ABSTRACT
Arm/Rmt methyltransferases have emerged recently in pathogenic bacteria as enzymes that confer high-level resistance to 4,6-disubstituted aminoglycosides through methylation of the G1405 residue in the 16S rRNA (like ArmA and RmtA to -E). In prokaryotes, nucleotide methylations are the most common type of rRNA modification, and they are introduced posttranscriptionally by a variety of site-specific housekeeping enzymes to optimize ribosomal function. Here we show that while the aminoglycoside resistance methyltransferase RmtC methylates G1405, it impedes methylation of the housekeeping methyltransferase RsmF at position C1407, a nucleotide that, like G1405, forms part of the aminoglycoside binding pocket of the 16S rRNA. To understand the origin and consequences of this phenomenon, we constructed a series of in-frame knockout and knock-in mutants of Escherichia coli, corresponding to the genotypes rsmF+, ΔrsmF, rsmF+ rmtC+, and ΔrsmF rmtC+. When analyzed for the antimicrobial resistance pattern, the ΔrsmF bacteria had a decreased susceptibility to aminoglycosides, including 4,6- and 4,5-deoxystreptamine aminoglycosides, showing that the housekeeping methylation at C1407 is involved in intrinsic aminoglycoside susceptibility in E. coli. Competition experiments between the isogenic E. coli strains showed that, contrary to expectation, acquisition of rmtC does not entail a fitness cost for the bacterium. Finally, matrix-assisted laser desorption ionization (MALDI) mass spectrometry allowed us to determine that RmtC methylates the G1405 residue not only in presence but also in the absence of aminoglycoside antibiotics. Thus, the coupling between housekeeping and acquired methyltransferases subverts the methylation architecture of the 16S rRNA but elicits Arm/Rmt methyltransferases to be selected and retained, posing an important threat to the usefulness of aminoglycosides worldwide.
INTRODUCTION
Aminoglycosides are used for the treatment of a wide range of infections due to both Gram-negative and Gram-positive bacteria and have been classified by the World Health Organization as critically important antibiotics in human medicine (34). They bind to the A site of the 16S rRNA and force A1492 and A1493 to “flip out” of helix 44 (29, 32). This molecular switch is similar to the one that these two universally conserved nucleotides adopt when there is a cognate fit between the mRNA codon and the tRNA anticodon, which is essential for the process of monitoring of the protein synthesis. Therefore, the conformation adopted when the aminoglycosides bind to their target causes a loss of translational fidelity, leading to accumulation of erroneous proteins and to bacterial death (26). Resistance to these antimicrobials is usually due to production of aminoglycoside-modifying enzymes (such as acetyltransferases, phosphorylases, and adenyltransferases) (30), reduced intracellular antibiotic accumulation, or mutation of ribosomal proteins or rRNA. An additional mechanism, methylation of the aminoacyl site of 16S rRNA, confers high-level resistance to clinically important aminoglycosides such as amikacin, tobramycin, and gentamicin. This phenomenon is mediated by a group of 16S rRNA methyltransferases that share similarity with those produced by actinomycetes, such as Streptomyces spp. and Micromonospora spp. In these species, 16S rRNA methylation protects their ribosomes from the toxic effects of their own aminoglycoside production (8). Six acquired 16S rRNA methyltransferase genes conferring resistance to these antimicrobials, i.e., armA in human, animal, and food isolates (15, 16, 17), rmtA (35), rmtB (11), rmtC (33), rmtD (12), and rmtE (10), have been identified in clinical bacterial strains and are now distributed worldwide. The target of these methyltransferases is the N7 position of nucleotide G1405 in 16S rRNA (Escherichia coli numbering), and it confers high-level resistance to the 4,6-disubstituted aminoglycosides such as gentamicin and tobramycin (24). The methylation takes place after the assembly of the 30S subunit, which increases the opportunity to methylate the target. These methyltransferases have the sole function of providing protection against aminoglycosides. In this aspect, they contrast with the numerous endogenous housekeeping methyltransferases that are important for the structure and function of mature rRNA (25, 1). However, the resistance and the housekeeping methylations are similar in that they cluster in phylogenetically conserved and functionally essential regions of the rRNA. At the ribosomal A site of E. coli 16S rRNA, where the aminoglycoside binding pocket is located, several nucleotides, including C1402 (21), C1407 (1), U1498 (4), and A1518 and A1519 (19), are posttranscriptionally methylated by housekeeping methyltransferases. It seems that there is an evolved order of methylation steps within endogenous housekeeping methyltransferases to avoid crowding and possible stereo-electronic effects on downstream methylations introduced by these modifications. Such constraints would require synchronization of resistance-mediating RNA methyltransferases with the other, housekeeping methyltransferases. It has recently been observed that the expression in E. coli of Sgm, a methyltransferase that confers aminoglycoside resistance by methylation of m7G1405 from an antibiotic-producing strain, decreases the methylation function of endogenous E. coli RsmF, an m5C1407-specific housekeeping methyltransferase (7). However, it is not known if similar methylation displacements occur with the emerging Arm/Rmt methyltransferases. Finally, the role of the presence or absence of aminoglycosides in the methylation of the aminoglycoside binding pocket of the ribosome, as well as the repercussion of these methylations in ribosome function and bacterial fitness, remains to be elucidated.
In this study, we have observed an interaction between the methylation at C1407 introduced by RsmF and the methylation at G1405 produced by acquired ArmA/Rmt methyltransferases. We have analyzed the consequences of this interaction for bacterial fitness in the presence and absence of antibiotics. Furthermore, we have identified that the housekeeping methyltransferase RsmF plays a role in intrinsic susceptibility to aminoglycosides in the presence and in the absence of the acquired methyltransferase RmtC. Finally, we have dissected the biological cost entailed by acquired, resistance-mediating aminoglycoside methyltransferases.
MATERIALS AND METHODS
Strains and plasmids.E. coli BW25113 (lacIq rrnBT14 ΔlacZWJ16 hsdR514 ΔaraBADAH33 ΔrhaBADLD78), here called BB1074 wt, was used for the construction of the mutants. BB1075 ΔrsmF, BB1076 rmtC+, and BB1077 ΔrsmF rmtC+ are the mutants constructed in this work. Salmonella Virchow H0 5366 0426 bearing rmtC was used as the DNA template for the construction of the rmtC insertion mutant (20). E. coli MUR050 bearing armA and E. coli bearing rmtB were used for matrix-assisted laser desorption ionization mass spectrometry (MALDI MS) (16).
The vector used for the cloning of the rmtB PCR product was pCRII-pTOPO (Invitrogen, Life Technologies, Carlsbad, CA), and E. coli Top10 was used as the receptor for the transformation of the pCRII-pTOPO cloning vector. plasmid pKD13 (GenBank Accession number AY048744) contains an FLP recombination target (FRT)-flanked kanamycin resistance (kan) gene and was used as a PCR template in order to construct the deletion mutants. pKD46 Red helper plasmid (GenBank Accession number AY048746) (9) was used for the homologous recombination of the generated PCR products with the chromosome of E. coli BW25113, and pCP20 FLP helper plasmid from BT340 (DH5a carrying pCP20 [5]) was used for the elimination of the FRT-flanked resistance gene.
Plasmid stability experiments.The curing rates of plasmid pTOPO and pTOPO bearing rmtB in E. coli Top10 transformants were determined as previously described (31). Strains were grown for 16 h at 37°C in LB medium, and then 106 CFU of each strain was grown in 50 ml of antibiotic-free LB medium at 37°C and 100 rpm. The cultures were diluted 1:1,000 into 50 ml of fresh LB medium every 24 h. Samples were taken every 24 h for 8 days. The proportion of bacteria harboring pTOPO was determined by plating subculture steps on brain heart infusion (BHI) agar plates and further replication on BHI agar plates and BHI agar plates containing ampicillin (25 μg/ml) or kanamycin (25 μg/ml). Colonies that had lost the plasmids were able to grow on BHI agar plates but not in BHI broth with ampicillin or kanamycin, while colonies harboring pTOPO or pTOPO bearing rmtB were able to grow on both plates. The rate of plasmid curing was calculated (data not shown).
Construction of single-gene deletion rsmF mutants and insertion rmtC mutants in E. coli.E. coli BW25113 carrying the Red helper plasmid pKD46 was used for the construction of the deletion and the insertion mutants. To carry out the in-frame rsmF knockout, a PCR was performed using pKD13 DNA as the template and Phusion high-fidelity DNA polymerase (Finnzymes, Woburn, MA) for the amplification of the FRT-flanked kanamycin resistance (kan) gene subsequently used for mutant selection (3). The 5′-terminal deletion primer had a 50-nucleotide (nt) homologous extension that include the rsmF initiation codon and the 20-nt 5′-ATTCCGGGGATCCGTCGACC-3′ priming site from pKD13, and the C-terminal deletion primer consisted of a 50-nt 5′-homologous extension that include 21 nt for the rsmF C-terminal region, the termination codon and 29 nt downstream, and the 20-nt 5′-TGTAGGCTGGAGCTGCTTCG-3′ priming site from pKD13.
Chromosomal rsmF was deleted using a PCR amplicon as previously described (9). pCP20 FLP helper plasmid was used for the elimination of the FRT-flanked resistance gene from the rsmF deletion mutants as previously described (9). To carry out the rmtC insertion, a PCR using rmtC-positive Salmonella Virchow DNA as the template was performed. The rmtC upstream insertion primer had a 50-nt 5′-homologous extension that includes the ygcE-ygcF intergenic region and the 20-nt 5′-GTTGCTCTGTGGATAACTTGC-3′ priming site located upstream of the rmtC gene, and the rmtC downstream insertion primer consisted of 50 nt homologous to the ygcE-ygcF intergenic region (23) and the 20-nt 5′-TGCAAGGCTAGAGTCAAGCCA-3′ priming site located downstream of the rmtC gene. Chromosomal rmtC was inserted using a PCR amplicon as previously described (9). PCRs and sequencing were used to confirm the genetic structures (data not shown).
Growth rate and growth competition experiments.Growth kinetics were determined for BB1074 wt and for the BB1075 ΔrsmF, B1076 rmtC+, and BB1077 ΔrsmF rmtC+ mutants. Volumes of 50 ml LB were inoculated independently with 107 CFU of each strain. Cultures were grown for 12 h at 37°C and 100 rpm, and the absorbance at 600 nm was measured every hour. Different pairs of these mutants were used in four independent competition experiments as previously described (27). Strains were grown for 16 h at 37°C in LB medium, and then 106 CFU of each strain was mixed in 50 ml of antibiotic-free LB medium. The mix was grown at 37°C and 100 rpm. The cultures were diluted 1:250 into 50 ml of fresh LB medium every 24 h. Samples were taken every 24 h for 10 days. For the competition experiments in which one competitor carried the aminoglycoside resistance gene rmtC for selection, i.e., BB1074 wt versus BB1076 rtmC+, BB1074 wt versus BB1077 ΔrsmF rmtC+, and BB1075 ΔrsmF versus BB1077 ΔrsmF rmtC+, aliquots were plated on nonselective LB agar, and the proportion of resistant colonies was deduced by replica plating of 100 colonies on LB agar plates containing 50 μg/ml of kanamycin. Antimicrobials were supplied by Sigma-Aldrich (Sigma Chemical Co., St. Louis, MO). For the competition experiment with BB1074 wt versus BB1075 ΔrsmF, which have no antibiotic resistance selective gene, aliquots were plated on nonselective LB agar, and the proportion of ΔrsmF knockout mutants was deduced by performing PCR of 100 colonies with the oligonucleotides rsmFant.F (5′-AACCGAAAGAGTGGCGTGAAT-3′) and rsmFpost.R (5′-TTCAGCCTAGTTGAGGCAGCC-3′) in order to check whether the rsmF gene was present or not. The competition index (CI) represents the relative fitness and was calculated as the ratio of the mean CFU in four independent competition experiments for the resistant and susceptible strains at t1 divided by the same ratio at t0. The selection coefficient s was calculated as the slope of the linear regression model s = ln(CI)/ln(d), where d is the dilution factor (27). The selection coefficient estimates the difference between the relative fitnesses of the two competitors over the entire competition experiment.
Antimicrobial susceptibility testing.Antimicrobial susceptibilities of E. coli strains were determined by broth microdilution in microtiter plates (Sensititre EUMVS2; Trek Diagnostics, Inc., Westlake, OH), except for arbekacin, tobramycin, and neomycin, for which susceptibilities were determined with in-house microtiter plates according to the CLSI guidelines (6). Antibiotic quality control of the panels was performed with collection strain ATCC 25922 (E. coli).
Analysis of rRNA by MALDI mass spectrometry.Total rRNA was extracted from ribosomal particles isolated from E. coli BB1074 wt, the E. coli BB1076 rmtC+ mutant, E. coli MUR050 (armA), and E. coli (rmtB). The 16S rRNA sequence from C1378 to G1432 was isolated by hybridization to complementary deoxyoligonucleotides, and 100 pmol of rRNA was heated with 500 pmol of deoxyoligonucleotide (complementary to the sequence shown in Fig. 1) for 1 min at 85°C, followed by slow cooling over 2 h to 45°C. Unhybridized regions of the rRNAs were digested with mung bean nuclease (NE Biolabs) and RNase A (Sigma), and the rRNA fragment protected by the deoxyoligonucleotide was isolated by gel electrophoresis (2). The rRNA fragment was digested at 37°C overnight in 2 μl of 50 mM 3-hydroxypicolinic acid containing 20 units of RNase T1 (Roche). Samples were analyzed by MALDI MS (Voyager Elite; PerSeptive Biosystems) recording in reflector and positive-ion mode (22). Spectra were analyzed using the program m/z (Proteometrics, Inc.).
(A) Three-dimensional structure of the E. coli 30S subunit structure (Protein Data Bank [PDB] code 3I1M) (36). Helix 44 (gray) and the ribosomal A site (red) on the 16S rRNA are indicated. (B) Representation of the E. coli 16S rRNA secondary structure showing the sequence in helix 44 that was isolated for analysis by mass spectrometry. The sites of posttranscriptional methylations in this region, including m5C1407, are indicated. The site of the m7G1405 modification added by the Arm/Rmt methyltransferase family is also shown. (C to E) Aminoglycoside interactions at the ribosomal A site on 16S rRNA (PDB codes 2ESI and 2ET4 [14] and 1YRJ [18]). The ribose skeleton of the 16S rRNA is shown in gray. Nucleotides A1492 and A1493 (dark gray), which participate in the fidelity of the cognate mRNA-tRNA codon-anticodon association, “flip out” of helix 44 in the presence of aminoglycosides. Nucleotides m7G1405 (blue) and m5C1407 (yellow) are shown. Methyl groups from both nucleotides (pink) are also indicated. Kanamycin 4,6-deoxystreptamine aminoglycoside (red) (C), neomycin B 4,5-deoxystreptamine (green) (D), and apramycin 4-deoxystreptamine aminoglycoside (brown) (E) are also represented. N4 of C1407 is implicated in the binding of these aminoglycosides, since it interacts both with ring III of kanamycin and neomycin B and with ring I of apramycin (dotted lines). Note that the methyl group in m7G1405 directly clashes only with ring III of kanamycin and not with neomycin B nor apramycin. These pictures were drawn with MacPyMOL (Schrödinger, LLC).
RESULTS AND DISCUSSION
The resistance methyltransferases ArmA, RmtB, and RmtC impede the RsmF housekeeping methylation.The methylation site of the widely spread 16S rRNA methyltransferases conferring high-level resistance to aminoglycosides ArmA and RmtB has been shown to be m7G1405 (24, 28). Here, we performed MALDI mass spectrometry to identify whether the emerging aminoglycoside resistance Arm/Rmt methyltransferases found in Enterobacteriaceae also impede the housekeeping methylation at m5C1407. As expected, in the bacteria bearing either ArmA or RmtB, methylation at G1405 and C1407 was evident (Fig. 2). Interestingly, a second fragment, namely, 4485.4/4467.4 mCmCCGmUCACACCAUG was identified, in which the methyl group at position m7G1405 (boldface) was present but the housekeeping methyl group at position m5C1407 was absent. Thus, as shown for the sisomicin-gentamicin resistance methyltransferase from the sisomicin-producing actinomycete Micromonospora zionensis Sgm (7), Arm/Rmt methyltransferases from pathogenic bacteria reduce methylation at the adjacent rRNA nucleotide m5C1407. However, the biological significance of this phenomenon remains unexplored. We decided to study the fitness costs of 16S rRNA methyltransferases of the Arm/Rmt family and the RsmF (rRNA small-subunit methyltransferase) housekeeping methyltransferase, which is responsible for the m5C1407 methylation in E. coli (1).
MALDI MS spectra of RNase T1 oligonucleotides from the E. coli 16S rRNA sequence C1378 to G1432. (A) Theoretical monoisotopic masses of the RNase digestion products. Two different values for the same fragments are indicated, depending on whether they have 3′-end cyclic phosphates or are linear fragments (boxes in gray). Only fragments that are trinucleotides and larger are shown. The theoretical and empirical masses (given above the peaks) match to within 0.2 Da. (B to D) Corresponding spectra of rRNAs from E. coli BB1074 wt with rsmF, showing the fragments at m/z 1307 and 3197 (B), E. coli with armA/rmtB in the presence of aminoglycosides (C), and the E. coli BB1076 rmtC+ insertion mutant in the absence of aminoglycosides (D). The experiments took place at different times in such a way that for E. coli BB1074 wt (B) and E. coli arm/rmt (C) we obtained fragments with the linear form and for the E. coli BB1076 rmtC+ mutant (without gentamicin) (D) (experiment performed later) we obtained fragments with 3′-end cyclic phosphates. Panels C and D show essentially the same mass spectra, in which the fragments at m/z 1307 and 3197 are missing and remain combined in the longer sequences at m/z 4467/4485 when G1405 is methylated and 4481/4499 when both G1405 and C1407 are methylated. The enlargement of this spectral region shows the multiple tops in these fragments that reflect the natural distribution of 12C and 13C isotopes.
To address this subject, we first analyzed the stability of the rmtB gene in E. coli when borne on a multicopy plasmid. As soon as 4 days after serial passages in the absence of selective pressure, only 20% of the cells still bore the rmtB gene, whereas the mock plasmid pTOPO remained present in all colonies tested (data not shown). Thus, further study of Arm/Rmt genes in high-copy-number plasmids was not optimal. In order to mimic natural conditions as much as possible and to assess the fitness cost of the resistance determinants without interference of plasmids, we decided not to artificially subclone these genes in other vectors or to use other antibiotics to maintain the plasmid stability. Instead, we took advantage of another 16S rRNA methyltransferase from pathogenic bacteria of the Arm/Rmt family, RmtC, that has recently been identified in enterobacteria in the United Kingdom (20) and the United States (13). In these clinical isolates, rmtC was located on the chromosome and transcribed from the constitutive promoter of the ISEcp1 element located immediately upstream of rmtC (33). We integrated rmtC and its native promoter into the chromosome of E. coli BB1074 wt, giving rise to strain BB1076 rmtC+. The integration site of the construct was the ygcE-ygcF locus, which has recently been shown to be suitable for knock-in of genes into E. coli (23). MALDI mass spectrometry of BB1076 rmtC+ showed that, as for ArmA and RmtB, RmtC methylation at m7G1405 impedes methylation of m5C1407 (Fig. 2).
The RsmF housekeeping methylation contributes to aminoglycoside resistance in E. coli.In order to assess the consequences of the loss of housekeeping methylation at m5C1407, the rsmF gene was deleted in frame in E. coli BB1074 wt, giving rise to BB1075 ΔrsmF. The growth curves of strains BB1074 wt and BB1075 ΔrsmF were indistinguishable, indicating that the loss of m5C1407 had no distinct effect on the growth kinetics of the bacterium. However, competition experiments offer a more discriminative and precise measurement of fitness, since they reflect the competitive disadvantage during all the phases of the growth cycle and in several consecutive cycles (27). Strains BB1074 wt and BB1075 ΔrsmF were grown in competition experiments, and the proportion of bacteria carrying the rsmF deletion was monitored every 24 h for 10 days. One hundred colonies were checked by PCR at each time point to determine the presence or absence of the rsmF gene. From day 2 on, a constant decrease in the proportion of the rsmF mutant strain was observed. At day 6, nearly all the colonies tested were BB1074 wt (Fig. 3). The selection coefficient (s) was established to estimate the difference in relative fitness between the strains in competition over the entire experiment. E. coli BB1075 ΔrsmF had a selection coefficient s of −0.0767, which means that it possesses a competitive disadvantage of ca. 7.7% per 10 generations relative to the parental strain E. coli BB1074 wt. As the methylation site of RsmF is adjacent to the binding site of aminoglycosides (Fig. 1), we hypothesized that BB1075 ΔrsmF could potentially possess an altered pattern of susceptibility to this family of antimicrobials. MIC measures with this strain compared to the parental strain revealed that the mutant BB1075 ΔrsmF was more resistant to kanamycin, tobramycin, gentamicin, neomycin, apramycin, and arbekacin (Table 1). These results reveal that the housekeeping methyltransferase RsmF is involved not only in the function of the ribosome but also in intrinsic susceptibility to aminoglycosides in E. coli. It is known that ring III of neomycin and kanamycin-like aminoglycosides as well as ring I of apramycin interact with N4 of the nucleotide C1407 (14, 18). The absence of the methyl group in the N5 position of C1407 seems to destabilize the contact between these aminoglycosides and the A site pocket, giving rise to an increase of resistance when the N5 1405 methylation is missing. The currently available crystal structures elucidate the involvement of RmtC methylation in the binding of these molecules to the aminoglycoside binding pocket of the 16S rRNA ribosome (Fig. 1).
Fitness cost of RsmF and RmtC in different backgrounds. Growth competition curves between BB1074 wt and BB1075 ΔrsmF (A), BB1074 wt and BB1077 ΔrsmF rmtC+ (B), BB1074 wt and BB1076 rmtC+ (C), and BB1075 ΔrsmF and BB1077 ΔrsmF rmtC+ (D) are shown. Competition curves were constructed with the averages from three independent experiments. The selection coefficient (s) was calculated from the competition experiment. s is the slope of the linear regression model ln(CI)/ln(d), where CI is the ratio of the CFU of the resistant and susceptible populations at t1 divided by the same ratio at t0 and where d is the dilution factor.
Aminoglycoside susceptibilities of E. coli with and without rsmF and/or rmtC
RmtC has no fitness cost in E. coli.We have identified that the Arm/Rmt methyltransferases impede methylation by the endogenous methyltransferase RsmF at position m5C1407 and that the absence of this methylation affects aminoglycoside resistance and entails a strong fitness cost for the bacterium. We next questioned whether methylation at position m7G1405 by Arm/RmtB methyltransferases could compensate for the competitive disadvantage of losing rsmF. To assess this, the rmtC gene was integrated into the chromosome of the rsmF mutant strain BB1075 ΔrsmF, generating strain BB1077 ΔrsmF rmtC+. This strain was used in fitness competition experiments against the original wild-type bacterium with rsmF. The fitness cost of lacking rsmF but bearing rmtC was 6% per 10 generations relative to the wild-type parental strain (Fig. 3). These results reveal that in the absence of rsmF, an acquired aminoglycoside resistance methyltransferase gene has no pronounced fitness cost.
To gain insight into the fitness cost of RmtC in a wild-type E. coli background, we then introduced the rmtC gene with its native promoter into the chromosome of the E. coli parental strain, creating BB1076 rmtC+. MICs of the corresponding 4,6-disubstituted deoxystreptamines gentamicin, kanamycin, and tobramycin and the Japanese aminoglycoside arbekacin increased to the levels for clinical isolates bearing RmtC (Table 1), showing that the gene was expressed as in pathogenic wild-type enterobacteria (20). BB1076 rmtC+ and its parental strain BB1074 wt mimic the environmental situation, in which E. coli becomes resistant to these clinically relevant aminoglycosides through acquisition of an Arm/Rmt methyltransferase. In the presence of antibiotics, it is evident that acquisition of the resistance gene is an evolutionary advantage for the bacterium. However, when the antimicrobial pressure disappears, e.g., when therapy is stopped, the fitness burden of resistance is in most cases so high that the antimicrobial resistance determinants are diluted in the bacterial population. In order to be able to predict whether Arm/Rmt genes would be counterselected in the absence of antibiotic pressure, competition experiments between the BB1074 wt and BB1076 rmtC+ were performed. Surprisingly, and contrary to the accepted idea that the fitness burden of antimicrobial resistance is high, the competitive disadvantage of bearing the rmtC gene was only 0.39% per 10 generations, showing that rmtC has a negligible fitness cost in E. coli.
Finally, to assess the effect of RmtC in the absence of the housekeeping RsmF methyltransferase, competition experiments were performed between strains BB1075 ΔrsmF and BB1077 ΔrsmF rmtC+. Interestingly, the fitness cost of RmtC was again only 0.69% per 10 generations, showing that the clash with the housekeeping methyltransferase RsmF does not influence the fitness cost entailed by RmtC (Fig. 3). However, the MIC of strain BB1077 ΔrsmF rmtC+ was overall higher for the 4,6-disubstituted deoxistreptamines gentamicin, kanamycin, tobramycin and arbekacin than that of the parental E. coli BB1076 rmtC+ (Table 1). The difference in the MICs for strains BB1076 rmtC+ and BB1077 ΔrsmF rmtC+ is due to the effect of RsmF on aminoglycoside resistance that we describe here, since this increment in resistance is also observed between E. coli BB1074 wt and BB1075 ΔrsmF.
RmtC methylates m7G1405 in the presence and absence of aminoglycosides.Since RmtC, like Sgm, in the presence of aminoglycosides impedes methylation at the N5 position of nucleotide C1407 when the N7 position of G1405 is methylated, we further questioned what would happen in the absence of antibiotics. Our model with the chromosomal rmtC gene is ideal to answer this question. We thus performed MALDI mass spectrometry experiments with total rRNA extracted from ribosomal particles from the strain BB1076 rmtC+ that had been grown in the presence and absence of gentamicin. Interestingly, in both experimental scenarios, the residue m7G1405 was always methylated, whereas the methylation in m5C1407 was impeded in both cases (Fig. 2). These results unequivocally demonstrate that RmtC has high affinity for methylation of m7G1405 and clashes with the housekeeping modification enzyme RsmF. This experiment shows how effectively a resistance methyltransferase that has been acquired as an exogenous genetic element can elbow aside the cell's own housekeeping methyltransferases to gain access to its target.
Thus, acquisition of an aminoglycoside resistance methyltransferase modifies the methylation pattern in the 16S rRNA but does not entail a major fitness cost for the host bacterium. This means that the evolution of the ribosome toward resistance implies complex counterintuitive phenomena that ultimately lead to selection and maintenance of resistance even in the absence of selective pressure.
ACKNOWLEDGMENTS
We acknowledge Natalia Montero for excellent technical assistance and Elisabeth Diago-Navarro for excellent assistance with MacPyMol. We thank Y. Arakawa for providing the rmtB gene.
We thank the Universidad Complutense de Madrid for a Ph.D. scholarship to B.G., the Spanish Ministry of Science and Innovation (MICINN) for supporting Ph.D. scholarships to L.C. and C.M.O., and the Comunidad de Madrid for a scholarship to L.H. This work was supported by grants from the Spanish Ministry of Science and Innovation (BIO 2010-20204, PRI-PIBIN-2011-0915, and BFU2011-14145-E), the EU FP7 Health Project EvoTAR, TRAIN-ASAP, and the Programa de Vigilancia Sanitaria 2009 AGR/4189 of the Comunidad de Madrid (Spain).
FOOTNOTES
- Received 3 November 2011.
- Returned for modification 24 December 2011.
- Accepted 5 February 2012.
- Accepted manuscript posted online 13 February 2012.
- Copyright © 2012, American Society for Microbiology. All Rights Reserved.
