Abstract
Zygotic genome activation (ZGA) is a critical developmental milestone characterized by the rapid and simultaneous activation of genome-widely silenced chromatin. Various active histone modifications accumulate upon ZGA and have long been implicated in ZGA. However, their biological relevance still remains unclear. Here, we comprehensively examined the functional significance of active histone modifications and their writers during ZGA in teleost embryos. Our data propose that developmental genes and housekeeping genes are distinctively regulated during ZGA; CBP/P300 activity is required for developmental gene activation, whereas housekeeping genes depend on non-CBP/P300 histone acetylations H3K9ac/H4K16ac/H3K14ac. Moreover, accumulation of H3K4me2/3 is not prerequisite for activation of all types of genes during ZGA, in contrast to previous reports with cell lines. Finally, temporal accumulation of H3.3S31ph greatly enhances CBP/P300 activity specifically at ZGA, ensuring the activation of developmental genes. Our data demonstrate that multiple histone modifications cooperatively shape ZGA-specific gene activation programs in non-mammalian vertebrates.
Similar content being viewed by others
Introduction
In non-mammalian vertebrates, embryos undergo rapid cleavages after fertilization1. During this period, development proceeds with maternally provided factors without zygotic transcription, and most histone modifications undergo erasure1. After about the tenth round of cleavages at the blastula stage, pioneer factors initiate genome-wide transcription from the zygotic genome within a very narrow time window1. This transition is referred to as zygotic genome activation (ZGA) and is essential for subsequent gastrulation and cellular differentiation. Histone modifications are mostly absent, or their accumulation is limited during cleavage stages, but re-accumulate in the zygotic genome during ZGA2,3,4,5,6,7. In mammals, however, the complete erasure of histone modifications is rare, but instead, the non-canonical pattern of histone modifications is established upon ZGA8.
One of the most intensively studied histone modifications during ZGA is H3K27ac whose accumulation is correlated to active gene promoters and enhancers9. H3K27ac accumulates during ZGA in zebrafish2,3,10, and the pattern of H3K27ac is drastically remodeled during ZGA in mice and human11,12. Previous studies also revealed the importance of H3K27ac and its writer CBP/P300 during ZGA in vertebrates. The inhibition of CBP/P300 activity impaired ZGA in mouse, zebrafish, and medaka (Oryzias latipes, Japanese killifish)3,4,10,11,13,14. On the other hand, the overexpression of CBP/P300 caused premature, but partial, ZGA in zebrafish10. However, how CBP/P300 and its mediated H3K27 acetylation regulate ZGA is not fully understood even in those species. Furthermore, other active modifications, such as H3K9ac, H4K16ac and H3K4me2/3, also accumulate during ZGA in non-mammalian vertebrates (zebrafish, medaka and Xenopus)2,3,4,5,6,7,15. Moreover, it was previously proposed that such accumulation of active marks primes genes to be activated during ZGA in non-mammalian vertebrates6,7. In spite of many previous studies, their necessity for ZGA has not been examined yet. Taken together, it is far from a complete understanding of the biological relevance of active histone modifications accumulated during ZGA.
In non-mammalian vertebrates, chromatin is inaccessible, canonical nucleosome patterning is absent, and Pol II does not accumulate on chromatin until ZGA2,5,10,16,17,18,19. Furthermore, most of active histone modifications are almost absent or their enrichment is limited before ZGA in non-mammalian vertebrates3,4,6,7. Nevertheless, their embryos must activate such tightly silenced genome within a very narrow window, i.e. almost simultaneously and rapidly, unlike relatively long-lasting ZGA in mammals1. These features are in striking contrast to differentiated cells in which some portions of chromatin are already activated or primed. Despite the importance of ZGA, it is not known how such transcriptional barriers are resolved during ZGA in non-mammalian vertebrates.
In this study, we examined the biological relevance of active histone modifications accumulated during ZGA in fish, medaka (Oryzias latipes) and zebrafish (Danio rerio). We revealed that developmental genes and housekeeping genes are distinctively regulated during ZGA; CBP/P300 activity is required for activation of developmental genes, but not for that of housekeeping genes during ZGA. CBP/P300 was found to exert its effect via regulation of Pol II activity along with BRD4, a reader of histone acetylations. In contrast, our data suggest that non-CBP/P300 histone acetyltransferases (HATs) are responsible for activation of housekeeping genes during ZGA. Furthermore, unlike previous studies with cell lines20,21,22, we found that accumulation of H3K4me2/3 is dispensable for proper ZGA in fish embryos. Finally, we observed that CBP/P300 activity is enhanced specifically during ZGA by the temporal accumulation of H3.3S31ph, as compared to that of other active mark writers, ensuring activation of developmental genes during ZGA. Our data reveal the coordinated functions of each active modification to achieve the ZGA-specific unique transcriptional program, providing insights into the biological relevance of histone marks for transcription during development.
Results
CBP/P300 activity is essential for activation of developmental genes but dispensable for activation of housekeeping genes during ZGA
Inhibition of CBP/P300 impairs zygotic genome activation and results in developmental arrest at the blastula stage in both medaka and zebrafish3,4,10. However, what kind of genes are targets of CBP/P300 and by what mechanism CBP/P300 regulates activation of those genes remained unknown. We first sought to identify CBP/P300-dependent and -independent genes in medaka. In medaka, ZGA occurs from the tenth round cell divisions (the early blastula stage, ~1000 cells, 6.5 hpf)19(Supplementary Fig. 1A). Blastomeres are known to be pluripotent until the late blastula stage ( ~ 4000 cells, 8.5 hpf)23 and cell differentiation begins from the pre-early gastrula stage onwards (6-7000 cells, 10.5 hpf). We attempted to reduce the activity of CBP/P300 by incubating embryos with chemicals, A48524, an inhibitor to HAT activity, and SGC-CBP3025, an inhibitor to Bromo-domain activity (Fig. 1A, B). Each treatment efficiently reduced the amount of H3K27ac in embryos at the late-blastula stage, resulting in development arrest before gastrulation (Fig. 1C–G). We also confirmed that A485 treatment did not affect the chromatin binding of CBP/P300 by preforming fractionation assay (Supplementary Fig. 1B, C). After the validation of RNA-seq reproducibility (Supplementary Fig. 1D), we found that both A485 and SGC-CBP30 treatment caused downregulation (Fig. 1H, I) of those that would otherwise be activated at ZGA (ZGA genes, Supplementary Fig. 1E, see “Methods”), including both “Maternal-ZGA” and “Pure ZGA” genes (Supplementary Fig. 1F–H, see “Methods”). Importantly, the downregulated genes are almost completely overlapped between the two experiments (hereafter referred to as “CBP/P300-dependent”) (Fig. 1J). Moreover, the principal component analysis (PCA) of the RNA-seq data revealed that samples from the CBP/P300 inhibitions are distinct from that of normal embryos and resemble each other (Supplementary Fig. 1I), suggesting that both inhibitors similarly affected the ZGA. However, almost half of the ZGA genes were not affected by inhibition of CBP/P300 activity (referred to as “CBP/P300-independent”) (Fig. 1J). Our conventional RNA-seq in this study was unable to distinguish between maternally provided and newly transcribed mRNAs, but the comparison with our previous data of nascent/zygotic mRNA expression in medaka19 indicated that ZGA genes examined in this study covered a major fraction of highly-expressed nascent/zygotic transcripts (Supplementary Fig. 1J-L). Moreover, our re-analysis of the previous nascent RNA-seq dataset of zebrafish embryos with or without SGC-CBP30 treatment10 identified many zygotic genes which are insensitive to CBP/P300 inhibition (Supplementary Fig. 1M, N). Thus, we conclude that the identification of CBP/P300-independent genes did not arise from the technical limitation of our study.
A Schematic showing the mechanism of CBP/P300 inhibitors. B Schematic showing CBP/P300 inhibitor treatment in medaka. C Phenotypes of embryos at 1 dpf. Magenta dashed lines indicate the location of cells. DMSO-treated embryos formed embryonic bodies normally, while others did not. D Representative Western blot of GAPDH and H3K27ac in DMSO (DM) or A485-treated medaka embryos. Linearity was tested using titrated samples from the wild type (WT) embryos. Same number of embryos (1×) were loaded to lanes. E Western blot of GAPDH and H3K27ac in SGC-CBP30-treated embryos. Linearity was tested using DMSO (DM)-treated embryos. Same number of embryos (1×) were loaded to lanes. F Quantification of (D). Each dot indicates the intensity of H3K27ac band normalized to GAPDH intensity. Two-sided Welch’s t test against DMSO. n = 3 biological replicates. Source data are provided as a Source Data file. G Quantification of (E). Each dot indicates the intensity of H3K27ac band normalized to GAPDH intensity. Two-sided unpaired Student’s t test. n = 3 biological replicates. Source data are provided as a Source Data file. H Scatter plots indicate RNA-seq expression levels in DMSO and 20 µM A485-treated embryos. Red and blue dots indicate significantly upregulated and downregulated genes ( | log2 fold change | > 1, adjusted p value < 0.01, RPKM > 1.5) in A485-treaed embryos, respectively. Pink, green, and light blue dots indicate Maternal, Maternal-ZGA, and Pure ZGA genes (see Supplementary Fig. 1F–H), respectively. n = 2 biological replicates. I Scatter plots indicate RNA-seq expression levels in DMSO and SGC-CBP30-treated embryos. Red and blue dots indicate significantly upregulated and downregulated genes ( | log2 fold change | > 1, adjusted p value < 0.01, RPKM > 1.5) in SGC-CBP30-treated embryos, respectively. Pink, green, and light blue dots indicate Maternal, Maternal-ZGA, and Pure ZGA genes (see Supplementary Fig. 1F–H), respectively. n = 2 biological replicates. J Venn diagram indicates overlap of ZGA genes (see Supplementary Fig. 1E) and significantly downregulated genes in A485 or SGC-CBP30-treated embryos. K Bubble plot indicates the results of GO analysis (FDR < 0.05, fold enrichment >2, top 10 GO terms). L Heatmap showing expression level of representative developmental genes. RNA-seq data of normal embryos19 and DMSO- or A485-treated embryos in this study are shown. M Heatmap showing RNA-seq transcript enrichment around TSS (bottom) and its average (top). *: p < 0.05, **: p < 0.01, ***: p < 0.001, NS: not significant. dpf /hpf: days/hours post fertilization. Error bars indicate the mean ± s.d.
To reveal what kind of genes are included in “CBP/P300-dependent”, we investigated the characters of CBP/P300-dependent genes in detail. Gene ontology (GO) analysis revealed that GO terms related to development/differentiation are enriched in CBP/P300-dependent genes, while those related to recombination and nucleosomes, which are essential for cell maintenance, are enriched in CBP/P300-independent genes (Fig. 1K). Indeed, following representative developmental genes are involved in CBP/P300-dependent genes in medaka (Fig. 1L). This raised the possibility that the activation of developmental genes is CBP/P300-dependent, while the activation of housekeeping genes is CBP/P300-independent during ZGA in medaka. Previous studies revealed that developmental genes use a single transcription start site (TSS) (sharp promoter) while housekeeping genes uses multiple TSSs (broad promoter) in mammals26 (Supplementary Fig. 1O). Consistent with this notion, aggregate plots of transcripts around TSS revealed that the boundary between untranscribed regions and transcribed regions around TSS of CBP/P300-dependent genes is relatively sharp compared to that of CBP/P300-independent genes (Fig. 1M), further supporting the above tendency. Moreover, our re-analysis of the previous dataset10 confirmed that developmental genes are sensitive to CBP/P300 inhibition also in zebrafish (Supplementary Fig. 1P). Taken together, these data suggest that CBP/P300 activity is indispensable for the activation of developmental genes, but dispensable for the activation of housekeeping genes during ZGA in fish.
CBP/P300 activity is required for proper establishment of epigenetic landscape during ZGA
We previously found that A485 treatment attenuated H3K4me2 accumulation and increased H3K27me3 deposition at some promoters4. Therefore, loss of active modifications or gain of repressive modifications could be the cause of downregulation after CBP/P300 inhibition. To test this possibility, we first compared the changes in histone modifications and zygotic transcription. Hereafter, we focused on CBP/P300-dependent genes with proximal H3K27ac peaks (TSS ± 3 kb), because downregulation of CBP/P300-dependent genes without H3K27ac proximal peaks is likely to be a secondary effect, i.e. downstream of direct-CBP/P300-target genes (Supplementary Fig. 2A, B). Re-analysis of our previous ChIP-seq data4 revealed that H3K27ac accumulates to both CBP/P300-dependent and -independent gene promoters, and A485 treatment decreased H3K27ac there (Fig. 2A–C). We also found the loss of H3K4me2 specifically in CBP/P300-dependent genes (Fig. 2A-F) and the premature gain of H3K27me3 at a fraction of CBP/P300-dependent genes at the ZGA (Fig. 2A–C). In the latter fraction of genes, H3K27me3 would normally accumulate at the onset of gastrulation (Supplementary Fig. 2C). Furthermore, we newly performed spike-in ChIP-seq of H2AK119ub1, another histone modifications related to H3K27me3-mediated repression27. After the validation of reproducibility (Supplementary Fig. 2D), we noticed that H2AK119ub deposition was attenuated after CBP/P300 inhibition in both CBP/P300-dependent and -independent genes (Fig. 2A–C). The expression level of chromatin modifiers related to H3K4 methylations, H3K27me3 and H2AK119ub was comparable between control and A485-treated embryos (Supplementary Fig. 2E). These data indicate that CBP/P300-activity is required for proper establishment of epigenetic landscape during the ZGA in medaka.
A Track view of RNA-seq and spike-in ChIP-seq after CBP/P300 inhibitor treatment. Data from the previous study4 are shown with *. B Heatmap showing histone modification enrichment (spike-in ChIP-seq) around H3K27ac proximal peaks after CBP/P300 inhibitor treatment. C Quantification of (B). Boxplots indicate average of spike-in normalized enrichment of each histone modification in proximal H3K27ac peaks. Values above boxes indicate p values. Two-sided Wilcoxon rank-sum test. n = 703, 398 for CBP/P300-dependent, CBP/P300-independent, respectively. D Heatmap showing correlations between expression changes (RNA-seq) and changes in histone modification enrichment (spike-in ChIP-seq) after CBP/P300 inhibitor treatment. E Heatmap showing fold change of gene expression level and H3K4me2 enrichment in H3K27ac proximal peaks after inhibitor treatments. F Boxplot indicates the fold change of H3K4me2 enrichment in H3K27ac proximal peak after A485 treatment. Two-sided Wilcoxon rank-sum test. n = 703, 398 for CBP/P300-dependent, CBP/P300-independent, respectively. G Track view of RNA-seq and spike-in ChIP-seq after α-amanitin injection. Data from the previous study4 are shown with *. H Heatmap showing histone modification enrichment (spike-in ChIP-seq) around H3K27ac proximal peaks after α-amanitin injection. I Quantification of (H). Boxplots indicate average of spike-in normalized enrichment of each histone modification in H3K27ac proximal peaks. Values above boxes indicate p values. Two-sided Wilcoxon rank-sum test. n = 703, 398 for CBP/P300-dependent, CBP/P300-independent, respectively. J Heatmap showing fold change of gene expression level and histone modification enrichment in H3K27ac proximal peaks after α-amanitin injection. Box plots indicate median (middle line), 25th, 75th percentile (box) and 5th and 95th percentile (whiskers), and outliers are not shown. NS: not significant.
Genome-wide analysis showed that loss of H3K4me2 best correlates with transcriptional changes after CBP/P300 inhibition (Fig. 2D–F), raising the possibility that loss of H3K4me2 after CBP/P300 inhibition is a primary cause of the downregulation of CBP/P300-dependent genes. Consistently, H3K4 methylations start to accumulate upon ZGA in non-mammalian vertebrates3,4,6,7. However, accumulation of active histone modifications could be a consequence, but not a cause of transcription during ZGA5,28. To distinguish between these possibilities, we investigated the pattern of histone modifications after inhibition of ZGA by injecting α-amanitin. After we confirmed that our new spike-in ChIP-seq well recapitulated the previous data (Supplementary Fig. 2F), we found that, although α-amanitin injection impaired all transcription19,29, i.e. both CBP/P300-dependent and -independent genes, the enrichment of H3K27ac and H3K4me2 was not attenuated (Fig. 2G–J). Taken together, these data demonstrate that CBP/P300 activity, but not the zygotic transcription itself, is required for the accumulation of H3K4me2 in developmental genes during ZGA.
Accumulation of H3K4me2/3 is not prerequisite for ZGA
To experimentally examine the hypothesis that CBP/P300 activates CBP/P300-dependent genes via H3K4me2 deposition, we sought to assess the necessity of H3K4me2 accumulation for ZGA. We attempted H3K4me2 depletion by injecting a human histone demethylase for H3K4me2/3, KDM5B28,30, or by treatment with MM-10231, an inhibitor of WDR5-MLL1 interaction which is required for activity of H3K4me2/3 writer complex (Fig. 3A). We first confirmed that both KDM5B injection and MM-102 treatment attenuated global H3K4me2 levels at the late blastula stage (Fig. 3B–D, Supplementary Fig. 3A). KDM5B injection also decreased global H3K4me3 levels (Supplementary Fig. 3B, C). However, RNA-seq at the late blastula stage revealed that KDM5B or MM-102 did not affect ZGA at all (Fig. 3E, F, Supplementary Fig. 3D). Consistent with this, KDM5B injection did not affect early development in medaka (Supplementary Fig. 3E). These data indicate that accumulation of H3K4me2/3 is not prerequisite for gene expression during ZGA. Therefore, we concluded that the accumulation of H3K4me2/3 and ZGA are both CBP/P300-dependent, but independent of each other.
A Schematic showing the depletion of H3K4me2/3. B Western blot of GAPDH and H3K4me2 in KDM5B-injected or MM-102-treated medaka embryos. Linearity was tested using control embryos. Same number of embryos (1×) were loaded to lanes. Quantification is shown in Supplementary Fig. 3A. C Representative immunofluorescence staining of FLAG and H3K4me2 using Control or KDM5B-injected medaka embryos. D Quantification of (C). Each dot indicates the average intensity of cells in a single broad field slice image of single embryo. Horizontal bars indicate the mean of dots. Two-sided Wilcoxon rank-sum test. **: p < 0.01. n = 10, 11, 10, and 14 embryos pooled from two biological replicates. Error bars indicate the mean ± s.d. Source data are provided as a Source Data file. E Scatter plots indicate RNA-seq expression levels in control and H3K4me2-depleted embryos. Red and blue dots indicate significantly upregulated and downregulated genes ( | log2 fold change | > 1, adjusted p value < 0.01, EPKM > 1.5) in H3K4me2-depleted embryos, respectively. n = 2 biological replicates. F PCA analysis of RNA-seq data from normal development19, A485 treatment, and H3K4me2 depletion. Each dot indicates one of the replicates. The dashed arrow indicates normal developmental process, and the data from late blastula, control of inhibitor treatment, and H3K4me2 depletion are clustered together at the arrowhead. CBP/P300 inhibition data are also included for comparison.
CBP/P300 and BRD4 induce Pol II initiation and elongation, respectively, during ZGA
Previous studies in cell lines demonstrated that CBP/P300 regulate Pol II initiation and elongation32,33. Therefore, we next hypothesized that CBP/P300 activity modulates Pol II activity at developmental genes during ZGA in medaka. Pol II spike-in ChIP-seq using A485-treated late blastula embryos showed that inhibition of CBP/P300 activity resulted in loss of Pol II occupancy in both TSSs and gene bodies of CBP/P300-dependent genes (Fig. 4A–C, Supplementary Fig. 4A–D), suggesting that CBP/P300 activity is required for Pol II initiation. Since BRD4, a reader of CBP/P300-dependent histone acetylations, is known to regulate pause-release of Pol II34,35 (Fig. 4D), we examined if this is the case during ZGA. For this, we inhibited the bromodomain activity of BRD4 by the chemical inhibitor JQ136 (Fig. 4D). RNA-seq of JQ1-treated embryos revealed that BRD4 inhibition impaired the activation of CBP/P300-dependent genes during ZGA, as did CBP/P300 inhibition (Fig. 4E, F, Supplementary Fig. 1C), suggesting that BRD4 activity is involved in CBP/P300-mediated gene activation during ZGA. Furthermore, Pol II ChIP-seq using JQ1-treated embryos revealed that the accumulation of Pol II was attenuated in gene bodies but not in TSSs of CBP/P300-dependent genes after JQ1 treatment (Fig. 4A–C, Supplementary Fig. 4B–D), suggesting that BRD4 activity is mainly required for Pol II elongation. It is known that Pol II phosphorylation at serine 2 (Ser2P) is associated with Pol II elongation. Therefore, we performed Pol II Ser2P ChIP-seq using DMSO, A485, or JQ1-treated late blastula embryos, and found that Ser2P was attenuated at the gene bodies of CBP/P300-dependent gene in A485 and JQ1-treated embryos (Supplementary Fig. 4E). Taken together, we concluded that CBP/P300 and BRD4 are both required for the activation of developmental genes, but function differently in Pol II initiation and elongation, respectively.
A Track view of RNA-seq and spike-in ChIP-seq after CBP/P300 and BRD4 inhibitor treatment. Data from the previous study4 are shown with *. B Heatmap showing Pol II enrichment (spike-in ChIP-seq) around gene bodies after treatment with CBP/P300 and BRD4 inhibitors. C Line plot showing the average of (B). D Schematic showing the function of BRD4 and mechanism of JQ1. E Scatter plots indicate RNA-seq expression levels in control and JQ1-treated embryos. Red and blue dots indicate significantly upregulated and downregulated genes ( | log2 fold change | > 1, adjusted p value < 0.01, RPKM > 1.5) in JQ1-treated embryos, respectively. Pink, green, and light blue dots indicate Maternal, Maternal-ZGA, and Pure ZGA genes (see Supplementary Fig. 1F–H), respectively. n = 2 biological replicates. F Venn diagram indicates overlap of ZGA genes (see Supplementary Fig. 1E), CBP/P300-dependent genes (Fig. 1J) and significantly downregulated genes in JQ1-treated embryos.
It is also possible that CBP/P300 increases chromatin accessibility, thereby activating CBP/P300-dependent genes during ZGA. However, re-analysis of ATAC-seq using DMSO or A485-treated late blastula embryos4 indicates that global loss of H3K27ac only modestly affected the gain of chromatin accessibility upon ZGA including promoter of CBP/P300-dependent genes (Supplementary Fig. 4F, G), which is consistent with the previous results during ZGA in Drosophila embryos37. Therefore, it is less likely that the transcriptional changes of CBP/P300-dependent genes were caused by an alteration in chromatin accessibility.
CBP/P300-independent histone acetylation accumulates at CBP/P300-independent ZGA genes
In addition to the above CBP/P300-dependent genes, A485 and JQ1 treatment also attenuated Pol II occupancy in CBP/P300-independent genes (Fig. 4C). However, the extent of reduction was much milder than that in CBP/P300-dependent genes (Fig. 4C, Supplementary Fig. 4B–D). Probably due to this, the expression of CBP/P300-independent genes was largely insensitive to A485 and JQ1 treatment (Fig. 4F). Furthermore, transcription of CBP/P300-independent genes was insensitive to the inhibition of CBP/P300 activity (Fig. 1H–J), even though CBP/P300 targets those gene promoters (Fig. 2B). Therefore, we speculated that other mechanisms are involved in the activation of CBP/P300-independent genes, i.e. housekeeping genes. To test this idea, we related gene expression changes after CBP/P300 inhibition with promoter enrichment of various histone modifications in normal late blastula embryos. We found that H3K9ac, H4K16ac, and H3K14ac are more highly enriched in the promoters of CBP/P300-independent gene in normal embryos (Fig. 5A–E). H3K9ac, H4K16ac, and H3K14ac are known to be deposited by non-CBP/P300 histone acetyltransferases38,39. Among the writers of those histone acetylation, GCN5, MOF, and HBO1 (encoded by kat2a, kat8, and kat7b in medaka, respectively) are expressed in medaka early development (Fig. 5F, G), raising the possibility that they are involved in the activation of CBP/P300-independent genes during ZGA in medaka.
A Heatmap showing correlations between expression changes (RNA-seq) after CBP/P300 inhibitor treatment and histone modification enrichment levels (ChIP-seq) in normal medaka embryos. Data from the previous study4 are shown with *. B Track view of RNA-seq after CBP/P300 inhibitor treatment and ChIP-seq in normal medaka embryos. C Heatmap showing histone modification enrichment (ChIP-seq) around H3K27ac proximal peaks in normal medaka embryos. D Line plots indicate average enrichment of each histone modification around H3K27ac proximal peaks in normal medaka embryos. E Quantification of (C, D). Boxplots indicate average of histone modification enrichment in H3K27ac proximal peaks in normal medaka embryos. Two-sided Wilcoxon rank-sum test. Box plots indicate median (middle line), 25th, 75th percentile (box) and 5th and 95th percentile (whiskers), and outliers are not shown. n = 703, 398 for CBP/P300-dependent, CBP/P300-independent, respectively. F Schematic showing HATs and histone acetylations. G Heatmap showing expression levels (RNA-seq)75 of genes encoding HATs in medaka. H Heatmap showing enrichment of transcription factor motifs that were specifically enriched (log2 fold change of p value > 1, log2 fold change of enrichment >0.1) in CBP/P300-dependent genes or -independent genes and the expression level of genes encoding the transcription factor in medaka embryos.
Although not clearly mentioned in previous papers, the substantial number of genes were CBP/P300-independent during ZGA in zebrafish and mouse embryos3,10,11,13. We thus examined the correlation between CBP/P300 dependency and the presence of H3K9ac and H4K16ac in zebrafish and mouse by re-analysis of the previous datasets10,11,40,41. We found in zebrafish that H3K9ac and H4K16ac were accumulated in the promoters of CBP/P300-independent genes, but not in those of CBP/P300-dependent genes (Supplementary Fig. 5A, B). This correlation, however, was never observed in the case of mouse (Supplementary Fig. 5C–F). Intriguingly, GO analysis in mouse revealed that, unlike in medaka (Fig. 1K) and zebrafish (Supplementary Fig. 1N), CBP/P300-dependent genes in mouse do not include developmental genes, although there is a clear separation of CBP/P300-dependent and -independent genes (Supplementary Fig. 5G). Furthermore, representative developmental genes which were CBP/P300-dependent ZGA genes in medaka (Fig. 1L) were not zygotically expressed during ZGA in mouse (Supplementary Fig. 5H). These data suggest that the GCN5/PCAF and MOF are involved in activation of CBP/P300-independent genes during teleost ZGA, but the alternative mechanism compensates for the activation of CBP/P300-independent genes during mammalian ZGA.
It was reported in zebrafish that pioneer transcription factors (TFs) recruit HATs to chromatin during ZGA40. Furthermore, inhibition of CBP/P300 HAT activity did not affect the binding of CBP/P300 to chromatin in medaka (Supplementary Fig. 1B, C). Therefore, we hypothesized that TFs recruiting CBP/P300 to CBP/P300-dependent genes are different from those recruiting GCN5/PCAF and MOF to CBP/P300-independent genes in teleosts. We performed the motif enrichment analysis and identified some motifs specifically enriched in each promoter of CBP/P300-dependent and -independent genes in medaka; PRMD9, HOXB13, SOX15 for CBP/P300-dependent-specific, and Twist2, Tcf12, Olig2, Neurod2 for CBP/P300-independent-specific (Fig. 5H). This result is consistent with our model in which CBP/P300-dependent and -independent genes are differentially activated by HATs recruited by each specific group of TFs during ZGA.
CBP/P300 activity is exceptionally enhanced during ZGA in teleosts
The finding that H3K27ac is indispensable, whereas H3K4me2/3 is dispensable, for ZGA in medaka led us to further investigate the regulation of active histone modifications during development in medaka. For this purpose, we first quantitatively compared the global enrichment of active histone modifications per cell using ZGA stage embryos (the late blastula stage), embryos after gastrulation (the late neurula, 6-somite, and the 24-somite stages) and the differentiated cell line (the fibroblast cell line19) in medaka by quantitative Western blot29 (Fig. 6A, Supplementary Fig. 6A). In this method, the number of cells in each sample was first estimated by measuring relative DNA copy number, and then the same number of cells were loaded to each lane, enabling comparison for the enrichment of histone modification per cell among different stages (Fig. 6A). We found that H3K27ac accumulation was 2–3-fold higher during ZGA as compared to other stages (Fig. 6B, C, Supplementary Fig. 6B). This ZGA-specific accumulation was also observed in the case of H3K18ac, another CBP/P300-dependent histone acetylation38,39 (Fig. 6B, C). By contrast, H3K9ac and H4K16ac, which are known to be induced by other HATs38,39 (Fig. 5F), did not show such ZGA-specific strong accumulation (Fig. 6B, C). Furthermore, active histone methylations, H3K4me2 and H3K4me3, did not accumulate specifically at ZGA, either (Supplementary Fig. 6C, D). These ZGA-specific strong accumulation was also observed by immunofluorescence staining (Supplementary Fig. 6E, F). In zebrafish, active histone modifications also accumulate during ZGA2,3,6,7. To expand our finding, we further compared H3K27ac enrichment at the sphere stage (4 hpf, around ZGA) and at the bud stage (gastrulation, 10 hpf) in zebrafish (Supplementary Fig. 6G), and found that the enrichment of H3K27ac at the sphere stage was 2–3 higher than that at the bud stage (Supplementary Fig. 6H, I). Taken together, these data suggest that the activity of CBP/P300 is enhanced, resulting in a strong enrichment of CBP/P300-dependent histone acetylation during ZGA.
A Developmental stages of medaka. B Representative Western blot of H3K27ac, H3K18ac, H3K9ac, and H4K16ac. Linearity was tested using titrated fibroblast samples. Same number of cells (1×) were loaded to lanes. C Quantification of (B). Each dot indicates the intensity of band. Two-sided unpaired Student’s t test. n = 4, 4, 4, and 5 biological replicates for H3K27ac, H3K18ac, H3K9ac and H4K16ac Western blot, respectively. Source data are provided as a Source Data file. D Schematics showing H3.3S31ph deposition and enhancement of CBP/P300 HAT activity. E Representative Western blot of H3.3S31ph. Linearity was tested using titrated late blastula samples. Same number of cells (1×) were loaded to lanes. F Quantification of (E). Each dot indicates the intensity of band. Two-sided Welch’s t test. n = 3 biological replicates. Source data are provided as a Source Data file. G Heatmap showing H3.3S31ph enrichment (ChIP-seq) around H3K27ac proximal peaks in normal medaka embryos. Average plot is shown in Supplementary Fig. 5N. H Schematic showing chek1 overexpression (top) and the results of Representative Western blot (bottom). I Quantification of (H). Each dot indicates the intensity of band. Two-sided unpaired Student’s t test and Two-sided Welch’s t test. n = 7 biological replicates. Source data are provided as a Source Data file. J Scatter plots indicate RNA-seq expression levels in no injection and Chk1-injected embryos. Red and blue dots indicate significantly upregulated and downregulated genes ( | log2 fold change | > 1, adjusted p value < 0.01, RPKM > 1.5) in Chk1-injected embryos, respectively. Pink, green, and light blue dots indicate Maternal, Maternal-ZGA, and Pure ZGA genes (see Supplementary Fig. 1F–H), respectively. n = 2 biological replicates. K Venn diagram indicates overlap of CBP/P300-dependent genes, CBP/P300-indepndent genes, and significantly upregulated genes in Chk1-injected embryos. L PCA analysis of RNA-seq data from normal development19, A485 treatment, and Chk1 injection. Each dot indicates one of the replicates. Dashed arrow indicates normal developmental process. No injection is located adjacent to pre-ZGA stage (late morula), while Chk1-injected is located closer to the post-ZGA stages (early blastula and late blastula). *: p < 0.05, **: p < 0.01, ***: p < 0.001, NS not significant. hpf hours post fertilization. Error bars indicate the mean ± s.d.
We next sought to examine the possible mechanism by which CBP/P300 activity is properly regulated during early development. The expression level of CBP/P300 is almost comparable between ZGA and post-gastrulation stages in both medaka and zebrafish, and the translation level of CBP/P300 is also comparable in zebrafish (Fig. 5G, Supplementary Fig. 6J). Changes in histone deacetylase (HDAC) activity may also not account for the enhanced CBP/P300 activity during ZGA. Its inhibition by TSA treatment at the late blastula stage resulted in a strong deposition of H3K27ac in medaka (Supplementary Fig. 6K–M), indicating that HDAC is active even during ZGA. Furthermore, the amount of histone acetylation substrate, acetyl-CoA, may not change, because H3K9ac and H4K16ac, histone acetylations mediated by other HATs, are not enriched during ZGA. One of the known mechanisms to increase the CBP/P300 activity is a phosphorylation of serine 31 residue of histone variant H3.3 (H3.3S31ph) deposited by kinases such as Chk142,43,44 (Fig. 6D). We thus quantitatively compared the level of H3.3S31ph per cell in embryos and found that the H3.3S31ph level during ZGA is much higher than that after ZGA in medaka (Fig. 6E, F). We also performed H3.3S31ph ChIP-seq and confirmed that H3.3S31ph accumulates to H3K27ac promoter peaks in both CBP/P300-dependent and -independent genes in medaka late blastula embryos (Fig. 6G, Supplementary Fig. 6N). Furthermore, overexpression of chek1 (a gene encoding Chk1 in medaka) increased both H3.3S31ph and H3K27ac (Fig. 6H, I) and caused precocious activation of CBP/P300-dependent genes in the late-morula (pre-ZGA) embryos, bringing the transcriptome closer to that of post-ZGA stages (Fig. 6J-L, Supplementary Fig. 6O, P). Therefore, these data propose a model in which a Chk1-mediated increase in H3.3S31ph enhances CBP/P300 activity during ZGA to properly proceed early teleost development (see Discussion).
Discussion
Various active epigenetic marks are established during ZGA2,3,4,5,6,7, but the previous studies only addressed the causal relationship between CBP/P300, H3K27ac and ZGA in non-mammalian vertebrates3,4,10, and the underlying molecular mechanism was not fully addressed even in the case of H3K27ac. In this study, we revealed that CBP/P300 does not activate all zygotic genes, but activates developmental genes via Pol II regulation during ZGA. The activation of housekeeping genes, on the other hand, could be mediated by other HATs and/or other types of histone acetylations, as suggested by accumulation of H3K9ac, H4K16ac, and H3K14ac (Fig. 5). Surprisingly, we found that H3K4me2/3 is not prerequisite for proper ZGA (Fig. 3A–C). Finally, our data suggest that the enhancement of CBP/P300 activity mediated by H3.3S31ph and Chk1 is essential for developmental gene activation during ZGA (Fig. 6). These data thus provide a comprehensive landscape of how active histone modifications and their writers cooperate to achieve ZGA in teleosts; developmental genes and housekeeping genes are activated by distinct mechanisms (Fig. 7).
During ZGA in teleost embryos, only developmental genes are activated by CBP/P300. CBP/P300 regulates Pol II initiation and deposits histone acetylations such as H3K27ac. BRD4 binds to H3K27ac and induces pause-release and elongation of Pol II. CBP/P300 activity also accumulates H3K4me2, but H3K4me2 is dispensable for ZGA. The activity of CBP/P300 is enhanced specifically during ZGA by temporal accumulation of H3.3S31ph. In contrast, activation of housekeeping genes is CBP/P300-independent. Alternatively, GCN5, MOF and HBO1 activity, or their substrates H3K9ac, H4K16ac and H3K14ac, are suggested to be involved in the activation of housekeeping genes. H3K4me2/3 accumulates at active gene promoters but is not prerequisite for their gene expression.
For developmental genes, H3K27ac was found to be specifically required for their activation. H3K27me3 and H3K4me2/me3 are known to regulate spatiotemporal repression and activation of developmental genes, respectively. Among histone acetylations, H3K27ac alone can directly regulate the deposition of these repressive and active modifications; H3K27ac antagonizes H3K27me345, while it promotes the accumulation of H3K4 methylation at some promoters46. Consistent with this, in our data, a certain proportion of the developmental genes showed gain of H3K27me3 after elimination of H3K27ac (Fig. 2A, B), and elimination of H3K27ac caused loss of H3K4me2 (Fig. 2A-D). Therefore, H3K27ac, deposited by CBP/P300, would be a suitable mediator for the spatiotemporal regulation of developmental genes. On the other hand, housekeeping genes, once activated, need to be continuously and robustly expressed throughout life. To ensure this, their activation should not depend solely on CBP/P300, but also on other HATs. Therefore, simultaneous inhibition of those multiple HATs would be required for disrupting activation of housekeeping genes during ZGA in medaka in the future studies. We also found that the difference in promoter characters (sharp vs broad promoter) (Fig. 1M) of those two types of genes. This would be caused by the difference in genomic features which might be reflected in the choice of the two types of gene regulation (CBP/P300 or non-CBP/P300 HATs). Our results are consistent with the previous study using mammalian cell lines showing that A485-sensitive genes are highly cell-type specific, while A485-insensitive genes are commonly expressed across all tissues, and that decreased promoter H3K27ac correlates with downregulation of gene expression32. Moreover, the use of different HATs for the activation of different genes is reminiscent of the recent study in Drosophila embryos, in which the activation of early developmental genes is CBP-dependent, while the activation of late developmental genes depends on other HATs47. Thus, there are various pathways that epigenetically promote gene activation. Despite the species specificity, developmental genes tend to use the CBP/P300-mediated pathway.
Our data suggest that CBP/P300 activity initiates the recruitment of Pol II, and CBP/P300-induced histone acetylations recruit BRD4, which in turn allows the elongation of Pol II at developmental genes (Fig. 3D–F). This stepwise regulation was also observed in cell lines32,33. Furthermore, CBP regulates Pol II initiation and elongation in Drosophila embryos during ZGA37, and CBP/P300 enhances Pol II elongation during ZGA in mouse embryos48. These indicate the common mechanism of CBP/P300-mediated regulation among many cell types, tissues, and species. However, as compared to differentiated cells (such as those in later stage embryos), the activity of CBP/P300 was extraordinarily enhanced during ZGA (late blastula) in teleosts (Fig. 5A–G), leading to high levels of H3K27ac accumulation. The levels of other active histone modifications did not globally change, irrespective of developmental stage and differentiation state. This enhanced CBP/P300 activity allows for rapid, genome-wide and simultaneous gene activation during ZGA of non-mammalian blastula embryos, which is in striking contrast to asynchronous activation of selected genes in differentiated cells. Because of this, the strong enhancement of the CBP/P300 activity may be indispensable for ZGA in non-mammalian vertebrates. Furthermore, our data also showed the correlation between the enhanced CBP/P300 activity and strong H3.3S31ph enrichment (Fig. 5I, J), the latter of which is known to promote the deposition of H3K27ac in cell lines and during mouse ZGA43,44,49. The deposition of H3.3S31ph is mediated by Chk1, one of the cell cycle regulators42,43,44, whose activity is temporarily and strictly regulated during mid-blastula transition in non-mammalian vertebrates50,51,52,53,54. Consistent with above, we and others previously found that overexpression of Chk1 caused precocious and partial initiation of ZGA in both medaka and zebrafish10,29. Hence, it is likely that non-mammalian embryos establish rapid and genome-wide genome activation (i.e. ZGA) by coupling two distinct biological mechanisms: epigenetic activation and developmental timing.
In contrast to histone acetylations, our data revealed for the first time that H3K4me2/3 are dispensable for gene activation during ZGA (Fig. 3A–C). This finding does not necessarily contradict the previous studies with cell lines that H3K4me3 is required for Pol II initiation20 and pause-release21,22, as they did not address the requirement of H3K4me3 for the activation of silenced genes, but rather for the maintenance of the active state. Therefore, it is likely that H3K4me2/3 is essential for the maintenance of activated genes during differentiation or gastrulation in medaka embryos. Alternatively, it is also possible that during ZGA, the levels of H3K27ac accumulation are sufficient to activate developmental genes. At later stages, however, H3K4me2/3 may be required for gene activation in addition to H3K27ac, as CBP/P300 activity is dramatically attenuated compared to that at ZGA stages (Fig. 5B, D). Further studies are warranted to elucidate the function of H3K4 methylations during early development.
Methods
Animal procedures
Medaka d-rR strain was used in this study. Fertilized embryos were raised according to standard protocols23 at 28 °C. Developmental stages were determined according to previously published guidelines (Iwamatsu 2004). 700 ng/µL of KDM5B mRNA, 20 ng/µL of chek1 mRNA, 280 ng/µL of FLAG-ep300a mRNA, and 10 ng/µL of α-amanitin (Sigma, A2263) were injected into one-cell stage embryos.
Zebrafish RW strain was used in this study. Fertilized embryos were raised according to standard protocols55 at 28.5 °C. Developmental stages were determined according to previously published guidelines55.
All experimental procedures and animal cares were performed under the approval of the animal ethics committee of the University of Tokyo (Approval No. 20-2).
Cell lines and culture condition
Medaka fibroblast19 and zebrafish fibroblast cell line BRF41 (RIKEN BioResource Center, Cell Bank No. RCB0804) were used in this study. Medaka fibroblasts were cultured in L-15 medium (Thermo, 11415-064) supplemented with 10% FBS and 1% Penicillin Streptomycin at 30 °C. BRF41 was cultured in L-15 medium (Thermo, 11415-064) supplemented with 15% FBS and 1% Penicillin Streptomycin at 33 °C.
Inhibitor treatment
Dechorionated medaka embryos were incubated in Yamamoto ringer supplemented with DMSO, 0.5-20 µM A485 (Tocris, 6387), 10 µM SGC-CBP30 (Sigma, SML1133), 20 or 100 nM TSA (Wako, 203-17561), 40 µM JQ1 (Sigma, SML1524) or 100 µM MM-102 (Selleck, S7265) from the four-cell stage (2 hpf) to the late blastula stage (8.5 hpf).
Constructions and in vitro transcription
Human KDM5B and medaka olep300a sequences were amplified by PCR from the human and medaka cDNA library, respectively, cloned into TOPO vector using TOPO TA Cloning Kit Dual Promoter (Invitrogen, 45-0640), and introduced into pCS2+ vector using NEBuilder HiFi DNA Assembly Master Mix (NEB, E2621). pCS2 + -Chk1 plasmid29 was used to produce chek1 mRNA. Templates for in vitro transcription were amplified by PCR, and mRNA was transcribed in vitro from the templates using mMESSAGE mMACHINE kit (Thermo, AM1340) or HiScribe T7 ARCA mRNA kit with tailing (NEB, E2060S). Primer sequences used for cloning and in vitro transcription are listed in Supplementary Data 1. The transcribed mRNA was purified using RNeasy mini kit (QIAGEN).
Fractionation
A previous cytoplasmic/chromatin fractionation protocol56 was modified in this study (Supplementary Fig. 1B). Briefly, medaka fibroblast or embryonic cells were collected and washed in PBS. The cells were lysed with fractionation buffer (25 mM Tris-HCl pH 7.5, 100 mM NaCl, 5 mM MgCl2, 10 % glycerol, 0.2 % NP-40 (Abcam, ab142227), 10 mM sodium butyrate, 1× protease inhibitor) by pipetting, and the lysates were placed on ice for 5-10 min. After we confirmed that cells were completely lysed, tubes were centrifuged at 1500 g for 5 min at 4 °C. The supernatant was transferred to new tubes, mixed with sample buffer, and used as a first soluble fraction (S). Fractionation buffer was added to the pellet, and tubes were centrifuged at 1500 × g for 5 min at 4 °C. After supernatant was removed, the pellet was mixed with sample buffer and used as pellet fraction (P). As shown in Supplementary Fig. 1B, cytoplasmic proteins are enriched in S, and chromatin is enriched in P. To purify genomic DNA from each fraction, mild sonication and DNA purification were performed as described in later in the section “Cell number estimation by qPCR”.
Western blot
Western blot was performed as previously described29 with modifications. Briefly, dechorionated embryos were homogenized by gentile pipetting up and down or passing through 21 G needle, centrifuged, and the pellets were snap frozen in liquid nitrogen. Samples were boiled in Laemmli sample buffer (62.5 mM Tris-HCl pH 6.8, 10% Glycerol, 2% SDS, 5% 2-mercaptoethanol, 0.0025% BPB) at 95 °C for 5 min and run on SDS-polyacrylamide gels. For the Western blot to compare different stages in Fig. 6C and Supplementary Fig. 6I, the cell number was estimated by measuring relative copy number of genome by qPCR (Supplementary Fig. 6A, primers are in Supplementary Data 1), and the same number of cells were loaded to each lane. For the Western blot to compare different stages in Fig. 6F and Supplementary Fig. 6C, the expected number of cells per embryo estimated from Supplementary Fig. 6A was used for loading the same number of cells to each lane. Otherwise, the same number of embryos were loaded to lanes. After transfer to PVDF membranes (Immobilon-FL, Millipore, IPFL00010), the membranes were incubated with blocking buffer (Intercept blocking buffer, LI-COR, 927-60001) for 1 h at room temperature (RT), incubated with primary antibody overnight at 4 °C (0.1% Tween-20 in blocking buffer), washed with TBST, incubated with secondary antibody (1/10000 × IRDye, 0.1% Tween-20, 0.01% SDS in blocking buffer) for 1 h at RT, washed with TBST, and air dried for a few hours at RT in the dark. Antibodies used for Western blot are listed in Supplementary Data 1. Chameleon Duo Pre-stained Protein Ladder was used as the size marker (LI-COR, 928-60000).
Immunofluorescence staining
We performed immunofluorescence staining using the late blastula stage (Fig. 3C) according to the previous protocols4,29 with modifications. Briefly, embryos were fixed with 4% PFA/PBS for 4 h at RT and overnight at 4 °C, manually dechorionated, permeabilized with 0.5% Triton X-100/PBS for 30 min at RT and washed with PBS. For antigen retrieval, samples were incubated with 4 M HCl for 15 min at RT, incubated with 100 mM Tris-HCl pH8.0 for 20 min at RT, and washed with PBS. Samples were blocked with blocking buffer (2% BSA, 1% DMSO, 0.2% Triton X-100, 1×PBS) for 1 h at RT, incubated with primary antibodies overnight at 4 °C, washed with PBSDT (1×PBS, 1% DMSO, 0.1% Triton X-100), blocked with blocking buffer for 1 h at RT, incubated with secondary antibodies for 1 h at RT, washed with PBSDT, and washed with PBS. The blastodisc was manually separated from yolk and mounted on slides.
The above fixation duration was too long for later stage embryos such at late neurula stage. Therefore, we further modified the protocol as follows for the immunofluorescence staining in Supplementary Fig. 6E. Briefly, the late blastula and late neurula embryos were fixed with 4% PFA/PBS for 2 h at RT, and manually dechorionated. Blastodisc and embryonic bodies were manually separated from yolk and transferred into 1% BSA/PBS on embryo dish. Samples were permeabilized with 0.5% Triton X-100/PBS for 20 min at RT, washed with 1% BSA/PBS, blocked with blocking buffer for 20 min at RT, incubated with primary antibodies overnight at 4 °C, washed with 1% BSA/PBS, incubated with secondary antibodies for 1 h at RT, washed with 1% BSA/PBS, and mounted on slides.
Antibodies used for immunofluorescence staining are listed in Supplementary Data 1. Imaging was performed with Leica Sp8.
Cell number estimation by qPCR
Dechorionated embryos were homogenized in microtubes by passing through a 21 G needle. After centrifugation, the supernatant was removed, and the pellet was stored in a freezer. The cell pellet was lysed with lysis buffer (50 mM Tris-HCl pH 8.0, 10 mM EDTA, 1% SDS) and sonicated briefly with Covaris (peak power: 105, duty factor: 4.0, cycles per burst: 200, duration: 180 sec). After RNaseA treatment and ProK treatment, DNA was purified by phenol: chloroform: isoamyl alcohol method and ethanol purification. The relative amount of DNA was measured by qPCR using AriaMx (Agilent). Primers used in this study are listed in Supplementary Data 1.
RNA-seq library preparation
RNA-seq was performed as previously described19 with modifications. Briefly, embryos were homogenized in 1 mL of ISOGEN, 200 µL of chloroform was added, and total RNA was purified using RNeasy mini kit. rRNA was depleted using RiboMinus Eukaryote System v2 with 1 µg of total-RNA as input. After rRNA depletion, RNA-seq libraries were generated using KAPA RNA HyperPrep Kit (KAPA Biosystems, KK8540). RNA-seq libraries were sequenced using the Illumina HiSeq1500 or HiSeqX system.
ChIP-seq library preparation
ChIP -seq was performed as previously described4 with modifications. Briefly, crosslinked and sonicated medaka embryonic cells were assembled with spike-in chromatin derived from zebrafish fibroblast cell line (BRF41), and incubated overnight with beads-antibody premix. The volumes of beads, antibodies, medaka chromatin, and spike-in chromatin are shown in Supplementary Data 1. For Pol II ser2P, H3K9ac, H4K16ac, H3K14ac, and H3.3S31ph ChIP-seq, spike-in chromatin was not added to the ChIP reaction. DNA was purified by phenol: chloroform: isoamyl alcohol method and ethanol purification. ChIP-seq libraries were prepared using KAPA Hyper Prep Kit (KAPA Biosystems, KK8502). ChIP-seq libraries were sequenced using the Illumina HiSeq1500, or HiSeqX system.
Quantification of Western blot
Imaging was performed using Odyssey CLx (LI-COR) and Image Studio (LI-COR). The samples in which linearity of signal intensity was confirmed and all blots were almost within the linear range were used for further analysis. For Western blots of different stages in Fig. 6A, the intensities were normalized by expected 1× intensity of fibroblast which was calculated using linear curve to minimize the difference between membranes, and the intensities were compared. For Western blots of different stages in other figures, raw intensities were compared. For the other Western blots, intensities were normalized to GAPDH intensity. Finally, intensities were normalized to the average of controls or late blastula, and the relative intensities were shown in dot plots.
Quantification of immunofluorescence staining
Images were quantified using Fiji57 according to the previous protocol4,29 with minor modifications. DAPI-dense regions were automatically selected as nuclei with manually defined thresholds, signal intensities in nuclei were measured, and the average intensity of cells in a single broad field slice image was normalized by that of DAPI.
RNA-seq data processing
Medaka and zebrafish RNA-seq data were processed as previously described19. Briefly, only the first 50 bp of forward strand reads were used in order to minimize the difference in sequencing method, low-quality reads and adapter-derived sequences were trimmed using Trimmomatic58, the trimmed reads were aligned to the medaka HdrR genome or zebrafish GRCz11 genome genome using STAR59, and alignments with mapping quality less than 20 were removed using SAMtools60. RPKM and adjusted p-value were calculated using DESeq261. Bigwig files were generated using deepTools bamCoverage and bigwigAverage62, and visualized using IGV63. Reproducibility of replicates was confirmed by measuring Pearson correlation coefficient of RPKM. For PCA analysis, top 500 genes with high variance in log2 RPKM were used for PCA analysis, and principal component (PC) 1 and PC2 were calculated from log2 RPKM using scikit-learn64. deepTools computeMatrix and plotHeatmap62 were used to generate heatmaps and line plots of transcript enrichment levels around gene bodies.
Raw reads from mouse RNA-seq were quality-checked using fastp65, aligned to the mm10 genome using STAR59 and RSEM66, FPKM was calculated using RSEM66, and adjusted p values were generated using DESeq261.
Identification of differentially expressed genes (DEGs)
Expressed (RPKM > 1.5) genes with significant changes (adjusted p-value < 0.01 and |log2 fold change | > 1) were used as DEGs in medaka. Expressed (FPKM > 2) genes with significant changes (adjusted p-value < 0.01 and |log2 fold change | > 1) were used as DEGs in mouse. To identify DEGs in zebrafish, original RPKM was normalized by ratio of total read count and read count aligned to mitochondria genes as described previously10, and log2 fold change (SGC-CBP30 / control) was calculated based on the normalized RPKM. Genes with log2 fold change less than −2 were used as CBP/P300-dependent genes in zebrafish. The gene lists are shown in Supplementary Data 1.
Identification of zygotically expressed genes
In medaka, upregulated DEGs in the late blastula stage compared to the late morula stage were used as “ZGA genes” in this study. To further classify expressed genes (Supplementary Fig. 1F), we first defined genes expressed (RPKM > 5) in the late morula stage (i.e. pre-ZGA) as “Maternal genes” (Supplementary Fig. 1G, H). We next classified Maternal genes overlapped with ZGA genes as “Maternal-ZGA genes”, and the others were named as “Pure ZGA genes” (Supplementary Fig. 1G, H). To strictly identify zygotically expressed genes in medaka, previous RNA-seq data of hybrid medaka embryos19 were used. Briefly, only the paternal reads were compared, and upregulated genes (normal late blastula embryos vs α-amanitin-injected embryos, log2 fold change > 1) were identified as “paternal ZGA genes” in this study. For mouse early 2-cell stage analysis, upregulated DEGs in DMSO early 2-cell stage embryos compared to DMSO zygote stage embryos were used as “zygotic genes”. In mouse late 2-cell stage analysis, upregulated DEGs in DMSO late 2-cell stage embryos compared to DMSO zygote stage embryos were used as “zygotic genes”. The gene lists are shown in Supplementary Data 1.
Gene ontology (GO) analysis
GO enrichment analysis was performed using PANTHER67. For GO analysis in medaka, top 10 GO terms (biological process) with FDR less than 0.05 and whose fold enrichment was more than 2.0 were shown. For GO analysis in zebrafish, top 20 GO terms (biological process) with FDR less than 0.01 and whose fold enrichment was more than 1.7 were shown. For GO analysis in mouse, top 10 GO terms (biological process) with FDR less than 0.01 and whose fold enrichment was more than 2.0 were shown.
ChIP-seq data processing
Medaka and zebrafish ChIP-seq data were processed as previously described4 with modifications. Briefly, only the first 50 bp of forward strand reads were used in order to minimize the differences in sequencing method, low quality reads and adapter-derived sequences were trimmed by Trimmomatic58, and the trimmed reads were aligned to the medaka (HdrR) and zebrafish (GRCz11) concatenated genome, medaka (HdrR) genome, or zebrafish (GRCz11) genome. Low mapping quality reads ( < 20) and PCR duplicates were removed using SAMtools60, and the reads were extended by 200 bp using bedtools68. Bigwig files were generated using deepTools bamCoverage and bigwigAverage62 and visualized by IGV63. If necessary, the bigwig files were multiplied by the spike-in factor4. To confirm reproducibility of replicates, ChIP enrichment of each 10 kb bin was obtained using deepTools multiBigwigSummary62 with the options “–binSize 10000 --outRawCounts”, data of unanchored contigs were removed, and Pearson correlation coefficient was calculated. deepTools computeMatrix and plotHeatmap62 were used to generate heatmaps and line plots of ChIP signal levels. To identify H3K27ac peaks in medaka, macs2 callpeak69 was performed with the following options “-g 600000000 -q 0.01 --keep-dup all –nomodel”. After removing peaks within unanchored contigs, the peaks with fold change >6.5 were extracted and merged by bedtools merge68 with the option “-d 150”. The merged peaks were annotated using Homer annotatePeaks70, and the peaks whose distance to the nearest TSS is less than 3 kb were used as promoter-proximal peaks in in this study.
Previous mouse ChIP-seq bigwig files were downloaded from GEO database under accession number GSE7278471 and GSE14352341. H3K27ac peaks were called using macs2 bdgpeakcall69 with the options “-c 10 -l 300 -g 300”, merged with the option “-d 150” using bedtools merge68, annotated using Homer annotatePeaks70, and the peaks whose distance to the nearest TSS is less than 3 kb were used as promoter-proximal peaks in this study.
Pol II occupancy analysis
To compare Pol II enrichment in CBP/P300-dependent and -independent genes, Pol II occupancy was quantified using Pol II spike-in ChIP-seq or Pol II Ser2P ChIP-seq data as follows. The average Pol II enrichment within TSS ± 250 bp was used as Pol II density around TSS. The average Pol II enrichment within the range “TSS + 500 bp to TES – 250 bp” was used as Pol II density within gene body. The short genes (gene body length ≤ 750 bp) were excluded from this analysis.
Motif enrichment analysis
Motif enrichment analysis was performed by AME72 with JASPAR2022 CORE vertebrates non-redundant v2 as input of motifs. Motifs that were specifically enriched (log2 fold change of p value > 1, log2 fold change of enrichment > 0.1) in CBP/P300-dependent genes or -independent genes are shown in Fig. 5H. Motif sequence log was downloaded from JASPAR website73.
ATAC-seq data processing
Previous ATAC-seq data4 was processed as previously described4,19 with modifications. First, the reads were trimmed and mapped, and low-quality reads and PCR duplicates were removed as described above in “ChIP-seq data processing”. Second, the reads were were extended to both directions by 100 bp using bedtools68. Bigwig files were generated using deepTools bamCoverage and bigwigAverage62. Peak calling was performed as described above, peaks within unanchored contigs were removed, the peaks with fold change > 4 were extracted, and peaks were merged by bedtools merge68 with the option “-d 150”. Gain and loss of chromatin accessibility in ATAC-seq peaks (p value < 0.05, |log2 fold change | > 1) were found by DiffBind74. deepTools computeMatrix and plotHeatmap62 were used to generate line plots of ATAC-seq signal.
Data quantification and statistical test
Box plots in this study indicate median (middle line), 25th, 75th percentile (box) and 5th and 95th percentile (whiskers), and outliers are not shown. For two-sample statistical tests, normality and equal variances were first tested using the Shapiro-Wilk test and the F-test, respectively, with p value = 0.05. If both null hypotheses were not rejected, a two-sided unpaired Student’s t test was performed. If the null hypothesis of F-test alone was rejected, we performed two-sided Welch’s t test. Otherwise, two-sided Wilcoxon rank-sum test was performed. *** p < 0.001, ** p < 0.01, * p < 0.05, NS: not significant, respectively.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
RNA-seq and ChIP-seq data newly generated in this study have been deposited in the DDBJ Sequenced Read Archive under the accession number PRJDB18921 [https://ddbj.nig.ac.jp/search/entry/bioproject/PRJDB18921]. Publicly available datasets used in this study were from the DDBJ Sequenced Read Archive accession number PRJDB5325 and PRJDB7492 (medaka RNA-seq data19,75); PRJDB13204 (medaka ChIP-seq and ATAC-seq data4); the Sequence Read Archive under accession number SRP184786 (zebrafish RNA-seq data10); SRP355652 (zebrafish ChIP-seq data40); the Gene Expression Omnibus under accession number GSE207222 (mouse RNA-seq data11); GSE143523 (mouse ChIP-seq data41). The raw immunofluorescence staining data in this study have been deposited in the SSBD repository at https://ssbd.riken.jp/repository/431/. Source data are provided with this paper. The raw Western blot data, results of quantification of immunofluorescence staining and Western blot are provided in the Source Data file. Source data are provided with this paper.
References
Vastenhouw, N. L., Cao, W. X. & Lipshitz, H. D. The maternal-to-zygotic transition revisited. Development 146, dev161471 (2019).
Sato, Y. et al. Histone H3K27 acetylation precedes active transcription during zebrafish zygotic genome activation as revealed by live-cell analysis. Development 146, dev179127 (2019).
Zhang, B. et al. Widespread enhancer dememorization and promoter priming during parental-to-zygotic transition. Mol. Cell 72, 673–686.e6 (2018).
Fukushima, H. S., Takeda, H. & Nakamura, R. Incomplete erasure of histone marks during epigenetic reprogramming in medaka early development. Genome Res. 33, 572–586 (2023).
Hontelez, S. et al. Embryonic transcription is controlled by maternally defined chromatin state. Nat. Commun. 6, 10148 (2015).
Vastenhouw, N. L. et al. Chromatin signature of embryonic pluripotency is established during genome activation. Nature 464, 922–926 (2010).
Lindeman, L. C. et al. Prepatterning of developmental gene expression by modified histones before zygotic genome activation. Dev. Cell 21, 993–1004 (2011).
Xia, W. & Xie, W. Rebooting the epigenomes during mammalian early embryogenesis. Stem Cell Rep. 15, 1158–1175 (2020).
Zhou, V. W., Goren, A. & Bernstein, B. E. Charting histone modifications and the functional organization of mammalian genomes. Nat. Rev. Genet. 12, 7–18 (2011).
Chan, S. H. et al. Brd4 and P300 confer transcriptional competency during zygotic genome activation. Dev. Cell 49, 867–881.e8 (2019).
Wang, M., Chen, Z. & Zhang, Y. CBP /p300 and HDAC activities regulate H3K27 acetylation dynamics and zygotic genome activation in mouse preimplantation embryos. EMBO J. 41, e112012 (2022).
Wu, K. et al. Dynamics of histone acetylation during human early embryogenesis. Cell Discov. 9, 29 (2023).
Sakamoto, M. et al. Detection of newly synthesized RNA reveals transcriptional reprogramming during ZGA and a role of Obox3 in totipotency acquisition. Cell Rep. 43, 114118 (2024).
Wike, C. L. et al. Chromatin architecture transitions from zebrafish sperm through early embryogenesis. Genome Res. 31, 981–994 (2021).
Akkers, R. C. et al. A Hierarchy of H3K4me3 and H3K27me3 Acquisition in Spatial Gene Regulation in Xenopus Embryos. Dev. Cell 17, 425–434 (2009).
Liu, G., Wang, W., Hu, S., Wang, X. & Zhang, Y. Inherited DNA methylation primes the establishment of accessible chromatin during genome activation. Genome Res. 28, 998–1007 (2018).
Pálfy, M., Schulze, G., Valen, E. & Vastenhouw, N. L. Chromatin accessibility established by Pou5f3, Sox19b and Nanog primes genes for activity during zebrafish genome activation. PLOS Genet. 16, e1008546 (2020).
Zhang, Y. et al. Canonical nucleosome organization at promoters forms during genome activation. Genome Res. 24, 260–266 (2014).
Nakamura, R. et al. CTCF looping is established during gastrulation in medaka embryos. Genome Res. 31, 968–980 (2021).
Lauberth, S. M. et al. H3K4me3 Interactions with TAF3 Regulate Preinitiation Complex Assembly and Selective Gene Activation. Cell 152, 1021–1036 (2013).
Wang, H. et al. H3K4me3 regulates RNA polymerase II promoter-proximal pause-release. Nature 615, 339–348 (2023).
Hu, S. et al. H3K4me2/3 modulate the stability of RNA polymerase II pausing. Cell Res. 33, 403–406 (2023).
Kinoshita, M., Murata, K., Naruse, K. & Tanaka, M. Medaka: Biology, Management, and Experimental Protocols. Medaka: Biology, Management, and Experimental Protocols (Wiley, 2009). https://doi.org/10.1002/9780813818849.
Lasko, L. M. et al. Discovery of a selective catalytic p300/CBP inhibitor that targets lineage-specific tumours. Nature 550, 128–132 (2017).
Hammitzsch, A. et al. CBP30, a selective CBP/p300 bromodomain inhibitor, suppresses human Th17 responses. Proc. Natl. Acad. Sci. 112, 10768–10773 (2015).
Sandelin, A. et al. Mammalian RNA polymerase II core promoters: Insights from genome-wide studies. Nat. Rev. Genet. 8, 424–436(2007).
Kim, J. J. & Kingston, R. E. Context-specific Polycomb mechanisms in development. Nat. Rev. Genet. 23, 680–695 (2022).
Zhang, B. et al. Allelic reprogramming of the histone modification H3K4me3 in early mammalian development. Nature 537, 553–557 (2016).
Fukushima, H. S., Ikeda, T., Ikeda, S. & Takeda, H. Cell cycle length governs heterochromatin reprogramming during early development in non-mammalian vertebrates. EMBO Rep. https://doi.org/10.1038/s44319-024-00188-5 (2024).
Hörmanseder, E. et al. H3K4 Methylation-Dependent Memory of Somatic Cell Identity Inhibits Reprogramming and Development of Nuclear Transfer Embryos. Cell Stem Cell 21, 135–143.e6 (2017).
Karatas, H. et al. High-Affinity, Small-Molecule Peptidomimetic Inhibitors of MLL1/WDR5 Protein–Protein Interaction. J. Am. Chem. Soc. 135, 669–682 (2013).
Narita, T. et al. Enhancers are activated by p300/CBP activity-dependent PIC assembly, RNAPII recruitment, and pause release. Mol. Cell 81, 2166–2182.e6 (2021).
Hsu, E., Zemke, N. R. & Berk, A. J. Promoter-specific changes in initiation, elongation, and homeostasis of histone H3 acetylation during CBP/p300 inhibition. Elife 10, (2021).
Yang, Z. et al. Recruitment of P-TEFb for stimulation of transcriptional elongation by the bromodomain protein Brd4. Mol. Cell 19, 535–545 (2005).
Jang, M. K. et al. The bromodomain protein Brd4 is a positive regulatory component of P-TEFb and stimulates RNA polymerase II-dependent transcription. Mol. Cell 19, 523–534 (2005).
Filippakopoulos, P. et al. Selective inhibition of BET bromodomains. Nature 468, 1067–1073 (2010).
Cardamone, F. et al. Chromatin landscape at cis-regulatory elements orchestrates cell fate decisions in early embryogenesis. Nat. Commun. 16, 3007 (2025).
Jin, Q. et al. Distinct roles of GCN5/PCAF-mediated H3K9ac and CBP/p300-mediated H3K18/27ac in nuclear receptor transactivation. EMBO J. 30, 249–262 (2011).
Shvedunova, M. & Akhtar, A. Modulation of cellular processes by histone and non-histone protein acetylation. Nat. Rev. Mol. Cell Biol. 23, 329–349 (2022).
Miao, L. et al. The landscape of pioneer factor activity reveals the mechanisms of chromatin reprogramming and genome activation. Mol. Cell 82, 986–1002.e9 (2022).
Yang, G. et al. Dux-mediated corrections of aberrant H3K9ac during 2-cell genome activation optimize efficiency of somatic cell nuclear transfer. Cell Stem Cell 28, 150–163.e5 (2021).
Chang, F. T. M. et al. CHK1-driven histone H3.3 serine 31 phosphorylation is important for chromatin maintenance and cell survival in human ALT cancer cells. Nucleic Acids Res. 43, 2603–2614(2015).
Martire, S. et al. Phosphorylation of histone H3.3 at serine 31 promotes p300 activity and enhancer acetylation. Nat. Genet. 51, 941–946 (2019).
Sitbon, D., Boyarchuk, E., Dingli, F., Loew, D. & Almouzni, G. Histone variant H3.3 residue S31 is essential for Xenopus gastrulation regardless of the deposition pathway. Nat. Commun. 11, 1256(2020).
Tie, F. et al. CBP-mediated acetylation of histone H3 lysine 27 antagonizes Drosophila Polycomb silencing. Development 136, 3131–3141 (2009).
Zhao, W. et al. Investigating crosstalk between H3K27 acetylation and H3K4 trimethylation in CRISPR/dCas-based epigenome editing and gene activation. Sci. Rep. 11, 15912 (2021).
Ciabrelli, F. et al. CBP and Gcn5 drive zygotic genome activation independently of their catalytic activity. Sci. Adv. 9, eadf2687 (2023).
Xiao, L. et al. DUX-mediated configuration of p300/CBP drives minor zygotic genome activation independent of its catalytic activity. Cell Rep. 44, 115544 (2025).
Zhang, J. et al. Maternal H3.3-mediated paternal genome reprogramming contributes to minor zygotic genome activation. bioRxiv 2023.11.07.566007 https://doi.org/10.1101/2023.11.07.566007 (2023).
Collart, C., Smith, J. C. & Zegerman, P. Chk1 inhibition of the replication factor Drf1 guarantees cell-cycle elongation at the xenopus laevis mid-blastula transition. Dev. Cell 42, 82–96.e3 (2017).
Shimuta, K. et al. Chk1 is activated transiently and targets Cdc25A for degradation at the Xenopus midblastula transition. EMBO J. 21, 3694–3703 (2002).
Gotoh, T., Kishimoto, T. & Sible, J. C. Phosphorylation of Claspin is triggered by the nucleocytoplasmic ratio at the Xenopus laevis midblastula transition. Dev. Biol. 353, 302–308 (2011).
Collart, C., Allen, G. E., Bradshaw, C. R., Smith, J. C. & Zegerman, P. Titration of four replication factors is essential for the xenopus laevis midblastula transition. Sci. (80-). 341, 893–896 (2013).
Zhang, M., Kothari, P., Mullins, M. & Lampson, M. A. Regulation of zygotic genome activation and DNA damage checkpoint acquisition at the mid-blastula transition. Cell Cycle 13, 3828–3838 (2014).
Kimmel, C. B., Ballard, W. W., Kimmel, S. R., Ullmann, B. & Schilling, T. F. Stages of embryonic development of the zebrafish. Dev. Dyn. 203, 253–310 (1995).
Minamino, M. et al. Esco1 acetylates cohesin via a mechanism different from that of Esco2. Curr. Biol. 25, 1694–1706 (2015).
Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: A flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120(2014).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Li, H. et al. The sequence alignment/map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Ramírez, F. et al. deepTools2: A next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 44, W160–W165 (2016).
Robinson, J. T. et al. Integrative genomics viewer. Nat. Biotechnol. 29, 24–26 (2011).
Pedregosa, F. et al. Scikit-learn: Machine learning in python. J. Mach. Learn. Res. 12, 2825–2830 (2011).
Chen, S., Zhou, Y., Chen, Y. & Gu, J. fastp: An ultra-fast all-in-one FASTQ preprocessor. Bioinformatics 34, i884–i890 (2018).
Li, B. & Dewey, C. N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinforma. 12, 323 (2011).
Mi, H., Muruganujan, A., Ebert, D., Huang, X. & Thomas, P. D. PANTHER version 14: more genomes, a new PANTHER GO-slim and improvements in enrichment analysis tools. Nucleic Acids Res. 47, D419–D426 (2019).
Quinlan, A. R. & Hall, I. M. BEDTools: A flexible suite of utilities for comparing genomic features. Bioinformatics 26, 841–842 (2010).
Zhang, Y. et al. Model-based analysis of ChIP-Seq (MACS). Genome Biol. 9, R137 (2008).
Heinz, S. et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell 38, 576–589 (2010).
Dahl, J. A. et al. Broad histone H3K4me3 domains in mouse oocytes modulate maternal-to-zygotic transition. Nature 537, 548–552(2016).
McLeay, R. C. & Bailey, T. L. Motif enrichment analysis: A unified framework and an evaluation on ChIP data. BMC Bioinforma. 11, 165 (2010).
Rauluseviciute, I. et al. JASPAR 2024: 20th anniversary of the open-access database of transcription factor binding profiles. Nucleic Acids Res. 52, D174–D182 (2024).
Ross-Innes, C. S. et al. Differential oestrogen receptor binding is associated with clinical outcome in breast cancer. Nature 481, 389–393 (2012).
Ichikawa, K. et al. Centromere evolution and CpG methylation during vertebrate speciation. Nat. Commun. 8, 1833 (2017).
Acknowledgements
We acknowledge all the laboratory members for everyday discussion and for their continuous support. In particular, we thank Ryohei Nakamura (Univ. of Tokyo) and Takafumi Ikeda (Kyoto Sangyo Univ.) for their assistance in sample preparation. This work was supported by JSPS KAKENHI Grant No. JP23K14121 and 24KJ0236 to HSF, and by Japan Agency for Medical Research and Development (AMED) under Grant No. JP18gm1110007 and JSPS KAKENHI Grant No. 23H02484 to HT.
Author information
Authors and Affiliations
Contributions
H.S.F. conceived the study, performed experiments, analyzed the data, and wrote the manuscript. H.T. supervised the work and wrote the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Fukushima, H.S., Takeda, H. Coordinated action of multiple active histone modifications shapes the zygotic genome activation in teleost embryos. Nat Commun 16, 5222 (2025). https://doi.org/10.1038/s41467-025-60246-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-025-60246-x
