Critical role of IRF1 and BATF in forming chromatin landscape during type 1 regulatory cell differentiation

Abstract

Type 1 regulatory T cells (Tr1 cells) are induced by interleukin-27 (IL-27) and have critical roles in the control of autoimmunity and resolution of inflammation. We found that the transcription factors IRF1 and BATF were induced early on after treatment with IL-27 and were required for the differentiation and function of Tr1 cells in vitro and in vivo. Epigenetic and transcriptional analyses revealed that both transcription factors influenced chromatin accessibility and expression of the genes required for Tr1 cell function. IRF1 and BATF deficiencies uniquely altered the chromatin landscape, suggesting that these factors serve a pioneering function during Tr1 cell differentiation.

Figure 1. IRF1 and BATF are required for Tr1 differentiation in vitro

Naïve CD4⁺ T cells isolated from wildtype mice were primed with IL-27 in the presence of anti-CD3 and anti-CD28 antibodies (a) Volcano plot analysis for samples collected at 2 hours post cell stimulation. Depicted is time point of log₂ fold-change (x-axis) versus -log₁₀ p-value (y-axis, representing the probability that the gene is differentially expressed). Black dotted line marks p-value 0.05 and red dots marks fold change higher or lower than two. Irf1 and Batf marked in blue. (b) Irf1 and Batf mRNA expression measured by qPCR over 72 hours following cell stimulation. Analysis of Tr1 differentiation in Irf1^−/− cells 72 hours after cell priming with IL-27 measured by (c) flow cytometry (d) qPCR (left, n=3 samples) and ELISA (right, n=5 samples). Analysis of Tr1 differentiation in Batf^−/− cells 72 hours after cell priming with IL-27 measured by (e) flow cytometry (f) qPCR (left, n=3 samples) and ELISA (right, n=5 samples). Dots represent biological replicates. Data are representative of three independent experiments (b), representative of four independent experiments (c, e), or are pooled from three independent experiments (d, f). *P < 0.001, **P < 0.0001 (unpaired t-test, error bars represent mean ±s.e.m.).

Figure 2. The effects of IRF1 and BATF deficiency on Tr1 function

(a) Irf1 mRNA expression in Stat1^−/− (left) and Stat3^−/− (right) cells primed in the presence of IL-27. (b) Batf mRNA expression in Stat1^−/− (left) and Stat3^−/− (right) cells primed in the presence of IL-27. (c) Effects of IRF1 and BATF retroviral overexpression on Il10, Il21, Maf and AhR expression in cells treated in T_H0 or Tr1 conditions; mRNA levels were quantified using qPCR. Data are pooled from 3 independent experiments (a (Stat1^−/−), b, c; n=3 samples, dots represent biological replicates) or representative of 4 independent experiments (a (Stat3^−/−); n=3 samples, dots represent technical replicates). NS, not significant (P > 0.05); *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001 (unpaired t-test, error bars represent mean ±s.e.m.).

Figure 3. IRF1 affects induction and resolution of autoimmune disease

(a) EAE scores in WT and Irf1^−/− mice were immunized with MOG_35-55, n=24 mice (WT) or n=27 mice (Irf1^−/−) (left); linear regression curve including the 95% confidence band of the regression line (middle); maximum scores (right). (b) Cytokine expression in CD4⁺ T cells isolated from the CNS of WT and Irf1^−/− mice at the peak of disease. (c) Frequency of IL-17A⁺ and IFN-γ⁺ cells within the CD4⁺ T cells isolated from the CNS (n=5 mice). (d) EAE course in Rag2^−/− mice with 2D2 or Irf1^−/−2D2 CD4⁺ T cells transferred, followed by MOG_35-55 immunization; n=8 mice/group. (e) Spontaneous EAE scores in 2D2 (n=5 mice) and Irf1^−/− 2D2 mice (n=7 mice). (f) Spleens and lymph nodes from WT and Irf1^−/− immunized mice were cultured with different concentration of MOG_35-55. ³H thymidine was added to assess cell proliferation; n=5 samples. (g) Cytokine expression in CD4⁺ T cells isolated from WT and Irf1^−/− mice immunized with MOG_35-55 and cultured with MOG_35-55±IL23. (h) Quantification of (g); n=5 mice. Data are pooled from four independent experiments (a), representative of two independent experiments (b, f, g), or pooled from two independent experiments (c, d, h). Dots represent individual mice. *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001. Data represent mean ±s.e.m (linear regression (a (EAE plot), d; unpaired t-test (a (max score), c, f, h).

Figure 4. IRF1 and BATF control generation of IL-10⁺ Tr1 cells in vivo

(a) Il10 expression in CD4⁺ T cells isolated from the CNS and dLNs (draining lymph nodes) of WT and Irf1^−/− mice immunized with MOG_35-55. Floating bars indicate min to max with a line at mean. (b) Gene expression fold change between CD4⁺ T cells isolated from wildtype mice immunized with MOG_35-55 and re-stimulated in the presence of MOG_35-55 ± IL-27. Genes with at least 2-fold change (MOG+IL-27/MOG) are shown. (c) Differential gene expression in WT, Irf1^−/− and Batf^−/− CD4⁺ T cells isolated from mice immunized with MOG_35-55 and restimulated with MOG_35-55 and IL-27. Gene-set created in (b) was used to generate the heatmap. (d) Tr1-polarized CD4⁺ T cells isolated from WT, Irf1^−/− or Batf^−/− mice were transferred into WT mice 10 days after EAE induction; n=7 mice (No Tr1, Batf^−/−), n=8 (Irf1^−/−), n=10 (WT Tr1). (e) IL-10 expression in mesenteric lymph nodes (MLN) from WT, Irf1^−/− and Batf^−/− mice injected with anti-CD3 antibody or an isotype control (IC). (f) Splenocytes from mice in (e) were cultured for 72 hours, ELISA was used to measure IL-10 in the cultures. Dots represent individual mice. Data are pooled from three independent experiments (a), pooled from two independent experiments (b, c, d), representative of two independent experiments with n=3–4 mice pooled/group per experiment (e) or representative of two independent experiments (f). *P < 0.05, **P < 0.01 and ***P < 0.0001. Data represent mean ±s.e.m (unpaired t-test (a, f); linear regression analysis (d)).

Figure 5. IRF1 and BATF bind to the Il10 locus

(a) Schematic drawing of the Il10 locus; known CNS and HSS sites are marked. ChIP analysis of (b) IRF1 (n=2 samples) and (c) BATF (n=3 samples) interactions within the Il10 promoter in WT Tr1 cells polarized for 72 hrs. (d) Sequential ChIP analysis of BATF and IRF1 interactions in the CNS-9 and HSS⁺2.98 regions of the Il10 promoter in WT and Batf^−/− cells differentiated in the presence of IL-27. (e) ChIP analysis of IRF1 binding in the CNS-9, HSS-0.12, HSS⁺2.98 sites of the Il10 promoter in WT and Batf^−/− cells differentiated in Tr1 conditions; n=2 samples. (f) ChIP analysis of BATF binding in the CNS-9, and HSS⁺2.98 sites of the Il10 promoter in WT and Irf1^−/− cells differentiated in Tr1 conditions; n=3 samples. (g) ChIP analysis of epigenetic marks recruitment to the Il10 promoter in WT, Irf1^−/− and Batf^−/− cells differentiated in Tr1 conditions. Data are representative of three independent experiments with similar results (b, c, e, f) or two independent experiments (d, g). *P < 0.05 (unpaired t-test, error bars represent mean ±s.e.m.).

Figure 6. IRF1 and BATF control the Il10 locus

Luciferase activity in 293T cells transfected with different IL-10 luciferase reporters in the presence of constructs encoding IRF1, IRF1 DNA-binding mutant W11R, c-Maf and BATF. RLU (Relative Light Units) are shown. (a) Proximal promoter indicates the proximal 1.5 kb region of the promoter. (b) CNS-9 and HSS⁺2.98 reporters contain the Il10 promoter fragments cloned upstream of the Il10 minimal promoter. (c) Il10 expression (mRNA) in WT and Irf1^−/− CD4⁺ T cells retrovirally transduced with c-Maf and cultured in T_H0 or Tr1 conditions (d) Il10 expression (mRNA) in WT and Batf^−/− CD4⁺ T cells retrovirally transduced with c-Maf and cultured in T_H0 or Tr1 conditions (e) ChIP analysis of c-Maf interactions with c-Maf-binding sites (MARE-1--4) in the Il10 promoter in WT and Irf1^−/− Tr1 polarized cells. (f) ChIP analysis of AhR interactions with AhR-binding site XRE-1 in the Il10 promoter in WT, Irf1^−/− and Batf^−/− Tr1 polarized cells. Data are representative of three independent experiments (a, b; n=3 samples), representative of two independent experiments (c, d; n=3 samples) or pooled from two independent experiments (e, f). Dots represent technical replicates (a, b, c, d) or biological replicates (e, f). NS, not significant (P > 0.05); *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001.

Fig 7. BATF- and IRF1-deficiences uniquely contribute to Tr1 chromatin and transcriptional landscapes

(a) Heatmap of normalized ATAC-seq peak intensities (log₂-fold-changes relative to the mean for each peak). Limited to (6017) peaks that are condition-dependent with log₂|FC|>3 and FDR=1% for at least one pairwise comparison of interest. (b) PCA of the 72h conditions, using all ATAC-seq peaks in the dataset (180,478 DESeq2-normalized peak intensities). (c) Number of differentially accessible peaks detected using DESeq2, comparing Tr1 KO (knockouts) to control cells at 72h, log₂|FC|>1 and FDR=10% (Subsampled, each comparison had n=2 (KO), n=6 (control)). Estimates of log₂-fold-changes in gene expression from RNA-seq data, comparing either Batf^−/− or Irf1^−/− and control Tr1 cells (log₂|FC|>1, FDR=10%) for DNA-binding proteins (d) and cytokines and cytokine receptors (e). RNA-seq and ATAC-seq datasets were integrated to generate putative transcriptional regulatory networks for Irf1^−/− (f–g) and Batf^−/− (h–i) Tr1 cells. Nodes represent differentially expressed transcription factors and target genes, colored according to relative gene expression (log₂(KO/Control)) at 72h; red and blue indicate high and low relative expression, respectively. Edges are drawn between TFs and putative gene targets if that differentially expressed TF’s motif was enriched in ATAC-seq peaks cis to genes that were increased or decreased in response to KO (P_raw<10⁻³, hypergeometric CDF). Edge colors (blue - inhibitory, red - activating) are based on correlation between TF and putative target gene expression. “ON” indicates genes de-repressed in the knockout cells, while “OFF” indicates genes repressed in the knockout cells.