Testis specific histone 2B is associated with sperm chromatin dynamics and bull fertility-a pilot study

Cryopreserved semen straws from Holstein bulls (Alta Genetics, Inc., Watertown, WI, USA) were used in the present study. Straws from 26 bulls were used for the flow cytometry experiment (Table. 1) and 10 bulls were used for immunofluorescence and Western blotting assays (Table. 2). Bulls are considered fertile if they repeatedly mount to serve females and bring about 90% pregnancies in 50 females in a 9-weeks time [24]. Therefore, high fertility (HF) bulls have 80–90% pregnancies in 50 females in a 9-weeks time, whereas low fertility (LF) bulls have 70–80% pregnancies in 50 females in a 9-weeks time and less than 40% is sub fertile. In our study, the bulls were divided as HF and LF as reported in detail in our previous publications [25, 26]. Non-return rate (NRR) on day 40 after insemination were calculated for all bulls and the fertility score was ranked as the deviation of each NRR from the average. In the present study, we grouped the bulls with positive deviation from average NRR as HF and those animals with negative deviation from average as LF. Fertility data of the selected bulls are periodically updated with information from partnering herds [25]. Then, we tested the hypothesis that different cellular intensities, expression levels, and localization of TH2B are associated with sperm chromatin dynamic and bull fertility using methods in immunofluorescence, Western blotting, flow cytometry and bioinformatics.

Flow cytometry experiments were performed according to methods described by Dogan et al. (2015) [27] and Odhiambo et al. (2011) [28], with modifications. Flow cytometry was used to quantify the TH2B cell population by passing individual sperm through a laser beam of the appropriate wavelength. Following settings for BD-FACSCalibur flow cytometer (BD Bioscience San Jose, CA 95131–1807, USA), were used; laser line 488 nm, emission filters 530/30 nm and fluorochrome 542 nm. TH2B molecules were conjugated to fluorescent antibodies to detect and quantify the presence of this protein. Semen straws from 26 Holstein bulls, 13 high fertility (HF) and 13 low fertility (LF), were removed from liquid nitrogen and thawed at 37 °C for 30 s. The extenders were then separated from the cells by centrifugation at 2000 x g at 4 °C for 5 min. And pellets were washed twice in washing buffer (WB: PBS with 0.1% Bovine Serum Albumin BSA) and again centrifuged at 2000 x g at 4 °C for 5 min. The pellets were then fixed in 1 ml of 4% formaldehyde at RT for 1 h in separate centrifuge tubes. The samples were then centrifuged at 3000 x g at 4 °C for 5 min and pellets were resuspended in 250 μl of PBS and immediately permeabilized in 250 μl of 0.1% Triton X-100 in 0.1% sodium citrate in PBS on ice for 2 min. The pellets were resuspended in 500 μl of PBS, filtered through a flow cytometric tube using a cell strainer cap (Becton Dickinson Labware; catalogue no. 352235), and then incubated with the primary antibody at 4 °C overnight. Primary antibody was TH2B (Rabbit polyclonal to Testes Specific Histone H2B; Abcam, Cambridge, MA, USA; catalog # 23913; 1/250 dilution). Next day, samples were centrifuged at 3000 x g at 4 °C for 5 min, washed once in 500 μl of washing buffer, centrifuged at 3000 x g at 4 °C for 5 min and incubated with secondary antibodies for 2 h at RT. The secondary antibody was donkey anti-rabbit IgG-FITC (Santa Cruz, Dallas, Texas, USA; catalog # 2090; 1/250 dilution). Following the incubation, the samples were washed twice in WB (3000 g at 4 °C for 5 min). Sperm samples were then analyzed using the BD-FACSCalibur flow cytometer (BD Bioscience San Jose, CA 95131–1807 USA).

Immunofluorescence was performed according to the methods described by Li et al. (2008) [29] and de Oliveira et al. (2013) [26], with modifications. Briefly, cryopreserved semen straws from five high fertility and five low fertility bulls were thawed in a water bath at 37 °C for 30 s (sec). Sperm samples were washed with PBS containing protease inhibitors (cOmplete; Roche, Indianapolis, IN, USA; catalog # 04693116001), and 10 mM ethylenediaminetetraacetic acid (EDTA). Then, the solution was centrifuged at 2000×g at room temperature (RT) for 5 min (min). In addition, the sperm pellets were incubated with 20 mM CHAPS at RT for 20 min. Sperm chromatin was then decondensed in 10 mM DTT and 1 mg/ml of heparin at RT for 30 min [30]. Moreover, sperm were fixed in 4% paraformaldehyde at 4 °C for 10 min. Following fixation, cells were permeabilized with 0.2% Triton X-100 and 0.1% bovine serum albumin (BSA) in PBS at RT for 15 min. Sperm were then washed in 50, 70, 95 and 100% ethanol at RT for 1 min each. The excess ethanol was removed by quick decanting followed by an additional step of fixation using 100% methanol at −20 °C for 20 min. Excess methanol was removed using washing buffer (WB: PBS containing 0.1% Triton X-100) and the sample was blocked with 1% BSA in the WB at RT for 1 h (h). Sperm were probed with primary antibodies against TH2B (Rabbit polyclonal to Testes Specific Histone H2B; Abcam, Cambridge, MA, USA; catalog # 23913; 1/200 dilution) at 4 °C overnight followed by a washing step and probing with secondary antibody of donkey anti-rabbit IgG-FITC against TH2B (Santa Cruz, Dallas, Texas, USA; catalog # 2090; 1/5000 dilution) at RT for 1 h, and then with 2.5 mg/ml of DAPI at RT for 10 min. Coverslips were placed onto the slides using a drop of an antifade mounting medium (VECTAshield, H-1000) and sealed using a nail polish border. The samples were examined under a confocal fluorescence microscope (Zeiss LSM 510) under 40X and 63X magnifications using immersion oil. The experiments were repeated three times and the data were statistically analyzed.

Sperm TH2B was extracted according to the methods of Aoki et al. (2005) [31] and de Oliveria et al. (2013) [26], with some modifications. Briefly, cryopreserved semen straws from five high fertility and five low fertility bulls were thawed at 37 °C for 30 s and washed twice in PBS with protease inhibitor, centrifuging each time at 700×g at 4 °C for 5 min. An aliquot containing 25–40 × 10⁶ sperm was used to extract the nuclear proteins. In order to lyse the cells, sperm samples were washed twice with 400 μL of 1 mM Phenyl methylsulfonyl fluoride (PMSF) in ddH₂O, centrifuged each time at 700×g at 4 °C for 5 min. Then, 100 μL of 20 mM EDTA, 1 mM phenylmethylsulfonyl fluoride (PMSF), and 100 mM Tris (pH 8.0) were added to the pellets followed by the addition of 100 μL of 6 M guanidine hydrochloride, 575 mM dithiothreitol (DTT), and 200 μL of 552 mM iodoacetamide. The samples were protected from light and incubated at 20 °C for 30 min. In addition, the samples were supplemented with 1 mL of cold ethanol (−20 °C), and each sample was then incubated at −20 °C for 1 min and centrifuged at 12000×g at 4 °C for 10 min. The ethanol wash was repeated once more and the pellet was resuspended in 1 mL of 0.5 M HCl and incubated at 37 °C for 15 min and centrifuged at 10,000×g at 25 °C for 10 min. The supernatant was kept and the nuclear proteins were precipitated by the addition of 300 μL of 100% trichloroacetic acid (TCA) to a final concentration of 20% TCA. The solution was incubated at 4 °C for 5 min and centrifuged at 12000×g for 10 min. The pellet was washed twice in 500 μL of 1% 2-mercaptoethanol in acetone. The final pellet was dried out and stored at −30 °C.

Protein concentration in samples containing sperm protein extracts was determined in triplicates using Quick Start™ Bradford Protein Assay Kit 2 (Bio Rad®, Hercules, CA, USA; catalogue # 5000202). In addition, sperm nuclear proteins were precipitated with cold acetone at −20 °C for 3 h, followed by centrifugation at 10,000×g at 4 °C for 10 min. The supernatants were discarded and the pellets were resuspended in 50 μL of 1× Laemmli sample buffer (Bio Rad®, Hercules, CA, USA) with 5% 2-mercaptoethanol, vortexed (10 s), boiled for 10 min, and stored −30 °C. An aliquot of sperm nuclear proteins (10 μg) was reduced, denatured and separated with a vertical polyacrylamide gel electrophoresis (4–20% SDS-PAGE; Mini-Protean TGX™ gel). Protein bands were transferred from the gels to an Immobilon®-P polyvinylidene difluoride (PVDF) membrane using HEP-1 semi-dry electro blotting (Thermo-Scientific Inc.®; Waltham, MA USA) set at 46 mA for 2.5 h. Binding sites were blocked with 5% BSA in PBS-0.1% Tween 20 (PBS-T) at RT under mild agitation for 1 h, followed by an incubation with primary antibodies against bovine TH2B (Abcam®, Cambridge, MA, USA; 1/1000 dilution; rabbit polyclonal IgG; catalog #23913). Lamin B1 was used as internal loading control (Abcam®, Cambridge, MA, USA 1: 2000, rabbit polyclonal IgG; catalog #16048; Abcam) in PBS-T with 1% of BSA at RT for 1 h. Because, the expression levels of Lamin B1 were not stable, therefore another protein band (~25 KDa) served as second internal control. Membranes were then washed three times for 10 min in PBS-T and incubated with secondary antibodies (1:10,000; donkey ant-rabbit IgG-HRP; sc-2313; Santa Cruz Biotechnology®) and Precision Protein™ StrepTactin-HRP Conjugate (Bio Rad®, Hercules, CA, USA; 1/10,000; catalog # #1610380) in PBS-T with 1% of BSA at RT for 1 h. Membranes were washed with PBS-T three times for 5 min each. The bands were revealed using a chemoluminescence reagent (Clarity™ Western ECL Substrate, Bio-rad®, Hercules, CA, USA) and Image Laboratory software (Bio-Rad®) for 30 s. The band intensity of TH2B histone protein was analyzed using ImageJ 1.x [32].

Emboss dotmatcher was used to compare protein sequences of predicted bovine TH2B with human TH2B, bovine H2B with human H2B, and bovine TH2B with bovine H2B (http://​www.​bioinformatics.​nl/​cgi-bin/​emboss) to ensure the sequence similarity of bovine TH2B for downstream computational biology analysis. Protein sequences of predicted TH2B and H2B of bovine were compared using computational tools. Protein sequences of TH2B and H2B were first extensively searched and aligned using UniProt for the bovine (www.​uniprot.​org, UniProt, 2012). The sequence alignment was performed using Clustal Omega method and the percent identity matrixes between the sequences were obtained (http://​www.​ebi.​ac.​uk/​Tools/​msa/​clustalo/​).

Predicted protein-protein interactions among TH2B, H2B, H2A, H3, H4 TNP1, TNP2, PRM1, BRD4 and CHD5 of bovine, human and mouse were obtained using STRING database (Search Tool for the Retrieval of Inter-Acting Proteins; http://​string-db.​org/) [33]. The testis specific histone 2B (TH2B) in mature sperm interacts with chromatin proteins H2A, H3, H4, H1, TNP1, TNP2, PRM1, BRD4 and CHD5. Thus, the interactome of TH2B using the biological networks gene ontology tool (BiNGO) within Cytoscape 3.3 (http://​www.​cytoscape.​org) was elucidated as well [34]. To analyze the interactome of HIST1H2B (TH2B), a merged-network was generated in Cytoscape by entering the query keywords “HIST1H2B for TH2B”, “H2A”, “H3”, “H4”, “TNP1”, “TNP2”, “PRM1”, “PRM2”, “H2B”, “BRD4” and “CHD5” into the search bar and results were retrieved from the BiNGO plugin to assess overrepresentation of GO categories.

The associations between in vivo fertility scores of bulls with sperm TH2B intensity (measured by flow cytometry) and expression levels of TH2B (estimated by Western blotting) were assessed using mixed model analysis with PROC MIXED from SAS for Windows 9.4. For this mixed model, in vivo fertility group was the fixed effect. The random statement included replicate within repeat, repeat within bull, and bull as random effects. The relationship between fertility score and TH2B intensity was assessed through mixed model linear regression using PROC MIXED with SAS for Windows 9.4. In an initial model, fertility score, fertility group, and their interaction term were included as fixed effects. The random statement included replicate within repeat, repeat within bull, and bull as random effects. Fitting this model indicated no significant interactions between fertility score and fertility group, but the fertility scores were widely separated between the two groups. Accordingly, separate models with fertility score as the fixed effect were fit for both the high and low fertility groups to better understand the relationships between fertility and TH2B intensity. Regression line plots were developed using SGPLOT with SAS for Windows 9.4. The distribution of the conditional residuals was assessed to ensure that assumptions for the statistical models have been met. An alpha level of 0.05 was used to determine statistical significance.

Total of 150,000 sperm per each bull were analyzed to investigate the expression of testis specific histone 2B (TH2B) in sperm from 13 high fertility and 13 low fertility bulls (Table. 1). Flow cytometric measurements of sperm TH2B showed different histogram profiles between HF and LF bulls (Fig. 1). As evaluated by regression analysis, the interactions between TH2B intensity and fertility scores of bulls were significant (P = 0.0182). The expression levels of TH2B was highly significant compared to fertility score in the high fertility group (P = 0.0055; y = −12.3828× + 56.9464) but no significance was found in the low fertility group (Fig. 2; y = 2.0450× + 28.0499).

We determined the cellular localization of TH2B in sperm of low vs. high fertile bulls using immunofluorescence and confocal microscopy.TH2B signal was detectable in sperm head, concentrated over the equatorial and subacrosomal area of sperm. The signal was visualized as a band just beneath the acrosome, with a clear space between the base of the band and the proximal end of tail of sperm (Fig. 3). The immunofluorescence signal for TH2B was brighter in sperm of the low fertile than that of the high fertile bulls (Fig. 4; Table 2). As evaluated by regression analysis, association between TH2B and bull fertility was not different. The fact that the signals measured by immunofluorescence were obtained from 45 spermatozoa per bull could be the reason for the non-significance.

An immunoblotting approach was used to detect the expression of TH2B in the sperm from low vs. high fertility bulls. TH2B protein band intensity was analyzed using ImageJ [32]. Our results showed that the TH2B was consistently detectable in sperm from all of bulls (Fig. 5). The expression levels of TH2B were significantly affected by an interaction between fertility score and fertility group (P = 0.0058). In high fertility bulls, as fertility decreased, TH2B expression increased (P = 0.0182; y = −22,315× + 89,049) signifying that the bulls with the highest fertility had the least TH2B retention in their spermatozoa. In low fertility bulls, as fertility decreased, TH2B expression decreased (P = 0.0737; y = 9603× + 50,993; Fig. 5) indicating that the bulls with the lowest fertility had the highest of retained TH2B in their spermatozoa.

Histone variant of TH2B was identified as the testis specific H2B [35, 36]. TH2B is encoded by the gene HIST1H2B [37]. Mouse TH2B gene as well as TH2B protein have been sequenced [19], and a predicted sequence of bovine TH2B gene HIST1H2BA (http://​www.​ncbi.​nlm.​nih.​gov/​nuccore/​XM_​010825421.​2) has been used in our study, because bovine TH2B gene as well as TH2B protein have not been sequenced yet. Therefore, the predicted gene HIST1H2BA (TH2B) had 85.7% similarity with known bovine H2B, using multiple sequence alignment method Clustal Omega (http://​www.​ebi.​ac.​uk/​Tools/​msa/​clustalo/​) [38]. Comparisons among predicted TH2B protein sequences in bovine and human revealed higher similarity patterns of diagonal lines (Fig. 6). The comparison of predicted bovine TH2B sequence with human TH2B showed that the match was a perfect diagonal line with three frame shift mutation sites, which might have resulted in change in sequence of amino acids (Fig. 6). Similarly, comparison between bovine H2B and human H2B showed the match was a perfect diagonal line having four frame shift mutation sites, resulted change in sequence of amino acids (Fig. 6). The sequence comparison of bovine TH2B with bovine H2B showed two areas of different amino acids, due to frame shift mutations resulting in different amino acid sequences (Fig. 6) [39].

We used the STRING database to depict the interactions of sperm proteins in human, mice, and bovines respectively. Analyses of the interactions of nuclear proteins of developed using STRING for human, mouse, and bovine (Fig. 7) showed the missing link as TH2B; which functions in mediating the transition of nuclear proteins from histone to protamine in bovine spermatozoa. In humans and mice, TH2B is known to interact with chromatin remodelers (BRD4 and CHD5) and is involved in replacement of histones with protamines but function(s) of bovine TH2B has yet to be discovered. However, we found the high prediction of likeliness that bovine TH2B is also involved in eviction of retained histones and replacement of those with highly basic protamines, while assisted by two chromatin remodelers BRD4 and CHD5 (Fig. 7).

The interactome networking of bovine TH2B with nuclear proteins (H2B, H2A, H3, H4 TNP1, TNP2, PRM1, BRD4 and CHD5) showed the significant gene ontology overexpression terms (P < 0.05; Table. 3). We showed that TH2B is involved in male gamete generation, chromosome organization, spermatogenesis, gamete generation, DNA packaging, DNA conformation change, chromatin organization, reproductive process, nucleosome organization, chromatin disassembly, spermatid nucleus elongation, nucleosome assembly, protein-DNA complex disassembly, spermatid nucleus differentiation, sperm motility, chromatin organization, chromatin condensation, single strand break repair, chromatin assembly, chromatin silencing, nucleus organization, spermatid development, spermatid differentiation, and chromatin remodeling (Fig. 8).