High prevalence of the A2058T macrolide resistance-associated mutation in Mycoplasma genitalium strains from the Netherlands

Sir, Mycoplasma genitalium causes sexually transmissible disease (STD). Approximately 4.5% of Dutch patients undergoing STD screening are infected with the bacterium.1 Although urogenital infections with M. genitalium are often treated with a single dose of 1 g azithromycin,2 the efficacy of this regimen has decreased worldwide from 85% to 60% over the past 7 years due to the spread of resistant M. genitalium strains.3 However, many of the samples used in studies of the prevalence of M. genitalium susceptibility were from STD clinics. The expected prevalence of STDs is higher in STD clinics, and thus these studies are probably not representative of the general population. Systematically collected data on the antibiotic susceptibility of M. genitalium from the general population are still scarce, which has implications for public health as reliable data are important for primary-care general practitioners. Mutations in the 23S rRNA gene at positions A2058 and A2059 (Escherichia coli numbering) in region 5 (V region) are associated with azithromycin resistance and treatment failure.4 The most prevalent mutations are A2058G and A2059G, but in a Dutch study by Nijhuis et al.5 an exceptionally high prevalence of the mutation A2058T (27.3%) was found. However, this study involved samples from a single laboratory and it is not clear if these data can be extrapolated nationally. To test this hypothesis, the prevalence of macrolide resistance-associated mutations in M. genitalium, especially A2058T, in the general population was determined. Samples sent to three laboratories [Canisius-Wilhelmina Hospital (CWZ), Nijmegen; PAMM, Veldhoven; Rijnstate Hospital, Velp] for routine molecular detection of M. genitalium (PCR performance essentially as described by Jensen et al.6) were collected between January 2014 and January 2015. CWZ and Rijnstate are both located in the east, while PAMM is located in the south of the Netherlands. Only patients directly referred by their family physician were selected and samples referred to the laboratories by STD clinics were excluded. In addition, only the first positive sample from a patient was included. M. genitalium-positive samples were subjected to 23S rRNA qPCR as described by Nijhuis et al.5 and amplicons were sequenced to identify macrolide resistance-associated mutations. Samples where no sequence was determined after the 23S rRNA qPCR were tested again using a reverse transcriptase nested qPCR (RTN–qPCR). The same primers were used for the RT–qPCR using TaqMan Fast Virus 1-Step Master Mix (Applied Biosystems, Palo Alto, CA, USA). Subsequently, the obtained RT–qPCR amplicons were subjected to a second qPCR using the forward primer 5′-AGGAGGTTAGCAATTTATTGCA-3′ and the reverse primer 5′-TCTCTACATGGTGGTGTTTTGA-3′. Next the 263 bp formed amplicon was used for sequencing. This study was approved by the Medical Ethics Committee of the University Medical Centre Utrecht (protocol number 16-238/C). Comparison between the three regions of all samples tested for macrolide resistance-associated mutations in M. genitalium. *Statistically difference: P < 0.05. In the present study, the most frequently detected mutations were A2058G (n = 18/46, 39.1%) and A2059G (n = 16/46, 34.8%), which is comparable to other countries.7–9 The high prevalence of A2058T (n = 10/46, 21.7%) in this study is rarely found in other countries. In the previously performed regional Dutch study by Nijhuis et al.,5 an even higher proportion (n = 12/44, 27.3%) of the mutation A2058T was found. Although the prevalence of the A2058T substitution was lower in the present study, this mutation was found in samples from all three regions. Since the A2058T mutation is rare in other countries, but commonly found in different regions in the Netherlands, this finding might be explained by clonal spread. However, the sequencing of the 23S rRNA gene revealed nucleotide differences between the variant A2058T strains, indicating that this is not a single clonal outbreak. Because of the high prevalence of resistance to azithromycin in M. genitalium, it is advisable to use molecular tests to detect macrolide resistance-associated mutations to guide therapy and to prevent further spread of resistant bacteria. Molecular methods such as qPCR should be designed to detect all macrolide resistance-associated mutations, including the mutation A2058T. Patients infected with proven macrolide-resistant strains can be empirically treated with moxifloxacin, as resistance to this antibiotic is still scarce in Europe.10,11 This study was supported by internal funding as part of our routine work. None to declare.