Revisiting the mechanism of macrolide-antibiotic resistance mediated by ribosomal protein L22

Sean D. Moore and Robert T. Sauer
PNAS November 25, 2008 105 (47) 18261-18266; https://doi.org/10.1073/pnas.0810357105

  1. Contributed by Robert T. Sauer, October 16, 2008 (received for review September 13, 2008)

Abstract

Bacterial antibiotic resistance can occur by many mechanisms. An intriguing class of mutants is resistant to macrolide antibiotics even though these drugs still bind to their targets. For example, a 3-residue deletion (ΔMKR) in ribosomal protein L22 distorts a loop that forms a constriction in the ribosome exit tunnel, apparently allowing nascent-chain egress and translation in the presence of bound macrolides. Here, however, we demonstrate that ΔMKR and wild-type ribosomes show comparable macrolide sensitivity in vitro. In Escherichia coli, we find that this mutation reduces antibiotic occupancy of the target site on ribosomes in a manner largely dependent on the AcrAB-TolC efflux system. We propose a model for antibiotic resistance in which ΔMKR ribosomes alter the translation of specific proteins, possibly via changes in programmed stalling, and modify the cell envelope in a manner that lowers steady-state macrolide levels.

Many antibiotics inhibit bacterial protein synthesis. Understanding how microbes become antibiotic resistant is important both for developing effective treatment regimens and designing new therapeutics. Macrolides consist of a 14- to 16-member lactone ring with different appended sugars and comprise a key group of inhibitors of bacterial translation (1, 2). The inhibitory activity of macrolides, including erythromycin, depends on binding to a site near the polypeptide exit tunnel of the large ribosomal subunit (3, 4). Because macrolides do not bind to ribosomes with an occupied exit tunnel and cause the synthesis of 2–10 residue peptides in translation assays in vitro, it has been proposed that drug binding physically blocks elongation of nascent proteins beyond this size (5–7).

Some macrolide-resistance mutations alter the ribosomal target site and prevent binding (4, 8). Intriguingly, other mutations confer resistance despite the fact that macrolides still bind the mutant ribosome well (8–10). For example, deletion of the M82K83R84 sequence in Escherichia coli ribosomal protein L22 (ΔMKR) allows growth in the presence of high levels of erythromycin and other macrolides (11–13). The same mutation makes Haemophilus influenzae resistant to numerous macrolides (14); different L22 mutations also confer macrolide resistance in other bacterial species (2). When binding has been measured, ribosomes with macrolide-resistant alterations in L22 bind erythromycin with near wild-type affinity (8, 10, 12, 15).

In a crystal structure of the E. coli ribosome, the MKR sequence is part of an extended L22 loop, which together with a similar loop in protein L4 forms a narrow constriction in the exit tunnel (Fig. 1A) (16, 17). Cryo-EM studies initially revealed a widened exit tunnel in E. coli ΔMKR ribosomes (18). This loop is also displaced to create an expanded tunnel in structures of ΔMKR ribosomes from Thermus thermophilus and Haloarcula marismortui (4, 19). These results explain the altered chemical reactivity in E. coli ΔMKR ribosomes of 23S-RNA bases (13). Together, these results support a prevailing model in which the L22 ΔMKR deletion confers antibiotic resistance by allowing nascent proteins to enter the ribosome exit tunnel despite bound macrolides (18, 19).

Here, we rule out the accepted model of L22-mediated macrolide resistance by showing that E. coli ΔMKR ribosomes are inhibited by erythromycin in vitro and in vivo. Instead, the ΔMKR mutation appears to reduce the intracellular concentration of macrolides. We present evidence that links changes in the activity of the AcrAB-TolC efflux system to antibiotic resistance in the ΔMKR strain and suggest that changes in translation of specific proteins are responsible for ΔMKR-linked macrolide resistance.

Results

ΔMKR Ribosomes Are Inhibited by Erythromycin in Vitro.

The E. coli S10 operon, encoding L22 and additional ribosomal proteins, can be deleted without substantially altering the growth rate if the cells contain a plasmid-borne operon encoding L22 or L22 fused to the titin-I27 domain (20). The L22-titin fusions are longer than unmodified L22 and this feature allowed us to establish unambiguously that mutant forms of L22 were the only versions present in cell (20). Using this system, we constructed a strain expressing only L22-titin with the ΔMKR mutation. We purified ribosomes from strains expressing just L22-titin or just ΔMKR-L22-titin; SDS-PAGE of these ribosomes revealed that the ΔMKR L22 variant had a slightly faster electrophoretic mobility (Fig. 1B). In coupled transcription/translation reactions, both types of ribosomes were equally active in the synthesis of 14C-labeled test proteins (Fig. 1C Left) and had activities comparable to ribosomes containing unfused L22 (data not shown). Surprisingly, however, increasing concentrations of erythromycin inhibited the ΔMKR-L22 ribosomes and the “wild-type” ribosomes (Fig. 1 C and D). Fitting these data gave apparent inhibition constants (Ki) of 41 ± 9 nM for the L22-titin ribosomes and 26 ± 5 nM for the ΔMKR-L22-titin ribosomes (Fig. 1D). The Ki values were within the range of KD values (≈10–100 nM) measured for erythromycin binding to ribosomes containing wild-type L22 or ΔMKR-L22 (21–24). At the highest concentrations of erythromycin tested, no translation by ΔMKR ribosomes was detected (Fig. 1C Bottom). Erythromycin also completely inhibited translation of another mRNA by L22-titin and ΔMKR-L22-titin ribosomes (data not shown). Thus, these experiments are inconsistent with models in which the ΔMKR deletion in L22 allows the ribosome to function when erythromycin is bound.

Fig. 1.
Fig. 1.

Ribosomal protein L22 and erythromycin inhibition of translation in vitro. (A) Crystal structure of L22 protein (green; M82K83R84 is shown in red), L4 protein (blue), and 23S and 5S RNA (orange) from the 50S subunit of the E. coli ribosome (PDB entry 2AWB). The exit tunnel is visible through the center of the subunit. (B) SDS-PAGE of proteins from purified ribosomes with wild-type L22, an L22-titin fusion, or a ΔMKR L22-titin fusion. Wild-type L22 (12 kDa) is not resolved from other ribosomal protein in the Left lane. The L22 fusion proteins are visible as separate bands (Center and Right lanes). (C) Autoradiograms of translation reactions containing 100 nM wild-type or ΔMKR ribosomes and increasing amounts of erythromycin. (D) Integrated band intensities from the autoradiogram shown in the Top and Middle of C are plotted as a percentage of the intensity of the band from the reaction without erythromycin. These data were fit to a quadratic binding equation to obtain the inhibition constant (Ki).

Erythromycin Sensitivity in Vivo.

As anticipated from previous studies, in liquid cultures of strains expressing only L22-titin or ΔMKR-L22-titin, the mutation increased the erythromycin minimal inhibitory concentration (MIC) from ≈125 to ≈850 μM (Fig. 2A, dashed lines), doubled the spiramycin MIC (≈500 to ≈1000 μg/ml; data not shown), and quadrupled the tylosin MIC (≈500 μM to ≈2 mM; data not shown). E. coli strains containing plasmid-borne L22 variants lacking the titin-fusion domain behaved similarly (data not shown) as did strains with chromosomal alleles of wild-type L22 or ΔMKR L22 (2, 11–13). Although sufficiently high concentrations of macrolides inhibited growth of both the L22-titin and ΔMKR-L22-titin strains, as expected from our results in vitro, the differential macrolide sensitivity in vivo was not recapitulated in the in vitro assays.

Fig. 2.
Fig. 2.

Erythromycin sensitivity increases when drug efflux is blocked. (A) Erythromycin resistance of strains containing wild-type L22-titin (triangles) or ΔMKR L22-titin (circles). The turbidity (600 nm) of 16 h cultures grown at 37 °C in the presence of erythromycin (dashed lines) or erythromycin and 30 μg/ml PAβN (solid lines) is plotted as a percentage of the value of a culture grown without drug. (B) Growth at 37 °C of cultures containing the wild-type L22-titin fusion in the absence of erythromycin (open circles), after addition of 200 μM erythromycin (filled circles), or after addition of 200 μM erythromycin and 30 μg/ml PAβN (filled triangles). (C) Growth of cultures containing the ΔMKR mutation in L22 with the same symbols as in B. (D) Strains containing wild-type (circles) or ΔMKR L22 (filled circles) and an induced ermC gene were grown in medium with 30 μg/ml PAβN to early log phase and erythromycin was added to 200 μM at the time indicated by the arrows. (E) ErmC methylation of A2058 in 23S RNA assayed by primer extension. An oligonucleotide complementary to 23S rRNA bases 2066–2102 was used to prime a reverse-transcription reaction containing dideoxy-ATP and RNA purified from strains containing wild-type or ΔMKR L22-titin fusions grown with or without induction of ermC. Reaction products were electrophoresed on a urea-acrylamide gel. In the presence of template RNA, reverse transcription proceeded until blocked by ErmC-mediated methylation of A2058 or until incorporation of dideoxy-ATP at the position corresponding to U2041.

Efflux Pumps and Macrolide Resistance.

Efflux pumps can dramatically alter the sensitivity of bacteria to antibiotics (2). For example, phenylanine-arginine-β-napthylamide (PAβN) inhibits multidrug efflux pumps and markedly lowers the MIC of erythromycin and other antibiotics (25, 26). In strains expressing only L22-titin or only ΔMKR-L22-titin, 30 μg/ml PAβN lowered the MIC for erythromycin >20-fold (Fig. 2A, solid lines) and the MICs for tylosin and spiramycin >10-fold (data not shown). Therefore, drug efflux plays a significant role in determining the macrolide resistance of both strains. In liquid culture, some growth of the L22-titin strain was observed 2 h after addition of 200 μM erythromycin (Fig. 2B). By contrast, growth ceased almost immediately with 30 μg/ml PAβN and 200 μM erythromycin (Fig. 2B), suggesting that active erythromycin efflux normally delays growth inhibition. For the ΔMKR-L22-titin strain, 200 μM erythromycin had little effect on growth in the absence of the efflux inhibitor but caused rapid cessation of growth in its presence (Fig. 2C).

Erythromycin Inhibits L22-ΔMKR Ribosomes in Vivo.

Is erythromycin sensitivity of the ΔMKR-L22-titin strain a consequence of drug binding to the mutant ribosome or a different cellular target? To answer this question, we used ErmC, which methylates A2058 in 23S rRNA and blocks erythromycin binding (27, 28). Constitutive expression of ErmC allowed L22-titin or ΔMKR-L22-titin strains to grow in the presence of 200 μM erythromycin and PAβN (Fig. 2D). The growth rates were approximately half those without erythromycin, which could be caused by incomplete ErmC methylation. Indeed, consistent with a previous report (28), a primer-extension assay, using reverse transcriptase showed that A2058 was methylated in only ≈60% of the ribosomes (Fig. 2E). Because ErmC-mediated resistance to erythromycin correlated approximately with the extent of A2058 methylation, we conclude that erythromycin inhibits growth of the ΔMKR-L22-titin strain by occupying its normal binding site in the exit tunnel of ΔMKR ribosomes.

Erythromycin Binding to Ribosomes in Vivo.

If ΔMKR cells accumulated or retained less erythromycin than wild-type cells, then ΔMKR ribosomes would appear to be more erthyromycin resistant in vivo. Because erythromycin protects A2058 and A2059 in the 23S rRNA of ribosomes from dimethyl sulfate (DMS) modification in vitro (22, 29), we used DMS reactivity in vivo to monitor erythromycin binding. L22-titin or ΔMKR-L22-titin cells were grown without drug, with 200 μM erythromycin, or with 200 μM erythromycin and PAβN, DMS was added, and total RNA was purified and used for primer-extension reactions. The bands corresponding to A2058/A2059 exhibited similar intensities in both strains grown without erythromycin and showed reduced but comparable intensities in L22-titin or ΔMKR-L22-titin cells treated with erythromycin and PAβN (Fig. 3A). By contrast, A2058/A2059 were modified to a greater extent in the ΔMKR-L22-titin strain than in the L22-titin strain after growth in erthyromycin alone (Fig. 3A). The reactivity of A2062 did not change substantially with or without erythromycin, allowing normalization of the A2058/A2059 intensities for quantitative comparisons (Fig. 3B). Irrespective of the L22 allele, A2058/A2059 reactivities were high in the absence of erythromycin and low in the presence of very high intracellular concentrations of erythromycin (mediated by PAβN inhibition of efflux). Importantly, A2058/A2059 reactivities were approximately twice as high in ΔMKR-L22-titin ribosomes than in L22-titin ribosomes during growth in erthyromycin alone, supporting a model in which the intracellular concentration of erythromycin is lower in the ΔMKR cells.

Fig. 3.
Fig. 3.

Reduced erythromycin binding to ribosomes with ΔMKR L22 in vivo. Cultures were treated with dimethyl sulfate (DMS), reactions were quenched, and total RNA was purified and used in primer-extension assays to detect base modifications near the erythromycin-binding site in 23S rRNA. (A) Gel of primer-extension products from the following RNA templates: (Lane 1) Non-DMS-treated ErmC-modified template (product terminates at the position corresponding to A2058). (Lanes 2 and 3) Templates from cultures grown without erythromycin. (Lanes 4 and 5) Templates from cultures grown with 200 μM erythromycin. (Lanes 6 and 7) Templates from cultures grown with 200 μM erythromycin and 30 μg/ml PAβN. (B) Plots of the ratio of band intensities corresponding to A2058 (filled bars) or A2059 (hatched bars) in A divided by the intensity of A2062.

Larger Deletions of the MKR Loop Do Not Confer Erythromycin Resistance.

Ribosomes containing L22 mutants with larger deletions in the MKR loop appear to be functional in merodiploid strains, because they are found with wild-type ribosomes in polysomes (30). To test whether these larger deletions allowed ribosome function in the presence of erythromycin, we constructed strains expressing just L22-titin variants missing residues 85–95 (Δloop1) or residues 82–100 (Δloop2) (Fig. 4A). In growth-rate studies, the doubling times were 34 min (L22-titin), 53 min (ΔMKR-L22-titin), 57 min (Δloop1-L22-titin), and 62 min (Δloop2-L22-titin) (data not shown). Thus, all of the deletions allow bacterial growth but with significant growth defects. The Δloop1 and Δloop2 strains did not grow on plates with 200 μM erythromycin (Fig. 4B) and had erythromycin MICs in liquid culture of ≈125 μM and ≈150 μM, respectively (data not shown). Thus, partial or complete removal of the MKR loop from the narrow constriction in the exit tunnel is not sufficient to confer significant increases in erythromycin resistance. Because strains with the extended loop mutants grew more slowly than the ΔMKR mutant, these results also show that L22-mediated slowing of the growth rate is not sufficient to confer enhanced macrolide resistance. We also constructed a strain in which M82K83R84 of L22 was replaced with A82A83A84. This strain grew as well as wild type in the absence of erythromycin and was equally sensitive to the antibiotic (data not shown). Thus, the chemical identity of the side chains at these residue positions is unrelated to the resistance phenotype caused by the ΔMKR deletion.

Fig. 4.
Fig. 4.

Extended L22 loop deletions. (A) Ribbon drawings of E. coli L22 with the M82-K83-R84 residues marked in red (PDB 2AWB). The Δloop1 and Δloop2 deletions remove residues 85–95 and 82–100, respectively. (B) Growth of strains on a plate containing 200 μM erythromycin.

acrAB-tolC System and ΔMKR Macrolide Resistance.

In E. coli, the AcrAB-TolC efflux system is primarily responsible for resistance to low levels of erythromycin and a variety of other toxic compounds (31–32). To determine whether this system was responsible for the increased resistance conferred by the ΔMKR mutation, we deleted either the AcrA/AcrB or TolC ORFs and assayed resistance to erythromycin. Deletion of either tolC or acrAB abolished much of the difference in erythromycin resistance between strains expressing L22-titin or ΔMKR-L22-titin and substantially increased the sensitivity of both strains (Fig. 5 A and B). Therefore, increased drug efflux mediated by AcrAB-TolC appears to be largely responsible for the increased erythromycin resistance conferred by the ΔMKR mutation. The tolC deletion reduced the difference in erythromycin sensitivity between the wild-type and ΔMKR strains to a greater extent than the acrAB deletion. This result can be rationalized because several TolC-dependent efflux pumps in addition to ArcAB-TolC have also been shown to plays roles in erythromycin resistance (32, 34).

Fig. 5.
Fig. 5.

Contribution of AcrAB-TolC to erythromycin resistance. The TolC or AcrA/AcrB components of the AcrAB-TolC efflux pump were deleted from strains with wild-type or ΔMKR ribosomes. (A) Erythromycin resistance of cells lacking tolC with wild-type (triangles) or ΔMKR (circles) ribosomes. (B) Erythromycin resistance of cells lacking acrA/acrB with wild-type (triangles) or ΔMKR (circles) ribosomes. Inhibition plots of tolC+ acrAB+ cells are shown for comparison (dashed lines).

Discussion

The discovery that bacterial resistance to erythromycin can be caused by mutations in ribosomal proteins was first reported in 1967, and the ΔMKR deletion in protein L22 was subsequently shown to cause this phenotype (8, 11, 12). Since these initial observations, consensus molecular mechanisms for macrolide inhibition of translation and for the effects of the ΔMKR mutation have emerged from biochemical and structural studies (3, 4, 7, 18, 19). According to the exit-tunnel blockade/bypass model, macrolides sterically block peptide elongation during translation by binding in the exit tunnel of the large ribosomal subunit, with the ΔMKR deletion serving to relieve this blockade by providing additional space for the nascent chain to bypass the drug and enter the exit tunnel.

Our results disprove the prevailing ΔMKR bypass model (18, 19), which predicts that ribosomes harboring the L22 ΔMKR deletion should be active in the presence of bound macrolides. By contrast, we find that erythromycin inhibits translation by ΔMKR ribosomes in vitro, with the mutant ribosomes being as sensitive to inhibition as wild-type ribosomes. In the cell, we find that ΔMKR ribosomes are also sensitive to macrolides added at sufficiently high concentrations, in agreement with previous studies (12). In E. coli expressing ΔMKR-L22-titin, the MIC for erythromycin is 850 μM, but this value drops to ≈10 μM when PAβN, an inhibitor of efflux pumps is also present. Similarly, mutations that inactivate the AcrAB-TolC efflux system reduce the MIC for erythromycin to 5–10 μM for ΔMKR strains. These inhibitory effects are caused by macrolide binding to the site near the exit tunnel of ΔMKR ribosomes, because ErmC methylation of a base within this site alleviates inhibition both in vitro (data not shown) and in vivo. These results do not support models that rely on macrolide-bound ΔMKR ribosomes functioning better than macrolide-bound wild-type ribosomes.

Macrolide concentrations appear to be lower in ΔMKR strains than in wild-type strains. For example, when ΔMKR or wild-type cells are grown in 200 μM erythromycin, footprinting experiments show lower occupancy of the macrolide-binding site in ΔMKR ribosomes, but similar occupancy in the presence of an efflux inhibitor. This result is unlikely to arise from differences in affinity, because we find that both types of ribosomes show similar sensitivity to erythromycin inhibition in vitro and previous studies show that erythromycin binds ΔMKR ribosomes similarly to wild-type ribosomes (8, 12). Moreover, small differences in the affinity of erythromycin for wild-type and ΔMKR ribosomes in the cell should have no appreciable effect on occupancy, because near stoichiometric binding would be expected in both cases because the intracellular concentration of ribosomes exceeds the KD for erythromycin binding by 100-fold or more (35). Two mechanisms could potentially give rise to reduced drug concentration in the ΔMKR strain: (i) a factor sequesters or destroys the antibiotic; or (ii) cellular uptake of the drug is reduced or efflux is increased. Although either of these mechanisms are feasible, we favor the second possibility, which is consistent with our findings that blocking efflux with PAβN makes cells more sensitive to erythromycin and deleting either the AcrA/AcrB or TolC components of the AcrAB-TolC efflux pump reduces the erythromycin resistance of a strain with ΔMKR-L22-titin ribosomes compared with a strain with L22-titin ribosomes.

The ΔMKR mutation is known to affect translation of certain mRNAs by reducing programmed ribosome stalling (36, 37). How might ΔMKR-linked translational changes affect macrolide uptake or efflux? It is possible that AcrAB and/or TolC activities are increased through direct effects on translation of these proteins or indirect effects on translation of a regulator. However, preliminary experiments, using Western blot analysis with anti-AcrA and anti-TolC antibodies did not reveal up-regulation of these proteins in the ΔMKR strain (data not shown). Moreover, our ΔMKR strain did not show enhanced resistance to SDS or ethidium bromide (data not shown), which are also pumped from the cell by the AcrAB-TolC system (38, 39), although this result could be complicated by the regulatory mechanisms by which these compounds are sensed by the cell (40). Because erythromycin must enter and cross the inner membrane to gain access to ribosomes, a delay in this step could result in the efflux pumps being more effective at shedding the drug in the ΔMKR strain, which would not necessarily require increased pump levels.

Although the mechanism by which the ΔMKR mutation increases macrolide resistance remains to be determined, it seems plausible that L22-mediated ribosome stalling plays a role in determining the balance of cell envelope components, which, in turn, affects macrolide resistance by altering the efficiency of efflux pumps. Moreover, given that the antibiotic resistance of wild-type and ΔMKR strains often changed in parallel when efflux systems were perturbed, it seems likely that the ΔMKR mutation simply enhances a ribosome-mediated defense mechanism against antibiotics that also occurs in wild-type cells.

We note that L22 mutations can cause macrolide resistance in Gram-positive and Gram-negative bacteria (2). Although Gram-positive strains lack TolC-mediated efflux systems, they do contain numerous membrane efflux pumps capable of transporting a wide variety of compounds including macrolides (2). Moreover, synergistic relationships between L22-mediated antibiotic resistance and efflux systems have been reported in strains of Haemophilus and Campylobacter (41, 42). Thus, the connection between L22 function and drug efflux seems to be highly conserved. Our experiments show that ΔMKR ribosomes are not inherently resistant to macrolides and support a new model in which changes in translation alter bacterial physiology and reduce macrolide accumulation. Because wild-type and ΔMKR strains are efficiently inhibited by low concentrations of macrolides when drug efflux is blocked, combination therapies, using macrolides and efflux inhibitors may be an appealing antibacterial strategy that precludes many normal routes to drug resistance.

Experimental Procedures

Strains and Plasmids.

L22 fusions containing an N-terminal His6 tag and a C-terminal titin-I27-ssrA domain were encoded on plasmid pS10 and maintained in clpX−, clpA−, rna− cells with a deletion of the chromosomal S10 operon (20). We confirmed that the macrolide resistance mediated by the L22 ΔMKR mutation was not altered in clpX+, clpA+, rna+ strains or when the C-terminal I27-ssrA fusion was omitted (data not shown). The tolC− strains were constructed by P1 transducing tolC::kan from strain JW5503–1 of the Keio collection (43) into strain SM1090 (X90, clpX−, clpA−, rna−), removing the kan marker with FLP recombinase (44), transforming with either pS10-H6-L22WT-I27-ssrA or pS10-H6-L22ΔMKR-I27-ssrA, and then deleting the chromosomal S10 operon (20). The acrAB− strains were constructed by replacing the acrAB ORF with a PCR product containing the cat gene, using the recombination plasmid pSIM-5 (45), and then transducing the cat gene by selection on chloramphenicol into strains SM1145 (wild-type L22) or SM1211 (L22 ΔMKR) (20). All constructs were verified using diagnostic PCR and the wild-type, ΔMKR, Δloop1, and Δloop2 L22 fusions constructs were verified as the only copies of L22 genes in the cell by Southern blot analysis with a probe complementary to the antisense strand of the rplV gene (20). A plasmid containing only the ORF of the ermC gene from Bacillus subtilis under control of an arabinose-inducible promoter (pBAD24-ermC) was a gift from C. Hayes (University of California, Santa Barbara, CA) and was introduced into wild-type and ΔMKR strains and grown in the presence of 0.2% arabinose for induction.

Antibiotics, Culture Growth, and MIC Determination.

Antibiotics and PAβN were purchased from SIGMA-Aldrich. Stocks of erythromycin were prepared in ethanol, and the concentration was determined in 10 mM bis-Tris (pH 6.5), using an extinction coefficient at 298 nm of 25.7 M−1·cm−1. MICs for antibiotics were determined in Luria–Bertani broth supplemented with 1דMops mixture” (pH 7.2, Teknova), using 150 μL of cultures in 96-well plates containing 1,000 log-phase colony-forming units per well and grown with agitation for 16 h at 37 °C. Culture densities were determined by measuring the turbidity at 600 nm.

Biochemical Assays.

Purification of ribosomes and rRNA from log-phase cultures and reverse-transcriptase extension of a 5′-TCAATGTTCAGTGTCAAGCTATAGTAAAGGTTCACG-3′ primer complementary to residues 2066–2101 of E. coli 23S rRNA were performed as described in ref. 20. Coupled transcription/translation assays were performed as described in ref. 20 by adding purified ribosomes to complete a ribosome-free translation extract. Erythromycin was then added to desired concentrations, the mixtures were incubated at room temperature for 15 min, and reactions were started by adding a plasmid encoding the SspB and β-lactamase proteins under control of a T7 promoter. After 1 h of transcription/translation at 37 °C, reactions were stopped and 14C-labeled proteins were visualized using SDS-PAGE and autoradiography. Some experiments were performed using a transcription template encoding a variant of the E. coli rbsK gene. These translation assays produced 14C-labeled proteins in a linear fashion for >120 min, and thus the intensity of the product bands at 60 min is a reasonable estimate of the translation rate. Inhibition constants (Ki) were obtained by fitting the erythromycin (E) dependence of ribosome (R) inhibition to the equation activity = max·(1 − ((([Etotal] + [Rtotal] + Ki) − SQRT(([Etotal] + [Rtotal] + Ki)2) − 4·[Etotal]·[Rtotal]))/(2·[Rtotal]))).

DMS modification of ribosomes in vivo was performed similarly to a published protocol (46). Cultures of wild-type and ΔMKR cells were grown in LB broth with shaking at 37 °C until early log phase and erythromycin was then added to 1 aliquot of each culture. Growth was continued for 2 h and cells were chilled on ice, harvested, and resuspended at equal cell densities (≈1.2 × 109 cfu per milliliter) in fresh medium with or without erythromycin. Aliquots of each culture were then placed in clean tubes in a 37 °C water bath and PAβN was added to 30 μg/ml in 1 sample of each strain containing erythromycin. After 10 min of preincubation, DMS solution (freshly diluted 1/6 in ethanol) was added (≈50 mM final) to each culture, and the cultures were incubated with shaking for an additional 15 min. Reactions were stopped by diluting an aliquot of each culture 20-fold into ice-cold stop solution (100 mM Tris·Cl (pH 7.5), 0.5 M 2-mercaptoethanol, 50 mM NaCl, 2 mM MgOAc) and mixing rapidly. Cells were then harvested at 4 °C, washed twice with ice-cold buffer (50 mM Tris·Cl (pH 7.5), 50 mM NaCl, 2 mM MgOAc), and resuspended in lysis buffer (B-Per II (Pierce) supplemented with 2 mM MgOAc, 0.5 mM CaCl2, 10 units/ml DNase I (Roche), and 0.1 mg/ml lysozyme). After lysis, total RNA was purified using acidic phenol/chloroform extraction and alcohol precipitation (20).

Acknowledgments

We thank K. Griffith, A. Grossman, and R. Britton for helpful discussions and C. Hayes for materials and discussions. This work was supported by National Institutes of Health Grants AI-15706 and AI-16892.

Footnotes

  • 1To whom correspondence may be addressed at the present address:
    Department of Molecular Biology and Microbiology, Burnett School of Biomedical Sciences, University of Central Florida, Orlando, FL 32816.
    E-mail: semoore{at}mail.ucf.edu
  • 2To whom correspondence may be addressed. E-mail: bobsauer{at}mit.edu

  • Author contributions: S.D.M. designed research; S.D.M. performed research; S.D.M. contributed new reagents/analytic tools; S.D.M. analyzed data; and S.D.M. and R.T.S. wrote the paper.

  • The authors declare no conflict of interest.

References

  1. ↵
    1. Vester B,
    2. Douthwaite S
    (2001) Macrolide resistance conferred by base substitutions in 23S rRNA. Antimicrob Agents Chemother 45:1–12.
    OpenUrlFREE Full Text
  2. ↵
    1. Roberts MC
    (2008) Update on macrolide-lincosamide-streptogramin, ketolide, and oxazolidinone resistance genes. FEMS Microbiol Lett 282:147–159.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Schlünzen F,
    2. et al.
    (2001) Structural basis for the interaction of antibiotics with the peptidyl transferase centre in eubacteria. Nature 413:814–821.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Tu D,
    2. Blaha G,
    3. Moore PB,
    4. Steitz TA
    (2005) Structures of MLSBK antibiotics bound to mutated large ribosomal subunits provide a structural explanation for resistance. Cell 121:257–270.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Contreras A,
    2. Vázquez D
    (1977) Cooperative and antagonistic interactions of peptidyl-tRNA and antibiotics with bacterial ribosomes. Eur J Biochem 74:539–547.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Andersson S,
    2. Kurland CG
    (1987) Elongating ribosomes in vivo are refractory to erythromycin. Biochimie 69:901–904.
    OpenUrl
  7. ↵
    1. Tenson T,
    2. Lovmar M,
    3. Ehrenberg M
    (2003) The mechanism of action of macrolides, lincosamides and streptogramin B reveals the nascent peptide exit path in the ribosome. J Mol Biol 330:1005–1014.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Wittmann HG,
    2. et al.
    (1973) Biochemical and genetic studies on two different types of erythromycin resistant mutants of Escherichia coli with altered ribosomal proteins. Mol Gen Genet 127:175–189.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Douthwaite S,
    2. Powers T,
    3. Lee JY,
    4. Noller HF
    (1989) Defining the structural requirements for a helix in 23 S ribosomal RNA that confers erythromycin resistance. J Mol Biol 209:655–665.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Zaman S,
    2. Fitzpatrick M,
    3. Lindahl L,
    4. Zengel J
    (2007) Novel mutations in ribosomal proteins L4 and L22 that confer erythromycin resistance in Escherichia coli. Mol Microbiol 66:1039–1050.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Apirion D
    (1967) Three genes that affect Escherichia coli ribosomes. J Mol Biol 30:255–275.
    OpenUrlPubMed
  12. ↵
    1. Chittum HS,
    2. Champney WS
    (1994) Ribosomal protein gene sequence changes in erythromycin-resistant mutants of Escherichia coli. J Bacteriol 176:6192–6198.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Gregory ST,
    2. Dahlberg AE
    (1999) Erythromycin resistance mutations in ribosomal proteins L22 and L4 perturb the higher order structure of 23 S ribosomal RNA. J Mol Biol 289:827–834.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Clark C,
    2. et al.
    (2002) In vitro selection of resistance in Haemophilus influenzae by amoxicillin-clavulanate, cefpodoxime, cefprozil, azithromycin, and clarithromycin. Antimicrob Agents Chemother 46:2956–2962.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Pardo D,
    2. Rosset R
    (1977) Properties of ribosomes from erythromycin resistant mutants of Escherichia coli. Mol Gen Genet 156:267–271.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Schuwirth BS,
    2. et al.
    (2005) Structures of the bacterial ribosome at 3.5 A resolution. Science 310:827–834.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Voss NR,
    2. Gerstein M,
    3. Steitz TA,
    4. Moore PB
    (2006) The geometry of the ribosomal polypeptide exit tunnel. J Mol Biol 360:893–906.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Gabashvili IS,
    2. et al.
    (2001) The polypeptide tunnel system in the ribosome and its gating in erythromycin resistance mutants of L4 and L22. Mol Cell 8:181–188.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Davydova N,
    2. Streltsov V,
    3. Wilce M,
    4. Liljas A,
    5. Garber M
    (2002) L22 ribosomal protein and effect of its mutation on ribosome resistance to erythromycin. J Mol Biol 322:635–644.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Moore SD,
    2. Baker TA,
    3. Sauer RT
    (2008) Forced extraction of targeted components from complex macromolecular assemblies. Proc Natl Acad Sci USA 105:11685–11690.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Pestka S
    (1974) Binding of [14C]erythromycin to Escherichia coli ribosomes. Antimicrob Agents Chemother 6:474–478.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Douthwaite S,
    2. Aagaard C
    (1993) Erythromycin binding is reduced in ribosomes with conformational alterations in the 23 S rRNA peptidyl transferase loop. J Mol Biol 232:725–731.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Lovmar M,
    2. Tenson T,
    3. Ehrenberg M
    (2004) Kinetics of macrolide action: The josamycin and erythromycin cases. J Biol Chem 279:53506–53515.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Petropoulos AD,
    2. Kouvela EC,
    3. Dinos GP,
    4. Kalpaxis DL
    (2008) Stepwise binding of tylosin and erythromycin to Escherichia coli ribosomes, characterized by kinetic and footprinting analysis. J Biol Chem 283:4756–4765.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Lomovskaya O,
    2. et al.
    (2001) Identification and characterization of inhibitors of multidrug resistance efflux pumps in Pseudomonas aeruginosa: Novel agents for combination therapy. Antimicrob Agents Chemother 45:105–116.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    1. Chollet R,
    2. Chevalier J,
    3. Bryskier A,
    4. Pagès JM
    (2004) The AcrAB-TolC pump is involved in macrolide resistance but not in telithromycin efflux in Enterobacter aerogenes and Escherichia coli. Antimicrob Agents Chemother 48:3621–3624.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Skinner R,
    2. Cundliffe E,
    3. Schmidt FJ
    (1983) Site of action of a ribosomal RNA methylase responsible for resistance to erythromycin and other antibiotics. J Biol Chem 258:12702–12706.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Denoya CD,
    2. Dubnau D
    (1987) Site and substrate specificity of the ermC 23S rRNA methyltransferase. J Bacteriol 169:3857–3860.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Moazed D,
    2. Noller HF
    (1987) Chloramphenicol, erythromycin, carbomycin and vernamycin B protect overlapping sites in the peptidyl transferase region of 23S ribosomal RNA. Biochimie 69:879–884.
    OpenUrl
  30. ↵
    1. Zengel JM,
    2. Jerauld A,
    3. Walker A,
    4. Wahl MC,
    5. Lindahl L
    (2003) The extended loops of ribosomal proteins L4 and L22 are not required for ribosome assembly or L4-mediated autogenous control. RNA 9:1188–1197.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Nikaido H
    (1998) Multiple antibiotic resistance and efflux. Curr Opin Microbiol 1:516–523.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Kobayashi N,
    2. Nishino K,
    3. Yamaguchi A
    (2001) Novel macrolide-specific ABC-type efflux transporter in Escherichia coli. J Bacteriol 183:5639–5644.
    OpenUrlAbstract/FREE Full Text
    1. Jellen-Ritter AS,
    2. Kern WV
    (2001) Enhanced expression of the multidrug efflux pumps AcrAB and AcrEF associated with insertion element transposition in Escherichia coli mutants Selected with a fluoroquinolone. Antimicrob Agents Chemother 45:1467–1472.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Nishino K,
    2. Senda Y,
    3. Yamaguchi A
    (2008) The AraC-family regulator GadX enhances multidrug resistance in Escherichia coli by activating expression of mdtEF multidrug efflux genes. J Infect Chemother 14:23–29.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Dennis PP,
    2. Bremer H
    (1974) Macromolecular composition during steady-state growth of Escherichia coli B-r. J Bacteriol 119:270–281.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Nakatogawa H,
    2. Ito K
    (2002) The ribosomal exit tunnel functions as a discriminating gate. Cell 108:629–636.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Vazquez-Laslop N,
    2. Thum C,
    3. Mankin AS
    (2008) Molecular mechanism of drug-dependent ribosome stalling. Mol Cell 30:190–202.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Otsuji N,
    2. Soejima T,
    3. Maki S,
    4. Shinagawa H
    (1982) Cloning of colicin E1 tolerant tolC (mtcB) gene of Escherichia coli K12 and identification of its gene product. Mol Gen Genet 187:30–36.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Kawabe T,
    2. Fujihira E,
    3. Yamaguchi A
    (2000) Molecular construction of a multidrug exporter system, AcrAB: Molecular interaction between AcrA and AcrB, and cleavage of the N-terminal signal sequence of AcrA. J Biochem 128:195–200.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Rosenberg EY,
    2. Bertenthal D,
    3. Nilles ML,
    4. Bertrand KP,
    5. Nikaido H
    (2003) Bile salts and fatty acids induce the expression of Escherichia coli AcrAB multidrug efflux pump through their interaction with Rob regulatory protein. Mol Microbiol 48:1609–1619.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Peric M,
    2. et al.
    (2004) Inability of L22 ribosomal protein alteration to increase macrolide MICs in the absence of efflux mechanism in Haemophilus influenzae HMC-S. J Antimicrob Chemother 54:393–400.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Cagliero C,
    2. Mouline C,
    3. Cloeckaert A,
    4. Payot S
    (2006) Synergy between efflux pump CmeABC and modifications in ribosomal proteins L4 and L22 in conferring macrolide resistance in Campylobacter jejuni and Campylobacter coli. Antimicrob Agents Chemother 50:3893–3896.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Baba T,
    2. et al.
    (2006) Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: The Keio collection. Mol Syst Biol 2:1–11.
    OpenUrlPubMed
  44. ↵
    1. Datsenko KA,
    2. Wanner BL
    (2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci USA 97:6640–6645.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. Datta S,
    2. Costantino N,
    3. Court DL
    (2006) A set of recombineering plasmids for gram-negative bacteria. Gene 379:109–115.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Moazed D,
    2. Robertson JM,
    3. Noller HF
    (1988) Interaction of elongation factors EF-G and EF-Tu with a conserved loop in 23S RNA. Nature 334:362–364.
    OpenUrlCrossRefPubMed
View Abstract
Back to top
Article Alerts
Email Article
Citation Tools
Request Permissions
Revisiting the mechanism of macrolide-antibiotic resistance mediated by ribosomal protein L22
Sean D. Moore, Robert T. Sauer
Proceedings of the National Academy of Sciences Nov 2008, 105 (47) 18261-18266; DOI: 10.1073/pnas.0810357105
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo
Proceedings of the National Academy of Sciences: 116 (38)
Current Issue

Submit

Jump to section

You May Also be Interested in

Habituation and training of the ravens in the experimental compartments, Haidlhof Research Station, Austria. Image courtesy of Jessie E. C. Adriaense.
Shared emotional states in ravens
Researchers report negative emotional contagion, or the coordination of emotional states between multiple individuals, in common ravens, suggesting convergent emotional evolution in birds and mammals.
Image courtesy of Jessie E. C. Adriaense.
Fishing fleet. Image courtesy of Pixabay/3282700.
Increase in global fishing fleets
A study examines trends in global fishing fleets and finds that by 2015, 68% of the global fishing fleet became motorized, and that the overall number of fleet vessels increased to 3.7 million, despite a consistent decrease in the catch per unit of effort.
Image courtesy of Pixabay/3282700.
Magnified views of fingerprints found on fragments of 11th-century AD jars from the U S Southwest enabled archaeologists to determine the sex of the potter. Image courtesy of John Kantner.
Fingerprints reveal gender roles in ancient society
A method to determine gender from fingerprints suggests pottery making was not a primarily female activity in ancient Puebloan society, challenging previous assumptions about gendered divisions of labor in ancient societies.
Image courtesy of John Kantner.

Similar Articles