NFYA regulates two sequential genome-wide transcriptional activation events during oocyte to embryo transition

Abstract

Primordial follicle oocyte activation (PFA) and zygotic genome activation (ZGA) represent two major waves of transcription activation respectively required for oocyte growth and preimplantation embryo development. Although many shared molecular hallmarks between PFA and ZGA suggest potential common factors and mechanisms driving both waves of transcriptional activation, such factors are yet to be identified. Here we demonstrate that the pioneer factor NFYA belongs to such regulators. Oocyte-specific Nfya deletion impairs open chromatin establishment and transcriptional activation during PFA, which triggers non-canonical ferroptosis leading to early folliculogenesis failure. Moreover, acute NFYA depletion in zygotes causes defective ZGA and predominantly two-cell embryo arrest. Mechanistically, although NFYA exhibits distinct chromatin-binding preferences predominantly targeting promoters during PFA and enhancers during ZGA, pre-occupied NFYA regulates chaperones and histone genes in both PFA and ZGA through conserved promoter binding. Together, our studies establish NFYA as a multifaceted regulator of genome activation during both PFA and ZGA.

Introduction

Neonatal mammalian non-growing oocytes are arrested in the diplotene stage of meiotic prophase I and maintained in a dormant state in primordial follicles ¹, characterized by widespread transcriptional quiescence ^2–4. During primordial follicle oocyte activation (PFA), the quiescent genome is sequentially activated to support oocyte growth and progression to primary follicle stage and then secondary follicle stage, which are called primordial-to-primary transition (PPT), and primary-to-secondary transition (PST), respectively ^5,6. Through this process, oocytes accumulate maternal RNAs and proteins required for subsequent oocyte growth and early embryonic development ^5,6. Studies on the mechanisms of quiescent primordial follicle maintenance have led to the identification of several “repressors”, including PTEN-PI3K ⁷, FOXO3A ⁸, and LHX8 ^9,10. Knockout these “repressors” leads to excessive activation or depletion of primordial follicles, resulting in premature exhaustion of primordial follicle pool. However, the identity of pioneer factors responsible for initiating PFA in vivo have remained elusive.

The secondary follicle oocytes continue to grow and accumulate maternal factors to develop into full-grown oocytes, which is accompanied by decline in transcriptional activity to a state of transcriptional quiescence and arrest at MII stage ^5,6. Following fertilization, the transcriptionally silenced genome becomes active again through a process called zygotic genome activation (ZGA) to enable the maternal-to-zygotic transition (MZT) for early embryogenesis ¹¹. ZGA is initiated at the late-zygotic stage with low levels of transcription and peaks at the late 2-cell (L2C) stage with large-scale transcription ¹². Notably, the two waves of genome activation, PFA and ZGA, are both characterized by ribosome biogenesis, rapid and massive protein synthesis and proper folding, and widespread chromatin remodeling ^13–16. Although these shared molecular features have been recognized, a unified pioneer factor that triggers both PFA and ZGA has not been identified, largely due to sample and technical limitations for comprehensive characterization of both PFA and ZGA processes.

In this study, we demonstrate that the pioneer factor NFYA is a unified regulator for both PFA and ZGA by utilizing an oocyte-specific knockout mouse model ¹⁷ and a targeted protein degradation tag (dTAG) mouse model ^18,19, combined with RNA-seq, ATAC-seq, and low-input CUT&RUN assays. Phenotypically, oocyte-specific deletion of Nfya impairs open chromatin establishment and transcriptional activation during PFA, which triggers non-canonical ferroptosis leading to early folliculogenesis failure. By generating NFYA^dTAG mice, we further demonstrate that NFYA depletion causes defective ZGA and predominantly 2-cell embryonic arrest. Mechanistically, although NFYA exhibits distinct chromatin-binding preferences predominantly targeting promoters during PFA and enhancers during ZGA, pre-occupied NFYA regulates chaperones and histone genes in both PFA and ZGA through conserved promoter binding. Inhibition of chaperones impairs both PFA and ZGA supporting their requirement for genome activation. Together, these findings establish the multifaceted NFYA as a unified regulator of both PFA and ZGA that orchestrate the oocyte-to-embryo transition. Our study advances the understanding of how the same transcription factor mediates activation of silenced genome in two critical developmental transitions.

Results

Identification of NFYA as a potential unified regulator for PFA and MZT

To identify potential regulators of oocyte-to-embryo transition, we performed integrative analyses combining our newly generated total RNA-seq and ATAC–seq datasets with previously published data, covering developmental stages from primordial follicle oocytes to 2-cell embryos ^14,20–22. Principal components analysis (PCA) revealed that both chromatin accessibility and transcription undergo two major transitions at PFA and MZT (Figure 1A and S1A). Notably, the magnitude of chromatin remodeling in PFA was as large as that in MZT, during which zygotic genome activation occurs ¹². Additionally, the lower global chromatin accessibility in primordial follicle oocytes was in line with its transcriptional quiescent state compared to that of growing-oocytes (Figure S1B–C). To identify candidate transcription factors (TFs) participating in both PFA and MZT, we first analyzed the chromatin accessibility changes during these two transitions. We identified 109,129 newly accessible chromatin regions (P2) during primordial-to-primary transition (Figure 1B), while similar analysis identified 3,806 and 5,084 newly accessible chromatin regions at zygote stage (Z2 and Z3) and late 2-cell (L2C) stage (Z4), respectively (Figure 1C). Motif enrichment analysis of these accessible regions revealed that OBOX ²³, NR5A2 ^24,25, and GABPA ¹⁹ are exclusively enriched during MZT, while LHX ¹⁰ and TCF3/12 ²⁶ motifs are specifically enriched during PFA, generally aligning with their expression dynamics (Figure 1D and S1D). Notably, NFYA binding motifs are highly enriched in the newly accessible regions during both PFA and MZT, suggesting that NFYA may act as a unified regulator orchestrating these two transitions.

Figure 1. Identification of NFYA as a potential regulator of both PFA and ZGA and loss of NFYA causes folliculogenesis failure.

(A) Upper panel: Schematic diagram of the oocyte-to- embryo transition, which initiates at the primordial follicle stage and ends at the 2-cell embryo (2C) stage with PFA and ZGA timing indicated. GV, germinal vesicle oocyte. Lower panel: principal component analysis (PCA) of global chromatin accessibility across different stages from primordial follicle to late 2-cell. (B) Heatmaps showing the ATAC-seq peaks classified based on their changes during PFA (P1-P3). P1 cluster (n=1,789) represents specific open chromatin in primordial follicle stage; P2 cluster (n=109,129) represents those that gained open chromatin in primary and secondary follicle stages; P3 cluster 3 (n=21,717) indicates the shared open chromatin across PFA. (C) Heatmaps showing the ATAC-seq peaks classified based on their changes during MZT (Z1-Z5). Z1 cluster (n=8,467) represents specific open chromatin in GV; Z2 cluster (n=3,650) represents the gained open chromatin in zygote; Z3 cluster (n=156) represents the gained open chromatin in both zygote and late 2-cell (L2C); Z4 cluster (n=5,048) represents the gained open chromatin in L2C; Z5 cluster (n=249) represents the shared open chromatin across MZT. (D) Enrichment of TF motifs at the ATAC–seq peak groups (P2, Z2, Z3, and Z4) during PFA and MZT. (E) The dynamics of Nfya expression at transcription (RNA-seq, FPKM) and translation (LiRibo-seq ¹⁴, RPKM) levels in mouse oocytes, early embryos, and tissues. (F) Immunostaining of NFYA (green) during mouse oocyte and early embryo development. Scale bar, 25 pm. Dashed white box indicates a higher-magnification view of the primordial follicle oocyte. (G) Immunostaining of NFYA (green) in WT and Nfya-cKO primordial/primary follicle oocytes. Scale bar, 25 pm. The white dashed circle marks the whole oocytes. (H) Fertility test showing the cumulative numbers of pups from crosses of WT (n=3) and Nfya-cKO (n=3) females with WT males. (I) The numbers of MII oocytes per female mice at 4 weeks- or 8 weeks-age after superovulation. The WT (n=3) and Nfya-cKO (n=3) female mice were used for the analysis. (J) The numbers of GV oocytes per female mice at 4 weeks- or 8 weeks-age after 48h PMSG induction. The WT (n=3) and Nfya-cKO (n=3) female mice were used for the analysis. (K-M), Upper panel: H&E staining of paraffin embedded ovarian sections showing the histology of WT and Nfya-cKO ovaries at the indicated postnatal days. Scale bar, 100 pm. A high-resolution view of the boxed area is shown in parallel. Arrows point to follicles at the indicated stages. PrF, primordial follicle; PF, primary follicle; ESF, early secondary follicle; LSF, late secondary follicle; AF, antral follicle. Lower panel: Bar chart showing the follicle numbers of WT (blue) and Nfya-cKO (red) ovaries at the indicated times. Quantitative data are shown as mean ± SEM; **p < 0.01, *p < 0.05, ns p > 0.05, Student’s t test.

Consistent with the notion that NFYA may serve as a unified regulator of these two transitions, its transcription is dramatically increased during PFA, peaked at the MII oocyte stage, and subsequently declined after fertilization (Figure 1E). However, its translation is maintained at high levels throughout MZT, with an upregulation during ZGA (Figure 1E). Notably, Nfya is highly expressed during both PFA and ZGA, but at a very low level in later-stage of embryogenesis, mESCs, and somatic tissues, suggesting a potential role of NFYA in these two genome activation events. Immunostaining showed that NFYA protein level increases during PFA and stays in nucleus from primordial to GV stages, and translocated from the cytoplasm into the nucleus during MZT (Figure 1F). These data are consistent with a potential role of NFYA in regulating both PFA and MZT. In fact, our previous studies have suggested that NFYA regulates ZGA and preimplantation development ¹⁵. Therefore, we first focused our effect on NFYA’s role in PFA.

Loss of NFYA causes early folliculogenesis failure

NFYA is the sequence-specific DNA binding subunit of the NFY/CCAAT complex with an evolutionarily conserved function in chromatin accessibility establishment or maintenance ^27,28. Previous studies have shown that Nfya knockout results in embryonic lethality ²⁹. To study its potential role in PFA, we generated Gdf9-Cre Nfya^flox/flox mice (referred to as Nfya-cKO hereafter) which depletes NFYA when primordial follicle oocytes activation just get started ¹⁷ (Figure S1E). We confirmed its successful depletion in primary follicle oocytes, but not in dormant primordial follicle oocytes (Figure 1G and S1F). Notably, Nfya-cKO mice are infertile (Figure 1H) and cannot generate matured oocytes or full-grown oocytes (Figure 1I–J), suggesting early folliculogenesis defects. Furthermore, significant ovarian developmental defects were observed in Nfya-cKO mice starting from 2-week (Figure S1G–H), corresponding to the stage when the first wave of antral follicle formation is completed in WT ovaries ^5,6. Detailed analysis revealed that ablation of Nfya leads to a drastic loss of primary follicles (PF), with some progressing to early secondary follicle (ESF) exhibiting abnormal morphology but failed to generate late secondary follicles (LSF) or antral follicles (AF) (Figure 1K). Notably, folliculogenesis defects are more severe in 4-week Nfya-cKO mice, with shrunken morphology of primary and early secondary follicle oocytes and significant loss of primary follicles, suggesting oocyte arrest and follicle degeneration and resorption (Figure 1L). By 8 weeks, almost all the primary follicles were degenerated in Nfya-cKO mice, with only primordial follicles (PrF) remaining, which are characterized by flat squamous pre-granulosa cells (Figure 1M and S1I), indicating defective PFA in Nfya-cKO mice. Taken together, these results demonstrate that NFYA is critical for early folliculogenesis.

NFYA deficiency impairs PFA in primary follicle oocytes

We next asked whether ablation of NFYA affects PFA. To this end, we performed RNA-seq for Nfya-cKO and WT oocytes at primordial, primary, and secondary follicle stages (Figure S2A). RNA-seq revealed that oocyte marker genes were highly expressed, while granulosa cell marker genes were not detectable (Figure S2B), indicating that the samples were not contaminated by granulosa cells. Consistent with protein depletion (Figure 1G), RNA-seq confirmed Nfya RNA depletion at primary and secondary follicle stages, but not at dormant primordial follicle stage (Figure S2C). PCA showed that while primordial follicle oocytes of WT and Nfya-cKO were clustered closely, the primary and secondary follicle oocytes of Nfya-cKO and WT were distinctively clustered (Figure 2A), suggesting extensive transcriptomic alterations in the Nfya-cKO primary and secondary follicle oocytes. To further pinpoint the defects, comparative analysis identified 809 downregulated and 281 upregulated genes in primary follicle oocytes in response to NFYA depletion (Figure 2B and Table S1). To identify differentially expressed genes (DEGs) directly targeted by NFYA, we performed CUT&RUN experiments using WT primary follicle oocytes and confirmed that NFYA peaks are highly enriched for NFYA binding motif (Figure S2D–E). Importantly, NFYA CUT&RUN in Nfya-cKO primary follicle oocytes revealed that all the NFYA peaks were diminished compared to that in WT control (Figure S2F), confirming the specificity of the CUT&RUN signal. Furthermore, integrative analysis showed that NFYA mainly occupies promoter regions of the downregulated genes with 402 out of the 809 genes bound by NFYA (Figure 2C and Table S1). Consistently, 95.2% (383/402) of the NFYA bound promoters contain at least one NFYA motif (Figure 2C and S2G). Gene Ontology (GO) analysis revealed that these direct targets are enriched for cell cycle transition and chromosome condensation (Figure 2D). Notably, the direct targets include many genes whose knockouts or mutations exhibit infertility phenotype such as Fancf, Hsp90b1, Rfwd3, and Prkaca (Figure 2E and S2H). Interestingly, Prkaca encodes the catalytic subunit of cAMP-dependent protein kinase A (PKA), and the cAMP-PKA signaling is pivotal for PFA and meiotic cell cycle arrest ^30,31, supporting a critical role of NFYA in PFA.

Figure 2. NFYA deficiency impairs PFA in primary follicle oocytes.

(A) PCA analysis reveals the effects of loss NFYA function on primary and secondary follicle oocytes. (B) Scatter plots comparing the gene expression profiles of WT and Nfya-cKO primary follicle oocytes from P7 ovaries. The x and y axis of the dot plots are Log2CPM (counts per million) from RNA-seq. Fold change > 2, false discovery rate (FDR) < 0.05. (C) Heatmaps showing the DEGs of primary follicles caused by NFYA loss, as well as NFYA binding and NFYA motif occurrence profiles around the TSS of the corresponding genes. The downregulated and upregulated genes are each separated into two groups based on whether they have direct NFYA binding. (D) Gene Ontology (GO) terms enriched for the NFYA-bound down-regulated genes (n=402) with examples. (E) Genome browser view of NFYA targeted infertility-related genes examples showing NFYA CUT&RUN and RNA-seq in WT and CKO primary follicle oocytes. (F) Heatmap showing early-PFA gene expression (z-scores) in WT and Nfya-cKO oocytes at the indicated stages. (G) Heatmaps showing the RNA-seq results of the downregulated early-PFA genes (n=389) at primary follicle stage upon NFYA loss, as well as the NFYA binding and NFYA motif occurrence around the TSS of the corresponding genes. The downregulated genes are separated into two groups based on whether they have direct NFYA binding. (H) Heatmaps and density plots showing the enrichment and average ATAC–seq signal fold change (ATAC_FC) in the promoter regions of DEGs of primary follicles upon NFYA loss, respectively. The downregulated or upregulated genes are separated into two groups based on whether they have direct NFYA promoter binding. P values are calculated with two-sided Student’s t-test. (I) Examples of genome browser view of NFYA binding (shaded), ATAC and RNA-seq signals in WT and cKO primary follicles.

To understand how NFYA regulates PFA, we compared the transcriptomes of primordial and primary follicle oocytes and identified 1,888 genes activated during the primordial-to-primary transition (PPT), which we termed as early-PFA genes (Figure 2F and Table S2). Of these early-PFA genes, 389 (20.6%) failed to be activated in response to NFYA depletion (Figure 2F), accounting for 48.1% of the 809 downregulated genes (Figure S2I). A global disruption of early-PFA genes expression in Nfya-cKO primary follicle oocytes was further demonstrated by gene set enrichment analysis (GSEA) (Figure S2J). Among the 389 affected early-PFA genes, 130 gene promoters are directly bound by NFYA (Figure 2G), including Ezh2, Kdm4a, and Sdl2l1 (Figure S2K). Since low-input CUT&RUN data can inherently miss peaks leading to underestimation, we performed motif analysis on the remaining 259 genes and found that 79.9% (207/259) contained at least one NFYA motif in their promoter regions (Figure S2L). Collectively, these data suggest a direct role of NFYA in PFA.

Since NFYA is a pioneer factor and essential for the establishment of open chromatin ^15,27, we next asked whether NFYA regulates early-PFA genes by promoting their chromatin accessibility. Comparative ATAC-seq analysis (Figure S2M) revealed that NFYA loss significantly reduced chromatin accessibility at the promoters of NFYA-bound downregulated genes (for example Prkaca, Hspa2, Kdm4a, and Akap8l) (Figure 2H–I). Taken together, these data indicate that NFYA contributes to establishment of promoter chromatin accessibility, and that loss of NFYA leads to defective PFA during primordial-to-primary follicle transition.

NFYA deficiency impairs PFA in secondary follicle oocytes

After reaching primary follicle, PFA continues into secondary follicles to rapidly accumulate maternal mRNAs and proteins required for subsequent oocyte growth and development ^2,3. We asked whether NFYA loss would affect gene activation in secondary follicle oocytes. To this end, transcriptomic analysis was performed, which revealed 1,844 downregulated and 1,517 upregulated genes in secondary follicle oocytes in response to NFYA loss (Figure 3A and Table S3). To identify the direct effects caused by NFYA loss, we performed NFYA CUT&RUN in secondary WT follicle oocytes (Figure S3A) and found that NFYA mainly occupies the promoter regions with highly enriched for NFYA binding motif (Figure S3B–C). We found that 1,064 out of the 1,844 down-regulated genes bound by NFYA (Figure S3D and Table S3), which is consistent with its motif enrichment in these promoters (Figure S3D). Since loss function of NFYA affects PFA in primary follicle oocytes (Figure 2F), to identify the genes specifically affected in the secondary follicle oocytes, we separated the affected genes into two groups: those affected in both primary and secondary follicle oocytes, and those affected only in secondary follicle oocytes (Figure S3E). Of the 1,231 secondary-specific down-regulated genes in response to NFYA loss, 658 are bound by NFYA with 87.5% (576/658) affected genes contain at least one NFYA binding motif in their promoters (Figure 3B and S3F). GO analysis revealed that these downregulated direct targets are enriched for meiotic cell cycle, response to endoplasmic reticulum (ER) stress, and chromosome segregation (Figure S3G). Notably, these secondary-specific direct targets include many genes whose knockouts or mutations exhibit infertility phenotype such as Calr, Swsap1, Nlrp14, and Piwil2 (Figure 3C and S3H). For example, CALR is important for quality control of nascent proteins in the ER and for oocyte development ³² and Piwil2 are important for functional oocyte generation ^33,34, consistent with a critical role of NFYA in oocyte development.

Figure 3. NFYA deficiency impairs PFA in secondary follicle oocytes.

(A) Scatter plots comparing the gene expression profiles of WT and Nfya-cKO secondary follicle oocytes from P10 ovaries. The x and y axis of the dot plots are Log2CPM from RNA-seq. Fold change > 2, FDR < 0.05. (B) Heatmaps showing the DEGs at secondary follicle stage upon NFYA loss, and the NFYA binding, NFYA motif occurrence around the TSS of corresponding genes. Left panel: the downregulated or upregulated genes are separated into two groups based on whether they are affected at primary follicle stage. Right panel: the secondary stage specifically downregulated or upregulated genes are separated into two groups based on whether they have direct NFYA binding. (C) Genome browser view of NFYA targeted infertility related gene examples of NFYA CUT & RUN and RNA-seq of secondary follicles. (D) Heatmap showing late-PFA gene (n=834) expression in WT and Nfya-cKO oocytes at the indicated stages. (E) Heatmaps showing the downregulated late-PFA genes (n=548) at secondary follicle stage upon NFYA loss, and the NFYA binding around the TSS of corresponding genes. The downregulated genes are separated into two groups based on whether they have direct NFYA promoter binding. (F) Genome browser view of NFYA targeted late-PFA gene examples of NFYA CUT&RUN and RNA-seq in secondary follicles. (G) Density plots showing the average ATAC_FC signals in the promoter regions of secondary follicle-specific DEGs with or without NFYA promoter binding. P values were calculated with two-sided Student’s t-test. (H) Genome browser view of examples of the NFYA binding (shaded), ATAC and RNA-seq signals in WT and CKO secondary follicles. (I) The percentage of early- and late-PFA genes regulated by NFYA in primary and secondary follicle oocytes.

To determine the role of NFYA in regulating the primary-to-secondary follicle transition, we first focused on the 1,888 early-PFA genes (Figure 2F). We found that these genes are activated at primary follicle oocytes and their expression is maintained in secondary follicle oocytes (Figure S3I). Of these early-PFA genes, 661 are affected by NFYA deficiency, including 389 affected in the primary follicles and 576 affected in the secondary follicles with 304 commonly affected in both stages (Figure S3J). Of the 272 newly affected early-PFA genes, 110 (40.4%) gene promoters are bound by NFYA, including Pde6d, Myh10, Rell1, and Fam72a (Figure S3K). Notably, PDE6D has been shown to be important for cAMP-PKA and PI3K-AKT signaling, which are critical for PFA ^31,35, suggesting a critical role of NFYA in regulating PFA. Similarly, motif analysis of the 162 genes revealed 82.1% (133/162) contain at least one NFYA binding motif in their promoter regions (Figure S3L).

In addition to the early-PFA genes, we also identified 834 genes specifically activated during the primary-to-secondary transition by comparing the transcriptomes of primary and secondary follicle oocytes, which we termed as late-PFA genes (Figure 3D and Table S2). Of these late-PFA genes, 548 (65.7%) failed to be activated due to NFYA depletion (Figure 3D). A global disruption of late-PFA genes expression in Nfya-cKO secondary follicle oocytes was further validated by GSEA (Figure S3M). Among the 548 affected late-PFA genes, 229 gene promoters are directly bound by NFYA (Figure 3E), including Katnal2 and various histone genes (Figure 3F). Further motif analysis on the 319 genes revealed 84% (268/319) contain at least one NFYA binding motif in their promoter regions (Figure S3N). Collectively, these data suggest a direct role of NFYA in late-PFA regulation.

To explore whether NFYA activates late-PFA genes by promoting chromatin accessibility in secondary follicle oocytes, we performed ATAC-seq in Nfya-cKO and WT secondary follicle oocytes (Figure S3O). Comparative analysis showed that NFYA deficiency significantly reduced chromatin accessibility at the promoters of NFYA-bound downregulated genes (Figure S3D, right). Importantly, promoter accessibility of NFYA-bound secondary follicle oocyte-specific downregulated genes (n=658) also significantly decreased (for example Hist1h2bk/j, Nfe2l1, Xbp1, Ccnb2, Cdc23, and Top2a) (Figure 3G–H). Collectively, these data suggest that NFYA is essential for gene activation in secondary follicle oocytes through promoting chromatin accessibility.

In summary, data presented above demonstrate that NFYA is responsible for the activation of 35.01% early-PFA genes and 65.71% late-PFA genes (Figure 3I), supporting that NFYA serves as a critical activator of PFA by driving transcriptional activation.

NFYA loss triggers non-canonical ferroptosis leading to Nfya-cKO oocyte degeneration

Next, we asked how NFYA deficiency leads to the degeneration and elimination of early follicles (Figure 1K–M). Previous studies have demonstrated that oocyte elimination can be caused by apoptosis, autophagy, and ferroptosis ^36–38. Interestingly, GSEA revealed significant enrichment of ferroptosis-related pathways, but not apoptosis- or autophagy-related pathways, in Nfya-cKO oocytes compared to WT controls (Figure 4A and S4A). Notably, many ferroptosis suppressors, including Pdia4, Hspa5, Nfe2l1, and G3bp1 ^39–42, are downregulated in Nfya-cKO secondary follicle oocytes and are directly bound by NFYA (Figure 4B), suggesting that NFYA plays a critical role in protecting oocytes from ferroptosis. Consistently, the protein levels of apoptosis-related markers (γH2AX and cleaved caspase3) or autophagy-related marker LC3 are not changed in Nfya-cKO oocytes (Figure S4B–C). Consistent with the involvement of ferroptosis, the labile iron ions (Fe²⁺), a key marker of ferroptosis detected by FerroOrange, exhibit a markedly altered cytoplasmic distribution pattern in Nfya-cKO secondary follicle oocytes, but not in primary follicle oocytes (Figure 4C and S4D). Interestingly, the overall signal intensity of FerroOrangeis comparable between Nfya-cKO and WT oocytes (Figure S4E), diverging from canonical ferroptosis, which typically features increased Fe² levels ^38,43. Since Fe²-driven lipid peroxidation of cellular membranes is a key step in ferroptosis ^43,44, we examined lipid peroxidation using the BODIPY 581/591 C11 probe (a lipid peroxidation sensor) and found that Nfya-cKO oocytes exhibit significantly increased lipid peroxidation, particularly along the cell membrane (Figure 4D). Furthermore, as mitochondrial and ER membrane is highly sensitive to lipid peroxidation ^43,44, we examined mitochondria and ER membrane structure by electron microscope. The results showed loss of cristae structure and disruption of mitochondrial membranes in Nfya-cKO secondary follicle oocytes (Figure 4E). We also observed disorganized ER, which stacked together, formed a big circler shape, and contacted with disrupted mitochondria (Figure 4F–G), similar to the previously reported ER disorganization induced by strong ER stress ^45,46. Consistently, GO analysis of NFYA-bound downregulated genes showed enrichment of ER stress in Nfya-cKO oocytes (Figure S3G).

Figure 4. NFYA loss triggers non-canonical ferroptosis in growing oocytes.

(A) Gene set enrichment analysis (GSEA) revealed the ferroptosis suppressors are significantly decreased in Nfya-cKO oocytes. (B) Genome browser view of examples of NFYA CUT&RUN and RNA-seq of NFYA targeted ferroptosis suppressors in secondary follicles. (C) Representative images of FerroOrange (an Fe²⁺ indicator) staining of WT and Nfya-cKO secondary follicle oocytes reveal a change in staining patterns. Scale bar, 10 gm. (D) Representative images of BODIPY 581/591 C11 probe staining of oxidized lipids and reduced lipids in WT and Nfya-cKO secondary follicle oocytes. Scale bar, 10 pm. (E) Electron Microscopy (EM) images showing the cross-section of mitochondria (M) in WT and Nfya-cKO secondary follicle oocytes. The arrowhead indicates damaged mitochondria membrane. Scale bar, 1 pm. (F) Representative images of Endoplasmic Reticulum (ER) tracker labeling in WT and Nfya-cKO secondary follicle oocytes reveal morphological changes in ER. Scale bar, 10 pm. (G) EM images showing the cross-section of ER in WT and Nfya-cKO secondary follicle oocytes. Right are the high-magnification EM images of left boxed areas showing the ER-mitochondria contact. Scale bar, 1 pm. (H) Schematic diagram of an in vitro 3D follicle culture experiment with the isolated WT follicles treated with DMSO and the Nfya-cKO follicles treated with DMSO or Fer-1. (I) Left panel: representative images of a WT or Nfya-cKO follicle at day 0 or day 6 after culturing with DMSO or Fer-1. Scale bar, 50 pm. Right panel: Survival curves of WT and Nfya-cKO follicles in culture supplemented with DMSO or Fer-1. P values, log-rank test.

To determine whether ferroptosis plays a causal role in Nfya-cKO oocyte degeneration, we performed a rescue experiment by culturing follicles in a 3D in vitro system for 8 days with or without the presence of a ferroptosis antagonist ferrostatin-1 (Fer-1) (Figure 4H). We found that the presence of Fer-1 significantly reduced NFYA deficiency-induced oocyte degeneration (Figure 4I) supporting a causal role of ferroptosis in Nfya-cKO oocyte degeneration. Taken together, these results suggest that NFYA loss triggers non-canonical form of ferroptosis in oocytes, characterized by decreased expression of ferroptosis suppressors, abnormal cytoplasmic Fe²⁺ distribution, ER disorganization, excessive lipid peroxidation, and rupture of cellular membrane.

dTAG-based NFYA degradation impairs ZGA

After demonstrating NFYA is a key regulator of PFA, we next asked whether NFYA is also required for ZGA. Although our previous studies have suggested a role of NFYA in ZGA through siRNA-mediated Nfya knockdown in cultured GV oocytes ¹⁵, the potential for off-target effect of siRNA or secondary effects of GV oocyte culturing could not be excluded. Despite the requirement of NFYA for normal oocyte development ruled out the use of the Nfya-cKO model to address its role in ZGA, the recently developed rapid protein degradation dTAG system and its successful application in mouse embryos ^18,19 have made the use of such approach to address the role of NFYA in ZGA possible.

To this end, we designed an N-terminal HA (hemagglutinin)- FKBP (FK506 binding proteins)-NFYA fusion construct (referred to as NFYA^dTAG) to enable conditional degradation (Figure S5A). After confirming the stability of this fusion protein and the efficiency of dTAG13-induced degradation in mouse ES cells (mESCs) (Figure S5B–C), we generated the Nfya^dTAG mouse line using the CRISPR technology ⁴⁷ (Figure S5D–E). Notably, these mice are fertile and developmentally normal, with an average little-size of 10 pups based on 11 litters obtained from Nfya^dTAG/dTAG intercrosses (Figure S5E), indicating that the dTAG knock-in does not impair NFYA function under physiological conditions. We then confirmed that dTAG13 treatment can efficiently deplete NFYA from cultured embryos at late 2-cell (L2C) stage (Figure 5A). Collectively, these results demonstrate the successful generation of a NFYA^dTAG mouse model.

Figure 5. dTAG13-mediated NFYA depletion causes ZGA defects.

(A) Immunostaining confirming NFYA degradation by dTAG13 treatment during ZGA. Anti-HA antibody was used for endogenous NFYA^dTAG fusion protein detection. Scale bar, 25 pm. (B) Left upper panel: schematic diagram of the experimental design. Left lower panel: The Nfya^dTAG/dTAG (Homo) embryo development rate upon DMSO or dTAG13 treatment. The data are presented as mean ± SEM. Right panel: representative images of blastocyst stage embryos after NFYA degradation (DMSO, n = 83; dTAG13, n = 90; n represents the total number of embryos from three independent experiments). Scale bar, 100 pm. Red arrows indicate representative arrested embryos in dTAG13 group. (C) Scatter plots comparing the gene expression profiles of Homo late-2cell embryos treated with dTAG13 or DMSO. The x and y axis of the dot plots are Log2CPM from RNA-seq. Fold change > 2, FDR < 0.05. A total of 2,773 major ZGA genes are analyzed (Table S2). (D) Heatmaps showing the Obox and Dux family gene expression in late 2cell embryos treated with dTAG13 or DMSO. (E) Venn diagram showing the overlapped downregulated ZGA genes in L2C embryos among Obox mzKnockout ²³, Dux knockout ⁵⁷, and NFYA depletion (NFYA-dTAG). (F) Heatmaps showing the commonly downregulated ZGA genes (n=625) in OBOX-mzKO and NFYA-dTAG, and the motif occurrence of NFYA and OBOX around the TSS of the corresponding genes. (G) Bar chart showing the downregulated gene numbers based on the distance from their TSS to nearest NFYA or OBOX motif. (H) Cumulative distributions of the distance from commonly downregulated gene TSS to the nearest NFYA or OBOX motif. (I) The genomic distribution of NFYA binding peaks generated by NFYA CUT&RUN using late 2-cell embryos. (J) Heatmaps showing the downregulated major ZGA genes (n=1,067) at late 2-cell embryos upon NFYA loss, and the NFYA binding, NFYA motif occurrence around the TSS of corresponding genes. The downregulated major ZGA genes are separated into two groups based on whether they have direct NFYA promoter binding. (K) Genome browser view examples of NFYA CUT&RUN and RNA-seq in NFYA promoter targeted ZGA genes at late-2cell stage. (L) Heatmaps showing the downregulated major ZGA genes (without NFYA promoter binding) with predicted P-E interactions (n=650) at late 2-cell stage upon NFYA loss, and the NFYA binding around the center of enhancers. The downregulated major ZGA genes are separated into two groups based on whether their putative enhancers have direct NFYA binding. (M) Genome browser view examples of NFYA CUT&RUN and RNA-seq in NFYA distal targeted ZGA genes at late-2cell stage.

To evaluate the role of NFYA in ZGA, we treated Nfya^dTAG/dTAG (Homo) embryos with dTAG13 and found that NFYA-depletion caused half of the embryos arrest at 2-cell stage and ~90% embryos failed to reach blastocyst stage (Figure 5B and S5F), suggesting a critical role of NFYA in ZGA and preimplantation development. To directly address the role of NFYA in ZGA, we collected dTAG13-treated Homo late 2-cell (L2C) embryos and performed RNA-seq (Figure S5G), which revealed 2,317 downregulated and 1,313 upregulated genes relative to that of the control (Figure 5C, left and Table S4). Among the downregulated genes, 1,067 are major ZGA genes (Figure 5C, right and Table S4), indicating that NFYA has a role in activating these ZGA genes. Importantly, NFYA degradation did not affect the expression of OBOX or DUX gene families (Figure 5D), suggesting that NFYA regulates ZGA independently of OBOX and DUX. Furthermore, an integrative analysis of DEGs showed that 40% of NFYA-regulated ZGA genes (432 out of 1,067) do not overlap with those of OBOX- or DUX-regulated ZGA genes, including Hspa2, Eif2ak3, G3bp1, and various histone genes (Figure 5E). Intriguingly, although we found that 625 major ZGA genes were regulated by both NFYA and OBOX (Figure 5E–F), motif analysis revealed distinct binding patterns with NFYA motifs mainly enriched near their core promoter regions (TSS ± 500bp), while OBOX motifs are dispersed throughout the gene body (Figure 5G–H), suggesting that NFYA most likely regulates these genes through promoter binding, whereas OBOX may act through enhancer binding. Collectively, these data demonstrate that NFYA is critical for ZGA and preimplantation development.

NFYA activates ZGA genes by promoter and distal binding

To determine whether NFYA directly binds to and regulates major ZGA genes, we performed low-input CUT&RUN for NFYA in L2C embryos. For quality control purpose, we confirmed the reproducibility and compared our low-input L2C dataset with that of bulk mESC dataset and found most of the major peaks are similar (Figure S6A–C). Moreover, the most enriched motif in L2C peaks is the classical NFY binding sequences CCAAT (Figure S6D). Interestingly, analysis of the CUT&RUN data revealed that more than half of the NFYA binding peaks are located in distal regions (putative enhancers) (Figure 5I), which is different from the binding patterns in oocytes (Figure S2E and S3B) or mESCs (Figure S6E) where it mostly binds to promoters. To verify that the distinct NFYA-binding patterns observed in L2C embryo are genuine, we performed NFYA CUT&RUN in NFYA-depleted L2C embryos and found that both promoter and distal NFYA-binding peaks disappeared in NFYA-depleted L2C embryos (Figure S6F), suggesting that NFYA indeed exhibits preferential binding to distal regions in L2C. To investigate whether NFYA-bound distal regions are enriched for transposable elements, we performed analysis which revealed that NFYA-bound sites are preferentially associated with LTR, SINEs, and snRNA repetitive elements (Figure S6G). Further analysis showed that NFYA binding is significantly more enriched on downregulated ZGA genes compared to other downregulated genes (Figure S6H), suggesting that NFYA preferentially binds and regulates ZGA genes. Integrative analyses of CUT&RUN and RNA-seq data revealed that many downregulated major ZGA genes (323 out of 1,067) have promoter NFYA binding, including Hspa2, Eif2ak3, G3bp1, Hmgn5, and various histone genes (Figure 5J–K and S6I). Further motif analysis of the 744 down-regulated genes without CUT&RUN signal revealed 72% (538/744) contain at least one NFYA motif in their promoter regions (Figure S6J). Collectively, these data suggest a direct role of NFYA-promoter binding in ZGA regulation.

To further characterize NFYA distal binding, we utilized the activity-by-contact (ABC) model ⁴⁸ to predict potential NFYA-mediated promoter-enhancer (P-E) interactions and found that 650 out of 744 downregulated ZGA genes are correlated with predicable enhancer activity (Figure S6K). Further analysis indicated that many NFYA-bound enhancers are linked to downregulated major ZGA genes (218 out of 650), including Ccne1, Hist1h2af, Sael, and Polr3g (Figure 5L–M and S6L). Collectively, our integrated analysis supports that NFYA directly regulates ZGA genes by both promoter and distal binding.

To explore whether NFYA regulates ZGA genes by promoting chromatin accessibility in L2C, we performed ATAC-seq in L2C embryos with or without dTAG13 treatment (Figure S6M). Comparative analysis showed that NFYA depletion reduced chromatin accessibility at the promoters of downregulated ZGA genes (Figure S6N), suggesting that NFYA contributes to gene activation during ZGA by promoting chromatin accessibility.

NFYA regulates PFA and ZGA by multifaceted chromatin binding

Having demonstrated that NFYA serves as a unified regulator for both PFA and ZGA through its promoter and distal chromatin binding, we next sought to examine the conserved and distinct roles of NFYA in these two major genome transcriptional activation events. By comparing NFYA binding profiles in oocytes and L2C embryos, we found that 60.9% oocyte-specific NFYA binding is located at promoter regions, whereas 85.5% L2C-specific NFYA binding is located at distal regions (Figure 6A), underscoring the distinct preference of NFYA-chromatin interactions during PFA and ZGA. Intriguingly, GO analysis of the stage-specific NFYA binding related genes revealed that oocyte-specific NFYA bound genes were associated with “response to hormone stimulus” and “meiotic cell cycle”, whereas L2C-specific NFYA bound genes were associated with “stem cell maintenance” and “RNA splicing” (Figure S7A), suggesting distinct functions of stage-specific NFYA-chromatin interactions. Furthermore, our integrative analysis of CUT&RUN and ATAC-seq revealed that the L2C-specific NFYA binding sites are largely inaccessible in oocytes (Figure S7B), suggesting that stage-specific chromatin accessibility may contribute to the differential NFYA binding landscape. Additionally, motif enrichment analysis of the NFYA oocyte- and L2C-specific binding regions revealed that, in addition to the enrichment of NFYA motif, TFs such as SP1, KLF15, and CTCF are enriched at oocyte-specific NFYA-binding regions, whereas NR5A2, RARG, and OTX2 are enriched at L2C-specific regions (Figure 6B), suggesting that these stage-specific factors may help shape the multifaceted NFYA chromatin binding profiles during PFA and ZGA.

Figure 6. NFYA regulates PFA and ZGA by multifaceted chromatin binding.

(A) Heatmaps showing the dynamics of NFYA genomic binding profile during PFA (primordial follicle oocytes, primary follicle oocytes, and secondary follicle oocytes) and ZGA (zygotes and late 2cell embryo) at promoter and distal regions. Pie chart showing the percentage of promoter and distal occupancy. (B) Enrichment and expression of top TF motifs at the CUT&RUN peak groups (oocyte-specific, L2C-specific, and common). (C) Heatmaps showing the gene expression of the 920 common binding associated genes (n=1,351) in primary follicle embryos, secondary follicle embryos, and late 2-cell embryo upon NFYA loss. These genes are separated into two groups: commonly downregulated genes in both oocytes and L2C (n=215) and other genes (n=1,136). (D) Representative GO (BP) terms enriched in the commonly downregulated genes (n=215). The example genes of the highlighted term are shown. (E) Representative GO (CC and MF) terms enriched in the commonly downregulated genes (n=215). (F) Genome browser view of NFYA CUT&RUN and RNA-seq of example NFYA target genes common in oocytes and L2C.

Pre-occupied NFYA regulates chaperones and histone genes common in both PFA and ZGA

In addition to the stage-specific binding, we also identified 920 common NFYA binding regions, predominantly (90.8%) located at promoter regions (Figure 6A). Intriguingly, the 920 common bound regions not only showed stronger NFYA binding signals and motif enrichment (Figure 6A–B), but also were already pre-occupied in both the primordial follicle stage and the zygotic stage before PFA or ZGA begins, whereas most of the oocyte- or L2C-specific NFYA bound regions were gradually established during transcriptional activation (Figure 6A and S7C). Consistent with the pre-occupancy, NFYA immunostaining also showed nuclear enrichment at primordial follicle stage and zygotic stage (Figure 1F). Consequently, these 920 common NFYA bound regions were accessible before PFA or ZGA begins (Figure S7B), and NFYA-depletion reduced their accessibility in both oocytes and L2C (Figure S7D), suggesting that NFYA pre-occupancy is important for the chromatin opening at those regions. Intriguingly, we found that chromatin accessibility was reduced more at the common bound regions compared with the oocyte- or L2C-specific bound regions (Figure S7E), suggesting that NFYA is more important for maintaining chromatin accessibility at the commonly bound regions than at stage-specific bound regions. In sum, these findings suggest that NFYA plays a more important role in the common and pre-occupied regions.

To further characterize the 920 common NFYA pre-bound regions, we performed integrative analyses of CUT&RUN and RNA-seq, and identified 1,351 associated genes. Among them, 215 are commonly downregulated in both PFA and ZGA upon NFYA-depletion (Figure 6C). GO analysis revealed that they are associated with Biological Processes, such as nuclear organization, ER response, and protein folding (Figure 6D); Cellular Components, such as chromosomal region and endoplasmic reticulum; and Molecular Functions, such as ER chaperone complex and heat shock protein binding (Figure 6E), including chaperones and histone genes (Figure 6F). Collectively, these results suggest that NFYA regulates both PFA and ZGA through a combination of divergent and conserved chromatin binding. Additionally, pre-occupied NFYA regulates chaperones and histone genes common in both PFA and ZGA.

Inhibition of chaperones impairs both PFA and ZGA

Previous studies have established a role of histone genes in genome activation by affecting nuclear organization ⁴⁹. However, the role of chaperones in PFA or ZGA has not been explored. To investigate the function of NFYA-regulated chaperones in PFA and ZGA, we utilized two widely used chaperone inhibitors Ganetespib (1 nM~150 nM for cells) ^50,51 and VER155008 (1 μM~50 μM for cells) ⁵² that respectively inhibit HSP90 or HSPA2/5. To optimize their concentration for mouse embryo treatment, we tested different concentration of Ganetespib (2 nM, 5 nM, 10 nM, and 25 nM) and VER155008 (1 μM, 5 μM, 10 μM, and 25 μM) (Figure S7F) and found 10 nM Ganetespib or 10 μM VER155008 caused 2-cell embryo arrest, while 5 nM Ganetespib and 5 μM VER155008 had only a minor effect on blastocysts development. Since NFYA-depletion downregulated both HSPA2/5 and HSP90, and to avoid potential off-target effects caused by high concentration inhibitor treatment, we performed the treatment by combining 5 nM Ganetespib and 5 μM VER155008 (referred to as G&V hereafter). We found that only the treatment before ZGA (G&V 1C-B), but not after ZGA (G&V 4C-B), resulted in most embryos arrest at the 2-4 cell stage (Figure 7A), suggesting a specific role of chaperones in ZGA. To directly address the role of chaperones in ZGA, we collected G&V-treated L2C embryos and performed RNA-seq (Figure S7G), which revealed 705 downregulated and 718 upregulated genes compared to that of the control (Figure 7B, left and Table S5). Among the downregulated genes, 181 are major ZGA genes (Figure 7B, right and Table S5), indicating that chaperones have a role in activating these ZGA genes. Moreover, half of the downregulated ZGA genes were overlapped with NFYA-activated ZGA genes, including cell fate regulator: Tead4 and Fgf4; and RNA processing factors: Ddx49, Rc3h2, and Mov10l1 (Figure 7C), suggesting that chaperones contribute to ZGA. Collectively, these data demonstrate that chaperones play an important role in ZGA and preimplantation development.

Figure 7. Inhibition of chaperones impairs both PFA and ZGA.

(A) Left upper panel: schematic diagram of the experimental design. Left lower panel: the embryo development rate upon DMSO or G&V treatment (Ganetespib 5nM and VER155008 5μΜ). The data are presented as mean ± SEM. [DMSO, n = 67; G&V (1C-B), n = 64; G&V (4C-B), n = 47. n represents the total number of embryos from two independent experiments]. Right panel: representative images of blastocyst stage embryos after chaperones inhibition (G&V). Scale bar, 100 pm. Red arrows indicate representative arrested embryos in G&V (1C-B) group. (B) Scatter plots comparing gene expression profiles of late-2cell embryos treated with G&V or DMSO. The x and y axis of the dot plots are Log2CPM from RNA-seq. Fold change > 2, FDR < 0.05. ZGA genes, n=2,773. (C) Venn diagram showing the overlapped downregulated ZGA genes between NFYA depleted (NFYA-dTAG) and G&V treated L2C embryos. P values (Fisher’s exact test, two-sided) for overlapped genes are also shown. (D) Upper panel: schematic diagram of the experimental design. Lower left panel: ovarian morphology of 6 days in vitro cultured P4 ovaries with DMSO and G&V treatment. DDX4 stained oocytes. The white arrowhead indicates representative growing oocytes. Scale bar, 50 pm. Lower right panel: numbers of growing oocytes (DDX4 positive oocytes bigger than 25 pm in diameter that were surrounded by cuboidal granulosa cells) in each ovary after 6 days culture with DMSO and G&V treatment. Quantitative data are shown as mean ± SEM; **p < 0.01, Student’s t test. (E) Scatter plots comparing the gene expression profiles of growing oocytes (GOs) from 6 days in vitro cultured P4 ovaries with G&V or DMSO. The x and y axis of the dot plots are Log2CPM from RNA-seq. Fold change > 2, FDR < 0.05. PFA genes, n=2,463. (F) Venn diagram showing the overlapped downregulated PFA genes between Nfya-cKO and G&V treated GOs. P values (Fisher’s exact test, two-sided) for overlapped genes are also shown. (G) Model illustrating the role of NFYA in PFA and ZGA. Before PFA (primordial), the PFA genes are silenced. Subsequently (primary and secondary), NFYA binding promotes open chromatin and activates PFA genes through promoter and distal binding. Loss of NFYA leads to reduced chromatin accessibility, defective PFA, and early follicular degeneration. Before ZGA (1C), the ZGA genes are silenced. Subsequently (L2C), NFYA activates ZGA genes through promoter and enhancer binding. Loss of NFYA leads to defective ZGA and predominantly embryo arrest at 2-cell stage. (H) NFYA pre-occupies and regulates a set of genes, including chaperones and histone genes, common in both PFA and ZGA through conserved promoter binding.

Similarly, to test chaperones’ role in PFA, we treated P4 neonatal ovaries, when PFA begins in vivo, with G&V and found a significant reduction in the numbers of growing oocytes (GOs), which were marked by DDX4 that were bigger than 25 μm in diameter and surrounded by cuboidal granulosa cells, compared to that of the controls at day 6 (Figure 7D), suggesting an important role of chaperones in PFA. Notably, folliculogenesis defects are more severe at day10 (Figure S7H) with significant loss of central growing follicles, suggesting defective PFA and oocyte arrest. To investigate whether PFA genes are affected by chaperone inhibition, we collected growing oocytes from G&V-treated (6 day) ovary and performed RNA-seq (Figure S7I), which revealed 1,244 downregulated genes relative to that of the control (Figure 7E, left and Table S6). Among the downregulated genes, 833 are PFA genes (Figure 7E, right and Table S6), indicating that chaperones have an important role in activating these PFA genes. Moreover, 266 out of the 833 downregulated PFA genes were overlapped with NFYA-activated PFA genes, including Oog2/3/4 and NLRP family (Figure 7F), suggesting that chaperones may contribute to the regulation of these PFA genes. Taken together, these data suggest that chaperones are critical for both PFA and ZGA.

Collectively, our data support a unified model for NFYA’s function in transcriptional activation during PFA and ZGA: 1) NFYA pre-occupies and regulates a group of common genes including those encoding chaperones and histones before genome activation; and 2) NFYA gradually establish bindings on oocyte- or L2C-specific genes to activate these stage-specific genes during the two genome activation events.

Discussion

Although many shared molecular hallmarks between PFA and ZGA have been noticed ^6,53, a unified regulator of these two events have not been identified. Here, we demonstrate that NFYA is the long sought-after TF involved in both PFA and ZGA, the two transcriptional activation events critical for oocyte-to-embryo transition (Figure 7G–H). Mechanistically, NFYA exhibits distinct chromatin-binding preferences with predominant promoter targeting during PFA and enhancers targeting during ZGA, thereby regulating transcriptional programs in a cell type- and developmental stage-specific manner. On the other hand, conserved NFYA-chromatin binding pre-occupies and regulates a set of common genes, including chaperones and histone genes, in both PFA and ZGA. Phenotypically, oocyte-specific deletion of Nfya compromised both chromatin opening and transcriptional activation during PFA, which triggers non-canonical ferroptosis leading to early folliculogenesis failure. Additionally, acute depletion of NFYA in zygotes impairs ZGA and causes predominantly embryo arrest at 2-4 cell stage. Intriguingly, inhibition of chaperones impairs both PFA and ZGA. Together, our study demonstrates NFYA as a unified and multifaced regulator of PFA and ZGA that orchestrates the oocyte-to-embryo transition.

Role of NFYA in quiescent primordial follicle activation

In mammalian females, one of the major functions of ovary is to maintain a balance between quiescent and activated primordial follicle pool. The maintenance of quiescent primordial follicles is controlled by a regulatory network, including transcription factors such as FOXO3A ⁸ and LHX8 ^9,10; and the PTEN-PI3K signaling pathway ⁷. Knockout these “repressors” results in excessive activation of PFA genes and subsequent depletion of primordial follicles. However, the identity of pioneer factors that directly activate PFA in vivo remains largely unknown. Although NOBOX has been reported to be critical for PPT ⁵⁴, extensive transcriptional defects and dysregulation of early germ cell genes have been observed in oocytes from newborn ovaries before primordial follicle formation ⁵⁵, suggesting that the defective PPT might be a secondary effects caused by earlier germ cell developmental abnormalities. In this study, we provide evidence demonstrating that NFYA directly activates a subset of PFA genes through direct promoter binding (Figure 2–3), making NFYA the first identified TF involved in mammalian PFA. Oocyte-specific NFYA deficiency impairs PFA resulting in depletion of primary and secondary follicles (Figure 1), supporting a critical role of NFYA in PFA. These observations raise several mechanistic questions, including how NFYA deals with the “repressors” to drive primordial follicles out quiescence. Intriguingly, an interaction between NFYA and FOXO3A has been reported in other cell types ⁵⁶, suggesting that NFYA and FOXO3A may coordinately regulate the balance between quiescence and activation of primordial follicles. Further study characterizing their common targets during the quiescence and activation transition may increase our understanding of how the balance of quiescence and activation of primordial follicles is controlled at the molecularly level.

Role of NFYA in ZGA

Great progress has been made in the past several years in identifying TFs important for mammalian ZGA, such as OBOX ²³, DUX ⁵⁷, NR5A2 ^24,25, GABPA¹⁹, and KLF17 ⁵⁸, although none of them has a demonstrated function in PFA. Despite our previous studies have implicated the function of NFYA in ZGA ¹⁵, definitive evidence has been lacking. Here we provide several lines of definitive evidence demonstrate its role in ZGA. First, using a dTAG protein degradation system, we demonstrate 1,067 ZGA genes are down-regulated in response to NFYA degradation (Figure 5C), a number similar to that affected by OBOX ²³. Second, many of the down-regulated genes are direct targets with NFYA bound to their promoters or distal regions (Figure 5J–M and S6I–L). Third, acute NFYA depletion before ZGA results in half of the embryos arrest at 2-cell stage, consistent with dysregulation of ZGA.

The identification of cell type-specific or developmental stage-specific TFs has advanced our understanding of the intricate and diverse regulatory networks governing gene expression, but the conserved and unified mechanisms underlying similar transcriptional programs across different contexts are often overlooked. Notably, the conserved NFYA-chromatin interactions for activating chaperone and nucleosome organization related genes that we uncovered underscore the importance of these processes in both PFA and ZGA (Figure 6–7). Intriguingly, it was recently demonstrated that successful ZGA is closely associated with proper nucleosome organization ⁴⁹. Although chaperones are reported as the first major products of mouse ZGA ⁵⁹, their roles in oocytes and early preimplantation development remain unclear. Here, our chaperone inhibition experiments demonstrate that chaperones are critical for both PFA and ZGA. Given that comprehensive epigenetic reprogramming accompanies both PFA and ZGA ^6,60, it’s imperative to understand the interplay between NFYA-chromatin binding and epigenetic reprogramming during these processes. Since NFYA is evolutionarily conserved in all eukaryotes ⁶¹, functional demonstration of its evolutionarily conserved regulator roles in ZGA or PFA in other species await to be shown.

NFYA activates transcription programs across diverse biological contexts

Beyond PFA and ZGA, transcription reactivation also takes place following mitosis, termed post-mitotic reactivation, which restores gene expression in the highly compacted and transcriptionally quiescent metaphase genome ⁶². Recent studies have suggested a role of NFYA in post-mitotic reactivation ⁶³. Combining with its role in PFA and ZGA we have shown in this study, NFYA appears to be a general regulator of genome reactivation across diverse cellular contexts. Notably, activation of dormant embryos also involves genome-wide transcriptional activation ^64,65, whereas it remains unclear whether NFYA plays a critical role in this process. Although the activation of quiescent hematopoietic stem cells (HSCs) and tissue regeneration does not involve genome-wide transcriptional activation to the extent observed in PFA and ZGA, these processes also involve transcriptional reprogramming, cell cycle transitions, and rapid protein synthesis and folding ^66,67. Notably, NFYA has also been shown to regulate these regenerative processes ^68,69, suggesting a broader role of NFYA in reactivating transcription programs. Intriguingly, while loss of NFYA in various biological processes has been associated with the induction of apoptosis ^46,68,69, our results showed that NFYA deficiency in oocytes triggers ferroptosis instead of apoptosis. This divergence underscores the context-dependent roles of NFYA in regulating cell death decisions.

Limitations of the study

We showed that NFYA loss triggers non-canonical ferroptosis in oocytes, which is characterized by decreased expression of ferroptosis suppressors, redistribution of cytoplasmic Fe2+, excessive lipid peroxidation, disorganization of ER, and mitochondria damage. We propose that NFYA deficient oocytes could serve as a good model for understanding non-canonical ferroptosis, which is featured by redistribution of cytoplasmic Fe2+ rather than elevation of cellular Fe2+ levels (Figure 4C and S4D–E). Although we found many suppressors of ferroptosis are down-regulated by NFYA loss, further studies are needed to elucidate the mechanisms by which NFYA loss leads to redistribution of cytoplasmic Fe2+ and ER disorganization. Additionally, the mechanisms linking altered cytoplasmic distribution of Fe2+ to the initiation of ferroptosis require further investigation.

While our analysis has been mainly focused on the down-regulated NFYA direct targets, we also observed NFYA binding on many upregulated genes in Nfya-cKO oocytes (Figure 2C and 3B), indicating that NFYA could also have transcriptional repressor function. Further investigations are needed to understand how NFYA functions as a transcriptional repressor in oocytes.

While we showed that even milder chaperone inhibition impairs both PFA and ZGA (Figure 7A–F), further genetic manipulation studies are needed to understand how chaperones regulate both PFA and ZGA.

Resource availability

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Yi Zhang (yzhang@genetics.med.harvard.edu)

STAR Methods

Nfya-cKO and Nfya-dTAG mouse generation

All animal experiments were performed in accordance with the protocols of the Institutional Animal Care and Use Committee at Harvard Medical School. All mice were kept under specific pathogen-free conditions within an environment controlled for temperature (20-22°C) and humidity (40-70%), and were subjected to a 12 h light/dark cycle. Nfya^flox/flox mice were obtained from Nobuyuki Nukina lab ⁴⁶. Gdf9-Cre [Tg(Gdf9-icre)5092Coo/J] mouse lines were purchased from the Jackson Laboratory (011062). The Gdf9-Cre mice were crossed with Nfya^flox/flox mice to generate Gdf9-Cre Nfya^flox/flox (Nfya-cKO) mice. All mice had a C57BL/6J genetic background. Primers used for genotyping are listed in Table S7.

Nfya-dTAG knock-in mice was generated following a previous protocol with modifications ¹⁹. Briefly, 2-cell embryo (20hpf) were injected with Nfya donor DNA (30 ng/μΐ), Cas9 protein (25 ng/μΐ) and sgRNA (50 ng/μl each) using a Piezo impact-driven micromanipulator (Primer Tech, Ibaraki, Japan). After 2 h KSOM incubation, 2-cell embryos were transferred into oviducts of pseudo-pregnant ICR strain mothers (Charles River). F0 chimera mice was backcross with wildtype CD1 mice for at least two generation. Genotyping was performed with mouse tail lysed in lysis buffer (50 mM Tirs-HCl, 0.5% Triton, 400 gg/ml Proteinase K) at 55°C overnight. For F0 and F1 mice genotyping, the primers outside the homology arm are used. For F2 and beyond mice genotyping, the inner primers are used. The primers are included in Table S7.

Droplet Digital PCR (ddPCR) was used for detecting the copy number of Nfya-dTAG knock-in allele in F1 mice. Briefly, 250 ng purified DNA template was digested by incubation with Haelll Enzyme (NEB) at 37°C for 1h, and then inactivated at 80°C for 5 mins. A final 30 ng DNA was used as template for PCR. Fkbp was used for knock-in detecting, mRPP30 was used as internal control of two copy genome. The PCR primers are listed in Table S7. The mice with single Fkbp copy were used for subsequent mating (Table S8).

Preparation and collection of mouse oocytes and embryos

All experiments involved in oocytes and embryo preparation were performed as previously described ^19,21, with minor modifications. Briefly, primary and secondary follicle oocytes were isolated from mice at postnatal day 7 or day 10. To capture the earlier molecular changes that lead to growing oocytes defects, we collected primordial and primary follicle oocytes at postnatal day 7, when the primary follicle oocytes exhibit no morphological difference between WT and Nfya-cKO. Similarly, we collected secondary follicle oocytes at postnatal day 10.

Female mice (7-8 weeks) were super-ovulated through an initial injection of 5 IU pregnant mare serum gonadotropin (PMSG, BioVendor, RP1782725000), followed by a 5 IU injection of human chorionic gonadotropin (hCG, Sigma, C1063) 48 hours later. Oocyte-cumulus complexes were collected 14 hours post hCG injection. Sperms were collected from the cauda epididymis of adult male mice (8-12 weeks). The sperm suspension was capacitated for 1 hour in 200 μl HTF medium (Millipore, MR-070-D). Oocytes were then incubated with spermatozoa for 6-hours. The time when sperms were added to oocytes was considered as 0 hpf. Two-nuclear zygotes were cultured in the KSOM medium (Millipore, MR-106-D) under a humidified atmosphere of 5% CO₂ at 37°C for further development.

dTAG13 treatment

dTAG13 (Tocris, 6605) was reconstituted in DMSO to a 50 mM stock. For Nfya^dTAG/dTAG mESCs treatment, dTAG13 was dilute in mESCs culture medium to 0.5 μM. For Nfya^dTAG/dTAG embryos treatment, dTAG13 was dilute in KSOM to 0.5 μM. Embryos were washed with KSOM with dTAG13 for at least three times, then cultured in KSOM with dTAG13 for further development.

Chaperone inhibitor treatment

Ganetespib (MCE, HY-15205) and VER155008 (Selleckchem, S7751) were reconstituted in DMSO to a 100 mM or 150 mM stocks, respectively. For embryos treatment, Ganetespib and VER155008 were dilute in KSOM to 5 nM and 5 μM, respectively. Embryos were washed with KSOM with inhibitors for at least three times, then cultured in KSOM with Ganetespib and VER155008 for further development. L2C embryos were collected for RNA-seq. For ovary treatment, P4 ovaries were dissected and cultured in media [DMEM-F12 media supplemented with 3 mg/ml bovine serum albumin (BSA), 10% fetal bovine serum (FBS) and 10 mIU/ml follicular stimulating hormone (FSH)] with or without 5 nM Ganetespib and 5 μM VER155008 for further development. Growing oocytes (diameter > 25 μm) from 6 days cultured P4 ovaries were collected for RNA-seq.

Cell culture and establishment of Nfya^dTAG mESC cell line

Undifferentiated E14 mESCs were cultured on 0.1% gelatin coated plates with 2i and LIF condition. E14 mESCs were grown in DMEM (Gibco, 11960069), supplemented with 15% fetal bovine serum (Sigma-Aldrich, F6178), 2 mM GlutaMAX (Gibco, 35050061), 1 mM sodium pyruvate (Gibco, 11360), 1x MEM NEAA (Gibco, 11140050), 0.084 mM 2-mercaptoethanol (Gibco, 21985023), 1 mM sodium pyruvate (Gibco, 11360070), 100 U/ml penicillinstreptomycin (Gibco, 15140122), 1000IU/ml LIF (Millipore, ESG1107), 0.5 μM PD0325901 (Tocris, 4192), and 3 μM CHIR99021 (Tocris, 4423). To establish Nfya^dTAG mESC cell line, Nfya donor DNA and Nfya-sgRNA inserted pX330.puro (addgene, #110403) plasmids were co-transfected with Lipofectamine^™ 2000 (Invitrogen, 11668030). Cells were then selected with puromycin (Gibco, A1113803), and the single cell-derived colonies were then picked for genotyping and further analysis. Genotyping primers are listed in Table S7. Cells were cultured in an incubator containing 5% CO₂ at 37 °C.

Western blot

Cells were lysed in 1 x LDS buffer (Invitrogen, NP0007) at 1 × 10⁴ cell/μL final concentration and heated at 100 °C for 10 min. Samples were run on NuPAGE 4-12% gel (Invitrogen, NP0322BOX) and transferred onto PVDF Transfer Membrane. Primary antibodies used included anti-NFYA (1:250, sc-17753X, lot#D0522, Santa Cruz Biotechnology), anti-HA (1:1000, 3724S, Cell Signaling Technology), and anti-GAPDH (1:50000, 60004-1-Ig, Proteintech). Protein bands were detected with ECL kit (Thermo Fisher Scientific, 32209) and imaged by iBright 1500.

FerroOrange and C11 staining

Staining procedures were based on the manufacturer’s manual. Briefly, oocytes were incubated in M2 medium (Sigma, M7167) with 1 mM FerroOrange (Dojindo, F374) and 1 μg/mL Hoechst 33342 (Invitrogen, H3570) for 30 min at 37°C. For imaging, oocytes were maintained in the M2 medium with 1 mM FerroOrange. For C11 staining, oocytes were incubated in M2 medium (Sigma, M7167) with 5 μM BODIPY^™ 581/591 C11 (Invitrogen, D3861) and 1 μg/mL Hoechst 33342 (Invitrogen, H3570) for 30 min at 37°C before imaged in M2 medium.

ER staining

The procedures were based on the manufacturer’s manual. Briefly, oocytes were incubated in M2 medium (Sigma, M7167) with ER-ID (1:1000, Enzo Life Sciences, ENZ-51025) and 1 μ g/mL Hoechst 33342 (Invitrogen, H3570) for 30 min at 37°C before imaged in M2 medium.

In vitro follicle culture

The experiments were performed as previously described with minor modifications ⁷⁰. Briefly, ovaries were removed from 10-day-old female mice and transferred to dissection medium (Leibovitz L-15 (Gibco, 11415064). Follicles were mechanically isolated with 30G needles in dissection medium supplemented with 1% FBS (Hyclone, SH3007003) and 0.5% penicillin-streptomycin (Gibco, 15140122)). Isolated follicles were washed and recovered in maintenance medium [MEMα with GlutaMAX (Gibco, 32561037) supplemented with 1% FBS (Hyclone, SH3007003) and 0.5% penicillin-streptomycin (Gibco, 15140122)] for 1 h at 37°C under 5% CO. High-quality follicles were selected and encapsulated in 0.5% alginate hydrogels via a droplet-based method. Briefly, follicles were suspended in alginate (Sigma, 71238) solution and droplets (~5 μL) were crosslinked in 50 mM CaCl (Sigma, 21115) for 2 min. Encapsulated follicles were briefly equilibrated and then cultured in growth medium [α-MEM/F12 (Gibco, 31765035 and 32561037) supplemented with 3 mg/mL bovine serum albumin (Jackson ImmunoResearch, 001-000-162), 1% insulin-transferrin-selenium (Gibco, 41400045), 10 mIU/mL recombinant follicle-stimulating hormone (Sigma, F4021), and 5 mg/mL fetuin (Sigma, F3385)] at 37°C under 5% CO. Half of the medium was changed every 48 hours with 10 μM Ferrostatin-1 (Sigma, SML0583) or an equivalent volume of DMSO (Sigma, 472301) added to the medium.

Immunofluorescence

Oocytes, embryos, or ovaries were fixed in 4% paraformaldehyde at room temperature 20min or 4 overnight and permeabilized in 0.5% Triton X-100 - 0.1% Polyvinylalcohol - PBS (PBST-PVA) at room temperature for 1 h. After blocking with 1% BSA in PBST-PVA, the samples were incubated in primary antibodies diluted in PBST-PVA containing 1% BSA overnight at 4. Antibodies were used as follows: anti-NFYA (1:200, sc-17753X, lot#D0522, Santa Cruz Biotechnology), anti-HA (1:200, 3724S or 2367S, Cell Signaling Technology), anti-caspase3 (1:500, 9664S, Cell Signaling Technology), anti-γH2AX (1:200, 05-636-25UG, Millipore), anti-LC3 (1:1000, 4108S, Cell Signaling Technology), anti-DDX4 (1:200, 8761T, Cell Signaling Technology). Following three washes with PBST-PVA, the samples were incubated with appropriate Alexa 488 or 594 fluorophores conjugated secondary antibodies (A-11012 or A-11029, Thermo Fisher Scientific) for 1 h at room temperature. Finally, the samples were counterstained with DAPI. The fluorescence signals were imaged using a confocal laser scanning microscope (Zeiss, LSM800). The fluorescent intensity was quantified using ImageJ. For oocyte quantification in the ovaries after in vitro culture, growing oocytes were defined by counting DDX4 positive oocytes bigger than 25 μm in diameter that were surrounded by cuboidal granulosa cells. Growing oocytes in the whole ovary were counted by imageJ analysis software.

Hematoxylin and eosin (HE) staining and electron microscopy

The ovary samples were fixed with 4% paraformaldehyde overnight. The fixed placentas were sent to the Rodent Histopathology Core (220 Longwood Ave., Boston, MA, USA) for paraffin embedding, section and HE staining. Electron microscopic imaging was done by the HMS Electron Microscopy Core.

CUT&RUN, ATAC and RNA-seq library preparation and sequencing

CUT&RUN assays were performed as previously described with minor modifications ¹⁹. For mESCs CUT&RUN with more than 1 × 10⁴ cells, cells were resuspended in 50 μl washing buffer (20 mM HEPES/pH=7.5, 150 mM NaCl, 0.5 mM spermidine and 1× protease inhibitor) with activated Concanavalin A Magnetic Beads (Polysciences, 86057-3) for 10 mins at room temperature (RT), then samples were incubated with anti-NFYA (1:100, Invitrogen, MA5-36198, lot#2L4586018) overnight at 4°C. Samples were then incubated with 2.8 ng/μl pA-MNase (home-made) for 2 hrs at 4 °C. Subsequently, samples were incubated with 200 μl precooled 0.5 μM CaCl2 for 20 mins at 4 °C and quench by the addition of 23 μl 10×stop buffer (1.7 M NaCl, 20 mM EGTA, 100 mM EDTA, 0.02% Digitonin, 250 µg/ml glycogen and 250 μg/ml RNase A). DNA fragments were released by incubation at 37°C for 15 mins. Then, 2.5 μl 10% SDS and 2.5 μl 20 mg/ml Protease K (Thermo Fisher) was added and incubated at 55 °C for at least 1 h. DNA was extracted by phenol-chloroform followed by ethanol precipitation. For low input mESCs and embryo CUT&RUN, some modifications were made. Briefly, mESCs, zona-free embryos or isolated oocytes were resuspended in 50 μΐ wash buffer with activated Concanavalin A Magnetic Beads for 10 mins at RT, then samples were incubated with anti-NFYA overnight at 4°C. The subsequent procedures are the same as those for mESCs described above. Sequencing libraries were prepared with the NEBNext Ultra II DNA library preparation kit for Illumina (New England Biolabs, E7645S).

ATAC-seq was performed as previously described with some modifications ^14,19,21. Briefly, isolated oocytes were digested with adapter-loaded Tn5 for 15 mins at 37°C, and stopped by stop buffer (100 mM Tris/pH=8.0, 100 mM NaCl, 40 µg/ml Proteinase K and 0.4% SDS) and incubated overnight at 55°C. Then, 5 µl of 25% Tween-20 was added to quench SDS. Sequencing libraries were prepared with NEBNext High-Fidelity 2×PCR Master Mix (NEB, M0541S).

RNA-seq library was prepared by SMART-Seq^® Stranded Kit (Takara Bio, 63444) using 5,000 mESCs, 20~50 isolated oocytes, and ~10 L2C embryos. All libraries were sequenced by NextSeq 550 system (Illumina) with paired-ended 75-bp reads (Table S9)

RNA-seq data processing

Raw sequencing reads were trimmed using Trimmomatic ⁷¹ (v0.39) to remove sequencing adaptors, and subsequently mapped to the GRCm38 genome using STAR ⁷² (v2.7.8a). Gene expression was quantified with featureCounts ⁷³ (v2.0.1) by counting reads mapped to each gene. Then, edgeR ⁷⁴ (v3.32.1) was employed for normalization and differential expression analysis. Genes with RPKM lower than 1 were defined as lowly expressed genes and excluded from the differential expression analysis. Differentially expressed genes (DEGs) were identified using likelihood ratio test (glmFit and glmLRT functions from edgeR). Genes with a false discovery rate (FDR) below 0.05 and an absolute value of fold change greater than 2 were defined as DEGs. Genes upregulated during primordial to primary follicle transition were classified as “early-PFA genes”, while those upregulated from primary to secondary follicle transition were designated as “late-PFA genes” (Table S2). The minor and major ZGA genes were defined as previously described ^19,75 (Table S2). Gene Ontology (GO) enrichment was performed using the R package clusterProfiler ⁷⁶. Gene Set Enrichment Analysis (GSEA) was performed using clusterProfiler and enrichplot. The gene set of Ferroptosis Suppressors was obtained from Ferrdb (http://www.zhounan.org/ferrdb/current/).

CUT&RUN data analysis

Raw sequencing reads were trimmed using Trimmomatic ⁷¹ (v0.39) to remove sequencing adaptors, and subsequently aligned to the GRCm38 reference genome using bowtie2 ⁷⁷ (v2.4.2) with parameters: --local --very-sensitive-local --no-unal --no-mixed --no-discordant --dovetail -I 10 -X 700 --soft-clipped-unmapped-tlen. PCR duplicates were removed by Picard MarkDuplicates ⁷⁸ (v2.23.4) and reads with a mapping quality below 30 were removed. The mapped reads were further filtered to only retain proper paired reads with fragment length between 10 and 120 bp. Then, MACS2 ⁷⁹ (v2.2.7.1) was used to call significant peaks with parameters “-f BAMPE -B --SPMR -q 0.05 -g mm --keep-dup all”. To obtain highly conserved NFYA binding sites, only peaks present in both biological replicates were defined as binding sites. The signal tracks were generated with deeptools ⁸⁰ bamCoverage (v3.5.1) with bin size of 1 and normalized using CPM. Peak annotation was performed using the ChIPseeker ⁸¹ R package. Peaks located within ±2,500 bp of transcription start sites (TSS) were classified as promoter peaks, whereas peaks outside this region were considered distal peaks. The heatmaps of binding profiles were calculated with deeptools ⁸⁰ computeMatrix (v3.5.1) using bigwig signal tracks as input and bin size of 10 and visualized in R with profileplyr and EnrichedHeatmap ⁸² packages.

ATAC–seq data analysis

ATAC–seq data were processed with the ENCODE ATAC–seq pipeline using default parameters (v2.1.2, https://github.com/ENCODE-DCC/atac-seq-pipeline). Peaks located within±2,500 bp of transcription start sites (TSS) were classified as promoter peaks, whereas peaks outside this region were considered distal peaks.

To identify differentially accessible peaks, the ATAC-seq read counts within each NFYA-binding site were quantified using multiBamCov ⁸³ (v2.30.0). The resulting read count matrix was imported into edgeR ⁷⁴ for downstream analysis. After normalization and dispersion estimation, a generalized linear model was fitted, and likelihood ratio tests were performed (glmFit and glmLRT functions). Peaks with a false discovery rate (FDR) less than 0.05 were defined as differentially accessible peaks.

Motif enrichment analysis

Motif enrichment analysis was performed using HOMER ⁸⁴ (v4.11) findMotifsGenome.pl with the mm10 reference and parameter -size 200, using peak regions as input. Briefly, HOMER identifies over-represented motifs in the input peak regions by comparing them to matched random background sequences. Enrichment P values for all motifs in all peaks were then corrected for multiple testing by calculating adjusted P value using the Benjamini-Hochberg method. Motifs with an adjusted p-value less than 0.05 were considered significantly enriched.

Motif occurrence analysis

Motif occurrence analysis was performed using HOMER ⁸⁴ (v4.11) annotatePeaks.pl with parameters mm10 -size -500,500 -hist 20 -ghist for regions around gene TSSs. Motif files were downloaded from the JASPAR database14 and manually converted to HOMER motif format. The JASPAR ⁸⁵ motif ID for TFs analyzed in this study is NFYA (MA0060.1) and OBOX (PH0121.1), and a log odds detection threshold of 6.0 was applied. The motif occurrence matrix was visualized in R with the EnrichedHeatmap ⁸² package.

Prediction of NFYA-mediated promoter-enhancer interactions

Multiple public dataset in the 2-cell stage were collected as the input for Activity-by-Contact (ABC) model (v0.2) ⁴⁸, including, RNA-seq, ATAC-seq ²⁰, H3K27ac CUT&RUN ⁷⁵, and L2C NFYA CUT&RUN, to predict NFYA-mediated promoter–enhancer interactions. Enhancers within 5 Mb and with scores ≥0.02 were assigned to target genes.

Supplementary Material

Supplement 1

Figure S1. Identification of NFYA as a potential regulator of PFA and ZGA and demonstration that NFYA loss impairs folliculogenesis. (A) Principal components analysis (PCA) of gene expression from primordial follicle to late 2-cell embryos. (B) Metaplot showing the fold change of ATAC–seq signals (ATAC_FC) around transcription start sites (TSSs) of primordial, primary, and secondary follicles. (C) Heatmaps showing the expression of upregulated genes from primordial to primary and secondary follicles, and the corresponding ATAC_FC signals at their promoters. (D) Enrichment of TF motifs at promoter ATAC–seq peaks of the indicated stages. (E) Schematic diagrams showing the early folliculogenesis and a conditional Nfya knockout strategy using Gdf9-Cre-mediated deletion of exons 4-8 harboring its DNA-binding domain. Gdf9-Cre is specifically active in oocytes starting from primordial follicle stage. (F) Upper panel: the grey scale of NFYA signals in WT and Gdf9-Cre Nfya^loxp/loxp (Nfya-cKO) oocytes from figure 1G. Scale bar, 25 μm. Lower panel: relative nuclear to cytoplasmic intensity ratio of NFYA in primordial and primary follicle oocytes from WT and Nfya-cKO female mouse. P values, Student’s t test. (G) Upper panel: The gross morphology of ovaries derived from WT and Nfya-cKO mice at postnatal day 7 (P7), 10, 2 weeks, 4 weeks, and 8 weeks. Lower panel: the relative size of ovaries from WT (n=4) and Nfya-cKO mice (n=4). Scale bar, 500 μm. (H) Images of 8-week ovaries from WT (Nfya^loxp/loxp), Gdf9-Cre Nfya^loxp/wt, and Gdf9-Cre Nfya^loxp/loxp (Nfya-cKO) mice. (I) H&E staining of paraffin embedded 8-week ovarian sections of WT and Nfya-cKO mice. Scale bar, 250 μm.

Figure S2. NFYA deficiency impairs PFA in primary follicle oocytes. (A) Correlations of the replicates of the total RNA-seq. The x and y axis of the dot plots are Log (CPM+1). (B) Heatmap showing the expression of oocyte- and granulosa cell-marker genes in the WT and Nfya-cKO primordial, primary, and secondary follicle oocytes. (C) Genome browser view of RNA-seq at the Nfya genomic regions in WT and Nfya-cKO oocytes at primordial, primary, and secondary follicle stages. (D) Correlation of replicates for NFYA CUT&RUN in primary follicle oocytes. Each dot represents a 5-kb bin. (E) Upper panel: the genomic distribution of NFYA binding peaks generated by NFYA CUT&RUN in primary follicle oocytes. Lower panel: the top 3 TF motifs enriched in the regions bound by NFYA in primary follicle oocytes. Note that based on the p-value, the NFYA motif is much more highly enriched compared to the other enriched motifs. (F) Heatmaps showing all the CUT&RUN signals of NFYA peaks at promoters and distal regions in WT or Nfya-cKO primary follicle oocytes. (G) NFYA motif numbers found in the promoters of the other 402 downregulated genes in primary follicle oocyte. (H) MGI phenotypic annotation for NFYA-bound downregulated genes (n=402) in primary follicles. Infertility related terms and example genes are shown. (I) The percentage of differentially expressed (Nfya-cKO vs WT) early-PFA genes in primary follicle oocytes. (J) Gene set enrichment analysis (GSEA) showing the early-PFA genes (n=1,888) are significantly decreased in Nfya-cKO oocytes. (K) Genome browser view of NFYA targeted early-PFA gene examples of NFYA CUT&RUN and RNA-seq signals in primary follicles. (L) NFYA motif numbers found in the promoters of the other 259 downregulated genes in early-PFA. (M) Correlation of replicates for ATAC-seq in WT or Nfya-cKO primary follicle oocytes. The x and y axis of the dot plots are log2(normalized_counts+1). Each dot represents a 5-kb bin.

Figure S3. NFYA is critical for PFA in secondary follicle oocytes. (A) Correlation of replicates for NFYA CUT&RUN in secondary follicle oocytes. The x and y axis of the dot plots are log2(normalized_counts+1). Each dot represents a 5-kb bin. (B) The genomic distribution of NFYA binding peaks generated by NFYA CUT&RUN in secondary follicle oocytes. (C) The top 3 TF motifs enriched in the regions bound by NFYA in secondary follicle oocytes. (D) Left panel: heatmaps showing the gene expression, NFYA CUT&RUN signals, NFYA motif occurrence, and ATAC–seq signal fold change (ATAC_FC) in the promoter regions of the DEGs in the secondary follicle oocytes upon NFYA loss. The downregulated or upregulated genes are separated into two groups based on whether they have direct NFYA promoter binding. Right panel: density plots showing the average ATAC–seq signal fold change (cKO/WT) in the promoter regions of the DEGs in different groups of the secondary follicle oocytes. P values are calculated with two-sided Student’s t-test. (E) A comparison of the downregulated and upregulated genes between primary follicle oocytes and secondary follicle oocytes after Nfya knockout. (F) NFYA motif numbers found in the promoters of the other 658 downregulated genes in secondary follicle oocytes. (G) GO terms enriched in the NFYA-bound down-regulated genes (n=658) in the secondary follicle oocytes. The example genes are shown in each GO term. (H) MGI phenotypic annotation for NFYA-bound newly downregulated genes (n=658) in secondary follicle oocytes. Infertility related terms and gene examples are shown. (I) Heatmaps showing the gene expression of early-PFA genes (n=1,888) in oocytes at primordial, primary, and secondary follicle stages. (J) A comparison of the downregulated early-PFA genes between primary follicle oocytes and secondary follicle oocytes after Nfya knockout. (K) Left panel: heatmaps showing the newly downregulated early-PFA genes (n=272) in secondary follicle oocytes upon NFYA loss, and the NFYA binding around the TSS of the corresponding genes. The downregulated genes are separated into two groups based on whether they have direct NFYA promoter binding. Right panel: Genome browser view of NFYA targeted newly affected early-PFA gene examples of NFYA CUT & RUN and RNA-seq in secondary follicle oocytes. (L) NFYA motif numbers found in the promoters of the 162 newly affected early-PFA genes. (M) GSEA enrichment plots showing the late-PFA genes (n=834) are significantly decreased in the Nfya-cKO oocytes. (N) NFYA motif numbers found in the promoters of the 319 downregulated late-PFA genes. (O) Correlation of replicates for ATAC-seq in WT and Nfya-cKO secondary follicle oocytes. The x and y axis of the dot plots are log2(normalized_counts+1). Each dot represents a 5-kb bin.

Figure S4. NFYA loss triggers non-canonical ferroptosis in growing oocytes. (A) Dotplots of GSEA enrichment showing pathways of ferroptosis, but not apoptosis or autophagy, are enriched in Nfya-cKO oocytes. (B) Cleaved caspase3 (red) and γH2AX (green) staining show no sign of apoptosis in Nfya-cKO oocytes. Scale bar, 25 μm. (C) LC3 (red) staining does not show sign of autophagy in Nfya-cKO secondary follicle oocytes. Scale bar, 25 μm. (D) Representative images of FerroOrange (an Fe²⁺ indicator) staining of WT and Nfya-cKO primary follicle oocytes. Scale bar, 10 μm. (E) Relative intensity of FerroOrange staining in primary and secondary follicle oocytes. P values, Student’s t test.

Figure S5. Generation of NFYA^dTAG mice and NFYA depletion affects major ZGA. (A) Diagram illustration of HA-FKBP (F36V) knock-in at the 5’ end of Nfya gene and the NFYA^dTAG fusion protein. (B) Western blot analysis confirming NFYA^dTAG knock-in in ES cells. The blots were incubated with anti-NFYA and anti-HA, respectively. GAPDH was used as a loading control. (C) Western blot analysis confirming dTAG13-mediated NFYA degradation in mES cells. The blots were incubated with anti-HA. GAPDH was used as a loading control. (D) Diagram illustration for Nfya^dTAG/dTAG mouse generation and HA staining confirming the dTAG knock-in cells in morula. Scale bar, 25 μm. (E) Statistics of pup numbers from NFYA^dTAG/dTAG × NFYA^dTAG/dTAG crosses, indicating normal function of the NFYA^dTAG fusion protein. (F) Representative images of 48 hpf (4-8 cell stage) Homo embryos with or without dTAG13 treatment. (G) Upper panel: schematic diagram of Nfya^dTAG/dTAG embryo generation and dTAG13 treatment. Lower panel: correlation of replicates for total RNA-seq in Nfya^dTAG/dTAG L2C embryos with DMSO or dTAG13 treatment. The x and y axis of the dot plots are Log (CPM+1).

Figure S6. NFYA affects ZGA by promoter and distal binding. (A) Genome browser view showing examples of NFYA CUT&RUN profiles in mESCs (1000 cells), WT L2C embryos, and maternal NFYA-depleted L2C embryos by dTAG13 (L2C-dTAG13). (B) Correlation between replicates of NFYA CUT&RUN in late 2cell embryos. The x and y axis of the dot plots are log2(normalized_counts+1). Each dot represents a 5-kb bin. (C) Correlation between replicates of NFYA CUT&RUN in mESCs. The x and y axis of the dot plots are Log2(normalized_counts+1). Each dot represents a 5-kb bin. (D) The top 3 TF motifs enriched in the regions bound by NFYA in L2C embryos. (E) The genomic distribution of NFYA binding peaks from NFYA CUT&RUN in mESCs. (F) Heatmaps showing all the CUT&RUN signals of NFYA peaks at promoters and distal regions in L2C with or without dTAG13 treatment. (G) A bar chart showing repeat class/family enrichment at NFYA-binding peaks in the distal regions of L2C embryos. (H) Density plots showing the average L2C NFYA CUT&RUN signal in the promoter regions of downregulated major ZGA genes (n=1,067) and non-ZGA genes (n=1,250) at late 2-cell stage upon NFYA depletion. P values were calculated with two-sided Student’s t-test. (I) Genome browser view of NFYA promoter targeted ZGA gene examples of NFYA CUT & RUN and RNA-seq in late 2-cell embryos. (J) NFYA motif numbers found in the promoters of the 744 downregulated ZGA genes. (K) Schematic diagram showing prediction of promoter-distal interactions with NFYA distal binding by ABC model ⁴⁸. (L) Genome browser view of NFYA distal targeted ZGA gene examples of NFYA CUT & RUN and RNA-seq in L2C embryos. (M) Correlation of replicates for ATAC-seq in L2C embryos with DMSO or dTAG13 treatment. The x and y axis of the dot plots are log2(normalized_counts+1). Each dot represents a 5-kb bin. (N) Density plots showing the average ATAC–seq_FC signal in the promoter regions of the downregulated major ZGA genes of the L2C embryos.

Figure S7. Conserved and divergent NFYA chromatin bindings regulate PFA and ZGA. (A) Representative GO (BP) terms enriched in the oocyte- and L2C-specific bindings related genes. (B) Heatmaps showing the dynamics of ATAC–seq signal in the NFYA binding regions during PFA (primordial follicle oocytes, primary follicle oocytes, and secondary follicle oocytes) and ZGA (zygotes and late 2-cell embryo) at promoter and distal regions. (C) Correlation between replicates of NFYA CUT&RUN in primordial follicle oocytes and zygotes (6hpf). The x and y axis of the dot plots are log2(normalized_counts+1). Each dot represents a 5-kb bin. (D) Density plots showing the average ATAC–seq_FC signals of the 920 NFYA common binding regions in oocytes or L2C embryos after NFYA-depletion. (E) Box plots showing the comparison of ATAC–seq_FC signal changes [Log2(KO/WT)] between the oocyte-specific, L2C-specific, and common groups in primary follicle oocytes, secondary follicle oocytes, and L2C embryos. (F) Upper panel: schematic diagram of the experimental design. Medium panel: the embryo development rate upon DMSO and Ganetespib: 2nM, 5nM, 10nM, 25nM treatment. Lower panel: the embryo development rate upon DMSO and VER155008: 1 μM, 5μΜ, 10μΜ, 25μΜ treatment. (G) Correlation of replicates for total RNA-seq in L2C embryos with DMSO or G&V treatment. The x and y axis of the dot plots are Log (CPM+1). (H) Ovarian morphology of 10 days in vitro cultured P4 ovaries with DMSO or G&V treatment. DDX4 stained oocytes. Scale bar, 50 μm. (I) Correlation of replicates for total RNA-seq in growing oocytes (GOs) from P4 ovaries after 6 days DMSO or G&V treatment. The x and y axis of the dot plots are Log (CPM+1).

Highlights.

• NFYA deficiency impairs primordial follicle oocyte activation (PFA) and triggers non- canonical ferroptosis resulting in early folliculogenesis failure

• NFYA depletion impairs zygotic genome activation (ZGA) and causes predominantly 2cell embryo arrest

• Conserved and distinct NFYA-chromatin interactions drive both PFA and ZGA

• Chaperones are pre-occupied and regulated by NFYA and their inhibition impairs both PFA and ZGA.