CoChat
Sign Up
Cited by 258Genomics and Chromatin DynamicsChromatin Remodeling and CancerCancer-related molecular mechanisms research

The DOI matches its Crossref record; 74 of the 74 references it cites were found in a registry or an index.

Dissecting super-enhancer hierarchy based on chromatin interactions

Open access
Huang, Jialiang, Li, Kailong, Cai, Wenqing, Liu, Xin, Zhang, Yuannyu, Orkin, Stuart H., Xu, Jian, Yuan, Guo-Cheng·2018·Nature Communications
DOI 10.1038/s41467-018-03279-9PMC PMC5838163

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Recent studies have highlighted super-enhancers (SEs) as important regulatory elements for gene expression, but their intrinsic properties remain incompletely characterized. Through an integrative analysis of Hi-C and ChIP-seq data, here we find that a significant fraction of SEs are hierarchically organized, containing both hub and non-hub enhancers. Hub enhancers share similar histone marks with non-hub enhancers, but are distinctly associated with cohesin and CTCF binding sites and disease-associated genetic variants. Genetic ablation of hub enhancers results in profound defects in gene activation and local chromatin landscape. As such, hub enhancers are the major constituents responsible for SE functional and structural organization.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-018-03279-9

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Definition of hierarchical SEs and hub enhancers based on Hi-C chromatin interactions in K562 cells. a Overview of pipeline. b Representative hierarchical (left) and non-hierarchical (right) SEs. For each 5 kb bin within SE, the frequency of chromatin interactions (left y-axis) of and the z-score (right y-axis) is shown. The dashed red line represents the threshold of z-score = 1.5. c The proportion of hierarchical and non-hierarchical SEs. (d) The ROSE ranking of hierarchical and non-hierarchical SEs. In box plots, the center line represents the median, the box limits represent the 25th and 75th percentiles and the whiskers represent the 5th and 95th percentiles. P values were calculated using Wilcoxon rank-sum test. *P < 0.05; **P < 0.01; ***P < 0.001. (e) GREAT functional analysis of hierarchical and non-hierarchical SEs
Fig. 1 Definition of hierarchical SEs and hub enhancers based on Hi-C chromatin interactions in K562 cells. a Overview of pipeline. b Representative hierarchical (left) and non-hierarchical (right) SEs. For each 5 kb bin within SE, the frequency of chromatin interactions (left y-axis) of and the z-score (right y-axis) is shown. The dashed red line represents the threshold of z-score = 1.5. c The proportion of hierarchical and non-hierarchical SEs. (d) The ROSE ranking of hierarchical and non-hierarchical SEs. In box plots, the center line represents the median, the box limits represent the 25th and 75th percentiles and the whiskers represent the 5th and 95th percentiles. P values were calculated using Wilcoxon rank-sum test. *P < 0.05; **P < 0.01; ***P < 0.001. (e) GREAT functional analysis of hierarchical and non-hierarchical SEsCC BY · PubMed Central
Fig. 2 Chromatin landscapes at hub enhancers in K562 cells. a–f Spatial distribution of chromatin marks centered by enhancers in four groups, hub (n = 444), non-hub (n = 2303) enhancers, enhancers in non-hierarchical SEs (n = 4290) and regular enhancers (n = 22477): H3K4me1 (a), H3K27ac (b), DNase I hypersensitivity (c), master regulators GATA1 (d) and TAL1 (e), coactivator p300 (f). P values were calculated using Student’s t-test based on the ChIP-seq signal intensity within 1 kb window centered by enhancers. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant
Fig. 2 Chromatin landscapes at hub enhancers in K562 cells. a–f Spatial distribution of chromatin marks centered by enhancers in four groups, hub (n = 444), non-hub (n = 2303) enhancers, enhancers in non-hierarchical SEs (n = 4290) and regular enhancers (n = 22477): H3K4me1 (a), H3K27ac (b), DNase I hypersensitivity (c), master regulators GATA1 (d) and TAL1 (e), coactivator p300 (f). P values were calculated using Student’s t-test based on the ChIP-seq signal intensity within 1 kb window centered by enhancers. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significantCC BY · PubMed Central
Fig. 3 CTCF binding at hub enhancers within hierarchical SEs in K562 cells. a–c Spatial distribution of two cohesin components SMC3 (a) and RAD21(b), and CTCF (c), centered by enhancers in four groups. P values were calculated using Student’s t-test based on the ChIP-seq signal intensity of 1 kb window centered by enhancers. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. d Percentage of hub enhancers with (purple) or without (red) overlapping with TAD boundaries collected from the literature15. The CTCF ChIP-seq peaks/motif-sites associated with hub enhancers overlapping with TAD boundaries were excluded for analysis in e–h. e,f CTCF binding consensus across cell types (e) and CTCF-motif-matching score (f) of CTCF peaks in different contexts: hub (red), non-hub enhancers (blue) and TAD boundaries (purple). For each CTCF peak in K562, the consensus score (y-axis) was quantified as the percentage of cell types containing the same CTCF peak. In box plots, the center line represents the median, the box limits represent the 25th and 75th percentiles and the whiskers represent the 5th and 95th percentiles. P-values were calculated using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. g Sequence conservation around CTCF motif sites. The sitepro plots were centered by CTCF motif sites. P values were calculated using Student’s t-test based on the PhastConst100way score (y-axis) within CTCF motif sites. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. h Somatic mutation rate in cancers collected from IGGC around CTCF motif sites. The sitepro plots were centered by CTCF motif sites with 10 bp smoothing window. P-values were calculated using Fisher’s exact test based on overlap between CTCF motif sites and somatic mutation sites. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. i Model of the hierarchical organization of SEs containing both hub and non-hub enhancers. A hub enhancer is highly enriched with CTCF and cohesin binding, and functions as an organization hub to coordinate the non-hub enhancers and other distal regulatory elements within and beyond the SE
Fig. 3 CTCF binding at hub enhancers within hierarchical SEs in K562 cells. a–c Spatial distribution of two cohesin components SMC3 (a) and RAD21(b), and CTCF (c), centered by enhancers in four groups. P values were calculated using Student’s t-test based on the ChIP-seq signal intensity of 1 kb window centered by enhancers. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. d Percentage of hub enhancers with (purple) or without (red) overlapping with TAD boundaries collected from the literature15. The CTCF ChIP-seq peaks/motif-sites associated with hub enhancers overlapping with TAD boundaries were excluded for analysis in e–h. e,f CTCF binding consensus across cell types (e) and CTCF-motif-matching score (f) of CTCF peaks in different contexts: hub (red), non-hub enhancers (blue) and TAD boundaries (purple). For each CTCF peak in K562, the consensus score (y-axis) was quantified as the percentage of cell types containing the same CTCF peak. In box plots, the center line represents the median, the box limits represent the 25th and 75th percentiles and the whiskers represent the 5th and 95th percentiles. P-values were calculated using Student’s t-test. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. g Sequence conservation around CTCF motif sites. The sitepro plots were centered by CTCF motif sites. P values were calculated using Student’s t-test based on the PhastConst100way score (y-axis) within CTCF motif sites. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. h Somatic mutation rate in cancers collected from IGGC around CTCF motif sites. The sitepro plots were centered by CTCF motif sites with 10 bp smoothing window. P-values were calculated using Fisher’s exact test based on overlap between CTCF motif sites and somatic mutation sites. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. i Model of the hierarchical organization of SEs containing both hub and non-hub enhancers. A hub enhancer is highly enriched with CTCF and cohesin binding, and functions as an organization hub to coordinate the non-hub enhancers and other distal regulatory elements within and beyond the SECC BY · PubMed Central
Fig. 4 Enrichment of genetic variants associated with cell-type-specific gene expression and diseases in hub enhancers in K562 cells. a–c Enrichment of the eQTLs curated in GTEx in the enhancers in three groups, using randomly selected genomic regions as control (see Methods). The GTEx eQTL identified in all tissues (a) were separated into two subsets, identified in whole blood (b) or other tissues (c). The number of enhancers overlap with eQTLs in each group was labelled on each bar. P values were calculated using Fisher’s exact test. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. d–f Enrichment of the disease or traits-associated SNPs curated in GWAS catalog in the enhancers in three groups, using randomly selected genomic regions as control. The GWAS SNPs associated all diseases/traits (d), were separated into two subsets, associated with blood-related diseases/traits (e) or other traits (f). The number of enhancers overlap with SNPs in each group was labelled on each bar. P values were calculated using Fisher’s exact test. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. g–i Enrichment of GWAS SNPs in hub and non-hub enhancers, which were defined based on chromatin interactions after filtering a specific subtype of chromatin interactions, enhancer-CTCF (g), enhancer-enhancer (h) or enhancer-promoter (i). The fold-change between hub and non-hub enhancers were labelled
Fig. 4 Enrichment of genetic variants associated with cell-type-specific gene expression and diseases in hub enhancers in K562 cells. a–c Enrichment of the eQTLs curated in GTEx in the enhancers in three groups, using randomly selected genomic regions as control (see Methods). The GTEx eQTL identified in all tissues (a) were separated into two subsets, identified in whole blood (b) or other tissues (c). The number of enhancers overlap with eQTLs in each group was labelled on each bar. P values were calculated using Fisher’s exact test. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. d–f Enrichment of the disease or traits-associated SNPs curated in GWAS catalog in the enhancers in three groups, using randomly selected genomic regions as control. The GWAS SNPs associated all diseases/traits (d), were separated into two subsets, associated with blood-related diseases/traits (e) or other traits (f). The number of enhancers overlap with SNPs in each group was labelled on each bar. P values were calculated using Fisher’s exact test. *P < 0.05; **P < 0.01; ***P < 0.001, n.s. not significant. g–i Enrichment of GWAS SNPs in hub and non-hub enhancers, which were defined based on chromatin interactions after filtering a specific subtype of chromatin interactions, enhancer-CTCF (g), enhancer-enhancer (h) or enhancer-promoter (i). The fold-change between hub and non-hub enhancers were labelledCC BY · PubMed Central
Fig. 5 In situ genome editing reveals distinct requirement of hub vs non-hub enhancers in SE function. a Chromatin signatures and TF occupancy at the MYO1D SE locus in K562 cells are shown. The identified hub and non-hub enhancers are depicted by red (hub) and blue (non-hub) lines, respectively. The Hi-C chromatin interaction z-score and frequency at 5 kb resolution is shown at the bottom. The positions of sgRNAs used for CRISPRi or CRISPR-Cas9-mediated knockout analyses are shown as arrowheads. b Chromatin signatures and TF occupancy at the SMYD3 SE locus in K562 cells are shown. c Schematic of CRISPRi-mediated repression of hub or non-hub enhancers. d, e Expression of MYO1D and SMYD3 mRNA in untreated (control), CRISPRi-mediated repression of hub or non-hub enhancers. The mRNA expression levels related to GAPDH determined by qRT-PCR are shown. Each colored circle represents an independent biological replicate experiment. Results are means ± s.e.m. of 3 or 6 independent experiments. P values were calculated by two-sided Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, n.s. not significant. f Schematic of CRISPR-Cas9-mediated knockout of hub or non-hub enhancers. g Expression of all genes within the SE-containing TAD domain in unmodified (control), CRISPR-Cas9-mediated knockout of hub, non-hub enhancers or the CTCF binding site within the hub enhancer. The mRNA expression levels relative to GAPDH are shown. Each colored circle represents an independent single-cell-derived homozygous enhancer knockout clone. A schematic of the SE-containing TAD domain and associated genes are shown on the top. Results are means ± s.e.m. of at least 4 independent experiments. P values were calculated by a two-sided Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, n.s. not significant
Fig. 5 In situ genome editing reveals distinct requirement of hub vs non-hub enhancers in SE function. a Chromatin signatures and TF occupancy at the MYO1D SE locus in K562 cells are shown. The identified hub and non-hub enhancers are depicted by red (hub) and blue (non-hub) lines, respectively. The Hi-C chromatin interaction z-score and frequency at 5 kb resolution is shown at the bottom. The positions of sgRNAs used for CRISPRi or CRISPR-Cas9-mediated knockout analyses are shown as arrowheads. b Chromatin signatures and TF occupancy at the SMYD3 SE locus in K562 cells are shown. c Schematic of CRISPRi-mediated repression of hub or non-hub enhancers. d, e Expression of MYO1D and SMYD3 mRNA in untreated (control), CRISPRi-mediated repression of hub or non-hub enhancers. The mRNA expression levels related to GAPDH determined by qRT-PCR are shown. Each colored circle represents an independent biological replicate experiment. Results are means ± s.e.m. of 3 or 6 independent experiments. P values were calculated by two-sided Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, n.s. not significant. f Schematic of CRISPR-Cas9-mediated knockout of hub or non-hub enhancers. g Expression of all genes within the SE-containing TAD domain in unmodified (control), CRISPR-Cas9-mediated knockout of hub, non-hub enhancers or the CTCF binding site within the hub enhancer. The mRNA expression levels relative to GAPDH are shown. Each colored circle represents an independent single-cell-derived homozygous enhancer knockout clone. A schematic of the SE-containing TAD domain and associated genes are shown on the top. Results are means ± s.e.m. of at least 4 independent experiments. P values were calculated by a two-sided Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, n.s. not significantCC BY · PubMed Central
Fig. 6 Effect on chromatin landscape and TF binding upon KO of the MYO1D hub enhancer. a ChIP-qPCR analysis of H3K27ac, H3K4me3, GATA1, TAL1 and IgG (negative control) in unmodified (control), hub, non-hub enhancer or the CTCF binding site within the hub enhancer knockout cells. Primers against MYO1D and TMEM98 promoters, hub and non-hub enhancers, and a negative control genome region (chr2:211,337,339–211,337,429) are used. The results are shown as fold enrichment of the ChIP signals against the negative control region as means ± s.e.m. of four independent experiments. It is important to note that the low or lack of ChIP signals at the non-hub or hub enhancer regions is due to the deletion of the non-hub or hub enhancers containing the primer binding sites. P values were calculated by a two-sided Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, n.s. not significant. b ChIP-seq analysis of H3K27ac, GATA1 and TAL1 in control (WT), hub or non-hub enhancer knockout K562 cells. Browser view of the ChIP-seq intensity from two independent ChIP-seq experiments (rep1 and rep2) is shown. The identified hub and non-hub enhancers are depicted by red (hub) and blue (non-hub) lines, respectively. The Hi-C chromatin interaction z-score and frequency at 5 kb resolution is shown at the bottom (see Methods). The positions of sgRNAs used for CRISPRi or CRISPR-Cas9-mediated knockout analyses are shown as arrowheads
Fig. 6 Effect on chromatin landscape and TF binding upon KO of the MYO1D hub enhancer. a ChIP-qPCR analysis of H3K27ac, H3K4me3, GATA1, TAL1 and IgG (negative control) in unmodified (control), hub, non-hub enhancer or the CTCF binding site within the hub enhancer knockout cells. Primers against MYO1D and TMEM98 promoters, hub and non-hub enhancers, and a negative control genome region (chr2:211,337,339–211,337,429) are used. The results are shown as fold enrichment of the ChIP signals against the negative control region as means ± s.e.m. of four independent experiments. It is important to note that the low or lack of ChIP signals at the non-hub or hub enhancer regions is due to the deletion of the non-hub or hub enhancers containing the primer binding sites. P values were calculated by a two-sided Student’s t-test. *P < 0.05, **P < 0.01, ***P < 0.001, n.s. not significant. b ChIP-seq analysis of H3K27ac, GATA1 and TAL1 in control (WT), hub or non-hub enhancer knockout K562 cells. Browser view of the ChIP-seq intensity from two independent ChIP-seq experiments (rep1 and rep2) is shown. The identified hub and non-hub enhancers are depicted by red (hub) and blue (non-hub) lines, respectively. The Hi-C chromatin interaction z-score and frequency at 5 kb resolution is shown at the bottom (see Methods). The positions of sgRNAs used for CRISPRi or CRISPR-Cas9-mediated knockout analyses are shown as arrowheadsCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 74 references

Cited by

Sorted by citations

See all 258

Related

Metadata

Authors
Huang, J., Li, K., Cai, W., Liu, X., Zhang, Y., Orkin, S. H., Xu, J., & Yuan, G.-C.
Affiliations
  • Boston Children's Hospital (US)
  • Children's Medical Center (US)
  • Harvard University (US)
Corresponding author
  • Xu, Jian
  • Yuan, Guo-Cheng
ORCID
Venue
Nature Communications
Publisher
Springer Science and Business Media LLC
Published
2018-03-05
Volume
9
Issue
1
Pages
943-943
ISSN
2041-1723 (ISSN-L)
DOI
10.1038/s41467-018-03279-9
PMC
PMC5838163
OpenAlex
W2952336494
Cited by
241
References
74
Citations by year
2026: 22 · 2025: 20 · 2024: 22 · 2023: 27 · 2022: 37 · 2021: 43 · 2020: 30 · 2019: 25 · 2018: 14
FWCI
11.38
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Digital Access to Scholarship at Harvard (DASH) (Harvard University) (submittedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • American Cancer Society (IRG‐02‐196, 02-196)
  • Cancer Prevention and Research Institute of Texas (RR140025)
  • University of Texas Southwestern Medical Center
  • National Institutes of Health
  • National Heart, Lung, and Blood Institute (R01HL119099)
  • National Human Genome Research Institute (R01HG009663)
  • National Institute of Diabetes and Digestive and Kidney Diseases (K01DK093543, R03DK101665, R01DK111430)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Chromatin Dynamics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Chromatin Remodeling and Cancer
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cancer Research › Cancer-related molecular mechanisms research
Keywords
super-enhancers, chromatin interactions, cohesin, CTCF binding sites, histone modifications, disease-associated variants, gene activation, chromatin landscape
MeSH terms
Gene Editing, CCCTC-Binding Factor, Cohesins, Chromatin, Chromosomal Proteins, Non-Histone, Disease, Enhancer Elements, Genetic, Gene Expression Regulation, Humans, Models, Genetic, Protein Binding, Genetic Variation, Cell Cycle Proteins, K562 Cells, Gene Knockout Techniques, Human Umbilical Vein Endothelial Cells, CRISPR-Cas Systems

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review