Article Text
PDF
Abstract
Objective A single 1 g dose regimen of azithromycin has been recommended for the treatment of Mycoplasma genitalium infections. The authors evaluated whether this regimen could select M genitalium strains with macrolide resistance after treatment for M genitalium-positive non-gonococcal urethritis.
Methods In seven men with non-gonococcal urethritis, who were infected with M genitalium without macrolide resistance-associated mutations but experienced microbiological azithromycin treatment failure, M genitalium DNAs in their post-treatment urine specimens were examined for mutations in the 23S rRNA gene and the ribosomal protein genes of L4 and L22. To assess the relatedness of M genitalium strains before and after treatment, their DNAs in pretreatment and post-treatment urine were genotyped by analysing short tandem repeats of an AGT/AAT unit in the MG309 gene and single nucleotide polymorphisms in the MG191 gene.
Results In four of seven patients, M genitalium in post-treatment urine had an A-to-G transition at nucleotide position 2071 or 2072, corresponding to 2058 or 2059 in the 23S rRNA gene of Escherichia coli. In one of the four strains, Pro81Ser in the ribosomal protein L4 accompanied the mutation in the 23S rRNA gene. The genotyping of M genitalium DNAs suggested that these four post-treatment strains were selected from the respective closely related or identical pretreatment strains without macrolide resistance-associated mutations by the treatment.
Conclusions The single 1 g dose treatment of azithromycin could select M genitalium strains harbouring macrolide resistance-associated mutations. For M genitalium, this regimen might increase the risk of macrolide resistance selection after treatment.
- Mycoplasma genitalium
- azithromycin
- macrolide resistance
- 23S rRNA
Statistics from Altmetric.com
Mycoplasma genitalium is recognised as one of the important pathogens in sexually transmitted infections.1 A single 1 g dose of azithromycin has been considered as one of several relevant treatments for the treatment of M genitalium infections.2 Recently, however, macrolide-resistant clinical strains of M genitalium have been isolated from some patients with M genitalium-positive non-gonococcal urethritis (NGU) experiencing treatment failure.3 4 In these clinical isolates, mutations in domain V of 23S rRNA were associated with high-level macrolide resistance, and mutations in ribosomal protein genes of L4 and L22 were also found.4 In our previous study,5 we examined M genitalium DNAs in urine specimens from 25 Japanese men with NGU for macrolide resistance-associated mutations in the 23S rRNA gene and in the ribosomal protein genes of L4 and L22. We found that M genitalium in the urine from one man (4%) harboured macrolide resistance-associated mutations, including an A2059G (Escherichia coli numbering) mutation in the 23S rRNA gene and a Gly93Glu change in L22. The M genitalium in urine specimens from the remaining 24 men had no mutations in domain V of 23S rRNA. However, 7 of these 24 men experienced treatment failure with a single 1 g dose regimen of azithromycin. In the present study, we evaluated whether azithromycin resistance was selected in M genitalium by a single 1 g dose regimen of azithromycin in these seven men by examining M genitalium DNAs in their post-treatment urine specimens for the emergence of mutations in the 23S rRNA gene and the ribosomal protein genes of L4 and L22.
Materials and methods
Patients and urine specimens
This retrospective study was approved by the Institutional Review Board of the Graduate School of Medicine, Gifu University, Gifu, Japan. Of the 24 men with M genitalium-positive NGU enrolled in our previous study5 and who had no mutations related to macrolide resistance in urinary M genitalium DNAs, 7 men who experienced microbiological treatment failure with a single 1 g dose regimen of azithromycin were enrolled in the present study. Four of these seven men also had persistent urethritis symptoms after treatment. Six of these seven men reported sexual abstinence during the period of time from the administration of azithromycin to their successive follow-up visits. The post-treatment urine specimens were collected 14–28 days after the administration of azithromycin. M genitalium DNAs in their stored pretreatment and post-treatment urine specimens were analysed.
Determination of domain V of 23S rRNA and ribosomal proteins of L4 and L22 of M genitalium
Domain V of the 23S rRNA gene and the ribosomal protein genes of L4 and L22 of M genitalium were amplified by PCR from the DNAs, which were purified from the post-treatment urine specimens from the seven men, as reported previously,4 and were then sequenced.
Assessment of the relatedness of M genitalium strains in the pretreatment and post-treatment urine specimens
To assess the relatedness of M genitalium strains before and after treatment, their DNAs present in the pretreatment and post-treatment urine specimens were genotyped by analysing short tandem repeats (STRs) of an AGT/AAT unit in the MG309 gene and single nucleotide polymorphisms (SNPs) in the MG191 gene as reported previously.6
Results
In four of the seven men who were infected with M genitalium without macrolide resistance-associated mutations but who experienced azithromycin treatment failure, M genitalium in the post-treatment urine specimens had an A-to-G transition at nucleotide position 2071 or 2072 in the 23S rRNA gene of M genitalium, corresponding to 2058 or 2059 in E coli (table 1). In M genitalium of the post-treatment urine specimen from patient 4, Pro81Ser in the ribosomal protein L4 accompanied the A2058G (E coli numbering) mutation in the 23S rRNA gene.
Mutations in 23S rRNA gene, amino acid alterations in ribosomal proteins of L4 and L22 and genotypes of Mycoplasma genitalium strains present in the urine specimens of seven men failing azithromycin treatment
In patient 1, the copy number of AGT/AAT in the MG309 gene of the M genitalium strain in the post-treatment urine specimen differed from that in the pretreatment urine specimen by only one, and the MG191-SNPs were identical. In this case, the post-treatment strain could be closely related to the pretreatment strain. In the remaining six patients, the MG309-STRs and MG191-SNPs of M genitalium strains in the pretreatment and post-treatment urine specimens were identical. In four of the seven patients, M genitalium strains having macrolide resistance-associated mutations in the 23S rRNA gene were selected from the respective closely related or identical strains without the mutations by a single 1 g dose of azithromycin.
Discussion
The mutations in the residues of the 23S rRNA gene of M genitalium found in the present study and which correspond to A2058 and A2059 in domain V of the 23S rRNA gene of E coli are critical for the binding of macrolides. These mutations in the 23S rRNA gene could be associated with a high-level resistance to macrolides, including azithromycin, in M genitalium, because this species possesses only one copy of the rRNA gene operon.7 Macrolide resistance-associated missense mutations in L4 tend to be localised to the region analogous to Gln62-Gly66 in L4 of E coli, which are closest to the macrolide-binding site.8 In some bacterial species, however, macrolide-resistant mutations outside this cluster have been reported.9 Therefore, Pro81Ser in the ribosomal protein L4 of M genitalium could be associated with macrolide resistance.
A single 1 g dose regimen of azithromycin selected M genitalium strains harbouring macrolide resistance-associated mutations after treatment in four of the seven patients. These four men reported sexual abstinence during the period of time from the administration of azithromycin to their successive follow-up visits. In addition, the findings from genotyping of the strains in the pretreatment and post-treatment urine specimens suggested that the four post-treatment strains were selected from the respective closely related or identical pretreatment strains without macrolide resistance-associated mutations by the treatment. One of the limitations of this study is the lack of analysis of other mechanisms of macrolide resistance, including the enhancement of active efflux and modification of target sites by methylation or mutation (domain II of 23S rRNA).7 10 In addition, we did not isolate the clinical strains and were therefore unable to determine their susceptibilities to azithromycin. However, M genitalium strains harbouring mutations in the 23S rRNA gene could be highly resistant to azithromycin,4 and the selection of such strains after treatment would be a matter of serious concern. The single 1 g dose regimen of azithromycin has been recommended as a presumptive treatment for non-gonococcal urogenital infections. For M genitalium, however, this regimen might increase the risk of macrolide resistance selection after treatment. We should continue to monitor antimicrobial susceptibilities of M genitalium strains. The non-culture approach used in the present study to assess antimicrobial resistance of M genitalium clinical strains will be helpful, until culturing of the mycoplasma from clinical specimens and testing of their in vitro antimicrobial susceptibility can be performed easily in laboratories.
References
Footnotes
Funding This study was supported in part by the Japan Society for the Promotion of Science (JSPS), Japan (Grant-in-Aid for Scientific Research (C) 22591788).
Competing interests None.
Ethics approval This study was approved by the Institutional Review Board of the Graduate School of Medicine, Gifu University, Gifu, Japan (reference number 22-11).
Provenance and Peer review Not commissioned; externally peer reviewed.