Mycoplasma genitalium causes sexually transmitted infecti.ons in men and women. Treatment failures to macrolides and fluoroquinolones have been reported worldwide. Although the mgpB typing method has often been used in M. genitalium-infected men who have sex with men (MSM), limited typing data are available for M. genitalium-infected women. In this study, we aimed to investigate the genetic relationship between M. genitalium strains and their antibiotic resistance profile in a cohort of MSM (86.2% on HIV preexposure prophylaxis [PrEP], 13.8% HIV positive) and a large cohort of women using mgpB/MG309 typing. The mgpB types were determined in 374 samples from 305 women and 65 MSM. Three MSM and one woman had two concurrent or subsequent samples. Macrolide and fluoroquinolone resistance-associated mutations were searched in the 23S rRNA as well as parC and gyrA genes. The mgpB phylogenetic construction revealed three large clusters that differed according to sexual practices and geographical origin of patients. The prevalence of macrolide and fluoroquinolone resistance was significantly higher in MSM compared with women (95.4% vs. 14.1% and 30.6% vs. 7.2%, p < 0.001, respectively). The macrolide resistance spread was polyclonal in both populations, but clonal diffusion of two dual-resistant types was observed in PrEP users in association with high antibiotic pressure and dense connectivity in this population.

Mycoplasma hominis is an opportunistic human pathogen associated with genital and neonatal infections. Until this study, the lack of a reliable transformation method for the genetic manipulation of M. hominis hindered the investigation of the pathogenicity and the peculiar arginine-based metabolism of this bacterium. A genomic analysis of 20 different M. hominis strains revealed a number of putative restriction-modification systems in this species. Despite the presence of these systems, a reproducible polyethylene glycol (PEG)-mediated transformation protocol was successfully developed in this study for three different strains: two clinical isolates and the M132 reference strain. Transformants were generated by transposon mutagenesis with an efficiency of approximately 10⁻⁹ transformants/cell/µg plasmid and were shown to carry single or multiple mini-transposons randomly inserted within their genomes. One M132-mutant was observed to carry a single-copy transposon inserted within the gene encoding P75, a protein potentially involved in adhesion. However, no difference in adhesion was observed in cell-assays between this mutant and the M132 parent strain. Whole genome sequencing of mutants carrying multiple copies of the transposon further revealed the occurrence of genomic rearrangements. Overall, this is the first time that genetically modified strains of M. hominis have been obtained by random mutagenesis using a mini-transposon conferring resistance to tetracycline.

Mycoplasma hominis is a human urogenital pathogen involved in gynaecological, neonatal and extra-genital infections. However, no versatile genetic tools are currently available to study the pathogenicity of this bacterium. Targeting-Induced Local Lesions IN Genomes (TILLING) is a reverse-genetic method that combines point mutations induced by chemical mutagenesis with a DNA screening technique. We used ethyl methanesulfonate (EMS) that introduces C-G to T-A transition mutations to generate a library of M. hominis mutants. As a proof of concept, mutagenized organisms were screened for mutations in two target genes previously associated with the mycoplasma pathogenicity, the vaa gene encoding an adhesin lipoprotein and the oppA gene encoding the main ectoATPase of the bacterium. The resulting mutants were evaluated using functional assays, an adhesion to HeLa cell assay for vaa-mutants and an ATPase activity test for oppA-mutants. A 1200-clone library was generated by exposing M. hominis PG21 to 9 mg/mL EMS for 3 h. To identify mutants of interest, targeted gene fragments were amplified, heat-denatured, slowly reannealed and digested with the mismatch-specific endonuclease ENDO1. If multiple alleles were present in the PCR amplicons, these alleles formed heteroduplexes during reannealing that were specifically cleaved by ENDO1 at mismatching positions. A total of four vaa-mutants and two oppA-mutants harbouring missense mutations were obtained and fully sequenced. Zero to eight additional mutations were identified in the genomes of each mutant. The vaa-mutants were tested for adhesion to immobilized HeLa cells but their adhesion was not significantly different from the adhesion of M. hominis PG21. One of the two oppA-mutants that were tested for ATPase activity presented a higher affinity for its ATP substrate than the parental strain. For the first time, we demonstrated that M. hominis gene-targeted mutants could be successfully obtained using this TILLING strategy. In the absence of robust genetic tools for studying M. hominis, the TILLING strategy that can target any gene of the genome could help to elucidate gene functions and to better understand the pathogenesis of this human pathogenic species.

Macrolide and fluoroquinolone resistance (MLr/FQr) in Mycoplasma genitalium (MG) infections is concerning worldwide. Current guidelines recommend performing MLr detection in MG-positive cases to adjust antimicrobial therapy. We aimed to evaluate the usefulness of PCR followed by pyrosequencing for MLr detection in comparison with a one-step commercial assay and to assess the prevalence of MLr and FQr in Badalona, Spain. A total of 415 MG-positive samples by Allplex STI-7 (Seegene) were analyzed for MLr detection by pyrosequencing. From those, 179 samples were further analyzed for MG and MLr by ResistancePlus^® MG kit (SpeeDx) and 100 of them also for fluoroquinolone resistance (FQr) by sequencing the parC gene. Regarding MG detection, Allplex and Resistance Plus^® showed an overall agreement of 87%, but this value rose to 95.4% if we compare them for MLr detection. Prevalence of MLr was 23.1% in Badalona, but this rate increased to 73.7% in the HIV-positive patients cohort. FQr detection showed 3% of resistant strains. Pyrosequencing is a convenient and cheap technique for MLr detection, but one-step tools should be considered in high-throughput laboratories. Despite the fact that MLr remained moderate and FQr was low in our study, simultaneous MG and MLr detection would improve patient’s management applying resistance-guided treatment strategies.

In women being screened for Chlamydia trachomatis (CT) infection, Mycoplasma genitalium (MG) testing is not recommended. Among women positive for CT, studies reported that 7% to 28% had concurrent urogenital MG. The aim of this study was to investigate the impact on MG of a single 1-g dose of azithromycin or a 7-day course of doxycycline regimens as part of CT treatment. MG was tested in vaginal swabs from the Chlazidoxy cohort collected at baseline and the 6-week follow-up visit after CT treatment. MG genotype and antibiotic resistance were analysed by sequencing the mgpB gene, and the parC, 23S and 16S rRNA genes. Among the 281 women included, 131 received azithromycin and 150 doxycycline. The proportion of MG-positive women was 16% (21/131) in the azithromycin group, and 12% (18/150) in the doxycycline group (P = 0.39). In each group, a similar proportion of macrolide (11.1% vs 26.7%, P = 0.37), and quinolone (11.7% vs 16.7%, P = 1) resistance-associated mutations were detected. Six weeks after treatment initiation, azithromycin eradicated MG in 85.7% (18/21) of the cases and doxycycline in 55.6% (10/18) (P = 0.07). In asymptomatic women infected with CT, the treatment with azithromycin or doxycycline eradicated MG in the majority of the cases.