Bacterial vaginosis toxins impair sperm capacitation and fertilization

Abstract

Study question: What effect do toxins produced by bacterial vaginosis (BV) bacteria have on sperm function?

Summary answer: BV toxins dysregulate sperm capacitation and intracellular calcium homeostasis, and impair the ability of sperm to fertilize oocytes.

What is known already: In BV, which is linked to infertility, overgrowth of Prevotella and Gardnerella in the vagina is accompanied by elevated concentrations of the toxins lipopolysaccharide (LPS) and vaginolysin (VLY).

Study design, size, duration: This was a laboratory study in which human semen samples were collected from consenting healthy donors with normal semen parameters. Mouse sperm samples were obtained from the caudal epididymis.

Participants/materials, setting, methods: Motile mouse and human sperm were isolated via swim-up and treated under non-capacitating or capacitating conditions. LPS from Escherichia coli was commercially available. VLY was produced by cloning the Gardnerella VLY protein in the ClearColi expression system. Mouse sperm were pre-incubated in IVF medium with LPS or VLY and then co-cultured with ovulated cumulus-oocyte complexes. The effects of LPS and VLY on sperm motility and hyperactivation were assessed with computer-assisted sperm analysis. Effects on viability were assessed by Hoechst staining. Acrosomal exocytosis was assessed in sperm from transgenic Acr-eGFP mice and in human sperm stained with Pisum sativum agglutinin FITC. Intracellular calcium concentration was measured by using the calcium-sensitive dye Fluo-4 AM and fluorescence microscopy. The effects of LPS on sperm from CatSper knockout mice were assessed. Additionally, sperm were treated with a Toll-like receptor 4 (TLR4) antagonist and further exposed to LPS.

Main results and the role of chance: Exposure of mouse sperm to LPS or VLY significantly decreased IVF (P < 0.05). Under capacitating conditions, both toxins initially increased mouse (P < 0.001) and human (P < 0.05) sperm hyperactivation, then significantly decreased sperm motility (P < 0.05), hyperactivation (P < 0.05), and acrosomal exocytosis (P < 0.01). These changes were accompanied by a rapid and irreversible increase in sperm intracellular calcium concentration. Effects of LPS, but not VLY, were prevented by polymyxin B, which binds LPS. The LPS-induced intracellular calcium increase required external calcium, but not the calcium channel CatSper, and was inhibited by a TLR4 antagonist.

Large scale data: N/A.

Limitations, reasons for caution: First, the commercially available LPS we used was isolated from Escherichia coli, rather than from the BV-associated bacteria Prevotella bivia. Second, we did not quantify the absolute sperm intracellular calcium concentration before or after LPS or VLY treatment. Third, all of our experiments were in vitro.

Wider implications of the findings: These studies suggest that BV-associated toxins contribute to infertility, in part, by impairing sperm capacitation and reducing their fertilizing ability.

Study funding/competing interest(s): This work was supported by the National Institutes of Health (grant number R01 HD069631 to C.M.S.). The authors declare that they have no conflict of interest.

Keywords: CatSper; acrosomal exocytosis; bacterial vaginosis; hyperactivation; infertility; lipopolysaccharide; sperm capacitation; toll-like receptor 4; vaginolysin.

Figure 1.

Lipopolysaccharide (LPS) and vaginolysin (VLY) impair mouse sperm fertility, hyperactivation, total motility, and acrosomal exocytosis in capacitating (CAP) conditions. Percentage of oocytes that reached the two-cell stage within 24 h after IVF using sperm previously treated for 3 h with 1 µg/ml (A) LPS (n = 5 biological replicates) or (B) VLY (n = 5 biological replicates) were calculated. CASA measurements were obtained for (C, D) hyperactivated motility (HA), (E, F) total motility, and (G, H) progressive motility of mouse sperm incubated under CAP conditions in the presence and absence of 1 µg/ml (C, E, G) LPS (n = 10 biological replicates) or (D, F, H) VLY (n = 9 biological replicates). Here, 0 min is the timepoint of bovine serum albumin (BSA)+sodium bicarbonate addition to initiate in vitro capacitation. Acrosomal exocytosis (AE) was quantified in non-capacitating (NC), and CAP sperm (200 each, from five biological replicates) incubated with (I) LPS or (J) VLY, for 0.5 and 1.5 h. Similarly, AE induced by 15 µM A23187 (A23) or 50 mM potassium chloride (KCl), was also quantified in CAP sperm. Sperm viability for NC and CAP sperm (n ≥ 3 biological replicates) was analyzed in the presence of 1 µg/ml (K) LPS or (L) VLY (10000 sperm each). Data are presented as mean and SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.001, ns, non-significant. For (A, B), a paired t-test was used. For (C–L), two-way ANOVA with Bonferroni’s multiple comparison test was used.

Figure 2.

Effects of lipopolysaccharide (LPS) and vaginolysin (VLY) on mouse sperm are specific. CASA measurements were obtained for hyperactivated motility (HA) of mouse sperm incubated under capacitating (CAP) conditions in the presence of (A) 1 µg/ml LPS or 100 µg/ml polymyxin B (PMB)+1 µg/ml LPS, (B) CAP media or 100 µg/ml PMB alone, (C) CAP media, 1 µg/ml VLY, or VLY empty vector control, and (D) 1 µg/ml VLY or 100 µg/ml PMB+1 µg/ml VLY. Here, 0 min is the timepoint of bovine serum albumin (BSA)+sodium bicarbonate addition to initiate in vitro capacitation. (E) Percentage of oocytes that reached the two-cell stage within 24 h after IVF using sperm (n = 4 biological replicates) previously treated for 3 h without LPS, or with 1 µg/ml LPS or 100 µg/ml PMB+1 µg/ml LPS, were also calculated. Data are presented as mean and SD (n = 4 biological replicates for all experiments). *P < 0.05, ****P < 0.0001, ns, non-significant. Data were analyzed by two-way ANOVA with Bonferroni’s multiple comparison test.

Figure 3.

Lipopolysaccharide (LPS) and vaginolysin (VLY) impair human sperm hyperactivation, total motility, progressive motility, and acrosomal exocytosis in capacitating (CAP) conditions. CASA measurements were obtained for (A, B) hyperactivated motility (HA), (C, D) total motility, and (E, F) progressive motility of human sperm incubated under CAP conditions in the presence and absence of 0.1 µg/ml (A, C, E) LPS (n = 12 biological replicates for all experiments) or (B, D, F) VLY or VLY vector control (n = 12 biological replicates for all experiments). Zero minute is the timepoint of bovine serum albumin (BSA)+sodium bicarbonate addition to initiate in vitro capacitation. Acrosomal exocytosis (AE) was quantified in non-capacitating (NC), and CAP sperm (200 each, n = 3 biological replicates for G, H) incubated with (G) LPS or (H) VLY for 0.5 and 3.5 h. Similarly, AE induced by 10 µM A23187 (A23) or progesterone (P4), was also quantified in CAP sperm. Sperm viability for NC and CAP sperm was analyzed in the presence of 0.1 µg/ml (I) LPS or (J) VLY (10000 sperm each, n ≥ 3 biological replicates). Data were analyzed by independent t-test with Holm–Sidak’s multiple comparison test or two-way ANOVA with Bonferroni’s multiple comparison test and are presented as mean and SD. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, non-significant.

Figure 4.

The initial lipopolysaccharide (LPS)-induced human sperm hyperactivation is dose-dependent and inhibited by polymyxin B (PMB). The percentage of sperm with hyperactivated motility (HA) at time 0 of capacitation measured by CASA in sperm samples incubated in the presence of increasing concentrations of LPS with or without 100 µg/ml PMB. Data are presented as mean and SD (n = 5 biological replicates). *P < 0.05 by unpaired t-test.

Figure 5.

Lipopolysaccharide (LPS) and vaginolysin (VLY) cause rapid increases in intracellular calcium ([Ca²⁺]_i) in capacitating (CAP) mouse sperm. Representative traces of normalized Fluo4-AM fluorescence in responding (red) versus non-responding (black) mouse sperm with 1 µg/ml (A, C) LPS or (B, D) VLY, incubated under (A, B) Non-capacitating (NC) or (C, D) CAP conditions. Each trace was normalized to its respective ionomycin (Iono, 5 µM) response. Percentages of mouse sperm responding to (E) LPS or (F) VLY, in NC or CAP conditions were calculated. (G, H) Amplitude and (I, J) tau of the [Ca²⁺]_i response in NC or CAP sperm, with 1 µg/ml (G, I) LPS and (H, J) VLY. Data are presented as mean and SD (n = 4 biological replicates for all experiments). *P < 0.05, ***P < 0.001, ****P < 0.0001, by unpaired t-test for (E–J).

Figure 6.

Lipopolysaccharide (LPS) and vaginolysin (VLY) cause rapid increases in intracellular calcium ([Ca²⁺]_i) in capacitating (CAP) human sperm. Representative traces of normalized Fluo4-AM fluorescence in responding (red) versus non-responding (black) human sperm with 0.1 µg/ml (A, C) LPS or (B, D) VLY, incubated under (A, B) Non-capacitating (NC) or (C, D) CAP conditions. Each trace was normalized to its respective ionomycin (Iono, 5 µM) response. Percentages of human sperm responding to (E) LPS or (F) VLY, in NC or CAP conditions were calculated. (G, H) Amplitude and (I, J) tau of the [Ca²⁺]_i response in NC or CAP sperm, with 0.1 µg/ml (G, I) LPS or (H, J) VLY was quantified. Data are presented as mean and SD (n = 5 biological replicates for LPS, n = 4 biological replicates for VLY; in all experiments). **P < 0.01, ***P < 0.001, ****P < 0.0001, by unpaired t-test for (E–J).

Figure 7.

The lipopolysaccharide (LPS)- and vaginolysin (VLY)-induced intracellular calcium ([Ca²⁺]_i) increases in mouse and human sperm are irreversible. Representative traces of normalized Fluo4-AM fluorescence in CAP (A, B) mouse or (C, D) human sperm, perfused with (A, C) LPS (denoted in green), or (B, D) VLY (denoted in pink), followed by media (Wash, denoted in blue), and subsequent perfusion with 5 µM ionomycin (Iono, denoted in black). Each trace was normalized to its respective ionomycin response. Data are presented as mean and SD (n = 3 biological replicates for all experiments).

Figure 8.

The lipopolysaccharide (LPS)-induced increase in sperm intracellular calcium ([Ca²⁺]_i) is dependent on external Ca²⁺, independent of the CatSper ion channel, and mediated by the toll-like receptor 4 (TLR4). Representative traces of normalized Fluo4-AM fluorescence from capacitating (CAP) (A) wild-type (WT) mouse or (B) human sperm perfused with 0 mM Ca²⁺, 0 mM Ca²⁺ with LPS and ionomycin (Iono) with 2 mM Ca²⁺ (n = 3 biological replicates). Representative traces for LPS responding (red) and non-responding (black) (C) WT or (D) CatSper knockout (KO) mouse sperm in the presence of 2 mM extracellular Ca²⁺. (E) Percentages of LPS-responding sperm with increased [Ca²⁺]_i from WT and CatSper KO mice (n = 5 biological replicates). The LPS-induced [Ca²⁺]_i responses are shown on the traces obtained from (F) WT or (G) CatSper KO mouse sperm. Red curves are calculated as a standard exponential fit. Representative traces of Fluo4-AM fluorescence from CAP (H) WT mouse or (I) human sperm, perfused with 10 µM TAK-242 (TLR4 inhibitor), TAK-242 with LPS, and ionomycin (Iono), in the presence of 2 mM extracellular Ca²⁺ throughout the experiment (n = 3 biological replicates). Each trace was normalized to its respective ionomycin (Iono, 5 µM) response. Data are presented as mean and SD. ns, non-significant, by unpaired t-test in (E).