Regulatory T cell vaccination without autoantigen protects against experimental autoimmune encephalomyelitis

Abstract

Regulatory T (T(reg)) cells show promise for treating autoimmune diseases, but their induction to elevated potency has been problematic when the most optimally derived cells are from diseased animals. To circumvent reliance on autoantigen-reactive T(reg) cells, stimulation to myelin-independent Ags may offer a viable alternative while maintaining potency to treat experimental autoimmune encephalomyelitis (EAE). The experimental Salmonella vaccine expressing colonization factor Ag I possesses anti-inflammatory properties and, when applied therapeutically, reduces further development of EAE in SJL mice. To ascertain T(reg) cell dependency, a kinetic analysis was performed showing increased levels of FoxP3(+)CD25(+)CD4(+) T cells. Inactivation of these T(reg) cells resulted in loss of protection. Adoptive transfer of the vaccine-induced T(reg) cells protected mice against EAE with greater potency than naive or Salmonella vector-induced T(reg) cells, and cytokine analysis revealed enhanced production of TGF-beta, not IL-10. The development of these T(reg) cells in conjunction with immune deviation by Th2 cells optimally induced protective T(reg) cells when compared those induced in the absence of Th2 cells. These data show that T(reg) cells can be induced to high potency to non-disease-inducing Ags using a bacterial vaccine.

FIGURE 1.

Oral Salmonella-CFA/I (H696) treatment reduced EAE clinical scores in PLP_139–151-challenged SJL/J mice (A) and decreased spinal cord demyelination (B) and inflammation (B and C). A, Mice therapeutically treated with H696 showed a significant reduction in their clinical scores with all mice recovering from EAE. Mice treated with Salmonella vector (strain H647) also recovered, but presented greater clinical scores when compared with PBS-treated mice. Depicted are the combined results from four separate experiments for a total of 20 mice/group: *, p < 0.001 for PBS vs H647 and PBS vs H696. B, H696-treated group showed minimal demyelination (LFB) and inflammatory (H&E stain) cell infiltration (indicated by arrows) at 14 days postchallenge when compared with PBS-dosed group; the H647-treated mice showed increased demyelination and inflammation. Representative samples from six mice are depicted. C, FACS analysis of spinal cord leukocytes was performed 14 days after challenge to assess the types of inflammatory cells: neutrophils, CD11b⁺SK208⁺; macrophages, CD11b⁺SK208⁻F4/80⁺; and T cells, TCRβ⁺. Minimal leukocytes were detected in H696-treated mice unlike that observed for PBS- or H647-treated mice; representative samples of five mice per group are depicted.

FIGURE 2.

Elevated Th2-type cytokines contribute to the therapeutic treatment by H696 vaccine by splenic (A) and CLN (B) lymphocytes with reductions in IFN-γ. Splenic (A) and CLN (B) lymphocytes were collected 14 days after EAE challenge (8 days after oral Salmonella vaccination) and cultured in the presence of OVA, purified CFA/I fimbriae, PLP_139–151, or medium. Data depict the mean of three different experiments ± SEM. *, p < 0.001 for H696 vs PBS and H696 vs H647.

FIGURE 3.

Oral Salmonella vaccination enhances the percentage of T_reg cells expressing FoxP3 in both the mucosal and systemic compartments. A, CLN lymphocytes taken from SJL mice 14 days after EAE challenge (treated with H696, H647, or PBS 6 days postchallenge) and were examined by FACS for the presence of CD25⁺CD4⁺ T cells. Each group showed stimulation of T_reg cells, but mice treated with H647 or H696 had increased numbers, and these were >90% FoxP3⁺. B, A kinetic analysis was performed on spleen, MLN, PP, SMLN, and deep CLN. Increases in FoxP3⁺ T_reg cells were observed for Salmonella-treated groups. *, p < 0.001 for PBS vs H647 and PBS vs H696. C and D, Oral vaccination with H696 only (no PLP_139–151 challenge) results in FoxP3⁺ T_reg cells in spleen, MLN, PP, SMLN, and (C and D) CLN peaking ~2 wk after immunization (mean of five mice per group ± SEM).

FIGURE 4.

The T_reg cells induced by Salmonella-CFA/I vaccination suppress effector T cells, and their in vivo inactivation results in enhanced clinical disease. A, CD25⁻CD4⁺ and CD25⁺CD4⁺ T cells from spleens, MLN, and HNLN were cell sorted from mice orally immunized with H696 for 2 wk. T_eff (1 × 10⁵) cells with feeder (1 × 10⁵) cells were cultured with CFA/I fimbriae and with or without T_reg (5 × 10⁴) cells for 4 days. During the last 18 h of culture, cells were pulsed with [³H]TdR. Proliferation in the absence of T_reg cells was set at zero inhibition. Depicted is the percent inhibition upon coculture with splenic, MLN, or HNLN T_reg cells. One of three experiments is depicted. B, One week before PLP_139–151 challenge, mice (five per group) orally immunized with H696 were treated in vivo with anti-CD25 mAb, rat IgG, or PBS on days −5 and −2 before challenge. Inactivation of T_reg cells provoked greater EAE than the PBS-treated group. Normal rat IgG-treated mice vaccinated with H696 showed minimal EAE. *, p < 0.001 for anti-CD25-treated group vs H696-vaccinated or H696-vaccinated plus normal rat IgG-treated mice.

FIGURE 5.

Adoptive transfer of T_reg cells obtained from Salmonella-CFA/I-immunized mice confers greater protection against EAE challenge than Salmonella vector-induced T_reg cells whereas those from naive mice failed to protect; protection is TGF-β dependent. Immune CD25⁺CD4⁺ T cells and CD25⁻CD4⁺ T cells from 2-wk H696 (A)- and H647 (B)-immunized SJL mice were obtained from pooled spleens, MLN, and HNLN, purified by cell sorting, and 6 × 10⁵ T_reg or T_eff cells were injected i.v. into naive SJL mice. PBS-, H696-, and H647-immunized mice were used as negative and positive controls, respectively. Recipient mice given immune H696 T_reg cells conferred total protection against EAE, whereas mice given T_eff cells showed only partial protection with greater clinical disease than H696-vaccinated mice. *, p < 0.001 for PBS vs T_reg cell group, CD25⁻CD4⁺ T cell group, or H696-vaccinated group. T_reg cells obtained from H647-vaccinated mice were less protective, and T_eff cells failed to protect against EAE. C, Naive SJL T_reg and T_eff cells were purified by cell sorting of lymphocytes from MLN, HNLN, and spleens. T_reg or CD25⁻CD4⁺ T cells (6 × 10⁵) were injected i.v. into naive recipients 1 day before PLP_139–151 challenge. While adoptive transfer of naive T_reg cells delayed EAE onset, these failed to protect against EAE development. *, p < 0.001 for PBS vs naive T_reg cell group or naive CD25⁻CD4⁺ T cell group.

FIGURE 6.

Oral vaccination with Salmonella-CFAH elicits TGF-β-producing T_reg cells and IL-4-, IL-10-, and IL-13-producing T_eff cells in contrast to IL-17-producing, but no Th2-type cytokine-producing, T_eff cells upon vaccination with Salmonella vector. Cell-sorted CD25⁺CD4⁺ and CD25⁻CD4⁺ T cells from mice orally immunized with H696 or H647 were evaluated for cytokine production following anti-CD3 and anti-CD28 costimulation. T_reg cells had reduced IFN-γ production but elevated TGF-β when compared with T_eff cells. IL-4, IL-10, and IL-13 segregated with the T_eff cells induced by vaccination with H696. T_eff cells from H647-vaccinated mice did not produce any Th2-type cytokines but rather produced elevated levels of IFN-γ. H647-induced T_reg cells did produce TGF-β (although less than H696-vaccinated mice) and IL-17. Thus, protection conferred by Salmonella-CFA/I-induced T_reg cells is because of reduced IFN-γ production and increased TGF-β, as well as in part supported by immune deviation by the T_eff cells. *, p < 0.001 represents differences in cytokine production between CD25⁺CD4⁺ and CD25⁻CD4⁺ T cells, and †, p < 0.001 represents differences in cytokine production between H696- and H647-sorted cells.