Glycogen availability and pH variation in a medium simulating vaginal fluid influence the growth of vaginal Lactobacillus species and Gardnerella vaginalis

Abstract

Background: Glycogen metabolism by Lactobacillus spp. that dominate the healthy vaginal microbiome contributes to a low vaginal pH (3.5-4.5). During bacterial vaginosis (BV), strict and facultative anaerobes including Gardnerella vaginalis become predominant, leading to an increase in the vaginal pH (> 4.5). BV enhances the risk of obstetrical complications, acquisition of sexually transmitted infections, and cervical cancer. Factors critical for the maintenance of the healthy vaginal microbiome or the transition to the BV microbiome are not well defined. Vaginal pH may affect glycogen metabolism by the vaginal microflora, thus influencing the shift in the vaginal microbiome.

Results: The medium simulating vaginal fluid (MSVF) supported growth of L. jensenii 62G, L. gasseri 63 AM, and L. crispatus JV-V01, and G. vaginalis JCP8151A at specific initial pH conditions for 30 d. L. jensenii at all three starting pH levels (pH 4.0, 4.5, and 5.0), G. vaginalis at pH 4.5 and 5.0, and L. gasseri at pH 5.0 exhibited the long-term stationary phase when grown in MSVF. L. gasseri at pH 4.5 and L. crispatus at pH 5.0 displayed an extended lag phase over 30 d suggesting inefficient glycogen metabolism. Glycogen was essential for the growth of L. jensenii, L. crispatus, and G. vaginalis; only L. gasseri was able to survive in MSVF without glycogen, and only at pH 5.0, where it used glucose. All four species were able to survive for 15 d in MSVF with half the glycogen content but only at specific starting pH levels - pH 4.5 and 5.0 for L. jensenii, L. gasseri, and G. vaginalis and pH 5.0 for L. crispatus.

Conclusions: These results suggest that variations in the vaginal pH critically influence the colonization of the vaginal tract by lactobacilli and G. vaginalis JCP8151A by affecting their ability to metabolize glycogen. Further, we found that L. jensenii 62G is capable of glycogen metabolism over a broader pH range (4.0-5.0) while L. crispatus JV-V01 glycogen utilization is pH sensitive (only functional at pH 5.0). Finally, our results showed that G. vaginalis JCP8151A can colonize the vaginal tract for an extended period as long as the pH remains at 4.5 or above.

Keywords: Gardnerella vaginalis; Glucose; Glycogen; Lactobacillus crispatus; Lactobacillus gasseri; Lactobacillus jensenii; Medium simulating vaginal fluid; pH.

Fig. 1

Growth patterns and changes in pH of G. vaginalis JCP8151A grown in MSVF and NYCB. G. vaginalis was grown overnight in NYCB at pH 7.3; cells were pelleted, washed, and resuspended in MSVF at pH 4.0, pH 4.5, and pH 5.0. One-mL aliquots were pipetted into the wells of a 24-well microtiter plate and inoculated with ~ 10⁴ CFU. The plates were sealed with breathable membrane and the cultures were incubated at 37 °C under 5% CO₂ in a humid chamber for 30 dpi. Samples were taken at 1-d intervals through 4 dpi and then every 2 d over the 30-d growth cycle and the CFU/mL and pH were determined. (a) Viability of G. vaginalis grown in MSVF for 30 d at starting pH of 4.0, 4.5, and 5.0; (b) pH of MSVF and NYCB at each time point throughout the growth cycle; (c) viability of G. vaginalis grown in NYCB for 30 d at starting pH of 4.0, 4.5, and 5.0. Each symbol represents the mean of three independent experiments ± SOM. Dotted lines (a, c) indicate starting CFU/mL. Arrowheads (b) indicate loss of viability. Two-way ANOVA with Tukey’s multiple comparisons posttest was done to determine significant differences between time points across the growth curves. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001

Fig. 2

Growth patterns and changes in pH of L. jensenii 62G grown in MSVF and MRSB. L. jensenii was grown overnight in MRSB at pH 6.3 and processed as described in Fig. 1. (a) Viability of L. jensenii grown in MSVF for 30 d at starting pH of 4.0, 4.5, and 5.0; (b) pH of MSVF and MRSB at each time point throughout the growth cycle; (c) viability of L. jensenii grown in MRSB for 30 d at starting pH of 4.0, 4.5, and 5.0. Each symbol represents the mean of three independent experiments ± SOM. Dotted lines (a, c) indicate starting CFU/mL. Arrowheads (b) indicate loss of viability. Two-way ANOVA with Tukey’s multiple comparisons posttest was done to determine significant differences between time points across the growth curves. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001

Fig. 3

Growth patterns and changes in pH of L. gasseri 63 AM grown in MSVF and MRSB. L. gasseri was grown overnight in MRSB at pH 6.3 and processed as described in Fig. 1. (a) Viability of L. gasseri grown in MSVF for 30 d at starting pH of 4.0, 4.5, and 5.0; (b) pH of MSVF and MRSB at each time point throughout the growth cycle; (c) viability of L. gasseri grown in MRSB for 30 d at starting pH of 4.0, 4.5, and 5.0. Each symbol represents the mean of three independent experiments ± SOM. Dotted lines (a, c) indicate starting CFU/mL. Arrowheads (b) indicate loss of viability. Two-way ANOVA with Tukey’s multiple comparisons posttest was done to determine significant differences between time points across the growth curves. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001

Fig. 4

Growth patterns and changes in pH of L. crispatus JV-V01 grown in MSVF and MRSB. L. crispatus was grown overnight in MRSB at pH 6.3 and processed as described in Fig. 1. (a) Viability of L. crispatus grown in MSVF for 30 d at starting pH of 4.0, 4.5, and 5.0; (b) pH of MSVF and MRSB at each time point throughout the growth cycle; (c) viability of L. crispatus grown in MRSB for 30 d at starting pH of 4.0, 4.5, and 5.0. Each symbol represents the mean of three independent experiments ± SOM. Dotted lines (a, c) indicate starting CFU/mL. Arrowheads (b) indicate loss of viability. Two-way ANOVA with Tukey’s multiple comparisons posttest was done to determine significant differences between time points across the growth curves. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001

Fig. 5

All strains require glycogen for growth except L. gasseri 63 AM, which uses glucose at pH 5.0. Strains were grown for 15 d in standard MSVF, which contains 10 g/L of glycogen and 10 g/L of glucose, MSVF with half the level of glycogen (MSVF_5Gly) at (a) pH 4.0, (b) pH 4.5, and (c) pH 5.0, and MSVF with no glycogen (MSVF_0Gly) at (d) pH 4.0, (e) pH 4.5, and (f) pH 5.0 Samples were taken at 5-d intervals and the CFU/mL were determined. Each symbol represents the mean of three independent experiments ± SOM. Two-way ANOVA with Tukey’s multiple comparisons posttest was done to determine significant differences between time points across the growth curves. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001

Fig. 6

 L. gasseri 63 AM utilizes alternative carbon sources within MSVF pH 5.0. Growth of L. gasseri in standard MSVF was compared to its growth in MSVF_0Gly, MSVF without glycogen or glucose (MSVF_0Gly0Glu), and without glycogen, glucose, or mucin (MSVF_0Gly0Glu0Muc) over 15 d, or until viability was lost. Each symbol represents the mean of three independent experiments ± SOM. Two-way ANOVA with Šídák’s multiple comparisons posttest was done to determine significant differences between time points across the growth curves; ****, P < 0.0001

Fig. 7

 L. jensenii 62G and G. vaginalis JCP8151A are metabolically active during the long-term stationary phase. (a) Strains were grown in MSVF at pH 5.0 for 6 and 12 dpi and carbenicillin sufficient to inhibit growth (64 µg/mL for L. jensenii or 1 µg/mL for G. vaginalis) was added. Incubation was continued for an additional 24 h and samples were taken at 7 and 13 dpi and the CFU/mL were determined. (b) Addition of exogenous glycogen to MSVF pH 5.0 at intervals during the long-term stationary phase did not affect growth of L. jensenii or G. vaginalis. A concentrated solution of glycogen in water to a total of 2.5 mg/mL was added to cultures of the strains in MSVF at pH 5.0 at 6 or 12 dpi. Incubation was continued for an additional 24 h and samples were taken at 7 and 13 dpi and the CFU/mL were determined. For both panels: each symbol represents the mean of three independent experiments ± SOM. Two-tailed unpaired t-tests were used to determine significant differences between treatments; ns, no significance; ****, P < 0.0001