PATL2 regulates mRNA homeostasis in oocytes by interacting with EIF4E and CPEB1

The proper accumulation, storage, translation and degradation of maternal mRNA is essential for oocyte maturation and embryonic development (Coticchio et al., 2015; Sha et al., 2020). After birth, oocytes are maintained in prophase I, which is the period of oocyte growth. During this phase, the oocyte grows rapidly in size and mass accompanied by a dramatic increase in transcriptional activity (Picton et al., 1998), and the oocyte accumulates the maternal mRNA required for subsequent maturation and embryonic development. To prevent premature activation of the developmental program, the maternal mRNAs are translationally silenced by shortening the length of the poly(A) tail, which causes them to be transported into the cytoplasm and sequestered into storage in cytoplasmic granules (Eichhorn et al., 2016; Liu et al., 2019; Passmore and Coller, 2022). Oocyte meiotic maturation and early embryogenesis occur in the absence of transcription and rely on maternal mRNAs stored in oocytes (Susor et al., 2015). Defects in this process can lead to oocyte maturation arrest and early embryonic development problems.

PATL2 (PAT1 homolog 2), belonging to the PAT1 family, was first identified in Xenopus eggs as an oocyte-specific RNA-binding protein (Rother et al., 1992; Ozgur et al., 2010) that binds to maternal mRNA in Xenopus eggs and regulates its translation (Nakamura et al., 2010). We, and others, have independently identified that bi-allelic mutations in PATL2 cause oocyte maturation arrest leading to female infertility (Chen et al., 2017; Maddirevula et al., 2017; Christou-Kent et al., 2018), and subsequent studies and case reports further confirmed this conclusion and expanded the early embryonic development arrest phenotype (Huang et al., 2018, 2022; Wu et al., 2019; Liu et al., 2020). However, the physiological function of PATL2 in the process of oocyte maturation and embryonic development is largely unknown.

In this study, we investigated the physiological function of PATL2 in oocyte maturation and embryonic development by using knockout and knock-in mouse models. We found that PATL2 coupled with CPEB1 and EIF4E to maintain mRNA stability in vitro, and Patl2 knockout impaired the maternal transcriptome and the protein synthesis of maternal mRNAs, thus leading to female infertility. Furthermore, we found that phosphorylation of PATL2 residue S279 plays an important role in its degradation. This study thus shows the physiological function of PATL2 in the process of oocyte maturation and embryonic development and clarifies that PATL2 phosphorylation regulates the protein level of PATL2 for regulating the expression of maternal mRNAs.

Previous transcriptome analyses have shown that Patl2 mRNA is highly expressed in mouse and human oocytes (Chen et al., 2017; Christou-Kent et al., 2018). To study the physiological functions of PATL2, we generated a knockout mouse strain using the CRISPR/Cas9 system to make an 11-bp deletion in exon 1 of the Patl2 gene (Fig. 1A). This deletion caused a reading-frame shift, and western blotting analysis confirmed that the PATL2 protein was deleted in the oocytes of Patl2^−/− mice (Fig. 1B). Patl2^−/− female mice were sterile, whereas the males exhibited normal fertility (Fig. 1C), and an in vitro maturation assay showed that the Patl2^−/− female mice had significantly decreased rates of first polar body extrusion (Fig. 1D). After in vitro fertilization, the metaphase II (MII) oocytes derived from ovulation stimulation in Patl2^−/− female mice could be fertilized, but they all arrested at the two-cell stage, whereas the control embryos from the wild-type (WT) and Patl2^+/− mice exhibited normal embryonic development (Fig. 1E,F). Immunostaining showed that PATL2 was specifically expressed in oocyte cytoplasm and that PATL2 protein level was increased in the oocytes of secondary follicles (Fig. 1G). These results confirmed the essential role of PATL2 in female infertility.

PATL2 is reported to be an oocyte-specific RNA-binding protein (Chen et al., 2017), and thus we explored the effect of PATL2 deletion on mRNA expression in the oocyte maturation process. Germinal vesicle (GV) and MII-stage oocytes were collected for RNA-seq analysis. First, we normalized the FPKM values of genes to the corresponding FPKM values of GFP or mCherry and analyzed the overall decay of maternal mRNA in WT and Patl2^−/− oocytes during meiotic maturation, and we observed an obvious decrease in total mRNA copies in Patl2^−/− oocytes at the GV stage compared with WT oocytes (Fig. 2A). This decrease seemed to be weakened in MII oocytes because 80% of the mRNAs were degraded during meiotic maturation (GV to MII) in WT oocytes (Fig. 2A). We confirmed the result by real-time quantitative reverse transcription PCR (qRT-PCR), with transcripts all showing significant decreases (Fig. 2B; Fig. S1A). We then proposed that deletion of PATL2 had a major effect on the decrease of mRNA expression in growing oocytes and in fully grown GV oocytes. To prove this, we collected growing oocytes from postnatal day (P)5 and P10 female mice, GV oocytes, MI and MII oocytes from 7-8-week-old female mice. We performed qRT-PCR analysis of three clearly downregulated genes, and all three genes showed the greatest decrease at the GV stage (Fig. 2C-E; Fig. S1B-D). To further confirm this, we re-performed the RNA-seq analysis of GV-stage oocytes of WT and Patl2^−/− mice in five biological replicates. The results showed a significant decrease in total mRNA in Patl2^−/− oocytes (Fig. 2F,G; Fig. S3). In addition, oligo(dT) in situ hybridization (ISH) staining also showed a significant decrease in poly(A) mRNA in Patl2^−/− oocytes (Fig. 3A,B; Fig. S2A). To examine the role of transcriptional activity, we performed a 5-ethynyl uridine (EU) staining experiment, and the results showed that there was no significant difference between WT and Patl2^−/− oocytes (Fig. 3C,D; Fig. S2B). The mRNA transporting assay showed that the cRNA in Patl2^−/− GV oocyte nucleus could transport to cytoplasm normally, as in WT oocytes, at 30 min and 60 min after nucleus cRNA injection (Fig. S4). We then analyzed protein synthesis in the WT and Patl2^−/− oocytes according to the L-homopropargylglycine (HPG) assay, which showed that the protein synthesis in Patl2^−/− oocytes decreased significantly (Fig. 3E,F). Together, these results demonstrate that PATL2 deletion leads to the decrease of maternal mRNA expression in GV stages.

Our study showed that PATL2 is an oocyte-specific RNA binding protein and is indispensable for maternal mRNA expression in GV oocytes. Previous studies have suggested that PATL2 might interact with CPEB1, DDX6 and EIF4E, which are typical RNA-binding proteins involved in oocyte mRNA stability in Xenopus and mice (Stebbins-Boaz et al., 1999). To confirm this, we co-expressed PATL2 with CPEB1, DDX6 and EIF4E in HeLa cells and performed a co-immunoprecipitation assay. The results showed that PATL2 co-interacted with EIF4E and CPEB1 independent of RNA (Fig. 4A,B) and that endogenous PATL2 in mouse ovaries bound to both CPEB1 and EIF4E (Fig. 4C,D). To explore the effect of the PATL2 on mRNA stability, an RNA decay assay was performed. The RNA probe was incubated with IP-enriched protein by PATL2 antibody and IgG in vitro. The results indicated that the IP-enriched protein (including CPEB1 and EIF4E) by PATL2 antibody from the mouse ovaries significantly increased the RNA half-life compared with the IgG control, suggesting that PATL2 plays a role in maintaining mRNA stability (Fig. 4E,F). These results suggested that PATL2 might couple CPEB1 and EIF4E to maintain mRNA stability.

Western blots of PATL2 at the GV, MI and MII stage showed a decreasing protein level of PATL2 when entering MI, and the electrophoresis band in MI oocytes appeared to shift to a greater molecular weight compared with GV oocytes (Fig. 5A,B). Thus, we speculated that PATL2 phosphorylation might be occurring after germinal vesicle breakdown (GVBD). To verify this hypothesis, we treated the MI oocyte lysates with λ-PPase, an Mn²⁺-dependent protein phosphatase that can remove the phosphorylation of serine, threonine, tyrosine and histidine residues. Western blot showed that λ-PPase treatment significantly decreased the protein level of PATL2 and p-ERK1/2 (a positive control) in MI oocytes compared with untreated MI oocytes (Fig. 5C-E). To further confirm the phosphorylation of PATL2 after GVBD, we compared the phosphorylation level between GV and MI oocytes using a Phos-Ser antibody. The results showed an obvious phosphorylation band in MI oocytes compared with GV oocytes (Fig. 5F). Together, these results showed that PATL2 was phosphorylated and degraded during the process of oocyte maturation. Next, we sought to determine the phosphorylation site of PATL2. We ectopically expressed PATL2 in HeLa cells and performed phosphorylation proteomics (Fig. 5G,H). We identified S279 as a phosphorylation site in PATL2 (Fig. 5I; Fig. S5; Table S4), and this amino acid is highly conserved across humans, chimps, mice, pigs, horses and dogs (Fig. 5I).

To determine the function of the phosphorylation of PATL2, we generated the S279D phosphomimetic mutant. We first confirmed that the S279D mutation had no effect on the PATL2 protein level in HeLa cells (Fig. 6A,B). We next injected WT and S279D cRNA into GV oocytes, and the western blotting assay indicated that the S279D mutation caused a sharp decrease in the PATL2 protein level in oocytes (Fig. 6C,D). In addition, PATL2 was sensitive to MG132, an inhibitor of the proteasome degradation (Fig. S6A,B), and thus we proposed that the S279D phosphomimetic mutant may affect the ubiquitin (Ub)-mediated degradation of PATL2. We found that the S279D mutation promoted the polyubiquitylation of PATL2 (Fig. 6E; Fig. S6C). All of these results indicate that phosphorylation of S279 promoted Ub-mediated degradation of PATL2.

To determine the function of the phosphorylation of PATL2 in vivo, we generated the PATL2^S279D knock-in mouse model using the CRISPR/Cas9 system (Fig. 7A). There were significantly fewer pups per litter in Patl2^S279D/S279D mice compared with Patl2^WT/S279D mice, indicating reduced fertility (Fig. 7B), whereas the male mice exhibited normal fertility. After in vitro fertilization, the zygotes from Patl2^S279D/S279D mice showed decreased rates of two cell, four- to eight-cell, morula and blastocyst stages compared with Patl2^WT/S279D mice (Fig. 7C,D). We then measured the protein level of PATL2 in the oocytes of PATL2^WT/S279D and PATL2^S279D/S279D mice. Similar to the in vitro result, the protein level of PATL2 in the oocytes of PATL2^S279D/S279D mice was sharply decreased (Fig. 7E,F), whereas the PATL2 mRNA level showed no obvious difference (Fig. 7G). Considering that the Oligo(dT) ISH showed similar weaker poly(A) probe binding in Patl2^S279D/S279D GV oocytes than in control mice, this indicated that a reduction of poly(A) mRNA may also happen (Fig. 7H,I). The EU staining showed no significant difference in transcriptional activity between Patl2^WT/S279D and Patl2^S279D/S279D mice (Fig. S7). Protein synthesis in Patl2^S279D/S279D oocytes also decreased significantly (Fig. 7J,K). These results indicated that phosphorylation of PATL2 was essential for its function in vivo.

Our work revealed the unrecognized roles of PATL2 in regulating the maternal transcriptome by recruiting CPEB1 and EIF4E in oocytes. Furthermore, we showed the essential role of PATL2 phosphorylation in oocyte maturation using Patl2^S279D/S279D knock-in mice, and we illustrated the mechanism through which PATL2 regulates its protein level via ubiquitin-mediated proteasomal degradation (Fig. 8).

The accumulation and storage of maternal mRNA is essential for oocyte maturation and embryonic development (Sha et al., 2020). To maintain the translationally silenced state, the lengths of the poly(A) tails of maternal mRNAs are shortened and the mRNAs are sequestered into cytoplasmic granules (Racki and Richter, 2006; Liu et al., 2022; Passmore and Coller, 2022). These shorter poly(A) tails present a big challenge for mRNA stability in the GV arrested stage. To maintain mRNA stability, P-bodies disappear during oocyte growth, and the factors related to mRNA decay are highly limited (Swetloff et al., 2009). For example, the decapping factors DCP1 and DCP2 and the deadenylating factor BTG4 are almost completely absent in GV stage (Ma et al., 2013; Yu et al., 2016). However, these mechanisms alone are not sufficient. In this study, we showed that PATL2 is highly expressed before GVBD and might be coupled with EIF4E and CPEB1 to regulate maternal mRNA expression in immature oocytes.

PATL2 belongs to the PAT1 family, and the PAT1 RNA-binding proteins are highly conserved from fungi to humans (Marnef et al., 2010; Marnef and Standart, 2010). Vertebrates have two paralog proteins, including PAT1A (PATL2), which is expressed in oocytes, and PAT1B (PATL1), which is expressed in embryos and somatic cells (Marnef et al., 2010). In cultured cells, PATL1 is reported to function in cytoplasmic mRNA decay by physically connecting deadenylation with decapping and by controlling the assembly of processing bodies (Ozgur et al., 2010). However, PATL2 seems to have little effect on RNA decay in cultured cells (Ozgur et al., 2010). Combined with previous genetic studies (Chen et al., 2017; Maddirevula et al., 2017; Christou-Kent et al., 2018), all these findings indicate that PATL2 might play a specific role in oocytes. A previously reported Patl2 knockout mouse model showed using Affymetrix arrays that PATL2 deletion causes the downregulation of a set of highly relevant genes involved in oocyte maturation and early embryonic development, thus leading to subfertility (Christou-Kent et al., 2018). However, the role of PATL2 in oocyte maturation has not been fully investigated. Our study used spiked-in transcriptome analysis and oligo (dT) ISH to show that PATL2 deletion caused the global decay of poly(A) mRNA in immature oocytes leading to female sterility. The level of decay increased following oocyte growth and reached its highest in GV stages, after which it slowed through the MI and MII stages. In eukaryotes, mRNA homeostasis begins with precursor messenger RNA (pre-mRNA) transcription and processing into mature mRNA in the nucleus. The mRNA is subsequently recognized and packaged by proteins into an mRNA ribonucleoprotein complex (mRNP) and exported to the cytoplasm for translation and degradation (Kang and Han, 2011). In this study, we found a reduction of total mRNA in KO oocytes. EU staining and an RNA transporting assay showed that PATL2 knockout had little effect on transcription and mRNA transporting. As PATL2 mainly localized in oocyte cytoplasm and had little signal in the nucleus, it was thought that PATL2 might function by affecting the stability of mRNA. However, an alternative possibility was that there was a difference in the amount of mRNA packaged in WT and knockouts in earlier stages of growing follicles. This demonstrated that PATL2 may be an oocyte-specific mRNA stability-maintaining protein during the oocyte growth process. The major effect of maternal mRNA in the GV stage might partly explain the GV arrests phenotype of human mutations. The phenotypic heterogeneity between humans and mice might be due to the following reasons. First, the inherent differences in fertility between humans and mice might have led the mice to exhibit a weaker phenotype. Second, among reported cases with PATL2 variants, a portion of them also caused early embryo development arrest, which might be due to the dose effect of the variants (Ma et al., 2013; Chen et al., 2017; Maddirevula et al., 2017; Christou-Kent et al., 2018; Huang et al., 2018, 2022; Liu et al., 2020). However, this needs to be further confirmed by research on oocytes in PATL2 variants.

In immature oocytes, messenger RNAs containing a cytoplasmic polyadenylation element (CPE) are translationally dormant (masked). In Xenopus, maternal mRNA is incorporated in a complex containing CPE-binding protein (CPEB), maskin and EIF4E (Stebbins-Boaz et al., 1999; Groisman et al., 2001; Minshall et al., 2007). However, TACC3, the homologous analog of maskin in mammals, appears to function as a spindle assembly protein in mouse and human oocytes rather than an RNA-binding protein (Ding et al., 2017; So et al., 2019; Wu et al., 2022). Here, we showed that PATL2 might substitute for the role of maskin as an adaptor protein to recruit EIF4E and CPEB1. PATL2 interacted with EIF4E and CPEB1 to maintain mRNA stability in immature oocytes (Fig. 8).

The translational activation of stored maternal mRNAs depends on the regulation of phosphorylation upon meiotic resumption (Stebbins-Boaz et al., 1996; Mendez et al., 2002; Setoyama et al., 2007; Sha et al., 2017). Once maturation begins, newly phosphorylated CPEB recruits the cleavage and polyadenylation specificity factor (CPSF) and poly(A) polymerase (PAP), which together function to elongate the poly(A) tail (Kang and Han, 2011; Clerici et al., 2018). Here, we showed that the RNA binding protein PATL2 is highly phosphorylated, and we showed that phosphorylation occurs at residue S279 upon meiotic resumption. In addition, we showed the essential role of PATL2 phosphorylation using Patl2^S279D/S279D knock-in mice, and we found that PATL2 regulates its protein level via ubiquitin-mediated proteasomal degradation (Fig. 8). The S279D mutant affected the protein levels in oocytes and had no effect in HeLa cells. This may be because: PATL2 is an oocyte-specific RNA binding protein and had no expression in HeLa cells, so the ectopic expression in HeLa cells might not mimic the situation in oocytes; the protein post-translational modification level (such as phosphorylation level and ubiquitylation modifications) might be different in HeLa cells and oocytes, and the HeLa cell might increase the phosphorylation level and ubiquitination modifications to clean up the ectopic expression of PATL2; or other factors specifically expressed in oocytes might also be involved in regulation of PATL2 protein level.

In this study, we confirmed the essential role of PATL2 in oocyte growth and oocyte maturation, demonstrated the effect of PATL2 on mRNA expression, and clarified the phosphorylation modification of PATL2. Our work thus sheds light on the novel mechanism of PATL2-related maternal mRNA storage and post-translational modifications in oocytes.

The full-length coding sequences of PATL2, CPEB1, EIF4E and DDX6 were amplified and cloned into the pCMV6 vector fusing with a C-terminal FLAG or GFP tag to obtain PATL2-FLAG, CPEB1-GFP, EIF4E-GFP and DDX6-GFP. The KOD-Plus Mutagenesis Kit (Toyobo) was used to introduce the S279D (AGC>GAT) mutant. The HeLa cell line was obtained from the Cell Bank of the Shanghai Institute for Biological Sciences, the Chinese Academy of Sciences (Shanghai, China). HeLa cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin (Gibco) at 37°C in an atmosphere of 5% CO₂. Cells were plated 24 h before transfection and maintained until the cell density reached 70-80% confluence. Before transfection, fresh complete culture medium containing serum and antibiotics was added to each well. Cell transfection was then performed using the PolyJet In Vitro DNA Transfection Reagent (SignaGen) according to standard protocols.

The Patl2 knockout mouse model (C57BL/6J) was generated by CRISPR/Cas-mediated genome engineering. For the Patl2 knockout mice, a region of 11 bp in exon 1 was deleted, which caused a frameshift of coding region. For the Patl2 knock-in mice, the p.S279D (AGC to GAT) variant site in the donor oligonucleotide was introduced by homology-directed repair. F0 founder animals were identified by PCR followed by sequence analysis and were bred with WT mice to get the F1 generation. WT and homozygous knockout and knock-in mice (6-8 weeks) were mated with C57BL/6J WT male mice at 10-12 weeks of age for three months. The vaginal plugs were checked to confirm successful mating. The number of litters and pups per litter were calculated. Genotyping primers used in this study are shown in Table S1.

GV oocytes were collected from ovaries of the 7- to 8-week-old female Institute of Cancer Research mice (Beijing Vital River Laboratory Animal Technology) by puncturing the antral follicles with a fine needle under the dissecting microscope, and the GV oocytes were cultured in M2 medium (Sigma-Aldrich). WT and mutant plasmids were linearized by digestion with the AgeI restriction enzyme (R0552S, New England Biolabs), and 1 μg of the linearized DNA was used as a template to transcribe cRNA with poly(A) tailing using the HiScribe T7 ARCA mRNA Kit (E2060S, New England Biolabs). cRNAs were then purified with the RNeasy MinElute Cleanup Kit (74204, Qiagen). For detailed microinjection methods refer to Sang et al. (2019).

HeLa cells ectopically expressing PATL2 were harvested at 36 h after transfection. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed, and we stained the polyacrylamide gel with Coomassie Brilliant Blue. The overexpressed band of PATL2 was cut out and collected, and phosphoproteomics was performed as previously described (Wang et al., 2017). Briefly, tandem mass tag-labeled phosphopeptides were enriched through immobilized metal affinity chromatography and consecutive titanium dioxide methods. Raw files were searched against the mouse protein sequences obtained from the Universal Protein Resource database using the MaxQuant software (version 1.3.0.5) (Cox and Mann, 2008). The Phospho (STY) site is shown in Table S7.

HeLa cells were harvested after transfection for 36 h with or without MG132 and washed with phosphate-buffered saline (PBS). They were then lysed in 1× SDS loading buffer (Shanghai Wei AO Biological Technology) and boiled for 10 min for denaturation. Mouse oocytes were denatured by boiling for 10 min in 1× SDS loading buffer. Equal amounts of protein were resolved on 10% SDS-PAGE gels and transferred to nitrocellulose filter membranes. Membranes were blocked in 5% non-fat milk for 1 h and incubated with the following primary antibodies overnight at 4°C: rabbit anti-PATL2 (Abclonal; 1:500), rabbit anti-GFP (Abclonal; 1:3000), rabbit anti-Vinculin (13901; Cell Signaling Technology; 1:1000) and α-Tubulin (Sigma-Aldrich; 1:1000). Goat anti-mouse or anti-rabbit IgG (Abmart; 1:3000) was used as the secondary antibody to detect the primary antibodies. The blots were captured using ECL Western Blotting Substrate (Tanon). Antibodies used in this study are shown in Table S2.

Oocytes were fixed with 2% paraformaldehyde (PFA; Sangon Biotech) for 30 min at room temperature and permeabilized in 0.5% Triton X-100 in PBS for 20 min. Oocytes were blocked in blocking buffer (3% bovine serum albumin, 0.1% Tween 20, and 0.01% Triton X-100 in PBS) for 1 h at room temperature, and all antibody incubations were performed in blocking buffer. Anti-PATL2 antibody (Abclonal; 1:100) was added for 1 h at 37°C to determine PATL2 localization. Hoechst 33342 (BD Biosciences; 1:1000) was used to label DNA for 15 min at room temperature. All images were captured on an LSM 880 laser scanning confocal microscopy (Zeiss).

PATL2-FLAG and mutant plasmids were co-transfected with the CPEB1-GFP, EIF4E-GFP and DDX6-GFP plasmid into HeLa cells. At 36 h after transfection, cell proteins were extracted with NP-40 lysis buffer (50 mM Tris, 150 mM NaCl, 0.5% NP-40, pH 7.5) containing 1% protease inhibitor cocktail (Bimake) and 50 μg/ml RNase A to digest the RNA. Total protein was incubated with anti-FLAG beads (Bimake) at 4°C for 3 h. The beads were washed with lysis buffer with 10 μg/ml RNase A three times and then boiled with SDS loading buffer for western blotting with anti-FLAG (1:3000) and anti-GFP (1:3000) antibodies.

The 7- to 8-week-old female mice were superovulated by injecting 10 IU of pregnant mare serum gonadotropin (PMSG, Ningbo Second Hormone Factory) followed by 10 IU of human chorionic gonadotropin (hCG, Ningbo Second Hormone Factory) 46-48 h later. After 12 h, sperm were collected from the cauda epididymis of male mice and capacitated in a drop of Human Tubal Fluid (HTF) at 37°C with 5% CO₂ for 1 h. Superovulated oocytes were transferred into a new HTF drop containing capacitated sperm for fertilization at 37°C in 5% CO₂ for 6 h. Subsequently, 2 pronuclear (2PN) zygotes were selected and transferred into KSOM medium (EasyCheck, M1430) at 37°C in 5% CO₂ for blastocyst culture.

Total RNA from mouse ovaries was extracted using a RNeasy Mini Kit (Qiagen). Genomic DNA was depleted, and RNA reverse transcription was used to obtain cDNA using a PrimeScript RT reagent Kit with gDNA Eraser (TaKaRa). qRT-PCR was performed using a Quant Studio 6 Flex Real-time system with TaKaRa TB Green Premix Ex Taq (Tli RnaseH Plus). Three repeated reactions were performed for each sample. The PCR reactions were performed as follows: 95°C for 2 min and 40 cycles of 95°C for 1 s and 60°C for 30 s. The relative expression level equals k.2^−ΔCt where ΔCt=Ct (Patl2) – Ct (Actin) and k is a constant (Pfaffl, 2001). All primers used in this study for qPCR are shown in Table S1.

The 200 nt Biotin-labeled-T7 probe was transcribed using a HiScribe T7 High Yield RNA Synthesis Kit (E2040) with a supplement of Bio-11-UTP (AM8450). The negative control (NC, free water), IgG (proteins immunoprecipitation from the ovary by IgG) and IP-PATL2 (proteins immunoprecipitation from the ovary by anti-PATL2 antibody) was adjusted to 11 µl using in vitro RNA decay buffer [50 mM HEPES-NaOH (pH 8.0), 150 mM NaCl, 2 mM MgCl₂, 10% glycerol, 1 mM DTT, and 1% protease inhibitors]. Then 1 μl (1000 ng) of T7 probe was incubated with NC, IgG or IP-PATL2 for 0 min, 10 min, 20 min and 30 min at 30°C. Reactions were stopped by the addition of 12 μl RNA loading buffer (95% formamide, 0.025% bromophenol blue, 0.025% xylene cyanol FF, 0.025% SDS and 5 mM EDTA) supplemented with 1 µl (400 U) RNA inhibitor. The mixture was then digested by 5 μg/ml protease K for 2 h at 65°C. Reaction products were heated for 3 min at 85°C and analyzed by 12% denaturing SDS-PAGE containing 7 M urea. The Biotin-labeled-RNA was transferred to nylon membranes and crosslinked using an ultraviolet crosslinker and was detected using a Chemiluminescent Biotin-labeled Nucleic Acid Detection Kit (Beyotime, D3308).

The GDF9 cRNA was transcribed using a HiScribe T7 High Yield RNA Synthesis Kit (E2060). The RNA (500 ng/μl) was injected into the nucleus of WT and KO oocytes. The oocytes were fixed by 2% PFA at 0 min, 30 min and 60 min. Then ISH was performed according to standard protocol (Jansova et al., 2021). Briefly, the Cy5-labeled oligo(dT) or specific DNA probes were generated. Oocytes were fixed with 2% PFA (Sangon Biotech) for 30 min at room temperature and permeabilized in 2% Triton X-100 in PBS for 10 min with RNase Inhibitor. Then oocytes were incubated with 100 µl hybridization buffer containing 100 nM hybridization probes overnight at 37°C. After hybridization, oocytes were washed three times by 2×SSC to remove extra probes.

GV oocytes were treated with 2 mM EU (Thermo Fisher Scientific) at 37°C with 5% CO₂ for 1 h and then fixed in 2% PFA for 1 h and permeabilized in 0.5% Triton X-100 in PBS for 20 min. Oocytes were stained using a Click-iT RNA Alexa Fluor 594 kit (Thermo Fisher Scientific) to detect nascent RNA. DNA was labeled using Hoechst 33342 for 10 min. To detect protein synthesis, oocytes were cultured in methionine-free medium (Gibco) supplemented with 1% dialyzed fetal bovine serum (10,000 MW; Sigma-Aldrich) and 50 µM HPG for 1 h. HPG was detected using a Click-iT HPG Alexa Fluor Protein Synthesis Assay Kit (Life Technologies) according to the manufacturer's instructions. All images were captured on an LSM 880 laser scanning confocal microscope (Zeiss).

GV and MII oocytes were collected from WT and Patl2^−/− female mice (7- to 8-weeks-old). Total RNA from oocytes was extracted using a RNeasy Mini Kit (Qiagen). The mRNA-GFP and mCherry were sparked into the lysed sample for RNA isolation according to a previously reported method (Yu et al., 2016). Then sequencing libraries were prepared according to the cDNA Synthesis, Amplification and Library Generation Kit (E6420, New England Biolabs). Library samples were sequenced on the Illumina Novaseq 6000 instrument by Genergy Biotechnology Co. The raw data were handled by Skewer, and data quality was checked by FastQC v0.11.2. The read length was 2×150 bp. Clean reads were aligned to the Mus musculus UCSCmm9 references using STAR and StringTie. The expression levels of the genes were quantified using fragments per kilobase of transcript per million mapped fragments (FPKM). The total mRNA copy was calculated as its FPKM normalized to the FPKM of exogenous GFP/mCherry. The FPKMs of the RNA-seq results are listed in Tables S3, S5 and S6.

Quantitation of western blotting results was performed with ImageJ software, and statistical analyses were performed using GraphPad Prism. P-values were calculated with two-tailed unpaired Student's t-test or chi-square test as described in the figure and table legends. *P<0.05; **P<0.01 and ***P<0.001 (P>0.05, not significant).

We thank Yinghua Jiang for providing help with confocal imaging, and the Public Technology Platform of Shanghai Medical College.

The peer review history is available online at https://journals.biologists.com/dev/lookup/doi/10.1242/dev.201572.reviewer-comments.pdf.