Regulation of the sperm calcium channel CatSper by endogenous steroids and plant triterpenoids

Progesterone (P4) and its closest analog 17-OH-P4 have been shown to trigger calcium influx into the sperm flagellum (19) via activation of the sperm-specific calcium channel CatSper (3, 5). However, CatSper specificity toward other steroid hormones has not been evaluated. To reveal CatSper regulation by endogenous steroids and steroid-like molecules, we have analyzed inward monovalent currents through CatSper (I_CatSper) of human sperm. Cells were first stimulated with either 1 μM testosterone, 17 beta-estradiol (E2), HC, or PregS alone. Subsequently, each preexposed cell was stimulated with 1 μM P4 in combination with the corresponding steroid (testosterone, E2, or HC). Testosterone, E2, or HC did not alter basal CatSper activity, whereas P4, as expected, caused a sevenfold potentiation of inward CatSper currents (Fig. 1 A–D and Table S1). However, coadministration of testosterone and P4 to spermatozoa, which were preexposed to testosterone, completely prevented I_CatSper potentiation by P4 (Fig. 1 A and B and Table S1). E2 and HC partially reduced CatSper activation by P4 to 3.6- and 4.1-fold, when applied simultaneously with P4 (Fig. 1 C and D and Table S1). To explore the efficiency of P4 inhibition by either testosterone, E2, or HC, we have determined the IC₅₀ values for each of them by measuring CatSper activation by 1 μM of P4 in the presence of various concentrations of the corresponding steroid. Human spermatozoa were first preexposed to corresponding concentrations of either testosterone, E2, or HC. Subsequently, cells were exposed to both the individual steroid concentration from the first stimulus plus 1 µM P4. The respective IC₅₀ values were 429 ± 73 nM (testosterone; Fig. 1 F and I and Table S1), 833 ± 280 nM (E2; Fig. 1 G and I and Table S1), and 153 ± 90 nM (HC; Fig. 1 H and I and Table S1). These data indicate that exposure to physiological concentrations of either testosterone or hydrocortisone prevents CatSper activation by P4, and therefore, human spermatozoa must be free from testosterone or hydrocortisone influence to achieve full CatSper activation. Estrogen, however, requires much higher concentrations, which are outside of the physiological range, to antagonize CatSper activation (Fig. 1I).

In contrast to the aforementioned steroids, PregS alone stimulated inward currents fourfold, which resembled I_CatSper properties (Fig. 1E and Table S1). To test whether PregS-elicited currents were identical to I_CatSper, we conducted the following experiments.

To evaluate whether PregS activated CatSper, we inhibited PregS-elicited currents with 1 μM NNC 55–0396, a known CatSper inhibitor (3) (Fig. 2 A and B and Table S2). Indeed, 1 μM NNC 55–0396 reduced PregS-elicited currents below basal levels indicating that PregS is another steroid capable of activating CatSper. The potentiation of I_CatSper by PregS was dose-dependent, with an EC₅₀ of 15.2 ± 5.0 nM (Fig. 2 C and D, black line and Table S2), which is close to the EC₅₀ obtained with P4 [7.7 ± 1.8 nM (3)]. The serum concentration of PregS in adults reaches levels of 130 nM (women) and 140 nM (men) (14). It is also abundant in testes, in which the total PregS concentrations can reach 2.8 µmol/kg (3.8 µM), whereas it drops to 0.76 µmol/kg (190 nM) in the epididymis (14). Such high PregS concentrations may result in premature CatSper activation, while spermatozoa are still in the male reproductive tract, which would be detrimental for sperm fertility. Because testosterone is also abundant in the male reproductive tract, we have explored whether it can block CatSper activation by PregS. Human sperm cells were first preincubated with 1 µM testosterone alone followed by 1 µM testosterone plus various concentrations of PregS (Fig. S1). Under such conditions, the EC₅₀ of PregS increased 10-fold and reached 176.0 ± 71.0 nM, indicating that testosterone indeed interferes with PregS activation of CatSper (Fig. 2D, red line and Table S2). In human sperm, P4 activates CatSper indirectly via binding to and then activating its membrane receptor ABHD2 (7). ABHD2 acts as a serine hydrolase, thus depleting the endogenous CatSper inhibitor 2-AG from the plasma membrane. Removal of 2-AG subsequently leads to the opening of CatSper. To determine whether the potentiation of I_CatSper by PregS was ABHD2-mediated, we inactivated ABHD2 by exposing human spermatozoa to 2 μM methyl arachidonoyl fluorophosphate (MAFP), a serine hydrolase inhibitor (Fig. 2 E and F and Table S2). As expected, MAFP exposure ablated CatSper activation by PregS, whereas basal I_CatSper was not affected. Taken together, these results suggest that in addition to P4, ABHD2 can also be activated by PregS. Because both P4 and PregS influence ABHD2 activity, we tested whether these compounds compete for the binding site. To do so, human sperm cells were stimulated with saturated concentrations of PregS (Fig. 3A, blue trace) followed by costimulation with a combination of saturated concentrations of P4 and PregS (Fig. 3 A and C, red trace and Table S3). The experiments were repeated in the reverse order (Fig. 3 B and C and Table S3). The increase of CatSper currents by PregS was not further potentiated upon addition of P4. The same result was obtained when sperm cells were first stimulated with P4 and then with a combination of P4 and PregS. These experiments suggest that P4 and PregS likely compete for the same ABHD2 binding site and that CatSper activation by PregS follows the same mechanism as CatSper activation by P4.

2-AG hydrolysis by monoacyl lipases, such as ABHD2, is required for CatSper activation. The plant triterpenoid pristimerin was reported to inhibit the activity of the ABHD-related enzyme monoacylglycerol lipase (MAGL) (18). If pristimerin can also inhibit ABHD2, then sperm exposure to this compound should prevent CatSper activation by P4 or PregS. To test this hypothesis, human spermatozoa were stimulated with pristimerin, followed by exposure to either a mixture of pristimerin and P4 or to a mixture of pristimerin and PregS. The mere presence of pristimerin did not affect basal I_CatSper (Fig. 4A, Fig. S2 A–D, and Table S4), which indicates that pristimerin did not target the channel directly. However, coapplication of pristimerin with P4 or PregS significantly reduced CatSper activation. Compared with stimulation with P4 or PregS alone, I_CatSper stimulation was reduced by 63% (pristimerin and P4) and by 48% (pristimerin and PregS; Fig. 4A, Fig. S2 A–D, and Table S4). To determine the IC₅₀ value of pristimerin’s inhibitory effect on CatSper activation by P4, human sperm cells were first preexposed to various concentrations of pristimerin. Then, cells were exposed to the respective pristimerin concentration of the first stimulus plus 1 µM P4. The IC₅₀ of pristimerin in the presence of 1 µM P4 was 116 ± 77 nM (Fig. 4B and Table S4).

Another pharmacologically active plant triterpenoid, which affects sperm functions when orally administered to rats, is lupeol (17). Comparable to pristimerin, lupeol alone did not affect basal I_CatSper (Fig. 5A, Fig. S2 E and F, and Table S5). However, administration of lupeol with P4 or PregS led to an even stronger inhibition of I_CatSper than pristimerin. Compared with CatSper stimulation elicited by P4 or PregS alone, CatSper currents were reduced by 71% and 68% in combination with P4 or PregS, respectively (Fig. 5A). The IC₅₀ value of lupeol to inhibit CatSper activation by 1 µM P4 was 109 ± 56 nM (Fig. 5C and Table S5).

Hyperactivated sperm motility is characterized by a highly asymmetrical bending of the sperm tail due to a CatSper-mediated rise in flagellar calcium concentration (20). Curvilinear velocity (VCL) is the average velocity of the sperm head through the sperm trajectory, which increases during capacitation. A CatSper-mediated calcium rise promotes hyperactivation and makes the sperm trajectory less linear, which results in an increase of VCL. Capacitation increases CatSper activity, which results in a higher percentage of sperm with hyperactivated motility. Thus, capacitated spermatozoa tend to have higher VCL values than noncapacitated spermatozoa (Fig. S3A and Table S6). Because both P4 and PregS activate CatSper, whereas pristimerin and lupeol inhibit it, we explored whether VCL values of human sperm were affected in the presence of steroid hormones or plant triterpenoids. As expected, neither of the compounds changed VCL values of noncapacitated sperm cells (Fig. S3, Bottom panels and Table S6). In capacitated spermatozoa, P4 stimulation increased VCL by 12%, comparable to numbers reported earlier (7, 21), but VCL values remained unchanged in the presence of PregS, pristimerin, or lupeol alone (Fig. S3A, Top and Inset and Table S6). When spermatozoa were stimulated with P4 + pristimerin or P4 + lupeol, VCL values were reduced by 39% and 48%, respectively, in comparison with VCL values obtained in the presence of P4, and the reduced VCL values were comparable to those of noncapacitated cells (Fig. S3B and Table S6). Stimulating capacitated sperm with PregS in combination with pristimerin or lupeol decreased VCL by 18% and 9%, respectively, compared with sperm, which were treated with PregS alone (Fig. S3C, Top and Table S6). VCL values of noncapacitated sperm remained unchanged under such conditions (Fig. S3C, Bottom and Table S6). To explore whether pristimerin and lupeol also affect basal sperm motility, we assessed the effect of these compounds on motility of capacitated and noncapacitated sperm cells. The number of both capacitated and noncapacitated motile spermatozoa remained unchanged when stimulated with either compounds (Fig. S4A and Table S7). However, when P4 was coapplied with pristimerin or lupeol, the percentage of motile capacitated sperm decreased by 19%, whereas the values of noncapacitated cells did not change (Fig. S4B and Table S7). Costimulation with PregS and pristimerin or lupeol, however, did not result in a significant reduction of motile capacitated or noncapacitated sperm (Fig. S4C and Table S7). These results demonstrate that both pristimerin and lupeol have the capacity to significantly reduce sperm hyperactivation by blocking the P4-mediated activation of CatSper but also have a minor effect on basal sperm motility of capacitated cells.

Progesterone and pristimerin were purchased from Calbiochem (EMD Millipore). MAFP was from Cayman Chemical Company, and NNC 55–0396 was from Tocris. All other compounds were from Sigma Aldrich. Testosterone was purchased in accordance with the controlled substance protocol (CS084484), as a collaborative effort with Yuriy Kirichok (University of California, San Francisco).

The participation of four healthy human sperm donor volunteers was approved by the Committee on Human Research at the University of California, Berkeley (protocol number 2013–06-5395). All donors provided informed consent. Freshly ejaculated semen samples were obtained by masturbation. Sperm were purified with the swim-up technique (3) using artificial human tubal fluid solution (HTF), containing (in mM) 21 Hepes, 21 lactic acid, 98 NaCl, 4.7 KCl, 3 glucose, 2 CaCl₂, 0.3 KH₂PO₄, 0.3 sodium pyruvate, and 0.2 MgSO₄, pH 7.4 (adjusted with NaOH).

All recordings were performed as described in ref. 3. Briefly, gigaohm seals between patch pipette and spermatozoa were formed at the cytoplasmic droplet in high saline (HS) solution containing (in mM) 135 NaCl, 20 Hepes, 10 lactic acid, 5 KCl, 5 glucose, 2 CaCl2, 1 MgSO₄, 1 sodium pyruvate, pH 7.4 (adjusted with NaOH), and ∼320 mOsm/L. Transition into whole-cell mode was achieved by applying suction and short voltage pulses. For CatSper recordings, the divalent-free bath solution contained (in mM) 140 Cs-methanesulfonate, 40 Hepes, 1 EDTA, pH 7.4 (adjusted with CsOH), and ∼325 mOsm/L. Pipettes (10–15 MΩ) were filled with 130 mM Cs-methanesulfonate, 70 mM Hepes, 3 mM EGTA, 2 mM EDTA, 0.5 mM Tris·HCl, pH 7.4 (adjusted with CsOH), and ∼335 mOsm/L. Access resistance was 42–60 MΩ. Cells were stimulated every 5 s, and data were sampled at 2–5 kHz and filtered at 1 kHz. All experiments were performed at room temperature and currents elicited by a voltage ramp from –80 mV to 80 mV with a holding potential of 0 mV. Data were analyzed with Clampfit 9.2 and OriginPro 9.0. To build dose–response curves, data were fitted with the Hill-based equation: y = I_min + (I_max – I_min)/(1 + (x/IC₅₀)^k), where I_max is close to 100% activation, I_min is close to inactivation, and k is the Hill slope factor.

Purified human spermatozoa were capacitated for 3.5 h at 37 °C and 5% CO₂ in capacitation media (HS supplemented with 15 mM NaHCO₃ and 5% BSA) as reported in ref. 7. Aliquots of the cell suspension were preincubated for 15 min with 3 μM pristimerin or 3 μM lupeol before exposure to steroids (3 μM progesterone or 3 μM PregS). Sperm motility was analyzed at 37 °C with an HTM-IVOS sperm analysis system (version 12.3, Hamilton Thorne Biosciences). We pipetted 10 μL of sperm suspension into a two-chamber slide (Leja), and sperm movement of a minimum of 300 cells was recorded. Parameters measured were the four motility classes (A–D), average path velocity (VAP, μm/s), straight line velocity (VSL, μm/s), and VCL (μm/s). Measurements on a given day were performed in duplicates and defined as one experiment.

Statistical data were calculated as the mean ± SEM, and n indicates number of individual cells analyzed unless stated otherwise. Statistical significance (unpaired t test) is indicated by *P < 0.05, **P < 0.005, ***P < 0.001, and ****P < 0.0001.