Trimethylation of histone H3 on lysine-4 (H3K4me3) is associated with gene-regulatory elements, but its transcription-independent function in cell division is unclear. CxxC-finger protein-1 (CFP1) is a major mediator of H3K4 trimethylation in mouse oocytes. Here we report that oocyte-specific knockout of Cxxc1, inhibition of CFP1 function, or abrogation of H3K4 methylation in oocytes each causes a delay of meiotic resumption as well as metaphase I arrest owing to defective spindle assembly and chromosome misalignment. These phenomena are partially attributed to insufficient phosphorylation of histone H3 at threonine-3. CDK1 triggers cell division–coupled degradation and inhibitory phosphorylation of CFP1. Preventing CFP1 degradation and phosphorylation causes CFP1 accumulation on chromosomes and impairs meiotic maturation and preimplantation embryo development. Therefore, CFP1-mediated H3K4 trimethylation provides 3a permission signal for the G2–M transition. Dual inhibition of CFP1 removes the SETD1–CFP1 complex from chromatin and ensures appropriate chromosome configuration changes during meiosis and mitosis.
Histone posttranslational modifications (PTMs)—and specific combinations they create—mediate a wide range of nuclear events. Methylation of histone H3 on lysine-4 (H3K4me) is evolutionarily conserved from yeast to humans and is widely associated with gene-regulatory elements1. Particularly, the trimethylated species (H3K4me3), which is found primarily in active promoters, is implicated in transcriptional regulation2. In contrast to lysine methylation, phosphorylation of serine and threonine residues in histone H3 fluctuates substantially throughout the cell cycle, rising sharply when cells enter mitosis3. Histone H3 phosphorylation at threonine-3 (hereafter referred to as H3T3ph) is carried out by the kinase haspin early in prophase4. The close proximity of residues T3 and K4 in histone H3 potentially enables functional and regulatory interplay between the instances of phosphorylation and methylation5,6. Nonetheless, the possible direct function of H3K4me3 in cell division has been elusive because deficiency in this histone modification in cultured cells affects transcription of a broad spectrum of genes and causes cell cycle arrest in the G1 or S phase7,8. In this respect, a fully grown mammalian oocyte is an ideal model for studies on transcription-independent functioning of histone modifications because transcription is neither active nor required for the two sequential meiotic divisions9,10.
An oocyte-specific knockout of one of the genes coding for histone H3K4 methyltransferases or demethylases leads to impaired meiotic cell cycle progression, including delayed germinal vesicle breakdown (GVBD), distorted spindle assembly, error-prone chromosome separation, and compromised polar-body emission rates11,12,13. It has been proposed that overexpression or underexpression of certain key cell cycle regulators is responsible for these meiosis-related phenotypes14,15. Nevertheless, the evidence has been indirect and weak. The transcriptomes of fully grown GV oocytes are only moderately affected by epigenetic factors because their genome is transcriptionally quiescent16. In addition, epigenetic regulation of transcription is genome-wide. It is unlikely that genes of certain cell cycle regulators are specifically targeted in growing oocytes. In contrast, we hypothesized that multiple methylation of histone H3 at lysine-4 directly drives oocyte meiotic progression by regulating chromosome behaviors.
There are six major histone methyltransferases that catalyze H3K4 methylation: SETD1A, SETD1B, MLL1, MLL2, MLL3, and MLL4, which are subdivided into three distinct groups according to their sequence similarity and complex formation: SETD1A/B, MLL1/2, and MLL3/417,18. In most cell types, SETD1-based complexes are the predominant H3K4 methyltransferases19,20. The SETD1 complex is driven to trimethylate H3K4 via its essential subunit CxxC-finger protein 1 (CFP1, encoded by the Cxxc1 gene in mammals), which engages in multivalent chromatin binding to recognize both nonmethylated DNA and existing H3K4me321,22. Zygotic knockouts of Setd1a, Setd1b, or Cxxc1 in mice each results in early embryonic mortality due to a failure in gastrulation23,24. These pleiotropic phenomena illustrate fundamental roles for the SETD1–CFP1 complex in mammalian development but, at the same time, prevent detailed studies of its functions in adult cell types, including oocytes.
Recently, we generated an oocyte-specific Cxxc1 knockout mouse strain using transgene Zp3-Cre, which is selectively expressed in growing oocytes, and demonstrated that SETD1–CFP1 is the major complex that mediates H3K4me3 accumulation in mouse oocytes16. Physiologically, maternal CFP1 and H3K4me3 are essential for the establishment of zygotic developmental competence. Meanwhile, spindle assembly was partially impaired in some CFP1-deleted oocytes that underwent meiotic maturation. Nevertheless, the oocyte transcriptome was only mildly affected in the CFP1-null oocytes; there was no sign of oocyte death and premature ovarian insufficiency in these mice before 6 months of age16. A conditional knockout of Setd1b in oocytes also caused oocyte maturation defects and female infertility, but the mechanism was investigated insufficiently25. We thus speculate that the SETD1–CFP1 complex and H3K4me3 may be involved in the regulation of meiotic cell division through a previously unidentified transcription-independent mechanism.
Given the fundamental role that the SETD1–CFP1 complex plays in depositing maternal H3K4me3 and sustaining female fertility, we investigate the transcription-independent function of CFP1 and its regulation during oocyte meiotic maturation. In contrast to our previous study in which Cxxc1 was selectively knocked out in growing oocytes by Zp3-Cre, Gdf9-Cre was employed in the current investigation to knock Cxxc1 out in oocytes as early as the primordial follicle stage, thereby giving the oocytes a longer time window to remove the existing CFP1 mRNA and protein after the gene knockout. We demonstrate that CFP1-mediated H3K4me3 accumulation in fully grown oocytes is a priming signal for meiotic cell cycle progression, particularly for T3 phosphorylation of histone H3. Moreover, we find that activity of the SETD1–CFP1 complex is dually inhibited by CDK1-triggered CFP1 phosphorylation and degradation at the onset of G2–M transition. Thus the downregulation of H3K4me3 and removal of CFP1 from chromatin are crucial for proper chromatin structure that is required for successful spindle assembly, accurate chromosome separation, and proper meiotic cell cycle progression.
Dynamic histone H3K4 trimethylation and CFP1 accumulation in cell cycles
In wild-type (WT) oocytes, chromatin-bound H3K4me3 levels increased during transition of chromatin configurations from the non-surrounded nucleolus (NSN) to surrounded nucleolus (SN) type, as determined by immunofluorescence analysis of GV chromatin spreads (Fig. 1a, b). During meiotic maturation, H3K4me3 levels gradually decreased after GVBD and reached the lowest point in anaphase I (Fig. 1c, d). In contrast, H3K4me1 and H3K4me2 levels increased during meiotic maturation (Fig. 1c, d). The meiotic maturation-coupled fluctuation of H3K4me3 levels suggested that this type of histone modification may perform a direct function in regulating meiotic cell cycle progression of oocytes.
It is noteworthy that the CFP1 protein level underwent a dramatic decrease during the G2–M transition and remained undetectable in maturing oocytes (Fig. 1c). Western blotting of oocyte samples collected at closer time points showed that CFP1 protein level quickly decreased within 1–2 h after oocytes were released from the GV arrest (Fig. 1e) and CFP1 remained undetectable in metaphase II (Fig. 1f). The CFP1 protein re-accumulated in zygotes after fertilization throughout the time period in which the fertilized eggs exited meiosis and entered interphase (Fig. 1f, g). Upon entering the first mitosis, CFP1 was degraded again (Fig. 1f, g). Furthermore, immunofluorescence results in 2–8-cell embryos revealed that CFP1 was degraded in each blastomere when it entered the second, third, and fourth mitosis (Fig. 1g).
In HeLa cells, CFP1 was detected in the nucleus of interphase cells but not in the cells that underwent mitotic divisions (Fig. 2a). In HeLa cells in which the cell cycle was arrested in metaphase by the microtubule disruptor nocodazole, the CFP1 protein was undetectable (Fig. 2b, c). When these cells were released from nocodazole treatment, CFP1 reappeared, accompanying the mitotic exit (Fig. 2b, c). Taken together, these results indicated that the CFP1 protein is transiently degraded during the meiotic and mitotic division. The observation of cell division–coupled CFP1 degradation inspired us to study the function of CFP1 and regulation of protein stability in oocytes and preimplantation embryos.
CFP1 is crucial for spindle assembly and chromosome in murine oocyte
CFP1 is expressed in the nucleus of oocytes as early as the primordial follicle stage (Fig. 3a). To study the function of CFP1 in oocyte maturation, we specifically knocked out Cxxc1 in mouse oocytes as early as the primordial follicle stage using the Gdf9-Cre. Efficient CFP1 protein depletion in oocytes was confirmed by immunohistochemistry (Fig. 3a). Cxxc1fl/fl;Gdf9-Cre female mice (hereafter referred to as Cxxc1oo–/–) are completely infertile. Nonetheless, the CFP1-null oocytes were able to develop to the fully grown GV stage (Fig. 3a and Supplementary Fig. 1a) and could be ovulated by superovulation treatment (Supplementary Fig. 1b). We collected these oocytes from oviducts and analyzed their potential defects in oocyte meiotic maturation, at 16 h after human chorionic gonadotropin (hCG) injection. Immunofluorescence and confocal microscopic results showed that the majority of ovulated CFP1-null oocytes contained distorted spindles and chromosomes were not properly aligned at the equatorial plate (Fig. 3b). To examine the meiotic maturation process more closely, we isolated fully grown GV oocytes from pregnant mare serum gonadotropin (PMSG)-primed Cxxc1oo–/– mice and cultured them in vitro. These oocytes resumed meiosis more slowly than WT oocytes did and manifested reduced GVBD and polar body 1 (PB1) emission rates (Fig. 3c). Similar to those that underwent meiotic maturation in vivo, the in vitro-matured CFP1-null oocytes failed to assemble bipolar spindles, and chromosomes were not aligned at the equatorial plates at both the MI and MII stages (Fig. 3d). We observed that 30% of CFP1-deleted oocytes released PB1s and developed to metaphase II (MII), but the spindles were distorted (Fig. 3d, e). In addition, their chromosomes were mostly aneuploid (Fig. 3f), as counted in chromosome spreads (Fig. 3g).
In Cxxc1fl/fl;Gdf9-Cre mice, CFP1 was deleted in oocytes as early as at the primordial follicle stage (Fig. 3a). Therefore, the defects of meiotic maturation in CFP1-null oocytes might have been caused by some indirect and nonspecific factors accumulated during oocyte growth. To test whether CFP1 is directly involved in oocyte meiotic maturation, we expressed a dominant-negative CFP1 mutant (CFP1C379A) in fully grown GV oocytes by mRNA microinjection. This mutant protein can bind to DNA but fails to recruit SETD1, and therefore has a dominant-negative effect26. The expression of hemagglutinin (HA)-tagged CFP1C379A at a level comparable with that of the endogenous CFP1 was detected by western blotting (Fig. 3h). In addition, western blotting and immunofluorescence results also indicated that the H3K4me3 level was significantly lower in CFP1-null and CFP1C379A-overexpressing oocytes (Fig. 3h, i). Notably, in the in vitro maturation experiment, the CFP1C379A-overexpressing oocytes had a low GVBD rate, similar to that of oocytes isolated from Cxxc1oo–/– mice (Fig. 3c, upper panel). In addition, the PB1 emission rate in CFP1C379A-overexpressing oocytes was significantly lower than that of the control oocytes (Fig. 3c, lower panel). Furthermore, overexpression of CFP1C379A impaired spindle formation at both stages MI and MII (Fig. 3d, e). Overall, overexpression of a dominant-negative CFP1 mutant in fully grown oocytes mimicked the phenotype of the oocyte-specific Cxxc1 knockout. Considering the fact that transcriptional activity is absent in fully grown oocytes, our results indicated that CFP1 has a transcription-independent role in the regulation of oocyte meiotic maturation.
To obtain more direct evidence, we ectopically expressed the H3.3K4R mutant protein in GV oocytes by mRNA microinjection. According to other studies, H3.3 is the major H3 variant that undergoes K4 trimethylation in mouse oocytes27,28. This H3.3K4R mutant deficient for methylation was expressed and efficiently incorporated into chromatin (Supplementary Fig. 2a). As predicted, the chromatin-bound H3K4me3 amount was low in oocytes expressing Flag-H3.3K4R but not in those expressing exogenous Flag-H3.3WT (Supplementary Fig. 2a). Furthermore, the H3.3K4R-expressing oocytes had meiosis defects resembling to those of CFP1-deleted oocytes: GVBD and PB1 emission rates decreased (Fig. 3j); and spindle assembly and equatorial chromosome alignment failed (Fig. 3d, bottom panels; and Fig. 3j). These results indicated that methylation of histone H3 at K4 is a prerequisite for successful meiotic maturation.
CFP1 deletion in oocytes caused meiotic cell cycle arrest in metaphase I
Employing live cell imaging microscopy, we next compared the dynamic spindle assembly and meiotic cell cycle progression between WT and CFP1-deleted oocytes. GV-arrested oocytes were microinjected with mRNA encoding green fluorescent protein (GFP)-tagged α-tubulin (to label spindles) and were cultured in a medium containing Hoechst-33342 (to label DNA). Oocytes that underwent GVBD within 4 h after being released from milrinone were subjected to live cell imaging. Results in Fig. 4a reveal that WT oocytes finished meiosis I and developed to MII at the end of cultivation. In contrast, bipolar spindle formation, equatorial chromosome alignment, and PB1 emission failed in Cxxc1oo−/− oocytes.
To assess the kinetics of metaphase-to-anaphase transition, we microinjected mRNA encoding mCherry-securin into WT and Cxxc1oo−/− oocytes and monitored the dynamics of mCherry fluorescence. As shown in Fig. 4a and quantified in Fig. 4b, securin was degraded in WT oocytes approximately 10–12 h after meiotic resumption but remained stable in CFP1-deleted oocytes up to 18 h after meiotic resumption, suggesting that the CFP1-deleted oocytes failed to enter anaphase I.
At the end of in vitro maturation culture, oocytes were collected for preparing chromosome spreads. Immunofluorescence results in Fig. 4c show that all WT oocytes under study (n = 26) contained 20 pairs of coherent sister chromatids, and the cohesion core subunit SMC3 was detected only at the kinetochore (labeled with CREST antibody). On the other hand, most Cxxc1oo−/− oocytes (21 out of 32) contained unseparated homologous chromosomes with remarkable SMC3 distribution on chromosome arms. Thus, in CFP1-null oocytes, the anaphase-promoting complex, which removes securin, and the separase, which in turn removes cohesin, were not activated, and the meiotic cell cycle was arrested at pre-MI.
CFP1-dependent H3K4me3 accumulation is a prerequisite for H3T3 phosphorylation
Phosphorylation of histone H3 at threonine-3 during the G2–M transition is required for oocyte meiotic maturation3. T3 of histone H3 is promptly phosphorylated during the G2–M transition. High H3T3ph levels were maintained throughout meiosis I (Fig. 1c, d). Blocking this K4-adjacent H3 modification causes similar defects of oocyte maturation as we observed in Cxxc1oo−/− oocytes3. Therefore, we hypothesized that CFP1-dependent H3K4 trimethylation is a priming signal for H3T3 phosphorylation. Consistent with this hypothesis, the H3T3ph level was significantly compromised in Cxxc1-null oocytes at 6 h after in vitro culture (Fig. 5a). Similarly, overexpression of the dominant-negative CFP1 (CFP1C379A) also blocked H3T3 phosphorylation on chromosomes (Fig. 5b). In contrast, phosphorylation of histone H3 at Ser-10 (H3S10ph), another cell division–related histone H3 modification, was not affected in Cxxc1oo−/− oocytes (Supplementary Fig. 2b, c). This result indicated that the role of CFP1 in regulating H3T3ph is specific.
More directly, when the oocytes overexpressing H3.3K4R were allowed to undergo meiotic maturation, total and chromosomal H3T3ph levels also decreased (Fig. 5c–e). The H3T3ph western blot and immunofluorescence results in Fig. 5b, c, e represent only endogenous protein because the anti-H3T3ph antibody does not recognize any K4-mutation-bearing H3 forms, due to the close proximity of these mutations to the T3ph epitope. A co-immunofluorescence assay of H3T3ph and H3K4me3 in MI oocytic chromosome spreads (using a rabbit- and a mouse-derived antibody, respectively) indicated that these two types of histone H3 modifications did not overlap strongly (Fig. 5f). Rather, H3T3ph preferentially accumulated at the interaction interface between homologous chromosomes, whereas H3K4me3 tended to be located at the surface of paired chromosomes (Fig. 5f). Collectively, these results strongly indicated that methylated H3K4 regulates T3 phosphorylation in an intermolecular manner.
Histone H3T3 phosphorylation partially reverses the defects in CFP1-null oocytes
The evidence so far indicated that CFP1-mediated H3K4 trimethylation is upstream of H3T3 phosphorylation in the control over meiotic progression. Therefore, we determined whether a histone H3.3 mutant that mimics T3 phosphorylation (H3.3T3E) can partially reverse the meiosis defects in CFP1-null oocytes. This histone mutant was expressed in GV oocytes of WT and Cxxc1oo−/− mice by mRNA microinjection. After being released from the GV arrest, the H3.3T3E expressing CFP1-null oocytes has a GVBD rate close to that of normal oocytes (Fig. 5g). Although the PB1 emission rate was not rescued by H3.3T3E, the spindles were assembled with better morphology than in CFP1-null oocytes expressing H3.3T3E (Fig. 5g, h). These results suggested that an important function of CFP1-mediated H3K4 trimethylation is to facilitate H3T3 phosphorylation and that these hierarchical histone modifications are crucial for meiotic cell cycle progression.
CDK1 triggers CFP1 phosphorylation and degradation
The results so far indicated that CFP1-dependent H3K4me3 accumulation in fully grown GV oocytes is a priming signal for proper meiotic cell cycle progression. In contrast, we also observed prompt CFP1 degradation during the G2–M transition in both oocytes and early blastomeres. Therefore, we further investigated the physiological importance and biochemical mechanism of CFP1 removal during meiotic and mitotic divisions.
To easily monitor CFP1 degradation in oocytes, we constructed a plasmid expressing a CFP1-GFP-HA fusion protein. After in vitro transcription (IVT) and microinjection of the mRNA into GV oocytes, the CFP1-GFP-HA fusion protein was abundantly expressed. Just as endogenous CFP1, this fusion protein was degraded within 3 h of meiotic resumption (Fig. 6a). Furthermore, its degradation was prevented by addition of the 26S proteasome inhibitor MG132 into the culture medium, suggesting that CFP1 was degraded via the canonical ubiquitin–proteasome pathway (Fig. 6b). After a series of domain-mapping analyses, we found that deleting the C-terminal 65 amino acids (aa) stabilized CFP1 in maturing oocytes (Fig. 6c, d). In HeLa cells transfected with a CFP1-expressing plasmid, deletion of the C-terminal 65 aa significantly inhibited the polyubiquitination of CFP1 (Fig. 6e) and prolonged half-life of the CFP1 protein when protein synthesis was inhibited by cycloheximide treatment (Fig. 6f). Moreover, we noticed an upshift of the CFP1 band on a gel (after sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)), specifically in MI oocytes, when CFP1 degradation was blocked by MG132 treatment (Fig. 6b) or C-terminal deletion (Fig. 6d). This band upshift was abrogated when the protein sample was pretreated with protein phosphatase (Fig. 6g), indicating that the upshift was caused by CFP1 phosphorylation.
Based on this observation, we attempted to identify the protein kinase that mediates meiotic maturation–coupled CFP1 phosphorylation and degradation. In a co-immunoprecipitation (co-IP) experiment, CFP1ΔC interacted with CDK1, the major kinase that triggers the G2–M transition in the cell cycle (Fig. 6h). We generated constitutively active CDK1, in which the inhibitory phosphorylation sites (Thr-14 and Tyr-15) were mutated to alanine and phenylalanine, respectively. This CDK1 mutant (CDK114A,15F) bound to CFP1ΔC as well (Fig. 6h). When we coexpressed CDK114A;15F and CFP1ΔC in GV oocytes by mRNA microinjection, CDK114A;15F triggered premature degradation of endogenous CFP1 and phosphorylation of CFP1ΔC (Fig. 6i). On the other hand, specific inhibition of haspin and aurora A, two kinases that also mediate the G2–M transition but downstream to CDK1, did not prevent the GVBD-associated CFP1 phosphorylation (Supplementary Fig. 3a, b). Taken together, these results strongly indicated that CDK1 directly interacts with CFP1 and phosphorylates it at the onset of the G2–M transition.
The CFP1 protein contains multiple consensus CDK1 phosphorylation sites (SP and TP) that are conserved among vertebrates (Fig. 6c). By a series of point mutagenesis analyses, we found that mutation of two SP sites (Ser-138 and Ser-143) close to, but not within, the CXXC finger domain of CFP1 completely blocked CFP1 phosphorylation in maturing oocytes (Fig. 6j). Nevertheless, the full-length CFP1 protein that bears these two point mutations (CFP12SA) was degraded normally after meiotic resumption (Supplementary Fig. 3c). This result suggests that, although CDK1 triggers both phosphorylation and degradation of CFP1 protein, phosphorylation of CFP1 by CDK1 is not a prerequisite for its degradation during cell division. Therefore, CDK1 may trigger CFP1 degradation through some indirect mechanisms rather than CFP1 phosphorylation. The biochemical and cellular function of CFP1 phosphorylation was further addressed below.
Preventing CFP1 degradation and phosphorylation impairs oocyte maturation
To assess the cellular importance of CFP1 degradation, we microinjected mRNAs encoding WT and nondegradable CFP1 (CFP1ΔC and CFP1ΔC;2SA) into GV oocytes and examined the process of their meiotic maturation. In addition, to test whether the potential function of the stabilized CFP1 depends on its DNA-binding ability, we generated a CFP1 mutation in which a key cysteine residue required for DNA binding (C173)29 was mutated to alanine in the CFP1ΔC background (CFP1ΔC;C173A). In GV oocytes, both CFP1WT and CFP1ΔC were relatively ubiquitously distributed in the nucleus except for the nucleolus (Fig. 7a). CFP1ΔC;C173A was also located in the nucleus but formed multiple foci that did not overlap with DNA (Fig. 7a). When the microinjected oocytes developed to the MI stage, CFP1WT was degraded and became undetectable (Fig. 7b). In contrast, CFP1ΔC and CFP1ΔC;2SA were stably expressed and enriched on chromosomes, particularly in the kinetochores (Fig. 7b). CFP1ΔC;C173A was also stabilized but existed as large cytoplasmic particles without co-localization with chromosomes (Fig. 7b).
Ectopic expression of CFP1WT in GV oocytes did not affect meiotic maturation (Fig. 8a, b). This finding is conceivable because CFP1 was quickly degraded during the G2–M transition. On the other hand, both GVBD and PB1 emission rates decreased in oocytes overexpressing stabilized CFP1ΔC (Fig. 8b). Confocal microscopy showed that spindle assembly was severely affected in these oocytes (Fig. 8a, middle panels, and Fig. 8b). Moreover, these phenotypes are remarkably strengthened in oocytes overexpressing CFP1ΔC;2SA (Fig. 8a, bottom panels, and 8b). Although CFP1ΔC;C173A was not degraded, it did not cause any abnormality during meiotic maturation (Fig. 8a, b), indicating that only the DNA-binding CFP1 is detrimental to meiotic progression of the oocyte.
We further examined the chromosome spreads made from MI oocytes. While the ectopically expressed CFP1WT was degraded and not detected on chromosomes, CFP1ΔC was present on the chromosome arms but, unexpectedly, was more concentrated in centromeres (Fig. 8c). Furthermore, the chromosomal H3K4me3 amount was high in oocytes expressing the degradation-resistant CFP1ΔC (Fig. 8d, e). CFP1ΔC;2SA overexpression not only caused a more significant increase in the H3K4me3 level than CFP1ΔC did (Fig. 8d, e) but also impaired the condensation and organization of chromosomes: poorly condensed chromosomes formed a clump in oocytes and single chromosomes were difficult to distinguish (Fig. 8a bottom panels; Figs. 8d, f; 7b, bottom panels). In contrast, CFP1ΔC;C173A was neither present on chromosomes (Fig. 8c) nor did it affect the H3K4me3 level in oocytes (Fig. 8d, e).
Phosphorylation of CFP1 decreases its chromatin-binding ability
Western blotting analysis of whole-oocyte lysates confirmed that the H3K4me3 level increased at the MI stage in CFP1ΔC-expressing but not in CFP1ΔC;C173A-expressing oocytes (Fig. 9a, b). Although CFP1ΔC was not degraded, it did not completely prevent the decrease of total H3K4me3 levels in MI oocytes (Fig. 9a–c). On the other hand, overexpression of phosphorylation site–mutated CFP1ΔC (CFP1ΔC;2SA) was more effective than CFP1ΔC at maintaining a high H3K4me3 level in MI oocytes (Fig. 9b, c). Moreover, we generated phosphorylation-mimicking mutations in CFP1ΔC (Ser-138 and Ser-143 were mutated to aspartic acid, CFP1ΔC;2SD). When CFP1ΔC;2SD was overexpressed in GV oocytes, it had a dominant-negative effect by decreasing the H3K4me3 level (Fig. 9d). Taken together, these results indicated that phosphorylation of CFP1 at Ser-138 and Ser-143 inhibits its ability to mediate H3K4 trimethylation, and this phenomenon is independent of CFP1 degradation.
We investigated the mechanism by which CFP1 phosphorylation inhibits H3K4me3 accumulation. In co-IP experiments, mutations of CFP1 phosphorylation sites did not affect CFP1 binding to SETD1 (Supplementary Fig. 3d) but reduced the interaction between CFP1 and histone H3 (Fig. 9e). We ectopically expressed CFP1WT or CFP12SD in GV oocytes at comparable levels (Fig. 9f). CFP1WT showed an uneven distribution pattern in the nucleus and was more concentrated in the chromatin area (Fig. 9g). On the other hand, CFP12SD was evenly distributed in the entire nucleus excluding the nucleolus. In addition, H3K4me3 levels slightly increased in oocytes expressing CFP1WT but remarkably decreased in oocytes expressing CFP12SD (Fig. 9g, h). This observation is in agreement with the western blot results in Fig. 9d. Furthermore, we prepared chromatin spreads from these oocytes. With this procedure, free proteins and proteins loosely bound to chromatin in the nucleus were removed. Only proteins tightly bound to chromatin remained. Immunofluorescence results showed that CFP1WT was detected in GV chromatin spreads, but the signal of CFP12SD was much weaker in comparison to CFP1WT (Fig. 9l, j).
The identified CFP1 phosphorylation sites are adjacent to the CXXC domain, which mediates DNA binding (Fig. 6c). Therefore, the potential influence of CFP1 phosphorylation on DNA-binding ability was evaluated by chromatin immunoprecipitation (ChIP) assay in HeLa cells transfected with plasmid expressing HA-CFP1 (WT, 2SD, and the non-DNA binding mutant C173A as a negative control). The binding of CFP1 with its reported target gene promoter regions22,30 was evaluated. For the four genes detected, CFP1WT and CFP12SD had comparable binding ability with the genomic DNA, whereas the same genomic DNA fragments were not co-precipitated with CFP1C173A (Supplementary Fig. 3e). These results confirmed that phosphorylation inhibited CFP1 by impairing its histone-binding, but not DNA-binding, ability.
CFP1 inhibition are crucial for embryo development
We found that CFP1 is degraded not only in dividing mouse oocytes but also in blastomeres and somatic cells in metaphase of the cell cycle. To evaluate the physiological importance of CFP1 inhibition for mitosis, we overexpressed WT or mutated CFP1 in mouse zygotes by mRNA microinjection and monitored the preimplantation developmental process for the next 72 h. While overexpression of CFP1WT in the zygote did not significantly affect embryo development to the 8-cell stage, the constitutively expressing CFP1 (CFP1ΔC;2SA) blocked embryo development beyond a 2-cell stage (Fig. 10a, b; arrows and hollow arrows indicated embryos arrested at 1- and 2-cell stages, respectively). In comparison, overexpression of CFP1ΔC or CFP12SA caused intermediate phenotypes: although CFP1ΔC decreased embryonic developmental rates to a lesser extent than CFP1ΔC;2SA did, CFP12SA impaired the development from the 1- to 2-cell stage as significantly as CFP1ΔC;2SA did but was less effective in the next two cell cycles (Fig. 10a, b). This is conceivable because the CFP12SA protein was degraded during the first mitotic division. Overall, these results indicated that dual inhibition of CFP1 by phosphorylation and degradation is not only important in oocytes but also crucial for successful preimplantation embryo development.
Although H3K4me3 is an evolutionarily conserved and common histone modification, its transcription-independent role in cell division has not been investigated before this study. Several research groups have demonstrated that the amount of H3K4me3 remarkably increases during oocyte maturation, and H3K4me3 is abundantly deposited at numerous broad regions on chromatin of mature oocytes31,32,33, but the functional significance of these phenomena remain unclear. It is well recognized that oocytes with an SN chromatin configuration have a better developmental potential than those with an NSN configuration34. This notion is in agreement with our finding that the H3K4me3 level in SN oocytes is higher than that in NSN oocytes. Therefore, H3K4me3 is a reliable marker of oocytes with high maturation rates and a post-fertilization developmental potential. Our results suggest that reinforcing H3K4me3 accumulation in human oocytes may be a promising approach to optimizing oocyte quality during human-assisted reproduction.
It is worth noticing that the adjacent threonine residue (T3) of H3K4 is modified by phosphorylation during cell division35. Mitotic and meiotic cell cycle–coupled phosphorylation of H3T3 by haspin kinase performs a pivotal function in spindle assembly and chromosome alignment4,36,37. The H3T3ph mark recruits survivin (a subunit of the chromosomal passenger complex) to chromatin and is essential for proper chromosome segregation and for modulation of aurora B kinase activity38. The haspin inhibitor 5-ITu was reported to impair spindle assembly and chromosome separation in mouse oocytes3. In our experiment, CFP1 deletion in mouse oocytes led to phenotypes similar to those in oocytes with inhibited haspin, including compromised GVBD and PB1 emission rates, defective spindle assembly, and failure of homologous chromosome separation. Evidently, meiotic maturation–coupled H3T3 phosphorylation was significantly decreased in Cxxc1oo−/− oocytes. According to these observations, we hypothesized that CFP1-mediated H3K4 trimethylation is a priming signal for H3T3 phosphorylation and facilitates meiotic resumption (Fig. 10c). This previously unrecognized scheme may ensure proper spindle assembly and chromosome organization during cell division.
During further testing of our hypothesis, ectopic expression of a H3.3 mutant (H3.3T3E; which mimics H3T3 phosphorylation) partially rescued GVBD and spindle assembly, but not PB1 emission rates, in Cxxc1oo−/− oocytes. The rescue effect was not very efficient possibly because it is difficult to control the expression level of H3.3T3E in oocytes, and abnormally low and high H3T3ph levels are all detrimental to meiotic spindle assembly4. Besides, H3.3T3E may not effectively mimic the precise chromosomal localization and function of H3T3ph.
Interplays between H3T3 phosphorylation and H3K4 methylation have been investigated before by biochemical studies, but contradictory results were obtained. Protein structural analyses reveals that the bulkiness of K4 methylation compromises H3 tail interactions with the narrow substrate-binding groove of haspin5. Correspondingly, in vitro kinase assay results indicated that H3K4 methylation impaired H3T3 phosphorylation by haspin5. This finding is consistent with the observation that H3T3ph was detected adjacent to, but not overlap with H3K4me3, on chromosome spreads of oocytes (Supplemental Fig. S3D of this study). In return, in vitro and in vivo results derived from nuclear magnetic resonance chemical shift perturbation analysis, fluorescence binding assays, pull-downs, peptide microarrays, and cell fluorescence microscopic studies revealed a strong anti-correlation between histone H3T3 phosphorylation and the ability of PHD finger-containing proteins (including CFP1) to recognize H3K4me3 or to retain in chromatin during mitosis6. Contradictorily, co-occurrence of H3T3ph/K4me, H3K9me/S10ph, and other phospho/methyl modifications has also been experimentally detected, and combinatorial PTM-specific antibody studies reveal that K4me3 and T3ph can be present on the same histone H3 tail39,40.
Our results revealed that H3K4 methylation affects H3T3 phosphorylation in an intermolecular manner because the ectopically expressed H3K4 mutant (H3.3K4R) decreased T3 phosphorylation of endogenous histone H3. Technically, we could not distinguish the intermolecular effects of monomethylation, dimethylation, and trimethylation of K4 on T3 phosphorylation in this study because all these modifications were abrogated in the H3.3K4R mutant. Nonetheless, only trimethylation, but not the other modifications of H3K4, was significantly repressed in Cxxc1oo−/− oocytes. Therefore, we concluded that H3K4me3, rather than H3K4me1 or H3K4me2, triggers H3T3 phosphorylation. In our previous study, a decreased H3K4me3 level in CFP1-null oocytes caused tightening of the chromatin structure and reduced the rate of histone H3 exchange16. As a result, the T3 sites in histone H3 on the chromatin of Cxxc1oo−/− oocytes may not be accessible to haspin. Although our results indicate that defective T3 phosphorylation is an important reason for meiotic maturation failure of Cxxc1oo−/− oocytes, it is not the only abnormality that matters here. Insufficient accumulation of maternal mRNAs and aberrant distribution of cytoplasmic organelles caused by CFP1 deletion may also indirectly influence meiotic cell cycle progression of oocytes16.
In this study, we observed that CFP1 is phosphorylated and degraded during the G2–M transition in both meiosis and mitosis. Chromatin undergoes extensive reorganization throughout the cell cycle, transitioning from a relatively relaxed state in interphase to a highly condensed state in mitosis and returning back to the interphase state after cell division is completed27. The degradation of CFP1 is a robust way to remove the SETD1–CFP1 complex from chromatin and therefore facilitates chromosome condensation. Otherwise, chromosome-bound CFP1 proteins, particularly those in the centromeric region, may become an obstacle for chromosome organization and microtubule attachment during spindle assembly. The finding of complete degradation of CFP1 during each cell division reveals a surprising fact that all SETD1–CFP1 H3K4 methyltransferase complexes need to be de novo assembled in the cell after each division, and the distribution pattern of this complex on chromatin needs to be re-established in every cell cycle. This knowledge is conceptually important for understanding the genome-wide regulation of H3K4 methylation dynamics in general. For instance, this concept implies that quickly proliferating cells tend to have a more dynamic H3K4me3 distribution pattern, whereas quiescent non-proliferating cells have relatively stable SETD1–CFP1 and H3K4me3 localization on chromatin.
Although our data indicate that CDK1 is involved in CFP1 degradation and phosphorylation, which ubiquitin E3 ligase directly binds to CFP1 and mediates its polyubiquitination is currently unclear. It is conceivable that this ubiquitin E3 ligase is activated by CDK1 during the G2–M transition and may directly interact with the C terminus of CFP1. Nor do we know the signal that recruits nondegradable CFP1 to the centromere region.
CDK1 is a multifunctional kinase and is essential for oocyte meiotic resumption41,42. Therefore, we could not test whether CDK1 inhibitors, such as roscovitine, prevent CFP1 phosphorylation and degradation, because the inhibitors would block oocyte GVBD as well. We cannot completely rule out the possibility that CDK1 activation leads to CFP1 degradation and phosphorylation indirectly as consequences of the G2–M transition. Nonetheless, our experiments showed that (1) CDK1 directly interacted with CFP1; (2) mutations at two putative CDK1 phosphorylation sites (S138 and S143) abrogated meiotic division–coupled CFP1 phosphorylation; and (3) inhibition of other G2–M transition–related kinases, such as haspin and aurora A43, did not prevent CFP1 phosphorylation. Together, these results strongly suggest that CDK1 directly phosphorylates CFP1.
Although CFP1 phosphorylation is not a prerequisite for CFP1 degradation, the dual inhibition of CFP1 is physiologically important. Stabilization of CFP1 in a maturing oocyte by C-terminal deletion did not completely prevent the reduction in H3K4me3 levels. Only when both CFP1 phosphorylation and degradation were abrogated, was H3K4me3 maintained at high levels in maturing oocytes, and strong defects in oocyte meiotic maturation were observed. The CFP1 phosphorylation sites are conserved among vertebrates and are close to the CxxC finger domain. Mutations that mimic phosphorylation (CFP12SD) did not impair the SETD1–CFP1 interaction but weakened the chromatin binding of CFP1. In agreement with this observation, CFP12SD has a dominant-negative effect on H3K4 trimethylation in GV oocytes. Therefore, phosphorylation of CFP1 provides a mechanism for SETD1–CFP1 regulation and may be involved in fine-tuning of genomic histone modification processes. CFP1 is also degraded in mitosis, and constitutively active CFP1 blocks preimplantation murine embryo development. These facts suggest that the dual inhibition of the SETD1–CFP1 complex is a widespread mechanism in cell cycle regulation that is not restricted to oocytes. Furthermore, unidentified extracellular signals, ubiquitin E3 ligases, or kinases other than CDK1 may regulate activity of the SETD1–CFP1 complex via this dual inhibition mechanism in interphase cells in a time- and location-specific manner.
Three-to-four-week-old C57BL/6 mice maintained under specific pathogen-free conditions were used for all experiments. Cxxc1fl/fl;Gdf9-Cre mice were produced by crossing mice bearing the previously reported Cxxc1fl allele with Gdf9-Cre transgenic mice8,44. Animal care and experimental procedures were in accordance with the Animal Research Committee guidelines of Zhejiang University.
Female mice were primed with PMSG (5 IU) and full-grown oocytes were collected 44 h later by ovarian puncture. Healthy looking fully grown oocytes were collected from antral follicles and cultured in M16 medium (M7292; Sigma-Aldrich) covered with mineral oil at 37 °C in a 5% CO2 atmosphere45.
Plasmid construction and IVT
cDNAs encoding human histone H 3.3 and CFP1 were subcloned into a Flag-tagged and a HA-tagged expression plasmid, respectively, and used for IVT. IVT was performed using the T7 and SP6 mMESSAGE mMACHINE Kit (Invitrogen, AM1344 and AM1340) according to the manufacturer’s instructions46.
Oocytes were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) at room temperature for 30 min. They were then permeabilized with 0.3% Triton X-100 in PBS for 15 min. Antibody staining was performed using standard protocols described previously. Antibodies used in the experiments are described in Supplementary Table 1. Imaging was performed on a Zeiss LSM710 confocal microscope. Semiquantitative analysis of the fluorescence signals was conducted using the NIH Image program Image-J, as previously described27.
Zona pellucida-free oocytes were fixed in a solution containing 1% paraformaldehyde, 0.15% Triton X-100, and 3 mmol/L dithiothreitol (Sigma-Aldrich) on glass slides for 30 min and air dried. Immunofluorescent staining was performed as in oocytes described above.
Microinjection of mRNAs and small interfering RNAs
All microinjections were performed using an Eppendorf transferman NK2 micromanipulator. Fully grown GV oocytes were harvested in M2 medium with 2 μM milrinone to inhibit spontaneous GVBD. Approximately 5 pl of 200 μg/ml mRNAs was microinjected into the ooplasm.
Live cell imaging
For live imaging, mRNAs encoding for GFP-tubulin and mCherry-securin were microinjected into WT and CFP1-deleted oocytes and released from milrinone after 8 h. Oocytes were cultured in M16 medium containing Hoechst-33342 (to label DNA) and subjected to live cell imaging at 5 h after meiotic resumption. Live oocytes were acquired on a DV ELITE High Resolution Invented Living Cell Work station. Image acquisition was performed using Zeiss LSM-780 confocal microscope (Zeiss) equipped with PC-Apochromat 20×/0.8 NA objective lenses at 6 min intervals for 13 h.
Oocytes were lysed in protein-loading buffer and heated at 95 °C for 5 min. SDS-PAGE and immunoblots were performed following the standard procedures using a Mini-PROTEAN Tetra Cell System (Bio-Rad, Hercules, CA). The antibodies used are listed in Supplementary Table 1. Uncropped scans of the most important blots are supplied as Supplementary Figures 4, 5.
HeLa cells were from the American Type Culture Collection and were grown in Dulbecco’s modified Eagle’s medium (Invitrogen) supplemented with 10% fetal bovine serum (Hyclone) and 1% penicillin–streptomycin solution (Gibco) at 37 °C in a humidified 5% CO2 incubator. Cells were in healthy condition but were not tested for mycoplasma contamination.
The immunoprecipitation (ChIP) assay in HeLa cells was performed using the SimpleChIP Enzymatic Chromatin IP Kit (CST, 9003) according to the manufacturer’s protocol. Approximately 4 × 106 cells were used for each immunoprecipitation. Immunoprecipitation of immunoglobulin G (CST#2729; 1 µg, as a negative control) or HA (CST#3724; 1 µg per sample) was performed at 48 h after transfection. In all, 10% total DNA was used for input evaluation. PCR products of immunoprecipitated and input samples were analyzed on a 2% agarose gel. Sequences of primers for PCR amplification are provided in Supplementary Table 2.
Two-tailed Student’s t test was used to calculate P values. Statistically significant values for <0.05, <0.01, and <0.001 are indicated by single, double, and triple asterisk, respectively.
The authors declare that all data supporting the findings of this study are available within the article and its supplementary information files or from the corresponding author upon reasonable request.
Howe, F. S., Fischl, H., Murray, S. C. & Mellor, J. Is H3K4me3 instructive for transcription activation? Bioessays 39, 1–12 (2017).
Clouaire, T., Webb, S. & Bird, A. Cfp1 is required for gene expression-dependent H3K4 trimethylation and H3K9 acetylation in embryonic stem cells. Genome Biol. 15, 451 (2014).
Wang, Q. et al. H3 Thr3 phosphorylation is crucial for meiotic resumption and anaphase onset in oocyte meiosis. Cell Cycle 15, 213–224 (2016).
Balboula, A. Z. et al. Haspin kinase regulates microtubule-organizing center clustering and stability through Aurora kinase C in mouse oocytes. J. Cell Sci. 129, 3648–3660 (2016).
Eswaran, J. et al. Structure and functional characterization of the atypical human kinase haspin. Proc. Natl Acad. Sci. USA 106, 20198–20203 (2009).
Gatchalian, J. et al. Chromatin condensation and recruitment of PHD finger proteins to histone H3K4me3 are mutually exclusive. Nucleic Acids Res. 44, 6102–6112 (2016).
Clouaire, T. et al. Cfp1 integrates both CpG content and gene activity for accurate H3K4me3 deposition in embryonic stem cells. Genes Dev. 26, 1714–1728 (2012).
Cao, W. et al. CXXC finger protein 1 is critical for T-cell intrathymic development through regulating H3K4 trimethylation. Nat. Commun. 7, 11687 (2016).
De La Fuente, R. Chromatin modifications in the germinal vesicle (GV) of mammalian oocytes. Dev. Biol. 292, 1–12 (2006).
Lin, C. J., Koh, F. M., Wong, P., Conti, M. & Ramalho-Santos, M. Hira-mediated H3.3 incorporation is required for DNA replication and ribosomal RNA transcription in the mouse zygote. Dev. Cell 30, 268–279 (2014).
Ancelin, K. et al. Maternal LSD1/KDM1A is an essential regulator of chromatin and transcription landscapes during zygotic genome activation. Elife 5, e08851 (2016).
Andreu-Vieyra, C. V. et al. MLL2 is required in oocytes for bulk histone 3 lysine 4 trimethylation and transcriptional silencing. PLoS Biol. 8, e1000453 (2010).
Kim, J. et al. LSD1 is essential for oocyte meiotic progression by regulating CDC25B expression in mice. Nat. Commun. 6, 10116 (2015).
Eymery, A., Liu, Z., Ozonov, E. A., Stadler, M. B. & Peters, A. H. The methyltransferase Setdb1 is essential for meiosis and mitosis in mouse oocytes and early embryos. Development 143, 2767–2779 (2016).
Kim, J. et al. Maternal Setdb1 is required for meiotic progression and preimplantation development in mouse. PLoS Genet. 12, e1005970 (2016).
Yu, C. et al. CFP1 regulates histone H3K4 trimethylation and developmental potential in mouse oocytes. Cell Rep. 20, 1161–1172 (2017).
Long, H. K., Blackledge, N. P. & Klose, R. J. ZF-CxxC domain-containing proteins, CpG islands and the chromatin connection. Biochem. Soc. Trans. 41, 727–740 (2013).
Takahashi, Y. H. et al. Structural analysis of the core COMPASS family of histone H3K4 methylases from yeast to human. Proc. Natl Acad. Sci. USA 108, 20526–20531 (2011).
Ardehali, M. B. et al. Drosophila Set1 is the major histone H3 lysine 4 trimethyltransferase with role in transcription. EMBO J. 30, 2817–2828 (2011).
Shilatifard, A. The COMPASS family of histone H3K4 methylases: mechanisms of regulation in development and disease pathogenesis. Annu. Rev. Biochem. 81, 65–95 (2012).
Lee, J. H. & Skalnik, D. G. CpG-binding protein (CXXC finger protein 1) is a component of the mammalian Set1 histone H3-Lys4 methyltransferase complex, the analogue of the yeast Set1/COMPASS complex. J. Biol. Chem. 280, 41725–41731 (2005).
Brown, D. A. et al. The SET1 complex selects actively transcribed target genes via multivalent interaction with CpG island chromatin. Cell Rep. 20, 2313–2327 (2017).
Bledau, A. S. et al. The H3K4 methyltransferase Setd1a is first required at the epiblast stage, whereas Setd1b becomes essential after gastrulation. Development 141, 1022–1035 (2014).
Carlone, D. L. & Skalnik, D. G. CpG binding protein is crucial for early embryonic development. Mol. Cell. Biol. 21, 7601–7606 (2001).
Brici, D. et al. Setd1b, encoding a histone 3 lysine 4 methyltransferase, is a maternal effect gene required for the oogenic gene expression program. Development 144, 2606–2617 (2017).
Butler, J. S., Lee, J. H. & Skalnik, D. G. CFP1 interacts with DNMT1 independently of association with the Setd1 Histone H3K4 methyltransferase complexes. DNA Cell Biol. 27, 533–543 (2008).
Akiyama, T., Suzuki, O., Matsuda, J. & Aoki, F. Dynamic replacement of histone H3 variants reprograms epigenetic marks in early mouse embryos. PLoS Genet. 7, e1002279 (2011).
Banaszynski, L. A., Allis, C. D. & Lewis, P. W. Histone variants in metazoan development. Dev. Cell 19, 662–674 (2010).
Lee, J. H., Voo, K. S. & Skalnik, D. G. Identification and characterization of the DNA binding domain of CpG-binding protein. J. Biol. Chem. 276, 44669–44676 (2001).
Thomson, J. P. et al. CpG islands influence chromatin structure via the CpG-binding protein Cfp1. Nature 464, 1082–1086 (2010).
Zhang, B. et al. Allelic reprogramming of the histone modification H3K4me3 in early mammalian development. Nature 537, 553–557 (2016).
Dahl, J. A. et al. Broad histone H3K4me3 domains in mouse oocytes modulate maternal-to-zygotic transition. Nature 537, 548–552 (2016).
Liu, X. et al. Distinct features of H3K4me3 and H3K27me3 chromatin domains in pre-implantation embryos. Nature 537, 558–562 (2016).
Tan, J. H. et al. Chromatin configurations in the germinal vesicle of mammalian oocytes. Mol. Hum. Reprod. 15, 1–9 (2009).
Kang, H., Park, Y. S., Cho, D. H., Kim, J. S. & Oh, J. S. Dynamics of histone H3 phosphorylation at threonine 3 during meiotic maturation in mouse oocytes. Biochem. Biophys. Res. Commun. 458, 280–286 (2015).
Nguyen, A. L. et al. Phosphorylation of threonine 3 on histone H3 by haspin kinase is required for meiosis I in mouse oocytes. J. Cell Sci. 127, 5066–5078 (2014).
Kelly, A. E. et al. Survivin reads phosphorylated histone H3 threonine 3 to activate the mitotic kinase Aurora B. Science 330, 235–239 (2010).
Wang, F. et al. Haspin inhibitors reveal centromeric functions of Aurora B in chromosome segregation. J. Cell Biol. 199, 251–268 (2012).
Garcia, B. A. et al. Modifications of human histone H3 variants during mitosis. Biochemistry 44, 13202–13213 (2005).
Markaki, Y., Christogianni, A., Politou, A. S. & Georgatos, S. D. Phosphorylation of histone H3 at Thr3 is part of a combinatorial pattern that marks and configures mitotic chromatin. J. Cell Sci. 122, 2809–2819 (2009).
Adhikari, D. & Liu, K. The regulation of maturation promoting factor during prophase I arrest and meiotic entry in mammalian oocytes. Mol. Cell Endocrinol. 382, 480–487 (2014).
Adhikari, D. et al. Inhibitory phosphorylation of Cdk1 mediates prolonged prophase I arrest in female germ cells and is essential for female reproductive lifespan. Cell Res. 26, 1212–1225 (2016).
Solc, P., Schultz, R. M. & Motlik, J. Prophase I arrest and progression to metaphase I in mouse oocytes: comparison of resumption of meiosis and recovery from G2-arrest in somatic cells Mol. Hum. Reprod. 16, 654–664 (2010).
Lan, Z. J., Xu, X. & Cooney, A. J. Differential oocyte-specific expression of Cre recombinase activity in GDF-9-iCre, Zp3cre, and Msx2Cre transgenic mice. Biol. Reprod. 71, 1469–1474 (2004).
Yu, C., Ji, S. Y., Sha, Q. Q., Sun, Q. Y. & Fan, H. Y. CRL4-DCAF1 ubiquitin E3 ligase directs protein phosphatase 2A degradation to control oocyte meiotic maturation. Nat. Commun. 6, 8017 (2015).
Yu, C. et al. BTG4 is a meiotic cell cycle-coupled maternal-zygotic-transition licensing factor in oocytes. Nat. Struct. Mol. Biol. 23, 387–394 (2016).
This study is funded by the National Key Research and Developmental Program of China (2017YFC1001500, 2016YFC1000600) and National Natural Science Foundation of China (31528016, 31371449, 31671558). We thank Dr. Wei Ma for sharing reagents related to H3T3 phosphorylation.
The authors declare no competing interests.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Sha, QQ., Dai, XX., Jiang, JC. et al. CFP1 coordinates histone H3 lysine-4 trimethylation and meiotic cell cycle progression in mouse oocytes. Nat Commun 9, 3477 (2018). https://doi.org/10.1038/s41467-018-05930-x
Dynamic mRNA degradome analyses indicate a role of histone H3K4 trimethylation in association with meiosis-coupled mRNA decay in oocyte aging
Nature Communications (2022)
Expression of the histone lysine methyltransferases SETD1B, SETDB1, SETD2, and CFP1 exhibits significant changes in the oocytes and granulosa cells of aged mouse ovaries
Histochemistry and Cell Biology (2022)
CFP1-dependent histone H3K4 trimethylation in murine oocytes facilitates ovarian follicle recruitment and ovulation in a cell-nonautonomous manner
Cellular and Molecular Life Sciences (2020)
The CRL4-DCAF13 ubiquitin E3 ligase supports oocyte meiotic resumption by targeting PTEN degradation
Cellular and Molecular Life Sciences (2020)
Cancer Cell International (2019)