Streptomycin-resistance of Euglena gracilis chloroplasts: identification of a point mutation in the 16S rRNA gene in an invariant position.

We sequenced the chloroplast 16S rRNA gene of two Euglena gracilis mutants which contain streptomycin-resistant chloroplasts (Smr 139.12/4 and Smr 139.20/2). These mutants are known to contain a single intact rrn operon per circular chloroplast genome. Nucleotide sequence comparison between a 16S rRNA gene of wild type Euglena gracilis, strain Z, with streptomycin-sensitive chloroplasts, and the 16S rRNA gene of both Smr-strains reveals a single base change (C to T) at position 876. This position is equivalent to the invariant position 912 of the E. coli 16S rRNA gene. The analogous position is also conserved in all chloroplast small subunit RNA genes from lower and higher plants sequenced so far. Light dependent protein synthesis with purified chloroplasts from streptomycin-resistant cells is not inhibited by streptomycin. Based on the results reported here we postulate linkage between the observed point mutation on the 16S rRNA gene and streptomycin-resistance of chloroplast 70S ribosomes.     

A mutation in the 530 loop of Escherichia coli 16S ribosomal RNA causes resistance to streptomycin.

Oligonucleotide-directed mutagenesis was used to introduce an A to C transversion at position 523 in the 16S ribosomal RNA gene of Escherichia coli rrnB operon cloned in plasmid pKK3535. E. coli cells transformed with the mutated plasmid were resistant to streptomycin. The mutated ribosomes isolated from these cells were not stimulated by streptomycin to misread the message in a poly(U)-directed assay. They were also restrictive to the stimulation of misreading by other error-promoting related aminoglycoside antibiotics such as neomycin, kanamycin or gentamicin, which do not compete for the streptomycin binding site. The 530 loop where the mutation in the 16S rRNA is located has been mapped at the external surface of the 30S subunit, and is therefore distal from the streptomycin binding site at the subunit interface. Our results support the conclusion that the mutation at position 523 in the 16S rRNA does not interfere with the binding of streptomycin, but prevents the drug from inducing conformational changes in the 530 loop which account for its miscoding effect. Since this effect primarily results from a perturbation of the translational proofreading control, our results also provide evidence that the 530 loop of the 16S rRNA is involved in this accuracy control.     

Modulation of 16S rRNA function by ribosomal protein S12

Ribosomal protein S12 is a critical component of the decoding center of the 30S ribosomal subunit and is involved in both tRNA selection and the response to streptomycin. We have investigated the interplay between S12 and some of the surrounding 16S rRNA residues by examining the phenotypes of double-mutant ribosomes in strains of Escherichia coli carrying deletions in all chromosomal rrn operons and expressing total rRNA from a single plasmid-borne rrn operon. We show that the combination of S12 and otherwise benign mutations at positions C1409-G1491 in 16S rRNA severely compromises cell growth while the level and range of aminoglycoside resistances conferred by the G1491U/C substitutions is markedly increased by a mutant S12 protein. The G1491U/C mutations in addition confer resistance to the unrelated antibiotic, capreomycin. S12 also interacts with the 912 region of 16S rRNA. Genetic selection of suppressors of streptomycin dependence caused by mutations at proline 90 in S12 yielded a C912U substitution in 16S rRNA. The C912U mutation on its own confers resistance to streptomycin and restricts miscoding, properties that distinguish it from a majority of the previously described error-promoting ram mutants that also reverse streptomycin dependence.     

Streptomycin-resistant and streptomycin-dependent mutants of the extreme thermophile Thermus thermophilus

We have isolated spontaneous streptomycin-resistant, streptomycin-dependent and streptomycin-pseudo-dependent mutants of the thermophilic bacterium Thermus thermophilus IB-21. All mutant phenotypes were found to result from single amino acid substitutions located in the rpsL gene encoding ribosomal protein S12. Spontaneous suppressors of streptomycin dependence were also readily isolated. Thermus rpsL mutations were found to be very similar to rpsL mutations identified in mesophilic organisms. This similarity affords greater confidence in the utility of the crystal structures of Thermus ribosomes to interpret biochemical and genetic data obtained with Escherichia coli ribosomes. In the X-ray crystal structure of the T. thermophilus HB8 30 S subunit, the mutated residues are located in close proximity to one another and to helices 18, 27 and 44 of 16 S rRNA. X-ray crystallographic analysis of ribosomes from streptomycin-resistant, streptomycin-pseudo-dependent and streptomycin-dependent mutants described here is expected to reveal fundamental insights into the mechanism of tRNA selection, translocation, and conformational dynamics of the ribosome.     

A single base substitution in 16S ribosomal RNA suppresses streptomycin dependence and increases the frequency of translational errors

A C to U substitution at position 1469 of 16S ribosomal RNA (rRNA) from Escherichia coli suppresses streptomycin dependence and causes increased translational error frequencies. Strains containing the rpsL252 or StrM287 streptomycin-dependent alleles are able to grow in the absence of streptomycin when transformed with plasmids containing the U1469 mutation in 16S rRNA. Ribosomes containing wild-type proteins and U1469 mutant 16S rRNA misincorporate leucine in vitro at elevated levels, comparable to that of some typical S4 ram ribosomes. These results provide additional support for the participation of 16S rRNA in maintaining translational accuracy.     

Single-base mutations at position 2661 of Escherichia coli 23S rRNA increase efficiency of translational proofreading.

Two single-base substitutions were constructed in the 2660 loop of Escherichia coli 23S rRNA (G2661-->C or U) and were introduced into the rrnB operon cloned in plasmid pKK3535. Ribosomes were isolated from bacteria transformed with the mutated plasmids and assayed in vitro in a poly(U)-directed system for their response to the misreading effect of streptomycin, neomycin, and gentamicin, three aminoglycoside antibiotics known to impair the proofreading control of translational accuracy. Both mutations decreased the stimulation of misreading by these drugs, but neither interfered with their binding to the ribosome. The response of the mutant ribosomes to these drugs suggests that the 2660 loop, which belongs to the elongation factor Tu binding site, is involved in the proofreading step of the accuracy control. In vivo, both mutations reduced read-through of nonsense codons and frameshifting, which can also be related to the increased efficiency in proofreading control which they confer to ribosomes.     

Affinity purification of ribosomes with a lethal G2655C mutation in 23 S rRNA that affects the translocation

A method for preparation of Escherichia coli ribosomes carrying lethal mutations in 23 S rRNA was developed. The method is based on the site-directed incorporation of a streptavidin binding tag into functionally neutral sites of the 23 S rRNA and subsequent affinity chromatography. It was tested with ribosomes mutated at the 23 S rRNA position 2655 (the elongation factor (EF)-G binding site). Ribosomes carrying the lethal G2655C mutation were purified and studied in vitro. It was found in particular that this mutation confers strong inhibition of the translocation process but only moderately affects GTPase activity and binding of EF-G.     

Fluorescence studies on a streptomycin-induced conformational change in ribosomes which correlates with misreading

The fluorescent reagent N-(iodoacetylaminoethyl)-5-naphthylamine-1-sulfonic acid (I-AEDANS) was employed to detect and study the previously reported conformational change in the Escherichia coli ribosome induced by streptomycin. Labeling of ribosomes with this probe, which results in the derivatization of proteins S18 and L31', described earlier, inhibits neither their ribosomal protein synthesizing nor misreading ability. To calculate the amount of streptomycin bound to the ribosome, we determined the K'D for streptomycin, which is 0.24 micron, indicating that under our conditions, bound streptomycin/ribosome molar ratios are low, not in excess of 1. Under these conditions, streptomycin addition induces fluorescence quenching by 15% but does not affect streptomycin-resistant ribosomes. Maximal misreading occurs at these same ratios. Removal of AEDANS-L31' from the ribosomes drastically reduces streptomycin-induced quenching indicating the involvement of the environment of this protein in streptomycin action. The finding that streptomycin decreases AEDANS-L31' affinity for the ribosome supports this view. Streptomycin has been shown to bind to the 30 S subunit protein S4 while the 50 S protein L31' has been shown to be localized at the subunit interface. Thus, the observation that streptomycin influences this 50 S subunit protein L31', combined with the tight correlation between the effects of streptomycin on quenching and on misreading, strongly suggests that this antibiotic induces a conformational change at the subunit interface of the ribosome, and that this results in misreading. Polyuridylic acid also induces a conformational change in the ribosome but the polynucleotide and streptomycin seem to act independently. Streptomycin-resistant ribosomes, which undergo neither streptomycin-induced fluorescence nor streptomycin-induced misreading, are resistant to misreading induced by high Mg2+ as well.     

Mutants with base changes at the 3'-end of the 16S RNA from Escherichia coli. Construction, expression and functional analysis

The functionally important 3' domain of the ribosomal 16S RNA was altered by in vitro DNA manipulations of a plasmid-encoded 16S RNA gene. By in vitro DNA manipulations two double mutants were constructed in which C1399 was converted to A and G1401 was changed to either U or C and a single point mutant was made wherein G1416 was changed to U. Only one of the mutated rRNA genes could be cloned in a plasmid under the control of the natural rrnB promoters (U1416) whereas all three mutations were cloned in a plasmid under the control of the lambda PL promoter. In a strain coding for the temperature-sensitive lambda repressor cI857 the mutant RNAs could be expressed conditionally. We could show that all three mutant rRNAs were efficiently incorporated into 30S ribosomes. However, all three mutants inhibited the formation of stable 70S particles to various degrees. The amounts of mutated rRNAs were quantified by primer extension analysis which enabled us to assess the proportion of the mutated ribosomes which are actively engaged in in vivo protein biosynthesis. While ribosomes carrying the U1416 mutation in the 16S RNA were active in vivo a strong selection against ribosomes with the A1399/U1401 mutation in the 16S RNA from the polysome fraction is apparent. Ribosomes with 16S RNA bearing the A1399/C1401 mutation did not show a measurable protein biosynthesis activity in vivo. The growth rate of cells harbouring the different mutations reflected the in vivo translation capacities of the mutant ribosomes. The results underline the importance of the highly conserved nucleotides in the 3' domain of the 16S RNA for ribosomal function.     

Genetic and structural analysis of base substitutions in the central pseudoknot of Thermus thermophilus 16S ribosomal RNA

Characterization of base substitutions in rRNAs has provided important insights into the mechanism of protein synthesis. Knowledge of the structural effects of such alterations is limited, and could be greatly expanded with the development of a genetic system based on an organism amenable to both genetics and structural biology. Here, we describe the genetic analysis of base substitutions in 16S ribosomal RNA of the extreme thermophile Thermus thermophilus, and an analysis of the conformational effects of these substitutions by structure probing with base-specific modifying agents. Gene replacement methods were used to construct a derivative of strain HB8 carrying a single 16S rRNA gene, allowing the isolation of spontaneous streptomycin-resistant mutants and subsequent genetic mapping of mutations by recombination. The residues altered to give streptomycin resistance reside within the central pseudoknot structure of 16S rRNA comprised of helices 1 and 27, and participate in the U13–U20–A915 base triple, the G21–A914 type II sheared G–A base pair, or the G885–C912 Watson–Crick base pair closing helix 27. Substitutions at any of the three residues engaged in the base triple were found to confer resistance. Results from structure probing of the pseudoknot are consistent with perturbation of RNA conformation by these substitutions, potentially explaining their streptomycin-resistance phenotypes.     

Genetic evidence against the 16S ribosomal RNA helix 27 conformational switch model

A mechanistic understanding of ribosome function demands knowledge of the conformational changes that occur during protein synthesis. One current model proposes a conformational switch in Helix 27 (H27) of 16S rRNA involved in the decoding of mRNA. This model was based on the behavior of mutations in the 912 region of H27 of Escherichia coli 16S rRNA, which were predicted to stabilize the helix in either of two alternative conformations. This interpretation was supported by evidence from both genetics and structural biochemistry. However, recently published X-ray crystallographic structures of the Thermus thermophilus 30S subunit at different stages of tRNA selection have raised doubts regarding the validity of this model. We have therefore revisited the model genetically by constructing a H27 quadruple mutation (C912G, C910G, G885C, and G887C), which would create multiple mismatches in the proposed alternative conformation without perturbing the native H27 conformation seen in the crystal structures. Inconsistent with the H27 switch model, cells containing pure populations of quadruple mutant ribosomes grow at essentially wild-type rates. The mutants used to construct the H27 switch model all carried A2058G in 23S rRNA and C1192U in 16S rRNA as selectable markers. The quadruple mutant carrying these additional marker mutations is deleterious, and we conclude that they have a synergistic effect when combined with other mutations and are not phenotypically silent. Their presence confounded the interpretation of the original mutant phenotypes and, in light of the viability of the quadruple mutant, we conclude that the genetic evidence no longer supports the model.     

Cross-linking of streptomycin to the 30S subunit of Escherichia coli with phenyldiglyoxal

[3H]Dihydrostreptomycin was covalently linked to the 30S subunit of Escherichia coli K12A19 with the bifunctional cross-linking reagent phenyldiglyoxal. The cross-linking was abolished under conditions that prevent the binding of streptomycin, which indicates that the cross-linking occurs at the specific binding site of streptomycin. The cross-linking involved 16S RNA and the ribosomal proteins S1, S5, S11, and S13. This suggests that the streptomycin binding site is located in the upper part of the 30S subunit, facing the 50S subunit. Unexpectedly, the same extent and pattern of cross-linking were observed with the 30S subunits from a streptomycin-resistant mutant. We have shown previously that streptomycin induces conformational changes in the ribosomes from sensitive bacteria but not from streptomycin-resistant mutants. From this and from the results in the present study, it is suggested that the binding of streptomycin to streptomycin-sensitive ribosomes is a two-step reaction wherein an initial loose interaction at the antibiotic binding site is followed by a conformational rearrangement of the ribosomal particle. The second step would tighten the association with streptomycin and cause interference with protein synthesis. That step would be lacking in streptomycin-resistant mutants.     

A novel site for streptomycin resistance in the “530 loop” of chloroplast 16S ribosomal RNA

The chloroplast gene for 16S rRNA was cloned from two maternally inherited streptomycin-resistant mutants ofNicotiana differing in degree of resistance at the whole plant and isolated chloroplast level. A single-nucleotide change in the 16S rRNA gene was detected for each mutant: a C to T transition at nucleotide 860 (Escherichia coli coordinate C₉₁₂) which is an often mutated site, and a novel transition of C to T at nucleotide 472 (E. coli coordinate C₅₂₅). The novel mutation is located in the phylogenetically conserved 530 loop.     

Loss of a conserved 7-methylguanosine modification in 16S rRNA confers low-level streptomycin resistance in bacteria

Streptomycin has been an important drug for the treatment of tuberculosis since its discovery in 1944. But numerous strains of Mycobacterium tuberculosis, the bacterial pathogen that causes tuberculosis, are now streptomycin resistant. Although such resistance is often mediated by mutations within rrs, a 16S rRNA gene or rpsL, which encodes the ribosomal protein S12, these mutations are found in a limited proportion of clinically isolated streptomycin-resistant M. tuberculosis strains. Here we have succeeded in identifying a mutation that confers low-level streptomycin resistance to bacteria, including M. tuberculosis. We found that mutations within the gene gidB confer low-level streptomycin resistance and are an important cause of resistance found in 33% of resistant M. tuberculosis isolates. We further clarified that the gidB gene encodes a conserved 7-methylguanosine (m(7)G) methyltransferase specific for the 16S rRNA, apparently at position G527 located in the so-called 530 loop. Thus, we have identified gidB as a new streptomycin-resistance locus and uncovered a resistance mechanism that is mediated by loss of a conserved m(7)G modification in 16S rRNA. The clinical significance of M. tuberculosis gidB mutation also is noteworthy, as gidB mutations emerge spontaneously at a high frequency of 10(-6) and, once emerged, result in vigorous emergence of high-level streptomycin-resistant mutants at a frequency more than 2000 times greater than that seen in wild-type strains. Further studies on the precise function of GidB may provide a basis for developing strategies to suppress pathogenic bacteria, including M. tuberculosis.     

Mutant ribosomes can generate dominant kirromycin resistance.

Mutations in the two genes for EF-Tu in Salmonella typhimurium and Escherichia coli, tufA and tufB, can confer resistance to the antibiotic kirromycin. Kirromycin resistance is a recessive phenotype expressed when both tuf genes are mutant. We describe a new kirromycin-resistant phenotype dominant to the effect of wild-type EF-Tu. Strains carrying a single kirromycin-resistant tuf mutation and an error-restrictive, streptomycin-resistant rpsL mutation are resistant to high levels of kirromycin, even when the other tuf gene is wild type. This phenotype is dependent on error-restrictive mutations and is not expressed with nonrestrictive streptomycin-resistant mutations. Kirromycin resistance is also expressed at a low level in the absence of any mutant EF-Tu. These novel phenotypes exist as a result of differences in the interactions of mutant and wild-type EF-Tu with the mutant ribosomes. The restrictive ribosomes have a relatively poor interaction with wild-type EF-Tu and are thus more easily saturated with mutant kirromycin-resistant EF-Tu. In addition, the mutant ribosomes are inherently kirromycin resistant and support a significantly faster EF-Tu cycle time in the presence of the antibiotic than do wild-type ribosomes. A second phenotype associated with combinations of rpsL and error-prone tuf mutations is a reduction in the level of resistance to streptomycin.     

Positions 13 and 914 in Escherichia coli 16S ribosomal RNA are involved in the control of translational accuracy.

Using a conditional expression system with the temperature-inducible lambda PL promoter, we previously showed that the single mutations 13U-->A and 914A-->U, and the double mutation 13U-->A and 914A-->U in Escherichia coli 16S ribosomal RNA impair the binding of streptomycin (Pinard et al., The FASEB Journal, 1993, 7, 173-176). In this study, we found that the two single mutations and the double mutation increase translational fidelity, reducing in vivo readthrough of nonsense codons and frameshifting, and decreasing in vitro misincorporation in a poly(U)-directed system. Using oligodeoxyribonucleotide probes which hybridize to the 530 loop and to the 1400 region of 16S rRNA, two regions involved in the control of tRNA binding to the A site, we observed that the mutations in rRNA increase the binding of the probe to the 530 loop but not to the 1400 region. We suggest that the mutations at positions 13 and 914 of 16S rRNA induce a conformational rearrangement in the 530 loop, which contributes to the increased accuracy of the ribosome.     

The involvement of base 1054 in 16S rRNA for UGA stop codon dependent translational termination.

The deletion of the highly conserved cytidine nucleotide at position 1054 in E. coli 16S rRNA has been characterized to confer an UGA stop codon specific suppression activity which suggested a functional participation of small subunit rRNA in translational termination. Based on this structure-function correlation we constructed the three point mutations at site 1054, changing the wild-type C residue to an A, G or U base. The mutations were expressed from a complete plasmid encoded rRNA operon (rrnB) using a conditional expression system with the lambda PL-promoter. All three altered 16S rRNA molecules were expressed and incorporated into 70S ribosomal particles. Structural analysis of the protein and 16S rRNA moieties of the mutant ribosomes showed no differences when compared to wild-type particles. The phenotypic analysis revealed that only the 1054G base change led to a significantly reduced generation time of transformed cells, which could be correlated with the inability of the mutant ribosomes to specifically stop at UGA stop codons in vivo. The response towards UAA and UAG termination codons was not altered. Furthermore, in vitro RF-2 termination factor binding experiments indicated that the association behaviour of mutant ribosomes was not changed, enforcing the view that the UGA stop codon suppression is a direct consequence of the rRNA mutation. Taken together, these results argue for a direct participation of that 16S rRNA motif in UGA dependent translational termination and furthermore, suggest that termination factor binding and stop codon recognition are two separate steps of the termination event.