Epigenetic regulation by DNA methylation, histone modifications and chromatin remodeling complexes in controlling spermatogenesis and their dysfunction with male infertility

Abstract

As key factors of cellular development, epigenetic regulation can accurately control gene expression through multiple manners, e.g., DNA methylation, histone modification, and chromatin remodeling complexes (CRCs). Epigenetic factors play pivotal roles in various kinds of cell processes, including cell proliferation, differentiation, and apoptosis, and diseases may be resulted from their dysfunction. Spermatogenesis refers to the complex process by which spermatogonial stem cells (SSCs) self-renew and differentiate into the differentiating spermatogonia that further develop to spermatocytes and mature spermatids. Significantly, epigenetic regulation has recently been shown to mediate fate determinations of SSCs to ensure normal spermatogenesis. Interestingly, much progress has recently been made in epigenetic regulation and their dysfunction in controlling spermatogenesis and male infertility, respectively. In this review, we address the dynamic expression patterns, functions and mechanisms of DNA methylation, histone modification, and CRCs in mediating the development of SSCs and spermatogenesis, and we also discuss the association between epigenetic dysfunction and male infertility. We further point out the perspectives in epigenetic regulation on human spermatogenesis. Our review on the in-depth analysis of epigenetic regulatory mechanisms in normal and abnormal spermatogenesis not only helps us better understand the etiology of male infertility but also provides novel targets for treating this disease.

Introduction

The infertility rate of couples has increased to 12%–18% worldwide, And male factors contribute to 40%−50% of this disease [1, 2]. Spermatogenesis comprises three main stages, namely, the mitosis of SSCs, meiotic division of spermatocytes to halve chromosome numbers while ensuring genetic diversity through homologous recombination, and spermiogenesis of round spermatids to undergo dramatic chromatin remodeling with histone replacement by protamine to achieve nuclear compaction. The precise regulation of spermatogenesis relies on synergistic interactions between genetic and epigenetic factors. Notably, spermatogenesis failure is resulted from epigenetic and genetic dysregulation [3–5], which underscores the importance of the epigenetic regulation in male germ cell development.

Epigenetic regulation involves heritable and reversible changes in gene expression without altering the DNA sequence. Key epigenetic regulatory factors include DNA methylation, histone modifications (phosphorylation, ubiquitination, methylation, and acetylation), CRCs, and non-coding RNAs (e.g., miRNAs, lncRNAs, and circRNAs) [6–8]. Recently advancements in multi-omics and single-cell technologies have unveiled molecular mechanisms by epigenetic regulation in spermatogenesis. For example, PRMT5 deficiency increases H3K9me2 and H3K27me2 levels and alters chromatin state of PLZF, which leads to SSC developmental defect and spermatogenesis disorder [9]. Similarly, histone methyltransferase Suv39h null mice exhibit spermatogenic failure with nonhomologous chromosome association [10]. We have previously discussed the regulation of non-coding RNAs in SSC fate determinations and spermatogenesis [11–13]. In this review, we address epigenetic regulation mediated by DNA methylation, histone modifications and CRCs in controlling spermatogenesis, and we also discuss associations between the dysfunction of epigenetic factors and male infertility. This review provides novel insights into the epigenetic mechanisms underlying spermatogenesis and etiology of male infertility.

DNA methylation in regulating spermatogenesis and its dysfunction in male infertility

Molecular basis of DNA methylation

DNA methylation refers to the process by which a methyl group, derived from S-adenosyl methionine (SAM), is covalently attached to specific DNA bases catalyzed by DNA methyltransferases (DNMTs). Since DNA methylation was discovered in 1925, its dynamic distribution, functions, And regulatory mechanisms across diverse cell types And species have been extensively characterized. In mammals, DNA methylation occurs at the 5th carbon of cytosines within CpG dinucleotides (5-methylcytosine, 5mC) in the 5’UTR and gene bodies [14]. Under normal physiological conditions, 70%−90% of CpG sites are methylated. In contrast, CpG islands, at the regions with high G + C content (> 50%) and dense CpG clustering remain largely unmethylated [15, 16]. These CpG islands are frequently located near promoter regions or transcriptional start sites (TSS), reflecting their critical roles in regulating gene expression.

The distribution of DNA methylation is precisely controlled by DNMTs and demethylating enzymes (Table 1). The DNMT family includes DNMT1, DNMT3A, DNMT3B, DNMT3C, and DNMT3L (the catalytically inactive cofactor). During the early embryogenesis and gametogenesis, the genome undergoes waves of global demethylation followed by de novo methylation, a process driven primarily by DNMT3A and DNMT3B [17, 18]. DNMT3L facilitates this process by enhancing the enzymatic activity of DNMT3A/B. Conversely, DNMT1 acts as the maintenance methyltransferase, ensuring the fidelity of methylation patterns during DNA replication by methylating hemimethylated CpG sites on nascent DNA strands.

Table 1.

DNA methylation functions and its dysfunction with male infertility

	Enzymes or Proteins	Functions	Loss of function
Writers	DNMT1	Maintenance of DNA methyltransferase	Apoptosis of germline stem cells [19]; Hypogonadism and meiotic arrest [20]
	DNMT3A	De novo DNA methyltransferase	Abnormal spermatogonial function [21]
	DNMT3B	De novo DNA methyltransferase	Fertility and no distinctive phenotype [22]
	DNMT3C	De novo DNA methyltransferase	Severe defect in DSB repair and homologous chromosome synapsis during meiosis [23]
	DNMT3L	De novo DNA methyltransferase cofactor	Decrease in quiescence SSCs [24]
Readers	MBD1	Methylated DNA binding protein	/
	MBD2	Methylated DNA binding protein	/
	MBD3	Methylated DNA binding protein	/
	MBD4	Methylated DNA binding protein	/
	MeCP2	Methylated DNA binding protein	/
Erasers	TET1	DNA demethylation	Fertile [25]
	TET2	DNA demethylation	Fertile [25]
	TET3	DNA demethylation	/

Generally, DNA methylation with regulatory elements correlates with transcriptional repression, whereas DNA demethylation is associated with transcriptional reactivation. The mechanism underlying DNA methylation is controversial. The prevailing view is that DNA methylation can alter chromatin accessibility, impede transcription factor binding, and thereby inhibit gene transcription [26]. Readers of DNA methylation, e.g., MBD protein family, recognize methylated DNA and recruit complexes containing histone deacetylases (HDACs) to regions of high methylation, while this recruitment results in histone deacetylation and subsequent transcriptional repression [27]. The interactions between MBD family proteins and histone methylation modifiers have been revealed. Histone H3 methyltransferase Suv39h1 And heterochromatin protein 1 (HP1) directly interact with MBD1, and Suv39h1 enhances MBD1-mediated transcriptional inhibition via MBD domain [28]. However, DNA methylation does not universally suppress gene transcription. It can also induce transcriptional activation by stabilizing RNA polymerase II elongation [29]. Furthermore, while DNA methylation typically occurs at CpG site, non-CpG methylation, e.g., CAG or CAA, has been observed particularly in stem cells [30, 31]. Although DNMTs are relatively conserved across cells and species, DNA methylation exhibits diverse distribution patterns and functions. It would be interesting to answer the questions on how DNA methylation regulates gene expression and what are the interactions of DNA methylation with other epigenetic factors in stem cells.

DNA methylation dynamics in spermatogenesis

DNA methylation plays pivotal roles in germ cell development, and its dynamics are tightly regulated during embryonic and postnatal stages [32]. Mouse primordial germ cells (mPGCs), the precursor cells of SSCs, undergo genome-wide DNA demethylation when they migrate to the gonads between embryonic days 8.5 (E8.5) And 13.5 (E13.5). During this period, 5mC levels in mPGCs are decreased to approximately 16.3%, which is significantly lower than the 75% of 5mC abundance in embryonic stem cells (ESCs) [33, 34]. This hypomethylation is driven by the repression of de novo methyltransferases DNMT3A/B [35, 36] and elevated activity of DNA demethylation factors (e.g., TET1) [37] leading to the erasure of methylation at transposable elements and imprinted loci. Subsequently, from E13.5 to E16.5, the de novo DNA methylation is gradually established until birth [38].

DNA methylation state is evolutionarily conserved in mice and humans [39]. Human PGCs (hPGCs) undergo global demethylation during gonadal colonization, and they have minimal DNA methylation by week10-11 with completion of sex differentiation [40, 41]. At 13.5 days post-fertilization, the male PGCs develop into prospermatogonia (PSG). After birth, a subset of PSG differentiates into spermatogonia [42], while other PSG develop into SSCs with the abilities of self-renewal and differentiation [42]. DNA methylation is restored during embryonic and prospermatogonial development [40, 41], as we illustrated in Fig. 1.

Fig. 1.

Distribution patterns of DNA methylation-related catalytic enzymes And histone modifications in prospermatogonial development. Schematic diagram displays the abundance of different histone modifications, 5mC and DNA methylation-related enzymes during embryonic and prospermatogonial period. The darker the color, the higher the expression level

Throughout spermatogenesis, there is a relative increase of DNA methylation. However, DNA methylation patterns differ in male germ cell types. Compared to the undifferentiated spermatogonia (Thy1⁺ cells, enriched for SSCs), the differentiating spermatogonia (c-Kit⁺ cells) exhibit higher levels of DNMT3A and DNMT3B [43], suggesting that DNA methylation may regulate the SSCs-to-differentiating spermatogonia transition. Genome-wide DNA methylation increases during this transition [44], whereas DNA demethylation occurs in preleptotene spermatocytes [45, 46]. DNA methylation gradually rises at leptotene and zygotene stages, and it reaches at a high level in pachytene spermatocytes [45], as we indicated in Fig. 2. The conservation of DNA methylation in mammals underscores its potential roles in regulating human SSC fate decisions and spermatogenesis. We have recently revealed that DNA methylation is involved in reprogramming human Sertoli cells to functional SSCs. Notably, single-cell DNA methylomics in human testes could explore the functions and mechanisms of DNA methylation by different subpopulations in regulating human spermatogenesis.

Fig. 2.

Distribution patterns of DNA methylation and histone modifications in the process of spermatogenesis. The scheme illustrates the relative abundance of DNA methylation and key histone modifications during spermatogenesis. H3K9ac is highly abundant in SSCs, while H3K4me3 peaks in leptotene spermatocytes. During zygotene stage, H4K5ac/H4K8ac/H4K12ac acetylation reaches their maximal levels. In contrast, DNA methylation becomes the predominantly epigenetic modification during meiotic prophase I, and it persists in subsequent stages. Notes: ‘Meiosis I’ stage encompasses prophase I sub-stages (leptotene, zygotene, pachytene, and diplotene). Meiosis II and post-diplotene stages of Meiosis I are not shown in this figure due to unavailable data on DNA methylation and histone modifications in these stages

Association of DNA methylation dysfunction with male infertility

Emerging evidence highlights the strong correlation between dysfunctional DNA methylation and the impaired spermatogenesis in both mice and humans, as we summarized in Table 1. Comparative analyses of testicular biopsies from patients with obstructive azoospermia (OA) with normal spermatogenesis and non-obstructive azoospermia (NOA) have revealed differential DNMT expression profiles [47]. Specifically, in NOA patients, including spermatocyte (SC) arrest, round spermatid (RS) arrest or Sertoli cell-only syndrome (SCOS), the expression levels of DNMT1 and DNMT3A are significantly lower compared to patients with hypospermatogenesis [47]. DNMT1 and DNMT3A are remarkably reduced in spermatogonia and spermatocytes from SC arrest and RS arrest [47], while DNMT3B is decreased in RS arrest and SCOS patients [47]. These changes are associated with global hypomethylation in the testes of NOA patients [47, 48], indicating that aberrant DNA methylation might contribute to spermatogenesis failure. Several studies have linked the reduced DNA methylation in spermatids to the decreased semen parameters in infertile men [49–52]. However, whether abnormal DNA methylation is a cause or consequence of spermatogenesis disorder remains unclear.

Interestingly, not all cases of spermatogenic arrest are accompanied by global methylation loss. For example, altered expression of DNMT3B has been detected in the testes of NOA patients without apparently global hypomethylation [53]. Instead, accumulating evidence highlights the importance of locus-specific DNA methylation aberration in male infertility. In Klinefelter syndrome patients, male germ cells exhibit normal transcription but have the impaired DNA methylation [54]. Among imprinting-associated loci, aberrant methylation at the imprinted gene H19 in spermatids is highly associated with idiopathic male infertility risk and represents a potential biomarker of spermatogenesis failure [55–57]. Notably, hypomethylation at CTCF-binding site 6 within the H19 DMR (differentially methylated region) disrupts CTCF-mediated chromatin looping, which elevates idiopathic oligospermia risk [58]. Aberrant methylation at CTCF/cohesin-dependent boundaries can disintegrate topologically associating domains (TADs) in germline chromatin [59], and this architectural perturbation permits ‘enhancer hijacking’ [60]. While deletion of TAD boundary may have the negative impact on spermatogenesis [61], the precise identity, dynamic regulation, and functional relevance of these elements in germ cells remain to be further defined.

In addition, similar allele-specific methylation error at the maternally imprinted gene MEST has also been linked to spermatogenesis disorder [62]. Furthermore, abnormal methylation at spermatogenesis-associated genes, e.g., SMARCB1, DDX27, CDH1, HOXB1, GATA3, and TCERG1L, has been implicated in disrupted germline development and represents potential epigenetic biomarkers of male infertility [50, 52]. These findings underscore the importance of precise, locus-specific DNA methylation in orchestrating spermatogenesis. The highly localized methylation patterns require tight spatiotemporal control, which is primarily governed by the coordinated actions of DNMTs and demethylation enzymes. Therefore, understanding the function and regulation of the DNMTs is essential to elucidate how locus-specific methylation landscapes are established and maintained during germline development. Future research should aim to dissect these regulatory networks across spatial, temporal, and functional dimensions to determine how methylation defects contribute to spermatogenic arrest.

Besides the imprinted genes, aberrant methylation of retrotransposons critically contributes to male infertility. During spermatogenesis, DNMT3C and the PIWI/piRNA pathway synergistically silence retrotransposons to maintain genomic stability [63, 64]. Dnmt3c deficiency in mice triggers germ cell apoptosis via retrotransposon insertions [63]. In Miwi2 deficient mice, long interspersed elements (LINEs) and intracisternal A particle (IAP) elements are demethylated, which is potentially linked to meiotic defect [65]. Clinical studies reveal that hypermethylation of MAEL promoter leads to the increased expression of LINE-1 and it is associated with spermatogenic failure [66].

As a primary regulator for maintaining methyltransferase activity, DNMT1 preserves DNA methylation patterns during SSC proliferation. Dnmt1 knockdown triggers apoptosis of SSCs [19], and DNMT1 mutation causes hypogonadism and meiotic arrest [20]. These phenotypes suggest that DNMT1-mediated epigenetic stability is essential for SSC self-renewal and entry into meiosis. Critical for SSC differentiation, Dnmt3a knockout mice exhibit progressive germ cell depletion due to SSC pool exhaustion [21, 22], potentially mimicking NOA progression in humans. This defect may be attributed to the disrupted paternal imprinting at loci H19 and Rasgrf1 in spermatogonia, highlighting the importance of DNMT3A in establishing epigenetic marks required for germline lineage progression [22]. The phenotypes observed in Dnmt3l mutation are strikingly similar to those in Dnmt3a conditional mutation. Dnmt3l mutant mice are infertile [67] and they have fewer SSCs, which leads to spermatogenic disorder [24] probably through interaction with DNMT3A and DNMT3B [68]. DNMT3L functions primarily as a regulator of imprinted gene rather than a DNA methyltransferase, since the imprinted loci methylation is lost by its deficiency [68, 69]. Another possibility for the sterility in Dnmt3l^−/− mice may be associated with retrotransposon reactivation [70, 71]. DNMT3C (de novo DNA methyltransferase) shares approximately 70% homology with DNMT3B, and notably, Dnmt3c^−/− mice are infertile probably by abnormal methylation of retrotransposons [63].

In contrast, the TET family including TET1, TET2, and TET3, catalyzes DNA demethylation by oxidizing 5mC to 5-hydroxymethylcytosine (5hmC). TET family genes are predominantly expressed in pachytene spermatocytes and round spermatids [72], which is consistent with their roles in mediating meiosis and spermiogenesis. Tets expression levels correlate with semen quality, and reduced TETs expression or activity has been implicated in male infertility [72]. In Tet1/Tet2 double knockout mice, 5hmC level is suppressed at imprinted loci without apparent reproductive defects [25]. As such, advanced technologies are required to explore the specific roles and mechanisms of TET family in controlling spermatogenesis and male fertility.

In summary, alterations in DNA methylation, whether affecting the genome globally or at specific sites/loci, both can contribute to male infertility. Future efforts should focus on the locus-specific regulatory networks to determine if DNA methylation anomalies define the subtypes of male infertility. Additionally, identification of diagnostic methylation signatures will enable precision mapping of the etiology for male infertility.

Histone modifications in the regulation of spermatogenesis and male fertility

Classification and functions of histone modifications

Histones are the principal protein components of chromatin, and they are composed of H2A, H2B, H3, and H4, which assemble with DNA into the nucleosome structure. Chromatin structure, nucleosome dynamics, and gene transcription are regulated by histones. Histones undergo diverse post-translational modifications (PTMs), which affects their interaction with DNA. Certain modifications disrupt histone-DNA interactions, cause nucleosome unwinding, enable DNA to bind transcription complexes, and lead to gene activation [73]. Conversely, modifications that enhance histone-DNA interactions produce tightly packed chromatin structure called heterochromatin, which results in gene silencing [73]. Therefore, histone modifications alter chromatin structure and regulate gene transcription. In addition to directly altering the structure by changing histone charge, histone modifications can indirectly affect DNA compaction by recruiting DNA-binding proteins [74].

At least eight types of histone modifications, including acetylation, methylation, phosphorylation, ubiquitination, N-acetylglucosamine glycosylation, citrullination, crotonylation, and isomerization, have been identified and characterized. Each histone modification involves the addition or removal of specific chemical groups to/from histone amino acid residue by the definite enzymes and cofactors (Table 2). A balance between specific enzymes maintains the normal transcriptional status of local genomic regions. The mutations or dysregulation of these enzymes or cofactors are associated with various diseases, including male infertility [75].

Table 2.

Histone modifications and their dysfunction with male infertility

Modifications	Writers	Erasers	Functions	Loss of function
H3K4me1/2/3	PRDM9, SET1A/B, KMT2A/B (MLL1/2)	KDM1A/B, KDM2A/B, KDM5A/B/C/D	Transcriptional activation	Prdm9 knockout results in meiotic arrest [76, 77]; Kmt2b is essential for SSCs-to-progenitors transition [78]; Kdm1a is required for SSC and progenitor maintenance and differentiation [79];
H3K9me1/2/3	G9a, SUV39H1, SETDB1	KDM1A, KDM3A (JMJD1A), KDM3B/C, KMD4A/B/C/D	Transcriptional activation (K9me1), repression (K9me2/3)	G9a mutant mice display sterility [80]; Suv39h1 knockout doesn’t affect fertility of mice [10]; Setdb1 deficiency leads to apoptosis of SSCs and meiotic arrest of spermatocytes [81]; Jhdm2a deficient mice are infertile with the impaired postmeiotic chromatin condensation [82].
H3K27me1/2/3	PRC2, EZH1/2	KDM6A/B (UTX/JMJD3), KDM7A, PHF8	Transcriptional activation (K27me1), Transcriptional silencing (K27me2/3)	Kdm6a knockout mice assume fertility [83].
H3K36me1/2/3	PRDM9, NSD1–3, SETD2/3, ASH1L, SETMAR, SMYD2	KDM2A/B, KDM4A/B/C/D, JHDM1A	Transcriptional repression	Nsd1 deficiency shows the severe defect in spermatogenesis [84]; Setd2 knockout causes aberrant spermiogenesis with acrosomal malformation [85].
H3K79me1/2/3	DOTL1		Transcriptional activation	Mice lacking Dot1l fail to maintain the pool of SSCs [86]
H4K20me1/2/3	PR-Set7, SUV4-20H1/2	LSD1n, DPY-21	Transcriptional activation (K20me1), Transcriptional silencing (K20me2/3)	/
H3K9ac	KAT2A (GCN5), KAT2B (PCAF), KAT6A/7/14, TIP60	SIRT1/6, HDAC1/2/3/8	Transcriptional activation	Kat2a mutation disrupts male fertility [87]; Sirt1 deficiency causes the reduced fertility [88]; Sirt6−/− male mice are infertile and the spermatogenesis is arrested at the elongated spermatid stage [89]; Tip60−/− mice appear to be arrested at the round spermatids [90].
H3K14ac	KAT5/6A/6B/7	SIRT1, HDAC1/3/6/8	Transcriptional activation	Hdac3 or Hdac6-deficient mice are fertile [91, 92]
H3K18ac	KAT3A/3B	SIRT7, HDAC1	Transcriptional activation	/
H3K23ac	KAT6A/6B	HDAC1	Transcriptional activation	/
H3K27ac	KAT3A/3B	HDAC1	Transcriptional activation	/
H4K5ac	KAT1/3A/3B/5/7/8/14	HDAC3/6	Transcriptional activation	Hdac3 or Hdac6-deficient mice are fertile [91, 92]
H4K8ac	KAT5/7/8		Transcriptional activation	/
H4K12ac	KAT1/5/7/14	HDAC2/3/6	Transcriptional activation	Hdac3 or Hdac6-deficient mice are fertile [91, 92]
H4K16ac	KAT5/8/14, TIP60	HDAC6, SIRT1/2/3	Transcriptional activation	Hdac6-deficient mice are fertile [92]; Tip60−/− mice appear to be arrested at the round spermatids [90].

Histone acetylation, one of the earliest identified PTMs affecting transcriptional regulation, adds negative changes to lysine residues within the N-terminal tail protruding from the nucleosome. These charges repel the negatively charged DNA. This interaction results in chromatin relaxation (euchromatin), which facilitates transcription factor binding and RNA polymerase II recruitment. Acetylated lysine residues also serve as docking sites for specific proteins, such as CRCs, which promotes open chromatin formation [93]. This process is dynamically regulated by histone acetyltransferases (HATs) and histone deacetylases (HDACs). HATs catalyze acetyl group of lysine residues on histones H3 and H4, while HDACs remove them [94]. Besides transcriptional activation, histone acetylation participates in DNA repair and apoptosis, and its dysregulation is linked to various kinds of diseases [95, 96].

Histone methylation occurs on lysine and arginine residues of H3 and H4. Lysine residues are mono-methylation (me1), di-methylation (me2), or tri-methylation (me3), while arginine can be mono-methylated (me1), symmetrically di-methylated (me2s), or asymmetrically di-methylated (me2a). Unlike histone acetylation, methylated histone residues exhibit context-dependent effect on transcription. Methylation at H3K4 [97] and H3K36 [98, 99] is involved in transcriptional activation. In contrast, H3K9 [100, 101] and H3K27 [98] methylation correlates with transcriptional inhibition. Mechanistically, histone methylation regulates transcription through recruitment of specific transcription factors, interaction with initiation and elongation factors, and influence on RNA processing [102]. Both arginine and lysine methylation are catalyzed by histone methyltransferases (HMTs) and reversed by histone demethylases (HDMs). Additionally, modified histones are recognized by specific reader proteins. These processes maintain a dynamic equilibrium under normal circumstances.

Histone ubiquitination primarily occurs on H2A and H2B, and it can mediate genome stability [103]. H2BK120ub promotes transcriptional elongation [104], while H2AK119ub controls PRC1-dependent gene silencing [105]. Histone phosphorylation typically emerges at serine, threonine, and tyrosine residues, predominantly within N-terminal tails, and it can induce chromatin decondensation and recruit DNA repair proteins [106]. With the development of advanced technologies on histone modifications, a range of new modifications have been found, such as citrullination and crotonylation [107, 108]. However, their specific roles and regulatory mechanisms of these histone modifications remain to be explored. Interestingly, histone modifications engage in extensive crosstalk. For instance, H3K27ac antagonizes PRC2-mediated H3K27me3 deposition [109], while DNA damage induces γH2AX and H2BK120 ubiquitination to recruit repair complexes [110]. Dysregulation of these networks with histone modifications has been implicated in diverse disorders, including male infertility [111].

Histone modification profiles during spermatogenesis

Spermatogenesis is precisely orchestrated by histone PTMs, which regulate chromatin accessibility, transcriptional program, and genome repackaging. These dynamic modifications of histone are essential for the development of PGCs and prospermatogonia (Fig. 1), SSC self-renewal and differentiation, proper meiotic of spermatocytes, and post-meiotic development. In E8.5 PGCs of mice, the reduction of H3K9me2 coincides with the increase of H3K4me2/3 and H3K27me3, establishing a permissive chromatin state for pluripotency maintenance [112, 113]. During gonadal colonization (E11.5), the heterochromatin-related histone modifications, including H3K9me3, H3K27me3, and H3K9ac, are lost, which facilitates genome-wide DNA demethylation and epigenetic resetting [112]. During E11.5 to E12.5, these heterochromatic marks are restored to safeguard genomic integrity [112]. Human PGC development exhibits analogous changes in H3K9me2 and H3K27me3 prior to gonadal migration (week6-8) [114, 115], reflecting the conserved reprogramming principles. However, there is a distinct dynamic of H3K27me3 during the late hPGC development. Following the arrival to genital ridges, hPGCs show the remarkable decline in H3K27me3 level [115, 116], whereas H3K9ac, H3K4me2, and H3K4me3 increase gradually until week 13 [115, 116], as we summarized in Fig. 1. Regarding self-renewal and differentiation of SSCs, high levels of H3K9ac, H3K18ac, and H3K23ac are present in SSCs and differentiating spermatogonia [117]. In contrast, H3K4me, H3K4me2, and H3K4me3 levels are increased during their differentiation into type B spermatogonia [117, 118]. Conversely, H3K9me3 is decreased dramatically during SSC differentiation [119]. This shift highlights the antagonistic regulation by activating H3K4me3 and repressive H3K9me3 modifications in SSC fate decisions.

Stage-specific PTMs coordinate chromatin reorganization during meiosis and spermiogenesis. Most histone methyltransferases are highly expressed in spermatocytes and round spermatids, with their expression gradually decreasing during spermatid elongation, particularly in spermatozoa [120]. Similarly, overall histone acetyltransferase levels are decreased during late spermiogenesis [120]. H3K4me, H3K4me2, and H3K4me3 levels enhance significantly in leptotene spermatocytes, diminish gradually in pachytene spermatocytes [117, 118], recover moderately in round spermatids, and remain low in elongated spermatids [117, 118]. H3K27me3 and H3K23ac are present at low levels in leptotene spermatocytes and persist through zygotene and pachytene spermatocytes [117]. Unlike H3K27me3, H3K23ac expression increases in diplotene spermatocytes but remains at a low level in spermatids [117]. Upon meiotic entry, the levels of H3K9ac gradually decrease and reach a basal level in round spermatids before they increase significantly in elongated spermatids [117]. H4 modifications also display a dynamic change during spermatogenesis. H4K5ac, H4K8ac, and H4K12ac show a moderate intensity in type A and type B spermatogonia. Their levels increase in leptotene spermatocytes but decrease in zygotene, pachytene, and diplotene spermatocytes [121]. In the late spermatids, H4K5ac, H4K8ac, and H4K12ac levels are highly elevated [121], as we illustrated in Fig. 2. This acetylation-deacetylation cycle of histone ensures proper DNA condensation while preserving paternal genome integrity. Histone modifications directly regulate the activation or repression of gene transcription, facilitating heterochromatin condensation and euchromatin opening formation. Given their dynamic changes during spermatogenesis, dysfunction in histone modifications can lead to spermatogenic failure and male infertility.

Association of histone modification dysfunction with male infertility

Histone modifications serve as dynamic regulators of chromatin states to drive transcriptional program essential for normal spermatogenesis (Table 2). While these modifications ensure precise control over SSC self-renewal, meiotic recombination, and chromatin compaction, their dysfunction is increasingly implicated in male infertility. Recent studies emphasize that global changes or locus-specific histone modification defects often underlie critical disorders in male germ cell development.

The balance between self-renewal and differentiation of SSCs is highly organized with dynamic changes in chromatin structure. Key histone modifications, e.g., H3K4me3 and H3K27me3, play pivotal roles in regulating these processes. H3K4 marks active gene promoters to sustain SSC self-renewal, while H3K27 ensures the silencing of genes crucial for SSC differentiation [78]. Disruption of histone modifications can destabilize this equilibrium, leading to spermatogenesis disorders. Notably, germ cell nuclear acidic protein (GCNA) has been identified as a histone chaperone, which can bind to H3-H4 tetramers or H2A-H2B dimers by interacting with core histone [122]. Both Gcna mutation and Gcna deficiency are linked to male infertility [123, 124], with the latter specifically impairing SSC maintenance [122]. These findings underscore an indispensable role of histone modification dynamics in mediating SSC development. PRMT5 conditional knockout in SSCs reduces male germ cells, which causes abnormal SSC proliferation and male infertility [9]. This defect is attributed to the disrupted H3K9me2/H3K27me2 deposition in Plzf promoter [9] and the impaired H2A/H4R3me2s-mediated LINEs and IAP transposons repression in PGCs [125]. Deletion of H3K9 demethylases JMJD1A/JMJD1B diminishes the numbers of male germ cells via H3K9me2 demethylation, reflecting its essential role in spermatogonial differentiation [126]. Conversely, JMJD3 mutation promotes SSC self-renewal through TET1-JMJD3 complex formation, which reduces H3K27me3 level at the Pramel3 promoter [127, 128]. Notably, HDAC3 catalytic activity is dispensable for male fertility [91].

In addition to regulate SSC self-renewal and differentiation, histone modifications play equally critical roles in ensuring meiotic fidelity during spermatogenesis. In mammals, the processes in meiosis include homologous chromosome pairing (synapsis), DNA double-stranded break (DSB), and recombination. During meiosis, PRDM9 directs sequence-specific H3K4me3 deposition, which recruits ZCWPW1 to repair DSBs. Zcwpw1^−/− mice fail to resolve breaks at these sites, and thus meiotic arrest occur [75, 129, 130]. Centromeric H3K9me2/3 deposition by Suv39h1/2 maintains genomic integrity, and their double knockout induces instability and spermatogenesis arrest [10, 131]. Another H3K9 methyltransferases SETDB1 has been reported to play an essential role in synapsis during meiosis with disrupting the distribution of H3K9me3 on chromosomes and inappropriate derepression of transposons [132, 133]. Concurrently, KDM2A demethylase precisely erases H3K36me2/3 at pachytene spermatocytes, while its conditional knockout disrupts crossing-over resolution and spermatogenesis arrest at zygotene spermatocytes [134]. Following meiotic completion, spermiogenesis requires precisely timed histone-protamine replacement. During chromatin condensation, H3K36 methyltransferase SETD2 is required for the acrosome biogenesis by mediating the H3K36me3 of Acrosin-binding protein 1 (Acrbp1) and protamine [85]. Concomitantly, histone demethylase JHDM2A (also known as JMJD1A) demethylates H3K9me2 at Tnp1 and Prm1 promoters, which enables chromatin decompaction. JMJD1A dysfunction disrupts elongation and leads to extensive apoptosis of germ cells and male infertility [135, 136].

In summary, the altered histone PTMs in male germ cells correlate with oligozoospermia in humans. Understanding the abnormal histone modifications in spermatogenesis can provides new theoretical clues and potential targets for NOA patients. Nevertheless, there are some concerns to be noted, since spermatogenesis involves complex interactions among histone modifications and the mechanisms of synergistic regulation among these modifications are not yet fully understood. Notably, many studies rely on in vitro cell experiments or animal models, species-specific differences necessitate caution when translating murine findings to human translational medicine. In the future, single-cell multi-omics approaches (scChIP-seq, scATAC-seq) should be employed to enable mapping of histone landscapes in infertile patients, which offers diagnostic biomarkers and therapeutic targets of male infertility.

CRCs in controlling the spermatogenesis and male fertility

Molecular mechanisms of chromatin remodeling

Chromatin remodeling refers to the dynamic process of altering chromatin structure between a condensed state and a transcriptionally accessible state. It encompasses multiple mechanisms, including covalent modifications of chromatin components (e.g., histones and DNA) and the non-covalent modifications of ATP-dependent chromatin remodeling by CRCs. Here we focus on the roles of CRCs in regulating spermatogenesis and its dysfunction with male infertility. Each chromatin remodeler family contains different functional domains and performs diverse functions, as we illustrated in Fig. 3.

Fig. 3.

Main domains and function of SWI/SNF, CHD, ISWI and INO80 families of CRCs. The left diagram illustrates the domains encompassed by distinct CRCs. Specifically, the SWI/SNF family comprises the HSA, DExx, Helicase, and Bromodomain domains. The ISWI family includes the DExx, Helicase, HAND, SANT, and SLIDE domains. The CHD family encompasses two chromodomains in addition to the DExx and Helicase domains. Furthermore, the INO80 family consists of the HSA, DExx, and Helicase domains. On the right diagrams are the mechanisms by which CRCs regulate alterations in chromatin structure, including nucleosome eviction, sliding, assembly, and H2A-H2B dimer exchange

CRCs are evolutionarily conserved multi-protein assemblies that contain ATPase domains. There are four families of remodelers for CRCs, including Switch/sucrose nonfer-mentable (SWI/SNF), chromodomain-helicase-DNA binding (CHD), imitation switch (ISWI), And Inositol requiring 80 (INO80) (Table 3). All families share a conserved SNF2-like ATPase subunit. These remodelers can alter chromatin composition and structure by promoting nucleosomes sliding, ejection of nucleosomes, or facilitating histone recombination through the removal of histones and replacement with histone variants [137]. The specific mechanisms of different CRCs are determined by the structure of their ATPase subunit.

Table 3.

CRCs functions and their dysfunction with male infertility

Members of CRCs		Functions	Loss of function
SWI/SNF family	BAF	Transcriptional activation, DNA repair and recombination	Catalytic subunit Brg1 deficiency causes mouse infertility with an arrest at meiotic prophase I [138].
	PBAF
ISWI family	NURF	Transcriptional regulation (activation or inhibition)	Baz1a, but not Baz1b, deficient mice have abnormal spermatogenesis and are sterility [139].
	ACF
	CHRAC
	NoRC
	RSF
	WICF
CHD family	CHD1	Transcriptional activation or inhibition	/
	CHD2		Chd2 knockdown inhibits proliferation of spermatogonia [140].
	CHD3		No distinctive phenotype [141]
	CHD4		Essential for the maintenance of SSCs and fertility [141]
	CHD5		Essential for normal spermatogenesis [142]
	CHD6		/
	CHD7		/
	CHD8		Chd8 deficiency leads to meiotic prophase I arrest [143].
	CHD9		/
INO80 family	INO80	DNA double strand break repair, DNA replication	Essential for successful meiosis [144]
	SWR1

The SWI/SNF family, also known as the BAF complex in mammals, alters the localization and structure of nucleosome [145]. This complex comprises canonical BAF (cBAF), polybromo-associated BAF (PBAF), and noncanonical BAF (ncBAF). All three complexes share the ATPase subunit BRM or BRG1, which hydrolyzes ATP to provide energy for nucleosome sliding or eviction. Typically, SWI/SNF family recognizes distal enhancers, promoters, or CTCF-binding sites to regulate chromatin accessibility, and it is involved in transcriptional activation, DNA repair and recombination [146, 147].

The ISWI family was first identified in Drosophila, and it is a homologue to yeast Swi2/Snf2 [148]. ISWI possesses a highly conserved ATPase domain that hydrolyzes ATP to promote chromatin remodeling, and it has a C-terminal HAND-SANT-SLIDE domain for DNA binding. Thus, the ISWI family facilitates nucleosome sliding for transcriptional regulation (activation or inhibition). In mammals, the catalytic ATPase within the ISWI complex is SNF2L (also named SMARCA1) or SNF2H (SMARCA5), which assembles with 1–3 noncatalytic subunits. There are at least 16 unique ISWI complexes, including ATP-utilizing chromatin assembly and remodeling factor (ACF), chromatin accessibility complex (CHRAC), nucleosome remodeling factor (NuRF), remodeling and spacing factor (RSF), WSTF-ISWI chromatin remodeling complex (WICH), nucleolar remodeling complex (NoRC), CECR2-containing-remodeling factor (CERF), BAZ2B-associated remodeling factor (BRF), and their variants.

The CHD family comprises CHD1-9 remodelers, which activates gene transcription by facilitating nucleosome sliding. CHD remodelers have double chromodomains at their N-terminal region and a SNF2 helicase-like ATPase domain that directly binds to methylated lysine residues [149]. CHD3 and CHD4 are capable of chromatin remodeling and histone deacetylation, thereby regulating transcriptional activation and inhibition.

The INO80 family catalyzes nucleosome sliding along DNA and promotes the exchange of canonical histones for histone variants, which helps DNA DSB repair, DNA replication and transcription [150]. This ATP-dependent CRC is highly conserved from yeast to humans, and it share core functional subunits, including Ino80, Rvb1/2, actin, actin-related proteins Arp4, Arp5 and Arp8, and Ies2/6 [151]. Among them, Ino80 ATPase subunit provides catalytic activity [150], while actin and actin-related proteins may help with histone binding [152].

Distinct expression patterns and functions of CRCs in spermatogenesis

During spermatogenesis, chromatin remodeling actively facilitates critical processes, e.g., chromosomal synapsis, DSB repair, homologous recombination, and the replacement of histone by protamine. These processes ensure genomic integrity, precise gene regulation, and functional sperm formation. To coordinate these multifaceted roles, CRCs exhibit stage-specific expression patterns during spermatogenesis.

The SWI/SNF family plays indispensable roles in orchestrating chromatin dynamics during spermatogenesis. Despite its functional versatility, the assembly of SWI/SNF subunits across developmental stages of male germ cells poses challenges in deciphering its regulatory mechanisms. Notably, mutations in SWI/SNF family are highly associated with spermatogenic failure, which underscores its importance in maintaining male fertility. The main catalytic ATPases of SWI/SNF family are BRG1 (SMARCA4) and BRM (SMARCA2), and their spatiotemporal expression patterns are tightly linked to male germ cell maturation [153]. BRG1 is present in SSCs (PLZF⁺ spermatogonia), spermatocytes, and spermatids, with peak expression level at the pachytene spermatocytes to support homologous recombination and transcriptional activation of meiosis-specific genes [138, 153]. When spermatocytes further differentiate into round spermatids, BRG1 expression is reduced and becomes undetectable in the elongated spermatids [138, 153]. Complementary to the expression profile of BRG1, BRM is expressed in pachytene spermatocytes and does not persist until round spermatids [153, 154]. Subunit composition further diversifies SWI/SNF functions. For instance, ARID2, a component of the PBAF subcomplex, is expressed in pachytene spermatocytes at a low level but absent in leptotene or zygotene spermatocytes [155], implicating the stage-specific expression patterns and roles of SWI/SNF family in regulating crossover resolution or chromatin compaction.

The ISWI family of ATP-dependent chromatin remodelers, including SNF2H (SMARCA5) and SNF2L (SMARCA1), exhibit complementary expression patterns during spermatogenesis. SNF2H is enriched in proliferating spermatogonia and Sertoli cells, and it is critical for promoting the self-renewal of SSCs and mitotic progression, as evidenced by its high abundance in mammalian testes compared to other tissues [156]. In contrast, SNF2L is exclusively expressed in spermatids, suggesting a specialized role of SNF2L in controlling chromatin compaction during spermiogenesis [156]. Another ISWI remodeler BAZ1A (also named ACF1) is predominantly expressed in human testicular tissue [157], and its expression level peaks at the pachytene spermatocytes and declines at elongating spermatids [139], which facilitates nucleosome spacing and crossover resolution. This dynamic expression pattern of BAZ1A reflects that it is a key modulator of meiotic chromatin architecture and post-meiotic remodeling.

Certain CHD family members also exhibit stage-specific expression and localization patterns, which underscores their specialized roles in coordinating chromatin dynamics throughout spermatogenesis. CHD4 and CHD5 have a higher expression level in testis than other tissues [158, 159]. CHD4, a core subunit of the NuRD complex, is prominently expressed in the undifferentiated spermatogonia (PLZF⁺), and it represses differentiation genes via H3K27me3 deposition and histone deacetylation for the maintenance of SSCs [141]. As male germ cells enter into meiosis, CHD4 level is declined for homologous pairing and sister chromatid cohesion, while it is undetectable in round spermatids [158, 159]. This dynamic expression profiles of CHD4 reflects its dual roles in safeguarding meiotic progression and genomic stability. Meanwhile, CHD3, a structural homolog of CHD4, is present in spermatogonia at a relatively lower level [160, 161]. In addition, CHD5 is absent in mitotic and meiotic male germ cells [142, 158], whereas it is highly expressed during spermiogenesis with the peak level in round spermatids (steps 7–8) [142, 158] to orchestrate histone-protamine exchange. Its expression vanishes in elongated spermatids, which is coinciding with the completion of chromatin hypercondensation. The Chd5 knockout mice exhibit defective sperm chromatin compaction and male infertility. Collectively, the CHD family operates as a choreographed network for regulating spermatogenesis. CHD4 dominates early stages to preserve stemness of SSCs and meiotic fidelity, while CHD3 provides auxiliary support, ensuring seamless transition across male germ cell development. CHD5 executes late-stage chromatin compaction.

In murine testes, INO80 is highly expressed in spermatocytes but undetectable in SSCs or spermatids [144, 162], which implicates that INO80 is involved in the regulation of meiosis of spermatocytes. INO80 is indispensable for repairing meiotic DNA DSBs. INO80 binds to γH2AX-marked DSBs via its Arp5/Arp8 subunits, which promotes nucleosome eviction to enable resection and RAD51-mediated homologous recombination [163, 164]. Germline-specific Ino80 knockout mice assume persistent DSBs, synapsis failure, and spermatocyte apoptosis, which directly links its activity to meiotic fidelity [144]. On the other hand, the regulatory networks governing INO80 recruitment and its interactions with other repair factors (e.g., BRCA1, DMC1) remain to be defined.

Histone-to-protamine replacement is important for normal spermatogenesis

The replacement of histones with protamines is a critical chromatin remodeling event for spermiogenesis, which ensures the extreme nuclear compaction required for sperm functionality. This process involves a stepwise substitution of histones by testis-specific histone variants, followed by the incorporation of transition proteins (TNPs), which are ultimately replaced by protamines [165]. Disruption in this histone-protamine exchange can lead to male infertility, as manifested by azoospermia, oligospermia, and teratospermia. Testis-specific histone variants (e.g., H1.6, H1.7, H1.9, H3.5, and H2BW1) play essential roles in the initial loosening of nucleosomal chromatin, which facilitates the access of transition proteins to DNA. Loss of functions of these variants can result in incomplete histone displacement and male infertility, as we summarized in Table 4.

Table 4.

Main factors for histone-to-protamine transition and its dysfunction with male infertility

Factors		Cellular locations	Loss of function
Histone variants	H1.6	Mid- to late pachytene spermatocytes, spermatids	H1t-null mice are fertile [166].
	H1.7	Round and elongating spermatids	Mice lacking H1t2 are infertile with aberrant elongation of spermatids [167].
	H1.9	Elongating and elongated spermatids	NA
	H2BW1	Pachytene spermatocytes to round spermatids	Mice lacking both Th2a and Th2b genes are sterile [168].
	H2AL2	Spermatids	H2al2-KO male mice are completely sterile [169].
	H2A.B	Pachytene spermatocytes to round spermatids	H2a.b.3 knockout mice display subfertility [170].
	H3.3	Spermatogonia to spermatids	H3.3B, but not H3.3 A, is required for male fertility [171].
	H3.5	Spermatogonia and spermatocytes	NA
	H3.4	Differentiating spermatogonia to spermatocytes	H3.4 knockout mice display infertility [172].
TNPs	TNP1/2	Spermatids	Mild effect on mouse fertility [173, 174]
PRMs	PRM1/2	Spermatids	Prm1 or Prm2 knockout mice assume infertility [175].

A total of 11 H1 variants have been found in humans and mice, including testis-specific subtypes (e.g., H1.6 and H1.7), and they orchestrate histone-to-protamine transition through stage-specific expression [176, 177]. H1.6 dominates pachytene spermatocytes to round spermatids, while its deficiency has undetectable defect in mouse spermatogenesis with the function compensation by other H1 subtypes [178]. In contrast, H1.7 level peaks in elongating spermatids, and this histone variant has unique domain which can increase DNA accessibility for TNP binding [179]. Mutations in H1.7 impair nuclear condensation during spermiogenesis and results in aberrant elongation of spermatids [167, 176].

H2A and H2B histone variants also contribute to aberrant spermatogenesis. Incorporation of H2BW1 destabilizes nucleosomes and promotes the eviction of histones H2A and H2B, which initiates global chromatin decondensation essential for spermiogenesis. Mice lacking H2BW1 are infertile, with the malformed elongating spermatids [180]. H2A.L.2 is expressed in elongating spermatids, and its assembly opens nucleosomes to allow TNP loading [169]. Loss of H2A.L.2 function significantly impairs TNP loading on chromatin and subsequent protamine incorporation [169].

Within the H3 variant family (e.g., H3.3, H3.4 and H3.5), H3.3 is incorporated into chromatin replication independently throughout cell cycle, which enriches active promoter, enhancer, and gene body regions to maintain an open chromatin state [181]. Aberrant expression of H3.3B leads to a progressive loss of post-meiotic germ cells, which is resulted from pan-sex chromosomal expression and RLTR10B/2 retrotransposon activation [171]. The germ cell-specific H3.4 and H3.5 are highly expressed during early spermatogenesis and they are involved in meiotic processes [172, 182]. These variants are crucial for male fertility, and their dysfunction is closely linked to male infertility. Depletion of H3.4 causes meiotic abnormalities and severe fertility defects [172], while the function of H3.5 remains to be explored.

TNPs act as molecular bridges during histone-protamine exchange. These proteins temporarily bind to DNA, which relieves torsional stress and facilitates the recruitment of protamines. Interestingly, knockout of either Tnp1 or Tnp2 in mice results in only mild fertility defect [173, 174], while the combined deletion of both genes leads to spermatid arrest at step 12, with the retained histones and fragmented DNA [183]. Similar to TNPs, protamines (PRM1 and PRM2) are essential for the final chromatin compaction in spermatids, which renders the chromatin transcriptionally inert. Mutations in either Prm1 or Prm2 disrupt normal sperm chromatin formation, which impairs sperm function [175], as we addressed in Table 4.

In summary, the histone-to-protamine replacement represents a paradigm of chromatin reprogramming during male germ cell differentiation. This process is not merely a structural transformation but the safeguard for paternal genome integrity. Nevertheless, the precise molecular mechanisms underlying the transition of histone-to-protamine remain to be elucidated. Its disruption underscores that the epigenetic choreography of spermiogenesis by CRCs is the same important as the genetic genes in ensuring normal spermatogenesis.

Associations between CRCs dysfunction and male infertility

CRCs regulate cell function by altering the structure and composition of chromatin. During spermatogenesis, CRCs precisely orchestrate gene expression of male germ cells by dynamically modulating chromatin accessibility. In addition, ATP-dependent chromatin remodelers, through dynamic nucleosome repositioning and histone variant incorporation, are involved in DSB repair by modulating chromatin accessibility. This exquisite regulation ensures the sequential progression of meiotic processes, and any perturbation of this regulatory machinery may give rise to aberrant development of male germ cells and male infertility, as we summarized in Table 3.

The SWI/SNF complex, driven by its catalytic ATPase BRG1, orchestrates both transcriptional activation and DNA DSB repair during meiosis. BRG1 facilitates chromatin decompaction at recombination hotspot, which enables SPO11-mediated DNA DSB formation and subsequent RAD51 recruitment for homologous recombination [184]. In BRG1-deficient mice, spermatocytes arrest at prophase I with the unresolved γH2AX foci and synaptic failure, which leads to complete infertility [138, 185]. Another SWI/SNF subunit, ARID2, ensures spindle assembly during spermatocyte meiosis [155]. Besides DSB repair, SWI/SNF maintains SSC self-renewal by activating stem cell genes (e.g., Plzf and Id4) and repressing differentiation signals [138, 185]. These studies reflect SWI/SNF integrates DNA DSB repair with transcriptional and mitotic fidelity, whereas its dysfunction results in male infertility.

Most members of CHD family are abundantly expressed in testis, which indicates their function in retaining male fertility. CHD4 is essential for the maintenance of SSCs [141, 186], and it represses the expression of differentiation genes (e.g., Kit, Stra8) via H3K27me3 deposition to maintain SSC stemness [141]. Chd4 knockout triggers premature meiosis entry and SSC depletion, which eventually leads to the Sertoli cell-only syndrome (SCOS) [141]. The function of CHD2 is identical to CHD4 to promote the self-renewal of SSCs [140]. Additionally, CHD3 supports SSC maintenance through partial functional overlap with CHD4, highlighting the CHD family’s adaptive redundancy [141]. In contrast, CHD5 emerges at spermiogenesis, and it modulates histone-protamine exchange in round spermatids [142, 187]. CHD8 is involved in both the mitosis of SSCs and the progression of meiotic prophase I [143].

The ISWI remodeler SNF2H/BAZ1A ensures precise nucleosome spacing during meiosis, which prevents ectopic recombination between non-homologous regions. While dispensable for DNA DSB repair, BAZ1A-deficient spermatocytes exhibit aberrant chromatin loops and crossover errors, which leads to aneuploidy of these cells [139]. The INO80 complex exemplifies the dual functionality of chromatin remodelers. It directly binds to γH2AX-marked DSBs via its Arp5/Arp8 subunits, which evicts nucleosomes to permit DNA end resection and RAD51-mediated recombination [188]. Germline-specific Ino80 knockout mice exhibit persistent DSBs, synaptonemal complex disintegration, and spermatocyte apoptosis, which mimics the pathology of idiopathic NOA [144, 189]. Concurrently, INO80 stabilizes RNA polymerase II at promoters, which triggers transcriptional activation by chromatin remodeling [190]. This dual role of INO80 complex ensures that meiotic gene expression is tightly synchronized with DNA DSB repair, which is the coordination critical for male fertility. Taken together, CRCs are multifunctional sentinels for maintaining normal spermatogenesis. The roles of CRCs include DNA DSB repair, transcriptional regulation, and structural reorganization. Notably, CRC dysfunction might result in stage-specific pathology (e.g., SSC depletion and spermatocyte meiotic arrest) and eventual male infertility.

Conclusions and prospectives

Epigenetic modifications are involved in the precise regulation of gene expression, which ensures the normal spermatogenesis and correct genetic inheritance. In recent years, great progress has been made in unveiling the functions and mechanisms of DNA methylation, histone modifications and CRCs in controlling normal spermatogenesis and their dysfunctions with male infertility: (i) the identification of genome-wide DNA methylation of testis [191–193]; (ii) different expression patterns, functions And mechanisms by DNA methylation, histone modifications, And CRCs; 3) association between epigenetic dysfunction and male infertility [55, 194]; and iv) epigenetic regulatory networks among DNA methylation, histone modifications, and CRCs [185].

Meanwhile, certain concerns remain to be noticed and questions require to be explored in the above epigenetic regulation in spermatogenesis. First of all, a major challenge in exploring epigenetic regulation in spermatogenesis lies in the intrinsic cellular heterogeneity of the testis where various kinds of male germ cells (SSCs, spermatocytes, spermatids) at distinctly developmental stages coexist with somatic cells. In bulk-tissue assays, this cellular complexity causes various kinds of effect, where DNA methylation, histone modification, or chromatin remodeling profiles may reflect an average signal that masks cell type-specific patterns. For instance, global hypomethylation observed in certain NOA patients may partially be resulted from loss of male germ cells rather than true hypomethylation within surviving cells. Currently, fluorescence-activated cell sorting (FACS) and magnetic-activated cell separation (MACS) are widely used for isolating and purifying cell types of germ cells based upon their specific markers [195, 196]. For instance, PLZF and GFRA1 are hallmarks for SSCs [197], while c-Kit has been regarded as a marker for the differentiating spermatogonia. In addition, STA-PUT allows size-based separation of male germ cells, but its resolution is limited for closely related cell types [198]. While these enrichment methods enable epigenetic analysis of specific cells, they require careful validation, and cross-contamination of cell types can still confound locus-specific findings. A notable example of resolving cell-type heterogeneity by the study [199] utilizing the synchronized spermatogenesis and FACS-based purification to isolate eleven germ cell types for epigenomic profiling and obtain a high-resolution atlas of histone modifications, DNA methylation, and chromatin accessibility across spermatogenesis. Combining fluorescence-reporter models with physical separation techniques, e.g., FACS, can mitigate certain limitations associated with purely size- or markers-based cell sorting. More recently, single-cell epigenomic technologies have emerged as powerful tools to resolve heterogeneity without physical cell sorting [200]. For example, recent applications in human fetal gonads have revealed the profile of DNA methylation of human fetal germ cells at different phases at single-cell resolution [201]. These approaches have revealed not only stage-specific epigenetic signatures but transcriptional heterogeneity within morphologically similar germ cell populations [202]. Such findings highlight the need to move beyond bulk measurements, as they may obscure the true epigenetic regulation within rare or transient cell types. Therefore, future studies can integrate cell-type enrichment techniques with single-cell epigenomic profiling to accurately map the epigenetic landscape across germ cell development. These approaches can establish causative links between cell-specific epigenetic dysregulation and male infertility phenotypes, which avoids misleading interpretation from mixed-cell data. Secondly, a critical frontier lies in elucidating the dynamic crosstalk between histone modifications, DNA methylation, and CRCs-multilayered regulatory networks that orchestrate stage-specific gene expression in male germ cells. The tracing technologies by fluorescence or probes can be employed to identify new epigenetic factors during spermatogenesis. Recent breaking advances in spatial technologies (e.g., Stereo-seq, DBiT-seq) have for the first time mapped spatiotemporal transcriptomic changes in developing Drosophila with spatial cellular state dynamics in larval testes [203]. The integration of these approaches enables comprehensive delineation of cell-specific dynamics in histone modifications and DNA methylation during spermatogenesis, which permits locus-specific analysis of histone modifications or DNA methylation patterns. For instance, the combined application of scCUT&Tag and CoBATCH can map co-evolutionary landscapes of histone modifications and DNA methylation. Most significantly, CRISPR-based epigenome editing technologies can merge CRISPR targeting capabilities with epigenetic reprogramming [204], which provide novel approaches to investigate epigenetic regulatory networks. Nevertheless, there are still critical limitations, such as spatial resolution remains insufficient to resolve tightly packed spermatid nuclei, and single-cell isolation causes epigenetic memory distortion in round spermatids. To optimize these methodologies for effective applications in reproductive biology, further improvement remains imperative. Finally, it remains to be demonstrated about whether the conclusions of epigenetic regulation in animal spermatogenesis could translate to primates or humans. Testicular organoids can serve as an excellent model for the in vivo studies of functions and mechanisms of epigenetic factors [205–207]. Large-scale epigenome-wide association studies (EWAS) could be used to identify epigenetic signatures for idiopathic infertility. Further studies might focus on these above questions to establish fully epigenetic regulation and networks of human spermatogenesis, which could provide novel discoveries and targets for the treatment of male infertility.

Author contributions

Y.C. wrote the manuscript. J. D., Y. Z., L. D., F. J., C. L., W. C., and H. Z. assisted with the manuscript writing. Z. H. were responsible for design and revising the manuscript. All authors approved the manuscript.

Funding

This work was funded by the grants from National Nature Science Foundation of China (32470904 and 32170862), Developmental Biology and Breeding (2022XKQ0205), Shanghai Key Laboratory of Reproductive Medicine, Natural Science Foundation of Hunan Province of China (2024JJ5284, 2023JJ30424, and 2024JJ5282), Research Team for Reproduction Health and Translational Medicine of Hunan Normal University (2023JC101), and Research Foundation of Education Bureau of Hunan Province for Outstanding Youth (23B0064).

Data availability

All data and materials used in this study are available from the corresponding author upon request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors approved this manuscript for publication.

Competing interests

The authors declared no competing financial interest.