KSper, a pH-sensitive K⁺ current that controls sperm membrane potential

Abstract

Mature mammalian spermatozoa are quiescent in the male reproductive tract. Upon ejaculation and during their transit through the female reproductive tract, they undergo changes that enable them to fertilize the egg. During this process of capacitation, they acquire progressive motility, develop hyperactivated motility, and are readied for the acrosome reaction. All of these processes are regulated by intracellular pH. In the female reproductive tract, the spermatozoan cytoplasm alkalinizes, which in turn activates a Ca²⁺-selective current (I_CatSper) required for hyperactivated motility. Here, we show that alkalinization also has a dramatic effect on membrane potential, producing a rapid hyperpolarization. This hyperpolarization is primarily mediated by a weakly outwardly rectifying K⁺ current (I_KSper) originating from the principal piece of the sperm flagellum. Alkalinization activates the pH_i-sensitive I_KSper, setting the membrane potential to negative potentials where Ca²⁺ entry via I_CatSper is maximized. I_KSper is one of two dominant ion currents of capacitated sperm cells.

Before fertilization, mammalian sperm must undergo a poorly understood process called capacitation (1). Capacitation refers to a series of biochemical and physiological changes in the sperm cell that enable it to reach and fertilize the egg. Among these changes is an increase in the intracellular pH (pH_i) (2) and hyperpolarization of the sperm plasma membrane potential (V_m) (3, 4). Resting pH_i of mammalian sperm is ≈6.5 as determined by pH-sensitive fluorescent probes (2, 5–7). After in vitro capacitation in bicarbonate-containing media at pH 7.4, pH_i increases >0.3 units (2, 8, 9). When intracellular alkalinization is prevented by glucose incubation, bovine sperm fail to capacitate (8, 10).

Measurements with voltage-sensitive fluorescent dyes indicate that noncapacitated murine sperm are relatively depolarized (approximately −30 mV) and hyperpolarize to approximately −60 mV during capacitation (4, 11, 12). Indirect measurements attribute the capacitation-associated hyperpolarization to an increase in K⁺ permeability (4) and a block of epithelial sodium channels (ENaCs) (13). Interestingly, addition of BaCl₂ (a nonspecific K⁺ channel blocker) to the capacitating medium prevented the hyperpolarization and the acrosome reaction (12). Several K⁺ channels have been reported in mammalian sperm (ref. 14; see also reviews in refs. 15 and 16), but only recently have patch-clamp methods been developed that allow whole-sperm-cell currents to be recorded under voltage clamp (17). To date, the inward Ca²⁺-selective current, I_CatSper, but not outward K⁺-selective current, has been described in detail (17, 18). By using whole-spermatozoan current clamp, we find that pH_i largely determines mouse sperm membrane potential. Under whole-sperm and whole-flagellum voltage clamp, a pH_i-sensitive K⁺ current (I_KSper) was identified that sets the epididymal spermatozoan membrane potential.

Results

pH_i Controls Sperm Resting Membrane Potential.

Mouse sperm cells are long (≈100 μm) and vary in width from 2 μm at the head to <0.5 μm at the tip of the tail. There also is a relatively tight constriction (≈1 μm) between the head and midpiece. To establish that the patch-clamp pipette accesses the entire sperm cell, Lucifer yellow dye was loaded into the pipette (see ref. 17). Within 10 sec after gigaseal formation and break-in, the Lucifer yellow dye (molecular weight, 522) was seen throughout the spermatozoon midpiece, head, and principal piece (Fig. 1a). In current-clamp mode, the resting sperm membrane potential could then be measured. At pH_i 6.0 and in a bath of physiological saline (see Materials and Methods), the resting sperm membrane potential was ≈0 mV at 1.5 min after break-in (Fig. 1b). Intracellular alkalinization, induced by 5 mM NH₄Cl in the bath, dramatically hyperpolarized the membrane potential to −45 mV within a few seconds.

Fig. 1.

pH_i sets the murine spermatozoan resting membrane potential. (a) Spermatozoan in whole-cell patch-clamp configuration. The pipette contains 1 mM Lucifer yellow, which fills the cytoplasmic space (Lower). Differential interference contrast image (Upper). (Scale bar, 10 μm.) (b) WT spermatozoan resting membrane potential (pH_i 6.0) before and after induction of intracellular alkalinization by bath perfusion with 5 mM NH₄Cl. Reduction of extracellular Cl⁻ ions, or the addition of 25 μM amiloride (ENaC channel antagonist), did not block the alkalinization-induced hyperpolarization. Representative of three similar recordings is shown. (c) WT I_KSper at pH_i 6.0 before (red) and after (black) addition of 5 mM NH₄Cl in normal saline (HS) or low Cl⁻ solution (green). The dashed line indicates 0 current level. Also shown is the voltage-clamp ramp protocol from −100 to +100 mV [1 sec; holding potential (HP), 0 mV]. (d) Mean membrane potential in varying pH_i. Alkalinization dramatically shifts the spermatozoan's resting membrane potential (averaged data from experiments similar to a).

Reduction of Cl⁻ ions in the bath solution (from 144 to 4 mM) and addition of the ENaC channel blocker amiloride (25 μM) did not prevent alkalinization-induced hyperpolarization, suggesting that the hyperpolarization does not result from active Cl⁻ or ENaC channels (Fig. 1b). By switching between current- and voltage-clamp modes during the experiment, the effect of intracellular alkalinization on the ion currents was determined. NH₄Cl-induced alkalinization increased outward current >7-fold (Fig. 1c), which reversed upon NH₄Cl washout. Decreasing extracellular Cl⁻ ions from 144 to 4 mM (Fig. 1c) did not affect this potentiation of outward current, again excluding a significant role for Cl⁻-permeable channels. When the pipette was buffered to pH 6, alkalinization induced by NH₄Cl dramatically shifted sperm membrane potential from −7 ± 1 to −54 ± 1.6 mV. In contrast, when the pipette pH was 7.0, alkalinization by NH₄Cl induced only a −5 mV shift, from −54 ± 1.1 to −58.6 ± 2 mV (Fig. 1d). Thus, pH_i sets the sperm membrane potential primarily by modifying a K⁺ conductance.

Properties of the Spermatozoan Potassium Current (I_KSper).

Immediately after break-in under voltage clamp, a constitutively active, weakly outwardly rectifying K⁺ current was recorded (symmetrical 160 mM [K⁺]) (Fig. 2a). To avoid contamination from monovalent currents via CatSper channels in nominal Ca²⁺-free conditions (17) [supporting information (SI) Fig. 5], spermatozoa from CatSper1^−/− mice were used in subsequent recordings (I_KSper was unaffected by the absence of CatSper1 protein). In step depolarizations, I_KSper exhibited no time or voltage dependence (Fig. 2b). I_KSper was relatively K⁺-selective; the reversal potential was 1.2 ± 0.34 mV in 160 mM [K⁺]_o and −72.5 ± 0.71 mV in 5 mM [K⁺]_o (160 mM [K⁺]_i; predicted E_K = −88 mV at 22°C) (Fig. 2c). P_Na/P_K could not be determined because of a contaminating Na⁺ current (SI Fig. 6a) under biionic conditions (SI Fig. 6b) and K⁺ accumulation in the sperm flagellum (SI Fig. 6c) (see also ref. 17).

Fig. 2.

Endogenous sperm cell K⁺ current (I_KSper). (a) Weakly outwardly rectifying K⁺ current from a WT spermatozoan measured in symmetrical 160 mM [K⁺] (pH_i 8.0) in response to a 1-sec ramp protocol from −100 to +100 mV [holding potential (HP), 0 mV]. (b) I_KSper of CatSper1^−/− spermatozoa in response to voltage steps. Whole-cell currents elicited by 1-sec steps from HP of 0 mV to test potentials between −100 and +100 mV (Δ = +20 mV; 5-sec intervals; pH_i 8.0). (c) Reversal potentials of a CatSper1^−/− spermatozoan were measured in response to changing [K⁺]_o (pH_i 8.0) from 160 mM (black trace; E_rev = 1.2 ± 0.34 mV; n = 30) to 5 mM (red trace; E_rev = −72.5 ± 0.71 mV; n = 21; predicted E_K = −87 mV). Green trace, 0 mM [K⁺]_o. (d) Head plus midpiece sperm fragment loaded with Lucifer yellow dye (Lower). Differential interference contrast (DIC) image (Upper). (Scale bar, 10 μm.) (e) Midpiece + principal piece sperm fragment loaded with Lucifer yellow dye (Lower). DIC image (Upper). (Scale bar, 10 μm.) (f) Current in CatSper1^−/− spermatozoa in response to a 1-sec voltage-clamp ramp from −100 to +100 mV from the head + midpiece (red) and from the midpiece + principal piece (black; pH_i 8.0).

One advantage of spermatozoa is that they can be divided into two intact fragments (head + midpiece or midpiece + principal piece) and patch-clamped separately (17). I_KSper originated from the principal piece of the sperm flagellum as shown by recordings from head + midpiece and midpiece + principal piece fragments (Fig. 2 e and f). Averaged outward currents of the two respective fragments were 11.9 ± 4.4 pA (n = 6; +100 mV) and 442.9 ± 32.7 pA (n = 4; +100 mV). In summary, I_KSper is a weakly outwardly rectifying K⁺ current originating from the principal piece of the sperm flagellum. Thus, I_KSper and I_CatSper are flagellar-specific currents.

I_KSper Is Strongly Potentiated by Intracellular Alkalinization.

Intracellular acidification inhibited, whereas alkalinization significantly potentiated, I_KSper (Fig. 3). Average I_KSper increased ≈8-fold when pH_i changed from 6.0 to 8.0 without affecting the overall shape of the I–V relationship (Fig. 3 a and b). Intracellular alkalinization induced by bath addition of 5 mM NH₄Cl, strongly potentiated I_KSper. This potentiation was independent of [K⁺]_o (Fig. 3b) and thus unlikely to be caused by increased permeation of NH₄Cl through I_KSper channels.

Fig. 3.

Intracellular alkalinization strongly potentiates I_KSper. (a) Steady-state whole-cell currents of CatSper1^−/− spermatozoa at pH_i 6.0 (red), 7.0 (blue), and 8.0 (black). (b) Initial current (red) of CatSper1^−/− spermatozoa in symmetrical 160 mM [K⁺] at pH_i 6.0. Addition of 5 mM NH₄Cl to the bath alkalinized the cytoplasm and induced I_KSper (black). Alkalinization-induced I_KSper in 5 mM (purple) or 80 mM (green) [K⁺]_o. (c) Average I_KSper amplitudes at −100 and +100 mV at varying pH_i. HP, holding potential.

Extracellular pH also affected the amplitude of I_KSper. On average, currents were smaller at acidic pH_o and larger at alkaline pH_o (SI Fig. 7). This effect was likely because of changing pH_i rather than direct effects by external protons; as long as the internal buffer capacity was sufficient to control pH_i, pH_o had little effect on I_KSper.

I_KSper was not affected by 2 mM membrane-permeant cAMP and cGMP analogs (SI Fig. 8), by increasing [Ca²⁺]_o to 2 mM or [Ca²⁺]_i (up to 10 μM), or by changes in bath osmolarity (230–350 mOsm) (SI Fig. 9). The response of I_KSper to pH was identical in WT and CatSper1^−/− sperm cells. Hence, I_KSper is a pH_i-sensitive channel that is activated in vivo by intracellular alkalinization.

I_KSper Antagonists and Their Effect on Sperm Membrane Potential.

Quinine (500 μM), an antagonist of many K⁺ channels, reversibly inhibited I_KSper by 88% (at +100 mV) (Fig. 4a). Clofilium (50 μM), a HERG (K_V11.1), K_V1.5, and TASK2 (K_2p5.1) blocker irreversibly reduced I_KSper by 92%. Unexpectedly, 50 μM 5-(N-ethyl-N-isopropyl)amiloride (EIPA) (an amiloride analog and Na⁺/H⁺ exchanger antagonist) reversibly inhibited I_KSper by 72%, whereas the ENaC antagonist, amiloride, had no effect (at up to 1 mM). Mibefradil (5 μM), an antagonist of Ca_V3 (T-type), two-pore K⁺ channels (K_2p), and K_V1.5, reversibly reduced I_KSper by 69%. In contrast, 2 mM BaCl₂, 10 mM tetraethylammonium-Cl, and 200 μM CdCl₂ (an inhibitor of K_V4.3) minimally inhibited I_KSper (<2%). Ba²⁺ inhibited I_KSper only minimally in symmetrical 160 mM [K⁺], perhaps because K⁺ is harder to displace from the pore under these conditions. However, when Na⁺ is the dominant extracellular cation, Ba²⁺ blocked I_KSper by ≈50% (SI Fig. 10). 4-Aminopyridine (4-AP), a nonselective K_V channel antagonist, enhanced I_KSper as well as I_CatSper, but this was most likely a consequence of its known alkalinization of cells (pK_a 9.2; at physiological pH, ≈1% is present as the uncharged membrane-permeant pyridine) (SI Fig. 11) (19). Thus, quinine, clofilium, EIPA, 4-AP, Ba²⁺, and mibefradil were tested for their effects on sperm membrane potential under current clamp.

Fig. 4.

I_KSper antagonists and their effects on sperm membrane potential. (a) I_KSper antagonist effect on CatSper1^−/− spermatozoa (symmetrical 160 mM [K⁺] at pH_i 7.5, +100mV). (b) Spermatozoan membrane potential (WT; pH_i 7.0) was reversibly depolarized by 500 μM quinine but irreversibly depolarized by 50 μM clofilium (washout, 15 min). Addition of 5 mM NH₄Cl had a small effect on membrane potential. A representative trace from four independent experiments is shown. (c) Bath perfusion with 2 mM BaCl₂, 5 μM mibefradil, and 50 μM EIPA reversibly depolarized the WT sperm cell membrane potential (pH_i 7.0). A representative trace from four independent experiments is shown. (d) EIPA (50 μM) blocked WT sperm hyperpolarization induced by 1 mM NH₄Cl or 4 mM 4-AP (pH_i 6.0). Amiloride (25 μM) had no effect on V_m. A representative trace from three independent experiments is shown.

Sperm membrane potential appears to be largely determined by pH_i-sensitive I_KSper. At resting pH_i (≈6.0–6.5), even very small currents can shift the membrane potential because the spermatozoan has an exceptionally large input resistance (R_m > 5 GΩ; C_m ≈ 2.5 pF). As expected, quinine transiently depolarized sperm membrane potential, whereas clofilium did so irreversibly (pH_i 7.0) (Fig. 4b). EIPA depolarized the sperm cell membrane potential (pH_i 7.0) (Fig. 4c) and blocked the hyperpolarization induced by NH₄Cl and 4-AP (pH_i 6.0) (Fig. 4d). Interestingly, Ba²⁺, which weakly blocks I_KSper, depolarized sperm cells (pH_i 7.0) (Fig. 4c and SI Fig. 12a) and blocked NH₄Cl-induced hyperpolarization (SI Fig. 12b). Because Ba²⁺ permeates CatSper and potentially could induce depolarization, the I_CatSper blocker Ni²⁺ (300 μM) was added together with Ba²⁺ in some experiments. I_CatSper block did not affect Ba²⁺-induced depolarization, suggesting that this effect is mainly through its block of I_KSper. We also tested the effect of [K⁺]_o on membrane potential. Addition of 5 mM KCl to the bath induced a small depolarization (≈3 mV) (Fig. 1b), whereas 150 mM [K⁺]_o depolarized the membrane by ≈60 mV (SI Fig. 12a). In summary, antagonists that reduce I_KSper also dramatically depolarize sperm V_m.

Discussion

Our results show that pH_i has a dramatic effect on sperm membrane potential, with alkalinization producing a rapid hyperpolarization. This hyperpolarization is primarily mediated by an endogenous weakly outwardly rectifying, pH_i-sensitive K⁺ current (I_KSper) originating from the principal piece of the sperm flagellum. Together with the effects of I_KSper antagonists on sperm membrane potential and pH_i sensitivity, we conclude that I_KSper is the dominant hyperpolarizing conductance within the physiological range and thus largely sets spermatozoan resting membrane potential. Sperm intracellular pH_i roughly follows pH_o (20). As sperm travel from the vagina (pH ≈5) to the cervical mucous (pH ≈8), they undergo intracellular alkalinization. Consequently, the sperm's membrane potential hyperpolarizes as pH_o increases in the female reproductive tract.

In this and previous work, we characterized two of the primary ion conductances of epididymal mouse spermatozoa, I_CatSper and I_KSper. I_CatSper requires at least four CatSper proteins to comprise this weakly voltage-gated Ca²⁺-selective conductance. The molecular identity of I_KSper is not yet determined.

Both I_CatSper and I_KSper are sensitive to pH_i in the physiological range and originate specifically from the principal piece of the sperm flagellum. Alkalinization hyperpolarizes sperm by activating I_KSper and at the same time dramatically shifts the activation potential of I_CatSper to the hyperpolarized range. Because at alkaline pH I_CatSper is active, hyperpolarization will increase the driving force for Ca²⁺ entry (E_Ca > +150 mV). The primary effect of this simple change is to increase intraflagellar [Ca²⁺] and induce hyperactivated motility. However, currents that might be induced by unknown native agonists, and the uncharacterized Na⁺-carrying current, could mediate other effects on sperm membrane potential and physiology.

The most likely gene responsible for I_KSper is mSlo3 (21). Like the CatSper channels, mSlo3 appears specific to testis and has not been functionally expressed in mammalian cell lines. mSlo3 expression in Xenopus oocytes yielded measurable currents, and these currents were activated by intracellular alkalinization (21, 22). Like I_KSper, Xenopus oocyte-expressed mSlo3 is also weakly voltage-sensitive (≈16 mV/e-fold), has relaxed K⁺-selectivity, and is insensitive to [Ca²⁺]_i and external tetraethylammonium. Available anti-mSlo3 antibodies were not specific enough to allow immunocytochemical identification or localization. Further evidence that I_KSper is mediated by mSlo3 will require deletion of this gene in mice.

Materials and Methods

Whole-cell recordings were made on sperm cells from the corpus epididymes from mice 3–8 months of age, as reported (17). The standard bath solution (HS) contained the following: 135 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 1 mM MgSO₄, 20 mM Hepes, 5 mM glucose, 10 mM lactic acid, 1 mM Na pyruvate, pH 7.4 (with NaOH). After break-in, the access resistance was 25–80 MΩ. The standard pipette solution was as follows: 115 mM K-methanesulfonate (K-MeSO₃), 5 mM KCl, 10 mM K₄-BAPTA, 20 mM Hepes, and 20 mM Mes (pH 8.0 with Trizma base or pH 6.0 with methanesulfonic acid). A weak pH-buffered pipette solution was used in some experiments and contained the following: 130 mM K-MeSO₃, 5 mM KCl, 1 mM K₄-BAPTA, 5 mM K-Hepes, and 5 mM K-Mes (pH 6.0). In symmetrical 160 mM [K⁺] experiments, the bath solution was 150 mM K-MeSO₃, 10 mM K-Hepes, and 10 mM Mes (pH 7.4). The bath solution for biionic experiments contained the following: 160 mM Na-MeSO₃, 5 mM K-MeSO₃, 10 mM Hepes, 10 mM Mes (pH 7.4); for E_rev measurements, it was as follows: 5 mM K-MeSO₃, 170 mM Hepes (pH 7.4). The pipette solution for current-clamp experiments contained the following: 130 mM K-MeSO₃, 5 mM KCl, 15 mM NaCl, 3 mM MgATP, 0.5 mM Na₂GTP, 1 mM K₄-BAPTA, 5 mM K-Hepes, and 5 mM K-Mes (pH 6.0 or 7.0). The low Cl⁻ solution was as follows: 150 mM Na-MeSO₃, 5 mM K-MeSO₃, 2 mM CaCl₂, 10 mM Hepes, and 10 mM Mes, pH 7.4. For tests of osmolarity, the bath solution was as follows: 110 mM K-MeSO₃, 10 mM Hepes, and 10 mM Mes (230 mOsm; pH 7.4), with mannitol being added to increase osmolarity.

Headless and tailless sperm cells (17) were prepared by incubating the sperm cell suspension in the presence of 0.2 mg/ml trypsin at 37°C for <5 min, followed by gentle trituration. All experiments were performed at 22–24°C. EIPA, amiloride hydrochloride hydrate, and quinine were dissolved in DMSO (final <0.1%). 4-AP, mibefradil, and clofilium were water-soluble. Antagonists were diluted in HS solution to their final concentration and then perfused into the recording chamber. All currents were recorded by using an Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA), acquired with Clampex (pClamp9; Molecular Devices), and analyzed with Origin software (OriginLab, Northampton, MA). Signals were low-pass-filtered at 2 kHz and sampled at 10 kHz. Data are given as mean ± SEM.

Abbreviations

pH_i

intracellular pH

ENaC

epithelial sodium channel

EIPA

5-(N-ethyl-N-isopropyl)amiloride

4-AP

4-aminopyridine.