JOURNAL
OF
BACTERIOLOGY, Feb. 1992, P. 1333-1338
Vol. 174, No. 4
0021-9193/92/041333-06$02.00/0 Copyright © 1992, American Society for Microbiology
Functional Interactions within 23S rRNA Involving the Peptidyltransferase Center STEPHEN DOUTHWAITE Department of Molecular Biology, Odense University, DK-5230 Odense M, Denmark Received 11 September 1991/Accepted 5 December 1991 A molecular genetic approach has been employed to investigate functional interactions within 23S rRNA. Each of the three base substitutions at guanine 2032 has been made. The 2032A mutation confers resistance to the antibiotics chloramphenicol and clindamycin, which interact with the 23S rRNA peptidyltransferase loop. All three base substitutions at position 2032 produce an erythromycin-hypersensitive phenotype. The 2032 substitutions were compared with and combined with a 12-bp deletion mutation in domain II and point mutations at positions 2057 and 2058 in the peptidyltransferase region of domain V that also confer antibiotic resistance. Both the domain II deletion and the 2057A mutation relieve the hypersensitive effect of the 2032A mutation, producing an erythromycin-resistant phenotype; in addition, the combination of the 2032A and 2057A mutations confers a higher level of chloramphenicol resistance than either mutation alone. 23S rRNAs containing mutations at position 2058 that confer clindamycin and erythromycin resistance become deleterious to cell growth when combined with the 2032A mutation and, additionally, confer hypersensitivity to erythromycin and sensitivity to clindamycin and chloramphenicol. Introduction of the domain II deletion into these double-mutation constructs gives rise to erythromycin resistance. The results are interpreted as indicating that position 2032 interacts with the peptidyltransferase loop and that there is a functional connection between domains II and V.
The coupling of amino acids to form proteins is an essential biological process catalyzed by the ribosomal 50S subunit. A region within domain V of the 23S rRNA, termed the peptidyltransferase center, has been conclusively shown to be involved in this process (6, 37). This region has a phylogenetically conserved secondary structure consisting of a loop formed at the junction of five helices (20). Highly conserved bases in the single-stranded loop have been shown by footprinting, photoaffinity labeling, and mutagenesis to interact with the aminoacyl terminus of tRNA (1, 26) and with antibiotics that inhibit peptide bond formation (6, 8, 17, 25, 37) (Fig. 1). Studies with one of these antibiotics, chloramphenicol, have played a major role in defining the 23S rRNA peptidyltransferase region. Chloramphenicol's only points of rRNA contact identified by footprinting are within the loop region (25), as are mutagenized nucleotides that confer drug resistance (Fig. 1). The antibiotic anisomycin, lincosamides, and macrolides bind to the same rRNA region as chloramphenicol, as shown by competition binding studies (35), overlapping drug footprints (11, 25), and multiple-drug resistance conferred by certain point mutations (16, 28) (Fig. 1). Peptidyltransferase is unlikely, however, to be exclusively supported by this region, as several lines of evidence indicate the involvement of other rRNA structures. The aminoacyl end of tRNA protects bases within the peptidyltransferase loop against chemical modification, but there are additional effects in neighboring structures (26). Resistance to erythromycin (a macrolide) is also conferred by deletion mutations in 23S rRNA domain II (12), and in tobacco chloroplasts, resistance to lincomycin (a lincosamide) is conferred by a mutation at 23S rRNA nucleotide 2032 adjacent to, but distinct from, the peptidyltransferase loop (7). A reliable secondary structure model, confirmed by phylogenetic sequence comparisons, is available for 23S rRNA (20). Presently, however, little is understood about how the regions discussed above are spatially oriented or how they
interact to form a functionally active peptidyltransferase. In the present study, each of the three nucleotide changes at position 2032 has been made in Escherichia coli 23S rRNA. The altered tolerances towards chloramphenicol, clindamycin (7-chloro-7-deoxylincomycin), and erythromycin caused by these mutations have been evaluated and compared with resistances conferred by domain II and peptidyltransferase loop mutations. Incorporation of combinations of mutations into the same 23S rRNA showed that position 2032, the peptidyltransferase loop, and a helix in domain II are functionally interconnected.
MATERIALS AND METHODS Bacterial strains and growth conditions. E. coli DH1 [FrecAl endAl gyrA96 thi-I hsdR17 (rK mK+) supE44 X] was used as the plasmid host. E. coli TG1 [A(lac-pro) supE thi hsdDS)IF' (traD36 proAB+ lacIq lacZAM15)] was employed as the host for M13 bacteriophage derivatives. The media for liquid cultures were LB for DH1 and YT medium for TG1 (27); all incubations were at 37°C. Antibiotics were purchased from Sigma, except for clindamycin (Upjohn). All plates and broth used for growing plasmid-transformed DH1 cells contained ampicillin at 50 mg/liter in the absence of other drugs and at 25 mg/liter when other drugs were added. The effect of antibiotics on growth was determined by diluting an ovemight culture of plasmid-transformed cells to approximately 10 cells per ml and spreading 50 RI directly onto antibiotic-containing agar plates. In vitro DNA manipulations. The enzymes used to digest, ligate, and phosphorylate DNA were purchased from Boehringer Mannheim, with the exception of Sequenase enzyme for dideoxy sequencing (U.S. Biochemical Corporation). Reagents for phosphorothioate mutagenesis were obtained from Amersham. All enzymes were used according to the suppliers' recommendations. Cell transformations, plasmid 1333
1334
J. BACTERIOL.
DOUTHWAITE
+
Cldr(E.C)
C-G Ansr(H) Cmr (Y)
2050
C C C G C G G I II * 1 II G G G U G CI C 2620
A G A I
G
I
C-
G( I
I
C(
U C U
A UC
U 746
G G
X-link
C
2600
tRJA* 3'-end -
\G
A4z U U\
Ac G\NAA
2590 G G AG \ UG PB-Pho tRMA (A-Site]
f
oolLtprint
-
A G
G G-
2580-U * G
PB-Phe tRNA [A- and P-sitsuI
FIG. 1. Secondary structure of the peptidyltransferase loop and adjacent structures (20). Arrows indicate the relevant cross-linked (X-link) bases (10) and the positions photoaffinity labeled by benzophenone-derivatized Phe tRNA (PB-Phe tRNA) (1) andp-azidopuromycin (6). Sites within this region where interaction of the tRNA 3'-end causes altered reactivities to chemical reagents (tRNA 3-end footprint) are shown (26). Positions where mutations confer drug resistance are encircled. Abbreviations: Ans, anisomycin; Cam, chloramphenicol; Cld, clindamycin; Ery, erythromycin; Lnc, lincomycin; Spr, spiramycin; r, resistant; hs, hypersensitive; Cc, Chlamydomonas chloroplasts (18); E.c, E. coli; H, halobacteria; Hu, human mitochondria; M, mouse mitochondria; T, Tetrahymena cytoplasmic ribosomes (33); tc, tobacco chloroplasts (7); Y, yeast mitochondria. Data for the following mutations are from the indicated references: 2057, 16; 2058U, 28; 2058G (E. coli), 36; 2058G (yeast), 30; 2062, 23; 2447A, 14; 2447C, 21; 2451, 22; 2452 (halobacteria), 21 and 23; 2452 (human), 2; 2452 (mouse), 29; 2453, 21; 2503 (E. coli), 37; 2503 (yeast), 14; 2504 (human), 2 and 21; 2504 (mouse), 3; and 2611 (yeast), 31.
DNA preparation, and gel electrophoresis were performed by standard procedures (27). Site-directed mutagenesis and plasmid construction. All plasmids used in this study are formed from plasmid pKK3535, which is a pBR322 derivative encoding ampicillin resistance, with the entire rmnB operon (Fig. 2) inserted into the tetracycline resistance gene (4). Mutations at position 2032 were introduced by the phosphorothioate method (34) after the deoxyoligonucleotide 5'-CTCTGCATNTTCAC AGCG-3' (where N = G, A, or T) was hybridized to the 2.7-kbp SalI-BamHI fragment of the 23S RNA gene cloned into M13 mpl9. Recloning of the mutagenized 23S rRNA gene SalI-BamHI fragments back into their original context in the rnB operon was carried out via plasmid pSK41 (Fig. 2), thereby circumventing the problem of duplicated restriction sites (13). The 23S RNA 2057G-+A and 2058A-T mutations on plasmids coding for the E. coli rnH operon (16, 28) were excised on 4.1-kb SalI-BamHI fragments, which were ligated with the 23S rRNA gene Sall site in rmnB to form the rrnB-rrnH fusions shown in Fig. 2b. The formation of the 2058G mutation and A12 (a 12-bp deletion of nucleotides 1219 to 1230) has been previously described (13, 36). 23S
rRNA genes with multiple mutations were constructed by combining fragments at the unique Sall and/or KspI sites in plasmid pSK41 (Fig. 2). In plasmids with 2058G, rnB is truncated at the BclI site and the rnB-rmnH operon fusions are longer than rmnB; these differences in plasmid size did not affect rates of cell growth on ampicillin (Table 1). Certain combinations of mutations proved difficult to isolate and had a slow-growth phenotype when expressed from the rnB promoters. This problem was overcome by using plasmid pLK35 (12) to modulate rRNA gene expression from the lambda PL promoter (19) (Fig. 1). The structures of all plasmids were verified by restriction enzyme analysis and by direct sequencing of the mutagenized regions
(5). RESULTS Effects of mutations at 23S RNA nucleotide 2032 on antibiotic tolerance. Cells transformed with plasmids encoding 23S RNA with a G-to-A transition at position 2032 grew readily on clindamycin at 25 mg/liter, whereas this level of drug inhibited growth of cells with the wild-type G or with C or U at this position (Table 1). There was no difference in growth
VOL. 174, 1992
23S rRNA INTERACTIONS INVOLVING PEPTIDYLTRANSFERASE
1
2
16S rRNA
tRNAGk
T1 T2
23S rRNA
I~~~~ ~ -.1Xbo
1335
FL-
5 1 1
K
Bom Hi
Sph
Sol
Ksp
rRNA Bcl
Bom HI
b 1
2
16S rRNA
tRNAu
Bom HI
0
Xba
1
2
3
T1 T2
23S rRNA
Sp Sol Sph
4
SS rRNA
K sp
5
6
rAn Bam Hi I
7
8
9 kbp
FIG. 2. All plasmids used in this study are composed of an E. coli rRNA operon inserted into the tetracycline resistance gene of pBR322 from which the 0.9-kb NaeI-BamHI fragment has been removed. All plasmids confer pBR322-encoded ampicillin resistance. (a) Plasmid pKK35 (Table 1) contains wild-type rrnB on a 7.5-kb BamHI fragment; pKK2032A, pKK2032C, and pKK2032T have the same structure with the 23S RNA point mutations indicated. Plasmid pKKA12 has a 12-bp deletion overlapping the SphI site (see text). Plasmid constructs with the 2058G mutation are truncated at the BclI site. In plasmid pLK35 and the derivatives thereof listed in Table 2, the rrnB promoters have been replaced by the lambdapL promoter (12, 19). Plasmid pSK41 contains the 4.3-kb XbaI-BamHI 3' portion of rrnB (13). (b) Structure of the r-nB-r-nH fusions on a 8.9-kb BamHI fragment. The 5' portion up to SalI is derived from rmnB, and the 3' portion is from rnH constructs containing the 2057A and 2058T mutations. The KspI site at position 2048 in the 23S RNA gene was used to combine the 2032 and 2057 (or 2058) mutations.
rates of wild-type, 2032C, or 2032U mutants at permissive
levels of clindamycin (10 mg/liter). E. coli is naturally resistant to lincomycin; therefore, the growth of the mutants on this drug was not assayed. Cells formed distinct colonies after 22 h on erythromycin at a concentration of 30 mg/liter, whereas there was no growth under these conditions when the G at position 2032 was changed to A, C, or U (Table 1). After prolonged incubation for 48 h, the 2032 mutants formed visible colonies, showing that this drug concentration is nonlethal. The 2032A mutation enabled E. coli to grow on chloramphenicol at 4 mg/liter, while cells with the 2032G or 2032C mutation were inhibited by 2 mg/liter. The 2032U mutation supported slow growth on 2 mg of chloramphenicol per liter. In the absence of antibiotics that bind to
the 50S ribosomal subunit, the 2032 mutations had no effect cell growth rate (Tables 1 and 2). Other 23S RNA mutations conferring resistance to chloramphenicol, lincosamides, and macrolides. An equivalent degree of chloramphenicol resistance is conferred by 2057A and 2032A, both mutations supporting growth on chloramphenicol at 4 mg/liter. The 2058G and 2058U mutations conferred a phenotype similar to that of the 2032U mutation, each providing a distinct growth advantage on chloramphenicol at 2 mg/liter, but could not support growth at 4 mg/liter. On clindamycin, the 2058G and 2058U mutations both confer greater resistance than the 2032A mutation (Table 1). The 2058G and 2058U mutations also confer higher levels of erythromycin resistance than the 2057A mutation, which in on
TABLE 1. Drug resistances conferred by plasmids on E. coli DH1 cells Growth on antibiotic
Plasmid
Ampicillin'
Chloramphenicolb
Clindamycin'
pKK35 (wild type) pKK2032A pKK2032T pKK2032C
+++ +++ +++ +++
++ +
pKK2057A pKK2058G pKK2058T pKKA12 pKK2032A + 2057A pKK2032A + 2058G(T) pKK2032A + A12
+++ +++ +++
++ + +
+ +++
+++
+
++
ND
++
+++ + ++
ND
Erythromycind
ND
++
+++ + ++ +
Cells formed clearly visible colonies after 14 (+ + +), 16 (+±), or 22 (+) h. + + +, growth at up to 6 mg/liter; + +, growth at up to 4 mg/liter; +, slow growth at 2 mg/liter;-, no growth at 2 mg/liter. ND, not determined. At 25 mg/liter, distinct colonies were observed after 20 (+++), 22 (++), or 24 (+) h; -, no growth at this drug concentration. ND, not determined. d Cells with the wild-type pKK35 plasmid formed distinct colonies after 22 h at 30 mg/liter (±) but did not grow at 50 mg/liter; -, no growth or severely retarded growth at 30 mg/liter. The A12, 2057A, and 2058 mutations support growth at 80 mg/liter, forming distinct colonies of comparable sizes after 40), 24, and 16 h (+? + +, and + + +, respectively). a
b
C
1336
J. BACTERIOL.
DOUTHWAITE
TABLE 2. Growth of cells transformed with plasmids in which rRNA genes are under control of the lambda PL promoter Growth on antibiotic Plasmid
Ampicillin'
pLK35 (wild type) pLK2032A pLK2058G pLKA12 pLK2032A + 2058G pLK2032A + A12 pLK2032A + 2058G and A12
Erythromycin"
+++ + ++ +++ +++ +++ +++ ++ +++ ++ +++
All cells grew at the same rate, forming distinct colonies in 13 h. b -, no growth at 30 mg/liter; +, growth at 30 mg/liter but no growth at 50 mg/liter; +, growth at 50 mg/liter after 36 h but no growth at 80 mg/liter; + +, growth at 50 mg/liter after 22 h and growth at 80 mg/liter after 36 h; +++ growth at 50 mg/liter after 20 h and growth at 80 mg/liter after 30 h.
turn confers better resistance than the A12 mutation (Table 1). However, when 23S rRNA was expressed from the lambdapL promoter, the A12 mutation enabled cells to grow at higher erythromycin levels than the 2058G (Table 2) and 2058U mutations. Combinations of drug resistance mutations in 23S RNA. The combination of mutations 2032A and 2057A in the same 23S RNA produced a phenotype that retains the advantageous growth characteristics of each individual mutation (Table 1). Cells grew better on chloramphenicol with both mutations than with either mutation alone; the 2032A-encoded clindamycin resistance is retained, and the 2057A-encoded erythromycin-resistant phenotype is dominant over the 2032A-induced hypersensitivity. When mutations 2032A and 2058G (or 2058U) were combined, however, cells grew slowly in the absence of 50S subunit-specific drugs and they were incapable of growing on erythromycin even after extended incubations; cells also failed to grow on chloramphenicol and clindamycin, despite the growth advantage on these drugs conferred by each of the two mutations on its own. The rate of cell growth on ampicillin was normal when 23S RNA containing the 2032A and 2058G mutations was expressed from the lambdapL promoter, but cells remained hypersensitive to erythromycin (Table 2). Introduction of the A12 mutation into plasmids either with the 2032A mutation alone or with a combination of the A2032 and G2058 mutations relieved the hypersensitivity effect and conferred an erythromycin-resistant phenotype.
DISCUSSION
The sequence of the 23S rRNA peptidyltransferase loop region depicted as single stranded in Fig. 1 is highly conserved (20). Many of these bases have been shown to interact both with the aminoacyl end of tRNA (1, 26) and with antibiotics that inhibit peptide bond formation (6, 25) (Fig. 1). The importance for rRNA function of the neighboring stem-loop (formed by nucleotides 2023 to 2040) is indicated by its presence in all 23S-like rRNAs, including the truncated versions in mitochondria (20). Nucleotide 2032 in the hairpin loop is conserved as a guanosine in all eubacteria and chloroplasts, and the first indication that this nucleotide is functionally connected with peptidyltransferase was revealed when Cseplo et al. identified an A transition mutation at this position in a lincomycin-resistant tobacco chloroplast (7). The clindamycin resistance conferred by the 2032A muta-
tion shows that the effect observed in chloroplast 23S rRNA can be extrapolated to E. coli (and most likely to all eubacteria and chloroplasts). The 2032A mutation also confers chloramphenicol resistance, emphasizing the involvement of this position in peptidyltransferase. It cannot be concluded, however, that these drugs directly interact with this position in wild-type rRNA, as the 2032C and 2032U mutations confer no resistance. The 2057G--A change confers chloramphenicol resistance (16), as do other mutations in the peptidyltransferase loop region of mitochondrial rRNA (Fig. 1). However, the latter mutations that have been reproduced at the corresponding sites in E. coli result in an inviable or slow-growth phenotype (37). The highest levels of clindamycin and erythromycin resistance observed in this study are conferred by the 2058 mutations, which is consistent with the idea that this base and its immediate vicinity form the target site recognized by the drugs used in this study (8, 25). The hypersensitivity to erythromycin in 2032 mutants was unexpected. Hypersensitivity to this drug has previously been encountered after genetic elimination of r protein Lii (reference 32 and references therein), which binds to 23S rRNA between bases 1055 and 1105 (15), and also after introduction of a mutation in 23S rRNA domain II adjacent to the site where deletions confer erythromycin resistance (9). The mutations could be affecting drug binding, or they could be enhancing the effectiveness of erythromycin in physically blocking peptide chain elongation (35) or in aggravating premature dissociation of peptidyl-tRNA from the ribosome (24). Nucleotide 2032 has been cross-linked by nitrogen mustard to 2055 (10). The proximity of these two positions suggests that the 2032A mutation could bring about its effect by altering the tertiary alignment of 2058A and/or the neighboring bases. Mutations at positions 2057 and 2058 probably also perturb the local structure. The drug resistances conferred (Table 1) show that any of these alterations on its own can be incorporated in a functionally active rRNA molecule. The combination of the 2032A mutation and the 2057A mutation is permissible and indeed advantageous in, for example, the case of the additive chloramphenicol resistance effects. However, the combination of the 2032A mutation and the 2058G or U mutation is deleterious, indicating that the bases are a part of a finely tuned mechanism which cannot accommodate these combined structural perturbations. UV and nitrogen mustard cross-links have been created between nucleotides 571 and 2031, 746 and 2613, and 979 to 984 and 2029 (10) (Fig. 1), orienting positions 2032 and 2058 in the midst of domain II. A functional connection between domains II and V based on the A12 mutation that confers erythromycin resistance has previously been reported (13). Expression of A12 23S RNA from the rnB promoters slows cell growth in the absence of erythromycin (Table 1), probably because the deletion has a side effect on 50S subunit assembly (12). However, the growth rate is normal when A12 23S RNA is expressed from the weaker lambdapL promoter (12), and this construct confers better erythromycin resistance (Table 2). The A12 mutation is dominant over the 2032A mutation, relieving the erythromycin-hypersensitive effect and conferring a drug-resistant phenotype essentially identical to that conferred by A12 alone (Tables 1 and 2). Expressed from thepL promoter, 23S rRNA with the 2032A and 2058G mutations does not cause appreciably slower growth in the absence of drugs, but erythromycin hypersensitivity is still observed (Table 2). Introduction of A12 into
23S rRNA INTERACTIONS INVOLVING PEPTIDYLTRANSFERASE
VOL. 174, 1992
this construct relieves the effect of the otherwise deleterious juxtaposition of the 2032A and 2058G mutations and produces a functionally active 23S rRNA that confers erythromycin resistance. It is evident, therefore, that a functional connection between domains II and V exists. ACKNOWLEDGMENTS Anne-Mette Wennermark is thanked for technical assistance, as are Claus Aagaard and Edel Rasmussen for art work. Ed Morgan
kindly provided rnH clones. Eric Cundliffe is thanked for advice on antibiotics, and Claus Aagaard, Mette Dam, Roger Garrett, and Gunnar Rosendahl are thanked for their comments on the manu-
script. The research was supported by The Carlsberg Foundation. REFERENCES 1. Barta, A., E. Kuechier, and G. Steiner. 1990. Photoaffinity labeling of the peptidyltransferase region, p. 358-365. In W. E. Hill, A. Dahlberg, R. A. Garrett, P. B. Moore, D. Schlessinger, and J. R. Warner (ed.), The ribosome: structure, function, and evolution. American Society for Microbiology, Washington, D.C. 2. Blanc, H., C. W. Adams, and D. C. Wallace. 1981. Different nucleotide changes in the large rRNA gene of the mitochondrial DNA confer chloramphenicol resistance on two human cell lines. Nucleic Acids Res. 9:5785-5795. 3. Blanc, H., C. T. Wright, M. J. Bibb, D. C. Wallace, and D. A. Clayton. 1981. Mitochondrial DNA of chloramphenicol-resistant mouse cells contains a single nucleotide change in the region encoding the 3' end of the large ribosomal RNA. Proc. Natl. Acad. Sci. USA 78:3789-3793. 4. Brosius, J., A. Ullrich, M. A. Raker, A. Gray, T. J. Dull, R. R. Gutell, and H. F. Noller. 1981. Construction and fine mapping of recombinant plasmids containing the rrnB ribosomal RNA operon of E. coli. Plasmid 6:112-118. 5. Chen, E. Y., and P. H. Seeburg. 1985. Supercoil sequencing: a fast and simple method for sequencing plasmid DNA. DNA 4:165-170. 6. Cooperman, B. S., C. J. Weitzmann, and C. L. Fernindez. 1990. Antibiotic probes of Escherichia coli ribosomal peptidyltransferase, p. 491-501. In W. E. Hill, A. Dahlberg, R. A. Garrett, P. B. Moore, D. Schlessinger, and J. R. Warner (ed.), The ribosome: structure, function, and evolution. American Society for Microbiology, Washington, D.C. 7. Cseplo, A., T. Etzold, J. Schell, and P. H. Schreier. 1988. Point mutation in the 23S rRNA genes of four lincomycin resistant Nicotiana plumbaginifolia mutants could provide new selectable markers for chloroplast transformation. Mol. Gen. Genet. 214:295-299. 8. Cundliffe, E. 1990. Recognition sites for antibiotics within rRNA, p. 479-490. In W. E. Hill, A. Dahlberg, R. A. Garrett, P. B. Moore, D. Schlessinger, and J. R. Warner (ed.), The ribosome: structure, function, and evolution. American Society for Microbiology, Washington, D.C. 9. Dam, M., and S. Douthwaite. Unpublished data. 10. Doring, T., B. Greuer, and R. Brimacombe. 1991. The three dimensional folding of ribosomal RNA; localization of a series of intra-RNA cross-links in 23S RNA induced by treatment of Escherichia coli 50S ribosomal subunits with bis-(2-chloroethyl)-methylamine. Nucleic Acids Res. 19:3517-3524. 11. Douthwaite, S. Unpublished data. 12. Douthwaite, S., T. Powers, J. Y. Lee, and H. F. Noller. 1989. Defining the structural requirements for a helix in 23S ribosomal RNA that confers erythromycin resistance. J. Mol. Biol. 209: 655-665. 13. Douthwaite, S., J. B. Prince, and H. F. Noller. 1985. Evidence for functional interaction between domains II and V of 23S ribosomal RNA from an erythromycin resistant mutant. Proc. Natl. Acad. Sci. USA 82:8330-8334.
1337
14. Dujon, B. 1980. Sequence of the intron and flanking exons of the mitochondrial 21S rRNA gene of yeast strains having different alleles at the omega and rib-i loci. Cell 20:185-197. 15. Egebjerg, J., S. R. Douthwaite, A. Liljas, and R. A. Garrett. 1990. Characterization of the binding sites of protein Lii and the L10(L12)4 pentameric complex in the GTPase domain of 23S ribosomal RNA from Escherichia coli. J. Mol. Biol. 213:275288. 16. Ettayebi, M., S. M. Prasad, and E. A. Morgan. 1985. Chloramphenicol-erythromycin resistance mutations in a 23S rRNA gene of Escherichia coli. J. Bacteriol. 162:551-557. 17. Gale, E. F., E. Cundliffe, P. E. Reynolds, M. H. Richmond, and M. J. Waring. 1981. The molecular basis of antibiotic action. John Wiley and Sons, London. 18. Gauthier, A., M. Turmel, and C. Lemieux. 1988. Mapping of chloroplast mutations conferring resistance to antibiotics in Chlamydomonas: evidence for a novel site of streptomycin resistance in the small subunit RNA. Mol. Gen. Genet. 214:192197. 19. Gourse, R. L., Y. Takebe, R. A. Sharrock, and M. Nomura. 1985. Feedback regulation of rRNA and tRNA synthesis and accumulation of free ribosomes after conditional expression of rRNA genes. Proc. Natl. Acad. Sci. USA 82:1069-1073. 20. Gutell, R. R., and G. E. Fox. 1988. A compilation of large subunit RNA sequences presented in structural format. Nucleic Acids Res. 16:r175-r269. 21. Hummel, H., and A. Bock. 1987. 23S ribosomal RNA mutations in halobacteria conferring resistance to the anti-80S ribosome targeted antibiotic anisomycin. Nucleic Acids Res. 15:24312443. 22. Kearsey, S. E., and I. W. Craig. 1981. Altered ribosomal RNA genes in mitochondria from mammalian cells with chloramphenicol resistance. Nature (London) 290:607-608. 23. Mankin, A. S., and R. A. Garrett. 1991. Chloramphenicol resistance mutations in the single 23S rRNA gene of the archaeon Halobacterium halobium. J. Bacteriol. 173:35593563. 24. Menninger, J. R., and D. P. Otto. 1982. Erythromycin, carbomycin, and spiramycin inhibit protein synthesis by stimulating the dissociation of peptidyl-tRNA from ribosomes. Antimicrob. Agents Chemother. 21:811-818. 25. Moazed, D., and H. F. Noller. 1987. Chloramphenicol, erythromycin, carbomycin and vernamycin B protect overlapping sites in the peptidyl transferase region of 23S ribosomal RNA. Biochimie 69:879-884. 26. Moazed, D., and H. F. Noller. 1991. Sites of interaction of the CCA end of peptidyl-tRNA with 23S rRNA. Proc. Natl. Acad. Sci. USA 88:3725-3728. 27. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 28. Sigmund, C. D., M. Ettatebi, A. Borden, and E. A. Morgan. 1988. Antibiotic resistant mutations in ribosomal RNA genes of Escherichia coli. Methods Enzymol. 164:673-690. 29. Slott, E. F., Jr., R. 0. Shade, and R. A. Lansman. 1983. Sequence analysis of mitochondrial DNA in a mouse cell line resistant to chloramphenicol and oligomycin. Mol. Cell. Biol. 3:1694-1702. 30. Sor, F., and H. Fukuhara. 1982. Identification of two erythromycin resistant mutations in the mitochondrial gene coding for the large ribosomal RNA in yeast. Nucleic Acids Res. 10:65716577. 31. Sor, F., and H. Fukuhara. 1984. Erythromycin and spiramycin resistance mutations of yeast mitochondria: nature of the rib2 locus of the large ribosomal RNA gene. Nucleic Acids Res.
12:8313-8318. 32. Stomfler, G., E. Cundliffe, M. Stoffler-Meilicke, and E. R. Dabbs. 1980. Mutants of Escherichia coli lacking ribosomal protein Lii. J. Biol. Chem. 255:10517-10522. 33. Sweeney, R., C.-H. Yao, and M.-C. Yao. 1991. A mutation in the large subunit ribosomal RNA gene of Tetrahymena confers anisomycin resistance and cold sensitivity. Genetics 127:327334.
1338
DOUTHWAITE
34. Taylor, J. W., J. Ott, and F. Eckstein. 1985. The rapid generation of oligonucleotide-directed mutations at high frequency using phosphorothioate-modified DNA. Nucleic Acids Res. 13:8764-8785. 35. V6zquez, D. 1979. Inhibitors of protein synthesis, p. 169-175. Springer-Verlag, Berlin. 36. Vester, B., and R. A. Garrett. 1987. A plasmid-coded and
J. BACTERIOL.
site-directed mutation in Eschenchia coli 23S RNA that confers resistance to erythromycin: implications for the mechanism of action of erythromycin. Biochimie 69:891-900. 37. Vester, B., and R. A. Garrett. 1988. The importance of highly conserved nucleotides in the binding region of chloramphenicol in the peptidyl transferase centre of Escherichia coli 23S ribosomal RNA. EMBO J. 7:3577-3587.
OF
BACTERIOLOGY, Feb. 1992, P. 1333-1338
Vol. 174, No. 4
0021-9193/92/041333-06$02.00/0 Copyright © 1992, American Society for Microbiology
Functional Interactions within 23S rRNA Involving the Peptidyltransferase Center STEPHEN DOUTHWAITE Department of Molecular Biology, Odense University, DK-5230 Odense M, Denmark Received 11 September 1991/Accepted 5 December 1991 A molecular genetic approach has been employed to investigate functional interactions within 23S rRNA. Each of the three base substitutions at guanine 2032 has been made. The 2032A mutation confers resistance to the antibiotics chloramphenicol and clindamycin, which interact with the 23S rRNA peptidyltransferase loop. All three base substitutions at position 2032 produce an erythromycin-hypersensitive phenotype. The 2032 substitutions were compared with and combined with a 12-bp deletion mutation in domain II and point mutations at positions 2057 and 2058 in the peptidyltransferase region of domain V that also confer antibiotic resistance. Both the domain II deletion and the 2057A mutation relieve the hypersensitive effect of the 2032A mutation, producing an erythromycin-resistant phenotype; in addition, the combination of the 2032A and 2057A mutations confers a higher level of chloramphenicol resistance than either mutation alone. 23S rRNAs containing mutations at position 2058 that confer clindamycin and erythromycin resistance become deleterious to cell growth when combined with the 2032A mutation and, additionally, confer hypersensitivity to erythromycin and sensitivity to clindamycin and chloramphenicol. Introduction of the domain II deletion into these double-mutation constructs gives rise to erythromycin resistance. The results are interpreted as indicating that position 2032 interacts with the peptidyltransferase loop and that there is a functional connection between domains II and V.
The coupling of amino acids to form proteins is an essential biological process catalyzed by the ribosomal 50S subunit. A region within domain V of the 23S rRNA, termed the peptidyltransferase center, has been conclusively shown to be involved in this process (6, 37). This region has a phylogenetically conserved secondary structure consisting of a loop formed at the junction of five helices (20). Highly conserved bases in the single-stranded loop have been shown by footprinting, photoaffinity labeling, and mutagenesis to interact with the aminoacyl terminus of tRNA (1, 26) and with antibiotics that inhibit peptide bond formation (6, 8, 17, 25, 37) (Fig. 1). Studies with one of these antibiotics, chloramphenicol, have played a major role in defining the 23S rRNA peptidyltransferase region. Chloramphenicol's only points of rRNA contact identified by footprinting are within the loop region (25), as are mutagenized nucleotides that confer drug resistance (Fig. 1). The antibiotic anisomycin, lincosamides, and macrolides bind to the same rRNA region as chloramphenicol, as shown by competition binding studies (35), overlapping drug footprints (11, 25), and multiple-drug resistance conferred by certain point mutations (16, 28) (Fig. 1). Peptidyltransferase is unlikely, however, to be exclusively supported by this region, as several lines of evidence indicate the involvement of other rRNA structures. The aminoacyl end of tRNA protects bases within the peptidyltransferase loop against chemical modification, but there are additional effects in neighboring structures (26). Resistance to erythromycin (a macrolide) is also conferred by deletion mutations in 23S rRNA domain II (12), and in tobacco chloroplasts, resistance to lincomycin (a lincosamide) is conferred by a mutation at 23S rRNA nucleotide 2032 adjacent to, but distinct from, the peptidyltransferase loop (7). A reliable secondary structure model, confirmed by phylogenetic sequence comparisons, is available for 23S rRNA (20). Presently, however, little is understood about how the regions discussed above are spatially oriented or how they
interact to form a functionally active peptidyltransferase. In the present study, each of the three nucleotide changes at position 2032 has been made in Escherichia coli 23S rRNA. The altered tolerances towards chloramphenicol, clindamycin (7-chloro-7-deoxylincomycin), and erythromycin caused by these mutations have been evaluated and compared with resistances conferred by domain II and peptidyltransferase loop mutations. Incorporation of combinations of mutations into the same 23S rRNA showed that position 2032, the peptidyltransferase loop, and a helix in domain II are functionally interconnected.
MATERIALS AND METHODS Bacterial strains and growth conditions. E. coli DH1 [FrecAl endAl gyrA96 thi-I hsdR17 (rK mK+) supE44 X] was used as the plasmid host. E. coli TG1 [A(lac-pro) supE thi hsdDS)IF' (traD36 proAB+ lacIq lacZAM15)] was employed as the host for M13 bacteriophage derivatives. The media for liquid cultures were LB for DH1 and YT medium for TG1 (27); all incubations were at 37°C. Antibiotics were purchased from Sigma, except for clindamycin (Upjohn). All plates and broth used for growing plasmid-transformed DH1 cells contained ampicillin at 50 mg/liter in the absence of other drugs and at 25 mg/liter when other drugs were added. The effect of antibiotics on growth was determined by diluting an ovemight culture of plasmid-transformed cells to approximately 10 cells per ml and spreading 50 RI directly onto antibiotic-containing agar plates. In vitro DNA manipulations. The enzymes used to digest, ligate, and phosphorylate DNA were purchased from Boehringer Mannheim, with the exception of Sequenase enzyme for dideoxy sequencing (U.S. Biochemical Corporation). Reagents for phosphorothioate mutagenesis were obtained from Amersham. All enzymes were used according to the suppliers' recommendations. Cell transformations, plasmid 1333
1334
J. BACTERIOL.
DOUTHWAITE
+
Cldr(E.C)
C-G Ansr(H) Cmr (Y)
2050
C C C G C G G I II * 1 II G G G U G CI C 2620
A G A I
G
I
C-
G( I
I
C(
U C U
A UC
U 746
G G
X-link
C
2600
tRJA* 3'-end -
\G
A4z U U\
Ac G\NAA
2590 G G AG \ UG PB-Pho tRMA (A-Site]
f
oolLtprint
-
A G
G G-
2580-U * G
PB-Phe tRNA [A- and P-sitsuI
FIG. 1. Secondary structure of the peptidyltransferase loop and adjacent structures (20). Arrows indicate the relevant cross-linked (X-link) bases (10) and the positions photoaffinity labeled by benzophenone-derivatized Phe tRNA (PB-Phe tRNA) (1) andp-azidopuromycin (6). Sites within this region where interaction of the tRNA 3'-end causes altered reactivities to chemical reagents (tRNA 3-end footprint) are shown (26). Positions where mutations confer drug resistance are encircled. Abbreviations: Ans, anisomycin; Cam, chloramphenicol; Cld, clindamycin; Ery, erythromycin; Lnc, lincomycin; Spr, spiramycin; r, resistant; hs, hypersensitive; Cc, Chlamydomonas chloroplasts (18); E.c, E. coli; H, halobacteria; Hu, human mitochondria; M, mouse mitochondria; T, Tetrahymena cytoplasmic ribosomes (33); tc, tobacco chloroplasts (7); Y, yeast mitochondria. Data for the following mutations are from the indicated references: 2057, 16; 2058U, 28; 2058G (E. coli), 36; 2058G (yeast), 30; 2062, 23; 2447A, 14; 2447C, 21; 2451, 22; 2452 (halobacteria), 21 and 23; 2452 (human), 2; 2452 (mouse), 29; 2453, 21; 2503 (E. coli), 37; 2503 (yeast), 14; 2504 (human), 2 and 21; 2504 (mouse), 3; and 2611 (yeast), 31.
DNA preparation, and gel electrophoresis were performed by standard procedures (27). Site-directed mutagenesis and plasmid construction. All plasmids used in this study are formed from plasmid pKK3535, which is a pBR322 derivative encoding ampicillin resistance, with the entire rmnB operon (Fig. 2) inserted into the tetracycline resistance gene (4). Mutations at position 2032 were introduced by the phosphorothioate method (34) after the deoxyoligonucleotide 5'-CTCTGCATNTTCAC AGCG-3' (where N = G, A, or T) was hybridized to the 2.7-kbp SalI-BamHI fragment of the 23S RNA gene cloned into M13 mpl9. Recloning of the mutagenized 23S rRNA gene SalI-BamHI fragments back into their original context in the rnB operon was carried out via plasmid pSK41 (Fig. 2), thereby circumventing the problem of duplicated restriction sites (13). The 23S RNA 2057G-+A and 2058A-T mutations on plasmids coding for the E. coli rnH operon (16, 28) were excised on 4.1-kb SalI-BamHI fragments, which were ligated with the 23S rRNA gene Sall site in rmnB to form the rrnB-rrnH fusions shown in Fig. 2b. The formation of the 2058G mutation and A12 (a 12-bp deletion of nucleotides 1219 to 1230) has been previously described (13, 36). 23S
rRNA genes with multiple mutations were constructed by combining fragments at the unique Sall and/or KspI sites in plasmid pSK41 (Fig. 2). In plasmids with 2058G, rnB is truncated at the BclI site and the rnB-rmnH operon fusions are longer than rmnB; these differences in plasmid size did not affect rates of cell growth on ampicillin (Table 1). Certain combinations of mutations proved difficult to isolate and had a slow-growth phenotype when expressed from the rnB promoters. This problem was overcome by using plasmid pLK35 (12) to modulate rRNA gene expression from the lambda PL promoter (19) (Fig. 1). The structures of all plasmids were verified by restriction enzyme analysis and by direct sequencing of the mutagenized regions
(5). RESULTS Effects of mutations at 23S RNA nucleotide 2032 on antibiotic tolerance. Cells transformed with plasmids encoding 23S RNA with a G-to-A transition at position 2032 grew readily on clindamycin at 25 mg/liter, whereas this level of drug inhibited growth of cells with the wild-type G or with C or U at this position (Table 1). There was no difference in growth
VOL. 174, 1992
23S rRNA INTERACTIONS INVOLVING PEPTIDYLTRANSFERASE
1
2
16S rRNA
tRNAGk
T1 T2
23S rRNA
I~~~~ ~ -.1Xbo
1335
FL-
5 1 1
K
Bom Hi
Sph
Sol
Ksp
rRNA Bcl
Bom HI
b 1
2
16S rRNA
tRNAu
Bom HI
0
Xba
1
2
3
T1 T2
23S rRNA
Sp Sol Sph
4
SS rRNA
K sp
5
6
rAn Bam Hi I
7
8
9 kbp
FIG. 2. All plasmids used in this study are composed of an E. coli rRNA operon inserted into the tetracycline resistance gene of pBR322 from which the 0.9-kb NaeI-BamHI fragment has been removed. All plasmids confer pBR322-encoded ampicillin resistance. (a) Plasmid pKK35 (Table 1) contains wild-type rrnB on a 7.5-kb BamHI fragment; pKK2032A, pKK2032C, and pKK2032T have the same structure with the 23S RNA point mutations indicated. Plasmid pKKA12 has a 12-bp deletion overlapping the SphI site (see text). Plasmid constructs with the 2058G mutation are truncated at the BclI site. In plasmid pLK35 and the derivatives thereof listed in Table 2, the rrnB promoters have been replaced by the lambdapL promoter (12, 19). Plasmid pSK41 contains the 4.3-kb XbaI-BamHI 3' portion of rrnB (13). (b) Structure of the r-nB-r-nH fusions on a 8.9-kb BamHI fragment. The 5' portion up to SalI is derived from rmnB, and the 3' portion is from rnH constructs containing the 2057A and 2058T mutations. The KspI site at position 2048 in the 23S RNA gene was used to combine the 2032 and 2057 (or 2058) mutations.
rates of wild-type, 2032C, or 2032U mutants at permissive
levels of clindamycin (10 mg/liter). E. coli is naturally resistant to lincomycin; therefore, the growth of the mutants on this drug was not assayed. Cells formed distinct colonies after 22 h on erythromycin at a concentration of 30 mg/liter, whereas there was no growth under these conditions when the G at position 2032 was changed to A, C, or U (Table 1). After prolonged incubation for 48 h, the 2032 mutants formed visible colonies, showing that this drug concentration is nonlethal. The 2032A mutation enabled E. coli to grow on chloramphenicol at 4 mg/liter, while cells with the 2032G or 2032C mutation were inhibited by 2 mg/liter. The 2032U mutation supported slow growth on 2 mg of chloramphenicol per liter. In the absence of antibiotics that bind to
the 50S ribosomal subunit, the 2032 mutations had no effect cell growth rate (Tables 1 and 2). Other 23S RNA mutations conferring resistance to chloramphenicol, lincosamides, and macrolides. An equivalent degree of chloramphenicol resistance is conferred by 2057A and 2032A, both mutations supporting growth on chloramphenicol at 4 mg/liter. The 2058G and 2058U mutations conferred a phenotype similar to that of the 2032U mutation, each providing a distinct growth advantage on chloramphenicol at 2 mg/liter, but could not support growth at 4 mg/liter. On clindamycin, the 2058G and 2058U mutations both confer greater resistance than the 2032A mutation (Table 1). The 2058G and 2058U mutations also confer higher levels of erythromycin resistance than the 2057A mutation, which in on
TABLE 1. Drug resistances conferred by plasmids on E. coli DH1 cells Growth on antibiotic
Plasmid
Ampicillin'
Chloramphenicolb
Clindamycin'
pKK35 (wild type) pKK2032A pKK2032T pKK2032C
+++ +++ +++ +++
++ +
pKK2057A pKK2058G pKK2058T pKKA12 pKK2032A + 2057A pKK2032A + 2058G(T) pKK2032A + A12
+++ +++ +++
++ + +
+ +++
+++
+
++
ND
++
+++ + ++
ND
Erythromycind
ND
++
+++ + ++ +
Cells formed clearly visible colonies after 14 (+ + +), 16 (+±), or 22 (+) h. + + +, growth at up to 6 mg/liter; + +, growth at up to 4 mg/liter; +, slow growth at 2 mg/liter;-, no growth at 2 mg/liter. ND, not determined. At 25 mg/liter, distinct colonies were observed after 20 (+++), 22 (++), or 24 (+) h; -, no growth at this drug concentration. ND, not determined. d Cells with the wild-type pKK35 plasmid formed distinct colonies after 22 h at 30 mg/liter (±) but did not grow at 50 mg/liter; -, no growth or severely retarded growth at 30 mg/liter. The A12, 2057A, and 2058 mutations support growth at 80 mg/liter, forming distinct colonies of comparable sizes after 40), 24, and 16 h (+? + +, and + + +, respectively). a
b
C
1336
J. BACTERIOL.
DOUTHWAITE
TABLE 2. Growth of cells transformed with plasmids in which rRNA genes are under control of the lambda PL promoter Growth on antibiotic Plasmid
Ampicillin'
pLK35 (wild type) pLK2032A pLK2058G pLKA12 pLK2032A + 2058G pLK2032A + A12 pLK2032A + 2058G and A12
Erythromycin"
+++ + ++ +++ +++ +++ +++ ++ +++ ++ +++
All cells grew at the same rate, forming distinct colonies in 13 h. b -, no growth at 30 mg/liter; +, growth at 30 mg/liter but no growth at 50 mg/liter; +, growth at 50 mg/liter after 36 h but no growth at 80 mg/liter; + +, growth at 50 mg/liter after 22 h and growth at 80 mg/liter after 36 h; +++ growth at 50 mg/liter after 20 h and growth at 80 mg/liter after 30 h.
turn confers better resistance than the A12 mutation (Table 1). However, when 23S rRNA was expressed from the lambdapL promoter, the A12 mutation enabled cells to grow at higher erythromycin levels than the 2058G (Table 2) and 2058U mutations. Combinations of drug resistance mutations in 23S RNA. The combination of mutations 2032A and 2057A in the same 23S RNA produced a phenotype that retains the advantageous growth characteristics of each individual mutation (Table 1). Cells grew better on chloramphenicol with both mutations than with either mutation alone; the 2032A-encoded clindamycin resistance is retained, and the 2057A-encoded erythromycin-resistant phenotype is dominant over the 2032A-induced hypersensitivity. When mutations 2032A and 2058G (or 2058U) were combined, however, cells grew slowly in the absence of 50S subunit-specific drugs and they were incapable of growing on erythromycin even after extended incubations; cells also failed to grow on chloramphenicol and clindamycin, despite the growth advantage on these drugs conferred by each of the two mutations on its own. The rate of cell growth on ampicillin was normal when 23S RNA containing the 2032A and 2058G mutations was expressed from the lambdapL promoter, but cells remained hypersensitive to erythromycin (Table 2). Introduction of the A12 mutation into plasmids either with the 2032A mutation alone or with a combination of the A2032 and G2058 mutations relieved the hypersensitivity effect and conferred an erythromycin-resistant phenotype.
DISCUSSION
The sequence of the 23S rRNA peptidyltransferase loop region depicted as single stranded in Fig. 1 is highly conserved (20). Many of these bases have been shown to interact both with the aminoacyl end of tRNA (1, 26) and with antibiotics that inhibit peptide bond formation (6, 25) (Fig. 1). The importance for rRNA function of the neighboring stem-loop (formed by nucleotides 2023 to 2040) is indicated by its presence in all 23S-like rRNAs, including the truncated versions in mitochondria (20). Nucleotide 2032 in the hairpin loop is conserved as a guanosine in all eubacteria and chloroplasts, and the first indication that this nucleotide is functionally connected with peptidyltransferase was revealed when Cseplo et al. identified an A transition mutation at this position in a lincomycin-resistant tobacco chloroplast (7). The clindamycin resistance conferred by the 2032A muta-
tion shows that the effect observed in chloroplast 23S rRNA can be extrapolated to E. coli (and most likely to all eubacteria and chloroplasts). The 2032A mutation also confers chloramphenicol resistance, emphasizing the involvement of this position in peptidyltransferase. It cannot be concluded, however, that these drugs directly interact with this position in wild-type rRNA, as the 2032C and 2032U mutations confer no resistance. The 2057G--A change confers chloramphenicol resistance (16), as do other mutations in the peptidyltransferase loop region of mitochondrial rRNA (Fig. 1). However, the latter mutations that have been reproduced at the corresponding sites in E. coli result in an inviable or slow-growth phenotype (37). The highest levels of clindamycin and erythromycin resistance observed in this study are conferred by the 2058 mutations, which is consistent with the idea that this base and its immediate vicinity form the target site recognized by the drugs used in this study (8, 25). The hypersensitivity to erythromycin in 2032 mutants was unexpected. Hypersensitivity to this drug has previously been encountered after genetic elimination of r protein Lii (reference 32 and references therein), which binds to 23S rRNA between bases 1055 and 1105 (15), and also after introduction of a mutation in 23S rRNA domain II adjacent to the site where deletions confer erythromycin resistance (9). The mutations could be affecting drug binding, or they could be enhancing the effectiveness of erythromycin in physically blocking peptide chain elongation (35) or in aggravating premature dissociation of peptidyl-tRNA from the ribosome (24). Nucleotide 2032 has been cross-linked by nitrogen mustard to 2055 (10). The proximity of these two positions suggests that the 2032A mutation could bring about its effect by altering the tertiary alignment of 2058A and/or the neighboring bases. Mutations at positions 2057 and 2058 probably also perturb the local structure. The drug resistances conferred (Table 1) show that any of these alterations on its own can be incorporated in a functionally active rRNA molecule. The combination of the 2032A mutation and the 2057A mutation is permissible and indeed advantageous in, for example, the case of the additive chloramphenicol resistance effects. However, the combination of the 2032A mutation and the 2058G or U mutation is deleterious, indicating that the bases are a part of a finely tuned mechanism which cannot accommodate these combined structural perturbations. UV and nitrogen mustard cross-links have been created between nucleotides 571 and 2031, 746 and 2613, and 979 to 984 and 2029 (10) (Fig. 1), orienting positions 2032 and 2058 in the midst of domain II. A functional connection between domains II and V based on the A12 mutation that confers erythromycin resistance has previously been reported (13). Expression of A12 23S RNA from the rnB promoters slows cell growth in the absence of erythromycin (Table 1), probably because the deletion has a side effect on 50S subunit assembly (12). However, the growth rate is normal when A12 23S RNA is expressed from the weaker lambdapL promoter (12), and this construct confers better erythromycin resistance (Table 2). The A12 mutation is dominant over the 2032A mutation, relieving the erythromycin-hypersensitive effect and conferring a drug-resistant phenotype essentially identical to that conferred by A12 alone (Tables 1 and 2). Expressed from thepL promoter, 23S rRNA with the 2032A and 2058G mutations does not cause appreciably slower growth in the absence of drugs, but erythromycin hypersensitivity is still observed (Table 2). Introduction of A12 into
23S rRNA INTERACTIONS INVOLVING PEPTIDYLTRANSFERASE
VOL. 174, 1992
this construct relieves the effect of the otherwise deleterious juxtaposition of the 2032A and 2058G mutations and produces a functionally active 23S rRNA that confers erythromycin resistance. It is evident, therefore, that a functional connection between domains II and V exists. ACKNOWLEDGMENTS Anne-Mette Wennermark is thanked for technical assistance, as are Claus Aagaard and Edel Rasmussen for art work. Ed Morgan
kindly provided rnH clones. Eric Cundliffe is thanked for advice on antibiotics, and Claus Aagaard, Mette Dam, Roger Garrett, and Gunnar Rosendahl are thanked for their comments on the manu-
script. The research was supported by The Carlsberg Foundation. REFERENCES 1. Barta, A., E. Kuechier, and G. Steiner. 1990. Photoaffinity labeling of the peptidyltransferase region, p. 358-365. In W. E. Hill, A. Dahlberg, R. A. Garrett, P. B. Moore, D. Schlessinger, and J. R. Warner (ed.), The ribosome: structure, function, and evolution. American Society for Microbiology, Washington, D.C. 2. Blanc, H., C. W. Adams, and D. C. Wallace. 1981. Different nucleotide changes in the large rRNA gene of the mitochondrial DNA confer chloramphenicol resistance on two human cell lines. Nucleic Acids Res. 9:5785-5795. 3. Blanc, H., C. T. Wright, M. J. Bibb, D. C. Wallace, and D. A. Clayton. 1981. Mitochondrial DNA of chloramphenicol-resistant mouse cells contains a single nucleotide change in the region encoding the 3' end of the large ribosomal RNA. Proc. Natl. Acad. Sci. USA 78:3789-3793. 4. Brosius, J., A. Ullrich, M. A. Raker, A. Gray, T. J. Dull, R. R. Gutell, and H. F. Noller. 1981. Construction and fine mapping of recombinant plasmids containing the rrnB ribosomal RNA operon of E. coli. Plasmid 6:112-118. 5. Chen, E. Y., and P. H. Seeburg. 1985. Supercoil sequencing: a fast and simple method for sequencing plasmid DNA. DNA 4:165-170. 6. Cooperman, B. S., C. J. Weitzmann, and C. L. Fernindez. 1990. Antibiotic probes of Escherichia coli ribosomal peptidyltransferase, p. 491-501. In W. E. Hill, A. Dahlberg, R. A. Garrett, P. B. Moore, D. Schlessinger, and J. R. Warner (ed.), The ribosome: structure, function, and evolution. American Society for Microbiology, Washington, D.C. 7. Cseplo, A., T. Etzold, J. Schell, and P. H. Schreier. 1988. Point mutation in the 23S rRNA genes of four lincomycin resistant Nicotiana plumbaginifolia mutants could provide new selectable markers for chloroplast transformation. Mol. Gen. Genet. 214:295-299. 8. Cundliffe, E. 1990. Recognition sites for antibiotics within rRNA, p. 479-490. In W. E. Hill, A. Dahlberg, R. A. Garrett, P. B. Moore, D. Schlessinger, and J. R. Warner (ed.), The ribosome: structure, function, and evolution. American Society for Microbiology, Washington, D.C. 9. Dam, M., and S. Douthwaite. Unpublished data. 10. Doring, T., B. Greuer, and R. Brimacombe. 1991. The three dimensional folding of ribosomal RNA; localization of a series of intra-RNA cross-links in 23S RNA induced by treatment of Escherichia coli 50S ribosomal subunits with bis-(2-chloroethyl)-methylamine. Nucleic Acids Res. 19:3517-3524. 11. Douthwaite, S. Unpublished data. 12. Douthwaite, S., T. Powers, J. Y. Lee, and H. F. Noller. 1989. Defining the structural requirements for a helix in 23S ribosomal RNA that confers erythromycin resistance. J. Mol. Biol. 209: 655-665. 13. Douthwaite, S., J. B. Prince, and H. F. Noller. 1985. Evidence for functional interaction between domains II and V of 23S ribosomal RNA from an erythromycin resistant mutant. Proc. Natl. Acad. Sci. USA 82:8330-8334.
1337
14. Dujon, B. 1980. Sequence of the intron and flanking exons of the mitochondrial 21S rRNA gene of yeast strains having different alleles at the omega and rib-i loci. Cell 20:185-197. 15. Egebjerg, J., S. R. Douthwaite, A. Liljas, and R. A. Garrett. 1990. Characterization of the binding sites of protein Lii and the L10(L12)4 pentameric complex in the GTPase domain of 23S ribosomal RNA from Escherichia coli. J. Mol. Biol. 213:275288. 16. Ettayebi, M., S. M. Prasad, and E. A. Morgan. 1985. Chloramphenicol-erythromycin resistance mutations in a 23S rRNA gene of Escherichia coli. J. Bacteriol. 162:551-557. 17. Gale, E. F., E. Cundliffe, P. E. Reynolds, M. H. Richmond, and M. J. Waring. 1981. The molecular basis of antibiotic action. John Wiley and Sons, London. 18. Gauthier, A., M. Turmel, and C. Lemieux. 1988. Mapping of chloroplast mutations conferring resistance to antibiotics in Chlamydomonas: evidence for a novel site of streptomycin resistance in the small subunit RNA. Mol. Gen. Genet. 214:192197. 19. Gourse, R. L., Y. Takebe, R. A. Sharrock, and M. Nomura. 1985. Feedback regulation of rRNA and tRNA synthesis and accumulation of free ribosomes after conditional expression of rRNA genes. Proc. Natl. Acad. Sci. USA 82:1069-1073. 20. Gutell, R. R., and G. E. Fox. 1988. A compilation of large subunit RNA sequences presented in structural format. Nucleic Acids Res. 16:r175-r269. 21. Hummel, H., and A. Bock. 1987. 23S ribosomal RNA mutations in halobacteria conferring resistance to the anti-80S ribosome targeted antibiotic anisomycin. Nucleic Acids Res. 15:24312443. 22. Kearsey, S. E., and I. W. Craig. 1981. Altered ribosomal RNA genes in mitochondria from mammalian cells with chloramphenicol resistance. Nature (London) 290:607-608. 23. Mankin, A. S., and R. A. Garrett. 1991. Chloramphenicol resistance mutations in the single 23S rRNA gene of the archaeon Halobacterium halobium. J. Bacteriol. 173:35593563. 24. Menninger, J. R., and D. P. Otto. 1982. Erythromycin, carbomycin, and spiramycin inhibit protein synthesis by stimulating the dissociation of peptidyl-tRNA from ribosomes. Antimicrob. Agents Chemother. 21:811-818. 25. Moazed, D., and H. F. Noller. 1987. Chloramphenicol, erythromycin, carbomycin and vernamycin B protect overlapping sites in the peptidyl transferase region of 23S ribosomal RNA. Biochimie 69:879-884. 26. Moazed, D., and H. F. Noller. 1991. Sites of interaction of the CCA end of peptidyl-tRNA with 23S rRNA. Proc. Natl. Acad. Sci. USA 88:3725-3728. 27. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 28. Sigmund, C. D., M. Ettatebi, A. Borden, and E. A. Morgan. 1988. Antibiotic resistant mutations in ribosomal RNA genes of Escherichia coli. Methods Enzymol. 164:673-690. 29. Slott, E. F., Jr., R. 0. Shade, and R. A. Lansman. 1983. Sequence analysis of mitochondrial DNA in a mouse cell line resistant to chloramphenicol and oligomycin. Mol. Cell. Biol. 3:1694-1702. 30. Sor, F., and H. Fukuhara. 1982. Identification of two erythromycin resistant mutations in the mitochondrial gene coding for the large ribosomal RNA in yeast. Nucleic Acids Res. 10:65716577. 31. Sor, F., and H. Fukuhara. 1984. Erythromycin and spiramycin resistance mutations of yeast mitochondria: nature of the rib2 locus of the large ribosomal RNA gene. Nucleic Acids Res.
12:8313-8318. 32. Stomfler, G., E. Cundliffe, M. Stoffler-Meilicke, and E. R. Dabbs. 1980. Mutants of Escherichia coli lacking ribosomal protein Lii. J. Biol. Chem. 255:10517-10522. 33. Sweeney, R., C.-H. Yao, and M.-C. Yao. 1991. A mutation in the large subunit ribosomal RNA gene of Tetrahymena confers anisomycin resistance and cold sensitivity. Genetics 127:327334.
1338
DOUTHWAITE
34. Taylor, J. W., J. Ott, and F. Eckstein. 1985. The rapid generation of oligonucleotide-directed mutations at high frequency using phosphorothioate-modified DNA. Nucleic Acids Res. 13:8764-8785. 35. V6zquez, D. 1979. Inhibitors of protein synthesis, p. 169-175. Springer-Verlag, Berlin. 36. Vester, B., and R. A. Garrett. 1987. A plasmid-coded and
J. BACTERIOL.
site-directed mutation in Eschenchia coli 23S RNA that confers resistance to erythromycin: implications for the mechanism of action of erythromycin. Biochimie 69:891-900. 37. Vester, B., and R. A. Garrett. 1988. The importance of highly conserved nucleotides in the binding region of chloramphenicol in the peptidyl transferase centre of Escherichia coli 23S ribosomal RNA. EMBO J. 7:3577-3587.
