Induction of CCR6 on Th17 cells in vitro. (A) Retroviral vector, pMxs-RORγt-IRES-GFP (RORγt-IG), or control pMxs-IRES-GFP (Control-IG), was transduced into BALB/c CCR6⁻CD4⁺ T cells, as previously described (reference 7), and the cells expressing GFP were stained for CCR6 on day 4 after transduction. (B) The mean percentages ± SD of CCR6⁺ or IL-17⁺ cells among CD4⁺GFP^high cells (as gated in A) in three independent experiments are shown. (C) CD4⁺CD25⁻CCR6⁻ T cells were in vitro driven to differentiate to Th17 cells in the indicated conditions, as previously described (reference 21). CCR6 and intracellular IL-17 gated on CD4⁺ T cells were stained on day 5. Results in A and C represent three independent experiments, and the percentage of cells in each quadrant is shown.

Production of IL-17 and CCL20 in RA joints. (A and B) The concentrations of IL-17 and CCL20 in synovial fluid from RA or OA patients were measured by ELISA. Vertical bars represent the means ± SD. (C–E) Scatter plots show the correlation between IL-17 and IFN-γ (C), CCL20 and IFN-γ (D), or CCL20 and IL-17 (E) in the synovial fluid of RA patients. RA and OA patients are 66.6 ± 11.6 and 74 ± 7.4 yr old, respectively. r² values of Pearson's product-moment correlation and p-values of their null hypothesis are shown. Similar analyses performed only on the samples with the amounts of IL-17, IFN-γ, or CCL20 detectable by ELISA yielded the following statistics: IL-17 versus CCL20, r² = 0.147 and P = 0.177 (n = 14); IFN-γ versus CCL20, r² = 0.0034 and P = 0.481 (n = 17); IL-17 versus IFN-γ, r² = 0.359 and P = 0.003 (n = 13).

Contribution of CCR6/CCL20 to trafficking of Th17 cells in SKG arthritis. (A) CD4⁺ T cells infiltrating arthritic joints were prepared from collagenase-digested ankle joints of SKG mice, and stained for CCR6 and intracellular IL-17 or IFN-γ. (B) SKG mice received an i.v. injection of 100 μg anti-CCR6 antibody or control IgG; LN cells were stained by anti–rat IgG and CD4 48 h after injection. The percentage of cells in each quadrant is shown in A and B. (C) SCID mice received an i.v. injection of 10⁶ SKG CD4⁺ T cells, and then injections of 100 μg anti-CCR6 mAb or control IgG once a week for 5 wk. The arrows indicate i.v. injections of anti-CCR6 mAb. The incidence and severity of arthritis was scored every week, as previously described (reference 8). Vertical bars represent the means ± SEM. The disease curves of arthritis scores are significantly different between the two groups (P < 0.005 according to the analysis of covariance test). The differences in scores are statistically significant (according to the Mann-Whitney U test) in the 8th (*, P = 0.04), 9th (*, P = 0.02), and 10th (*, P = 0.04) wk. Results in A and B represent three independent experiments.

Cross-regulation of synoviocyte CCL20 production by cytokines and preferential migration of Th17 cells in response to CCL20. (A) 2.5 × 10⁴ adherent synoviocytes, prepared as previously described (reference 10), were cultured with 10 ng/ml of the indicated cytokine for 24 h. The amounts of CCL20 in culture supernatants were measured by ELISA. Data are shown as the mean ± SD of triplicate wells. (B and C) Purified CCR6⁺ or CCR6⁻CD4⁺ T cells from SKG mice were stimulated with PMA/ionomycin for 3 h; CCL20 messenger RNA was assessed by quantitative RT-PCR (B), and CCL20 protein was measured by ELISA (C). Data are shown as the mean ± SD of three independent experiments. (D) CD4⁺ T cells were in vitro driven to differentiate to Th17 cells in the presence of IL-6 and TGF-β, as previously described (reference 21). Cells were restimulated with PMA/ionomycin for 3 h on day 4, and the amounts of CCL20 were measured by ELISA. (E) SKG LN cells were placed with 100 μg/ml anti-CCR6 antibody or control IgG on the upper well of a Transwell system. The migration assay was performed in the presence of designated concentrations of CCL20 added to the bottom well. See Materials and methods for details of the experiments. Data are shown as the mean ± SD of triplicate wells. (F) The migration assay was performed in the absence or the presence of 50 ng/ml CCL20, as in E. CD4⁺ cells before being added to the upper well and CD4⁺ T cells that had migrated to the lower well were stained for intracellular IL-17 and IFN-γ (top). Percentages of IL-17⁺ or IFN-γ⁺ cells among CD4⁺ cells in the lower wells in three independent migration assays with (+) or without (−) CCL20 are shown (bottom). Vertical bars represent the means ± SD. Results in A, D, and E represent three independent experiments.

Gene microarray analysis of Th17 cells and their predominant expression of CCR6. (A) LN CD4⁺ T cells from 10-mo-old BALB/c or SKG mice were stained for intracellular IL-17 and IFN-γ, or these cytokines and cell-surface FR4. (B) CD4⁺ T cells from IFN-γ^−/− or IL-6^−/− mice were transferred to RAG2^−/− mice or IL-6^−/−RAG2^−/− mice, respectively. Intracellular IL-17 and IFN-γ in recipient splenic CD4⁺ T cells were stained on day 7. (C) Purified FR4⁻CD4⁺ T cells from BALB/c or SKG mice or CD4⁺ T cells after homeostatic proliferation, as shown in A and B, were stimulated with PMA/ionomycin for 3 h, and the amounts of IL-17 in culture supernatants were measured by ELISA. (D and E) Total RNA extracted from activated cells, as shown in C, was subjected to gene microarray for SKG versus BALB/c FR4⁻CD4⁺ T cells or IFN-γ^−/− versus IL-6^−/−CD4⁺ T cells after homeostatic proliferation. The expression of each gene was averaged from three independent experiments and analyzed by GeneSpring software. The Venn diagram shown in D represents a group of up-regulated genes in SKG FR4⁻CD4⁺ T cells and IFN-γ^−/−CD4⁺ T cells after homeostatic proliferation. Genes for cytokines, chemokines, or their receptors commonly up-regulated in the two groups of genes are shown in E. Graded colors represent relative expression levels. (F) LN CD4⁺ T cells from 2- or 10-mo-old SKG mice were stained for CCR6 and intracellular IL-17 or IFN-γ. (G) CD4⁺ T cells from IFN-γ^−/−, IL-6^−/−, or WT mice were stained for CCR6 and intracellular IL-17 or IFN-γ after homeostatic proliferation, as shown in B. (H) Total RNA extracted from purified SKG CCR6⁺ or CCR6⁻CD4⁺ T cells stimulated with PMA/ionomycin for 3 h was subjected to quantitative RT-PCR for the indicated genes and normalized for hypoxanthine-guanine phosphoribosyltransferase (HPRT) messenger RNA expression, as previously described (reference 7). Data are shown as the mean ± SD of three independent experiments. Results in A–C, F, and G represent three to five independent experiments. The percentage of cells in each quadrant is shown in A, B, F, and G.