Saccharomyces cerevisiae-based probiotic as novel anti-microbial agent for therapy of bacterial vaginosis

Abstract

In this study, we demonstrate, for the first time, that Saccharomyces cerevisiae-based probiotic shows an inhibitory effect on Gardnerella vaginalis infection. This effect is likely due to several actions: direct interference with adherence to vaginal tissues, inhibition of sialidase activity, reduction of vaginal epithelial exfoliation. Gardnerella vaginalis does not induce vaginal inflammation and no inflammatory cytokines were, indeed, produced, by the mouse vagina, neither by Gardnerella vaginalis and by the probiotic. Collectively, our data incite to further investigations on Saccharomyces cerevisiae probiotic as a potential prophylactic or therapeutic agent in the vaginosis caused by Gardnerella vaginalis.

Keywords: G. vaginalis; S. cerevisiae; adherence; bacterial vaginosis; displacement; exfoliation; probiotic; sialidase.

Figure 1.

Effect of GI treatment on G. vaginalis infection. C57/Bl6 mice, under pseudoestrus condition, were treated intravaginally with 10 μl of Saline, or L. crispatus (2 × 10⁹/ml) or GI (10⁸/ml) two days before the challenge with G. vaginalis (5 × 10⁷/20μl/mouse) and once a day for 3 days beginning the day of infection (A). G. vaginalis load were determined by enumerating colony forming units (CFU) in vaginal washes (B), in tissue (D) and uterine horn homogenates (F) at days 1 and 3 post-infection. Percentage of G. vaginalis CFU decrease (C, E, G) was quantified relative to G. vaginalis-infected mice treated with Saline. Data are the mean ± SEM from 2 independent experiments each with 6 mice/group. *p < 0.05 L. crispatus- or GI-treated mice vs Saline-treated mice.

Figure 2.

Effect of GI treatment on G. vaginalis sialidase activity and epithelial exfoliation G. vaginalis-induced. Sialidase activity and epithelial exfoliation were determined in vaginal washes of mice, treated intravaginally with 10 μl of Saline, or L. crispatus (2 × 10⁹/ml) or GI (10⁸/ml) and infected with G. vaginalis (5 × 10⁷ /20 μl/mouse) as described in Materials and Methods, at days +1 and +3 post-infection. (A, B, C) Optical density, of sialidase activity, was determined as described in Materials and Methods. (A, B) Lines are representative of experiments (n = 2) with similar results. (C) Bars are the mean ± SEM from 2 independent experiments each with 6 mice/group. The dashed line represents the optical density of sialidase activity from vaginal washes of not-infected mice. *p < 0.05 L. crispatus- or GI-treated mice vs Saline-treated mice. (D) Epithelial exfoliation score has been evaluated by assigning a value from 0 to 3, with 0 = cells number < 25 and 3 = cells number > 75. The dashed line represents the exfoliation score from vaginal washes of not-infected mice. Data are the mean ± SEM from 2 independent experiments each with 6 mice/group. *p < 0.05 L. crispatus- or GI-treated mice vs Saline-treated mice. (E) Percentage of epithelial exfoliation decrease was quantified in respect to mice treated with Saline.

Figure 3.

Evaluation of immune response to probiotic treatment. (A) Pro-(IL-1β and TNF-α) and anti-(IL-10) inflammatory cytokines levels have been determined in vaginal washes of mice, treated intravaginally with 10 μl of Saline, or L. crispatus (2 × 10⁹/ml) or GI (10⁸/ml) and infected with G. vaginalis (5 × 10⁷/20 μl/mouse), at days +1 and +3 post-infection. Data are the mean ± SEM from 2 independent experiments each with 6 mice/group. (B) Histological inflammation was assessed by haematoxylin-eosin staining of formalin-fixed, paraffin-embedded vaginal tissue sections. Images (Bar = 200 µm, Magnification 10x) are representative of 2 separate experiments with similar results.

Figure 4.

Effect of GI treatment on G. vaginalis adherence to vaginal (A-431) and cervix (HeLa) epithelial cells. (A) Adhesion of L. crispatus or GI to A-431 and to HeLa cell lines. L. crispatus or GI (both 2 × 10⁷/ml or 2 × 10⁸/ml) were added to monolayer of A-431 or HeLa cells for 4 h at 37°C in anaerobic conditions. After incubation, cells were washed 2 times and microorganisms adhered were quantified as number of CFU/ml. Data are the mean ± SEM from 2 independent experiments. *p < 0.05, 2 × 10⁸/ml (L. crispatus or GI) vs 2 × 10⁷/ml (L. crispatus or GI). (B) Interference of L. crispatus or GI on G. vaginalis initial adhesion onto A-431 and HeLa cell lines. Two inocula (2 × 10⁷/ml or 2 × 10⁸/ml) of L. crispatus or GI were pre-adhered to epithelial cells, as above described, and subsequently G. vaginalis (2 × 10⁸/ml) has been added to the co-culture for 30 min at 37°C in anaerobic conditions. G. vaginalis adhered were quantified as number of CFU/ml. *p < 0.05 L. crispatus or GI vs Saline treatment. Percentage adherence inhibition was quantified in respect to Saline. (C) Reduction of G. vaginalis adherent to epithelial cells. G. vaginalis (2 × 10⁸/ml) was incubated with the monolayers for 30 min at 37°C in anaerobic conditions. Then, non-adherent bacteria were removed by washing and probiotics (2 × 10⁷/ml or 2 × 10⁸/ml) were added to co-cultures for 30 min at 37°C in anaerobic conditions. G. vaginalis displacement was expressed as CFU/ml as described in Material and Methods. *p < 0.05 L. crispatus or GI vs Saline treatment.

Figure 5.

Co-aggregation between GI or L. crispatus and G. vaginalis. G. vaginalis or FITC-G. vaginalis (1 × 10⁹/ml) in PBS were mixed with equal volume of L. crispatus or RhB-L. crispatus (1 × 10⁹/ml) or with equal volume of GI or RhB-GI (10⁸/ml). The samples were vortexed for at least 10 sec and incubated in a 24 well plate for 4 h at 37°C under agitation. The suspensions were, then, observed by inversion light microscopy to evaluate the aggregation degree or photographed by fluorescence microscopy. (A) Scores, from 0 (no aggregation) to 4 (maximum aggregation), and mean are shown. Data are from replicate samples of 3 different experiments. (B) Images are representative of 3 different experiments with similar results (Scale Bar = 50 µm, Magnification 20x). BF = bright field; G. vaginalis (GV) = green; L. crispatus (LC) = red.

Figure 6.

Schematic representation of GI mechanism on G. vaginalis infection. GI, by inhibition of G. vaginalis adhesion and by displacement of G. vaginalis adhered to epithelial cells, reduces G. vaginalis vaginal load and its key virulence factors.