Loss of UBE2S causes meiosis I arrest with normal spindle assembly checkpoint dynamics in mouse oocytes

The process of meiosis is a unique form of cell division that results in the production of haploid gametes from diploid cells (Hassold and Hunt, 2001; Mogessie et al., 2018). Mammalian oocytes enter meiosis during the embryonic stage and are arrested in the diplotene stage of the first meiotic prophase, which is marked by the existence of the germinal vesicle (GV). Only after the arrival of luteinizing hormone peaks during puberty (Adhikari and Liu, 2014) do the fully grown GV oocytes resume meiosis, manifested by germinal vesicle breakdown (GVBD) (Sun et al., 2009), along with spindle organization and homologous chromosome segregation. The first meiotic division (MI) is accomplished with the first polar body extrusion (PBE) (Hampl and Eppig, 1995). Then, the oocytes are arrested at the metaphase of the second meiotic division (MII), awaiting fertilization.

The ubiquitin-proteasome system (Kleiger and Mayor, 2014), which consists of ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzyme (E2) and ubiquitin-ligase enzyme (E3), plays an important role in oocyte maturation. In oocytes, the cell cycle progression requires the activity of maturation promoting factor (MPF), which is a complex of cyclin B1 (CCNB1) and cyclin-dependent kinase 1 (CDK1). Securin (PTTG1) is an inhibitory protein of separase (ESPL1), which cleaves the cohesin complex to segregate chromosomes. In meiosis, the degradation of CCNB1 or securin is triggered by APC/C. Activation of APC/C requires recruitment of its active factor, Cdc20 (Choi et al., 1991). Spindle assembly checkpoint (SAC) proteins (such as BUBR1, BUB3) target Cdc20 (Sun and Kim, 2012) to form the mitotic checkpoint complex (MCC), which inhibits APC/C activity (Sudakin et al., 2001). With all chromosomes correctly attached to the spindle, SAC is inactivated and Cdc20 is released, enabling APC/C activation to facilitate the transition from metaphase to anaphase (Mikwar et al., 2020; Alfieri et al., 2016).

E3 has been extensively studied owing to its substrate specificity. APC/C (Pesin and Orr-Weaver, 2008) and the Skp1-Cullin 1-F-box (SCF) complex (Kinterova et al., 2022) are cullin RING E3 ubiquitin ligases that catalyze the degradation of cyclin and cytostatic factor (CSF) to ensure normal meiosis. Emi1 (Wang and Kirschner, 2013) and Emi2 (FBX43) (Sako et al., 2014) are two SCF substrates that inhibit the activity of APC/C (Tunquist and Maller, 2003), and are degraded by SCF to ensure the meiotic progression. Depletion one of the SCF subunits, FBXO43 (Zhao et al., 2021), results in abnormal SAC and defective APC/C activity. Direct destruction of APC/C subunits or inhibition of activity (Pesin and Orr-Weaver, 2008) can also lead to meiosis failure.

In eukaryotes, APC/C assembles Lys11-linked ubiquitin chains on substrates (Wu et al., 2010; Jin et al., 2008) using two different E2 ubiquitin-conjugating enzymes, UBE2C and UBE2S (Williamson et al., 2009; Chang et al., 2015). UBE2C is primarily responsible for mono-ubiquitylation of APC/C substrates, and UBE2S catalyzes ubiquitin chain elongation. The depletion of both UBE2C and UBE2S results in non-degradation of APC/C substrates in human cells and delayed mitotic processes. However, UBE2S may only be necessary in a subset of cell types, although it may influence SAC activation (Garnett et al., 2009). A recent study (Sako et al., 2014) showed that UBE2C-catalyzed mono-ubiquitylation at multiple sites was sufficient to degrade CCNB1 in Xenopus egg extracts. In porcine oocytes, the absence of UBE2C and UBE2S (Fujioka et al., 2018) had no effect on meiosis. In mouse oocytes, antisense morpholinos were microinjected to knockdown Ube2s, and caused a reduction in PBE (Ben-Eliezer et al., 2015).

In this study, a Ube2s knockout mouse model was used to investigate the role of UBE2S in the meiotic process of mouse oocytes. Our results showed that female Ube2s knockout mice were infertile. Female mice lacking Ube2s were able to conceive normally, but their fertilized eggs were triploid and the embryos died at embryonic day (E) 10.5, resulting in infertility in female mice. Another finding was that SAC functioned normally in Ube2s knockout oocytes, whereas CCNB1 and securin levels were always high, resulting in the failure of chromosome separation and PBE. Our results reveal that a lack of UBE2S affects the progress of oocyte meiosis and leads to female infertility in mice.

To explore the functions of UBE2S in female meiosis, its expression was first analyzed by immunoblotting of oocyte extracts from GV to MII stages (Fig. 1A,B). The results showed that the expression of UBE2S was highest in MI-stage and lowest in GV-stage oocytes. We then checked the subcellular localization of UBE2S by microinjection of Ube2s-mCherry mRNA (100 ng/μl). Exogenously expressed Ube2s mRNA (1.5 μg/μl) did not affect meiotic resumption and PBE in oocytes (Fig. S1). We cultured oocytes for different meiotic stages after microinjecting. The subcellular localization of UBE2S was then examined by immunofluorescence staining. At GV stage, UBE2S was concentrated in the germinal vesicle. From the GVBD stage to the metaphase of MII stage, UBE2S was distributed in the cytoplasm (Fig. 1C).

Ube2s^−/− mice were created by obtaining conditional knockout mice using a gene target method (Fig. S2), and mating with EIIA cre mice to produce knockout mice, which have been described previously (Huang et al., 2020). After the Ube2s knockout mice were obtained, the knockout efficiency was verified first. Western blotting of extracts from Ube2s^−/− oocytes yielded no UBE2S signal, indicating that the protein was absent (Fig. 1D,E). To investigate the fertility of knockout mice, female Ube2s^+/− and Ube2s^−/− mice were naturally mated with wild-type (WT) male mice for more than 6 months. As shown in Fig. 2A, we found that female Ube2s^−/− mice were infertile. Pregnancies were observed but no pups were born from female Ube2s^−/− mice after natural mating. We measured the ovaries and uterus of Ube2s^+/− and Ube2s^−/− mice and found no abnormalities (Fig. S3A-D). The number of naturally ovulated or superovulated oocytes from female Ube2s^−/− mice were similar to that of female Ube2s^+/− mice (Fig. S4).

To investigate the causes of this phenomenon, we examined the implantation sites of female Ube2s^−/− mice at different time points. We collected pregnant uteri from female Ube2s^−/− mice to detect the post-implantation development of embryos. There was no change in the number of implantation sites at E5.5 (Fig. S3E). There was no difference between E9.5 embryos of Ube2s^+/− and Ube2s^−/− mice (Fig. S3F). However, the embryos collected at E10.5 from Ube2s^−/− mice had degenerated (Fig. 2B; Fig. S3G). These results demonstrate that knockout of Ube2s leads to female infertility caused by embryonic lethality.

In previous reports (Niemierko, 1981), mouse triploid embryos could develop up to E10.5 and then failed to grow. This is similar to the time point at which embryos died in female Ube2s^−/− mice. We performed chromosome spreading of zygotes from Ube2s^+/− mice and Ube2s^−/− mice that had been arrested at metaphase by treatment with nocodazole. As expected, most of the zygotes from female Ube2s^−/− mice were triploid (Fig. 2C,D).

To investigate whether the triploid zygotes were caused by oocytes, MII oocytes were obtained by superovulation, and the rates of oocytes with polar bodies were checked. As shown in Fig. 2E,F, the oocytes of female Ube2s^−/− mice rarely exhibited PBE at 12 h after human chorionic gonadotropin (hCG) injection, indicating that the Ube2s^−/− oocytes might be arrested at MI.

To test the chromosome status in Ube2s^−/− oocytes, we performed chromosome spreading of superovulated oocytes. Chromosome spreading from Ube2s^+/− oocytes revealed the expected 20 pairs of sister chromatids. In contrast, Ube2s^−/− oocytes had 20 bivalent chromosomes with no homologous chromosome separation (Fig. 3A,B).

We further explored the SMC3 protein, a subunit of cohesin, to examine the adhesion of chromosomes. GV oocytes from female Ube2s^+/− and Ube2s^−/− mice were cultured and collected at 8 h after GVBD. As shown in Fig. 3C,D, we observed that SMC3 was absent in Ube2s^+/− oocytes, whereas it was still present on the chromosome arms in Ube2s^−/− oocytes. These results indicate that the Ube2s^−/− oocytes were arrested at MetI with inseparable homologous chromosomes.

To investigate why oocytes are arrested in MetI, we analyzed the meiotic maturation process of Ube2s^−/− oocytes. We found that fully grown Ube2s^−/− GV oocytes resumed meiosis, but failed to extrude polar bodies (Fig. 4A). Moreover, spindles of Ube2s^−/− oocytes were well organized at 4 h and 6 h after GVBD (Fig. 4B, Fig. S5).

To determine whether the MetI arrest was caused by sustained MPF activity in Ube2s^−/− oocytes, we examined the dynamics of the APC/C substrates (CCNB1 and securin) by live imaging of oocytes microinjected with mRNAs coding for protein fused to fluorescent tags. As shown in Fig. 4C,D (also see Movies 1 and 2), the green fluorescence intensity (representing expression of CCNB1) showed a significant decline at 8 h after GVBD with PBE in Ube2s^+/− oocytes. However, the green fluorescence intensity of Ube2s^−/− oocytes showed only a slight decrease during MI progression. The securin-EGFP in Ube2s^−/− oocytes was degraded more slowly and remained at a higher level at the end of the analysis than that of control oocytes (Fig. 4E,F, Movies 3 and 4). These results indicate that there was reduced APC/C activity, leading to incomplete degradation of APC/C substrates in Ube2s^−/− oocytes.

Separase is a cysteine protease that cleaves subunits of the cohesin complex to separate homologous chromosomes and sister chromatids during meiosis. Securin is a known inhibitor of separase. To illustrate the effect of UBE2S on separase activity, we observed chromosome separation in Ube2s^−/− oocytes using a separase activity sensor (Fig. S6A), which is similar to a sensor reported previously (Li et al., 2019a). It is a plasmid containing H2B, Rad21 and two fluorescent (EGFP and mCherry) tag sequences, which glows yellow on chromosomes due to the superimposition of red and green. The mRNA was injected into GV oocytes and expressed protein was located on the chromosome. When separase is active, Rad21 is cleaved, the EGFP is degraded owing to instability, and the fluorescent signal changes from yellow to red. We observed the fluorescent signal on chromosomes changing suddenly from yellow to red when chromosomes separated in Ube2s^+/− oocytes at 12 h after GVBD (Fig. 4G, Movies 5 and 6). In contrast, yellow fluorescent signals were maintained on chromosomes in Ube2s^−/− oocytes, which indicated that separase cannot be activated in Ube2s^−/− oocytes. These results explain why Ube2s knockout oocytes failed to separate homologous chromosomes and extrude the polar bodies.

Moreover, to investigate whether exogenous UBE2S could recover the MetI arrest in Ube2s^−/− oocytes, we injected Ube2s mRNA into Ube2s^−/− oocytes at the GV stage (Fig. S7). Unexpectedly, our results showed that replenishment of Ube2s mRNA could not reverse this phenotype.

Activation of APC/C requires timely deactivation of SAC (Li et al., 2009; Zhang et al., 2005). To evaluate whether the failure of APC/C activation was due to sustained activity of the SAC, we used immunofluorescence to examine the activity of the SAC proteins BUB3 and MAD1 (MD1L1) in MetI (4 h culture after GVBD) and AnaI (8 h culture after GVBD) oocytes of Ube2s^+/− mice. Because Ube2s^−/− oocytes cannot reach AnaI, we collected Ube2s^−/− oocytes 8 h culture after GVBD for the assay. The results showed that both BUB3 and MAD1 were observed at MetI, but they were absent from the kinetochores in Ube2s^+/− and Ube2s^−/− oocytes collected at 8 h after GVBD (Fig. 5A-F). Reversine has been shown to be an inhibitor of monopolar spindle 1 (MPS1) kinase (Santaguida et al., 2010), which inhibits SAC activity. After 2 h culture in medium, we collected GVBD oocytes for treatment with 0.5 μM reversine for 12 h (Fig. S6B). The PBE rate of Ube2s^+/− oocytes was significantly higher than that of Ube2s^−/− oocytes, and the homologous chromosomes remained undivided in Ube2s^−/− oocytes (Fig. 5G,H). These results indicate that Ube2s knockout does not affect the timely deactivation of SAC.

After fertilization, oocyte activation begins with [Ca²⁺] oscillations. Strontium is a parthenogenetic agent for mouse oocytes that induces [Ca²⁺] oscillations in a similar fashion to normal fertilization (Ma et al., 2005). As Ube2s^−/− oocytes could extrude polar bodies after fertilization, we were curious about whether MetI-arrested oocytes would exit from the first meiosis after parthenogenetic activation (PA). We collected MetII oocytes of Ube2s^+/− mice and MetI-arrested oocytes of Ube2s^−/− mice at 16 h after hCG injection. Mouse oocytes were activated by strontium chloride in Ca²⁺-free CZB medium after 6 h of culture. More than 80% of Ube2s^+/− and Ube2s^−/− oocytes (87.5% and 83.7%) had extruded a polar body (Fig. 6A,B), indicating that Ube2s^−/− oocytes can proceed beyond MI by extruding the first polar body. The results also showed that Sr²⁺-treated Ube2s^−/− oocytes had segregated homologous chromosomes, indicating effects of Sr²⁺ activation on overcoming MetI arrest (Fig. 6C). After Sr²⁺ treatment, PA embryos with a pronucleus were cultured to the blastocyst stage in KSOM medium. The results showed that Ube2s^−/− oocytes could develop to blastocyst stage after PA (Fig. S8), which demonstrates that the ability of Ube2s^−/− PA embryos to develop was normal before implantation in the uterus.

APC/C is a RING E3 ubiquitin ligase that regulates the processes of chromosome separation and cell division. Previous research has demonstrated that UBE2S acts as a co-factor to enhance the function of APC/C and is essential for mitosis (Wu et al., 2010). Errors in oocyte meiosis are common in women. An estimated 20% of human oocytes are aneuploid (Gruhn et al., 2019), and meiosis errors account for one-third of all pregnancy losses (Ma et al., 2020; Qiao et al., 2014). In this study, we verified the important role of UBE2S in female fertility, and proved that UBE2S deletion leads to a failure of the MetI-to-AnaI transition of oocyte meiosis. This was due to a lack of CCNB1 and securin degradation in the presence of SAC deactivation, resulting in the inhibition of separase activation, and subsequent failure of homologous chromosome separation (Fig. 6D).

Ube2s knockout mice survived normally. Female mice lacking UBE2S were unable to produce offspring, but we observed no abnormalities in oocyte ovulation or embryo implantation in the knockout mice (Fig. S3E), and embryo death occurred at E10.5. Previous studies (Niemierko, 1981) have shown that triploid embryos induced in vitro can implant and develop up to E10.5 after transplantation into the uterus of mice. Therefore, we conducted chromosome spreading on the fertilized eggs and found that the majority of them were triploid. Previous research has demonstrated that triploid fertilized eggs generated following Ccnb3 (Li et al., 2019b) or G9a (Ehmt2) (Meng et al., 2022) knockout can be successfully implanted and developed in vivo, but female mice were unable to produce offspring. A previous study (Wang et al., 2022) showed that mouse embryonic stem cells exhibited severe aneuploidy and DNA damage after knockdown of Ube2s. There is a 1% chance of human embryos being triploid (XXX, 69; or XXY, 69), and the fact that some triploid fetuses can even be born (Toufaily et al., 2016) suggests that triploid embryos might have developmental potential. We tried to find some genetic data from human samples to learn the function of UBE2S in human reproduction, but were unable to identify any. We will try to investigate whether there are any mutations or abnormalities associated with UBE2S in human triploid fetuses in further studies.

Meiosis is a cellular process that facilitates the production of haploid gametes from diploid cells, and its proper execution is crucial for successful fertilization and continued development of these gametes (Hassold and Hunt, 2001). According to previous studies (Marangos and Carroll, 2008), CCNB1 binds to CDK1 to activate MPF activity, and its degradation deactivates CDK1, thereby reducing MPF activity and shifting meiosis to the metaphase and anaphase stages. Securin functions as an inhibitor of separase. Upon its degradation, separase is activated, leading to the separation of homologous chromosomes and extrusion of the first polar body through cytoplasmic division (Ishiguro, 2019). However, complete degradation of neither CCNB1 nor securin was observed in UBE2S-deficient oocytes (Fig. 4C-F, Movies 1-4). We initially examined whether the persistent activation of SAC exerts an impact on APC/C activity, leading to the failure of substrate degradation. Surprisingly, deletion of UBE2S did not affect the normal inactivation of the SAC, meaning that Cdc20 was successfully released from SAC. ​In mitosis, ubiquitylation of CCNB1 (Kirkpatrick et al., 2006) or securin is mediated by the E3 enzyme APC/C (Lu et al., 2015), with the K-11 ubiquitin chain acting as a degradative signal for the 26S proteasome to recognize (Brown et al., 2014; Stewart et al., 2016). At the beginning of ubiquitylation of CCNB1 and securin, UBE2C-mediated monoubiquitylation is responsible for degradation. ​In some cell types, knockout of UBE2S alone does not affect the mitosis process (Garnett et al., 2009). When we examined the degradation of CCNB1 and securin in Ube2s-deficient oocytes, we found that they could be partially degraded, suggesting that there might be a threshold for CCNB1 and securin degradation in mouse oocytes.

Our observations in oocytes in which Ube2s had been deleted are similar to those of a previous study (Ben-Eliezer et al., 2015) in which antisense morpholinos were microinjected to knock down Ube2s, causing arrest of the first meiosis in mouse oocytes; this phenomenon was caused by the inability of the substrates of APC/C (cyclin B1 and securin) to degrade. This interesting study used morpholino knockdown of Ube2s, and showed that spindle assembly and SAC inactivation were affected. By contrast, we used knockout mice and found that spindle assemble of MetI oocytes was normal (Fig. S5) and SAC was deactivated (Fig. 5A-F). We cannot rule out the possibility that some unknown factor might compensate for the lack of UBE2S under physiological conditions. We also found that this deletion leads to the creation of triploid zygotes. We suspect that the different approaches (knockdown and knockout) are responsible for some differences in the results of these studies. UBE2S may not be strictly necessary for degradation of APC/C substrates because some oocytes (about 10%) could complete meiosis and extruded a polar body, so it may be that oocytes lacking Ube2s occasionally had enough APC/C activity to cross the threshold and thus completed the transition to anaphase. We tried to rescue the phenomenon of MetI arrest by injecting Ube2s mRNA into Ube2s^−/− oocytes at the GV stage, but with no success. We hypothesize that the absence of UBE2S might cause some unknown change, making it impossible to remedy this phenomenon even if the Ube2s mRNA is supplemented.

​Unexpectedly, Ube2s-deficient oocytes were able to extrude polar bodies upon fertilization and form fertilized eggs for continued development. After PA of Ube2s-deficient oocytes, polar bodies could be extruded in the form of the first polar body (homologous chromosome segregation). Ube2s^−/− zygotes showed no defects in preimplantation development, suggesting that UBE2S is required for MI, not MII. Previous studies (Yoon et al., 2017; Ducibella and Buetow, 1994) have shown that MI-arrested oocytes can be fertilized and develop into blastocysts, because the ability of oocytes to undergo exocytosis of cortical granules in response to Ca²⁺ occurs between MI and MII. Moreover, no live birth was obtained from their MI-arrested oocytes, similar to our observations. Although not all Ube2s-deficient oocytes failed to extrude the first polar body, we were unable to obtain offspring from our knockout female mice. We speculate that, in addition to the problem of oocytes, there were other problems contributing to infertility in female Ube2s^−/− mice.

Overall, our data show that UBE2S is an essential protein for the metaphase–anaphase transition in MI mouse oocytes, and suggest that insufficient APC/C activity with normal SAC deactivation result in MI arrest.

Female Ube2s^−/−mice were generated by crossing male Ube2s^−/− mice with female Ube2s^+/− mice. All mice were generated and housed in specific pathogen-free barrier facilities at the Institute of Zoology, Chinese Academy of Sciences (Beijing, China). Light was provided between 07:00 and 19:00. Animal care and handling were conducted according to the guidelines of the Animal Research Committee of the Institute of Zoology, Chinese Academy of Sciences.

GV oocytes were isolated from ovaries of 6- to 8-week-old female mice and cultured in M2 medium (M7167, Sigma-Aldrich) under liquid paraffin oil at 37°C in an atmosphere of 5% CO₂ in air. Oocytes were collected at different time points of culture for immunofluorescence staining, western blotting and chromosome spreads, respectively. PA embryos were cultured in KSOM medium (MR-121-D, Sigma-Aldrich) under liquid paraffin oil at 37°C in an atmosphere of 5% CO₂ in air.

The following primary antibodies were used: rabbit polyclonal anti-UBE2S antibody (1:1000 for western blotting; ABclonal, A4658); mouse monoclonal anti-α-tubulin-FITC antibody (1:1000; Sigma-Aldrich, F2168); anti-centromere antibody (1:50; Antibodies Incorporated, 15-234), anti-SMC3 antibody (1:50; Abcam, ab128919), anti-BUB3 antibody (1:300; Abcam, ab133699), and rabbit polyclonal anti-MAD1 antibody (1:200; GeneTex, GTX105079). The following secondary antibodies were used: Alexa Fluor 488-conjugated goat anti-rabbit IgG (H+L) and 594-conjugated goat anti-rabbit IgG (H+L) (1:800; Thermo Fisher Scientific, A-11008 and A-11012); Cy5-conjugated donkey anti-human IgG (H+L) (1:100; Jackson ImmunoResearch, 709-175-149).

Total RNAs were extracted from 100 GV mouse oocytes using Dynabeads mRNA DIRECT (Invitrogen, 61021) and first-strand cDNA was generated with a cDNA synthesis kit (abm, G490) using poly(dT) primers. PCR was used to amplify the full-length amplify cDNA for each gene. PCR products were digested using FseI and AscI (New England Biolabs, Inc.). Each fusion plasmid was transfected into T1 competent cells (TransGen Biotech). Each modified plasmid was then extracted (TransGen) and linearized. Capped mRNAs were produced using mMESSAGE T7 (AM1344, Invitrogen), and a poly(A) tail was added with a poly(A) polymerase tailing kit (AP-31220, Epicenter). Finally, the RNeasy cleanup kit (74004, QIAGEN) was used to purify the mRNAs.

Microinjections were performed using a Narishige TE300 microinjector and completed within 30 min. For Ube2s-mCherry overexpression, 100 ng/μl Ube2s-mCherry mRNA solution was injected into the cytoplasm of the GV oocytes. For CCNB1–GFP dynamics analysis, 20 ng/μl Ccnb1 mRNA solution and 20 ng/μl H2B-mCherry mRNA were injected into the cytoplasm of the GV oocytes. For securin-GFP dynamics analysis, 50 ng/μl securin mRNA solution and 20 ng/μl H2B-mCherry mRNA were injected into the cytoplasm of the GV oocytes. For separase dynamics analysis, 400 ng/μl Separase-sensor mRNA solution was injected into the cytoplasm of the GV oocytes. For mRNA injection, oocytes were arrested at the GV stage in M2 medium containing 2.5 μm of the inhibitor 3-isobutyl-1-methylxanthine (IBMX) for 4 h, respectively. The oocytes were then transferred to M2 medium and cultured for 14 h.

For the natural ovulation assay, 6- to 8-week-old female mice were mated with fertile males overnight. Presence of vaginal plugs was used to confirm successful mating. Fertilized eggs were harvested from oviducts, and counted and analyzed after removal of the cumulus mass with 1 mg/ml hyaluronidase (Sigma-Aldrich) in M2 medium.

For the superovulation assay, each female mouse was injected with 10 IU pregnant mare serum gonadotrophin (PMSG; SANGSHENG, PCL0120) followed by 10 IU hCG (SANGSHENG, PCL0121) 48 h later to promote ovulation. PMSG and hCG were dissolved in 10 ml of normal saline per 1000 units. Mice were killed after 12-14 h of hCG treatment and cumulus–oocyte complexes were recovered from each oviduct. After treatment with hyaluronidase (1 mg/ml) in M2 medium, oocytes were collected and counted.

Female mice (6-10 weeks old) were mated with fertile males overnight. The presence of vaginal plugs was designated E0.5. Embryos of E10.5 from female mice were fixed in 4% paraformaldehyde (pH 7.5) overnight at 4°C, dehydrated and embedded in paraffin. The treated embryos were sectioned consecutively at 8 μm for Hematoxylin and Eosin staining. After dewaxing and hydration, the sections were stained with Hematoxylin and 1% Eosin and imaged with a Nikon ECLIPSE Ti microscope. Embryos from more than three mice of each genotype were used for the analysis.

Oocytes for immunofluorescence staining were fixed in 4% paraformaldehyde in PBS for 30 min at room temperature. Fixed oocytes were then transferred to membrane permeabilization solution (0.5% Triton X-100) for 20 min followed by blocking buffer [1% bovine serum albumin (BSA) in PBS] for 1 h. Finally, oocytes were incubated overnight at 4°C with the antibodies described above at appropriate dilutions. After three washes with washing buffer (PBS containing 0.1% Tween-20 and 0.01% Triton X-100), oocytes were stained with DAPI (D9542, Sigma-Aldrich) for 15 min at room temperature. Oocytes were then mounted on glass slides and examined with a laser-scanning confocal microscope (Zeiss LSM 880 META).

For chromosome spreading, oocytes were placed in acid Tyrode's solution (T1788, Sigma-Aldrich) to remove the zona pellucida. A 1 cm×1 cm frame was drawn on a glass slide with a hydrophobic pen. After a brief recovery in M2 medium, the oocytes were transferred onto glass slides and fixed in a solution of 1% paraformaldehyde in distilled H₂O (pH 9.2) containing 0.15% Triton X-100 and 3 mM dithiothreitol as previously reported (Hodges and Hunt, 2002). Immunofluorescent staining was then performed using BUB3, MAD1 or human anti-centromere protein (ACA) antibodies incubated overnight at 4°C, and DAPI was used to stain the chromosomes.

CCNB1-EGFP, securin-EGFP and separase dynamics were filmed on a confocal imaging system. Exposure time was set ranging from 200 to 300 ms depending on the EGFP or mCherry fluorescence levels. The acquisition of digital time-lapse images was controlled by living cell workstation (Andor Dragonfly 200) software packages. Confocal images of CCNB1, securin or separase in live oocytes were acquired with a 20× objective on a spinning-disc confocal microscope.

A total of 150 mouse oocytes were collected in 2× SDS sample buffer and boiled for 5 min at 100°C. Western blotting was performed as described previously (Li et al., 2022). The proteins were separated by SDS-PAGE and then transferred onto PVDF membranes. Next, the membranes were blocked in Tris-buffered saline with 0.1% Tween 20 (TBST) containing 5% BSA for 2 h at room temperature, followed by incubation overnight at 4°C with primary antibody, then incubation with specific secondary antibody at room temperature for 1 h. Finally, the membranes were processed using the enhanced chemiluminescence detection system (Bio-Rad, ChemiDoc XRS System).

For nocodazole treatment, nocodazole (M1404, Sigma-Aldrich) was added at a final concentration of 10 ng to KSOM medium to inhibit mitosis of zygotes. For reversine treatment, reversine (HY14711, MCE) was dissolved in DMSO and stored at −20°C and added into the M2 medium at a final concentration of 0.5 μM after GVBD. For Sr²⁺ treatment, collected oocytes were parthenogenetically activated in Ca²⁺-free CZB (Chatot Ziomek Bavister medium) containing 10 mM Sr²⁺ (204463, Sigma-Aldrich) for 6 h.

Images were acquired using a confocal laser-scanning microscope (LSM 880; Zeiss) equipped with a C-Apochromat 40× water immersion objective. Data analysis was performed using ZEN 2012 LSM software (Zeiss) and ImageJ software.

Images were analyzed with ImageJ software (National Institutes of Health), composed of Illustrator CC5 (Adobe). Statistical parameters, including statistical tests and P-values, are reported in the figure legends. Statistical analyses were performed using Prism Software (GraphPad). For statistical comparison, unpaired Student's t-test was used. Differences with P<0.05 were considered as significant.

We thank all members of the Wang lab for their help and discussions. We thank Shi-Wen Li and Xi-Li Zhu for their excellent technical confocal and live imaging support.