SPIDR is required for homologous recombination during mammalian meiosis

Abstract

Meiotic recombinases RAD51 and DMC1 mediate strand exchange in the repair of DNA double-strand breaks (DSBs) by homologous recombination. This is a landmark event of meiosis that ensures genetic diversity in sexually reproducing organisms. However, the regulatory mechanism of DMC1/RAD51-ssDNA nucleoprotein filaments during homologous recombination in mammals has remained largely elusive. Here, we show that SPIDR (scaffold protein involved in DNA repair) regulates the assembly or stability of RAD51/DMC1 on ssDNA. Knockout of Spidr in male mice causes complete meiotic arrest, accompanied by defects in synapsis and crossover formation, which leads to male infertility. In females, loss of Spidr leads to subfertility; some Spidr^−/− oocytes are able to complete meiosis. Notably, fertility is rescued partially by ablation of the DNA damage checkpoint kinase CHK2 in Spidr^−/− females but not in males. Thus, our study identifies SPIDR as an essential meiotic recombination factor in homologous recombination in mammals.

INTRODUCTION

Meiosis is the process of cell division of diploid progenitors into haploid gametes in sexually reproducing organisms. Erroneous meiosis generates aberrant gamete which can cause fertility defects, or even infertility (1). During meiosis, homologous chromosomes pair, synapse and undergo homologous recombination (HR) in order to ensure proper segregation (2). Recombination is initiated by programmed DNA double-strand breaks (DSBs) which are generated by SPO11 (3). SPO11 is subsequently removed together with an oligonucleotide chain under the action of endonuclease MRE11, after which DSB ends are further resected by the exonuclease to generate single strand 3’ overhangs (4,5). This single stranded DNA (ssDNA) is protected by replication protein homologs (RPAs). Subsequently, the RecA-like proteins RAD51 and its meiosis-specific homolog DMC1 are recruited by BRCA2, which will replace RPA and mediate strand-invasion and promote the formation of displacement loop (D-loop) (6,7). After D-loop formation, meiotic DSBs are repaired by the synthesis-dependent strand annealing (SDSA) pathway or by the double Holiday-Junction (dHJ) pathway (8,9). In the dHJ pathway, the extended D-loop is captured by another terminal single chain of DSB, and the new chain is synthesized with a homologous chromosome as a template. Finally, it forms an intermediate called dHJ, a CO-specific form of recombination intermediate specifically resolved to COs by the MLH1–MLH3 complex (10,11).

The meiotic recombinases RAD51 and DMC1 mediate homologous recombination by promoting the nucleoprotein filament to search and invade its homologous chromosome. The mechanisms regulating DMC1/RAD51-ssDNA nucleoprotein filaments remain to be clarified. SPIDR is considered as a highly conserved protein with no identified motifs (Supplementary Figure S1E), and was initially proposed to function as a scaffolding protein. Accordingly, this protein would promote the assembly of subnuclear RAD51 foci and regulate the ratio of crossover-to-non-crossover recombinants via BLM helicase, in order to promote HR efficiency (12). SPIDR was shown to function as an efficient HR factor acting through several HR-related molecules such as FIGNL1 (fidgetin-like 1), SWSAP1, and SWS1 in somatic cells (13–16). The SWS1–SWSAP1 complex is a recently identified complex that functions in promoting the stable assembly of both RAD51 and DMC1 nucleoprotein filaments during meiosis (17). SPIDR is a subunit of the SWS1–SWSAP1–SPIDR complex regulating inter-homolog homologous recombination (IH-HR) (18). However, its role in homologous recombination during meiosis is not well understood.

In this study, we knocked out Spidr in mice, seeking to characterize its function(s) in meiotic recombination. We found that during meiosis SPIDR recruits RAD51 and DMC1 for the formation or stabilization of nucleoprotein filaments. Deletion of Spidr completely blocked the progression of meiosis, and this was accompanied by defects in DSB repair and synapsis, which rendered failure in male spermatogenesis. Primary ovarian insufficiency (POI) is a subclass of ovarian dysfunction in which an unknown mechanism leads to premature exhaustion of the primordial follicle pool (PFP) (19). We found that Spidr-deficient females showed defects of meiosis and decrease in fertility similar to human POI—the phenotype resulting from severe exhaustion of PFP (19–23). Remarkably, Spidr-deficient fertility could be restored by the ablation of the DNA damage checkpoint kinase CHK2 in females, but not in males.

MATERIALS AND METHODS

Generation of spidr and chk2 knockout mice

The mouse Spidr gene is located on chromosome 16. Spidr knockout mice in a B6D2F2 genetic background were constructed by deleting genomic DNA fragments covering exon 2 to exon 3 using CRISPR/Cas9 system entrusted to the Institute of Zoology, Chinese Academy of Sciences.

Chk2 knockout mice in a C57BL/6JGpt genetic background were constructed by deleting genomic DNA fragments covering exon 3 to exon 6 using CRISPR/Cas9 system. The founders were genotyped by polymerase chain reaction (PCR), followed by DNA sequencing analysis.

Mouse care

The care and use of mice in this study were carried out according to the protocol approved by the animal management center of Shandong University. Mice were raised in the animal management center of Shandong University in accordance with the guidelines of the animal ethics committee and the Animal Care and Use Committee of Shandong University. All experimental protocols were approved by the Ethics Committee of Shandong University.

Genotyping

Genotyping was performed by PCR amplification of genomic DNA extracted from mouse tails. The PCR primers for Spidr-deficient genotyping are as follows: the mutant alleles were 5′- GGA ACC AGA GGC CAG ATA GCC CAG AGA C-3′ and 5′-CCA CTT GGA GTT GAG CTT TGT GCA AGG TGA CAA AT-3′, yielding a 653–base pair (bp) fragment; the wild-type alleles were 5′-GGA ACC AGA GGC CAG ATA GCC CAG AGA C-3′ and 5′-CCA CTT GGA GTT GAG CTT TGT GCA AGG TGA CAA AT-3′, yielding a 657–base pair (bp) fragment.

The PCR primers for Chk2-deficient genotyping are as follows: the mutant alleles were 5′- GGA CTT TTG GGA TGG CAT TTG AA-3′ and 5′-GCA TAG CAC TTA GGT CAG GAT TAA ACC-3′, yielding a 287–base pair (bp) fragment; the wild-type alleles were 5′-GCC CTT GAT TGT CTT CTT ACT GCT GT-3′ and 5′-TCC GAG GAG AGA AGC AAC CGTTA-3′, yielding a 316-base pair (bp) fragment.

Tissue collection and histological analysis

Immediately after euthanasia, the testes and ovaries were dissected and fixed in 4% paraformaldehyde (p1110, Solarbio, Beijing, China) for 24 h. After dehydration, they were embedded in paraffin. Prepare sections (5 μm) on glass slides. After dewaxing and hydration, the slides were stained with hematoxylin and eosin, then dehydrated and mounted for histological analysis.

Fertility test

Males: Each male mouse (8–12 weeks, n = 3) was caged with two wild-type C57BL/6 females (4–6 weeks), that were checked for vaginal plugs every morning. Once a vaginal plug was identified (day 1 postcoitus), the male was allowed to rest for 2 days, after which another female was placed in the cage for another round of mating. The plugged female was separated and singly caged, and the pregnancy was recorded. If a female did not generate any pups by day 22 postcoitus, it was deemed as not pregnant and euthanized to confirm result. The fertility test lasted for 2 months.

Females: Every three female mice (5 weeks) were caged with one wild-type C57BL/6 male mouse having normal fertility. The number of pups was recorded for each pregnancy. The cumulative pups of each female were calculated. The fertility test lasted for 6 months.

Immunocytology and antibodies

Spermatocyte and oocyte chromosome spreading was performed as previously described (24). Produced antibody used for western blotting: Mouse anti-SPIDR polyclonal antibody was prepared against mouse SPIDR protein by Dai-an Biological Technology Incorporation (Wuhan, China). Primary antibodies used for immunofluorescence were as following: rabbit anti-SYCP1 (1:500 dilution; Abcam #ab15090), mouse anti-SYCP3 (1:500 dilution; Abcam #ab97672), rabbit anti-RPA2 (1:200 dilution; Abcam #ab76420), rabbit anti-RAD51 (1:200 dilution; Thermo Fisher Scientific #PA5-27195), rabbit anti-DMC1 (1:200 dilution; homemade); mouse anti-γH2AX (pSer139) (1:1000 dilution; Millipore #05–636), mouse anti-MLH1 (1:100 dilution; BD Biosciences #550838), rabbit anti-BLM (1:200 dilution; Invitrogen #PA5-27384), rabbit anti-MSH4(1:200 dilution; Abcam #ab58666); rat anti-HEI10 (1:100 dilution; homemade); Rabbit anti-MSY2(1:200 dilution; Abcam #ab33164). Primary antibodies were detected with Alexa Fluor 488- or 594-conjugated secondary antibodies (1:500 dilution; Abcam #ab150080, #ab150077, #ab150113 and #ab150116) for 1 h at room temperature. The slides were mounted using Mounting Medium with DAPI (Abcam #ab104139).

Western blotting

Testes were separated from the mice, washed with PBS. According to the invent Kit (Invent #SD-001/SN-002), total protein was extracted. After denaturation, equal amounts of protein were electrophoresed on 10% SDS polyacrylamide gels, and the bands were transferred to polyvinylidene fluoride membranes (Millipore, USA). Immunoreactive bands were detected and analyzed with a Bio-Rad ChemiDoc MP imaging System and Image Lab Software (Bio-Rad, USA). The primary antibodies for immunoblotting included anti-Actin (1:5000 dilution; Cell Signaling Technology, #66009-1-Ig) and anti-SPIDR (1:1000 dilution; produced as described above).

RT-PCR

Total RNA was extracted from the testicular sorting cells of wild-type mice. cDNA was synthesized by PrimeScript RT reagent Kit with gDNA Eraser (Takara). Qualitative PCR was performed using TB Green Premix Ex Taq (Takara) and specific forward and reverse primers as follows: Actin primer pair, 5′-GGC TGT ATT CCC CTC CAT CG-3′ (forward) and 5′-CCA GTT GGT AAC AAT GCC ATG T-3′ (reverse). Spidr primer pair, 5′-CA G CAG TTG AGA CGG GGA A-3′ and 5’-ATC CTT CTC CAC ACT TGA GCC-3′ (forward). All reverse transcription PCR reactions were performed with an initial denaturation at 95°C for 10 min followed by 25 cycles of denaturation at 95°C for 30 s, annealing at 60°C for 30 s, extension at 72°C for 30 s, and a final extension at 72°C for 5 min.

Plasmids and CO-IP

Full-length cDNAs encoding Spidr, Rad51 and Dmc1 were amplified by RT-PCR from murine testis RNA. Full-length cDNAs were cloned into the pcDNA3.1 mammalian expression vectors. HEK 293T was maintained in DMEM supplemented with 10% FBS and 1% penicillin and streptomycin in a 37°C incubator with 5% CO2. Cell transfection was performed using lipofectamine 3000 (Invitrogen), following the manufacturer's protocol. The same amount of antibody and IgG were added into the cell extract and the mixture was incubated at 4°C overnight. After incubation, immunocomplexes were isolated by adsorption to protein A/G Sepharose beads for 2 h. The beads were added loading buffer in proportion, boiled at 95°C, and loaded onto 10% Tris-Glycine Mini Gels (Invitrogen).

Yeast two hybrid assay

Mouse Spidr cDNA was subcloned into pGBKT7 vector as bait. Mouse Rad51, Dmc1 cDNA were separately subcloned into pGADT7 vector as prey. The bait and prey plasmids were co-transformed into a Y2H gold strain and the interaction was tested on DDO/TDO/3AT/QDO medium.

Oocyte in vitro maturation and in situ chromosome counting

Oocytes were isolated from female mice at the age of 3–4 weeks. To get fully grown GV-stage oocytes, mice were super-stimulated with 5 IU pregnant mare's serum gonadotropin (PMSG) injection. After 48 h of PMSG injection, cumulus oocyte complexes were obtained by manually rupturing the ovarian follicle structure. The oocytes were cultured in the small drops of M16 (M7292; Sigma-Aldrich) and maintained in 5% CO₂ at 37°C for 8 and 16 h until they arrived to metaphase I and metaphase II. Oocytes were exposed to Tyrode's buffer (pH 2.5) for 30 s at 37°C to remove zona pellucida and fixed in a drop of 1% paraformaldehyde with 0.15% Triton X‐100 on a glass slide. After air drying, oocytes were stained with the Topoisomerase II antibody (1:400 dilution; Abcam #ab109524) to detect chromosomes.

Immunohistochemistry

The ovaries of E13.5, E15.5, PD1, PD8 and PD15 were immunohistochemical. Immunohistochemistry and quantification of ovarian follicles were performed as previously (24).

Imaging

Chromosome spreads were imaged by confocal microscopy (Andor Dragonfly spinning disc confocal microscope driven by Fusion Software) and Imaris software was used to process the images. Histological samples were taken by fluorescence microscope (bx52, Olympus), Photoshop (Adobe), and ImageJ (v.1.52a) were used for image processing to count foci and quantify fluorescence.

Statistical analysis

Data were analyzed by GraphPad prism 8.0.1. Data difference between the two groups was tested by unpaired, two-tailed Student t-test. The error bars stand for means ± SD. The sample size has been described in the Figure legend. The significance threshold was set at P value <0.05.

RESULTS

The spidr knockout mice phenotype is sexually dimorphic

To investigate the role of Spidr in mice, we disrupted Spidr in B6D2F2 mice by deleting exons 2–3 using CRISPR-Cas9 genome editing; this was performed in order to generate Spidr-deficient (Spidr^−/−) mice (Figure 1A). Heterozygous Spidr^+/− mice were mated and gave birth to Spidr^−/− offspring. Spidr^−/− mice were born according to Mendelian proportions and had normal body weight (Supplementary Figure S2A and B). Immunoblotting using total protein extracts from the testes of both wild type (WT) and Spidr^−/− males confirmed that SPIDR was successfully disrupted (Figure 1B).

Figure 1.

SPIDR is required for both male and female fertility. (A) Schematic representation of the CRISPR-Cas9 editing genome editing system to the Spidr-deficient mice. (B) Western blot of SPIDR in testes of WT and Spidr^−/− mice at PD42. β-Actin was used as the loading control. PD, postnatal day. (C) Morphology of testes at PD42 in WT and Spidr^−/− male mice. Spidr^−/− testes were obviously smaller than WT at PD42. (D) Testis to body weight ratios of adult mice. Spidr^−/− testes had significantly lower testis weight to body weight ratio compared to WT. Error bars, mean ± s.d. Student's t-test, two-tailed. (E) Spidr^−/− male mice between 8–12 weeks were sterile with no offspring. n = 3 for each genotype. Error bar, mean ± s.d. (F) H&E staining of adult testis sections. The complete arrest of spermatogenesis in Spidr^−/− mice is indicated by lack of spermatids or mature spermatozoa in the testis or epididymis. (G) Cumulative numbers of pups per female during the defined period. n = 3 mice per genotype. (H) Histological analysis of ovaries from WT and Spidr^−/− females. Ovaries in 2-month-old Spidr^−/− mice have fewer follicles than WT. At around 4 months of age, mutant ovaries exhibited a near-complete loss of developing follicles. At 6 months, Spidr^−/− ovaries lose normal morphology.

We found that the Spidr^−/− mice phenotype is sexually dimorphic since male knockout mice were sterile but females displayed subfertility. 6-week-old Spidr^−/− males had much smaller testes than those of WT mice (Figure 1C). Adult testis weights from Spidr^−/− mice were 3- to 4-fold lighter than that in WT mice (Figure 1D). Spidr^−/− male mice were completely infertility (Figure 1E). Histological analysis of Spidr^−/− testes showed impaired spermatogenesis in which no post-meiotic round or elongating spermatids appeared in the seminiferous tubules (Figure 1F, upper panel). No spermatozoa were found in the cauda epididymis of Spidr^−/− males (Figure 1F, lower panel). This phenotype was similar with SWS1 and SWSAP1 null mutants (17).

Spidr^−/− females also showed fertility defects. The number of pups from the WT females increased incrementally as expected, while fertility tests in female Spidr^−/− mice exhibited significantly lower fertility than WT mice, and became completely sterile after a 3-month period (Figure 1G). Morphological studies showed that ovaries of 2-month-old Spidr^−/− mice contained fewer oocytes than WT mice ovaries. At around 4 months, the Spidr^−/− mice ovaries exhibited a near-complete loss of developing oocytes. At 6 months, Spidr^−/− mice ovaries displayed complete loss of normal morphology and were fibrotic (Figure 1H). Collectively, these results show that SPIDR is required for fertility in both males and females.

SPIDR is highly expressed in tissues and meiotic germ cells

A significant reduction of post-meiotic germ cells in testes from Spidr^−/− mice resembled that of other early HR-defective mice, such as Dmc1, Meilb2, and Brme1 knockout mice (25–27). In order to study the role of SPIDR in homologous recombination during meiosis, we first analyzed SPIDR expression and localization. SPIDR was expressed in various mouse tissues, with especially high expression in adult testis, where meiotic prophase I occur (Supplementary Figure S1A). We therefore focused on the function of SPIDR in testis to more closely examine its expression patterns during spermatogenesis. Immunoblotting of total protein extracts from WT testes at increasing ages revealed that the SPIDR protein level started to increase at postnatal day 14 (PD14) (Supplementary Figure S1B). PD14 is a developmental stage when a great number of spermatocytes enter zygotene stage (28) and this thus suggests that SPIDR has an important function in the zygotene stage. These findings were consistent with the analysis of average SPIDR protein abundance in isolated germ cells based on data from previous studies (Supplementary Figure S1C) (29). Further, qPCR analysis of isolated germ cells verified that Spidr was highly transcribed in the zygotene stages (Supplementary Figure S1D). Taken together, these results suggest that SPIDR may function in homologous recombination during spermatogenesis.

Spidr-deficient spermatocytes show pachytene arrest with defects of synapsis and DSB repair

Considering that SPIDR was highly expressed in zygotene spermatocytes, and knock out of Spidr resulted in mice infertility, we performed fluorescence-activated cell sorting in order to explore the exact defects during Spidr^−/− spermatogenesis. We found that Spidr^−/− mice lack spermatocytes beyond the pachytene stage (Supplementary Figure S2C). We next investigated whether SPIDR is required for homologous recombination in meiosis. We first examined chromosomes during meiosis prophase I by immunostaining of the synaptonemal complex central element (SYCP1) and axial/lateral element (SYCP3); these are components of the synaptonemal complex, a tripartite proteinaceous structure that forms between the homologs (30). No diplotene spermatocytes were detectable in Spidr^−/− testes, indicating the pachytene stage arrest (Figure 2A). We further quantified the percentage of spermatocytes at different stages of meiosis prophase I in WT and Spidr^−/− testes. We found that spermatocytes failed to proceed beyond the early-pachytene stage and only 4.8% of spermatocytes reached early-pachytene stage in Spidr^−/− testes. This was in sharp contrast to that 23.08% of spermatocytes had reached the early-pachytene stage at PD42 in WT testes (Figure 2B).

Figure 2.

SPIDR is indispensable for meiotic progression. (A) Spermatocyte chromosome spreads from WT and Spidr^−/− testes at PD45. Immunofluorescent signals for SYCP1 (green) and SYCP3 (red) indicate pachytene arrest in spermatocytes of Spidr^−/− mice. Scale bars, 5 μm. (B) Proportions of WT and Spidr^−/− spermatocytes in different meiotic stages, determined by calculating the number of spermatocytes at each stage divided by the total number of spermatocytes counted in adult testes. P values were calculated by Student's t-test. Error bars, mean ± s.d. (C) Representative images of spermatocyte chromosome spreads from WT and Spidr^−/− testes at PD38 showing leptotene, zygotene, pachytene, and diplotene stages. Immunofluorescent signals for SYCP3 (red) and γH2AX (green) show that γH2AX persists around autosomes at pachytene in Spidr^−/− spermatocytes, indicating incomplete DSB repair. Scale bars, 5 μm. (D, E) Representative images of WT and Spidr^−/− spermatocytes immunostained for SYCP3 (red) and MLH1 crossover markers (green) (D) and quantification of foci for image (E). No MLH1 signal was detected in Spidr^−/− spermatocytes compared to WT. Scale bars, 5μm. Error bars, mean ± s.d. (F, G) Representative images of WT and Spidr^−/− spermatocytes immunostained for SYCP3 (red) and MSH4 (green) at the early-zygotene, late-zygotene, and pachytene stages are shown in (F). MSH4 foci counts were reduced in Spidr^−/− spermatocytes. Quantification of foci for images in (G). P values were calculated by Student's t-test. Error bars, mean ± s.d. Scale bars, 5 μm.

Additionally, abnormal synapsis was observed in Spidr^−/− zygotene spermatocytes (Supplementary Figure S4A). It is observed as different regions of one chromosome synapsed with multiple partners. This results in frequent synapsis between non-homologous chromosomes, which has been reported in several early HR-defective mice (26,27,31). In Spidr^−/− spermatocytes, 49.7% of zygotene cells showed abnormal synapses of heterologous chromosomes (Supplementary Figure S4B). Possibly, these zygotene spermatocytes with severe abnormal synapses were unable to enter the pachytene stage, which is supported by the high proportion of zygotene spermatocytes in the Spidr^−/− testes (Figure 2B).

The completion of synapsis at pachytene stage is always accompanied by the completion of DSB repair on autosomes. Considering the defects of synapsis in Spidr^−/− testes, we speculated that DSB repair would be impaired as well. The strength of the γH2AX signal can represent the degree of DSB damage (32). It appears in the leptotene stage upon DSB formation (32), and disappears from autosomes till the pachytene stage, but remains concentrated in the XY body from pachytene to diplotene stage (33). We therefore evaluated meiotic DSB repair by immunostaining γH2AX in spermatocytes and found that γH2AX signals were comparable between WT and Spidr^−/− spermatocytes at the leptotene stage, suggesting that DSB formation was unaffected in Spidr^−/− spermatocytes. However, γH2AX signals were still maintained on autosomes at pachytene-like stage in Spidr^−/− spermatocytes, while γH2AX signals were only concentrated in the XY body at pachytene stage in WT spermatocytes (Figure 2C) (Supplementary Figure S3A). Although it seems that chromosomes synapsed normally in Spidr^−/− early pachytene stage spermatocytes, a considerable number of meiotic DSBs remained unrepaired in these Spidr^−/− spermatocytes. Further, terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assays revealed an increase in the number of apoptotic cells in the Spidr^−/− testes compared to that of the WT (Supplementary Figure S2D). This indicates that the repair of DBS disruption could trigger the pachytene checkpoint, subsequently followed by the apoptosis of defective spermatocytes (34). SPIDR is therefore essential for the progression of meiotic recombination.

Crossover complexes and HR intermediates are defective in spidr-deficient spermatocytes

A small proportion of Spidr^−/− spermatocytes still entered the pachytene stage, which led us to ask whether these spermatocytes can complete homologous recombination and form crossovers, which are marked by MLH1/HEII10 at middle to late pachytene stage (35–37). Approximately 20–30 MLH1/HEI10 foci per cell could be detected in WT pachytene spermatocytes, whereas no signal was observed in Spidr^−/− pachytene spermatocytes (Figure 2D, E and Supplementary Figure S4C, D). Thus, crossover formation is completely defective in Spidr^−/− spermatocytes.

It has been reported that the recombination factor MSH4 stabilizes DNA-strand exchange intermediates during homologous recombination, some of which will become crossovers mostly marked by MLH1 (11,38). We next immunostained MSH4 to investigate the effects of SPIDR deletion on the HR intermediates during meiosis prophase I. We detected reductions in the number of MSH4 foci in Spidr^−/− spermatocytes compared to the WT from the early-zygotene stage to the pachytene stage: early-zygotene (26.667 ± 14.129 in Spidr^−/− versus 54.952 ± 21.306 in WT), late-zygotene (55.442 ± 17.602 in Spidr^−/− versus 99.476 ± 13.941 in WT), and pachytene (57 ± 6.751 in Spidr^−/− versus 89.125 ± 4.390 in WT) (Figure 2F, G). Taken together, these results suggest that SPIDR is required for HR intermediates stabilization and crossover formation during meiosis homologous recombination.

SPIDR is indispensable for the recruitment of recombinase RAD51 and DMC1

To support meiotic recombination and synapsis progression, recombinase RAD51 and its meiosis-specific paralog DMC1 are recruited and assembled to form ssDNA filaments that facilitate scanning for homologous pairing regions (6,39,40). RAD51 and DMC1 foci were counted to assess the effects of Spidr-deficient on early homologous recombination. We found that Spidr^−/− spermatocytes exhibited a reduction in RAD51 foci at leptotene stage (28 ± 7.892 in Spidr^−/− versus 150.6 ± 42.88 in WT), zygotene stage (67.82 ± 27.84 in Spidr^−/− versus 184.7 ± 55.67 in WT), and pachytene stage (8.857 ± 6.388 in Spidr^−/− versus 34.30 ± 11.04 in WT) (Figure 3A, B). Similarly, the number of DMC1 foci was also greatly reduced in Spidr^−/− spermatocytes compared to WT at leptotene stage (67.1 ± 10.567 in Spidr^−/− versus 121.1 ± 33.099 in WT), zygotene stage (82.846 ± 23.954 in Spidr^−/− versus 163.538 ± 27.235 in WT), and pachytene stage (26.429 ± 10.143 in Spidr^−/− versus 68.7 ± 13.174 in WT) with no significant change in the fluorescence intensity (Figure 3C, D and Supplementary Figure S3C, D). The reduction of RAD51 and DMC1 foci in Spidr^−/− spermatocytes is also supported by a previous study (18).

Figure 3.

SPIDR is indispensable for the recruitment of recombinase RAD51 and DMC1. (A–D) Representative images of WT and Spidr^−/− spermatocytes immunostained for SYCP3 (red) and DMC1/RAD51 (green) at the leptotene, zygotene, and pachytene stages are shown in (A, C). DMC1/RAD51 foci counts were reduced in Spidr^−/− spermatocytes. Quantification of foci for images in (B, D). P values were calculated by Student's t-test. Error bars, mean ± s.d. Scale bars, 5 μm. (E) Co-Immunoprecipitation analysis of SPIDR and RAD51/DMC1 from 293T cells extracts which overexpressed these three proteins. (F) Yeast-two-hybrid: SPIDR(1–500aa) interacts with RAD51/DMC1/BLM in the second panels, while SPIDR (501–933aa) interacts with DMC1/BLM in the third panels with SPIDR as bait and RAD51/DMC1/BLM as prey. The degree of dilution of fungus solution increased from left to right.

Before the formation of ssDNA filaments, the long single strand 3’ overhangs were coated by RPA complex, which subsequently were replaced by RAD51 and DMC1 (7). An increase in RPA foci at leptotene stage (290 ± 26.291 in Spidr^−/− versus 217.375 ± 41.071 in WT), zygotene stage (284.1 ± 49.253 in Spidr^−/− versus 257 ± 61.514 in WT), and pachytene stage (185 ± 9.472 in Spidr^−/− versus 148.6 ± 17.422 in WT) were observed in Spidr^−/− spermatocytes compared to WT (Supplementary Figure S4E, F), indicating accumulated end-resected intermediates. Further, the RPA foci exhibited brighter in Spidr^−/− spermatocytes at zygotene stage (Supplementary Figure S3B). There is a possible explanation for this phenotypic difference. In Spidr^−/− spermatocytes, the inefficient loading of RAD51 and DMC1 on ssDNA delayed the homology searching and strand exchange, and resulted in the RPA foci’ s aberrant retention.

To determine whether SPIDR interacts with the homologous recombinase RAD51 and DMC1, we carried out co-immunoprecipitation (co-IP) in 293T cells, and found that both MYC-RAD51 and MYC-DMC1 were pulled-down by FLAG-SPIDR; noteworthy, this was not disrupted upon the addition of DNase, suggesting that the detected interaction is not dependent on DNA (Figure 3E). Further, a yeast-two-hybrid (Y2H) assays also supported the idea that SPIDR interacts with RAD51 and DMC1. SPIDR’s N terminus (1–500aa) mediates the SPIDR-RAD51 interaction (Figure 3F, second panels), while the interaction of SPIDR-DMC1 is mediated by both SPIDR’s N (1–500aa) and C termini (501–933aa) (Figure 3F, third panels).

A previous study found that BLM focus formation was severely impaired in SPIDR-depleted U2OS cells (12), which motivated us to examine whether BLM foci formation is impaired in Spidr^−/− spermatocytes during meiotic recombination. Immunostaining against BLM showed no defects in Spidr^−/− spermatocytes compared with WT, which is quite different from that observed in human somatic cells (Supplementary Figure S4G, H). Thus, SPIDR is required for the assembly of RAD51 and DMC1 foci, but not of BLM foci, during meiotic recombination.

Spidr^−/− oocytes exhibit defects in meiotic recombination

Chromosome spreads from different stages of embryonic female Spidr^−/− mice showed that meiosis initiates normally, and progress to pachytene stage at embryonic day17.5 (E17.5) (Figure 4A). However—and similar to our findings in males showing that few Spidr^−/− spermatocytes proceeded to pachytene—we found that only 7.5% of Spidr^−/− female oocytes reach pachytene stage at E17.5, whereas 58.8% of WT female oocytes reach pachytene stage (Figure 4B). These results clearly suggested that the meiotic progression was delayed in Spidr^−/− female oocytes.

Figure 4.

Defective meiotic progression in Spidr^−/− oocytes. (A) Representative images of oocytes immunostained for SYCP3 (red) and SYCP1 (green) at the leptotene, zygotene, and pachytene in WT and Spidr^−/− mice. Scale bars, 5 μm. Error bars, mean ± s.d. (B) Proportions of WT and Spidr^−/− oocytes in different meiotic stages, determined by calculating the number of oocytes at each stage divided by the total number of oocytes counted in E17.5 ovaries. P values were calculated by Student's t-test. Error bars, mean ± s.d. (C–H) Representative images of WT and Spidr^−/− oocytes immunostained for SYCP3 (red) and RAD51/DMC1/MLH1 (green) in E17.5 ovaries (C, E, G). RAD51/DMC1/MLH1 focus counts are reduced in Spidr^−/− oocytes. Quantification of foci for images in (D, F, H). Scale bars, 5μm. P values were calculated by Student's t-test. Error bars, mean ± s.d.

To investigate whether the SPIDR-dependent assembly of RAD51 and DMC1 is essential for meiosis progression in oocytes, immunostaining of RAD51 and DMC1 were implemented in E17.5 oocytes. Spidr^−/− oocytes display reduction in RAD51 (75.67 ± 13.78 in Spidr^−/− versus 107.1 ± 25.64 in WT) and DMC1 (106.3 ± 36.15 in Spidr^−/− versus 150.1 ± 42.35 in WT) foci at zygotene stage (Figure 4C–F). The degree of RAD51 and DMC1 foci reduction is milder in oocytes as compared with that in spermatocytes.

Remarkably, MLH1 foci were detectable in Spidr^−/− oocytes with a reduction at pachytene stage (16.00 ± 2.507 in Spidr^−/− versus 22.09 ± 3.330 in WT) (Figure 4G, H), indicating that a small proportion of CO-defective oocytes entered the mid-pachytene stage. We further studied whether such CO-defective oocytes could develop into normal follicles after birth. The absence of chiasma causes the failure of univalent chromosomes to align properly while attaching to microtubules at the metaphase plate, which can trigger a spindle assembly checkpoint (SAC) and block anaphase progression (41,42). Only 5.8% of metaphase I (MI) oocytes showed univalent chromosomes (arrowhead) marked by Topoisomerase II (TOP2) which localized on chromosome arms in Spidr^−/− oocytes, indicating that most CO-defective oocytes can escape SAC (Supplementary Figure S5C and D). However, in above 5.8% oocytes, abnormal bivalent formation of a few chromosome pairs is inefficient for triggering the SAC, allowing such oocytes to survive and thus predisposing to aneuploidy (43). Thus, we analyzed the percentage of aneuploid metaphase II oocytes and found that there was only 3.8% aneuploid oocytes in Spidr^−/− mice (Supplementary Figure S5E and F). These findings suggest that most CO-defective oocytes can complete meiosis and develop backward into follicles to support the fertility of Spidr^−/− females after birth; this is quite different from complete pachytene arrest in Spidr^−/− males.

Knockout of spidr leads to POI

A homozygous nonsense mutation in the SPIDR was reported to be related to an autosomal recessive premature ovarian insufficiency (POI) (44). POI, characterized by cessation of menstruation before menopause with elevated gonadotropin levels, is mostly caused by premature exhaustion of the PFP with multiple mechanisms (19,20). The PFP based on oocytes arresting at the diplotene stage of meiosis prophase I represents the ovarian reserve that determines fertility (20–22). To determine whether the loss of follicles within three months in Spidr^−/− female mice was caused by the defective meiosis progression, ovaries dissected from E13.5 (embryonic day 13.5), E15.5 (embryonic day 15.5) and PD1 (postnatal day 1) mice were immunostained with mouse vasa homolog (MVH) antibody in which MVH marks for primordial germ cells (45). Quantification of relative numbers of oocytes showed that there were comparable numbers of oocytes in WT and Spidr^−/− ovaries at E13.5 when meiosis is initiated (Figure 5A–C) and at E15.5 when the oocytes are just entering the pachytene stage (Figure 5D–F) (46). However, there was only 8.2% oocytes in Spidr^−/− ovaries compared to WT ovaries at PD1 (Figure 5G–I). Thus, oocytes failing to proceed to pachytene stage were likely the reason for the exhaustion of PFP in Spidr^−/− mice. Due to the depletion of PFP, the primary follicles decreased greatly on PD8 and PD15 in Spidr^−/− ovaries (Supplementary Figure S5A), and that 3–5w Spidr^−/− female mice showed apparently fewer follicles (Supplementary Figure S5B), which led to POI at 3–4 months (Figure 1H).

Figure 5.

Knockout of Spidr leads to POI. (A, B) Representative WT (A) and Spidr^−/− (B) ovary sections from E13.5 females immunostained for mouse vasa homolog (MVH) with hematoxylin counterstaining. (C) Average oocyte counts showed that there were similar numbers of oocytes in E13.5 WT and Spidr^−/− females before entering into meiosis. (D, E) Representative WT (D) and Spidr^−/− (E) ovary sections from E15.5 females immunostained for MVH with hematoxylin counterstaining. (F) Average oocyte counts showed that Spidr^−/− ovaries contained insignificantly fewer oocytes than WT ovaries when just entering pachytene. (G, H) Representative WT (G) and Spidr^−/− (H) ovary sections from PD1 females immunostained for MVH with hematoxylin counterstaining. (I) Average oocyte counts showed that Spidr^−/− ovaries contained significantly fewer follicles than WT ovaries after birth. In all cases, MVH-positive cells were counted. Counts were made for every section (5 μm per section) and summed to calculate the total number of oocytes per ovary. For each genotype, three ovaries were analyzed. P values were calculated by Student's t-test. Error bars, mean ± s.d. Scale bars, 200 μm.

Considering that Spidr^−/− oocytes also exhibit defects in meiotic recombination, we concluded that SPIDR is also required for recombinase RAD51 and DMC1 recruitment and/or stabilization during female meiotic recombination. However, the defects of Spidr^−/−oocytes meiotic recombination were milder than that in males, and consequently led to premature ovarian insufficiency in Spidr^−/− female mice.

Spidr^−/−chk2^−/− females but not males are fertile

It has been reported that oocyte loss in other Shu complex subunits, SWS1 and SWSAP1, could be completely rescued by eliminating the DNA damage checkpoint kinase CHK2 (17). By contrast, we observed that oocyte loss was only partially rescued in Spidr^−/−Chk2^−/− adult mice (Figure 6A). Further, Spidr^−/−Chk2^−/− females were able to produce more offsprings than Spidr^−/− females within 6 months (Figure 6B). These results suggest that the rescue by CHK2 ablation is likely better in Sws1 and Swsap1 mutants than that observed in Spidr^−/− mice, and CHK2 is indispensable for oocyte elimination in the Shu complex subunit mutants (17).

Figure 6.

Chk2 ablation rescues oogenesis and spermatogenesis of Spidr-deficient mice. (A) Histological analysis of ovaries showed that adult Spidr^−/−Chk2^−/− ovaries have follicles at various stages of oocyte development, unlike Spidr^−/− ovaries. (B) Cumulative numbers of pups per female within 6 months of age. n = 3 for each genotype. (C) Proportions of WT, Spidr^−/− and Spidr^−/−Chk2^−/− spermatocytes in different meiotic stages, determined by calculating the number of spermatocytes at each stage divided by the total number of spermatocytes counted in adult testes. (D) H&E staining of adult testis sections showed occasionally round and elongated spermatids in Spidr^−/−Chk2^−/− tubules. (E) Proposed model for SPIDR function during meiotic recombination.

In contrast to females, Spidr^−/−Chk2^−/− males exhibit only minimal rescue. There were no spermatozoa in the Spidr^−/−Chk2^−/− epididymis (Supplementary Figure S6A). However, the meiosis process continued to move forward upon deletion of CHK2, which is demonstrated by the observations of diplotene and metaphase I cells in Spidr^−/−Chk2^−/− testes (Figure 6C). Post-meiotic cells marked by MSY2 which is highly expressed in post meiotic round spermatids (47) were present in Spidr^−/−Chk2^−/− testes but not Spidr^−/− testes (Supplementary Figure S6B). These post-meiotic cells which escaped the checkpoint elimination even developed into round or elongating spermatids in Spidr^−/−Chk2^−/− testes (Figure 6D). Thus, CHK2 is critical for oocyte and spermatocyte elimination in Spidr^−/− mice, but the rescue by CHK2 ablation exhibits a sex-dependent manner between Spidr^−/− male and female.

DISCUSSION

In this study, we identified a function of SPIDR in meiotic recombination. The reduction in the number of RAD51 and DMC1 foci in the Spidr^−/− mice resulted in abnormal repair of DSBs and abnormal synapsis. SPIDR is a subunit of the SWS1-SWSAP1-SPIDR complex and exerts regulatory effects on recombinases during homologous recombination. It appears that during meiotic recombination the SWS1-SWSAP1-SPIDR complex regulates the recruitment and/or stability of RAD51/DMC1 (Figure 6E).

Proper homologous recombination requires efficient assembly and stabilization of DMC1/RAD51 on ssDNA. Many accessory HR factors have been reported to regulate this process (31,48–51). How these accessory HR factors promote or regulate DMC1/RAD51–ssDNA nucleoprotein filaments remain to be elucidated. In our study, Spidr-deficient mice have a reduction in RAD51 foci and DMC1 foci formation in the early step of meiosis which ultimately resulted in DSB repair defects. These defects are sufficient to render pachytene arrest while spermatocytes can progress to diplotene in Brme1-deficient and Sws1-Swsap1-deficient mice in which DSB defect also exist (17,25). These differences position SPIDR as an effective mediator protein for RAD51 and DMC1 nucleoprotein filament formation/stabilization, and DSB repair.

SPIDR specifically interacts with RAD51 and DMC1. In Spidr-deficient mice, RAD51/DMC1 foci are detected at a lower level compared with that of the WT mice. Even if it is not clear in mice, RAD51 has a supporting effect on DMC1 in catalytic meiotic recombination in budding yeast and arabidopsis (39,52). Thus, whether SPIDR directly supports DMC1 function or acts through RAD51 remains to be elucidated.

Although SPIDR is required for the formation of DMC1/RAD51 foci in both males and females, knockout of spidr yielded sex-dependent phenotypes: meiotic progression was fully arrested in Spidr^−/− males, whereas some (albeit obviously reduced) COs were formed in Spidr^−/− females, and this allowed oocytes to develop into primordial follicles that supported fertility in females up to 3 months. Several recombination accessory factors regulating DMC1/RAD51 are known to have sex-dependent impacts: HSF2BP and ATR only affect male meiosis and fertility but have no or little effects on female meiosis or fertility (51,53); this is in contrast to accessory factors that exert similar effects in both males and females (e.g. HOP2-MND1, MEIOB and the Shu complex SWS1-SWSAP1) (17,31,54). Since adult Spidr^−/− males are sterile and females exhibit subfertility, its function depends on sex to some degree. This difference appears possible because the checkpoint is less stringent in females than that in males, thus meiotic errors could be tolerated in oogenesis but not in spermatogenesis (55,56).

We found that loss of the DNA damage checkpoint kinase CHK2 partially restores fertility to Spidr-deficient females but not males. This type of sex-specific, CHK2-mediated rescue has also been reported for Swsap1 and Trip3 knockout mice (17,57). Our study adds another layer of evidence supporting that CHK2 can be informatively conceptualized as a meiotic DNA damage checkpoint for the efficient elimination of abnormal oocytes but not spermatocytes, and this deeper understanding of DSB repair checkpoint should inform therapeutic developments towards treating POI.

In summary, our study provides a comprehensive analysis of meiotic phenotypes and function of SPIDR in homologous recombination. Moreover, distinct regulatory and sex-dependent phenotypes mediated by SPIDR allow us to uncover the association between function and biological environment.

DATA AVAILABILITY

No new data were generated or analysed in support of this research.

SUPPLEMENTARY DATA

Supplementary Data are available at NAR Online.

FUNDING

National Key R&D Program of China [2022YFC2702600, 2018YFC1003400]; Research Unit of Gametogenesis and Health of ART-Offspring, Chinese Academy of Medical Sciences [2020RU001]; Academic Promotion Programme of Shandong First Medical University [2019U001]; Major Innovation Projects in Shandong Province [2021ZDSYS16]; Science Foundation for Distinguished Yong Scholars of Shandong [ZR2021JQ27]; Taishan Scholars Program for Young Experts of Shandong Province [tsqn202103192]; Basic Science Center Program of NSFC [31988101]; Shandong Provincial Key Research and Development Program [2020ZLYS02]. Funding for open access charge: National Key R&D Program of China [2022YFC2702600, 2018YFC1003400]; Research Unit of Gametogenesis and Health of ART-Offspring, Chinese Academy of Medical Sciences [2020RU001]; Academic Promotion Programme of Shandong First Medical University [2019U001]; Major Innovation Projects in Shandong Province [2021ZDSYS16]; Science Foundation for Distinguished Yong Scholars of Shandong [ZR2021JQ27]; Taishan Scholars Program for Young Experts of Shandong Province [tsqn202103192]; Basic Science Center Program of NSFC [31988101]; Shandong Provincial Key Research and Development Program [2020ZLYS02].

Conflict of interest statement. None declared.