Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ochi, K. Articles by Kawamura, F. Search for Related Content PubMed PubMed Citation Articles by Ochi, K. Articles by Kawamura, F.

 Previous Article  |  Next Article 

Antimicrobial Agents and Chemotherapy, January 2009, p. 193-201, Vol. 53, No. 1
0066-4804/09/$08.00+0     doi:10.1128/AAC.00873-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Inactivation of KsgA, a 16S rRNA Methyltransferase, Causes Vigorous Emergence of Mutants with High-Level Kasugamycin Resistance

Kozo Ochi,^1* Ji-Yun Kim,^1,2 Yukinori Tanaka,^1,3 Guojun Wang,¹ Kenta Masuda,⁴ Hideaki Nanamiya,⁴ Susumu Okamoto,¹ Shinji Tokuyama,³ Yoshikazu Adachi,² and Fujio Kawamura⁴

National Food Research Institute, Tsukuba, Ibaraki 305-8642, Japan,¹ Laboratory of Animal Health, School of Agriculture, Ibaraki University, Ami, Ibaraki 300-0393, Japan,² Department of Applied Biological Chemistry, Faculty of Agriculture, Shizuoka University, Ohya, Shizuoka 422-8529, Japan,³ Laboratory of Molecular Genetics and Research Information Center for Extremophiles, College of Science, Rikkyo University, Tokyo 171-8501, Japan⁴

Received 2 July 2008/ Returned for modification 2 September 2008/ Accepted 31 October 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The methyltransferases RsmG and KsgA methylate the nucleotides G535 (RsmG) and A1518 and A1519 (KsgA) in 16S rRNA, and inactivation of the proteins by introducing mutations results in acquisition of low-level resistance to streptomycin and kasugamycin, respectively. In a Bacillus subtilis strain harboring a single rrn operon (rrnO), we found that spontaneous ksgA mutations conferring a modest level of resistance to kasugamycin occur at a high frequency of 10^–6. More importantly, we also found that once cells acquire the ksgA mutations, they produce high-level kasugamycin resistance at an extraordinarily high frequency (100-fold greater frequency than that observed in the ksgA⁺ strain), a phenomenon previously reported for rsmG mutants. This was not the case for other antibiotic resistance mutations (Tsp^r and Rif^r), indicating that the high frequency of emergence of a mutation for high-level kasugamycin resistance in the genetic background of ksgA is not due simply to increased persistence of the ksgA strain. Comparative genome sequencing showed that a mutation in the speD gene encoding S-adenosylmethionine decarboxylase is responsible for the observed high-level kasugamycin resistance. ksgA speD double mutants showed a markedly reduced level of intracellular spermidine, underlying the mechanism of high-level resistance. A growth competition assay indicated that, unlike rsmG mutation, the ksgA mutation is disadvantageous for overall growth fitness. This study clarified the similarities and differences between ksgA mutation and rsmG mutation, both of which share a common characteristic—failure to methylate the bases of 16S rRNA. Coexistence of the ksgA mutation and the rsmG mutation allowed cell viability. We propose that the ksgA mutation, together with the rsmG mutation, may provide a novel clue to uncover a still-unknown mechanism of mutation and ribosomal function.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The bacterial ribosome is a major target for antibiotics (6, 25). The aminoglycoside antibiotic kasugamycin, first reported in 1965 (31), has been important in agriculture because of its potent activity against rice blast. In the early 1970s, it was reported that bacterial resistance to kasugamycin involves silencing of the ksgA gene (also known as rsmA) through natural mutation, leading to inactivation of KsgA and resulting in the loss of dimethylation of two adjacent adenosine bases in 16S rRNA (12, 13). KsgA and the resulting modified adenosine bases (A1518 and A1519) appear to be conserved in all microorganisms examined to date. More recently, using Escherichia coli strains in which all of the rRNA is transcribed from a plasmid-encoded rrn operon, Vila-Sanjurjo and coworkers (32) reported that three 16S rRNA mutations (A794G, G926A, and A1519C) confer resistance to kasugamycin. The KsgA protein is homologous to another family of RNA methyltransferases, Erm, the members of which methylate a single adenosine base in 23S rRNA and confer resistance to the macrolide group of antibiotics. Recent work on the crystal structure of KsgA demonstrated a strong resemblance between KsgA and the crystal structure of ErmC' (21).

Previously, we demonstrated that loss of the m⁷G modification in 16S rRNA due to rsmG (renamed from gidB) mutations results in a low-level resistance to streptomycin as shown in E. coli, Bacillus subtilis, and Streptomyces coelicolor (19, 20, 23). The methyltransferase RsmG methylates the N7 position of nucleotide G527 (numbered according to E. coli). The nucleotide G527 is situated within a hairpin loop (the so-called 530 loop) that is one of the most highly conserved features of 16S rRNA, and mutations in this loop have been shown to be associated with resistance to streptomycin (17, 26, 30). This region of 16S rRNA is situated close to the ribosomal protein S12, and both of these ribosomal components play major roles in translational fidelity (22, 26). The phylogenetic conservation of RsmG and of the 16S rRNA sequence in the 530 loop suggests that methylation at this rRNA site should confer some selective advantage. However, a growth competition assay revealed that there are no differences in growth fitness between the rsmG mutant and the parent strain. Thus, the apparent lack of a disadvantage in cells that can no longer methylate the G527 position raised questions regarding the biological importance of this modification. The rsmG mutations arise spontaneously at a high frequency, ranging from 10^–4 to 10^–6 (19, 20, 23). Most importantly, in the rmsG mutant background, rpsL (encoding ribosomal protein S12) mutants with high-level streptomycin resistance emerged at a frequency 200-fold greater than that in the wild-type strain. This elevated frequency in the emergence of high-level streptomycin resistance was facilitated by a mutation pattern in rpsL more varied than that obtained by selection of the wild-type strain. As rsmG mutation (conferring a low level of streptomycin resistance) and ksgA mutation (conferring a modest level of kasugamycin resistance) share a common characteristic—failure to methylate the 16S rRNA bases—we examined whether ksgA mutants also display such a peculiar phenotype as that observed in rsmG mutants. To characterize ksgA mutation, we chose B. subtilis strain 168, as genomic information and numerous tools for genetic, biochemical, and physiological analyses are available for this well-characterized system (7, 24). Here, we report that acquisition of high-level resistance at an extraordinarily high frequency is common for rsmG mutants and ksgA mutants, while ksgA but not rsmG mutants display a disadvantage in overall fitness compared to the parent strain.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains and culture conditions. Strains of B. subtilis and E. coli were grown in LB medium at 37°C, except for ksgA mutants that were grown in LB medium without NaCl because the absence of NaCl rendered cells more sensitive to kasugamycin. LB medium consisted of 1% tryptone (Difco, Detroit, MI), 0.5% yeast extract (Difco), and 0.5% NaCl. S. coelicolor was grown in glucose-yeast extract-malt extract medium (29) at 30°C. Mycobacterium smegmatis was grown on rich (R) agar at 37°C. R agar contained 1% peptone (Difco), 0.5% yeast extract (Difco), 0.2% beef extract (Difco), 0.2% glycerol, 0.1% MgSO₄·7H₂O, 2% agar, and 50 mg/liter of Tween 80. The strains used in this study are listed in Table 1. Spontaneous B. subtilis mutants with low-level kasugamycin resistance were generated from the RIK543 strain (MIC, 500 µg/ml) and wild-type strain 168 (MIC, 1,500 µg/ml) on agar plates of LB medium minus NaCl containing 1,000 µg/ml and 3,000 µg/ml kasugamycin, respectively. Drug-resistant colonies developed after 2 to 3 days of incubation at 37°C. On the other hand, spontaneous E. coli mutants with low-level kasugamycin resistance were generated from the BW25113 strain (MIC, 200 µg/ml) on agar plates of LB medium minus NaCl containing 500 µg/ml kasugamycin. Spontaneous mutants with low-level resistance to streptomycin, thiostrepton, or rifampin (i.e., KO-826, KO-819, and KO-809) were generated from the RIK543 strain on LB medium containing 5 µg/ml streptomycin, 0.02 µg/ml thiostrepton, or 0.01 µg/ml rifampin, respectively. Spontaneous mutants with low-level resistance to capreomycin were generated from the Mycobacterium smegmatis wild-type strain JCM5866 (MIC, 5 µg/ml) on R medium containing 10 µg/ml of capreomycin. Serial dilutions of the cell suspension were also plated on media without any antibiotic to determine the number of viable cells in the original suspension. To measure the frequency of resistant mutants, single colonies were isolated, and cells originating from each of about 10 clones were examined separately. MICs to fully inhibit the growth were determined by spotting cell suspensions (10⁶) onto drug-containing plates, followed by incubation for 24 h to 48 h at 37°C.

View this table: [in this window] [in a new window]

TABLE 1. Strains used in this study

Construction of the B. subtilis strain harboring a single rrn operon. B. subtilis strain RIK543 contains only a single rrn operon (rrnO) in the genome, deleting the other nine rrn operons. The detailed procedures for the construction of RIK543 will be reported elsewhere. In brief, two methods were utilized for the construction of the deletion mutations. One method is to delete the target rrn operon by gene conversion, as follows. First, the target operon was disrupted by a chloramphenicol resistance gene (cat), and then the cat gene was deleted by the recombinant plasmid carrying the upstream and downstream sequence of the target rrn operon. A shuttle vector, pCHE11 (18), derived from pBR322 and pUB110ts-2 carrying a temperature-sensitive replication mutation in the repU gene of pUB110, was used as a vector for deleting the cat gene in the target rrn operon via gene conversion. After the selection for a kanamycin-resistant and chloramphenicol-sensitive phenotype at 30°C on LB plates, the resulting transformants were grown at 37°C on LB plates without any added antibiotics to eliminate the recombinant plasmid. Then, the transformant exhibiting a kanamycin- and chloramphenicol-sensitive phenotype at 30°C on LB plates was selected. The deletion mutation of rrnHG (rrnHG1) was obtained in this way.

Another method used for the construction of the deletion mutation was based on the "ampicillin screening" technique. To utilize this technique, kanamycin and erythromycin resistance genes were first integrated in tandem into the trpB trpA region in the genome, and then point mutation was introduced into either one of the resistance genes, resulting in the kanamycin-resistant, erythromycin-sensitive phenotype (Km^r Em^s) or kanamycin-sensitive, erythromycin-resistant phenotype (Km^s Em^r). Using these markers, the deletion mutation of the target rrn operon was constructed as follows. First, the cat gene was inserted downstream of the promoter region of the target rrn operon in the strain carrying Km^s Em^r markers. Next, a PCR fragment with the deletion mutation of the target rrn operon and a DNA fragment containing Km^r Em^s markers were simultaneously added to the competent-cell culture of the strain constructed as described above. After incubation at 37°C for 90 min, the culture was diluted with LB medium containing kanamycin and incubated at 37°C overnight. The overnight culture was again diluted with LB medium containing kanamycin and incubated at 37°C with shaking. Two types of transformants existed in this culture; one carries both kanamycin resistance and chloramphenicol resistance genes, and the other carries the kanamycin resistance gene but lacks the chloramphenicol resistance gene as a result of cotransformation with the kanamycin resistance gene and the deletion of the target rrn operon. To obtain the transformants carrying the deletion of the target rrn operon more efficiently, chloramphenicol was added to the culture at a final concentration of 50 µg/ml when cells were growing exponentially in the medium. Then, the culture was further incubated for 30 min to cause the growth arrest of the transformants carrying the deletion of the target rrn operon. In contrast, transformants carrying the cat gene grew normally in the presence of chloramphenicol. Next, ampicillin was added to the culture at a final concentration of 1 mg/ml. Since this antibiotic exhibits a bactericidal activity toward only the growing cells, most of the transformants carrying the cat gene were expected to be killed by this treatment. After further incubation for 2 or 3 h, cells were plated on LB plates containing kanamycin. The transformants which exhibited a kanamycin-resistant, chloramphenicol-sensitive phenotype were selected, and then proper introduction of the deletion mutation of the target rrn operon was confirmed by PCR amplification and DNA sequencing. The deletion mutations rrnD (rrnD1), rrnE (rrnE1), rrnB (rrnB2), rrnA (rrnA1), rrnI (rrnI2), and rrnW (rrnW2) were obtained in this way.

To introduce the deletion mutations into one strain successively, cotransformation experiments were carried out using both trpC2 and hisC101 genetic markers. In brief, the strain carrying the cat gene inserted downstream of the promoter region of the target rrn operon was used as the recipient, and the strain carrying the deletion of the target rrn operon was used as the donor strain, respectively. If the recipient strain contained the trpC2 mutation, the hisC101 mutation was introduced into the donor strain. The recipient strain was transformed with the DNA extracted from the donor strain, and Trp⁺ transformants were first selected. Among the Trp⁺ transformants, the chloramphenicol-sensitive transformant, which is generated as a result of the cotransformation event, was selected. In most cases, the resulting transformants also showed a His^– phenotype because of the close genetic linkage between trpC and hisC genes. Thus, the obtained transformant containing the hisC101 mutation was subjected to the next transformation experiment as described above, using the donor DNA carrying the trpC2 mutation. As a result of these successive transformation experiments, RIK543 (rrnHG1 rrnD1 rrnE1 rrnB2 rrnA1 rrnI2 rrnW2 rrnJ1::cat trpC2), which contained only a single rrnO operon, was obtained. The existence of only one rrnO operon in the genome of RIK543 was confirmed by Southern blot analysis (data not shown).

Mutation analysis. The primers used to amplify the candidate DNA fragments are listed in Table 2. PCR amplification was carried out with ExTaq (Takara, Otsu, Japan). Purified PCR products were directly sequenced with BigDye Terminator cycle sequencing kits (Perkin-Elmer Applied Biosystems, Foster City, CA). The sequence data were aligned using the GENETIX program (Software Development Co., Tokyo, Japan). Comparative genome sequencing, a new method of identifying unknown mutations, was performed as described previously (19).

View this table: [in this window] [in a new window]

TABLE 2. The primers used in this study

Analysis of intracellular polyamine profiles. Polyamines were identified as described previously (8, 27). Aliquots of bacterial cultures were pelleted, washed, and extracted with 0.2 M perchloric acid. Polyamines were subsequently dansylated and extracted with toluene for analyses by thin-layer chromatography (TLC). TLC was performed on silica gel G plates (Merck) and developed in ethyl acetate/cyclohexane (2:3, vol/vol). Spots were visualized under UV light.

Competition assay between mutant and parent strains. A competition assay was performed to compare the overall fitness of each strain. Briefly, equal numbers of cells from drug-resistant mutant and parent strains were inoculated into LB medium or into sterilized soil. After incubation (with shaking in LB medium but with standing in sterilized soil) for appropriate times as indicated above, the mixed cultures were inoculated (inoculation size, 1%) into new medium or new sterilized soil. This procedure was repeated 5 to 10 times. Finally, the ratios of mutant and parent strains were determined by spreading the cultured broth on LB medium with or without the corresponding antibiotic.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Development of ksgA mutations in B. subtilis. We initially assumed that high-level kasugamycin resistance may involve a mutation within the rrn gene encoding the 16S rRNA, as Vila-Sanjurjo and coworkers successfully identified kasugamycin resistance mutations (A794G, G926A, and A1519C) in the 16S rRNA gene of engineered E. coli (32). However, it is generally difficult to identify rRNA mutations because wild-type E. coli and B. subtilis strains have 7 and 10 chromosomal copies of the rRNA genes, respectively. One way to circumvent the difficulties due to the presence of multiple rrn operons is to use an engineered strain harboring only a single set of rrn. Thus, Kawamura and coworkers constructed the B. subtilis strain RIK543, in which all nine rrn genes other than rrnO have been deleted. The construction of the strain is described in Materials and Methods. As expected, RIK543 grew more slowly than the wild-type strain 168 and showed increased sensitivity to the antibiotic tested (Table 3). The details of the procedure for strain construction will be reported elsewhere. Using the B. subtilis wild-type (168) and engineered (RIK543) strains, we first attempted to isolate ksgA mutants. Mutants with low-level kasugamycin resistance developed at a high frequency (10^–6) in each strain on plates containing a twofold MIC amount of kasugamycin. These kasugamycin-resistant mutants (19 mutants derived from RIK543 were tested) were all found to have a mutation within the ksgA gene and were characterized by the frequent appearance of deletion or insertion mutations that resulted in stop codons just downstream of the mutations (Table 4). Likewise, the wild-type strain 168 also produced a wide variety of ksgA mutants at a high frequency (data not shown), indicating independence of this phenomenon of the number of rRNA gene copies. Unexpectedly, none of the RIK543-derived kasugamycin-resistant mutants had a mutation in the rRNA gene (i.e., rrnO) as determined by DNA sequencing. These results suggest that the frequency of emergence of rRNA mutation conferring kasugamycin resistance (if any in B. subtilis) is much lower than that of the ksgA mutation. In fact, mutants with a 16S rRNA mutation (A802G or G935A, corresponding to A794G and G926A in E. coli, respectively) were readily detected, though at a low frequency of 10^–8 to 10^–9, when selection was done for high-level (3,000 µg/ml or more) kasugamycin resistance, under the conditions which the ksgA mutants could no longer develop (data not shown). We noted that the ksgA mutation conferred no resistance to any of the other antibiotics tested, including streptomycin, kanamycin, spectinomycin, gentamicin, thiostrepton, lincomycin, erythromycin, and fusidic acid. Strains KO-827 (derived from RIK543) and KO-847 (derived from 168) were used for further study of the ksgA mutation. We also isolated mutants with low-level streptomycin resistance (KO-826), thiostrepton resistance (KO-819), and rifampin resistance (KO-809) from RIK543 for use as the control mutants in the following experiments (Table 1).

View this table: [in this window] [in a new window]

TABLE 3. Antibiotic susceptibility of B. subtilis strains 168 and RIK543

View this table: [in this window] [in a new window]

TABLE 4. Location of mutation in the ksgA gene and resulting amino acid exchange in KsgAa

High-frequency emergence of high-level kasugamycin resistance in ksgA mutants. Spontaneous mutations that lead to high-level antibiotic resistance (often a 100-fold increase in the MIC) generally emerge at a low frequency in bacteria (10^–8). Consistently, in B. subtilis RIK543 (MIC, 500 µg/ml), spontaneous mutants conferring high-level kasugamycin resistance (MIC, at least 5,000 µg/ml) arose at a low frequency, between 8 x 10^–9 and 5 x 10^–8. Strikingly, but consistent with previous observations for streptomycin resistance (20, 23; this study), the B. subtilis ksgA mutant produced spontaneous mutants showing resistance to a high level of kasugamycin (5,000 µg/ml) at a frequency on the order of 10^–6 to 10^–7. The data for ksgA mutant KO-827 are shown in Table 5 and indicate that there was a 100-fold greater frequency of mutations to high-level kasugamycin resistance than in the wild-type strain. Likewise, high-frequency emergence was detected when the ksgA mutant KO-847, derived from the wild-type strain 168, was used (a 110-fold greater frequency) or when the E. coil ksgA mutant KO-845 was compared with its parent strain, BW25113 (a 200- to 500-fold-greater frequency) (Table 5). Thus, this peculiar event occurs irrespective of the number of rRNA gene copies. Despite such marked increases in the frequency of high-level kasugamycin resistance in the genetic background of ksgA, only slight increases in the frequency of high-level resistance to the corresponding antibiotics were detected in the genetic background of Tsp^r or Rif^r, mutations of which confer a low level of resistance to thiostrepton and rifampin, respectively (Table 5). In agreement with previous work (20), the rsmG mutant KO-826 with low-level resistance to streptomycin produced spontaneous mutants showing resistance to a high level of streptomycin (1,000 µg/ml) at a 100-fold greater frequency. It is notable that the ksgA mutation did not lead to a greater frequency of the appearance of high-level streptomycin-resistant mutants and vice versa for the rsmG mutation (Table 5). In addition, the ksgA mutation did not affect the frequency at which mutants resistant to antibiotics other than kasugamycin (spectinomycin, lincomycin, erythromycin, chloramphenicol, and rifampin were tested) emerged, indicating that the observed effect of the ksgA mutation is limited to kasugamycin.

View this table: [in this window] [in a new window]

TABLE 5. Effect of ksgA mutation on the emergence of mutants with high-level kasugamycin resistance

Importantly, although the collective importance of rRNA modifications for protein synthesis has been shown (11), rsmG mutants developed even in the genetic background of ksgA when selected for low-level streptomycin resistance (e.g., KO-924 to KO-926) (Table 1), suggesting that ksgA and rsmG mutations can coexist without a loss of cell viability. Allowance of coexistence was confirmed by transformation (using KO-847 as the recipient and KO-750 as the donor DNA), in which ksgA rsmG transformants readily developed (e.g., KO-927) (Table 1). The ksgA rsmG double mutants not only grew as well as the wild-type strain 168 but also sporulated well (see Table 8).

View this table: [in this window] [in a new window]

TABLE 8. Ability of B. subtilis mutant strains to sporulate in certain media

speD mutation is responsible for high-level kasugamycin resistance. We next determined mutations leading to high-level kasugamycin resistance from low-level resistance. As expected, the mutants with high-level resistance derived from KO-809 (Rif^r) and KO-826 (Sm^r) were all found to have a mutation within the rpoB and rpsL genes, respectively, although the majority of mutants with high-level thiostrepton resistance showed no mutations within the rplK gene (data not shown). Unexpectedly, no mutations were found in the 16S rRNA gene when the mutants with high-level kasugamycin resistance (more than 20 strains were tested) derived from KO-827 (ksgA) were subjected to DNA sequencing. Therefore, we utilized comparative genome sequencing (19), a new method that uses microarray-based DNA sequencing to identify single-nucleotide polymorphisms and insertion-deletion sites within the genome. The mutant KO-875, derived from KO-847 (ksgA), with high-level kasugamycin resistance, was utilized as the source of mutant genomic DNA, and the stain KO-847 was utilized as the source of reference genomic DNA. As a result, we successfully identified a putative single-nucleotide polymorphism within the speD gene (formerly called ytcF but renamed speD by Sekowska et al. [28], which was confirmed by direct sequencing to be a point mutation [19GT representing Gly 7Trp]). The speD gene encodes S-adenosylmethionine decarboxylase (28). Strikingly, four other isolates with high-level kasugamycin resistance (KO-876 to KO-879) (Table 6) were all found to carry a mutation within the speD gene, including a frameshift mutation, thus strongly suggesting that the speD mutations are responsible for the acquisition of high-level resistance to kasugamycin (Table 6). The causal relationship was confirmed by speD transformation; the speD transformants (e.g., KO-894) showed kasugamycin resistance as did the ksgA mutant KO-847 (Table 6). This observation also indicated that the speD mutation alone is able to confer resistance to kasugamycin. Importantly, in contrast to the ksgA mutants, the speD mutant KO-894 did not produce mutants with high-level kasugamycin resistance at a high frequency (Table 5). Unlike the case for the B. subtilis ksgA mutant, none of the mutants with high-level kasugamycin resistance derived from E. coli ksgA strain KO-845 carried a mutation in speD and speE (coding for spermidine synthase), although the ksgA strain KO-845 also produced mutants with high-level kasugamycin resistance at a high frequency, which was more than 200-fold greater than that in the parental strain BW25113 (Table 5). The mutation(s) that arose in these E. coli mutants was not studied further.

View this table: [in this window] [in a new window]

TABLE 6. Location of mutation in the speD gene and resistance level to kasugamycin in B. subtilis mutants

speD mutation causes a marked reduction of intracellular spermidine. Sekowska et al. (28) recently reported that the B. subtilis SpeD protein (encoded by the speD gene), together with SpeE, participates in spermidine biosynthesis in this organism; speD-disrupted mutants exhibit a total lack of spermidine. Therefore, we performed TLC analyses of polyamine contents in wild-type and speD mutant cells. As shown in Fig. 1, spermidine concentrations in both the speD and ksgA speD mutant cells were significantly lower than those in wild-type cells, although they were not entirely diminished, perhaps due to the uptake of spermidine present in the medium. As expected, the kasugamycin resistance acquired in the speD transformant KO-894 was negated entirely or to some extent when 5 mM or 1 mM spermidine, respectively, was included in the medium (data not shown). In contrast, the addition of putrescine (5 to 10 mM) did not negate the speD mutation-induced kasugamycin resistance in KO-894. Thus, we concluded that the reduced spermidine concentrations caused by speD mutations are responsible for the phenotype with high-level kasugamycin resistance observed in the ksgA speD double mutants.

View larger version (83K): [in this window] [in a new window]

FIG. 1. Identification of polyamines by TLC. Dansyl polyamine derivatives of cell extracts were analyzed by TLC in ethyl acetate-cyclohexane (2:3, vol/vol). Lane 1, wild-type (168); lane 2, ksgA (KO-847); lane 3, speD (KO-894); lane 4, ksgA speD (KO-875); lane 5, ksgA speD (KO-876); lane 6, spermidine standard; lane 7, spermine standard; and lane 8, putrescine standard.

Inactivation of TlyA does not cause a high-frequency emergence of mutants with high-level capreomycin resistance. Like RsmG and KsgA, which methylate the base moiety of 16S rRNA, TlyA, encoded by the tlyA gene, methylates the ribose moiety of nucleotide C1409 of 16S rRNA and C1920 of 23S rRNA in Mycobacterium tuberculosis (14, 16). Inactivation of the tlyA gene confers low-level resistance to the ribosome-targeting drug capreomycin, which has been used as a second-line antibiotic for tuberculosis chemotherapy. Although many bacterial species lack the tlyA gene, including E. coli, B. subtilis, and S. coelicolor, we studied using M. smegmatis whether or not inactivation of tlyA (i.e., failure to methylate the ribose moiety of rRNA) results in a high-frequency emergence of mutants with high-level capreomycin resistance. The tlyA mutant KO-935, derived from the wild-type strain JCM5866 (MIC, 7 µg/ml) by spontaneous mutation (Table 1), revealed a low-level resistance (MIC, 20 µg/m) to capreomycin. The M. smegmatis tlyA mutant, however, did not give rise to mutants with high-level capreomycin resistance at a high frequency; the emergence of mutants with high-level resistance (MIC, 100 µg/ml) was as low as 2 x 10^–10 or less in either wild-type or tlyA mutant strains, thus discriminating the tlyA mutation from the rsmG and ksgA mutations in resistance emergence (data not shown).

ksgA, speD, and tlyA mutations result in a fitness disadvantage. Although mutations that confer drug resistance often have a biological cost, causing mutant bacteria to grow more slowly (2), rsmG mutants grow as well as the parental strain, as demonstrated by E. coli, B. subtilis, and S. coelicolor (19, 20, 23). Moreover, the growth competition assay demonstrated that the B. subtilis rsmG mutants are as fit as the wild-type strain. These findings were in contrast with previous studies of several other 16S rRNA methylases (1, 4, 15), which showed that knockout mutants were less fit than the wild-type strain. Although the ksgA mutants (and also the speD mutants) of B. subtilis and E. coli grew as well as the parental strain in LB medium, we examined whether the B. subtilis ksgA and speD mutants were as fit as the parental strain. The results of growth competition assays using strains of B. subtilis, E. coli, S. coelicolor, and M. smegmatis are summarized in Table 7. It is evident that the ksgA mutations resulted in a substantial disadvantage in growth fitness and that this was especially pronounced in speD mutations and rpsL mutations. A modest level of disadvantage in growth fitness was also detected in the M. smegmatis tlyA mutant. In contrast, no substantial disadvantages resulting from the rsmG mutation were detected in any bacteria tested, irrespective of culture conditions using LB medium (representing a rich medium) or sterilized soil (representing a poor medium) (Table 7). The B. subtilis mutant strains used here all displayed abundant sporulation, the ability of which might affect the overall growth fitness of each strain. In particular, rsmG mutant KO-750 displayed an ability for sporulation that was superior to that of the wild-type strain 168 (Table 8), which may account for the observation of greater fitness of this strain than that of the wild-type 168 (Table 7).

View this table: [in this window] [in a new window]

TABLE 7. Growth fitness of ksgA and speD mutants compared to that of rpsL and rsmG mutants

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The bacterial enzyme KsgA catalyzes the transfer of a total of four methyl groups from S-adenosylmethionine to two adjacent adenosine bases (A1518 and A1519 in the loop of helix 45) to produce N⁶,N⁶-dimethyladenosine in 16S rRNA (12, 13). To date, 10 methylatable nucleosides within E. coli 16S rRNA (1), including position G527, which is methylated by the enzyme RsmG (20), are reported. Although the collective importance of these rRNA modifications for protein synthesis has been demonstrated (11), the functions of individual methylations are still unclear, as inactivation of the genes encoding their cognate methyltransferases does not affect cell viability (2, 4, 15). In the present study, we demonstrated the following four characteristic aspects of ksgA mutations: (i) ksgA mutations emerge at a high frequency of 10^–6, conferring a modest level of resistance to kasugamycin that is apparently due to the dispensability of this gene, allowing cells to remain viable; (ii) once cells acquire the ksgA mutation, they produce a mutation conferring a high level of resistance to kasugamycin at an extraordinarily high frequency; (iii) the mutation conferring a high level of kasugamycin resistance occurs solely within the speD gene (at least in B. subtilis), which encodes S-adenosylmethionine decarboxylase; and (iv) unlike the rsmG mutation, the ksgA mutation gives rise to a disadvantage in overall growth fitness. These results clearly indicate similarities (i and ii) and dissimilarities (iii and iv) between ksgA and rsmG mutations, both of which share common characteristics as represented by the failure to methylate the 16S rRNA bases (not ribose moieties). The mechanism underlying the high-frequency emergence of high-level kasugamycin (and also high-level streptomycin) resistance is not yet clear, but it is unlikely that KsgA (and RsmG) functions as an anti-mutator-like protein, as the ksgA (and rsmG) mutation did not affect the frequency at which mutants resistant to antibiotics other than kasugamycin or streptomycin emerged (23; this study). In addition, we can exclude the possibility that the observed high frequency of the emergence of mutations conferring high-level kasugamycin resistance is caused by an increase in persistence due to ksgA mutation, as mutations conferring a low level of resistance to thiostrepton or rifampin gave rise to only a slight increase (2.5-fold at most) in emergence, possibly due to the increased persistence (Table 5). This conclusion was further supported by the observation that speD mutants did not produce mutants with high-level kasugamycin resistance at a high frequency (Table 5), despite the fact that the speD mutant and ksgA mutant have similar resistance levels to kasugamycin (Table 6). Thus, the mechanism underlying the observed peculiar phenomena, common to ksgA and rsmG mutants, remains to be studied at the molecular level. Nonetheless, the emergence of mutants with high-level kasugamycin resistance at an extraordinarily high frequency due to ksgA mutation is of considerable importance from an agricultural viewpoint, given that kasugamycin has been widely used as a potent anti-rice blast drug. Apart from the kasugamycin resistance, it is also notable that the high-frequency emergence of mutants with high-level streptomycin resistance was detected even in the genetic background of relA (coding for ppGpp synthetase) as examined using the E. coli rsmG relA double mutant (Y. Tanaka, S. Okamoto, and K. Ochi, unpublished data), indicating that ppGpp is irrelevant to the high-frequency emergence observed.

Polyamines as represented by spermidine, spermine, and putrescine are extremely important for the cell, although they are dispensable under routine laboratory growth conditions. They are involved in macromolecular syntheses and in particular in the modulation of translation accuracy at steps which may be essential for survival of the cell populations (5). Polyamine auxotrophy in E. coli due to SpeD inactivation has been implicated in resistance to aminoglycoside antibiotics, including kasugamycin (9, 10). Ribosomes of a polyamine auxotrophic E. coli mutant starved of polyamines showed a reduced affinity for streptomycin, and their protein synthesis activity was less sensitive to the drug (10). These previous findings indicate a causal relationship between kasugamycin resistance and intracellular polyamine contents, and in turn account for our observation that B. subtilis ksgA speD double mutants showed a higher level of kasugamycin-resistant phenotypes than ksgA single mutants (Table 6). Although putrescine did not negate the speD mutation-induced kasugamycin resistance in B. subtilis (see Results), it is likely that putrescine (present at a high concentration in E. coli) (28) participates in the level of kasugamycin resistance in E. coli, accounting for the question of why speD mutants were not found in E. coli with the same screening procedure (see Results). It is also notable that no speD mutants were detected in the study when B. subtilis ksgA⁺ strains 168 and RIK543 were used, although speD mutation alone could confer resistance to kasugamycin (Table 6). It is conceivable that the emergence of speD mutation (in the genetic background of ksgA⁺) was much lower than that of the ksgA mutation, although these genes are both dispensable.

Homologs of ksgA (and also rsmG) are highly conserved among eubacteria, so it was somewhat surprising that, despite the apparent important contribution made by KsgA and RsmG to ribosomal function, ksgA mutations (this study) and disruption of rsmG (20, 23) had no effect on growth of E. coli and B. subtilis. However, it is notable that ksgA mutations carried fitness costs, as demonstrated by the growth competition assay (Table 7), implying an important role of 16S rRNA methylation in survival. In this regard, the complete lack of fitness cost associated with rsmG mutation is distinctive, as demonstrated by several bacteria (Table 7). Given the possible involvement of the rRNA methyltransferases in ribosome homeostasis (5a), the ksgA mutation, together with the rsmG mutation, may provide new insights to uncover unknown mechanisms of mutation and ribosomal function.

 
This work was supported by grants to K.O. from the Effective Promotion of Joint Research of Special Coordination Funds (the Ministry of Education, Culture, Sports, Science, and Technology of the Japanese government).

We gratefully acknowledge Takayoshi Tamura (Meiji Seika Kaisha) for the generous supply of kasugamycin and Genefrontier Corp., Tokyo, Japan, for supporting the mutation search using the comparative genome sequencing technique.

 
* Corresponding author. Mailing address: National Food Research Institute, 2-1-12 Kannondai, Tsukuba, Ibaraki 305-8642, Japan. Phone: 81-29-838-8125. Fax: 81-29-838-7996. E-mail: kochi{at}affrc.go.jp

Published ahead of print on 10 November 2008.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Andersen, N. M., and S. Douthwaite. 2006. YebU is a m⁵C methyltransferase specific for 16S rRNA nucleotide 1407. J. Mol. Biol. 359:777-786.[CrossRef][Medline]
2
2. Andersson, D. I., and B. R. Levin. 1999. The biological cost of antibiotic resistance. Curr. Opin. Microbiol. 2:489-493.[CrossRef][Medline]
3
3. Baba, T., T. Ara, M. Hasegawa, Y. Takai, Y. Okumura, M. Baba, K. A. Datsenko, M. Tomita, B. L. Wanner, and H. Mori. 2006. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol. Syst. Biol. 2:2006.0008. doi:10.1038/msb4100050.
4
4. Basturea, G. N., K. E. Rudd, and M. P. Deutscher. 2006. Identification and characterization of RsmE, the founding member of a new RNA base methyltransferase family. RNA 12:426-434.[Abstract/Free Full Text]
5
5. Cohen, S. 1998. A guide to the polyamines. Oxford University Press, Oxford, United Kingdom.
5
6. Connolly, K., J. P. Rife, and G. Culver. 2008. Mechanistic insight into the ribosome biogenesis functions of the ancient protein KsgA. Mol. Microbiol. 70:1062-1075.[Medline]
6
7. Cundliffe, F. 1990. Recognition sites for antibiotics within rRNA, p. 479-490. In W. E. Hill, A. Dahlberg, R. A. Garrett, P. B. Moore, D. Schlessinger, and J. R. Warner (ed.), The ribosome: structure, function, and evolution. American Society for Microbiology, Washington, DC.
7
8. Dworkin, J., and R. Losick. 2001. Linking nutritional status to gene activation and development. Genes Dev. 15:1051-1054.[Free Full Text]
8
9. Flores, H. E., and A. W. Galston. 1982. Analysis of polyamines in higher plants by high performance liquid chromatography. Plant Physiol. 69:701-706.[Abstract/Free Full Text]
9
10. Goldemberg, S. H., and I. D. Algranati. 1981. Polyamine requirement for streptomycin action on protein synthesis in bacteria. Eur. J. Biochem. 117:251-255.[Medline]
10
11. Goldemberg, S. H., J. G. Fernandez-Velasco, and I. D. Algranati. 1982. Differential binding of streptomycin to ribosomes of polyamine-dependent bacteria grown in the absence and presence of putrescine. FEBS Lett. 142:275-279.[Medline]
11
12. Green, R., and H. F. Noller. 1999. Reconstitution of functional 50S ribosomes from in vitro transcripts of Bacillus stearothermophilus 23S rRNA. Biochemistry 38:1772-1779.[CrossRef][Medline]
12
13. Helser, T. L., J. E. Davies, and J. E. Dahlberg. 1971. Change in methylation of 16S ribosomal RNA associated with mutation to kasugamycin resistance in Escherichia coli. Nat. New Biol. 233:12-14.[Medline]
13
14. Helser, T. L., J. E. Davies, and J. E. Dahlberg. 1972. Mechanism of kasugamycin resistance in Escherichia coli. Nat. New Biol. 235:6-9.[Medline]
14
15. Johansen, S. K., C. E. Maus, B. B. Plikaytis, and S. Douthwaite. 2006. Capreomycin binds across the ribosomal subunit interface using tlyA-encoded 2'-O-methylations in 16S and 23S rRNAs. Mol. Cell 23:173-182.[CrossRef][Medline]
15
16. Lesnyak, D. V., J. Osipiuk, T. Skarina, P. V. Sergiev, A. A. Bogdanov, A. Edwards, A. Savchenko, A. Joachimiak, and O. A. Dontsova. 2007. Methyltransferase that modifies guanine 966 of the 16S rRNA: functional identification and tertiary structure. J. Biol. Chem. 282:5880-5887.[Abstract/Free Full Text]
16
17. Maus, C. E., B. B. Plikaytis, and T. M. Shinnick. 2005. Mutation of tlyA confers capreomycin resistance in Mycobacterium tuberculosis. Antimicrob. Agents Chemother. 49:571-577.[Abstract/Free Full Text]
17
18. Melançon, P., C. Lemieux, and L. Brakier-Gingras. 1988. A mutation in the 530 loop of Escherichia coli 16S ribosomal RNA causes resistance to streptomycin. Nucleic Acids Res. 16:9631-9639.[Abstract/Free Full Text]
18
19. Nanamiya, H., K. Kasai, A. Nozawa, C. S. Yun, T. Narisawa, K. Murakami, Y. Natori, F. Kawamura, and Y. Tozawa. 2008. Identification and functional analysis of novel (p)ppGpp synthetase genes in Bacillus subtilis. Mol. Microbiol. 67:291-304.[Medline]
19
20. Nishimura, K., T. Hosaka, S. Tokuyama, S. Okamoto, and K. Ochi. 2007. Mutations in rsmG, encoding a 16S rRNA methyltransferase, result in low-level streptomycin resistance and antibiotic overproduction in Streptomyces coelicolor A3(2). J. Bacteriol. 189:3876-3883.[Abstract/Free Full Text]
20
21. Nishimura, K., S. K. Johansen, T. Inaoka, T. Hosaka, S. Tokuyama, Y. Tahara, S. Okamoto, F. Kawamura, S. Douthwaite, and K. Ochi. 2007. Identification of the RsmG methyltransferase target as 16S rRNA nucleotide G527 and characterization of Bacillus subtilis rsmG mutants. J. Bacteriol. 189:6068-6073.[Abstract/Free Full Text]
21
22. O'Farrell, H. C., J. N. Scarsdale, and J. P. Rife. 2004. Crystal structure of KsgA, a universally conserved rRNA adenine dimethyltransferase in Escherichia coli. J. Mol. Biol. 339:337-353.[CrossRef][Medline]
22
23. Ogle, J. M., and V. Ramakrishnan. 2005. Structural insights into translational fidelity. Annu. Rev. Biochem. 74:129-177.[CrossRef][Medline]
23
24. Okamoto, S., A. Tamaru, C. Nakajima, K. Nishimura, Y. Tanaka, S. Tokuyama, Y. Suzuki, and K. Ochi. 2007. Loss of a conserved 7-methylguanosine modification in 16S rRNA confers low-level streptomycin resistance in bacteria. Mol. Microbiol. 63:1096-1106.[CrossRef][Medline]
24
25. Piggot, P. J., and R. Losick. 2002. Sporulation genes and intercompartmental regulation, p. 483-517. In A. L. Sonenshein, J. A. Hoch, and R. Losick (ed.), Bacillus subtilis and its closest relatives: from genes to cells. ASM Press, Washington, DC.
25
26. Poehlsgaard, J., and S. Douthwaite. 2005. The bacterial ribosome as a target for antibiotics. Nat. Rev. Microbiol. 3:870-881.[CrossRef][Medline]
26
27. Powers, T., and H. F. Noller. 1991. A functional pseudoknot in 16S ribosomal RNA. EMBO J. 10:2203-2214.[Medline]
27
28. Sekowska, A., P. Bertin, and A. Danchin. 1998. Characterization of polyamine synthesis pathway in Bacillus subtilis 168. Mol. Microbiol. 29:851-858.[CrossRef][Medline]
28
29. Sekowska, A., J. Y. Coppée, J. P. Le Caer, I. Martin-Verstraete, and A. Danchin. 2000. S-Adenosylmethionine decarboxylase of Bacillus subtilis is closely related to archaebacterial counterparts. Mol. Microbiol. 36:1135-1147.[CrossRef][Medline]
29
30. Shima, J., A. Hesketh, S. Okamoto, S. Kawamoto, and K. Ochi. 1996. Induction of actinorhodin production by rpsL (encoding ribosomal protein S12) mutations that confer streptomycin resistance in Streptomyces lividans and Streptomyces coelicolor A3(2). J. Bacteriol. 178:7276-7284.[Abstract/Free Full Text]
30
31. Springer, B., Y. G. Kidan, T. Prammananan, K. Ellrott, E. C. Böttger, and P. Sander. 2001. Mechanisms of streptomycin resistance: selection of mutations in the 16S rRNA gene conferring resistance. Antimicrob. Agents Chemother. 45:2877-2884.[Abstract/Free Full Text]
31
32. Umezawa, H., Y. Okami, T. Hashimoto, Y. Suhara, M. Hamada, and T. Takeuchi. 1965. A new antibiotic, kasugamycin. J. Antibiot. (Tokyo) 18:101-103.[Medline]
32
33. Vila-Sanjurjo, A., C. L. Squires, and A. E. Dahlberg. 1999. Isolation of kasugamycin resistant mutants in the 16 S ribosomal RNA of Escherichia coli. J. Mol. Biol. 293:1-8.[CrossRef][Medline]

---

Antimicrobial Agents and Chemotherapy, January 2009, p. 193-201, Vol. 53, No. 1
0066-4804/09/$08.00+0     doi:10.1128/AAC.00873-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Kim, J.-Y., Inaoka, T., Hirooka, K., Matsuoka, H., Murata, M., Ohki, R., Adachi, Y., Fujita, Y., Ochi, K. (2009). Identification and Characterization of a Novel Multidrug Resistance Operon, mdtRP (yusOP), of Bacillus subtilis. J. Bacteriol. 191: 3273-3281 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ochi, K. Articles by Kawamura, F. Search for Related Content PubMed PubMed Citation Articles by Ochi, K. Articles by Kawamura, F.