Fig 2: Induction of NFIL3 by acute ERS contributes to repression of genes involved in xenobiotic metabolism RT–qPCR analyses of Hlf, Tef, Dbp, and Nfil3 expression monitoring changes induced by acute ERS in MPH (3–5 independent experiments) (left panel) or mouse liver (five mice per group) (right panel). The bar graphs show means ± SD (standard deviations). One-sample t-test with BH correction for multiple testing was used to determine whether the mean Log2 FC ERS/control is statistically different from 0, *P < 0.05.Total protein extracts from MPH (left panel) or mouse liver (right panel) were subjected to Western blot with an antibody against NFIL3. ACTB was used as loading control. See Appendix Fig S7B for antibody validation.Chromatin fractions from MPH (upper panel) or mouse liver (lower panel) were subjected to Western blot with an antibody against NFIL3. Histone H3 was used as loading control.Enrichment scores from GSEA performed using LIVER-ID genes repressed by acute ERS in MPH (MPH ERS DOWN) as the gene set and liver transcriptomic changes induced by acute ERS and/or deletion of Nfil3 (NFIL3 KO) as the ranked gene lists were integrated and corrected for multiple testing using the BubbleGUM tool. For the NFIL3 KO ERS vs WT ERS comparison, the Core Enrichment genes (i.e., the subset of genes that contributes most to the enrichment result) were subjected to functional enrichment analyses using the ToppGene Suite. The top ranked KEGG Pathway with its Bonferroni-corrected P-value is shown.Box plots showing mRNA expression for 5 genes from the Core Enrichment from D involved in xenobiotic metabolism issued from the transcriptomic analyses. Shown are Log2 FC relative to the mean normalized expression in the WT control group (five mice per group). Box plots are composed of a box from the 25th to the 75th percentile with the median as a line and min to max as whiskers. Two-way ANOVA with Bonferroni's post hoc test was used to assess statistical significance, *P < 0.05.The Integrated Genome Browser (IGB) was used to visualize ChIP-seq profiles for NFIL3 (green) and several LIVER-ID TFs (red) in the mouse liver at the Gsta3 gene locus. Levels of H3K27ac in MPH and cells from the non-parenchymal fraction (NPC) are shown in blue. The grey bar indicates the position of a BRD4 SE.

Fig 3: Brd4 is dispensable for the maintenance of cell identity gene expression in adipocytes. Brd4 f/f mice were crossed with Adipoq-Cre to generate adipocyte-specific Brd4 KO mice. a qRT-PCR confirmation of Adipoq-Cre-mediated Brd4 deletion in mouse adipose tissues (n = 6 per group). b Western blot analysis of Brd4 in BAT and eWAT. c Representative morphology of mouse adipose tissues. d Adipose tissue mass was normalized to body weight (n = 6 per group). Statistical comparison between groups was performed using Student’s t -test. The data are presented as means ± SEM. e qRT-PCR of Pparg, Cebpa, Fabp4, and Ucp1 expression in adipose tissues (n = 4 per group). Quantitative PCR data in this figure are presented as means ± SEM. f, g Deletion of Brd4 does not affect adipocyte gene expression in adipose tissues. Equal amounts of total RNA from BAT and eWAT of four 12-week-old male mice were combined for RNA-Seq analysis. f Genome browser view of RNA-Seq analysis on Brd4 gene locus. The targeted exon 3 is highlighted in red box. g Heat map of RNA-Seq values of adipogenesis/BAT marker genes

Fig 4: Brd4 is required for Pol II binding on active enhancers during adipogenesis. a Schematic of genomic locations of Brd4+ enhancers (e1–e8) on Pparg, Cebpa, and Fabp4 loci. b Brd4 is dispensable for C/EBPß, MLL4, H3K4me1, and H3K27ac enrichment, but is essential for MED1, TBP, Pol II, and CDK9 binding, on adipogenic enhancers. ChIP-qPCR analyses of indicated factors are shown on adipogenic enhancers (e1–e8). Enhancers of constitutively expressed genes Arid1a (n1) and Jak1 (n2) were chosen as negative control regions. c Brd4 is required for eRNA transcription on adipogenic enhancers. The data are presented as means ± SD. Three technical replicates from a single experiment were used. d Model depicting that sequential actions of LDTFs, H3K4 mono-methyltransferases MLL3/MLL4, H3K27 acetyltransferases CBP/p300, epigenomic reader Brd4, transcription coactivator Mediator, and Pol II on enhancers control cell identity gene expression during differentiation

Fig 5: ERSEs are highly active in ERa-positive breast cancer. (A) MCF7 cells treated with estrogen (E2, 10-7 M, 1 h) were subjected to ChIP-seq with anti-ERa-specific antibody. Genomic distribution of ERa-binding sites is shown by a pie chart. (B) Normalized ERa ChIP-seq tag density in enhancer regions after clustering is shown. Estrogen-induced genes in the vicinity of representative ERSEs are shown. The number in parentheses indicates the rank of ChIP-seq tag density. (C) Metagene representation of ERa occupancy at typical enhancers (TEs) and super-enhancers (SEs) is shown. The x-axis shows the start and end of the TE (left) or SE (right) regions flanked by a 3 kb sequence. The y-axis shows the normalized tag density. (D) The characteristics of TEs and SEs defined by ERa are shown. (E) Metagene representation of H3K4me1, H3K4me2, H3K4me3, MED1, H3K27Ac, P300, H3K9me3 and H3K27me3 occupancy at TEs and SEs is shown. The x-axis shows the start and end of the TE (left) or SE (right) regions flanked by ± 3 kb sequence. The y-axis shows the normalized tag density. (F) UCSC Genome browser views of ERa, BRD4, H3K4me1, H3K4me2, H3K4me3, H3K27Ac, P300, MED1, H3K9me3 and H3K27me3 ChIP-seq in the presence or absence of estrogen on the SE region in the vicinity of the estrogen-induced gene, XBP1. The boxed region indicates SEs. (G) MCF7 cells treated or not with estrogen (E2, 10-7 M, 1 h) were subjected to Gro-seq. Genes positively and negatively regulated by estrogen based on Gro-seq, and those with an SE nearby are shown.