Skip to main content
Fig. 7 | Genome Biology

Fig. 7

From: Insulator-based loops mediate the spreading of H3K27me3 over distant micro-domains repressing euchromatin genes

Fig. 7

Insulator protein mutants impair H3K27me3 micro-domains depending on CP190 recruitment. a Upper: scheme representing the 3D-based formation of micro-domains involving the indicated molecular players of long-range interactions (LRIs-3; see panels b, c, e). Lower: scheme representing the impact of Beaf32 looping mutants on insulator-mediated LRIs by GAF /dCTCF and CP190 co-factors that results in both distant spreading onto micro-domains and (gain) in local spreading at borders. b Averaged H3K27me3 levels of previously identified micro-domains using normalized ChIP-seq data from cells expressing Beaf32 mutants compared to control cells (see “Methods”). c Upper panel: Box plot quantifying CP190 binding as normalized ChIP-seq reads in cells expressing Beaf32 mutants compared to control (see “Methods”) in two groups of micro-domains harboring decreasing CP190 levels (left) or not (right). Middle panel: Box plot showing the levels of H3K27me3 binding (normalized ChIP-seq reads) in Beaf32 mutants compared to control for the same micro-domains harboring decreasing CP190 levels or not (as defined in the upper panel). H3K27me3 levels were measured depending on proximity of micro-domains (< 1 kbp; lower panel) or not (> 5 kb; middle panel) to borders. d Box plot representing the differential gene expression analysis from RNAseq (as a log ratio of normalized RNAseq reads; see “Methods”) in Beaf32 mutants compared to control cells. A global negative influence of micro-domains was evidenced by a significant upregulation of the flanked genes (p value 1e−4; by the Wilcoxon pairwise test; for a total of 147 TSSs flanked by a micro-domain; see Tables S1-S2) as compared to control genes not flanked by a micro-domain or to genes near a Beaf32 site (“Insulator type 1”). Insulator-2 represents GAF or dCTCF. e Quantification of local spreading (at borders) depending on no bracketing (left), one-side bracketing by either Beaf32 (second box) or GAF/dCTCF CP190 (third box), and (as depicted in panel a) bracketing on both sides by Beaf32 and GAF/dCTCF CP190 (right; genomic context depicted in panel a). p values: Wilcoxon pairwise test for normalized levels in Beaf-KD/control cells (see Additional file 1: Fig. S6-S7)

Back to article page