Reagents and media
Unless otherwise stated, reagents were purchased from Sigma Aldrich (St. Louis, MO, USA). Fixable Live/Dead Red Stain, rabbit anti-arylsulfatase F (ARSF) polyclonal antibody, goat anti-rabbit IgG Alexa-555-conjugated secondary antibody, 10% normal goat serum, SlowFade™ Diamond Antifade Mounting Solution with DAPI, and methanol-free 16% paraformaldehyde were acquired from Thermo Fisher Scientific (Waltham, MA, USA). Accumax® Cell Detachment Solution was purchased from Stemcell™ Technologies Inc., (Cambridge, MA, USA). Acridine Orange stain was obtained from Polysciences Inc., (Warrington, PA, USA). CryoMax Lactose-EDTA® semen freezing extender (20% egg-yolk + 2% glycerol and 3% methyl formamide) was obtained from Animal Reproduction Systems (Chino, CA, USA). A silane-coated silica particle solution (Redigrad®) for density gradient centrifugation was acquired from Global Life Sciences Solutions (Marlborough, MA, USA). The base medium, MW-HEPES, used for sperm washing by density gradient centrifugation was a modified Whitten’s medium27 and consisted of 110 mM NaCl, 4.7 mM KCl, 1.2 mM MgCl2, 1.9 mM CaCl2, 22 mM HEPES, and 50 µL/mL gentamicin sulfate. The medium used for sperm in vitro incubation under capacitating conditions (Lac-MW) was prepared as reported previously6, 27 and consisted of modified Whitten’s medium with 25 mM HCO3− instead of HEPES, 7 mg/mL bovine serum albumin – heat shock fraction (BSA), and 10 mM sodium-DL lactate (60% syrup). All media were adjusted with NaCl to an osmolality of 280–290 mOsm/kg. On the day of the experiment, the pH of each medium was adjusted to 7.25 using NaOH or HCl. The Lac-MW medium was maintained at 38.2°C in 5% CO2 for a minimum of 2 hours before use.
Stallions and semen collection
Texas A&M University Institutional Animal Care and Use Committee (IACUC 2021-0007) approved all the procedures performed in this study. All stallions enrolled (n = 6) were Thoroughbred, sexually active, and 7–15 years old. Hair samples were procured to determine the presence of the susceptibility genotype for IAE, A/A-A/A in the gene FKBP6 exon 5, as previously reported7, 8. The FKBP6 genotype, in vivo fertility rates, and conventional sperm quality parameters of the six stallions used (three fertile and three subfertile) are presented in Table 1. Before semen collection, each stallion was exposed to an ovariectomized mare (when available), or a mare in standing estrus. Once erect, the penis was rinsed thoroughly with warm water and dried with paper towels. The ejaculates from two fertile TB stallions and one subfertile TB stallion were collected using a Colorado-type artificial vagina (Animal Reproduction Systems, Chino, CA, USA), while ejaculates from the other three TB stallions (one fertile and two subfertile) were collected using a Missouri-Model artificial vagina (Nasco, Ft. Atkinson, WI, USA). For both artificial vagina types, an in-line nylon micromesh filter (Animal Reproduction Systems) was placed between the artificial vagina and the semen collection receptacle to separate the gel fraction from the gel-free semen. Following semen collection, the gel-free semen was transported to an adjacent laboratory and placed in an incubator (37°C) before processing.
Initial semen processing and analyses
The gel-free semen volume was estimated based on sample weight, while sperm concentration and plasma membrane intactness (i.e., viability) were measured using a fluorescence-based cell counter (NucleoCounter SP-100™, Chemometec A/S, Allerød, Denmark), following a previously described methodology28. Sperm motion characteristics were determined using computer-assisted sperm analysis (CASA; Hamilton-Thorne IVOS II, Hamilton-Thorne Inc., Beverly, MA, USA), as reported previously29. The preset values for the instrument consisted of the following: frames acquired, 45/s; frame rate, 60 Hz; minimum contrast, 70; minimum cell size, four pixels; minimum static contrast, 30; straightness (STR) threshold for progressive motility, 50%; average path velocity (VAP) threshold for progressive motility, 30%; VAP threshold for static cells, 15 µm/s; cell intensity, 106 pixels; static head size, 0.60 to 2.00 µm; static head intensity, 0.20 to 2.01; static elongation, 40 to 85; illumination intensity, 2200. Sperm motility parameters included the percent of total motility (TMOT), progressive motility (PMOT), and the mean curvilinear velocity (µm/s; VCL). For sperm morphology analysis, samples of raw semen were fixed with buffered-formal saline (BFS; 4.75% formaldehyde) and analyzed using differential interference contrast (DIC) microscopy (1,563x, Olympus BX-60, Olympus Corporation, Melville, NY, USA). Sperm morphological classification was done as previously reported30. All morphologic abnormalities were counted for each spermatozoon to determine the incidence rate. A total of 100 sperm were counted for each ejaculate, and the percentage of morphologically normal sperm was recorded. Sperm DNA quality was determined in flash-frozen/thawed samples obtained from raw semen, using the Sperm Chromatin Structure Assay (SCSA), as previously described31. The percentage of Cells Outside the Main Population (COMPα−t), was used as an endpoint to determine the extent of the susceptibility of sperm DNA to denaturation.
Semen cryopreservation and stimulation of lactate-induced spontaneous AE in viable frozen/thawed sperm
In a previous study, we determined that stallion sperm stored at 5°C for 24 hours, or frozen/thawed stallion sperm do undergo spontaneous AE in viable sperm at the same rate as fresh semen after being incubated in a Lac-MW medium6. Because all the subfertile TB stallions that have been identified by our group as carrying the IAE susceptibility genotype were either located far from our laboratory, had been already castrated, or their sperm were cryopreserved several years previously, in the current study we used frozen/thawed sperm from these stallions to perform the acrosome function and subsequent DIA-MS analyses. When these ejaculates were obtained, immediately after semen collection and initial sperm analysis, the raw semen was diluted 1:1 (v/v) with INRA-96® extender (IMV Technologies, L’Aigle, France) and subjected to cushioned centrifugation at 1000 x g for 20 min, as described previously32. After centrifugation, the supernatant was removed, and the sperm pellet was resuspended with the EZ-Freezin CryoMax LE® semen freezing extender (Animal Reproduction Systems, Chino, CA) at a final sperm concentration of 200 x 106 sperm/mL. Sperm diluted with the freezing extender was loaded into 0.5-mL plastic straws, sealed ultrasonically, and frozen in a controlled rate freezer (CBS 2100; Custom Biogenic Systems, Bruce Township, MI, USA) using the following cooling curve: − 2.0°C/min from 25 to 20°C; − 0.1°C/min from 20 to 5°C; hold for 5 min; − 60°C/min from 4°C to − 140°C33. The straws were plunged directly into liquid nitrogen and stored in a liquid nitrogen tank. Frozen straws from each stallion were thawed for 30 sec in a water bath set at 37°C and the thawed semen was processed through density gradient centrifugation using 40% Redigrad® (Global Life Science Solutions, Marlborough, MA34) to remove seminal plasma, debris, and semen extender. After centrifugation, the sperm pellet was diluted to 30 x 106 sperm/mL in Lac-MW medium6, 27 and incubated for up to 6 hours at 38.2°C in 5% CO2. Sperm aliquots were analyzed after 0, 2, 4, and 6 h of incubation (T0h, T2h, T4h, and T6h, respectively) in Lac-MW medium for viability/acrosomal exocytosis (AE-Viable), as previously reported6. At each time point (T0h, T2h, T4h, and T6h), a 30 x 106 sperm aliquot was also flash-frozen in dry ice for further proteomic analysis using DIA-MS.
Analysis of sperm viability/acrosomal exocytosis (Fixable Live/Dead Red stain + FITC-PSA)
The intactness of both the plasma membranes (viability) and the acrosome membranes (AE) was evaluated simultaneously, as previously described35, with some modifications. An aliquot (50 µL) of frozen/thawed semen diluted (30 x 106 sperm/mL) in Lac-MW was added to 1 mL of Lac-MW medium. This dilution resulted in a final sperm concentration of approximately 1.5 x 106 sperm/mL. One µL of Fixable Live/Dead Red stain (Excitation: 488 nm, Emission: 617 nm; final concentration: 50 µg/mL) was added to the sperm sample and incubated at 38.2°C in an air atmosphere for 20 min. Then, 140 µL of methanol-free paraformaldehyde (paraformaldehyde final concentration: 1.88% v/v) was added, and the sperm sample was stored at 5°C in the dark for 30 min. Subsequently, the samples were centrifuged (400 x g x 5 min) using BSA in DPBS (2 mg/mL), permeabilized with Triton-X100 (1% v: v), centrifuged again with BSA in DPBS, diluted in 133 µL Accumax to avoid sperm clumping, and incubated with 10 µL Pisum sativum agglutinin (PSA)-FITC conjugate (excitation, 488 nm; emission, 517 nm; final concentration, 0.0375 mg/mL) for 20 min at room temperature in the dark. The Fixable Live/Dead Red stain binds to free amines both in the intracellular space and the surface of cells with a disrupted plasma membrane (“non-viable” cells)36. In the case of the Fixable Live/Dead Red stain, non-viable sperm will allow the internalization of this dye and emit red fluorescence35. PSA binds to the glycoconjugates of the acrosomal matrix, particularly to the α-D-glucosyl and the α-D-mannosyl residues at the inner acrosomal membrane37. When the acrosomal membrane is disrupted, either as a consequence of acrosome damage or during AE, FITC-PSA will bind to such residues and emit a green, fluorescent signal. In both fixed and permeabilized sperm, four subpopulations are identified using Fixable Live/Dead Red stain and FITC-PSA: 1) sperm with both intact plasma and acrosomal membranes [FITC-PSA (+)/Fixable Live/Dead Red (–)]; 2) sperm with intact plasma membrane and disrupted acrosomal membrane [FITC-PSA (–)/Fixable Live/Dead Red (–)]; 3) sperm with damaged plasma membrane and intact acrosomal membrane [FITC-PSA (+)/Fixable Live/Dead Red (+)]; 4) sperm with both damaged plasma and acrosomal membranes [FITC-PSA (–)/Fixable Live/Dead Red (+)]. Following incubation with FITC-PSA, the samples were diluted in 150 µL Accumax and processed immediately using a flow cytometer (FACScan, Beckton Dickinson, Mountain View, CA) equipped with a 488-nm argon laser at 20 mW and three fluorescent detectors (FL1, bandpass 530/30nm; FL2, bandpass 585/42nm; and FL3, long pass 670 nm). The voltage settings on the flow cytometer were: FSC-H, 553; SSC, 240; FL1, 741, and FL2, 821. The compensation was set on FL2 as 98% of FL1. The FITC-PSA signal was acquired using the FL1 filter, while the Fixable Live/Dead Red stain signal was acquired using the FL2 filter. The flow rate was 200–400 sperm/s and a minimum of 5000 sperm were analyzed per sample. To identify doublets and clumps that could affect the analysis and interpretation of sperm events, a manual gating strategy was applied whereby the FSC and SSC were plotted. This methodology has been validated in previous reports from our laboratory6, 27. Flow cytometry data were analyzed by WinList™ software (Verity Software House, Topsham, ME, USA). The percentage of AE in viable sperm (AE/Viable) was considered as the experimental endpoint.
Sperm preparation for protein extraction
Aliquots of frozen/thawed sperm (30 x 106 sperm) incubated at either 0h, 2h, 4h, or 6h in Lac-MW medium were flash/frozen in dry ice until further processing, and then thawed for 1 min at 37°C in a water bath. Immediately after thawing, 10 µL of EDTA-free protease inhibitor cocktail (Pierce™ Protease Inhibitor Tablets, Thermo Fisher Scientific, Waltham, MA) was added per each sperm aliquot. The sperm were washed two times in calcium-free PBS at 400 x g for 5 min, pelleted, maintained frozen at − 80°C, and sent to the Institutional Mass Spectrometry Core Laboratory at the University of Texas Health Science Center at San Antonio (San Antonio, Texas) for further analysis.
Protein identification and relative quantification by data-independent acquisition mass spectrometry (DIA-MS)
Protein aliquots corresponding to 70 µg protein, as quantified using the EZQ™ Protein Quantitation Kit (Thermo Scientific, Waltman, MA), were reduced with tris(2-carboxyethyl)phosphine hydrochloride (TCEP-HCl; Pierce™, Thermo Fisher Scientific), alkylated in the dark with iodoacetamide and applied to S-Traps spin columns (Protifi™, Farmingdale, NY) for tryptic digestion (sequencing grade; Promega) in 50 mM triethylammonium bicarbonate (TEAB; Thermo Fisher Scientific). Peptides were eluted from the S-Traps spin columns with 0.2% formic acid in 50% aqueous acetonitrile and quantified using the Pierce™ Quantitative Fluorometric Peptide Assay (Thermo Fisher Scientific). Data-independent acquisition mass spectrometry (DIA-MS) was conducted on an Orbitrap Fusion Lumos mass spectrometer (Thermo Fisher Scientific). On-line HPLC separation was accomplished with an RSLC NANO HPLC system (Thermo Fisher Scientific/Dionex) and a PicoFrit™ nanospray column (75 µm i.d.; New Objective, Littleton, MA) packed to 15 cm with C18 adsorbent column (218MS 5 µm, 300 Å; Vydac®, W.R. Grace & Co., Columbia, MA); mobile phase A, 0.5% acetic acid (Hac)/0.005% trifluoroacetic acid (TFA; Pierce™, Thermo Fisher Scientific) in water; mobile phase B, 90% acetonitrile/0.5% Hac/0.005% TFA/9.5% water; gradient 3 to 42% B in 120 min; flow rate: 0.4 µL/min. A pool was made of all samples, and 2 µg aliquots of the digests were analyzed using gas-phase fractionation and 4-m/z windows [three mass ranges (395-605mz, 595-805mz, 795-1005mz), staggered; 30k resolution for precursor and product ion scans, all in the orbitrap] to create a DIA chromatogram library38 by searching against a Prosit-generated predicted spectral library39 based on the UniProt Equus caballus protein sequence database (44,488 sequences, downloaded on 12-11-2020): peptide mass tolerance, 10.0 ppm; fragment mass tolerance, 10.0 ppm; fixed modification, carbamidomethylation (C); enzyme, trypsin with a maximum of one missed cleavage; peptide charge state, + 2 – +3; peptide length, 6–30; protein FDR, 1%; minimum of two identified peptides; peptide quantification, Encyclopedia (0.8.1) based on the five highest quality fragment ions. Experimental samples were blocked by replicate and randomized within each replicate for sample preparation and analysis, employing 2 µg of peptides and the 2-h HPLC gradient described above. MS data for experimental samples were acquired in the orbitrap using 8-m/z windows (390-1010mz, staggered; 30k resolution for precursor and product ion scans) and searched against the chromatogram library using the same parameters as described above for generation of the chromatogram library. Scaffold DIA (v3.2.1; Proteome Software) was used for all DIA-MS data processing. The data files were converted to mzML format using ProteoWizard (3.0.19254)40. Deconvolution of the staggered windows was performed. Peptides identified in each sample were filtered by Percolator (3.01.nightly-13-655e4c7-dirty)41 to achieve a maximum false discovery rate (FDR) of 0.01. Individual search results were combined, and peptide identifications were assigned posterior error probabilities and filtered to an FDR threshold of 0.01 by Percolator. Peptide quantification was performed by Encyclopedia (version 1.12.31)38. For each peptide, the five highest-quality fragment ions were selected for quantification. Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis were grouped to satisfy the principles of parsimony. Quartile normalization was applied on the log10 peptide intensities across replicates.
Bioinformatic analyses
Scaffold DIA was used to assess differences in the results for samples from fertile and subfertile stallions (A/A-A/A) across the selected time periods (T0h, T2h, T4h, or T6h). A two-way ANOVA design, available in the Scaffold DIA software, was used in which the primary category of analysis was the stallion phenotype (fertile vs. subfertile), while the second category of analysis was the time point (T0h vs. T2h vs. T4h vs. T6h). Significance was assessed using Benjamini-Hochberg multiple testing correction42, resulting in a cutoff of P < 0.0069. Proteins that exhibited log2(fold change) ≤ − 0.585 or ≥ 0.585 (i.e., fold change in protein abundance between fertile and subfertile stallions 1.5x up or down) were used to filter the results for further analysis.
The proteins identified as differentially abundant between stallion groups and among time periods were queried for gene ontology (GO) terms using g:Profiler43, according to cellular component (CC), biological process (BP), and molecular function (MF). Overrepresentation of the proteins of differential abundance was queried using Equus caballus orthologs and a g:SCS threshold value of 0.05. To analyze the overrepresentation of biological pathways, the Kyoto Encyclopedia of Genes and Genomes (KEGG) database was also used in g:Profiler. In addition, pathway enrichment analyses were conducted utilizing both the g:Profiler and Reactome servers using Homo sapiens orthologs, given the increased depth of the human proteome in terms of annotation. To identify functionally grouped GO terms, a network analysis of the differentially abundant proteins was performed using the Cytoscape (version 3.9.1) plugin ClueGo (version 2.5.8)44.
Detection of acrosome proteins of differential abundance between fertile and subfertile TB stallions by immunofluorescence
Indirect immunofluorescence of the protein arylsulfatase F (ARSF) was performed, as previously described6, with some modifications. Briefly, frozen/thawed semen from both fertile and subfertile TB stallions (n = 3 stallions per group) were processed by density gradient centrifugation using 40% Redigrad and diluted to 30 x 106 sperm in Lac-MW. Next, the sperm were centrifuged (600 x g for 5 min) three times in DPBS, applied to poly-L-lysine coated slides for 2h at room temperature, and exposed to 4% paraformaldehyde in DPBS for 20 min at 4°C. As ARSF is a protein of acrosome origin, we did not expose the sperm to any permeabilizing agent (i.e., triton X-100) to avoid any loss of proteins located at the outer acrosomal membrane or acrosomal matrix. Following fixation, the slides were washed again three times using DPBS, blocked with 10% normal goat serum for 1h at room temperature, and washed again three times using DPBS. A rabbit anti-ARSF, diluted 1:100 in 10% normal goat serum, was added to the slides and incubated overnight at 4°C. A negative control was produced in which the slides were not exposed to the primary antibody (i.e., anti-ARSF). Following overnight incubation, the sets of slides were washed three times with DPBS and stained for 1h at room temperature with a goat anti-rabbit IgG Alexa-555-conjugated secondary antibody (excitation: 555 nm; emission: 572 nm) at a 1:100 dilution in 10% normal goat serum. A final washing step was performed using DPBS; then, 3.5 µL of the SlowFade Mountant with DAPI was added, to both prevent fluorescence bleaching and to counterstain the sperm nuclei. Then, a coverslip was carefully applied to the slide, and the slides were evaluated using an Olympus BX-60 fluorescence microscope at a 1,563x magnification. A total of 100 sperm were analyzed per ejaculate, and the localization of ARSF within each spermatozoon was recorded and compared between fertile and subfertile TB stallions.
Heterologous zona pellucida-binding assay
Due to the difficulties in establishing a repeatable method for conventional in vitro fertilization of equine oocytes using frozen/thawed stallion sperm, we performed a heterologous zona pellucida-binding assay to determine the ability of stallion sperm incubated in Lac-MW to undergo acrosomal exocytosis and bind to the zona pellucida (ZP)45, 46. Porcine oocytes were used to determine the ability of frozen/thawed sperm from subfertile and fertile TB stallions to bind to the ZP. Ovaries were procured from a slaughterhouse 1 hour away from our laboratory, and the specimens were maintained at 37°C in an insulated container for transport. Upon arrival at our laboratory, the ovaries were washed with warm PBS and dried with cotton gauze. All visible follicles were sliced using a scalpel blade, and the follicular cavity was flushed using Vigro® Complete Flush Medium (Vetoquinol, Pullman, WA, USA) over a sterile Petri dish. Groups of 50 cumulus-oocyte complexes (COCs) were identified using a stereoscope and transferred into a Petri dish containing 150-µL droplets of maturation medium [M199 with Earle’s salts, supplemented with 5 mU/mL FSH (Sioux Chemicals, Sioux Center, IA, USA), 10 IU/mL hCG (Chorulon, Merck Animal Health, Rahway, NJ, USA), 10% FBS, and 25 µg/mL gentamicin] under light mineral oil, at 38.2°C in a humidified atmosphere of 5% CO2 for 22h. After this period, the COCs were placed into a fresh droplet of maturation medium and incubated for an additional 22h, under the same conditions. Following the maturation period, COCs were denuded of cumulus by repeated pipetting in M199 medium with Hank’s salts containing 10% FBS and 0.5 mg/mL hyaluronidase. Denuded oocytes were considered as matured if a polar body was extruded and present at the periphery of the oolemma. To perform the heterologous zona-binding assay, frozen/thawed semen from three fertile and three subfertile TB stallions was thawed, processed by density gradient centrifugation using 40% Redigrad, and resuspended to 30 x 106 sperm/mL in Lac-MW. A total of 20 mature oocytes were placed in 50-µL droplets of Lac-MW, inseminated with 5 x 104 sperm from each stallion, and incubated for 4 h at 38.2°C in a humidified atmosphere of 6% CO2, 5% O2, and 89% N2. Following the co-incubation period, to remove any loosely bound sperm, the oocytes were washed by repeated pipetting in M199 with Hank’s salts, transferred to a 50-µL droplet of 2% paraformaldehyde in DPBS for 10 min at room temperature and placed into a 50-µL droplet of 2% BSA + 10 µM Hoechst 33342 in DPBS for 10 min at room temperature. The oocytes were loaded onto polylysine-coated slides, carefully covered with a coverslip, and examined at 400x using a BX-60 Olympus fluorescence microscope. For each oocyte, the total number of sperm bound to the ZP was quantified and compared between fertile and subfertile TB stallions.
Statistical analysis
Statistical analyses were performed using commercial software (SAS Version 9.4; SAS Institute, Inc., Corp., Cary, NC, USA). The Shapiro-Wilk test (PROC UNIVARIATE) was conducted to test data distribution. Within periods (T0h, T2h, T4h, or T6h), Student t-tests (PROC TTEST) or the Wilcoxon Ranked Sum test (PROC NPAR1WAY) were used to compare the percent of AE in viable sperm (AE-Viable), the immunolocalization of ARSF in sperm from fertile and subfertile stallions, and the number of sperm bound to the ZP between fertile and subfertile stallions. Group differences were set at P < 0.05.