This current study was approved by the Research Center of Experimental Animals (Approval ID: JkXK (Hb) 200980006) of the Ethical Committee. In this present study, Kunming mice were obtained from the local Central Animal Laboratory and were kept in the experimental animal center at the University at a temperature of 22°C with access to water and food ad libitum. In this study, all of the experimental procedures were performed according the guidelines of the Committee of the Animal Research Institute of the University.

Frozen buffalo bull semen was purchased from a local Breeding Bull station of Hubei Province, Chinawere used for the present study. The Bull spermatozoa were frozen in liquid nitrogen and, for the experiments, were thawed out in a water bath at 37°C and were separated by percoll density gradient centrifugation, as described previously (Marchiani et al., 2014).

Kunming mice sperm were obtained from the cauda epididymis by mincing the epididymis tissue in PBS. This washed mouse sperm were attached to poly-lysine coated slides for further use in experiments.

The polyclonal anti-SUMO-1 antibody (catalog no: sc-9060) was purchased from Santa Cruz biotechnology, Inc. (Texas, U.S.A), the FITC-PNA staining kit was from GENMED SCIENTIFIC (Shanghai, China) and all other reagents were purchased from Sigma-Aldrich Chemical Co., USA, unless stated otherwise.

Assessments of buffalo bull acrosomal status were conducted essentially as previously described (Hutt et al., 2005). Briefly, fixed spermatozoa were collected by centrifugation at 800 × g for 5 min and washed two times in 100 mM ammonium acetate, pH 9.0 (1 ml). The spermatozoa were subsequently resuspended in 0.5 ml of 100 mM ammonium acetate, pH 9.0, the sample was smeared and allowed to air-dry on glass slides. After air drying, the slides were rinsed with tap water, methanol, and water for 5 min each. The sperm were stained for 10 min at RT with Coomassie Brilliant Blue G-250 (50% methanol and 10% acetic acid). The slides were washed three times with dH2O, air-dried, and mounted in glycerol (30% in PBS) before imaging with a Nikon inverted fluorescence microscope to analyze acrosomal staining.

The sperm acrosome and nuclei were assessed using FITC-PNA and 4, 6-diamino-2-phenylindole (DAPI) staining, respectively. The sperm acrosome status was assessed by applying FITC-PNA staining kit (GENMED SCIENTIFIC, Shanghai, China), which is specific for sperm acrosome staining. Briefly, after incubation at 37°C for 30 min, the bull spermatozoa specimen was centrifuged for 5 min at 600 × g, the sample was collected and resuspended in a preservative solution at a final concentration of 2 × 10⁷ sperms cell/ml. A 20 μl sperm sample was smeared over a microscope glass slide at room temperature and was fixed using an ice cold fixing solution for 60 s. After drying, 200 μl of FITC-PNA stain (25 mg/ml) were spread gently over the smear and the slide was incubated in a dark chamber for 20 min at room temperature. The prepared glass slide was further washed with 200 μl PBS or cleaning solution to displace surplus stain. The sperm acrosome status was assessed and photographs were taken using an epifluorescence microscope (Olympus) at wavelengths of 480 and 530 nm.

Sperm cells, recovered from the Percoll gradient, were washed in PBS. The washed buffalo bull sperm were attached to slides and fixed in 4% formaldehyde (Beyotimes, China) for 20 min at RT and the slides were rinsed three times in PBS. The fixed sperm cells were treated for 10 min with 0.3% Triton x100, and the slides were pre-blocked for 60 min with 5% BSA placed at 4°C. The sperm cells were washed with PBS and were incubated with anti SUMO-1 antibody (1:250 in PBS/1% BSA) overnight at 4°C in a dark chamber. The cells were washed three times with PBS and were then incubated with CY-3 labeled secondary antibody (Boster Co., China) at a 1:150 dilution for 60 min at 37°C in a dark room. The cells were then washed 3 × 5 min with PBS before nuclear staining for 5 min with (DAPI, 1:5,000 in PBS) in a dark room. The specimens were washed, and the glass slides were mounted in DABCO. The images were finally analyzed with a Nikon inverted fluorescence microscope with 60 and 100x objective lenses and DAPI, FITC, and CY-3 filter settings. Minimally, 50 cells were analyzed per slide. In order to determine the co-localization of SUMO-1 with acrosome in bull sperm, after immunostaining the sperm with SUMO-1, the slides were further incubated with FITC-PNA as described above.

Protein extraction was performed as previously described (Sang et al., 2011; Kumar et al., 2014). Briefly, after the final wash in PBS, the sperm pellets were resuspended in 400 μl 2x Laemmli buffer [126 mM Tris/HCl, 20% glycerol, 4% SDS, 10 mM PMSF, protease inhibitor cocktail (10 mg/ml; Santa Cruz Biotechnology, CA, USA)], containing isopeptidase inhibitor NEM (Sigma), to inhibit SUMO cleavage from modified proteins. The samples were heated to 100°C (5 min), placed on ice (5 min) and centrifuged at 12,000 × g for 2 min at 4°C before collecting the supernatants. The protein concentration was determined by the BCA assay with a BSA standard (Pierce, Rockford, USA). The samples were stored at −20°C for later use.

Anti-SUMO-1 antibodies and protein A+G Agarose (Beyotimes, China) were used for the Co-Immunoprecipitation according a manufacturer's instructions. Briefly, the sperm protein extracts were incubated with anti-SUMO-1 polyclonal antibody over-night at 4°C on constant shaker. The next day, 40 μl Protein A+G Agarose were added, and the samples were incubated at 4°C with constant shaking. The resin was washed five times with ice cold PBS, and the pellets were resuspended in 2x SDS loading buffer and were heated for 5 min at 100°C before SDS-PAGE electrophoresis for western blotting or mass spectrometry.

After measuring the protein concentration (determined by BCA assay), the sperm lysates were mixed with 2x SDS loading buffer and were loaded over 10% polyacrylamide gels. After SDS-PAGE, the proteins were transferred to PVDF membranes (Millipore Corp., Bedford, MA). The membranes were blocked with 5% skim milk (Sigma-Aldrich), in TBST [10 mM Tris (pH 7.5), 150 mM NaCl and 0.05% Tween 20], and were then incubated with the polyclonal rabbit anti-SUMO-1 IgG (1:500) diluted in blocking buffer overnight at 4°C. Afterward the membrane was rinsed three times in TBST and was then incubated with HRP conjugated anti-rabbit IgG antibody in TBST for 60 min at room temperature and was finally rinsed three times in TBST. Detection of the bands was accomplished with a BM-enhanced chemiluminescence system (Amersham Biosciences, Piscataway, NJ).

The acrosome reactions were stimulated as previously described (Kumar et al., 2014). In brief, the sperm were capacitated in vitro by thawing of frozen semen at 37°C for 40 s and centrifugation at 1,500 rpm for 25 min. The pellets were reconstituted in 2 ml of a modified Tyrode's Hepes-buffered (Sp-TALPH) washing medium and centrifuged twice, at 1,500 rpm for 10 min. After centrifugation, the pellets were washed with a modified Tyrode's bicarbonate-buffered (Sp-TALP). The pellets were re-suspended in Sp-TALP containing concentration of 25 × 10⁶ spermatozoa in a final volume of 0.5 ml in an Eppendorf tube and capacitation was induced by the presence of 10 μg/ml heparin with an incubation for 4 h at the 38.5°C in 5% CO₂. The capacitated spermatozoa were treated for 30 min with 1 mM calcium ionophore A23187 (Sigma Chemical Co.). The acrosome reaction was confirmed by FITC-PNA staining.

Protein digestion was conducted in accordance with the FASP method, essentially as earlier reported (Wisniewski et al., 2009). Briefly, the protein pellet (~30 μg) was solubilized in 30 μl SDT buffer (4% SDS, 100 mM DTT, 150 mM Tris-HCL pH 8.0) at 90°C for 5 min. The detergent, DTT and other low molecular weight components were removed using 200 μl UA buffer (8 M Urea, 150 mM Tris Hcl pH 8.0) by repeated ultrafiltration (Microcon units, 30 kD). Then 100 μl of 0.05 M iodacetamide in UA buffer was added to block the reduced cysteine residues and the samples were incubated in the dark for 20 min. The filters were rinsed with 100 μl of UA buffer three times and then with 100 μl of 25 mM NH₄HCO₃ twice. Finally, the protein suspension was digested with 2 μg trypsin (Promega) in 40 μl 25 mM NH₄HCO₃ overnight at 37°C, and the ensuing peptides were collected. The mass spectrometry investigations were conducted in a Q Exactive MS that was linked to easy nLC (Proxeon Biosystem, now Thermo Fisher Scientific). Six microliters of each fraction was injected for the nanoLC-MS/MS investigation. A peptide admixture of 5 μg was loaded onto a C18 reverse phase column (Thermo scientific easy column, 10 cm long, 75 μm inner diameter, 3 μm resin) into buffer A (Formic acid 0.1%) and was separated by a linear gradient of buffer B (acetonitrile 80% and Formic acid 0.1%) at a 250 nl/min flow rate controlled by Inelli Flow technology over 140 min. The mass spectrometry data were procured utilizing a data dependent top10 procedure energetically selecting the most abundant precursor ions from the survey, scan (300–1800 m/z) for HCD fragmentation. The fortitude of the target value was based pAGC. The dynamical elimination period was 1 min. The investigating scans were acquired by data determination in 70,000 at m/z 200, and the resolutions for the HCD spectra were fixed in 17,500 at m/z 200. The normalized collision energy was 30 eV under a fill ratio that specializes the minimum percentage of target value probable to be attained at the maximum fill time, which was definite as 0.1%. The equipment was run in an enabled mode of peptide identification.

MS/MS spectrometry was investigated utilizing a MASCOT engine (Matrix Science, London, UK; version 2.2) with a non-redundant International Protein Index Arabidopsis sequence database v3.85, declared and released from the European Bioinformatics Institute (http://www.ebi.ac.uk/). For the identification of proteins, the following adoptions were utilized: peptide mass tolerance = 20 ppm; MS/MS tolerance = 0.1 Da; Enzyme = Trypsin; Missed cleavage = 2; fixed modification: Carbamidomethyl (C), and Variable modification: Oxidation (M). The entire described data were established on a 99% assurance for protein discovery as resolute by a false discovery rate (FDR) of ≤ 1%.

Bio-analysis was performed under Ubuntu 16.04 LTS Docker image (Hung et al., 2016) within a custom Python (Python Software Foundation, 2015) scripts using the packages biopython, pyexcel, csv, collection, panda, and matplotlib and R v3.3.1 (R Core Team, 2016) scripts within the ggplot2 and plotrix libraries. Based on the MS/MS identified proteins listed (Table S1), sequences matching with Bubalus bubalis genome (41,499, protein, assembly UMD_CASPUR_WB_2.0) were searched against the NCBI databases (September-2016, https://www.ncbi.nlm.nih.gov) (Geer et al., 2010). The Prediction of Sub Cellular Location (SCL) was performed using two packages for prediction in eukaryotic and animal proteins with single or multiple sites as follows: (I) iLoc-Euk (http://www.jci-bioinfo.cn/iLoc-Euk) (Chou et al., 2011) and (II) iLoc-Animal (http://www.jci-bioinfo.cn/iLoc-Animal) (Lin et al., 2013). The protein sequences (in fasta format) were sent via user interfaces of the dedicated package's web servers. For further understanding of the B. bubalis sperm ejaculated proteome profile, we classified proteins of interest following three strategies: (I) COG (Clusters of Orthologous Group) classification; (II) Protein function classification; and (III) GO Slim (Huntley et al., 2014) classification. The first strategy, the COG analysis, was completed with EggNOG v4.5 (a database of Orthologous Groups and functional annotation covering prokaryotic and eukaryotic species, http://eggnogdb.embl.de) (Huerta-Cepas et al., 2016). The protein sequences were submitted to the web server API. Sequences longer than 800aa were processed with HMMER v3.1b2 (A software suite widely used for analysis of homologous protein and nucleotide sequences with high speed and sensitivity, http://hmmer.org) (Jiang and Ganesan, 2016). HMMER was used for homology sequences, searching within the EggNOG mammalian database annotation (maNOG.hmm). The results were filtered based on score and E-value significance.

The second strategy was the protein function classification and Gene Ontology (GO), based. Analysis was processed by PANTHER v11 (Protein Analysis Through Evolutionary Relationships, http://pantherdb.org) (Mi et al., 2016) with a statistical overrepresentation test and default setting criteria. The third strategy consisted of GO Slim (Huntley et al., 2014) classification. The GO terms reported by PANTHER were classified by the web server tool CateGOrizer v3.218 (May06-2010, A Web-Based Program to Batch Analyze Gene Ontology Classification Categories: http://www.animalgenome.org) (Von Mering et al., 2003).

Conscious about the crucial role of protein-protein interaction (PPI) in spermatozoa process, additional information about PPIs, pathways and COG classification were collected by STRING v10. 0. (A Database of predicted functional associations between proteins, http://string-db.org) (Hu et al., 2008).

The tracking of the PTMs in the protein list previously identified by MS/MS focused on SUMOylation, which one of the most important PTMs affecting the sperm proteome of water buffalo. In order to identify SUMOylated modified protein and SUMO-1 targets, we used several tools for predicting SUMOylation sites, motifs and interactions. Thus, the protein sequences were submitted to (i) GPS-SUMO-2.0 (a tool for the prediction of sumoylation sites and SUMO-interaction motifs, http://sumosp.biocuckoo.org) (Zhao et al., 2014) and (ii) JASSA (a comprehensive tool for prediction of SUMOylation sites and SIMs, http://www.jassa.fr) (Beauclair et al., 2015). The peptide sequence visualization was performed by Seq2Logo (Thomsen and Nielsen, 2012) using the following Parameters: PSSM_LOG; Halfbits and 800 × 800.

In the sperm from Buffalo bull, the acrosome was clearly identified by intense blue staining with Coomassie Brilliant Blue (Figure 1A image a), and the acrosome reaction was confirmed by loss of the Coomassie staining in the acrosome region of the sperm (Figure 1A image a). The evaluation of nuclear and acrosome morphology of the frozen-thawed bull spermatozoa was assessed with DAPI and FITC-PNA staining, respectively, indicating that the acrosome was enveloped with green fluorescence of FITC in the bull spermatozoa (Figure 1A images b–d). The acrosome status of the bull spermatozoa was evaluated through FITC-PNA (Figure 1B) and was classified within four categories. In category I (Figure 1B image a), the spermatozoa showed a strong, bright fluorescence in the acrosomal cap-like structure, resulting from the PNA binding, demonstrating an intact acrosome. In category II (Figure 1B image b), the fluorescence of the acrosomal cap was disrupted or there was fluorescence in the equatorial plane. Category III (Figure 1B image c) displayed fluorescence in the post acrosomal region of the sperm, and category IV (Figure 1B image d) spermatozoa displayed no fluorescence in the acrosome area, indicating a reacted acrosome. The results demonstrated that Coomassie Brilliant Blue staining and the FITC-PNA staining of sperm could be used to evaluate the acrosome status of water buffalo sperm.

The immunofluorescence staining demonstrated the mid piece, neck, head, and acrosome localization of SUMO-1 in bull spermatozoa (Figure 2A). At high magnification on the optimal immune stained samples, the SUMO-1 signal was observed in the acrosome region of sperm heads (Figure 2A). The SUMO-1 staining was not observed in the controls incubated with rabbit IgG instead of anti- SUMO-1 antibody. Additionally, signal specificity was confirmed using a secondary antibody alone under the same fluorescence intensity condition. For comparison, mouse epididymal spermatozoa were also immune-stained within anti SUMO1 antibodies. The SUMO-1 expression was similar to the bull spermatozoa (Figure 2A).

In order to further reveal the relationship of SUMO-1 localization with the acrosome of bull sperm, FITC-PNA staining of bull sperm was also used to co-localize the expression SUMO-1 and the acrosome in bull sperm (Figure 2B). The results demonstrated that SUMO-1 immunosignal was observed in the mid piece, neck, head, and acrosome region of the sperm head, overlapping with the PNA lectin acrosomal labeling (Figure 2B panel a). The immunosignal in the bull spermatozoa was not observed in the control sperm stained with rabbit IgG. The use of secondary antibody alone confirmed the specificity of the signal (Figure 2B panel b).

We further examined whether acrosome-localized SUMO-1 is released from buffalo bull spermatozoa during the acrosome reaction. Capacitated spermatozoa were induced to undergo acrosome reaction by Ca²⁺ ionophore A23187 and the acrosome status was monitored by FITC-PNA staining. The sperm showed FITC-PNA and SUMO-1 staining on an acrosomal region in spermatozoa with an intact acrosome (Figure 2B panel c, left), and this disappeared in the acrosome-reacted spermatozoa (Figure 2B panel c, right and also see Figure 2B panel a, arrow indicated sperm without an acrosome).

To assess the expression profile of SUMO-1 between capacitated sperm and non-capacitated sperm in water buffalo after in vitro capacitation, protein samples extracted from the bull spermatozoa were run analyzed by western blot. After in vitro capacitation, SUMO-1-modifications were observed compared to the non-capacitated sperm (Figure 3A). After capacitation, bands at ~130, 37, and 27 kDa showed a lower intensity, a band at 30 kDa appears and bands at 85 and 50 kDa are enhanced (Figure 3A). Furthermore, after capacitation, the spermatozoa were induced to undergo the acrosome reaction by calcium ionophore A23187, and the expression of SUMO-1 in acrosome reacted sperm was also detected and was compared with the acrosome intact sperm (Figure 3B). There were subtle changes in the SUMO-1 expression profile after the acrosome reaction was induced. The acrosome reaction caused the band at 37 kDa to decrease and a smear of bands below 25 kDa disappeared (Figure 3B).

The SUMO-1 conjugates in water buffalo spermatozoa were first pulled down by Co-Immunoprecipitation using protein A+G agarose and the SUMO-1 antibody, and this was confirmed by Coomassive Brilliant Blue staining of the SDS-PAGE and by a Western blot analysis (Figures 3C,D). The results demonstrated that specific protein bands in the sample with SUMO-1 pull down, but not in the negative control sample (normal rabbit IgG) (Figures 3C,D), indicating that SUMO-1 specific conjugates were pulled down. These bands from the Coomassie blue stained gel (Figure 3C) were analyzed by mass spectrometry, resulting in the identification of 60 proteins that were unique to the SUMO-1 antibody sample compared to the negative control (Table S1). Ten of these proteins were identified in previous studies as being sumoylated, and these are indicated by an asterisk in Table S1.

The identified proteins have multiple functions in germ cells and sperm, including proteins associated with sperm motility, such as tubulin alpha-1C chain andalpha-1B chain, relaxin receptors RXFP4 relaxin (RLN)/insulin like family peptide receptor 4, RLN-3/INSL7 receptor 2, and LOC102283812 RLN-3 receptor 1-like, alpha and gamma actins ACTC1 and ACTG2.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.