Distinct Functional Traits of Lactobacilli from Women with Asymptomatic Bacterial Vaginosis and Normal Microbiota

Asymptomatic bacterial vaginosis (BV) in reproductive-age women has serious obstetric and gynecological consequences. Despite its high incidence, the behavior of vaginal lactobacilli in asymptomatic BV is unknown. We analyzed the functional properties of previously isolated vaginal lactobacilli from asymptomatic women with normal, intermediate, and BV microbiota. Lactic acid and antimicrobial activity against seven urogenital pathogens were evaluated from lactobacilli cell-free culture supernatants (CFCs) (n = 207) after 48 h incubation in MRS. Lactobacilli isolates were used to evaluate H2O2, autoaggregation and coaggregation with C. albicans. Lactobacilli from normal microbiota produced more d-lactate than lactobacilli from intermediate and asymptomatic BV (p = 0.007). L. plantarum, L. fermentum and L. reuteri produced greater d-lactate whereas L. rhamnosus, L. crispatus, L. johnsonii were greater producers of l-lactate. Interspecies positive correlation was observed in the lactic acid contents of CFCs. Distribution of H2O2-producing lactobacilli did not vary significantly among the groups. When lactic acid isomers were considered, species from intermediate and BV microbiota clustered together with each other and distinctly from species of normal microbiota. Broad-spectrum antagonism (≥90% inhibition) against E. coli, C. albicans, S. aureus, P. aeruginosa, G. vaginalis, N. gonorrhoeae, S. agalactiae were displayed by 46.86% (97) of isolates. Our study highlights the differential functional properties of vaginal lactobacilli from women with normal microbiota and asymptomatic BV.


Introduction
Vaginal microbiota are one of the critical features in maintaining vaginal homeostasis and providing protection against urogenital infections. Dysbiosis of the vaginal microbial composition could lead to bacterial vaginosis (BV) and other associated infections such as vulvovaginal candidiasis, trichomoniasis, and sexually transmitted infections (STIs) [1][2][3]. Bacterial vaginosis is the most common reproductive tract infection in women which could be symptomatic (with symptoms) or asymptomatic (without manifestation of clinical symptoms) [4]). Apart from symptomatic cases, incidences of asymptomatic BV can be 14.5% to as high as 84% [4][5][6]. Both symptomatic and asymptomatic BV has been associated with severe obstetrics' and gynecological consequences [7][8][9][10]. Asymptomatic BV has been identified as a risk factor for late miscarriage and preterm delivery [8], and persistent infection of HPV [11]. Despite the increased risk of other infections, reproductive morbidity, and complications, asymptomatic BV is understudied.
We previously reported that vaginal microbiota of asymptomatic BV are different from healthy microbiota. Predominant Lactobacillus communities in eubiosis were L. iners, L. crispatus, L. reuteri, L. jensenii, and L. gasseri while women with asymptomatic BV harbored L. iners, L. rhamnosus, L. reuteri, L. salivarius and L. johnsonii [6]. Lactobacilli in the vaginal ecosystem play a protective role by limiting the growth, proliferation, and colonization of pathogens. These beneficial microbes contribute to the control of infections by producing antimicrobial compounds which include organic acids [12,13], hydrogen peroxide [14], bacteriocins and biosurfactants [15]. The presence of H 2 O 2 -producing lactobacilli strains during pregnancy has been associated with reduced risk of BV and adverse gynecological consequences [16]. Besides, lactobacilli can compete with pathogens for adherence to vaginal epithelial cells and prevent their colonization [17]. The mechanism of competitive exclusion could be due to the coaggregation of lactobacilli with the pathogenic microorganisms thus hindering the adherence and colonization of pathogens on the vaginal epithelium [18]. Another exclusion mechanism could be due to autoaggregation where Lactobacillus can form multi-cellular aggregates with bacteria from the same species and their adherence to epithelial cells and surfaces of mucus creates a barrier for pathogens [19].
Thus, Lactobacillus plays a protective role in the vaginal microenvironment, and reduction in their abundances and diversity leads to dysbiosis [6]. However, it is uncertain if the probiotic properties of lactobacilli, inhabiting the vaginal tract, vary during eubiosis and dysbiosis. So, we aimed to study the probiotic functional properties of lactobacilli isolated from normal and BV microbiota of asymptomatic women.

Isolation of Lactobacilli
In this study, 207 Lactobacillus isolates were previously recovered from 145 healthy premenopausal, regularly menstruating participants of reproductive age asymptomatic for any vaginal complaints [6].  [20]. We adopted the standardized 0-10 point Nugent scoring system based on identifying proportions of three bacterial morphotypes on Gram-stained vaginal smears: large Gram-positive rods (Lactobacillus spp.), small Gram-negative or Gram-variable coccobacilli (Gardnerella and anaerobic spp.), and curved Gram-variable rods (Mobiluncus spp.). A score of 0-3 was considered normal, 4-6 was intermediate, and scores ≥7 indicated BV. The lactobacilli were isolated from 116 women with normal microbiota, 11 with intermediate microbiota, and 18 women with asymptomatic BV. All the Lactobacillus strains had been isolated as colonies on MRS (deMan Rogosa and Sharpe) (HiMedia, Mumbai, India) agar medium after incubation for 24 to 48 h in a candle jar. The isolates further grown in MRS broth (HiMedia, Mumbai, India) were stored as glycerol stocks (20%) at −80 • C. For the present study, glycerol stocks of the 207 lactobacilli isolates were streaked on MRS agar and incubated at 37 • C for 48 h in a candle jar.

Preparation of Cell-Free Culture Supernatants
Colonies of lactobacilli obtained on MRS agar were inoculated in MRS broth followed by incubation at 37 • C for 48 h in a candle jar. About 100 µL of lactobacilli cultures was used to measure bacterial growth at 600 nm using Synergy H1 Hybrid Multi-mode reader (BioTek Instruments, Winooski, VT, USA). Cell-free culture supernatants (CFCs) were obtained by centrifuging the culture medium at 10,000× g for 10 min at 4 • C. Centrifuged CFCs were passed through a sterile Millex GS filter unit (0.22 µm) (Millipore, Darmstadt, Germany). The pH of CFCs was measured using HiIndicator TM pH papers (Himedia, Mumbai, India).

Determination of Lactic Acid
Briefly, 100 µL of lactobacilli CFCs was used to quantify d-/l-and total lactic acid using the d-/l-Lactic Acid test kit (Megazyme, K-DLATE, Co. Wicklow, Ireland) according to the manufacturer's instructions. The amount of d-lactate and l-lactate was calculated after taking the absorbance at 340 nm using a microplate reader (BioTek Instruments, Winooski, VT, USA).

Determination of Hydrogen Peroxide
The Lactobacillus isolates were plated onto MRS agar containing horseradish peroxidase (HRP) (Sigma-Aldrich, St Louis, MO, USA) and 3,3 ,5,5 -tetramethylbenzidine (TMB) (Sigma-Aldrich, St Louis, MO, USA) using the protocol mentioned elsewhere [21]. After incubation, colonies were exposed to ambient air. Colonies turning to blue color were considered H 2 O 2 producers. Depending on the intensity of blue, isolates were graded semi-quantitatively as strong, moderate, and weak producers.

Semi-Quantitative Estimation Lactobacilli Agglutination with C. albicans
Slide agglutination of a clinical C. albicans with lactobacilli was carried out as per a previous protocol with slight modifications [22]. Briefly cells were washed and resuspended in sterile PBS to achieve 1.0 OD 600 nm . About 20 µL of previously isolated vaginal C. albicans CA119 (Pramanick et al., 2019) suspension was added to 20 µL of Lactobacillus cell suspension, along with 5 µL of crystal violet. Visual and microscopic agglutination was observed after 10 min of incubation.

Quantitative Evaluation of Coaggregation and Autoaggregation
Overnight cultures of lactobacilli and C. albicans CA119 grown in MRS and Sabourauds broth (HiMedia, Mumbai, India), respectively, were centrifuged at 10,000× g for 10 min, followed by washing the cell pellet twice with sterile PBS. After washing cells were resuspended in sterile PBS and adjusted to 0.1 OD 620 nm . For coaggregation assay, 100 µL each of Lactobacillus and C. albicans cell suspensions were added on a 96-well microplate. For autoaggregation, only 200 µL of Lactobacillus was added to each well. After 3 h of incubation at 37 • C, 100 µL of the upper supernatant was removed and OD 620 nm was recorded.
Coaggregation and autoaggregation were calculated using the following equations: A 0 refers to the initial OD 620 nm taken immediately after both the cultures were mixed. A s refers to the OD 620 of the supernatant after 3 h incubation.
A 0 refers to the initial OD 620 , and A s refers to the OD 620 determined after 3 h [23].
G. vaginalis, N. gonorrhoeae and S. agalactiae were grown in Brain-heart infusion (BHI) broth (HiMedia, Mumbai, India). Mueller Hinton (MH) broth (HiMedia, Mumbai, India) was used to grow E. coli, S. aureus, and P. aeruginosa and Sabouraud's broth (HiMedia, Mumbai, India) was used to culture C. albicans. The overnight grown cultures of pathogens were, washed and resuspended in phosphate-buffered saline (PBS) to obtain 0.1 OD 600 nm cell suspension. The pathogen cell suspensions were further diluted 1:100 in the respective growth medium to prepare the inoculum. In a 96-well microtiter plate containing 100 µL lactobacilli CFCs, 100 µL pathogen inoculum was added into each well. Media control with no microbial culture and a positive control containing pathogen inoculum and MRS broth were used as negative control and positive control for growth. CFCs of each pathogen were used to determine the effect of any nutrient exhaustion in the metabolites. The plates were incubated under anaerobic conditions at 37 • C for 48 h. The growth of the cultures was measured at OD 600 nm after 4 h, 18 h, 24 h, and 48 h incubation. Percent growth inhibition was calculated using the following formula, % Growth = [(Positive control OD 600 − Test well OD 600 )/Positive control OD 600 ] * 100 % Inhibition = 100 − % Growth (4)

Statistical Analysis
Data were analyzed using the Graphpad Prism 8.0.1 software. Results are the mean of experimental readings taken at least in duplicates. Lactic acid concentrations are represented in mM. Readings are expressed as mean ± std.deviations. Data of qualitative variable H 2 O 2 was analyzed using the chi-square test and, Fisher's exact test. Other variables were analyzed using Kruskal-Wallis one-way analysis of variance followed by pairwise multiple comparisons using Dunn's multiple comparison test. Spearman's rank correlation was used to determine the relationship between the variables. Statistical significance was considered at p < 0.05.

Lactobacillus Isolates Diversity
For this study, 207 lactobacilli from 145 reproductive age women with no vaginal complaints and infections were assessed. These samples were previously classified as normal microbiota (116), intermediate (11)

Acidification of Medium
Lactobacilli in the vaginal milieu determine the vaginal pH. Herein, we evaluated the pH-buffering capacity of lactobacilli isolated from normal, intermediate, and BV microbiota in growth medium. The maximal OD600 nm of lactobacilli cultures in MRS ranged from 0.7 to 0.9. The CFCs of lactobacilli from normal, intermediate, and BV microbiota had an average pH of 4.17, 4.27, 3.96, respectively, with no significant difference (p = 0.4). The acidifying ability of Lactobacillus species differed significantly from each other ( Figure 1). The average pH of L. plantarum, L. fermentum, L. rhamnosus, L. johnsonii CFCs were below 4. L. acidophilus, L. delbrueckii, and L. vaginalis isolates acidified the medium poorly. Acidifying potential of different isolates of the same species was similar regardless of vaginal microbiota ( Figure S1).

Species-Specific Lactic Acid Production
Our studies showed significant difference in lactic acid isomer production from lactobacilli isolated from the three groups. Hence, we further investigated whether the production of lactic acid and its isomers could be attributed to a particular species. L. plantarum isolates produced the highest lactic acid in the medium (74.11 ± 15.72) followed by L. johnsonii  (Table S1).

Intraspecies Comparison of Lactic Acid Production from Different Microbiota
We further investigated whether there was a strain-to-strain difference among the major lactobacillus isolates from the eubiotic and dysbiotic vaginal state. Metabolites of L. jensenii isolates from healthy samples had significantly higher d-/l-lactic acid ratios as compared to L. jensenii isolated from intermediate and BV samples (p = 0.04) (Figure 4a), but the sample size is small. Though not statistically significant, metabolites of other major Lactobacillus species from healthy samples had higher d-/l-lactic acid ratio than isolates of same spp. from other groups (Figure 4b-f).

Species-Specific Correlation of Lactic Acid
Normal vaginal microbiota have a heterogeneous lactobacilli population. Our previous studies had shown that majority of normal microbiota harbored at least two Lactobacillus species simultaneously [6]. Heterogeneity of lactobacilli population is reduced during dysbiosis (6). To determine the plausible synergistic effect of different Lactobacillus species with each other, we did a correlation analysis of lactic acid produced by different Lactobacillus isolates from normal microbiota. The concentrations of d-, l-Lactic acid isomers in metabolites of certain Lactobacillus species positively correlated with each other while few others negatively correlated (Figure 5a-h).

Coaggregation of Lactobacillus with C. albicans
About 119 (70.41%), 9 (60%) and 13 (56.5%) lactobacilli from normal, intermediate and BV microbiota, respectively, could agglutinate C. albicans. Agglutination of C. albicans by lactobacilli was further semi-quantified as strong, medium, and weak based on the size of aggregates ( Figure S2 (Table S1). The % Candida coaggregating ability of lactobacilli when evaluated quantitatively were not statistically significant among the groups as well as among the species (Figure 6c,d).

Antagonistic Effect on Pathogens
Growth of pathogens was tested after incubation with lactobacilli metabolites at different time intervals (4 h, 18 h, 24 h). Data represented is of 24 h growth inhibition, since there were no significant differences between readings of 18 h and 24 h. The pathogen inhibitory effects of lactobacilli were similar across the three groups for all the pathogens ( Figure S3). Growth of S. agalactiae was inhibited up to 30% when grown in its own CFCs.

Association of Probiotic Properties from Different Microbiota
Hierarchical clustering analysis showed different clustering of patterns of lactic acid in normal microbiota compared to intermediate and BV groups ( Figure S4a-c). In BV microbiota, d-lactic acid clustered separately from total lactic acid and l-lactic acid ( Figure S4c).
We further performed hierarchical clustering of lactic acid isomers produced by Lactobacillus species from normal, intermediate and BV microbiota. We could see the majority of lactobacilli species from intermediate and asymptomatic BV clustered together rather than with the species from normal microbiota ( Figure 8). L. jensenii, L. johnsonii from normal microbiota clustered distinctly from L. jensenii, L. johnsonii, respectively, from intermediate and asymptomatic BV microbiota (Figure 8).
Inhibition of the pathogens was positively correlated to pH and lactic acid content of metabolites. The inhibitory effects on S. aureus, C. albicans and N. gonorrhoeae were positively correlated to total and l-lactic acid present in the CFCs. Inhibition of the pathogens positively correlated to the l-lactic acid amount (Figure 9). Coaggregating and autoaggregating ability of lactobacilli isolates positively correlated with the growth inhibition of C. albicans.

Discussion
The efficacy of asymptomatic BV treatment with antibiotics is often debatable [7,24]. With no risk of antimicrobial resistance, the use of probiotic lactobacilli as a prophylaxis for symptomatic BV appears coherent. Asymptomatic BV is also highly prevalent in women, associated with a lactobacilli-deficient condition [6,25] and an increased risk of infection due to disruption of the vaginal epithelium [26]. However, it is largely unknown whether the lactobacilli present during eubiosis have different functional properties than lactobacilli during asymptomatic BV. To decipher the probiotic properties of 207 lactobacilli from contrasting vaginal niche, we examined their antimicrobial metabolites and antagonistic potential towards various urogenital pathogens.
Acidification is the primary mechanism by which Lactobacillus protects the vaginal microenvironment from pathogenic bacteria [13,27]. In this study, the total lactic acid in lactobacilli metabolites did not vary significantly between the groups. These observations are in contrast to previous reports on lactic acid from eubiotic and dysbiotic microbiota [12,28]. This discrepancy could be due to the detection of lactic acid from cervicovaginal fluid samples in the earlier reports and the use of axenic cultures in the present study. This implies that during asymptomatic BV, which is characterized by reduction in Lactobacillus abundance and diversity, the remaining lactobacilli strains are capable of producing lactic acid when suitable environmental conditions are provided. d-lactic acid, which is an isomer of l-lactic acid, has a greater protective role than l-lactic acid [13]. We report remarkably higher levels of d-LA and high d/l lactic acid ratios in metabolites of lactobacilli from normal microbiota compared to those from asymptomatic BV. d-lactic acid is exclusively contributed by the bacteria, whereas l-lactate is produced both by the bacteria and vaginal epithelial cells [29]. Recently higher d-lactic acid was reported in axenic cultures and cervicovaginal mucus samples from women with normal microbiota [30,31]. Thus, presence of d-lactic acid producing lactobacilli in the vaginal milieu is more important than any lactic acid producing species. We found isolates of L. fermentum, L. plantarum, L. reuteri were highest d-LA producers and L. rhamnosus, L. crispatus, L. johnsonii were the highest l-LA producers and. Furthermore, L. jensenii isolates from normal microbiota produced higher d-lactic acid than L. jensenii from intermediate and asymptomatic BV. Studies have associated L.jensenii with normal microbiota [32,33] and presence of d-lactic acid producing L. jensenii suggest protective effect of the species to maintain vaginal homeostasis. Thus, though lactobacilli strains from asymptomatic BV could acidify the medium and produce total lactic acid comparable to lactobacilli from normal microbiota, they could not produce an equivalent amount of d-lactic acid like their counterparts from normal microbiota.
In healthy vaginal microbiota, which is characterized by a heterogeneous Lactobacillus population, the protective outcome of these lactobacilli could be the result of their synergistic effects [6]. We noted certain lactobacilli species positively correlated in lactic acid and its isomer production. Additionally, the correlation of lactic acid concentrations produced by L. gasseri and L. reuteri strains from normal microbiota differed from dysbiotic microbiota. This difference between normal and dysbiotic microbiota in lactobacilli co-species relation indicates the presence of different strains of species. Thus, the interaction and relation among the Lactobacillus species may differ in a healthy environment and perturbed condition like BV. It will be interesting to note the composition of the metabolites when co-cultured. Such synergistic efficacy among the species forms the basis of selection of a multistrain consortia instead of a single probiotic strain to restore vaginal homeostasis.
Hydrogen peroxide is another metabolite of lactobacilli reported to have antimicrobial potential and immunomodulatory effect [33]. Studies have associated the presence of H 2 O 2 -producing lactobacilli with normal microbiota [34,35]. We did a semi-quantitative evaluation of H 2 O 2 on the MRS-TMB-HRP medium, due to its instability in CFCs. We observed no significant difference in H 2 O 2 -producing lactobacilli between normal and dysbiotic microbiota. We did a qualitative or semi-quantitative estimation of H 2 O 2 from the lactobacilli on solid medium. However, a quantitative estimation of hydrogen peroxide in liquid medium probably would confirm whether there was any difference in the production of H 2 O 2 by lactobacilli isolated from normal, intermediate and BV microbiota.
Adherence to vaginal epithelial cells is another mechanism by which lactobacilli can colonize and sustain in the vaginal microenvironment. Autoaggregation of lactobacilli is an indicator of adherence ability and biofilm formation on host mucosa [36,37]. We observed species-specific lactobacilli self-aggregation and no significant difference among the microbiota groups. This observation was not surprising, because strains with poor adherence properties will not be able to colonize and will be lost from the vaginal microbiota and not recovered from the vaginal samples. Isolates of L. crispatus, L. fermentum, L. acidophilus, and L. delbrueckii were not present in women with asymptomatic BV [6]. A compromised adhering strength could be one of the plausible reasons for their absence from asymptomatic BV. Among the species that were recovered, isolates of L. gasseri and L. johnsonii had better autoaggregating abilities than isolates of L. fermentum. Moreover, L. johnsonii, L. gasseri, and L. salivarius displayed better autoaggregation than other species.
Coaggregation of lactobacilli with pathogens results in the physical elimination of pathogens (Pino, 2019). In this study, isolates of L. acidophilus, L. johnsonii and L. rhamnosus demonstrated the commendable coaggregating strength of C. albicans. Earlier L. crispatus was reported to have the highest C. albicans aggregating property [38]. Similar to autoaggregation, the C. albicans coaggregating property of lactobacilli did not vary among the groups. Desirable probiotic lactobacilli should have a strong adhesive strength for efficient pathogen displacement [18]. C. albicans is the main causative agent for vulvovaginal candidiasis [39], which is the second most gynecological infection of women in reproductive age [40]. Exploring these Candida coagglutinating lactobacilli strains seems a promising probiotic strategy against vulvovaginal candidiasis.
We further tested the antimicrobial effects of vaginal lactobacilli from normal and dysbiotic microbiota on the major urogenital pathogens causing aerobic vaginitis and urinary tract infections (E. coli, S. aureus, P. aeruginosa), bacterial vaginosis (G. vaginalis), vulvovaginal candidiasis (C. albicans), gonorrhea (N. gonorrhoeae) and preterm birth (S. agalactiae) [41]. Studies have reported lactic acid as the key lactobacilli defense factor against various pathogens [12,28,42]. Since total lactic acid did not vary between lactobacilli from normal and dysbiotic microbiota, their antagonistic activity might have remained similar.
Nearly half of the isolates exhibited broad-spectrum (>90% inhibition) antimicrobial activity and about three quarters of these lactobacilli belonging to normal microbiota predominantly consisted of L. rhamnosus, L. reuteri and L. gasseri. Recently, L. rhamnosus from vaginal samples have been reported to demonstrate broad-spectrum antagonistic activity [43]. Different studies have demonstrated the antimicrobial effect of selected lactobacilli on some pathogens [44,45]. However, an extensive evaluation of the probiotic properties of such a large number of lactobacilli was not reported earlier.
We observed a positive correlation of lactobacilli and C. albicans coaggregation with inhibition of C. albicans by lactobacilli metabolites. Besides lactic acid, metabolites of lactobacilli contain exopolysachharides (EPS) and membrane vesicles (MVs). Both EPS and MVs from lactobacilli can affect the adhesion of C. albicans on host epithelial cells [46]. Exoploysachharides present in lactobacilli metabolites have been reported to affect the growth of C. albicans by extending its lag phase. Thus, lactobacilli with higher C. albicans coaggregating ability may have a better growth inhibitory effect on C. albicans due to the presence of exopolysachharides.
Since lactic acid isomers were the quantitative variable that differed between the groups, we further analyzed the clustering pattern of the species. The species picked from intermediate and asymptomatic BV clustered together and were separated from the species from normal microbiota. This indicates that, though lactobacilli from asymptomatic BV may be similar to lactobacilli from normal microbiota in certain traits, they are different in some properties at the species and strain level.
Our study is limited by the fact that we have not evaluated other functional attributes of Lactobacillus such as other organic acids, bacteriocin, biofilm formation, or coaggregation with other pathogens. Despite being the prevalent vaginal lactobacilli, we were unable to evaluate L. iners due to its inability to grow on MRS [6]. In spite of these shortcomings, we could depict a broad view of the distinct functional traits of vaginal lactobacilli from normal, intermediate and asymptomatic BV microbiota which was not reported earlier with such an extensive sample size. It will be interesting to note the behavior of these lactobacilli along with other microbial co-members to further decipher their potential. Additionally, other functional traits such as their anticancer effects needs to be explored. The meritorious shortlisted lactobacilli will be evaluated for these traits in the future.

Conclusions
Numerous studies have reported the diversity of vaginal microbiota, but the functional aspects of lactobacilli, especially in asymptomatic BV, was overlooked. Our findings suggest that asymptomatic BV is characterized by the absence of beneficial strains of lactobacilli and should not be left untreated. A carefully selected microbial consortium of lactobacilli should be considered for the treatment of asymptomatic cases of BV to prevent any future episodes of other RTIs and STIs.  Table S1: Functional properties of different species of Lactobacillus used in the study. Kruskal-Wallis test was used to determine statistical significance.