TYK2 inhibition enhances Treg differentiation and function while preventing Th1 and Th17 differentiation

Abstract

Janus kinase (JAK) inhibitors are widely used to inhibit inflammatory cytokine signaling in autoimmune and inflammatory diseases, but their effect on regulatory T cells (Tregs) is poorly characterized. We investigated the effect of a JAK inhibitor, upadacitinib, on human Treg differentiation and phenotype in comparison to BMS-986202, a selective Tyrosine kinase 2 (TYK2) inhibitor. Both upadacitinib and BMS-986202 blocked naive CD4⁺ T cell differentiation into Th1/17 cells, but only BMS-986202 and a related TYK2 inhibitor, deucravacitinib, spared interleukin-2 (IL-2) signaling and Treg induction. BMS-986202 also increased Treg suppressive function and stability under Th1/17-polarizing conditions, whereas upadacitinib significantly impaired the phenotype and viability of ex vivo Tregs. In lamina propria mononuclear cells from patients with inflammatory bowel disease cultured under Th17-polarizing conditions, BMS-986202 redirected CD4⁺ T cells toward a Treg phenotype. The Treg-sparing and enhancing properties of TYK2 inhibition suggest that TYK2 inhibitors are a promising pharmacological approach for tolerance induction.

Keywords: JAK inhibition; TYK2 inhibition; Th1 differentiation; Th17 differentiation; Treg differentiation; autoimmunity; immune tolerance; inflammatory bowel disease; regulatory T cell.

Graphical abstract

Figure 1

BMS-986202 inhibits TYK2-mediated cytokine signaling in Tregs Tregs were treated with varying concentrations of BMS-986202 (B202) or upadacitinib (UPA) for 1 h and then stimulated with cytokine for 15 min. STAT phosphorylation (pSTAT) was quantified by flow cytometry. (A) Schematic of IFN-α signaling. (B and C) Representative histograms and quantification of pSTAT1 geometric mean fluorescence intensity (gMFI) relative to unstimulated Tregs after treatment with B202 (B; n = 3) or UPA (C; n = 3) and stimulation with IFN-α. (D) Schematic of IL-12/23 signaling. (E and F) Representative contour plots and quantification of % pSTAT3/4+ Tregs after treatment with B202 or UPA and stimulation with IL-12 (E; n = 3) or IL-23 (F; n = 3). (G) Schematic of IL-2/15 signaling. (H and I) Representative histograms and quantification of pSTAT5 gMFI relative to unstimulated Tregs after treatment with B202 or UPA and stimulation with IL-2 (H; n = 3) or IL-15 (I; n = 3). Statistical significance between DMSO (0 μM) and B202/UPA-treated cells was determined by repeated-measures one-way ANOVA with Dunnett’s multiple comparisons test (B and C) or repeated-measures two-way ANOVA with Sidak’s multiple comparisons test (E, F, H, and I). Points indicate mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.

Figure 2

TYK2 inhibition reduces Th1 and Th17 differentiation of naive CD4⁺ T cells but preserves Treg induction (A–D) Naive CD4⁺ T cells were stimulated with anti-CD3/CD28 for 7 days in the presence of a Th17-inducing cytokine cocktail ± BMS-986202 (B202) or upadacitinib (UPA) to induce Th17 differentiation. (A) Schematic of Th17 differentiation. (B) Representative histograms and quantification of CCR4, CCR6, and RORγt expression relative to undifferentiated CD4⁺ T cells at day 7 (n = 3). (C) Representative plot and quantification of intracellular IL-17A/F expression after a 4 h stimulation with PMA/ionomycin (n = 6). (D) IL-17A and IL-17F concentration in supernatant after a 24 h re-stimulation with anti-CD3/CD28 (n = 6). (E–H) Naive CD4⁺ T cells were stimulated with anti-CD3/CD28 for 7 days in the presence of IL-12 ± B202/UPA to induce Th1 differentiation. (E) Schematic of Th1 differentiation. (F) Representative histograms and quantification of TBET and CXCR3 expression relative to undifferentiated CD4⁺ T cells at day 7 (n = 6). (G) Representative plot and quantification of intracellular IFN-γ expression after a 4 h stimulation with PMA/ionomycin (n = 6). (H) IFN-γ concentration in supernatant after a 24 h re-stimulation with anti-CD3/CD28 (n = 6). (I–K) Naive CD4⁺ T cells were stimulated with anti-CD3/CD28 for 7 days in the presence of TGF-β ± B202/UPA to induce Treg differentiation. (I) Schematic of Treg differentiation. (J and K) Representative histograms and quantification of FOXP3 expression at day 7 with treatment with B202 (J; n = 5) or UPA (K; n = 2–5). Statistically significant differences compared to DMSO-treated cells were determined by a repeated-measures one-way ANOVA with Dunnett’s multiple comparisons test (B–D, F–H, and J). Bars indicate mean ± SEM. Points indicate individual replicates. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.

Figure 3

TYK2 inhibition maintains Treg phenotype and enhances suppressive function Tregs (CD4⁺CD25^hiCD127^lo) were isolated from peripheral blood of healthy human donors, stimulated with anti-CD3/CD28, and cultured for 7 days in the presence of BMS-986202 (B202) or upadacitinib (UPA). (A) Schematic of Treg culture. (B–E) Treg viability (B); percentage Ki67⁺ (C); percentage FOXP3⁺/Helios⁺ (D); and CD25, CTLA-4, and CD39 (E) expression were determined after 7 days (n = 4–8). (F and G) Responder PBMCs were stimulated with anti-CD3/CD28 Dynabeads in the presence of varying ratios of Tregs and cultured for 96 h. (F) Percent suppression of CD4⁺ and CD8⁺ responder T cell proliferation was determined relative to PBMCs alone (dotted line). Representative histograms and quantification shown (n = 7). (G) CD80 and CD86 expression on B cells relative to PBMC alone condition (dotted line). Representative histograms and quantification shown (n = 6). Statistically significant differences compared to DMSO-treated cells were determined by one-way ANOVA with Dunnett’s multiple comparisons test (B–E) or an uncorrected Fisher’s least significant difference (LSD) test (F and G). Bars/lines indicate mean ± SEM. Points indicate individual replicates (C–E). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.

Figure 4

TYK2 inhibition enhances Treg stability in Th17-polarizing conditions Tregs (CD4⁺CD25^hiCD127^lo) were isolated from peripheral blood of healthy human donors, stimulated with anti-CD3/CD28, and cultured for 7 days in a Th17-polarizing cocktail in the presence of BMS-986202 (B202). (A) Schematic of polarization protocol. (B) The proportion of FOXP3⁺ cells was assessed after 7 days. Representative plots and quantification are shown (n = 9). (C) Expression of RORC2, CCR4, and CCR6 was determined after 7 days relative to non-polarized, control Tregs (n = 3). (D) Methylation of the Treg-specific demethylated region (TSDR) after 7 days of culture (n = 5). (E) Representative plot and quantification of intracellular IL-17A/F expression after a 4 h stimulation with PMA/ionomycin (n = 5). (F) IL-17A, IL-17F, and TNF-α concentration in supernatant after a 24 h re-stimulation with anti-CD3/CD28 (n = 6). (G) Representative figures and quantification of TIGIT and CD226 expression on Tregs on day 7 (n = 9). Statistically significant differences compared to DMSO-treated cells were determined by a repeated-measures one-way ANOVA with Dunnett’s multiple comparisons test (B–G). Bars indicate mean ± SEM. Points indicate individual replicates. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.

Figure 5

TYK2 inhibition enhances Treg stability in Th1-polarizing conditions Tregs (CD4⁺CD25^hiCD127^lo) were isolated from peripheral blood of healthy human donors, stimulated with anti-CD3/CD28, and cultured for 7 days with IL-12 and in the presence of BMS-986202 (B202). (A) Schematic of polarization protocol. (B) The proportion of FOXP3⁺ cells was assessed after 7 days. Representative plots and quantification are shown (n = 3–8). (C) Representative plots and quantification of intracellular IFN-γ expression after a 4 h stimulation with PMA/ionomycin (n = 3–7). (D) IFN-γ concentration in supernatant after a 24 h re-stimulation with anti-CD3/CD28 tetramer (n = 3). (E) Proportion of Tregs with intracellular expression of IFN-γ⁺IL-17A/F⁺ Tregs after a 4 h stimulation with PMA/ionomycin (n = 3–5). (F) Surface expression of CD226 and TIGIT on Tregs on day 7 (n = 3–6). Statistical differences between DMSO and other conditions were determined by a repeated-measures one-way ANOVA with Dunnett’s multiple comparisons test (B–G). Bars indicate mean ± SEM. Points indicate individual replicates. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.

Figure 6

TYK2 inhibition redirects Th17-polarized LPMCs toward a regulatory phenotype (A–G) Lamina propria mononuclear cells (LPMCs) were isolated from colon biopsies of healthy or IBD patients, stimulated with anti-CD3/CD28 tetramer in the presence of a Th17 cytokine cocktail and BMS-986202 (B202), and cultured for 7 days. (A) Schematic of LPMC culture. (B) Heatmap of cytokine content in LPMC supernatant collected after 3 days of culture. (C) Surface expression of CD226 and TIGIT on LPMC CD4⁺ T cells. (D–F) Representative histograms and quantification of PD-1 (D), CTLA-4 (E), and CD39 (F) expression on LPMC CD4⁺ T cells. (G) Representative plots and quantification of FOXP3⁺ and FOXP3⁺Helios⁺ cells as a proportion of LPMC CD4⁺ T cells. (H) Primary human blood-derived naive CD4⁺ T cells were stimulated with anti-CD3/CD28 and cultured in various cytokine combinations for 7 days with increasing concentrations of BMS-986202. All conditions were supplemented with 100 U/mL IL-2 and anti-IFN-γ and anti-IL-4. FOXP3 expression was determined at day 7 (n = 2–4). Open squares (□) represent healthy patients. Closed circles (•) represent IBD patients (C–G). Statistically significant differences compared to DMSO-treated cells were determined by repeated-measures one-way ANOVA with Dunnett’s multiple comparisons test. Bars indicate mean ± SEM. Points indicate individual replicates. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.