In the present study, it was shown that
M. genitalium is a frequent cause of male urethritis in Moscow, Russia and that the
M. genitalium eradication rate in 2006-2008 was 93.5% with the recommended first-line treatment in Russia, that is, josamycin 500 mg three times daily, 10 days [
21]. These are the first data published regarding efficacy using josamycin for treatment of
M. genitalium infections in the international (English language) literature. Accordingly, no evidence-based data supporting josamycin, compared to e.g. azithromycin, for treatment of
M. genitalium infections and exceedingly limited
in vitro data regarding josamycin susceptibility in
M. genitalium, have been internationally published. To our best knowledge, the only
in vitro antimicrobial susceptibility data were published in 1992 (only seven isolates were tested) and the activity was slightly lower compared to azithromycin [
25]. In the present study, it was also shown that josamycin resistance resulting in treatment failures can rapidly be selected during treatment. The eradication rate of
M. genitalium after the internationally widely used azithromycin 1 g therapy has substantially declined during the last decade [
1,
10,
12-
14,
16]. Treatment with azithromycin 1 g was also recently associated with selection of macrolide resistant
M. genitalium [
14], and it was also hypothesized that treatment failures are frequently associated with a higher load of
M. genitalium [
17]. The results of the present study, instead examining the macrolide josamycin, are in full concordance with these recent findings. Accordingly, three failures to treat male
M. genitalium urethritis with josamycin were verified. All of these patients had a high pre-treatment
M. genitalium load (>6 geq/mL [log
10]) and
M. genitalium cells with mutations in the macrolide resistance-determining region V of 23S rRNA [
1,
10,
13,
14,
18] were selected during treatment. Thus, despite that no 23S rRNA gene mutations were identified in the pre-treatment
M. genitalium specimens the term selected resistance instead of induced resistance was used in the present study. Accordingly, it was considered that an undetected minority population of macrolide resistant
M. genitalium cells might have been present already pre-treatment. These resistant cells, in the specimens with high
M. genitalium load, could subsequently be selected during treatment with the bacteriostatic josamycin. The probability of this selection scenario was considered higher particularly because it is unknown how frequently josamycin might be able to directly induce 23S rRNA gene mutations, especially during treatment of
M. genitalium with a very long generation time. The three identified treatment failures were further supported by the rebound in
M. genitalium loads in all three patients and the recrudescence of symptoms in two of the three patients. Thus, the pre-treatment
M. genitalium load might be an effective predictor of eradication efficacy with macrolides (and possibly additional antimicrobials) and selection of macrolide resistance due to mutations in the 23S rRNA gene. Unambiguously, additional data are crucial to support the recommendation of using josamycin regimen as first-line treatment in the Russian treatment guidelines for
M. genitalium infections.
The most frequent 23S rRNA mutations identified in
M. genitalium strains resistant to the 15-membered macrolide azithromycin have been A2058G and A2059G [
1,
13,
14,
17,
18,
30,
31]. In the present study, it was for the first time shown that the A2059G mutation is associated also with
M. genitalium resistance to 16-membered macrolides, e.g. josamycin. It has been earlier demonstrated that the A2059G mutation results in high-level resistance to 16-membered macrolides in the closely related species
M. pneumoniae [
32]. The A2062G mutation found in the remaining treatment failure has not been previously described in
M. genitalium. However,
M. hominis and
M. pneumoniae cells with A2062G and A2062T mutations in the 23S rRNA gene have been previously selected
in vitro using josamycin and shown to cause high-level josamycin resistance [
33,
34]. Interestingly, previous studies have suggested that mutations in the A2062 position of the 23S rRNA gene result in resistance to the 16-membered macrolides (e.g., josamycin), but not necessarily to the 15-membered macrolides (e.g., azithromycin), indicating a slightly different binding site [
35-
37]. In
M. pneumoniae, the A2062G mutation, detected after
in vitro exposure of josamycin, has also been shown to significantly increase the MICs of the streptogramin pristinamycin [
34]. This is of grave concern because pristinamycin is highly active against
M. genitalium infections and currently the only treatment option for infections resistant to both azithromycin and moxifloxacin [
38]. Nevertheless, it remains unknown if an A2058G/A2059G mutation (resulting in azithromycin resistance) and an A2062G mutation (resulting in resistance to pristinamycin (and josamycin)) can co-exist in the 23S rRNA gene of
M. genitalium, i.e. without severely decreasing the biological fitness and possibilities for transmission of the
M. genitalium strain
.