Spatial enhancer clustering and regulation of enhancer-proximal genes by cohesin - PubMed (original) (raw)

Spatial enhancer clustering and regulation of enhancer-proximal genes by cohesin

Elizabeth Ing-Simmons et al. Genome Res. 2015 Apr.

Abstract

In addition to mediating sister chromatid cohesion during the cell cycle, the cohesin complex associates with CTCF and with active gene regulatory elements to form long-range interactions between its binding sites. Genome-wide chromosome conformation capture had shown that cohesin's main role in interphase genome organization is in mediating interactions within architectural chromosome compartments, rather than specifying compartments per se. However, it remains unclear how cohesin-mediated interactions contribute to the regulation of gene expression. We have found that the binding of CTCF and cohesin is highly enriched at enhancers and in particular at enhancer arrays or "super-enhancers" in mouse thymocytes. Using local and global chromosome conformation capture, we demonstrate that enhancer elements associate not just in linear sequence, but also in 3D, and that spatial enhancer clustering is facilitated by cohesin. The conditional deletion of cohesin from noncycling thymocytes preserved enhancer position, H3K27ac, H4K4me1, and enhancer transcription, but weakened interactions between enhancers. Interestingly, ∼ 50% of deregulated genes reside in the vicinity of enhancer elements, suggesting that cohesin regulates gene expression through spatial clustering of enhancer elements. We propose a model for cohesin-dependent gene regulation in which spatial clustering of enhancer elements acts as a unified mechanism for both enhancer-promoter "connections" and "insulation."

© 2015 Ing-Simmons et al.; Published by Cold Spring Harbor Laboratory Press.

PubMed Disclaimer

Figures

Figure 1.

Figure 1.

Cohesin facilitates the regulated expression of genes located near enhancer elements. (A) Enhancers and super-enhancers (Supplemental Fig. S1) emerge as major determinants of cohesin-dependent gene regulation. The “Next to enhancer” category includes genes that are nearest neighbors of conventional enhancers. The “Near SE” category includes genes that are positioned within 40 kb of an SE, including genes that overlap SEs and nearest neighbors of SEs. “Next to SE” includes only nearest neighbors of SEs. “Diff. Hi-C interaction” indicates genes in 100-kb regions that interact differentially in control and _Rad21_-deficient thymocytes (Seitan et al. 2013). Other categories refer to promoter binding of the listed factors (TSS ± 2.5 kb). Multivariate multinomial regression analysis was done as described in Seitan et al. (2013). The statistical significance of regression model coefficients is indicated and only significant associations are shown. (B) Proximity to enhancers accounts for ∼50% of genes deregulated in _Rad21_-deficient thymocytes. (Enhancer) Genes that are nearest neighbors or overlap conventional enhancers; (SE) genes ±40 kb of a super-enhancer, overlapping a super-enhancer, or that are nearest neighbors of super-enhancers. (C) Cohesin is required for the regulation of genes near enhancers and super-enhancers. (Genome average) Expressed genes (17,003 genes); (Next to enhancer) nearest neighbors of conventional enhancers (3540 genes); (Overlapping enhancer) genes that overlap conventional enhancers (198 genes); (Near SE) ±40 kb of a super-enhancer (1036 genes); (Overlapping SE) overlap a super-enhancer (493 genes); (Next to SE) nearest neighbors of super-enhancers (447 genes). Additional data are shown in Supplemental Table 1.

Figure 2.

Figure 2.

Enhancer elements are maintained in cohesin-deficient thymocytes. The Cd8 (A) and the Ppp1r16b region (B) illustrate the maintenance of H3K27ac in cohesin-deficient thymocytes. Binding of the cohesin subunits RAD21 and SMC1A as well as CTCF is shown for reference (see below). (C) Rad21 deletion does not abolish H3K27ac marking of conventional enhancers (left), newly established developmentally regulated enhancers (red, middle), or super-enhancers (right). (D) Enhancers and super-enhancers were stratified according to the ratio of H3K27ac in _Rad21_-deficient thymocytes over wild type. The frequency of deregulated genes is shown for the top, middle, and lower third of enhancers (left) and super-enhancers (right).

Figure 3.

Figure 3.

Enhancer elements are enriched for CTCF and cohesin binding. (A) CTCF, RAD21, and SMC1A ChIP-seq signal enrichment in thymocytes at conventional enhancers. Enhancers were defined based on H2K27ac ChIP-seq data (Methods) and enhancer length was normalized in order to align the start and end of the enhancer. (B) CTCF, RAD21, and SMC1A ChIP-seq signal enrichment in thymocytes at super-enhancers. Super-enhancers were defined based on H2K27ac ChIP-seq data (Methods), enhancer length was normalized in order to align the start and end of the super-enhancers, and flanking regions of equal size to the super-enhancer are shown for reference. “Start” and “end” are based on the genomic coordinates. (C) Classification of super-enhancers on the basis of CTCF binding at both boundaries, one boundary, or neither boundary. Heatmaps of ChIP-seq signal enrichment in 100-kb windows around super-enhancer centers, grouped according to CTCF binding at the super-enhancer boundaries. Histone modifications, the cohesin loading factor NIPBL, and the cohesin subunits RAD21 and SMC1A are shown for reference.

Figure 4.

Figure 4.

Cohesin mediates the spatial clustering of enhancer elements. (A) The mouse Cd3 super-enhancer is flanked and punctuated by CTCF and cohesin binding (refer to Fig. 2A,B for additional examples of the relationship between super-enhancers, CTCF, and cohesin). Restriction fragments used for 3C analysis are indicated by gray bars. “A” is a proximity ligation control that demonstrates comparable efficiency of 3C experiments. Enhancer-enhancer interactions B, C, and D across the Cd3 super-enhancer were significantly reduced in _Rad21-_deficient (red) compared to control (black) thymocytes. n = 3, mean ± SD, (*) P < 0.05. (B) Long-range interactions between the Cd3 super-enhancer and enhancer elements outside the Cd3 locus. “E” is a control used to demonstrate background interactions with a downstream genomic fragment lacking H3K27ac marks and CTCF binding. The position of the Cd3 super-enhancer is marked by dashed lines. Interactions F, G, H, I, and J link the Cd3 super-enhancer with downstream enhancer elements outside the Cd3 region and were significantly reduced in _Rad21-_deficient (red) compared to control (black) thymocytes. n = 3, mean ± SD, (*) P < 0.05. (C) Structured interaction matrix analysis (SIMA) of long-range interactions between chromatin features based on Hi-C data for control (gray) and _Rad21_-deficient (red) thymocytes. Interactions between constituent elements of super-enhancers (left) were analyzed by SIMA within super-enhancer regions sized 100 kb and larger. Interactions between other chromatin features were analyzed by SIMA within open chromatin compartments. Note that chromatin features show increased self-interactions, while interactions of random regions (Seitan et al. 2013) conformed to the level of interactions predicted by a background model based on genomic distance and sequencing depth (dashed red line). “Interaction strength” refers to the strength of interactions between 10-kb regions within super-enhancers (SEs) or 10-kb regions within chromosomal compartments (Enh, RAD21 sites, TSS, and random regions after normalization to the background model). _P_-values shown are from a Wilcoxon signed-rank test. (D) Reduced enhancer-enhancer interactions in _Rad21_-deficient thymocytes based on SIMA analysis of Hi-C data are compared for all open chromosomal compartments and separately for compartments with or without super-enhancers. _P_-values shown are from a Wilcoxon signed-rank test.

Figure 5.

Figure 5.

Cell-type specificity of CTCF association and spatial clustering of super-enhancers. (A) CTCF (top), RAD21 (middle), and SMC1A (bottom) ChIP-seq signal in thymocytes at super-enhancers that are active in thymocytes (black), macrophages (purple), myoblasts (blue), or ES cells (green). (B) Preferential interactions within cell-type-specific super-enhancers. SIMA analysis of thymocyte Hi-C data was used to compare interactions within super-enhancers active in thymocytes or other cell types. (Left) Hi-C interactions in control (gray) and _Rad21_-deficient thymocytes (red) within thymocyte super-enhancers sized 50 kb or more (n = 105). (Right) Hi-C interactions in WT (light gray) and _Rad21_-deficient thymocytes (orange) within super-enhancers of 50 kb or more that are active in pro-B cells, macrophages, or ES cells (n = 20) (Whyte et al. 2013). _P_-values shown are from a Wilcoxon signed-rank test. (C) Model for the impact of cohesin-dependent enhancer-enhancer interactions on gene expression. Spatial clustering between enhancer elements can affect promoter activity positively and negatively. The promoters P1 and P3 are both distal to a super-enhancer. P1 is isolated as a result of the spatial clustering between the enhancer elements within the SE, while P3 is connected to the super-enhancer by enhancer-enhancer contacts (left). Removal of cohesin (right) decreases the spatial constraint on enhancer elements so that P1 is contacted more readily by enhancer elements, while P3 dissociates from the super-enhancer.

Similar articles

Cited by

References

    1. Akalin A, Franke V, Vlahoviček K, Mason CE, Schübeler D. 2014. genomation: a toolkit to summarize, annotate and visualize genomic intervals. Bioinformatics 10.1093/bioinformatics/btu775. - DOI - PubMed
    1. Anders S, Huber W. 2010. Differential expression analysis for sequence count data. Genome Biol 11:R106. - PMC - PubMed
    1. Dixon JR, Selvaraj S, Yue F, Kim A, Li Y, Shen Y, Hu M, Liu JS, Ren B. 2012. Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature 485:376–380. - PMC - PubMed
    1. Dowen JM, Fan ZP, Hnisz D, Ren G, Abraham BJ, Zhang LN, Weintraub AS, Schuijers J, Lee TI, Zhao K, et al.2014. Control of cell identity genes occurs in insulated neighborhoods in mammalian chromosomes. Cell 159:374–387. - PMC - PubMed
    1. Faure AJ, Schmidt D, Watt S, Schwalie PC, Wilson MD, Xu H, Ramsay RG, Odom DT, Flicek P. 2012. Cohesin regulates tissue-specific expression by stabilizing highly occupied cis-regulatory modules. Genome Res 22:2163–2175. - PMC - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources