Modulation of IL-17 and Foxp3 expression in the prevention of autoimmune arthritis in mice

Abstract

Background: Rheumatoid Arthritis (RA) is a chronic immune mediated disease associated with deregulation of many cell types. It has been reported that different T cell subsets have opposite effects in disease pathogenesis, in particular Th17 and Treg cells.

Methodology and findings: We investigated whether non-depleting anti-CD4 monoclonal antibodies, which have been reported as pro-tolerogenic, can lead to protection from chronic autoimmune arthritis in SKG mice--a recently described animal model of RA--by influencing the Th17/Treg balance. We found that non-depleting anti-CD4 prevented the onset of chronic autoimmune arthritis in SKG mice. Moreover, treated mice were protected from the induction of arthritis up to 60 days following anti-CD4 treatment, while remaining able to mount CD4-dependent immune responses to unrelated antigens. The antibody treatment also prevented disease progression in arthritic mice, although without leading to remission. Protection from arthritis was associated with an increased ratio of Foxp3, and decreased IL-17 producing T cells in the synovia. In vitro assays under Th17-polarizing conditions showed CD4-blockade prevents Th17 polarization, while favoring Foxp3 induction.

Conclusions: Non-depleting anti-CD4 can therefore induce long-term protection from chronic autoimmune arthritis in SKG mice through reciprocal changes in the frequency of Treg and Th17 cells in peripheral tissues, thus shifting the balance towards immune tolerance.

Figure 1. SKG mice develop chronic autoimmune arthritis upon induction with curdlan.

(A) Arthritis incidence in female SKG and BALB/c mice after a single i.p. injection of 3 mg curdlan (n = 6). Curdlan was able to induce chronic arthritis in SKG but not in BALB/c mice. Each point represents the median per group. (B) Clinical score of 5 month-old male (M) and female (F) SKG mice 90 days following disease induction with curdlan. Data are from two independent experiments.

Figure 2. Non-depleting anti-CD4 MAb prevents the onset of autoimmune arthritis.

(A) Female SKG mice were immunized with 3mg curdlan i.p. together with 1 mg non-depleting anti-CD4 or an isotype control. The MAb was administered again on days 2 and 4. Anti-CD4 treated mice were protected from the development of autoimmune arthritis (n = 6, P<0.001). Data, represented as mean ± SEM, are from two independent experiments. (B) Serum concentration of rheumatoid factor (RF) was measured by ELISA. Mice treated with anti-CD4 showed significantly lower levels of RF (n = 6, P<0.001) (C) Histological sections stained with eosin-hematoxilin from the ankle and metatarso-phalangical joints from SKG mice in the absence and in the presence of anti-CD4 treatment, 90 days following curdlan immunization. (D) Serum concentration of IL-6, IFN-γ, TNF and MCP-1 in naive SKG mice, SKG mice exposed to curdlan, or curlan and anti-CD4. Naive SKG mice were age matched and did not develop arthritis in the absence of curdlan immunization (no induction). The serum levels of IL-6 (P<0.05), IFN-γ (P<0.01) and MCP-1 (P<0.01) were significantly lower in anti-CD4 treated mice compared with animals injected with curldan in the absence of tolerizing MAbs. Differences in TNF concentration did not reach statistical significance. (E) The serum concentration of IL-10 and IL-17 in SKG mice exposed to curdlan, or curlan and anti-CD4 remained similar in the different experimental groups. Culture supernatants from Th17 cell culture were used as positive control.

Figure 3. Anti-CD4 MAb influences the local balance of Th17/Treg cells.

(A) Frequency of splenic CD4⁺ T cells or CD4⁺Foxp3⁺ Treg cells from SKG mice exposed to curdlan, or curdlan + anti-CD4 treatment. No significant difference was observed between the two populations of animals. (B) Representative dot plots showing the frequency of splenic CD4⁺Foxp3⁺ T cells from SKG mice exposed to curdlan, or curdlan + anti-CD4 treatment. No significant difference was observed, as represented in panel A. (C) Foxp3 and IL-17 mRNA expression from the synovial membrane of SKG mice exposed to curdlan, or curdlan + anti-CD4 (mRNA expression levels were normalized to CD3 expression). (D) Frequency of Foxp3⁺ and IL-17⁺ T cells within draining LNs of SKG mice exposed to curdlan, or curdlan + anti-CD4.

Figure 4. CD4-blockade prevents Th17 polarization.

(A) Sorted TCR-transgenic cells were stimulated in vitro with peptide-loaded DCs under culture conditions known to preferentially polarize Th17 cells, with the addition of recombinant TGF-β, IL-6, IL-1β, and anti-IFN-γ. After 5 days of culture we observed a significant reduction of IL-17⁺ cells in the presence of anti-CD4 (n = 6, P<0.05). In contrast, anti-CD4 addition led to an increased frequency of Foxp3⁺ T cells (n = 6, P<0.05). (B) Representative dot plots from the two different culture conditions. An independent experiment was performed with a peptide dose of 0.3 µM with similar results.

Figure 5. Long-term protective effect of non-depleting anti-CD4 treatment does not affect immune competence.

(A) Female SKG mice were injected with 3 mg curdlan i.p. on day 0 and anti-CD4 on days 0, 2, and 4. Animals treated with anti-CD4 displayed long term protection from arthritis (n = 5, P<0.05). On day 60, anti-CD4 treated mice and age matched SKG were challenged with 3 mg curdlan i.p. Mice previously treated with anti-CD4 remained protected from the development of autoimmune arthritis (n = 5, P<0.05). (B) Female SKG mice were injected with 3 mg curdlan i.p. and, when the clinical score reached 0.5, some of the animals initiated treatment with three i.p. administrations of 1 mg anti-CD4 on alternate days. Control animals rapidly progressed to severe arthritis unlike the anti-CD4 treated mice (n = 6, P<0.05). (C) Mice were treated with non-depleting anti-CD4 30 days before immunization with OVA-alum i.p. (αCD4…OVA). One week following sensitization the serum levels of OVA-specific IgG1 were quantified, and compared with untreated (PBS…OVA) and non-immunized controls (PBS…PBS). Mice treated with anti-CD4 MAbs were competent to produce OVA-specific IgG1 to titres similar to untreated controls, and considerably higher than the non-immunized mice (n = 5, P<0.05). (D) OVA-specific IgG1 in mice where OVA was initially administered at the time of anti-CD4 treatment (tOVA…OVA), compared with animals that were not initially treated with anti-CD4 (PBS…OVA). Mice treated with anti-CD4 at the time of sensitization with OVA, became tolerant to the subsequent OVA immunization (n = 6, P<0.05).