H2A.Z deposition at meiotic prophase I underlies homologous recombination and pachytene genome activation during male meiosis

Accurate meiotic progression is important for gamete formation and the generation of genetic diversity. However, little is known about the identity of chromatin regulators that underlie mammalian meiosis in vivo. Here, we identify the multifaceted functions of the chromatin remodeler Znhit1 in governing meiosis. We observe a gradual increase in Znhit1 expression during the meiotic prophase. Znhit1 deficiency in spermatocytes results in arrested pachytene development, impaired DNA double-strand break repair, and defective homologous recombination. Single-cell RNA sequencing and transcriptome analysis reveal that Znhit1 loss downregulates the transcription of pachytene genome activation (PGA) genes globally. Chromatin immunoprecipitation data show that Znhit1 is needed for the incorporation of the histone variant H2A.Z into pachytene chromatin. Moreover, we find that H2A.Z cooperates with the transcription factor A-MYB to co-bind DNA elements and control enhancer activity. Our findings provide insights into the regulatory mechanisms governing meiotic progression and highlight Znhit1 as a critical regulator of meiotic recombination and PGA.


Introduction
Meiosis is a fundamental and conserved process that plays a crucial role in gamete formation and the generation of genetic diversity.By undergoing one round of DNA replication and two rounds of cell division, the meiotic process ensures the production of haploid cells, each with a unique combination of genetic materials (Handel and Schimenti, 2010;Lascarez-Lagunas et al., 2020).Any disruptions to the normal progression of meiosis would have significant consequences, such as aneuploidy, infertility, spontaneous abortion, and congenital diseases (Hassold and Hunt, 2001).
Therefore, understanding the molecular mechanisms underlying meiosis will provide valuable insights for the diagnosis and treatment of reproductive and developmental diseases.
Following the formation of DSBs, the DNA undergoes resection, leading to the generation of single-stranded DNA overhangs.These overhangs are then coated by RPA, DMC1, and RAD51, facilitating the production of recombination intermediates.
Subsequently, these intermediates are processed and resolved, ultimately forming either meiotic crossovers or non-crossovers (Gray and Cohen, 2016;San Filippo et al., 2008;Symington, 2014).The tightly coordinated timing and spatial arrangement of these meiotic events are of utmost importance for proper germ cell development.
Zinc finger HIT-type containing 1 (Znhit1), an evolutionarily conserved subunit of the SRCAP chromatin remodeling complex, acts as a key regulator for the histone variant H2A.Z deposition to control gene expression and cell fate determination (Cai et al., 2005;Cuadrado et al., 2007;Feng et al., 2022).Recent studies have shown that Znhit1 is involved in a wide range of developmental processes, including muscle and lens differentiation, heart development, lung branching, and adult tissue stem cell maintenance (Cuadrado et al., 2010;Lu et al., 2022;Sun et al., 2020;Wei et al., 2022;Xu et al., 2021;Zhao et al., 2019).In our most recent study, we deleted Znhit1 in early male germ cells and found that Znhit1 is required for the mitosis-to-meiosis transition by controlling Meiosin expression (Sun et al., 2022).However, the in vivo function of Znhit1 in regulating meiotic progression remains unknown.
To delve into the function of Znhit1 in meiosis, we analyzed Znhit1 expression and observed upregulated Znhit1 expression specifically during meiotic prophase.Here, we examine the role of Znhit1 in multiple meiotic events and show that the deletion of Znhit1 in spermatocytes causes impaired HR and defective PGA, thereby resulting in meiotic developmental arrest.Hence, our study highlights the essential role of Znhit1 in ensuring meiotic progression, implicating its potential as a therapeutic target for meiosis-related diseases.

Znhit1 expression is upregulated during meiotic prophase and Znhit1 knockout in spermatocytes disrupts spermatogenesis
To identify potential chromatin regulators in meiosis, we first queried for chromatin factors highly expressed during the zygotene stage based on published transcription data (Fig S1A and Table S1) (Chen et al., 2018).Quantitative analysis showed that the chromatin remodeler Znhit1 was among the most highly expressed factors (Figs 1A,1B,S1B,and S1C).RNA in situ hybridization validated that Znhit1 expression in male germ cells decreased after meiotic initiation, followed by a gradual increase in spermatocytes from the zygotene stage to the metaphase stage (Fig 1C).These results suggest that Znhit1 is a potential chromatin regulator of meiotic progression.

Znhit1 deletion leads to meiotic arrest at the pachytene stage
To dissect spermatocyte development in Znhit1-sKO testes, we co-stained SYCP3, a marker of primary spermatocytes, and HSPA2, a testis-specific HSP70 family member that accumulates from the pachytene stage onward.As shown in Figure 2A, both control and Znhit1-sKO testis sections exhibited SYCP3 + HSPA2 + primary spermatocytes.
Phospho-histone H3 (Thr3) (pH3) is a marker of mitotic and meiotic metaphase cells, and Znhit1 deletion caused the complete absence of pH3 + metaphase spermatocytes within the cavity of seminiferous tubules (Fig 2B).These results suggest that Znhit1 deletion causes spermatocyte arrest at meiotic prophase I.
To pinpoint the onset of meiotic failure, we performed immunostaining against SYCP3 in spermatocyte chromosome spreads.The wild-type testes displayed typical spermatocytes at consecutive substages, including leptotene, zygotene, pachytene, diplotene, and diakinesis (Fig 2C).However, spermatocytes at the diplotene and diakinesis stages were rarely observed in Znhit1-sKO testes (Fig 2D and 2E).We also examined the expression of H1T, a middle pachytene marker, and no difference in H1T expression was observed in Znhit1 -/-spermatocytes (Fig S2A).Moreover, Znhit1 knockout resulted in increased TUNEL-positive spermatocytes, suggesting that these defective spermatocytes were eliminated by the apoptotic pathway (Fig 2F and 2G).These results indicate that Znhit1 deletion leads to meiotic arrest at the pachytene stage.

Znhit1 is required for DSB repair and meiotic recombination
To investigate the impact of Znhit1 on meiotic events, we first performed immunostaining of SYCP1 and SYCP3, the essential components of the synaptonemal complex, to examine whether Znhit1 regulated homologous synapses.As shown in Figure 3A, pachytene nuclei with fully synapsed autosomes were observed in spermatocytes from control and Znhit1-sKO testes.During the pachytene stage, the X and Y chromosomes undergo pairing and synapses between the short pseudoautosomal regions (PARs).We found an increase in the percentage of unsynapsed X-Y chromosomes in Znhit1-sKO spermatocytes (Fig 3B).Thus, deletion of Znhit1 results in impaired sex chromosomal synapsis.
Next, we examined the programmed formation and repair of DSBs by co-staining SYCP3 and γH2AX, a DSB marker, on spermatocyte chromosome spreads.In both control and Znhit1-sKO spermatocytes, γH2AX signals were evident in the leptotene and zygotene stages (Fig 3C).However, in Znhit1-deficient pachynema, we still observed diffuse γH2AX signals on the autosomes, while in control pachynema, γH2AX signals only accumulated on the X-Y chromosomes.Moreover, we detected the presence of another DSB marker, phospho-ATM (pATM), on the autosomes in Znhit1-sKO pachytene spermatocytes (Fig 3D).Together, these results demonstrate that Znhit1 is required for completing DSB repair.
To understand how Znhit1 regulates DSB repair, we performed an immunostaining analysis using the markers RPA2 and DMC1, the essential factors involved in DSB repair by coating resected DSBs.We found an increase in RPA2 and DMC1 counts in both zygonema and pachynema of Znhit1-deficient testes (Fig 3E,3F,and S3A), indicating that Znhit1 deletion delayed recombinational repair, resulting in defective resolution of autosomal DSBs in Znhit1 -/-pachynema.The formation and repair of DSBs play crucial roles in initiating and facilitating meiotic recombination, respectively.To examine the consequences of unrepaired DSBs on meiotic recombination in Znhit1 mutants, we performed an immunostaining analysis against MLH1, a marker of meiotic crossover formation.Indeed, we found an almost absence of MLH1 foci in Znhit1 -/-pachytene spermatocytes compared with controls (Fig 3G and 3H), indicating that the loss of Znhit1 severely disrupted meiotic crossover formation.
Collectively, these results argue that Znhit1 is essential for multiple meiotic events, including sexual chromosome synapsis, DSB repair, and homologous recombination.
Of the changed genes between zygotene spermatocytes and pachytene spermatocytes, 1,560 genes were upregulated (referred to as pachytene-activated genes or PGA genes).
Moreover, we identified differentially expressed genes (DEGs) in each spermatocyte type between the control and Znhit1-sKO groups.The number of DEGs gradually increased from the preleptotene stage to the pachytene stage, with downregulated genes being predominant (Fig 5B and Table S4).In particular, Znhit1 -/-pachytene spermatocytes exhibited 2,785 downregulated genes, among which were most of the pachytene-activated genes (1,094 out of 1,560, 70.1%; Figs 5B-5D, and S5A).Gene Ontology (GO) enrichment analysis revealed that upregulated genes following Znhit1 deletion were associated with apoptotic pathways, supporting an observation that Znhit1 deletion resulted in germ cell clearance through apoptosis.Moreover, downregulated GO terms included germ cell development and DNA recombination (Fig 5E and Table S5).
Interestingly, recent studies showed that cilium organization was essential for spermatocyte development (Mytlis et al., 2022), and Znhit1 deletion reduced the expression of cilium-related genes in zygotene and pachytene spermatocytes (Fig 5E).
We further conducted bulk RNA-seq experiments with P14 control and Znhit1-sKO testicular cells and identified 882 DEGs (812 downregulated genes and 70 upregulated genes) in Znhit1-sKO testicular cells (Figs 5F and S6A, and Table S6).Through integrative analysis of bulk and single-cell RNA-seq data, we found that the expression of these 812 genes was activated during the zygotene-to-pachytene transition and Znhit1 deletion significantly reduced the expression of these genes (Figs 5G, 5H, and S5B).The downregulated genes following Znhit1 loss were enriched for germ cell-related GO terms, such as cilium formation and spermatid development (Fig S6B).Together, these data argue that Znhit1 is required for meiotic transcriptional activation.

Znhit1 deletion impairs H2A.Z incorporation into pachytene chromatin
We then asked how Znhit1 regulates transcription.Znhit1 is a subunit of the SRCAP complex that facilitates the incorporation of the histone variant H2A.Z by replacing H2A (Fig S7A).Immunostaining against H2A.Z showed a gradual increase in H2A.Z accumulation on autosomes from the leptotene to the diakinesis stage, while the H2A.Z signal was markedly low on the X-Y body (Fig 6A).Znhit1 loss decreased H2A.Z staining on the pachytene cells, suggesting that Znhit1 is required for maintaining H2A.Z integrity during meiosis.
To probe the H2A.Z targets, we performed chromatin immunoprecipitation followed by sequencing (ChIP-seq) against H2A.Z in P14 testicular cells.Using the ChromHMM model to annotate H2A.Z signals with meiocyte chromatin states defined by published epigenetic markers (Spruce et al., 2020), we found that H2A.Z sites were primarily enriched in chromatin state 1 and state 2 (promoters and enhancers) but were less

H2A.Z deposition cooperates with A-MYB to regulate transcription
We further asked how Znhit1/H2A.Z deposition facilitates lineage-specific gene expression.We performed TF motif analysis using active promoters/enhancers with Previous studies have shown the fundamental role of A-MYB in activating meiotic enhancers (Maezawa et al., 2020).Enhancer activity is linked to an increased formation of enhancer single-stranded DNA (ssDNA), which can be evaluated using KAS-seq (Wu et al., 2020).We found that the removal of Znhit1 resulted in abolished ssDNA at active enhancers (Fig 7I and 7J).Interestingly, the chromatin accessibility detected by ATACseq at active enhancers increased after Znhit1 deletion.These findings suggest that Znhit1/H2A.Z deposition is essential for enhancer activation.
Together, these results demonstrate the critical role of the Znhit1/H2A.Z/A-MYB axis in governing transcriptional activation and enhancer activity.

Discussion
Meiosis plays a crucial role in the production of haploid gametes.Any disturbances during meiotic progress are known to lead to infertility and congenital diseases.Here we delineate the multifaceted functions of the chromatin remodeler Znhit1 in regulating meiotic progression.Our study provides convincing data to show that the Znhit1/H2A.Z/A-MYB axis controls spermatocyte development, meiotic recombination, and PGA.Thus, these findings underscore the significance of specific chromatin structures in governing appropriate transcriptional reprogramming and precise meiotic recombination.
The role of H2A.Z in DNA injury and repair has been extensively studied (Colino-Sanguino et al., 2022;Dong et al., 2014;Xu et al., 2012;Yamada et al., 2018), but its function in programmed DSB formation and repair during mammalian meiosis remains unknown.To address this question, we generated Znhit1 conditional knockout mice specifically during male meiosis to examine the role of Znhit1-mediated H2A.Z deposition in the formation and repair of programmed DSB.One new observation in this study is that Znhit1 loss is dispensable for DSB formation but necessary for DNA break repair during meiosis.Intriguingly, unlike the observations in plant meiosis, we found that H2A.Z does not directly bind DNA recombination sites in mammalian spermatocytes (Choi et al., 2013;Spruce et al., 2020;Wang and Copenhaver, 2018).Moreover, we showed that Znhit1 deletion delays recombinational repair, but has limited impacts on DNA resection, ultimately causing defective meiotic crossover formation.Importantly, our study demonstrates that Znhit1 is essential for the expression of meiotic recombination-related genes, such as Ccnb1ip1 and Rnf212, implying that Znhit1/H2A.Z modulates meiotic recombination through transcriptional regulation.Therefore, our data reveals a different mechanism underlying meiotic recombination between plants and mammals.
Pachytene spermatocytes utilize a highly specific mechanism of meiotic surveillance, the pachytene checkpoint, to prevent aneuploidy formation by removing abnormal germ cells with incomplete chromosome synapsis and defective homologous recombination (Huang and Roig, 2023;Roeder and Bailis, 2000;Subramanian and Hochwagen, 2014).Two checkpoint pathways exist in male meiosis: one responding to unrepaired DSBs and the other triggered by MSCI failure (Royo et al., 2013).However, whether epigenetic regulation is involved in meiotic surveillance is not well defined.Previous studies have reported that upon stimulation by unrepaired DSBs, the MRE11-ATM-CHK2 pathway is activated to eliminate aberrant germ cells (Marcet-Ortega et al., 2017).Our study found that Znhit1 deletion impairs the removal of ATM phosphorylation signals on autosomes in pachynema, ultimately causing meiotic pachytene arrest and apoptosis.Therefore, these findings demonstrate that Znhit1 acts as an important chromatin factor involved in the regulation of the pachytene checkpoint.
It has long been noticed that in meiotic prophase I, chromatin that has not yet completed chromosomal synapses is transcriptionally inactive, known as meiotic silencing of unsynapsed chromatin (MSUC) (Turner et al., 2005).When germ cells enter the pachytene stage, a large number of protein-coding genes and non-coding RNAs begin to be actively expressed, known as PGA.Previous studies have shown that transcription factors A-MYB and BRDT are involved in transcriptional activation during meiotic prophase, but how PGA takes place in the chromatin context is unclear.In this study, we found that the expression of the chromatin remodeler Znhit1 is specifically upregulated during the zygotene-to-pachytene transition.We also observed that H2A.Z deposition is enriched in the autosomes but less in the sex chromosome at the pachytene stage.Znhit1 deletion in spermatocytes repressed pachytene gene activation globally and reduced chromosome-wide H2A.Z deposition, supporting a central role of Znhti1 in PGA regulation and chromatin remodeling.It has been known that A-MYB is a transcription factor that controls pachytene transcription activation.We found that Znhit1-deficient spermatocytes phenocopied abnormal meiotic phenotypes in A-MYB mutants, such as X-Y synapsis failure, impaired DSB repair, and defective HR (Alexander et al., 2023;Bolcun-Filas et al., 2011;Li et al., 2013;Maezawa et al., 2020).Our results also revealed that Znhit1/H2A.Z cooperates with A-MYB to regulate PGA gene activation.Therefore, our study illuminates the molecular mechanisms underlying the fundamental question of how PGA is regulated.
The histone variant H2A.Z is enriched at gene promoters and regulatory regions, but there is ongoing debate regarding its involvement in transcriptional regulation.A recent paper reported that H2A.Z knockout in post-mitotic muscle cells has limited effects on gene expression (Belotti et al., 2020).In this study, we utilized the meiotic prophase as a model system (where DNA replication does not occur) to study H2A.Z's function in transcriptional regulation.We found that Znhit1-mediated H2A.Z deposition, independent of DNA replication, is indispensable for the transcriptional activation of a large number of meiotic genes.One explanation of this conflicting phenomenon is that H2A.Z dynamics but not stable H2A.Z accumulation is essential for priming transcriptional changes.
Through the integration of functional and molecular evidence, our findings establish the critical involvement of Znhit1-dependent chromatin remodeling in the orchestration of meiotic progression and coordination of various meiotic processes, such as HR and PGA.Furthermore, we pinpoint the pivotal role played by the Znhit1/H2A.Z/A-MYB axis in driving transcriptional reprogramming during meiotic prophase.Taken together, this study deepens our understanding of the interplay between epigenetic regulation and mammalian meiosis.

Materials and Methods
Animals.Znhit1 fl/fl mice have been previously described (Zhao et al., 2019) and are available from the Model Animal Research Center of Nanjing University (MARC, Nanjing, China).The Stra8-cre knock-in mouse line was kindly provided by Dr. Ming-Han Tong (Lin et al., 2017).All mice were maintained on the C57BL/6J background.
Germ cell-specific Znhit1 knockout mice (Znhit1 fl/fl ; Stra8-cre) were obtained by crossing Znhit1 fl/+ ; Stra8-cre mice with Znhit1 fl/fl mice.All mice were housed in the SPF (Specific-Pathogen-Free) animal facility with standard 12 h light/dark cycles and standard temperature (22 to 24°C).All mice were provided with ad libitum access to standard laboratory food and water.All experiments in this study were performed following relevant guidelines and approved by the Animal Care and Use Committee of Fudan University.
Images were analyzed using the confocal microscope.
TUNEL staining was carried out using the DeadEnd TM Fluorometric TUNEL System (Promega, G3250) according to the manufacturer's instructions.
The clumps of tubules were then transferred to 30 μL of ice-cold 100 mM sucrose and mashed gently using tweezers to obtain a cell suspension.An additional 30 μL of ice-cold 100 mM sucrose was added to the cell suspension and mixed several times.Positively charged slides were incubated in the ice-cold fixation solution (2% paraformaldehyde, 0.1% Triton X-100, and 0.02% SDS, pH 9.2) for 3 min.30 μL of the diluted cell suspension was applied to the slides and incubated in humid chambers at room temperature for 2 h.The slides were washed with 0.4% Photo-Flo 200 and stored at -80°C.
Znhit1 in situ hybridization.Testis sections (5 μM) at the indicated times were prepared for Znhit1 in situ hybridization with the RNAscope kit (Advanced Cell Diagnostics, 323100) according to the manufacturer's instructions.

Quantitative reverse transcription PCR (RT-qPCR).
Total RNA was isolated using the RNeasy Mini-plus Kit (Qiagen, 74134) and then reverse-transcribed into complementary DNA (cDNA) with the GoScript Reverse Transcription System (Promega, A5003).cDNA was used as the template for the quantitative PCR assay using 2×SYBR Green qPCR Master Mix (Bimake, B21202).Quantitative PCR primers are listed in Table S8.
RNA-seq library generation and sequencing.Total RNA from fresh testes was isolated and mRNA was purified with magnetic beads (Vazyme, N401-01).Then, mRNA was fragmented and processed to generate RNA-seq libraries (Vazyme, NR605-01).Over 40 million reads were obtained per sample using the Illumina NovaSeq platform for 2 independent biological replicates.
Single-cell RNA sequencing (scRNA-seq).P16 control and Znhit1-sKO testes were digested by collagenase IV and trypsin at 37°C for 10 min to obtain testicular cell suspensions.scRNA-seq libraries were constructed using a 10 × Genomics kit and sequenced on the Illumina platform.
ChIP-seq library generation and sequencing.ChIP-seq library generation was performed as previously described.Briefly, P14 control and Znhit1-sKO testes were digested by collagenase IV and trypsin at 37°C for 10 min, crosslinked with 1% formaldehyde for 10 min, and quenched with glycine.The cells were lysed and sheared with a Bioruptor Plus machine for 20 min.Then, 2 μg of anti-H2A.Z antibody (Abcam, ab4174) was added to the sonicated chromatin and incubated overnight at 4°C.The following day, 20 μL of protein G beads were added and incubated for 2 h at 4°C.The beads were washed and reverse-crosslinked for 4 h at 65°C.DNA was extracted using phenol-chloroform and subjected to library construction using the VAHTS Universal DNA Library Prep Kit for Illumina (Vazyme, ND607-01) according to the manufacturer's instructions.Over 30 million reads were obtained per sample using the Illumina NovaSeq platform for 2 independent biological replicates.

KAS-seq library and ATAC-seq library generation and sequencing.
Testes were digested by collagenase IV and trypsin at 37°C for 10 min.Pachytene cells were sorted using fluorescence-activated cell sorting (FACS) as previously described (Long et al., 2017).For KAS-seq, pachytene cells were labeled with N3-kethoxal, and DNA was isolated using the DNA Clean and Concentrator kit (Zymo, D4013) and subjected to library construction using Q5 high-fidelity DNA polymerase (New England Biolabs, construction.For ATAC-seq, pachytene nuclei were isolated and fragmented with Tn5 transposase.DNA was isolated using the DNA Clean and Concentrator kit (Zymo, D4013) and subjected to library construction using Q5 high-fidelity DNA polymerase (New England Biolabs, M0544S).For high-throughput sequencing, over 30 million reads were obtained per sample using the Illumina NovaSeq platform for 2 independent biological replicates.
Gene Set Enrichment Analysis (GSEA) was carried out using GSEA software (Subramanian et al., 2005).GO analysis was performed using clusterProfiler (v4.6.2) (Wu et al., 2021).scRNA-seq data analysis.FASTQ files were run through CellRanger (v7.1.0)software with default parameters for de-multiplexing, aligning reads with STAR software to mm10, and counting unique molecular identifiers (UMIs).As input files of the Seurat R package (v4.4.0), the filtered gene expression matrices were then used for downstream analyses (Butler et al., 2018).Low-quality cells were filtered (expressing < 500 or >6,000 unique gene counts and >15% mitochondrial reads).Principal component analysis was performed on SCT-transformed data using 3,000 variable genes.The top 50 principal components were used for clustering and visualized using the UMAP algorithm in the Seurat R package.The "FindAllMarkers" function of the Seurat R package was used to calculate cluster-specific genes.Marker genes for each cluster are shown in Table S2.
Then, the output loom file was opened by the SCENIC R package (v1.3.1), and AUC values of regulons were extracted for visualization of downstream transcription factor activity.
For visualization, bam files were converted to bigWig files using deepTools (v3.3.1)(Ramirez et al., 2014), and bigWig files were used to calculate tag density under 50 bp resolution.In addition, CPM was used to normalize the number of reads per bin.We used the R package DiffBind (v3.8.4) (http://bioconductor.org/packages/release/bioc/html/DiffBind.html) for differential peak analysis and the R package rGREAT (v2.1.11)(Gu and Hubschmann, 2023) for peak annotation.We used ComputeMatrix and plotHeatmap of deepTools to generate count matrices and heatmaps, respectively.H2A.Z enrichment within chromatin states was annotated using the ChromHMM model (Ernst and Kellis, 2012;Spruce et al., 2020).TF motif enrichment was calculated using HOMER (v4.11) (Heinz et al., 2010).We used deepTools to draw Pearson's correlation scatter plots for the correlation between replicates of each experiment.
For promoter and enhancer type analysis, H3K27ac ChIP-seq data in wild-type pachytene spermatocytes were downloaded from the published data (GSE107398) (Adams et al., 2018); ChIP-seq data for H3K4me1, H3K4me3, and H3K27me3 in wild-type pachytene spermatocytes were downloaded from the published data (GSE132446) (Chen et al., 2020).

Supplementary Materials
This manuscript file includes Figs.S1 to S9 and Tables S1 and S8.
). Examination of germ cell scRNA-seq profiles via unsupervised clustering revealed decreased pachytene spermatocytes after Znhit1 deletion (Figs 4D, 4E, and S4B).Pou5f2 is expressed explicitly at the diplotene stage, and we found a lack of Pou5f2 + diplotene spermatocytes in Znhit1-sKO testes (Fig 4F).These results indicate that Znhit1 is required for male germ cell development beyond the pachytene stage.
decreased H2A.Z signals.This analysis identified several MYB family TFs, including MYB, A-MYB, and B-MYB (Fig 7A).Transcriptional network analysis conducted by SCENIC further identified A-MYB (encoded by Mybl1 gene) as the core regulatory TF in spermatocytes (Fig 7B).We reanalyzed RNA-seq data in P14 testes from A-MYBdeficient mice(Li et al., 2013) and found that A-MYB-deficient DEGs correlated with those observed in Znhit1-deficient testicular cells(Fig 7C).Although Znhit1 deletion didn't affect Mybl1 mRNA expression, transcriptome analyses showed that the expression of A-MYB target genes was significantly reduced in Znhit1 -/-pachytene spermatocytes (Figs 7D, 7E, S8A, and S8B).These results suggest that Znhit1 -/-and Mybl1 -/-pachytene spermatocytes share common gene expression alterations.Next, we compared the genome binding sites of H2A.Z and A-MYB.Utilizing published A-MYB ChIP-seq data from P14 testicular cells(Li et al., 2013), we identified 6,088 A-MYB binding signals, with 78.1% (4755) of these coinciding with H2A.Z peaks (Fig 7F).Annotation of genomic features showed that approximately 66.52% and 12.02% of these overlapping sites occupied active promoters and enhancers, respectively, representing a stronger enrichment compared to other regulatory elements (Fig 7G).Genes associated with H2A.Z and A-MYB binding included essential HR-related genes, such as Ccnb1ip1 (encode the Hei10 protein) (Fig 7H).Moreover, we showed that Znhit1 deletion resulted in a marked reduction in H2A.Z signals and nascent transcription identified by KAS-seq signals (Fig 7I and 7J).
Statistics and reproducibility.Statistical analyses were performed using R and Prism 9. Data are presented as means ± s.d.unless otherwise indicated.The two-tailed, unpaired, Student's t-test, Mann-Whitney test, or one-way or two-way ANOVA were performed to analyze statistical significance.

Figure 4 |Figure 5 |
Figure 4 | Spermatocyte transcriptome analysis using scRNA-seq.(A) UMAP plot showing the annotated germ cells captured from both control and Znhit1-sKO testicular cells.(B) Expression of selected markers identifying major somatic cell types and germ cell types on the UMAP plot.UMAP plot showing the expression patterns of Znhit1 in control or Znhit1-sKO testicular cells.(C) Heatmap showing the top marker genes of each cell cluster from scRNA-seq data.U-Sg, undifferentiated spermatogonia; D-Sg, differentiating spermatogonia; PreL, preleptotene; Lep, leptotene; Zyg, zygotene; Pac, pachytene.(D) Top: UMAP plot showing the annotated germ cells captured from control or Znhit1-sKO testicular cells.Down: UMAP plot showing the expression patterns of Znhit1 in control or Znhit1-sKO testicular cells.(E) Bar plot showing the percentage of major spermatocyte types for control or Znhit1-sKO testicular cells.(F) UMAP plot

Figure 6 |
Figure 6 | Znhit1 regulates H2A.Z binding on promoter and enhancer regions.(A) Immunostaining of SYCP3 and H2A.Z in spermatocyte chromosome spreads of control or Znhit1-sKO mice.The dashed circle indicates X-Y chromosomes.Scale bar, 10 μm.(B) Representative tracks of END-seq, DMC1-SSDS, and H2A.Z ChIP-seq in wild-type, control, or Znhit1-sKO testicular cells.(C) Heatmaps of chromatin states produced by ChromHMM and showing different enrichment for H2A.Z in control or Znhit1-sKO testicular cells.(D) Heatmaps and mean plots of H2A.Z ChIP-seq signal in control or Znhit1-sKO testicular cells (n = 2).(E) Venn diagram showing the overlap between decreased H2A.Z-bound genes and downregulated genes in Znhit1-sKO testicular cells.

Figure 7 |
Figure 7 | Znhit1/H2A.Z/A-MYB axis regulates enhancer activation and gene expression.(A) Motif analysis of active promoters or active enhancers with decreased H2A.Z binding for putative transcription factor (TF)-binding sites using the HOMER database.(B) Ridge map showing the A-MYB activity (AUC) at each cell type.(C) Scatter plot with marginal histograms comparing the fold change of DEGs from A-MYBrelated RNA-seq data with Znhit1-related RNA-seq data in P14 testicular cells.(D)

Figure S4 .
Figure S4.Single-cell RNA-seq of testicular cells.(A) Dot plot showing the relative expression and the percentage of cells expressing selected markers across scRNA-seq clusters.(B) Violin plot showing the normalized expression of Znhit1 in control and Znhit1-sKO spermatocytes at consecutive stages.

Figure S5 .
Figure S5.Single-cell transcriptomic analysis of meiotic genome activation.(A) Violin plots showing the average expression level of upregulated or downregulated genes between two consecutive spermatocyte stages.(B) Violin plots showing the average expression level of upregulated or downregulated genes between P14 control and Znhit1-sKO testicular cells.PreL, preleptotene; Lep, leptotene; Zyg, zygotene; Pac, pachytene.

Figure S6 .
Figure S6.differentially expressed genes (DEGs) in P14 testicular cells between control and Znhit1-sKO mice.(B) Gene ontology (GO) analysis of DEGs showing changed biological processes.(C and D) Bar charts showing HR-related gene expression in P14 control and Znhit1-sKO testes.Data from bulk RNA-seq are shown in (C).Validated data from RT-qPCR are shown in (D).Ctrl, control; sKO, Znhit1-sKO.

Figure S7 .
Figure S7.H2A.Z is enriched at meiotic gene promoters and enhancers.(A) Schematic representation of the components of the SRCAP chromatin remodeling complex.(B) Heatmaps showing H2A.Z ChIP-seq signals surrounding different types of promoters.(C) Heatmaps showing H2A.Z ChIP-seq signals surrounding different types of enhancers.

Figure S8 .
Figure S8.Znhit1 deletion downregulates the expression of A-MYB target genes.(A) Violin plot showing the normalized expression of Mybl1 in control and Znhit1-sKO spermatocytes at consecutive stages.(B) Cluster annotation on the basis of the expression of A-MYB target genes in control or Znhit1-sKO testicular cells.

Figure S9 .
Figure S9.Reproducibility of replicates in this study.(A) Scatter plots showing H2A.Z enrichments in P14 control or Znhit1-sKO testes for two independent H2A.Z ChIP-seq replicates.Pearson correlation coefficients are indicated.(B) Scatter plots showing KASseq signals in P14 control or Znhit1-sKO pachytene cells for two independent H2A.Z ChIP-seq replicates.Pearson correlation coefficients are indicated.(C) Scatter plots showing ATAC-seq signals in P14 control or Znhit1-sKO pachytene cells for two independent H2A.Z ChIP-seq replicates.Pearson correlation coefficients are indicated.