Advertisement

Homeodomain-interacting protein kinase HIPK4 regulates phosphorylation of manchette protein RIMBP3 during spermiogenesis

Open AccessPublished:August 01, 2022DOI:https://doi.org/10.1016/j.jbc.2022.102327
      Nonobstructive azoospermia (NOA) is the most serious form of spermatogenesis abnormalities in male infertility. Genetic factors are important to consider as elements leading to NOA. Although many pathogenic genes have been reported, the causative genes of NOA for many patients are still unknown. In this study, we found ten point mutations in the gene encoding homeodomain-interacting protein kinase 4 (HIPK4) in patients with NOA, and using in vitro studies, we determined a premature termination point mutation (p. Lys490∗, c.1468A>T) that can cause decreased expression of HIPK4. Our phosphoproteomic analysis of Hipk4−/− testes revealed phosphorylation of multiple proteins regulated by HIPK4 during spermiogenesis. We also confirmed that a substrate of HIPK4 with four downregulated phosphorylation sites matching the xSPx motif is the known manchette-related protein RIMS-binding protein 3, which is required for sperm head morphogenesis. Therefore, we conclude HIPK4 regulates the phosphorylation of manchette protein RIMS-binding protein 3 and plays essential roles in sperm head shaping and male fertility.

      Keywords

      Abbreviations:

      ES (ectoplasmic specialization), F-actin (filamentous actin), HIPK4 (homeodomain-interacting protein kinase 4), NOA (nonobstructive azoospermia), OAT (oligoasthenoteratozoospermia), RIMBP3 (RIMS-binding protein 3), TMT (tandem mass tags)
      Infertility is defined as the inability to conceive after 12 months of unprotected intercourse (
      • Irvine D.S.
      Epidemiology and aetiology of male infertility.
      ) and is a worldwide health problem affecting approximately 15% of couples (
      • Boivin J.
      • Bunting L.
      • Collins J.A.
      • Nygren K.G.
      International estimates of infertility prevalence and treatment-seeking: potential need and demand for infertility medical care.
      ). Male infertility often presents as oligoasthenoteratozoospermia (OAT) or the complete absence of sperm (azoospermia) in semen (
      • Sudhakar D.V.S.
      • Shah R.
      • Gajbhiye R.K.
      Genetics of male infertility - present and future: a narrative review.
      ). Nonobstructive azoospermia (NOA) is the most severe phenotype of spermatogenesis abnormalities, accounting for about 60% of azoospermia according to Jarow et al. (
      • Jarow J.P.
      • Espeland M.A.
      • Lipshultz L.I.
      Evaluation of the azoospermic patient.
      ) and affecting about 10% of infertile men according to Matsumiya et al. (
      • Matsumiya K.
      • Namiki M.
      • Takahara S.
      • Kondoh N.
      • Takada S.
      • Kiyohara H.
      • et al.
      Clinical study of azoospermia.
      ). Although the etiology is known for some cases, the majority are idiopathic. Among the causes, genetic factors contribute to 21% to 29% of NOA (
      • Berookhim B.M.
      • Schlegel P.N.
      Azoospermia due to spermatogenic failure.
      ), including chromosome number defects, Y-chromosome microdeletions, karyotype aberrations, and gene mutations (
      • Lee J.Y.
      • Dada R.
      • Sabanegh E.
      • Carpi A.
      • Agarwal A.
      Role of genetics in azoospermia.
      ). The identification of novel mutations in genes related to spermatogenesis is an important aspect of the diagnosis of the causes of infertility.
      Spermatogenesis is a complex process, involves mitosis of spermatogonia, meiosis of spermatocytes, and spermiogenesis of spermatids to generate sperm (
      • Grootegoed J.A.
      • Baarends W.M.
      • Hendriksen P.J.
      • Hoogerbrugge J.W.
      • Slegtenhorst-Eegdeman K.E.
      • Themmen A.P.
      Molecular and cellular events in spermatogenesis.
      ,
      • Jan S.Z.
      • Hamer G.
      • Repping S.
      • de Rooij D.G.
      • van Pelt A.M.
      • Vormer T.L.
      Molecular control of rodent spermatogenesis.
      ). During spermiogenesis, spermatids undergo dramatic morphological changes, including reshaping and condensation of the nucleus; removal of cytoplasm; and formation of mitochondrial sheath, acrosome, and flagella (
      • Yan W.
      Male infertility caused by spermiogenic defects: lessons from gene knockouts.
      ,
      • O'Donnell L.
      Mechanisms of spermiogenesis and spermiation and how they are disturbed.
      ). Many microtubule-based structures are involved in shaping the falciform spermatid head and assembly of the flagellum (
      • Sperry A.O.
      The dynamic cytoskeleton of the developing male germ cell.
      ). Manchette is a transient microtubule and filamentous actin (F-actin)–containing structure, appears in spermatids at step 8 and disappears at around steps 13 to 14, and is critical for head shaping and flagellum formation (
      • O'Donnell L.
      • O'Bryan M.K.
      Microtubules and spermatogenesis.
      ). Manchette connects nucleus via two potential nucleation sites, the perinuclear ring, and the centrosome, with a “grass skirt-like” structure formed by parallel microtubule bundles (
      • Lehti M.S.
      • Sironen A.
      Formation and function of the manchette and flagellum during spermatogenesis.
      ), which is crucial for sperm head shaping. For example, deficiency of the spermatid nuclear membrane protein SUN4 led to a disconnection between the manchette microtubules and nuclear envelope, which caused round-headed sperm (
      • Pasch E.
      • Link J.
      • Beck C.
      • Scheuerle S.
      • Alsheimer M.
      The LINC complex component Sun4 plays a crucial role in sperm head formation and fertility.
      ). While the absence of RIMS-binding protein 3 (RIMBP3), a manchette-associated protein, would lead to abnormal sperm head morphology and male infertility (
      • Zhou J.
      • Du Y.R.
      • Qin W.H.
      • Hu Y.G.
      • Huang Y.N.
      • Bao L.
      • et al.
      RIM-BP3 is a manchette-associated protein essential for spermiogenesis.
      ). However, the delicate function of manchette is under debate in terms of its role in spermatid head shaping and flagellum formation, and the molecular mechanisms and factors involved in this process remain poorly understood.
      Protein phosphorylation is an important cellular regulatory mechanism. Proteins can be phosphorylated on serine, threonine, or tyrosine residues by protein kinases (
      • Wang Z.
      • Cole P.A.
      Catalytic mechanisms and regulation of protein kinases.
      ). Some kinases have been found to be essential for spermiogenesis and male fertility, for example, the mitogen-activated protein kinases (
      • Wong C.H.
      • Cheng C.Y.
      Mitogen-activated protein kinases, adherens junction dynamics, and spermatogenesis: a review of recent data.
      ), the cell cycle regulators POLO-like kinases (
      • Qi L.
      • Liu Z.
      • Wang J.
      • Cui Y.
      • Guo Y.
      • Zhou T.
      • et al.
      Systematic analysis of the phosphoproteome and kinase-substrate networks in the mouse testis.
      ), the androgen receptor p21-activated kinase 6 (
      • Liu X.
      • Busby J.
      • John C.
      • Wei J.
      • Yuan X.
      • Lu M.L.
      Direct interaction between AR and PAK6 in androgen-stimulated PAK6 activation.
      ), and the testis-specific serine/threonine-protein kinase family (
      • Kueng P.
      • Nikolova Z.
      • Djonov V.
      • Hemphill A.
      • Rohrbach V.
      • Boehlen D.
      • et al.
      A novel family of serine/threonine kinases participating in spermiogenesis.
      ) were all shown to be essential for male infertility. Previously, we systematically characterized protein phosphorylation during spermiogenesis, and the function of a kinase, homeodomain-interacting protein kinase 4 (HIPK4), in spermiogenesis and male infertility has been less documented (
      • Zhao H.
      • Li T.
      • Zhao Y.
      • Tan T.
      • Liu C.
      • Liu Y.
      • et al.
      Single-cell transcriptomics of human oocytes: environment-driven metabolic competition and compensatory mechanisms during oocyte maturation.
      ). HIPK4 belongs to the homeodomain-interacting protein kinases family. This serine/threonine kinase is predominantly expressed in the testis (
      • Ogurtsov A.Y.
      • Marino-Ramirez L.
      • Johnson G.R.
      • Landsman D.
      • Shabalina S.A.
      • Spiridonov N.A.
      Expression patterns of protein kinases correlate with gene architecture and evolutionary rates.
      ) and has a cytoplasmic localization because it lacks a typical nuclear localization signal (
      • van der Laden J.
      • Soppa U.
      • Becker W.
      Effect of tyrosine autophosphorylation on catalytic activity and subcellular localisation of homeodomain-interacting protein kinases (HIPK).
      ). Previous single-cell sequencing showed its expression in spermatids in human testis (
      • Guo J.
      • Grow E.J.
      • Mlcochova H.
      • Maher G.J.
      • Lindskog C.
      • Nie X.
      • et al.
      The adult human testis transcriptional cell atlas.
      ,
      • Wang M.
      • Liu X.
      • Chang G.
      • Chen Y.
      • An G.
      • Yan L.
      • et al.
      Single-cell RNA sequencing analysis reveals sequential cell fate transition during human spermatogenesis.
      ). In a recent study, deletion of Hipk4 was found to lead to male sterility with defects in the F-actin–scaffolded acroplaxome during spermatid elongation and abnormal head morphologies in mature spermatozoa (
      • Crapster J.A.
      • Rack P.G.
      • Hellmann Z.J.
      • Le A.D.
      • Adams C.M.
      • Leib R.D.
      • et al.
      HIPK4 is essential for murine spermiogenesis.
      ); however, the direct substrates of HIPK4 are still not known.
      Here, we examined whether HIPK4 mutations are present in patients with NOA and identified an HIPK4 heterozygous truncating mutation that leads to decreased protein expression. Phosphoproteomic profiling of Hipk4 knockout (KO) testis and experimental validation further identified manchette-associated protein RIMS-binding protein 3 as a phosphorylation substrate of HIPK4. Thus, we identified HIPK4 as a novel manchette function regulator with essential roles in spermatid head shaping and male fertility.

      Results

      Identification of deleterious HIPK4 mutations from patients with NOA

      To identify the HIPK4-associated mutations in human with spermatogenic defects, we conducted Sanger sequencing in coding regions of HIPK4 from a cohort of 620 NOA patients and 2678 fertile controls (Figs. 1A and S1A). Intriguingly, 10 rare (absent in the 2678 fertile controls) and deleterious (combined annotation–dependent depletion [CADD score of >15]) variants were identified in 10 unrelated NOA-affected men (Table S2), including two heterozygous nonsense mutations (NP_653286.2: p.Lys490∗, c.1468A>T and NP_653286.2: p.Arg541∗, c.1621C>T) and eight heterozygous missense mutations, three of which were in the kinase catalytic domain of HIPK4 (Figs. 1B and S1B). To further evaluate the effects of these mutations in vitro, we overexpressed the full-length wildtype and mutant cDNA constructs in HEK293T cells. Two heterozygous nonsense mutations harboring in HIPK4 were truncated, and the protein level of one variant (p.Lys490∗, c.1468A>T) was significantly decreased compared with that of the wildtype, FLAG-tagged EGFP as transfection efficiency control (Fig. 1, C and D). Thus, the NOA-associated mutation (NP_653286.2: p.Lys490∗, c.1468A>T) could lead to decreased protein expression of HIPK4.
      Figure thumbnail gr1
      Figure 1HIPK4 mutations were detected in patients with NOA. A, the number of NOA cases and fertile controls included in the mutation analysis. B, sequencing graphs of five mutations identified from patients with NOA with the detected mutations annotated in red. C and D, western blots (C) and statistics analysis (D) of the overexpression of HA-HIPK4 with different mutations from three independent replicates, with FLAG-EGFP as a transfection control and ACTIN as a loading control. HIPK4 proteins were detected via the anti-HA antibody. Data are presented with the mean ± SD. ∗∗p < 0.01 by Student’s t test. HIPK4, homeodomain-interacting protein kinase 4; NOA, nonobstructive azoospermia.

      HIPK4 deficiency led to defective manchette formation and retained spermatids

      According to previous reports, HIPK4 was largely expressed in the testis and first detected in germ cells at postnatal day 21, coinciding with the appearance of step 2 to 3 round spermatids (
      • Crapster J.A.
      • Rack P.G.
      • Hellmann Z.J.
      • Le A.D.
      • Adams C.M.
      • Leib R.D.
      • et al.
      HIPK4 is essential for murine spermiogenesis.
      ), with a cytoplasmic location (
      • van der Laden J.
      • Soppa U.
      • Becker W.
      Effect of tyrosine autophosphorylation on catalytic activity and subcellular localisation of homeodomain-interacting protein kinases (HIPK).
      ). To study in vivo functions of Hipk4, we used CRISPR/Cas9 system to generate Hipk4 KO mice. Among the Hipk4 mutants, female Founder #3 with a 49 bp deletion was chosen for subsequent studies (Fig. S2). Western blotting analysis detected no HIPK4 in adult Hipk4−/− mouse testis (Fig. 2A). Hipk4−/− females displayed normal fertility, while Hipk4−/− males were sterile (Fig. 2B). Compared with wildtype, dramatically decreased numbers of sperm, sperm motility, and sperm progressive motility (Fig. 2C, Table S3 and Movie S1) were observed.
      Figure thumbnail gr2
      Figure 2Hipk4 knockout male mice are infertile with abnormal manchette. A, protein expression of HIPK4 in different genotypes, with GAPDH as a loading control. B, litter sizes of Hipk4−/− male and female mice (ANOVA with post-hoc Dunnett’s t test). C, comparison of sperm count (i), sperm motility (ii), and sperm progressive motility (iii) of 8-week-old Hipk4+/+ and Hipk4−/− mice (n = 3) (two-tailed Student’s t test). D, immunofluorescence analysis between seminiferous tubules from Hipk4+/+ and Hipk4−/− testes with acrosome stained by PNA and DNA stained by DAPI. Scale bar: 20 μm. E, PAS-stained different stages of seminiferous tubules from adult Hipk4+/+ and Hipk4−/− testes. The green arrow points to abnormal nuclei of spermatids. Scale bar, 20 μm. F, ultrastructural analysis of Hipk4+/+ and Hipk4−/− spermatids using transmission electron microscopy. Hipk4+/+ elongating spermatids (steps 9–12) show normal assembly of the manchette and perinuclear ring (a and d). The perinuclear ring (red arrow) of elongating spermatids (steps 9–12) in Hipk4−/− is ectopically placed (detached from the nucleus) (b and f). The acroplaxome (green arrow) is distorted (c and f). The yellow arrow indicates that manchette is not attached to the nucleus (c). The manchette is elongated and the perinuclear ring shows abnormal constriction in spermatids (e). The mutant condensed spermatids frequently contain deformed nuclei and detached acrosomes (green arrow), with an expanded perinuclear space (h and i). Scale bars: 1 μm. Arrow (red), perinuclear ring. Box is high amplification. G, spermatids isolated from Hipk4+/+ and Hipk4−/− mice were stained for α-TUBULIN (green) at different manchette-containing steps. The distance between the perinuclear ring and the caudal side of the head is indicated by double-headed arrows. Scale bar: 5 μm. H, manchette clearance is delayed in Hipk4−/− mice. During stages II–III the manchette has disappeared in the Hipk4+/+ mouse but was retained in the Hipk4−/− mouse. α-TUBULIN (green) was used to visualize manchette microtubules, with DNA stained by DAPI (blue), and acrosome stained by PNA (red). Data are presented with the mean ± SD. ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. D, diplotene; ES, elongated spermatids; HIPK4, homeodomain-interacting protein kinase 4; L, leptotene; M, metaphase; Nu, nucleus; P, pachytene; RS, round spermatids; Acr, acrosome; M, manchette; Z, zygotene.
      In addition, Hipk4−/− sperm lacked a typical hook-shaped appearance and showed different degrees of acrosome defects (Fig. 2D). Abnormal nuclei in the elongated and condensing Hipk4−/− spermatids from steps 10 to 16 were observed (Fig. 2E). Similar to previous report (
      • Crapster J.A.
      • Rack P.G.
      • Hellmann Z.J.
      • Le A.D.
      • Adams C.M.
      • Leib R.D.
      • et al.
      HIPK4 is essential for murine spermiogenesis.
      ), varying degrees of defects were observed using transmission electron microscope during nuclear elongation, including invaginated and/or evaginated acroplaxome containing region (Fig. 2F, c and f), elongated manchette along with constricted nuclear around perinuclear ring (Fig. 2F, e), and deviated perinuclear ring (Fig. 2F, b and f). At steps 15 to 16, most condensed spermatids exhibited detached acrosome, deformed nucleus, and expanded perinuclear space (Fig. 2F, h and i), suggesting defects in the transport along the spermatid manchette. Moreover, abnormally elongated manchette and distal movement along the nuclear surface were found in Hipk4−/− elongating spermatids (Figs. 2G and S3A) and persistent in steps 13 to 14, whereas the manchette had already disappeared in wildtype (Fig. 2H).
      The increased unreleased spermatids with delayed spermiation in stage IX of Hipk4−/− mice (Fig. 3, A and B) caused us to focus on apical ectoplasmic specialization (ES). ES is a kind of adhesion junction based on microfilaments (F-actin) (
      • Cheng C.Y.
      • Mruk D.D.
      A local autocrine axis in the testes that regulates spermatogenesis.
      ,
      • Mruk D.D.
      • Cheng C.Y.
      Cell-cell interactions at the ectoplasmic specialization in the testis.
      ,
      • Vogl A.W.
      • Young J.S.
      • Du M.
      New insights into roles of tubulobulbar complexes in sperm release and turnover of blood-testis barrier.
      ), and retained apical ES between spermatids and Sertoli cells in stage IX has been shown to lead to delayed spermiation (
      • O'Donnell L.
      Mechanisms of spermiogenesis and spermiation and how they are disturbed.
      ). Staining of ESPIN, a marker for ES, showed retained spermatids in stage IX–X without ES structures in Hipk4−/− mice (Figs. 3C and S3B), suggesting the spermatids had completed the earlier phases of spermiation but failed disengagement. Therefore, a significant proportion of spermatids were not released from Sertoli cell and instead phagocytosed, resulting in fewer sperm entering the epididymis. As expected, TUNEL staining revealed increased numbers of apoptotic cells in Hipk4−/− seminiferous tubules, which were predominantly present in stage IX to X (Figs. 3, D and E and S3C). TUNEL staining and cyclical variations of the apoptotic cells strongly supported an origin from phagocytized spermatids.
      Figure thumbnail gr3
      Figure 3Hipk4 knockout testes show retained spermatids. A, cross sections of Hipk4+/+ and Hipk4−/− mouse testes with stages of seminiferous tubules labeled, scale bar, 50 μm. Retained spermatids in stage IX tubules are shown in red arrow. scale bar, 20 μm. B, statistics analysis of retained spermatids from 25 tubules in three replicates (Student’s t test). C, Immunofluorescence of seminiferous tubules from control and Hipk4−/− testes using mouse anti-ESPIN antibody with acrosome stained by PNA and DNA stained by DAPI. The arrows indicate unreleased spermatids. Scale bars: 10 μm. D, TUNEL in stage IX–X Hipk4+/+ and Hipk4−/− seminiferous tubules (30 round tubules analyzed for each genotype). TUNEL (green), DNA (blue), and H4K16ac (red), a marker for elongating spermatids. Arrows (green) indicate TUNEL-positive spermatids. Scale bar, 20 μm. Box is high amplification. E, quantitative analysis of TUNEL staining in stage IX–X Hipk4+/+ and Hipk4−/− seminiferous tubules (Student’s t test, 30 round tubules analyzed for each genotype). Data are presented as the mean ± SD. ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. ES, elongated spermatid; L, leptotene; P, pachytene

      Phosphoproteomic analysis of Hipk4−/− testes revealed downstream phosphorylation substrates

      To explore the mechanisms by which HIPK4 functions during spermiogenesis, we used multiplexed tandem mass tags (TMT) labeling and a tandem mass spectrometry (LC-MS/MS) approach to quantify the proteome and phosphoproteome of testes from Hipk4−/− and Hipk4+/+ mice. Altogether, 10,621 proteins were quantified, and only 47 proteins showed differential expression levels after Hipk4 KO, comprising six upregulated proteins and 41 downregulated proteins (Fold change > 1.5 and p < 0.05) (Fig. S4, A and B and Table S4). Gene ontology (GO) analysis of the 47 differentially abundant proteins showed no enrichment of any terms in biological process or cellular localization. Phosphoproteomic profiling identified 31,331 phosphorylation sites, corresponding to 6142 phosphorylated proteins, with 513 differential phosphorylation sites corresponding to 359 proteins between the Hipk4−/− and Hipk4+/+ groups after normalization against the protein expression levels (localization probability > 0.75, multiplicity = 1, fold change > 1.5 and p < 0.05). The number of differentially phosphorylated proteins induced by Hipk4 KO was significantly higher than that of differentially abundant proteins, suggesting that HIPK4 mainly affected the phosphorylation of proteins. Among the differential phosphorylation sites, 475 sites corresponding to 332 phosphoproteins were downregulated, but only 38 sites were upregulated (Fig. 4, A and B and Table S4), consistent with the loss of kinase function after Hipk4 KO. The proteins with downregulated phospho-sites can be potential phosphorylation substrates of the HIPK4 kinase. GO analysis of these 332 downregulated phosphoproteins showed significant enrichment in “microtubule” (20/332), “microtubule organizing center part (18/332)”, “motile cilium” (16/332), “acrosomal vesicle” (10/332), “sperm flagellum” (8/332), “microtubule associated complex” (8/332), and “sperm principal piece” (3/332) (Fig. 4C and Table S5). Besides, “microtubule-based movement” in cellular component terms (16/332), “nucleocytoplasmic transport” (14/332), “spermatid differentiation” (22/332), “microtubule bundle formation” (8/332), “sperm motility” (7/332), “fertilization” (13/332), and “cilium assembly” (23/332) were the enriched biological process terms (Fig. 4D and Table S5). The analysis demonstrated regulatory roles of HIPK4 in spermiogenesis and sperm functions through phosphorylation. Kinase specificity is defined by the amino acid sequences surrounding the phosphorylation sites (
      • Ubersax J.A.
      • Ferrell Jr., J.E.
      Mechanisms of specificity in protein phosphorylation.
      ); therefore, we employed the Motif-X algorithm (
      • Cox J.
      • Mann M.
      MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification.
      ) using a sequence window of ±six amino acids around the identified phosphorylated site to predict consensus phosphorylation site motifs. The most significantly enriched motif was xSPx (p = 9.50E-12) (amino acids with low probability in modification site motif visualization were replaced with letter “x” in phospho-peptides); the corresponding conserved sequence logos were generated by WebLogo (
      • Schwartz D.
      • Gygi S.P.
      An iterative statistical approach to the identification of protein phosphorylation motifs from large-scale data sets.
      ) (Fig. 4E). Interestingly, this most enriched motif, xSPx, is also recognized by HIPK2, another member of HIPK family (
      • Ann E.J.
      • Kim M.Y.
      • Yoon J.H.
      • Ahn J.S.
      • Jo E.H.
      • Lee H.J.
      • et al.
      Tumor suppressor HIPK2 regulates malignant growth via phosphorylation of Notch1.
      ,
      • Yamada D.
      • Perez-Torrado R.
      • Filion G.
      • Caly M.
      • Jammart B.
      • Devignot V.
      • et al.
      The human protein kinase HIPK2 phosphorylates and downregulates the methyl-binding transcription factor ZBTB4.
      ), which suggested that the kinase specificity might be conserved among the HIPKs.
      Figure thumbnail gr4
      Figure 4Phosphoproteomic profiling of Hipk4 knockout testicular proteins. A, the volcano plot of quantified protein phosphorylation sites between Hipk4+/+ and Hipk4−/− mice. The cutoff values (localization probability > 0.75, multiplicity = 1, fold change >1.5, and p < 0.05) were utilized to identify phosphorylation sites with significantly differential levels. B, heatmap of differentially regulated phosphorylation sites. The Euclidean distance metric and the complete linkage clustering algorithm were used. C and D, Gene ontology annotations of the downregulated phosphorylated proteins in Hipk4−/− mice testes. Enriched terms of cellular components (C) and biological processes (D) are shown as a bar graph and a network by Cytoscape, respectively. E, the most enriched motif among 475 downregulated phosphorylation sites in Hipk4−/− testes was identified using the Motif-X algorithm. The height of the residues represents the frequency with which they appear at the respective positions. The color of the residues represents their physicochemical properties.

      Manchette-associated protein RIMBP3 is a phosphorylation substrate of HIPK4

      The quantitative phosphoproteomic profiling identified four phosphorylation sites (Ser15, Ser21, Ser263, and Ser347) as xSPx motif in RIMBP3, which were all significantly downregulated after Hipk4 KO and located in the N-terminal region of RIMBP3 (Fig. 5, A and B). RIMBP3 is a manchette-associated protein, previously reported to be essential for manchette formation, playing a key role in sperm head morphogenesis during the late stages of sperm development (
      • Zhou J.
      • Du Y.R.
      • Qin W.H.
      • Hu Y.G.
      • Huang Y.N.
      • Bao L.
      • et al.
      RIM-BP3 is a manchette-associated protein essential for spermiogenesis.
      ). Rimbp3 deletion phenocopied the malformed nucleus of Hipk4−/− spermatids (
      • Zhou J.
      • Du Y.R.
      • Qin W.H.
      • Hu Y.G.
      • Huang Y.N.
      • Bao L.
      • et al.
      RIM-BP3 is a manchette-associated protein essential for spermiogenesis.
      ), which suggested that RIMBP3 might be a potential HIPK4 phosphorylation substrate. To validate whether the N terminus of RIMBP3 is phosphorylated by HIPK4, we performed Phos-tag Western blotting, which allows the detection of phosphorylation by reducing the electrophoretic mobility of phosphorylated protein using Phos-tag, a phosphorylated amino acid chelator. When Rimbp3 encoding an N-terminally truncated (1–460 AA) protein was cotransfected with Hipk4 into HEK293T cells, phosphorylated RIMBP3 and autophosphorylated HIPK4 showed altered migration behavior using Phos-tag, but no change in migration of RIMBP3 and HIPK4 occurred when using kinase-dead mutant HIPK4 (Y175F) (
      • van der Laden J.
      • Soppa U.
      • Becker W.
      Effect of tyrosine autophosphorylation on catalytic activity and subcellular localisation of homeodomain-interacting protein kinases (HIPK).
      ) (Fig. 5C). To detect the potential interaction between RIMBP3 and HIPK4, we overexpressed HA-tagged HIPK4 with FLAG-tagged full-length RIMBP3 in HEK293T cells, and co-immunoprecipitation (IP) analysis indicated that RIMBP3 interacted with HIPK4 in HEK293 cells (Fig. 5D). Furthermore, we immunoprecipitated RIMBP3 from testis lysates using anti-RIMBP3 antibodies, which also identified HIPK4 as an interacting protein of RIMBP3 in testis (Fig. 5E). Thus, RIMBP3 is a phosphorylation substrate of HIPK4. HIPK4 was reported to be expressed in steps 3 to 8 spermatids (
      • Crapster J.A.
      • Rack P.G.
      • Hellmann Z.J.
      • Le A.D.
      • Adams C.M.
      • Leib R.D.
      • et al.
      HIPK4 is essential for murine spermiogenesis.
      ), while RIMBP3 was expressed in the cytoplasm of spermatocytes and round spermatids and was mainly located in the manchette of elongating spermatids (
      • Zhou J.
      • Du Y.R.
      • Qin W.H.
      • Hu Y.G.
      • Huang Y.N.
      • Bao L.
      • et al.
      RIM-BP3 is a manchette-associated protein essential for spermiogenesis.
      ) (Figs. 6A and S3D). RIMBP3 plays a critical role in manchette organization and function; therefore, we wanted to know whether RIMBP3 localization and the manchette were affected in Hipk4 null mice. Immunofluorescence analysis showed that RIMBP3 and α-TUBULIN were co-localized in the manchette (Fig. 6B). After Hipk4 KO, RIMBP3 and α-TUBULIN co-localized manchette showed abnormal shape in spermatids of steps 11 to 12. Collectively, these results suggested that HIPK4 regulates manchette formation and function at least partly through phosphorylation of RIMBP3, thus playing an essential role in sperm head shaping (Fig. 6C).
      Figure thumbnail gr5
      Figure 5RIMBP3 is a phosphorylation substrate of HIPK4. A, four downregulated phosphor-peptides from RIMBP3 present an xSPx motif. B, the level of different phosphorylation sites in RIMBP3 from the quantitative phosphoproteomics data. Data are shown as the mean ± SD (n = 3), ∗∗∗∗p < 0.0001. C, Western blotting analysis of overexpressed HA-HIPK4 or HA-HIPK4 (Y175F) in HEK293T cell together, with or without FLAG-RIMBP3 (1–460) using a Phos-tag gel with GAPDH as a loading control. D, interaction between HA-HIPK4 and FLAG-RIMBP3 in vitro validated in HEK293T cell. E, interaction between HIPK4 and RIMBP3 in vivo in mouse testis. HIPK4, homeodomain-interacting protein kinase 4; RIMBP3, RIMS-binding protein 3.
      Figure thumbnail gr6
      Figure 6HIPK4 regulates sperm head shaping via phosphorylating RIMBP3. A, localization of RIMBP3 in seminiferous tubules from Hipk4+/+ mice. RIMBP3 was mainly located in the manchette of elongating spermatids and also expressed in spermatocytes and round spermatids. IF staining of RIMBP3 (red) and α-TUBULIN (green) in seminiferous tubules from Hipk4+/+ mice with nuclei stained by DAPI. Scale bar, 10 μm. B, the localization of RIMBP3 in the manchette of steps 11 to 12 spermatids were affected after deletion of Hipk4. Co-immunofluorescence of RIMBP3 together with manchette stained by anti-α-TUBULIN, acrosome stained by PNA, and DNA stained by DAPI in Hipk4+/+ and Hipk4−/− spermatids from step 8 to step 11 to 12 during spermiogenesis. Schematic diagrams of Hipk4+/+ and Hipk4−/− spermatids at different steps showing shape of the nucleus (blue), acrosome (purple), and manchette (green) are on the right. Scale bar: 5 μm. C, a theoretical model shows that HIPK4 phosphorylates RIMBP3 and regulates manchette during spermatogenesis. HIPK4, homeodomain-interacting protein kinase 4; RIMBP3, RIMS-binding protein 3.

      Discussion

      HIPK4 is a serine/threonine kinase that is predominantly expressed in mouse testis. Despite Hipk4-defective mice displaying male infertility, the exact roles of HIPK4 in spermatogenesis remain unclear. In this study, we identified ten heterogeneous mutations in HIPK4 from patients with NOA and demonstrated that nonsense mutant (p. Lys490∗, c.1468A>T) decreased protein expression. The Hipk4 KO mice showed male infertility with defects in spermiogenesis which led to malformed sperm heads, which was consistent with Crapster et al.’s findings (
      • Crapster J.A.
      • Rack P.G.
      • Hellmann Z.J.
      • Le A.D.
      • Adams C.M.
      • Leib R.D.
      • et al.
      HIPK4 is essential for murine spermiogenesis.
      ). In Hipk4−/− mice, phosphoproteomic profiling mainly showed downregulated phosphorylation, suggesting essential roles of HIPK4 in spermiogenesis. Moreover, we also demonstrated that a manchette-associated protein, RIMBP3, was a substrate of HIPK4. Overall, our study showed that HIPK4 is associated with human spermatogenesis, which is an important regulator of manchette formation and function, and essential for sperm head shaping.
      Previous studies have shown that a proportion of idiopathic human male infertility, often accompanied by azoospermia or severe oligospermia, is caused by genetic defects (
      • de Kretser D.M.
      Male infertility.
      ,
      • Don J.
      • Stelzer G.
      The expanding family of CREB/CREM transcription factors that are involved with spermatogenesis.
      ). In this study, we found that the Hipk4 KO mice presented OAT, while we detected deleterious HIPK4 mutations in patients with NOA, which is a more severe form of spermatogenesis abnormality. Actually, human azoospermia does not mean that the testis cannot produce sperm. Although no sperm could be observed in semen, studies show that sperm could be retrieved surgically from 43 to 63% of patients with NOA and could be used for assisted reproduction (
      • Westlander G.
      Utility of micro-TESE in the most severe cases of non-obstructive azoospermia.
      ). Idiopathic NOA is a complex, highly polygenic, and environmentally affected disease (
      • Cannarella R.
      • Condorelli R.A.
      • Duca Y.
      • La Vignera S.
      • Calogero A.E.
      New insights into the genetics of spermatogenic failure: a review of the literature.
      ). Environmental factors can cause spermatogenesis abnormalities (
      • Zafar A.
      • Eqani S.A.
      • Bostan N.
      • Cincinelli A.
      • Tahir F.
      • Shah S.T.
      • et al.
      Toxic metals signature in the human seminal plasma of Pakistani population and their potential role in male infertility.
      ,
      • Alfano M.
      • Ferrarese R.
      • Locatelli I.
      • Ventimiglia E.
      • Ippolito S.
      • Gallina P.
      • et al.
      Testicular microbiome in azoospermic men-first evidence of the impact of an altered microenvironment.
      ) and might interact with genetic mutations, leading to more severe phenotypes in humans. Common mutations in the human genome, mainly single nucleotide polymorphisms and copy number variations, might also play roles in spermatogenesis failure (
      • Cervan-Martin M.
      • Castilla J.A.
      • Palomino-Morales R.J.
      • Carmona F.D.
      Genetic landscape of nonobstructive azoospermia and new perspectives for the clinic.
      ) and might interact with the mutations of HIPK4. However, the mice used in this study were inbred with a pure genetic background. Different phenotypes between mice and humans have been reported for many other genes. For example, mutations of DNAH1 can lead to multiple morphological abnormalities of the sperm flagella and male sterility in humans (
      • Ben Khelifa M.
      • Coutton C.
      • Zouari R.
      • Karaouzene T.
      • Rendu J.
      • Bidart M.
      • et al.
      Mutations in DNAH1, which encodes an inner arm heavy chain dynein, lead to male infertility from multiple morphological abnormalities of the sperm flagella.
      ); however, invalidation of the mouse ortholog, Mdhc7, caused male sterility with the decreased motility, but no morphological defects of the sperm tail were observed (
      • Neesen J.
      • Kirschner R.
      • Ochs M.
      • Schmiedl A.
      • Habermann B.
      • Mueller C.
      • et al.
      Disruption of an inner arm dynein heavy chain gene results in asthenozoospermia and reduced ciliary beat frequency.
      ). Therefore, it is reasonable to see HIPK4 mutations in NOA and OAT in Hipk4 KO mice.
      A previous study revealed F-actin scaffolded acroplaxome defects in elongating spermatids and abnormal sperm head morphologies in Hipk4-null mice (
      • Crapster J.A.
      • Rack P.G.
      • Hellmann Z.J.
      • Le A.D.
      • Adams C.M.
      • Leib R.D.
      • et al.
      HIPK4 is essential for murine spermiogenesis.
      ), but the distinct substrates and regulatory mechanisms of HIPK4 in spermiogenesis are still unknown. Similar to that, we also found a comprehensive abnormality of the acrosome–acroplaxome–manchette complex after Hipk4 KO, including a distorted acroplaxome, a detached acrosome, elongated and symmetric conical manchette, and the ectopic positioning of the perinuclear ring. The acrosome–acroplaxome–manchette complex was thought to define the specific shape of the sperm head (
      • Kierszenbaum A.L.
      • Tres L.L.
      The acrosome-acroplaxome-manchette complex and the shaping of the spermatid head.
      ). The acroplaxome is tightly bound to the acrosome. It shapes the apical portion of the spermatid head and might provide a scaffold to regulate exogenous constriction forces produced by Sertoli cell F-actin hoops (
      • Russell L.D.
      • Russell J.A.
      • MacGregor G.R.
      • Meistrich M.L.
      Linkage of manchette microtubules to the nuclear envelope and observations of the role of the manchette in nuclear shaping during spermiogenesis in rodents.
      ). The manchette shapes the basal portion of the spermatid head and moves distally in conjunction with the constriction of the perinuclear ring (
      • Russell L.D.
      • Russell J.A.
      • MacGregor G.R.
      • Meistrich M.L.
      Linkage of manchette microtubules to the nuclear envelope and observations of the role of the manchette in nuclear shaping during spermiogenesis in rodents.
      ). Further mechanistic studies revealed a substrate of HIPK4, the adaptor molecule RIMBP3, deletion of which showed similar acrosome–acroplaxome–manchette complex abnormalities (
      • Zhou J.
      • Du Y.R.
      • Qin W.H.
      • Hu Y.G.
      • Huang Y.N.
      • Bao L.
      • et al.
      RIM-BP3 is a manchette-associated protein essential for spermiogenesis.
      ). Previous studies also showed that RIMBP3 regulates the development and function of manchette by interacting with hook microtubule tethering protein 1 (HOOK1) and kinesin family member 3B (
      • Zhou J.
      • Du Y.R.
      • Qin W.H.
      • Hu Y.G.
      • Huang Y.N.
      • Bao L.
      • et al.
      RIM-BP3 is a manchette-associated protein essential for spermiogenesis.
      ). HOOK1 also is a manchette-associated protein, whose mutation caused abnormal sperm heads (
      • Mendoza-Lujambio I.
      • Burfeind P.
      • Dixkens C.
      • Meinhardt A.
      • Hoyer-Fender S.
      • Engel W.
      • et al.
      The Hook1 gene is non-functional in the abnormal spermatozoon head shape (azh) mutant mouse.
      ). Kinesin family member 3B is microtubule-based molecular motor protein responsible for intramanchette transport (
      • Kierszenbaum A.L.
      Intramanchette transport (IMT): managing the making of the spermatid head, centrosome, and tail.
      ). Thus, HIPK4 might regulate spermatid head morphogenesis by regulating the acrosome–acroplaxome–manchette complex through RIMBP3. Besides, Kif3a, Cep131, Lrguk1, and Spef2 mutant mice (
      • Lehti M.S.
      • Sironen A.
      Formation and function of the manchette and flagellum during spermatogenesis.
      ) showed that the manchette also plays an important role in flagellum formation; however, the ultrastructure of Hipk4−/− sperm flagellum is normal and similar to that of Rimbp3 and Hook1 defective mice. It seems that abnormalities of certain manchette proteins may not necessarily lead to a disorder of the flagellum structure. HIPK4 mainly regulates RIMBP3, a manchette protein essential for sperm head but not flagella formation.
      Previous studies have shown that the C terminus of RIMBP3 interacts with HOOK1 to regulate the development of the manchette (
      • Zhou J.
      • Du Y.R.
      • Qin W.H.
      • Hu Y.G.
      • Huang Y.N.
      • Bao L.
      • et al.
      RIM-BP3 is a manchette-associated protein essential for spermiogenesis.
      ); however, the function of its N terminus is unknown. Protein phosphorylation can change its overall structure and regulate its functions (
      • Alonso A.
      • Sasin J.
      • Bottini N.
      • Friedberg I.
      • Friedberg I.
      • Osterman A.
      • et al.
      Protein tyrosine phosphatases in the human genome.
      ). Here, we found that HIPK4 could phosphorylate the N terminus of RIMBP3. HIPK4 and RIMBP3 were both expressed in steps 3 to 8 of spermatid development, and RIMBP3 continued to be expressed until step 13 (
      • Zhou J.
      • Du Y.R.
      • Qin W.H.
      • Hu Y.G.
      • Huang Y.N.
      • Bao L.
      • et al.
      RIM-BP3 is a manchette-associated protein essential for spermiogenesis.
      ). From step 8, RIMBP3 was highly expressed in the manchette of elongating spermatids. Phosphorylation of RIMBP3 might be regulated in an early step by HIPK4, with its functions regulated in subsequent stages. Our phosphoproteomic analysis showed a decreased phosphorylation level of HIPK4 in the Hipk4 KO testis. Therefore, we speculated that phosphorylation of RIMBP3 by HIPK4 plays important roles in the functions of the manchette during spermiogenesis.
      In summary, our work demonstrates that phosphorylation is essential for spermatogenesis, and decreased phosphorylation of microtubule-associated proteins might affect microtubule structure and function within this process. Furthermore, as a kinase predominantly expressed in spermatids, HIPK4 might serve as a potential target for male contraception.

      Experimental procedures

      All primer sequences and antibodies are given in Table S1.

      NOA patients’ information

      Approval was granted by research ethics committee of Nanjing Medical University before sample collection. Studies in this work abide by the Declaration of Helsinki principles. Six hundred and twenty NOA patients and 2678 male controls with informed consent were recruited from Clinical Center of Reproductive Medicine in Nanjing. All infertile male subjects were genetically unrelated Han Chinese and screened based on andrological examinations, including a detailed medical history, physical examination, semen analysis, scrotal ultrasound, hormone assessment, karyotyping, and Y chromosome microdeletion screening. Those with a history of cryptorchidism, vascular trauma, orchitis, obstruction of the vas deferens, abnormalities in chromosome number, or microdeletions of the azoospermia factor region on the Y chromosome were excluded. Semen analysis for sperm concentration, motility, and morphology was performed according to the guidelines of World Health Organization (WHO) (2010) (
      • Cooper T.G.
      • Noonan E.
      • von Eckardstein S.
      • Auger J.
      • Baker H.W.
      • Behre H.M.
      • et al.
      World Health Organization reference values for human semen characteristics.
      ). NOA patients lack detectable sperm in their ejaculate; therefore, to differentiate NOA from obstructive azoospermia, only those with idiopathic azoospermia with small and soft testis, normal fructose and neutral alpha glucosidase in seminal plasma were included, and vasectomized patients were excluded. For a reliable diagnosis, each subject was examined twice, only those with an absence of spermatozoa from both replicates were taken to indicate azoospermia. While each control subject had fathered one healthy child or more. A 5-ml whole blood sample was obtained from each participant as a source of genomic DNA for further Sanger sequencing analysis.

      Mice and cell line

      Approval was granted by Animal Ethical and Welfare Committee of Nanjing Medical University before animal studies. To generate Hipk4 KO mice, paired single guide RNAs (sgRNAs, Fig. S2B) were designed to target exon 3 of Hipk4. The oligonucleotides used to generate the sgRNA expression plasmid were annealed and cloned into the pGL3-U6-sgRNA-PGK-puromycin (Addgene plasmid # 51133). Transcription and microinjection of CRISPR/Cas9 were performed in vitro as described previously (
      • Wang D.
      • Cheng L.
      • Xia W.
      • Liu X.
      • Guo Y.
      • Yang X.
      • et al.
      LYPD4, mouse homolog of a human acrosome protein, is essential for sperm fertilizing ability and male fertilitydagger.
      ). Briefly, the Cas9 plasmid (pST1374-NLS-flag-linker-Cas9, Addgene plasmid # 44758) was linearized using AgeI and transcribed using a T7 Ultra Kit (Ambion), pGL3-T7-sgRNA-PGK-puromycin expression vectors were linearized by BsaI and transcribed using the MEGAshortscript Kit in vitro (Ambion). A mixture of Cas9 mRNA and two sgRNAs were injected into the cytoplasm and male pronucleus of the zygote via electroporation. Embryos were implanted into pseudo-pregnant C57BL/6J females according to standard procedures. Founder mice were backcrossed to C57BL/6J. All animal experiments in this study were approved by the Institutional Animal Care and Use Committee of Nanjing Medical University, Nanjing, China. All mice were housed in a specific pathogen-free animal facility under standard conditions.
      The cell line HEK-293T (Homo sapiens) cells were used by experiments for this paper.

      Point mutation plasmid construction

      In vitro point mutation plasmid construction used a ClonExpress II One Step Cloning Kit (Vazyme, C112-02), with specific primers according to the manufacturer’s instructions.

      Histological analysis

      For hematoxylin and eosin staining, the epididymis was fixed with modified Davidson fluid (30% of a 37–40% formaldehyde stock solution, 15% ethanol, 5% glacial acetic acid, and 50% distilled water) and embedded in paraffin. Sections were cut at 5-μm thickness. The sections were dewaxed with xylene, hydrated, stained, dehydrated with ethanol (70%, 80%, 90%, 100%), and blocked with resin. For PAS staining, the sections were deparaffinized, hydrated, stained with PAS reagent, counterstained with hematoxylin, dehydrated, and blocked.

      Electron microscopy

      Samples were fixed at 4 °C overnight with 2.5% glutaraldehyde in 0.2 M cacodylate buffer, washed with 0.2 M cacodylate buffer, dehydrated in a graded series of ethanol, embedded, and polymerized using an automated microwave tissue processor (Leica EM AMW). After polymerization, the samples were cut into 1 mm3 pieces using a LEICA Ultracut UCT ultramicrotome. Ultrathin sections were stained with uranyl acetate and lead citrate and then examined by transmission electron microscope (JEOL, JEM-1010).

      Immunofluorescence and TUNEL assay

      For immunofluorescence, the sections were subjected to heat-induced antigen retrieval with citrate buffer (1.8 mM citric acid, 8.2 mM sodium citrate, pH 6.0) and washed with phosphate-buffered saline (PBS), immersed in 0.1% PBS-Triton X-100, blocked with 5% donkey serum, and incubated with primary antibodies overnight at 4 °C. After incubated with secondary antibodies and Hoechst 33258, the samples were sealed with 50% glycerol. The TUNEL staining assay kit (Vazyme) was used for apoptosis detection according to the manufacturer's instructions.

      Western blotting

      Cells or tissues were homogenized in cold radioimmunoprecipitation assay buffer containing protease inhibitor cocktail (Bimake, B14002). While for sperm samples, the sperm were rinsed with PBS and treated with hypotonic lysate (0.1% SDS, 0.5% Triton X-100) and resuspended with lysis buffer (250 mM Tris-HCl pH 8.8, 50 mM EDTA, 500 mM dithiothreitol, 10% SDS, 50% glycerol). Lysate was centrifuged at 10,000g at 4 °C for 1 h. The supernatant was collected, and the protein concentration was determined using bicinchoninic acid assay.
      Equivalent amounts of proteins were separated using SDS-PAGE, transferred onto nitrocellulose membrane, blocked using 5% skim milk, and incubated with primary antibodies overnight at 4 °C. Secondary antibodies conjugated with were used to visualize specific protein bands.
      The phos-tag gels contained 20 μM Phos-tag (APExBIO, F4002) and 40 μM MnCl2. Before transferring onto nitrocellulose Membrane, phos-tag gel was immersed in transmembrane buffer containing 10 mM EDTA and washed to eliminate manganese ions.

      Sperm analysis

      Sperm from the epididymal tail were incubated in HTF medium (Irvine Scientific) supplemented with 10% fetal bovine serum at 37 °C for 5 min. Sperm motility assessments were performed using the Hamilton Thorne’s Ceros II system (Hamilton-Thorne Research).

      Protein sample preparation, digestion, and TMT labeling

      Testes from three mice from each of the KO and control groups were subjected to protein extraction, digestion, and TMT labeling. In brief, testicular tissues were lysed with protein extraction buffer (8 M urea, 75 mM NaCl, 50 mM Tris, pH 8.2, 1% (vol/vol) EDTA-free protease inhibitor, 1 mM NaF, 1 mM β-glycerophosphate, 1 mM sodium orthovanadate, 10 mM sodium pyrophosphate) (Sigma-Aldrich) followed by sonication and centrifugation, with protein concentrations determined by the Bradford assay (Beyotime). Cysteine residues were reduced using 5 mM dithiothreitol (Thermo Fisher Scientific) and alkylated using 14 mM iodoacetamide (Sigma-Aldrich). Proteins were digested overnight at 37 °C with 5 ng/μl trypsin (Promega) with 1 mM CaCl2 (Sigma-Aldrich). The peptides were subjected to TMT 6-Plex (Thermo Fisher Scientific) labeling, mixed, and desalted using a Waters tC18 Sep-Pak column (Waters).

      LC-MS/MS measurements and data processing

      The TMT-labeled peptide mixture was fractionated using an ACQUITY UPLC M-Class with XBridge BEH C18 column (300 μm × 150 mm, 1.7 μm; 130 Å, Waters) with 30 fractions collected by nonadjacent pooling scheme using a 128-min gradient of 3% buffer B (A: 10 mM ammonium formate, pH 10; B: 100% acetonitrile) for 14 min, 3% to 8% B for 1 min, 8% to 29% B for 71 min, 29% to 41% B for 12 min, 41% to 100% B for 1 min, 100% B for 8 min, 100% to 3% B for 1 min, followed by 20 min at 3% B.
      For phosphoproteomic quantification, the TMT-labeled peptide mixture was fractionated using Agilent 1260 system with XBridge BEH300 C18 column (10 × 250 mm, 5 μm; Waters) into ten fractions using a nonadjacent pooling scheme. TMT-labeled phospho-peptides were enriched through Ti4+-IMAC (immobilized metal affinity chromatography, J&K Scientific). In short, peptides from each fraction were dissolved in loading buffer (80% acetonitrile, 6% trifluoroacetic acid (TFA)) and incubated with IMAC beads for 30 min, washed with wash buffer I (50% acetonitrile, 200 mM NaCl, 6% TFA) and II (30% acetonitrile, 0.1% TFA) for 30 min, respectively, and eluted using elution buffer (10% NH4OH) for 15 min. The eluates of phospho-peptides were dried and desalted using C18 StageTips (Thermo Fisher Scientific).
      For LC-MS/MS analyses, TMT-labeled peptides or enriched TMT-labeled phospho-peptides were resuspended in 0.1% FA and analyzed using an Orbitrap Fusion Lumos Tribird mass spectrometer (Thermo Fisher Scientific) coupled to the Easy-nLC 1200 high performance liquid chromatography, using a 95 min linear gradient (3%–5% buffer B for 5 s, 5%–15% buffer B for 40 min, 15%–28% buffer B for 34.8 min, 28%–38% buffer B for 12 min, 30% buffer to 100% buffer B for 5s and to 100% buffer B for 8 min) (buffer A: 0.1% FA; buffer B: 80% acetonitrile, 0.1% FA). The Orbitrap Fusion Lumos Tribrid Mass Spectrometer was operated in the data-dependent mode. A full survey scan was obtained for the m/z range of 350 to 1500. For both protein expression and phosphorylation quantification, the resolution of HCD MS/MS was 15,000.

      Bioinformatics analysis

      Raw files were searched against the mouse protein sequences obtained from the Universal Protein Resource (UniProt, 2018/07/18) database with MaxQuant software (version 1.6.5.0) (
      • Tyanova S.
      • Temu T.
      • Cox J.
      The MaxQuant computational platform for mass spectrometry-based shotgun proteomics.
      ) under default parameters, except that the optional modification of phosphorylation (S/T/Y) was used for phosphorylation data. The expression level of each phosphorylated site was normalized against the abundances of the corresponding protein. Only type I phosphorylation sites with a localization probability >0.75 were used for the downstream analysis. To obtain an overview of the function of proteins identified by our proteomic analyses, GO enrichment analysis was performed using “clusterProfiler” package (
      • Yu G.
      • Wang L.G.
      • Han Y.
      • He Q.Y.
      clusterProfiler: an R package for comparing biological themes among gene clusters.
      ) in R. The network of biological process was constructed using Cytoscape (version 3.7.1) (
      • Shannon P.
      • Markiel A.
      • Ozier O.
      • Baliga N.S.
      • Wang J.T.
      • Ramage D.
      • et al.
      Cytoscape: a software environment for integrated models of biomolecular interaction networks.
      ). p-values were adjusted considering the false discovery rate using the Benjamini-Hochberg method. A false discovery rate <0.05 was considered significant. Motif analysis was conducted using MOMO (
      • Cheng A.
      • Grant C.E.
      • Noble W.S.
      • Bailey T.L.
      MoMo: discovery of statistically significant post-translational modification motifs.
      ) with a score-threshold of 1.0E-6.

      Co-immunoprecipitation from cell culture extracts and mouse testes

      PCDNA3.1-CMV-3XHA-HIPK4 and PCDNA3.1-CMV-3XFLAG-RIMBP3 plasmids were transfected into HEK293T cells using Lipofectamine 2000. Two days after transfection, the cells were lysed using Pierce IP Lysis Buffer (Thermo Fisher Scientific) supplemented with 1% protease inhibitor cocktail for 1 h at 4 °C and centrifuged at 12,000g for 30 min. The supernatant was removed and precleared using Protein A/G beads (Millipore) at 4 °C for 1 h. Mouse testicular lysates were prepared as described above. Subsequent experimental procedures were performed using the Pierce Co-Immunoprecipitation Kit (Thermo Scientific), and the eluted proteins were subjected to Western blotting.

      Statistical analyses

      All results are presented as the mean ± SD values. The statistical significance of the differences was determined using two-tailed Student’s t test or one way ANOVA with Dunnett’s t test. Each experiment was performed at least three times, and p-values <0.05 were considered significant.

      Data availability

      All MS raw files have been deposited to the PRIDE archive (https://www.ebi.ac.uk/pride/archive) and could be accessed with the dataset identifier PXD019303.

      Supporting information

      This article contains supporting information.

      Conflict of interest

      The authors declare no conflict of interest with contents of this article.

      Acknowledgments

      We would like to thank Dr Yarui Du and Guoliang Xu from SIBCB, CAS, for a kind gift of RIMBP3 antibodies.

      Author contributions

      X. G., J. Z., X. L., Y. G., and Z. H. methodology; X. G., J. Z., and Z. H. supervision; X. L., C. Z., Y. W., R. M., J. Z., Y. C., T. J., C. W., X. Y., C. S., and Y. G. investigation; X. L., T. J., C. W., C. Z., R. M., and Y. W. data curation; X. L., C. Z., R. M., and Y. W. writing-original draft; Y. C., Y. W., C. S., X. G., and Z. H. formal analysis; T. J., C. W., and X. Y. resources; X. G., J. Z., Z. H., X. L., C. Z., R. M., Y. C., Y. W., C. S., T. J., C. W., X. Y., and Y. G. writing-review and editing; X. L., C. Z., and Y. W. visualization; R. M. and Y. C. validation; Z. H., J. Z., and X. G. conceptualization; Z. H., J. Z., and X. G. project administration; Z. H., J. Z., and X. G. funding acquisition.

      Funding and additional information

      This work was supported by grants from the National Key R&D Program of China ( 2021YFC2700200 ) and National Natural Science Foundation of China ( 31530047 , 31871445 , 31501211 , 81971439 and 32071133 ).

      Supporting information

      • Loading ...

      References

        • Irvine D.S.
        Epidemiology and aetiology of male infertility.
        Hum. Reprod. 1998; 13 Suppl 1: 33-44
        • Boivin J.
        • Bunting L.
        • Collins J.A.
        • Nygren K.G.
        International estimates of infertility prevalence and treatment-seeking: potential need and demand for infertility medical care.
        Hum. Reprod. 2007; 22: 1506-1512
        • Sudhakar D.V.S.
        • Shah R.
        • Gajbhiye R.K.
        Genetics of male infertility - present and future: a narrative review.
        J. Hum. Reprod. Sci. 2021; 14: 217-227
        • Jarow J.P.
        • Espeland M.A.
        • Lipshultz L.I.
        Evaluation of the azoospermic patient.
        J. Urol. 1989; 142: 62-65
        • Matsumiya K.
        • Namiki M.
        • Takahara S.
        • Kondoh N.
        • Takada S.
        • Kiyohara H.
        • et al.
        Clinical study of azoospermia.
        Int. J. Androl. 1994; 17: 140-142
        • Berookhim B.M.
        • Schlegel P.N.
        Azoospermia due to spermatogenic failure.
        Urol. Clin. North Am. 2014; 41: 97-113
        • Lee J.Y.
        • Dada R.
        • Sabanegh E.
        • Carpi A.
        • Agarwal A.
        Role of genetics in azoospermia.
        Urology. 2011; 77: 598-601
        • Grootegoed J.A.
        • Baarends W.M.
        • Hendriksen P.J.
        • Hoogerbrugge J.W.
        • Slegtenhorst-Eegdeman K.E.
        • Themmen A.P.
        Molecular and cellular events in spermatogenesis.
        Hum. Reprod. 1995; 10 Suppl 1: 10-14
        • Jan S.Z.
        • Hamer G.
        • Repping S.
        • de Rooij D.G.
        • van Pelt A.M.
        • Vormer T.L.
        Molecular control of rodent spermatogenesis.
        Biochim. Biophys. Acta. 2012; 1822: 1838-1850
        • Yan W.
        Male infertility caused by spermiogenic defects: lessons from gene knockouts.
        Mol. Cell. Endocrinol. 2009; 306: 24-32
        • O'Donnell L.
        Mechanisms of spermiogenesis and spermiation and how they are disturbed.
        Spermatogenesis. 2014; 4e979623
        • Sperry A.O.
        The dynamic cytoskeleton of the developing male germ cell.
        Biol. Cell. 2012; 104: 297-305
        • O'Donnell L.
        • O'Bryan M.K.
        Microtubules and spermatogenesis.
        Semin. Cell Dev. Biol. 2014; 30: 45-54
        • Lehti M.S.
        • Sironen A.
        Formation and function of the manchette and flagellum during spermatogenesis.
        Reproduction. 2016; 151: R43-R54
        • Pasch E.
        • Link J.
        • Beck C.
        • Scheuerle S.
        • Alsheimer M.
        The LINC complex component Sun4 plays a crucial role in sperm head formation and fertility.
        Biol. Open. 2015; 4: 1792-1802
        • Zhou J.
        • Du Y.R.
        • Qin W.H.
        • Hu Y.G.
        • Huang Y.N.
        • Bao L.
        • et al.
        RIM-BP3 is a manchette-associated protein essential for spermiogenesis.
        Development. 2009; 136: 373-382
        • Wang Z.
        • Cole P.A.
        Catalytic mechanisms and regulation of protein kinases.
        Methods Enzymol. 2014; 548: 1-21
        • Wong C.H.
        • Cheng C.Y.
        Mitogen-activated protein kinases, adherens junction dynamics, and spermatogenesis: a review of recent data.
        Dev. Biol. 2005; 286: 1-15
        • Qi L.
        • Liu Z.
        • Wang J.
        • Cui Y.
        • Guo Y.
        • Zhou T.
        • et al.
        Systematic analysis of the phosphoproteome and kinase-substrate networks in the mouse testis.
        Mol. Cell. Proteomics. 2014; 13: 3626-3638
        • Liu X.
        • Busby J.
        • John C.
        • Wei J.
        • Yuan X.
        • Lu M.L.
        Direct interaction between AR and PAK6 in androgen-stimulated PAK6 activation.
        PLoS One. 2013; 8e77367
        • Kueng P.
        • Nikolova Z.
        • Djonov V.
        • Hemphill A.
        • Rohrbach V.
        • Boehlen D.
        • et al.
        A novel family of serine/threonine kinases participating in spermiogenesis.
        J. Cell Biol. 1997; 139: 1851-1859
        • Zhao H.
        • Li T.
        • Zhao Y.
        • Tan T.
        • Liu C.
        • Liu Y.
        • et al.
        Single-cell transcriptomics of human oocytes: environment-driven metabolic competition and compensatory mechanisms during oocyte maturation.
        Antioxid. Redox Signal. 2019; 30: 542-559
        • Ogurtsov A.Y.
        • Marino-Ramirez L.
        • Johnson G.R.
        • Landsman D.
        • Shabalina S.A.
        • Spiridonov N.A.
        Expression patterns of protein kinases correlate with gene architecture and evolutionary rates.
        PLoS One. 2008; 3: e3599
        • van der Laden J.
        • Soppa U.
        • Becker W.
        Effect of tyrosine autophosphorylation on catalytic activity and subcellular localisation of homeodomain-interacting protein kinases (HIPK).
        Cell Commun. Signal. 2015; 13: 3
        • Guo J.
        • Grow E.J.
        • Mlcochova H.
        • Maher G.J.
        • Lindskog C.
        • Nie X.
        • et al.
        The adult human testis transcriptional cell atlas.
        Cell Res. 2018; 28: 1141-1157
        • Wang M.
        • Liu X.
        • Chang G.
        • Chen Y.
        • An G.
        • Yan L.
        • et al.
        Single-cell RNA sequencing analysis reveals sequential cell fate transition during human spermatogenesis.
        Cell Stem Cell. 2018; 23: 599-614.e4
        • Crapster J.A.
        • Rack P.G.
        • Hellmann Z.J.
        • Le A.D.
        • Adams C.M.
        • Leib R.D.
        • et al.
        HIPK4 is essential for murine spermiogenesis.
        Elife. 2020; 9e50209
        • Cheng C.Y.
        • Mruk D.D.
        A local autocrine axis in the testes that regulates spermatogenesis.
        Nat. Rev. Endocrinol. 2010; 6: 380-395
        • Mruk D.D.
        • Cheng C.Y.
        Cell-cell interactions at the ectoplasmic specialization in the testis.
        Trends Endocrinol. Metab. 2004; 15: 439-447
        • Vogl A.W.
        • Young J.S.
        • Du M.
        New insights into roles of tubulobulbar complexes in sperm release and turnover of blood-testis barrier.
        Int. Rev. Cell Mol. Biol. 2013; 303: 319-355
        • Ubersax J.A.
        • Ferrell Jr., J.E.
        Mechanisms of specificity in protein phosphorylation.
        Nat. Rev. Mol. Cell Biol. 2007; 8: 530-541
        • Cox J.
        • Mann M.
        MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification.
        Nat. Biotechnol. 2008; 26: 1367-1372
        • Schwartz D.
        • Gygi S.P.
        An iterative statistical approach to the identification of protein phosphorylation motifs from large-scale data sets.
        Nat. Biotechnol. 2005; 23: 1391-1398
        • Ann E.J.
        • Kim M.Y.
        • Yoon J.H.
        • Ahn J.S.
        • Jo E.H.
        • Lee H.J.
        • et al.
        Tumor suppressor HIPK2 regulates malignant growth via phosphorylation of Notch1.
        Cancer Res. 2016; 76: 4728-4740
        • Yamada D.
        • Perez-Torrado R.
        • Filion G.
        • Caly M.
        • Jammart B.
        • Devignot V.
        • et al.
        The human protein kinase HIPK2 phosphorylates and downregulates the methyl-binding transcription factor ZBTB4.
        Oncogene. 2009; 28: 2535-2544
        • de Kretser D.M.
        Male infertility.
        Lancet. 1997; 349: 787-790
        • Don J.
        • Stelzer G.
        The expanding family of CREB/CREM transcription factors that are involved with spermatogenesis.
        Mol. Cell. Endocrinol. 2002; 187: 115-124
        • Westlander G.
        Utility of micro-TESE in the most severe cases of non-obstructive azoospermia.
        Ups. J. Med. Sci. 2020; 125: 99-103
        • Cannarella R.
        • Condorelli R.A.
        • Duca Y.
        • La Vignera S.
        • Calogero A.E.
        New insights into the genetics of spermatogenic failure: a review of the literature.
        Hum. Genet. 2019; 138: 125-140
        • Zafar A.
        • Eqani S.A.
        • Bostan N.
        • Cincinelli A.
        • Tahir F.
        • Shah S.T.
        • et al.
        Toxic metals signature in the human seminal plasma of Pakistani population and their potential role in male infertility.
        Environ. Geochem. Health. 2015; 37: 515-527
        • Alfano M.
        • Ferrarese R.
        • Locatelli I.
        • Ventimiglia E.
        • Ippolito S.
        • Gallina P.
        • et al.
        Testicular microbiome in azoospermic men-first evidence of the impact of an altered microenvironment.
        Hum. Reprod. 2018; 33: 1212-1217
        • Cervan-Martin M.
        • Castilla J.A.
        • Palomino-Morales R.J.
        • Carmona F.D.
        Genetic landscape of nonobstructive azoospermia and new perspectives for the clinic.
        J. Clin. Med. 2020; 9: 300
        • Ben Khelifa M.
        • Coutton C.
        • Zouari R.
        • Karaouzene T.
        • Rendu J.
        • Bidart M.
        • et al.
        Mutations in DNAH1, which encodes an inner arm heavy chain dynein, lead to male infertility from multiple morphological abnormalities of the sperm flagella.
        Am. J. Hum. Genet. 2014; 94: 95-104
        • Neesen J.
        • Kirschner R.
        • Ochs M.
        • Schmiedl A.
        • Habermann B.
        • Mueller C.
        • et al.
        Disruption of an inner arm dynein heavy chain gene results in asthenozoospermia and reduced ciliary beat frequency.
        Hum. Mol. Genet. 2001; 10: 1117-1128
        • Kierszenbaum A.L.
        • Tres L.L.
        The acrosome-acroplaxome-manchette complex and the shaping of the spermatid head.
        Arch. Histol. Cytol. 2004; 67: 271-284
        • Russell L.D.
        • Russell J.A.
        • MacGregor G.R.
        • Meistrich M.L.
        Linkage of manchette microtubules to the nuclear envelope and observations of the role of the manchette in nuclear shaping during spermiogenesis in rodents.
        Am. J. Anat. 1991; 192: 97-120
        • Mendoza-Lujambio I.
        • Burfeind P.
        • Dixkens C.
        • Meinhardt A.
        • Hoyer-Fender S.
        • Engel W.
        • et al.
        The Hook1 gene is non-functional in the abnormal spermatozoon head shape (azh) mutant mouse.
        Hum. Mol. Genet. 2002; 11: 1647-1658
        • Kierszenbaum A.L.
        Intramanchette transport (IMT): managing the making of the spermatid head, centrosome, and tail.
        Mol. Reprod. Dev. 2002; 63: 1-4
        • Alonso A.
        • Sasin J.
        • Bottini N.
        • Friedberg I.
        • Friedberg I.
        • Osterman A.
        • et al.
        Protein tyrosine phosphatases in the human genome.
        Cell. 2004; 117: 699-711
        • Cooper T.G.
        • Noonan E.
        • von Eckardstein S.
        • Auger J.
        • Baker H.W.
        • Behre H.M.
        • et al.
        World Health Organization reference values for human semen characteristics.
        Hum. Reprod. Update. 2010; 16: 231-245
        • Wang D.
        • Cheng L.
        • Xia W.
        • Liu X.
        • Guo Y.
        • Yang X.
        • et al.
        LYPD4, mouse homolog of a human acrosome protein, is essential for sperm fertilizing ability and male fertilitydagger.
        Biol. Reprod. 2020; 102: 1033-1044
        • Tyanova S.
        • Temu T.
        • Cox J.
        The MaxQuant computational platform for mass spectrometry-based shotgun proteomics.
        Nat. Protoc. 2016; 11: 2301-2319
        • Yu G.
        • Wang L.G.
        • Han Y.
        • He Q.Y.
        clusterProfiler: an R package for comparing biological themes among gene clusters.
        OMICS. 2012; 16: 284-287
        • Shannon P.
        • Markiel A.
        • Ozier O.
        • Baliga N.S.
        • Wang J.T.
        • Ramage D.
        • et al.
        Cytoscape: a software environment for integrated models of biomolecular interaction networks.
        Genome Res. 2003; 13: 2498-2504
        • Cheng A.
        • Grant C.E.
        • Noble W.S.
        • Bailey T.L.
        MoMo: discovery of statistically significant post-translational modification motifs.
        Bioinformatics. 2019; 35: 2774-2782