Mastl is required for timely activation of APC/C in meiosis I and Cdk1 reactivation in meiosis II

Meiotic maturation of mammalian oocytes consists of two consecutive M phases, meiosis I and meiosis II (MII), without an intervening S phase. As in mitosis, entry into both of the meiotic M phases is driven by the activity of Cdk1 in association with cyclin B (Jones, 2004). The resumption of oocyte meiosis I, as characterized by germinal vesicle (GV) breakdown (GVBD), is considered to be equivalent to the mitotic G2/M transition (Eppig et al., 2004; Jones, 2004).

After GVBD in mouse oocytes, Cdk1 activity increases gradually during the lengthy prometaphase of meiosis I (Polanski et al., 1998; Kitajima et al., 2011; Davydenko et al., 2013). Once metaphase I is reached, a transient decrease in Cdk1 activity leads to the extrusion of the first polar body (PB1) as a result of the degradation of cyclin B1. This degradation is mediated by the ubiquitin ligase anaphase-promoting complex/cyclosome (APC/C) that is loaded with its targeting subunit Cdc20. In mitosis, APC/C activity is regulated by the modulation of Cdc20 through phosphorylation, degradation, or the direct binding to inhibitory proteins (Pesin and Orr-Weaver, 2008). The regulation of APC/C activity in mammalian meiosis remains poorly understood, but recent findings show that mitogen-activated protein kinase and Cdk1 are involved in the regulation of APC/C during meiosis I progression in mouse oocytes (Nabti et al., 2014). Once anaphase I is completed, Cdk1 activity is up-regulated rapidly, and this is essential for entry into MII. The elevated Cdk1 activity peaks at metaphase II (MetII) and remains high until fertilization (Kubiak et al., 1992).

Recent studies have indicated that an increase in Cdk1 activity alone is not sufficient for mitotic progression and that a simultaneous suppression of antagonizing protein phosphatase activity is also required (Virshup and Kaldis, 2010). Greatwall (Gwl) kinase or its mammalian orthologue microtubule-associated serine/threonine kinase like (Mastl) is the key regulator that suppresses protein phosphatase activity and thus plays an important role in mitotic progression in all model systems tested so far, including Drosophila melanogaster, Xenopus laevis egg extracts, and mammalian cell lines (Yu et al., 2004, 2006; Archambault et al., 2007; Burgess et al., 2010; Voets and Wolthuis, 2010; Glover, 2012; Lorca and Castro, 2012; Álvarez-Fernández et al., 2013; Wang et al., 2013). In addition, Gwl function has also been shown to be important for the entry into meiosis I in Xenopus oocytes (Dupré et al., 2013), the entry and progression of meiosis I in Drosophila (Kim et al., 2012; Von Stetina et al., 2008), and in in vitro–studied porcine oocytes (Li et al., 2013). Gwl is also required for chromosome segregation during meiosis I in starfish oocytes (Okumura et al., 2014) and for the temporal order of anaphase and cytokinesis at the end of mitotic M phase (Cundell et al., 2013).

In Xenopus egg extracts, Gwl promotes mitotic entry and progression by phosphorylating endosulfine α (Ensa) and cAMP-regulated phosphoprotein 19 (Arpp19), which in their phosphorylated states bind and inhibit protein phosphatase 2A (PP2A)–B55 and prevent dephosphorylation of Cdk1 substrates (Gharbi-Ayachi et al., 2010; Mochida et al., 2010; Virshup and Kaldis, 2010). This pathway is conserved and plays an important role in regulating both mitosis and meiosis in Drosophila (Von Stetina et al., 2008; Rangone et al., 2011; Wang et al., 2011). It was also reported in Xenopus oocytes that protein kinase A maintains the meiotic arrest at prophase I by phosphorylating Arpp19 at serine 109 (Dupré et al., 2013, 2014).

To understand the functions of Mastl in the meiotic divisions of mammalian oocytes, we generated a mutant mouse model in which Mastl was deleted specifically in oocytes. Our results demonstrate that although Mastl is not required for the resumption of meiosis and progression to metaphase I, it plays an important role in the timely activation of APC/C at the end of meiosis I. Moreover, Mastl is indispensable for the rapid Cdk1 reactivation that is essential for MII entry.

We first investigated the expression pattern of the Mastl protein during oocyte maturation and transition into embryos. As shown in Fig. S1 A, Mastl was expressed in mouse oocytes throughout the maturation process and in early embryos. Notably, Mastl protein levels were elevated in the oocytes at MetII, and Mastl exhibited slower migration (Fig. S1 A) as a result of its phosphorylation status in prometaphase I and MetII oocytes (Fig. S1 B). Immunofluorescence microscopy indicated that Mastl was localized in the GV before meiotic resumption but was found throughout the ooplasm after GVBD and in MetII oocytes (Fig. S1 C).

To determine the functions of Mastl during oocyte maturation, we generated a mouse model in which exon 4 of the Mastl gene is flanked by two loxP sequences (Mastl^FLOX; Fig. S1, D and E). We crossed these mice with Zona pellucida 3 (Zp3)–Cre mice (de Vries et al., 2000) to specifically inactivate the Mastl gene in mouse oocytes during the early stages of oocyte growth (Fig. S1 F). The resulting mice (Mastl^FLOX/FLOX;Zp3-Cre) are referred to as OoMastl^−/− mice, and Mastl^FLOX/FLOX mice are labeled as OoMastl^+/+. Immunoblotting confirmed that the Mastl protein was already completely absent in GV-stage OoMastl^−/− oocytes (Fig. S1 G).

OoMastl^−/− females were found to be infertile (Fig. S2 A), but the ovarian development and ovulation of OoMastl^−/− females were normal (not depicted). The requirement of Mastl/Gwl during the entry into meiosis I in Xenopus oocytes (Dupré et al., 2013), the entry and progression of meiosis I in Drosophila (Von Stetina et al., 2008; Kim et al., 2012) and porcine oocytes (Li et al., 2013), and the progression of meiosis in starfish oocytes (Okumura et al., 2014) led us to assume that the Mastl-deficient oocytes would fail to progress through meiosis I. However, we found that in vitro–cultured OoMastl^−/− oocytes underwent GVBD with kinetics and efficiencies that were indistinguishable from those of OoMastl^+/+ oocytes (Fig. 1 A). OoMastl^−/− oocytes progressed to metaphase I with chromosome condensation and spindle formation comparable with those of OoMastl^+/+ oocytes (Fig. 1 B, Fig. S2 B, and Videos 1 and 2). The spindles migrated toward the cortex normally in Mastl-deficient oocytes (Fig. 2 D, right; and Video 2). These results suggest that Mastl has no essential role during meiotic resumption or prometaphase I progression in mouse oocytes.

In OoMastl^+/+ oocytes, PB1 extrusion (PBE), an event marking the completion of meiosis I, had occurred by 12 h after GVBD (n = 60) in vitro. In contrast, only ∼60% of OoMastl^−/− oocytes extruded PB1 at 16 h after GVBD (n = 57; Fig. 1 C). Live cell imaging confirmed normal metaphase I entry but delayed PBE in OoMastl^−/− oocytes (Fig. 2 D, right; and Video 2). However, it is worth noting that the in vivo PBE rate of ovulated OoMastl^−/− oocytes (97.03%, n = 236) was comparable with OoMastl^+/+ oocytes (97.77%, n = 90) when collected at 16 h after human chorionic gonadotropin (hCG) treatment (Fig. 1 D). Thus, OoMastl^−/− oocytes are capable of completing meiosis I after a delay. It is possible that oocytes complete meiosis I faster under more optimal in vivo conditions than those in more challenging in vitro cultures. Thus the differences in the timing of completing meiosis I between the control and mutant oocytes might be missed in vivo.

To study whether the delayed PBE rate in OoMastl^−/− oocytes is caused by a delayed activation of APC/C during meiosis I, we monitored the timing of endogenous securin degradation, which mirrors APC/C activity during oocyte maturation (McGuinness et al., 2009). We found that in OoMastl^+/+ oocytes, securin levels had already decreased at 9 h after GVBD (Fig. 1 E). In contrast, the securin level in OoMastl^−/− oocytes still remained high at 9 h after GVBD but was decreased at 14 h after GVBD, suggesting a delayed degradation of securin in OoMastl^−/− oocytes (Fig. 1 E). Thus, Mastl is required for the timely activation of APC/C that is needed for the completion of meiosis I in mouse oocytes.

We checked whether the suppressed APC/C activity in the OoMastl^−/− oocytes was caused by spindle assembly checkpoint (SAC) activation because of possible subtle defects in spindle and/or chromosome alignment. As an indicator of the SAC status, we examined the localization of the checkpoint protein Mad2, which binds to unattached kinetochores but is released upon the formation of stable microtubule–kinetochore attachments at metaphase I (Wassmann et al., 2003). As shown in Fig. S2 C, we detected Mad2 staining in virtually every kinetochore in both control and mutant oocytes at prometaphase I (3 h after GVBD). Subsequently, Mad2 was completely dissociated from kinetochores of OoMastl^−/− oocytes at metaphase I (7 h after GVBD) when stable kinetochore–microtubule attachments were established, and this was similar to the OoMastl^+/+ oocytes (Fig. S2 C). Therefore, we propose that the activation of SAC in Mastl-null oocytes occurs normally as in the wild-type oocytes.

In addition, we found that in contrast to OoMastl^+/+ oocytes in which the level of Cdc20 was increased at 9 h after GVBD when the activity of APC/C is elevated, the level of Cdc20 in OoMastl^−/− oocytes remained low at this time (Fig. 1 E). A previous study has shown that PBE is delayed when the amount of Cdc20 protein is limited in mouse oocytes (Jin et al., 2010), and this appears to be a similar phenomenon to what we observed in the Mastl-null oocytes. Thus, we hypothesize that Mastl might regulate APC/C activity directly during meiosis I by regulating the levels of Cdc20. Because the regulation of the multisubunit complex of APC/C in mouse oocytes is far from being completely understood, however, the possibility of other regulatory mechanisms being involved cannot be discounted.

Although OoMastl^−/− oocytes completed meiosis I in vivo with normal cytokinesis (Fig. 1 D) and extruded morphologically normal PB1s (Fig. 2 B), confocal microscopy revealed that after PBE, OoMastl^−/− oocytes contained distinct nuclei with decondensed chromatin in both the oocytes and the PB1 (Fig. 2, B [“PB1” and “N” in insets] and C). This was in sharp contrast to the typical bipolar MetII spindles and aligned chromosomes seen in the OoMastl^+/+ oocytes (Fig. 2 A, inset). In 33% (n = 229) of the OoMastl^−/− oocytes, residual central spindle microtubules were still visible between the decondensed chromatin in the PB1 and the nucleus (Fig. 2 B, left inset, ovulated oocytes 16 h after hCG). Moreover, live cell imaging revealed that after PBE, OoMastl^−/− oocytes decondensed their chromatin and did not reform the MetII bipolar spindle (Fig. 2 D and Video 2). These results suggest that Mastl is essential for MII entry. The OoMastl^−/− oocytes subsequently entered S phase after PBE (Fig. S2 D), and within 24 h of in vitro culture, most of the OoMastl^−/− oocytes were cleaved into two cells and subsequently degenerated (not depicted).

To investigate the activity of Cdk1 in OoMastl^−/− oocytes throughout the maturation process, we performed in vitro kinase assays. From the GV stage to prometaphase I (3 h after GVBD) and metaphase I (6 h after GVBD) stages, both OoMastl^+/+ and OoMastl^−/− oocytes displayed similar increases in Cdk1 activity (Fig. 3 A, left of the dashed line). Consistent with the similar levels of Cdk1 activity, both OoMastl^+/+ and OoMastl^−/− oocytes displayed similar levels of inhibitory phosphorylation on Cdk1 (Y15) at the GV stage and subsequent dephosphorylation after GVBD (Fig. 3 B). We also found that during prometaphase I the endogenous Cdk1 substrate, PP1, was phosphorylated on residue T320 to a similar degree in both OoMastl^+/+ and OoMastl^−/− oocytes (Fig. 3 C). These results demonstrate that Mastl-deficient oocytes are capable of phosphorylating Cdk1 substrates through normal up-regulation of Cdk1 activity; thus, the progression through prometaphase I in the OoMastl^−/− oocytes appears to be normal.

We measured PP2A activity in the oocytes (as described in Materials and methods) and found that the PP2A activities in both OoMastl^+/+ and OoMastl^−/− oocytes were comparable at the GV and GVBD stages (Fig. S3 A). Thus, our results suggest that mouse oocytes progress through prometaphase I without suppressing PP2A activity and, therefore, that Mastl regulation of PP2A does not play a major role during prometaphase I.

Nevertheless, Cdk1 activity in OoMastl^+/+ oocytes sharply decreased at the time of anaphase I onset (9 h after GVBD), which is caused by cyclin B1 degradation mediated by APC/C as has been shown previously (Madgwick et al., 2006; Jin et al., 2010). In contrast, the Cdk1 activity in OoMastl^−/− oocytes remained high at this point (Fig. 3 A), which is likely caused by low APC/C activity (as shown in Fig. 1 E) in OoMastl^−/− oocytes. This result suggests that Mastl plays an essential role in triggering the activation of APC/C that is required for the down-regulation of Cdk1 activity and that this mediates the timely onset of anaphase I in mouse oocytes.

The activation of Cdk1 that is required for the entry into MII and the following arrest at MetII (Kubiak et al., 1992) was observed in OoMastl^+/+ oocytes at 14 h after GVBD and after further culture for up to 20 h (Fig. 3 A, top right). However, the Cdk1 activity in OoMastl^−/− oocytes failed to increase after PBE and remained at low levels throughout the extended culture period (Fig. 3 A, bottom right). This indicates that in the absence of Mastl, the rapid elevation in Cdk1 activity, which is needed for MII entry, was completely inhibited. As another indicator of Cdk1 activity, the level of phosphorylation of lamin A/C, an endogenous Cdk1 substrate (Peter et al., 1990; Haas and Jost, 1993), was decreased after PBE in OoMastl^−/− oocytes (Fig. 3 D, ovulated oocytes), which might directly trigger the reformation of the nuclear membrane in the mutant oocytes (Fig. 2 B).

To determine the reasons of the low Cdk1 activity in OoMastl^−/− oocytes after PBE, we measured the levels of cyclin B1. In fact, the levels of cyclin B1 in OoMastl^−/− oocytes were higher compared with OoMastl^+/+ oocytes (Fig. 3 D, ovulated oocytes), and this seems to be the result of low APC/C activity in OoMastl^−/− oocytes (Fig. 1 E). However, the inhibitory phosphorylation of Cdk1 at Y15 was significantly increased in OoMastl^−/− oocytes after PBE in both in vitro–cultured oocytes (Fig. 3 B, 14 h after GVBD) and in ovulated oocytes (Fig. 3 D). The increased inhibitory phosphorylation of Cdk1 in OoMastl^−/− oocytes is most likely the reason for the low activity of Cdk1 after PBE (Fig. 3 A).

Collectively, our results indicate that Mastl depletion in mouse oocytes results in Cdk1 inactivation through its inhibitory phosphorylation at MII entry, but not during meiosis I. We propose that during MII entry, the suppression of PP2A activity by Mastl might prevent the dephosphorylation of Wee1B at its inhibitory sites and prevent the dephosphorylation of Cdc25B at its activating sites (Pal et al., 2008; Vigneron et al., 2009), both of which promote the activation of Cdk1.

To investigate whether the suppressed Cdk1 activity and failure of MII entry in OoMastl^−/− oocytes resulted from elevated PP2A activity, we measured PP2A activity in oocytes immediately after PBE. We found that PP2A activity in OoMastl^+/+ oocytes was very low at this point, but PP2A activity in the OoMastl^−/− oocytes was about nine times higher compared with OoMastl^+/+ oocytes (Fig. S3 B). These results indicate that the deletion of Mastl in mouse oocytes results in a dramatic elevation in PP2A activity specifically after the completion of meiosis I, and this prevents the entry into MII. Based on these findings, we suggest that Mastl-mediated suppression of PP2A plays an essential role in the rapid activation of Cdk1 and the phosphorylation of Cdk1 substrates required for MII entry. However, Mastl-mediated suppression of PP2A is not required for the entry and progression of meiosis I in mouse oocytes. At the moment, the identity of the specific PP2A isoform that is regulated by Mastl in mouse oocytes remains unclear because of the lack of specific antibodies against the B55 subunits. It would also be of interest to measure the PP2A activity using a known physiological substrate of Cdk1 during mouse oocyte maturation.

We tested whether pharmacological activation of Cdk1 or inhibition of PP2A could lead to MII entry in OoMastl^−/− oocytes. Treatment with the Wee1 inhibitor PD166285 (Li et al., 2002; Hashimoto et al., 2006) led to reduced levels of Cdk1 (Y15) phosphorylation and dramatically increased H1 kinase activity in OoMastl^−/− oocytes (Fig. 4 A). This resulted in the breakdown of the nuclear membranes and chromosome condensation in all of the oocytes (n = 33), and 73% of the oocytes formed bipolar MII spindles (Fig. 4, B and C; and Fig. S3 E). Because PP2A activity is higher in OoMastl^−/− oocytes after PBE (Fig. S3 B), we also treated the OoMastl^−/− oocytes after PBE with the protein phosphatase inhibitor okadaic acid (OA; 50 nM; Cohen et al., 1989; Chang et al., 2011; Tay et al., 2012). Inhibition of phosphatases resulted in 42% (n = 121) of the OoMastl^−/− oocytes forming bipolar spindles with aligned chromosomes (Fig. 4, B and C; and Fig. S3 E). Oocytes that failed to form MII spindles after culture with OA still contained nuclei (Fig. S3 E, OA). The OA treatment also led to an increase in Cdk1 activity and decreased Cdk1 phosphorylation on residue Y15 (Fig. 4 A). Finally, we found that the combination of both inhibitors led to the breakdown of the nuclear membrane in all the oocytes and had a moderate additive effect, with 86% of the OoMastl^−/− oocytes forming bipolar spindles with aligned chromosomes (Fig. 4, B and C). This suggested that both Cdk1 and PP2A are involved in an interdependent manner during MII entry.

We measured Ensa and Arpp19 protein levels in oocytes after PBE to investigate their involvement in the failure of OoMastl^−/− oocytes to enter MII. Similar levels of Arpp19 were expressed in OoMastl^−/− and OoMastl^+/+ oocytes, but the Ensa protein could not be detected in OoMastl^−/− oocytes (Fig. 5 A and Fig. S3 C) despite comparable levels of Ensa mRNA (Fig. 5 B). We found that the Ensa protein levels were increased after GVBD in OoMastl^+/+ oocytes but Ensa was absent in OoMastl^−/− oocytes (Fig. S3 C). A previous study in Xenopus egg extracts suggested that Mastl phosphorylates and activates Ensa, leading to PP2A inhibition (Mochida et al., 2010), but our data indicate the possibility that the translation or stability of Ensa is regulated by Mastl in mouse oocytes. We hypothesize that Mastl might be essential for the stability of Ensa through its phosphorylation.

Recent studies have shown that Mastl/Gwl kinase is essential for the initiation and progression of prometaphase during mitosis by suppressing the antagonizing activity of PP2A (Gharbi-Ayachi et al., 2010; Mochida et al., 2010; Domingo-Sananes et al., 2011; Mochida and Hunt, 2012; Álvarez-Fernández et al., 2013). In our study, however, we found that the Mastl-deficient oocytes entered meiosis I with normal timing and efficiency and progressed through prometaphase and metaphase I normally. These oocytes completed meiosis I with a delay in PBE as the result of insufficient APC/C activation. Moreover, we demonstrated that the Mastl-deficient oocytes failed to proceed to MII and instead entered interphase, which was caused by a failure to increase Cdk1 activity and was associated with an increase in PP2A activity.

Our results suggest that in mammalian oocytes Mastl activity is essential for entry into MII. In contrast, Mastl is not required for meiosis I entry but becomes important at the end of meiosis I for the activation of APC/C. Because PP2A has been shown to inhibit APC/C-dependent degradation of securin in HeLa cells (Hellmuth et al., 2014), it might be possible that increased PP2A activity in Mastl-null oocytes inhibits APC/C and delays the onset of anaphase I and PBE. Previous experiments have shown that the dual-specificity phosphatase Cdc14B negatively regulates the resumption and progression of meiosis in an APC/C-dependent manner (Schindler and Schultz, 2009). In the absence of Mastl, Cdc14B might act to counteract Cdk1 and thereby inhibit the transition from prophase to M phase.

The prometaphase of oocyte meiosis I is a lengthy process that takes around 6–8 h in mice (Polanski et al., 1998). This is in sharp contrast to the <30 min required for prometaphase in mitotic cells (Rieder et al., 1994). Our findings indicate that progression to metaphase I in mouse oocytes does not require Mastl-mediated PP2A suppression. Our data raise the possibility that the sustained PP2A activity causes the slow increase in Cdk1 activity and at the same time dephosphorylates Cdk1 substrates, which slows the progression of meiosis I.

In summary, Mastl is required in mouse oocytes for the timely activation of APC/C at the exit of meiosis I and is essential for the reactivation of Cdk1 needed for entry into MII. This study indicates a switch in the mechanisms that regulate Cdk1 activity between the two meiotic divisions in mouse oocytes and provides the starting point for a more detailed exploration of the mechanisms behind Mastl’s role in mammalian meiosis.

The Mastl^FLOX mouse strain was generated essentially as described previously for Cdk1^FLOX mice (Diril et al., 2012). Mouse genomic DNA harboring the Mastl locus was isolated from the BAC clone pBACe3.6 RP23-280K (Invitrogen). Using recombineering techniques (Lee et al., 2001), the LoxP recombination sites and a neomycin-selection cassette were introduced flanking the fourth coding exon of the mouse Mastl genomic locus. The 5′ homology arm comprised 3,690 bp (Mus musculus strain C57BL/6J GRCm38.p2 assembly, Chr2:23, 149, 361-23, 145, 671), whereas the 3′ homology arm included 5,899 bp (23, 138, 962-23, 144, 861). The resulting targeting vector (PKB926) was linearized by NotI digestion, and embryonic stem (ES) cells were electroporated. After positive and negative selection with Geneticin and ganciclovir, respectively, genomic DNA of surviving ES cell colonies was screened for homologous recombination by Southern hybridization using 5′ and 3′ probes (M. musculus strain C57BL/6J GRCm38.p2 assembly, Chr2:23, 153, 022-23, 150, 791 [PKO719/PKO720] and 23, 138, 362-23, 138, 961 [PKO756/PKO757]) located outside of the targeting vector. Correctly targeted ES cell clones (3002, 3003, 3004, and 3009) were identified and used for the generation of the Mastl conditional knockout mouse strain. To generate the Mastl^FLOX allele, the neomycin cassette was removed by crossing Mastl conditional knockout mice with β-actin–Flpe transgenic mice (Rodríguez et al., 2000; strain name: B6.Cg-Tg [ACTFLPe] 9205Dym/J; stock no. 005703; The Jackson Laboratory). Mastl^FLOX/FLOX mice of a mixed background (129S1/SvlmJ and C57BL/6) were backcrossed to C57BL/6J mice for four generations and crossed with transgenic mice carrying Zp3 promoter–mediated Cre recombinase in a C57BL/6J background (de Vries et al., 2000). PCR genotyping primers sequences can be found in Table S1. After multiple rounds of crossing, we obtained homozygous mutant female mice lacking Mastl in their oocytes (OoMastl^−/− mice). Littermates that did not carry the Cre transgene are referred to as OoMastl^+/+ mice and were used as controls.

All mice were housed under controlled environmental conditions with free access to water and food. Illumination was on between 6 am and 6 pm. Experimental protocols were approved by the regional ethical committee of the University of Gothenburg, Sweden, and by the Institutional Animal Care and Use Committee at the Biological Research Centre mouse facility at Biopolis, Singapore.

Rabbit pAbs against Mastl were raised using an N-terminal 6-His–tagged peptide fragment from the mouse Mastl protein (residues 461–694, PKB908) as the antigen using a published protocol (Berthet et al., 2003). Rabbit pAbs against Arpp19 were raised using an N-terminal GST-tagged fusion protein of mouse Arpp19 (residues 25–112, PKB1506) as the antigen using a published protocol (Berthet et al., 2003). Mouse mAb against securin and rabbit pAb against Cdc20 were from Abcam. Human anti-CREST antibody was from Antibodies Inc., and rabbit anti-Mad2 antibody was obtained from Covance. Mouse mAbs against Cdk1 and HA were purchased from Santa-Cruz Biotechnology, Inc. Rabbit pAbs against Ensa, phospho-PP1α (Thr320), phospho–lamin A/C (Ser22), lamin A/C, cyclin B1, and phospho-Cdk1 (Y15) were obtained from Cell Signaling Technology. Mouse mAbs against β-actin, M2 and M16 media, dibutyryl-cAMP (dbcAMP), pregnant mare serum gonadotropin (PMSG), hCG, hyaluronidase, DAPI, mouse monoclonal anti–α-tubulin–FITC, and mineral oil were purchased from Sigma-Aldrich. OA and PD166285 were obtained from Tocris Bioscience. MK-1775 was obtained from Selleck Chemicals. Western blots were performed according to the instructions of the suppliers of the different antibodies and visualized using the ECL Prime Western Blotting Detection System (GE Healthcare). Alkaline phosphatase treatment was performed in 100 mM NaCl, 50 mM Tris-HCl, pH 7.9, 10 mM MgCl₂, and 1 mM DTT with the addition of 1 U rAPid alkaline phosphatase (Roche) for 1 h at 37°C.

Zp3-Cre males were mated with R26R females (003474; The Jackson Laboratory) to generate the Zp3-Cre; R26R mice. Whole-mount β-galactosidase staining was performed to visualize the labeled oocytes in the postnatal day 10 (PD10) ovaries of Zp3-Cre; R26R mice. In brief, ovaries were fixed in 4% PFA for 1 h at 4°C and then rinsed three times for 15 min each in a buffer consisting of 1× PBS, pH 7.4, 2 mM magnesium chloride, 5 mM ethylene glycol tetraacetic acid, 0.01% sodium deoxycholate, and 0.02% NP-40 at RT. The ovaries were then incubated with a staining solution consisting of 1× PBS, pH 7.4, 2 mM magnesium chloride, 0.01% sodium deoxycholate, 0.02% NP-40, 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, and 1 mg/ml X-gal at 37°C overnight. After staining, the ovaries were refixed in 4% PFA for 8 h, embedded in paraffin, serially cut into 8-µm sections, and counterstained with 0.1% nuclear fast red solution. The sections were examined under an Axio Scope A1 upright microscope (Carl Zeiss).

To obtain ovulated oocytes for observation of PBE, immunostaining, and biochemical analysis, 3–4-wk-old female mice were injected intraperitoneally with 7.5 IU PMSG and 48 h later with 5 IU hCG. The ovulation usually occurred 12–14 h after the hCG injection (Hsieh et al., 2002). Ovulated oocytes were collected from the oviducts 16 h after hCG injection unless otherwise noted. Cumulus cells were removed by treatment with 0.3 mg/ml hyaluronidase in M2 medium. For Western blots, the oocytes were lysed in buffer containing 50 mM Tris-HCl, pH 8.0, 120 mM NaCl, 20 mM NaF, 20 mM β-glycerophosphate, 1 mM EDTA, 6 mM EGTA, pH 8.0, 1% NP-40, 1 mM DTT, 5 mM benzamidine, 1 mM PMSF, 250 µM sodium orthovanadate, 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 1 µg/ml pepstatin.

For inhibitor experiments, the protein phosphatase inhibitor OA (50 nM) and the Wee1 inhibitor PD166285 (10 µM) were added in M16 medium. Ovulated oocytes (those that contained a PB1 and a nucleus) collected 16 h after hCG treatment were treated for 6 h with PD166285, OA, or a combination of PD166285 and OA. When treated with OA, oocytes were treated for 10 h. For controls, oocytes were cultured with M16 medium.

3–4-wk-old female mice were injected with 7.5 IU PMSG intraperitoneally, and 42–44 h later the mice were sacrificed and their ovaries were collected in 100 µg/ml dbcAMP-containing M2 medium. Fully grown GV-stage oocytes surrounded by cumulus cells were released by puncturing the ovaries in M2 medium supplemented with dbcAMP to maintain meiotic arrest during the in vitro operation. Oocytes were freed from the attached cumulus cells by repetitive pipetting through a narrow-bore glass pipette. When GV-stage oocytes were required for Western blots, the denuded oocytes were counted and lysed as described above. For immunofluorescence microscopy, GV-stage oocytes were fixed in PFA in PBS, pH 7.4, at RT.

To follow GVBD in vitro, the denuded oocytes were washed twice in dbcAMP-free M2 medium followed by a single washing in M16 medium. The oocytes were then cultured in M16 medium at 37°C in a humidified atmosphere of 5% CO₂ under mineral oil. 20 oocytes were cultured in a 50-µl drop of M16 medium. GVBD rates were recorded every 15 min after culturing in M16 medium.

To obtain GVBD oocytes for biochemical experiments, oocytes were collected 3 h after in vitro culture or at different time points of in vitro culture as mentioned in the respective figure legends. Oocytes were collected 3 h after GVBD for analysis at the prometaphase I stage and 6 h after GVBD for analysis at the metaphase I stage. The oocytes were lysed as described above.

To determine the localization of Mastl protein in mouse oocytes, wild-type oocytes at GV, GVBD, and MetII stages were fixed in 4% PFA. After the fixed oocytes were permeabilized with 0.5% Triton X-100 at RT for 20 min, they were blocked with 1% BSA-supplemented PBS for 1 h and incubated at 4°C with Mastl antiserum overnight, followed by an incubation with Alexa Fluor 594 goat anti–rabbit IgG (Invitrogen) for 1 h at RT. OoMastl^−/− oocytes were used as negative controls.

For spindle and DNA staining of oocytes at metaphase I, OoMastl^−/− and OoMastl^+/+ oocytes that were cultured in M16 medium for 6 h after GVBD were fixed in 4% PFA at RT. For MetII spindle staining, ovulated oocytes harvested from the oviduct 16 h after hCG injection were fixed in 4% PFA. For assessing MII entry after inhibitor treatments, the cultured oocytes were fixed in 4% PFA at the end of their respective incubation periods. After the fixed oocytes were permeabilized with 0.5% Triton X-100 and blocked with 1% BSA-supplemented PBS as mentioned above, they were incubated with anti–α-tubulin–FITC antibody for 1 h at RT. After three washes in PBS containing 0.1% Tween 20 and 0.01% Triton X-100, the oocytes were costained with DAPI (1 µg/ml in PBS). Finally, the oocytes were mounted on glass slides with DABCO-containing mounting medium and examined by laser-scanning confocal microscopy (LSM 700; Carl Zeiss) with a Plan-Apochromat 20×/0.8 objective with the following band pass emission filters (in nm): 385–470 (DAPI), 505–530 (Alexa Fluor 488 and FITC), and 585–615 (Alexa Fluor 594). Z sections were analyzed and projected into one picture using the LSM image browser (Carl Zeiss). The images were then assembled by Illustrator CS2 (Adobe).

Capped RNAs were synthesized in vitro with the T7 mMessage mMachine kit (Ambion) using linearized plasmids provided by M. Anger (Veterinary Research Institute in Brno, Brno, Czech Republic) and purified with the RNeasy Mini kit (QIAGEN). RNAs prepared at 150 ng/µl and 350 ng/µl for H2B-mCherry (for DNA visualization) and Map7-EGFP (for spindle visualization), respectively, were mixed in a 1:1 ratio and injected with an IM-6 (Narishige) into GV-stage oocytes placed in M2 medium supplemented with 0.2 mM 3-isobutyl-1-methylexanthine (IBMX; Sigma-Aldrich) at RT. Injected oocytes were cultured in M16 medium for 6 h in an incubator at 37°C and 5% CO₂. Time-lapse live epifluorescence imaging was performed with a DMI6000 B (Leica) equipped with an HCX PL FLUOTAR 20×/0.40 CORR objective, dichroic filters L5 and TX2, a heating stage at 37°C, and a stage cover maintaining 5% CO₂. Images were taken every 15 min for 16 h with an ORCA-ER charge-coupled device camera (Hamamatsu) operated by the MetaMorph software (Molecular Devices) and processed with ImageJ (National Institutes of Health).

Histone H1 kinase activity assays were performed using five oocytes in a 10-µl reaction volume based on a published protocol (Kudo et al., 2006). In brief, the reaction was performed in a buffer containing 25 mM HEPES, pH 7.4, 15 mM MgCl₂, 20 mM EGTA, 50 mM NaCl, 0.05% NP-40, 0.1 mM ATP, 100 mM PMSF, 1 mM DTT, 1 µg histone H1 (Roche), and 3 µCi γ-[³²P] ATP (PerkinElmer) and incubated at 37°C for 30 min. Samples were boiled in SDS sample buffer and then separated on a 15% SDS–polyacrylamide gel. The radioactive signal was detected by exposing the gels on an FLA 3000 phosphorimager (Fujifilm).

PP2A activity was measured with the PP2A Immunoprecipitation Phosphatase Assay kit (EMD Millipore). In brief, PP2A was immunoprecipitated from lysates of oocytes (400 oocytes for each time point) in lysis buffer containing 20 mM imidazole-HCl, pH 7.0, 2 mM EDTA, 2 mM EGTA, 0.1% NP-40, and freshly added protease inhibitors (Roche). The PP2A protein was pulled down using anti-PP2A antibodies (clone 1D6; EMD Millipore) and protein A agarose beads, after which it was incubated with the synthetic phosphopeptide K-R-pT-I-R-R at 30°C for 10 min before detection with malachite green phosphate detection solution (EMD Millipore) according to the manufacturer’s instructions. The resulting color intensity was measured at 650 nm with a POLARstar Omega microplate reader (BMG LABTECH GmbH). Although this assay has been routinely used for measuring PP2A activity (Tay et al., 2012; Fan et al., 2013; Wei et al., 2013), it does not use a peptide substrate derived from a known physiological substrate of Cdk1 that is dephosphorylated by PP2A-B55. Previous experiments have used such a substrate in experiments with Xenopus egg extracts (Mochida et al., 2009, 2010), but such methods have not yet been optimized for use in mouse oocytes. Because the substrate specificity of PP2A is regulated by its associated B subunits, optimization of such a method for use with mouse oocytes would be able to characterize the activity of B55-associated PP2A in Mastl-null oocytes in future studies.

After GVBD, oocytes were allowed to mature for 14 h in M16 medium, and oocytes with PB1 were transferred to M16 medium containing 100 µM BrdU (Sigma-Aldrich) and allowed to mature for another 10 h. Oocytes were fixed in 4% PFA. After permeabilization with 0.5% Triton X-100 at RT for 20 min, oocytes were incubated in 4 M HCl solution at RT for 10 min. After neutralization for 10 min in 100 mM Tris-HCl, pH 8.0, and a second fixation with 4% PFA for 20 min at RT, oocytes were blocked overnight at 4°C with 1% BSA-supplemented PBS. Oocytes were labeled with BrdU mouse mAb (clone MoBU-1; 1:200 dilution) and Alexa Fluor 488 conjugate (Life Technologies) for 2 h at RT. After three washes in PBS containing 0.1% Tween 20 and 0.01% Triton X-100, the oocytes were costained with DAPI (1 µg/ml in PBS). Finally, the oocytes were mounted on glass slides with DABCO-containing mounting medium and examined by laser-scanning confocal microscopy (LSM 700 Inverted).

Total RNA was isolated from oocytes using the RNeasy Mini kit according to the manufacturer’s instructions. Oligo (dT)–primed cDNA was synthesized using qScript reverse transcription (Quanta Biosciences). Real-time quantitative PCR for Ensa was performed using primers 5′-CCTGAGAAAGCTGAGGAGGC-3′ and 5′-AGGTTCTTGTCTGCTCCTGC-3′ and for Arpp19 using primers 5′-CCGCAACCTTTGAGGGAAAG-3′ and 5′-CCTCCAGGCTTTTGTCCCAA-3′ on an ABI 7500 Real-Time PCR System (Applied Biosystems) using SYBR Green PCR master mix (Applied Biosystems). The reactions were performed in triplicate. Threshold cycle (Ct) values were obtained, and the ΔΔCt method was used to calculate the fold changes. All of the values were normalized to Gapdh.

Mouse Ensa and Arpp19 cDNAs were amplified by RT-PCR from RNA extracted from mouse embryonic fibroblasts. After the introduction of N-terminal HA tags by PCR, the cDNAs were cloned into pBOBI mammalian expression vectors and the sequences were verified. For expressing the Ensa and Arpp19 proteins, pBOBI-HA-Ensa (PKB1481) and pBOBI-HA-Arpp19 (PKB1483) plasmids were transfected into HEK293 cells using Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer’s instructions. Cell lysates were prepared in EBN buffer (80 mM β-glycerophosphate, pH 7.3, 20 mM EGTA, 15 mM MgCl₂, 150 mM NaCl, 0.5% NP-40, 1 mM DTT, and protease inhibitors [20 µg/ml each of leupeptin, chymostatin, and pepstatin; EI8, EI6, and EI10; EMD Millipore]) for 20 min with constant shaking at 1,200 rpm. Lysates were centrifuged for 30 min at 18,000 g at 4°C, and supernatants were snap-frozen in liquid nitrogen and stored at −80°C.

All experiments were repeated at least three times. For comparisons between groups, the differences were calculated with Student’s t test, and a difference was considered to be significant if P < 0.05.

Fig. S1 shows the phosphorylation and localization of Mastl in mouse oocytes and generation of Mastl conditional knockout mice. Fig. S2 shows that although the Mastl conditional knockout females are infertile, the spindle, chromosome condensation, and SAC activation in the oocyte are normal and Mastl-null oocytes enter S phase after PBE. Fig. S3 shows the PP2A activities and Ensa protein levels during oocyte maturation. Table S1 lists PCR genotyping primers. Video 1 shows the normal MII entry in an OoMastl^+/+ oocyte. Video 2 shows the failure of MII entry in an OoMastl^−/− oocyte by live imaging.

We are thankful to Eileen Southon and Susan Reid (National Cancer Institute–Frederick, Frederick, MD) for help in generating the Mastl^FLOX/FLOX mice and to the staff at the University of Gothenburg core facility for cellular imaging for assistance with confocal microscopy.

This study was supported by grants (to K. Liu) from the Jane and Dan Olssons Foundation, the LUA/ALF-medel Västra Götalandsregionen, AFA Insurance, the Swedish Research Council, and the Swedish Cancer Foundation; by funding (to P. Kaldis) from the Biomedical Research Council of A*STAR, Singapore; and by funding (to N.R. Kudo) from the Medical Research Council and the Genesis Research Trust.

The authors declare no competing financial interests.