Distinct IL‐1α‐responsive enhancers promote acute and coordinated changes in chromatin topology in a hierarchical manner

Abstract How cytokine‐driven changes in chromatin topology are converted into gene regulatory circuits during inflammation still remains unclear. Here, we show that interleukin (IL)‐1α induces acute and widespread changes in chromatin accessibility via the TAK1 kinase and NF‐κB at regions that are highly enriched for inflammatory disease‐relevant SNPs. Two enhancers in the extended chemokine locus on human chromosome 4 regulate the IL‐1α‐inducible IL8 and CXCL1‐3 genes. Both enhancers engage in dynamic spatial interactions with gene promoters in an IL‐1α/TAK1‐inducible manner. Microdeletions of p65‐binding sites in either of the two enhancers impair NF‐κB recruitment, suppress activation and biallelic transcription of the IL8/CXCL2 genes, and reshuffle higher‐order chromatin interactions as judged by i4C interactome profiles. Notably, these findings support a dominant role of the IL8 “master” enhancer in the regulation of sustained IL‐1α signaling, as well as for IL‐8 and IL‐6 secretion. CRISPR‐guided transactivation of the IL8 locus or cross‐TAD regulation by TNFα‐responsive enhancers in a different model locus supports the existence of complex enhancer hierarchies in response to cytokine stimulation that prime and orchestrate proinflammatory chromatin responses downstream of NF‐κB.


APPENDIX FIGURE LEGENDS
Appendix Figure S1. Direct modulation of heterochromatin decondensation and chromatin accessibility by NF-κB.
(A) Schematic representation visualizing the U2OS cell system, where a lacO array is stably integrated into centromeric heterochromatin of chromosome 2 (Jegou, Chung et al., 2009), allowing the visualization of binding by lacI-GFP fusion proteins as a single spot (green).  Appendix Figure S3. Cytokine array-based secretome analysis of enhancer mutant and ΔRELA HeLa cells.
(A) Design and representative example of the cytokine array used in this study. Vector controls cells were left untreated or were stimulated with IL-1α for 8 h. The supernatants were used to probe cytokine arrays coated with antibodies recognizing 80 cytokines or other inflammatory factors. After hybridizing to secondary HRP-streptavidin-conjugated antibodies signals were quantified using a chemiluminescence scanner and images were recorded. The red circles mark the top 10 most abundant factors.
(B) Quantification of inter-array normalized signals from two independent arrays. The left graph shows all signals (normalized to the mean signals of six biotinylated antibody positive controls on each array) and the right graphs shows the fold change upon IL-1α stimulation. Signals are sorted according to intensity in exp. 2. The red line marks the mean background defined by the mean of negative controls (no antibodies spotted) on all arrays. These data show that IL-6 and IL-8 are the most abundant secreted proteins and that most signals were close to or below background.
(C) IL-8 and IL-6 secretion was determined from vector controls, Δp65 eIL8 and ΔRELA cell lines that were treated as described in (A). Graphs show the results from specific ELISA of supernatants (mean ± S.D., n=2). Protein concentrations of IL-8 and IL-6 were normalized to total RNA obtained from the cell pellets.
(D) All supernatants shown in (C) were also subjected to cytokine array analysis as described in detail in (A). The only additional factor that was found to be significantly present above background in the IL-1α-stimulated conditions and to be down-regulated in both independent experiments in the mutant cells was MIP3-α (CCL20). The table shows normalized signal values (secretion) and ratios of signals between the stimulated conditions. Appendix Figure S4. Immuno-RNA FISH in p65 enhancer-mutant HeLa cells shows defects in NF-κB translocation and gene expression.
(A-C) Control (empty vector) and p65-deletion (Δp65 eIL8 , Δp65 eCXCL2 , and Δp65 eIL8+eCXCL2 ) HeLa cell lines ± IL-1α stimulation for 60 min were analyzed for nuclear translocation of p65 (top) and expression of IL8 (middle) or NFKBIA (bottom) transcripts. Data from three independent experiments were pooled. (A) i4C profiles in the 1 Mbp around the CXCL2 locus on chromosome 4 (ideogram) from control (empty vector) and enhancer-mutant (Δp65 eIL8 and Δp65 eCXCL2 ) HeLa lines ± IL-1α stimulation for 60 min. The average of two biological replicates is plotted, generated using the CXCL2 promoter (blue highlight) or enhancer (pink highlight) as a viewpoint. Below each profile significant strong (brown), medium (red) or weaker interactions (orange) called via foursig are indicated. All profiles are shown aligned to gene models (blue) and to CTCF, H3K4me1, H3K4me3, H3K27ac, and RNA polymerase II ENCODE ChIP-seq profiles from HeLa-S3 cells. The breadth of topologically-associating domains (TADs) in the locus is indicated above (rectangles).
(B) Meta-plots showing coverage of H3K27ac ChIP-seq signal at i4C fragments ± 1 kbp contacted by the CXCL2 promoter or enhancer in control cells (empty vector) in the presence (magenta) or absence (gray) of IL-1α stimulation for 60 min, and in enhancer-mutant cells (Δp65 eIL8 , blue; Δp65 eCXCL2 , green) after IL-1α stimulation.
(A) i4C profiles in the 1 Mbp around the CXCL2 and IL8 loci on chromosome 4 (ideogram) from p65knockout (ΔRELA) HeLa cells ± IL-1α stimulation for 60 min. Data were generated using the CXCL2 enhancer (blue highlight) or IL8 promoter (pink highlight) as viewpoints, and profiles are shown aligned to gene models (blue) and to CTCF, H3K4me1, H3K4me3, H3K27ac, and RNA polymerase II ENCODE ChIP-seq profiles from HeLa-S3 cells. The breadth of topologically-associating domains (TADs) in the locus is indicated above (rectangles).
(B) As in panel A, but using the promoters of the inducible BMP4 (pink highlight) and SAMD4A genes (blue highlight) on chromosome 14 as viewpoints ± TNFα stimulation for 60 min.