Fig. 4: BRD4 is required for genome-wide enhancer transcription.

Heatmap (a) and boxplot quantifications (b) of BRD4 occupancy (FE) at BRD4 peaks associated with promoter regions, extragenic-, and intragenic enhancers (one-sided Wilcoxon rank sum test; ****p < 2.2e−16, ns: 0.72). ns: not significant. c Quantification of TPM normalized Pol II occupancy measured by HiS-NET-seq (two biological replicate measurements) at putative enhancer regions and FANTOM5 enhancers depending on their respective BRD4 levels. Boxplot shows four quantiles (Q1-Q4), from the lowest to the strongest BRD4 signal (one-sided Wilcoxon rank sum test; ****p < 2.2e−16). a–c Mean BRD4 ChIP-Rx signals are shown for two biological replicate measurements. b, c See Fig. 3d legend for boxplot definition. d Scheme for BRD4-specific degradation using the PROTAC degrader dTAG7¹²⁰. E2: ubiquitin-conjugating enzyme; E3: ubiquitin ligase; Ub: ubiquitin. Created with Biorender.com. e Immunoblot for both BRD4 isoforms upon treatment with 500 nM dTAG7 and the DMSO control. An antibody against the HA tag, which was integrated along with the degron tag, was used. GAPDH served as a loading control. Experiment was performed in duplicate. Source data are provided as a Source Data file. f–h Pol II occupancy changes (log2) between 2 h of dTAG7 treatment and the DMSO control. The significance reports the FDR-adjusted p-values (padj) from the Wald test calculated by DEseq2 as described in the Method section. The red and blue data points mark significant changes with an padj smaller than 0.05. The number of genes or enhancers with decreased or increased Pol II occupancy are indicated within blue or red frames, respectively. Data is RLE normalized (see Methods) to spike-in controls from mouse NIH/3T3 cells. NET-seq data was re-analyzed from Arnold et al.⁴⁷. Analyzed are active genes (n = 10,789) (f), extragenic- (n = 5900) and intragenic enhancers (n = 4954) (g, h). h The Boxplot distinguishes between Pol II occupancy changes (log2; two-sided Wilcoxon rank sum test; ****p < 2.2e−16) at putative enhancers with significant (blue, padj <0.05) and non-significant (gray, padj >0.05) reductions in DNA-bound BRD4 levels measured by ChIP-Rx upon 40 or 120 min of dTAG7 treatment. The padj values are derived from Wald tests calculated by DiffBind as described in the Method section. The boxplot shows the median values as the center. The box is defined by the first to the third interquartile range. The whiskers extend this interquartile range by a factor of 1.5, not exceeding the minimum or maximum values. Outlier measurements that exceed the whiskers are depicted as individual dots. a–h Data were obtained for human K562 dTAG-BRD4 cells.