Recombinant streptolysin O (SLO) was obtained from Dr. Bhakdi (University of Mainz, Mainz, Germany). The rabbit polyclonal antibody directed towards human α-synuclein (purified on protein A) was from Axxora, LLC (Farmingdale, NY). The mouse monoclonal anti-synaptobrevin-2 antibody (clone 69.1, purified IgG) and the rabbit polyclonal anti-complexin I/II (purified IgG) were from Synaptic Systems (Göttingen, Germany). HisTrap columns, FF, Cytiva (formerly GE healthcare Life Sciences) were purchased from ALLSCIENCE, LLC (Doral, FL). FITC-coupled Pisum sativum agglutinin (FITC-PSA) was from Vector Labs (BIOARS S.A, Buenos Aires, Argentina). PSA coupled to 20 nm colloidal gold was from glycoMatrix (Dublin, OH). Horseradish peroxidase- and Cy™3-conjugated goat anti-rabbit, as well as Cy™3-conjugated goat anti-mouse IgGs (H + L) were from Jackson ImmunoResearch (West Grove, PA). O-nitrophenyl EGTA-acetoxymethyl ester (NP-EGTA-AM) was purchased from Life Technologies (Buenos Aires, Argentina). Ni-NTA-agarose was from GE Healthcare. Prestained molecular weight markers were from Bio-Rad (Tecnolab, Buenos Aires, Argentina). All other chemicals were from Sigma-Aldrich™ Argentina S.A., Genbiotech, or Tecnolab (Buenos Aires, Argentina).

The light chains of BoNT/B and TeTx fused to His₆ in a pQE3 vector were generously provided by Dr. T. Binz (Medizinische Hochschule Hannover, Hannover, Germany). The cDNA encoding human α-synuclein fused to His₆ and to the TAT-transduction domain of HIV virus in a pET11a plasmid was a kind gift of Dr. L. Pollegioni (Università degli Studi dell’Insubria, Varese, Italy). Purified, recombinant Rab3A-22A fused to glutathione S-transferase was a kind gift from Dr. M.F. Quevedo (IHEM, Mendoza, Argentina).

Expression and purification of recombinant His₆-BoNT/B and His₆-TeTx were as in (De Blas et al., 2005). The plasmid encoding His₆-TAT-α-synuclein was transformed into Escherichia coli BL21 (DE3) and protein expression was induced with 0.5 mM IPTG 4 h at 37°C. Recombinant-α-synuclein is mostly located in the periplasmic space (Ren et al., 2007). We recovered the periplasm of transformed bacteria by osmotic shock as in (Caldinelli et al., 2013), made it 20 mM TrisHCl, pH 7.4 and run it through a metal-chelating HisTrap column. Twenty volumes of 20 mM TrisHCl, pH 7.4, 1 M NaCl, and 5 volumes each of 20 mM TrisHCl, pH 7.4 plus 125 mM, 250 mM and 500 mM imidazole were applied to the column. Purified His₆-TAT-α-synuclein eluted with 500 mM imidazole was added 10% glycerol, aliquoted and stored at −80°C until use.

Semen samples were donated by normal healthy men. Semen was allowed to liquefy by incubating at 37°C for 30–60 min. Highly motile cells were isolated from semen following a swim-up protocol. Sperm concentrations were adjusted to 7 × 10⁶/mL before incubating for at least 2 h under capacitating conditions (37°C, 5% CO₂/95% air) in Human Tubal Fluid media (Serendipia Lab, Vedia, Argentina) supplemented with 0.5% bovine serum albumin (BSA, HTF media). Sperm were permeabilized with SLO as follows: after washing twice with PBS, sperm were incubated for 15 min at 4°C in cold PBS containing 3 U/mL SLO. Cells were washed once with PBS, resuspended in ice-cold sucrose buffer (20 mM Hepes-K, pH 7, 250 mM sucrose, 0.5 mM EGTA) and permeabilized by activating the SLO bound to their plasma membrane with 2 mM DTT. The reagents to test were added sequentially as indicated in the figure legends, and incubated for 8–15 min at 37°C after each addition. Samples were processed for AE assays, immunofluorescence and for transmission electron microscopy. To obtain the results shown in Figure 2B, photosensitive NP-EGTA-AM was supplied to SLO-permeabilized sperm at the beginning of the experiments, ensuring that all subsequent procedures were carried out in the dark. After the last incubation, photolysis was induced by exposing twice (1 min each) to an UV lamp, mixing, and incubating for 5 min at 37°C.

Sperm were spotted on teflon-printed slides, air dried, and fixed/permeabilized in ice-cold methanol for 20 s. Acrosomal status was assessed by staining with 25 μg/mL FITC-PSA in PBS for 40 min at room temperature and washing for 20 min in water (Mendoza et al., 1992). PSA binds α-methyl mannoside moieties in proteins present in the acrosomal matrix (Cross et al., 1986). At least 200 cells per condition were scored manually using an upright Zeiss microscope equipped with epifluorescence optics. All experiments included a basal (no stimulation) and a positive (0.5 mM CaCl₂ [corresponding to 10 μM free calcium]) controls. AE indices were calculated by subtracting the number of spermatozoa that lost their acrosome spontaneously (“control”; assigned 0% for normalization) from all values and expressing the results as a percentage of the AE observed with 0.5 mM CaCl₂ (“Ca”, assigned 100% for normalization). Our analysis only included results derived from experiments that produced similar responses and where the difference between basal (7%–19% before normalization) and stimulated conditions was of at least 8–10 percentage points (Supplementary Figure S1B). SLO did not affect the PSA staining patterns (see images of PSA-stained SLO-permeabilized sperm in Supplementary Figure S1A; compare “-SLO and “+SLO” FITC-PSA panels in Figure 1B). Data were evaluated before normalization with the program GraphPad Prism 8, using the one way Anova, Dunnett’s multiple comparisons test. Bar plots represent the mean ± SEM of at least three independent experiments. Different letters indicate statistical significance (p < 0.05).

SLO-permeabilized sperm were bathed in 30 μg/mL FITC-PSA for all the incubation time. Cells were loaded with 5 nM anti-α-synuclein antibodies followed by 0.5 mM CaCl₂ and incubated for 15 min at 37°C after each addition. At the end of the second incubation, sperm were spotted on teflon-printed slides, air dried overnight and fixed/permeabilized in 2% paraformaldehyde for 10 min, followed by a 5 min-wash in water. AE indices and statistical analysis were as described above.

For anti-synaptobrevin-2 immunofluorescence, capacitated and SLO-permeabilized sperm samples were processed as described for indirect AE assays, adding reagents and CaCl₂ sequentially, and incubating for 10 min at 37°C after each addition. α-Synuclein immunofluorescence was performed on cells subjected or not to SLO permeabilization. Samples were fixed in 2% paraformaldehyde in PBS for 15 min at room temperature, centrifuged and resuspended in PBS containing 100 mM glycine to neutralize the fixative. Cells (5 × 10⁵ per condition) were attached to poly-L-lysine-coated, 12 mm round coverslips by incubating for 30 min at room temperature in a moisturized chamber. Exposure to 0.1% Triton X-100 in PBS for 10 min at room temperature was used to permeabilize the plasma membrane. After the detergent, sperm were washed three times with PBS containing 0.1% polyvinylpyrrolidone (PVP, average M.W. = 40,000; PBS/PVP). Incubation in 3% BSA in PBS/PVP for 1 h at 37°C was used to reduce nonspecific staining. Anti-α-synuclein (10 μg/mL) and anti-synaptobrevin-2 (25 μg/mL) antibodies were diluted in 2% BSA in PBS/PVP, added to the coverslips, and incubated for 1 h at 37°C in a moisturized chamber. After washing (three times, 6 min each) with PBS/PVP, 2.5 μg/mL Cy™3-conjugated anti-rabbit (synuclein) or mouse (synaptobrevin) IgGs in 1% BSA in PBS/PVP) were added and incubated for 1 h at room temperature protected from light. Coverslips were washed (three times, 6 min each) with PBS/PVP. The acrosomal membrane was permeabilized with ice-cold methanol for 1 min, and the acrosomal contents stained with FITC-PSA as described in “Indirect AE assay” (but without air drying). Coverslips were washed (three times, 10 min each) with PBS/PVP, mounted with Mowiol^® 4–88 in PBS containing 2 μM Hoechst 33342, and stored at room temperature in the dark. Samples were examined with an 80i Nikon microscope equipped with a Plan Apo 60x/1.40 oil objective. Images were captured with a Nikon DS-Fi1 camera operated with NIS software (Nikon). ImageJ (freeware from N.I.H.) was used to subtract background and adjust brightness/contrast to render all-or nothing labeling patterns. The presence of immunostaining in the acrosomal region was scored in digital images from at least ≈100 cells contained in 10 fields. Data were evaluated before normalization with the program GraphPad Prism 8 using the one way Anova, Dunnett’s multiple comparisons test.

Proteins were resolved by electrophoresis on 15% Tris-glycine-SDS-gels and electro-transferred to 0.22 μm nitrocellulose membranes (Hybond, GE Healthcare) on a semi-dry apparatus (Amersham Biosciences) for 35 min at 25 mA. Because α-synuclein monomers tend to detach from blotted membranes, resulting in no or very poor detection (Lee and Kamitani, 2011), membranes were treated with 0.4% paraformaldehyde for 30 min at room temperature immediately after transfer. Non-specific reactivity was blocked with 2% BSA dissolved in washing buffer (PBS, pH 7.6, 0.2% Tween 20) for 1 h at room temperature. Blots were incubated with 1 μg/mL anti-α-synuclein in blocking solution overnight at 4°C. Horseradish peroxidase-conjugated goat-anti-rabbit IgG (0.1 μg/mL in washing buffer) was used as secondary antibody with 1 h incubation at room temperature. Excess first and second antibodies were removed by rocking three times, 10 min each in washing buffer. Detection was accomplished with a chemiluminescence kit from Kalium Technologies (Biolumina, Buenos Aires, Argentina) on a Luminescent Image Analyzer LAS-4000 (Fujifilm, Tokyo, Japan).

For experiments summarized in Figure 4, we processed capacitated and SLO-permeabilized human sperm (5 × 10⁶ cells in 250 µL per condition) as described for “Indirect AE assays”. For experiments summarized in Figure 5, we incubated the cells with 20 μg/mL PSA-colloidal gold as described in the legend. Reactions were stopped by fixing in 2.5% glutaraldehyde in 0.1 M sodium phosphate buffer and incubating overnight at 4°C or 2 h at room temperature. Samples were washed in PBS (three times, 20 min each) and postfixed in 1% OsO₄ in PBS for 1 h at room temperature, washed as before, dehydrated in a graded acetone series and embedded in a low viscosity epoxy resin (Pelco International, Fresno, CA) (Spurr, 1969). Polymerization was performed at 70°C for 48 h. Ultrathin sections with interference color gray were cut with an ultramicrotome (Ultracut R; Leica, Wien, Austria), mounted on grids and stained with uranyl acetate and lead citrate (Reynolds, 1963). Grids were examined by transmission electron microscopy at 80 kV in a 900 Zeiss microscope (Jena, Germany). We included negative (not stimulated) and positive (stimulated with 0.5 mM CaCl₂) controls in all experiments.

We conducted umbrella sampling molecular dynamics using GROMACS (GROningen Machine for Chemical Simulations)-2020.5 (Van Der et al., 2005; Pronk et al., 2013), PLUMED-2.7.2 (Tribello et al., 2014) and the Martini 3 coarse-grained model (Souza et al., 2021). In all cases, we modeled two 1024-lipid bilayers patches composed of 87.5% of 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (POPC), 10% of 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine (POPS) and 2.5% of the recently developed phosphatidylinositol 4,5-bisphosphate (PI(4,5)P₂, or PIP₂) model (Borges-Araujo et al., 2022). This arrangement, previously used in our laboratory (Di Bartolo et al., 2022; Di Bartolo and Masone, 2022) follows the experimental membrane composition reported by Jahn and collaborators to trap Syt1 to the plasma membrane in the presence of calcium (Perez-Lara et al., 2016). These bilayer patches of ∼17 × 17 nm ensure negligible finite-size effects due to interactions between periodic images of the fusion pore (Caparotta et al., 2020b). Solvation fulfilled the ample water condition for Martini (Ingolfsson et al., 2014) and the inter-membrane separation was adjusted to equilibrate at ∼3.9 nm (Di Bartolo and Masone, 2022). See Supplementary Material for more technical details.

We investigated the presence of α-synuclein in human sperm by Western blot using a polyclonal antibody raised against recombinant human α-synuclein as probe. Immunoblot analysis of whole cell extracts detected a single protein band with an apparent molecular mass of 15 kDa (Figure 1A, lane 2). The recombinant protein used as positive control showed decreased electrophoretic mobility compared to that of the endogenous form (Figure 1A, lane 1). This is because α-synuclein expressed in E. coli contains the TAT peptide and therefore has a higher molecular weight (≃ 17.8 kDa) than sperm α-synuclein.

To explore the localization of α-synuclein in sperm, we performed indirect immunofluorescence experiments using anti-α-synuclein as primary and anti-rabbit-Cy3 IgGs as secondary antibodies (Figure 1B, red, left panels). FITC-coupled PSA, a lectin that binds the acrosomal matrix, allowed visualization of the acrosomes (Figure 1B, green, central panels). Hoechst 33342 stained the nuclei and accounted for all cells in each field (Figure 1B, blue, right panels). α-Synuclein was detected in the acrosomal region of the sperm head in over 60% of the cells (Figure 1C); this is the expected localization for a protein with a role in exocytosis. No head staining was observed in sperm that lost their acrosomes either spontaneously (6%) or upon inducing AE with the calcium ionophore A23187 (20%) (not shown). These results suggest that the subcellular compartments that bear α-synuclein shed in cells that undergo exocytosis. Very few cells unexposed to primary antibodies showed fluorescence in the acrosomal domain, indicating that the staining was specific (Figures 1B, C). The percentage of cells labeled by the anti-α-synuclein antibodies was identical whether or not the plasma membrane had been permeabilized with SLO, which indicated that despite its small size, α-synuclein did not diffuse out of cells upon treatment with SLO (Figure 1C).

To investigate the requirement of α-synuclein for AE, we introduced anti-α-synuclein antibodies into SLO-permeabilized sperm before inducing exocytosis with calcium. The antibodies inhibited the AE in a dose-response fashion, reaching 77% inhibition at 5 nM and 90% inhibition at 7.5 nM (Figure 2A). Preincubation of the antibody with recombinant α-synuclein abolished the IgG’s ability to prevent the AE (Supplementary Figure S2). His₆-TAT-α-synuclein did not affect either the spontaneous or calcium-induced AE. Nor did imidazole, used in the purification of the protein, or a nonimmune rabbit IgG (Supplementary Figure S2). Antibodies to complexin bind the endogenous protein and decorate the acrosomal domain of the head in human sperm (Roggero et al., 2007). In contrast to anti-α-synuclein antibodies, 7.5 nM anti-complexin (cpx) antibodies did not affect the AE (Supplementary Figure S2). Thus, we concluded that the negative effect of the anti-α-synuclein antibodies on sperm secretion was due to binding to endogenous α-synuclein and therefore specific. Our results revealed that α-synuclein exhibits a role in human sperm AE.

The acrosome behaves as an internal calcium reservoir; release from this store is necessary for the AE [reviewed in (Darszon et al., 2011; Correia et al., 2015)] after the outer acrosomal membrane docks to the plasma membrane and before they fuse (De Blas et al., 2005; Zanetti and Mayorga, 2009; Rodriguez et al., 2011). NP-EGTA-AM is a membrane permeant, photolabile calcium chelator that crosses the plasma (because it is permeabilized with SLO) and outer acrosomal (due to the AM moiety) membranes. After this, the AM group is removed and NP-EGTA accumulates inside the acrosome, where it halts the AE by chelating intra-acrosomal calcium for as long as the system is kept in the dark. Photolysis of the chelator by exposure to UV light rapidly replenishes the acrosomal calcium pool and allows exocytosis to resume (De Blas et al., 2002; Ackermann et al., 2008; Hu et al., 2010; Sosa et al., 2015). In sperm loaded with NP-EGTA-AM and kept in the dark, challenging with CaCl₂ allowed exocytosis to advance up to the stage that requires intracellular calcium mobilization. Subsequent incubation with anti-α-synuclein antibodies prevented exocytosis even after illuminating the tubes to photolyse the chelator (Figure 2B, black bar). These results indicate that α-synuclein is required at a late stage of the AR, likely after docking and intra-acrosomal calcium release. Interestingly, this is the same timeframe in which the chimeric protein Rab3A-22A halts the AE (Bustos et al., 2014; Quevedo et al., 2016).

When introduced into SLO-permeabilized human sperm, recombinant α-synuclein reversed the exocytotic block imposed by the anti-α-synuclein antibodies on the AE (Figure 3A, black bar). The reversibility of the effect of the antibodies by bacterially expressed α-synuclein was instrumental to interpret the findings described below.

Results summarized in Figure 2B indicated that α-synuclein controls a post-docking stage during the human sperm AR. Molecularly, docking corresponds to a state where the outer acrosomal and plasma membranes are bridged together by partially assembled trans SNARE complexes in preparation for fusion. When in this configuration, SNARE proteins are sensitive to BoNTs but resistant to TeTx. To scrutinize the configuration of the SNARE complex at the post-docking stage governed by α-synuclein, we combined anti-α-synuclein antibodies, BoNT/B and TeTx in a functional assay (Figure 3A). When reagents were introduced into SLO-permeabilized sperm in the sequence anti-α-synuclein → calcium → BoNT/B or TeTx → α-synuclein, the AE proceeded normally (Figure 3A, striped bars). These results revealed that synaptobrevin-2 was protected from either toxin cleavage, which only happens when engaged in SNARE complexes in a cis configuration.

We have established in the past that monitoring SNARE proteins’ sensitivity to neurotoxins by indirect immunofluorescence is a suitable assay to determine their configuration. Following a triple staining strategy similar to that used to gather the data compiled in Figures 1B, C, we scrutinized if the introduction of anti-α-synuclein antibodies into SLO-permeabilized sperm affected the susceptibility of synaptobrevin-2 to BoNT/B. The monoclonal antibody used in these experiments recognizes an epitope located in a portion of the molecule severed by the toxin and therefore detects intact—but not BoNT/B-cleaved—synaptobrevin-2. Representative images are shown in Figure 3B. As we have reported before, the anti-synaptobrevin-2 antibody decorated the acrosomal region despite pretreatment with BoNT/B because SNAREs are in a cis configuration in resting sperm. When sperm were treated with the toxin plus CaCl₂, the proportion of cells exhibiting synaptobrevin-2 labeling dropped significantly because the initiation of the AE sensitized this SNARE to BoNT/B. In contrast, when anti-α-synuclein antibodies were introduced into sperm before CaCl₂, synaptobrevin was insensitive to BoNT/B cleavage (Figure 3C). These results suggest that when CaCl₂ advanced the AE until it reached the stage blocked by the anti-α-synuclein antibodies, synaptobrevin was in a BoNT/B-resistant configuration. Once again, our findings indicate that the anti-α-synuclein antibodies halted the AE at a stage when SNAREs were likely engaged in a cis configuration. This arrangement is characteristic of two exocytotic stages: an early one in resting cells and a late, post-fusion, one. Only the latter is consistent with the findings reported here and therefore we conclude that α-synuclein exhibits its role in the AE after the opening of fusion pores between the acrosomal and plasma membranes.

Anti-α-synuclein antibodies froze sperm SNAREs in toxin-resistant, cis complexes (Figure 3). We used fluorescence and transmission electron microscopies to further explore the premise that the antibodies affect a post-fusion event. FITC-PSA reveals the presence of the acrosomal matrix after fixation in the standard (indirect) functional assay: sperm without acrosomes are not stained whereas sperm with acrosomes, are (see diagram). On the contrary, in the direct functional assay, PSA reaches the acrosomal matrix of sperm undergoing exocytosis without fixation: only these sperm stain. This is because PSA enters the acrosomes through the fusion pores and remains attached to the matrix (see diagram) (Sanchez-Cardenas et al., 2014; Pelletan et al., 2015; Sosa et al., 2015). Thus, the read outs of both protocols are complementary. Sperm loaded with anti-α-synuclein antibodies and challenged with CaCl₂ evinced fluorescent acrosomes under both staining methods: cells incorporated the lectin through the fusion pores (direct protocol, Figure 4B); yet, they retained their acrosomal matrix (indirect staining, Figure 4A). Taken together these results support our hypothesis that the exocytotic cascade advanced to a post-fusion—and before acrosomal dispersion—stage. Similar results are obtained when permeabilized sperm are incubated with the chimaeric protein Rab3A-22A before challenging with CaCl₂ (Quevedo et al., 2016).

We resorted to transmission electron microscopy to gain further insights into the exocytotic stage halted by the anti-α-synuclein antibodies. Micrograph labeled “intact” in Figure 4C shows the typical aspect of the heads of a resting cell, where the acrosome is planar. This pattern is the most abundant in unperturbed cells (Figure 4D, I). The cell photographed in the panel “fully exocytosed” has completed the AE. As expected, this stage was predominantly observed in samples challenged with calcium (Figure 4D, IV). Panel “swollen and waving” illustrates a cell that underwent acrosomal swelling/deformation in response to external calcium, but it could not complete the AE. This pattern is seldom observed in cells undergoing exocytosis spontaneously and those challenged with calcium, because swelling is transient (Figure 4D, II) and (Sosa et al., 2015). Interestingly, this was the predominant pattern in cells preloaded with anti-α-synuclein antibodies before adding CaCl₂ (Figure 4D, black bars). This response resembled that observed when using recombinant Rab3A-22A instead of anti-α-synuclein antibodies (Quevedo et al., 2016). Few cells completed the AE because the antibodies prevented it. We visualized diffuse, localized and electron dense material adhered to the surface of a subpopulation of cells with swollen and waving acrosomes, exclusively upon treatment with anti-α-synuclein antibodies and calcium (Figure 4C, pattern “swollen, waving and diffuse (III)”, arrows, quantified in Figure 4D, III).

Considering that the anti-α-synuclein antibodies stabilized open—but not expanded—fusion pores (Figure 4B), we hypothesized that the electron dense material characteristic of pattern III might correspond to acrosomal components that diffuse out of the cells through those pores. To test this premise, we adapted for electron microscopy experiments similar to those summarized in Figure 4B, only that PSA coupled to 20 nm colloidal gold particles substituted for FITC-PSA. The membranes of intact sperm (pattern I in Figure 4C) showed absence of gold nanoparticles (images not shown but included in quantifications to compose Figure 5B), which indicates that they were not ruptured by SLO-permeabilization or processing for transmission electron microscopy. Likewise, PSA-gold did not bind sperm that had completed the AE (pattern IV in Figure 4C; images not shown but included in quantifications to compose Figure 5B), which suggests that no acrosomal matrix remained associated to the inner acrosomal membrane. Interestingly, one or more gold particles were consistently found adjacent to the surface of cells exhibiting swollen and waving acrosomes (patterns II and III in Figure 4C, undistinguishable here). A gallery of typical images illustrating this subpopulation of sperm is shown in Figure 5A, with arrows pointing to gold nanoparticles. Quantification of hundreds of electron micrographs showed that the proportion of gold-labelled cells (+AuNP) was low in sperm that underwent no treatment (Figure 5B, control) or were exposed to the AE inducer (Figure 5B, Ca). This is consistent with the observations that bins II and III were the least abundant of the histogram shown in Figure 4C. In contrast, gold nanoparticles decorated the extracellular face of the acrosomal region in a high percentage of cells that had initiated the AE but could not complete it because the anti-α-synuclein prevented it (Figure 5B, anti-α-syn → Ca). This is consistent with the distribution shown in Figure 4C, where bins II and III were the most abundant under these experimental conditions. Because colloidal gold was coupled to a probe that exclusively binds the acrosomal matrix, these results support our hypothesis that the acrosomal contents targeted by PSA had diffused out of the sperm (diffuse material, arrows in pattern III, Figure 4C). Similar patterns of detection of extracellular acrosomal contents attributed to efflux through fusion pores have been described with anti-acrosin antibodies (Tesarik et al., 1988) and PSA coupled to paramagnetic beads (Kohn et al., 1996). Colloidal gold present in the medium bathing the cells was seldom observed inside acrosomes, which reflects its inability to traverse the pores opened between the plasma and acrosomal membranes. Soluble PSA permeated through fusion pores and reached the acrosomal membrane (Figure 4B) but that bound to gold nanoparticles did not; these observations suggest that the latter was excluded because of size. The pores were large enough to allow efflux of acrosomal material but too small to allow influx of 20 nm gold nanoparticles. Thus, we conclude that the fusion pores stabilized by anti-α-synuclein antibodies were smaller than 20 nm, the size of the gold nanoparticles used in this study.

We have shown that the inhibition of the AE by anti-α-synuclein antibodies is a post-fusion event because they freeze sperm SNAREs in toxin-resistant, cis complexes (Figure 3) and allow the influx of an acrosomal probe (Figure 4B) as well as the efflux of the acrosomal matrix (Figure 5). Is α-synuclein required after bilayer fusion to dilate the fusion pores? We tested this hypothesis with computational and experimental approaches. We have recently implemented into PLUMED the reaction coordinate ξ_e to study the thermodynamics of the expansion of the fusion pore using restrained molecular dynamics (Di Bartolo et al., 2022). ξ_e is a dimensionless parameter approximately proportional to the radius (R, in nm) of a toroid-shaped fusion pore (see Eq. 1) [for mathematical details on ξ_e see Masone and collaborators (Di Bartolo et al., 2022)] R ≃ 0.75 ξ e + 0.2 (1)

To investigate the influence of α-synuclein, we expanded a fusion pore nucleated between two lipid bilayers using molecular dynamics simulations and the collective variable ξ_e. We inserted a Martini model of α-synuclein between the bilayers (in the cytosolic space) generated from the human micelle-bound α-synuclein (PDB ID: 1xq8) (Ulmer et al., 2005). Figure 6A shows the Gibbs free energy needed to expand a fusion pore with (red) and without (black) α-synuclein in system. A significant reduction (∼ 20 kJ/mol) of the free energy occurred as the fusion pore expanded from ξ_e ∼ 0.4 (R ≃ 0.5 nm) to ξ_e ∼ 0.75 (R ≃ 0.75 nm) in the presence of the protein. Snapshots of the expanding pore at three different values of ξ_e are shown in panel B. Notice that the protein, initially situated in the “cytoplasm” (between the bilayers), freely adsorbed on the phospholipid surface of the pore. This location is consistent with experimental data that show strong binding of α-synuclein to highly curved lipid membranes (Middleton and Rhoades, 2010; Caparotta et al., 2020a). These findings suggest that α-synuclein facilitates the expansion of fusion pores that connect two pure lipid membranes.

When introduced into SLO-permeabilized human sperm, the chimeric protein Rab3A-22A halts the AE elicited by calcium at a stage after membrane fusion/pore opening but before pore expansion/membrane vesiculation (Quevedo et al., 2016). Addition of recombinant α-synuclein after Rab3A-22A and calcium rescued the exocytotic block imposed by the chimaera (Figure 6C). Taken together, our results suggest that α-synuclein is required at a post-fusion stage during sperm exocytosis and that its unavailability inhibits the AE because pores do not expand.