Lactobacilli are an important part of the vaginal microbiota. The presence of lactic acid, H2O2 and other by-products of these bacteria are beneficial in controlling the overgrowth of potentially pathogenic bacteria (i.e., bacterial vaginosis). For example, Antonio et al. reported that women who were not colonised with H2O2-producing lactobacilli, such as L. crispatus, L. iners, L. jensenii, L. gasseri and L. vaginalis, were 15 times more likely to have bacterial vaginosis than woman who were colonised by these strains. Despite their importance to women’s health, vaginal lactobacilli have not been extensively studied in the Mexican population. Historically, L. acidophilus was considered to be the dominant species in the human vagina. It is now known that the group of organisms previously known as L. acidophilus is highly heterogeneous, and includes at least six separate species [29, 31]. In addition, it has been shown that the majority of vaginal Lactobacillus strains from women of geographically separated countries belong to the four species L. iners, L. crispatus, L. gasseri and L. jensenii, indicating a high degree of species consistency in vaginal lactobacilli among women worldwide . The identity of the vaginal lactobacilli in Mexican women has not been well-studied, and the majority of published papers have used phenotypic approaches. Therefore, we based our approach on 16S rRNA gene sequencing and ERIC-PCR analysis.
The isolation of lactobacilli in the present study was performed using a general selective medium, and relatively few strains were isolated. This result is consistent with other reports, which indicate that lactobacilli may go undetected in the laboratory because their growth requires unique media and an extended incubation time . In 2001, Angeles-Lopez et al.  reported the isolation of lactobacilli in only 87 of 156 samples inoculated on MRS agar. In addition, it has been reported that L. iners, a species belonging to the L. acidophilus group, does not grow on MRS agar. Therefore, in the present study and other studies using MRS, this species was not evaluated and has been misidentified . Other authors have noted that even after recovery, strain misidentification can occur because the strains morphologically resemble those of other genera, including Corynebacterium, Clostridium and Streptococcus. Given the difficulty in isolating these bacteria, and the possibility that they may often be misidentified, culture-independent genetic approaches are now preferred over culture-dependent methods . Moreover, several types of vaginal microbiota exist in healthy women. Although Lactobacillus is often the predominant genus, the vaginal microbiota also includes a diverse assemblage of anaerobic microorganisms, which likely occur within the mestizo and Mexican populations.
When strains can be isolated, their identification can be aided by molecular techniques to distinguish between closely related species within the Lactobacillus genus, which can be impossible by phenotypic methods alone. Although a large number of molecular methodologies are currently available to study these bacteria, some of the techniques, such as PCR, denaturing gradient gel electrophoresis (DGGE) and thermal gradient gel electrophoresis (TGGE) [9, 34–37], still require improvement, especially with regard to the sensitivity, cost and quantitative power. Among all available molecular techniques, 16S rRNA sequencing analysis has been accepted as the most reliable method. Therefore, this method was used in the present work to provide the first genetic identification of the indigenous microbiota of the vaginal cavity of non-pregnant Mexican women.
The majority of the species detected in the Mexican population in the present study belong to the L. acidophilus group, although strains of the L. reuteri, L. casei and L. buchneri groups have also been identified . The majority of the lactobacilli found in the vaginal communities were phylogenetically related to L. gasseri, L. fermentum, L. rhamnosus, L. jensenii, L. crispatus, L. fornicalis or L. brevis, which is at least partially consistent with previous reports on the species distribution in other countries [6, 17, 32, 38]. However, other lactobacilli species have been described at a lower frequency, including L. vaginalis, L. fermentum, L. mucosae, L. paracasei and L. rhamnosus, in reports that have indicated the variability among women of specific regions [6, 32, 39, 40]. This trend is consistent with our results, in which L. fermentum, L. rhamnosus and L. brevis were also identified. These findings are not surprising since it has become increasingly apparent in recent years that ethnicity can affect the number and type of organisms present in the vaginal cavity [17, 18, 37].
A previous study published by Hernández-Rodríguez et al. did not identify L. crispatus using a culture-independent method (DGGE-PCR) in samples from pregnant Mexican women, which is in contrast with our study, in which this species was isolated, albeit at a low frequency [17, 18, 38, 40].
The absence of L. iners isolates in our study is significant because it is one of the dominant species reported worldwide. The lack of detection of L. iners could be attributed to the limitations of the methodology used in the current analysis, and must be confirmed in future studies without these limitations (e.g. DGGE of PCR-amplified 16S rRNA fragments can overcome the limitation of traditional cultivation techniques to retrieve the vaginal econiche diversity).
Correct species identification is dependent on the reliability of the reference strains and accuracy of the database used. Because many reference strains were previously characterised using non-genetic methods, it is possible that they were misidentified and, thus, the reference strains themselves may be unreliable . For example, the reference strain previously identified as L. acidophilus should be re-labelled as L. jensenii based on the genetic data obtained in the present study (Table 1). Both species belong to the L. acidophilus group and are genetically related.
The majority of the strains isolated in this study were identified as species of the L. acidophilus group. Because Lactobacillus species are considered to be critical for protection against pathogens in the female genital tract, this set of strains could be useful in future studies on probiotic properties , especially given the potential differences in the protective capabilities of vaginal Lactobacillus species.
The intraspecific variability among the L. gasseri and L. fermentum species, detected by ERIC-PCR analysis, concurs with a previous report by Stephensen et al. . This previous study showed that ERIC-PCR analysis was capable of typing Lactobacillus isolates at the strain level.
The main limitations of this study were the small sample size and problems associated with culture-dependent methods. In addition, because of the small number of isolates, it was not possible to correlate the observed species with a healthy or unhealthy (BV) status. Our results should be corroborated using a larger cohort and culture-independent methods (DGGE or sequencing of cloned 16S rRNA molecules) to describe the relative abundance of the species described herein. Despite these limitations, the results of this study concur with previously published findings showing that L. crispatus, L. gasseri and L. jensenii are consistently present in the healthy vaginal ecosystem, and provide additional information regarding the Mexican (Hispanic-mestizo) non-pregnant population [6, 17, 32, 38, 40, 42]. Additional species or phylotypes not common in other countries were found in this study, which furthers our understanding of vaginal colonisation by lactobacilli, and the Lactobacillus species diversity in vaginal communities in the mestizo population.