Abstract
The horse breeding industry relies upon optimal stallion fertility. Conventional sperm assessments provide limited information regarding ejaculate quality and are not individually predictive of fertilizing potential. The aim of this study was to harness mass spectrometry to compare the proteomic profiles of high- and low-quality stallion spermatozoa, with the ultimate goal of identifying fertility biomarker candidates. Extended stallion semen (n = 12) was fractionated using Percoll density gradients to isolate low-quality and high-quality sperm populations. Motility and morphological assessments were carried out, and proteomic analyses was conducted using UHPLC-MS/MS. High-quality spermatozoa recorded higher total (95.2 ± 0.52% vs 70.6 ± 4.20%; P ≤ 0.001) and progressive motilities (43.4 ± 3.42% vs 27.3 ± 4.32%; P ≤ 0.05), and a higher proportion of morphologically normal cells (50.2 ± 4.34% vs 38.8 ± 2.72%; P ≤ 0.05). In total, 1069 proteins were quantified by UHPLC-MS/MS, of which 22 proteins were significantly more abundant in the high-quality sperm population (P ≤ 0.05). A-kinase anchor protein 4 (AKAP4) and Hexokinase 1 (HK1) were considered possible biomarker candidates and their differential expression was confirmed by immunoblot. Protein expression was significantly correlated with total (AKAP4 R2 = 0.38, P ≤ 0.01; HK1 R2 = 0.46, P ≤ 0.001) and progressive motilities (AKAP4 R 2 = 0.51, P ≤ 0.001; HK1 R2 = 0.55, P ≤ 0.01), percentage rapid (AKAP4 R2 = 0.29, P ≤ 0.05; HK1 R2 = 0.58, P ≤ 0.001), straight-line velocity (HK1 R2 = 0.50, P ≤ 0.01) and straightness (HK1 R2 = 0.40, P ≤ 0.01). Furthermore, AKAP4 was highly susceptible to adduction by 4-hydroxynonenal (4HNE), which resulted in a global reduction in the phosphorylation profiles following capacitation. In conclusion, the proteomic profiles of high- and low-quality stallion spermatozoa differ substantially, and proteins such as AKAP4 and HK1 could serve as biomarkers of ejaculate quality.
Introduction
The selection of stallions for breeding focuses on pedigree and racetrack success rather than heritable reproductive traits, therefore, stallions experience lower per cycle conception (PCC) rates compared to other livestock species (Colenbrander et al. 1993, Parlevliet & Colenbrander 1999, Morris & Allen 2002, Nath et al. 2010). The Thoroughbred stud book prohibits the use of assisted reproductive technologies, while also imposing an official foal birthdate of January 1st in the northern hemisphere, and August 1st in the southern hemisphere; effectively reducing the commercial breeding season to a 4-month period. Such measures mean that breeders experience significant pressure to achieve pregnancies as close to the start of the breeding season as possible, as larger and more mature horses will reach greater success both at yearling sales and on the racetrack (Pagan et al. 2005, Takahashi 2015).
The result of a given breeding cannot be ascertained until approximately 14 days post-breeding or insemination. Conventional sperm assessments, such as sperm concentration, viability, motility, and morphological assessment, although valuable, provide only limited information regarding ejaculate quality, and are not entirely predictive of the fertilizing potential (Parlevliet & Colenbrander 1999, Colenbrander et al. 2003, Petrunkina et al. 2007). Moreover, our knowledge of the underlying molecular mechanisms leading to the formation of defective spermatozoa – and therefore periods of subfertility – is lacking compared to certain other species (Park et al. 2012, Intasqui et al. 2013, Kwon et al. 2015a,b, Cui et al. 2016, Netherton et al. 2017).
Unravelling the events that govern a conception success or failure, and understanding the molecular cascades that regulate fertility, requires a systems biology approach utilizing high throughput omics technologies. The advent of increasingly sophisticated proteomic technologies has enabled many detailed investigations, not only into the composition of mammalian spermatozoa (Baker et al. 2008, Somashekar et al. 2015, Swegen et al. 2015, Sergeant et al. 2016, Martín-Cano et al. 2020), but also into the identification of numerous proteomic fertility biomarker candidates in mice (Baker et al. 2005), men (de Mateo et al. 2007, Netherton et al. 2017), bulls (Park et al. 2012, Somashekar et al. 2017) and boars (Kwon et al. 2015a,b). To our knowledge, only one study has utilized omics technologies in identifying such biomarkers in stallion ejaculates (Novak et al. 2010), and reported that sperm protein expression accounted for only 10% of fertility variability. This study analyzed the full complement of cells within ejaculates, with no selection criteria to remove the dead or indeed low-quality cells; unlikely to reach the site of fertilization. Consequently, it remains unclear how a high-quality and high functioning spermatozoon, capable of fertilization, differs in composition from a low-quality cell.
Density gradient centrifugation is currently the most effective method of isolating sperm populations of differing quality within an ejaculate. Sperm populations isolated using this technique are associated with greater yields of motile, viable, morphologically normal spermatozoa, with improved fertilizing capacity, and higher pregnancy success rates (Guerin et al. 1989, Punjabi et al. 1990, Shalika et al. 1995, Sharma & Agarwal 1996, Carbone et al. 1997, Morrell et al. 2011, Karamahmutoglu et al. 2014). This technique also enables the direct comparison of sperm parameters and molecular attributes, facilitating the identification of biomarkers of fertility (Park et al. 2012, Cui et al. 2016, Netherton et al. 2017).
We hypothesize that by harnessing mass spectrometry technologies to compare the proteomic profile of high-quality and low-quality spermatozoa, separated using discontinuous density gradient centrifugation, we will identify robust biomarkers of ejaculate quality.
Materials and methods
Reagents
Unless specified, research-grade chemical reagents were obtained from Sigma-Aldrich. A modified Biggers, Whitten and Whittingham (BWW) medium (Biggers et al. 1971) containing 95 mM NaCl, 4.7 mM KCl, 1.7 mM CaCl2.2H2O, 1.2 mM KH2PO4, 1.2 mM MgSO4.7H2O, 25 mM NaHCO3, 5.6 mM d-glucose, 275 mM sodium pyruvate, 3.7 μL/mL 60% sodium lactate syrup, 50 U/mL penicillin, 50 mg/mL streptomycin, 20 mM HEPES (GE Healthcare) and 0.1% (w/v) polyvinyl alcohol, with an osmolarity of 310 mOsm/kg, was utilized throughout this study. The following antibodies were purchased to characterize proteins of interest: mouse monoclonal anti-AKAP4 (clone 7E10) from 4BioDx (Lille, France), mouse monoclonal anti-Hexokinase1 (Abcam), monoclonal anti-Tubulin (Sigma-Aldrich), rabbit polyclonal anti-4HNE (HNE11-S; Alpha Diagnostic International, San Antonio, TX, USA) rabbit polyclonal anti-GAPDH antibodies (G9545; Sigma-Aldrich), monoclonal anti-phosphotyrosine (P5872; Sigma-Aldrich). Percoll, nitrocellulose membranes and enhanced chemiluminescence (ECL) detection reagents were purchased from GE Healthcare, and Tris was purchased from Astral Scientific (Taren Point, NSW, Australia). Chloroform and methanol were purchased from Fronine (Riverstone, NSW, Australia) at the highest purity available. Ultrapure water was from Fluka (Castle Hill, NSW Australia). 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS) was obtained from Research Organics (Cleveland, OH, USA). Dynabeads Protein G and 3,3′-dithiobis[sulfosuccinimidyl propionate] (DTSSP) were purchased from Thermo Fisher Scientific. Mini-PROTEAN TGX Precast (4–20%) Protein Gels were purchased from Bio-Rad Laboratories. Paraformaldehyde was obtained from ProSciTech (Thuringowa, Qld, Australia).
Preparation of Stallion Spermatozoa
Institutional and New South Wales State Government ethical approval was secured for the use of animal material in this study. Semen samples were collected on three occasions from four normozoospermic Shetland and Miniature crossbred pony stallions (between six and nine years of age) of proven fertility, housed on institution-approved premises. Stallions had access to native pasture 24 h a day and were provided with supplementary feed consisting of grass and lucerne hay once daily.
Semen was collected using a pony-sized Missouri or Colorado artificial vagina (Minitube Australia, Ballarat, VIC, Australia) with an inline semen filter. Ejaculates were immediately diluted (2:1, extender:semen) using EquiPlus extender (Minitube Australia) and transported to the laboratory (~1 h). On arrival at the laboratory, the extended semen was fractionated using 40%/80% Percoll gradients, as previously described by Mitchell et al. (2011). This technique is frequently utilized to isolate spermatozoa of differing quality – in terms of motility, morphology and DNA integrity – within an ejaculate (Aitken et al. 2013, Xavier et al. 2018). Moreover, Percoll gradients were recently used to successfully identify proteins of differential expression between high- and poor-quality spermatozoa in men (Netherton et al. 2017). Following centrifugation (500 g, 25 min), purified spermatozoa were recovered from the 40%/80% Percoll interface (low-quality sperm), and from the 80% pellet (high-quality sperm). Samples were washed in BWW medium (500 g, 20 min) and resuspended at a concentration of 40 × 106 cells/mL in BWW, for use in all assays.
Motility analysis
Sperm motility was objectively determined using computer-assisted sperm analysis (CASA; IVOS, Hamilton Thorne, Danvers, MA), with the following settings: negative phase-contrast optics, recording rate of 60 frames/s, minimum contrast of 70, minimum cell size of 4 pixels, low-size gate of 0.17, high-size gate of 2.9, low-intensity gate of 0.6, high-intensity gate of 1.74 nonmotile head size of 10 pixels, nonmotile head intensity of 135, progressive smoothed path velocity (VAP) threshold of 50 μm/s, slow (static) cell VAP threshold of 20 μm/s, slow (static) cell straight-line velocity (VSL) threshold of 0 μm/s, and threshold straightness (STR) of 75%. Cells exhibiting a VAP of ≥50 μm/s and a STR of ≥75% were considered progressive. Cells with a VAP greater than that of the mean VAP of progressive cells were considered ‘rapid’. A minimum of 200 spermatozoa, in a minimum of five fields were assessed using 20 μm Leja standard count slides (Gytech, Hawthorne East, VIC, Australia) and a stage temperature of 37°C.
Morphology
Approximately 4 × 106 spermatozoa were fixed in 2% paraformaldehyde for 10 min at 4°C, washed in PBS, and stored in 0.1 M glycine in PBS. Morphology was assessed using phase-contrast microscopy (Olympus CX40) at 1000× magnification under oil immersion. A total of 100 cells were classified per sample as either being normal or having abnormalities of the head, midpiece, tail, or proximal or distal cytoplasmic droplets, as previously described (Sieme 2009).
Measurement of sperm dimensions
Approximately 4 × 106 spermatozoa were fixed in 2% paraformaldehyde, as described previously. Cells were visualized using phase-contrast microscopy (Zeiss Axio Imager A1) at 1000× magnification under oil immersion and the following measurements were recorded on 100 cells per sample; length of head, width of head, length of midpiece and length of tail using Olympus cellSense Standard (v2.3) software.
Cell lysis
High-quality (n = 12) and low-quality spermatozoa (n = 12) isolated from the ejaculates of four stallions, collected on three occasions, were snap frozen in aliquots containing 40 × 106 cells, and stored at −80°C until all samples could be assessed collectively. Approximately 80 μL of lysis buffer consisting of 1% (w/v) C7BzO [3-(4-Heptyl) phenyl-(3-hydroxypropyl) dimethylammoniopropanesulfonate], 7 M urea, 2 M thiourea, and 40 mM Tris (pH 10.4) was added to a pellet of 40 × 106 spermatozoa in Eppendorf tubes, for 1 h at 4°C with constant rotation. Samples were centrifuged (18,000 g, 15 min), and supernatants were recovered and transferred to fresh tubes. Protein quantification was performed using a 2D quant kit (GE Healthcare) following the manufacturer’s protocol.
Protein precipitation
Lysate was collected and reduced using 10 mM 1,4-dithiolthreitol (DTT) for 1 h at room temperature (RT) and then alkylated with 20 mM iodoacetamide (IAA) for one h at RT, in the dark. Approximately 100 μg of protein per sample were precipitated using the methanol/chloroform protocol, as described previously (Wessel & Flügge 1984). Briefly, lysates were adjusted to 400 μL with ultrapure water. Methanol and chloroform were then added to lysates at a ratio of 2:2:1 (lysate:MeOH:CHCl3), inverted and centrifuged briefly (17,000 g, 1 min). The upper aqueous phase was removed (leaving the interfaced, protein precipitate layer intact), and 300 μL methanol was added. Lysates were gently inverted and centrifuged (17,000 g) for 15 min. Supernatants were discarded and lysates were air-dried for 15 min at RT.
In-solution trypsin digestion
Protein pellets were solubilized in 200 μL of 5M urea and mixed with 100 μL of 25 mM ammonium bicarbonate buffer (pH 8.5). Proteins were reduced by adding 30 µL of 10 mM DTT (20 min, 56°C) and alkylated by adding 30 µl of 20 mM IAA (30 min, RT) in the dark. Digestion was performed by adding 1 μL of proteomics grade trypsin (1 μg/μL in trypsin 1 mM HCl) and samples were incubated overnight (37°C). The reaction was stopped using 10 μL of 0.1% formic acid, and filtering samples through 0.2 μm (hydrophilic PTFE) into a 2 mL dark glass vial. Finally, samples were dried using a nitrogen current at 35°C and resuspended in 20 μL of H2O/acetonitrile/formic acid buffer (94.9:5:0.1).
UHPLC-MS/MS analysis
The separation and analysis of the samples were performed using the ultra high performance liquid chromatography tandem mass spectrometry (UHPLC-MS/MS); consisting of an Agilent 1290 Infinity II Series HPLC (Agilent Technologies), equipped with an automated multi-sampler module and a high speed binary pump, and coupled to an Agilent 6550 Q-TOF Mass Spectrometer (Agilent Technologies), using an Agilent Jet Stream Dual electrospray (AJS-Dual ESI) interface. The control of the UHPLC-MS and Q-TOF were made using MassHunter Workstation Data Acquisition software (Agilent Technologies, Rev. B.06.01). Samples were injected onto an Agilent AdvanceBio Peptide Mapping HPLC column (2.7 μm, 150 × 2.1 mm, Agilent technologies), at 55°C, at a flow rate of 0.4 mL/min. The gradient program began with 2% of buffer B (H2O/acetonitrile/formic acid buffer, 10:89.9:0.1) in isocratic mode (5 min), increasing linearly to 45% (40 min), followed by 95% (15 min), remaining constant for five min. Buffer A: H2O/acetonitrile/formic acid buffer, 89.9:10:0.1. The mass spectrometer was operated in the positive mode. The nebuliser gas pressure was set to 35 PSI, drying gas flow was set to 10 L/min at a temperature of 250°C, and the sheath gas flow was set to 12 L/min at a temperature of 300°C. The capillary spray, fragmentor and octopole RF Vpp voltages were 3500, 340 and 750 V, respectively. Profile data were acquired for both MS and MS/MS scans in extended dynamic range mode. MS and MS/MS mass range was 50–1700 m/z, and scan rates were eight spectra/sec for MS, and three spectra/s for MS/MS. Auto MS/MS mode was used with precursor selection by abundance and a maximum of 20 precursors selected per cycle. A ramped collision energy was used with a slope of 3.6 and an offset of −4.8. The same ion was rejected after two consecutive scans.
Proteomic data processing and analysis
Data processing was performed using Spectrum Mill MS Proteomics Workbench (rev B.04.01, Agilent Technologies). Briefly, raw data were extracted under default conditions as follows: non fixed or variable modifications were selected; [MH]+ 50–10,000 m/z; maximum precursor charge +5; retention time and m/z tolerance ± 60 s; minimum signal-to-noise MS (S/N) 25; finding 12C signals. The MS/MS was searched against the UniProt horse database (downloaded September 2019) and performed with the following criteria; no fixed modifications, carbamidomethylated cysteines (variable modification); tryptic digestion with a maximum of five missed cleavages; ESI-Q-TOF instrument; minimum matched peak intensity 50%; maximum ambiguous precursor charge +5; monoisotopic masses; peptide precursor mass tolerance 20 ppm; product ion mass tolerance 50 ppm; and calculation of reversed database scores. Validation of peptide and protein data was performed using auto thresholds with a false discovery rate of 1.2% and summarized using the validations; score >5 and 60% SPI (Scored Peak Intensity: the percentage of the extracted spectrum that is explained by the database search result).
Exported protein lists were imported to Perseus (version 1.6.10.43; Tyanova et al. 2016), and z-scoring normalization was performed on transformed data, followed by Student’s t-tests. From this data, volcano plots were created between high- and low-quality populations harvested from each individual stallion, where the x-axis is the log2 fold change and the y-axis is the –log10 P value score (Nixon et al. 2019b). Using the refined protein list (Supplementary Table 1, see section on supplementary materials given at the end of this article), accessions were mapped to UniProt to collate information on level of protein evidence and associated Gene Ontology classifications of biological processes, molecular function and cell localization. In cases of obsolete or deleted protein entries (109 proteins), a Basic Local Alignment Search Tool (BLAST) search was carried out to identify putative genes, based on a minimum 50% shared identity (Supplementary Table 1). The initial protein list was further refined, based on the inclusion of proteins with a quantitative value in at least two technical replicates of one stallion, resulting in the identification of 499 proteins (denoted by fold change and P-values; Supplementary Table 1).
Proportion of morphological abnormalities present in high- (80%-Percoll fraction) and low-quality (40%-Percoll fraction) sperm populations.
| Classification | High quality sperm fraction (%) | Low quality sperm fraction (%) | P-value |
|---|---|---|---|
| Morphologically normal | 50.2 ± 4.34 | 38.8 ± 2.72 | ≤0.05 |
| Head abnormality | 17.6 ± 2.59 | 24.2 ± 2.97 | N/S |
| Mid-piece abnormality | 21.6 ± 3.22 | 21.5 ± 1.96 | N/S |
| Tail abnormality | 7.2 ± 2.27 | 9.5 ± 2.41 | N/S |
| Proximal cytoplasmicdroplet | 1.9 ± 0.78 | 4.2 ± 1.05 | N/S |
| Distal cytoplasmic droplet | 1.6 ± 0.66 | 1.9 ± 0.59 | N/S |
SDS-PAGE and immunoblotting
Candidate proteins that demonstrated a statistically significant fold-change between sperm fractions, and a spectral count >100, were identified. These proteins were cross-checked against a literature search of studies identifying fertility biomarker proteins in various species. A-kinase anchor protein 4 (AKAP4) and Hexokinase 1 (HK1) were repeated implicated (Intasqui et al. 2013, Frapsauce et al. 2014, Cui et al. 2016, Blommaert et al. 2019) and were therefore chosen for validation by immunoblotting. All cells were boiled (100°C, 5 min) in sodium dodecylsulphate (SDS) extraction buffer (0.375 M Tris pH 6.8, 2%, w/v SDS, 10%, w/v sucrose, protease inhibitor cocktail) to extract protein. Samples were centrifuged (17,000 g, 15 min, 4°C) and soluble protein remaining in the supernatant was quantified using a DC protein assay kit (Bio-Rad). A total of 10 μg of protein from each sample was boiled (5 min, 100°C) in SDS-PAGE (PAGE) sample buffer (2%, v/v mercaptoethanol, 2%, w/v SDS, and 10%, w/v sucrose in 0.375 M Tris, pH 6.8, with bromophenol blue). Protein was resolved by SDS-PAGE (150 V, 1 h), transferred to nitrocellulose membranes (350 mA, 1 h), and blocked (1 h, RT) in 5% skimmed milk, or 3% BSA in Tris-buffered saline (TBS; 100 mM Tris–HCl, pH 7.6, and 150 mM NaCl) supplemented with 0.1% Tween-20 (TBST). Primary antibodies for AKAP4 (1:1000) and 4HNE (1:500) were diluted in 1% skim milk/TBST, while those for HK1 (1:1000), Tubulin (1:3000) and GAPDH (1:4000) were diluted in 1% BSA/TBST. Membranes were incubated in primary antibody overnight (4°C) under rotation and probed (1 h, RT) with appropriate horseradish peroxidase-conjugated secondary antibodies diluted in 1% skim milk/TBST or BSA/TBST (anti-rabbit 1:2500; anti-mouse 1:2000). Following washing (three times for 5 min in TBST), cross-reactive proteins were visualized using an enhanced chemiluminescence (ECL) kit, according to the manufacturer’s instructions, and exposed to a digital camera LAS-3000 system (GE Healthcare). Semi-quantification was achieved using ImageJ software (NIH). Sperm SDS lysate blots were stripped and re-probed with anti-Tubulin or anti-GAPDH antibody as a loading control.
List of identified proteins with significantly higher abundance in high-quality Percoll-derived spermatozoa, compared to low-quality Percoll-derived spermatozoa.
| Uniprot accession no. | Protein | Gene | P-value | Fold change |
|---|---|---|---|---|
| F6Y1A5 | Phospholipase A2 receptor 1 | PLA2R1 | 0.001 | 104.35 |
| F7CZM1 | Cytochrome c oxidase subunit 5A | COX5A | 0.001 | 13.50 |
| Q6H321 | Kallikrein-1E2 | KLK1E2 | 0.002 | 9.69 |
| K9K2H8 | Mitochondrial 3-ketoacyl-CoA thiolase-like protein | ACAA2 | 0.002 | 16.50 |
| F6TCZ6 | GRB2 associated regulator of MAPK1 subtype 2 | GAREM2 | 0.002 | 195.25 |
| F7DN04 | RPTOR independent companion of MTOR complex 2 | RICTOR | 0.003 | 82.37 |
| F6ZWQ8 | Parkin coregulated | PACRG | 0.005 | 8.32 |
| F7ANR3 | Proteasome subunit alpha type | PSMA3 | 0.014 | 57.75 |
| F6RBE2 | Glyceraldehyde-3-phosphate dehydrogenase | GAPDHS | 0.014 | 1.06 |
| F7BUS7 | TSC22 domain family member 4 | TSC22D4 | 0.016 | 205.33 |
| F6ZKT4 | Adenylate kinase 8 | AK8 | 0.016 | 38.32 |
| F6Y0B1 | A-kinase anchoring protein 4 | AKAP4 | 0.016 | 1.18 |
| F6ZMI7 | Lysozyme A | LYZL4 | 0.017 | 4.43 |
| F7AAD0 | Sperm acrosome associated 3 | SPACA3 | 0.021 | 2.91 |
| F6ZD33 | Septin 8 | SEPTIN8 | 0.026 | 37.69 |
| B3IVM0 | Pyruvate kinase | PKM | 0.028 | 1.71 |
| F6WSG0 | Oxoglutarate dehydrogenase | OGDH | 0.028 | 29.52 |
| F7BXF3 | Histone H2A | H2AX | 0.030 | 11.55 |
| F6WY50 | NADH:ubiquinone oxidoreductase subunit S4 | NDUFS4 | 0.032 | 92.22 |
| F7E1S0 | Rho GTPase activating protein 11A | ARHGAP11A | 0.049 | 428.08 |
| F7BA85 | Hexokinase 1 | HK1 | 0.050 | 1.53 |
| K9KEM4 | ADP/ATP translocase 2-like protein | SLC25A5 | 0.051 | 3.39 |
Co-immunoprecipitation
Co-immunoprecipitation (IP) was utilized to assess the covalent modification of AKAP4 by 4HNE in high and low-quality sperm populations. Spermatozoa were lysed in IP lysis buffer (10 mM CHAPS, 10 mM HEPES, 137 mM NaCl and 10% glycerol) on constant rotation (2 h, 4°C). Lysates were recovered following centrifugation (13,800 g, 20 min, 4°C) and 100 µL was added to 50 µL aliquots of washed Dynabeads and incubated under rotation to preclear (1 h, 4°C). Anti-AKAP4 antibody (10 µg in 200 µL of PBS) was conjugated to fresh aliquots of washed Dynabeads (supernatant removed) by incubation under rotation (2 h, 4°C). Cross-linking was performed by adding 3,3′-dithiobis[sulfosuccinimidyl propionate] (DTSSP) at a final concentration of 2 mM (30 min, RT), after which 2 mM Tris was added (15 min, RT) to quench the reaction. Dynabeads were then washed three times in 200 µL of lysis buffer and supernatants were retained. Immunoprecipitation was performed by adding 100 µL of lysates to AKAP4 antibody-bound beads and incubating on constant rotation overnight (4°C). Supernatants were transferred to fresh tubes and washed three times in 200 µL of PBS. Beads were resuspended in 100 µL of PBS and transferred to fresh tubes to avoid co-elution of proteins bound to the tube. Antibody-antigen-bound beads, precleared beads, 10 µL of fresh protein G beads (bead-only control), 5 µL anti-AKAP4 (antibody-only control) and 5 µL of the final PBS wash of the antibody bound beads (negative control) were boiled in SDS loading buffer (containing 8% β-mercaptoethanol) for 5 min to elute proteins. The eluted proteins were then loaded onto a 4–20% Tris-Glycine gel, resolved by SDS-PAGE (150 V, 1 h) transferred to nitrocellulose membranes (350 mA, 1 h) and blocked as detailed previously. The blots of eluted sperm proteins were probed with anti-AKAP4, then stripped and re-probed with anti-4HNE antibody to assess 4HNE adduction.
4HNE treatment and capacitation
4HNE was used to assess the effects of aldehyde exposure on tyrosine phosphorylation. Spermatozoa recovered from 40%/80% Percoll interface, and from the 80% pellet were incubated (30 min, 37°C) with 4HNE (0, 50 or 100 µM), washed in BWW (500 g, 3 min) and resuspended in 500 µL of BWW capacitation medium (0.5 mM methyl-β-cyclodextrin, 3.0 mM pentoxifylline, 5 mM dibutyryl cyclic AMP, 3 mg BSA in 10 ml BWW; Bromfield et al. 2014) or normal BWW medium in flat-bottom tubes. All samples were incubated under an atmosphere of 5% CO2 and 95% air for 4 h (37°C). Throughout the incubation period, sperm suspensions were gently inverted at 30 min intervals to prevent settling of the cells. Samples were gently centrifuged (500 g, 3 min), supernatant was removed, and the resulting pellet was frozen (−80°C). Immunoblotting was performed as detailed previously. Membranes were blocked (3% BSA/TBST) and then probed with anti-phosphotyrosine antibody (α-PT66) diluted in 1% BSA/TBST (1:2000) under constant rotation overnight (4°C). Membranes were washed, probed with anti-mouse HRP secondary antibody (1:2000 in 1% BSA/TBST) and visualized.
Statistical analyses
Normality checks were conducted on all data using Shapiro–Wilk test in SPSS version 25 software (IBM Corp.) and where data was not normally distributed, appropriate data transformations were performed depending on the degree and direction of skewness. T-tests and Pearson correlations were performed using SPSS v 25 software (IBM Corp.).
Results
Assessment and characterization of high- and low-quality sperm fractions
Purified poor-quality spermatozoa were recovered from the 40%/80%-Percoll interface and high-quality spermatozoa were recovered from the 80% pellet. Significant differences in motility and cell morphology were evident between high- and low-quality spermatozoa. An average of 43.4% (±3.42%) of the high-quality sperm fraction demonstrated forward, progressive motility, in comparison with only 27.3% (±4.32%) of the low-quality sperm fraction (P ≤ 0.05; Fig. 1A). Specific morphological assessments of both populations revealed 50.2% (±4.34%) of the high-quality sperm fraction were classed as morphologically normal, in comparison to 38.8% (±2.72%) of the low-quality sperm fraction (P ≤ 0.05; Fig. 1B). Morphological abnormalities detected between populations are presented in Table 1.
Total and progressive motility (A), and the percentage of morphologically normal cells (B) in high-quality (closed bar) and low-quality (open bar) sperm fractions (n = 12). Data correspond to mean values ± s.e.m..*P ≤ 0.05, ***P ≤ 0.001.
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Total and progressive motility (A), and the percentage of morphologically normal cells (B) in high-quality (closed bar) and low-quality (open bar) sperm fractions (n = 12). Data correspond to mean values ± s.e.m..*P ≤ 0.05, ***P ≤ 0.001.
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Total and progressive motility (A), and the percentage of morphologically normal cells (B) in high-quality (closed bar) and low-quality (open bar) sperm fractions (n = 12). Data correspond to mean values ± s.e.m..*P ≤ 0.05, ***P ≤ 0.001.
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Proteomic characterization of high- and low-quality sperm fractions
Data obtained from UHPLC-MS/MS identified 1069 proteins (Supplementary Table 1), and the abundance of 22 of these proteins were significantly higher in the high-quality Percoll-fractionated spermatozoa than in the low-quality spermatozoa (Table 2). Based on our analysis using the UniProt Equine database, of the 1069 proteins we identified, only 1% have reported evidence at protein level (Fig. 2A). Of the remaining; 66 have evidence at transcript level only; 309 are inferred from homology; while 578 are predicted, with no evidence at protein, transcript, or homology levels. A total of 109 protein entries were obsolete or deleted, and were replaced with putative genes following a BLAST search (Supplementary Table 1).
Compositional analysis of proteomic data, and inter-stallion variability in protein expression. (A) Classification of protein existence, according to UniProt. (B) Gene Ontology annotation of biological processes. The ten most highly represented categories (highest number of assigned proteins) are shown. (C, D, E and F) Volcano plots corresponding to stallions 1–4 highlight inter-stallion variability in protein expression; x-axis (log2 fold change) and y-axis (–log10 P value; cut off ≥1.3).
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Compositional analysis of proteomic data, and inter-stallion variability in protein expression. (A) Classification of protein existence, according to UniProt. (B) Gene Ontology annotation of biological processes. The ten most highly represented categories (highest number of assigned proteins) are shown. (C, D, E and F) Volcano plots corresponding to stallions 1–4 highlight inter-stallion variability in protein expression; x-axis (log2 fold change) and y-axis (–log10 P value; cut off ≥1.3).
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Compositional analysis of proteomic data, and inter-stallion variability in protein expression. (A) Classification of protein existence, according to UniProt. (B) Gene Ontology annotation of biological processes. The ten most highly represented categories (highest number of assigned proteins) are shown. (C, D, E and F) Volcano plots corresponding to stallions 1–4 highlight inter-stallion variability in protein expression; x-axis (log2 fold change) and y-axis (–log10 P value; cut off ≥1.3).
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
To determine the key biological processes and molecular functions involved in the development of low-quality spermatozoa, protein accessions were mapped to UniProt. Classification according to biological process (Fig. 2B) revealed the dominant categories to be those associated with positive regulation (365 proteins), negative regulation (246 proteins), and the response to stimuli (187 proteins). When the category of positive regulation was analyzed further, the most represented processes included the regulation of transcription and translation (35 proteins); the regulation of protein activity (including expression, localization and post-translational modifications – 64 proteins); and the regulation of cellular processes (including growth, differentiation and apoptosis – 33 proteins). The category of negative regulation mapped to similar primary categories representing 33 (transcription and translation); 32 (regulation of protein activity); and 52 proteins (regulation of cellular processes), while the principal categories comprising ‘response to stimuli’ were; response to stress (including heat and oxidative stress – 32 proteins); pathogenic response (27 proteins); protein response (namely the folding of proteins – 12 proteins); and finally, response to DNA damage (8 proteins). Volcano plots were constructed to highlight the inter-individual variability in protein abundance between high- and low-quality cell populations collected from stallions 1–4 (Fig. 2C, D, E andF, respectively).
Immunoblotting validation
Two candidate proteins were chosen for validation by immunoblotting using the criteria of a statistically significant fold-change between sperm fractions, and a spectral count >100. These included Hexokinase 1 (HK1) and A-kinase anchor protein 4 (AKAP4). In both cases, these proteins were identified as significantly more abundant in the high-quality sperm fraction when compared to the low-quality fraction (P ≤ 0.05). An immunoblot using the anti-HK1 antibody demonstrated a major band of ~100 kDa (Fig. 3A), which could be normalized to the loading control (Fig. 3B); only two representative blots are shown. Band density quantification of all high-quality (n = 12) and low-quality (n = 12) samples analyzed, confirmed our MS analysis findings (Fig. 3C). To highlight the functional significance of these results, the relationship between HK1 band density and sperm motility (n = 12) was assessed. Positive correlations existed between the HK1 abundance in high- and low-quality spermatozoa and total motility (R2 = 0.46, P ≤ 0.001), progressive motility (R2 = 0.55, P ≤ 0.001; Fig. 3D), rapid motility (R2 = 0.58, P ≤ 0.001), VSL (R2 = 0.50, P ≤ 0.01), STR (R2 = 0.40, P ≤ 0.01) and LIN (R2 = 0.32, P ≤ 0.05).
Assessment of HK1 expression in sperm populations. (A) Cells were lysed in an SDS-based extraction buffer, resolved by SDS-PAGE and prepared for immunoblotting with anti-HK1 antibodies (n = 12). (B) Blots were subsequently stripped and probed with anti-Tubulin antibodies as a loading control. Representative blots of lysates from high-quality (‘high’) and low-quality (‘low’) sperm fractions are shown. (C) Band densitometry analysis of HK1 (100 kDa); mean values (±s.e.m.) are presented relative to the corresponding Tubulin control. (D) Correlation between progressive motility and HK1 abundance. **P ≤ 0.01.
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Assessment of HK1 expression in sperm populations. (A) Cells were lysed in an SDS-based extraction buffer, resolved by SDS-PAGE and prepared for immunoblotting with anti-HK1 antibodies (n = 12). (B) Blots were subsequently stripped and probed with anti-Tubulin antibodies as a loading control. Representative blots of lysates from high-quality (‘high’) and low-quality (‘low’) sperm fractions are shown. (C) Band densitometry analysis of HK1 (100 kDa); mean values (±s.e.m.) are presented relative to the corresponding Tubulin control. (D) Correlation between progressive motility and HK1 abundance. **P ≤ 0.01.
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Assessment of HK1 expression in sperm populations. (A) Cells were lysed in an SDS-based extraction buffer, resolved by SDS-PAGE and prepared for immunoblotting with anti-HK1 antibodies (n = 12). (B) Blots were subsequently stripped and probed with anti-Tubulin antibodies as a loading control. Representative blots of lysates from high-quality (‘high’) and low-quality (‘low’) sperm fractions are shown. (C) Band densitometry analysis of HK1 (100 kDa); mean values (±s.e.m.) are presented relative to the corresponding Tubulin control. (D) Correlation between progressive motility and HK1 abundance. **P ≤ 0.01.
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
An immunoblot using the anti-AKAP4 antibody demonstrated major bands of ~82 and ~110 kDa corresponding with AKAP4 and its precursor molecule proAKAP4, respectively (Fig. 4A), which could be normalized to the loading control (Fig. 4B); only three representative blots are shown. Plotting the ~82 kDa AKAP4 band densities of all high-quality (n = 12) and low-quality (n = 12) samples assessed, confirmed the findings of our MS analysis (P ≤ 0.05; Fig. 4C). ProAKAP4 abundance was also significantly higher in the high-quality compared to the low-quality sperm fraction (P ≤ 0.05, n = 12). Strong, positive correlations existed between AKAP4 expression (n = 12) and total motility (R2 = 0.38, P ≤ 0.01), progressive motility (R2 = 0.49, P ≤ 0.001; Fig. 4D) and rapid motility (R2 = 0.29, P ≤ 0.05).
Assessment of AKAP4 expression in sperm populations. (A) Cells were lysed in an SDS-based extraction buffer, resolved by SDS-PAGE and prepared for immunoblotting with anti-AKAP4 antibodies (n = 12). (B) Blots were subsequently stripped and probed with anti-Tubulin antibodies as a loading control. Representative blots of lysates from high-quality (‘high’) and low-quality (‘low’) sperm fractions are shown. (C) Band densitometry analysis of AKAP4 (82 kDa); mean values (±s.e.m.) are presented relative to the corresponding Tubulin control. (D) Correlation between progressive motility and AKAP4 abundance. ***P ≤ 0.001.
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Assessment of AKAP4 expression in sperm populations. (A) Cells were lysed in an SDS-based extraction buffer, resolved by SDS-PAGE and prepared for immunoblotting with anti-AKAP4 antibodies (n = 12). (B) Blots were subsequently stripped and probed with anti-Tubulin antibodies as a loading control. Representative blots of lysates from high-quality (‘high’) and low-quality (‘low’) sperm fractions are shown. (C) Band densitometry analysis of AKAP4 (82 kDa); mean values (±s.e.m.) are presented relative to the corresponding Tubulin control. (D) Correlation between progressive motility and AKAP4 abundance. ***P ≤ 0.001.
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Assessment of AKAP4 expression in sperm populations. (A) Cells were lysed in an SDS-based extraction buffer, resolved by SDS-PAGE and prepared for immunoblotting with anti-AKAP4 antibodies (n = 12). (B) Blots were subsequently stripped and probed with anti-Tubulin antibodies as a loading control. Representative blots of lysates from high-quality (‘high’) and low-quality (‘low’) sperm fractions are shown. (C) Band densitometry analysis of AKAP4 (82 kDa); mean values (±s.e.m.) are presented relative to the corresponding Tubulin control. (D) Correlation between progressive motility and AKAP4 abundance. ***P ≤ 0.001.
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Sperm dimensions
To confirm that the underexpression of AKAP4 and HK1 in low-quality sperm fractions was not the result of sperm fragmentation or differences in cell size, we measured the mid-piece length, tail length, head length and head width of 100 cells from each sample. No significant differences were present between the head length (7.0 ± 0.04 µm vs 7.0 ± 0.04 µm), head width (3.7 ± 0.04 µm vs 3.7 ± 0.03 µm), mid-piece length (10.4 ± 0.06 µm vs 10.6 ± 0.06 µm), tail length (48.0 ±0.25 µm vs 47.9 ±0.19 µm), or mid-piece:tail length ratio (ratio 4.6 ± 0.03 vs 4.6 ± 0.03) of spermatozoa isolated from high- and low-quality sperm fractions.
Co-immunoprecipitation
To explore the conservation and extent of AKAP4 aldehyde adduction in stallion sperm fractions, co-immunoprecipitation was performed, whereby sperm lysates were precipitated with anti-AKAP4 antibody, and the eluates were subsequently examined for the presence of AKAP4 and 4HNE. As illustrated in Fig. 5A, this technique effectively isolated AKAP4, and following subsequent stripping and re-probing of these blots, predominant bands of 4HNE were detected at an equivalent molecular weight to AKAP4 (82 kDa), thus confirming AKAP4 as a target for 4HNE adduction in stallion spermatozoa (Fig. 5B). Moreover, densitometry analysis confirmed 4HNE adduction was significantly higher in high-quality spermatozoa (which possess a higher quantity of the AKAP4 substrate; Fig. 5A and C) than low-quality spermatozoa (P ≤ 0.05; Fig. 5C). The inclusion of immunobead-only and antibody-only controls, together with preclear bead eluates, and bead-washes prior to elution, verified the specificity of each immunoprecipitation.
Analysis of 4HNE adduction to AKAP4. (A) Confirmation of AKAP4/4HNE interaction in both high-quality and low-quality sperm populations and was sought using immunoprecipitation, whereby AKAP4 was used to pull down interacting partners. The captured sperm proteins were eluted from protein G Dynabeads and resolved on SDS-PAGE gels alongside an antibody-only control (AB only), a bead-only control (bead only), sperm lysates (high-quality and low-quality fraction elutions) and precleared controls (high-quality and low-quality fraction preclears). (B) Re-probing these blots with anti-4HNE revealed heavy adduction of 4HNE to AKAP4 at 82 kDa in both elutions. (C) Band densitometry analysis of 4HNE bound to AKAP4 (82 kDa); mean values (±s.e.m.) are presented relative to the corresponding AKAP4 band density. *P ≤ 0.05
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Analysis of 4HNE adduction to AKAP4. (A) Confirmation of AKAP4/4HNE interaction in both high-quality and low-quality sperm populations and was sought using immunoprecipitation, whereby AKAP4 was used to pull down interacting partners. The captured sperm proteins were eluted from protein G Dynabeads and resolved on SDS-PAGE gels alongside an antibody-only control (AB only), a bead-only control (bead only), sperm lysates (high-quality and low-quality fraction elutions) and precleared controls (high-quality and low-quality fraction preclears). (B) Re-probing these blots with anti-4HNE revealed heavy adduction of 4HNE to AKAP4 at 82 kDa in both elutions. (C) Band densitometry analysis of 4HNE bound to AKAP4 (82 kDa); mean values (±s.e.m.) are presented relative to the corresponding AKAP4 band density. *P ≤ 0.05
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Analysis of 4HNE adduction to AKAP4. (A) Confirmation of AKAP4/4HNE interaction in both high-quality and low-quality sperm populations and was sought using immunoprecipitation, whereby AKAP4 was used to pull down interacting partners. The captured sperm proteins were eluted from protein G Dynabeads and resolved on SDS-PAGE gels alongside an antibody-only control (AB only), a bead-only control (bead only), sperm lysates (high-quality and low-quality fraction elutions) and precleared controls (high-quality and low-quality fraction preclears). (B) Re-probing these blots with anti-4HNE revealed heavy adduction of 4HNE to AKAP4 at 82 kDa in both elutions. (C) Band densitometry analysis of 4HNE bound to AKAP4 (82 kDa); mean values (±s.e.m.) are presented relative to the corresponding AKAP4 band density. *P ≤ 0.05
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
4HNE adduction and stallion sperm capacitation
Given that AKAP4 was heavily adducted by 4HNE in high-quality stallion spermatozoa, we extended our analysis to assess the consequences of 4HNE adduction on the ability of these cells to undergo capacitation as assessed by phosphotyrosine labelling; a key correlate of the capacitation process. Immunoblotting revealed that 4HNE treatment (both 50 and 100 µM concentrations) caused a global reduction in the tyrosine phosphorylation profile (Fig. 6A, C and D), particularly seen in bands resolving at 64–70 kDa, with significant under-expression or absence in cells which had been subject to 4HNE insult prior to capacitation (Fig. 6E).
4HNE treatment attenuates capacitation-associated tyrosine phosphorylation. High-quality and low-quality sperm populations were either exposed to 4HNE (50 or 100 μM) or untreated (‘UT’), and incubated for 1 h at 37°C. Spermatozoa were induced to capacitate (‘Cap’) or held in a non-capacitated state (‘N/C’) using standard protocols. (A and B) Spermatozoa were subsequently prepared for immunoblotting with anti-phosphotyrosine antibodies and re-probed with a loading control (representative blots shown). (C and D) A densitometric trace of representative immunoblots is included to highlight changes in phospho-labelling in untreated (UT; black trace) and 50 μM 4HNE treated (4HNE; red trace) spermatozoa isolated from the high- and low-quality Percoll fractions. (E) Densitometry analysis of bands 64–70 kDa; mean values (±s.e.m.) are presented relative to the corresponding loading control. Mean values without a common superscript differ significantly (P ≤ 0.001).
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
4HNE treatment attenuates capacitation-associated tyrosine phosphorylation. High-quality and low-quality sperm populations were either exposed to 4HNE (50 or 100 μM) or untreated (‘UT’), and incubated for 1 h at 37°C. Spermatozoa were induced to capacitate (‘Cap’) or held in a non-capacitated state (‘N/C’) using standard protocols. (A and B) Spermatozoa were subsequently prepared for immunoblotting with anti-phosphotyrosine antibodies and re-probed with a loading control (representative blots shown). (C and D) A densitometric trace of representative immunoblots is included to highlight changes in phospho-labelling in untreated (UT; black trace) and 50 μM 4HNE treated (4HNE; red trace) spermatozoa isolated from the high- and low-quality Percoll fractions. (E) Densitometry analysis of bands 64–70 kDa; mean values (±s.e.m.) are presented relative to the corresponding loading control. Mean values without a common superscript differ significantly (P ≤ 0.001).
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
4HNE treatment attenuates capacitation-associated tyrosine phosphorylation. High-quality and low-quality sperm populations were either exposed to 4HNE (50 or 100 μM) or untreated (‘UT’), and incubated for 1 h at 37°C. Spermatozoa were induced to capacitate (‘Cap’) or held in a non-capacitated state (‘N/C’) using standard protocols. (A and B) Spermatozoa were subsequently prepared for immunoblotting with anti-phosphotyrosine antibodies and re-probed with a loading control (representative blots shown). (C and D) A densitometric trace of representative immunoblots is included to highlight changes in phospho-labelling in untreated (UT; black trace) and 50 μM 4HNE treated (4HNE; red trace) spermatozoa isolated from the high- and low-quality Percoll fractions. (E) Densitometry analysis of bands 64–70 kDa; mean values (±s.e.m.) are presented relative to the corresponding loading control. Mean values without a common superscript differ significantly (P ≤ 0.001).
Citation: Reproduction 160, 5; 10.1530/REP-20-0284
Discussion
This is the first study to identify differentially accumulated proteins between the high- and low-quality sperm fractions of stallion ejaculates. A total of 1069 proteins were identified using UHPLC-MS/MS (Supplementary Table 1), of which 22 exhibited significantly higher abundance in the high-quality sperm fraction (Table 2). The identification of these proteins will lead to many future investigations regarding sperm function, quality, composition and overall stallion fertility. Furthermore, we have provided experimental evidence for 1063 stallion sperm proteins which were otherwise predicted, inferred or shown at transcript level only (Fig. 2A).
The differential expression of two of these proteins – Hexokinase 1 (HK1) and A-kinase anchor protein 4 (AKAP4) – was confirmed by immunoblotting analysis. Hexokinases are a family of enzymes functioning in the first step of glucose metabolism, by catalysing the phosphorylation of glucose to glucose-6-phosphate (G6P; Wilson 2003). HK1, the most ubiquitous of the major hexokinase isoforms, is localized to the sperm midpiece and fibrous sheath, via the porin binding domain in its hydrophobic N-terminal (Polakis & Wilson 1985, Wilson 2003, Sun et al. 2008). HK1 has previously been implicated as a marker of fertility in men (Intasqui et al. 2013, Cui et al. 2016, Fanny et al. 2018). Cui et al. (2016) reported a significantly higher abundance of HK1 in the high-quality sperm fraction, while Intasqui et al. (2013) reported higher HK1 abundance in spermatozoa with low levels of DNA damage. In the present study, the functional significance of HK1 was reinforced by the existence of a strong, positive relationship between HK1 expression and numerous motility parameters (Fig. 3D); a result that is not surprising considering the involvement of HK1 in energy metabolism (Wilson 2003).
In spermatozoa, AKAP4 is the most abundant member of the AKAP family – a conserved dynasty of diverse scaffolding proteins – which conduct pivotal roles in fibrous sheath formation, motility, capacitation and acrosome reaction, while orchestrating the integration of cAMP and various components of signal-transduction pathways (Colledge & Scott 1999, Moss et al. 1999, Miki et al. 2002, Brown et al. 2003, Welch et al. 2010, Luconi et al. 2011, Langeberg & Scott 2015, Rahamim Ben-Navi et al. 2016). Localized to the principal piece of the sperm flagellum, AKAP4 accounts for almost 50% of all fibrous sheath proteins (Carrera et al. 1994, Eddy et al. 2003, Krisfalusi et al. 2006). Considering the vital roles played by AKAP, it is not surprising that its underexpression has been repeatedly implicated in infertility in men (Xu et al. 2012, Frapsauce et al. 2014) bulls (Peddinti et al. 2008, Legare et al. 2017, Singh et al. 2019) and stallions (Blommaert et al. 2019). AKAP4 is synthesized from its precursor molecule proAKAP4, (Delehedde et al. 2019, Sergeant et al. 2019), which we identified at ~110 kDa (Fig. 4A). Recently, proAKAP4 expression has been strongly correlated with total and progressive motility in stallion spermatozoa (Blommaert et al. 2018, Blommaert et al. 2019), and has, therefore, become the focus of many investigations into its use as a biomarker of ejaculate quality in dogs, bulls, boars, stallions and men (Delehedde et al. 2018, 2019, Le Couazer et al. 2019, Ruelle et al. 2019, Sergeant et al. 2019). Differences in AKAP4 abundance in the present study most likely stem from defective proAKAP4 expression. Herein, we confirmed previous findings that AKAP4 is significantly more abundant in high-quality spermatozoa isolated using density gradient centrifugation (Cui et al. 2016), and that AKAP4 expression is significantly correlated with motility parameters (Fanny et al. 2018, Blommaert et al. 2019) (Fig. 4D).
Although protein abundance was greater in all high-quality sperm fractions analyzed, we found that AKAP4 was a statistically stronger marker of sperm quality in samples collected from half of the stallions, while HK1 was statistically stronger in samples collected from the other half. Indeed, Fig. 2C, D and F highlight the extent of protein expression variability between stallions. These results indicate first; that the formation of low-quality sperm fractions in an ejaculate is not governed by one singular protein, and second; that some stallions may be more adversely affected by the under-expression of distinct proteins, giving rise to the formation of low-quality spermatozoa. It may be possible that molecular interactions can overcome particular defects that arise during protein synthesis. For instance, sperm tail proteins SPAG5 and ODF1 interact via leucine zipper motifs; deletion of which obliterates this interaction (Shao et al. 2001). However, SPAG5 contains an alternate leucine zipper motif that does not interact with ODF1, suggesting its ability to interact with other proteins to maintain functional capacity (Shao et al. 2001). In the case of results presented here, some stallions may possess the necessary machinery to surmount interruptions to the synthesis of a particular protein (e.g. AKAP4, but not HK1), thereby explaining the differences in statistical strength of sperm quality markers between individuals. It appears that the use of a single biomarker to predict stallion fertility may not be the most appropriate strategy. Rather, there is likely benefit in developing a robust, multi-biomarker diagnostic platform, which accounts for inter-individual variability, and can simultaneously quantify a suite of biomarkers, increasing the predictive power of the assay.
The cytotoxic aldehyde 4-hydroxynonenal (4HNE) is a major end-product of lipid peroxidation, formed following the oxidation of polyunsaturated fatty acids (PUFA). Once produced, 4HNE covalently binds to the sidechains of amino acids, perturbing protein function (Esterbauer et al. 1991, Doorn & Petersen 2002, Dalleau et al. 2013) and leading to pronounced dysfunction in both male (Baker et al. 2015, Bromfield et al. 2015b, Moazamian et al. 2015) and female germlines (Lord et al. 2015, Mihalas et al. 2017). ProAKAP4 and AKAP4 have been identified as primary targets of 4HNE adduction in both round spermatids and mature spermatozoa (Baker et al. 2015, Nixon et al. 2019a). After treating cells with exogenous 4HNE, Baker et al. (2015) identified a number of AKAP4 peptides containing 4HNE carbonyl adducts, and reported a significant 11-fold increase in the number of modified residues present. These modified residues exist in close proximity to numerous important phosphorylation sites (Nixon et al. 2019a). In the present study, immunoprecipitation revealed significant adduction of AKAP4 by 4HNE (Fig. 5), while treatment with exogenous 4HNE resulted in a global reduction in phosphorylation profile (Fig. 6). Equine spermatozoa have a unique matrix of antioxidants that neutralize free radicals, but this ability is lost over time when antioxidant defences become overwhelmed (Gibb et al. 2016). The greater reduction in phosphorylation profile exhibited by the low-quality sperm fraction (Fig. 6C and D) may result from an inferior defense system exposing cells to detrimental oxidative attack more rapidly than higher quality cells.
Stallion spermatozoa are distinct in that their high dependence on oxidative phosphorylation means that markers of oxidative stress are positively correlated with sperm function (Gibb et al. 2014). As AKAPs are highly vulnerable to 4HNE adduction (Nixon et al. 2019a), in functional, high-quality cells, this may explain why equine sperm prove difficult to capacitate, and consequently why IVF in this species has rarely been achieved.
Immunoblotting revealed a significant loss or degradation of bands resolving at 64–70 kDa in non-capacitated, 4HNE treated samples (Fig. 6A and E). We suspect that these bands may represent a multiprotein complex involving heat shock-related 70 kDa protein 2 (HSPA2), arylsulfatase A (ARSA; 68 kDa) and the testis-associated serine proteases testisin (PRSS21; 70 kDa), among others. HSPA2 has been implicated as one of the initial targets of 4HNE-mediated modification (Bromfield et al. 2015b). HSPA2 functions in preventing the aggregation of misfolded proteins (Mayer & Bukau 2005), and therefore protection of other vital cell proteins from oxidative attack. In equine spermatozoa, adduction of 4HNE to HSPA2 disrupts the HSPA2/ARSA/SPAM1 complex that participates in sperm-egg recognition; ultimately rendering the sperm incapable of fertilization (Bromfield et al. 2015a). ARSA – a protein vital for zona pellucida adhesion and binding (Carmona et al. 2002, Redgrove et al. 2012a,b, Bromfield et al. 2015a) – has been identified resolving at 68 kDa in capacitated boar sperm (Jiménez et al. 2006), and shares 90% sequence identity with equine ARSA. Testisin was recently characterized in stallion spermatozoa, and holds a functional relationship with HSPA2 (Swegen et al. 2019). Localized to both the sperm tail and the inner acrosomal membrane, testisin has been implicated as an integral member of the zona pellucida-binding complex (Netzel-Arnett et al. 2009, Swegen et al. 2019). The significant degradation of bands resolving at 64–70 kDa in the present study confirms previous findings that, even in low concentrations, cytotoxic aldehydes can have a significant effect on overall mammalian cell function (Bromfield et al. 2015a,b, Hall et al. 2017).
Driven by the limitations imposed by the Thoroughbred breeding and racing industries, there is increasing pressure to produce foals close to the commencement of the breeding season. Decreases in stallion fertility generate a host of economic and welfare implications concerning both mares and stallions. Increasing our knowledge of the molecular factors that govern stallion fertility, will pave the way toward identifying putative biomarkers of sperm function. Such breakthroughs would facilitate the close monitoring of the molecular changes – predictive of future fertility outcomes – via the regular collection of post-coital dismount semen samples. Sperm populations collected from dismount samples are highly representative of whole ejaculate quality (Gibb et al. 2014). The results of such analyses will enable the timely adjustment of management strategies, well before mares begin to return negative pregnancy diagnoses.
In conclusion, we identified several proteins whose expression is distinctly different between high- and low-quality stallion spermatozoa isolated from the same ejaculate, and established their functional significance by highlighting significant correlations present between protein abundance and motility. The research findings presented herein will go a long way toward the development of a robust, multi-biomarker diagnostic platform, capable of accurately predicting sperm function – the development of which could potentially improve overall horse welfare and the economic conditions surrounding Thoroughbred ownership. Further research is needed to investigate the usefulness of AKAP4 and HK1 for predicting times of high and low fertility in the field.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/REP-20-0284.
Declaration of interest
Maryse Delehedde and Nicolas Sergeant are both co-founders of the SPQI company (Lille, France) that commercialise the AKAP4 antibody used in this study. Robert Aitken is on the editorial board of Reproduction. Robert Aitken was not involved in the review or editorial process for this paper, on which he is listed as an author.
Funding
This work was supported by Ministerio de Ciencia-FEDER (Madrid, AGL2017-83149-R), Junta de Extremadura-FEDER (IB16030 and GR18008) and the Australian Research Council in conjunction with the following partner organisations under Linkage Grant LP160100824: the Hunter Valley Equine Research Council, Memphasys Ltd, Texas A&M University, the University of Sydney, EquiBreed NZ Ltd, CellOxess, Harness Racing Australia and Scone Equine Hospital.
Author contribution statement
Z G, A S and R J A initiated and designed the study with input from R A G, M B, B N, M D and N S. R A G performed experiments and data collection. UHPLC-MS/MS was facilitated by F J P and performed by A S R and F E M-C. Data analysis was performed by R A G with assistance from M B and D A S-B; facilitated by B N. R A G drafted the original manuscript which was sent to all authors for correction.
Acknowledgements
The authors gratefully acknowledge the technical assistance of Shenae L Cafe and Stephanie Waugh.
References
Aitken RJ, Smith TB, Lord T, Kuczera L, Koppers AJ, Naumovski N, Connaughton H, Baker MA, De Iuliis GN2013 On methods for the detection of reactive oxygen species generation by human spermatozoa: analysis of the cellular responses to catechol oestrogen, lipid aldehyde, menadione and arachidonic acid. Andrology 1 192–205. (https://doi.org/10.1111/j.2047-2927.2012.00056.x)
Baker MA, Witherdin R, Hetherington L, Cunningham-Smith K, Aitken RJ2005 Identification of post-translational modifications that occur during sperm maturation using difference in two-dimensional gel electrophoresis. Proteomics 5 1003–1 01 2. (https://doi.org/10.1002/pmic.200401100)
Baker MA, Hetherington L, Reeves GM, Aitken RJ2008 The mouse sperm proteome characterized via IPG strip prefractionation and LC-MS/MS identification. Proteomics 8 1720–1730. (https://doi.org/10.1002/pmic.200701020)
Baker MA, Weinberg A, Hetherington L, Villaverde AI, Velkov T, Baell J, Gordon CP2015 Defining the mechanisms by which the reactive oxygen species by-product, 4-hydroxynonenal, affects human sperm cell function. Biology of Reproduction 92 108. (https://doi.org/10.1095/biolreprod.114.126680)
Biggers J, Whitten W, Whittingham D, 1971 The culture of mouse embryos in vitro. In Methods in Mammalian Embryology. Ed Daniel JC. San Francisco: Freeman.
Blommaert D, Sergeant N, Delehedde M, Franck T, Lejeune JP, Serteyn D2018 Significant correlation between the proAKAP4 concentration and the total and progressive motility in stallion sperm after thawing. Journal of Equine Veterinary Science 66 43. (https://doi.org/10.1016/j.jevs.2018.05.019)
Blommaert D, Sergeant N, Delehedde M, Jouy N, Mitchell V, Franck T, Donnay I, Lejeune JP, Serteyn D2019 Expression, localization, and concentration of A-kinase anchor protein 4 (AKAP4) and its precursor (proAKAP4) in equine semen: promising marker correlated to the total and progressive motility in thawed spermatozoa. Theriogenology 131 52–60. (https://doi.org/10.1016/j.theriogenology.2019.03.011)
Bromfield EG, Aitken RJ, Gibb Z, Lambourne SR, Nixon B2014 Capacitation in the presence of methyl-β-cyclodextrin results in enhanced zona pellucida-binding ability of stallion spermatozoa. Reproduction 147 153–1 66. (https://doi.org/10.1530/REP-13-0393)
Bromfield E, Aitken RJ, Nixon B2015 a Novel characterization of the HSPA2-stabilizing protein BAG6 in human spermatozoa. Molecular Human Reproduction 21 755–769. (https://doi.org/10.1093/molehr/gav041)
Bromfield EG, Aitken RJ, Anderson AL, Mclaughlin EA, Nixon B2015 b The impact of oxidative stress on chaperone-mediated human sperm–egg interaction. Human Reproduction 30 2597–2613. (https://doi.org/10.1093/humrep/dev214)
Brown PR, Miki K, Harper DB, Eddy EM2003 A-kinase anchor protein 4 binding proteins in the fibrous sheath of the sperm flagellum. Biology of Reproduction 68 2241–224 8. (https://doi.org/10.1095/biolreprod.102.013466)
Carbone DJ Jr, Seifarth K, Thomas AJ Jr, Agarwal A1997 Improved motile sperm recovery by a hyperosmotic Percoll gradient. Journal of Assisted Reproduction and Genetics 14 394–397. (https://doi.org/10.1007/BF02766147)
Carmona E, Weerachatyanukul W, Soboloff T, Fluharty AL, White D, Promdee L, Ekker M, Berger T, Buhr M, Tanphaichitr N2002 Arylsulfatase A is present on the pig sperm surface and is involved in sperm-zona pellucida binding. Developmental Biology 247 182–1 96. (https://doi.org/10.1006/dbio.2002.0690)
Carrera A, Gerton GL, Moss SB1994 The major fibrous sheath polypeptide of mouse sperm: structural and functional similarities to the A-kinase anchoring proteins. Developmental Biology 165 272–284. (https://doi.org/10.1006/dbio.1994.1252)
Colenbrander B, Feitsma H, Grooten HJ1993 Optimizing semen production for artificial insemination in swine. Journal of Reproduction and Fertility : Supplement 48 207–2 15.
Colenbrander B, Gadella BM, Stout TA2003 The predictive value of semen analysis in the evaluation of stallion fertility. Reproduction in Domestic Animals 38 305–3 11. (https://doi.org/10.1046/j.1439-0531.2003.00451.x)
Colledge M, Scott JD1999 AKAPs: from structure to function. Trends in Cell Biology 9 216–221. (https://doi.org/10.1016/s0962-8924(99)01558-5)
Cui Z, Sharma R, Agarwal A2016 Proteomic analysis of mature and immature ejaculated spermatozoa from fertile men. Asian Journal of Andrology 18 735–746. (https://doi.org/10.4103/1008-682X.164924)
Dalleau S, Baradat M, Guéraud F, Huc L2013 Cell death and diseases related to oxidative stress: 4-hydroxynonenal (HNE) in the balance. Cell Death and Differentiation 20 1615–16 30. (https://doi.org/10.1038/cdd.2013.138)
de Mateo S, Martinez-Heredia J, Estanyol JM, Dominguez-Fandos D, Vidal-Taboada JM, Ballesca JL, Oliva R2007 Marked correlations in protein expression identified by proteomic analysis of human spermatozoa. Proteomics 7 4264–42 77. (https://doi.org/10.1002/pmic.200700521)
Delehedde M, Bloomaert D, Jouy N, Scabello J, Miersman H, Franck T, Serteyn D, Mitchell V, Sergeant N2018 Concentration of proAKAP4 as a pertinent read-out of sperm quality in mammals. Animal Reproduction Science 194 24. (https://doi.org/10.1016/j.anireprosci.2018.04.053)
Delehedde M, Carracedo S, Selleslagh M, Amirat-Briand L, Sergeant N2019 ProAKAP4 polypeptide as a biomarker of sperm functionality and male fertility disorders. International Journal of Gynecology and Reproductive Sciences 2 13–19.
Doorn JA, Petersen DR2002 Covalent modification of amino acid nucleophiles by the lipid peroxidation products 4-hydroxy-2-nonenal and 4-oxo-2-nonenal. Chemical Research in Toxicology 15 1445–14 50. (https://doi.org/10.1021/tx025590o)
Eddy EM, Toshimori K, O'brien DA2003 Fibrous sheath of mammalian spermatozoa. Microscopy Research and Technique 61 103–1 1 5. (https://doi.org/10.1002/jemt.10320)
Esterbauer H, Schaur RJ, Zollner H1991 Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radical Biology and Medicine 11 81–128. (https://doi.org/10.1016/0891-5849(91)90192-6)
Fanny J, Julien S, Francisco-Jose FG, Sabiha E, Sophie DD, Luc B, Hélène B, Nicolas S, Valérie M2018 Gel electrophoresis of human sperm: a simple method for evaluating sperm protein quality. Basic and Clinical Andrology 28 10–10. (https://doi.org/10.1186/s12610-018-0076-0)
Frapsauce C, Pionneau C, Bouley J, Delarouziere V, Berthaut I, Ravel C, Antoine JM, Soubrier F, Mandelbaum J2014 Proteomic identification of target proteins in normal but nonfertilizing sperm. Fertility and Sterility 102 372–380. (https://doi.org/10.1016/j.fertnstert.2014.04.039)
Gibb Z, Lambourne SR, Aitken RJ2014 The paradoxical relationship between stallion fertility and oxidative stress. Biology of Reproduction 91 77. (https://doi.org/10.1095/biolreprod.114.118539)
Gibb Z, Lambourne SR, Curry BJ, Hall SE, Aitken RJ2016 Aldehyde dehydrogenase plays a pivotal role in the maintenance of stallion sperm motility. Biology of Reproduction 94 133. (https://doi.org/10.1095/biolreprod.116.140509)
Guerin JF, Mathieu C, Lornage J, Pinatel MC, Boulieu D1989 Improvement of survival and fertilizing capacity of human spermatozoa in an IVF programme by selection on discontinuous Percoll gradients. Human Reproduction 4 798–804. (https://doi.org/10.1093/oxfordjournals.humrep.a136989)
Hall SE, Aitken RJ, Nixon B, Smith ND, Gibb Z2017 Electrophilic aldehyde products of lipid peroxidation selectively adduct to heat shock protein 90 and arylsulfatase A in stallion spermatozoa. Biology of Reproduction 96 107–121. (https://doi.org/10.1095/biolreprod.116.145292)
Intasqui P, Camargo M, Del Giudice PT, Spaine DM, Carvalho VM, Cardozo KHM, Cedenho AP, Bertolla RP2013 Unraveling the sperm proteome and post-genomic pathways associated with sperm nuclear DNA fragmentation. Journal of Assisted Reproduction and Genetics 30 1187–1202. (https://doi.org/10.1007/s10815-013-0054-6)
Jiménez I, Fierro R, González-Márquez H, Mendoza-Hernández G, Romo S, Betancourt M2006 Carbohydrate affinity chromatography indicates that arylsulfatase-A from capacitated boar sperm has mannose and N-acetylglucosamine/sialic acid residues. Archives of Andrology 52 455–4 62. (https://doi.org/10.1080/01485010600694690)
Karamahmutoglu H, Erdem A, Erdem M, Mutlu MF, Bozkurt N, Oktem M, Ercan DD, Gumuslu S2014 The gradient technique improves success rates in intrauterine insemination cycles of unexplained subfertile couples when compared to swim up technique; a prospective randomized study. Journal of Assisted Reproduction and Genetics 31 1139–11 45. (https://doi.org/10.1007/s10815-014-0274-4)
Krisfalusi M, Miki K, Magyar PL, O'brien DA2006 Multiple glycolytic enzymes are tightly bound to the fibrous sheath of mouse spermatozoa. Biology of Reproduction 75 270–27 8. (https://doi.org/10.1095/biolreprod.105.049684)
Kwon WS, Oh SA, Kim YJ, Rahman MS, Park YJ, Pang MG2015 a Proteomic approaches for profiling negative fertility markers in inferior boar spermatozoa. Scientific Reports 5 13821. (https://doi.org/10.1038/srep13821)
Kwon WS, Rahman MS, Lee JS, Yoon SJ, Park YJ, Pang MG2015 b Discovery of predictive biomarkers for litter size in boar spermatozoa. Molecular and C ellular Proteomics 14 1230–1240. (https://doi.org/10.1074/mcp.M114.045369)
Langeberg LK, Scott JD2015 Signalling scaffolds and local organization of cellular behaviour. Nature Reviews : Molecular Cell Biology 16 232–2 44. (https://doi.org/10.1038/nrm3966)
Le Couazer D, Delehedde M, Jouy N, Loyens A, Ruelle I, Sergeant N, Michaud S, Lamia BA, Bencharif D2019 ProAKAP4 as a valuable marker to assess sperm quality in dogs. Reproduction in Domestic Animals 54 91–99. (https://doi.org/10.1111/rda.13449)
Legare C, Akintayo A, Blondin P, Calvo E, Sullivan R2017 Impact of male fertility status on the transcriptome of the bovine epididymis. Molecular Human Reproduction 23 355–369. (https://doi.org/10.1093/molehr/gax019)
Lord T, Martin JH, Aitken RJ2015 Accumulation of electrophilic aldehydes during postovulatory aging of mouse oocytes causes reduced fertility, oxidative stress, and apoptosis. Biology of Reproduction 92 33. (https://doi.org/10.1095/biolreprod.114.122820)
Luconi M, Cantini G, Baldi E, Forti G2011 Role of a-kinase anchoring proteins (AKAPs) in reproduction. Frontiers in Bioscience 16 1315–13 30. (https://doi.org/10.2741/3791)
Martín-Cano FE, Gaitskell-Phillips G, Ortiz-Rodríguez JM, Silva-Rodríguez A, Román Á, Rojo-Domínguez P, Alonso-Rodríguez E, Tapia JA, Gil MC, Ortega-Ferrusola C et al.2020 Proteomic profiling of stallion spermatozoa suggests changes in sperm metabolism and compromised redox regulation after cryopreservation. Journal of Proteomics 221 103765. (https://doi.org/10.1016/j.jprot.2020.103765)
Mayer MP, Bukau B2005 Hsp70 chaperones: cellular functions and molecular mechanism. Cellular and Molecular Life Sciences 62 670–6 84. (https://doi.org/10.1007/s00018-004-4464-6)
Mihalas BP, De Iuliis GN, Redgrove KA, Mclaughlin EA, Nixon B2017 The lipid peroxidation product 4-hydroxynonenal contributes to oxidative stress-mediated deterioration of the ageing oocyte. Scientific Reports 7 6247. (https://doi.org/10.1038/s41598-017-06372-z)
Miki K, Willis WD, Brown PR, Goulding EH, Fulcher KD, Eddy EM2002 Targeted disruption of the Akap4 gene causes defects in sperm flagellum and motility. Developmental Biology 248 331–342. (https://doi.org/10.1006/dbio.2002.0728)
Mitchell LA, De Iuliis GN, Aitken RJ2011 The TUNEL assay consistently underestimates DNA damage in human spermatozoa and is influenced by DNA compaction and cell vitality: development of an improved methodology. International Journal of Andrology 34 2–13. (https://doi.org/10.1111/j.1365-2605.2009.01042.x)
Moazamian R, Polhemus A, Connaughton H, Fraser B, Whiting S, Gharagozloo P, Aitken RJ2015 Oxidative stress and human spermatozoa: diagnostic and functional significance of aldehydes generated as a result of lipid peroxidation. Molecular Human Reproduction 21 502–515. (https://doi.org/10.1093/molehr/gav014)
Morrell JM, Johannisson A, Rodriguez-Martinez H2011 Effect of osmolarity and density of colloid formulations on the outcome of SLC-selection of stallion spermatozoa. ISRN Veterinary Science 2011 128984. (https://doi.org/10.5402/2011/128984)
Morris LHA, Allen WR2002 Reproductive efficiency of intensively managed Thoroughbred mares in Newmarket. Equine Veterinary Journal 34 51–60. (https://doi.org/10.2746/042516402776181222)
Moss SB, Turner RMO, Burkert KL, Vanscoy Butt H, Gerton GL1999 Conservation and function of a bovine sperm A-kinase anchor protein homologous to mouse AKAP82. Biology of Reproduction 61 335–342. (https://doi.org/10.1095/biolreprod61.2.335)
Nath LC, Anderson GA, Mckinnon AO2010 Reproductive efficiency of Thoroughbred and Standardbred horses in north-east Victoria. Australian Veterinary Journal 88 169–1 75. (https://doi.org/10.1111/j.1751-0813.2010.00565.x)
Netherton JK, Hetherington L, Ogle RA, Velkov T, Baker MA2017 Proteomics analysis of good and poor quality human sperm demonstrates several proteins are routinely aberrantly regulated. Biology of Reproduction 99 395–408. (https://doi.org/10.1093/biolre/iox166)
Netzel-Arnett S, Bugge TH, Hess RA, Carnes K, Stringer BW, Scarman AL, Hooper JD, Tonks ID, Kay GF, Antalis TM2009 The glycosylphosphatidylinositol-anchored serine protease PRSS21 (testisin) imparts murine epididymal sperm cell maturation and fertilizing ability. Biology of Reproduction 81 921–9 32. (https://doi.org/10.1095/biolreprod.109.076273)
Nixon B, Bernstein IR, Cafe SL, Delehedde M, Sergeant N, Anderson AL, Trigg NA, Eamens AL, Lord T, Dun MD et al.2019a A-kinase anchor protein 4 is vulnerable to oxidative adduction in male germ cells. Frontiers in Cell and Developmental Biology 7 319. (https://doi.org/10.3389/fcell.2019.00319)
Nixon B, Johnston SD, Skerrett-Byrne DA, Anderson AL, Stanger SJ, Bromfield EG, Martin JH, Hansbro PM, Dun MD2019b Modification of crocodile spermatozoa refutes the tenet that post-testicular sperm maturation is restricted to mammals. Mol ecular and Cell ular Proteomics 18 S59–S76. (https://doi.org/10.1074/mcp.RA118.000904)
Novak S, Smith TA, Paradis F, Burwash L, Dyck MK, Foxcroft GR, Dixon WT2010 Biomarkers of in vivo fertility in sperm and seminal plasma of fertile stallions. Theriogenology 74 956–9 67. (https://doi.org/10.1016/j.theriogenology.2010.04.025)
Pagan JD, Koch A, Caddel SE, 2005 Size matters at the sales. In Advances in Equine Nutrition III, pp. 433–437. Ed Pagan JD. Nottingham, UK: Nottingham University Press.
Park YJ, Kwon WS, Oh SA, Pang MG2012 Fertility-related proteomic profiling bull spermatozoa separated by Percoll. Journal of Proteome Research 11 4162–416 8. (https://doi.org/10.1021/pr300248s)
Parlevliet JM, Colenbrander B1999 Prediction of first season stallion fertility of 3-year-old Dutch Warmbloods with prebreeding assessment of percentage of morphologically normal live sperm. Equine Veterinary Journal 31 248–2 51. (https://doi.org/10.1111/j.2042-3306.1999.tb03181.x)
Peddinti D, Nanduri B, Kaya A, Feugang JM, Burgess SC, Memili E2008 Comprehensive proteomic analysis of bovine spermatozoa of varying fertility rates and identification of biomarkers associated with fertility. BMC Systems Biology 2 19. (https://doi.org/10.1186/1752-0509-2-19)
Petrunkina AM, Waberski D, Gunzel-Apel AR, Topfer-Petersen E2007 Determinants of sperm quality and fertility in domestic species. Reproduction 134 3–17. (https://doi.org/10.1530/REP-07-0046)
Polakis PG, Wilson JE1985 An intact hydrophobic N-terminal sequence is critical for binding of rat brain hexokinase to mitochondria. Archives of Biochemistry and Biophysics 236 328–337. (https://doi.org/10.1016/0003-9861(85)90633-2)
Punjabi U, Gerris J, Van Bijlen J, Delbeke L, Gielis M, Buytaert P1990 Comparison between different pre-treatment techniques for sperm recovery prior to intrauterine insemination, GIFT or IVF. Human Reproduction 5 75–83. (https://doi.org/10.1093/oxfordjournals.humrep.a137046)
Rahamim Ben-Navi L, Almog T, Yao Z, Seger R, Naor Z2016 A-kinase anchoring protein 4 (AKAP4) is an ERK1/2 substrate and a switch molecule between cAMP/PKA and PKC/ERK1/2 in human spermatozoa. Scientific Reports 6 37922–37922. (https://doi.org/10.1038/srep37922)
Redgrove KA, Anderson AL, Mclaughlin EA, O'bryan MK, Aitken RJ, Nixon B2012a Investigation of the mechanisms by which the molecular chaperone HSPA2 regulates the expression of sperm surface receptors involved in human sperm–oocyte recognition. Molecular Human Reproduction 19 120–135. (https://doi.org/10.1093/molehr/gas064)
Redgrove KA, Nixon B, Baker MA, Hetherington L, Baker G, Liu DY, Aitken RJ2012b The molecular chaperone HSPA2 plays a key role in regulating the expression of sperm surface receptors that mediate sperm-egg recognition. P L o S O NE 7 e50851. (https://doi.org/10.1371/journal.pone.0050851)
Ruelle I, Charreaux F, Bencharif D, Thorin C, Michaud S, Schmitt E, Sergeant N, Briand-Amirat L, Delehedde M2019 Assessment of the sperm specific protein proAKAP4 as a marker to evaluate sperm quality and fertility in Holstein bulls. Revista Brasileira de Reprodução Animal 43 472.
Sergeant N, Jumeau F, Eddarkaoui S, Sigala J, Dossou GF, Delehedde M, Buee L, Yvoz JF, Mitchell V2016 Investigating proteomic methods and tools to assess sperm quality. Animal Reproduction Science 169 125–126. (https://doi.org/10.1016/j.anireprosci.2016.03.072)
Sergeant N, Briand-Amirat L, Bencharif D, Delehedde M2019 The sperm specific protein proAKAP4 as an innovative marker to evaluate sperm quality and fertility. Journal of Dairy and Veterinary Sciences 11 1–6. (https://doi.org/10.19080/JDVS.2019.11.555803)
Shalika S, Dugan K, Pelesh D, Padilla S1995 A mono-Percoll separation technique improves sperm recovery of normal and male factor specimens when compared with the swim-up technique. Human Reproduction 10 3195–319 7. (https://doi.org/10.1093/oxfordjournals.humrep.a135886)
Shao X, Xue J & van der Hoorn F A 2001 Testicular protein Spag5 has similarity to mitotic spindle protein Deepest and binds outer dense fiber protein Odf1. Molecular Reproduction and Development 59 410-416. (https://doi.org/10.1002/mrd.1047)
Sharma RK, Agarwal A1996 Sperm quality improvement in cryopreserved human semen. Journal of Urology 156 1008–1 01 2. (https://doi.org/10.1016/S0022-5347(01)65687-8)
Sieme H, 2009 Semen evaluation. In Equine Breeding Management and Artificial Insemination, 2nd ed. Ed Samper JC. St. Louis, MO: Saunders Elselvier.
Singh R, Junghare V, Hazra S, Singh U, Sengar GS, Raja TV, Kumar S, Tyagi S, Das AK, Kumar A et al.2019 Database on spermatozoa transcriptogram of catagorised Frieswal crossbred (Holstein Friesian X Sahiwal) bulls. Theriogenology 129 130–145. (https://doi.org/10.1016/j.theriogenology.2019.01.025)
Somashekar L, Selvaraju S, Parthipan S, Ravindra JP2015 Profiling of sperm proteins and association of sperm PDC-109 with bull fertility. Systems Biology in Reproductive Medicine 61 376–3 87. (https://doi.org/10.3109/19396368.2015.1094837)
Somashekar L, Selvaraju S, Parthipan S, Patil SK, Binsila BK, Venkataswamy MM, Karthik Bhat S, Ravindra JP2017 Comparative sperm protein profiling in bulls differing in fertility and identification of phosphatidylethanolamine-binding protein 4, a potential fertility marker. Andrology 5 1032–1051. (https://doi.org/10.1111/andr.12404)
Sun L, Shukair S, Naik TJ, Moazed F, Ardehali H2008 Glucose phosphorylation and mitochondrial binding are required for the protective effects of hexokinases I and II. Molecular and Cellular Biology 28 1007–1017. (https://doi.org/10.1128/MCB.00224-07)
Swegen A, Curry BJ, Gibb Z, Lambourne SR, Smith ND, Aitken RJ2015 Investigation of the stallion sperm proteome by mass spectrometry. Reproduction 149 235–2 44. (https://doi.org/10.1530/REP-14-0500)
Swegen A, Smith ND, Gibb Z, Curry BJ, Aitken RJ2019 The serine protease testisin is present on the surface of capacitated stallion spermatozoa and interacts with key zona pellucida binding proteins. Andrology 7 199–212. (https://doi.org/10.1111/andr.12569)
Takahashi T2015 The effect of age on the racing speed of Thoroughbred racehorses. Journal of Equine Science 26 43–4 8. (https://doi.org/10.1294/jes.26.43)
Tyanova S, Temu T, Sinitcyn P, Carlson A, Hein MY, Geiger T, Mann M, Cox J2016 The Perseus computational platform for comprehensive analysis of (prote)omics data. Nature Methods 13 731–7 40. (https://doi.org/10.1038/nmeth.3901)
Welch EJ, Jones BW, Scott JD2010 Networking with AKAPs: context-dependent regulation of anchored enzymes. Molecular Interventions 10 86–97. (https://doi.org/10.1124/mi.10.2.6)
Wessel D, Flügge UI1984 A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Analytical Biochemistry 138 141–143. (https://doi.org/10.1016/0003-2697(84)90782-6)
Wilson JE2003 Isozymes of mammalian hexokinase: structure, subcellular localization and metabolic function. Journal of Experimental Biology 206 2049–20 57. (https://doi.org/10.1242/jeb.00241)
Xavier MJ, Nixon B, Roman SD, Aitken RJ2018 Improved methods of DNA extraction from human spermatozoa that mitigate experimentally-induced oxidative DNA damage. PL o S O NE 13 e0195003. (https://doi.org/10.1371/journal.pone.0195003)
Xu W, Hu H, Wang Z, Chen X, Yang F, Zhu Z, Fang P, Dai J, Wang L, Shi H et al.2012 Proteomic characteristics of spermatozoa in normozoospermic patients with infertility. Journal of Proteomics 75 5426–5436. (https://doi.org/10.1016/j.jprot.2012.06.021)
