IL-35 stimulation of CD39+ regulatory T cells confers protection against collagen II-induced arthritis via the production of IL-10

Abstract

IL-35 is produced by regulatory T cells, and this novel cytokine can downregulate Th17 cell development and inhibit autoimmune inflammation. In this work, an rIL-35, as a single-chain fusion between murine IL-12p35 and EBV-induced gene 3, was expressed in yeast. This rIL-35 inhibited OVA-specific cellular and Ab responses in OVA-challenged recipients of DO11.10 CD4+ T cells. Likewise, IL-35 inhibited clinical manifestation of collagen-induced arthritis or could cease further disease exacerbation upon initiation of IL-35 treatment. Exogenous IL-35 treatments suppressed Th1 and Th17 cells and promoted CD39 expression by CD4+ T cells. Sorted CD25-CD39+CD4+ T cells from IL-35-treated mice produced IL-10 and, upon adoptive transfer, were sufficiently potent to inhibit subsequent development of inflammation in mice with collagen-induced arthritis, whereas sorted CD25+CD39+CD4+ T cells showed reduced potency. IL-35 treatments of IL-10-/- mice failed to induce protective CD39+CD4+ T cells, demonstrating the effector role of IL-10 by IL-35 immunosuppression.

FIGURE 1

rIL-35 suppresses OVA-specific immune responses. A, Schematic of murine IL-35 and Western blot of purified protein detected with anti-histag mAb, goat anti–IL-12 Ab, or anti-EBI3 mAb. B, OVA-specific DTH response. Sorted DO11.10 CD4⁺ T cells (>95% purity) were adoptively transferred to sex- and age-matched BALB/c mice. One day later, recipients were challenged with 100 μg OVA, and mice were treated daily with 0.75 μg IL-35 or 200 μl PBS. On day 5, DTH test was performed. *p <0.05; mean change in ear thickness of 5 mice per group ± SEM is shown. C, Serum OVA-specific Ab titers from treated recipients on day 7. *p <0.001; **p < 0.005 between IL-35– and PBS-treated groups. D, Proliferation of OVA-specific CD4⁺ T cells. CD4⁺ T cells were purified on day 7 and restimulated with OVA_323–339 in presence of syngenic irradiated APCs for 72 h. Mean SI ± SEM is shown from quadruplicate cultures. *p <0.001. E, OVA-specific CD4⁺ T cell cytokine production. Purified from recipients’ LNs, CD4⁺ T cells (10⁶ cell/ml) were restimulated with OVA_323–339 for 4 d in presence of syngenic irradiated APCs. Mean cytokine concentrations ± SEM from triplicate cultures are depicted. *p < 0.001; **p < 0.05.

FIGURE 2

IL-35 inhibits development of CIA in C57BL/6 mice. A, CIA was induced with s.c. injection of 100 μg chick CII emulsified in CFA. Beginning on day 21 postchallenge, IL-35 (0.75 μg/dose; n = 20) or 200 μl sterile PBS (n = 30) was given daily until day 28. Average clinical score per treatment group represents severity of the disease, and incidence of arthritis illustrates percent mice with affected joints in treatment group. The sum of three experiments is depicted (mice/group): *p < 0.001; ^♮p < 0.005; ^#p < 0.01. B, H&E-(left panels) and toluidine blue-stained (right panels) knee from naive, PBS-, or IL-35–treated mice at termination of study (original magnification ×100). C, Histology score was evaluated on scale 0–3 on H&E-stained sections, and cartilage loss was graded from toluidine blue sections on scale 0–3 for each paw plus knee sections (total scores of 18 possible per mouse). *p < 0.05 between PBS- and IL-35–treated groups.

FIGURE 3

IL-35 suppresses CII-specific Ab and CD4⁺ T cell responses. A, Serum CII-specific Ab titers from PBS- or IL-35–treated mice as described in Fig. 2. Individual serum samples were collected on day 21 (prior first treatment) and on day 35 (1 wk after last treatment). *p < 0.01; **p < 0.05 as compared with PBS group on day 35. B, Cytokine production by CII-specific LN CD4⁺ T cells. Purified CD4⁺ T cells (10⁶ cell/ml, >95% purity) were restimulated with CII in presence of syngenic irradiated APCs for 4 d. Mean cytokine concentrations from triplicate cultures ± SEM are depicted. *p < 0.001; **p < 0.005. Data are representative of three experiments. ND, not detected.

FIGURE 4

IL-35 ceases further progression of established CIA. CIA was induced as described in Fig. 2, and mice were treated daily i.p. with sterile PBS or IL-35 (0.75 μg) beginning day 35 through day 38 post CII challenge. A, Average clinical score per group; arrows indicate treatment time points. *p < 0.05 between groups. B, Δ Cumulative swelling calculated as a difference between cumulative swelling of individual mice in PBS- (n = 10 mice) or IL-35–treated (n = 11 mice) groups and average cumulative swelling in naive mice. Measurements were performed in anesthetized mice on day 42 post CII challenge. *p < 0.05.

FIGURE 5

IL-35 stimulates induction of regulatory CD39⁺CD4⁺ T cells. CIA was induced, and IL-35 or PBS treatments were conducted as described in Fig. 2. On day 28, flow cytometry analysis was performed on stained splenic and draining LN lymphocytes, assessing CD39, CD25, and Foxp3 expression. A, Enhanced CD39 expression by splenic CD4⁺ T cells following IL-35 treatments. B, Frequencies of splenic and LN CD39⁺CD4⁺ T cells (six individual mice per group). *p < 0.001; **p < 0.05. C, FoxP3 expression in LN CD39⁺CD4⁺ T cells. Shaded histogram is isotype control (rat IgG2a). Mean fluorescence intensity (MFI) is 1023 ± 79 and 1731 ± 52 for PBS- and IL-35–treated mice, respectively. p = 0.002. D, Frequencies of FoxP3⁺CD39⁺CD4⁺ T cells in spleens and LNs of PBS- or IL-35–treated mice (six individual mice per group). *p < 0.001. E, CD39⁺CD4⁺ T cells include CD25⁻CD4⁺ and CD25⁺CD4⁺ T cell subsets in PBS- and IL-35–treated mice. Gated LN CD39⁺CD4⁺ T cells are depicted. F, FoxP3 expression by LN CD25⁺CD4⁺ and CD25⁻ CD4⁺ T cells following treatment with PBS or IL-35. MFI for FoxP3⁺CD25⁺CD4⁺ T cells is 1127 ± 82 and 2108 ± 42 for PBS- and IL-35–treated mice, respectively. p < 0.001 between treatment groups. MFI for FoxP3⁺CD25⁻CD4⁺ T cells is 494 ± 23 and 875 ± 63 for PBS- and IL-35–treated mice, respectively. p = 0.005 between treatment groups. Shaded histogram is rat PE-IgG2a stained control. G, At 42 d post CII challenge, a T cell proliferation assay using cell-sorted CD39⁺CD4⁺ and CD39⁻CD4⁺ T cells (>90% purity) from mice treated with IL-35 or PBS was performed. Mean SI ± SEM is depicted from triplicate cultures. *p < 0.001; **p < 0.01.

FIGURE 6

Cytokine profiles of CD39⁺ and CD39⁻CD4⁺ T cells from IL-10^+/+ and IL-10^−/− mice after IL-35 or PBS treatment. Naive mice were given PBS or IL-35 for 7 d, and then splenic and LN CD4⁺ T cells were sorted for CD39⁺ and CD39⁻ CD4⁺ T cell subsets (>93% purity). Each (10⁶ cells/ml) was stimulated with plate-bound anti-CD3 and soluble anti-CD28 mAbs for 4 d. Mean concentration of cytokine in supernatant ± SEM from triplicate cultures is shown. A, CD39⁻CD4⁺ T cells’ cytokine profile. *p < 0.001; **p < 0.005; ***p < 0.01. B, CD39⁺CD4⁺ T cells’ cytokine profile. *p < 0.001; **p < 0.01; ***p < 0.05 between IL-10^+/+ and IL-10^−/− mice.

FIGURE 7

IL-35–induced CD39⁺CD4⁺ T cells from IL-10^+/+, not IL-10^−/− mice, confer protection to CIA. A, Scheme for adoptive transfer experiment. CIA was induced in recipient groups on day 0. On days 7–14, IL-35 was given to donor mice. On day 15 postinduction of CIA, recipient mice were adoptively transferred with 10⁶ of cell-sorted CD39⁺ or CD39⁻ CD4⁺ T cells (>93% purity) from IL-10^+/+ or IL-10^−/− mice. B, Severity of arthritis and incidence of the disease in recipients of CD39⁻CD4⁺ T cells are shown. C, Severity of arthritis and incidence of the disease in recipients of CD39⁺CD4⁺ T cells are shown. Data represent mean clinical score per group of 10 mice ± SEM. *p < 0.001; ^✢p < 0.01. One of three experiments is depicted. CD4⁺ T cell cytokine production by recipients given CD39⁻CD4⁺ T cells (D) and by recipients given CD39⁺CD4⁺ T cells (E) is shown. LN cells were restimulated with CII for 4 d, as previously described. *p < 0.001; **p < 0.005; ***p < 0.05.