Isolates collected from two mother-neonate pairs were also analysed (Additional file 1: Table S1). In the first case, isolates obtained from the bloodstream of the mother and her new born infant belonged to the same MLVA type 8. In the second case, isolates obtained from the blood culture and cervix of the mother yielded the MLVA type GS 1101 7, as well as isolates collected from blood culture and throat sample of

her neonate. The genetic relationships of the 210 isolates were deduced by construction of an MST (Figure 1). This population model highlighted one major clonal complex, composed of 207 isolates belonged to 38 MLVA types. A second clonal complex could be defined for one urogenital isolate (Mh-3560) collected in 2003 and yielding the MLVA type 24. A third clonal complex was represented by two respiratory isolates (Mh-2327, Mh-2477) collected six months apart in 1996–1997 from the same patient and were of the MLVA type 33. Interestingly, the two VNTRs that were not used for the MLVA assay (due

Ferrostatin-1 to lack of discrimination) presented size variation only with these two isolates and harboured identical TR numbers between both isolates. The MST population modelling indicated the genetic diversity among the isolates tested. Unfortunately, there was no obvious link between the MLVA type and the isolate year of collection, the patient’s age or sex, the anatomical site of collection, or the antibiotic resistance. Indeed, the tetracycline or ofloxacin-resistant isolates were dispersed into 25 MLVA types (Figure 1). It should be noted that no significant difference in the repartition of MLVA types was identified between M. hominis cervical isolates present in ≥ 104 CCU /ml in patients without BV and in patients with BV (Additional file 1: Table S1). Figure 1 Minimum spanning tree of 210  M. hominis isolates based on categorical analysis of five however VNTRs. Each circle represents a unique MLVA profile, as indicated by

a number. The size of each circle is proportional to the number of isolates belonging to the indicated MLVA type. Thick connecting lines represent one marker difference. Regular connecting lines represent two marker differences. MLVA types connected by a same background could be considered a clonal complex. Asterisks in the MLVA types indicate the presence of tetracycline-and/or ofloxacin-resistant isolates. Discussion In this study, we present an MLVA-based molecular typing system for the discrimination of M. hominis isolates. We used this method on a group of 210 temporally separated isolates from French patients and obtained from a variety of urogenital and extragenital clinical circumstances. This effort represents the most comprehensive M. hominis molecular typing study because, until now, other studies were realised only on urogenital isolates and few isolates were tested [7–10]. MLVA typing of M. hominis is important both individually and epidemiologically.