Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A novel posttranslational modification of histone, H3 S-sulfhydration, is down-regulated in asthenozoospermic sperm

A novel posttranslational modification of histone, H3 S-sulfhydration, is down-regulated in... Oxidative stress is one of the major causes leading to male infertility including asthenozoospermia. Hydrogen sulfide (H S) has been widely recognized to be a potent antioxidant whose role is partially implemented by protein S-sulfhydration. How- ever, protein S-sulfhydration has not been reported in germ cells. Therefore, we investigated whether asthenozoospermia could be associated with sperm protein S-sulfhydration. S-sulfhydrated proteins in human sperm were enriched via biotin- switch assay and analyzed using LC-MS/MS spectrometry. Two hundred forty-four S-sulfhydrated proteins were identified. Importantly, we validated that sperm histones H3.1 and H3.3 were the S-sulfhydrated proteins. Their S-sulfhydrated amino acid residue was Cysteine111. Abundances of S-sulfhydrated H3 (sH3) and S-sulfhydrated H3.3 (sH3.3) were significantly down-regulated in asthenozoospermic sperm, compared with the fertile controls, and were significantly correlated with pro - gressive motility. Retinoic acid (RA) up-regulated level of sH3.3 in primary round spermatids and the C18-4 cells (a mouse spermatogonial stem cell line). Overexpression of the mutant H3.3 (Cysteine111 was replaced with serine) affected expression of 759 genes and raised growth rate of C18-4 cells. For the first time, S-sulfhydration H3 and H3.3 were demonstrated in the present study. Our results highlight that aberrant S-sulfhydration of H3 is a new pathophysiological basis in male infertility. Keywords S-sulfhydrated proteome · H3S-sulfhydration · Asthenozoospermia · Spermatogenesis · H3.3 Abbreviations ACRBP Acrosin binding protein ANKRD1 Ankyrin repeat domain 1 ASTH Asthenozoospermic Mr. Wang and Mr. Li both are the responding authors of this paper. ATP5A ATP synthase F1 subunit alpha Biotin-HPDP N-(6-(biotinamido) * Runsheng Li hexyl)-3′-(2′-pyridyldithio)-propionamide runshengli2007@163.com CABLES1 Cdk5 and Abl enzyme substrate 1 Jian Wang CCND Cyclin D wangjiansippr@126.com CENP-A Centromere protein A NHC Key Laboratory of Reproduction Regulation dpp Days postpartum (Shanghai Institute of Planned Parenthood Research), DEG Differentiall y expressed genes Pharmacy School, Fudan University, 2140 Xietu Road, DVL3 Dishevelled segment polarity protein 3 Shanghai 200032, China eST elongating/condensed spermatids NHC Key Laboratory of Reproduction Regulation (Shanghai GAPDH Glyceraldehyde-3-phosphate Institute of Planned Parenthood Research), 2140 Xietu Road, dehydrogenase Shanghai 200032, China GDNF Glial cell–derived neurotrophic factor State Key Laboratory of Genetic Engineering, Institute GO Gene ontology of Biostatistics and Computational Biology, School of Life Sciences, Fudan University, Shanghai 200438, China H O Peroxide 2 2 HS Hydr ogen sulfide NHC Key Laboratory of Reproduction Regulation (Shanghai Institute of Planned Parenthood Research), School H3 Histone3 of Life Sciences, Fudan University, 2140 Xietu Road, HirA1 Hira domain-containing protein Shanghai 200032, China Vol.:(0123456789) 1 3 3176 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 IAA Iodoace tamide motile (PR%) spermatozoa < 32% [3]. The molecular basis IAP Iodoacetyl-polyethylene glycol of asthenozoospermia is largely elusive. IGF-1R Insulin-lik e growth factor 1 receptor Human spermatozoa are extremely vulnerable to oxida- IgG Immunoglobulin G tive attack because they contain little cytoplasm sequestering IP Immunopr ecipitation antioxidants. The oxidative stress (OS)–mediated damage to LC-MS/MS Liquid chromatography-tandem mass sperm has been considered as one of the leading causes for spectrometry male infertility [4]. Defective mouse spermatogenesis can LDHA Lactate dehydrogenase A be caused by knockout of antioxidative genes, leading to an MMTS Methyl-methanethiosulfonate excessive production of reactive oxygen species (ROS) [5]. NORM Normozoospermic Asthenozoospermic sperm expressed some down-regulated ODF1/2 Outer dense fiber protein 1/2 antioxidative genes [6]. OS is accepted as the target for clini- OS Oxidativ e stress cal treatment of asthenozoospermia [7]. PTM Posttranslational modifications Comparative proteomic analysis widely revealed altered RA R etinoic acid expression of some proteins in asthenozoospermic sperm rST Round spermatids [8–10], indicating that sperm with poor motility might be RNA-seq RNA sequencing caused by the lower expression of tubulin with structural ROS Reactive oxygen species defects in sperm flagellum. An altered expression of his- SDS-PAGE SDS–polyacrylamide gel electrophoresis tone was detected in asthenozoospermic sperm by proteomic sH3.3 S-sulfhydrated H3.3 study [11]. Additionally, the aberrant expression of PTMs of sH3 S-sulfhydrated H3 proteins including phosphorylation [12], sumoylation [13], SSC Spermatogonial stem cells glutarylation [14], and hydroxyisobutyrylation [15] was Wisp1 WNT1 inducible signaling pathway pro- associated with poor sperm motility. Interestingly, charac- tein 1 terization of human sperm lysine acetylproteome revealed that protein acetylation was essential for sperm motility [16]. Study of the global protein phosphorylation land- Introduction scape of spermiogenesis showed wide phosphoregulation across a diverse range of processes during spermiogenesis Mammalian spermatogenesis, a precisely regulated [17]. However, these proteomic studies did not show any developmental process generating sperm, consists of specific protein whose function was actually regulated by three distinct phases. The first phase refers to the mitotic these PTMs in germ cells. Additionally, the links between division of spermatogonia physiologically resulting in ROS and the PTMs of proteins have not been established in accumulation of germ cells depending on renewal and asthenozoospermic sperm. differentiation of spermatogonial stem cells (SSC). The Hydrogen sulfide (H S) exerted a wide range of physi- second stage is meiosis, in which spermatocytes undergo ological and cytoprotective functions in the biological sys- two rounds of mitosis to produce haploid spermatids. The tems via its potent antioxidative capability [18]. The asthe- final one is spermiogenesis, wherein the round spermatids nozoospermic patients exhibited decreased concentration of (rST), known as haploid spermatids, undergo a complex H S in their seminal plasma, while supplying exogenous H S 2 2 differentiation process to develop into spermatozoa, to semen improved sperm motility of the asthenozoospermic including chromatin remodeling, nuclear elongation, and patients [19]. However, the mechanism remains unknown flagellum development. A hallmark of mammalian sperm is regarding how H S exerts its roles in germ cells. Signal- the highly compact and condensed structure of chromatin, ing by H S has been widely found in eukaryotic cells via in which depending on the species, approximately 90–99 protein S-sulfhydration [20–22], a PTM on thiol group of % of histones are replaced by protamines [1]. Distinct cysteine residues that converts Cys-SH to Cys-SSH. About posttranslational modifications (PTMs) of histones in 10–25% of proteins extracted from liver are S-sulfhydrated spermatogenesis are currently accepted to facilitate the in physiological conditions [20]. An accumulating number chromatin remodeling and histone-to-protamine transition of proteins have been recently validated to be S-sulfhydrated [1, 2]. Impaired spermatogenesis causes male infertility. proteins with the identified sulfhydrated cysteine residues, Around 15% of couples at reproductive age present with and their sulfhydration has wide and important functions infertility, and about half of the infertility are associated including regulating redox balance [21, 23]. However, pro- with male partner. Asthenozoospermia, a common male tein S-sulfhydration has not been reported in germ cells. We infertility, is characterized by both reduced sperm motility revealed S-sulfhydrated proteome consisting of 244 proteins and normal concentrations of sperm (>15 million per of human sperm in the present study. They included most matozoa/ml), and is defined as percentage of progressively of ROS-associated human sperm reported elsewhere [24]. 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3177 Importantly, for the first time, we demonstrated that histone Direct swim‑ up of spermatozoa H3 was an S-sulfhydrated protein in the present study. We further observed that the level of sH3 and sH3.3, a H3 vari- The direct swim-up of spermatozoa from semen was per- ant, was positively correlated with percentage of sperm with formed according to the “WHO Laboratory Manual” (5th progressive motility, respectively. Our findings highlighted a edition). Briefly, place 2 ml of liquefied semen in a sterile novel pathophysiological basis for asthenozoospermia. 15-ml conical centrifuge tube, and gently layer 2 ml of mod- ified HTF medium (#ART-1023, SAGE) with 5% human serum albumin solution (#10064, Vitrolife) over it. After Experimental procedures sperm were incubated at 37 °C with 5% C O in humidified air for 1 h, gently collected the uppermost 1 ml of medium Reagents, cells, and mice containing highly motile sperm cells and 1 ml semen at bot- tom of the tube, and used them as sperm with high motility Anti-H3.3antibody(#ab176840), anti-H3 antibody(#ab1971), and low motility, respectively. Centrifuged at 500g for 5 min anti-biotin antibody(#ab1227), anti-H3K4me3(#ab185637), and discarded the supernatant. The pellets were used for and anti-H3K9me3(#ab8898) were purchased from Abcam preparation of lysates. (USA). GDNF was purchased from Peprotech (#450-44, USA). Male C57BL/6J mice were purchased from SIPPR- Biotin‑ switch assay BK Animal Company (Shanghai, China). The C18-4 cells was kindly provided by Prof. Zuping S-sulfhydrated proteins in human spermatozoa were detected He, who was a principal investigator in Renji-Med X Clini- using biotin-switch assay as described previously with minor cal Stem Cell Research Center, Renji Hospital, School of modifications [ 20]. Briefly, 20 million spermatozoa were Medicine, Shanghai Jiao Tong University. The C18-4 cells centrifuged at 2000 g for 5 min and supernatant removed, were grown as described by He et al. [25]. and spermatozoa were then resuspended in the lysis buffer (250 mM HEPES pH 7.7, 1 mM EDTA, 0.1 mM neocu- proine, 1% Triton, 2.5% SDS) added with a cocktail of pro- Semen sample collection tease inhibitors (Sigma). The lysates were incubated for 5 min at room temperature and then centrifuged at 2000 g This study (PJ2019-05) was approved by the Ethics Commit- for 5 min. The supernatant was collected, and its protein tee of Shanghai Institute of Planned Parenthood Research/ concentration was adjusted to less than 0.5 mg/ml in each World Health Organization (WHO) Collaborating Center sample. Proteins were then precipitated using 4 volumes of on Human Research. Written informed constructs were ice-cold acetone for 20 min at −20°C, centrifuged at 2000 obtained from the semen donors involved in the study. All g for 5min at 4 °C, washed twice with 70% acetone, and the methods used in the present study were performed in dried out. The pellets were resuspended in HEN medium accordance with the Declaration of Helsinki. The donors (250 mM HEPES pH 7.7, 1 mM EDTA, 0.1 mM neocu- were recruited in compliance with the “WHO Laboratory proine) containing 2.5% SDS. Free thiols of proteins were Manual for the Examination and Processing of Human blocked with a rapidly thiol-reactive agent MMTS (20 mM) Semen” (Fifth edition). Semen samples were obtained by (#23011, Thermo Scientific, CHE) for 30 min at 50 °C. After masturbation after 3–5 days of sexual abstinence. Semen the reaction, the proteins were precipitated with acetone as samples which contained leukocytes were excluded from our described above in order to remove excess MMTS, resus- study. Spermatozoa motility was assessed by the computer- pended in HEN medium containing 1% SDS. Next, the pro- assisted sperm assay (CASA) method according to World teins solution was added with 1mM biotin-HPDP (#A8008, Health Organization guidelines, equipped with a camera APExBIO, USA) and incubated for 1 h at 25 °C to achieve (acA780-75gc, Basler, Germany), and a 20-fold objective, a biotinylation. The biotinylated proteins were separated by camera adaptor (Eclipse E200, Nicon, Japan), operated by an SDS-PAGE and finally detected with anti-biotin antibody or SCA sperm class analyzer (MICROPTIC S.L.). Finally, the the indicated antibodies in Western blotting analysis. semen samples were collected from 26 normozoospermic men (24–45 years old, mean ± SEM: 31.54 ± 5.04 years old) Cysteinyl labeling assay and 24 asthenozoospermic patients (24–39 years old, mean ± SEM: 32.25 ± 4.78 years old) (Table S3), and used in the We also detected S-sulfhydrated proteins in human sperma- present study. The normozoospermic men had known repro- tozoa using cysteinyl labeling assay as described elsewhere ductive histories in the past 2 years and progressive motility [26] with minor modifications. Briefly, 20 million sper - ≥32%, while the asthenozoospermic men had a progressive matozoa were lysated in lysis buffer. The lysate was added motility <32% (Table S3). with 2mM IAA for 1 h at room temperature. Cold acetone 1 3 3178 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 of double volume was then added into the sample. Then, Identification of the sulfhydrated amino acid the samples were precipitated at −20 °C for 20 min. After residue in H3.1 and H3.3 centrifuged at 12000 rpm at 4 °C for 10 min, the precipita- tion was diluted in HEN buffer with 1 mM DTT at room All the constructs that expressed wild-type and mutant temperature for 30 min. A total of 3 mM biotinylated IAP H3.3 and H3 were purchased from Genomeditech (Shang- was next added into the sample at room temperature for 1 h. hai, China). H3.1 was mutated in the three ways: C97 was Biotinylated proteins were enriched by using streptavidin- replaced with serine (C97S), C111 was replaced by ser- Sepharose beads for 16 h at 4 °C on a rotating wheel, with ine (C111S), and both of C111 and C97 were mutated sequential rounds of centrifugation (12,000 g, 1 min, 4 °C) to serine (double mutations). H3.3 was mutated at C111, using PBS to wash the beads. The beads were resuspended which was also replaced with serine. All the expression in 20 μl of 4×Laemmli sample buffer and heated at 90 °C constructs were generated based on pCMV2-FLAG tag for 1 min. The biotinylated proteins were separated by SDS- (Promega, USA). PAGE and finally detected with anti-biotin antibody or the Transfection of the C18-4 cells with the H3 expression indicated antibodies in Western blotting analysis. constructs via Lipofectamine 3000 (Invitrogen, Shanghai, China) was carried out according to the manufacturer’s pro- Analysis of immunoprecipitation tocol. Forty-eight hours later, the cells were harvested for preparation of lysate using the above lysis buffer. Wild-type Proteins from sperm or the C18-4 cells were biotinylated as and mutant H3.1 or H3.3 were immunoprecipitated using described above. Protein concentration was adjusted to 0.1 anti-FLAG antibody, and next subjected to biotin-switch mg/ml using HEN/10 media (10× dilution of HEN) contain- assay. The biotinylated H3 was enriched via the IP using ing 1% SDS. Three volumes of neutralization buffer (20 mM anti-biotin antibody, and finally detected using anti-FLAG HEPES pH 7.7, 100 mM NaCl, 1 mM EDTA and 0.5% Tri- antibody in Western blotting assay as described above. ton X-100) were added. The mixture was separated into two parts. One part was added 2× SDS sample buffer for load- ing control, while another was incubated overnight at 4 °C Primary germ cell preparation and treatment of rST with a specific antibody (1:1000) and 50 μl of protein A/G with RA (#ab193262, Abcam, USA) per ml. Beads were previously washed twice with the neutralization buffer and centrifuged SG cells were isolated from 8 days postpartum (dpp) mice at 200 g for 10 s. Once the incubation terminated, the beads [27]. The method of STA-PUT was used to isolate pacSC, were washed 5 times with 500 μl of Wash buffer (the neu- rST, and eST. They were characterized as previously tralization buffer containing 600 mM NaCl). Proteins were described [27, 28]. pacSC were from 17 dpp mice. rST and eluted with 1× SDS sample buffer containing 2 mM DTT. eST were from 56–70 dpp mice. After separated via grav- Samples were boiled for 5 min at 100 °C and centrifuged at ity sedimentation, rST were pelleted via a centrifugation at 14000 g for 5 min. The supernatant was collected, separated 500 g for 5 min, cultured in DMEM (10% PBS) medium by SDS-PAGE (10%), and detected via Western blotting with 2 mM L-glutamine, 100 units/ml penicillin, and 100 analysis or silver staining. The gel with silver staining was mg/ml streptomycin at 37 °C with 5% CO in humidified excised, and applied for proteomic analyses which were per- air. Three hours later, RA (#r2625, Sigma-Aldrich, USA) formed as described below. For each of these experiments, 3 diluted in ethanol was added to the culture medium to make fertile ejaculates (from different donors) were pooled. a final concentration of 0.3 μM or 1 μM. Twenty-four hours later, the cells were harvested for measurement of sH3.3 Silver staining expression. After proteins were separated by SDS-PAGE, whole gel was washed with water for 5 min, and then soaked into blocking Treatment of testis with NaHS buffer (50% ethanol, 8% acetic acid, 0.4% formaldehyde) for 2 h. After washed with 35% ethanol for 3 times, the gel Mice aged 4 weeks were sacrificed by cervical dislocation was soaked into staining buffer (10 mg/ml silver nitrate, and then put them into 75% ethanol. The abdominopelvic 0.4% formaldehyde) for 30 min. After washed with water cavity was opened using sterile scissors and forceps, and for 2 times, it was then soaked into cultivating buffer (0.12 then the testis was pulled out. The testicular tunica albuginea g/ml sodium carbonate, 0.4% formaldehyde) until bands was removed by puncturing the tissue and the loose seminif- appeared. Finally, the gel was placed into termination buffer erous tubules were collected. The tubules were transferred (50% ethanol, 8% acetic acid) for 5 min. The PAGE was to a new petri dish containing 2 ml of DMEM/F12 with scanned by Tanon scanner 5200. gentamicin (0.02 g/l; Sigma-Aldrich). The seminiferous 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3179 tubules were cut up, dispersed, and evenly distributed into min over 60 min. The gradient was set as the following: 3 dishes in a humidified atmosphere at 34°C with 5% CO . 5–8% buffer B from 0 to 2min, 8 to 23% buffer B from Three hours later, NaHS (50 or 100 mM; Sigma-Aldrich) 2 to 42 min, 23 to 40% buffer B from 42 to 50 min, 40 was added to the medium. After 24-h treatment, the pieces to 100% buffer B from 50 to 52 min, 100% buffer B kept of testis were collected for measurement of sH3.3 expression until 60 min. MS data were acquired using a data-depend- ent top20 method dynamically choosing the most abun- Infection of the C18‑ 4 cells, cell number dant precursor ions from the survey scan (350–1800m/z) determination, qRT‑ PCR, and Western blotting for HCD fragmentation. A lock mass of 445.120025 Da assay was used as internal standard for mass calibration. The full MS scans were acquired at a resolution of 60,000 at The recombinant Lentivirus that expressed the chimeric m/z 200, and 15,000 at m/z 200 for MS/MS scan. The human H3.3 protein (wild-type or the mutant H3.3 with maximum injection time was set to 50 ms for MS and 45 C111S) with a FLAG tag in the N-terminal of H3.3 was ms for MS/MS. Normalized collision energy was 27 and purchased from Kangchen Bio-tech (Shanghai, China). The the isolation window was set to 1.5 Th. Dynamic exclu- ectopic expression of H3.3 and the mutant H3.3 by viral sion duration was 30 s. infection in the C18-4 cells were performed according to the manufacturer’s instructions. Cell number determination, Database search qRT-PCR, and Western blotting assay were carried out as described previously [29]. The MS data were analyzed using MaxQuant software ver- sion 1.5.8.3. MS data were searched against the UniProtKB Mass spectrometry experimental design Human database (157600 total entries, downloaded in July, and statistical rationale 2017). The trypsin was selected as digestion enzyme. The maximal two missed cleavage sites and the mass tolerance Sample preparation of 4.5 ppm for precursor ions and 20 ppm for fragment ions were defined for database search. Carbamidomethyla - Human sperm lysate was prepared from ten pooled sperm tion of cysteines was defined as fixed modification, while samples and separated by SDS-PAGE. Gel pieces were cut, acetylation of protein N-terminal and lysine and oxidation destained for 20 min in 100 mM NH HCO with 30% ace- 4 3 of methionine were set as variable modifications for data - tonitrile, and washed with Milli-Q water until the gels were base searching. The database search results were filtered and fully destained. The spots were then lyophilized in a vacuum exported with <1% false discovery rate (FDR) at peptide centrifuge. The in-gel proteins were reduced with dithio- level and protein level, respectively. threitol (10 mM DTT/100 mM NH HCO ) for 30 min at 56 4 3 ° C, then alkylated with iodoacetamide (200 mM IAA/100 Protein structure modeling mM NH HCO ) in the dark at room temperature for 30 min. 4 3 Gel pieces were briefly rinsed with 100 mM NH HCO3 and The three-dimensional structure of H3.3 in nucleosome ACN, respectively. Gel pieces were digested overnight in was generated using data from human nucleosome struc- 12.5 ng/μl trypsin in 25 mM NH HCO . The peptides were 4 3 ture containing H3.3 (PDB ID: 5X7X) at 2.18 Å generated extracted three times with 60% ACN/0.1% TFA. The extracts by PyMOL-1.5.0.3. were pooled and dried completely by a vacuum centrifuge. LC-MS/MS RNA sequencing and analysis The peptide of each sample was desalted on C18 Car- Total RNA was extracted from mutated and normal mice tridges (Empore™ SPE Cartridges, Sigma), then con- sample by RNeasyPlus Micro Kit (Qiagen, Wetzlar, Ger- centrated by vacuum centrifugation and reconstituted in many) following manufacturer’s instructions and reverse 10 μl of 0.1% (v/v) formic acid. MS experiments were transcribed into cDNA libraries using the Ovation® RNA- performed on a Q ExactiveHF mass spectrometer that was Seq System V2 kit (NuGEN). Samples were sequenced coupled to Easy nLC (Thermo Scientific). Peptide was with paried-ends reads (PE150) using IlluminaHiseq X-ten first loaded onto a trap column (100 μm×20 mm, 5 μm, platform. The QC (quality control) analysis of the RNA C18) with 0.1% formic acid, then separated by an analyti- sequencing data was performed using FastQC. The raw cal column (75 μm×100 mm, 3 μm, C18) with a binary sequencing reads were pre-processed as follows: (1) remov- gradient of buffer A (0.1% formic acid) and buffer B (84% ing adapter sequences, (2) removing reads with over 20 bp acetonitrile and 0.1% formic acid) at a flow rate of 300 nl/ of low quality (Phred quality score < 20). The filtered clean 1 3 3180 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 reads were aligned to mouse reference genome (mm10) dehydrogenase (GAPDH) [31], and L-lactate dehydrogenase using Tophat2 and then the uniquely mapped reads were A chain (LDHA) [34] were elsewhere reported. In addition, assigned to each annotated gene using featureCount. Sta- we observed S-sulfhydration of Outer dense fiber protein 1 tistical significant test of differentially expressed genes was (ODF1) and ODF2, two proteins that stabilize the axoneme performed by NOISeq with R [30]. Genes with absolute to maintain sperm motility [35]. Interestingly, we detected log2-transformed fold changes greater than 2 were regarded that sperm acrosin and acrosin-binding protein (ARCBP), as differentially expressed genes and a threshold of p value which is needed for biogenesis of acrosome [36], was also < 0.05 was used. For significant DE genes, GO term path- S-sulfhydrated in sperm (Table S1), suggesting that the post- way enrichment analysis was performed using the DAVID translational modification is important for successful ferti- functional annotation tool. lization. Gene ontology (GO) analysis of the sulfhydrated proteome indicated that proteins were ontologically enriched for a series of functional clusters whose top three are oxida- Results tion-reduction process, binding of sperm to zonapellucida, and tricarboxylic acid cycle (Fig. 1C). Analysis of sperm protein S‑sulfhydration Expression of 74 sperm proteins was associated with a high level of ROS in seminal ejaculates [24]. We analyzed No specific antibody recognizing S-sulfhydrated proteins the relationship of the ROS-associated sperm proteome with has been reported. Biotin-switch assay has been widely our S-sulfhydrated sperm proteome. The results showed used for analysis of S-sulfhydrated proteins [20, 31, 32], that 75.7% (56/74) of ROS-associated proteins were S-sulf- in which free thiols (-SH) of proteins were blocked by a hydrated proteins (Fig. 1D), implying that altered protein highly specific free sulfhydryl-reactive compound, methyl- S-sulfhydration is the way for spermatozoa to respond to OS. methanethiosulfonate (MMTS), which did not interact with sulfhydrated thiols (-SSH) or any other forms of oxidized Levels of sH3 and sH3.3 in asthenozoospermic thiols (S-S, for example). The sulfhydrated thiols were then spermatozoa selectively labeled with N-(6-(biotinamido)hexyl)-3′-(2′- pyridyldithio)-propionamide (biotin-HPDP), a compound Spermatozoa histone H3 was identified as one of the that interacts with sulfhydrated thiols in this assay, so that S-sulfhydrated proteins (Fig. 1B). To further validate that the sulfhydrated proteins were biotinylated. In order to H3 is an S-sulfhydrated protein, the crude sperm extract examine protein S-sulfhydration in sperm, the extracts from prepared from five pooled sperm samples from fertile men mouse and human sperm were applied for the biotin-switch was applied for biotin-switch assay. The sulfhydrated pro- assay. The S-sulfhydrated proteins were enriched by immu- teins were immunoprecipitated using anti-biotin antibody. noprecipitation (IP) using biotin antibody, separated via The IP was further analyzed with H3 antibody in Western SDS–polyacrylamide gel electrophoresis (SDS-PAGE), and blotting assay. The result showed that H3 was detected in finally detected by anti-biotin antibody in Western blotting the IP (Fig. 2A), indicating the presence of S-sulfhydrated assay (Fig. 1A, left) and via silver staining (Fig. 1A, right), H3 (sH3) in the extract. Additionally, the validation also respectively. The results showed that sperm S-sulfhydrated started with enrichment of H3 via IP of the crude sperm proteins were detected both via Western blotting assay and extract using H3 antibody. The IP was next analyzed in silver staining. biotin-switch assay, and applied for Western blotting assay To identify S-sulfhydrated proteins in human sperm, the using anti-biotin antibody. The result showed that H3 was S-sulfhydrated proteins stained with silver were subjected recognized by anti-biotin antibody (Fig. 2A), indicating that to in-gel trypsin digestion, next analyzed through liquid sperm H3 is S-sulfhydrated. chromatography-tandem mass spectrometry (LC-MS/MS) The presence of S-sulfhydration of H3 was addition- followed by protein database searching of the acquired spec- ally evaluated by with cysteinyl labeling assay that utilizes tra. To control for nonspecific IP, IgG preimmune complex a biotinylated iodoacetic acid (IAA) probe, which reacts was also analyzed. The experiments were performed in three through a nucleophilic substitution of the halide group by replicates, and each replicate was run through LC-MS/MS the H3-reactive thiol group, resulting in a stable thio-ether three times. Two hundred forty-four proteins were identified bond [26] (Fig S1A). We detected the presence of H3 after (Table S1) in the IP complex after (1) removing proteins that the lysate was applied for cysteinyl labeling assay in anti- were not found in the replicate experiments and (2) subtract- H3 antibody-based immunoblotting analysis (Fig S1B). ing common proteins that were found in the IgG preimmune Similarly, H3 was detected by anti-biotin antibody in the complex. WB assay, after the H3 was enriched from sperm lysate via In the list of proteins, S-sulfhydration of ATP synthase IP, and next treated in cysteinyl labeling assay. Again, the subunit alpha (ATP5a) [31, 33], glyceraldehyde-3-phosphate results demonstrated that sperm H3 is a sulfhydrated protein. 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3181 Fig. 1 The S-sulfhydrated proteome of human sperm. (A) Detection (right). (B) LC-MS/MS of a subset of the S-sulfhydrated proteins in of S-sulfhydrated proteins in human and mouse sperm. The lysates (A) identifies top notable S-sulfhydrated proteins, including GAPDH, were prepared from pooled samples of human sperm (from ten fer- GSTM3, and H3. (C) Biological processes enriched in the S-sulf- tile men) and mouse sperm (from three male mice), next subjected to hydrated proteins of human sperm. (D) Vann diagram depicting the the modified biotin-switch assay. The numerous sulfhydrated proteins relationship of sulfhydrated proteins and ROS-associated proteins of were detected with anti-biotin antibody (left) or with silver staining human sperm. The presence of S-sulfhydration of H3 was additionally through a nucleophilic substitution of the halide group evaluated by with cysteinyl labeling assay that utilizes a by the H3-reactive thiol group, resulting in a stable thio- biotinylated iodoacetic acid (IAA) probe, which reacts ether bond [26] (Fig S1A). We detected the presence of 1 3 3182 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3183 ◂Fig. 2 Association of sH3 and sH3.3 with asthenozoospermic sperm. Expressions of sH3 and S-sulfhydrated H3.3 (sH3.3) (A) The validation of sH3.S-sulfhydrated proteins of human sperm were measured in sperm with different motility. H3.3, a lysates were biotinylated via biotin-switch assay, enriched via the IP H3 variant played an important role in spermatogenesis based on anti-biotin antibody, and detected in Western blotting anal- [37], was thus selected in the analysis as well. H3 and H3.3 ysis using anti-H3 antibody (top panel). H3 in the lysate of pooled human sperm (n=5) was immunoprecipitated via anti-H3 antibody, in sperm lysate were enriched via their antibodies-based and subjected to biotin-switch assay. The S-sulfhydrated H3 was IPs, respectively. The IPs were divided into two portions: detected in Western blotting analysis using anti-biotin antibody (bot- one was applied for biotin-switch assay, and followed by tom panel). (B) Detection of overall S-sulfhydrated proteins in sperm an immunoblotting analysis using biotin antibody; the with different motility. Subpopulation of sperm with high and low motility from three fertile men was separated via swim-up assay, other was directly used as the loading control in Western respectively, and their protein lysates were subjected to biotin-switch blotting assay. Our results showed that abundances of assay, detected with anti-biotin antibody in Western blotting assay sH3 and sH3.3 were significantly higher in sperm with (left panel). As a loading control, the proteins in the sperm lysates high mobility than with low mobility (Fig.  2C). We stained with Coomassie blue after separated via SDS-PAGE (right panel). (C) Expression of sH3 and sH3.3 in sperm with different also measured levels of other PTMs like H3K4me3 and motility. The biotinylated proteins in (B) were immunoprecipitated H3K9me3 in the two subpopulations of normozoospermic with biotin antibody, and further analyzed with anti-H3 and anti-H3.3 sperm samples. However, no significant difference was antibodies in Western blotting assay. In addition, the biotinylated detected. Together, these results indicated that expression protein in (B) from sperm with different motility was subjected in Western blotting assay using the indicated antibodies. The experi- of S-sulfhydration of proteins including sH3 and sH3.3 is ments were replicated in three fertile individuals. Error bar denotes positively associated with sperm motility. mean ± SEM. *P < 0.05 and **P < 0.01. (D) H3 and H3.3 were Association of sH3 and sH3.3 with asthenozoospermia enriched via IP with their antibodies from lysates of clinical sperm was next addressed. Levels of sH3 were measured in semen samples with different progressive motility (PR%), as indicated, next subjected to biotin-switch assay. One portion of the biotinylated pro- samples with different percentage of progressive motility teins, as a loading control, were analyzed using anti-H3 and anti-H3.3 (PR%) in Western blotting assay. Lower mean levels of antibodies in Western blotting assay, while another was analyzed sH3 and sH3.3 were found in asthenozoospermic men for detection of sH3 or sH3.3 using biotin antibody in Western blot- compared with fertile men (Fig. 2D). Statistical analysis ting assay. (E) Relative expression of sH3 and sH3.3 in asthenozoo- spermic (ASTH) sperm (N=19 for sH3; N=24 for sH3.3) compared showed that the mean levels of sH3 in patients (n=19) were with the normozoospermic (NORM) controls (N=16 for sH3; N=26 53.1% of that in the fertile controls (n=16). Additionally, for sH3.3). Error bar denotes mean ± SEM. ***P < 0.001. (F) Cor- mean levels of sH3.3 in patients (n=24) were only 42.3% relations among sperm sH3 (n=35), sH3.3 (n=50), and progressive of that in the fertile controls (n=26) (Fig. 2E). Correlation motility were analyzed by linear regression. (G) Effects of H O and 2 2 H Son expression of human sperm sH3.3. The cultured sperm was analysis results revealed that sperm levels of sH3 and sH3.3 added with indicated concentration of H O and NaHS, respectively, 2 2 correlated positively with progressive motility (Fig. 2F). and 1 h later, subjected to analysis of sH3.3 expression. The analy- Collectively, our study demonstrated that expression of sis represented one of three independent experiments with almost the sH3 and sH3.3 was down-regulated in asthenozoospermic same results. sperm. The effect of sperm redox status on levels of sH3 was H3 after the lysate was applied for cysteinyl labeling next addressed via treating sperm with the oxidative agent assay in anti-H3 antibody-based immunoblotting analysis hydrogen peroxide (H O ) and NaHS, which has been 2 2 (Fig S1B). Similarly, H3 was detected by anti-biotin widely used as a H S donor in culture. H S was shown to 2 2 antibody in the WB assay, after the H3 was enriched from be an oxidants scavenger in sperm [19]. The results showed sperm lysate via IP, and next treated in cysteinyl labeling that H O reduced the level of sH3 in a dose-dependent way 2 2 assay. Again, the results demonstrated that sperm H3 is a (Fig. 2G). By contrast, NaHS raised the level of sH3 in a sulfhydrated protein. dose-dependent way. These results indicated that the level We next studied the association of level of of sperm sH3 is under the control of redox status in a cel- S-sulfhydrated protein with sperm motility. lular context. Normozoospermic sperm subpopulations with high motility and low motility were separated via the Dynamics of sH3 and sH3.3 in spermatogenesis “swim-up” assay. The S-sulfhydrated proteins in their lysates were enriched using biotin antibody after biotin- We next approached the dynamic level of H3 S-sulfhydra- switch assay, next detected with biotin antibody in tion during mouse spermatogenesis. Germ cells at differ - Western blotting assay. As shown in Fig. 2B, a number ent phases of spermatogenesis were first isolated, includ - of S-sulfhydrated proteins (around 15–100 KDa) were ing spermatogonial cells (SG), pachytenespermatocytes observed in the two sperm subpopulations. The overall (pacSC), which are at the prophase of the first meiotic divi - level of S-sulfhydrated proteins was higher in sperm with sion, round spermatids (rST), and elongating/condensed high motility than in sperm with low motility. spermatids (eST). Both rST and eST are haploid germ cells. 1 3 3184 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 Fig. 3 Analysis of expression of sH3 and sH3.3 in spermatogenesis sH3.3. (B) The incubated rST were treated with two indicated doses and effect of RA and NaHS on level of sH3 in germ cells. (A) Sper - of RA for 24 h, then subjected for analysis of sH3.3 expression. (C) matognia (SG), pachytene spermatocytes (pacSC), round spermatids The incubated pieces of testis were treated with the indicated con- (rST), and elongating/condensed spermatids (eST) were isolated and centrations of NaHS for 24 h, and next subjected for measurement of characterized as described in “Experimental procedures.” These germ expression of sH3.3. The experiments were repeated independently cells, together with sperm from caput epididymis (SPM (cau)) and for three (for B and C) to four (for A) times. Error bar denotes mean from cauda epididymis (SPM (cau)), were applied for preparation for ± SEM. *P < 0.05 and **P < 0.01. protein lysates, subjected for measurement of expression of sH3 and 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3185 In addition, sperm from caput epididymis (SPM (cap)) and abolished when its C111 was mutated. These results indi- cauda epididymis (SPM (cau)) was isolated, respectively. cated that C111 is the site for S-sulfhydration both in H3.1 Levels of sH3 and sH3.3 in the six types of germ cells were and H3.3. measured (Fig.  3A). These cells were applied for meas- We failed to directly detect S-sulfhydrated peptide urement of sH3 and sH3.3. Our results showed that all the digested from human sperm H3 via MS analysis. We first types of germ cells expressed sH3 and sH3.3. A small but asked the question whether S-sulfhydration H3 could be significant up-regulation of sH3 (by 151.2%) and sH3.3 directly observed by MS analysis. The key is whether the (by 164.7%) was detected in rST, compared with pacSC. peptides containing the sulfhydrated C111 could be iden- Notably, levels of sH3 and sH3.3 were significantly higher tified by MS after trypsin digestion. We next checked the by 3.02 folds and 4.11 folds, respectively, in eST than in PeptideAtlas database (http://www.peptideatlas.org/), rST (Fig. 3A). As a control, expression of H3K4me3 was which is a multi-organism, publicly accessible compen- also measured and found to drop much in eST, compared to dium of peptides identified in a large set of tandem mass rST. No statistically significant difference was observed in spectrometry proteomics experiments, for all possible expression of sH3, and sH3.3 was observed between eST and peptides that could be identified by MS for H3 ( https:// db. SPM (cap). Together, the results indicated spermiogenesis syste msbio logy. net/ sbeams/ cgi/ Pepti deAtl as/ GetPr otein? is the main stage for H3 and H3.3 to be S-sulfhydrated in atlas_ build_ id= 337& prote in_ name= P8424 3& action= spermatogenesis. QUERY). As shown in the above link and in Supplemen- RA is a key physiological factor triggering differentia - tary Fig.  2, the peptide between N-terminal 81 and 137 tion of rST to eST [38]. We next investigated the effect of amino acid residues of H3 belongs to the category which RA on H3.3S-sulfhydration in spermatids. The cultured is unlikely to be identified due to its length. C111 happens rST were treated with RA soon after they were separated to fall into this peptide sequence, and S-sulfhydration of from testis. The results showed that 0.3μM and 1.0 μM RA C111 thus could not be identified directly by MS. raised expression of sH3.3 by 144.0% and 266.2% (Fig. 3B), In order to analyze the effect of S-sulfhydration of H3.3 respectively. Together, these results indicated that S-sulfhy- on a nucleosomal structure, the structure of human H3.3- dration of H3.3 is induced by RA in rST. containing nucleosome was next stimulated based on its We next studied whether H3.3 was susceptible to crystal structure solved at 2.8 Å [PDB ID: 3AV2] [40]. H S-induced sulfhydration in a testicular context. The pieces As shown in the nucleosome contains two H3.3 (Fig. 4C of testis were incubated and treated with NaHS. The results and D), the two C111 are located in the alpha helix (85- showed that sH3.3 expression was raised significantly upon 114) of H3.3, facing each other. The range between two treatment with the H S donor (Fig.  3C), suggesting that sulfur atoms is 6.31 Å (Fig. 4D). Given an average dis- H3.3 S-sulfhydration is under the control of H S signaling tance of 2.04 Å between the two sulfur atoms which gen- in spermatogenesis. erates a disulfide bond [ 41], it is still too far for the two cysteines to form a disulfide bond. However, when both Analysis of S‑ sulfhydrated amino acid residue in H3 of C111 are S-sulfhydrated, the distance between the two outer sulfur atoms is narrowed to approximately 2.23 Å, At least 6 variants of H3 have been reported (Fig. 4A). The which is probably close enough to form a disulfide bond canonical H3.1 and H3.2 are expressed and deposited on (Fig.  4E). Therefore, our simulation analysis suggested nucleosomes during DNA replication [39]. The expression that S-sulfhydration of H3.3 is beneficial to a formation of H3.1t is testis-specific. Centromere protein A (CENP- of an inter-molecular disulfide bond between two nucleo - A), a highly specialized variant, is only present at the cen- somal H3.3 proteins. tromere. Mammalian H3.3 is expressed throughout the cell On the other hand, our analysis of structure of two cycle, and deposited by a DNA replication–independent H3.3-containing nucleosome [40] also revealed another nucleosome assembly pathway [39]. Only two cysteines, potential effect of S-sulfhydration of C111. The two alpha Cysteine 111 (C111) and Cysteine 97 (C97), are present helixes (85-114) containing two C111 are close to another in H3. two helixes (120-132) of H3.3, which contains two Argi- H3.1 has two cysteine residues. We studied which nine128 (R128) (Fig.  4F). The region around C111 in cysteine mutation could disrupt H3.1 sulfhydration in tran- these four helixes is important to form the H3-H3 hydro- sient transfection assay (Fig. 4B). Serine was used to replace phobic four-helix bundle tetramer interface so as to hold cysteine in the constructs expressing the mutant H3.1. Our together two histone H2A-H2B-H3-H4 tetramers [42]. results showed that the mutation of C111, but not C97, com- Our analysis showed that the two nitrogen atoms of two pletely disrupted H3.1 sulfhydration. H3.3 has only one R128 are very close to two sulfur atoms of two C111, and cysteine, C111. Similarly, H3.3 S-sulfhydration was fully their distance is 3.68~4.60 Å (Fig.  4F). Most of N-H-S hydrogen bonds can form with a distance of 3.25~3.55 1 3 3186 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3187 ◂Fig. 4 Analysis of effect of H3.3S-sulfhydration on structure of recombinant Lentivirus that expressed wild-type H3.3 and nucleosomal H3.3. (A) Sequence alignment of human H3 family. a C111S-containing H3.3, respectively. No significant dif - H3 family only contains two cysteine residues which were framed in ference in the intensity of bands was detected with FLAG the sequences. Note that C111 is conservative in five members of H3 antibody between the two samples (Fig. 5B), indicating that family. (B) Effect of mutation of C97 and C111 on S-sulfhydration of H3.1 and H3.3. The C18-4 cells were transiently transfected with the mutation did not affect H3.3 expression based on virus wild-type and mutant, as indicated, H3.1 and H3.3 expression con- infection. Importantly, we detected that expression of sH3.3 structs, respectively. Forty-eight hours later, the cells were harvested was significantly higher in the C18-4 cells overexpressing for measurement of levels of sulfhydrated wild and mutant H3.1 and wild-type H3.3 than those overexpressing the mutant H3.3 H3.3. H3.1DM: the mutant H3.1 contains both C97S and C111S. (C) The structure of H3.3 in nucleosome, based on crystal structure (Fig. 5B). solved at 2.8 Å [PDB ID:3AV2]. H3.3 (pink) forms as dimer in nucle- We next measured the cell number at different times osome, binding with H2A, H2B, and two H4. Cys111 (blue) locates after the infected and untreated C18-4 cells were seeded. in the center. (D) Range between two sulfur atoms of two C111. C111 We also did not detect any significant difference in cell locates in alpha helixes (85-114, pink), and is surrounded by alpha helix (120-132, orange). The distance between two sulfur atoms of number between C18-4 cells infected with wild-type H3.3- two Cys111 is 6.31Å, as indicated. (E) Simulated structure of H3.3 expressing virus and the untreated C18-4 cells (Fig. 5C). when both Cys111 are S-sulfhydrated. The simulated electron cloud Strikingly, the mutant H3.3-expressing C18-4 cells grew of outer sulfur atoms (red) is overlapped. (F) Distance between two faster than wild-type H3.3-expressing cells. The mutant sulfur atoms of two Cys111 and two nitro atoms of two R128. R128 locates in alpha helix (120-132, orange). The distance between each H3.3-expressing cells were more than the controls, sig- R128 and C111 is 3.68~4.60Å, showed by blue and red line. (G) The nificantly by 25.8% and 36.1%, at the third and fifth day presence of overlapping electron cloud of the four atoms when both after the plating of cells (Fig. 5C). The results strongly C111 are S-sulfhydrated. suggested that sH3.3 is inhibitory to the growth of C18-4 cells, consistent with the repressive effect of GDNF on sH3.3 expression. We next performed RNA sequencing analysis (RNA- Å [43]. When both of C111 are S-sulfhydrated, the dis- seq) in order to study the mechanism underlying the pro- tance between two C111 and two R128 is much closer, so moting effect of C111S of H3.3 on the C18-4 cell growth that the simulated electron cloud of 4 atoms overlaps each rate. We found that expressions of 487 genes were down- other (Fig. 4G). Therefore, these four residues can prob- regulated, while the other 272 genes were up-regulated ably generate multiple inter-helical hydrogen bonds, thus (Fig. 5D, Table S2). Validation by quantitative RT-PCR stabilizing the structure of four-helix bundle tetramer, and was performed for some differentially expressed genes the whole nucleosome. (DEGs), and our quantitative RT-PCR analysis confirmed the RNA-seq data (Fig. 5E). Eec ff t of the C111 mutation of H3.3 on growth rate Introduction of the mutated H3.3 reduced relative and gene expression of C18‑ 4 cells expression of Cyclin D1 (Ccnd1). Ccnd1 expression was inhibitory to growth of SSCs [45]. Rassf8 reduced the The presence of sH3.3 in spermatogonia (Fig.  3A) sug- expression of ccnd1 when overexpressed in SSC [46], gested that sH3.3 could play an important role in the phase and its expression was also unregulated in the presence of mitosis in spermatogenesis. We studied the hypothesis of the mutated H3.3 (Fig.  5E). Therefore, sH3.3 could using SSC C18-4 cell line. Glial cell-line-derived neuro- regulate renewal of SSC via targeting the two genes. trophic factor (GDNF) is bona fide self-renewal factors of Some growth-inhibitory genes including Cdk5 and Abl SSC, and promotes proliferation of C18-4 cells [25]. By con- enzyme substrate 1 (Cables1) [47] and ankyrin repeat trast, RA signaling, which is a key physiological regulator domain 1 (Ankrd1) [48] were shown to be down-regulated of SSC differentiation, is also present in C18-4 cells [ 44]. in the presence of C111S (Fig.  5E). Among the list of The impact of GDNF and RA on sH3.3 expression was next unregulated genes, insulin-like growth factor 1 receptor investigated. GDNF was found to upgrade sH3.3 expres- (IGF-1R) is essential for the proliferation of mouse SSC sion in a time-dependent way (Fig. 5A). A significant rise by promoting the G2/M progression of the cell cycle [49]. in level of sH3.3 was observed as early as 2 h after GDNF Wnt1 inducible signaling pathway protein 1 (Wisp1) is treatment. However, sH3.3 expression was reduced in the required for proliferation of mesenchymal stem cells [50]. C18-4 cells when treated with RA in a time-dependent way. Dishevelled segment polarity protein 3 (Dvl3) repressed These results suggested that an altered sH3.3 expression is differentiation of mesenchymal stem cells inhibited via an important downstream event in signaling of GDNF and up-regulating Ccnd1 [51], and unregulated upon the RA. ectopic expression of the mutant H3.3. Together, the In order to explore the role of sH3.3 in the germ cells, we mutation could promote the growth of C18-4 cells by studied the effect of the H3.3 C111S mutant on growth rate regulating expression of these genes. of C18-4 cells. The C18-4 cells were thus infected with the 1 3 3188 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3189 ◂Fig. 5 Effect of ect opic expression of the mutant H3.3 with C111S on are S-sulfhydrated proteins in human sperm (Fig.  1D). cell growth rate and gene expression in the C18-4 cells. (A) Effects Expression of sH3.3 in male germ cells including sperm of GDNF and RA on sH3.3 expression in the C18-4 cells. Twenty- is unregulated by H S, a potent antioxidant (Figs. 2G and four hours after plating, the C18-4 cells were added with GDNF (50 3C), while sperm sH3.3 was down-regulated by H O ng/ml) and RA (1.0μM), respectively. The cells were harvested for 2 2 measurement of sH3.3 expression at the indicated times after the (Fig. 2G), strongly suggesting that expression of sH3.3 is treatment. Error bar denotes mean ± SEM. *P < 0.05 and **P < under control of redox status. H S raised enzyme activities 0.01. (B) Expression of sH3.3 and H3.3 in the C18-4 cells infected of ATP synthase [32] and LDHA [34] via up-regulation with the recombinant Lentivirus that expressed wild-type and mutant of their S-sulfhydration. Collectively, it is plausible that (C111S) H3.3. The cultured C18-4 cells were harvested for the analy- sis when their confluence reached approximately 80 –90%. These ROS represses H S signaling, which in turn causes hypo- results represented one of three independent experiments with the sulfhydration of proteins including H3/H3.3 in a subtype of similar data. (C) The growth rates of C18-4 cells. Untreated C18-4 asthenozoospermic sperm. Therefore, our study highlights cells and infected C18-4 cells which overexpressed wild-type or that sH3.3/sH3 is potentially a novel biomarker for diag- mutant H3.3 were seeded as described in “Experimental proce- dures,” and harvested for the counting of cell number at the indicated nosing etiology of asthenozoospermia. times after seeding. (D) Volcano plot showing differential expres - To our knowledge, both protein S-sulfhydration in germ sion of protein-coding genes between mutated and normal samples. cells and H3/H3.3 S-sulfhydration have not been reported Red and blue dots indicate significantly down-regulated ( p<0.05 and before. More than ten different PTMs of H3 were elsewhere log2FC<-1) and significantly up-regulated ( p<0.05 and log2FC>1) differential expression, respectively. (E) RNA-seq data validation by reported [53, 54], Therefore, this work extends the catalogue quantitative RT-PCR. Expression of select up-regulated and down- of histone PTM sites in mammalian cells. Oxidative stress regulated genes from the RNA-seq analysis was measured by quan- and ROS are emerging as important players, shaping the epi- titative RT-PCR in the C18-4 cells. (F) Biologic processes that are genetic landscape of the entire genome via different mecha - enriched in genes down-regulated (left) and up-regulated (right). nisms including modification of H3 methylation and acety - lation [55, 56]. Our study strongly suggests that sperm H3 GO analysis on the down-regulated DEGs found many S-sulfhydration is under the control of redox homeostasis, proteins involved in positive regulation of cell death, and unravelling epigenetic mechanisms underlying the patho- positive regulation of apoptotic process that were highly physiology of male infertility. Some interesting questions expressed in C18-4 cells with overexpression of the mutant have emerged from our study. For example, how do ROS- H3.3. Keeping in line with it, some significant terms asso - producing factors including smoking, alcohol, and inflam - ciated with up-regulated mRNAs in the presence of the mation affect sperm H3 S-sulfhydration? It is known that mutant H3.3 were cell cycle, mitotic sister chromatin seg- high levels of ROS can cause male infertility through not regation, and cellular macromolecule biosynthetic process. only by lipid peroxidation or DNA damage but also reduced total antioxidative capability in spermatozoa. What are their relationships with altered H3 S-sulfhydration? A few anti- Discussion oxidant medicines have been used to treat male infertility with different curative effects [ 57]. Can investigation of We reported the human sperm S-sulfhydrated proteome sperm sH3 before and after treatment allow a better under- including 244 proteins in the present study. GO analysis standing, monitoring, or selection of alternative antioxidant suggested that S-sulfhydrated proteins played important medicines? They are issues worth of investigation. roles in spermatogenesis, spermiogenesis, and fertilization. H3.3 has been reported to be important for spermatogen- S-sulfhydration of GAPDH significantly raised its enzyme esis [37, 58, 59]. We reported that S-sulfhydration of H3 activity [31]. Male mice with deficiency of GAPDH were and H3.3 was detected throughout spermatogenesis in the infertile and had profound defects in sperm motility [52]. present study. Noteworthy, RA, which is known to induce S-sulfhydration of ATP synthase [32] and LDHA [34] rST to differentiate into eST [ 38], also up-regulated their raised mitochondrial bioenergetics. Therefore, S-sulfhydra- sH3.3 expression (Fig. 3B). Keeping in line with the obser- tion of these proteins is probably required for maintenance vation, the level of sH3.3 was significantly higher in eST of optimal motility of sperm via regulating energy metabo- than in rST (Fig. 3A). Deficiency of mouse H3.3 resulted in lism. Our study also revealed a new mechanism regarding an aberrant spermiogenesis including an impaired develop- why addition of exogenous H S to semen improved the ment of round spermatids and poor motility of sperm [58, asthenozoospermic sperm motility [19]. Importantly, over- 59]. Collectively, sH3.3 is probably required for RA-induced all expression of sulfhydrated proteins including sH3/H3.3 spermiogenesis. Distinct aberrant PTMs of histones in sper- is higher in sperm with high motility than with low motility miogenesis resulted in infertile phenotypes including poor (Fig. 2B–E), and levels of sH3 and sH3.3 are positively sperm motility, strongly suggesting they can affect sperm associated with sperm progressive motility (Fig. 2F). Our motility in the way depending on their roles in modulating analysis also revealed that most of ROS-associated proteins gene transcription or disturbing sperm chromatin remodeling 1 3 3190 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 [60–64]. It is plausible that hypo-sulfhydration of H3.3 may sH3.3 expression in the C18-4 cells is oppositely regu- cause asthenozoospermia in a similar way. lated by GDNF and RA (Fig.  5A), two key factors for A hallmark of mammalian spermiogenesis is the step- SSC self-renewal and differentiation. Our study is con- wise completion of transition from histones to protamines sistent with the hypothesis that sH3.3 is important for the in spermatids [1, 2]. During the process, vast majority control of fate determination of SSCs. Ectopic expression of not only total histones but also levels of differentially of mutant H3.3 with C111S unregulated growth rate of modified histones were much reduced in eST, or mature C18-4 cells, partially by modulating expression of posi- sperm compared with rST [54, 62, 65–67]. Consistent with tive regulation of cell death and positive regulation of the reports, we also detected an expression of H3K9me3 apoptotic process–related genes and cell cycle–related in rST but it largely disappeared in eST (Fig. 3A). By con- genes (Fig.  5F). Therefore, sH3.3 is likely a suppressor trast, the presence of increased abundance of sH3/sH3.3 for the mitotic division of differentiating spermatogonia. in eST (Fig. 3A) strongly suggests that sH3/sH3.3 marks The role of sH3.3 in differentiation of SSC should be next the retained nucleosomes. The retained nucleosomes addressed. Nevertheless, our study revealed a regulatory have been revealed to distribute in genomic DNA in a role of sH3.3 in transcriptome in the SSC line. well-organized manner [68, 69], implying the existence Genomic distribution of H3.3 is critical to its regulat- of machinery protecting retained histones from eviction. ing role in gene transcription [37, 75], and is regulated However, little is known currently regarding the mecha- by RA in the way depending on raising turnover of H3.3 nisms. Considering that histone acetylation per se attenu- in the differentiation of embryonic stem cells [76, 77]. A ates the interplay between histone and DNA to facilitate high turnover of H3.3 was detected in male meiosis [37] histone removal [1], it is naturally tempting to speculate and spermiogenesis [69]. Considering that RA unregulated that the nucleosomes with an extra stabilizing mechanism expression of sH3.3 in male germ cells found in the pre- can probably be exempted from histone removal. The most sent study, whether/how S-sulfhydration of H3.3 affects members of the mammalian H3 family contain one or two genomic distribution of H3.3 should be next investigated cysteine(s) in their protein core, and this feature is a hall- in the different phases of spermatogenesis. C111 of H3.3 mark property of H3, given all other histone proteins lack is located inside the protein, making it difficult to be acces- cysteine. Intriguingly, the mammalian H3 variants contain sible for interaction of modified nucleosomal C111 with C111 that is located in their helix (85-114), the region any non-histone proteins. Therefore, H3.3 is likely to be where both H3 proteins are closely apposed in the nucleo- sulfhydrated largely outside nucleosomes. Some other some core particle [70]. The region immediately surround- PTMs of H3 were finished also outside nucleosomes [78]. ing C111 is important to hold together two histone H2A- In this regard, one can envision that H3.3 S-sulfhydration H2B-H3-H4 tetramers, because mutations of C111, for selectively regulates the turnover rate and distribution of example, destabilized the H3-H3 hydrophobic four-helix H3.3 probably via modulating interaction of H3.3 and its bundle tetramer interface in vitro [71]. Therefore, Hake chaperone proteins that were detected in spermatogenesis and Allis proposed that two C111 form an intermolecular [37, 79]. The hypothesis is worth a further approach. disulfide bond within two H3 proteins in the same nucleo- In conclusion, for the first time, H3.3 and H3 are some, adding stability to the H3-H4 tetramer [42]. Our showed to be S-sulfhydrated proteins in the present analysis showed the two C111 of nucleosomal H3.3 are study. We demonstrated that levels of sH3.3 and sH3 6.31 Åapart (Fig. 4D), basically excluding the possibility were down-regulated in asthenozoospermic sperm, sug- that they form the disulfide bond. However, once the two gesting that hypo-sulfhydration of H3 and H3.3 is a new Cys111 are thio-modified, the distance is narrowed to 2.3 biomarker for male infertility. sH3 has been detected Å (Fig. 4E), thus probably generating an intra-nucleoso- in all the different mouse organs examined (data not mal disulfide bond. In addition, our analysis also suggested shown). It is well known that oxidative stress is involved that S-sulfhydration of C111 is favorable to the formation in initiation and progression of diabetes, neurodegen- of multiple inter-helical hydrogen bonds between them and erative diseases, vascular disease, hypertension, aging, R128 (Fig. 4G). The four-residue-based multiple hydrogen and many other pathologies. Therefore, it could be bonds have been reported to exist in the structures of the speculated that aberrant regulation of sH3 can shape four-helix bundle tetramer [72, 73], causing the forma- epigenetic landscape, eventually making a significant tion of a super-secondary structure of four-stranded coiled contribution to the initiation and progression of distinct coil [74], and thus probably adds stability to the H3–H4 chronic diseases. tetramer. Collectively, S-sulfhydration of C111 is likely to Supplementary Information The online version contains supplemen- benefit the exemption of some nucleosomes from histone tary material available at https://doi. or g/10. 1007/ s10815- 021- 02314-x . removal via strengthening nucleosomal stability. 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3191 Acknowledgements We are very thankful for Prof. Biaoyang Lin, 9. Macleod G, Varmuza S. The application of proteomic approaches who works in Hangzhou Proprium Biotech Company, for his proteomic to the study of mammalian spermatogenesis and sperm function. analysis and comments. We are also very thankful for Shanghai Biopro- FEBS J. 2013;280(22):5635–51. file Company for its technical support on LC-MS analysis. 10. Saraswat M, et al. Human spermatozoa quantitative proteomic signature classifies normo- and asthenozoospermia. Mol Cell Pro - teomics. 2017;16(1):57–72. Funding This work was supported by grants from the National Nat- 11. Martinez-Heredia J, et  al. Identification of proteomic differ- ural Science Foundation of China (grant no. 81971443) (to R.L.), ences in asthenozoospermic sperm samples. Hum Reprod. Shanghai Municipal Committee of Science and Technology (grant no. 2008;23(4):783–91. 21140903600) (to R.L.), Shanghai Municipal Science and Technology 12. Parte PP, et al. Sperm phosphoproteome profiling by ultra per - Commission (21S11901000) (to R.L.), Innovation-Oriented Science formance liquid chromatography followed by data independent and Technology Grant from NHC Key laboratory of Reproduction Reg- analysis (LC-MS(E)) reveals altered proteomic signatures in ulation (CX2017–07) (to R.L.), and Shanghai Municipal Health and asthenozoospermia. J Proteome. 2012;75(18):5861–71. Health Commission clinical research project (20194Y0237) (to M.Z). 13. Marchiani S, et al. Sumo1-ylation of human spermatozoa and its relationship with semen quality. Int J Androl. 2011;34(6 Pt Data availability The data used to support the findings of this study are 1):581–93. available from the corresponding author upon request. 14. Cheng YM, et al. Lysine glutarylation in human sperm is associ- ated with progressive motility. Hum Reprod. 2019;34(7):1186–94. Declarations 15. Cheng YM, et al. Posttranslational lysine 2-hydroxyisobutyryla- tion of human sperm tail proteins affects motility. Hum Reprod. 2020;35(3):494–503. Conflict of interest The authors declare no competing interests. 16. Sun G, et al. Insights into the lysine acetylproteome of human sperm. J Proteome. 2014;109:199–211. Open Access This article is licensed under a Creative Commons Attri- 17. Li Y, et  al. The protein phosphorylation landscape of bution 4.0 International License, which permits use, sharing, adapta- mouse spermatids during spermiogenesis. Proteomics. tion, distribution and reproduction in any medium or format, as long 2019;19(11):e1900055. as you give appropriate credit to the original author(s) and the source, 18. Xie ZZ, Liu Y, Bian JS. Hydrogen sulfide and cellular redox provide a link to the Creative Commons licence, and indicate if changes homeostasis. Oxidative Med Cell Longev. 2016;2016:6043038. were made. The images or other third party material in this article are 19. Wang J, et al. Hydrogen sulfide as a potential target in preventing included in the article’s Creative Commons licence, unless indicated spermatogenic failure and testicular dysfunction. Antioxid Redox otherwise in a credit line to the material. If material is not included in Signal. 2018;28(16):1447–62. the article’s Creative Commons licence and your intended use is not 20. Mustafa AK, et al. H2S signals through protein S-sulfhydration. permitted by statutory regulation or exceeds the permitted use, you will Sci Signal. 2009;96(2):ra72. need to obtain permission directly from the copyright holder. To view a 21. Zhang D, et al. H2S-induced sulfhydration: biological function copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . and detection methodology. Front Pharmacol. 2017;8:608. 22. Ju Y et al. H(2)S-mediated protein S-sulfhydration: a prediction for its formation and regulation. Molecules. 2017 22(8) 23. Yang G, et al. Hydrogen sulfide protects against cellular senes - cence via S-sulfhydration of Keap1 and activation of Nrf2. Anti- References oxid Redox Signal. 2013;18(15):1906–19. 24. Sharma R, et al. Proteomic analysis of human spermatozoa pro- 1. Bao J, Bedford MT. Epigenetic regulation of the histone-to- teins with oxidative stress. Reprod Biol Endocrinol. 2013;11:48. protamine transition during spermiogenesis. Reproduction. 25. He Z, et al. Gdnf upregulates c-Fos transcription via the Ras/ 2016;151(5):R55–70. Erk1/2 pathway to promote mouse spermatogonial stem cell pro- 2. Wang T, et al. Essential role of histone replacement and modifica - liferation. Stem Cells. 2008;26(1):266–78. tions in male fertility. Front Genet. 2019;10:962. 26. Krishnan N, et al. H2S-Induced sulfhydration of the phosphatase 3. Lu WH, Gu YQ. Insights into semen analysis: a Chinese perspec- PTP1B and its role in the endoplasmic reticulum stress response. tive on the fifth edition of the WHO laboratory manual for the Sci Signal. 2011;4(203):ra86. examination and processing of human semen. Asian J Androl. 27. Pan H, et al. Ndrg3 gene regulates DSB repair during meiosis 2010;12(4):605–6. through modulation the ERK signal pathway in the male germ 4. Agarwal A, Gupta S, Sikka S. The role of free radicals and cells. Sci Rep. 2017;7:44440. antioxidants in reproduction. Curr Opin Obstet Gynecol. 28. Buchou T, et al. Purification and analysis of male germ cells from 2006;18(3):325–32. adult mouse testis. Methods Mol Biol. 2017;1510:159–68. 5. Smith TB, et  al. Functional deletion of Txndc2 and Txndc3 29. Wang W, et al. NDRG3 is an androgen regulated and prostate increases the susceptibility of spermatozoa to age-related oxida- enriched gene that promotes in vitro and in vivo prostate cancer tive stress. Free Radic Biol Med. 2013;65:872–81. cell growth. Int J Cancer. 2009;124(3):521–30. 6. Chen K, et  al. Low NRF2 mRNA expression in spermato- 30. Tarazona S, et  al. Data quality aware analysis of differential zoa from men with low sperm motility. Tohoku J Exp Med. expression in RNA-seq with NOISeq R/Bioc package. Nucleic 2012;228(3):259–66. Acids Res. 2015;43(21):e140. 7. Meybodi AM, et al. Importance of sperm gluthatione treatment in 31. Jarosz AP, et  al. Glyceraldehyde 3-phosphate dehydrogenase ART. J Assist Reprod Genet. 2012;29(7):625–30. (GAPDH) is inactivated by S-sulfuration in vitro. Free Radic Biol 8. Cao X, et al. Proteomic profile of human spermatozoa in healthy Med. 2015;89:512–21. and asthenozoospermic individuals. Reprod Biol Endocrinol. 32. Modis K, et al. S-Sulfhydration of ATP synthase by hydrogen 2018;16(1):16. sulfide stimulates mitochondrial bioenergetics. Pharmacol Res. 2016;113(Pt A):116–24. 1 3 3192 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 33. Paul BD, Snyder SH. H(2)S signalling through protein sulfhydra- 54. Tan M, et  al. Identification of 67 histone marks and histone tion and beyond. Nat Rev Mol Cell Biol. 2012;13(8):499–507. lysine crotonylation as a new type of histone modification. Cell. 34. Untereiner AA, et al. H2S-induced S-sulfhydration of lactate dehy- 2011;146(6):1016–28. drogenase a (LDHA) stimulates cellular bioenergetics in HCT116 55. Thomas K, et al. The epigenetic landscape related to reactive oxy- colon cancer cells. Biochem Pharmacol. 2017;136:86–98. gen species formation in the cardiovascular system. Br J Pharma- 35. Zhao W, et al. Outer dense fibers stabilize the axoneme to maintain col. 2017;174(12):1533–54. sperm motility. J Cell Mol Med. 2018;22(3):1755–68. 56. Sarah K, et al. Oxidative stress signaling to chromatin in health 36. Kanemori Y, et al. Biogenesis of sperm acrosome is regulated by and disease. Epigenomics. 2016;8(6):843–62. pre-mRNA alternative splicing of Acrbp in the mouse. Proc Natl 57. Barati E, et al. Oxidative stress and male infertility: current knowl- Acad Sci U S A. 2016;113(26):E3696–705. edge of pathophysiology and role of antioxidant therapy in disease 37. van der Heijden GW, et  al. Chromosome-wide nucleosome management. Cell Mol Life Sci. 2020;77(1):93–113. replacement and H3.3 incorporation during mammalian meiotic 58. Yuen BT, et al. Histone H3.3 regulates dynamic chromatin states sex chromosome inactivation. Nat Genet. 2007;39(2):251–8. during spermatogenesis. Development. 2014;141(18):3483–94. 38. Endo T, et al. Periodic production of retinoic acid by meiotic and 59. Tang MC, et  al. Contribution of the two genes encoding his- somatic cells coordinates four transitions in mouse spermatogen- tone variant h3.3 to viability and fertility in mice. PLoS Genet. esis. Proc Natl Acad Sci U S A. 2017;114(47):E10132–41. 2015;11(2):e1004964. 39. Henikoff S, Ahmad K. Assembly of variant histones into chroma- 60. Stiavnicka M, et al. H3K4me2 accompanies chromatin immaturity tin. Annu Rev Cell Dev Biol. 2005;21:133–53. in human spermatozoa: an epigenetic marker for sperm quality 40. Tachiwana H, et al. Structures of human nucleosomes containing assessment. Syst Biol Reprod Med. 2020;66(1):3–11. major histone H3 variants. Acta Crystallogr D Biol Crystallogr. 61. Yap DB, et al. Mll5 is required for normal spermatogenesis. PLoS 2011;67(Pt 6):578–83. One. 2011;6(11):e27127. 41. Craig DB, Dombkowski AA. Disulfide by Design 2.0: a web- 62. Lu LY, et  al. RNF8-dependent histone modifications regu- based tool for disulfide engineering in proteins. BMC Bioinfor - late nucleosome removal during spermatogenesis. Dev Cell. matics. 2013;14:346. 2010;18(3):371–84. 42. Hake SB, Allis CD. Histone H3 variants and their potential role 63. Liu S, et al. Chromodomain protein CDYL acts as a crotonyl-CoA in indexing mammalian genomes: the “H3 barcode hypothesis”. hydratase to regulate histone crotonylation and spermatogenesis. Proc Natl Acad Sci U S A. 2006;103(17):6428–35. Mol Cell. 2017;67(5):853–66 e5. 43. Adman E, Watenpaugh KD, Jensen LH. NH---S hydrogen bonds 64. Yu YE, et al. Abnormal spermatogenesis and reduced fertility in in Peptococcus aerogenes ferredoxin, Clostridium pasteurianum transition nuclear protein 1-deficient mice. Proc Natl Acad Sci U rubredoxin, and Chromatium high potential iron protein. Proc Natl S A. 2000;97(9):4683–8. Acad Sci U S A. 1975;72(12):4854–8. 65. Luense LJ, et al. Comprehensive analysis of histone post-transla- 44. Koli S, Mukherjee A, Reddy KVR. Retinoic acid triggers c-kit tional modifications in mouse and human male germ cells. Epi- gene expression in spermatogonial stem cells through an enhan- genetics Chromatin. 2016;9:24. ceosome constituted between transcription factor binding sites 66. Dottermusch-Heidel C, et al. H3K79 methylation: a new con- for retinoic acid response element (RARE), spleen focus forming served mark that accompanies H4 hyperacetylation prior to his- virus proviral integration oncogene (SPFI1) (PU.1) and E26 trans- tone-to-protamine transition in Drosophila and rat. Biol Open. formation-specific (ETS). Reprod Fertil Dev. 2017;29(3):521–43. 2014;3(6):444–52. 45. He Z, et al. MiRNA-20 and mirna-106a regulate spermatogonial 67. Song N, et al. Immunohistochemical analysis of histone H3 modi- stem cell renewal at the post-transcriptional level via targeting fications in germ cells during mouse spermatogenesis. Acta His- STAT3 and Ccnd1. Stem Cells. 2013;31(10):2205–17. tochem Cytochem. 2011;44(4):183–90. 46. Wang Y, et al. MicroRNA-322 regulates self-renewal of mouse 68. Hammoud SS, et  al. Distinctive chromatin in human spermatogonial stem cells through Rassf8. Int J Biol Sci. sperm packages genes for embryo development. Nature. 2019;15(4):857–69. 2009;460(7254):473–8. 47. Arnason T, et al. Cables1 is a tumor suppressor gene that regulates 69. Erkek S, et al. Molecular determinants of nucleosome retention at intestinal tumor progression in Apc(Min) mice. Cancer Biol Ther. CpG-rich sequences in mouse spermatozoa. Nat Struct Mol Biol. 2013;14(7):672–8. 2013;20(7):868–75. 48. Jimenez AP, et  al. The tumor suppressor RASSF1A induces 70. Luger K, et al. Crystal structure of the nucleosome core particle the YAP1 target gene ANKRD1 that is epigenetically inacti- at 2.8 A resolution. Nature. 1997;389(6648):251–60. vated in human cancers and inhibits tumor growth. Oncotarget. 71. Banks DD, Gloss LM. Folding mechanism of the (H3- 2017;8(51):88437–52. H4)2 histone tetramer of the core nucleosome. Protein Sci. 49. Wang S, et al. IGF-1R signaling is essential for the proliferation of 2004;13(5):1304–16. cultured mouse spermatogonial stem cells by promoting the G2/M 72. Haworth NL, et  al. Cross-strand disulfides in the hydrogen progression of the cell cycle. Stem Cells Dev. 2015;24(4):471–83. bonding site of antiparallel beta-sheet (aCSDhs): forbidden 50. Cernea M, et al. Wisp1 mediates Bmp3-stimulated mesenchymal disulfides that are highly strained, easily broken. Protein Sci. stem cell proliferation. J Mol Endocrinol. 2016;56(1):39–46. 2019;28(1):239–56. 51. He H, et al. miR-204-5p promotes the adipogenic differentiation 73. Liu J, et al. A seven-helix coiled coil. Proc Natl Acad Sci U S A. of human adipose-derived mesenchymal stem cells by modulating 2006;103(42):15457–62. DVL3 expression and suppressing Wnt/beta-catenin signaling. Int 74. Harbury PB, et  al. A switch between two-, three-, and four- J Mol Med. 2015;35(6):1587–95. stranded coiled coils in GCN4 leucine zipper mutants. Science. 52. Miki K, et  al. Glyceraldehyde 3-phosphate dehydroge- 1993;262(5138):1401–7. nase-S, a sperm-specific glycolytic enzyme, is required for 75. Szenker E, Ray-Gallet D, Almouzni G. The double face of the sperm motility and male fertility. Proc Natl Acad Sci U S A. histone variant H3.3. Cell Res. 2011;21(3):421–34. 2004;101(47):16501–6. 76. Schlesinger S, et al. A hyperdynamic H3.3 nucleosome marks 53. Sabari BR, et al. Metabolic regulation of gene expression through promoter regions in pluripotent embryonic stem cells. Nucleic histone acylations. Nat Rev Mol Cell Biol. 2017;18(2):90–101. Acids Res. 2017;45(21):12181–94. 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3193 77. Chen P, et al. H3.3 actively marks enhancers and primes gene 80. Schon SB, et  al. Histone modification signatures in human transcription via opening higher-ordered chromatin. Genes Dev. sperm distinguish clinical abnormalities. J Assist Reprod Genet. 2013;27(19):2109–24. 2019;36(2):267–75. 78. Loyola A, et  al. PTMs on H3 variants before chromatin assembly potentiate their final epigenetic state. Mol Cell. Publisher’s note Springer Nature remains neutral with regard to 2006;24(2):309–16. jurisdictional claims in published maps and institutional affiliations. 79. Rogers RS, et al. SUMO modified proteins localize to the XY body of pachytene spermatocytes. Chromosoma. 2004;113(5):233–43. 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Assisted Reproduction and Genetics Springer Journals

A novel posttranslational modification of histone, H3 S-sulfhydration, is down-regulated in asthenozoospermic sperm

Loading next page...
 
/lp/springer-journals/a-novel-posttranslational-modification-of-histone-h3-s-sulfhydration-id9zhTRAq2

References (85)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
ISSN
1058-0468
eISSN
1573-7330
DOI
10.1007/s10815-021-02314-x
Publisher site
See Article on Publisher Site

Abstract

Oxidative stress is one of the major causes leading to male infertility including asthenozoospermia. Hydrogen sulfide (H S) has been widely recognized to be a potent antioxidant whose role is partially implemented by protein S-sulfhydration. How- ever, protein S-sulfhydration has not been reported in germ cells. Therefore, we investigated whether asthenozoospermia could be associated with sperm protein S-sulfhydration. S-sulfhydrated proteins in human sperm were enriched via biotin- switch assay and analyzed using LC-MS/MS spectrometry. Two hundred forty-four S-sulfhydrated proteins were identified. Importantly, we validated that sperm histones H3.1 and H3.3 were the S-sulfhydrated proteins. Their S-sulfhydrated amino acid residue was Cysteine111. Abundances of S-sulfhydrated H3 (sH3) and S-sulfhydrated H3.3 (sH3.3) were significantly down-regulated in asthenozoospermic sperm, compared with the fertile controls, and were significantly correlated with pro - gressive motility. Retinoic acid (RA) up-regulated level of sH3.3 in primary round spermatids and the C18-4 cells (a mouse spermatogonial stem cell line). Overexpression of the mutant H3.3 (Cysteine111 was replaced with serine) affected expression of 759 genes and raised growth rate of C18-4 cells. For the first time, S-sulfhydration H3 and H3.3 were demonstrated in the present study. Our results highlight that aberrant S-sulfhydration of H3 is a new pathophysiological basis in male infertility. Keywords S-sulfhydrated proteome · H3S-sulfhydration · Asthenozoospermia · Spermatogenesis · H3.3 Abbreviations ACRBP Acrosin binding protein ANKRD1 Ankyrin repeat domain 1 ASTH Asthenozoospermic Mr. Wang and Mr. Li both are the responding authors of this paper. ATP5A ATP synthase F1 subunit alpha Biotin-HPDP N-(6-(biotinamido) * Runsheng Li hexyl)-3′-(2′-pyridyldithio)-propionamide runshengli2007@163.com CABLES1 Cdk5 and Abl enzyme substrate 1 Jian Wang CCND Cyclin D wangjiansippr@126.com CENP-A Centromere protein A NHC Key Laboratory of Reproduction Regulation dpp Days postpartum (Shanghai Institute of Planned Parenthood Research), DEG Differentiall y expressed genes Pharmacy School, Fudan University, 2140 Xietu Road, DVL3 Dishevelled segment polarity protein 3 Shanghai 200032, China eST elongating/condensed spermatids NHC Key Laboratory of Reproduction Regulation (Shanghai GAPDH Glyceraldehyde-3-phosphate Institute of Planned Parenthood Research), 2140 Xietu Road, dehydrogenase Shanghai 200032, China GDNF Glial cell–derived neurotrophic factor State Key Laboratory of Genetic Engineering, Institute GO Gene ontology of Biostatistics and Computational Biology, School of Life Sciences, Fudan University, Shanghai 200438, China H O Peroxide 2 2 HS Hydr ogen sulfide NHC Key Laboratory of Reproduction Regulation (Shanghai Institute of Planned Parenthood Research), School H3 Histone3 of Life Sciences, Fudan University, 2140 Xietu Road, HirA1 Hira domain-containing protein Shanghai 200032, China Vol.:(0123456789) 1 3 3176 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 IAA Iodoace tamide motile (PR%) spermatozoa < 32% [3]. The molecular basis IAP Iodoacetyl-polyethylene glycol of asthenozoospermia is largely elusive. IGF-1R Insulin-lik e growth factor 1 receptor Human spermatozoa are extremely vulnerable to oxida- IgG Immunoglobulin G tive attack because they contain little cytoplasm sequestering IP Immunopr ecipitation antioxidants. The oxidative stress (OS)–mediated damage to LC-MS/MS Liquid chromatography-tandem mass sperm has been considered as one of the leading causes for spectrometry male infertility [4]. Defective mouse spermatogenesis can LDHA Lactate dehydrogenase A be caused by knockout of antioxidative genes, leading to an MMTS Methyl-methanethiosulfonate excessive production of reactive oxygen species (ROS) [5]. NORM Normozoospermic Asthenozoospermic sperm expressed some down-regulated ODF1/2 Outer dense fiber protein 1/2 antioxidative genes [6]. OS is accepted as the target for clini- OS Oxidativ e stress cal treatment of asthenozoospermia [7]. PTM Posttranslational modifications Comparative proteomic analysis widely revealed altered RA R etinoic acid expression of some proteins in asthenozoospermic sperm rST Round spermatids [8–10], indicating that sperm with poor motility might be RNA-seq RNA sequencing caused by the lower expression of tubulin with structural ROS Reactive oxygen species defects in sperm flagellum. An altered expression of his- SDS-PAGE SDS–polyacrylamide gel electrophoresis tone was detected in asthenozoospermic sperm by proteomic sH3.3 S-sulfhydrated H3.3 study [11]. Additionally, the aberrant expression of PTMs of sH3 S-sulfhydrated H3 proteins including phosphorylation [12], sumoylation [13], SSC Spermatogonial stem cells glutarylation [14], and hydroxyisobutyrylation [15] was Wisp1 WNT1 inducible signaling pathway pro- associated with poor sperm motility. Interestingly, charac- tein 1 terization of human sperm lysine acetylproteome revealed that protein acetylation was essential for sperm motility [16]. Study of the global protein phosphorylation land- Introduction scape of spermiogenesis showed wide phosphoregulation across a diverse range of processes during spermiogenesis Mammalian spermatogenesis, a precisely regulated [17]. However, these proteomic studies did not show any developmental process generating sperm, consists of specific protein whose function was actually regulated by three distinct phases. The first phase refers to the mitotic these PTMs in germ cells. Additionally, the links between division of spermatogonia physiologically resulting in ROS and the PTMs of proteins have not been established in accumulation of germ cells depending on renewal and asthenozoospermic sperm. differentiation of spermatogonial stem cells (SSC). The Hydrogen sulfide (H S) exerted a wide range of physi- second stage is meiosis, in which spermatocytes undergo ological and cytoprotective functions in the biological sys- two rounds of mitosis to produce haploid spermatids. The tems via its potent antioxidative capability [18]. The asthe- final one is spermiogenesis, wherein the round spermatids nozoospermic patients exhibited decreased concentration of (rST), known as haploid spermatids, undergo a complex H S in their seminal plasma, while supplying exogenous H S 2 2 differentiation process to develop into spermatozoa, to semen improved sperm motility of the asthenozoospermic including chromatin remodeling, nuclear elongation, and patients [19]. However, the mechanism remains unknown flagellum development. A hallmark of mammalian sperm is regarding how H S exerts its roles in germ cells. Signal- the highly compact and condensed structure of chromatin, ing by H S has been widely found in eukaryotic cells via in which depending on the species, approximately 90–99 protein S-sulfhydration [20–22], a PTM on thiol group of % of histones are replaced by protamines [1]. Distinct cysteine residues that converts Cys-SH to Cys-SSH. About posttranslational modifications (PTMs) of histones in 10–25% of proteins extracted from liver are S-sulfhydrated spermatogenesis are currently accepted to facilitate the in physiological conditions [20]. An accumulating number chromatin remodeling and histone-to-protamine transition of proteins have been recently validated to be S-sulfhydrated [1, 2]. Impaired spermatogenesis causes male infertility. proteins with the identified sulfhydrated cysteine residues, Around 15% of couples at reproductive age present with and their sulfhydration has wide and important functions infertility, and about half of the infertility are associated including regulating redox balance [21, 23]. However, pro- with male partner. Asthenozoospermia, a common male tein S-sulfhydration has not been reported in germ cells. We infertility, is characterized by both reduced sperm motility revealed S-sulfhydrated proteome consisting of 244 proteins and normal concentrations of sperm (>15 million per of human sperm in the present study. They included most matozoa/ml), and is defined as percentage of progressively of ROS-associated human sperm reported elsewhere [24]. 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3177 Importantly, for the first time, we demonstrated that histone Direct swim‑ up of spermatozoa H3 was an S-sulfhydrated protein in the present study. We further observed that the level of sH3 and sH3.3, a H3 vari- The direct swim-up of spermatozoa from semen was per- ant, was positively correlated with percentage of sperm with formed according to the “WHO Laboratory Manual” (5th progressive motility, respectively. Our findings highlighted a edition). Briefly, place 2 ml of liquefied semen in a sterile novel pathophysiological basis for asthenozoospermia. 15-ml conical centrifuge tube, and gently layer 2 ml of mod- ified HTF medium (#ART-1023, SAGE) with 5% human serum albumin solution (#10064, Vitrolife) over it. After Experimental procedures sperm were incubated at 37 °C with 5% C O in humidified air for 1 h, gently collected the uppermost 1 ml of medium Reagents, cells, and mice containing highly motile sperm cells and 1 ml semen at bot- tom of the tube, and used them as sperm with high motility Anti-H3.3antibody(#ab176840), anti-H3 antibody(#ab1971), and low motility, respectively. Centrifuged at 500g for 5 min anti-biotin antibody(#ab1227), anti-H3K4me3(#ab185637), and discarded the supernatant. The pellets were used for and anti-H3K9me3(#ab8898) were purchased from Abcam preparation of lysates. (USA). GDNF was purchased from Peprotech (#450-44, USA). Male C57BL/6J mice were purchased from SIPPR- Biotin‑ switch assay BK Animal Company (Shanghai, China). The C18-4 cells was kindly provided by Prof. Zuping S-sulfhydrated proteins in human spermatozoa were detected He, who was a principal investigator in Renji-Med X Clini- using biotin-switch assay as described previously with minor cal Stem Cell Research Center, Renji Hospital, School of modifications [ 20]. Briefly, 20 million spermatozoa were Medicine, Shanghai Jiao Tong University. The C18-4 cells centrifuged at 2000 g for 5 min and supernatant removed, were grown as described by He et al. [25]. and spermatozoa were then resuspended in the lysis buffer (250 mM HEPES pH 7.7, 1 mM EDTA, 0.1 mM neocu- proine, 1% Triton, 2.5% SDS) added with a cocktail of pro- Semen sample collection tease inhibitors (Sigma). The lysates were incubated for 5 min at room temperature and then centrifuged at 2000 g This study (PJ2019-05) was approved by the Ethics Commit- for 5 min. The supernatant was collected, and its protein tee of Shanghai Institute of Planned Parenthood Research/ concentration was adjusted to less than 0.5 mg/ml in each World Health Organization (WHO) Collaborating Center sample. Proteins were then precipitated using 4 volumes of on Human Research. Written informed constructs were ice-cold acetone for 20 min at −20°C, centrifuged at 2000 obtained from the semen donors involved in the study. All g for 5min at 4 °C, washed twice with 70% acetone, and the methods used in the present study were performed in dried out. The pellets were resuspended in HEN medium accordance with the Declaration of Helsinki. The donors (250 mM HEPES pH 7.7, 1 mM EDTA, 0.1 mM neocu- were recruited in compliance with the “WHO Laboratory proine) containing 2.5% SDS. Free thiols of proteins were Manual for the Examination and Processing of Human blocked with a rapidly thiol-reactive agent MMTS (20 mM) Semen” (Fifth edition). Semen samples were obtained by (#23011, Thermo Scientific, CHE) for 30 min at 50 °C. After masturbation after 3–5 days of sexual abstinence. Semen the reaction, the proteins were precipitated with acetone as samples which contained leukocytes were excluded from our described above in order to remove excess MMTS, resus- study. Spermatozoa motility was assessed by the computer- pended in HEN medium containing 1% SDS. Next, the pro- assisted sperm assay (CASA) method according to World teins solution was added with 1mM biotin-HPDP (#A8008, Health Organization guidelines, equipped with a camera APExBIO, USA) and incubated for 1 h at 25 °C to achieve (acA780-75gc, Basler, Germany), and a 20-fold objective, a biotinylation. The biotinylated proteins were separated by camera adaptor (Eclipse E200, Nicon, Japan), operated by an SDS-PAGE and finally detected with anti-biotin antibody or SCA sperm class analyzer (MICROPTIC S.L.). Finally, the the indicated antibodies in Western blotting analysis. semen samples were collected from 26 normozoospermic men (24–45 years old, mean ± SEM: 31.54 ± 5.04 years old) Cysteinyl labeling assay and 24 asthenozoospermic patients (24–39 years old, mean ± SEM: 32.25 ± 4.78 years old) (Table S3), and used in the We also detected S-sulfhydrated proteins in human sperma- present study. The normozoospermic men had known repro- tozoa using cysteinyl labeling assay as described elsewhere ductive histories in the past 2 years and progressive motility [26] with minor modifications. Briefly, 20 million sper - ≥32%, while the asthenozoospermic men had a progressive matozoa were lysated in lysis buffer. The lysate was added motility <32% (Table S3). with 2mM IAA for 1 h at room temperature. Cold acetone 1 3 3178 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 of double volume was then added into the sample. Then, Identification of the sulfhydrated amino acid the samples were precipitated at −20 °C for 20 min. After residue in H3.1 and H3.3 centrifuged at 12000 rpm at 4 °C for 10 min, the precipita- tion was diluted in HEN buffer with 1 mM DTT at room All the constructs that expressed wild-type and mutant temperature for 30 min. A total of 3 mM biotinylated IAP H3.3 and H3 were purchased from Genomeditech (Shang- was next added into the sample at room temperature for 1 h. hai, China). H3.1 was mutated in the three ways: C97 was Biotinylated proteins were enriched by using streptavidin- replaced with serine (C97S), C111 was replaced by ser- Sepharose beads for 16 h at 4 °C on a rotating wheel, with ine (C111S), and both of C111 and C97 were mutated sequential rounds of centrifugation (12,000 g, 1 min, 4 °C) to serine (double mutations). H3.3 was mutated at C111, using PBS to wash the beads. The beads were resuspended which was also replaced with serine. All the expression in 20 μl of 4×Laemmli sample buffer and heated at 90 °C constructs were generated based on pCMV2-FLAG tag for 1 min. The biotinylated proteins were separated by SDS- (Promega, USA). PAGE and finally detected with anti-biotin antibody or the Transfection of the C18-4 cells with the H3 expression indicated antibodies in Western blotting analysis. constructs via Lipofectamine 3000 (Invitrogen, Shanghai, China) was carried out according to the manufacturer’s pro- Analysis of immunoprecipitation tocol. Forty-eight hours later, the cells were harvested for preparation of lysate using the above lysis buffer. Wild-type Proteins from sperm or the C18-4 cells were biotinylated as and mutant H3.1 or H3.3 were immunoprecipitated using described above. Protein concentration was adjusted to 0.1 anti-FLAG antibody, and next subjected to biotin-switch mg/ml using HEN/10 media (10× dilution of HEN) contain- assay. The biotinylated H3 was enriched via the IP using ing 1% SDS. Three volumes of neutralization buffer (20 mM anti-biotin antibody, and finally detected using anti-FLAG HEPES pH 7.7, 100 mM NaCl, 1 mM EDTA and 0.5% Tri- antibody in Western blotting assay as described above. ton X-100) were added. The mixture was separated into two parts. One part was added 2× SDS sample buffer for load- ing control, while another was incubated overnight at 4 °C Primary germ cell preparation and treatment of rST with a specific antibody (1:1000) and 50 μl of protein A/G with RA (#ab193262, Abcam, USA) per ml. Beads were previously washed twice with the neutralization buffer and centrifuged SG cells were isolated from 8 days postpartum (dpp) mice at 200 g for 10 s. Once the incubation terminated, the beads [27]. The method of STA-PUT was used to isolate pacSC, were washed 5 times with 500 μl of Wash buffer (the neu- rST, and eST. They were characterized as previously tralization buffer containing 600 mM NaCl). Proteins were described [27, 28]. pacSC were from 17 dpp mice. rST and eluted with 1× SDS sample buffer containing 2 mM DTT. eST were from 56–70 dpp mice. After separated via grav- Samples were boiled for 5 min at 100 °C and centrifuged at ity sedimentation, rST were pelleted via a centrifugation at 14000 g for 5 min. The supernatant was collected, separated 500 g for 5 min, cultured in DMEM (10% PBS) medium by SDS-PAGE (10%), and detected via Western blotting with 2 mM L-glutamine, 100 units/ml penicillin, and 100 analysis or silver staining. The gel with silver staining was mg/ml streptomycin at 37 °C with 5% CO in humidified excised, and applied for proteomic analyses which were per- air. Three hours later, RA (#r2625, Sigma-Aldrich, USA) formed as described below. For each of these experiments, 3 diluted in ethanol was added to the culture medium to make fertile ejaculates (from different donors) were pooled. a final concentration of 0.3 μM or 1 μM. Twenty-four hours later, the cells were harvested for measurement of sH3.3 Silver staining expression. After proteins were separated by SDS-PAGE, whole gel was washed with water for 5 min, and then soaked into blocking Treatment of testis with NaHS buffer (50% ethanol, 8% acetic acid, 0.4% formaldehyde) for 2 h. After washed with 35% ethanol for 3 times, the gel Mice aged 4 weeks were sacrificed by cervical dislocation was soaked into staining buffer (10 mg/ml silver nitrate, and then put them into 75% ethanol. The abdominopelvic 0.4% formaldehyde) for 30 min. After washed with water cavity was opened using sterile scissors and forceps, and for 2 times, it was then soaked into cultivating buffer (0.12 then the testis was pulled out. The testicular tunica albuginea g/ml sodium carbonate, 0.4% formaldehyde) until bands was removed by puncturing the tissue and the loose seminif- appeared. Finally, the gel was placed into termination buffer erous tubules were collected. The tubules were transferred (50% ethanol, 8% acetic acid) for 5 min. The PAGE was to a new petri dish containing 2 ml of DMEM/F12 with scanned by Tanon scanner 5200. gentamicin (0.02 g/l; Sigma-Aldrich). The seminiferous 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3179 tubules were cut up, dispersed, and evenly distributed into min over 60 min. The gradient was set as the following: 3 dishes in a humidified atmosphere at 34°C with 5% CO . 5–8% buffer B from 0 to 2min, 8 to 23% buffer B from Three hours later, NaHS (50 or 100 mM; Sigma-Aldrich) 2 to 42 min, 23 to 40% buffer B from 42 to 50 min, 40 was added to the medium. After 24-h treatment, the pieces to 100% buffer B from 50 to 52 min, 100% buffer B kept of testis were collected for measurement of sH3.3 expression until 60 min. MS data were acquired using a data-depend- ent top20 method dynamically choosing the most abun- Infection of the C18‑ 4 cells, cell number dant precursor ions from the survey scan (350–1800m/z) determination, qRT‑ PCR, and Western blotting for HCD fragmentation. A lock mass of 445.120025 Da assay was used as internal standard for mass calibration. The full MS scans were acquired at a resolution of 60,000 at The recombinant Lentivirus that expressed the chimeric m/z 200, and 15,000 at m/z 200 for MS/MS scan. The human H3.3 protein (wild-type or the mutant H3.3 with maximum injection time was set to 50 ms for MS and 45 C111S) with a FLAG tag in the N-terminal of H3.3 was ms for MS/MS. Normalized collision energy was 27 and purchased from Kangchen Bio-tech (Shanghai, China). The the isolation window was set to 1.5 Th. Dynamic exclu- ectopic expression of H3.3 and the mutant H3.3 by viral sion duration was 30 s. infection in the C18-4 cells were performed according to the manufacturer’s instructions. Cell number determination, Database search qRT-PCR, and Western blotting assay were carried out as described previously [29]. The MS data were analyzed using MaxQuant software ver- sion 1.5.8.3. MS data were searched against the UniProtKB Mass spectrometry experimental design Human database (157600 total entries, downloaded in July, and statistical rationale 2017). The trypsin was selected as digestion enzyme. The maximal two missed cleavage sites and the mass tolerance Sample preparation of 4.5 ppm for precursor ions and 20 ppm for fragment ions were defined for database search. Carbamidomethyla - Human sperm lysate was prepared from ten pooled sperm tion of cysteines was defined as fixed modification, while samples and separated by SDS-PAGE. Gel pieces were cut, acetylation of protein N-terminal and lysine and oxidation destained for 20 min in 100 mM NH HCO with 30% ace- 4 3 of methionine were set as variable modifications for data - tonitrile, and washed with Milli-Q water until the gels were base searching. The database search results were filtered and fully destained. The spots were then lyophilized in a vacuum exported with <1% false discovery rate (FDR) at peptide centrifuge. The in-gel proteins were reduced with dithio- level and protein level, respectively. threitol (10 mM DTT/100 mM NH HCO ) for 30 min at 56 4 3 ° C, then alkylated with iodoacetamide (200 mM IAA/100 Protein structure modeling mM NH HCO ) in the dark at room temperature for 30 min. 4 3 Gel pieces were briefly rinsed with 100 mM NH HCO3 and The three-dimensional structure of H3.3 in nucleosome ACN, respectively. Gel pieces were digested overnight in was generated using data from human nucleosome struc- 12.5 ng/μl trypsin in 25 mM NH HCO . The peptides were 4 3 ture containing H3.3 (PDB ID: 5X7X) at 2.18 Å generated extracted three times with 60% ACN/0.1% TFA. The extracts by PyMOL-1.5.0.3. were pooled and dried completely by a vacuum centrifuge. LC-MS/MS RNA sequencing and analysis The peptide of each sample was desalted on C18 Car- Total RNA was extracted from mutated and normal mice tridges (Empore™ SPE Cartridges, Sigma), then con- sample by RNeasyPlus Micro Kit (Qiagen, Wetzlar, Ger- centrated by vacuum centrifugation and reconstituted in many) following manufacturer’s instructions and reverse 10 μl of 0.1% (v/v) formic acid. MS experiments were transcribed into cDNA libraries using the Ovation® RNA- performed on a Q ExactiveHF mass spectrometer that was Seq System V2 kit (NuGEN). Samples were sequenced coupled to Easy nLC (Thermo Scientific). Peptide was with paried-ends reads (PE150) using IlluminaHiseq X-ten first loaded onto a trap column (100 μm×20 mm, 5 μm, platform. The QC (quality control) analysis of the RNA C18) with 0.1% formic acid, then separated by an analyti- sequencing data was performed using FastQC. The raw cal column (75 μm×100 mm, 3 μm, C18) with a binary sequencing reads were pre-processed as follows: (1) remov- gradient of buffer A (0.1% formic acid) and buffer B (84% ing adapter sequences, (2) removing reads with over 20 bp acetonitrile and 0.1% formic acid) at a flow rate of 300 nl/ of low quality (Phred quality score < 20). The filtered clean 1 3 3180 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 reads were aligned to mouse reference genome (mm10) dehydrogenase (GAPDH) [31], and L-lactate dehydrogenase using Tophat2 and then the uniquely mapped reads were A chain (LDHA) [34] were elsewhere reported. In addition, assigned to each annotated gene using featureCount. Sta- we observed S-sulfhydration of Outer dense fiber protein 1 tistical significant test of differentially expressed genes was (ODF1) and ODF2, two proteins that stabilize the axoneme performed by NOISeq with R [30]. Genes with absolute to maintain sperm motility [35]. Interestingly, we detected log2-transformed fold changes greater than 2 were regarded that sperm acrosin and acrosin-binding protein (ARCBP), as differentially expressed genes and a threshold of p value which is needed for biogenesis of acrosome [36], was also < 0.05 was used. For significant DE genes, GO term path- S-sulfhydrated in sperm (Table S1), suggesting that the post- way enrichment analysis was performed using the DAVID translational modification is important for successful ferti- functional annotation tool. lization. Gene ontology (GO) analysis of the sulfhydrated proteome indicated that proteins were ontologically enriched for a series of functional clusters whose top three are oxida- Results tion-reduction process, binding of sperm to zonapellucida, and tricarboxylic acid cycle (Fig. 1C). Analysis of sperm protein S‑sulfhydration Expression of 74 sperm proteins was associated with a high level of ROS in seminal ejaculates [24]. We analyzed No specific antibody recognizing S-sulfhydrated proteins the relationship of the ROS-associated sperm proteome with has been reported. Biotin-switch assay has been widely our S-sulfhydrated sperm proteome. The results showed used for analysis of S-sulfhydrated proteins [20, 31, 32], that 75.7% (56/74) of ROS-associated proteins were S-sulf- in which free thiols (-SH) of proteins were blocked by a hydrated proteins (Fig. 1D), implying that altered protein highly specific free sulfhydryl-reactive compound, methyl- S-sulfhydration is the way for spermatozoa to respond to OS. methanethiosulfonate (MMTS), which did not interact with sulfhydrated thiols (-SSH) or any other forms of oxidized Levels of sH3 and sH3.3 in asthenozoospermic thiols (S-S, for example). The sulfhydrated thiols were then spermatozoa selectively labeled with N-(6-(biotinamido)hexyl)-3′-(2′- pyridyldithio)-propionamide (biotin-HPDP), a compound Spermatozoa histone H3 was identified as one of the that interacts with sulfhydrated thiols in this assay, so that S-sulfhydrated proteins (Fig. 1B). To further validate that the sulfhydrated proteins were biotinylated. In order to H3 is an S-sulfhydrated protein, the crude sperm extract examine protein S-sulfhydration in sperm, the extracts from prepared from five pooled sperm samples from fertile men mouse and human sperm were applied for the biotin-switch was applied for biotin-switch assay. The sulfhydrated pro- assay. The S-sulfhydrated proteins were enriched by immu- teins were immunoprecipitated using anti-biotin antibody. noprecipitation (IP) using biotin antibody, separated via The IP was further analyzed with H3 antibody in Western SDS–polyacrylamide gel electrophoresis (SDS-PAGE), and blotting assay. The result showed that H3 was detected in finally detected by anti-biotin antibody in Western blotting the IP (Fig. 2A), indicating the presence of S-sulfhydrated assay (Fig. 1A, left) and via silver staining (Fig. 1A, right), H3 (sH3) in the extract. Additionally, the validation also respectively. The results showed that sperm S-sulfhydrated started with enrichment of H3 via IP of the crude sperm proteins were detected both via Western blotting assay and extract using H3 antibody. The IP was next analyzed in silver staining. biotin-switch assay, and applied for Western blotting assay To identify S-sulfhydrated proteins in human sperm, the using anti-biotin antibody. The result showed that H3 was S-sulfhydrated proteins stained with silver were subjected recognized by anti-biotin antibody (Fig. 2A), indicating that to in-gel trypsin digestion, next analyzed through liquid sperm H3 is S-sulfhydrated. chromatography-tandem mass spectrometry (LC-MS/MS) The presence of S-sulfhydration of H3 was addition- followed by protein database searching of the acquired spec- ally evaluated by with cysteinyl labeling assay that utilizes tra. To control for nonspecific IP, IgG preimmune complex a biotinylated iodoacetic acid (IAA) probe, which reacts was also analyzed. The experiments were performed in three through a nucleophilic substitution of the halide group by replicates, and each replicate was run through LC-MS/MS the H3-reactive thiol group, resulting in a stable thio-ether three times. Two hundred forty-four proteins were identified bond [26] (Fig S1A). We detected the presence of H3 after (Table S1) in the IP complex after (1) removing proteins that the lysate was applied for cysteinyl labeling assay in anti- were not found in the replicate experiments and (2) subtract- H3 antibody-based immunoblotting analysis (Fig S1B). ing common proteins that were found in the IgG preimmune Similarly, H3 was detected by anti-biotin antibody in the complex. WB assay, after the H3 was enriched from sperm lysate via In the list of proteins, S-sulfhydration of ATP synthase IP, and next treated in cysteinyl labeling assay. Again, the subunit alpha (ATP5a) [31, 33], glyceraldehyde-3-phosphate results demonstrated that sperm H3 is a sulfhydrated protein. 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3181 Fig. 1 The S-sulfhydrated proteome of human sperm. (A) Detection (right). (B) LC-MS/MS of a subset of the S-sulfhydrated proteins in of S-sulfhydrated proteins in human and mouse sperm. The lysates (A) identifies top notable S-sulfhydrated proteins, including GAPDH, were prepared from pooled samples of human sperm (from ten fer- GSTM3, and H3. (C) Biological processes enriched in the S-sulf- tile men) and mouse sperm (from three male mice), next subjected to hydrated proteins of human sperm. (D) Vann diagram depicting the the modified biotin-switch assay. The numerous sulfhydrated proteins relationship of sulfhydrated proteins and ROS-associated proteins of were detected with anti-biotin antibody (left) or with silver staining human sperm. The presence of S-sulfhydration of H3 was additionally through a nucleophilic substitution of the halide group evaluated by with cysteinyl labeling assay that utilizes a by the H3-reactive thiol group, resulting in a stable thio- biotinylated iodoacetic acid (IAA) probe, which reacts ether bond [26] (Fig S1A). We detected the presence of 1 3 3182 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3183 ◂Fig. 2 Association of sH3 and sH3.3 with asthenozoospermic sperm. Expressions of sH3 and S-sulfhydrated H3.3 (sH3.3) (A) The validation of sH3.S-sulfhydrated proteins of human sperm were measured in sperm with different motility. H3.3, a lysates were biotinylated via biotin-switch assay, enriched via the IP H3 variant played an important role in spermatogenesis based on anti-biotin antibody, and detected in Western blotting anal- [37], was thus selected in the analysis as well. H3 and H3.3 ysis using anti-H3 antibody (top panel). H3 in the lysate of pooled human sperm (n=5) was immunoprecipitated via anti-H3 antibody, in sperm lysate were enriched via their antibodies-based and subjected to biotin-switch assay. The S-sulfhydrated H3 was IPs, respectively. The IPs were divided into two portions: detected in Western blotting analysis using anti-biotin antibody (bot- one was applied for biotin-switch assay, and followed by tom panel). (B) Detection of overall S-sulfhydrated proteins in sperm an immunoblotting analysis using biotin antibody; the with different motility. Subpopulation of sperm with high and low motility from three fertile men was separated via swim-up assay, other was directly used as the loading control in Western respectively, and their protein lysates were subjected to biotin-switch blotting assay. Our results showed that abundances of assay, detected with anti-biotin antibody in Western blotting assay sH3 and sH3.3 were significantly higher in sperm with (left panel). As a loading control, the proteins in the sperm lysates high mobility than with low mobility (Fig.  2C). We stained with Coomassie blue after separated via SDS-PAGE (right panel). (C) Expression of sH3 and sH3.3 in sperm with different also measured levels of other PTMs like H3K4me3 and motility. The biotinylated proteins in (B) were immunoprecipitated H3K9me3 in the two subpopulations of normozoospermic with biotin antibody, and further analyzed with anti-H3 and anti-H3.3 sperm samples. However, no significant difference was antibodies in Western blotting assay. In addition, the biotinylated detected. Together, these results indicated that expression protein in (B) from sperm with different motility was subjected in Western blotting assay using the indicated antibodies. The experi- of S-sulfhydration of proteins including sH3 and sH3.3 is ments were replicated in three fertile individuals. Error bar denotes positively associated with sperm motility. mean ± SEM. *P < 0.05 and **P < 0.01. (D) H3 and H3.3 were Association of sH3 and sH3.3 with asthenozoospermia enriched via IP with their antibodies from lysates of clinical sperm was next addressed. Levels of sH3 were measured in semen samples with different progressive motility (PR%), as indicated, next subjected to biotin-switch assay. One portion of the biotinylated pro- samples with different percentage of progressive motility teins, as a loading control, were analyzed using anti-H3 and anti-H3.3 (PR%) in Western blotting assay. Lower mean levels of antibodies in Western blotting assay, while another was analyzed sH3 and sH3.3 were found in asthenozoospermic men for detection of sH3 or sH3.3 using biotin antibody in Western blot- compared with fertile men (Fig. 2D). Statistical analysis ting assay. (E) Relative expression of sH3 and sH3.3 in asthenozoo- spermic (ASTH) sperm (N=19 for sH3; N=24 for sH3.3) compared showed that the mean levels of sH3 in patients (n=19) were with the normozoospermic (NORM) controls (N=16 for sH3; N=26 53.1% of that in the fertile controls (n=16). Additionally, for sH3.3). Error bar denotes mean ± SEM. ***P < 0.001. (F) Cor- mean levels of sH3.3 in patients (n=24) were only 42.3% relations among sperm sH3 (n=35), sH3.3 (n=50), and progressive of that in the fertile controls (n=26) (Fig. 2E). Correlation motility were analyzed by linear regression. (G) Effects of H O and 2 2 H Son expression of human sperm sH3.3. The cultured sperm was analysis results revealed that sperm levels of sH3 and sH3.3 added with indicated concentration of H O and NaHS, respectively, 2 2 correlated positively with progressive motility (Fig. 2F). and 1 h later, subjected to analysis of sH3.3 expression. The analy- Collectively, our study demonstrated that expression of sis represented one of three independent experiments with almost the sH3 and sH3.3 was down-regulated in asthenozoospermic same results. sperm. The effect of sperm redox status on levels of sH3 was H3 after the lysate was applied for cysteinyl labeling next addressed via treating sperm with the oxidative agent assay in anti-H3 antibody-based immunoblotting analysis hydrogen peroxide (H O ) and NaHS, which has been 2 2 (Fig S1B). Similarly, H3 was detected by anti-biotin widely used as a H S donor in culture. H S was shown to 2 2 antibody in the WB assay, after the H3 was enriched from be an oxidants scavenger in sperm [19]. The results showed sperm lysate via IP, and next treated in cysteinyl labeling that H O reduced the level of sH3 in a dose-dependent way 2 2 assay. Again, the results demonstrated that sperm H3 is a (Fig. 2G). By contrast, NaHS raised the level of sH3 in a sulfhydrated protein. dose-dependent way. These results indicated that the level We next studied the association of level of of sperm sH3 is under the control of redox status in a cel- S-sulfhydrated protein with sperm motility. lular context. Normozoospermic sperm subpopulations with high motility and low motility were separated via the Dynamics of sH3 and sH3.3 in spermatogenesis “swim-up” assay. The S-sulfhydrated proteins in their lysates were enriched using biotin antibody after biotin- We next approached the dynamic level of H3 S-sulfhydra- switch assay, next detected with biotin antibody in tion during mouse spermatogenesis. Germ cells at differ - Western blotting assay. As shown in Fig. 2B, a number ent phases of spermatogenesis were first isolated, includ - of S-sulfhydrated proteins (around 15–100 KDa) were ing spermatogonial cells (SG), pachytenespermatocytes observed in the two sperm subpopulations. The overall (pacSC), which are at the prophase of the first meiotic divi - level of S-sulfhydrated proteins was higher in sperm with sion, round spermatids (rST), and elongating/condensed high motility than in sperm with low motility. spermatids (eST). Both rST and eST are haploid germ cells. 1 3 3184 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 Fig. 3 Analysis of expression of sH3 and sH3.3 in spermatogenesis sH3.3. (B) The incubated rST were treated with two indicated doses and effect of RA and NaHS on level of sH3 in germ cells. (A) Sper - of RA for 24 h, then subjected for analysis of sH3.3 expression. (C) matognia (SG), pachytene spermatocytes (pacSC), round spermatids The incubated pieces of testis were treated with the indicated con- (rST), and elongating/condensed spermatids (eST) were isolated and centrations of NaHS for 24 h, and next subjected for measurement of characterized as described in “Experimental procedures.” These germ expression of sH3.3. The experiments were repeated independently cells, together with sperm from caput epididymis (SPM (cau)) and for three (for B and C) to four (for A) times. Error bar denotes mean from cauda epididymis (SPM (cau)), were applied for preparation for ± SEM. *P < 0.05 and **P < 0.01. protein lysates, subjected for measurement of expression of sH3 and 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3185 In addition, sperm from caput epididymis (SPM (cap)) and abolished when its C111 was mutated. These results indi- cauda epididymis (SPM (cau)) was isolated, respectively. cated that C111 is the site for S-sulfhydration both in H3.1 Levels of sH3 and sH3.3 in the six types of germ cells were and H3.3. measured (Fig.  3A). These cells were applied for meas- We failed to directly detect S-sulfhydrated peptide urement of sH3 and sH3.3. Our results showed that all the digested from human sperm H3 via MS analysis. We first types of germ cells expressed sH3 and sH3.3. A small but asked the question whether S-sulfhydration H3 could be significant up-regulation of sH3 (by 151.2%) and sH3.3 directly observed by MS analysis. The key is whether the (by 164.7%) was detected in rST, compared with pacSC. peptides containing the sulfhydrated C111 could be iden- Notably, levels of sH3 and sH3.3 were significantly higher tified by MS after trypsin digestion. We next checked the by 3.02 folds and 4.11 folds, respectively, in eST than in PeptideAtlas database (http://www.peptideatlas.org/), rST (Fig. 3A). As a control, expression of H3K4me3 was which is a multi-organism, publicly accessible compen- also measured and found to drop much in eST, compared to dium of peptides identified in a large set of tandem mass rST. No statistically significant difference was observed in spectrometry proteomics experiments, for all possible expression of sH3, and sH3.3 was observed between eST and peptides that could be identified by MS for H3 ( https:// db. SPM (cap). Together, the results indicated spermiogenesis syste msbio logy. net/ sbeams/ cgi/ Pepti deAtl as/ GetPr otein? is the main stage for H3 and H3.3 to be S-sulfhydrated in atlas_ build_ id= 337& prote in_ name= P8424 3& action= spermatogenesis. QUERY). As shown in the above link and in Supplemen- RA is a key physiological factor triggering differentia - tary Fig.  2, the peptide between N-terminal 81 and 137 tion of rST to eST [38]. We next investigated the effect of amino acid residues of H3 belongs to the category which RA on H3.3S-sulfhydration in spermatids. The cultured is unlikely to be identified due to its length. C111 happens rST were treated with RA soon after they were separated to fall into this peptide sequence, and S-sulfhydration of from testis. The results showed that 0.3μM and 1.0 μM RA C111 thus could not be identified directly by MS. raised expression of sH3.3 by 144.0% and 266.2% (Fig. 3B), In order to analyze the effect of S-sulfhydration of H3.3 respectively. Together, these results indicated that S-sulfhy- on a nucleosomal structure, the structure of human H3.3- dration of H3.3 is induced by RA in rST. containing nucleosome was next stimulated based on its We next studied whether H3.3 was susceptible to crystal structure solved at 2.8 Å [PDB ID: 3AV2] [40]. H S-induced sulfhydration in a testicular context. The pieces As shown in the nucleosome contains two H3.3 (Fig. 4C of testis were incubated and treated with NaHS. The results and D), the two C111 are located in the alpha helix (85- showed that sH3.3 expression was raised significantly upon 114) of H3.3, facing each other. The range between two treatment with the H S donor (Fig.  3C), suggesting that sulfur atoms is 6.31 Å (Fig. 4D). Given an average dis- H3.3 S-sulfhydration is under the control of H S signaling tance of 2.04 Å between the two sulfur atoms which gen- in spermatogenesis. erates a disulfide bond [ 41], it is still too far for the two cysteines to form a disulfide bond. However, when both Analysis of S‑ sulfhydrated amino acid residue in H3 of C111 are S-sulfhydrated, the distance between the two outer sulfur atoms is narrowed to approximately 2.23 Å, At least 6 variants of H3 have been reported (Fig. 4A). The which is probably close enough to form a disulfide bond canonical H3.1 and H3.2 are expressed and deposited on (Fig.  4E). Therefore, our simulation analysis suggested nucleosomes during DNA replication [39]. The expression that S-sulfhydration of H3.3 is beneficial to a formation of H3.1t is testis-specific. Centromere protein A (CENP- of an inter-molecular disulfide bond between two nucleo - A), a highly specialized variant, is only present at the cen- somal H3.3 proteins. tromere. Mammalian H3.3 is expressed throughout the cell On the other hand, our analysis of structure of two cycle, and deposited by a DNA replication–independent H3.3-containing nucleosome [40] also revealed another nucleosome assembly pathway [39]. Only two cysteines, potential effect of S-sulfhydration of C111. The two alpha Cysteine 111 (C111) and Cysteine 97 (C97), are present helixes (85-114) containing two C111 are close to another in H3. two helixes (120-132) of H3.3, which contains two Argi- H3.1 has two cysteine residues. We studied which nine128 (R128) (Fig.  4F). The region around C111 in cysteine mutation could disrupt H3.1 sulfhydration in tran- these four helixes is important to form the H3-H3 hydro- sient transfection assay (Fig. 4B). Serine was used to replace phobic four-helix bundle tetramer interface so as to hold cysteine in the constructs expressing the mutant H3.1. Our together two histone H2A-H2B-H3-H4 tetramers [42]. results showed that the mutation of C111, but not C97, com- Our analysis showed that the two nitrogen atoms of two pletely disrupted H3.1 sulfhydration. H3.3 has only one R128 are very close to two sulfur atoms of two C111, and cysteine, C111. Similarly, H3.3 S-sulfhydration was fully their distance is 3.68~4.60 Å (Fig.  4F). Most of N-H-S hydrogen bonds can form with a distance of 3.25~3.55 1 3 3186 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3187 ◂Fig. 4 Analysis of effect of H3.3S-sulfhydration on structure of recombinant Lentivirus that expressed wild-type H3.3 and nucleosomal H3.3. (A) Sequence alignment of human H3 family. a C111S-containing H3.3, respectively. No significant dif - H3 family only contains two cysteine residues which were framed in ference in the intensity of bands was detected with FLAG the sequences. Note that C111 is conservative in five members of H3 antibody between the two samples (Fig. 5B), indicating that family. (B) Effect of mutation of C97 and C111 on S-sulfhydration of H3.1 and H3.3. The C18-4 cells were transiently transfected with the mutation did not affect H3.3 expression based on virus wild-type and mutant, as indicated, H3.1 and H3.3 expression con- infection. Importantly, we detected that expression of sH3.3 structs, respectively. Forty-eight hours later, the cells were harvested was significantly higher in the C18-4 cells overexpressing for measurement of levels of sulfhydrated wild and mutant H3.1 and wild-type H3.3 than those overexpressing the mutant H3.3 H3.3. H3.1DM: the mutant H3.1 contains both C97S and C111S. (C) The structure of H3.3 in nucleosome, based on crystal structure (Fig. 5B). solved at 2.8 Å [PDB ID:3AV2]. H3.3 (pink) forms as dimer in nucle- We next measured the cell number at different times osome, binding with H2A, H2B, and two H4. Cys111 (blue) locates after the infected and untreated C18-4 cells were seeded. in the center. (D) Range between two sulfur atoms of two C111. C111 We also did not detect any significant difference in cell locates in alpha helixes (85-114, pink), and is surrounded by alpha helix (120-132, orange). The distance between two sulfur atoms of number between C18-4 cells infected with wild-type H3.3- two Cys111 is 6.31Å, as indicated. (E) Simulated structure of H3.3 expressing virus and the untreated C18-4 cells (Fig. 5C). when both Cys111 are S-sulfhydrated. The simulated electron cloud Strikingly, the mutant H3.3-expressing C18-4 cells grew of outer sulfur atoms (red) is overlapped. (F) Distance between two faster than wild-type H3.3-expressing cells. The mutant sulfur atoms of two Cys111 and two nitro atoms of two R128. R128 locates in alpha helix (120-132, orange). The distance between each H3.3-expressing cells were more than the controls, sig- R128 and C111 is 3.68~4.60Å, showed by blue and red line. (G) The nificantly by 25.8% and 36.1%, at the third and fifth day presence of overlapping electron cloud of the four atoms when both after the plating of cells (Fig. 5C). The results strongly C111 are S-sulfhydrated. suggested that sH3.3 is inhibitory to the growth of C18-4 cells, consistent with the repressive effect of GDNF on sH3.3 expression. We next performed RNA sequencing analysis (RNA- Å [43]. When both of C111 are S-sulfhydrated, the dis- seq) in order to study the mechanism underlying the pro- tance between two C111 and two R128 is much closer, so moting effect of C111S of H3.3 on the C18-4 cell growth that the simulated electron cloud of 4 atoms overlaps each rate. We found that expressions of 487 genes were down- other (Fig. 4G). Therefore, these four residues can prob- regulated, while the other 272 genes were up-regulated ably generate multiple inter-helical hydrogen bonds, thus (Fig. 5D, Table S2). Validation by quantitative RT-PCR stabilizing the structure of four-helix bundle tetramer, and was performed for some differentially expressed genes the whole nucleosome. (DEGs), and our quantitative RT-PCR analysis confirmed the RNA-seq data (Fig. 5E). Eec ff t of the C111 mutation of H3.3 on growth rate Introduction of the mutated H3.3 reduced relative and gene expression of C18‑ 4 cells expression of Cyclin D1 (Ccnd1). Ccnd1 expression was inhibitory to growth of SSCs [45]. Rassf8 reduced the The presence of sH3.3 in spermatogonia (Fig.  3A) sug- expression of ccnd1 when overexpressed in SSC [46], gested that sH3.3 could play an important role in the phase and its expression was also unregulated in the presence of mitosis in spermatogenesis. We studied the hypothesis of the mutated H3.3 (Fig.  5E). Therefore, sH3.3 could using SSC C18-4 cell line. Glial cell-line-derived neuro- regulate renewal of SSC via targeting the two genes. trophic factor (GDNF) is bona fide self-renewal factors of Some growth-inhibitory genes including Cdk5 and Abl SSC, and promotes proliferation of C18-4 cells [25]. By con- enzyme substrate 1 (Cables1) [47] and ankyrin repeat trast, RA signaling, which is a key physiological regulator domain 1 (Ankrd1) [48] were shown to be down-regulated of SSC differentiation, is also present in C18-4 cells [ 44]. in the presence of C111S (Fig.  5E). Among the list of The impact of GDNF and RA on sH3.3 expression was next unregulated genes, insulin-like growth factor 1 receptor investigated. GDNF was found to upgrade sH3.3 expres- (IGF-1R) is essential for the proliferation of mouse SSC sion in a time-dependent way (Fig. 5A). A significant rise by promoting the G2/M progression of the cell cycle [49]. in level of sH3.3 was observed as early as 2 h after GDNF Wnt1 inducible signaling pathway protein 1 (Wisp1) is treatment. However, sH3.3 expression was reduced in the required for proliferation of mesenchymal stem cells [50]. C18-4 cells when treated with RA in a time-dependent way. Dishevelled segment polarity protein 3 (Dvl3) repressed These results suggested that an altered sH3.3 expression is differentiation of mesenchymal stem cells inhibited via an important downstream event in signaling of GDNF and up-regulating Ccnd1 [51], and unregulated upon the RA. ectopic expression of the mutant H3.3. Together, the In order to explore the role of sH3.3 in the germ cells, we mutation could promote the growth of C18-4 cells by studied the effect of the H3.3 C111S mutant on growth rate regulating expression of these genes. of C18-4 cells. The C18-4 cells were thus infected with the 1 3 3188 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3189 ◂Fig. 5 Effect of ect opic expression of the mutant H3.3 with C111S on are S-sulfhydrated proteins in human sperm (Fig.  1D). cell growth rate and gene expression in the C18-4 cells. (A) Effects Expression of sH3.3 in male germ cells including sperm of GDNF and RA on sH3.3 expression in the C18-4 cells. Twenty- is unregulated by H S, a potent antioxidant (Figs. 2G and four hours after plating, the C18-4 cells were added with GDNF (50 3C), while sperm sH3.3 was down-regulated by H O ng/ml) and RA (1.0μM), respectively. The cells were harvested for 2 2 measurement of sH3.3 expression at the indicated times after the (Fig. 2G), strongly suggesting that expression of sH3.3 is treatment. Error bar denotes mean ± SEM. *P < 0.05 and **P < under control of redox status. H S raised enzyme activities 0.01. (B) Expression of sH3.3 and H3.3 in the C18-4 cells infected of ATP synthase [32] and LDHA [34] via up-regulation with the recombinant Lentivirus that expressed wild-type and mutant of their S-sulfhydration. Collectively, it is plausible that (C111S) H3.3. The cultured C18-4 cells were harvested for the analy- sis when their confluence reached approximately 80 –90%. These ROS represses H S signaling, which in turn causes hypo- results represented one of three independent experiments with the sulfhydration of proteins including H3/H3.3 in a subtype of similar data. (C) The growth rates of C18-4 cells. Untreated C18-4 asthenozoospermic sperm. Therefore, our study highlights cells and infected C18-4 cells which overexpressed wild-type or that sH3.3/sH3 is potentially a novel biomarker for diag- mutant H3.3 were seeded as described in “Experimental proce- dures,” and harvested for the counting of cell number at the indicated nosing etiology of asthenozoospermia. times after seeding. (D) Volcano plot showing differential expres - To our knowledge, both protein S-sulfhydration in germ sion of protein-coding genes between mutated and normal samples. cells and H3/H3.3 S-sulfhydration have not been reported Red and blue dots indicate significantly down-regulated ( p<0.05 and before. More than ten different PTMs of H3 were elsewhere log2FC<-1) and significantly up-regulated ( p<0.05 and log2FC>1) differential expression, respectively. (E) RNA-seq data validation by reported [53, 54], Therefore, this work extends the catalogue quantitative RT-PCR. Expression of select up-regulated and down- of histone PTM sites in mammalian cells. Oxidative stress regulated genes from the RNA-seq analysis was measured by quan- and ROS are emerging as important players, shaping the epi- titative RT-PCR in the C18-4 cells. (F) Biologic processes that are genetic landscape of the entire genome via different mecha - enriched in genes down-regulated (left) and up-regulated (right). nisms including modification of H3 methylation and acety - lation [55, 56]. Our study strongly suggests that sperm H3 GO analysis on the down-regulated DEGs found many S-sulfhydration is under the control of redox homeostasis, proteins involved in positive regulation of cell death, and unravelling epigenetic mechanisms underlying the patho- positive regulation of apoptotic process that were highly physiology of male infertility. Some interesting questions expressed in C18-4 cells with overexpression of the mutant have emerged from our study. For example, how do ROS- H3.3. Keeping in line with it, some significant terms asso - producing factors including smoking, alcohol, and inflam - ciated with up-regulated mRNAs in the presence of the mation affect sperm H3 S-sulfhydration? It is known that mutant H3.3 were cell cycle, mitotic sister chromatin seg- high levels of ROS can cause male infertility through not regation, and cellular macromolecule biosynthetic process. only by lipid peroxidation or DNA damage but also reduced total antioxidative capability in spermatozoa. What are their relationships with altered H3 S-sulfhydration? A few anti- Discussion oxidant medicines have been used to treat male infertility with different curative effects [ 57]. Can investigation of We reported the human sperm S-sulfhydrated proteome sperm sH3 before and after treatment allow a better under- including 244 proteins in the present study. GO analysis standing, monitoring, or selection of alternative antioxidant suggested that S-sulfhydrated proteins played important medicines? They are issues worth of investigation. roles in spermatogenesis, spermiogenesis, and fertilization. H3.3 has been reported to be important for spermatogen- S-sulfhydration of GAPDH significantly raised its enzyme esis [37, 58, 59]. We reported that S-sulfhydration of H3 activity [31]. Male mice with deficiency of GAPDH were and H3.3 was detected throughout spermatogenesis in the infertile and had profound defects in sperm motility [52]. present study. Noteworthy, RA, which is known to induce S-sulfhydration of ATP synthase [32] and LDHA [34] rST to differentiate into eST [ 38], also up-regulated their raised mitochondrial bioenergetics. Therefore, S-sulfhydra- sH3.3 expression (Fig. 3B). Keeping in line with the obser- tion of these proteins is probably required for maintenance vation, the level of sH3.3 was significantly higher in eST of optimal motility of sperm via regulating energy metabo- than in rST (Fig. 3A). Deficiency of mouse H3.3 resulted in lism. Our study also revealed a new mechanism regarding an aberrant spermiogenesis including an impaired develop- why addition of exogenous H S to semen improved the ment of round spermatids and poor motility of sperm [58, asthenozoospermic sperm motility [19]. Importantly, over- 59]. Collectively, sH3.3 is probably required for RA-induced all expression of sulfhydrated proteins including sH3/H3.3 spermiogenesis. Distinct aberrant PTMs of histones in sper- is higher in sperm with high motility than with low motility miogenesis resulted in infertile phenotypes including poor (Fig. 2B–E), and levels of sH3 and sH3.3 are positively sperm motility, strongly suggesting they can affect sperm associated with sperm progressive motility (Fig. 2F). Our motility in the way depending on their roles in modulating analysis also revealed that most of ROS-associated proteins gene transcription or disturbing sperm chromatin remodeling 1 3 3190 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 [60–64]. It is plausible that hypo-sulfhydration of H3.3 may sH3.3 expression in the C18-4 cells is oppositely regu- cause asthenozoospermia in a similar way. lated by GDNF and RA (Fig.  5A), two key factors for A hallmark of mammalian spermiogenesis is the step- SSC self-renewal and differentiation. Our study is con- wise completion of transition from histones to protamines sistent with the hypothesis that sH3.3 is important for the in spermatids [1, 2]. During the process, vast majority control of fate determination of SSCs. Ectopic expression of not only total histones but also levels of differentially of mutant H3.3 with C111S unregulated growth rate of modified histones were much reduced in eST, or mature C18-4 cells, partially by modulating expression of posi- sperm compared with rST [54, 62, 65–67]. Consistent with tive regulation of cell death and positive regulation of the reports, we also detected an expression of H3K9me3 apoptotic process–related genes and cell cycle–related in rST but it largely disappeared in eST (Fig. 3A). By con- genes (Fig.  5F). Therefore, sH3.3 is likely a suppressor trast, the presence of increased abundance of sH3/sH3.3 for the mitotic division of differentiating spermatogonia. in eST (Fig. 3A) strongly suggests that sH3/sH3.3 marks The role of sH3.3 in differentiation of SSC should be next the retained nucleosomes. The retained nucleosomes addressed. Nevertheless, our study revealed a regulatory have been revealed to distribute in genomic DNA in a role of sH3.3 in transcriptome in the SSC line. well-organized manner [68, 69], implying the existence Genomic distribution of H3.3 is critical to its regulat- of machinery protecting retained histones from eviction. ing role in gene transcription [37, 75], and is regulated However, little is known currently regarding the mecha- by RA in the way depending on raising turnover of H3.3 nisms. Considering that histone acetylation per se attenu- in the differentiation of embryonic stem cells [76, 77]. A ates the interplay between histone and DNA to facilitate high turnover of H3.3 was detected in male meiosis [37] histone removal [1], it is naturally tempting to speculate and spermiogenesis [69]. Considering that RA unregulated that the nucleosomes with an extra stabilizing mechanism expression of sH3.3 in male germ cells found in the pre- can probably be exempted from histone removal. The most sent study, whether/how S-sulfhydration of H3.3 affects members of the mammalian H3 family contain one or two genomic distribution of H3.3 should be next investigated cysteine(s) in their protein core, and this feature is a hall- in the different phases of spermatogenesis. C111 of H3.3 mark property of H3, given all other histone proteins lack is located inside the protein, making it difficult to be acces- cysteine. Intriguingly, the mammalian H3 variants contain sible for interaction of modified nucleosomal C111 with C111 that is located in their helix (85-114), the region any non-histone proteins. Therefore, H3.3 is likely to be where both H3 proteins are closely apposed in the nucleo- sulfhydrated largely outside nucleosomes. Some other some core particle [70]. The region immediately surround- PTMs of H3 were finished also outside nucleosomes [78]. ing C111 is important to hold together two histone H2A- In this regard, one can envision that H3.3 S-sulfhydration H2B-H3-H4 tetramers, because mutations of C111, for selectively regulates the turnover rate and distribution of example, destabilized the H3-H3 hydrophobic four-helix H3.3 probably via modulating interaction of H3.3 and its bundle tetramer interface in vitro [71]. Therefore, Hake chaperone proteins that were detected in spermatogenesis and Allis proposed that two C111 form an intermolecular [37, 79]. The hypothesis is worth a further approach. disulfide bond within two H3 proteins in the same nucleo- In conclusion, for the first time, H3.3 and H3 are some, adding stability to the H3-H4 tetramer [42]. Our showed to be S-sulfhydrated proteins in the present analysis showed the two C111 of nucleosomal H3.3 are study. We demonstrated that levels of sH3.3 and sH3 6.31 Åapart (Fig. 4D), basically excluding the possibility were down-regulated in asthenozoospermic sperm, sug- that they form the disulfide bond. However, once the two gesting that hypo-sulfhydration of H3 and H3.3 is a new Cys111 are thio-modified, the distance is narrowed to 2.3 biomarker for male infertility. sH3 has been detected Å (Fig. 4E), thus probably generating an intra-nucleoso- in all the different mouse organs examined (data not mal disulfide bond. In addition, our analysis also suggested shown). It is well known that oxidative stress is involved that S-sulfhydration of C111 is favorable to the formation in initiation and progression of diabetes, neurodegen- of multiple inter-helical hydrogen bonds between them and erative diseases, vascular disease, hypertension, aging, R128 (Fig. 4G). The four-residue-based multiple hydrogen and many other pathologies. Therefore, it could be bonds have been reported to exist in the structures of the speculated that aberrant regulation of sH3 can shape four-helix bundle tetramer [72, 73], causing the forma- epigenetic landscape, eventually making a significant tion of a super-secondary structure of four-stranded coiled contribution to the initiation and progression of distinct coil [74], and thus probably adds stability to the H3–H4 chronic diseases. tetramer. Collectively, S-sulfhydration of C111 is likely to Supplementary Information The online version contains supplemen- benefit the exemption of some nucleosomes from histone tary material available at https://doi. or g/10. 1007/ s10815- 021- 02314-x . removal via strengthening nucleosomal stability. 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3191 Acknowledgements We are very thankful for Prof. Biaoyang Lin, 9. Macleod G, Varmuza S. The application of proteomic approaches who works in Hangzhou Proprium Biotech Company, for his proteomic to the study of mammalian spermatogenesis and sperm function. analysis and comments. We are also very thankful for Shanghai Biopro- FEBS J. 2013;280(22):5635–51. file Company for its technical support on LC-MS analysis. 10. Saraswat M, et al. Human spermatozoa quantitative proteomic signature classifies normo- and asthenozoospermia. Mol Cell Pro - teomics. 2017;16(1):57–72. Funding This work was supported by grants from the National Nat- 11. Martinez-Heredia J, et  al. Identification of proteomic differ- ural Science Foundation of China (grant no. 81971443) (to R.L.), ences in asthenozoospermic sperm samples. Hum Reprod. Shanghai Municipal Committee of Science and Technology (grant no. 2008;23(4):783–91. 21140903600) (to R.L.), Shanghai Municipal Science and Technology 12. Parte PP, et al. Sperm phosphoproteome profiling by ultra per - Commission (21S11901000) (to R.L.), Innovation-Oriented Science formance liquid chromatography followed by data independent and Technology Grant from NHC Key laboratory of Reproduction Reg- analysis (LC-MS(E)) reveals altered proteomic signatures in ulation (CX2017–07) (to R.L.), and Shanghai Municipal Health and asthenozoospermia. J Proteome. 2012;75(18):5861–71. Health Commission clinical research project (20194Y0237) (to M.Z). 13. Marchiani S, et al. Sumo1-ylation of human spermatozoa and its relationship with semen quality. Int J Androl. 2011;34(6 Pt Data availability The data used to support the findings of this study are 1):581–93. available from the corresponding author upon request. 14. Cheng YM, et al. Lysine glutarylation in human sperm is associ- ated with progressive motility. Hum Reprod. 2019;34(7):1186–94. Declarations 15. Cheng YM, et al. Posttranslational lysine 2-hydroxyisobutyryla- tion of human sperm tail proteins affects motility. Hum Reprod. 2020;35(3):494–503. Conflict of interest The authors declare no competing interests. 16. Sun G, et al. Insights into the lysine acetylproteome of human sperm. J Proteome. 2014;109:199–211. Open Access This article is licensed under a Creative Commons Attri- 17. Li Y, et  al. The protein phosphorylation landscape of bution 4.0 International License, which permits use, sharing, adapta- mouse spermatids during spermiogenesis. Proteomics. tion, distribution and reproduction in any medium or format, as long 2019;19(11):e1900055. as you give appropriate credit to the original author(s) and the source, 18. Xie ZZ, Liu Y, Bian JS. Hydrogen sulfide and cellular redox provide a link to the Creative Commons licence, and indicate if changes homeostasis. Oxidative Med Cell Longev. 2016;2016:6043038. were made. The images or other third party material in this article are 19. Wang J, et al. Hydrogen sulfide as a potential target in preventing included in the article’s Creative Commons licence, unless indicated spermatogenic failure and testicular dysfunction. Antioxid Redox otherwise in a credit line to the material. If material is not included in Signal. 2018;28(16):1447–62. the article’s Creative Commons licence and your intended use is not 20. Mustafa AK, et al. H2S signals through protein S-sulfhydration. permitted by statutory regulation or exceeds the permitted use, you will Sci Signal. 2009;96(2):ra72. need to obtain permission directly from the copyright holder. To view a 21. Zhang D, et al. H2S-induced sulfhydration: biological function copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . and detection methodology. Front Pharmacol. 2017;8:608. 22. Ju Y et al. H(2)S-mediated protein S-sulfhydration: a prediction for its formation and regulation. Molecules. 2017 22(8) 23. Yang G, et al. Hydrogen sulfide protects against cellular senes - cence via S-sulfhydration of Keap1 and activation of Nrf2. Anti- References oxid Redox Signal. 2013;18(15):1906–19. 24. Sharma R, et al. Proteomic analysis of human spermatozoa pro- 1. Bao J, Bedford MT. Epigenetic regulation of the histone-to- teins with oxidative stress. Reprod Biol Endocrinol. 2013;11:48. protamine transition during spermiogenesis. Reproduction. 25. He Z, et al. Gdnf upregulates c-Fos transcription via the Ras/ 2016;151(5):R55–70. Erk1/2 pathway to promote mouse spermatogonial stem cell pro- 2. Wang T, et al. Essential role of histone replacement and modifica - liferation. Stem Cells. 2008;26(1):266–78. tions in male fertility. Front Genet. 2019;10:962. 26. Krishnan N, et al. H2S-Induced sulfhydration of the phosphatase 3. Lu WH, Gu YQ. Insights into semen analysis: a Chinese perspec- PTP1B and its role in the endoplasmic reticulum stress response. tive on the fifth edition of the WHO laboratory manual for the Sci Signal. 2011;4(203):ra86. examination and processing of human semen. Asian J Androl. 27. Pan H, et al. Ndrg3 gene regulates DSB repair during meiosis 2010;12(4):605–6. through modulation the ERK signal pathway in the male germ 4. Agarwal A, Gupta S, Sikka S. The role of free radicals and cells. Sci Rep. 2017;7:44440. antioxidants in reproduction. Curr Opin Obstet Gynecol. 28. Buchou T, et al. Purification and analysis of male germ cells from 2006;18(3):325–32. adult mouse testis. Methods Mol Biol. 2017;1510:159–68. 5. Smith TB, et  al. Functional deletion of Txndc2 and Txndc3 29. Wang W, et al. NDRG3 is an androgen regulated and prostate increases the susceptibility of spermatozoa to age-related oxida- enriched gene that promotes in vitro and in vivo prostate cancer tive stress. Free Radic Biol Med. 2013;65:872–81. cell growth. Int J Cancer. 2009;124(3):521–30. 6. Chen K, et  al. Low NRF2 mRNA expression in spermato- 30. Tarazona S, et  al. Data quality aware analysis of differential zoa from men with low sperm motility. Tohoku J Exp Med. expression in RNA-seq with NOISeq R/Bioc package. Nucleic 2012;228(3):259–66. Acids Res. 2015;43(21):e140. 7. Meybodi AM, et al. Importance of sperm gluthatione treatment in 31. Jarosz AP, et  al. Glyceraldehyde 3-phosphate dehydrogenase ART. J Assist Reprod Genet. 2012;29(7):625–30. (GAPDH) is inactivated by S-sulfuration in vitro. Free Radic Biol 8. Cao X, et al. Proteomic profile of human spermatozoa in healthy Med. 2015;89:512–21. and asthenozoospermic individuals. Reprod Biol Endocrinol. 32. Modis K, et al. S-Sulfhydration of ATP synthase by hydrogen 2018;16(1):16. sulfide stimulates mitochondrial bioenergetics. Pharmacol Res. 2016;113(Pt A):116–24. 1 3 3192 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 33. Paul BD, Snyder SH. H(2)S signalling through protein sulfhydra- 54. Tan M, et  al. Identification of 67 histone marks and histone tion and beyond. Nat Rev Mol Cell Biol. 2012;13(8):499–507. lysine crotonylation as a new type of histone modification. Cell. 34. Untereiner AA, et al. H2S-induced S-sulfhydration of lactate dehy- 2011;146(6):1016–28. drogenase a (LDHA) stimulates cellular bioenergetics in HCT116 55. Thomas K, et al. The epigenetic landscape related to reactive oxy- colon cancer cells. Biochem Pharmacol. 2017;136:86–98. gen species formation in the cardiovascular system. Br J Pharma- 35. Zhao W, et al. Outer dense fibers stabilize the axoneme to maintain col. 2017;174(12):1533–54. sperm motility. J Cell Mol Med. 2018;22(3):1755–68. 56. Sarah K, et al. Oxidative stress signaling to chromatin in health 36. Kanemori Y, et al. Biogenesis of sperm acrosome is regulated by and disease. Epigenomics. 2016;8(6):843–62. pre-mRNA alternative splicing of Acrbp in the mouse. Proc Natl 57. Barati E, et al. Oxidative stress and male infertility: current knowl- Acad Sci U S A. 2016;113(26):E3696–705. edge of pathophysiology and role of antioxidant therapy in disease 37. van der Heijden GW, et  al. Chromosome-wide nucleosome management. Cell Mol Life Sci. 2020;77(1):93–113. replacement and H3.3 incorporation during mammalian meiotic 58. Yuen BT, et al. Histone H3.3 regulates dynamic chromatin states sex chromosome inactivation. Nat Genet. 2007;39(2):251–8. during spermatogenesis. Development. 2014;141(18):3483–94. 38. Endo T, et al. Periodic production of retinoic acid by meiotic and 59. Tang MC, et  al. Contribution of the two genes encoding his- somatic cells coordinates four transitions in mouse spermatogen- tone variant h3.3 to viability and fertility in mice. PLoS Genet. esis. Proc Natl Acad Sci U S A. 2017;114(47):E10132–41. 2015;11(2):e1004964. 39. Henikoff S, Ahmad K. Assembly of variant histones into chroma- 60. Stiavnicka M, et al. H3K4me2 accompanies chromatin immaturity tin. Annu Rev Cell Dev Biol. 2005;21:133–53. in human spermatozoa: an epigenetic marker for sperm quality 40. Tachiwana H, et al. Structures of human nucleosomes containing assessment. Syst Biol Reprod Med. 2020;66(1):3–11. major histone H3 variants. Acta Crystallogr D Biol Crystallogr. 61. Yap DB, et al. Mll5 is required for normal spermatogenesis. PLoS 2011;67(Pt 6):578–83. One. 2011;6(11):e27127. 41. Craig DB, Dombkowski AA. Disulfide by Design 2.0: a web- 62. Lu LY, et  al. RNF8-dependent histone modifications regu- based tool for disulfide engineering in proteins. BMC Bioinfor - late nucleosome removal during spermatogenesis. Dev Cell. matics. 2013;14:346. 2010;18(3):371–84. 42. Hake SB, Allis CD. Histone H3 variants and their potential role 63. Liu S, et al. Chromodomain protein CDYL acts as a crotonyl-CoA in indexing mammalian genomes: the “H3 barcode hypothesis”. hydratase to regulate histone crotonylation and spermatogenesis. Proc Natl Acad Sci U S A. 2006;103(17):6428–35. Mol Cell. 2017;67(5):853–66 e5. 43. Adman E, Watenpaugh KD, Jensen LH. NH---S hydrogen bonds 64. Yu YE, et al. Abnormal spermatogenesis and reduced fertility in in Peptococcus aerogenes ferredoxin, Clostridium pasteurianum transition nuclear protein 1-deficient mice. Proc Natl Acad Sci U rubredoxin, and Chromatium high potential iron protein. Proc Natl S A. 2000;97(9):4683–8. Acad Sci U S A. 1975;72(12):4854–8. 65. Luense LJ, et al. Comprehensive analysis of histone post-transla- 44. Koli S, Mukherjee A, Reddy KVR. Retinoic acid triggers c-kit tional modifications in mouse and human male germ cells. Epi- gene expression in spermatogonial stem cells through an enhan- genetics Chromatin. 2016;9:24. ceosome constituted between transcription factor binding sites 66. Dottermusch-Heidel C, et al. H3K79 methylation: a new con- for retinoic acid response element (RARE), spleen focus forming served mark that accompanies H4 hyperacetylation prior to his- virus proviral integration oncogene (SPFI1) (PU.1) and E26 trans- tone-to-protamine transition in Drosophila and rat. Biol Open. formation-specific (ETS). Reprod Fertil Dev. 2017;29(3):521–43. 2014;3(6):444–52. 45. He Z, et al. MiRNA-20 and mirna-106a regulate spermatogonial 67. Song N, et al. Immunohistochemical analysis of histone H3 modi- stem cell renewal at the post-transcriptional level via targeting fications in germ cells during mouse spermatogenesis. Acta His- STAT3 and Ccnd1. Stem Cells. 2013;31(10):2205–17. tochem Cytochem. 2011;44(4):183–90. 46. Wang Y, et al. MicroRNA-322 regulates self-renewal of mouse 68. Hammoud SS, et  al. Distinctive chromatin in human spermatogonial stem cells through Rassf8. Int J Biol Sci. sperm packages genes for embryo development. Nature. 2019;15(4):857–69. 2009;460(7254):473–8. 47. Arnason T, et al. Cables1 is a tumor suppressor gene that regulates 69. Erkek S, et al. Molecular determinants of nucleosome retention at intestinal tumor progression in Apc(Min) mice. Cancer Biol Ther. CpG-rich sequences in mouse spermatozoa. Nat Struct Mol Biol. 2013;14(7):672–8. 2013;20(7):868–75. 48. Jimenez AP, et  al. The tumor suppressor RASSF1A induces 70. Luger K, et al. Crystal structure of the nucleosome core particle the YAP1 target gene ANKRD1 that is epigenetically inacti- at 2.8 A resolution. Nature. 1997;389(6648):251–60. vated in human cancers and inhibits tumor growth. Oncotarget. 71. Banks DD, Gloss LM. Folding mechanism of the (H3- 2017;8(51):88437–52. H4)2 histone tetramer of the core nucleosome. Protein Sci. 49. Wang S, et al. IGF-1R signaling is essential for the proliferation of 2004;13(5):1304–16. cultured mouse spermatogonial stem cells by promoting the G2/M 72. Haworth NL, et  al. Cross-strand disulfides in the hydrogen progression of the cell cycle. Stem Cells Dev. 2015;24(4):471–83. bonding site of antiparallel beta-sheet (aCSDhs): forbidden 50. Cernea M, et al. Wisp1 mediates Bmp3-stimulated mesenchymal disulfides that are highly strained, easily broken. Protein Sci. stem cell proliferation. J Mol Endocrinol. 2016;56(1):39–46. 2019;28(1):239–56. 51. He H, et al. miR-204-5p promotes the adipogenic differentiation 73. Liu J, et al. A seven-helix coiled coil. Proc Natl Acad Sci U S A. of human adipose-derived mesenchymal stem cells by modulating 2006;103(42):15457–62. DVL3 expression and suppressing Wnt/beta-catenin signaling. Int 74. Harbury PB, et  al. A switch between two-, three-, and four- J Mol Med. 2015;35(6):1587–95. stranded coiled coils in GCN4 leucine zipper mutants. Science. 52. Miki K, et  al. Glyceraldehyde 3-phosphate dehydroge- 1993;262(5138):1401–7. nase-S, a sperm-specific glycolytic enzyme, is required for 75. Szenker E, Ray-Gallet D, Almouzni G. The double face of the sperm motility and male fertility. Proc Natl Acad Sci U S A. histone variant H3.3. Cell Res. 2011;21(3):421–34. 2004;101(47):16501–6. 76. Schlesinger S, et al. A hyperdynamic H3.3 nucleosome marks 53. Sabari BR, et al. Metabolic regulation of gene expression through promoter regions in pluripotent embryonic stem cells. Nucleic histone acylations. Nat Rev Mol Cell Biol. 2017;18(2):90–101. Acids Res. 2017;45(21):12181–94. 1 3 Journal of Assisted Reproduction and Genetics (2021) 38:3175–3193 3193 77. Chen P, et al. H3.3 actively marks enhancers and primes gene 80. Schon SB, et  al. Histone modification signatures in human transcription via opening higher-ordered chromatin. Genes Dev. sperm distinguish clinical abnormalities. J Assist Reprod Genet. 2013;27(19):2109–24. 2019;36(2):267–75. 78. Loyola A, et  al. PTMs on H3 variants before chromatin assembly potentiate their final epigenetic state. Mol Cell. Publisher’s note Springer Nature remains neutral with regard to 2006;24(2):309–16. jurisdictional claims in published maps and institutional affiliations. 79. Rogers RS, et al. SUMO modified proteins localize to the XY body of pachytene spermatocytes. Chromosoma. 2004;113(5):233–43. 1 3

Journal

Journal of Assisted Reproduction and GeneticsSpringer Journals

Published: Dec 1, 2021

Keywords: S-sulfhydrated proteome; H3S-sulfhydration; Asthenozoospermia; Spermatogenesis; H3.3

There are no references for this article.