Previous Article | Next Article ![]()
Journal of Bacteriology, June 2005, p. 3708-3712, Vol. 187, No. 11
0021-9193/05/$08.00+0 doi:10.1128/JB.187.11.3708-3712.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Sean R. Connell,3 and
Diane E. Taylor1,2*
Department of Medical Microbiology and Immunology,1 Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada,2 Institut für Medizinische Physik und Biophysik, Universitätsklinikum Charite, Berlin, Germany3
Received 5 November 2004/ Accepted 25 February 2005
|
|
|---|
|
|
|---|
Tetracycline has been intensively used since the 1950s, and many bacterial pathogens have acquired resistance to tetracycline (8). In most species, tetracycline resistance (Tcr) is conferred by resistance genes with two main modes of action. The first group encodes efflux systems that transport the drug from inside to outside the bacterial cell; in the second group a ribosomal protection protein removes tetracycline from the ribosome (8, 9). These Tcr determinants are often associated with transmissible genetic elements such as plasmids, transposons and integrons (8). In the case of H. pylori, Tcr is not observed as frequently as in other species (17, 19, 23). However, tetracycline-resistant H. pylori isolates have started to emerge, and their prevalence is increasing, especially in Asian countries such as China, Korea, and Japan (15, 16, 26).
Tcr in H. pylori has not been associated with efflux or ribosomal protection proteins but rather with mutations in the 16S rRNA gene (10, 13, 21, 25). The 16S rRNA is a component of the 30S ribosomal subunit, which is the target of tetracycline (8, 9). Binding of tetracycline to the 30S ribosomal subunit inhibits protein synthesis (8, 9) by interfering with the binding of aminoacyl-tRNA to the A site of the ribosome (8, 9). Recently, two groups have reported that X-ray diffraction analysis of the tetracycline-bound 30S ribosomal subunit showed the presence of one primary tetracycline-binding site and up to five secondary binding sites (6, 20)
Tcr mediated by mutations in the 16S rRNA was first found in Propionibacterium acnes and it was reported that several isolates of P. acnes contained a G-to-C mutation at position 1058 (Escherichia coli numbering) in their 16S rRNA genes (22). Recently, four groups have reported that a triplet mutation in the same 16S rRNA domain (965 to 967; E. coli numbering) mediated Tcr in H. pylori (10, 13, 21, 25). This mutation is located in loop of helix 31, which is a component of the primary tetracycline binding site observed by X-ray crystallography (6, 20). Three reports identified the triple mutation AGA965 to 967TTC to be responsible for high-level Tcr in H. pylori (13, 21, 25), while the fourth reported low-level Tcr conferred by single and double mutations in the same region (10). These results suggest that the triple mutation at 965 to 967 confers Tcr by decreasing the affinity of the ribosome for tetracycline (10, 13, 21, 25). Nevertheless, there was no direct evidence of a change in the affinity of resistant ribosomes towards tetracycline.
In this study, the involvement of positions 965 to 967 of 16S rRNA in tetracycline binding to the ribosome and Tcr were investigated in an E. coli-based system. E. coli was used owing to the technical limitation of growing sufficient quantities of H. pylori to produce ribosomes for in vitro studies, as well as to facilitate genetic manipulation of the rRNA genes. Importantly an E. coli strain which harbors only a single plasmid-borne copy of the rRNA genes (3) was used to provide a homogeneous population containing only the tetracycline-resistant ribosomes.
|
|
|---|
Site-directed mutagenesis. All mutations were introduced by site-directed mutagenesis at positions 965 to 967 and G1058 (E. coli numbering) on plasmid pKK3535 (11,864 bp), which encodes a single rRNA operon (rrnB) (7). Site-directed mutagenesis was carried out with the primers listed in Table 1 using the QuickChange XL-site-directed mutagenesis kit (Stratagene, La Jolla, Calif.). The manufacturer's recommended conditions for Pfx turbo polymerase (Stratagene, La Jolla, Calif.) were followed for 95°C, 1 min, and 18 cycles for 60°C, 50 s; 68°C, 24 min; and 68°C for 7 min. Products were treated with DpnI (Invitrogen, Carlsbad, Calif.) for 1 h at 37°C to digest the parental DNA used as the template, and the resulting plasmids were transformed into E. coli JM109.
|
View this table: [in a new window] |
TABLE 1. Primers used in this study
|
Plasmid replacement. Each plasmid containing different mutations was transformed into E. coli TA527. To replace the endogenous 16S rRNA-harboring plasmid (pHK-rrnC+), the transformed E. coli TA527 was cultured overnight in the presence of ampicillin and spectinomycin selection. This step was repeated after 100,000 dilution in phosphate-buffered saline to select for transformants harboring pKK3535. Individual colonies were restreaked on both LB plates in the presence and absence of kanamycin to confirm the absence of pHK-rrnC. The success of the plasmid replacement was confirmed by the extraction of the plasmids from E. coli TA527 containing different mutations and the size of the plasmids was confirmed, using 0.8% agarose gel electrophoresis in 0.04 Tris-acetate-0.001 M EDTA, to be approximately 6.8 kb and 12 kb, corresponding to pTRNA66 and mutants of pKK3535, respectively.
Tetracycline susceptibility. The MICs of tetracycline were determined using agar plates containing twofold serial dilutions of tetracycline from 256 to 0.125 µg/ml. E-test (AB Biodisk, Solna, Sweden) was also used to detect small differences between mutants. The MICs determined by the E-test were lower compared to those detected by the agar dilution method. To confirm the MICs, the agar dilution method was performed three times. Each strain was incubated overnight and the cell concentration was determined by optical density (optical density at 600 nm) and adjusted to 0.8 to 1.0 in LB broth, and 10 µl of each sample was diluted with 5 ml of phosphate-buffered saline. A 10 µl sample of this dilution was spotted onto agar plates and incubated for 3 days at 37°C.
Extraction of ribosomes. To extract the 70S ribosomes, a 500 ml culture of E. coli mutants was grown for four h until the cell density reached an optical density at 600 nm of 0.4. The cells were harvested and then washed in 50 ml of buffer A (20 mM HEPES-KOH, 10 mM Mg acetate, 100 mM NH4Cl, 4 mM 2-mercaptoethanol), and ribosomes were extracted and purified as described by Blaha et al. (5).
Tetracycline binding. tetracycline binding was measured using [3H]tetracycline in a nitrocellulose-binding assay (24). Each assay was performed in triplicate. Ribosomes (24 pmol) were preincubated in binding buffer (20 mM HEPES-KOH, pH 7.6, 6 mM Mg acetate, 150 mM NH4Cl, 4 mM 2-mercaptoethanol, 0.05 mM spermine, 2 mM spermidine) at 37°C for 10 min. [3H]tetracycline was added at various concentrations and the reaction was incubated for 15 min before dilution with 2 ml of wash buffer (20 mM HEPES-KOH, pH 7.6, 6 mM Mg acetate, 150 mM KCl, 4 mM 2-mercaptoethanol, 0.05 mM spermine, 2 mM spermidine) and immediately filtered under vacuum through 0.45-µm nitrocellulose filters. Five ml of scintillation fluid was added to each filter, followed by overnight shaking. The amount of [3H]tetracycline bound was determined using a Wallac 1414 Winspectral liquid scintillation counter (Wallac, Helsinki, Finland).
|
|
|---|
|
View this table: [in a new window] |
TABLE 2. Mutants involved in this study and their tetracycline MICs determined by the serial dilution method and E-test
|
Tetracycline susceptibility of 16S rRNA mutants. The role of helix 31 alterations in conferring Tcr was investigated by assaying the susceptibility of the mutants to tetracycline. The MICs of tetracycline for the mutants ranged from 0.25 to 8 µg/ml by the serial agar dilution method and from 0.125 to 6 µg/ml using E-test strips (Table 2). The MIC of the strain exhibiting the wild-type sequence of E. coli (965TGC967 substitution) was 2 µg/ml using the agar dilution method. The MICs of the strains with 965AGA967 (the wild-type sequence of H. pylori), 965AAA967, 965AAG967, 965ACA967, and 975AGG967 substitutions were determined to be 0.25 µg/ml by the agar dilution method, which was the lowest MIC observed (Table 2). In comparison, the strains with helix 31 mutations previously found in highly resistant H. pylori isolates (10, 13, 21, 25) all demonstrated elevated levels of Tcr ranging from 1 to 8 µg/ml by the agar dilution method. The mutant containing the G1058C mutation, found in the 16S rRNA of the tetracycline-resistant P. acnes, displayed increased Tcr (16 µg/ml by the agar dilution method).
The similar effect of mutations in E. coli and H. pylori demonstrates a conserved role of helix 31 in conferring Tcr. In contrast, the variation in the extent of resistance in the two hosts suggests that the role of specific bases differs, perhaps due to differences imposed by the surrounding rRNA. Interestingly, the wild-type E. coli strain showed an intermediate MIC of 2 µg/ml, whereas five of the six most resistant mutants (8 µg/ml) were represented by sequences not yet observed in tetracycline-resistant H. pylori clinical isolates (Table 2). The six mutants having an MIC of 8 µg/ml all contained three substitutions relative to the highly susceptible wild-type H. pylori strain, whereas the most susceptible mutants displayed only single substitutions (Table 2).
Although no absolute rule describing the effect of the substitutions in helix 31 on Tcr can be made, it was generally seen that a shift to pyrimidine-rich sequences led to an increase in the resistance level (Table 2). To eliminate the possibility that secondary mutations (spontaneous or generated during PCR steps) contributed to Tcr, the 1,639-bp DNA fragment harboring the 965 to 967 mutations from eight representative mutants was excised, religated into wild-type pKK3535, and sequenced. The nature of the 16S rRNA sequences was confirmed, and the MICs were identical to those determined previously.
Effect of 16S rRNA mutations on tetracycline binding. To examine the ability of mutant ribosomes to bind tetracycline, ribosomes were extracted from the wild-type strains of E. coli and H. pylori as well as four other helix 31 mutants expressed in E. coli TA527. In these assays, the concentration of tetracycline was kept below 5 µM to maximize specific binding to the high-affinity tetracycline-binding site, which has a Kd of 1 to 20 µM (14). At these tetracycline concentrations, the ribosomes harboring the sequence corresponding to tetracycline-susceptible H. pylori (965AGA967 mutant) demonstrated the highest tetracycline binding (Fig. 1A). In contrast, the ribosomes with the 965TTC967 triple mutation, which is equivalent to that found in the highly resistant TcR H. pylori strains, bound approximately 2.5-fold less tetracycline than the 965AGA967 ribosomes (Fig. 1A). The mutant with the 965TCC967 triple substitutions displayed the weakest binding to tetracycline. The ribosomes from the mutants with intermediate MIC levels (965GGA967 and 965AGC967) showed moderate ability to bind tetracycline (Fig. 1A).
![]() View larger version (22K): [in a new window] |
FIG. 1. Binding of [3H]tetracycline to ribosomes extracted from mutants with different nucleotide substitutions at positions 965 to 967 (A) and 1058 (B) of the 16S rRNA gene. Each symbol indicates the E. coli strains containing the following sequences in the 16S rRNA gene: 965AGA967 ( ), 965TTC967 ( ), 965AGC967 ( ), 965GGA967( ), 965TCC967 ( ), 965TGC967 (A) or G1058 (B) wild-type E. coli ( ), and G1058C (x).
|
|
|
|---|
Using ribosomes isolated from several of these E. coli mutants, we showed that helix 31 mutations conferring Tcr in H. pylori decreased the ability of the drug to bind the ribosome. Dailidiene et al. (10) isolated several H. pylori strains which contained one or two nucleotide substitutions at 965 to 967. In these isolates, single nucleotide substitutions conferred only weak resistance, whereas double substitutions conferred a somewhat higher-level resistance and triple substitutions conferred the highest level of resistance. This suggested that the strength of resistance tended to be proportionate to the severity of change at positions 965 to 967 (10). This appears to be true according to our data (Table 2), although some notable exceptions were observed. For example, the triple mutation 965CAG967 (relative to the wild-type H. pylori sequence) conferred only low-level resistance, whereas the double mutant 965AAC967 was one of the most resistant isolates. Our data showed another trend: most tetracycline-susceptible mutants belonging to the lowest MIC group had purine-rich sequences in the loop of helix 31, while the highly resistant strains had pyrimidine-rich loop.
Structures derived from X-ray diffraction studies of the 30S ribosomal subunit bound by tetracycline showed that helix 31 in combination with helix 34 forms the tetracycline binding pocket (6, 20). In particular, there is a hydrogen bond between the 2'-OH of A965 and tetracycline in addition to a hydrogen bond between the phosphate oxygen of G966 and tetracycline (6). Both of these groups are part of the backbone structure and are invariant regardless of the nucleotide base. The nucleotide substitutions conferring resistance would not directly prohibit formation of these interactions but they may alter the conformation or flexibility of the backbone at positions 965 to 967 to weaken tetracycline binding. Rearrangements in the architecture of helix 31 during tetracycline binding were suggested by the fact that UV induced cross-linking between C1400 and C967 is inhibited completely by tetracycline (18). In terms of the resistance mutations, it is possible that pyrimidine-rich sequences in helix 31 loop are not compatible with the tetracycline-induced conformation, leading to weakened tetracycline binding.
The mutation G1058C found in tetracycline-resistant P. acnes (22) mediated eightfold increase in Tcr level (MIC was 16 µg/ml by the agar dilution method) relative to that of the strain with the wild-type sequence of E. coli. This is in exact agreement with a previous report (4). Additionally, the substitution G1058C significantly decreased the binding of tetracycline to the ribosome to less than half of that seen in the control (Fig. 1). This mutation does not directly affect the tetracycline-binding site, but it had been suggested that disruption of the G1058-U1199 base pair may lead to a long-range conformational change at the tetracycline-binding pocket (20).
This study confirms that the 16S rRNA mutations found in tetracycline-resistant H. pylori confer high-level Tcr in E. coli and goes on to identify novel mutations that could appear in emerging tetracycline-resistant strains. Furthermore, we demonstrated that Tcr mutations in the 16S rRNA lead to a decrease in the binding of tetracycline to the ribosome in vitro. Our results confirm that the nucleotide patterns at positions 965 to 967, which form the primary tetracycline-binding site, influence the binding ability of tetracycline and consequently change the level of bacterial resistance to tetracycline.
This work was supported by a grant from the National Sciences and Engineering Research Council of Canada. L.N received a Fellowship from the Japanese Society for the Promotion of Science and was partly supported by grants-in-aid from the 21st Century COE program. S.R.C. received a studentship from the Alberta Heritage Foundation for Medical Research (AHFMR). D.E.T. is an AHFMR Scientist.
Present address: Center for Marine Environmental Studies, Ehime University, 3 Bunkyo-cho, Matsuyama 790-8577, Japan. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»