All animal experiments were conducted in compliance with the ethical standards of the Institutional Animal Care and Use Committee (IACUC) in Zhongnan Hospital of Wuhan University (approval number: ZN2025039). Mice were maintained under specific pathogen-free (SPF) conditions with a 12 h light/dark cycle and ad libitum access to food and water. All procedures were performed in accordance with the guidelines of the National Institutes of Health for the Care and Use of Laboratory Animals.

To investigate the physiological function of NSMCE2 in male germ cells, we generated a germ cell–specific Nsmce2 conditional knockout (cKO) mouse model based on the canonical Nsmce2 transcript (RefSeq: NM_001374768.1). The Nsmce2 ^flox/flox line was established by inserting two loxP sites flanking exon 3 of the Nsmce2 gene (Cyagen Biosciences, China). These mice were crossed with Stra8-Cre transgenic mice, in which Cre recombinase is expressed in pre-meiotic spermatogonia starting from 3 days postpartum (3 dpp) (Li et al., 2024). Offspring with the genotype Nsmce2 ^flox/- ; Stra8-Cre were designated as cKO mice, while Nsmce2 ^flox/- littermates were used as controls. Genotyping was performed by polymerase chain reaction (PCR) using tail DNA and specific primers: Nsmce2-F1: 5′-ACG​GTA​GTT​GAG​CTT​TCT​TCT​GTA​T-3′, Nsmce2-R1: 5′-AGC​ACC​ATA​TCT​GGC​TAA​GTG​AG-3′, Nsmce2-F2: 5′-CCTCTTTA GTA​GTG​CCA​TGT​TGA​G-3′, Nsmce2-R2: 5′-ATA​AAG​AAA​CCA​GAA​GTG​CAA​GGCA-3′.

To evaluate reproductive performance, 8 week-old cKO and WT male mice were each paired with two adult WT females for at least six consecutive months. The number of litters and the total number of pups sired by each male were recorded. To assess fertility capacity, the average litter size and the cumulative number of offspring per female were calculated and statistically analyzed.

Testes and epididymides from adult mice were collected, fixed in Bouin’s solution, embedded in paraffin, and sectioned at 5 μm thickness. Slides were stained with hematoxylin and eosin (H&E) following standard protocols. The morphology of seminiferous tubules, germ cell organization, and epididymal sperm content were examined using a Nikon light microscope (Tokyo, Japan). At least 50 seminiferous tubules per animal were evaluated to ensure reproducibility.

For immunofluorescence, testes were fixed in 4% paraformaldehyde, dehydrated in 30% sucrose, embedded in OCT compound, and cryosectioned at 7 μm. After antigen retrieval and blocking in 5% bovine serum albumin (BSA), sections were incubated overnight at 4 °C with primary antibodies, followed by fluorophore-conjugated secondary antibodies. The following primary antibodies were used: anti-NSMCE2 (1:100, Proteintech, 13627-1-AP), anti-DDX4 (1:200, Abcam, ab13840), anti-SOX9 (1:500, Abclonal, A19710), anti-PCNA (1:200, Proteintech, 10205-2-AP), anti-γH2AX (1:200, Solarbio, K001451M). After counterstaining with DAPI (Sigma-Aldrich), sections were mounted and imaged using a confocal laser scanning microscope (Zeiss LSM880, Germany). Co-localization analyses were performed using ImageJ software (NIH).

Total protein extracts were obtained from testes and other tissues using RIPA buffer (Beyotime Biotechnology) supplemented with PMSF protease inhibitor. After homogenization and centrifugation at 12,000 rpm for 10 min at 4 °C, the supernatants were mixed with SDS loading buffer, boiled for 10 min, and subjected to SDS-PAGE. Proteins were transferred onto PVDF membranes (Millipore) and blocked in 5% non-fat milk in TBST for 1 h, and incubated overnight at 4 °C with the primary antibody. After three washes in TBST, the membranes were incubated with HRP-conjugated secondary antibody (Proteintech) at room temperature for 1 h. The primary antibodies used were as follows: anti-GAPDH (1:100000, Proteintech, 60004-1-Ig), anti-βTubulin (1:2500, Proteintech, 10094-1-AP), anti-NSMCE2 (1:1000, Proteintech, 13627-1-AP), anti-SUMO1 (1:1000, Abclonal, A19121), anti-SUMO2/3 (1:500, Abclonal, A5066), anti-PIAS1 (1:3000, Proteintech, 23395-1-AP), anti-PIAS4 (1:500, Proteintech, 14242-1-AP), anti-ZNF451 (1:500, Proteintech, 25228-1-AP).

Cauda epididymides were dissected from 8 week-old mice and minced in 1 mL of phosphate-buffered saline (PBS), followed by incubation at 37 °C for 30 min to release sperm. Sperm were stained using Giemsa solution (Servicebio, China) and analyzed under a light microscope. Sperm concentration and motility were quantified using a hemocytometer, and morphological abnormalities were evaluated in at least 200 sperm per sample.

Total RNA from tissues was extracted with TRIzol reagent, and then reverse transcribed with HiScript® II Q RT SuperMix for qPCR (+gDNA wiper) kit (Vazyme), and amplified with ChamQ SYBR qPCR Master Mix kit (Vazyme). The expression level was normalized to Gapdh expression. The primer sequence used was as follows: Gapdh-F: 5′-AGG​TCG​GTG​TGA​ACG​GAT​TTG-3′, Gapdh-R: 5′-TGTAGACCATGTAGTTGAGGTCA-3′, Nsmce2-F: 5′-GTG​GGA​CAG​CCA​GGT​AAC​AA-3′, Nsmce2-R: 5′-AGC​AAA​CTC​AAC​CAT​GGC​CT-3′.

Spermatocyte nuclear spreads were prepared as described previously (Fang et al., 2021). Briefly, seminiferous tubules were incubated in hypotonic extraction buffer (30 mM Tris, 50 mM sucrose, 17 mM trisodium citrate, 5 mM EDTA, pH 8.2) for 30 min and gently dispersed. Cell suspensions were dropped onto slides coated with 1% paraformaldehyde containing 0.15% Triton X-100 and air-dried overnight. For immunostaining, slides were blocked with 5% BSA and incubated with primary antibodies against SYCP3, SYCP1, γH2AX, SUMO1, MDC1, and RNA pol II at 4 °C overnight, followed by Alexa Fluor–conjugated secondary antibodies (Invitrogen). Nuclei were counterstained with DAPI and mounted with antifade reagent. Images were captured using a Zeiss LSM880 confocal microscope, and at least 100 pachytene spermatocytes per animal were analyzed for synapsis and recombination foci using ImageJ software. The primary antibodies used were as follows: anti-γH2AX (1:200, Solarbio, K001451M), anti-SYCP1 (1:200, Abcam, ab15090), anti-SYCP3 (1:200, Abcam, ab15093), anti-SUMO1 (1:100, Abclonal, A19121), anti-MDC1 (1:50, Abclonal, A12714), anti-RNA pol II (1:50, Abcam, ab193468).

Testis samples were collected from 8 week-old testes, followed by RNA extraction, RNA detection, library construction, sequencing, and data analysis. Differentially expressed genes were filtered using the criteria |log2FoldChange| ≥ 1 and FDR <0.05. Above processes were performed by Metware Company.

All experiments were independently repeated at least three times using biological replicates. Data calculations and graphing were performed using GraphPad Prism 9. Data were expressed as mean ± standard error of the mean (SEM). Statistical analysis was conducted using Student’s t-test. P values <0.05 were considered statistically significant.

Multiple sequence alignment revealed that NSMCE2 is highly conserved across species. Specifically, the sequence similarity between human and mouse is as high as 84.21%, which indicates the conservation of biological function (Figure 1A). To investigate the expression pattern of Nsmce2, we validated its expression in various tissues of adult mice using quantitative reverse-transcription PCR (qRT-PCR) and Western blotting. We found that Nsmce2 expression, both at the mRNA and protein levels, was significantly higher in the testes than in other organs such as the heart, brain, and lung (Figures 1B,C).

Furthermore, we investigated changes in Nsmce2 expression across different postnatal testicular stages (Figures 1D,E). From postpartum day 8 (8dpp) to 14dpp, both mRNA and protein levels of NSMCE2 remained low and relatively stable. A striking increase was observed at 18dpp. Subsequently, from 21dpp to 56dpp, NSMCE2 expression remained relatively stable at high levels. The sharp upregulation of NSMCE2 at 18dpp coincides with a prominent phase within the first wave of meiosis, a period during which spermatocytes are predominantly in the pachytene and diplotene stages. This suggests that NSMCE2 may play a potential role in these specific meiotic phases. Consistently, single-cell sequencing data from the FertilityOnline database (FertilityOnline, 2021) indicate that NSMCE2 exhibits higher expression during the meiotic stages than in spermatids and somatic cells in the testis of adult mice. By co-staining NSMCE2 with γ-H2AX (a DSB marker) and DAPI in testicular immunofluorescence, it was found that NSMCE2 was specifically expressed in the nuclei of pachytene and diplotene spermatocytes (Figure 1F; Supplementary Figure S1).

To investigate the physiological role of NSMCE2 in spermatogenesis, we generated germ cell–specific Nsmce2 conditional knockout (cKO) mice. Using CRISPR-Cas9 technology, we designed a strategy to target exon 3 of the Nsmce2 gene on chromosome 15 in testicular germ cells, aiming to generate a 1750 bp gene fragment deletion. This deletion is predicted to lead to the loss of Nsmce2 gene function in mice (Figure 2A). Specifically, Nsmce2 ^flox/flox mice were generated by inserting a LoxP site on both sides of exon 3. Crossbreeding Nsmce2 ^flox/flox mice with Stra8-Cre mice yielded Nsmce2 ^flox/- ; Stra8-Cre mice (hereafter termed Nsmce2-cKO) and Nsmce2 ^flox/- mice (hereafter termed control) after two generations (Figure 2B).

Following polymerase chain reaction (PCR) amplification of genomic DNA extracted from mouse tails, genotypes were identified via agarose gel electrophoresis. Nsmce2 ^flox/- mice exhibited only the floxed allele band (213 bp), and wildtype (WT) mice showed only the WT allele band (146 bp), while Nsmce2 ^flox/+ mice displayed both the floxed allele band (213 bp) and the WT allele band (146 bp) (Figure 2C). Furthermore, RT-PCR analysis of testicular cDNA confirmed the presence of the specific deletion allele (about 300–400 bp) in cKO mice, verifying the successful excision of exon 3 at the transcript level (Supplementary Figure S2). Western blotting and immunofluorescence also confirmed the successful knockout of NSMCE2 in germ cells (Figures 2D,E). These results confirmed the successful generation of germline-specific Nsmce2 knockout mice.

To assess the impact of Nsmce2 deficiency on male fertility, we conducted a comprehensive analysis of reproductive organ morphology and function. Macroscopic examination revealed no discernible differences in testicular and epididymal size or morphology between Nsmce2 cKO mice and control mice (Figure 3A). Consistent with this, hematoxylin-eosin (H&E) staining revealed normal seminiferous tubule architecture in cKO mouse testes, with comparable sperm content in both the caput and cauda epididymides compared to controls (Figure 3B). The testis-to-body weight ratio remained unchanged in cKO mice (Figure 3C).

Crucially, functional fertility assessments revealed that Nsmce2 cKO male mice exhibited full reproductive capacity. Their average litter size (Figure 3D) and cumulative number of offspring produced within 6 months post-mating (Figure 3E) showed no statistically significant differences compared to controls. Furthermore, sperm analysis revealed no significant defects in key parameters, including sperm count (Figure 3F), percentage of forward-moving sperm (Figure 3G), and sperm with normal morphology (Figure 3H). Collectively, these data indicate that germ cell-specific knockout of Nsmce2 does not result in apparent impairment of testis development, spermatogenesis, or male fertility.

To investigate whether NSMCE2 deficiency disrupts testicular cellular architecture or induces cellular stress responses, we performed immunofluorescence staining for key molecular markers. Immunofluorescence staining for DDX4 (a pan-germ cell marker) confirmed the presence of germ cells at all stages within the seminiferous tubules of Nsmce2 cKO mice, with distribution and density comparable to controls (Figures 4A,B). PNA staining (marking the acrosome of developing spermatids) revealed normal patterns of acrosome formation and elongation in spermatids at specific stages (Figure 4A). Furthermore, SOX9 staining (a marker specific to Sertoli cells) showed no alteration in the number or distribution of Sertoli cells in cKO testes (Figures 4C,D).

To directly assess potential cellular damage, we performed TUNEL staining to detect apoptosis. Results showed no significant increase in TUNEL-positive cells per seminiferous tubule in Nsmce2 cKO testes compared to controls (Figures 4E,F). This indicates that NSMCE2 deficiency in germ cells does not cause widespread germ cell death. Finally, immunofluorescence staining for PCNA (a proliferating cell marker) revealed that the proliferation index of spermatogonia remained normal in the absence of NSMCE2 (Figures 4G,H). Collectively, analysis of these key cellular markers indicates that spermatogenesis proceeds following NSMCE2 deficiency with minimal impact on germ cell homeostasis, Sertoli cell presence, or overall testicular cytological architecture.

Given the time-specific expression of NSMCE2 in testes and its known role in genomic stability, we next investigated its function during meiosis. Immunofluorescence analysis of chromosome spreads revealed the spatiotemporal localization of NSMCE2 in WT spermatocytes. During the leptotene and zygotene stages, NSMCE2 was undetectable on chromosomes. Strikingly, it was specifically enriched within the sex body throughout the pachytene and diplotene stages (Figure 5A). This unique localization pattern suggests that NSMCE2 may play a potential role in the sex body-specific chromatin environment during prophase I of meiosis.

Despite this specific localization, NSMCE2 deficiency in germ cells did not disrupt fundamental meiotic events. We first quantified the proportions of spermatocytes at different stages of prophase I. Our statistical analysis revealed that the proportion of leptotene, zygotene, pachytene, and diplotene cells in Nsmce2 cKO mice was comparable to that in control groups, indicating that meiotic progression was not grossly impaired (Figure 5B). Consistent with this, staining of SYCP3, a lateral component of the synaptonemal complex, and SYCP1, a central component protein, revealed that homologous chromosome assembly and synapsis proceeded normally in Nsmce2 cKO spermatocytes, indistinguishable from controls (Figure 5C). Furthermore, the repair of meiotic DNA double-strand breaks (DSBs), monitored by the formation and subsequent disappearance of γ-H2AX foci, remained unaffected. In control spermatocytes, γ-H2AX signaling was broadly present during the leptotene and zygotene stages and concentrated on the sex body during the pachytene and diplotene stages. This pattern was precisely reproduced in Nsmce2 cKO spermatocytes (Figure 5D), indicating that signaling and repair of meiotic DSBs proceed efficiently in the deficiency of NSMCE2. To investigate whether Nsmce2 deficiency compromises Meiotic Sex Chromosome Inactivation (MSCI), we examined the expression of key markers, including SUMO1, MDC1 and RNA Pol II (Figures 6A–F). During pachytene stage in cKO mice, SUMO1 and MDC1 were correctly localized to sex body, and RNA Pol II was effectively excluded from sex body, with fluorescence intensities of these markers all comparable to those in controls (p > 0.05). Collectively, these data demonstrate that Nsmce2 is dispensable for the establishment and maintenance of MSCI.

In summary, although NSMCE2 exhibits a highly specific spatiotemporal localization pattern during meiotic prophase I, its deficiency in germ cells does not significantly impair chromosome synapsis, DSB repair or MSCI, possibly suggesting the existence of compensatory pathways that maintain meiotic integrity.

To determine whether Nsmce2 loss induces transcriptional changes, RNA-sequencing was performed on adult testes from WT and cKO mice. Principal component analysis showed minimal separation between the two groups, indicating overall transcriptional stability (Figure 7A). Only a small set of genes exhibited differential expression (Figure 7B). Among these 135 altered genes, only 13 possess known biological functions (including Nsmce2). Genes related to meiosis, DNA repair, or spermatid differentiation remained largely unchanged (Figures 7C,D). Gene Ontology (GO) and KEGG analyses confirmed that only few differentially expressed genes (DEGs) were enriched in terms related to meiosis or chromatin maintenance. (Figures 7E–G). These findings suggest that the germline compensates for Nsmce2 loss without widespread transcriptional remodeling, further reflecting the intrinsic robustness of the spermatogenic system.

Collectively, the absence of morphological, cytological, and transcriptional abnormalities in Nsmce2 cKO testes suggests that meiotic progression is preserved through redundant or compensatory mechanisms. To investigate the molecular basis of this resilience, we examined the global SUMOylation landscape in mouse testes. Western blotting analysis revealed that the overall levels of SUMO1 and SUMO2/3 remained comparable between control and cKO groups (Figures 8A,B), indicating that Nsmce2 deficiency does not compromise the global SUMO environment.

We hypothesized that other SUMO E3 ligases might functionally compensate for NSMCE2. We screened several candidate ligases known to operate in the germline, including PIAS1 (Constanzo et al., 2016; Shah et al., 2024), PIAS4 (Gallardo-Chamizo et al., 2025; Han et al., 2022), and ZNF451 (Bouchard et al., 2025; Cappadocia et al., 2015). While the protein levels of PIAS4 (Figures 8D,G) and ZNF451 (Figures 8E,H) remained unchanged, quantitative analysis revealed a significant upregulation of PIAS1 in cKO testes compared to controls (p < 0.05; Figures 8C,F). These findings suggest that PIAS1 may be upregulated to compensate for the deficiency of NSMCE2, potentially contributing to the maintenance of SUMO-dependent DNA repair and chromatin stability. Such potential functional redundancy of NSMCE2 likely supports the stability of the meiotic program under physiological conditions, highlighting the robust and fail-safe design of the male germline.

Thus, our findings reveal that Nsmce2 is not essential for meiosis or fertility but instead exemplifies the system-level resilience of spermatogenesis, wherein multiple molecular pathways collectively safeguard genomic integrity.