Mei4 deficiency induces sexual dimorphism in early meiosis and promotes production of unreduced eggs in zebrafish

Abstract

Polyploidy frequently originates from meiotic failure, with mis-segregation of homologous chromosomes commonly arising from impaired formation of programmed DNA double-strand breaks (DSBs). Meiotic double-stranded break formation protein 4 (Mei4) is essential for recruiting the Sporulation-specific protein 11 (Spo11) nuclease to chromosome axes to initiate DSB formation in yeast and mice, yet its role in fish remains uncharacterized. In this study, mei4^−/− zebrafish mutants were generated and exhibited severe defects in synapsis, DSB formation, homologous recombination, and crossover (CO) formation during meiosis in both male and female germ cells. Notably, pronounced sexual dimorphism in responses to Mei4 deficiency was observed during meiotic prophase I, with males exhibiting complete sterility due to meiotic arrest, while females produced numerous aneuploid oocytes and a minority of unreduced eggs due to defective chromosome segregation. Remarkably, crosses between mei4^−/− females and wild-type males yielded a substantial number of viable triploid progeny. These findings demonstrate functional conservation of Mei4 in mediating meiotic DSB formation across vertebrates, reveal sex-specific divergence in meiotic responses to recombination failure in this vertebrate model, and highlight a critical role for DSB-dependent crossover failure in the genesis of polyploidy.

INTRODUCTION

Accurate segregation of homologous chromosomes during meiosis is essential for the production of viable gametes and depends on the induction and repair of programmed DNA double-strand breaks (DSBs). These breaks initiate homolog pairing and promote synaptonemal complex (SC) assembly, a three-part protein-based scaffold composed of paired lateral elements and an intervening central region, which together align homologous chromosomes along their length (Gao & Colaiácovo, 2018; Mercier et al., 2015; Zickler & Kleckner, 2015). The SC provides a structural platform that facilitates DSB repair through homologous recombination, ensuring the formation of at least one crossover (CO) per homolog pair (Hunter, 2015; Zickler & Kleckner, 2015). These crossovers establish chiasmata, which serve as physical linkages required for proper orientation and segregation of homologs during meiosis Ⅰ (Hunter, 2015; Petronczki et al., 2003; Zickler & Kleckner, 2015). To balance recombination efficiency with genomic stability, DSB formation is under tight spatiotemporal regulation (Dereli et al., 2021; Keeney et al., 2014; Thacker et al., 2014).

In diverse eukaryotes, the formation of meiotic DSBs is catalyzed by Sporulation-specific protein 11 (SPO11), a topoisomerase-like enzyme, in complex with its co-factor TOPOVIBL, a Topoisomerase VI (TopoVI) subunit B-like protein (Bergerat et al., 1997; Keeney et al., 1997; Robert et al., 2016; Vrielynck et al., 2016). This core machinery operates alongside a suite of auxiliary pro-DSB proteins, including meiotic double-stranded break formation protein 4 (MEI4) (Claeys Bouuaert et al., 2021; Kumar et al., 2010; Kumar et al., 2015; Lam & Keeney, 2015). First characterized in yeast, MEI4 is required for SPO11 recruitment and DSB formation, and its disruption leads to recombination failure and meiotic arrest (Arora et al., 2004; Keeney, 2001; Maleki et al., 2007; Menees & Roeder, 1989; Menees et al., 1992; Prieler et al., 2005). However, due to low primary sequence conservation, MEI4 homologs remained elusive in vertebrates until its functional role in recruiting SPO11 was demonstrated in mice (De Muyt et al., 2009; Keeney, 2008; Kumar et al., 2010; Kumar et al., 2015; Lam & Keeney, 2015; Richard et al., 2005). In mouse models, MEI4 deficiency abrogates DSB formation and synapsis, leading to early meiotic arrest in both spermatogenesis and oogenesis (Kumar et al., 2010; Kumar et al., 2015). However, whether MEI4 serves a conserved role in teleost meiosis remains unknown. Although the functions of DSB-associated proteins appear conserved, the primary sequences of genes such as MEI4 and its partner REC114 Meiotic Recombination Protein (REC114) show weak homology across eukaryotes—from yeast to plants to mammals—reflecting rapid evolutionary divergence of the meiotic DSB machinery (Daccache et al., 2023; De Muyt et al., 2009; Keeney, 2008; Laroussi et al., 2023; Richard et al., 2005).

Polyploidy often arises from the production of unreduced gametes due to rare failures during mitosis or meiosis (Comai, 2005). Although multiple polyploid lineages have been documented in fish, the underlying cytogenetic mechanisms and molecular determinants remain largely unknown (Burgess, 2015; Gui et al., 2022; He et al., 2018; Ou et al., 2024; Ren et al., 2024; Van De Peer et al., 2017). Recent genomic studies have revealed that amphitriploid Carassius gibelio generates unreduced eggs via an alternative ameiotic program that likely involves a reconfigured regulatory network of meiosis-related genes (Gui, 2024; Wang et al., 2022). In the present study, the functional role of Mei4 in zebrafish was investigated through targeted gene editing. mei4-deficient zebrafish exhibited pronounced meiotic defects in synapsis, DSB formation, homologous repair, and crossover formation, with clear sex-specific differences in meiotic progression. Notably, mutant mei4^−/− males were sterile due to early meiotic arrest, while females generated abundant aneuploid oocytes and a subset of unreduced eggs. Remarkably, crosses between mei4^−/− females and wild-type (WT) males yielded a significant number of viable triploid progeny, revealing a mechanistic link between recombination failure and polyploid formation in a vertebrate model.

MATERIALS AND METHODS

Zebrafish (Danio rerio) lines and maintenance

Wild-type AB strain zebrafish were maintained in a recirculating aquaculture system at the National Aquatic Biological Resource Center (NABRC). Breeding, husbandry, and developmental staging followed established protocols (Kimmel et al., 1995). All procedures were conducted under the approval of the Institutional Animal Care and Use Committee of the Institute of Hydrobiology, Chinese Academy of Sciences (Approval number IHB-20191001).

Generation of mei4 mutant zebrafish

Targeted disruption of mei4 was achieved using CRISPR-Cas9 mutagenesis protocols, as previously described (Hwang et al., 2013). Single-guide RNA (sgRNA) targeting mei4 was designed via ZiFiT Targeter (http://zifit.partners.org/ZiFiT/CSquare9Nuclease.aspx) and transcribed in vitro using a Transcript Aid T7 High-Yield Transcription Kit (Thermo Fisher Scientific, USA). Cas9 mRNA was synthesized from XbaI-digested pCS2-Cas9 using an mMESSAGE mMACHINE T7 ULTRA Kit (Ambion, USA). The primers for sgRNA template amplification were 5′-GTAATACGACTCACTATAGGAGGATCTGTTGAAGGAGAGTTTTAGAGCTAGAAATAGC-3′ and 5′-AAAAGCACCGACTCGGTGCC-3′.

Cas9 mRNA (500 ng/µL) and gRNA (50 ng/µL) were co-injected into AB strain embryos at the 1- or 2-cell stage. Injected founders were subsequently outcrossed with WT AB zebrafish, and F1 progeny were genotyped using primers 5′-TGGTTTGGGATTCAGTAGTGC-3′ and 5′-ATTTGCCTTGGTGATGTTGG-3′. Two distinct mei4 mutant alleles were identified: a 2-bp deletion (∆2) and a +20∆2 frameshift mutation in exon 3 (Figure 1B). Heterozygous carriers were intercrossed to generate homozygous mutants, which occurred at the expected frequency of 25% (data not shown) and exhibited normal external morphology. Fertility was evaluated by crossing adult mei4^−/− females with WT males. The number of unfertilized eggs and gastrula-stage embryos was counted at 6 h post-fertilization (hpf) under 28°C conditions. Fry counts were performed after hatching. Fertilization rate was calculated as the proportion of fertilized embryos out of the total embryos, and hatching rate as the proportion of hatched larvae relative to fertilized embryos, both expressed as percentages (Brion et al., 2004).

Figure 1.

Generation and characterization of mei4 mutant zebrafish

A: Tissue-specific expression of mei4 in adult zebrafish. ef1α served as the internal control. B: CRISPR/Cas9-mediated generation of mei4 mutants. The gRNA target sequence in exon 3 is underlined, and the protospacer adjacent motif is highlighted in green. Conserved Mei4 domain is indicated by a red box, frameshift amino acids by a gray box, and six conserved short similarity motifs (SSMs) by small boxes. C: Comparison of external morphology (left) and gonad morphology (right) between adult WT and mei4^−/− zebrafish. Scale bars: 10 mm (left) and 1 mm (right) in C.

RNA isolation and quantitative real-time polymerase chain reaction (PCR)

Total RNA was extracted from adult zebrafish testes and ovaries using the SV Total RNA Isolation System (Promega, USA). Complementary DNA (cDNA) was synthesized using the GoScript Reverse Transcription System (Promega, USA). Quantitative real-time PCR was performed using primers specific for mei4 (5′-GCTACTGCTAACCAACCAACAACT-3′ and 5′-GAGGGAAAACTCATCCGCCA-3′) and elongation factor 1α (ef1α) as the internal reference (5′-AGGCTGACTGTGCTGTGCTGA-3′ and 5′-CCAGGGTGAAAGCCAGGAGG-3′). Reactions were conducted in triplicate, and relative transcript abundance was calculated using the 2^−ΔΔCt method.

Histological analysis and apoptosis assays

Gonads from 4-month-old WT and mei4^−/− adults were fixed in 4% paraformaldehyde at 4°C for at least 24 h, dehydrated, and embedded in paraffin. Sections (4 μm) were subjected to hematoxylin and eosin (H&E) staining following standard procedures. Histological evaluation and staging of spermatogenesis and oogenesis were conducted according to established criteria (Figure 2) (Imai et al., 2021; Selman et al., 1993).

Figure 2.

Characterization of spermatogenesis and oogenesis in mei4 mutant zebrafish

A: Histological comparison of adult gonads from WT and mei4^−/− zebrafish. SPG, spermatogonia; SPC Ⅰ, primary spermatocytes; SPC Ⅱ, secondary spermatocytes; SPT, spermatids; SPZ, spermatozoa; MⅠ, meiotic metaphase Ⅰ; MⅡ, meiotic metaphase Ⅱ. Lumens lacking spermatozoa and containing arrested spermatocytes with abnormal chromosome segregation are indicated by red asterisks and outlined by dashed lines, respectively. PO, primary-growth oocyte; PVO, previtellogenic oocyte; VO, vitellogenic oocyte; MO, mature oocyte. B: Histological comparisons of spermatocytes at MⅠ and MⅡ in WT and mei4^−/− zebrafish. C: Quantification of testicular cell composition (left) and meiotic metaphase cells (right) in WT and mei4^−/− males. D: Distribution of oocytes across developmental stages in WT and mei4^−/− females. E: TUNEL staining of adult gonads from WT and mei4^−/− zebrafish. Apoptotic SPC Ⅰ (solid outlines), SPC Ⅱ (dashed outlines), SPT+SPZ (arrows), and MⅠ or MⅡ cells (arrowheads) are shown. F: Quantification of apoptotic testicular cells in WT and mei4^−/− zebrafish. Scale bars: 50 μm in A, E and 10 μm in B. Data are presented as mean±SEM. Each group includes three biological replicates. Statistical significance was assessed using unpaired two-tailed Student’s t-test. ^*: P<0.05; ^**: P<0.01; ^****: P<0.0001; ns: Not significant.

For morphometric assessment, all identifiable testicular or ovarian cell types were quantified. Testicular cells were counted across the entire cross-sectional area of seminiferous tubules in randomly selected, non-overlapping microscopic fields of view under 400× magnification. Ovarian follicles and interstitial cells were enumerated at 200× magnification under comparable sampling conditions. Each group included at least three biological replicates (n≥3).

Apoptotic cells were detected in 4-μm-thick sections based on terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) using an In-Situ Cell Death Detection Kit, Fluorescein (Roche, Switzerland). TUNEL-positive cells were quantified by cell type across the entire testis section at 400× magnification.

Testis and ovary chromosome spreads

For spermatocyte spreads, 4–6 intact testes from 2–3-month-old males were freshly dissected following anesthesia. Chromosome preparations were generated using a modified spreading protocol as previously described (Blokhina et al., 2019).

For oocyte spreads, ovaries from approximately four 2-month-old females were freshly dissected after anesthesia. Gonadal tissue was placed in a 1.5 mL tube containing 60 μL of phosphate-buffered saline (PBS), mechanically dissociated, and pelleted at 150 rpm for 90 s at room temperature. The supernatant was transferred to a clean 1.5 mL tube and mixed with three volumes of 75 mmol/L KCl and one-quarter volume of 4% paraformaldehyde. Cell suspensions were then spread onto slides, air-dried, and either stored at –20°C or processed immediately for immunostaining.

Immunostaining of chromosome spreads

Slides were subjected to antigen retrieval in boiling Citrate-EDTA Antigen Retrieval Solution (Beyotime, China) for 20 min, followed by permeabilization in PBS containing 0.1% Tween 20 and 0.1% Triton X-100 (Beyotime, China) for 10 min. Nonspecific binding was minimized by incubation in blocking solution (10% goat serum, 3% bovine serum albumin (BSA), 0.05% Triton X-100 in PBS, stored at –20°C) for 10 min at room temperature. Primary antibodies were applied overnight at 4°C in a humidified chamber. After washing, secondary antibodies and DAPI (Sigma, USA) were applied for 1 h at 37°C in the dark. All antibodies were diluted in blocking solution, as detailed in Supplementary Table S1. Following each incubation, slides were washed three times for 10 min each in PBS containing 0.04% Photo-Flo 200 and 0.03% Triton X-100. Finally, slides were mounted with VECTASHIELD Antifade Mounting Medium (Vector Labs, USA) and imaged using a Leica SP8 confocal microscope (Leica, Germany).

Image quantification

Sycp1 fragments were manually enumerated in all mid-zygotene/mid-zygotene-like and pachytene/pachytene-like spermatocytes on a per-nucleus basis (Figure 3), excluding punctate Synaptonemal complex protein 1 (Sycp1) signals. Synaptonemal complex protein 3 (Sycp3) axis length in spermatocytes and oocytes was measured from traced photographs using ImageJ. For Ser-139 phosphorylated histone H2AX (γH2AX) analysis, binary images were generated using a uniform threshold across all samples within each experiment to remove background fluorescence. Total γH2AX fluorescence intensity was calculated as the product of stained area and mean fluorescence intensity across all nuclear stages (Supplementary Figure S1).

Figure 3.

Synaptonemal complex (SC) assembly in mei4^−/− zebrafish

A: Immunofluorescence analysis of chromosome spreads from WT and mei4^−/− spermatocytes stained with anti-Sycp3 (green) and anti-Sycp1 (magenta). Meiotic substages were assigned based on Sycp3 and Sycp1 patterns: leptotene (L); early zygotene (EZ); mid zygotene (MZ); late zygotene (LZ); pachytene (P); leptotene-to-early zygotene-like (L/EZ-like); mid zygotene-like (MZ-like); pre-pachytene-like-to-pachytene-like (Post-LZ-like). Insets show magnified views of boxed regions. B: Immunofluorescence staining of WT and mei4^−/− oocyte chromosome spreads using anti-Sycp3 (green) and anti-Sycp1 (magenta). C: Total SC length across 25 chromosome pairs in pachytene spermatocytes and oocytes. Each group included 10 individuals. D: Proportion of gametocytes lacking Sycp1 or exhibiting detectable Sycp1 stretches during prophase Ⅰ in WT and mei4^−/− zebrafish. E: Self-assembly of Sycp1 filaments in mei4^−/− spermatocytes. Short Sycp1 stretches are indicated by white arrows; enlarged views are shown below. F: Frequency distribution of Sycp1 stretch numbers in mid-zygotene to pachytene (-like) spermatocytes with fully condensed axes in WT and mei4^−/− spermatocytes. G: Average Sycp3 axis lengths in mei4^−/− spermatocytes and oocytes during Post-LZ-like. Each group contained at least three cells. n denotes number of cells analyzed. Scale bars: 5 μm in A, B, E and 1 μm in enlarged views. Data are presented as mean±SEM. Statistical significance was assessed using unpaired two-tailed Student’s t-test. ^****: P<0.0001; ns: Not significant.

For Rad51 recombinase (Rad51) and Replication protein A2 (Rpa2) quantification, chromosome axis-proximal regions were defined as within 1 μm of Sycp3-stained axes, following criteria established for human DNA meiotic recombinase 1 (Dmc1) analysis (Pratto et al., 2014). Total Rad51 and Rpa2 foci per nucleus were counted using ImageJ (Figure 4; Supplementary Figure S2), applying identical thresholds to all images within each experiment. Rad51 and Rpa2 foci located within axis-proximal regions were additionally scored manually across all meiotic stages. All measurements were restricted to manually selected DAPI-positive nuclear areas.

Figure 4.

Formation of Rad51 foci is impaired in mei4^−/− zebrafish

A: Immunofluorescence analysis of chromosome spreads from WT and mei4^−/− spermatocytes stained with anti-Sycp3 (green), anti-Rad51 (red), and anti-Sycp1 (blue). B: Quantification of Rad51 fluorescence intensity per nucleus in WT and mei4^−/− spermatocyte spreads. C: Immunofluorescence staining of chromosome spreads from WT and mei4^−/− oocytes stained with anti-Sycp3 (green), anti-Rad51 (red), and anti-Sycp1 (blue). D: Quantification of Rad51 fluorescence intensity per nucleus in WT and mei4^−/− oocyte spreads. E: Distribution of DSBs in pachytene spermatocytes (left) and oocytes (right) in WT zebrafish. Bar plots depict Rad51 counts within successive 5% intervals of SC length measured from each end toward the center; “0” on the x-axis denotes either SC terminus. A total of 43 Rad51 foci across 28 male SCs and 46 Rad51 foci across 25 female SCs were detected. F: Comparison of Rad51 focus frequency per SC between WT pachytene spermatocytes and oocytes. n denotes number of cells analyzed. Scale bars: 5 μm in A, C and 1 μm in enlarged views. Data are presented as mean±SEM. Statistical significance was assessed using unpaired two-tailed Student’s t-test. ^**: P<0.01; ^****: P<0.0001; ns: Not significant.

DNA content analysis by flow cytometry

Embryonic cells were obtained from mechanically dissociated gastrula-stage embryos, while red blood cells were collected from adult caudal fin clips. Cell suspensions were incubated in 200 μL of CyStain® DNA 1 Step solution (Sysmex Corporation, Japan) at 4°C. DNA content was measured using a CytoFLEX S Flow Cytometer (Beckman, USA). Ploidy was determined by comparison with WT zebrafish controls.

Metaphase chromosome spreads

Metaphase chromosome preparations from kidney tissue followed established procedures (Zhu & Gui, 2007). For embryonic spreads, 4 hpf embryos were dechorionated and incubated in colchicine for 60 min to arrest cells in metaphase. Following hypotonic treatment with 75 mmol/L KCl, embryos were fixed in methanol:acetic acid solution (3:1). Blastocyst-stage cells were then dispersed, applied to glass slides, air-dried, and stained with Giemsa.

Statistical analysis

All experiments were independently repeated at least three times. For each phenotype, a minimum of three zebrafish were analyzed per group. Comparisons were performed between WT and mutant siblings derived from the same brood. Statistical significance of Sycp1 fragment counts, Rad51 and Rpa2 foci counts, and γH2AX signal intensities was assessed using the two-tailed Mann-Whitney U test in GraphPad Prism v.6 (GraphPad Software, USA). All statistical data are expressed as mean±standard error of the mean (SEM). Data were visualized using standard graphical representations.

RESULTS

Molecular characterization and expression profiles of zebrafish mei4

Zebrafish mei4 (zmei4) (Gene ID: 568897) encoded a 366 amino acid protein (RefSeq: XP_073789139) containing a conserved MEI4 domain (Supplementary Figure S3A). Although overall sequence identity with mammalian Mei4 was low (28.50% with human and 26.32% with mouse orthologs), zMei4 retained six conserved short similarity motifs (SSMs) shared across Mei4 proteins from yeast to mammals (Supplementary Figure S3B) (Kumar et al., 2010). Transcript analysis revealed strong gonadal enrichment, with testis expression exceeding ovary levels by 11.4-fold (Figure 1A), indicating marked sexual dimorphism.

Spermatogenesis is blocked but oogenesis appears normal in zebrafish mei4 mutants

To define Mei4 function in zebrafish, targeted CRISPR/Cas9 genome editing was applied to ablate the conserved Mei4 domain through disruption of exon 3 in mei4 (Figure 1B). Both the mei4^∆2/∆2 and mei4^{+20∆2/+20∆2} mutations introduced premature stop codons, generating truncated proteins retaining only the N-terminal SSM1 and SSM2 motifs. Because both alleles produced indistinguishable phenotypes, subsequent analyses focused on the mei4^∆2/∆2 line. Homozygous mutants of both sexes displayed normal body morphology (Figure 1C), as well as typical secondary sexual characteristics and balanced sex ratios (data not shown). Mutant testes appeared more transparent than WT testes, whereas mutant ovaries exhibited grossly normal morphology.

Histological examination revealed profound defects in spermatogenesis. WT testicular lobules contained the full complement of germ cell stages, including spermatogonia (SPG, singly distributed basal cells with large ovoid nuclei), primary and secondary spermatocytes (SPC Ⅰ and SPC Ⅱ, clustered cells with the second- and third-largest nuclei, respectively), spermatids (SPT, luminal clusters of cells with smallest nuclei), and spermatozoa (SPZ, similar to SPT but devoid of discernible cytoplasm) (Figure 2A). In contrast, SPT and SPZ were completely absent in mei4^−/− testes, and many SPC exhibited abnormally condensed nuclei (Figure 2A). Higher-magnification analysis demonstrated rod-shaped chromatin characteristic of metaphase arrest, with distinct metaphase I and II populations identified based on nuclear size and position relative to SPC I and SPC II (Figure 2B), consistent with previous reports (Feitsma et al., 2007; Ou et al., 2024; Selman et al., 1993). Unlike WT cysts, which contained only occasional metaphase cells, mutant cysts consisted almost entirely of metaphase-arrested germ cells (Figure 2C, right). In contrast to mouse mei4 mutants, which lack SPC Ⅱ (Kumar et al., 2010), zebrafish mei4^−/− testes retained substantial SPC Ⅱ populations, with significantly elevated numbers (858.0±116.9) compared to WT (338.7±55.8). Moreover, apoptosis was markedly increased in mutant testes (Figure 2E, F), predominantly affecting SPC I and SPC Ⅱ, as opposed to SPT and SPZ in WT testes.

Unlike testes, ovarian architecture remained largely intact in mei4^−/− females. Mutant ovaries closely resembled WT controls in both structure and apoptotic index (Figure 2A, D, E), containing abundant oocytes across all developmental stages, including primary-growth (PO), previtellogenic (PVO), vitellogenic (VO), and mature (MO) oocytes. These findings suggest that while mei4 deficiency blocks spermatogenesis, oogenesis proceeds with minimal overt disruption in zebrafish.

Synapsis failure and sexually divergent meiotic responses in mei4^−/− zebrafish

To further analyze gametogenesis in mei4 mutants, SC assembly was assessed by immunolabeling lateral and transverse elements using Sycp3 and Sycp1 antibodies, respectively (Hunter, 2015; Zickler & Kleckner, 2015). In WT spermatocytes and oocytes, Sycp3 initially appeared as short axial segments during leptotene (L) (Figure 3A, B). At zygotene (Z), discrete Sycp1 fragments emerged and colocalized with Sycp3-labeled axes, marking the onset of homolog synapsis. These synapsed regions, characterized by colocalization of Sycp1 and Sycp3 signals, progressively elongated, ultimately forming 25 fully assembled SCs at pachytene (P). Consistent with prior observations (Kochakpour & Moens, 2008), WT pachytene oocytes exhibited longer SCs compared to WT spermatocytes. Total Sycp3 axis length ranged from 155 to 208 μm in males and from 325 to 738 μm in females (Figure 3C). Mean SC length reached 19.44±1.62 μm in pachytene oocytes, representing a 2.71-fold increase relative to pachytene spermatocytes (7.17±0.22 μm). These results indicate pronounced sexual dimorphism in SC architecture, with female SCs displaying both greater length and intercellular variability compared to male SCs.

In mei4 mutants, Sycp3 deposition along chromosome axes proceeded comparably to WT (Figure 3A, B), enabling subdivision of prophase I into leptotene/early zygotene-like (L/EZ-like), mid-to-late zygotene-like (MZ-LZ-like), and pre-pachytene-like-to-pachytene-like (Post-LZ-like) stages based on axial morphology. Despite normal axis formation, Sycp1-labeled stretches were largely absent in both mei4^−/− spermatocytes and oocytes (Figure 3D), indicating a profound defect in homolog synapsis. In contrast to oocytes, approximately 26% of mutant spermatocytes retained sparse Sycp1 fragments, appearing either along single Sycp3 axes, bridging adjacent axes, or as a single filament (Figure 3D–F). Quantitative analysis further revealed sex-specific consequences of synapsis failure. Average Sycp3 axis length per nucleus increased significantly in mei4^−/− females relative to WT controls, whereas axis length in males remained largely unchanged (Figure 3G). These findings indicate that chromosome axes in females, which are intrinsically longer than those in males, are differentially affected by synapsis failure induced by mei4 deficiency, with significant axis elongation observed in mutant females but not in mutant males.

Mei4 knockout impairs meiotic DSB formation in both females and males

To define the contribution of Mei4 to meiotic DSB formation in zebrafish, γH2AX was used as a marker of DSB induction (Rogakou et al., 1999; Rogakou et al., 1998), together with Rad51 and Rpa2 as indicators of homologous recombination repair (Dernburg et al., 1998; Tessé et al., 2017). In WT cells, γH2AX signals initially appeared as polarized domains at the L-EZ stage, subsequently expanded across the nucleus during the MZ-LZ stage, and gradually declined by the P stage (Supplementary Figure S1), consistent with earlier observations (Blokhina et al., 2019; Takemoto et al., 2020). In parallel, abundant Rad51 and Rpa2 foci emerged along or adjacent to Sycp3-defined chromosome axes at the L-EZ stage, followed by a gradual reduction as synapsis advanced and SC assembly was completed (Figure 4; Supplementary Figure S2).

Distinct sex-specific DSB patterns were also evident in WT meiotic germ cells. In males, 51.16% of Rad51 foci were concentrated within the terminal 20% of chromosome length, compared with only 23.91% in females (Figure 4E), indicating preferential distal enrichment in spermatocytes, whereas oocytes displayed a more uniform axial distribution. Pachytene spermatocytes contained an average of 1.54 Rad51 foci per SC, lower than the 1.84 foci per SC observed in pachytene oocytes, and the proportion of SCs harboring more than one Rad51 focus was reduced in males relative to females (39.29% vs. 52.00%) (Figure 4F). However, when normalized to SC length, Rad51 density was substantially higher in pachytene spermatocytes (0.10 foci/μm) than in oocytes (0.035 foci/μm), reflecting the markedly longer SCs in female meiosis.

In mei4 mutants, γH2AX signals were markedly reduced, accompanied by near-complete loss of Rad51 and Rpa2 foci in both spermatocytes and oocytes (Figure 4; Supplementary Figure S1, S2). This global reduction in DSB-associated markers demonstrates that Mei4 is indispensable for efficient meiotic DSB induction in zebrafish of both sexes.

Crossover formation is abolished in mei4^−/− zebrafish

Given the profound impairment of DSB formation in mei4^−/− zebrafish, crossover formation was investigated using MutL homolog 1 (Mlh1) as a cytological marker (Baker et al., 1996; Hunter & Borts, 1997). In WT cells, Mlh1 signals appeared on the Sycp3-labeled chromosome axes at the Z-P stage, with most SCs harboring one, and occasionally two, discrete Mlh1 foci (Figure 5A). Total Mlh1 foci per nucleus ranged from 6 to 27 in WT pachytene spermatocytes and from 9 to 26 in WT pachytene oocytes (Figure 5B). Consistent with sex-specific DSB patterns, crossover sites were more uniformly distributed along SCs in females than in males (Figure 5C). In spermatocytes, 50.52% of Mlh1 foci were concentrated within the terminal 20% of chromosome length, compared to only 9.52% in oocytes.

Figure 5.

Abolished crossover formation in mei4^−/− zebrafish

A: Immunofluorescence staining of chromosome spreads from WT and mei4^−/− spermatocytes and oocytes stained with anti-Sycp3 (green) and anti-Mlh1 (red). B: Quantification of Mlh1 focus numbers per nucleus in WT and mei4^−/− P/Post-LZ-like spermatocyte and oocyte spreads. C: Distribution pattern of crossovers along SCs in WT pachytene spermatocytes (left) and oocytes (right). Bar plots indicate the number of Mlh1 foci within successive 5% intervals of SC length measured from each end toward the center; “0” on the x-axis denotes either SC terminus. A total of 97 Mlh1 foci across 95 male SCs and 105 Mlh1 foci across 93 female SCs were detected. n denotes number of spermatocytes and oocytes analyzed. Scale bars: 5 μm in A and 1 μm in enlarged views. Data are presented as mean±SEM. Statistical significance was assessed using unpaired two-tailed Student’s t-test. ^****: P<0.0001.

In contrast, Mlh1 foci were completely absent in both spermatocytes and oocytes of mei4^−/− zebrafish (Figure 5A, B), demonstrating complete loss of crossover formation following mei4 disruption. The absence of crossovers provides a direct mechanistic basis for homolog mis-segregation during meiosis Ⅰ.

mei4^−/− females produce abundant aneuploid eggs and rare unreduced eggs

Fertility assays revealed a striking sexual asymmetry in reproductive outcome. Consistent with histological observations (Figure 2A), mei4^−/− males were able to induce spawning in WT females but failed to achieve fertilization (Figure 6A), confirming complete male sterility. In contrast, mei4^−/− females produced morphologically normal oocytes that could be fertilized by WT sperm. However, embryonic development was severely compromised: most embryos exhibited pronounced malformations and high mortality within the first few days of embryogenesis, with progressive lethality during development. Notably, approximately 7.2% of progeny derived from crosses between mei4^−/− females and WT males survived to adulthood and developed without overt abnormalities (Figure 6A).

Figure 6.

Abnormal progeny derived from mei4^−/− female zebrafish

A: Fertilization and survival rates of progeny from different breeding combinations. Each group included more than eight individuals. B: DNA content histograms of embryonic cells from WT ♀×WT ♂ and mei4^−/− ♀×WT ♂ crosses. Each group contained 30 pooled embryos. C: Representative metaphase chromosome spreads from blastocyst-stage cells. WT embryos exhibited the normal somatic chromosome number of 50, whereas embryos derived from mei4^−/− females displayed variable chromosome counts. N denotes number of chromosomes per cell. D: Quantification of metaphase chromosome numbers in blastocyst-stage cells. Each group contained more than 50 cells. Scale bars: 10 μm in C. Data are presented as mean±SEM. Statistical significance was assessed using unpaired two-tailed Student’s t-test. ^****: P<0.0001; ns: Not significant.

To determine the underlying cytogenetic basis, DNA content was measured in 30 WT ♀×WT ♂ and mei4^−/− ♀×WT ♂ embryos (Figure 6B). WT embryos displayed two discrete peaks corresponding to uniform ploidy cells in the normal or replicated (2C or 4C) state. In contrast, embryos derived from mei4^−/− ♀×WT ♂ crosses exhibited a broad and irregular distribution of DNA content, indicative of heterogeneous aneuploidy. Karyotype analysis of blastocyst-stage cells further substantiated these findings (Figure 6C). WT embryos predominantly contained the expected diploid complement of 50 chromosomes, whereas embryos from mei4^−/− females crossed with WT males exhibited extensive numerical variation, ranging from 28 to 75 chromosomes (Figure 6D). Notably, among the 61 analyzed blastocysts, 59 exhibited chromosome numbers between 28 and 72, while only two contained exactly 75 chromosomes. These data demonstrate that mei4 deficiency disrupts faithful chromosome segregation during female meiosis, leading to the production of abundant aneuploid eggs and, at low frequency, unreduced gametes capable of generating triploid progeny.

Viable triploid progeny derived from mei4^−/− females crossed with WT males

Given the capacity of mei4^−/− females to produce unreduced eggs, surviving offspring from crosses with WT males were predicted to be triploids. This hypothesis was evaluated by assessing DNA content in peripheral blood cells and by metaphase chromosome analysis of kidney-derived cells from adult survivors. Flow cytometric profiles revealed that all surviving individuals exhibited a 1.5-fold increase in DNA content relative to WT controls (Figure 7A). Consistently, karyotype analysis confirmed the presence of 75 chromosomes in every survivor, establishing triploidy (Figure 7B). These triploid adults exhibited a normal sex ratio (Figure 7C). Collectively, these findings demonstrate that mei4^−/− females can produce viable triploid progeny when crossed with WT males.

Figure 7.

Viable triploid progeny derived from mei4^−/− females

A: DNA content histograms of peripheral blood cells from WT ♂, mei4^−/− ♀, and healthy offspring generated from mei4^−/− ♀×WT ♂ crosses. B: Representative metaphase chromosome spreads from kidney-derived cells. WT cells exhibited a normal somatic chromosome number of 50, whereas cells from triploid females contained 75 chromosomes. N denotes number of chromosomes per chromosome set. C: Sex ratio of healthy offspring derived from WT ♀×WT ♂ and mei4^−/− ♀×WT ♂ crosses. Scale bars: 10 μm in B. Data are presented as mean±SEM. Statistical significance was assessed using unpaired two-tailed Student’s t-test. Ns: Not significant.

DISCUSSION

Meiosis is an evolutionarily conserved process that produces haploid gametes and depends critically on proper DSB formation to ensure accurate homolog segregation (Subramanian & Hochwagen, 2014; Thacker et al., 2014; Zickler & Kleckner, 2015). Rare failures in this process can generate unreduced gametes and represent a major route to polyploidization. In the present study, Mei4 was examined as a core regulator of meiotic recombination in zebrafish. Results established functional conservation of Mei4 in vertebrate DSB induction and revealed pronounced sex-specific behaviors following recombination failure. Notably, although most oocytes produced by mei4-deficient females were aneuploid, a small proportion of unreduced eggs were generated, permitting the generation of viable triploid progeny after mating with WT males.

Loss of mei4 in zebrafish abolished DSB formation, disrupted synapsis and crossover establishment, and resulted in complete spermatogenic arrest with male sterility (Figures 2–5; Supplementary Figures S1, S2), consistent with phenotypes reported for mei4 deficiency in yeast and mouse models (Kumar et al., 2010; Kumar et al., 2015; Menees et al., 1992). In contrast to the ovarian arrest observed in mei4- and spo11-mutant mice (Baudat et al., 2000; Blokhina et al., 2019; Kumar et al., 2010), mei4^−/− and spo11^−/− zebrafish females retained the capacity to complete oogenesis but exhibited impaired fertility. Notably, mutant females produced mature oocytes that underwent extensive chromosome mis-segregation, yielding predominantly aneuploid embryos and, at low frequency, triploid progeny (Figures 6, 7). These findings suggest that male and female zebrafish respond differently to identical meiotic disturbances. Comparable sexual dimorphism has also been documented in mice, with spermatogenesis exhibiting more stringent quality control mechanisms that lead to developmental arrest or delay following meiotic perturbation (Hunt & Hassold, 2002).

Nevertheless, this relaxed tolerance during oogenesis occurred at the expense of gamete fidelity, as evidenced by elevated aneuploidy (Figure 6). In mice, aneuploid gametes arise primarily from chromosome mis-segregation associated with crossover failure, premature loss of cohesion, and impaired spindle assembly checkpoint (SAC) activity (Lane & Kauppi, 2019). Multiple studies have shown that SAC is less stringent in oocytes than in spermatocytes (Lane & Kauppi, 2019; Santaguida & Amon, 2015). In budding yeast, SAC activation delays anaphase onset to facilitate correction of chromosome alignment errors during meiosis (Barisic et al., 2021). In contrast, Caenorhabditis elegans exhibits an inverted pattern, with meiotic checkpoints robustly inducing apoptosis in oocytes while remaining inactive in spermatocytes (Bhalla, 2010). Together, these observations highlight pronounced sex-specific divergence in checkpoint regulation and signaling, permitting persistence of certain gametocytes despite meiotic disruption, independent of the conserved core machinery governing DSB formation and recombination.

As previously reported in mlh1 and spo11 mutants (Blokhina et al., 2019; Feitsma et al., 2007; Zhang et al., 2020), the present study demonstrated that loss of mei4 led to the production of predominantly aneuploid oocytes together with a rare subset of unreduced (diploid) oocytes capable of generating triploid progeny. In the absence of crossovers, homologous chromosomes failed to establish chiasmata, eliminating the physical connections required for stable bivalent alignment at the metaphase I plate and subsequent accurate segregation during anaphase I. This breakdown in pairing and orientation drove pervasive chromosome mis-segregation and consequently yielded high frequencies of aneuploid gametes. The most extreme outcome of this defective asymmetric division was complete failure of polar body extrusion, which directly produced unreduced (diploid) oocytes. Although these relatively rare unreduced oocytes likely arose through stochastic consequences of profound chromosome mis-segregation due to crossover deficiency, contribution from additional regulatory mechanisms governing asymmetric division cannot be excluded.

However, mei4-deficient females generated viable triploid offspring with a normal adult sex ratio, whereas progeny derived from spo11^−/− females crossed with WT males, as well as triploids produced by temperature shock, were predominantly male (Blokhina et al., 2019; Delomas & Dabrowski, 2018; Zhang et al., 2020). Sex determination in zebrafish is typically influenced by multiple factors, including regulation of sex-related gene networks, primordial germ cell (PGC) abundance, and masculinization secondary to oogenic arrest. In contrast to WT triploids, which typically exhibit arrested oogenesis, triploid progeny derived from mei4^−/− females retained normal PGC numbers (data not shown) and supported ovarian differentiation in females, consistent with balanced sex ratios. This divergence suggests that triploids generated from WT backgrounds, carrying one functional and two truncated mei4 copies, may provide sufficient gene dosage to disrupt oogenesis through aberrant crossover-associated recombination. Although Mei4 functions primarily to recruit Spo11 for DSB initiation, Spo11 may exert additional roles during oogenesis. Perturbation of these pathways could drive abnormal gametogenesis, resulting in either masculinization or differential expression of sex-determining genes.

Similar to that observed in mei4 mutants, defects in DSB formation, synapsis, recombination, and crossover during oogenesis also characterize unisexual systems (Newton et al., 2016). In C. gibelio, extensive genetic variation is enriched in meiosis-associated genes, which may facilitate production of unreduced eggs through inhibition of meiosis I, thereby promoting reproductive persistence in this unisexual polyploid lineage (Wang et al., 2022). Together, these observations indicate that disruption of meiotic programs—particularly pathways governing chromosome segregation via DSB-mediated crossovers—represents an early and critical step in neopolyploid establishment (Lu et al., 2022; Lu et al., 2023a; Lu et al., 2023b). Consistent with this model, cyclin N-terminal domain containing 1 (cntd1) mutant zebrafish have been shown to generate predominantly aneuploid progeny together with rare diploid and triploid offspring, with unreduced gametes arising by chance (Ou et al., 2024). Collectively, these findings provide direct evidence that failure of DSB-dependent crossover formation constitutes a central driver of polyploidy occurrence.

Elucidating the underlying mechanisms may enable development of strategies to reliably generate unreduced oocytes, thereby advancing triploid production and unisexual breeding approaches (Lu et al., 2023b). In plants, identification of core genes critical for polyploidy has led to the implementation of the Mitosis-to-Meiosis (MiMe) system, in which meiosis is replaced by mitosis, bypassing genetic segregation and yielding unreduced gametes (D'erfurth et al., 2009; Khanday et al., 2019; Wang et al., 2024). These genes broadly participate in gametogenic control, including recombination, cell-cycle regulation, spindle formation, and chromatid segregation (Chelysheva et al., 2005; D'erfurth et al., 2010; D'erfurth et al., 2008; De Storme & Geelen, 2011; Mercier et al., 2001; Mieulet et al., 2016). Given the comparatively high natural incidence of polyploidy in fish, targeted modulation of meiotic pathways to induce unreduced gametes represents a promising framework for unisexual breeding in vertebrates (Lu et al., 2024).

In summary, these results uncover a conserved role for Mei4 in zebrafish meiosis and reveal pronounced sexual dimorphism in meiotic responses to DSB disruption. Further investigation into sex-specific regulatory architectures in gametogenesis will advance understanding of meiotic control and illuminate fundamental mechanisms underlying polyploid formation.

CONCLUSIONS

This study establishes a conserved requirement for Mei4 in vertebrate meiosis and delineates a mechanistic connection between meiotic DSB failure and polyploid formation. Notably, disruption of mei4 in zebrafish caused profound meiotic abnormalities, including impaired synapsis, reduced DSB formation, and compromised homologous recombination repair and crossover formation in both male and female germ cells. Despite these shared molecular defects, meiotic outcomes diverged sharply between the sexes. While mei4^−/− males exhibited complete sterility associated with spermatogenic arrest, females produced abundant aneuploid oocytes and a small proportion of unreduced eggs due to chromosome mis-segregation. Strikingly, when mei4^−/− females were crossed with WT males, a considerable number of viable triploid offspring were obtained. Collectively, these findings demonstrate the conserved function of Mei4 in DSB formation across vertebrates, provide new insights into sex-specific meiotic regulation, and highlight the critical role of meiotic DSB-based crossover defects in polyploidy occurrence.

SUPPLEMENTARY DATA

Supplementary data to this article can be found online.

Funding Statement

This work was supported by the National Natural Science Foundation of China (32273125, 32473149), Science and Technology Major Project in Hubei Province (2023BBA0010), and Natural Science Foundation of Wuhan (2024BBSA02)