Small-molecule RORγt antagonists inhibit T helper 17 cell transcriptional network by divergent mechanisms

Abstract

We identified three retinoid-related orphan receptor gamma t (RORγt)-specific inhibitors that suppress T helper 17 (Th17) cell responses, including Th17-cell-mediated autoimmune disease. We systemically characterized RORγt binding in the presence and absence of drugs with corresponding whole-genome transcriptome sequencing. RORγt acts as a direct activator of Th17 cell signature genes and a direct repressor of signature genes from other T cell lineages; its strongest transcriptional effects are on cis-regulatory sites containing the RORα binding motif. RORγt is central in a densely interconnected regulatory network that shapes the balance of T cell differentiation. Here, the three inhibitors modulated the RORγt-dependent transcriptional network to varying extents and through distinct mechanisms. Whereas one inhibitor displaced RORγt from its target loci, the other two inhibitors affected transcription predominantly without removing DNA binding. Our work illustrates the power of a system-scale analysis of transcriptional regulation to characterize potential therapeutic compounds that inhibit pathogenic Th17 cells and suppress autoimmunity.

Figure 1. RORγt inhibitors suppress Th17 cell differentiation and maintenance and ameliorate EAE

(A) Chemical structures of TMP778, TMP920, and Digoxin. (B) Naïve CD4⁺ T cells were activated with anti-CD3 and CD28 under Th17 cell-polarizing conditions in the presence of optimal doses (not toxic but with maximal IL-17 inhibition) of TMP778 (2.5 μM), TMP920 (10 μM), Digoxin (10 μM), or DMSO. After 4 days, IL-17 and IFNγ production were measured by intracellular cytokine staining. Data are representative of 5-8 experiments. (C) Draining LN cells from mice immunized with MOG35-55 plus CFA for the development of EAE were re-stimulated with MOG35-55 in the presence of IL-23 plus TMP778, TMP920, Digoxin, or DMSO. After 4 days, production of IL-17 and IFNγ in CD4⁺ T cells was determined by intracellular cytokine staining. Left panel shows representative FACS plots the frequencies of IFNγ and IL17 producing cells in gated CD4⁺ T cells from samples treated with DMSO and TMP778; right panel shows the statistical data (n=5). Error bars represent the mean ± SD. * p<0.01. (D) C57BL/6 mice were immunized with MOG35–55 plus CFA, and RORγt inhibitor (TMP778, 200 μg per injection, n=19; TMP920, 500 μg per injection, n=7; Digoxin, 50 μg per injection, n=5, >100 μg caused mouse death; DMSO, n=19) were subcutaneously injected twice daily starting from day 0. Mice were evaluated daily for signs of EAE. Error bars represent the mean ± SD. * p<0.05 when 11 days after groups of mice treated with different RORγt inhibitors were compared with the group of mice with DMSO (vehicle control) treatment. (E) CNS-infiltrating mononuclear cells were isolated from brains and spinal cords of the mice on day 21 after EAE induction. IL-17 and IFNγ production of CNS-infiltrating CD4⁺ T cells were determined by intracellular staining. Data are representative of 4-5 mice in each group.

Figure 2. Effects of RORγt inhibitors on the Th17 cell transcriptome

(A) Heat map displaying the fold changes of genes (rows) in the various perturbations (columns). Displayed are only genes that were differentially expressed (fold>1.5) in at least one condition. On the right, differential expression of selected genes encoding transcriptional regulators (above) and cytokines or cytokine receptors (below). (B) Enrichment of the differentially expressed genes in gene signatures of different T-cell subsets. The height of the bars indicates fold enrichment; the color of the bars indicates the percentage of genes in the overlap (i.e. genes that are both differentially expressed and belong to the respective signature) that are over expressed (from blue to red). (C) The overlap between the sets of genes affected by each compound and the sets of genes affected by RORγt deficiency is evaluated using an F-score: the harmonic mean of their specificity (% of compound-affected genes that are affected by RORγt deficiency) and sensitivity (% of RORγt deficiency-affected genes that are affected by the compound). Results are presented for different fold change cutoffs for calling differential expression in the compounds.

Figure 3. The binding landscape of RORγt in Th17 cells

(A) RORγt binding at key Th17 cell gene loci. (B) RORγt binding motif (bottom) highly matches the known RORα binding motif (top). (C) Overlap between the set of genes bound by RORγt and the genes affected by RORγt deficiency.

Figure 4. RORγt selectively targets genes associated with Th17 cell function and signatures of other CD4⁺ T cells

(A) Enrichment (blue bars) and significance (brown bars) of RORγt target genes in signatures of different subsets of T cells. (B) Percentage of RORγt binding sites that are also occupied by other TFs in Th17 cells and other CD4⁺ T-cell subsets. (C) RORγt binding sites overlap with STAT3 (Yang et al., 2011), IRF4, and Batf in Th17 cells (Glasmacher et al., 2012), and Foxp3 in iTreg cells at selected key target gene loci.

Figure 5. Distinct effects of inhibitors on RORγt-DNA Interactions

(A) RORγt binding near transcription start sites (TSS). Every line depicts the 6kb region around a TSS (center) in 300bp windows. Shown are the 1544 TSS that contain a binding peak (p<10^-8; Methods) in at least one condition (compounds, DMSO, untreated Th17 cells) and do not contain any signal (Z score>0.1) in the control (RORγt-deficient) cells. Color intensity is proportional to the number of reads mapped to each window (normalized separately for each condition using Z-scores). Effects of RORγt inhibitors on RORγt occupancy at Il17a and Il17f (B) and Gata3 (C) loci were validated by ChIP-PCR. Naïve CD4⁺ T cells were cultured under Th17 cell-polarizing conditions in the presence of indicated doses of RORγt inhibitors. After 96 h, ChIP was performed with anti-RORγt, followed by real-time PCR analysis. Th17 cells for ChIP-seq were cultured in the presence of 2.5 μM TMP778, 10 μM TMP920, 10 μM Digoxin, or DMSO for 96 h. The RORγt binding sites in Il17a and Il17f and Gata3 loci are as indicated in the ChIP-seq binding tracks. ChIP-PCR was used to confirm binding at selected sites (shown below ChIP-Seq tracks) and the RORγt occupancy (% of input) is shown as “Enrichment”. Data are representative of two experiments. TMP778-caused RORγt binding site in Gata3 locus in Th17 cells overlaps with STAT3, IRF4, Batf in Th17 cells and Foxp3 in iTreg cells.

Figure 6. Effects of RORγt inhibitor GSK805 on Th17 cells and Th17 cell-mediated autoimmune diseases

(A) Chemical structure of GSK805. (B) Naïve CD4⁺ T cells were activated under Th17 cell-polarizing conditions in the presence of GSK805 (0.5 μM), TMP778 (2.5 μM), or DMSO. After 4 days, IL-17 and IFNγ production were measured by intracellular cytokine staining. Data are representative of 3 experiments. (C) C57BL/6 mice were immunized with MOG35-55 plus CFA, and RORγt inhibitor GSK805 (10 mg/kg) were orally given daily starting from day 0. Mice (n=8) were evaluated daily for signs of EAE. Error bars represent the mean ± SD. * p<0.01 by repeated ANOVA test. (D) C57BL/6 mice were induced for EAE and treated with GSK805 (30 mg/kg). On day 14, CNS-infiltrating cells were isolated and measured for IL-17 and IFNγ production by intracellular staining. Error bars represent the mean ± SD. * p<0.001. (E) Comparison of gene expression under the various perturbations (indicated in the legend) with WT DMSO. The figure depicts the average fold change of the genes in clusters #1-#5 from Figure 2A. Results were obtained by profiling transcripts 3′ end.

Figure 7. RORγt is a regulatory hub in a densely inter-connected network of CD4⁺ T-cell regulation

Depicted are TFs that share a significant (p<10^-3) number of common targets with RORγt (inner circle), and a subset of these common targets that are also differentially expressed under perturbation of RORγt (outer circle; shown are only genes that are associated with immune response; see Tables S1 and S2 for the complete lists). Edges indicate TF binding in a target gene. Node colors reflect the modulation of mRNA levels in RORγt-deficient cells (A) or under TMP778 treatment (B).