Fig. 4: Histone modifications at CXCL super-enhancer and CXCL promoter sites modulate chemokine gene expression.
From: Super enhancer regulation of cytokine-induced chemokine production in alcoholic hepatitis
a Schematic of dCas9-KRAB binding with sgRNA leading to epigenetic silencing. b dCas9-KRAB fusion protein targeting the selected NFκB site within CXCL super-enhancer suppressed CXCL expression in LSECs. A sgRNA with specificity for a predicted NFκB binding site on the CXCL enhancer decreased CXCL1, 6, and 8 expressions revealed by qPCR but did not affect the expression of MTHFD2L, a nearby noninflammatory gene (negative control) (n = 8, biologically independent samples from four independent experiments). Changes in chemokine expression were calculated as fold change over basal expression and log10 (fold change) was plotted on the y-axis. ELISA for CXCL1 was performed on supernatants and mirrored the pattern seen from qPCR (n = 6, biologically independent samples from three independent experiments). Data represented as mean ± SD. Two-way matched-pairs ANOVA was performed on log-transformed fold change values with post hoc Tukey’s multiple comparison correction. c Another sgRNA targeting a predicted NF-κB binding site on CXCL1 promoter decreased expression of CXCL1, but not other CXCLs by qPCR (n = 8, biologically independent samples from four independent experiments) and CXCL1 ELISA (n = 6, biologically independent samples from three independent experiments). Log10 (fold change) was plotted. Data represented as mean ± SD. Two-way matched-pairs ANOVA was performed on log-transformed fold change values, with post hoc Tukey’s multiple comparison correction. d ChIP-qPCR for H3K9me3 on dCas9-KRAB treated cells. dCas9-KRAB was co-transduced with sgRNA targeting CXCL promoter (P), CXCL super-enhancer (E), or empty vector (C), and treated with or without TNFα. ChIP-qPCR was performed with anti-H3K9me3 antibody or isotype control. Enrichment for either CXCL1 promoter or CXCL super-enhancer sequence was examined. Y-axis plots percent input (n = 3 from biologically independent samples). Data represented as mean ± SD. Two-way matched-pairs ANOVA was performed with post hoc Dunnett’s multiple comparison correction.
