Characterization of a plasmid-specified ribosome methylase associated with macrolide resistance.

The ermC gene of plasmid pE194 specifies resistance to the macrolidelincosamide-streptogramin B antibiotics. This resistance, as well as synthesis of the 29,000 dalton protein product of ermC, has been shown to be induced by erythromycin. Weisblum and his colleagues have established that macrolide resistance is associated with a specific dimethylation of adenine in 23 S rRNA. We show that pE194 specifies an RNA methylase that can utilize either 50 S ribosomes or 23 S rRNA as substrates. Synthesis of this methylase is induced by low concentrations of erythromycin, and the enzyme is produced in elevated amounts by strains carrying a high copy number mutant of pE194. The methylase comigrates with the 29K ermC product on polyacrylamide gels. The purification and some properties of this methylase are described.     

Domain V of 23S rRNA contains all the structural elements necessary for recognition by the ErmE methyltransferase.

The ErmE methyltransferase from the erythromycin-producing actinomycete Saccharopolyspora erythraea dimethylates the N-6 position of adenine 2058 in domain V of 23S rRNA. This modification confers resistance to erythromycin and to other macrolide, lincosamide, and streptogramin B antibiotics. We investigated what structural elements in 23S rRNA are required for specific recognition by the ErmE methyltransferase. The ermE gene was cloned into R1 plasmid derivatives, providing a means of inducible expression in Escherichia coli. Expression of the methyltransferase in vivo confers resistance to erythromycin and clindamycin. The degree of resistance corresponds to the level of ermE expression. In turn, ermE expression also correlates with the proportion of 23S rRNA molecules that are dimethylated at adenine 2058. The methyltransferase was isolated in an active, concentrated form from E. coli, and the enzyme efficiently modifies 23S rRNA in vitro. Removal of most of the 23S rRNA structure, so that only domain V (nucleotides 2000 to 2624) remains, does not affect the efficiency of modification by the methyltransferase. In addition, modification still occurs after the rRNA tertiary structure has been disrupted by removal of magnesium ions. We conclude that the main features that are specifically recognized by the ErmE methyltransferase are displayed within the primary and secondary structures of 23S rRNA domain V.     

Ribosomal RNA methylation in Staphylococcus aureus and Escherichia coli: effect of the "MLS" (erythromycin resistance) methylase

Classical acquired resistance to erythromycin in Staphylococcus aureus ("MLS," or macrolide-lincosamide-streptogramin, resistance) was shown by Weisblum and colleagues to be a direct consequence of the conversion of one or more adenosine residues of 23S rRNA, within the subsequence(s) GA3G, to N6-dimethyladenosine (m62A). The methylation reaction is effected by a class of methylase, whose genes are typically plasmid- or transposon-associated, and whose synthesis is inducible by erythromycin. Using a recently obtained clinical MLS isolate of S. aureus, we have further defined the methylation locus as YGG X m62A X AAGAC; and have shown that this subsequence occurs once in the 23S RNA and that it is essentially completely methylated in all copies of 23S RNA that accumulate in induced cultures. Similar findings were obtained with laboratory S. aureus strains containing two well-characterized evolutionary variants (ermB, ermC) of MLS methylase genes. Analyses of a strain of E. coli containing the ermC gene indicated that the specificity of the methylase gene was unchanged, but that its expression was muted. Even after prolonged periods of induction, the strain manifested only partial resistance to erythromycin, and only about one-third of the copies of the MLS subsequence were methylated in such "induced" cultures. Since the E. coli 23S RNA sequence is known in its entirety, localization of the MLS subsequence is in this case unambiguous; as inferred by homology arguments applied earlier to the S. aureus data, the subsequence is in a highly conserved region of 23S RNA considered to contribute to the peptidyl transferase center of the ribosome.     

Mono- and dimethylating activities and kinetic studies of the ermC 23 S rRNA methyltransferase

The ermC 23 S rRNA methyltransferase converts a single adenine residue to N6,N6-dimethyladenine, both in vivo and in vitro. The ermC methyltransferase was demonstrated to produce both N6-mono and N6,N6-dimethylated adenine residues in Bacillus subtilis 23 S rRNA during the course of the reaction in vitro. An almost total conversion of monomethylated intermediates into dimethylated products was observed upon completion of the reaction. Data presented here demonstrate that the addition of the two methyl groups to each 23 S rRNA molecule takes place through a monomethylated intermediate and suggest that the enzyme dissociates from its RNA substrate between the two consecutive methylation reactions. The enzyme is able to utilize monomethylated RNA as substrate for the addition of a second methyl group with an efficiency approximately comparable to that obtained when unmethylated RNA was the initial substrate. Initial-rate data and inhibition studies suggest that the ermC methylase reaction involves a sequential mechanism occurring by two consecutive Random Bi Bi reactions.     

Regulation of a macrolide resistance-beta-galactosidase (ermC-lacZ) gene fusion in Escherichia coli.

A fusion constructed between the putative attenuator plus the first 219 nucleotides of the ermC (erythromycin resistance) structural gene and a 5' terminally deleted lacZ gene produced a moderate, basal level of beta-galactosidase which was increased by erythromycin addition. Another construction containing an intact ermC gene in addition to the fusion produced lower levels of beta-galactosidase, suggesting that the ermC gene product exerts negative feedback control on expression.     

The ermC leader peptide: amino acid alterations leading to differential efficiency of induction by macrolide-lincosamide-streptogramin B antibiotics.

The inducibility of ermC by erythromycin, megalomicin, and celesticetin was tested with both wild-type ermC and several regulatory mutants altered in the 19-amino-acid-residue leader peptide, MGIFSIFVISTVHYQP NKK. In the model test system that was used, the ErmC methylase was translationally fused to beta-galactosidase. Mutational alterations that mapped in the interval encoding Phe-4 through Ile-9 of the leader peptide not only affected induction by individual antibiotics, but did so differentially. The subset of mutations that affected inducibility by the two macrolides erythromycin and megalomicin overlapped and were distinct from the subset of mutations that affected induction by celesticetin. These studies provide a model system for experimentally varying the relative efficiencies with which different antibiotics induce the expression of ermC. The possibility that antibiotics with inducing activity interact directly with the nascent leader peptide was tested by using a chemically synthesized decapeptide, MGIFSIFVIS--, attached at its C-terminus to a solid-phase support. This peptide, however, failed to bind erythromycin in vitro.     

Mutations in conserved helix 69 of 23S rRNA of Thermus thermophilus that affect capreomycin resistance but not posttranscriptional modifications

Translocation during the elongation phase of protein synthesis involves the relative movement of the 30S and 50S ribosomal subunits. This movement is the target of tuberactinomycin antibiotics. Here, we describe the isolation and characterization of mutants of Thermus thermophilus selected for resistance to the tuberactinomycin antibiotic capreomycin. Two base substitutions, A1913U and mU1915G, and a single base deletion, DeltamU1915, were identified in helix 69 of 23S rRNA, a structural element that forms part of an interribosomal subunit bridge with the decoding center of 16S rRNA, the site of previously reported capreomycin resistance base substitutions. Capreomycin resistance in other bacteria has been shown to result from inactivation of the TlyA methyltransferase which 2'-O methylates C1920 of 23S rRNA. Inactivation of the tlyA gene in T. thermophilus does not affect its sensitivity to capreomycin. Finally, none of the mutations in helix 69 interferes with methylation at C1920 or with pseudouridylation at positions 1911 and 1917. We conclude that the resistance phenotype is a consequence of structural changes introduced by the mutations.     

A 50S ribosomal subunit precursor particle is a substrate for the ErmC methyltransferase in Staphylococcus aureus cells

Macrolide antibiotics like erythromycin can induce the synthesis of a specific 23S rRNA methyltransferase which confers resistance to cells containing the erm gene. Erythromycin inhibits both protein synthesis and the formation of 50S subunits in bacterial cells. We have tested the idea that the 50S precursor particle that accumulates in antibiotic-treated Staphylococcus aureus cells is a substrate for the methyltransferase enzyme. Pulse-chase labeling studies were conducted to examine the rates of ribosomal subunit formation in control and erythromycin-induced cells. Erythromycin binding to 50S subunits was examined under the same conditions. The rate of 50S subunit formation was reduced for up to 30 min after antibiotic addition, and erythromycin binding was substantial at this time. A nuclease protection assay was used to examine the methylation of adenine 2085 in 23S rRNA after induction. A methyl-labeled protected RNA sequence was found to appear in cells 30 min after induction. This protected sequence was found in both 50S subunits and in a subunit precursor particle sedimenting at about 30S in sucrose gradients. 23S rRNA isolated from 50S subunits of cells could be labeled by a ribosome-associated methlytransferase activity, with (3)H-S-adenosylmethionine as a substrate. 50S subunits were not a substrate for the enzyme, but the 30S gradient region from erythromycin-treated cells contained a substrate for this activity. These findings are consistent with a model that suggests that antibiotic inhibition of 50S formation leads to the accumulation of a precursor whose 23S rRNA becomes methylated by the induced enzyme. The methylated rRNA will preclude erythromycin binding; thus, assembly of the particle and translation become insensitive to the inhibitory effects of the drug.     

Methyltransferase that modifies guanine 966 of the 16 S rRNA: functional identification and tertiary structure

N(2)-Methylguanine 966 is located in the loop of Escherichia coli 16 S rRNA helix 31, forming a part of the P-site tRNA-binding pocket. We found yhhF to be a gene encoding for m(2)G966 specific 16 S rRNA methyltransferase. Disruption of the yhhF gene by kanamycin resistance marker leads to a loss of modification at G966. The modification could be rescued by expression of recombinant protein from the plasmid carrying the yhhF gene. Moreover, purified m(2)G966 methyltransferase, in the presence of S-adenosylomethionine (AdoMet), is able to methylate 30 S ribosomal subunits that were purified from yhhF knock-out strain in vitro. The methylation is specific for G966 base of the 16 S rRNA. The m(2)G966 methyltransferase was crystallized, and its structure has been determined and refined to 2.05A(.) The structure closely resembles RsmC rRNA methyltransferase, specific for m(2)G1207 of the 16 S rRNA. Structural comparisons and analysis of the enzyme active site suggest modes for binding AdoMet and rRNA to m(2)G966 methyltransferase. Based on the experimental data and current nomenclature the protein expressed from the yhhF gene was renamed to RsmD. A model for interaction of RsmD with ribosome has been proposed.     

A 50S Ribosomal Subunit Precursor Particle Is a Substrate for the ErmC Methyltransferase in <Emphasis Type="Italic">Staphylococcus aureus</Emphasis> Cells

Macrolide antibiotics like erythromycin can induce the synthesis of a specific 23S rRNA methyltransferase which confers resistance to cells containing the erm gene. Erythromycin inhibits both protein synthesis and the formation of 50S subunits in bacterial cells. We have tested the idea that the 50S precursor particle that accumulates in antibiotic-treated Staphylococcus aureus cells is a substrate for the methyltransferase enzyme. Pulse-chase labeling studies were conducted to examine the rates of ribosomal subunit formation in control and erythromycin-induced cells. Erythromycin binding to 50S subunits was examined under the same conditions. The rate of 50S subunit formation was reduced for up to 30 min after antibiotic addition, and erythromycin binding was substantial at this time. A nuclease protection assay was used to examine the methylation of adenine 2085 in 23S rRNA after induction. A methyl-labeled protected RNA sequence was found to appear in cells 30 min after induction. This protected sequence was found in both 50S subunits and in a subunit precursor particle sedimenting at about 30S in sucrose gradients. 23S rRNA isolated from 50S subunits of cells could be labeled by a ribosome-associated methlytransferase activity, with ³H-S-adenosylmethionine as a substrate. 50S subunits were not a substrate for the enzyme, but the 30S gradient region from erythromycin-treated cells contained a substrate for this activity. These findings are consistent with a model that suggests that antibiotic inhibition of 50S formation leads to the accumulation of a precursor whose 23S rRNA becomes methylated by the induced enzyme. The methylated rRNA will preclude erythromycin binding; thus, assembly of the particle and translation become insensitive to the inhibitory effects of the drug. Received: 21 June 2002 / Accepted: 21 August 2002     

Effect of ermC leader region mutations on induced mRNA stability.

Induction of translation of the ermC gene product in Bacillus subtilis occurs upon exposure to erythromycin and is a result of ribosome stalling in the ermC leader peptide coding sequence. Another result of ribosome stalling is stabilization of ermC mRNA. The effect of leader RNA secondary structure, methylase translation, and leader peptide translation on induced ermC mRNA stability was examined by constructing various mutations in the ermC leader region. Analysis of deletion mutations showed that ribosome stalling causes induction of ermC mRNA stability in the absence of methylase translation and ermC leader RNA secondary structure. Furthermore, deletions that removed much of the leader peptide coding sequence had no effect on induced ermC mRNA stability. A leader region mutation was constructed such that ribosome stalling occurred in a position upstream of the natural stall site, resulting in induced mRNA stability without induction of translation. This mutation was used to measure the effect of mRNA stabilization on ermC gene expression.     

Induction of ermC requires translation of the leader peptide.

ermC confers resistance to macrolide-lincosamide streptogramin B antibiotics by specifying a ribosomal RNA methylase, which results in decreased ribosomal affinity for these antibiotics. ermC expression is induced by exposure to erythromycin. We have previously proposed a translational regulation model in which erythromycin causes stalling of a ribosome, which is translating a leader peptide. Stalling causes a conformation shift in the ermC mRNA which in turn unmasks the methylase ribosomal binding site. A prediction of this translational attenuation model for ermC induction was tested by replacing the second codon of the putative ermC leader peptide coding region by TAA. As expected, the introduction of this mutation resulted in an uninducible phenotype which was suppressible by two ochre suppressor mutations in Bacillus subtilis. It is concluded that translation through the leader peptide coding region, in frame with the predicted leader peptide, is required for ermC induction.     

Site and substrate specificity of the ermC 23S rRNA methyltransferase.

The purified ermC methyltransferase described here incorporates two methyl groups per Bacillus subtilis 23S rRNA molecule in vitro. The Km for S-adenosyl-L-methionine was 12 microM, and for B. subtilis 23S rRNA the Km was 375 nM. In vivo methylation specified by several related resistance determinants prevented in vitro methylation by the ermC enzyme, suggesting that methylation specified by all of these determinants occurs at homologous sites. Since methyl groups were incorporated in protein-free 23S rRNA molecules, the structure of rRNA alone must contain sufficient information to specify the methylation site.