香京julia种子在线播放

All chemicals were purchased from Sigma-Aldrich (Zwijndrecht, Netherlands) unless otherwise stated. (±)-2-(1H-benzimidazol-2-ylthio) propanoicacid2-[(5-bromo-2 hydroxyphenyl)methylene] hydrazide (KH7) was from Sanbio (13243-10, Uden, Netherlands), 6-chloro-N4-cyclopropyl-N4-[(thiophen-3-yl)methyl]pyrimidine-2,4-diamine (LRE1; HY-100524) was obtained from MedChemExpress (NJ, United States). Annexin–V-FLUOS was purchased from Merck (11828681001, Darmstadt, Germany), lectin from Arachis hypogea (peanut) conjugated to Alexa Fluor^™ 647 (PNA-AlexaFluor 647) was obtained from ThermoFisher Scientific (L32460; Waltham, MA, United States), and sodium; 3-[(2E)-2-[(E)-4-(1,3-dibutyl-4,6-dioxo-2-sulfanylidene-1,3-diazinan-5-ylidene) but-2-enylidene]-1,3-benzoxazol-3-yl] propane-1-sulfonate (merocyanine 540 abbreviated here to M540) was from Molecular Probes (M24571, Eugene, OR, United States). 4-(5-(4-Methylpiperazin-1-yl)-1H,1′H-[2,5′-bibenzo [d]imidazol]-2′-yl) phenol trihydrochloride (Hoechst 33258; 8861405) and 5.5‘,6.6‘-tetrachloro-1,1‘,3,3‘-tetraethylbenzimidazol-carbocyanine iodide (JC-1; T4069) were obtained from Sigma-Aldrich, and duramycin-cy5 was from Molecular Targeting Technologies, Inc., (D- 1002, West Chester, PS, United States). ADCY10 Polyclonal Antibody was purchased from Bioss Antibodies Inc., (bs-3916R, Woburn, MA, United States) and Goat anti-Rabbit IgG conjugated to Alexa Fluor^™ 488 was from Life Technologies (A-11008, Bleiswijk, Netherlands). The semen extender was a commercial skim milk-based product (INRA 96) purchased from IVM technologies (016441, l’Aigle, France).

The basic variant of Tyrode’s medium (Tyr_Control) consisted of 111 mM NaCl, 20 mM HEPES, 5 mM glucose, 3.1 mM KCl, 0.4 mM MgSO₄, 0.3 mM KH₂PO₄, 100 µg/ml gentamycin sulfate, 1.0 mM sodium pyruvate, 21.7 mM sodium lactate. In the bicarbonate containing variant (Tyr_Bic) a defined amount of NaCl was replaced by 30 mM of NaHCO₃. The pH was adjusted to 7.40 ± 0.05 at room temperature with NaOH or HCl and the osmolality to 300 ± 5 mOsmol/kg. All media were passed through a polyethersulfone syringe filter (PES membrane, pore size 0.22 µm; Merck Millipore, Amsterdam, Netherlands) for sterile filtration. Both media contained 1 mg/ml of bovine serum albumin (BSA; A6002, Sigma-Aldrich) and 2 mM of Ca²⁺ supplemented as CaCl₂. The bicarbonate containing medium (Tyr_Bic) and its variants were kept in an incubator with 5% CO₂ and 100% humidity at 37°C for at least 24 h for equilibration prior to experimentation. Incubations of spermatozoa in bicarbonate containing media took place in the same incubator used for equilibration. Incubations of spermatozoa in control medium (Tyr_Control) were carried out in a metal heating block at 37°C.

Semen was collected using an artificial vagina (Hanover model) from stallions attending the Faculty of Veterinary Medicine at Utrecht University for routine breeding soundness examination, or from stallions located at nearby horse farms (Stal Schep and Stal van Vliet) with the written consent of the owners. After collection, semen was filtered through gauze to remove the gel fraction and gross debris. A smear of raw semen with Aniline Blue-Eosin was prepared to assess sperm morphology. Sample concentration was measured with a Bürker Türk haemocytometer and ejaculates were diluted in INRA 96^® to a concentration of 30 × 10⁶ spermatozoa/mL. Motility was checked objectively using a computer-assisted semen analysis (CASA) system (SpermVision 3.5, Minitüb, Tiefenbach, Germany) as described by Brogan et al. (2015). Only samples with greater than or equal to 70% (total) motile sperm in the diluted semen were used for experiments. Diluted semen was kept at room temperature until further processing. For each experiment, semen from a minimum of three different stallions was used, with the exact number of replicates stated in each figure caption.

Density gradient centrifugation was performed to separate the spermatozoa from the semen extender and seminal plasma prior to experimentation. Diluted semen (6 ml) was layered on top of a discontinuous gradient consisting of 2 ml of isotonic 70% Percoll^®-saline solution and 4 ml of isotonic 35% Percoll^®-saline solution in a 15-ml centrifugation tube, as described by Harrison et al. (1993). Tubes were centrifuged for 20 min at room temperature; 10 min at 300 g followed by 10 min at 750 g, without stopping in between. After centrifugation, the supernatant was removed and the remaining pellet was resuspended in 1 ml of Tyr_Control without CaCl₂ and BSA. The sperm concentration was adjusted to 30 × 10⁶ sperm/mL, unless otherwise stated. The sperm suspension was used within 30 min of preparation.

A live imaging approach was used to demonstrate whether Annexin-V staining was present in viable stallion sperm under capacitating conditions. To this end, 10 µL of Percoll-washed sperm (120 × 10⁶ sperm/mL) was incubated in 500 µL of either Tyr_Bic or Tyr_Bic supplemented with 1 mM caffeine. Previous experiments indicated that these conditions yielded the highest abundance of Annexin-V positive spermatozoa. Both media contained Hoechst 33258 and PNA-AlexaFluor647, as described for flow cytometry. Samples were evaluated after total incubation times of 30 and 60 min, respectively. 15 min before measurements took place, 100 µL of the sample was transferred to a pre-warmed Eppendorf tube, and 2 µL of Annexin-V FLUOS was added. Live imaging was performed on a NIKON STORM/A1Rsi/TIRF microscope (Nikon, NY, United States), with a preheated stage at 37 °C. After incubation, the samples were centrifuged for 2 min at 1000 x g and then resuspended in 10 µL of the respective medium. Next, a 1 µL droplet was placed in a FluoroDish (FD35-100, World Precision Instruments, Friedberg, Germany) and covered with a round coverslip (diameter: 8 mm). Pre-warmed mineral oil was placed around the coverslip to prevent the sample from drying. The autofocus function of the microscope was used to locate the imaging plane. General imaging settings in the acquisition software were: 40x objective, scan speed ½; image size 1,024 × 1,024 pixels, pinhole 5.0 and zoom 1. At least five large scans were performed in a 2 × 2 panel series with laser lines at 405 nm, 488 nm, and 647 nm and corresponding detection channels for the dyes (Hoechst 33258, Annexin-V-FLUOS, AlexaFluor 647, DIC). Images were analyzed with NIS Elements Viewer software (Nikon, NY, United States). An automated detection of the different fluorescent signals in each cell was performed. Subsequently, the staining patterns were validated visually and the location of Annexin V staining (PS exposure) was determined. At least 200 viable cells were scored for each medium.

Stallion spermatozoa were sorted on a FACS Influx (Becton Dickinson, San Jose, Canada). A total of 4 × 10⁷ spermatozoa were incubated for 60 min in either 0.5 ml Tyr_Control or Tyr_Bic media supplemented with 2 µL Hoechst 33258. M540 was added 15 min before sorting. Hoechst 33258 was excited with a 405 nm Laser. Emission was captured with a 460/50 nm filter. M540 was excited with a 561 nm laser, and emission was captured with a 585/42 nm filter. Spermatozoa were analyzed at a rate of between 8,000 and 10,000 events per second. Only events with forward and side scatter characteristics of single spermatozoa were considered for further analysis. During sorting, the sample-input tube on the FACS Influx was kept at 38°C to maintain the sample’s temperature during the entire sorting procedure. Phosphate-buffered saline served as sheath fluid. Two subpopulations were sorted: 1) viable spermatozoa (Hoechst 33258 negative) with low membrane fluidity (M540 fluorescence low) from Tyr_Control, and 2) viable spermatozoa with high membrane fluidity (M540 fluorescence high) from Tyr_Bic. A total of 250,000 spermatozoa from a specific subpopulation were sorted into a single tube. The sorting time per tube ranged from 8 min to 15 min. Immediately after sorting, the tube was centrifuged at 11,000 x g for 10 min and the supernatant discarded.

Pellets of sorted spermatozoa were diluted to ∼3 × 10⁶ cells/mL in phosphate buffered saline. Approximately 3 µL of cell suspension was applied to glow-discharged Quantifoil R 2/1 200-mesh holey carbon grids. Approximately 1 µL of BSA-gold (Aurion, Wageningen, Netherlands) was added, after which grids were blotted manually from the back for 3–4 s and immediately plunged into a 37% liquid ethane/propane mix cooled to liquid nitrogen temperature. Grids were stored under liquid nitrogen until imaging. Imaging was performed on a Talos Arctica (ThermoFisher) operated at 200 kV and equipped with a post-column energy filter (Gatan) in zero-loss imaging mode with a 20-eV energy-selecting slit. All images were recorded on a ∼ 4k × 4k K2 Summit direct electron detector (Gatan) in counting mode with dose-fractionation. Tilt series were collected with SerialEM, using a grouped dose-symmetric tilt scheme covering an angular range of ± 56° in 2° increments. Tilt series were acquired with a Volta phase plate (VPP) at a target defocus of −0.75 µm and with a pixel size of 3.514 Å. The total dose was limited to <100 e⁻/Ų. Frames were aligned using Motioncor2 1.2.1. Tomograms were reconstructed in IMOD 4.10.25 using weighted back-projection. Contrast Transfer Function (CTF) correction was not performed because tilt series were acquired close to focus with the VPP. For segmentation and presentation, 6x-binned tomograms were reconstructed with a simultaneous iterative reconstruction technique (SIRT)-like filter corresponding to 20 iterations. Membrane thickness and intermembrane distances from at least five selected tomograms for each sample group were measured in Fiji (Schindelin et al., 2012). For all selected tomograms, ten measurements were performed for each compartment (plasma membrane, intermembrane distance and outer acrosomal membrane when possible). Measurement locations were spaced by at least 100 nm.

Data were analyzed using the Statistical Analysis System software (SAS^®, version 9.4; SAS Inst. Inc., Cary, NC, United States). Parameters were tested for normal distribution using the Shapiro-Wilk test. Where applicable, a multivariate analysis of variance (ANOVA) for repeated measurements was performed. Comparisons between individual treatments or time points were carried out using Student’s t-test for paired observations. All data are presented as mean ± standard deviation (SD). Differences were considered significant when p ≤ 0.05.

Initial experiments were conducted to confirm that the experimental conditions would stimulate an increase in membrane fluidity (as previously demonstrated by Rathi et al. (2001)). Our results indicated that incubating stallion spermatozoa in Tyr_Bic medium resulted in a significantly increased population of viable spermatozoa with high membrane fluidity after a 15 min incubation time (Figure 1). In the absence of bicarbonate no change was observed. The combination of a cAMP analogue (db-cAMP) and a PDE inhibitor (caffeine) was able to mimic the bicarbonate effect and induced an equally large population of viable sperm with high membrane fluidity (Figure 1; Supplementary Figure S1I). Nonetheless, the response was slower than for Tyr_Bic and it took at least 30 min before this population was detectable (Figure 1). It is thought that bicarbonate is essential for cAMP upregulation and in other eutherian mammals this response is mediated by adenylyl cyclase. Indeed, in spermatozoa, the presence of soluble adenylyl cyclase (sAC) has primarily been studied in human and mouse sperm (Uguz et al., 1994; Harrison and Miller, 2000; Lefievre et al., 2000; Baxendale and Fraser, 2003; Spehr et al., 2004; Tardif et al., 2004; Wertheimer et al., 2013). In stallion sperm, there is limited available data regarding the nature of sACs and its involvement in the specific steps of capacitation. Thus, the following studies were designed to determine whether the impact of bicarbonate on stallion sperm membrane fluidity is mediated by sAC.

To investigate the presence and localization of sAC in stallion spermatozoa, immunoblotting and immunofluorescence with an anti-ADCY10 antibody was performed. The results were compared sAC localization in boar sperm, where it is known to have high activity (Leemans et al., 2019b), and shares 87.03% sequence similarity to equine ADCY10. Immunoblotting demonstrated that the sAC is present in both boar and stallion spermatozoa as an immunoreactive band at approximately 55 kDa (Figure 2A). This is the predicted size of the testis-specific form of sAC. In both species, some additional bands were detected at approx. 35 and 50 kDa. Immunofluorescent labelling of sAC revealed that this adenylyl cyclase is distributed along the tail of spermatozoa from both species, with bright labelling noted in the endpiece of the tail. However, a higher signal intensity for both species was noted in the neck region. In the sperm head, species-specific staining patterns were observed with sAC localized over the whole acrosomal area in stallion spermatozoa, whereas the sAC signal was limited to a distinct band across the post-equatorial region of boar spermatozoa (Figure 2B). Secondary controls where Anti-ADCY10 was omitted revealed no non-specific fluorescence (Supplementary Figure S1).

Following confirmation that sAC was present in stallion spermatozoa, the effect of sAC inhibition was then assessed in this species. Indirect inhibition of sAC with the inhibitor KH7 in the presence of bicarbonate (Tyr_Bic) appeared to be effective in a large proportion of viable spermatozoa at concentrations of 60 and 120 µM (Supplementary Figure S2A). However, off-target effects were also observed, with KH7 at 60 and 120 µM resulting in an increase in viable sperm populations with high membrane fluidity in Tyr_Control but abolishing the high mitochondrial transmembrane potential in virtually all spermatozoa (Supplementary Figures S2B–D). Further observations confirmed that spermatozoa stopped tail beating within 1 min after exposure to KH7 at all concentrations. This is in line with a previous report suggesting that KH7 acts as a mitochondrial uncoupler (Jakobsen et al., 2018). Given these challenges, a direct inhibitor of sAC, LRE1, which specifically competes with the bicarbonate binding site of sAC (Ramos-Espiritu et al., 2016), was used in equivalent experiments. Using LRE1, the expected increase in membrane fluidity could be prevented in most of the viable spermatozoa after incubation for 15 min in the presence of 30 mM bicarbonate (Tyr_Bic; Figure 3A). The strength of the inhibition lessened over time, but was still significant after 60 min of incubation (Figure 3A). A small, but significant increase in the percentage of viable spermatozoa with high membrane fluidity was noted in the absence of bicarbonate (Tyr_Control). However, this population stayed below 10% (Figure 3B). In contrast to the observations for KH7, there were no adverse effects of LRE1 in Tyr_Bic or Tyr_Control on the percentage of spermatozoa with high mitochondrial membrane potential, and the cells remained motile (Figures 3C,D).

To visualize structural changes associated with high membrane fluidity, we imaged whole, unfixed, unstained stallion sperm using cryo-electron tomography (cryo-ET). Cryo-ET yields three-dimensional reconstructions of subcellular structures within the context of fully-hydrated cells, avoiding artefacts from dehydration and chemical fixation. To establish baseline membrane morphology, we imaged non-sorted spermatozoa that had been incubated for 60 min in Tyr_Control (Figures 4A,B). In these non-capacitated cells, the plasma membrane (PM) and outer acrosomal membrane (OAM) were smooth and parallel (10/12 tomograms, each from a different cell, from two stallions), with a regular intermembrane distance of ∼11 ± 2 nm. To directly correlate structural changes to membrane fluidity, we imaged flow-sorted viable sperm with either low or high M540 staining. The majority of sorted, viable sperm with low membrane fluidity were similar to non-sorted control cells, with the PM and OAM both intact, smooth, and running parallel to each other (13/19 tomograms, each from a different cell, from three stallions) (Figures 4C,D). In contrast, most of the sorted, viable spermatozoa with high membrane fluidity showed evidence of membrane destabilization (23/26 tomograms, each from a different cell, from three stallions). A range of phenotypes could be distinguished. Some cells showed a clear approximation of the PM and OAM (Figure 4E), in particular at sites where the OAM becomes discontinuous and bends upwards towards the OM (Figure 4F) (6 tomograms, each from a different cell). This was reflected by a reduced average intermembrane distance (Figure 4G). Finally, in some spermatozoa the presence of membrane vesicles was observed in regions close to sites of membrane disruption (Supplementary Figure S1).

Given that the cryo-ET analysis revealed modulated membrane organization in sperm cells with high membrane fluidity, we next sought to determine whether changes in membrane organization were reflective of changes in either the phospholipidome or alterations in cholesterol efflux caused by bicarbonate treatment. While lipidomic analysis of FACS-sorted spermatozoa revealed 15 distinct phospholipid and glycolipid classes (Figure 5A) consisting of >150 unique phospholipid/glycolipid species (Supplementary File S1), no significant differences in the relative intensity of these lipid classes, or in individual molecular species, were detected between low and high membrane fluidity stallion sperm samples (n = 6 stallions; Figure 5A). Similarly, when the levels of cholesterol and desmosterol were measured in the same samples, equivalent pmole amounts/sample (∼500,000 sperm cells) were detected in low and high membrane fluidity sorted cell populations (Figure 5B). The absence of differences in lipid composition between the high and low membrane fluidity sperm subpopulations indicates that differences in membrane fluidity are not caused by compositional differences in sperm membrane lipids. Therefore, further experiments were carried out to investigate whether or not the organization of lipids in the sperm membrane were altered in a cAMP-dependent manner.

In the sperm cells of most species studied, rearrangement of the sperm lipid membrane bilayer is essential to increase its fluidity prior to fertilization (Visconti et al., 1995a; Gadella and Harrison, 2000; Cross, 2003). The following experiment was performed to understand whether the exposure of PS and PE phospholipids is required for the activation of the sAC/cAMP/PKA pathway leading to an increase in membrane fluidity as previously described in boar sperm (Flesch and Gadella, 2000). In a preliminary experiment we observed that M540 staining may have a small, but significant impact on the number of viable spermatozoa detected as Annexin-V positive (data not shown). Consequently, the following results for PS and PE exposure in viable spermatozoa were obtained in the absence of M540 staining so as not to bias the analysis (Figure 6). Samples with M540 staining were run in parallel to ensure that most of the viable spermatozoa were showing increased membrane fluidity. Annexin-V labelling demonstrated that PS exposure could be detected in Tyr_Bic in a subset of viable spermatozoa (Figure 6A). However, exposure of PE in viable spermatozoa was barely observed (Figure 6B).

Bicarbonate was required to significantly increase the proportion of viable sperm with PS exposure after 30 and 60 min incubation (Tyr_Control (no addition) Figure 6C versus Tyr_Bic (no addition) Figure 6E). Maximal PS exposure in Tyr_Bic (no addition) was reached after 60 min incubation (Figure 6E). Caffeine, db-cAMP, and the combination of both compounds were able to increase the population of viable spermatozoa with PS exposure after 15 min and/or 30 min in Tyr_Bic, but not after 60 min (Figure 6E). In the absence of bicarbonate (Tyr_Control), either caffeine or the combination of caffeine and db-cAMP initiated an increase in the percentage of viable sperm with PS exposure (Figure 6C). Values in Tyr_Control with db-cAMP and caffeine were identical to those from Tyr_Bic (no addition; Figures 6C,E). The exposure of PE in viable sperm was always limited to less than 5% of the spermatozoa and could not be stimulated by db-cAMP, caffeine, or the combination of the two compounds (Figures 6D,F).

As flow cytometry cannot specify where Annexin-V binds within cells, live imaging was used to visualize Annexin-V staining patterns in stallion sperm under capacitating conditions in Tyr_Bic. Three main Annexin-V staining patterns were observed in viable stallion spermatozoa (Figure 7; arrow indicated). The dominant pattern was an homogenous labelling of the acrosomal area in the sperm head (Figure 7A), followed by a labelling of the entire head and midpiece (Figure 7C). A smaller subset of spermatozoa stained Annexin-V positive over the whole sperm head (Figures 7B,D). Statistical analysis revealed significant differences in the abundance of these three staining patterns (i.e., number of cells for each Annexin-V labelling pattern), but not significant influence of time or treatment on the number of cells demonstrating each pattern.