In the interphase of the cell cycle, chromatin is arranged in a hierarchical structure within the nucleus1,2, which has an important role in regulating gene expression3,4,5,6. However, the dynamics of 3D chromatin structure during human embryogenesis remains unknown. Here we report that, unlike mouse sperm, human sperm cells do not express the chromatin regulator CTCF and their chromatin does not contain topologically associating domains (TADs). Following human fertilization, TAD structure is gradually established during embryonic development. In addition, A/B compartmentalization is lost in human embryos at the 2-cell stage and is re-established during embryogenesis. Notably, blocking zygotic genome activation (ZGA) can inhibit TAD establishment in human embryos but not in mouse or Drosophila. Of note, CTCF is expressed at very low levels before ZGA, and is then highly expressed at the ZGA stage when TADs are observed. TAD organization is significantly reduced in CTCF knockdown embryos, suggesting that TAD establishment during ZGA in human embryos requires CTCF expression. Our results indicate that CTCF has a key role in the establishment of 3D chromatin structure during human embryogenesis.
Subscribe to Journal
Get full journal access for 1 year
only $3.90 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Data generated for this study have been deposited to the Genome Sequence Archive with the accession number CRA000852 and CRA000108. Hi-C data of unmixed human sperm sample have been deposited in CRA000108, and the other data have been deposited in CRA000852. Raw image files used in the figures that support the findings of this study are available from the corresponding authors upon reasonable request.
Codes used for the analysis reported in this study are available at https://github.com/ChenXP0310/2019-humanembryo3D.
Fullwood, M. J. et al. An oestrogen-receptor-α-bound human chromatin interactome. Nature 462, 58–64 (2009).
Atlasi, Y. & Stunnenberg, H. G. The interplay of epigenetic marks during stem cell differentiation and development. Nat. Rev. Genet. 18, 643–658 (2017).
Rao, S. S. et al. A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping. Cell 159, 1665–1680 (2014).
Tang, Z. et al. CTCF-mediated human 3D genome architecture reveals chromatin topology for transcription. Cell 163, 1611–1627 (2015).
Bonev, B. et al. Multiscale 3D genome rewiring during mouse neural development. Cell 171, 557–572 (2017).
Hsieh, T. H. et al. mapping nucleosome resolution chromosome folding in yeast by micro-C. Cell 162, 108–119 (2015).
Du, Z. et al. Allelic reprogramming of 3D chromatin architecture during early mammalian development. Nature 547, 232–235 (2017).
Ke, Y. et al. 3D chromatin structures of mature gametes and structural reprogramming during mammalian embryogenesis. Cell 170, 367–381 (2017).
Dixon, J. R. et al. Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature 485, 376–380 (2012).
Lieberman-Aiden, E. et al. Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326, 289–293 (2009).
Dixon, J. R. et al. Chromatin architecture reorganization during stem cell differentiation. Nature 518, 331–336 (2015).
Giorgetti, L. et al. Structural organization of the inactive X chromosome in the mouse. Nature 535, 575–579 (2016).
Li, C. et al. DNA methylation reprogramming of functional elements during mammalian embryonic development. Cell Discov. 4, 41 (2018).
Smith, Z. D. et al. DNA methylation dynamics of the human preimplantation embryo. Nature 511, 611–615 (2014).
Guo, H. et al. The DNA methylation landscape of human early embryos. Nature 511, 606–610 (2014).
Jung, Y. H. et al. Chromatin states in mouse sperm correlate with embryonic and adult regulatory landscapes. Cell Rep. 18, 1366–1382 (2017).
Baranello, L., Kouzine, F. & Levens, D. CTCF and cohesin cooperate to organize the 3D structure of the mammalian genome. Proc. Natl Acad. Sci. USA 111, 889–890 (2014).
Nora, E. P. et al. Targeted degradation of CTCF decouples local insulation of chromosome domains from genomic compartmentalization. Cell 169, 930–944 (2017).
Hug, C. B., Grimaldi, A. G., Kruse, K. & Vaquerizas, J. M. Chromatin architecture emerges during zygotic genome activation independent of transcription. Cell 169, 216–228 (2017).
Xue, Z. et al. Genetic programs in human and mouse early embryos revealed by single-cell RNA sequencing. Nature 500, 593–597 (2013).
Hendrickson, P. G. et al. Conserved roles of mouse DUX and human DUX4 in activating cleavage-stage genes and MERVL/HERVL retrotransposons. Nat. Genet. 49, 925–934 (2017).
Yan, L. et al. Single-cell RNA-seq profiling of human preimplantation embryos and embryonic stem cells. Nat. Struct. Mol. Biol. 20, 1131–1139 (2013).
Schmitt, A. D. et al. A compendium of chromatin contact maps reveals spatially active regions in the human genome. Cell Rep. 17, 2042–2059 (2016).
Wu, J. et al. Chromatin analysis in human early development reveals epigenetic transition during ZGA. Nature 557, 256–260 (2018).
Gao, L. et al. Chromatin accessibility landscape in human early embryos and its association with evolution. Cell 173, 248–259 (2018).
Flyamer, I. M. et al. Single-nucleus Hi-C reveals unique chromatin reorganization at oocyte-to-zygote transition. Nature 544, 110–114 (2017).
Roy, T. K., Bradley, C. K., Bowman, M. C. & McArthur, S. J. Single-embryo transfer of vitrified-warmed blastocysts yields equivalent live-birth rates and improved neonatal outcomes compared with fresh transfers. Fertil. Steril. 101, 1294–1301 (2014).
Zuin, J. et al. Cohesin and CTCF differentially affect chromatin architecture and gene expression in human cells. Proc. Natl Acad. Sci. USA 111, 996–1001 (2014).
Guillou, E. et al. Cohesin organizes chromatin loops at DNA replication factories. Genes Dev. 24, 2812–2822 (2010).
Ramírez, F. et al. High-resolution TADs reveal DNA sequences underlying genome organization in flies. Nat. Commun. 9, 189 (2018).
Wolff, J. et al. Galaxy HiCExplorer: a web server for reproducible Hi-C data analysis, quality control and visualization. Nucleic Acids Res. 46 (W1), W11–W16 (2018).
Imakaev, M. et al. Iterative correction of Hi-C data reveals hallmarks of chromosome organization. Nat. Methods 9, 999–1003 (2012).
Durand, N. C. et al. Juicer provides a one-click system for analyzing loop-resolution Hi-C experiments. Cell Syst. 3, 95–98 (2016).
Ursu, O. et al. GenomeDISCO: a concordance score for chromosome conformation capture experiments using random walks on contact map graphs. Bioinformatics 34, 2701–2707 (2018).
Gassler, J. et al. A mechanism of cohesin-dependent loop extrusion organizes zygotic genome architecture. EMBO J. 36, 3600–3618 (2017).
Naumova, N. et al. Organization of the mitotic chromosome. Science 342, 948–953 (2013).
Heinz, S. et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell 38, 576–589 (2010).
Li, G. et al. Genome wide abnormal DNA methylome of human blastocyst in assisted reproductive technology. J. Genet. Genom. 44, 475–481 (2017).
Langmead, B., Trapnell, C., Pop, M. & Salzberg, S. L. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 10, R25 (2009).
DePristo, M. A. et al. A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nat. Genet. 43, 491–498 (2011).
Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
John, S. et al. Chromatin accessibility pre-determines glucocorticoid receptor binding patterns. Nat. Genet. 43, 264–268 (2011).
Ramírez, F. et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 44 (W1), W160–W165 (2016).
Eisenberg, E. & Levanon, E. Y. Human housekeeping genes, revisited. Trends Genet. 29, 569–574 (2013).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Gu, Z., Gu, L., Eils, R., Schlesner, M. & Brors, B. circlize Implements and enhances circular visualization in R. Bioinformatics 30, 2811–2812 (2014).
Consortium, E. P.; ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57–74 (2012).
This work was supported by the grants from The Ministry of Science and Technology of China (2018YFC1004000 and 2018YFC1003300), National Natural Science Foundation of China (81430029, 91731312, 81871171, 31425015, 31630040, 31871454, 81622021 and 81871168), CAS funding (QYZDY-SSW-SMC016) and Strategic Priority Research Program of the Chinese Academy of Sciences (XDB13040200).
The authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
a, Interaction heat maps for mouse embryonic day (E)3.5 embryos. Top, pooled mouse E3.5 embryos Hi-C8. Bottom, a single mouse E3.5 embryo with ultra-low-input Hi-C from this study. b, A track snapshot of PC1 values for chromosome 1 in pooled mouse E3.5 embryos and single mouse E3.5 embryo along with public mouse ES cell (mESC) data9. c, Interaction heat maps for pooled mouse E3.5 embryos and single mouse E3.5 embryo for a zoomed-in region from chromosome 1 overlaid with TADs, directional index, boundaries and TAD separation scores. d, GenomeDISCO reproducibility (Methods) among the Hi-C data for each chromosome in pooled human blastocysts and two human single-blastocyst replicates. e, The strength of average TADs in pooled human blastocyst data, single human blastocyst data and metaphase HeLa data36. TAD positions are annotated in combined human blastocyst data. TAD structure strength (Methods) is shown on each corresponding panel. f, GenomeDISCO reproducibility for biological replicates of human sperm and embryos. n = 2–3 independent biological replicates for each stage. Each dot represents one comparison between two biological replicates. Data are mean ± s.d.
a, Interaction heat maps for human embryos with equal reads along with human H1 ES cells. b, An example for TAD calling by using TAD separation score method in human blastocysts with different read depths. The orange track under each heat map represents TAD domains. The bottom panels are TAD boundaries and TAD separation score tracks. M, million reads. c, Venn diagram for the overlap of TAD boundaries of human blastocyst between 10 million read depth data (10 M, blue) and total combined data (total, red). d, An equal-read-depth TAD signal track snapshot for human sperm and embryos along with human H1 ES cells. e, Average directional index around human TAD boundaries(0.5 Mb) at different embryonic stages. Directional index generated by a random blastocyst valid read pair dataset is also shown as a control. f, g, Interaction heat map examples at 50-kb resolution for unmixed human embryos and human embryos from mixed samples. h, The strength for average TADs in mouse morula embryos from mixed samples (six mouse morulae for each mixed sample), unmixed mouse morula and mouse MII oocyte. TAD positions are annotated in combined mouse E3.5 embryos. TAD structure strength is shown on each corresponding panel. i, Relative variance of TAD signal for replicates of human embryos including human embryos from mixed samples. We used the chromosome as the unit for this analysis. Two-cell (n = 3), 8-cell (n = 3), morula (n = 3), blastocyst (n = 3) and 6-week embryos (n = 2). Boxes represent the 25th, 50th and 75th percentiles and whiskers show 1.5× the interquartile range. ***adjusted P < 0.001 for all pairwise comparison between two stage samples (two-sided Wilcoxon rank-sum test with Benjamini–Hochberg multiple testing correction). NS, not significant.
a, Pearson correlation heat maps of chromosome 1 at 500-kb resolution in human 2-cell, 8-cell, morula and blastocysts, 6-week embryos and H1 ES cells (equal no. of reads generated from 2–3 biological independent replicates for each stage). b, Top, whole-genome PC1 value circle plots for human 2-cell embryos and human blastocysts (with R package circlize46). Bottom, PC1 tracks in mouse zygotes and mouse E3.5 embryos. c, Pearson correlation heat maps at 500-kb resolution of chromosome 1 in mouse zygotes and early embryos8 (equal number of reads generated from 2–3 biological independent replicates for each stage). d, e, Five-by-five contact enrichment maps of A–B compartments averaged over genomic positions using GC content in mouse early embryo Hi-C data8 (d) and ref. 7 (e). f, PC1 value distribution in human embryos and mitotic HeLa cells36.
a, The percentage of genomic regions with A/B compartment switches from morula to 6-week stage, and the percentage of single A/B switched regions relative to total switched regions. b, Left, the box plot illustrates the gene expression dynamics with A/B compartment switches between human blastocysts and 6-week embryos (biological replicates pooled; blastocyst, n = 3; 6-week embryo, n = 2). P values are shown (one-sided Wilcoxon rank-sum test). Boxes represent the 25th, 50th and 75th percentiles and whiskers show 1.5× the interquartile range (b, d–f). Middle, PC1 heat map for A/B compartment switched regions. Heat map is sorted according to A/B compartment switched PC1 values. Right, GO enrichment for genes with A/B compartment switches between blastocysts and 6-week embryos using DAVID v.6.8. c, A/B compartment status around the NEFM–NEFL locus in human blastocysts and 6-week embryos overlaid with gene expression. d, Unmethylated CpG (methylation level <0.25) density within A or B compartments in human sperm and embryos (biological replicates pooled; sperm, n = 3; morula, n = 2; blastocyst, n = 3; 6-week embryos, n = 2). P values are shown (one-sided Wilcoxon rank-sum test). e, Reduced CpG methylation level (ML) (MLsperm − MLblastocyst) in A compartments and B compartments of human blastocyst (biological replicates pooled; n = 3). P value is calculated by one-sided Wilcoxon rank-sum test. f, Reduced methylation level (MLsperm − MLblastocyst) for regions with A-to-A, A-to-B, B-to-A, and B-to-B compartment switches from human sperm to blastocysts (replicates pooled; n = 3). ***P < 0.001 for comparisons between ‘A-to-A’ and other groups (two-sided Wilcoxon rank-sum test). Right, randomly shuffled A/B compartment switched regions as control.
a, Interaction heat maps around Hoxa cluster in mouse sperm8,16 and mouse E3.5 embryos8. b, Density plot of interaction insert size for human sperm, human blastocysts and mouse sperm (Methods). c, Contact probability decay curve for human sperm, mouse sperm and metaphase HeLa cells. d, e, Examples of interaction heat maps at human and mouse conserved syntenic regions from unmixed sperm and mixed human and mouse sperm sample. Interaction heat maps in d (right) and e (right) are for HeLa and HT22 cells somatic cell mixed samples, respectively. f, Pearson correlation heat maps of chromosome 2 at 500-kb resolution in human sperm. Left, unmixed human sperm Hi-C. Right, human sperm from the mixed sperm sample (replicates pooled). g, RAD21 western blots and Ponceau staining of samples from somatic cell lines and human and mouse sperm. Black arrow indicates RAD21 band. Experiment repeated on two biologically independent replicates per sample. For gel source data, see Supplementary Fig. 1. h, Right, CTCF western blots in the control and knockdown cell lines (control and siCTCF HEK 293; control and siCTCF HeLa). Left, Ponceau staining. Black arrow indicates CTCF band. Experiment repeated on two biologically independent replicates per sample. For gel source data, see Supplementary Fig. 1.
a, An example for insulated boundaries in human embryos as yellow arrows point in the 2-cell stage. Dark coloured regions in TAD separation score heat maps mean strong insulation. The bottom blue bars are insulated boundaries of each stage. The red bars are 200-kb collapsed boundaries. Grey boxes highlight the 200-kb collapsed boundaries. b, Heat map illustrating the insulation score at stage-specific gained insulated boundaries in mouse embryos. c, d, An example of shared ZGA boundaries in human embryos (c) and mouse embryos (d). Grey boxes highlight the shared 200-kb collapsed ZGA boundaries.
a, Cumulative distribution function plot for the distance of stage-specific gained boundaries to the closest housekeeping genes in mouse embryos8 (pooled data from 2–3 biological replicates). The dash line marks a distance of 200 kb; 2-cell versus 4-cell, P = 1.25 × 10−9; 2-cell versus 8-cell, P = 1.59 × 10−11 (two-sided Kolmogorov–Smirnov test). b, Gene expression (expression data from ref. 22; 8-cell, n = 20; morula, n = 16; blastocyst, n = 30) for housekeeping genes within 200 kb of the stage-specific gained boundaries and the other housekeeping genes at the stages when insulated boundaries are being formed. Boxes represent the 25th, 50th and 75th percentiles and whiskers show 1.5× the interquartile range. P values are shown (two-sided Wilcoxon rank-sum test). c–e, Enrichment analysis of repeat elements at stage-specific gained insulated boundaries in human embryos. Enrichment analysis of AluY repeats (c), LINE repeats (d) and MIR repeats (e). f–h, Enrichment analysis of repeat elements at stage-specific gained insulated boundaries in mouse embryos. Enrichment analysis of Alu repeats (f), LINE repeats (g) and MIR repeats (h). i, Examples of the earlier stage gained insulated boundary locating around Alu dense regions in human and mouse. j, Left, snapshot of insulated boundaries overlaid with insulation score, AluS density and housekeeping genes. Grey box highlights the 200-kb collapsed boundary. Right, expression of AluS repeats at this boundary. k, Expression z-score of AluS repeats locating at human 2-cell gained boundaries during human embryonic development (repeat expression data from ref. 21; n = 2 for each stage; for repeat expression calculation, see Methods). Average gene expression of AluS repeats within a boundary was used. The number of 2-cell gained boundaries with AluS repeats is shown (n = 442).
a, ZGA gene expression in human embryos and α-amanitin treated 8-cell embryos (pooled replicates; for data from ref. 22: oocyte, n = 3; zygote, n = 3; 2-cell, n = 6; 4-cell, n = 12; 8-cell, n = 20; morula, n = 16; blastocyst, n = 30; for data from this study: 8-cell, n = 2; α-amanitin treated 8-cell, n = 2). ZGA genes were identified as in ref. 22 (Methods). P values are calculated by the one-sided Wilcoxon rank-sum test. Boxes represent the 25th, 50th and 75th percentiles and whiskers show 1.5× the interquartile range. b, Gene expression for human ZGA genes DUXA and ZSCAN5B in control 8-cell and α-amanitin treated 8-cell embryos. c, The Spearman correlation for two α-amanitin treated 8-cell Hi-C replicates. d, Interaction heat maps of human 2-cell, 8-cell, unmixed α-amanitin treated 8-cell embryos and α-amanitin 8-cell from the mixed sample (Methods).
a, Gene expression for different cohesin complex subunits during human embryonic development. Expression data from ref. 20: oocyte, n = 3; zygote, n = 2; 2-cell, n = 3; 4-cell, n = 4; 8-cell, n = 11; morula, n = 3. Data are mean ± s.e.m. b, Dynamics of CTCF expression during human embryonic development. Expression data from refs. 20,21. Data are mean ± s.e.m. c, CTCF gene expression in control 8-cell and α-amanitin treated 8-cell embryos. d, Interaction heat maps of untreated control morula and siCTCF morula with equal reads at 50-kb resolution. Right, contrast heat map between control and siCTCF morula. Dark colour represents increased interactions in untreated morula compared with siCTCF morula. e, Violin plot for relative variance of TAD signal in 8-cell, control morula and siCTCF morula (equal number of reads generated from 2–3 biological replicates for each stage). P values are shown (two-sided Wilcoxon rank-sum test). The white boxes in the violin plot represent median values. f, Enrichment heat maps for the insulation score of control morula and siCTCF morula at control morula TAD boundaries (\(\pm \)600 kb). b, TAD boundary centre.
a, Snapshot of FIREs in human blastocysts, 6-week embryos and H1 human ES cells overlaid with PC1 tracks and FIRE score tracks. One zoomed-in region in human blastocysts is also shown. b, Snapshot of FIREs in human morula and blastocysts overlaid with DHSs25 and A/B compartments. c, Bar plot showing FIREs enrichment and depletion in A/B compartments (replicates pooled; blastocyst, n = 3; 6-week, n = 2; H1 human ES cell, n = 2). P values are also shown (χ2 test). d, Box plot showing ATAC-seq reads24 signal at FIREs, non-FIREs and whole genome in human blastocysts. Boxes represent the 25th, 50th and 75th percentiles and whiskers show 1.5× the interquartile range. P values are also shown (two-sided Wilcoxon rank-sum test). e, Box plot showing DNase-seq reads25 signal at FIREs, non-FIREs and whole genome in human blastocysts. Boxes represent the 25th, 50th and 75th percentiles and whiskers show 1.5× the interquartile range. P values are also shown (two-sided Wilcoxon rank-sum test). f, Histograms of FIRE score difference between siCTCF and untreated control morula (siCTCF − untreated) in FIREs and non-FIREs for siCTCF morula no. 1, siCTCF morula no. 2 and siCTCF morula no. 3. P values for differential FIREscore (siCTCF − untreated) in FIREs and non-FIREs are also shown (one-sided t-test). g, The hierarchical cluster of gene expression in control morula and siCTCF morula analysed using the package ggdendro. h, Scatter plot of gene expression between control morula and siCTCF morula. Red dots refer to upregulated genes (254 genes) in siCTCF morula. Blue dots refers to downregulated genes (565 genes) in siCTCF morula. i, Heat map for human ES cell CTCF ChIP–seq signal47 around gene TSSs downregulated in siCTCF morula. Strongly downregulated genes (log2(fold change(siCTCF/control)) < −5).
a, Immunofluorescence confocal images of CTCF in α-amanitin treated 8-cell embryos (n = 2) and 8-cell (α-amanitin plus CTCF mRNA) embryos (n = 3). Scale bar, 40 μm. b, Track snapshot for TAD structures in 8-cell (α-amanitin plus CTCF mRNA) embryos along with untreated 2-cell, 8-cell and morula embryos. c, d, Interaction heat map examples at 50-kb resolution for human 8-cell embryos and 8-cell (α-amanitin plus CTCF mRNA) embryos. e, Interaction heat map examples for mouse morula embryos without mix and mouse morula from the α-amanitin plus CTCF mRNA mixed sample (Methods). f, Enrichment heat maps for the insulation score of human 2-cell embryos, 8-cell embryos, α-amanitin 8-cell embryos, 8-cell (α-amanitin plus CTCF mRNA) embryos and morula embryos around boundaries (± 600 kb). b, boundary centre. Left, boundaries gained at 2-cell stage; right, boundaries gained at 8-cell stage. g, Bar plots for TAD structure strength of human embryos and mouse morula from mixed samples. Left y axis, human embryos; right y axis, mouse morula.
Supplementary Figure 1: gel source data.
Table S1: Data summary for human sperm and embryos Hi-C.
Table S2: TADs in human embryos based on TAD separation score method.
Table S3: GO terms for genes with A/B compartment switches.
Table S4: Human embryo stage-gained insulated boundaries.
Table S5: Mouse embryo stage-gained insulated boundaries.
Table S6: Data summary for the NGS of amanitin-treated embryos and CTCF siRNA knockdown embryos.
About this article
Cite this article
Chen, X., Ke, Y., Wu, K. et al. Key role for CTCF in establishing chromatin structure in human embryos. Nature 576, 306–310 (2019). https://doi.org/10.1038/s41586-019-1812-0
Stem Cells International (2021)
Protein & Cell (2021)
Advance in the Role of Epigenetic Reprogramming in Somatic Cell Nuclear Transfer-Mediated Embryonic Development
Stem Cells International (2021)
The asynchronous establishment of chromatin 3D architecture between in vitro fertilized and uniparental preimplantation pig embryos
Genome Biology (2020)
Current Opinion in Cell Biology (2020)