Nicotinic Acetylcholine Receptor Subunits and Associated Proteins in Human Sperm*

We demonstrated previously the involvement of a nicotinic acetylcholine receptor containing an (cid:1) 7 subunit in the human sperm acrosome reaction (a modified exo-cytotic event essential to fertilization). Here we report the presence in human sperm of (cid:1) 7, (cid:1) 9, (cid:1) 3, (cid:1) 5, and (cid:2) 4 nicotinic acetylcholine receptor subunits and the following proteins known to be associated with the receptor in the somatic cell: rapsyn and the tyrosine kinases c-SRC and FYN. The (cid:1) 7 subunit appears to exist as a homomer in the posterior post-acrosomal and neck regions of sperm and is probably linked to the cytoskeleton via rapsyn. The (cid:1) 3, (cid:1) 5, and (cid:2) 4 subunits are present in the sperm flagellar mid-piece of sperm and possibly exist as (cid:1) 3 (cid:1) 5 (cid:2) 4 and/or (cid:1) 3 (cid:2) 4 channels. The (cid:1) 9 subunit is present in the sperm mid-piece. We detected the FYN and c-SRC tyrosine kinases in the flagellar mid-piece region. Both co-precipitated only with the nicotinic acetylcholine receptor (cid:2) 4 subunit. Immunolocalization with a C-terminal SRC kinase antibody, which recognizes several members of SRC kinase family, detected a SRC kinase co-localized with the (cid:1) antibody for localization of the receptor as described above, and the remaining sperm were assessed for acrosomal status using concanavalin A-FITC (45). We also compared (cid:1) 7 subunit localization before and after treatment with cytochalasin D. Capacitated sperm were treated with 20 (cid:3) M cytochalasin D an inhibitor of actin polymerization fo r 3 h asdescribed previously (46), and sperm were then stained with mAb306 antibody for localization of the receptor as described above. We identified regions of localization on the basis of well established descriptions of human sperm morphology (47).

Many proteins have been shown to be functionally associated with the nAChR (8). One such protein is a 43-kDa peripheral membrane protein, rapsyn (8), that is involved in the association of the receptor to the cytoskeleton (9 -11). Rapsyn coprecipitates with the receptor and interacts with all the subunits (12,13) and is essential for nAChR clustering in muscle (14). Rapsyn has also been detected in non-muscle cells, including neurons of the ciliary ganglia, fibroblasts, cardiac cells, and Leydig cells (15)(16)(17).
Protein tyrosine phosphorylation plays an important role in the clustering and cytoskeletal anchoring of the receptor at the neuromuscular junction (18,19), and the SRC family of kinases has been reported to be involved in this mechanism (20). The tyrosine kinases c-SRC and FYN associate with the ␣3␤4 receptor in chromaffin cells and are involved in the cholinergic stimulation of catecholamine secretion by those cells (21,22). In cortical neurons FYN associates with the ␣7 subunit (23). Among the SRC family kinases, c-YES is present in the acrosomal region of sperm (24,25) and HCK in sperm extracts (26). Other members of SRC family kinases, c-SRC, FYN, BLK, FRK, and LCK, are present in testis (see Refs. 27 and 28 and the NCBI data base).
There are several lines of evidence that support the presence of nAChRs in mammalian sperm. ␣-Bungarotoxin has been shown earlier to inhibit motility in sperm, and its binding sites were localized to the head and tail of sperm (29 -31). By using a fluorescent ligand coupled to ␣-bungarotoxin, Bacetti et al. (32) 2 have shown that nicotinic receptor-like molecules are present in the post-acrosomal and mid-piece region of the rabbit, ram, and human sperm. Partial transcripts for the ␣5, ␣7, and ␤4 subunits have been detected in human testis (NCBI data base).
The mammalian acrosome reaction (AR), a modified exocytotic event involving fusion of the outer membrane of the sperm head secretory granule-like organelle known as the acrosome with the overlying plasma membrane, is an essential step in fertilization (33,34). Recent studies have shown the importance of sperm nAChRs to the AR. It is generally accepted that the AR is initiated in vivo by the glycoprotein ZP3 (ZPC) found in the egg envelope known as the zona pellucida (34). Bray et al. (35) and Son and Meizel (36) have used the nAChR antagonists ␣-bungarotoxin, ␣-conotoxin IMI, and methyllycaconitine to show that a receptor containing the ␣7 subunit (␣7nAChR) is involved in the AR. These antagonists inhibited the AR initiated by acetylcholine in mouse and human sperm, by the mouse egg envelope in mouse sperm, and by recombinant human ZP3 in human sperm.
The current study was undertaken to further characterize the nAChRs in human sperm, including the presence and localization of nAChR subunits and signaling molecules associated with them.

Methods
Human Sperm Preparation and Capacitation-Protocols for human sperm studies were approved by the Human Subjects Committee at the University of California, Davis. Semen samples were obtained by masturbation from a pool of healthy donors. A population of Ͼ95% motile sperm was obtained by centrifugation of semen samples through a discontinuous Percoll gradient and subsequent washing as described previously (37). For capacitation, the washed sperm were diluted to 6 ϫ 10 6 sperm/ml in a medium containing a balanced salt solution, 26 mg/ml BSA, 25 mM bicarbonate, metabolites (lactate, pyruvate, and glucose), and antibiotics (penicillin and streptomycin) and capacitated by incubation of 500-l aliquots in 15-ml polypropylene centrifuge tubes for 24 h at 37°C in a 5% CO 2 /air atmosphere (37). Objective counts of the percentage of motile sperm and subjective estimates of sperm quality (using a range of 1 (twitching nonprogressive motion) to 4 (most motile sperm displayed vigorous forward motility) were carried out as described previously (37).
Antibodies-The antibodies used for Western blot analyses and immunolocalizations described in this paper were as follows: mAb306 antibody specific for the ␣7 subunit of acetylcholine receptor was purchased from Sigma; mouse monoclonal antibody to rapsyn, rap-1579 (for immunolocalization), was a gift from Dr. Stanley Froehner (University of Washington, Seattle) (38); rabbit polyclonal antibody to rapsyn, rap-1 (for Western blotting), was a gift from Dr. Michael Ferns (University of California, Davis) (12); rabbit polyclonal antibody to ␤4 subunit was a gift from Dr. John Forsayeth (University of California, San Francisco) (39); rabbit polyclonal antibodies to ␣9, ␣3, ␣5, and ␤2 subunits were gifts from Dr. Sergei Grando and Dr. Juan Arredondo (University of California, Davis) (40); rabbit polyclonal antibodies to N-terminal c-SRC and FYN were purchased from BioLegend (San Diego, CA); rabbit polyclonal C-terminal c-SRC sc18 antibody (from Santa Cruz Biotechnology) was a gift from Dr. Fumio Matsumura (University of California, Davis); fluorescein isothiocyanate (FITC)-goat anti-mouse IgG was purchased from Caltag Laboratories (Burlingame, CA); FITCgoat anti-rabbit IgG was purchased from Zymed Laboratories Inc. Hybridoma supernatant containing IgG against an ␣-115-kDa bovine lens intermediate filament-like cytoskeletal protein was a gift from Dr. Paul Fitzgerald (University of California, Davis) (41). All Western blots and immunolocalizations using these antibodies shown in the present report are representative of at least three separate experiments.
␣-Conotoxin IMI-Sepharose Batch Affinity Purification of the nAChRs-The sperm preparations were obtained by Percoll gradient centrifugation and washed with medium containing 1 mg/ml PVA as described previously (42). The final pellet was resuspended in medium A (10 mM Tris, pH 7.4, 150 mM NaCl, plus the protease inhibitors 1 mM EDTA, 20 M leupeptin, 1 g/ml aprotinin, 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride, 1 mM benzamidine HCl, 1 M pepstatin A, and 1 M E-64) and 1% (v/v) Triton X-100 at a sperm concentration of 250 ϫ 10 6 cells/ml. The sperm suspensions were incubated on ice for 1 h and centrifuged at 10,000 ϫ g for 30 min. NHS-␣conotoxin IMI-Sepharose slurry was prepared by linking 2 mol of ␣-conotoxin IMI/ml of NHS-activated Sepharose, per the manufacturer's protocol (Amersham Biosciences). Briefly, 40 l of 50% NHS-Sepharose slurry was activated by washing in cold 1 mM HCl, washed once with 0.1 M NaHCO 3 , 0.5 M NaCl, pH 8.3 (binding buffer), and incubated overnight at 4°C with ␣-conotoxin IMI. The beads were washed three times with binding buffer and incubated with 0.2 M glycine, pH 8.0, overnight at 4°C. The beads were then washed four times with binding buffer alternating with wash buffer (0.1 M acetate, 0.5 M NaCl, pH 4.0) and then washed twice with medium A plus 1% (v/v) Triton X-100. The supernatant obtained after Triton X-100 extraction of sperm was incubated with NHS-␣-conotoxin IMI slurry for 6 h at 4°C and spun at 500 ϫ g. The pellet was washed three times with medium A and resuspended in LDS sample buffer. The sample was boiled for 3 min and loaded onto a 7% precast NuPAGE Tris acetate gel, and Western blot analysis was performed as described below.
Preparation of Detergent-resistant Membranes-Human sperm was prepared by Percoll density gradient, and one-half of the sperm sample was capacitated under the conditions described above. The other half of the sample was treated as uncapacitated sperm. Both the uncapacitated and capacitated sperm samples were washed once with phosphate-buffered saline (PBS), pH 7.4, and the detergent-resistant membranes were prepared as described previously (43) with some modifications. Sperm pellets were resuspended in 500 l of 50 mM MES, pH 6.5, 0.15 M NaCl containing 0.5% Triton X-100 and protease inhibitors as described above for medium A. The cells were solubilized for 20 min at 4°C, homogenized with 10 strokes of a Dounce tight-fitting homogenizer, and passed five times through a 22-gauge needle. The homogenates were adjusted to 40% sucrose in MES plus protease inhibitors in a final volume of 1 ml, placed at the bottom of a 4-ml ultracentrifuge tube, and overlaid with 1 ml of 30% sucrose and 1 ml of 5% sucrose. The gradients were centrifuged at 42,000 rpm for 20 h in an SW60 Ti rotor (Beckman Instruments, Fullerton, CA). 500-l fractions were collected from the top and added to the ␣-conotoxin IMI slurry prepared as described above and analyzed by Western blotting as described below.
Immunoprecipitation of SRC and FYN-For immunoprecipitation of the kinases, the supernatant from the Triton X-100 extraction of ϳ75 ϫ 10 6 cells (as described above) was pre-cleared (to remove proteins that bind nonspecifically to protein A) by incubating the lysate with 25 l of 50% protein A for 1 h at 4°C and then spun at 10,000 ϫ g for 2 min. Then 1 g of antibody was added to the supernatant, and the mixture was incubated for 4 h at 4°C. Subsequently, a 25-l aliquot of the immobilized protein A (50% slurry) was added, and the suspension was further incubated for 1 h at 4°C and then spun at 10,000 ϫ g for 30 s. The pellet was washed three times with medium A and resuspended in LDS sample buffer. The sample was boiled for 3 min and loaded onto a 7% precast NuPAGE Tris acetate gel, and Western blot analysis was performed as described below.
Western Blot Analysis-For analysis of total sperm lysate, sperm preparations were obtained by Percoll gradient centrifugation and washed (all solutions without BSA but with 1 mg/ml PVA) as described previously (42). The final pellet was resuspended in LDS sample buffer and boiled for 3 min. The suspension was passed through a 22-gauge needle and spun at 10,000 ϫ g for 10 min. For immunoblotting, the supernatant was loaded onto a 7% Tris acetate gel, and the proteins were transferred onto nitrocellulose membrane at 100 mA constant current for 2 h. The membrane was blocked for 2 h with 3% BSA in TBST (10 mM Tris, pH 7.4, 150 mM NaCl, 0.05% Tween 20) for anti-␣7 antibody or 5% Blotto in TBST for all other antibodies and then incubated overnight at 4°C with a 1:5000 dilution of primary antibody diluted in blocking buffer. The membrane was washed three times with TBST and incubated with a 1:2000 dilution of sheep anti-mouse or donkey anti-rabbit antibody conjugated to horseradish peroxidase for 1 h at room temperature. The membrane was subsequently washed extensively with TBST and developed using the Amersham Biosciences ECL kit using Hyperfilm ECL. The films were scanned using a flatbed scanner (Canoscan N1240U, Canon Inc., Lake Success, NY), and Adobe Photoshop (Adobe Systems Inc., San Jose, CA) was used to crop and adjust the intensity levels to optimize contrast.
Immunolocalization of the Subunits of Acetylcholine Receptor, c-SRC, and FYN-For immunolocalization experiments, uncapacitated and capacitated human sperm were fixed with 4% paraformaldehyde (PFA) (for ␣7, rapsyn, ␣3, ␣5, and ␣9 subunits, and c-SRC sc-18) or with 1% PFA (for ␤4 subunit, c-SRC, and FYN) for 10 min at room temperature. The cells were then washed three times with PBS, pH 7.4, and plated on poly-L-lysine-coated slides. The cells were allowed to stick for 15 min, washed twice with PBS, incubated with 0.1% Triton X-100 for 5 min, and washed four times at 5 min each with PBS. The slides were then blocked with 0.2% gelatin in PBS for 2 h and incubated overnight at 4°C with primary antibody: 100 g/ml of mAb306, 100 g/ml of control ␣-115 antibody; 1:500 dilution of rap1579 antibody; 1:250 dilution of ␣3, ␣5, and ␣9 subunit antibodies; 1:500 dilution of ␤4 antibody; and 50 g/ml of c-SRC sc18, c-SRC, and FYN antibodies. The slides were then washed with PBS, incubated for 1 h with 6 g/ml FITC-rabbit antimouse IgG or 10 g/ml FITC-goat anti-rabbit IgG, and then washed as described above. The slides were then dried and mounted with Vectashield, and sperm was examined using a Nikon E800 epifluorescence microscope at ϫ1000 magnification equipped with Q imaging and a digital Retiga 1300i camera. Adobe Photoshop was used to crop the images, and the intensity levels of experimental and control were adjusted equally to optimize contrast.
We also compared ␣7 subunit localization before and after treatment with an AR initiator, the Ca 2ϩ ionophore ionomycin. Capacitated sperm were incubated with 3 M ionomycin for 5 min as described previously (44). At the end of the reaction, 5-l aliquots were removed to check for motility (37). One-half of the sperm were stained with mAb306 antibody for localization of the receptor as described above, and the remaining sperm were assessed for acrosomal status using concanavalin A-FITC (45). We also compared ␣7 subunit localization before and after treatment with cytochalasin D. Capacitated sperm were treated with 20 M cytochalasin D an inhibitor of actin polymerization for 3 h as described previously (46), and sperm were then stained with mAb306 antibody for localization of the receptor as described above. We identified regions of localization on the basis of well established descriptions of human sperm morphology (47).
Preparation of ␣-Conotoxin IMI-Rhodamine Conjugate and Localization with the Conjugate-Preparation of the ␣-conotoxin IMI-rhodamine conjugate was performed as described previously for conjugation of proteins to fluorescent probes (48,49). Briefly, 0.1 mol of ice-cold ␣-conotoxin IMI was added dropwise with mixing to 1 mol of ice-cold NHS-rhodamine in 0.1 M KPO 4 , pH 7.0, in a 150-l reaction mixture and incubated at room temperature for 4 h. The conjugate was partially purified on a Sephadex G-15 (25 ϫ 1 cm) column equilibrated in 50 mM ammonium acetate buffer, pH 7.0. Fractions (1 ml each) were collected at a flow rate of 0.5 ml/min, and the fractions in the void volume that had the peak 214 nm absorbance were pooled and lyophilized. The conjugate was resuspended in 250 l of PBS, pH 7.4, and checked by mass spectrometry for labeling (Molecular Structure Facility, University of California, Davis). The molecular mass was ϳ1763 Da as expected.
For localization with the conjugate, the cells were fixed with either 2 or 4% PFA, and 1:100 dilution of the conjugate was added followed by overnight incubation at 4°C. The cells were washed with PBS and mounted as described above for the immunolocalization of receptor subunits. For control, the cells were incubated with the conjugate together with 2.5 M unlabeled ␣-conotoxin IMI. Localization with ␣-bungarotoxin-rhodamine conjugate was performed as described earlier (32) with modifications. Briefly, capacitated sperm were fixed with either 2 or 4% PFA, and 5 g/ml of conjugate was added and incubated overnight at 4°C. For control, the cells were first preincubated for 2 h with 50 g/ml unlabeled ␣-bungarotoxin. The cells were rinsed with PBS and mounted as described above, and the images were processed as described above for immunolocalization.
Tyrosine Kinase Inhibitors Lavendustin A and PP2 Treatment of Sperm and AR Assay-Sperm samples were prepared and capacitated under conditions described above. Aliquots of capacitated sperm (100 l each) were first preincubated with either lavendustin A or PP2 or with their inactive analogs lavendustin B or PP3, respectively, for 5 min followed by the addition of 250 M acetylcholine and further incubation for 10 min. The reaction was stopped by adding 4% formaldehyde, and at least 200 sperm per sample were assessed for acrosomal status in a blind fashion using concanavalin A-FITC (45). AR percentage data were transformed to the arcsine of their square roots. The Duncan new multiple range test was used to compare group mean differences, and statistical significance was determined at p Ͻ 0.05.

RESULTS
Presence of ␣7 Subunit and Its Association with the Cytoskeleton-A 57 Ϯ 1.2-kDa ␣7 subunit was detected in the human sperm extracts by Western blotting, after NHS-Sepharose-␣-conotoxin IMI batch affinity purification (Fig. 1, panel a, lane 1). No protein bound to the NHS-Sepharose control (Fig. 1,  panel a, lane 2).
Immunofluorescence experiments using an ␣7 subunit-specific antibody showed that this subunit was localized in the neck region and posterior post-acrosomal region in both uncapacitated and capacitated sperm (Fig. 1, panels b and c, respectively). No signal was detected when using a control antibody or when the primary antibody was omitted (Fig. 1, panels d and  e, respectively). To test if there was a change in localization after the AR, capacitated sperm were treated with ionomycin and then immunostained. The percentages of sperm acrosome reacted in those experiments were as follows: 33.1 Ϯ 2.8 and 11.6 Ϯ 0.23 for ionomycin and Me 2 SO-treated sperm, respectively (n ϭ 3). The motility and quality of the sperm were ϳ75% and 3, respectively, for both ionomycin-treated and control sperm. No change in the localization of the ␣7 subunit was observed in acrosome-reacted sperm samples (Fig. 1, panel g) compared with its Me 2 SO-treated control (Fig. 1, panel f).
To determine whether the sperm ␣7 subunit was present in lipid rafts, membrane fractions from a sucrose density step gradient of uncapacitated and capacitated sperm were purified by ␣-conotoxin IMI affinity purification and analyzed by Western blotting. In uncapacitated sperm, the ␣7 subunit was present in the 5% low density membrane fraction 2 (Fig. 1, panel h). For Western blot analysis of the ␣7 subunit in sucrose density gradient membrane fractions, the fractions were batch-purified using ␣-conotoxin IMI conjugated to NHS-Sepharose, and eluates were run on an SDS-polyacrylamide gel and blotted with ␣7 subunit-specific antibody. Panel h is uncapacitated sperm, and panel i is capacitated sperm.
In capacitated sperm, the ␣7 subunit was absent in the low density fractions but present in the high density membrane fractions (Fig. 1, panel i).
To test whether actin was involved in the immobility of the ␣7 subunit, capacitated sperm were treated with cytochalasin D, an inhibitor of actin polymerization. This treatment led to a very diffuse localization of the receptor (Fig. 1, panel k) compared with its Me 2 SO-treated control (Fig. 1, panel j).
Western blotting of whole sperm cell lysates showed the presence of rapsyn, which runs at the expected molecular mass of 43 kDa (Fig. 2, panel a, lane 1) similar to the control lysate containing rapsyn (Fig. 2, panel a, lane 2). Immunolocalization of rapsyn in capacitated sperm showed that it is co-localized with the ␣7 subunit in the neck region and is also present in the flagellar mid-piece of sperm (Fig. 2, panel b). Cells stained with secondary antibody alone showed no signal (Fig. 2, panel c).
Presence of ␣9 Subunit-Localization studies with ␣-conotoxin IMI-rhodamine and ␣-bungarotoxin-rhodamine conjugate probes showed the presence of nAChRs in the posterior post-acrosomal and neck regions of capacitated sperm with 2% PFA fixation (Fig. 3, panels a and c, respectively) and neck and mid-piece regions of capacitated sperm with 4% PFA fixation (Fig. 3, panels e and g, respectively). Control samples where samples were incubated with unconjugated probes together with labeled probes showed no signal (Fig. 3, panels b, d, f,  and h).
Western blot analysis of proteins bound to ␣-conotoxin IMI-Sepharose showed the presence of a ␣9 subunit with a molecular mass of 61.2 Ϯ 0.81 (Fig. 3, panel i). Immunolocalization of the ␣9 subunit showed that it is present in the mid-piece region of sperm (Fig. 3, panel j). No signal was detected when the primary antibody was omitted (Fig. 3, panel k).
Presence of ␣3, ␣5, and ␤4 Subunits and Binding to ␣-Conotoxin IMI-Western blotting of whole sperm cell lysates showed the presence of an ␣3 subunit that runs at a molecular mass of 56.3 Ϯ 2.2-kDa, an ␣5 subunit that runs at a molecular mass of 51.7 Ϯ 2.5 kDa, and a ␤4 subunit that runs at a molecular mass of 66 kDa (Fig. 4, panel a, lanes 1-3, respectively). The ␣3, ␣5, and ␤4 bands were not detected when the primary antibody was omitted during Western blotting (Fig. 4, panel a, lane 5). The ␤2 subunit was not detected in the lysates (Fig. 4, panel a,  lanes 4). A higher molecular weight band seen with the ␣5 antibody may be due to cross-reactivity of the antibody. Immu-nolocalization studies showed that ␣3, ␣5, and ␤4 subunits are present in the mid-piece region of the sperm (Fig. 4, panels b-d,  respectively). Control samples in which primary antibody was not included showed no signal (Fig. 4, panel e). Western blotting of proteins bound to ␣-conotoxin IMI-Sepharose showed that in addition to the ␣7 and ␣9 subunits, as reported above, both the ␣3 and ␤4 subunits bound to the ␣-conotoxin IMI (Fig.  4, panel f).
Involvement of Tyrosine Phosphorylation in the Acetylcholine-initiated AR-Lavendustin A, a tyrosine kinase inhibitor, but not its inactive analog lavendustin B (50) significantly inhibited the acetylcholine-initiated AR at concentrations ranging from 2 to 10 nM (Fig. 5, panel a). PP2, a SRC kinase inhibitor, but not PP3, its inactive analog (51), also significantly inhibited the acetylcholine-initiated AR at a concentration of 10 nM (Fig. 5, panel b).
Presence of c-SRC and FYN Tyrosine Kinase-Both c-SRC and FYN kinases co-elute with the nicotinic receptor subunits during ␣-conotoxin IMI affinity purification in the low density membrane fraction 2 (Fig. 1, panel h) along with the ␤4 subunit and ran at a molecular mass of 66 kDa (Fig. 6, panel a). Immunoprecipitation studies, using antibodies specific to the N-terminal region of the kinases, also showed the presence of FYN (Fig. 6, panel b, lane 1) and c-SRC (Fig. 6, panel c, lane 1) that ran at a molecular mass of 66 kDa. Those studies also showed that both FYN and c-SRC are associated with the ␤4 subunit (Fig. 6, panel b and c, lane 3) but not with ␣7 and ␣9 subunits (Fig. 6, panels b and c, lanes 2 and 4, respectively). Only rabbit IgG was detected in the anti-rabbit secondary antibody-alone control (Fig. 6, panels b and c, lane 5) and nothing at all in the anti-mouse secondary antibody-alone control (Fig. 6, panels b and c, lane 6). Immunolocalization studies using the same antibodies showed that both c-SRC and FYN are present in the flagellar mid-piece region of the sperm (Fig.  6, panels d and e, respectively) co-localized with the ␤4 subunit as observed in Fig. 4, panel d. Control sample where primary antibody was omitted showed no signal (Fig. 6, panel f).
Association of a SRC Kinase with the ␣7 Subunit-Immunolocalization studies using a c-SRC sc-18 antibody against the C-terminal region that recognizes several members of the SRC family showed the presence of SRC kinases in the acrosome, neck, and mid-piece regions of sperm (Fig. 7, panel a). Control sample where primary antibody was omitted showed no signal (Fig. 7, panel b). Immunoprecipitation studies using this antibody precipitated the ␣7 subunit (Fig. 7, panel c, lane 1). This band was not detected in the anti-mouse secondary antibodyalone control (Fig. 7, panel c, lane 2). DISCUSSION Our Western blotting and cytochemical results provide the first direct evidence for the existence of the ␣7nAChR in mammalian sperm. In this study, using an ␣-conotoxin IMI affinity purification step followed by Western blotting, we show the presence of an ␣7 subunit in human sperm with a similar molecular mass as that reported earlier for the brain ␣7 subunit (4,52). The ␣7 subunit has been reported previously to be associated with 5% low density membrane fraction containing lipid rafts in PC12 cells and ciliary neurons (53,54). Because capacitation involves the destabilization of lipid rafts (55,56), it is not surprising to find that the ␣7 subunit in capacitated sperm is in the high density membrane fractions, whereas it is in the low density membrane fraction of the uncapacitated sperm. Immunolocalization experiments showed no change in the ␣7 subunit localization in the uncapacitated versus the capacitated sperm or in acrosome-reacted sperm. The latter result is as expected because the regions of sperm containing the ␣7nAChR would not be lost during the AR (33). Our evidence suggests that the ␣7 subunit is tethered to the sperm actin cytoskeleton during capacitation. This view is supported by the absence of any difference in the ␣7 subunit immunolocalization results between uncapacitated, capacitated, and acrosome-reacted sperm, the Western blot results indicating subunit change from low density to high density membrane fractions during capacitation, and the diffuse local-ization of the receptor after cytochalasin D treatment. It has been shown previously that ␣7nAChRs in neurons of the ciliary ganglia associate with actin (57) and that actin is present in the acrosome, post-acrosomal, and neck regions in capacitated human sperm (46).
In ciliary ganglion neurons, rapsyn was reported to be involved in the cytoskeletal linkage of the ␣7 subunit (10). We  1-4, respectively) or secondary antibody alone (lane 5). The standards in decreasing molecular weight are as described in the legend of Fig. 1. Immunolocalization of the subunits were studied in capacitated sperm using the same antibodies: panel b, ␣3 subunit; panel c, ␣5 subunit; panel d, ␤4 subunit; and panel e secondary antibody-alone control. Arrows in each panel indicate localization in the flagellar midpiece of sperm. Bars ϭ 5 m. Panel f, extracts from NHS-Sepharose ␣-conotoxin IMI affinity-purified receptor, as described under "Experimental Procedures," were run on a 7% Tris acetate gel, and blots were probed with ␣3 and ␤4 subunit antibodies. show here by Western blot that rapsyn is present in sperm and co-localized with the ␣7 subunit in the neck region of the sperm. Several attempts to co-precipitate rapsyn with the ␣7 subunit failed. This could be due to low extractability because of its strong association with the insoluble cytoskeleton and in addition its high sensitivity to proteases (14). But the co-localization of rapsyn with the ␣7 subunit and the linkage of the ␣7 subunit to the cytoskeleton strongly suggest that rapsyn is highly likely to be associated with the acetylcholine receptor in the neck region of the human sperm.
The localization results with the ␣-bungarotoxin-FITC conjugate were in agreement with earlier observations in unca-pacitated sperm by Baccetti et al. (32) 2 with that probe except that the neck region localization was included as part of the postacrosomal results in those earlier studies. Previous studies have shown that both ␣7 and ␣9 subunits bind to ␣-bungarotoxin and ␣-conotoxin IMI (58,59). Our Western blot results demonstrated the presence of the ␣9 subunit with an apparent molecular mass similar to that reported earlier for keratinocytes (40). Taken together with the present immunolocalization results demonstrating the ␣7 subunit in the neck region and posterior post-acrosomal region but not in the flagellar midpiece, the results obtained with the fluorescent probe-conjugated nAChR antagonists suggested that the ␣9 subunit might be present in the mid-piece region. Immunolocalization showed that the ␣9 subunit was indeed present in the mid-piece region of sperm.
The ␣7 subunit forms a heteromeric channel with the ␤2 subunit when co-expressed in Xenopus (60). Our Western blot results did not detect the ␤2 subunit but did demonstrate the presence of ␤4, ␣3, and ␣5 subunits in sperm with molecular masses similar to those detected in the cerebellum, brain, and ciliary ganglia, respectively (39,61,62).
␣-Conotoxin IMI binds with high specificity to the ␣7 and ␣9 subunits (59). While purifying the ␣7 subunit using ␣-conotoxin IMI-Sepharose, we found the ␣3 and ␤4 subunits co-eluted with the ␣7 subunit, suggesting that it might exist as a heteromer. In bovine chromaffin cells the ␣3␤4 nAChR has been reported to bind to ␣-conotoxin IMI at a concentration of 2 M (63), and recently ␣-conotoxin IMI was shown to bind to the ␣3␤4 receptor at an IC 50 of 3.39 M (64). In our study we used a concentration of ϳ20 M; hence, it is more likely that in addition to the binding of the ␣7 subunit, the ␣3␤4 receptor also binds to ␣-conotoxin IMI at that high concentration. Linkage studies have shown the localization of ␤4 subunit in a tight gene cluster with the ␣3 and ␣5 subunits (65,66) with shared regulatory elements suggesting that all the three subunits could be co-expressed. We find that all three of these subunits are present in sperm and co-localized in the mid-piece, and none are detected in the neck or the posterior post-acrosomal region. All these results taken together suggest that the ␣7 subunit most probably exists as a homomer as reported earlier for neuronal receptor (67) in the posterior post-acrosomal and neck regions, and the ␣3, ␣5, and ␤4 subunits probably exist as an ␣3␤4 (from our ␣-conotoxin IMI affinity purification study) receptor and/or ␣3␣5␤4 receptor in the mid-piece of sperm.
Unlike the ␣7 subunit receptor, the ␣9 subunit has both nicotinic and muscarinic pharmacological properties, and like the ␣7 subunit receptor, when expressed alone in Xenopus, it forms a homomeric channel; it does not form a heteromeric channel when expressed with the ␤4 subunit (6). Hence, even though the ␣9 subunit is found co-localized with the ␤4 subunit in the mid-piece of sperm, it most likely exists as a homomer. The mid-piece region of sperm plays a major role in the motility of sperm, and there are reports that show the inhibition of sperm motility by ␣-bungarotoxin. Interaction of that antagonist with the ␣9 subunit in the mid-piece could contribute to this inhibition.
Localizations with antagonist-conjugated probes were carried out with fixed nonpermeabilized sperm, but sperm were permeabilized in immunolocalization experiments. Because nAChRs are plasma membrane receptors in neurons and muscle (68), we assume that all the nAChR subunits detected in permeabilized sperm were on the plasma membrane of the identified regions. The presence of several different nAChR subunits in the sperm flagellar mid-piece or the neck suggests that various combinations of these subunits, including the apparently homomeric ␣7nAChR, may be involved in helping to FIG. 6. Western blot analysis and immunolocalization of c-SRC and FYN tyrosine kinases. Western blot of c-SRC and FYN: panel a, extracts from ␣-conotoxin IMI affinity-purified low density membrane fraction 2 from uncapacitated sperm (as shown in Fig. 1, panel h) were blotted with ␤4, c-SRC, and FYN antibodies. Panel b, immunoprecipitates using anti-FYN antibody were loaded onto a 7% Tris acetate gel and blotted onto nitrocellulose membrane as described under "Experimental Procedures." The blots were probed with anti-FYN, ␣7, ␤4, or ␣9 antibodies (lanes 1-4, respectively) or anti-rabbit or anti-mouse secondary antibody alone (lanes 5 and 6, respectively). Panel c, immunoprecipitates using N-terminal anti-c-SRC antibody blotted with anti-c-SRC, ␣7, ␤4, or ␣9 antibodies (lanes 1-4, respectively), anti-rabbit or anti-mouse secondary antibody alone (lanes 5 and 6, respectively).  control different forms of sperm motility under different conditions in the female tract. Indeed, recent studies from this laboratory 3 have shown that sperm from mice with a defective ␣7nAChR exhibit reduced hyperactivation in vitro, a form of motility important to fertilization (69).
c-SRC and FYN, members of the SRC family of tyrosine kinases, are associated with acetylcholine receptors in neurons and chromaffin cells (20 -22). Here, both c-SRC and FYN were detected in sperm but ran at a higher molecular mass than the expected molecular mass of 60 kDa. Isoforms of both SRC and FYN have been reported (70,71), and the isoform that is present in sperm needs further characterization. An earlier report (23) showed the association of the ␣7 subunit with the FYN kinase in cortical neurons, but in the present study the FYN kinase did not associate with ␣7 or ␣9 subunits but associates with the ␤4 subunit. The presence of the ␤4 subunit in the mid-piece region along with its association with FYN and SRC kinases suggest that ␣3␤4 and/or ␣3␣5␤4 receptors could be involved in sperm motility control because tyrosine phosphorylation is known to be important to sperm motility (72). Tyrosine kinase inhibitors lavendustin A and PP2 had no effect on the percentage of motile sperm by objective estimation and on the quality of sperm by subjective estimation. Computerassisted sperm analysis may be required to detect more subtle effects on motility.
In our previous studies, we reported the involvement of the ␣7nAChR in the human AR initiated by acetylcholine or recombinant human egg zona pellucida protein ZP3 and in the mouse AR initiated by the egg zona pellucida (35,36). The ␣7nAChR in sperm could be activated directly by acetylcholine in an autocrine/paracrine fashion (73) and/or indirectly by cross-talk as a result of the activation of a zona pellucida receptor. An increase in intracellular Ca 2ϩ due to release from intracellular stores and to depolarization-mediated opening of a voltage-operated Ca 2ϩ channel is required for the mammalian AR (74,75). Because the opening of nAChRs in neurons allows an influx of cations (mainly Na ϩ along with some Ca 2ϩ ) that depolarizes the membrane (68), membrane depolarization of human sperm required for the AR may be due at least in part to the influx of Na ϩ and/or Ca 2ϩ via the ␣7nAChR (35,73). Furthermore, intracellular Ca 2ϩ stores have been reported to be present in various sites in the sperm of different mammalian species, including the acrosome, post-acrosomal, neck, and mid-piece regions of human sperm (76), the neck region of bovine sperm (77), and the acrosome of mouse sperm (78). ␣7nAChRs, shown here to be present in the post-acrosomal and neck regions of human sperm, have been reported to be highly permeable to Ca 2ϩ (79). In hippocampal astrocytes, Ca 2ϩ influx via the ␣7nAChR was sufficient to increase further the intracellular Ca 2ϩ by Ca 2ϩ -induced Ca 2ϩ release from intracellular stores involving ryanodine and inositol 1,4,5-trisphosphate receptors (80). Most interestingly, ryanodine receptor has been shown to be present in the neck region in human sperm (81). Promoting such a release of intracellular stores may be one of the functions of the ␣7nAChR in the human sperm AR.
Tyrosine phosphorylation is also important to the mammalian AR (72), and we have shown here that tyrosine phosphorylation is involved in the acetylcholine-initiated AR. Two tyrosine kinase inhibitors, lavendustin A and PP2, significantly inhibited that event at 2-10 and 10 nM, respectively. These effective concentrations are within the range reported for the inhibition of epidermal growth factor receptor tyrosine kinase by lavendustin A (50) and several SRC kinases by PP2 (82).
Those AR results, the co-localization of an SRC kinase and the ␣7nAChR in the post-acrosomal and neck regions, and the fact that the ␣7nAChR is immunoprecipitated along with the kinase by a SRC kinase antibody suggest that this kinase may be involved in the acetylcholine-initiated AR. The identity of the SRC kinase has yet to be determined, but of the 11 SRC family kinases (83), we can rule out c-SRC and FYN (based on the present study) and c-YES since it was earlier shown to be present only in the acrosomal region of sperm (25). Our results indicate that different tyrosine kinases are associated with different human sperm nAChRs. This relationship of sperm nAChRs with SRC kinases could be a key regulator of signaling pathways important to the AR and motility in human sperm.