Interleukin-23 is critical for full-blown expression of a non-autoimmune destructive arthritis and regulates interleukin-17A and RORgammat in gammadelta T cells

Abstract

Introduction: Interleukin (IL)-23 is essential for the development of various experimental autoimmune models. However, the role of IL-23 in non-autoimmune experimental arthritis remains unclear. Here, we examined the role of IL-23 in the non-autoimmune antigen-induced arthritis (AIA) model. In addition, the regulatory potential of IL-23 in IL-17A and retinoic acid-related orphan receptor gamma t (RORgammat) expression in CD4+ and TCRgammadelta+ T cells was evaluated systemically as well as at the site of inflammation.

Methods: Antigen-induced arthritis was induced in wild-type, IL-23p19-deficient and IL-17 Receptor A - knockout mice. At different time points, synovial cytokine and chemokine expression was measured. At days 1 and 7 of AIA, splenocytes and joint-infiltrating cells were isolated and analyzed for intracellular IL-17A and interferon (IFN)-gamma ex-vivo by flow cytometry. In splenic CD4+ and TCRgammadelta+ T cells gene expression was quantified by flow cytometry and quantitative PCR.

Results: IL-23 was critical for full-blown AIA. Lack of IL-23 did not prevent the onset of joint inflammation but stopped the progression to a destructive synovitis. IL-23 regulated IL-17A expression in CD4+ T cells in the spleen. Of note, IL-17A and IFN-gamma expression was reduced in CD4+ T cells in the inflamed joints of IL-23p19-deficient mice. Interestingly, IL-23 was also critical for the induction of IL-17A and RORgammat but not IFN-gamma in TCRgammadelta+ T cells in the inflamed joints. The importance of the IL-23/IL-17 axis was further confirmed using IL-17 Receptor A knockout mice showing significantly milder AIA compared to control mice, with a disease course comparable to that of IL-23p19-deficient mice.

Conclusions: These data show that IL-23 is critical for full-blown expression of a non-autoimmune destructive arthritis and regulates the proportion of IL-17A and IFN-gamma-positive CD4+ T cells at the site of inflammation. Furthermore, IL-23 regulates IL-17A and RORgammat expression in TCRgammadelta T cells in arthritis. These findings indicate that regulating the IL-23 pathway may have therapeutic potential in non-autoimmune arthritis.

Figure 1

IL-23 has a critical role in the progression of antigen-induced arthritis. WT and IL-23p19KO mice were immunized with mBSA/CFA and one week later mono-arthritis was induced by injecting mBSA directly into the knee joints. A. Arthritis score was determined macroscopically at different time points. Mean values and SEM are given for 8 to 31 mice per group. Data are obtained from three separate experiments. * P < 0.001, WT vs IL-23p19KO. B, Histological analyses of joint inflammation and, C, bone erosion after H&E staining. B and C, Mean values and SEM are given from two separate experiments with a total of 20 knee joints per group. D-F. Cytokine levels in synovial washouts taken at days 1, 2 and 7. G and H. IL-17A and IFN-γ levels in synovial washouts taken at day 1. Mean values and SEM are given for 5 to 10 washouts obtained from two separate experiments.

Figure 2

The induction of Th17 cells in AIA is IL-23 mediated. A. WT mice were immunized with mBSA/CFA and 10 days later splenocytes were isolated and stimulated for four hours with PMA/Ionomycin, gated for CD4⁺ T cells and analyzed for intracellular IL-17A and IFN-γ expression. Numbers indicate percentage of positive cells within each quadrant. B-D. Antigen-induced arthritis was induced in WT and IL-23p19KO mice and at days 1 and 7, after i.a. mBSA injection, the splenocytes were isolated and stimulated for 4 h with PMA/Ionomycin and analyzed for intracellular IL-17A and IFN-γ expression on a CD4⁺ T cell gate. B. Numbers indicate percentage of positive cells within each quadrant. C and D. Quantification of flow cytrometric analyses from B; mean values and SEM are given for 6 to 12 mice per group from two to four independent experiments. E and F. On day 7 of AIA, splenic CD3⁺CD4⁺ T cells were FACS-sorted and gene expression was analyzed by quantitative RT-PCR for IL-17F and IL-22 respectively. Mean values and SEM are given for three mice. P-values were calculated using the student t-test.

Figure 3

IL-23 is critical for the induction of IL-17A and RORγt in TCRγδ T cells. A and B. At days 1 and 7 of AIA, splenocytes were isolated and stimulated for four hours with PMA/Ionomycin, gated for CD3⁺TCRγδ⁺ T cells and analyzed for intracellular IL-17A and IFN-γ respectively. Numbers in quadrants indicate percentage of cytokine-positive cells. C and D. Quantification of flow cytrometric analyses from A and B for IL-17A and IFN-γ; mean values and SEM are given for three to six mice for day 1 and for nine mice for day 7. E. Cell counts of total amount of TCRγδ⁺ T cells present in spleen on day 1 and 7 of AIA in WT and IL-23p19KO mice. F. On day 7 of AIA, splenic CD3⁺TCRγδ⁺ T cells were FACS-sorted and IL-17F gene expression was analyzed by quantitative RT-PCR. Mean values and SEM are given for three mice. P-values were calculated using the student's t-test.

Figure 4

IL-23 is essential for RORγt expression in TCRγδ ⁺ T cells. At day 7 of AIA, WT and IL-23p19KO mice were sacrificed and splenocytes were isolated. A. CD3⁺CD4⁺ T cells were FACS-sorted and RORγt gene expression was analyzed by quantitative RT-PCR (left panel); RORγt was measured intracellular and CD4⁺ T cells were gated (right panel). B T-bet expression was measured intracellular and CD4⁺ T cells were gated. C. CD3⁺TCRγδ⁺ T cells were FACS-sorted and RORγt gene expression was analyzed by quantitative RT-PCR (left panel); RORγt was measured intracellular and γδ⁺ T cells were gated (right panel). D. T-bet expression was measured intracellular and γδ⁺ T cells were gated. E. Comparison of the mRNA quantification of RORγt expression between FACS-sorted CD4⁺ and γδ⁺ T cells. Mean values and SEM are given for three mice and P-values were calculated using the student's t-test.

Figure 5

IL-23 deficiency results in less IL-17 production in the inflamed joint. At day 7 of AIA, cells from the arthritic joints of WT and IL-23p19KO mice were isolated and stimulated for four hours with PMA/Ionomycin. A. Intracellular cytokine staining of TCRγδ⁺ T cells for IL-17. B. Intracellular cytokine staining of TCRγδ⁺ T cells for IFN-γ; data from A and B are representatives from three mice per group. C. Quantification of flow cytrometric analyses from A and B. D. MFI was calculated for TCRγδ⁺ IL-17⁺ T cells. E. MFI was calculated for TCRγδ⁺ IFN-γ⁺ T cells. F. Total numbers of TCRγδ⁺ T cells in the joints of arthritic WT and IL-23p19KO mice. C-F, Mean values and SEM are given for three mice and P-values were calculated using the student t-test. G. Intracellular cytokine staining of CD4⁺ T cells for IL-17A and IFN-γ. Data are representatives of six mice per group. H. Quantification of flow cytrometric analyses from G. I. Total numbers of CD4⁺ T cells present in arthritic joints from WT and IL-23p19KO mice.

Figure 6

IL-17 receptor signaling is critical for the progression of AIA. A. WT, IL-23p19KO and IL-17RAKO mice were immunized with mBSA/CFA and mono-arthritis was induced one week later by an i.a. injection of mBSA. At days 1, 2, 4, 7 and 10 macroscopic scores were assessed. Mean values and SEM are given for six mice per group. * P < 0.01 WT vs IL-23P19KO and ^#, P < 0.01 WT vs IL-17RAKO. B. At day 7 of AIA, splenocytes from WT, IL-23p19KO and IL-17RAKO were isolated and stimulated for four hours with PMA/Ionomycin and analyzed for intracellular IL-17A and IFN-γ expression on a CD4⁺ and TCRγδ⁺ T cell gate. Numbers indicate percentage of positive cells within each quadrant and each dot plot shows a representative mouse from three mice per group analyzed.