Glycogen synthase kinase-3 promotes T helper type 17 differentiation by promoting interleukin-9 production

Abstract

T helper type 17 (Th17) cells are recognized as important contributors to the deleterious effects of several neurological and psychiatric diseases. Clarifying mechanisms that control the production of Th17 cells may therefore provide new strategies for developing novel interventions in a broad spectrum of disorders. Th17 cell differentiation is promoted by glycogen synthase kinase-3 (GSK3), but the mechanisms for this are only beginning to be understood. Using T-cell-selective depletion of GSK3β and multiple selective pharmacological GSK3 inhibitors, we found that GSK3 inhibition decreased C-C motif chemokine (ccl)20, C-C motif chemokine receptor (ccr)6, interleukin (IL)-9, Runt-related transcription factor (Runx)1, interferon regulatory factor (Irf)4 and c-maf mRNA expression after 2 days of Th17 cell differentiation in vitro. These effects were found to be independent of the master regulator transcription factor retinoic acid receptor-related orphan receptor γT (RORγT), as GSK3 inhibition still reduced Th17 cell differentiation in RORγT-depleted cells. Because IL-9 was approximately ninefold down-regulated in GSK3β^-/- CD4 cells, we tested if reintroduction of IL-9 during Th17 cell differentiation abolished the inhibition by GSK3 deficiency of Th17 cell differentiation. We found that IL-9 over-expression was sufficient to reverse the inhibition of Th17 cell differentiation by GSK3 inhibition or depletion. We found that IL-9 enhances Th17 cell differentiation in part through signal transducer and activator of transcription 3 (STAT3) activation, and IL-9 also enhances STAT3 binding to the IL-17a promoter. Altogether, these findings suggest that IL-9 might be an important mediator of GSK3β-dependent enhancement of Th17 cell differentiation.

Keywords: T helper type 17 cells; glycogen synthase kinase-3; interleukin-9.

Figure 1

IL‐17A⁺ CD4⁺ (T helper type 17; Th17) cells and the expression of several Th17 differentiation‐related genes are lower in GSK3β ^−/− (knockout; KO) CD4 cells compared with wild‐type (WT) CD4 cells. WT (GSK3β ^Flox/Flox) and glycogen synthase kinase‐3β (GSK3β) KO (GSK3β ^Flox/Flox CD4‐CRE) CD4 cells were differentiated in vitro towards Th17 cells for 2–4 days. (a) IL‐17A⁺ CD4⁺ and IRF4⁺ CD4⁺ cells after 4 days of differentiation, analyzed by flow cytometry. (b) Gene expression after 2 days of differentiation towards Th17 cells, identified using Th17 RT² profiler PCR array. (c) Fold changes of gene expression of GSK3β ^−/− CD4 cells after 2 days of differentiation towards Th17 cells, analyzed by RT‐PCR. Fold change was calculated as (2^{−Δ Ct} of a sample)/(average 2^{−Δ Ct} of WT CD4 cells). Gapdh was used as the endogenous control gene to normalize individual gene expression. Each symbol in (c) represents an individual GSK3β ^−/− mouse (n = 3). *P < 0·05, **P < 0·01, unpaired t‐test compared with respective gene level in WT CD4 cells.

Figure 2

Effects of glycogen synthase kinase‐3 (GSK3) inhibitors on gene expression during T helper type 17 (Th17) cell differentiation. Wild‐type (WT) CD4 cells were differentiated toward Th17 cells in the presence of (a) LiCl (10 mm); (b) CHIR99021 (1 μm); or (c) TDZD8 (1 μm) for 2 days. Fold change was calculated as (2^{−Δ Ct} of a sample)/(average 2^{−Δ Ct} of untreated CD4 cells). Gapdh was used as the endogenous control gene to normalize individual gene expression. Each symbol represents an independent cell differentiation from the pool of CD4⁺ cells from two mice (n = 4). *P < 0·05, **P < 0·01, unpaired t‐test as compared with respective gene levels in untreated CD4 controls.

Figure 3

Glycogen synthase kinase‐3 (GSK3) inhibitors still inhibit T helper type 17 (Th17) ‐dependent genes in RORγT^−/− CD4 cells. RORγT^−/− CD4 cells were differentiated toward Th17 cells for 2 days with or without treatment with LiCl (10 mm) or CHIR99021 (1 μm). Gene expression of (a) Ccl20, (b) Ccr6, (c) IL‐9, (d) Irf4, or (e) Runx1. Fold change was calculated as (2^{−Δ Ct} of a sample)/(average 2^{−Δ Ct} of wild‐type CD4 cells). Gapdh was used as the endogenous control gene to normalize individual gene expressions. Each symbol represents an independent cell differentiation from the pool of CD4⁺ cells from two RORγT^−/− mice (n = 3 or n = 4). One‐way analysis of variance F(2, 9) = 154·2 (a), F(2, 9) = 62·31 (b), F(2, 8) = 36·75 (c), F(2, 8) = 14·76 (d), F(2, 9) = 1·940 (e). *P < 0·05, **P < 0·01, Tukey’s post‐hoc test, compared with the respective gene levels in wild‐type CD4 cells, ^# P < 0·05, ^## P < 0·05 compared with the gene expression levels in untreated RORγT^−/− CD4 cells.

Figure 4

Effects of interleukin‐9 (IL‐9) and/or glycogen synthase kinase‐3 (GSK3) inhibitors on T helper type 17 (Th17) cell differentiation. (a) IL‐17A⁺ CD4⁺ cells at day 4 of culture of cells from wild‐type (WT) mice. WT CD4 cells were differentiated towards Th17 cells and treated or not with IL‐9 (40 ng/ml) and/or GSK3 inhibitors [LiCl (10 mm), CHIR99021 (1 μm), and TDZD8 (1 μm)] for 4 days. (b) WT (GSK3β ^Flox/Flox) and GSK3β ^−/− CD4 cells were differentiated toward Th17 cells with or without IL‐9 (40 ng/ml) for 4 days. IL‐17A⁺CD4⁺ cells were analyzed by flow cytometry and were gated on CD4⁺ cells. Fold change was calculated as percentage of cells of a sample/average percentage of cells of untreated WT (GSK3β ^Flox/Flox) CD4 cells. WT CD4 cells cultured without GSK3 inhibitors and IL‐9 were used as controls. Each symbol represents an independent cell differentiation (n = 7 to n = 9 for a, n = 3 to n = 7 for b). Two‐way analysis of variance F(1, 28) = 22·68 (a – GSK3 inhibitors), F(1, 28) = 23·74 (a – IL‐9); F(1, 18) = 63·93 (b – mice type), F(1, 18) = 159·6 (b – IL9), F(1, 18) = 17·53 (b – interaction), *P < 0·05, **P < 0·01, Tukey’s post‐hoc test, compared with controls.

Figure 5

Signal transducer and activator of transcription 3 (STAT3) is required for the interleukin‐9 (IL‐9) effect on T helper type 17 (Th17) cell differentiation. Wild‐type (WT) CD4 cells were transfected with 50 moi of Ad‐GFP (GFP) or 50 moi of Ad‐STAT3C (STAT3C) and differentiated towards Th17 cells with or without IL‐9 (40 ng/ml) and/or glycogen synthase kinase‐3 (GSK3) inhibitors [LiCl (10 mm), CHIR99021 (1 μm), and TDZD8 (1 μm)] for 4 days. IL‐17A⁺ CD4⁺ cells were analyzed by flow cytometry and gated on CD4⁺ cells. Each symbol represents an independent cell differentiation from the pool of CD4⁺ cells from three mice (n = 3–4). One‐way analysis of variance F(7, 18) = 34·80,**P < 0·01, Tukey’s post‐hoc test, compared with GFP CD4 cells.

Figure 6

Proteins binding to the IL‐17a promoter determined by chromatin immunoprecipitation (ChIP). Wild‐type (WT) CD4 cells were differentiated towards Th17 cells and treated or not with interleukin‐9 (IL‐9; 40 ng/ml) and/or glycogen synthase kinase‐3 (GSK3) inhibitors [LiCl (10 mm), CHIR99021 (1 μm), and TDZD8 (1 μm)] for 4 days. ChIP was performed by immunoprecipitating with either signal transducer and activator of transcription 3 (STAT3) (a), STAT5 (b), or GSK3β (c), and DNA from the IL‐17a promoter region was amplified. All results were normalized by respective input values. Each symbol represents an independent cell differentiation (n = 3 to n = 6). Two‐way analysis of variance for a–c, (a) F(1, 13) = 7·561 (GSK3 inhibitors), F(1, 13) = 7·236 (IL‐9), F(1, 13) = 5·910 (interaction); (b) F(1, 13) = 38·16 (GSK3 inhibitors), F(1, 13) = 4·487 (IL9); (c) F(1, 21) = 13·44 (GSK3 inhibitors), F(1, 21) = 32·09 (IL‐9), F(1, 21) = 9·510 (interaction). *P < 0·05, **P < 0·01, Tukey’s post‐hoc test for (a)–(c).