Dynamic nucleosome organization after fertilization reveals regulatory factors for mouse zygotic genome activation

Abstract

Chromatin remodeling is essential for epigenome reprogramming after fertilization. However, the underlying mechanisms of chromatin remodeling remain to be explored. Here, we investigated the dynamic changes in nucleosome occupancy and positioning in pronucleus-stage zygotes using ultra low-input MNase-seq. We observed distinct features of inheritance and reconstruction of nucleosome positioning in both paternal and maternal genomes. Genome-wide de novo nucleosome occupancy in the paternal genome was observed as early as 1 h after the injection of sperm into ooplasm. The nucleosome positioning pattern was continually rebuilt to form nucleosome-depleted regions (NDRs) at promoters and transcription factor (TF) binding sites with differential dynamics in paternal and maternal genomes. NDRs formed more quickly on the promoters of genes involved in zygotic genome activation (ZGA), and this formation is closely linked to histone acetylation, but not transcription elongation or DNA replication. Importantly, we found that NDR establishment on the binding motifs of specific TFs might be associated with their potential pioneer functions in ZGA. Further investigations suggested that the predicted factors MLX and RFX1 played important roles in regulating minor and major ZGA, respectively. Our data not only elucidate the nucleosome positioning dynamics in both male and female pronuclei following fertilization, but also provide an efficient method for identifying key transcription regulators during development.

Fig. 1. Nucleosome occupancy is quickly established in mouse male pronuclei after fertilization.

a Schematic representation showing the collection of pronucleus samples for ULI-MNase-seq. b Confocal microscopy images of H2B-RFP mRNA-injected embryos shortly after fertilization. Newly incorporated H2B was present in the male PN (arrowhead) as early as 1 hpf. c Bar plots showing the fraction of nucleosome-occupied 1-kb bins in each PN sample. Error bars represent 1.96 × SD. d Line charts showing the fraction of nucleosome-occupied 1-kb bins in sex chromosomes and autosomes of each PN and ESC samples. Chr, chromosome.

Fig. 2. Asynchronous establishment of canonical nucleosome positioning in mouse parental pronuclei.

a Nucleosome profiles around TSSs of Refseq genes at each PN stage. b, c Bar plots showing the nucleosome phasing periodicity (b; illustrated in Supplementary information, Fig. S5a) or phasing intensity (c; calculated as the spectral intensity corresponding to the periodicity after fast Fourier transform) around TSSs of Refseq genes at each PN stage. d Nucleosome profiles around TSSs of all Refseq genes and ZGA genes in 3-hpf and 6-hpf parental PN.

Fig. 3. GC content and the histone acetylation level influence nucleosome establishment and repositioning in mouse pronuclei, respectively.

a Boxplots showing the GC content of newly established nucleosome regions at each PN stage. Dashed lines represent the average GC content in genome. b Bar plot showing the association of different histone modifications and GC content with promoter NDR scores. c–e Nucleosome profiles around TSSs of all Refseq genes at 6 hpf (c) and 12 hpf (d), or genes of the indicated promoter NDR clusters at 6 hpf (e; defined in Supplementary information, Fig. S6f) in parental PN from groups under different treatments. JQ1-B1, JQ1 batch 1; JQ1-B2, JQ1 batch 2; DMSO, 0.05% dimethyl sulfoxide-treated; control, water-injected.

Fig. 4. Dynamics of nucleosome organization on motif regions predict TF binding landscapes in mouse pronuclei.

a Nucleosome profiles around CTCF motifs at each PN stage. b Heatmap showing the k-means clustering (k = 3) of TFs based on NDR scores on motifs at each PN stage. M, male PN; F, female PN; h, hpf. c Boxplots showing the enrichment of ZGA gene promoters on motifs of TFs in different clusters defined in b (***P < 0.001; *P < 0.05). d, e Nucleosome profiles around TSSs of different classes of genes in 8-hpf male PN from KD groups. Genes were classified according to whether motifs of MLX (d) or RFX1 (e) are present in promoters. f Boxplots showing the expression level of ZGA genes in KD embryos (*P < 0.05; **P < 0.01; ***P < 0.001; N.S., P > 0.05). g Stacked bar plots showing the percentage of 1-cell and 2-cell embryos in KD groups at the indicated time points. n = 3 biological replicates with ~50 embryos each. Error bars represent SD.