Epicutaneous immunization with type II collagen inhibits both onset and progression of chronic collagen-induced arthritis

Abstract

Epicutaneous immunization is a potential non-invasive technique for antigen-specific immune-modulation. Topical application of protein antigens to barrier-disrupted skin induces potent antigen-specific immunity with a strong Th2-bias. In this study, we investigate whether the autoimmune inflammatory response of chronic collagen-induced arthritis (CCIA) in DBA/1-TCR-beta Tg mice can be modified by epicutaneous immunization. We show that epicutaneous immunization with type II collagen (CII) inhibited development and progression of CCIA and, importantly, also ameliorated ongoing disease as indicated by clinical scores of disease severity, paw swelling and joints histology. Treated mice show reduced CII-driven T cell proliferation and IFN-gamma production, as well as significantly lower levels of CII-specific IgG2a serum antibodies. In contrast, CII-driven IL-4 production and IgE antibody levels were increased consistent with skewing of the CII response from Th1 to Th2 in treated mice. IL-4 production in treated mice was inversely correlated with disease severity. Moreover, T cells from treated mice inhibited proliferation and IFN-gamma production by T cells from CCIA mice, suggesting induction of regulatory T cells that actively inhibit effector responses in arthritic mice. The levels of CD4(+)CD25(+) T cells were however not increased following epicutaneous CII treatment. Together, these results suggest that epicutaneous immunization may be used as an immune-modulating procedure to actively re-programme pathogenic Th1 responses, and could have potential as a novel specific and simple treatment for chronic autoimmune inflammatory diseases such as rheumatoid arthritis.

Figure 1. Epicutaneous immunization prevents evolution of arthritis.

DBA/1-TCR-β Tg mice were epicutaneously immunized with CII 3 weeks prior to induction of disease by injection of CII in CFA. Mice were evaluated and clinical score (A) and hind paw thickness (B) were assessed every second day. Data represents the mean+1 SEM (n = 9). The epicutaneous immunization had a highly significant effect on disease severity as analysed by two-way ANOVA. The entire in vivo experiment was repeated twice with very similar results. (•) = treated with epicutaneous CII, (▴) = sham-treated CCIA controls.

Figure 2. Histopathological damage of joints is reduced following epicutaneous immunization.

Hind paws from healthy mice, CCIA control mice and epicutaneously treated mice were sectioned and stained with H&E (A). One representative example is shown. Note that skin and muscle remain unaffected in CCIA mice. Histological sections were evaluated and arbitrary scores assigned in 5 categories (4 being the most severe score) for extent of inflammatory infiltrate, cartilage and bone erosion. The number of affected hind paws assigned each severity score is shown (B). The mean histopathological severity score was significantly lower in treated mice, p<0.003 (n = 18). Healthy = un-CFA-immunized non-arthritic littermate, CCIA (▪) = sham-treated arthritic mouse, Treated (▪) = epicutaneously immunized mouse.

Figure 3. Epicutaneous immunization inhibits CII-specific proliferation, reduces Th1 and enhances Th2-type cytokines.

Splenocytes were collected from each mouse and cultured for 90 h with CII or control antigen. Proliferation was determined by [³H]-thymidine incorporation (A) and cytokine production by ELISA (B). Results of proliferative experiments are expressed as mean cpm+1 SEM (n = 9). Background proliferation when no antigen was present has been subtracted. No proliferation was observed to control antigen but the two groups responded with similar [cpm] when stimulated with ConA. Production of IFN-γ, IL-4, IL-13, IL-10 and TGF-β is expressed as mean pg/ml+1 SEM (n = 9). No cytokines were produced in response to a control antigen or when cells were not stimulated. Dashed horizontal lines (where visible) represents limit of detection. (□) = un-immunized healthy controls, (▪) = treated with epicutaneous CII, (▪) = sham-treated CCIA controls.

Figure 4. CII-driven production of IL-4 correlates inversely with severity of arthritis.

The linear correlation between Th2/Th1 cytokine production and severity of arthritis following epicutaneous immunization was analysed using linear regression. Production of IL-4 was negatively correlated to the clinical score (A) while production of IFN-γ was positively correlated to clinical score (B). Each point represents one mouse. (•) = treated with epicutaneous CII, (▴) = sham-treated CCIA controls.

Figure 5. Epicutaneous immunization inhibits development of CII-specific IgG2a and IgG.

CII-specific IgG, IgG1 and IgG2a as well as total IgE were measured in serum at the end of the experiment. Serum samples were diluted 1∶1500 for IgG, IgG1 and IgG2a and 1∶20 for IgE prior to analysis. Each bar represent mean antibody level+1 SEM (n = 9). (▪) = treated with epicutaneous CII, (▪) = sham-treated CCIA controls.

Figure 6. Lymphocytes from epicutaneously immunized mice inhibit proliferation and IFN-γ production by lymphocytes from CCIA mice.

Splenocytes from a sham-treated CCIA control were cocultured with splenocytes from an epicutaneously treated mouse at varying ratios (A). Total number of cells in the culture was kept constant. Cocultures were stimulated with 50 µg CII/ml and CII-specific splenocyte proliferation was assayed as in Figure 3. Black bars demonstrate the expected proliferation [cpm] in the culture, calculated by the formula: [cpm] = a f+b(1−f), where a = [cpm] with 100% control cells, b = [cpm] with 100% treated cells, f = fraction of control cells in the culture. The observed [³H]-thymidine incorporation is shown in white bars and represents the mean of triplicate cultures. Similar results were obtained in four individual experiments. Production of IFN-γ and IL-4 in 1∶1 ratio cocultures of splenocytes from control and treated mice were assessed by ELISA and results are expressed as mean pg/ml+1 SEM (n = 4) (B). No cytokines were produced in response to a control antigen or when cells were not stimulated. The proportion of CD4⁺CD25⁺ T cells in freshly isolated splenocytes (C) from CCIA control or epicutaneously immunized mice and after 90 h in vitro stimulation with 50 µg CII/ml (D) were assessed by flow cytometry. Results are expressed as the mean [%] CD4⁺CD25⁺ T cells out of the total CD4⁺ population+1 SEM (n = 9). (▪) = treated with epicutaneous CII, (▪) = sham-treated CCIA controls.

Figure 7. Treatment with epicutaneous immunization ameliorate ongoing arthritis.

Arthritis were induced in Tg mice by injection of CII in CFA and 1 week later mice were treated with epicutaneous CII. Mice were evaluated and clinical score (A) and hind paw thickness (B) were assessed every second day for 49 days. Data represents the mean+1 SEM (n = 9). Treatment by epicutaneous immunization had a significant effect on disease severity as analysed by two-way ANOVA. (•) = treated with epicutaneous CII, (▴) = sham-treated CCIA controls.

Figure 8. Treatment with epicutaneous immunization reduces CII-driven proliferation and IFN-γ production while IL-4 production is enhanced.

Splenocytes were collected from all mice on day 49 after induction of disease and cultured with CII or control antigen for 90 h. CII-driven proliferation (A) and production of IFN-γ, IL-4, IL-13, IL-10 and TGF-β (B) were assayed as described in Figure 3. Results are expressed as mean+1 SEM (n = 9). (□) = un-immunized healthy controls, (▪) = treated with epicutaneous CII, (▪) = sham-treated CCIA controls.

Figure 9. Treatment with epicutaneous immunization reduces levels of CII-specific IgG2a and IgG. CII-specific IgG, IgG1 and IgG2a as well as total IgE were measured in serum at the end of the experiment.

Serum samples were diluted 1∶1500 for IgG, IgG1 and IgG2a and 1∶20 for IgE prior to analysis. Each bar represent mean antibody level+1 SEM (n = 9). (▪) = treated with epicutaneous CII, (▪) = sham-treated CCIA controls.