Journal: The Journal of Experimental Medicine

Article Title: IFNγ-induced memory in human macrophages is sustained by the durability of cytokine signaling itself

doi: 10.1084/jem.20250976

Figure Lengend Snippet: IFNγ induces long-lasting transcription factor activity and chromatin accessibility after washout. Macrophages were treated with LPS, IFNγ, and LPS in the presence of ruxolitinib for 8 h, as in . Cells were washed and cultured for an additional 88 h. ATACseq was performed after 8 h of stimulation and 4 days after washout. (A) Heatmap of Z-scored reads within ATAC peaks induced by either LPS or IFNγ (L2FC > 2, FDR < 0.01). Clusters were generated by unsupervised k-means clustering. Each column represents a biological replicate from the same human donor. (B) Top enriched motifs in clusters from A. (C) Boxplot quantifying log2 cpm of reads within IFNγ-induced ATAC peaks before and after cytokine washout. (D) Boxplot quantifying log2 cpm of reads within LPS-induced ATAC peaks before and after cytokine washout. (E) Boxplot quantifying log2 cpm of reads within ATAC peaks induced by both IFNγ and LPS (L2FC > 2, FDR < 0.01 for each) peaks before and after cytokine washout. (F) Barplot quantifying percent of transcription factor-bound motifs within STAT1 and IRF1 (IFNγ) and IRF1 and NF-κB (LPS) within induced ATAC peaks in C and D for unstimulated, IFNγ/LPS-stimulated macrophages, and stimulated macrophages 4 days after washout. Motif binding predicted using TOBIAS ATACseq footprinting analysis. Results are average of two technical replicates from a single subject; error bars display standard deviation. (G) Human macrophages were stimulated with IFNγ (100 ng/ml), LPS (100 ng/ml), or IFNβ (10 ng/ml) for 8 h, washed, and then cultured for an additional 66 h. Cells were collected, and whole cell western blotting for phosphorylated STAT1 was performed at the indicated time points. Blot is representative of three replicates from two separate human donors. All box/whisker plots indicate interquartile range and 1.5× interquartile range. Statistical tests were determined by paired Wilcoxon test. ****P < 0.0001. Source data are available for this figure: .

Article Snippet: The following primary antibodies were used: pSTAT1 pY701.4A (#136229; Santa Cruz Biotechnology, RRID:AB_2019074) diluted at 1:10,000, IRF1 D5E4 (#8478; Cell Signaling Technologies, RRID:AB_10949108) diluted at 1:1,000, β-tubulin TUB2.1 (T5201; Sigma-Aldrich, RRID:AB_609915) diluted at 1:10,000, and GAPDH H-12 (#166574; Santa Cruz Biotechnology, RRID:AB_2107296) diluted at 1:10,000.

Techniques: Activity Assay, Cell Culture, Generated, Binding Assay, Footprinting, Standard Deviation, Western Blot, Whisker Assay

Journal: The Journal of Experimental Medicine

Article Title: IFNγ-induced memory in human macrophages is sustained by the durability of cytokine signaling itself

doi: 10.1084/jem.20250976

Figure Lengend Snippet: Cell surface–bound IFNγ mediates persistent JAK/STAT signaling even after cytokine washout. (A) Human macrophages were stimulated with IFNγ (100 ng/ml) for 8 h, washed, and then cultured in regular media or media containing ruxolitinib (1 µM) or increasing concentrations of anti-IFNγ neutralizing antibody for an additional 28 h. Cells were collected, and whole cell western blotting for phosphorylated STAT1 and IRF1 was performed at indicated time points. Representative blot of duplicates from two separate subjects. (B) Human macrophages were stimulated with 100 ng/ml IFNγ for 3 h, washed, and then cultured in either ruxolitinib (1 µM), anti-IFNγ neutralizing antibody (10 µg/ml), or isotype control antibody (10 µg/ml) for 2 h. Samples were collected at the indicated times for immunoblot. Representative blot of duplicates from two separate subjects. (C) Quantification of pSTAT1 band intensities from B. (D) Human macrophages were stimulated with 100 ng/ml IFNγ for 8 h, washed, and cultured for an additional 88 h in regular media. Supernatants from stimulated macrophages were collected after the 8-h stimulation and 88 h after washout. This supernatant was used to stimulate fresh macrophages for 1 h in the presence/absence of ruxolitinib (1 µM) or anti-IFNγ neutralizing antibody (10 µg/ml). As a control, fresh macrophages were stimulated with media supplemented with 1 ng/ml IFNγ for 1 h. Representative blot of duplicates from two separate subjects. (E) Macrophages were left in regular media or pre-treated with 10 µg/ml CHX for 15 min and stimulated with 100 ng/ml IFNγ for 3 h. Treated macrophages were washed and subsequently cultured for 2 h in regular media, media supplemented with 10 µg/ml CHX, or anti-IFNγ neutralizing antibody (10 µg/ml) and collected for immunoblot. Duplicates from one subject are shown. (F) Quantification of pSTAT1 band intensities in E normalized to band intensity of macrophages treated with IFNγ for 3 h. (G) Human macrophages were stimulated with 100 ng/ml IFNγ for 8 h, washed, and cultured in regular media or media supplemented with 1 µM ruxolitinib for 16 h. After 16 h, cells were washed again and cultured in regular media for an additional 24 h. Cells were collected for immunoblot at indicated times. Representative blot of four replicates from two subjects. (H) Quantification of pSTAT1 band intensities in G normalized to band intensity of macrophages treated with IFNγ for 3 h. Statistical tests were determined by a single-tailed t test. *P < 0.05, **P < 0.01, and ***P < 0.001. Source data are available for this figure: .

Article Snippet: The following primary antibodies were used: pSTAT1 pY701.4A (#136229; Santa Cruz Biotechnology, RRID:AB_2019074) diluted at 1:10,000, IRF1 D5E4 (#8478; Cell Signaling Technologies, RRID:AB_10949108) diluted at 1:1,000, β-tubulin TUB2.1 (T5201; Sigma-Aldrich, RRID:AB_609915) diluted at 1:10,000, and GAPDH H-12 (#166574; Santa Cruz Biotechnology, RRID:AB_2107296) diluted at 1:10,000.

Techniques: Cell Culture, Western Blot, Control

Journal: The Journal of Experimental Medicine

Article Title: IFNγ-induced memory in human macrophages is sustained by the durability of cytokine signaling itself

doi: 10.1084/jem.20250976

Figure Lengend Snippet: Extracellular IFNγ signaling sustains chromatin accessibility and ISG expression even after cytokine washout. (A) Schematic of experiments: Human macrophages were stimulated with 100 ng/ml IFNγ for 8 h, washed, and then cultured for an additional 88 h in regular media or media with 1 µM ruxolitinib for an additional 88 h. Cells were collected for ATACseq and RNAseq at the indicated time points. (B) Heatmap of Z-scored reads within ATAC peaks induced by IFNγ (L2FC > 2, FDR < 0.01) after 8 h of stimulation for 4 days after washout when cultured in regular media or media with 1 µM ruxolitinib. Clusters were generated by unsupervised k-means clustering. Each column represents a biological replicate from the same human donor. (C) Boxplot of log2CPM of reads within each peak for each cluster in B. (D) Heatmap of Log2 fold change in RNAseq reads of genes induced at least fivefold after 8 h of IFNγ stimulation. Log2 fold changes are shown after washout for cells cultured in regular media and media containing 1 µM ruxolitinib. Genes are clustered by persistent level of expression after washout (CPM after wash as percent of CPM at 8-h simulation). (E) Boxplot showing Log2 fold changes of individual genes by cluster in D. Box/whisker plots indicate interquartile range and 1.5× interquartile range. Statistical tests were determined by paired Wilcoxon test. (F) Macrophages were stimulated and washed as above in A; after washout, cells were cultured in media alone, media with 1 µM ruxolitinib, or 10 µg/ml anti-IFNγ neutralizing antibody for 88 h. Cells were collected 88 h after washout, and qPCR was performed for IDO1. Boxplots indicate 2 ΔΔCt normalized to HPRT. Error bars indicate standard deviation. Statistical tests determined by ordinary one way ANOVA. (G) qPCR for IRF1 as in F. ** P < 0.01; ***P < 0.001; ****P < 0.0001.

Article Snippet: The following primary antibodies were used: pSTAT1 pY701.4A (#136229; Santa Cruz Biotechnology, RRID:AB_2019074) diluted at 1:10,000, IRF1 D5E4 (#8478; Cell Signaling Technologies, RRID:AB_10949108) diluted at 1:1,000, β-tubulin TUB2.1 (T5201; Sigma-Aldrich, RRID:AB_609915) diluted at 1:10,000, and GAPDH H-12 (#166574; Santa Cruz Biotechnology, RRID:AB_2107296) diluted at 1:10,000.

Techniques: Expressing, Cell Culture, RNA sequencing, Generated, Whisker Assay, Standard Deviation

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Experimental design for multiomic assessment of WT and IRF1 KO bone marrow–derived macrophages (BMDMs) response to IFNγ stimulation. Time-resolved profiling by ATAC-seq, ChIP-seq, Hi-ChIP, SLAM-seq and metabolomics via GC/LC-MS is performed. (B) Venn diagram summarizing ATAC-seq–identified accessible chromatin regions, filtered for high-confidence peaks and IFNγ-responsiveness (n=38,564); this set is used for downstream clustering and differential analyses. (C) Heatmap of normalized ATAC-seq signal (rows = individual accessible site; columns = time points), grouped into eight clusters by k-means clustering. Clusters C1-C3 show IRF1-depedent increase in accessibility in response to IFNγ; highlighted in red. PU.1 ChIP-seq binding signal is also shown, with Cluster C1 lacking detectable PU.1 occupancy. (D) Ribbon plots of relative ATAC-seq peak height (each peak scaled to its maximum) over matched time points; lines indicate mean accessibility and shaded ribbons show ± SD for WT (black) and IRF1 KO (red). (E) Boxplots of normalized ATAC-seq counts in WT BMDMs at heterochromatin regions, and at clusters C1–C8 and unresponsive ATAC-seq sites; median with interquartile range are shown.

Article Snippet: Membranes were blocked in 5% skim milk in PBS Tween-20 0.1% and incubated with primary antibodies against IRF1 (R&D Systems, Cat. #AF4715, 1:200) or GAPDH (Cell Signaling Technologies, Cat. #2118) as loading control, followed by appropriate HRP-conjugated secondary antibodies.

Techniques: Derivative Assay, ChIP-sequencing, HiChIP, Liquid Chromatography with Mass Spectroscopy, Binding Assay

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Network diagrams of transcription factor motif frequency (node size) and co-occurrence (edge thickness) within ±100 bp of ATAC-seq peak centre for clusters C1-C3 and unresponsive sites. “IRF1–IRF1” denotes sites with ≥2 IRF motifs, and node/edge scales reflect motif frequency and co-occurrence. (B) Volcano plots of TOBIAS differential binding scores for 879 mammalian TFs in WT BMDMs comparing 0.5, 3 and 48 h post-IFNγ versus non-treated (0 h); significant TFs are highlighted [Bonferroni-corrected FDR < 0.05; log2 FC > |0.5|]. (C) Heatmap of centered TOBIAS TF footprinting intensity in WT BMDMs, grouped into four clusters by k-means clustering. (D) Density plots (top) and motif-centered TOBIAS footprint heatmaps (bottom) in WT and IRF1 KO BMDMs showing aggregated IRF1-centered footprinting signal at Cluster 1 sites (rows = individual sites; columns = base position around motif). (E) Representative Western blots in WT BMDMs showing IRF1 protein and GAPDH control across time points (0–48 h) post-IFNγ stimulation.

Article Snippet: Membranes were blocked in 5% skim milk in PBS Tween-20 0.1% and incubated with primary antibodies against IRF1 (R&D Systems, Cat. #AF4715, 1:200) or GAPDH (Cell Signaling Technologies, Cat. #2118) as loading control, followed by appropriate HRP-conjugated secondary antibodies.

Techniques: Binding Assay, Footprinting, Western Blot, Control

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Heatmaps of IRF1 occupancy (ChIP-seq), ATAC-seq accessibility and H3K4me1, H3K4me3 and H3K27ac signals across in response to IFNγ for sites in Clusters 1–3 in WT and IRF1 KO BMDMs. (B) Hi-ChIP arc plots showing loop contacts (arc width represents number of contacts) between IRF1-bound enhancers and promoters at C8 and unresponsive sites in WT and IRF1 KO BMDMs. [FitHiChIP thresholds FDR < 0.1; loop FC > 6, CPM > 6] (C) Graph of the temporal changes for ChIP-seq and ATAC-seq signals at Cluster 1. Half-time (t½) to reach 50% of each signal’s maximum was calculated by normalizing each trajectory to its maximum and extracting the pseudo-time at half-max. (D) Heatmap of ChIP-seq for IRF1, BRG1, ARID1A, BRD9 and PHF10 across Clusters 1–3 at 0, 1, and 4 h post TLR4 activation. ( E ) BRG1 ChIP–qPCR enrichment (fold over input) at four enhancers ( Wdr7 (C1), Shtn1 (C2), Clic5 (C2) , Nos2 (C3)) in WT and IRF1 KO BMDMs, untreated and 4 h post-IFNγ. (F) Boxplots of normalized ATAC-seq counts in Clusters 1–3 in WT BMDMs TLR4 activated with LipidA, with or without and BRG1 inhibition (BRM014).

Article Snippet: Membranes were blocked in 5% skim milk in PBS Tween-20 0.1% and incubated with primary antibodies against IRF1 (R&D Systems, Cat. #AF4715, 1:200) or GAPDH (Cell Signaling Technologies, Cat. #2118) as loading control, followed by appropriate HRP-conjugated secondary antibodies.

Techniques: ChIP-sequencing, HiChIP, Activation Assay, ChIP-qPCR, Inhibition

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Heatmaps of IRF1 ChIP–seq and normalized ATAC–seq at sites grouped by IRF1 signal strength (very strong to weak) in response to IFNγ. (B) Line plots of average IRF1 ChIP–seq signal in WT BMDMs for each binding-strength category. (C) Heatmap of relative enrichment of IRF1 binding classes across ATAC clusters (enrichment is relative to the maximum site overlap). (D) Stacked bar plots showing proportions of sites with 0, 1, 2 or ≥3 IRF1 motifs per ATAC cluster. (E) Aggregate plots of IRF1 motif frequency across ±100 bp around IRF1 peaks for each ATAC cluster. (F) Heatmap of IRF1 ChIP–seq signal at 3 h post–IFNγ for sites stratified by IRF1 motif count, as determined in D). (G) Scatter plot of fraction of sites forming IRF1 Hi-ChIP loops versus motif count, with a fitted trend line shown. [FitHiChIP thresholds FDR < 0.1; loop FC > 6, CPM > 6] (H) Genome browser tracks at the Jdp2 locus showing IRF1, PU.1 and H3K27ac ChIP–seq, Hi-ChIP interactions and ATAC–seq in WT and IRF1 KO BMDMs. The cluster to with each ATAC-seq peak belong is indicated [C1 = cluster 1; UR = Unresponsive]. Insets display the array of IRF1 motifs at the C1 site and a SLAM-seq Jdp2 expression plot across the IFNγ time course.

Article Snippet: Membranes were blocked in 5% skim milk in PBS Tween-20 0.1% and incubated with primary antibodies against IRF1 (R&D Systems, Cat. #AF4715, 1:200) or GAPDH (Cell Signaling Technologies, Cat. #2118) as loading control, followed by appropriate HRP-conjugated secondary antibodies.

Techniques: ChIP-sequencing, Binding Assay, HiChIP, Expressing

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Line plots of nascent RNA-seq log2 fold-change (FC) for genes within ±10 kb of ATAC cluster regions in WT and IRF1 KO BMDMs in response to IFNγ stimulation. (B) Heatmap of GO biological process enrichment for genes within ±50 kb of ATAC peaks. Categories with clusterProfiler FDR < 0.05 for at least one cluster are shown. (C) Line plots of nascent RNA counts per million (CPM; mean ± SD) for selected genes across IFNγ time points; WT vs IRF1 KO comparison by two-way ANOVA and pairwise post-hoc testing at each time point. (D) Genome browser tracks at the Kmt2c locus showing IRF1 and PU.1 ChIP-seq, Hi-ChIP arcs and ATAC-seq signal for WT and IRF1 KO BMDMs. [UR = Unresponsive] (E) Line plots of RNA-seq CPM (mean ± SD) for selected genes at 0, 1 and 4 h post-LipidA treatment in WT BMDMs, with BRM014 treatment at the 4 h time point. [Student T-test; n = 3] (F) Bar plot of log2 odds ratio of downregulated genes (FC < 0.5 and FDR < 0.05) after BRM014 treatment (4 h Lipid A) across clusters. * < 0.05, ** < 0.01, *** < 0.001

Article Snippet: Membranes were blocked in 5% skim milk in PBS Tween-20 0.1% and incubated with primary antibodies against IRF1 (R&D Systems, Cat. #AF4715, 1:200) or GAPDH (Cell Signaling Technologies, Cat. #2118) as loading control, followed by appropriate HRP-conjugated secondary antibodies.

Techniques: RNA Sequencing, Comparison, ChIP-sequencing, HiChIP

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Bar plot of the proportion of genes in selected metabolic pathways that harbor IRF1 ChIP-seq peaks; red intensity denotes average number of peaks per gene in each pathway. (B) Genome browser tracks of the Hk1 locus showing normalized IRF1 and PU.1 ChIP-seq, Hi-ChIP interactions and ATAC-seq; an inset shows the annotated intragenic enhancer and promoter contact. [UR = Unresponsive] (C) Line plots of nascent RNA CPM (mean ± SD) for selected genes in glycolysis, PPP and TCA pathways WT and IRF1 KO BMDMs; two-way ANOVA and post-hoc testing; * < 0.05, ** < 0.01, *** < 0.001. (D) Oxygen consumption rates (OCR; fmol mm⁻² s⁻¹) for untreated and IFNγ–stimulated WT and IRF1 KO BMDMs [n = 4/group]; adjacent heatmap shows Student t-test p-values for each time point measured. ( E ) Ribbon plots of relative glycolysis metabolite intensity (mean ± SD) detected by GC-MS in response to IFNγ in WT and IRF1 KO BMDMs [n = 3/group]. (F) Diagram of glycolysis, pentose phosphate pathway (PPP) and Krebs cycle highlighting genes and pathway components significantly dysregulated in IRF1 KO BMDMs for at least 1 timepoint.

Article Snippet: Membranes were blocked in 5% skim milk in PBS Tween-20 0.1% and incubated with primary antibodies against IRF1 (R&D Systems, Cat. #AF4715, 1:200) or GAPDH (Cell Signaling Technologies, Cat. #2118) as loading control, followed by appropriate HRP-conjugated secondary antibodies.

Techniques: ChIP-sequencing, HiChIP, Gas Chromatography-Mass Spectrometry

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Volcano plots from differential metabolite abundance analysis for GC-MS data (n=3/group), comparing 48 h versus 0 h in WT cells (left) and WT versus IRF1 KO at 48 h (right). (B) Top: bar plots of normalized GC-MS intensity for sedoheptulose 7-P at 3 h post IFNγ, xylulose at 12 h, and erythrose 4-P at 48 h. Bottom: ribbon plots of normalized MS signal over time with mean ± SD. (C) Top: normalized LC-MS GSH intensity at 24 h post-IFNγ stimulation. Bottom: ribbon plots of GSH/GSSG ratios over time (mean ± SD) calculated from normalized LC-MS intensities [n=3/group]. (D) Genome browser tracks at the Acod1 locus showing normalized IRF1 and PU.1 ChIP-seq, Hi-ChIP interactions and ATAC-seq [UR = Unresponsive]. Adjacent panels show Acod1 nascent RNA expression and itaconic acid levels. (E) Ribbon plots of normalized GC-MS signal for TCA metabolites in response to IFNγ in WT and IRF1 KO BMDMs. (F) Diagram of glycolysis, PPP and TCA cycle metabolic pathways with dysregulated intermediates denoted in red.

Article Snippet: Membranes were blocked in 5% skim milk in PBS Tween-20 0.1% and incubated with primary antibodies against IRF1 (R&D Systems, Cat. #AF4715, 1:200) or GAPDH (Cell Signaling Technologies, Cat. #2118) as loading control, followed by appropriate HRP-conjugated secondary antibodies.

Techniques: Gas Chromatography-Mass Spectrometry, Liquid Chromatography with Mass Spectroscopy, ChIP-sequencing, HiChIP, RNA Expression

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Schematic of experimental timeline for the long-term wash-and-rest assay. Cells are plated for seven days, pulsed with 24 h IFNγ (400 U/mL) at specified times (24 h, 48 h, 6 d) with defined washout intervals and a final 1 h re-stimulation. On day 7, cells are harvested for ChIP-seq (IRF1, H3K4me1, H3K27ac and H3K9me2). (B) Heatmaps of normalized ChIP-seq signal for IRF1, H3K4me1, and H3K27ac at Clusters 1–3. (C) Aggregate coverage plots of H3K4me1 ±1 kb from ATAC peak centers for UT, 24 h IFNγ, 6 d washout and 6 d + 1 h restimulation; insets show putative nucleosomal configurations. (D) Bar plots of fold-change in H3K27ac (mean ± SEM) comparing naïve and IFNγ-trained cells after 1 h restimulation; statistical comparison using Wilcoxon test. (E) Volcano plot of H3K4me1 differential enrichment for Cluster 1–3 (control versus IFNγ washout); points = enhancers, color key: red = increased, blue = decreased, yellow = pioneered genes; labeled enhancers meet log₂FC > 1 and CPM > 5. (F) Hif1a locus showing normalized IRF1, H3K27ac, and H3K4me1 ChIP-seq, and ATAC-seq in WT and IRF1 KO BMDMs [UR = Unresponsive]. Normalized SLAM-seq nascent RNA expression for Hif1a is shown; * p < 0.05.

Article Snippet: Membranes were blocked in 5% skim milk in PBS Tween-20 0.1% and incubated with primary antibodies against IRF1 (R&D Systems, Cat. #AF4715, 1:200) or GAPDH (Cell Signaling Technologies, Cat. #2118) as loading control, followed by appropriate HRP-conjugated secondary antibodies.

Techniques: ChIP-sequencing, Comparison, Control, Labeling, RNA Expression

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Experimental design for multiomic assessment of WT and IRF1 KO bone marrow–derived macrophages (BMDMs) response to IFNγ stimulation. Time-resolved profiling by ATAC-seq, ChIP-seq, Hi-ChIP, SLAM-seq and metabolomics via GC/LC-MS is performed. (B) Venn diagram summarizing ATAC-seq–identified accessible chromatin regions, filtered for high-confidence peaks and IFNγ-responsiveness (n=38,564); this set is used for downstream clustering and differential analyses. (C) Heatmap of normalized ATAC-seq signal (rows = individual accessible site; columns = time points), grouped into eight clusters by k-means clustering. Clusters C1-C3 show IRF1-depedent increase in accessibility in response to IFNγ; highlighted in red. PU.1 ChIP-seq binding signal is also shown, with Cluster C1 lacking detectable PU.1 occupancy. (D) Ribbon plots of relative ATAC-seq peak height (each peak scaled to its maximum) over matched time points; lines indicate mean accessibility and shaded ribbons show ± SD for WT (black) and IRF1 KO (red). (E) Boxplots of normalized ATAC-seq counts in WT BMDMs at heterochromatin regions, and at clusters C1–C8 and unresponsive ATAC-seq sites; median with interquartile range are shown.

Article Snippet: 20 million BMDMs from WT and Irf1−/− mice were plated in 15 cm tissue culture-treated dishes as described above and stimulated with IFNγ (400 U/mL, R&D Systems) for 0 – 48 h. Crosslinking and chromatin immunoprecipitation were performed following the Dovetail Genomics Hi-ChIP protocol using the IRF1 antibody (R&D Systems, AF4715).

Techniques: Derivative Assay, ChIP-sequencing, HiChIP, Liquid Chromatography with Mass Spectroscopy, Binding Assay

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Network diagrams of transcription factor motif frequency (node size) and co-occurrence (edge thickness) within ±100 bp of ATAC-seq peak centre for clusters C1-C3 and unresponsive sites. “IRF1–IRF1” denotes sites with ≥2 IRF motifs, and node/edge scales reflect motif frequency and co-occurrence. (B) Volcano plots of TOBIAS differential binding scores for 879 mammalian TFs in WT BMDMs comparing 0.5, 3 and 48 h post-IFNγ versus non-treated (0 h); significant TFs are highlighted [Bonferroni-corrected FDR < 0.05; log2 FC > |0.5|]. (C) Heatmap of centered TOBIAS TF footprinting intensity in WT BMDMs, grouped into four clusters by k-means clustering. (D) Density plots (top) and motif-centered TOBIAS footprint heatmaps (bottom) in WT and IRF1 KO BMDMs showing aggregated IRF1-centered footprinting signal at Cluster 1 sites (rows = individual sites; columns = base position around motif). (E) Representative Western blots in WT BMDMs showing IRF1 protein and GAPDH control across time points (0–48 h) post-IFNγ stimulation.

Article Snippet: 20 million BMDMs from WT and Irf1−/− mice were plated in 15 cm tissue culture-treated dishes as described above and stimulated with IFNγ (400 U/mL, R&D Systems) for 0 – 48 h. Crosslinking and chromatin immunoprecipitation were performed following the Dovetail Genomics Hi-ChIP protocol using the IRF1 antibody (R&D Systems, AF4715).

Techniques: Binding Assay, Footprinting, Western Blot, Control

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Heatmaps of IRF1 occupancy (ChIP-seq), ATAC-seq accessibility and H3K4me1, H3K4me3 and H3K27ac signals across in response to IFNγ for sites in Clusters 1–3 in WT and IRF1 KO BMDMs. (B) Hi-ChIP arc plots showing loop contacts (arc width represents number of contacts) between IRF1-bound enhancers and promoters at C8 and unresponsive sites in WT and IRF1 KO BMDMs. [FitHiChIP thresholds FDR < 0.1; loop FC > 6, CPM > 6] (C) Graph of the temporal changes for ChIP-seq and ATAC-seq signals at Cluster 1. Half-time (t½) to reach 50% of each signal’s maximum was calculated by normalizing each trajectory to its maximum and extracting the pseudo-time at half-max. (D) Heatmap of ChIP-seq for IRF1, BRG1, ARID1A, BRD9 and PHF10 across Clusters 1–3 at 0, 1, and 4 h post TLR4 activation. ( E ) BRG1 ChIP–qPCR enrichment (fold over input) at four enhancers ( Wdr7 (C1), Shtn1 (C2), Clic5 (C2) , Nos2 (C3)) in WT and IRF1 KO BMDMs, untreated and 4 h post-IFNγ. (F) Boxplots of normalized ATAC-seq counts in Clusters 1–3 in WT BMDMs TLR4 activated with LipidA, with or without and BRG1 inhibition (BRM014).

Article Snippet: 20 million BMDMs from WT and Irf1−/− mice were plated in 15 cm tissue culture-treated dishes as described above and stimulated with IFNγ (400 U/mL, R&D Systems) for 0 – 48 h. Crosslinking and chromatin immunoprecipitation were performed following the Dovetail Genomics Hi-ChIP protocol using the IRF1 antibody (R&D Systems, AF4715).

Techniques: ChIP-sequencing, HiChIP, Activation Assay, ChIP-qPCR, Inhibition

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Heatmaps of IRF1 ChIP–seq and normalized ATAC–seq at sites grouped by IRF1 signal strength (very strong to weak) in response to IFNγ. (B) Line plots of average IRF1 ChIP–seq signal in WT BMDMs for each binding-strength category. (C) Heatmap of relative enrichment of IRF1 binding classes across ATAC clusters (enrichment is relative to the maximum site overlap). (D) Stacked bar plots showing proportions of sites with 0, 1, 2 or ≥3 IRF1 motifs per ATAC cluster. (E) Aggregate plots of IRF1 motif frequency across ±100 bp around IRF1 peaks for each ATAC cluster. (F) Heatmap of IRF1 ChIP–seq signal at 3 h post–IFNγ for sites stratified by IRF1 motif count, as determined in D). (G) Scatter plot of fraction of sites forming IRF1 Hi-ChIP loops versus motif count, with a fitted trend line shown. [FitHiChIP thresholds FDR < 0.1; loop FC > 6, CPM > 6] (H) Genome browser tracks at the Jdp2 locus showing IRF1, PU.1 and H3K27ac ChIP–seq, Hi-ChIP interactions and ATAC–seq in WT and IRF1 KO BMDMs. The cluster to with each ATAC-seq peak belong is indicated [C1 = cluster 1; UR = Unresponsive]. Insets display the array of IRF1 motifs at the C1 site and a SLAM-seq Jdp2 expression plot across the IFNγ time course.

Article Snippet: 20 million BMDMs from WT and Irf1−/− mice were plated in 15 cm tissue culture-treated dishes as described above and stimulated with IFNγ (400 U/mL, R&D Systems) for 0 – 48 h. Crosslinking and chromatin immunoprecipitation were performed following the Dovetail Genomics Hi-ChIP protocol using the IRF1 antibody (R&D Systems, AF4715).

Techniques: ChIP-sequencing, Binding Assay, HiChIP, Expressing

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Line plots of nascent RNA-seq log2 fold-change (FC) for genes within ±10 kb of ATAC cluster regions in WT and IRF1 KO BMDMs in response to IFNγ stimulation. (B) Heatmap of GO biological process enrichment for genes within ±50 kb of ATAC peaks. Categories with clusterProfiler FDR < 0.05 for at least one cluster are shown. (C) Line plots of nascent RNA counts per million (CPM; mean ± SD) for selected genes across IFNγ time points; WT vs IRF1 KO comparison by two-way ANOVA and pairwise post-hoc testing at each time point. (D) Genome browser tracks at the Kmt2c locus showing IRF1 and PU.1 ChIP-seq, Hi-ChIP arcs and ATAC-seq signal for WT and IRF1 KO BMDMs. [UR = Unresponsive] (E) Line plots of RNA-seq CPM (mean ± SD) for selected genes at 0, 1 and 4 h post-LipidA treatment in WT BMDMs, with BRM014 treatment at the 4 h time point. [Student T-test; n = 3] (F) Bar plot of log2 odds ratio of downregulated genes (FC < 0.5 and FDR < 0.05) after BRM014 treatment (4 h Lipid A) across clusters. * < 0.05, ** < 0.01, *** < 0.001

Article Snippet: 20 million BMDMs from WT and Irf1−/− mice were plated in 15 cm tissue culture-treated dishes as described above and stimulated with IFNγ (400 U/mL, R&D Systems) for 0 – 48 h. Crosslinking and chromatin immunoprecipitation were performed following the Dovetail Genomics Hi-ChIP protocol using the IRF1 antibody (R&D Systems, AF4715).

Techniques: RNA Sequencing, Comparison, ChIP-sequencing, HiChIP

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Bar plot of the proportion of genes in selected metabolic pathways that harbor IRF1 ChIP-seq peaks; red intensity denotes average number of peaks per gene in each pathway. (B) Genome browser tracks of the Hk1 locus showing normalized IRF1 and PU.1 ChIP-seq, Hi-ChIP interactions and ATAC-seq; an inset shows the annotated intragenic enhancer and promoter contact. [UR = Unresponsive] (C) Line plots of nascent RNA CPM (mean ± SD) for selected genes in glycolysis, PPP and TCA pathways WT and IRF1 KO BMDMs; two-way ANOVA and post-hoc testing; * < 0.05, ** < 0.01, *** < 0.001. (D) Oxygen consumption rates (OCR; fmol mm⁻² s⁻¹) for untreated and IFNγ–stimulated WT and IRF1 KO BMDMs [n = 4/group]; adjacent heatmap shows Student t-test p-values for each time point measured. ( E ) Ribbon plots of relative glycolysis metabolite intensity (mean ± SD) detected by GC-MS in response to IFNγ in WT and IRF1 KO BMDMs [n = 3/group]. (F) Diagram of glycolysis, pentose phosphate pathway (PPP) and Krebs cycle highlighting genes and pathway components significantly dysregulated in IRF1 KO BMDMs for at least 1 timepoint.

Article Snippet: 20 million BMDMs from WT and Irf1−/− mice were plated in 15 cm tissue culture-treated dishes as described above and stimulated with IFNγ (400 U/mL, R&D Systems) for 0 – 48 h. Crosslinking and chromatin immunoprecipitation were performed following the Dovetail Genomics Hi-ChIP protocol using the IRF1 antibody (R&D Systems, AF4715).

Techniques: ChIP-sequencing, HiChIP, Gas Chromatography-Mass Spectrometry

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Volcano plots from differential metabolite abundance analysis for GC-MS data (n=3/group), comparing 48 h versus 0 h in WT cells (left) and WT versus IRF1 KO at 48 h (right). (B) Top: bar plots of normalized GC-MS intensity for sedoheptulose 7-P at 3 h post IFNγ, xylulose at 12 h, and erythrose 4-P at 48 h. Bottom: ribbon plots of normalized MS signal over time with mean ± SD. (C) Top: normalized LC-MS GSH intensity at 24 h post-IFNγ stimulation. Bottom: ribbon plots of GSH/GSSG ratios over time (mean ± SD) calculated from normalized LC-MS intensities [n=3/group]. (D) Genome browser tracks at the Acod1 locus showing normalized IRF1 and PU.1 ChIP-seq, Hi-ChIP interactions and ATAC-seq [UR = Unresponsive]. Adjacent panels show Acod1 nascent RNA expression and itaconic acid levels. (E) Ribbon plots of normalized GC-MS signal for TCA metabolites in response to IFNγ in WT and IRF1 KO BMDMs. (F) Diagram of glycolysis, PPP and TCA cycle metabolic pathways with dysregulated intermediates denoted in red.

Article Snippet: 20 million BMDMs from WT and Irf1−/− mice were plated in 15 cm tissue culture-treated dishes as described above and stimulated with IFNγ (400 U/mL, R&D Systems) for 0 – 48 h. Crosslinking and chromatin immunoprecipitation were performed following the Dovetail Genomics Hi-ChIP protocol using the IRF1 antibody (R&D Systems, AF4715).

Techniques: Gas Chromatography-Mass Spectrometry, Liquid Chromatography with Mass Spectroscopy, ChIP-sequencing, HiChIP, RNA Expression

Journal: bioRxiv

Article Title: Pioneer factor IRF1 unlocks latent enhancers to rewire chromatin and immunometabolism in inflammatory macrophages

doi: 10.64898/2026.02.27.708404

Figure Lengend Snippet: (A) Schematic of experimental timeline for the long-term wash-and-rest assay. Cells are plated for seven days, pulsed with 24 h IFNγ (400 U/mL) at specified times (24 h, 48 h, 6 d) with defined washout intervals and a final 1 h re-stimulation. On day 7, cells are harvested for ChIP-seq (IRF1, H3K4me1, H3K27ac and H3K9me2). (B) Heatmaps of normalized ChIP-seq signal for IRF1, H3K4me1, and H3K27ac at Clusters 1–3. (C) Aggregate coverage plots of H3K4me1 ±1 kb from ATAC peak centers for UT, 24 h IFNγ, 6 d washout and 6 d + 1 h restimulation; insets show putative nucleosomal configurations. (D) Bar plots of fold-change in H3K27ac (mean ± SEM) comparing naïve and IFNγ-trained cells after 1 h restimulation; statistical comparison using Wilcoxon test. (E) Volcano plot of H3K4me1 differential enrichment for Cluster 1–3 (control versus IFNγ washout); points = enhancers, color key: red = increased, blue = decreased, yellow = pioneered genes; labeled enhancers meet log₂FC > 1 and CPM > 5. (F) Hif1a locus showing normalized IRF1, H3K27ac, and H3K4me1 ChIP-seq, and ATAC-seq in WT and IRF1 KO BMDMs [UR = Unresponsive]. Normalized SLAM-seq nascent RNA expression for Hif1a is shown; * p < 0.05.

Article Snippet: 20 million BMDMs from WT and Irf1−/− mice were plated in 15 cm tissue culture-treated dishes as described above and stimulated with IFNγ (400 U/mL, R&D Systems) for 0 – 48 h. Crosslinking and chromatin immunoprecipitation were performed following the Dovetail Genomics Hi-ChIP protocol using the IRF1 antibody (R&D Systems, AF4715).

Techniques: ChIP-sequencing, Comparison, Control, Labeling, RNA Expression