A new prostaglandin E receptor mediates calcium influx and acrosome reaction in human spermatozoa

Abstract

Zona pellucida protein 3, a protein of the egg’s extracellular matrix, and progesterone secreted by granulosa cells surrounding the oocyte are regarded as physiological stimuli of sperm acrosome reaction. Signal transduction steps initiated by both stimuli result in influx of Ca²⁺ from the extracellular space. Herein, we propose a role for prostaglandin (PG) E as a physiological inducer of Ca²⁺ influx and acrosome reaction in human spermatozoa. PGE₁ specifically binds to human sperm membranes (K_d = 20.4 nM; B_max = 88 fmol/mg protein) and induces a pertussis toxin-insensitive, transient increase in intracellular Ca²⁺ concentrations, which can be blocked by μM concentrations of La³⁺, Gd³⁺, and Zn²⁺. The kinetic profile was similar to that observed after progesterone challenge. Sequential application of both agonists did not lead to cross-desensitization. E prostaglandins were found to be the only prostanoids with agonistic properties (EC₅₀ values for PGE₁ and PGE₂: <10 nM and 300 nM, respectively). Pharmacological characteristics were not compatible with those of cloned prostanoid receptors indicating the expression of a distinct membrane receptor. Activation of the sperm E prostanoid receptor stimulates incorporation of [α-³²P]GTP azidoanilide into immunoprecipitated Gα_q/11 subunits. Thus, in human sperm, PG induces Ca²⁺ influx and acrosome reaction via a G_q/11-coupled E prostanoid receptor. The block of PGE₁-induced Ca²⁺ transients and acrosome reaction by physiological Zn²⁺ concentrations highlights a role of Zn²⁺ as an endogenous Ca²⁺ channel blocker present in seminal plasma protecting sperm from premature PGE₁-evoked increases in intracellular Ca²⁺ concentrations.

In vivo, ejaculated and epididymal mammalian spermatozoa are not able to fertilize an egg immediately but have to undergo a process of maturation in the female reproductive tract. This time-dependent acquisition of fertilizing capacity called “capacitation” is correlated with changes in sperm motility, metabolism, plasma membrane fluidity, and intracellular ion concentrations (1). In capacitated spermatozoa, local stimuli acting in vicinity of the oocyte induce the acrosome reaction (AR), an exocytotic event leading to release of hydrolytic enzymes and substantial reorganization of the sperm plasma membrane (1). To date, two physiological inducers of the AR are known: Subsequent to species-specific binding of sperm to the zona pellucida (ZP), the oocyte’s extracellular matrix (2–4), one of the three major proteins forming the mouse ZP, ZP3, elicits the AR (5). Progesterone secreted by ovarian follicular cells surrounding the ovulated egg also initiates the AR (6), and a priming role of the steroid for the induction of the AR by the ZP has been suggested (7).

Signal transduction steps in sperm resulting in AR are understood poorly. One of the essential features of the AR is an influx of Ca²⁺ from the extracellular space required to promote the fusion between the outer acrosomal membrane and the overlying sperm plasma membrane (1). The lack of knowledge on signaling steps leading to Ca²⁺ influx and AR contrasts with the detailed information about the expression and subcellular localization of classical signal transduction components like G protein-coupled receptors (8), receptor kinases (9), G proteins (10, 11), and effectors such as enzymes (11) and ion channels (12). Thus, the delineation of signaling cascades in sperm is a crucial first step to understand the physiology of fertilization at the molecular level. In the present study, we identified a new role for E prostaglandins as physiological inducers of the AR in humans and provide evidence for the expression of a G protein-coupled E prostanoid (EP) receptor in human sperm.

MATERIALS AND METHODS

Sperm Sample Preparation.

Human semen samples were obtained from couples undergoing in vitro fertilization because of female infertility. Sperm samples with normal parameters of sperm count, motility, and morphology were pooled and included in this study. Motile sperm fractions were isolated by a swim-up procedure in hypertonic Biggers Whitten and Whittingham (BWW) medium (for composition, see refs. 13 and 14) supplemented with 10% fetal calf serum (BWW-FCS). Capacitation was promoted by incubating sperm suspensions (0.5–2 × 10⁷ per ml) for 6–8 h at 37°C in a humidified atmosphere containing 5% CO₂.

Assessment of Intracellular Ca²⁺ Concentrations in Cell Suspensions and Immobilized Single Cells.

For determination of intracellular Ca²⁺ concentrations in sperm suspensions, the fluorescent indicator fluo-3/AM (2 μM; Molecular Probes) was added during the final 30 min of capacitation followed by a 15-min incubation at room temperature. To prevent precipitation of insoluble salts, bicarbonate, phosphate, and sulfate ions present in the original BWW medium were replaced by chloride ions when lanthanum, gadolinium, or zinc ions were added. Fluorescence (excitation wavelength 506 nm; emission wavelength 526 nm) was monitored at 37°C with an LS50B dual wavelength fluorescence spectrophotometer (Perkin–Elmer). To obtain F_max and F_min, Triton X-100 (reduced form; final concentration, 0.1%) and EGTA (pH 8.4, 20 mM) were added to the incubation buffer, respectively. Intracellular Ca²⁺ concentrations were calculated by assuming a K_d of 400 nM (37°C) for fluo-3 (15). Stock solutions (10 mM) of lipophilic agonists were prepared in ethanol. Human erythroleukemia (HEL) 92.1.7 cells resuspended in culture medium were loaded with fluo-3/AM (2 μM) as outlined for fura-2/AM (16).

For single-cell Ca²⁺ determinations, capacitated, fura-2/AM-loaded sperm suspensions (Molecular Probes) were washed in incubation buffer (IC) containing 138 mM NaCl, 6 mM KCl, 1.25 mM CaCl₂, 1.25 mM MgCl₂, 5.5 mM glucose, and 10 mM Hepes (pH 7.4) and layered onto poly(l-lysine)-coated coverslips, which were subsequently washed with incubation buffer twice. Intracellular Ca²⁺ concentrations in single sperm were monitored with a digital imaging system (TILL Photonics, Planegg, Germany) using an inverted microscope (Zeiss Axiovert 100). Fura-2 fluorescence was excited alternately at 340 and 380 nm. Regions of interest were defined manually over single sperm. Background fluorescence was assessed by monitoring fluorescence signals in the presence of digitonin (50 μM) and MnCl₂ (10 mM) and subtracted from fluorescence readings.

Assessment of Sperm Acrosome Reaction and Determination of cAMP Levels.

Acrosome reaction assay was performed as described in ref. 17. Aliquots of capacitated sperm suspensions (1 × 10⁶ cells/ml) were resuspended in BWW supplemented with 1 mM 3-isobutyl-1-methylxanthine and incubated with or without agonists for the times indicated. Cells were pelleted and supernatants were removed. cAMP was extracted from sperm pellets by resuspending cells in 100 μl of ice-cold water containing 0.1% Triton X-100. Proteins were denatured by boiling samples for 10 min. Cellular debris was removed by centrifugation, and the cAMP content of the supernatant was determined by radioimmunoassay (16).

Membrane Preparation, SDS/PAGE, Immunoblotting, and G Protein Antibodies.

Membranes were prepared from washed and capacitated sperm according to Althouse et al. (18). Harvested membrane pellets were resuspended in Hepes-phosphate-buffered saline (pH 7.4) and stored at −70°C until analysis.

Membranes prepared from mouse L cells (16) served as positive controls. Membrane proteins (50 μg per lane) were resolved by SDS/PAGE performed on 13% (wt/vol) acrylamide gels and blotted onto nitrocellulose filters. Immunoreactive bands were visualized as described (19). The following antisera were used: AS 348 (α_s), AS 233 (α₁₂), AS 343 (α₁₃), and AS 368 (α_q/11) (20, 21).

Photolabeling of Sperm Membrane G Proteins.

The photolabeling of G_q/11 proteins with [α-³²P]GTP azidoanilide and immunoprecipitation was performed as described (22). Immunoprecipitated proteins (AS 368) were resolved by 10% SDS/PAGE and were visualized by blotting onto nitrocellulose membranes and autoradiography followed by analysis with a phosphorimager (Fuji BAS 1500). To test for comparable loading of immunoprecipitates, nitrocellulose membranes were immunostained with a biotinylated IgG fraction prepared from AS 370 (α_q/11) and horseraddish peroxidase-conjugated streptavidin.

[³H]PGE₁-Binding.

Binding of [³H]PGE₁ to human, bovine, and porcine sperm membranes was assessed as described (23). Sixty micrograms of sperm membranes were incubated for 1 h at room temperature with ³[H]PGE₁ at the concentrations indicated. Bound and free ligand were separated by filtration through GF/B glass fiber filters (Whatman). Nonspecific binding was determined in the presence of 10 μM unlabeled PGE₁.

RESULTS

Prostanoids Inducing Ca²⁺ Influx in Human Sperm.

Prostaglandins produced by ovarian follicular cells and locally within the oviduct are assumed to be involved in the process of ovulation and in the regulation of tubal function, respectively (24, 25). Therefore, we set out to study the effect of prostanoids on human sperm. PGE₁ induced a rapid and transient increase in [Ca²⁺]_i in capacitated human spermatozoa displaying a kinetic profile similar to that evoked by progesterone (Fig. 1 A and B). The peak response was reached within 10–20 sec. Sequential application of maximally effective concentrations of PGE₁ and progesterone did not lead to cross-desensitization but resulted in two independent Ca²⁺ transients unaffected by the order in which agonists were applied (Fig. 1 A and B). A costimulation of sperm with PGE₁ and progesterone had an additive effect with respect to the peak [Ca²⁺]_i. The decline of elevated [Ca²⁺]_i to baseline levels, however, was substantially retarded (Fig. 1C). On the contrary, a second challenge of sperm with 17α-hydroxyprogesterone (1 μM) subsequent to a primary stimulation with progesterone (1 μM) did not give rise to a second Ca²⁺ transient, and the simultaneous addition of both steroids at maximally effective concentrations (1 μM) did not yield a synergistic response (data not shown). Similar results were obtained on sequential or simultaneous application of PGE₁ and PGE₂. These data indicate that distinct signaling mechanisms are responsible for progesterone- and PGE₁-mediated Ca²⁺ influx in human sperm.

Figure 1.

Time courses of intracellular Ca²⁺ concentrations in human spermatozoa stimulated with different agonists. (A and B) Human sperm suspensions were loaded with fluo-3/AM and stimulated with 1 μM concentrations of PGE₁ and progesterone as indicated. (C) Recordings of intracellular Ca²⁺ concentrations after stimulation with 1 μM PGE₁ (1) or 1 μM progesterone (2) were superimposed. Trace 3 was obtained by simultaneous application of both agonists. The dotted lined represents a calculated addition of increases in [Ca²⁺]_i over basal values elicited by PGE₁ and progesterone in (1) and (2), respectively. One representative experiment of five is shown. (D) Capacitated human sperm loaded with fura-2/AM were immobilized on poly(l-lysine)-coated coverslips and stimulated with PGE₁ (1 μM) and progesterone (1 μM) as indicated. Fluorescence ratios (F₃₄₀/F₃₈₀) obtained from measurements of 10 single sperm were calculated after subtraction of autofluorescence. One representative experiment of three is depicted.

To exclude a potential contribution of contaminating round cells in sperm preparations to prostaglandin-induced Ca²⁺ transients and to determine the proportion of sperm responsive to prostanoids, single fura-2-loaded sperm were examined by fluorescence microscopy. Sperm were immobilized on poly(l-lysine)-coated glass coverslips and sequentially were stimulated with PGE₁ and progesterone (Fig. 1D). Image analysis of Ca²⁺ transients in single sperm heads revealed that the cellular response to PGE₁ was ubiquitous, and all of the fluorescent sperm reacted to agonist application with Ca²⁺ transients localized over the acrosomal region (n = 64 in three experiments). More than 90% of immobilized sperm responded to a subsequent progesterone challenge (1 μM) with a second increase in [Ca²⁺]_i. The kinetics of Ca²⁺ transients, however, displayed some variability in that sustained elevations of [Ca²⁺]_i were observed after addition of the first agonist (Fig. 1D) in a minor cell fraction.

In contrast to human sperm, bull and boar spermatozoa were unaffected by PGE₁ stimulation (data not shown). These findings are in accord with results from binding studies (Fig. 2). Human sperm membranes displayed specific and saturable [³H]PGE₁ binding sites (Fig. 2A). Scatchard analysis of the data revealed a K_d value of 20.4 nM and a B_max value of 88 fmol/mg protein (Fig. 2 B and C). Specific binding was detected only in human but not in bull and boar sperm membranes (Fig. 2C).

Figure 2.

Specific binding of [³H]PGE₁ to human sperm membranes. (A) Sperm membranes (60 μg) were incubated with increasing concentrations of [³H]PGE₁, and total (○), nonspecific (▿), and specific (•) binding was determined. (B) Scatchard transformation of binding data presented in A. (C) Human, bull, and boar sperm membranes were incubated with 50 nM of [³H]PGE₁, and specific binding was determined. The dotted line indicates the upper limit of the 95% confidence interval of [³H]PGE₁ binding to glass fiber membranes. Data represent means ± SD of three independent experiments each performed in duplicate.

To characterize prostaglandin-evoked Ca²⁺ transients, various prostanoids were tested for agonistic properties on human sperm (Fig. 3A). Apart from PGE₁, PGE₂ was the only additional prostaglandin found to raise [Ca²⁺]_i in a concentration-dependent manner (Fig. 3A), whereas PGF_2α, PGD₂, PGA₁, PGI₂, and the synthetic prostanoids cicaprost, U46619 (data not shown), iloprost and sulprostone, an EP₃ and EP₁ receptor-selective agonist, were ineffective (Fig. 3A). Unexpectedly, PGE₁ was found to be more potent and effective than PGE₂ (Fig. 3A). One-half maximally effective concentrations (EC₅₀) of PGE₂ were ≈300 nM; the EC₅₀ value for PGE₁ was below 10 nM. HEL cells express both EP₁ and EP₃ receptors (26, 27), which mediate phosphoinositide breakdown and Ca²⁺ mobilization. To control for the bioactivity of agonist preparations, prostanoid-induced Ca²⁺ transients were measured in HEL cells (Fig. 3B) by using the same agonist dilutions as in experiments with human sperm. In HEL cells, PGE₁, PGE₂, and sulprostone caused concentration-dependent rises of [Ca²⁺]_i characterized by EC₅₀ values around 80 nM. Iloprost, a full agonist for IP and partial agonist for EP₁ receptors, also was effective at 10 μM (Fig. 3B). These findings indicate that human sperm express a EP receptor whose activation results in rapid Ca²⁺ transients.

Figure 3.

Pharmacological characterization of prostaglandin effects on human spermatozoa. (A) Human sperm were loaded with the fluorescent dye fluo-3/AM and stimulated with different concentrations of PGE₁ (•), PGE₂ (▾). Sulprostone (⧫) was tested at 0.1, 1, and 10 μM and iloprost (○) was tested at 1 and 10 μM. Agonist-induced increases in [Ca²⁺]_i are indicated. Data represent means of three experiments. (B) HEL cells were loaded with fluo-3/AM and stimulated with agonists used in A. Agonist-induced rises in [Ca²⁺]_i were recorded. Iloprost (○) was tested at a concentration of 10 μM only. Data represent means of three experiments.

Blockers of Ca²⁺ Transients.

The Ca²⁺ signal in HEL cells appears to rely on agonist-dependent phospholipase C activation because the aminosteroid U73122 (10 μM) completely inhibited prostanoid-induced rises of [Ca²⁺]_i (data not shown). When applied to human sperm, U73122 and its inactive analog U73343 evoked substantial and irreversible rises of [Ca²⁺]_i. However, U73122 (10 μM) reduced agonist-dependent Ca²⁺ transients in PGE₁-challenged spermatozoa by ≈50% (data not shown). Considering that known EP₃ receptors primarily couple to G_i proteins (28), human sperm were treated with 1 μg/ml of pertussis toxin (PTX) for 6 h during the capacitation period. Such a regimen has been reported previously to efficiently ADP ribosylate G_i/o proteins in human sperm (29). We observed that PTX treatment did not affect progesterone- and PGE₁-induced Ca²⁺ transients, arguing against a participation of G_i/o proteins (data not shown).

Progesterone- and PGE₁-evoked Ca²⁺ transients were carried by extracellular Ca²⁺ ions. Preincubation in Ca²⁺-free medium containing 1 mM EGTA precluded any rise of [Ca²⁺]_i. Mn²⁺-quenching experiments with fura-2-loaded sperm incubated in Ca²⁺-free medium demonstrated that progesterone- and PGE₁-induced increase in [Ca²⁺]_i mainly relied on influx of Ca²⁺ as opposed to mobilization from internal stores. In addition to Ca²⁺ and Mn²⁺, other divalent cations like Sr²⁺ and Ba²⁺ also entered human spermatozoa upon progesterone and PGE₁ stimulation. Agonist-induced Ca²⁺ influx could be attenuated effectively by lanthanids (La³⁺, Gd³⁺) and Zn²⁺ in a concentration-dependent fashion (data not shown), whereas we observed Ni²⁺, Cd²⁺, and Co²⁺ to be ineffective up to a concentration of 1 mM. The IC₅₀ values for La³⁺, Gd³⁺, and Zn²⁺ were 2, 10, and 30 μM, respectively. Zn²⁺ concentrations physiologically occurring in seminal plasma (≈2 mM; 30) completely block progesterone- and PGE₁-mediated Ca²⁺ influx (data not shown). In these cell preparations, ≈90% of sperm were viable as tested with the dye Hoechst 33258 and even high Zn²⁺ concentrations did not affect sperm viability. Blockers of voltage-gated L-type Ca²⁺ channels like nicardipine (10 μM) and methoxyverapamil (10 μM) as well as pimozide (5 μM), a blocker of T-type voltage-gated Ca²⁺ channels, did not affect progesterone- or PGE₁-elicited Ca²⁺ transients. Incubation of sperm suspensions with 100 μM 8-bromo-cAMP (100 μM) or 8-bromo-cGMP (100 μM) did not raise [Ca²⁺]_i, and agonist-induced Ca²⁺ peaks were not attenuated by l-cis-diltiazem (data not shown), proven to be an effective blocker of cGMP-gated cation channels.

Stimulation of Acrosome Reaction.

The percentage of spontaneously acrosome-reacted sperm after a capacitation period of 5–8 h was 2.2 ± 0.2 (mean ± SEM, n = 24, four independent experiments) (Fig. 4). Treatment of sperm with PGE₁ (1 μM) or progesterone (1 μM) increased the fraction of acrosome-reacted sperm by a factor of 3.4–3.7 (Fig. 4). The simultaneous application of both agonists had an additive effect, and the proportion of acrosome-reacted sperm increased to nearly 20%. The reduced potency and efficacy of PGE₂ to increase [Ca²⁺]_i was reflected in the comparably weak effect of this prostaglandin (1 μM) to stimulate AR (≈1.4-fold, Fig. 4). The G_i/o protein-activating peptide mastoparan (50 μM) and the Ca²⁺ ionophor A 23187 (10 μM) served as positive controls and induced AR in 49 and 53% of spermatozoa examined (Fig. 4).

Figure 4.

Agonist-induced acrosomal exocytosis in capacitated human sperm. Capacitated human spermatozoa were incubated for 30 min in the presence of modulators as indicated. Mastoparan (50 μM) and the Ca²⁺ ionophore A23187 (10 μM) served as positive controls. All of the other agents were applied at 1-μM concentrations. The percentage of acrosome-reacted sperm is indicated. Two hundred sperm were evaluated per single assay. Data represent means ± SEM (n = 16–32) of four independent experiments.

Capacitation had no effect on the ability of progesterone and PGE₁ to elicit a rapid Ca²⁺ transient in human sperm (data not shown) but markedly affected the ability of these agents to evoke AR (Fig. 5 A–C). We observed a time-dependent increase in the proportion of sperm having undergone AR in response to progesterone and PGE₁. Optimal effects were reached after capacitating spermatozoa for 6 h (Fig. 5C). At all of the time points tested, however, costimulation of sperm with both agonists yielded additive results. When sperm were stimulated in bicarbonate, phosphate, and sulfate ion-free BWW medium substituted with chloride ions, results were similar to those obtained in original BWW medium (Fig. 5D) demonstrating that bicarbonate is not required for progesterone- and PGE₁-induced AR in capacitated human spermatozoa. In the presence of 200 μM Zn²⁺, however, progesterone- and PGE₁-induced AR was abolished completely (Fig. 5E).

Figure 5.

Effect of capacitation and Zn²⁺ on agonist-induced AR. Swim-up human sperm were capacitated for 2 (A), 4 (B), or 6 (C) h in BWW-FCS medium, incubated for 30 min with progesterone (1 μM), PGE₁ (1 μM), or a combination of the two agonists (application of agonist indicated by +), and the percentage of acrosome-reacted sperm was determined as outlined in the legend of Fig. 4. (D) After a 6-h capacitation period in BWW-FCS, sperm were washed and resuspended in BWW-chloride ions. (E) Sperm were treated as in D but were stimulated in the presence of 200 μM Zn²⁺. Data represent means ± SEM of four independent experiments each performed in quadruplicate.

cAMP Levels in Human Sperm.

Because the known EP₂ and EP₄ receptors couple to the G_s/adenylyl cyclase system in somatic cells, we examined the effect of PGE₁ and progesterone on intracellular cAMP formation in human sperm. The cAMP content of unstimulated cells was 101 ± 15 (after 5 min) and 117 ± 7 fmol per 10⁶ sperm (after 15 min; mean ± SEM, n = 3). Intracellular cAMP concentrations were not altered significantly 5 or 15 min subsequent to stimulation of cells with either 1 μM PGE₁ (5 min: 102 ± 7; 15 min: 101 ± 7 fmol/0⁶ sperm) or 1 μM progesterone (5 min: 102 ± 14; 15 min: 113 ± 16 fmol/10⁶ sperm).

PTX-Insensitive G proteins in Human Sperm Membranes.

Because PGE₁-dependent Ca²⁺ transients were unaffected by PTX pretreatment, we hypothesized that PTX-insensitive G proteins may be involved in PGE₁-evoked cellular effects. Immunoblotting with an antiserum specific for the G protein α subunits α_q and α₁₁ (AS 368) revealed the expression of corresponding 42-kDa proteins in sperm and in L cell membranes, the latter serving as a positive control (Fig. 6A). Under the separation conditions chosen, the upper band represents α₁₁ and the lower band represents α_q. Immunoblotting of membranes with two antisera raised against the C-terminal sequence of α₁₂ (AS 233) and α₁₃ (AS 343) revealed one distinct protein for each antibody in L cell membranes, whereas no specific signal was recognized in equivalent amounts (50 μg) of sperm membranes (Fig. 6A). Thus, human sperm do not express appreciable amounts of G proteins belonging to the G_12/13 family. In addition, we did not obtain any indication for the expression of Gα_s in human spermatozoa.

Figure 6.

Immunoblot analysis of PTX-insensitive G proteins and PGE₁-stimulated photolabeling of G protein α subunits in human sperm membranes. (A) Membrane proteins (50 μg per lane) of human sperm (lanes 1, 3, and 5) and of mouse L cells (lanes 2, 4, and 6) were resolved by 13% SDS/PAGE, blotted, and incubated with antisera recognizing α_q/11 (lanes 1 and 2), α₁₂ (lanes 3 and 4), and α₁₃ (lanes 5 and 6). Molecular masses (kDa) are indicated on the left. (B Left) Membranes (150 μg per tube) were photolabeled with [α-³²P]GTP azidoanilide at 4°C (lane 1) or 30°C (lanes 2 and 3) in the absence (lanes 1 and 2) or presence (lane 3) of 1 μM prostaglandin E₁ and immunoprecipitated with antiserum AS 370 (α_q/11). Immunoprecipitates were resolved by 10% SDS/PAGE, and labeled α subunits were visualized by autoradiography. One of seven independent experiments is shown. (B Right) The amount of immunoprecipitated α_q/11 was determined by immunoblotting with a biotinylated IgG fraction of AS 370 and peroxydase-conjugated streptavidin. Lanes 4, 5, and 6 correspond to lanes 1, 2, and 3, respectively.

Photolabeling of Receptor-Activated G Proteins.

To study coupling of activated EP receptors to G proteins of the G_q/11 family, sperm membranes were photolabeled with [α-³²P]GTP azidoanilide in the absence and presence of 1 μM PGE₁. Labeled G protein α subunits were immunoprecipitated with a specific antiserum and resolved by SDS/PAGE (Fig. 6B Left). In human sperm membranes, PGE₁ led to increased incorporation of radioactivity into Gα_q/11 proteins (239 ± 43%, mean ± SEM, n = 7). Thus, we show that in human spermatozoa PGE₁ activates G_q/11 proteins. Probing the immunoprecipitates with a biotinylated Gα_q/11-specific antibody showed that different loads of immunoprecipitated protein were not responsible for the autoradiographic results (Fig. 6B Right). Incubation of membranes with 1 μM progesterone did not result in increased photolabeling of Gα_q/11 proteins (data not shown).

DISCUSSION

To date, ZP proteins and progesterone are regarded as physiological stimuli leading to AR in human sperm (5, 6). There is evidence that multiple binding proteins or receptors function to mediate the dual biological functions of ZP3, i.e., sperm binding and induction of the AR (5). Signal transduction cascades that are initiated following binding of sperm to ZP3 employ distinct sperm G_i proteins as critical transducing elements (29, 31, 32). On the contrary, discrete signaling steps set in motion by progesterone acting at a plasma membrane receptor are less understood, and a diverse array of molecular mechanisms has been proposed (33). In contrast to stimulation of sperm by ZP, there is currently no evidence for the involvement of G proteins in the progesterone-induced Ca²⁺ influx in human sperm (34, 35). The present study proposes the concept of E prostaglandins as physiological inducers of the AR in human sperm. We demonstrate that both progesterone and PGE₁ evoke rapid Ca²⁺ transients with similar pharmacological properties, thus indicating that progesterone- and PGE₁-elicited signaling events converge on a common distal signaling molecule, for instance a Ca²⁺-permeable cation channel. In accord with recent observations on the percentage of human sperm responding to progesterone (36), we detected that the cellular reaction to PGE₁ challenge also was ubiquitous. The observation that the two agonists do not cross-desensitize indicates that progesterone and PGE₁ exert their rapid action via different receptors.

Early reports on hamster and guinea pig sperm suggested a role for E prostaglandins and PGF_2α for the mammalian sperm AR (37, 38). Applying the Ca²⁺ indicator quin-2, Aitken et al. (39) observed a rise of [Ca²⁺]_i in human sperm subsequent to exposure to PGE₁ and PGE₂. Because these effects were detected with high prostaglandin concentrations (170 μM) only, an ionophore-like action of E prostaglandins was proposed. On the contrary, we present evidence for the expression of a G protein-coupled EP prostaglandin receptor in human spermatozoa. Among the cloned prostanoid receptors, only the IP receptor shows a predilection for PGE₁ over PGE₂ (28, 40). The ineffectiveness of the IP receptor-selective agonist cicaprost, however, excludes this receptor species as a candidate for the PGE₁ action on human sperm. Because PGF_2α, PGD₂, and U-46619 have no agonistic properties on human sperm, F prostaglandin, D prostaglandin, and thromboxane receptors do not have to be considered either. All of the known EP receptors, however, display a similar or even lower affinity for PGE₁ as compared with PGE₂ (28, 40). In addition, the complete lack of a functional response upon iloprost stimulation rules out the EP₁ receptor because this compound behaves as a partial EP₁ receptor agonist in somatic cells (40), and the ineffectiveness of sulprostone strongly argues against EP₃ receptors. Thus, the biological effect of E prostaglandins on human sperm most likely is mediated by an unknown, PGE₁-preferring EP receptor. PGE₁ not only evokes rapid Ca²⁺ transients in human sperm, but also is capable of inducing AR. The simultaneous application of PGE₁ and progesterone had an additive effect on the induction of AR in capacitated spermatozoa. Whereas capacitation was not required for progesterone (34) and PGE₁ to elicit rapid Ca²⁺ transients, agonist-induced AR strictly was capacitation-dependent.

In accord with early findings by Blackmore et al. (34), we found that concentrations of blockers of L-type voltage-gated Ca²⁺ channels like dihydropyridines and benzothiazepines in excess of what is required to block classical Ca²⁺ channels had no significant effect on progesterone- and PGE₁-induced Ca²⁺ transients. Whereas pimozide, a selective blocker of T-type Ca²⁺ channels, effectively attenuated increases in [Ca²⁺]_i in ZP3-stimulated capacitated mouse sperm (41), this compound did not affect Ca²⁺ transients in human spermatozoa incubated with progesterone and PGE₁. In summary, these results do not favor the concept of L- or T-type Ca²⁺ channels in rapid Ca²⁺ transients induced by progesterone and PGE₁. Incubation of capacitated human sperm with membrane-permeable cyclic nucleotides did not evoke Ca²⁺ influx, and l-cis-diltiazem, a selective blocker of cyclic nucleotide-gated cation channels, was ineffective in decreasing agonist-elicited [Ca²⁺]_i rises. Thus, cyclic nucleotide-gated channels appear not to be involved in the generation of progesterone- and PGE₁-induced Ca²⁺ transients. Our own results described here as well as observations by Foresta et al. (35) that progesterone causes Na⁺ influx accompanied by depolarization and Ca²⁺ influx through the same channel, indicate that progesterone and PGE₁ both open nonselective cation channels that can be blocked by lanthanids and Zn²⁺.

Mammalian sperm express PTX-sensitive G proteins of the G_i/o-family as well as G_z (10). In the present study, we analyzed the expression of PTX-insensitive G proteins in membranes from human sperm and showed that Gα_q and Gα₁₁ can be detected in sperm membranes whereas no appreciable amounts of G proteins of the G_12/13 family nor G_s were observed. Photolabeling experiments showed that in human sperm membranes the activated EP receptor stimulates G proteins of the G_q/11 family. Thus, in human sperm, classical receptor G protein-meditated signaling cascades are employed to achieve a biological effect.

In human semen, spermatozoa are exposed to high PGE₁ concentrations of ≈80 μM (30). The simultaneous presence of mM Zn²⁺ concentrations protects sperm from a massive Ca²⁺ influx which has been shown to profoundly impair the fertilizing potential of spermatozoa (36). It has been suggested that Zn²⁺ may stabilize sperm membranes during storage and ejaculation, and that Zn²⁺ removal may be required to prepare sperm for fertilization (42). The latter notion is supported by the fact that incubation of sperm with Zn²⁺ chelators capacitates hamster sperm (43). An inhibitory effect of Zn²⁺ on human sperm motility and AR has been observed (44). Our findings indicate a role of Zn²⁺ as an endogenous cation channel blocker to protect and maintain spermatozoa in a transitory quiescent state. During their ascent in the female reproductive tract, spermatozoa escape millimolar seminal fluid Zn²⁺ concentrations and are further on exposed to serum levels of Zn²⁺ (≈20 μM), nearly all of which is bound to albumin. When approaching the ovulated egg, sperm are exposed again to rising concentrations of E prostaglandins produced locally within the fallopian tube and by granulosa cells surrounding the oocyte (24). Thus, in the vicinity of the oocyte, a simultaneous stimulation of spermatozoa by E prostaglandins and by micromolar concentrations of progesterone within the cumulus oophorus may serve to prime spermatozoa to undergo effectively AR after binding to the ZP.

ABBREVIATIONS

AR

acrosome reaction

PG

prostaglandin

PTX

pertussis toxin

ZP

zona pellucida

ZP3

zona pellucida protein 3

EP

E prostanoid

HEL

human erythroleukemia

BWW

Biggers Whitten and Whittingham