Stephen Douthwaite

The macrolide-ketolide antibiotic binding site is formed by structures in domains II and V of 23S ribosomal RNA.

The macrolide antibiotic erythromycin interacts with bacterial 23S ribosomal RNA (rRNA) making contacts that are limited to hairpin 35 in domain II of the rRNA and to the peptidyl transferase loop in domain V. These two regions are probably folded close together in the 23S rRNA tertiary structure and form a binding pocket for macrolides and other drug types. Erythromycin has been derivatized by replacing the L‐cladinose moiety at position 3 by a keto group (forming the ketolide antibiotics) and by an alkyl‐aryl extension at positions 11/12 of the lactone ring. All the drugs footprint identically within the peptidyl transferase loop, giving protection against chemical modification at A2058, A2059 and G2505, and enhancing the accessibility of A2062. However, the ketolide derivatives bind to ribosomes with widely varying affinities compared with erythromycin. This variation correlates with differences in the hairpin 35 footprints. Erythromycin enhances the modification at position A752. Removal of cladinose lowers drug binding 70‐fold, with concomitant loss of the A752 footprint. However, the 11/12 extension strengthens binding 10‐fold, and position A752 becomes protected. These findings indicate how drug derivatization can improve the inhibition of bacteria that have macrolide resistance conferred by changes in the peptidyl transferase loop.

Basis for prokaryotic specificity of action of aminoglycoside antibiotics

The aminoglycosides, a group of structurally related antibiotics, bind to rRNA in the small subunit of the prokaryotic ribosome. Most aminoglycosides are inactive or weakly active against eukaryotic ribosomes. A major difference in the binding site for these antibiotics between prokaryotic and eukaryotic ribosomes is the identity of the nucleotide at position 1408 (Escherichia coli numbering), which is an adenosine in prokaryotic ribosomes and a guanosine in eukaryotic ribosomes. Expression in E.coli of plasmid‐encoded 16S rRNA containing an A1408 to G substitution confers resistance to a subclass of the aminoglycoside antibiotics that contain a 6′ amino group on ring I. Chemical footprinting experiments indicate that resistance arises from the lower affinity of the drug for the eukaryotic rRNA sequence. The 1408G ribosomes are resistant to the same subclass of aminoglycosides as previously observed both for eukaryotic ribosomes and bacterial ribosomes containing a methylation at the N1 position of A1408. The results indicate that the identity of the nucleotide at position 1408 is a major determinant of specificity of aminoglycoside action, and agree with prior structural studies of aminoglycoside–rRNA complexes.

Macrolide–ketolide inhibition of MLS-resistant ribosomes is improved by alternative drug interaction with domain II of 23S rRNA

The macrolide antibiotic erythromycin and its 6‐O‐methyl derivative (clarithromycin) bind to bacterial ribosomes primarily through interactions with nucleotides in domains II and V of 23S rRNA. The domain II interaction occurs between nucleotide A752 and the macrolide 3‐cladinose moiety. Removal of the cladinose, and substitution of a 3‐keto group (forming the ketolide RU 56006), results in loss of the A752 interaction and an ≈u200a100‐fold drop in drug binding affinity. Within domain V, the key determinant of drug binding is nucleotide A2058 and substitution of G at this position is the major cause of drug resistance in some clinical pathogens. The 2058G mutation disrupts the drug‐domain V contact and leads to a further >u200a25u2003000‐fold decrease in the binding of RU 56006. Drug binding to resistant ribosomes can be improved over 3000‐fold by forming an alternative and more effective contact to A752 via alkyl–aryl groups linked to a carbamate at the drug 11/12 position (in the ketolide antibiotics HMR 3647 and HMR 3004). The data indicate that simultaneous drug interactions with domains II and V strengthen binding and that the domain II contact is of particular importance to achieve binding to the ribosomes of resistant pathogens in which the domain V interaction is perturbed.

Mapping posttranscriptional modifications in 5S ribosomal RNA by MALDI mass spectrometry

We present a method to screen RNA for posttranscriptional modifications based on Matrix Assisted Laser Desorption/Ionization mass spectrometry (MALDI-MS). After the RNA is digested to completion with a nucleotide-specific RNase, the fragments are analyzed by mass spectrometry. A comparison of the observed mass data with the data predicted from the gene sequence identifies fragments harboring modified nucleotides. Fragments larger than dinucleotides were valuable for the identification of posttranscriptional modifications. A more refined mapping of RNA modifications can be obtained by using two RNases in parallel combined with further fragmentation by Post Source Decay (PSD). This approach allows fast and sensitive screening of a purified RNA for posttranscriptional modification, and has been applied on 5S rRNA from two thermophilic microorganisms, the bacterium Bacillus stearothermophilus and the archaeon Sulfolobus acidocaldarius, as well as the halophile archaea Halobacterium halobium and Haloarcula marismortui. One S. acidocaldarius posttranscriptional modification was identified and was further characterized by PSD as a methylation of cytidine32. The modified C is located in a region that is clearly conserved with respect to both sequence and position in B. stearothermophilus and H. halobium and to some degree also in H. marismortui. However, no analogous modification was identified in the latter three organisms. We further find that the 5 end of H. halobium 5S rRNA is dephosphorylated, in contrast to the other 5S rRNA species investigated. The method additionally gives an immediate indication of whether the expected RNA sequence is in agreement with the observed fragment masses. Discrepancies with two of the published 5S rRNA sequences were identified and are reported here.

Characterization of the binding sites of protein L11 and the L10.(L12)4 pentameric complex in the GTPase domain of 23 S ribosomal RNA from Escherichia coli.

Ribonuclease and chemical probes were used to investigate the binding sites of ribosomal protein L11 and the pentameric complex L10.(L12)4 on Escherichia coli 23 S RNA. Protein complexes were formed with an RNA fragment constituting most of domains I and II or with 23 S RNA and they were investigated by an end-labelling method and a reverse transcriptase procedure, respectively. The results demonstrate that the two protein moieties bind at adjacent sites within a small RNA region. The L11 binding region overlaps with those of the modified peptide antibiotics thiostrepton and micrococcin and is constrained structurally by a three-helix junction while the L10.(L12)4 site is centred on an adjacent internal loop. The secondary structure of the whole region was determined in detail by the phylogenetic sequence comparison method, and the results for the L11 binding region, together with the experimental data, were used in a computer graphics approach to build a partial RNA tertiary structural model. The model provides insight into the topography of the L11 binding site. It also provides a structural rationale for the mutually co-operative binding of protein L11 with the antibiotics thiostrepton and micrococcin, and with the L10.(L12)4 protein complex.

Antibiotic interactions at the GTPase-associated centre within Escherichia coli 23S rRNA.

A comprehensive range of chemical reagents and ribonucleases was employed to investigate the interaction of the antibiotics thiostrepton and micrococcin with the ribosomal protein L11‐23S RNA complex and with the 50S subunit. Both antibiotics block processes associated with the ribosomal A‐site but differ in their effects on GTP hydrolysis, which is inhibited by thiostrepton and stimulated by micrococcin. The interaction sites of both drugs were shown to occur within the nucleotide sequences A1067‐A1098 within the protein L11 binding site on 23S RNA. This region of the ribosome structure is involved in elongation factor‐G‐dependent GTP hydrolysis and in the stringent response. No effects of drug binding were detected elsewhere in the 23S RNA. In general, the two drugs afforded 23S RNA similar protection from the chemical and nuclease probes in accord with their similar modes of action. One important exception, however, occurred at nucleotide A1067 within a terminal loop where thiostrepton protected the N‐1 position while micrococcin rendered it more reactive. This difference correlates with the opposite effects of the two antibiotics on GTPase activity.

The tylosin resistance gene tlrB of Streptomyces fradiae encodes a methyltransferase that targets G748 in 23S rRNA

tlrB is one of four resistance genes encoded in the operon for biosynthesis of the macrolide tylosin in antibiotic‐producing strains of Streptomyces fradiae. Introduction of tlrB into Streptomyces lividans similarly confers tylosin resistance. Biochemical analysis of the rRNA from the two Streptomyces species indicates that in vivo TlrB modifies nucleotide G748 within helix 35 of 23S rRNA. Purified recombinant TlrB retains its activity and specificity in vitro and modifies G748 in 23S rRNA as well as in a 74 nucleotide RNA containing helix 35 and surrounding structures. Modification is dependent on the presence of the methyl group donor, S‐adenosyl methionine. Analysis of the 74‐mer RNA substrate by biochemical and mass spectrometric methods shows that TlrB adds a single methyl group to the base of G748. Homologues of TlrB in other bacteria have been revealed through database searches, indicating that TlrB is the first member to be described in a new subclass of rRNA methyltransferases that are implicated in macrolide drug resistance.

Tandem DNA‐bound cAMP‐CRP complexes are required for transcriptional repression of the deoP2 promoter by the CytR repressor in Escherichia coli

We have studied the deoP2 promoter in Escherichia coli to define features important for its interaction with the CytR repressor. As is characteristic for CytR‐regulated promoters, deoP2 encodes tandem binding sites for the activating complex cAMP‐CRP. One of these sites, CRP‐1, overlaps the ‐ 35 region, and is sufficient for activation; the second site, CRP‐2, centred around‐93, is indispensable for repression. Here we demonstrate, by means of in vivo titration, that CytR interaction with deoP2 depends not only on CRP‐2, but also on CRP‐1 and the length and possibly the sequence separating these two sites. Also, point mutations in either CRP site reduce or abolish CytR titration; however, no co‐operativity is observed in the interaction of CytR with the two CRP binding sites. Furthermore, the reduction in CytR titration parallels the reduction in binding of cAMP‐CRP to the mutated CRP sites in vitro. These observations are not easily explained by current models for the action of prokaryotic repressors; instead we favour a model in which the interaction of CytR with deoP2 depends on the presence of tandem DNA‐bound cAMP–CRP complexes.

Design of camp–CRP-activated promoters in Escherichia coli

We have studied the doeP2 promoter of Escherichia coli to define features that are required for optimal activation by the complex of adenosine 3′, 5′ monophosphate (cAMP) and the cAMP receptor protein (CRP). Systematic mutagenesis of deoP2 shows that the distance between the CRP site and the ‐10 hexamer is the crucial factor in determining whether the promoter is activated by camp–CRP. Based on these observations, we propose that camp–CRP‐activated promoters can be created by correctly aligning a CRP target and a ‐ 10 hexamer. This idea has been successfully tested by converting both a CRP‐in‐dependent promoter and a sequence resembling the consensus ‐10 hexamer to strongly camp–CRP‐activated promoters.

Specific structural probing of plasmid-coded ribosomal RNAs from Escherichia coli

The preferred method for construction and in vivo expression of mutagenised Escherichia coli ribosomal RNAs (rRNAs) is via high copy number plasmids. Transcription of wild-type rRNA from the seven chromosomal rrn operons in strains harbouring plasmid-coded mutant rRNAs leads to a heterogeneous ribosome population, which consequently hinders direct probing of mutant rRNAs. Here, we describe how nonconserved helical regions of plasmid-coded rRNA have been altered in a manner that preserves their secondary structures while creating new sites for primer extension of mutant rRNAs. This facilitates specific biochemical probing of mutagenised rRNA regions despite the background of wild-type molecules. Four priming sites have been made to investigate the structural effects of mutations in the GTPase centre, helix 1200-1250, the peptidyl transferase region and the alpha-sarcin loop of 23S rRNA.