HDAC6 deacetylates alpha tubulin in sperm and modulates sperm motility in Holtzman rat

Abstract Histone deacetylase 6 (HDAC6) is an alpha (α)- tubulin deacetylase and its over-expression has been demon- strated to promote chemotactic cell movement. Motility in sperm is driven by the flagella, the cytoskeletal structure comprising the microtubules, which are heterodimers of α- and β-tubulins. We have hypothesized that HDAC6, by virtue of being an α-tubulin deacetylase, might modulate sperm motility. However, the presence of HDAC6 on sperm has hitherto not been reported. In this study, we have demonstrat- ed, for the first time, the presence of HDAC6 transcript and protein in the testicular and caudal sperm of rat. We have observed a significantly overlapping expression of HDAC6 with acetyl α-tubulin (Ac α-tubulin) in the mid-piece and principal piece of sperm flagella, and the co-precipitation of α-tubulin and Ac α-tubulin together with HDAC6 and vice versa in sperm lysates. This indicates that HDAC6 interacts with α-tubulin. The HDAC6 activity of sperm, sperm motility and status of Ac α-tubulin investigated in the presence of HDAC inhibitors Trichostatin A, Tubastatin A and sodium butyrate demonstrate that HDAC6 in sperm is catalytically active and that inhibitors of HDAC6 increase acetylation and restrict sperm motility. Thus, we show that (1) active HDAC6 enzyme is present in sperm, (2) HDAC6 in sperm is able to deacetylate α-tubulin, (3) inhibition of HDAC6 results in increased Ac α-tubulin expression and (4) HDAC6 inhibition affects sperm motility. This evidence suggests that HDAC6 is involved in modulating sperm movement.

Keywords: Acetylated alpha-tubulin . Deacetylase activity . HDAC6 . HDAC inhibitor . Sperm motility . Rat

Introduction

Sperm flagellar motility is well-orchestrated and is attrib- utable to a highly organized microtubule-based structure called the axoneme. The axoneme is composed of 9 doublet microtubules and 2 singlet microtubules running along the length of the flagellum. The axonemal structure is surrounded by auxiliary dense fibres and the fibrous sheath that have no clear active role in the sliding of micro- tubules and flagellar movement. The microtubules are com- posed of α- and β-tubulins, which undergo several post- translational modifications, namely, polyglutamylation, polyglycylation, tyrosylation/detyrosylation and acetyla- tion/deacetylation. Whereas the polyglutamylation of the lateral chain of α-tubulin has been shown to have a role in flagellar motility, detyrosination and acetylation are thought to be important for the assembly of the axoneme (Gagnon et al. 1996). The distribution of acetylated (Ac) α- tubulin is tightly controlled and stereotypic. Ac α-tubulin is most abundant in stable microtubules but is absent from dynamic cellular structures (e.g. neuronal growth cones, leading edges of fibroblasts). Reversible acetylation of α-tubulin has been implicated in
regulating microtubule sta- bility and function (LeDizet and Piperno 1987).

Although several enzymes, namely, the ARD1 subunit of the ARD1/NAT1 complex, N-acetyltransferase (NAT) 10, elongator acetyltransferase complex subunit 3 (ELP3) and MEC17/αTAT1, have been proposed as tubulin ace- tyltransferases (Akella et al. 2010; Creppe et al. 2009; Kalebic et al. 2013a, 2013b; Ohkawa et al. 2008; Shen et al. 2009), histone deacetylase 6 (HDAC6) and Sirtuin 2 (SIRT2) have been identified as tubulin deacetylases (Hubbert et al. 2002; Matsuyama et al. 2002; North et al. 2003; Zhang et al. 2003). SIRT2 is dependent on HDAC6 for tubulin deacetylation, whereas interaction between HDAC6 and tubulin has been established as being independent of other proteins (Zhang et al. 2008; Zhao et al. 2010). This designates HDAC6 as a key player in α-tubulin deacetylation.

The study of Hubbert et al. (2002) with A549 cells has demonstrated that the over-expression of HDAC6 leads to the deacetylation of α-tubulin and that this promotes chemotactic cell movement supporting the idea that HDAC6-mediated deacetylation regulates microtubule- dependent cell motility. This alteration in cell motility has been subsequently demonstrated to be attributable to alterations in the degree of tubulin acetylation or to the acetylation of some unidentified protein (Palazzo et al. 2003).

Of the 10 HDACs known, only HDAC 1 and 6 have been reported in the testis and are implicated to have a role in histone acetylation during spermatogenesis (Hazzouri et al. 2000). In the testis, their presence has been shown in spermatogenic cells, pachytenes, round and elongating spermatids and condensing spermatids. Fractions enriched in condensing spermatids, residual bodies and spermatozoa show a decreased expression of both the HDACs.
There is a dearth of literature on the presence of HDAC6 and the role of acetylation/deacetylation in sperm function. The only reported study in humans has shown hypoacetylation of α-tubulin and poor sperm motility in a man with retinal degeneration (Gentleman et al. 1996). Our own observations demonstrate significantly reduced acetylation of α-tubulin in asthenozoospermic individuals (Bhagwat et al. 2014). As HDAC6 has been shown to deacetylate α-tubulin with a role in cell movement, and since sperm motility involves flagellar activity mediated by the axonemal microtubules that are composed of tubu- lin and are highly acetylated, we have hypothesized that this α-tubulin post-translational modification has a role in sperm motility. The present study reports, for the first time, the existence of HDAC6 on the sperm flagella and demonstrates that it is catalytically active and is involved in regulating sperm motility.

Materials and methods

Experimental animals

Neonatal, pre-pubertal, pubertal and adult male Holtzman rats were used. Food and water were provided ad libitum. Rats were housed in groups of four/cage under conditions of 14 h light and 14 h dark. For immunizations, two adult Belgium White female rabbits were used. The study was approved by the Institutional Animal Ethics Committee (IAEC).

Materials

Trichostatin A (TSA; a general HDAC inhibitor) and sodium butyrate (NaB; inhibits all HDACs except HDAC6) were procured from Sigma-Aldrich (Saint Louis, Mo., USA). Tubastatin A (TBSA; HDAC6 specific inhibitor) was obtained from BioVision (Calif., USA). Monoclonal antibodies to Ac α-tubulin (Clone 6-11B-1) and α-tubulin (B-5-1-12) were acquired from Sigma- Aldrich. HDAC6 antibody used in the immunofluores- cence experiments was obtained from Thermo Fisher Scientific (Ill., USA). Horseradish peroxidase (HRP)-la- belled swine anti-rabbit antibody, rabbit anti-mouse anti- body and fluorescein isothiocyanate (FITC)-labelled swine anti-rabbit antibody were procured from Dako (Denmark); rodamine-labelled goat anti-mouse antibody was obtained from Invitrogen (Carlsbad, Calif., USA). Commonly used reagents, unless otherwise specified, were purchased from Qualigens or SRL India and were analytical grade.

Generating antibody to HDAC6

Polyclonal antibodies were raised in rabbit to the chimeric peptide “DPSVLYVSLYVSLHRYGGYMNEGELR” comprising a B-cell epitope and a T-cell epitope of rat HDAC6 separated by two glycine residues and designed by using EMBOSS: antigenic software (Rice et al. 2000). Briefly, after collection of pre-immune sera, rabbits were immunized with 200 μg peptide in 1 ml Freund’s Complete Adjuvant followed by three booster doses of 100 μg peptide in Freund’s Incomplete Adjuvant at 10-day intervals. Antibody titres for the sera collected after every booster were monitored by indirect enzyme-linked immunosorbent assay (ELISA) by titrating serial dilutions of the pre- and post-immune sera against 1 μg chimeric peptide coated onto the microtitre plate. Binding was detected by using HRP-conjugated swine anti- rabbit IgG and tetramethyl benzidine/H2O2 as substrate. Absorbance was determined at 450 nm. Specificity of the antibody was determined by competitive ELISA by preincu- bating the antisera (1:4000) with 0, 0.25, 0.5, 1, 2, 4 or 8 μg of either the corresponding peptide (DPSVLYVSLYVSLHRYG GYMNEGELR) or an unrelated peptide (VVDSEDLPLN) and then by using the preincubated mixtures to probe the peptide immobilized onto microtitre plates. Preimmune sera preincubated with peptide served as a control (see supplemen- tal data, Fig. S1a, b). Western blot analysis was performed by using the preabsorbed antibody to confirm the specificity of the antipeptide antibody. Caudal sperm proteins resolved by SDS polyacrylamide gel electrophoresis (SDS-PAGE) and transblotted onto nitrocellulose membrane were probed with the antipeptide (HDAC6) antibody (1:4000) or with a 1:4000 dilution of the antipeptide antibody pre-absorbed with 16 μg blocking peptide or with antipeptide antibody pre-absorbed with 16 μg unrelated peptide. Negative control was probed with pre-immune (1:4000) sera. Beta-actin was used as a loading control (see supplemental data, Fig. S1c).

Isolation of rat testicular and epididymal sperm

Testicular, caput and caudal sperm were isolated from the respective tissues by making 2–3 cuts and allowing the release of sperm into 0.1 M phosphate-buffered saline (PBS, pH 7.4) by incubating the teased tissue at 34 °C for 30 min. The supernatants were collected and washed three times with
0.1 M PBS by centrifugation at 800g for 20 min at 4 °C. The sperm pellet thus obtained was used for all the analyses performed in this study. If testicular sperm were used for Western blot analysis, they were first purified from the other testicular cell types by Percoll gradient centrifugation (Suryawanshi et al. 2011). The homogeneity of the testicular sperm population thus obtained was confirmed by microscop- ic analysis of smears of the sample stained by the Papanicolaou method (Fig. 1a). However, testicular sperm were not Percoll-purified if they were to be used for immuno- fluorescence localization studies.

Reverse transcription plus polymerase chain reaction

The presence of HDAC6 transcript in testicular and caudal sperm was determined by reverse transcription followed by the polymerase chain reaction (RT-PCR). Sperm from the testis and caudal region of epididymis were isolated and pelleted as described above. RNA extraction was carried out by using guanidinium-thiocyanate-chloroform extraction (TRIzol, Invitrogen). The purity and concentration of the RNA were determined spectrophotometrically at 260 and 280 nm. The respective RNA (1 μg) was reverse transcribed to cDNA by using the ImProm-II Reverse Transcription System (Promega, Madison, Wis., USA) and 250 ng of this cDNA was amplified by PCR with the Clontech Advantage 2 PCR kit (Clontech Laboratories, Mountain View, Calif.,USA). CAGCTAACCAGACCACGTCA and TAGTAGGCCCTCCTCGGATT were the forward and reverse primers, respectively, for HDAC6, and AGAGGGAAATCGTGCG TGAC and GCCGGACTCATCGTACTCCT were the for- ward and reverse primers for β-actin (housekeeping gene). The reactions were set for denaturation at 94 °C for 1 min, annealing at 59 °C for HDAC6 and at 62 °C for β-actin for 1 min, followed by extension at 72 °C for 1 min. Final extension was at 72 °C for 10 min. Reagent controls and no- reverse-transcriptase controls were examined to check for any contamination from reagents used or for genomic HDAC6 amplification, respectively. The experiments were repeated three times to ensure reproducibility of the results.

Western blot analysis

HDAC6 protein expression in sperm derived from the testis and caudal region of the epididymis was studied by Western blot analysis. Rat testicular and caudal sperm pellets were lysed in 15 mM TRIS–HCl buffer, pH 7.4, containing 0.34 M sucrose, 60 mM KCl, 15 mM NaCl, 0.65 mM spermidine, 2 mM EDTA, 0.5 mM EGTA, 0.05 % Triton X-100, 1 mM dithiothreitol, 0.5 mM phe nylmethane sulfonyl-fluoride as de scribed (Seigneurin-Berny et al. 2001), and the protein concentra- tion was quantified by using Bradford’s method (Bradford 1976). Total proteins (60 μg and 10 μg, respectively) were loaded for analysis of HDAC6 and of α- and Ac α-tubulin. Protein lysates were resolved by electrophoresis on 10 % SDS-polyacrylamide gels by using the standard protocol (Laemmli 1970) and transblotted to nitrocellulose mem- branes (GE healthcare, UK) in duplicate. These blots were further incubated with blocking at room temperature for 1 h in buffer. One blot was probed with a 1:500 dilution of the polyclonal antibody to HDAC6 raised in-house and with monoclonal antibodies to Ac α-tubulin (1:10,000) and α-tubulin (1:10,000), respectively, at room tempera- ture for 1 h. The corresponding other blot was used as negative control and was incubated with only the antibody diluent for Ac α-tubulin and α-tubulin and preimmune- sera in the case of HDAC6 antibody. These controls were carried out to determine any non-specific binding of the antibodies. The blots were washed three times with 0.1 M PBS containing 0.1 % Tween 20 (0.1 % PBST). The blots were then incubated with a 1:3000 dilution of HRP- labelled swine anti-rabbit antibody for HDAC6 and rabbit anti-mouse antibody in the cases of Ac α-tubulin and of α- tubulin followed by three washes with 0.1 % PBST. Chemiluminescent-based detection of the proteins of inter- est was undertaken by using the ECL plus Western blotting detection kit (GE healthcare, UK) following the kit proto- col. Western blot analysis was performed by using Gene Tools version 3.06.

Immunohistochemistry

Sections (5 μm thick) of Bouin-fixed paraffin-embedded testis (cut transversely) and epididymis (cut sagittally to cover the caput, corpus and cauda) were probed to study the expression of HDAC6 protein in the respective tissues. The sections were deparaffinized and rehydrated. Endogenous peroxidase activ- ity was quenched with 0.3 % H2O2 in 70 % methanol for 30 min at room temperature followed by three washes with 0.1 M PBS. Heat-induced antigen retrieval of the sections was carried out in 10 mM sodium citrate buffer, pH 6.0, followed by three washes in 0.1 M PBS. Non-specific sites were blocked by incubating the sections with blocking solution ( Ve ctastain ABC System peroxidase kit; Vector Laboratories, Burlingame, Calif., USA) for 30 min. Sections were incubated with either a 1:100 dilution of HDAC6 anti- body raised in-house or with a 1:100 dilution of pre-immune serum to serve as the negative control to account for non- specific binding of the antibody. Following overnight incuba- tions at 4 °C, sections were washed three times and then incubated with a 1:50 diluted rabbit anti-goat antibody. Signal was detected by using the Vectastain ABC System as described in the manufacturer’s protocol. Sections were coun- terstained with haematoxylin. Ten fields each from the caput epithelium, caput lumen, caudal epithelium and caudal lumen from duplicate slides were randomly selected for measure- ment of the integrated optical density (IOD). The staining intensities were quantified by using the image analysis soft- ware, Aperio ImageScope (Version v11.2.0.780 Aperio, Vista, Calif., USA).

Indirect immunofluorescence

Adult rat testes were fixed by whole body perfusion by using 4 % paraformaldehyde and transferred to a 60 % sucrose solution for 3 days at 4 °C. Tissue blocks were made by using cryo-protectant solution and processed as described by Upadhyay et al. (2011). The distribution of HDAC6 was studied in 8-μm-thick sections that were probed with a 1:100 diluted HDAC6 antibody raised by us. Pre-immune serum at the same dilution was used as a negative control. To study the presence and distribution of HDAC6 and to investigate its status of co-localization with respect to Ac α- tubulin in rat testicular, caput, and caudal epididymal sperm, the sperm were isolated from respective tissues as described earlier, fixed with chilled 95 % ethanol, permeabilized by using 0.1 % Triton X-100 and probed with a 1:10 diluted rabbit polyclonal anti-rat HDAC6 antibody (Thermo Fisher Scientific, Ill., USA.) and a 1:100 dilution of monoclonal Ac α-tubulin antibody. The secondary antibodies, namely FITC- conjugated swine anti-rabbit and rodamine-labelled goat anti- mouse, respectively, were used at a 1:100 dilution. DAPI (4,6- diamidino-2-phenylindole) was used as the counterstain to stain the sperm nucleus. Co-localization was observed by using an LSM 510 Meta Confocal microscope (Carl Zeiss, Oberkochen, Germany). Z stack images were obtained and a three-dimensional (3D) image was constructed, which was analysed by using LSM 510 Meta software. For statistical evaluation, average signal intensities of HDAC6 and Ac α- tubulin expression per micrometer length of the sperm flagella were measured for 10 sperm per group. The overlap coeffi- cient for the two proteins was also determined and cut-mask images showing only the co-localized regions were obtained.

Co-immunoprecipitation

The interaction between HDAC6 and α-tubulin was deter- mined by co-immunoprecipitation of the corresponding interacting proteins by using HDAC6, α-tubulin or Ac α- tubulin as bait. Aliquots containing 200 μg caudal sperm lysates prepared in NP40 lysis buffer were incubated with antibodies to HDAC6 (10 μl polyclonal HDAC6 antibody raised in-house or pre-immune sera), α-tubulin, Ac α-tubulin or mouse IgG (4 μg mouse monoclonal antibody) at 4 °C for 4 h. Protein G beads (30 μl) were then added to all the tubes and further incubated for 2 h. The bound antigen-antibody complexes were separated by centrifugation at 12,000g for 5 min and eluted in 30 μl Laemmli buffer at 95 °C for 10 min. The eluted proteins and the input protein lysates were resolved by SDS-PAGE, transblotted onto nitrocellulose membranes and probed with antibodies to α-tubulin, Ac α-tubulin and HDAC6 as described earlier.

Sperm motility

Sperm from the caudal region of the epididymis were isolated and 5×106 sperm were incubated in 5 % CO2-equilibrated Dulbecco’s modified Eagle’s medium with or without TSA (5 μM), TBSA (5 μM) or NaB (5 μM or 5 mM) at 37 °C for 3 h. At the end of the incubation, sperm viability was evaluated by using 0.5 % eosin in 0.9 % NaCl. Equal volumes of eosin solution and sperm suspension were mixed and observed immediately under a microscope. Sperm heads stained dark pink were counted as being dead. Sperm motility was assessed by using Computer Assisted Sperm Analysis (CASA, Hamilton Thorne, Mass., USA) following which the sperm were fixed in 95 % chilled ethanol to determine the status of Ac α-tubulin by flow cytometry. The co-localization of HDAC6 and Ac α-tubulin in sperm in the presence of the various inhibitors was determined by confocal microscopy.

Flow cytometry

Flow cytometry was performed on ethanol-fixed sperm. Sperm were permeabilized with 1 % Triton X-100 for 10 min, washed and then incubated with a 1:100 dilution of antibody to Ac α-tubulin at 37 °C for 1 h followed by three washes by centrifugation in sheath fluid (Becton Dickinson Biosciences, San Jose, Calif., USA) at 800g for 5 min at room temperature. The cells were then incubated with a 1:100 dilu- tion of rabbit anti-mouse FITC for 1 h in the dark followed by three washes of 5 min each with sheath fluid and analysed on a flow cytometer (Becton Dickinson FAC Sort Flow cytometer, San Jose, Calif., USA). A total of 10,000 cells per sample were recorded. Cells were gated so as to avoid debris and select only the intact sperm for analysis (Malkov et al. 1998). The median intensities for each of these gated cells were noted. Results were analysed by using CellQuestPro (Version 4.0).

HDAC6 enzyme assay

HDAC6 enzyme assay was performed by using the Fluor-de- Lys HDAC6 fluorimetric drug discovery kit (Enzo Life Sciences, Farmingdale, N.Y., USA) designed essentially to evaluate drugs for the inhibition of the deacetylase activity of recombinant human HDAC6 provided in the kit. We used this kit to determine the deacetylase activity in sperm lysate. Additionally, in order to demonstrate whether the activity, if observed, was attributable to HDAC6 or any other HDAC, we used various HDAC inhibitors. Briefly, 50 μg sperm lysates used as a source of HDAC6 were incubated with 10 μM Fluor de Lys SIRT1 substrate in the absence or presence of 5 μM HDAC inhibitors TSA and TBSA and of two concentrations (5 μM and 5 mM) of NaB at 37 °C for 2.5 h. At the end of incubation, 50 μl 1× Fluor-de-Lys developer was added and the assay was quantified fluorimetrically at excitation and emission wavelengths of 360 nm and 460 nm, respectively. Fluor de Lys deacetylated standard and Fluor de Lys substrate incubated with recombinant human HDAC6 were always included in the assay as positive controls together with the test samples. The experiment was performed three times.

Statistical analysis

All the experiments were performed ≥3 times and the signif- icance of the differences between the groups was determined by one way analysis of variance (ANOVA) with Bonferroni’s post-test correction. The level of significance was set at P≤ 0.05. Analyses were performed by using Graphpad Prism software (Version 5.0).

Results

HDAC6 transcript and protein is present in sperm

For HDAC6 to have a role in sperm motility, it is imperative that it be present in the sperm. In order to avoid any ambiguity with respect to the homogeneity of testicular sperm popula- tion, the purity of the testicular sperm population was ensured as described earlier (Fig. 1a). RT-PCR performed to detect the presence of the transcript in rat testicular and caudal sperm showed a 209-bp amplicon in testicular and caudal sperm and in the respective tissue RNA (Fig. 1b). The absence of a band in the no-reverse-transcriptase and reagent control indicated that the observed bands specifically demonstrated the pres- ence of the HDAC6 transcript. β-Actin used as housekeeping gene showed a band of 200 bp in testicular and caudal sperm. The presence of HDAC6 protein in sperm was demonstrated by Western blot analysis. A prominent band (~130 kDa) for HDAC6 was observed to be present in testicular and caudal sperm but not in the negative controls for the same samples. A weak but specific band (not seen in the negative control) was also observed at ~70 kDa. This, we presumed, represented fragmented HDAC6 (Fig. 1c, supplementary material in Fig. S1c). However, non-specific reactivity at ~100 kDa was observed for testicular sperm, but not for caudal sperm. The same band was also seen in the negative control for testicular sperm. Expression of HDAC6 in caudal sperm was clearly lower than that in testicular sperm. The band for Ac α-tubulin showed an apparent increase in caudal sperm in comparison with testicular sperm. Therefore, the band intensities for Ac α- tubulin in testicular and caudal sperm lysates were quantified and normalized to those of its corresponding α-tubulin ex- pression. However, after normalization, this increase was found not to be statistically significant (Fig. 1d, e).

Distribution of HDAC6 in adult rat testis and epididymis

Immunohistochemical analysis of adult rat testicular sections demonstrated specific localization of HDAC6 in round and elongating spermatids (Fig. 2b, c). Indirect immunofluores- cence to further verify the distribution pattern of HDAC6 in adult rat testis, showed the presence of HDAC6 in round,α-tubulin shows a significant decrease in HDAC6 expression in flagellar region of caput sperm (Cp sp) and caudal sperm (Cd sp) as compared with testicular sperm (T sp) and a significant increase in Ac α-tubulin in caput and caudal sperm as compared with testicular sperm (p). The graphical representation of Manders overlap coefficient shows a significant in- crease in the degree of overlap of HDAC6 and acetyl α-tubulin in caudal sperm as compared with testicular sperm and caput sperm (q). The experiment was performed twice on duplicate slides. Ten of each sperm were analysed for quantification of intensities. Statistical significance was determined by using a one way analysis of variance (ANOVA) with significance level being set at P≤0.05. *, **, ***Significance of P≤ 0.05, P≤0.01, and P≤0.001, respectively. Values are mean±S.E.M.

In order to trace the pattern of HDAC6 expression during the functional maturation of sperm, sagittal sections of the epididymis encompassing the caput, corpus and cauda regions were obtained on single slides ensuring uniform processing for immunohistochemical localization. Although we per- formed immunohistochemistry on all the sagittal sections of the epididymis, quantification of the staining intensity was performed only for the caput and caudal regions of epididymis as the corpus region is too narrow to be able to discern between the epithelium and lumen. In the epididymis of the adult rat, HDAC6 expression was observed mainly in the lumen of the caput (Fig. 2h, i) and caudal (Fig. 2j, k) region indicating its presence in caput and caudal sperm. The specific localization of HDAC6 was also observed in the cytoplasm of epithelial cells of the caput and caudal region of the epididy- mis (Fig. 2i, k). The staining intensity in caudal epithelial cells was significantly higher than that in the caput epithelial cells (P ≤0.001). Although an observable decrease appeared to occur in the staining intensity in the luminal region of the cauda as compared with the caput, this was not statistically significant (Fig. 2l).

Co-localization of HDAC6 and Ac α-tubulin in rat sperm

The possible overlapping distribution of the two proteins HDAC6 and Ac α-tubulin was investigated in rat testicu- lar and caudal epididymal sperm. Whereas HDAC6 ex- pression was observed mainly in the mid-piece region (Fig. 3a), Ac α-tubulin expression was observed mainly towards the end-piece region of the flagella in testicular sperm (Fig. 3b). In caput and caudal sperm, the expres- sion of HDAC6 was seen in the mid-piece and principal piece region of sperm flagella (Fig. 3f, k), whereas Ac α- tubulin was localized throughout the flagella (mainly in the mid- and end-piece of sperm flagella; Fig. 3g, l). Although the two proteins were co-localized in testicular, caput and caudal sperm, the expression of HDAC6 and Ac α-tubulin was not uniform along the length of the flagellum (Fig. 3d, l, n). The co-localization of HDAC6 and Ac α-tubulin can be best appreciated in the cut-mask images of the testicular, caput and caudal sperm (Fig. 3c, h, m). HDAC6 and Ac α-tubulin were co-localized main- ly in the mid-piece region of sperm flagella from testicu- lar to caudal sperm (Fig. 3c, d, h, i, m, n). A significant decrease (P ≤ 0.01) was observed in the signal intensity of HDAC6 in caput and caudal sperm flagella as compared with that in testicular sperm flagella, whereas the signal intensity of Ac α-tubulin was significantly increased in caput (P ≤ 0.05) and caudal (P ≤ 0.001) sperm flagella with respect to testicular sperm flagella (Fig. 3p). The overlap coefficient between HDAC6 and Ac α-tubulin showed a significant increase (P ≤ 0.05) in caudal sperm flagella with respect to caput and testicular sperm flagella (Fig. 3q).

Co-immunoprecipitation of HDAC6, α-tubulin and Ac α-tubulin in sperm

HDAC6 and Ac α-tubulin are observed to be co-localized on the flagella of testicular and caudal sperm. To ascertain that HDAC6 a nd α-tubulin indeed interacted, c o- immunoprecipitation studies were carried out by using caudal sperm. HDAC6, α-tubulin, Ac α-tubulin and their respective interacting proteins were immunoprecipitated by using their respective antibodies and probed for the presence of α-tubulin, Ac α-tubulin and HDAC6 in the eluted interactome. Bands for α-tubulin and Ac α-tubulin at ~55 kDa and for HDAC6 at ~130 kDa were identified in caudal sperm immunoprecipitates for all three proteins. The eluted interac- tome of pre-immune sera/mouse IgG used to account for any non-specificity did not show these bands (Fig. 4a, b).

Impact of HDAC6 in α-tubulin acetylation and sperm movement

HDAC inhibitors were used to determine whether the enzyme was catalytically active in sperm and whether deacetylation had any influence on sperm motility. Deacetylase activity was measured by using the Fluor- de-Lys HDAC6 fluorimetric drug discovery kit. This kit allowed the determination of the general deacetylase ac- tivity of sperm lysates by using Fluor de Lys SIRT1 as substrate. Whether this activity was attributable to HDAC6 or any other HDAC present or to a combination of both in the sperm lysate was discerned by using the three inhibitors; a general HDAC inhibitor (TSA at 5 μM), an HDAC6 specific inhibitor (TBSA at 5 μM) and an inhibitor to which HDAC6 is resistant (NaB at 5 μM and 5 mM). Sperm lysates used as the source of the enzyme were incubated with 10 μM Fluor de Lys SIRT1 in the absence or presence of the inhibitors at 37 °C for 2.5 h. The deacetylase activity was significantly inhibited with 5 μM TSA and TBSA and 5 mM NaB as compared with the control and with 5 μM NaB (Fig. 5a).

Rat caudal sperm were incubated with 5 μM TSA, TBSA (5 μM) or NaB (5 μM or 5 mM) and effect of these inhibitors on α-tubulin acetylation and on sperm motility parameters were investigated. Sperm viability was verified and was approximately 80 % in all groups. Acetylation of α-tubulin in these sperm was determined by flow cytometry analysis. α-Tubulin acetylation was significantly increased in sperm treated with TBSA as compared with the control and 5 mM NaB (Fig. 5b), i.e. the acetylation of α-tubulin was significantly lower with 5 mM NaB compared to that of TBSA. The signal inten- sities in sperm treated with TSA and NaB were compara- ble with that of the control. Progressive motility was significantly reduced in sperm treated with TBSA as compared with the control, TSA and both concentrations of NaB (Fig. 5c). Beat frequency in contrast was signifi- cantly higher in TBSA treated sperm as compared with the control and TSA (Fig. 5d). Other motility kinetics parameters, namely path velocity, progressive velocity, track speed, lateral amplitude, straightness and linearity, were not affected by any of the inhibitors (Fig. 5e, f). Studies to determine the status of HDAC6 and Ac α- tubulin localization in sperm treated with the above- mentioned inhibitors showed an apparent increase in the expression of Ac α-tubulin in the sperm incubated with TSA and TBSA (Fig. 6f, j, respectively). However, an apparent increase in HDAC6 localization on the flagella was also noted in sperm treated with the HDAC6-specific inhibitor TBSA (Fig. 6i). Relatively weak staining for HDAC6 on the flagella was seen in the TSA- and NaB- treated sperm (Fig. 6e, m, respectively).

Discussion

Reversible acetylation of α-tubulin has been implicated in the regulation of microtubule stability and function (LeDizet and Piperno 1987). HDAC6 has been identified as an α-tubulin deacetylase and its overexpression leads to the deacetylation of α-tubulin and promotes chemotactic cell movement supporting the idea that HDAC6-mediated deacetylation reg- ulates microtubule-dependent cell motility (Hubbert et al. 2002). Analogous to this, the deacetylation of α-tubulin in the microtubules of sperm flagella might regulate sperm mo- tility; this is the hypothesis tested in this study. The available literature demonstrates the presence of HDAC6 in testicular tissue (Hazzouri et al. 2000; Seigneurin-Berny et al. 2001). We have now shown the presence of the HDAC6 transcript and protein in rat testicular and caudal sperm (Fig. 1b, c). This is a novel observation as the presence of HDAC6 has not been previously reported on sperm. We have further observed that, as sperm mature, the acetylation of α-tubulin increases. HDAC6 expression, although reduced, persists in mature sperm (Fig. 3a–p).

Investigating the ontogenic expression of HDAC6 in rat testis, we have observed the presence of its transcript and protein right from birth until adulthood (data not shown). In adult testis, it is present in round, elongating and elongated spermatids. Its localization in adult epidid- ymis in the lumen of the caput and caudal region indicates its presence on epididymal sperm. A significant increase has been observed in the localization of HDAC6 in caudal epithelial cells as compared with that in caput epithelial cells (Fig. 2). As a part of the ubiquitinylation complex, HDAC is involved in the cellular management of misfolded proteins with the help of its ubiquitin-binding domain (Kawaguchi et al. 2003). The increased HDAC6 expression in the caudal epithelial cells might reflect its role in the degradation of misfolded or unfolded proteins derived from defective sperm phagocytosed by the epi- didymal epithelial cells (Sutovsky et al. 2001).

HDAC6 and Ac α-tubulin have been observed to be increasingly co-localized in the flagella of testicular and caudal sperm. Notably, the overlap coefficient is signifi- cantly higher in the caudal sperm in comparison with that in testicular and caput sperm (Fig. 3). The significant co-localization of HDAC6 and Ac α-tubulin in caudal sperm and the co-elution of α-tubulin, Ac α-tubulin and HDAC6 as seen from the pull-down experiments (Fig. 4) demon- strate that the two proteins interact. These data suggest that dynamic deacetylation and acetylation of α-tubulin occurs in mature sperm. The interaction of HDAC6 with α- and β-tubulin has been reported previously in NIH 3 T3 cells (Zhang et al. 2003).

To determine the specificity and relevance of the HDAC6/α-tubulin interaction, we incubated caudal sperm with inhibitors TSA, which inhibits all HDACs (Yoshida et al. 1990), TBSA, which specifically inhibits HDAC6 (Butler et al. 2010), or NaB to which HDAC6 is resistant (Candido et al. 1978; Kruh 1982), and assessed the effect of these inhibitors on sperm motility and α-tubulin acety- lation. In order to determine HDAC6 activity, sperm ly- sates were incubated with the inhibitors as described earli- er. At equimolar concentrations, HDAC6 activity was sig- nificantly inhibited in the presence of TBSA and TSA, as compared with the control and NaB (5 μM). When NaB was used at a concentration of 5 mM, the deacetylase activity was inhibited significantly compared with the con- trol and 5 μM NaB (Fig. 5a). Notably, with respect to the deacetylase activity assay, the kit used allows the determi- nation of the total deacetylase activity of sperm lysates by using Fluor de Lys SIRT1 as the substrate (SIRT1 is a substrate for most Class 1 and Class 2b HDACs). Whether this activity is attributable to HDAC6 or any other HDAC present or a combination of both in the sperm lysate was discerned by using the three inhibitors, namely TSA, TBSA and NaB. As this activity was measured by using acetylated SIRT1 peptide (Fluor de Lys SIRT1) as a sub- strate, we saw an enhanced reduction of activity with TSA compared with TBSA suggesting the presence of other HDACs. However, this was not statistically significant. With NaB (5 μM), we see HDAC6 activity that is compa- rable with that in the control. Interestingly, with NaB at a concentration of 5 mM, deacetylase activity was signifi- cantly inhibited compared with the control and 5 μM NaB, further substantiating the contribution of other HDACs, most likely HDAC1, which is a known histone-specific deacetylase and which has earlier been reported in the testis during spermatogenesis (Hazzouri et al. 2000). Whereas the inhibition of deacetylase activity with 5 μM TSA and 5 mM NaB suggests the presence of other HDACs in sperm, the inhibition of deacetylase activity with 5 μM TBSA certainly demonstrates the presence of catalytically active HDAC6 in sperm. Sperm progressive motility was significantly inhibited by TBSA with respect to the control, TSA and NaB (5 μM and 5 mM). Of note, motility in presence of NaB was comparable with that in the control and significantly increased compared with that in TBSA. The finding that HDAC6 is resistant to inhibition by NaB only consolidates our observations with TBSA (Fig. 5c). A significant increase in the signal intensity for Ac α-tubulin as determined by flow cytometry and as also apparent by IIF has been observed in sperm treated with TBSA (Figs. 5b,d 6j). An anticipated decrease in motility and increase in acetylation of α-tubulin was not observed with TSA at the concentration used, although, at this dose, deacetylase activity was inhibited. As a general HDAC inhibitor, TSA would probably be needed at much higher concentrations to inhibit HDAC6 specifically (Butler et al. 2010). Additionally, whereas activity was studied by using sperm lysate, an effect on motility and Ac α-tubulin ex- pression was studied by using intact sperm. This means that, in the lysate, the drugs had direct access to HDAC6/ other HDACs, whereas in the intact sperm, the drugs had to penetrate the membrane to access its site of action. Notably, the acetylation of the α-tubulin was significantly lower with 5mM NaB compared with that with TBSA. Given that HDAC6 is resistant to inhibition by NaB, this observation in conjunction with the significantly increased acetylation seen with the HDAC6-specific inhibitor TBSA provides evidence that HDAC6 regulates a tubulin acety- lation in sperm. At equimolar concentrations, an effect on progressive motility, beat frequency and increase in a tu- bulin acetylation is observed to be significant only with the HDAC6-specific inhibitor TBSA. Neither with TSA nor NaB was motility affected, even at the 5mM concentration of NaB, thereby providing evidence suggesting that HDAC6 deacetylates α-tubulin in sperm and is involved in modulating sperm movement. Intriguingly, an increased intensity for HDAC6 localization on the sperm flagella was seen in the TBSA treated sperm (Fig. 6i). This suggests that, although the HDAC6-specific inhibitor reduces HDAC6 activity, as can be seen from Fig 5a, it prevents dissociation of HDAC6 from the microtubules, as a con- sequence of which HDAC6 persists on the flagella. The relatively weak expression of HDAC6 in the untreated (Control), TSA and NaB-treated sperm (Fig 6a, e, m) indicates that HDAC6 is able to dissociate from the micro- tubules in the presence of these inhibitors. On inhibition by TBSA, HDAC6 possibly undergoes a conformational change that prevents its dissociation from its binding site on the microtubule, thus physically blocking the site. This creates stearic hindrance for other MAPs to bind to the microtubules, thereby interfering with their molecular pro- cesses, with a consequent affect on sperm movement. Our present study shows that, with regard to the pharmacolog- ical inhibition of HDAC6 activity in sperm, α-tubulin acetylation increases and sperm motility decreases. Work by Zilberman et al. (2009) in B16F1 cells has demonstrated that, whereas HDAC6 knockdown does not affect micro- tubule dynamics, HDAC6 with impaired enzymatic activ- ity can influence microtubule stability. A similar study in MCF-7 cells has shown that the inhibition of HDAC6 deacetylase activity increases its binding with microtu- bules leading to increased acetylation of α-tubulin and increased stability of the microtubules (Asthana et al. 2013). Observations from α-tubulin acetyltransferase 1 (Atat1) knockout mice indicate that, although these mice are viable and develop normally, they exhibit significantly reduced sperm motility and fertility. Thus, although acety- lation of α-tubulin is absent, microtubule stability in- creases (Kalebic et al. 2013b). The status of HDAC6 in these mice is not known. Our data from individuals with poor sperm motility show significantly reduced acetylation of α-tubulin in the sperm of these individuals (Bhagwat et al. 2014). In both cases, acetylation and sperm motility are reduced. The status of HDAC6 in these individuals is being investigated.

Taken together, these data suggest that the acetylation status might not be the determinant of sperm motility. We propose that, instead, the stability of the microtubules defines sperm motility; the more stable the microtubules are, the lower the flagellar motility is. The persistent association of HDAC6 with the flagella, even in the presence of HDAC6 inhibitor as seen in our study, and the increased binding of HDAC6 with microtu- bules in the presence of the inhibitor and the consequent in- crease in stability of these microtubules reported in the B16F1 cells (Zilberman et al. 2009) and MCF7 cells (Asthana et al. 2013) suggests that HDAC6 functions as a MAP and plays an important role in maintaining the dynamic instability in the sperm flagellar microtubules. Recently, MAP7 domain- containing protein 3 (Mdp3) has been shown to regulate HDAC6 activity and control microtubule stability through its binding to tubulin and microtubules (Tala et al. 2014). We propose that dynamic instability exists in sperm and is essential for normal sperm motility; investigations on this aspect are ongoing.

In summary, we have demonstrated that (1) an active HDAC6 enzyme is present in sperm, (2) HDAC6 in sperm is able to deacetylate α-tubulin, (3) the inhibition of HDAC6 activity results in increased α-tubulin acetylation and (4) HDAC6 inhibition affects sperm motility. These pieces of evidence suggest that HDAC6 is the α-tubulin-specific deacetylase in sperm and is involved in modulating sperm movement.