., S. R. Douthwaite, A. Liljas, and R. A. Garrett. 1990. Characterization of the binding sites of protein L11 and the L10.(L12)4 pentameric complex in the GTPase domain of 23S ribosomal RNA from Escherichia coli. J. Mol. Biol. 213:275-288.[CrossRef][Medline]
    8. 8
  8. Holmberg, L., and H. F. Noller. 1999. Mapping the ribosomal RNA neighborhood of protein L11 by directed hydroxyl radical probing. J. Mol. Biol. 289:223-233.[CrossRef][Medline]
    9. 9
  9. Jemiolo, D. K., F. T. Pagel, and E. J. Murgola. 1995. UGA suppression by a mutant RNA of the large ribosomal subunit. Proc. Natl. Acad. Sci. USA 92:12309-12313.[Abstract/Free Full Text]
    10. 10
  10. Lu, M., and D. E. Draper. 1994. Bases defining an ammonium and magnesium ion-dependent tertiary structure within the large subunit ribosomal RNA. J. Mol. Biol. 244:572-585.[CrossRef][Medline]
    11. 11
  11. Murgola, E. J., A. L. Arkov, N. S. Chernyaeva, K. O. F. Hedenstierna, and F. T. Pagel. 2000. rRNA functional sites and structures for peptide chain termination, p. 509-518. In R. A. Garrett, S. R. Douthwaite, A. Liljas, A. T. Matheson, P. B. Moore, and H. F. Noller (ed.), The ribosome: structure, function, antibiotics, and cellular interactions. ASM Press, Washington, D.C.
    12. 12
  12. Murgola, E. J., F. T. Pagel, K. A. Hijazi, A. L. Arkov, W. Xu, and S. Q. Zhao. 1995. Variety of nonsense suppressor phenotypes associated with mutational changes at conserved sites in Escherichia coli ribosomal RNA. Biochem. Cell Biol. 73:925-931.[Medline]
    13. 13
  13. Pagel, F. T., S. Q. Zhao, K. A. Hijazi, and E. J. Murgola. 1997. Phenotypic heterogeneity of mutational changes at a conserved nucleotide in 16S ribosomal RNA. J. Mol. Biol. 267:1113-1123.[CrossRef][Medline]
    14. 14
  14. Porse, B. T., and R. A. Garrett. 1995. Ribosomal mechanics, antibiotics, and GTP hydrolysis. Cell 97:423-426.
    15. 15
  15. Rosendahl, G., and S. Douthwaite. 1993. Ribosomal proteins L11 and L10.(L12)4 and the antibiotic thiostrepton interact with overlapping regions of the 23S rRNA backbone in the ribosomal GTPase centre. J. Mol. Biol. 234:1013-1020.[CrossRef][Medline]
    16. 16
  16. Rosendahl, G., and S. Douthwaite. 1994. The antibiotics micrococcin and thiostrepton interact directly with 23S rRNA nucleotides 1067A and 1095A. Nucleic Acids Res. 22:357-363.[Abstract/Free Full Text]
    17. 17
  17. Tate, W. P., H. Schulze, and K. H. Nierhaus. 1983. The Escherichia coli ribosomal protein L11 suppresses release factor 2 but promotes the release factor 1 activities in peptide chain termination. J. Biol. Chem. 259:12816-12820.
    18. 18
  18. Thompson, J. 1996. Ribosomal RNA, translocation, and elongation factor-associated GTP hydrolysis, p. 311-325. In R. A. Zimmermann and A. E. Dahlberg (ed.), Ribosomal RNA: structure, evolution, processing, and function in protein biosynthesis. CRC Press, Boca Raton, Fla.
    19. 19
  19. Wimberly, B. T., R. Guymon, J. P McCutcheon, S. W. White, and V. Ramakrishnan. 1999. A detailed view of a ribosomal active site: the structure of the L11-RNA complex. Cell 97:491-502.[CrossRef][Medline]
    20. 20
  20. Xing, Y., and D. E. Draper. 1996. Cooperative interactions of RNA and thiostrepton antibiotic with two domains of ribosomal protein L11. Biochemistry 35:1581-1588.[CrossRef][Medline]
    21. 21
  21. Xu, W. 1998. Ph.D. thesis. The University of Texas at Houston Graduate School of Biomedical Sciences, Houston, Tex.
    22. 22
  22. Xu, W., and E. J. Murgola. 1996. Functional effects of mutating the closing G·A base-pair of a conserved hairpin loop in 23S ribosomal RNA. J. Mol. Biol. 264:407-411.[CrossRef][Medline]
    23. 23
  23. Yusupov, M. M., G. Zh. Yusupova, A. Baucom, K. Lieberman, T. N. Earnest, J. H. D. Cate, and H. F. Noller. 2001. Crystal structure of the ribosome at 5.5 Å resolution. Science 292:883-896.

Journal of Bacteriology, February 2002, p. 1200-1203, Vol. 184, No. 4
0021-9193/01/$04.00+0     DOI: 10.1128/jb.184.4.1200-1203.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Scarlett, D.-J. G., McCaughan, K. K., Wilson, D. N., Tate, W. P. (2003). Mapping Functionally Important Motifs SPF and GGQ of the Decoding Release Factor RF2 to the Escherichia coli Ribosome by Hydroxyl Radical Footprinting. IMPLICATIONS FOR MACROMOLECULAR MIMICRY AND STRUCTURAL CHANGES IN RF2. J. Biol. Chem. 278: 15095-15104 [Abstract] [Full Text]  
  • Arkov, A. L., Hedenstierna, K. O. F., Murgola, E. J. (2002). Mutational Evidence for a Functional Connection between Two Domains of 23S rRNA in Translation Termination. J. Bacteriol. 184: 5052-5057 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xu, W.
Right arrow Articles by Murgola, E. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xu, W.
Right arrow Articles by Murgola, E. J.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»