In this study, we demonstrated that IFNβ treatment of naïve T cells in Th17 differentiation conditions resulted in decreased expression of IL-17 and RORγt. In contrast, treatment of IFNα/β led to increased number of both IL-10+ and IL-17+IL-10+ T cells, consequently enhanced IL-10 production in Th17 polarizing culture. We also found that IFNβ induced IL-10 production and reduced IL-17 production in antigen specific T cells derived from EAE mice. Furthermore, we found that treatment of myelin-specific T cells with IFNβ reduced their pathogenic function, and caused less severity of EAE in an adoptive transfer model. Results from this study imply that IFNβ may induce antigen-specific T cells to produce IL-10, thereby forming a negative feedback loop to regulate inflammatory and autoimmune response mediated by self-reactive T cells, probably including both Th17 and Th1 cells.
Our results suggest that type I IFN could suppress Th17-associated inflammation through multiple mechanisms, though further investigation is needed. Type I IFN may directly suppress the differentiation of Th17 cells, as evidenced by the inhibition of RORγt and IL-17 expression. During Th17 development, TGFβ and IL-6 induce naïve T cells to secrete IL-21, which functions as an autocrine factor to upregulate Th17 lineage-specific transcription factor RORγt and expression IL-23R 
. It would be interesting to know if components of IFN signaling pathways can directly interact with molecules regulating Th17 differentiation. Another possible mechanism is that IFNβ may inhibit Th17 differentiation via induction of IL-10, which serves as a negative regulator for Th17 cells. IFNβ-mediated upregulation of IL-10 may suppress Th17 cells through inhibition of RORγt and IL-17 gene expression. Alternatively, IFN may promote the survival or expansion of IL-10 producing T cells. However, we observed similar level of dividing capacity in IL-10 producing T cells and IL-17 producing T cells when treated with IFNβ in CFSE labeling experiments, indicating that increased IL-10 producing T cells unlikely resulted from the outgrowth or survival of IL-10+
populations in our culture system. However, we can't absolutely exclude the possibility that IFN may selectively induce a particular IL-10+
T cell sub-population, especially during in vivo
immune response. Therefore, further studies are required to define and track cell populations responsible for IFN-induced IL-10 production during immune response.
Previous work from our laboratory has shown that activation of type I IFN induction pathway constrains Th17 cells and EAE development. Our data suggest that IFNβ might indirectly inhibit Th17 cells via induction of IL-10 and IL-27 from macrophage and DCs. While induction of IL-27 may be one of important mechanisms for clinical benefits of IFNβ treatment, we hypothesize that the IFNβ-mediated IL-10 induction in T cells may represent an additional mechanism to inhibit Th17 cells and EAE. IL-27R deficient mice develop severe immunopathology in several infection and autoimmune models because of excessive inflammation, reminiscent the phenotypes of IL-10 and IFNAR deficient mice 
. In the context of autoimmunity, the overlapping EAE phenotypes of IL-27, IL-10, and IFNAR deficient mice suggest that these molecules are probably functionally linked in the regulation of Th17 development. Our data also show that IL-27 can induce IL-10 production from activated T cells. Whether Th17 cells can produce IL-10 is still controversial. A number of studies show that IL-6 plus TGFβ, IL-27 alone or with TGFβ, could induce T cells under Th17 conditions to upregulate IL-10 expression. However, some published studies suggest that no IL-10 is produced by Th17 cells 
. Nevertheless, our data suggest that Th17 cells could produce IL-10 in our experimental conditions. We also found either IFNβ or IL-27 could enhance IL-10 production from T cells. At present, it remains unclear regarding the relative contribution of IL-10 produced by different types of immune cells in normal and disease conditions.
Emerging evidence points to the heterogeneity and plasticity in Th17 cells. Our data suggest that IL-10 production from T cells, possibly Th17 cells and Tr1 cells, may act as a negative regulator to dampen inflammatory response mediated by antigen specific T cells or self-reactive T cells during autoimmune diseases. Although Th17 cells have been implicated in a number of autoimmune diseases, several recent reports suggest that Th17 cells have protective roles under certain conditions in inflammatory and autoimmune diseases. For example, McGeachy et al. reported that pathogenic Th17 cells cultured with TGFβ and IL-6 led to the generation of IL-10+
cells, which inhibited the pathogenic potential of Th17 cells and suppress the development of EAE in an adoptive transfer experiment of EAE 
. The authors further suggest that IL-10 producing Th17 cells may have bystander suppressive effects to inhibit fully differentiated pathogenic Th17 populations, and the development of neuronal inflammation. O'Connor et al. also reported that IL-17 had a protective function in the development of T cell-mediated colitis 
. It is possible that IFN-mediated IL-10 production from T cells could contribute to the inhibition of EAE development. We also noticed that IFNβ treatment led to an increase of IL-10+
double positive cells among CD4 T cells under Th17 polarizing conditions. The induction of a population of IL-10/IL-17 double positive T cells suggests that IFNβ was able to elicit the expression and production of IL-10 from T cells that have differentiated into the Th17 cells. However, at present, it is not clear if IL-10/IL-17 double positive T cells are transit populations to IL-10 single positive cells, or a Th17 population transiently expressing IL-10. Nevertheless, our results show that IFNα/β could induce a small population of IL-10 and IL-17 double positive cells in T cells cultured under Th17 polarizing conditions. These results suggest that potential source of IL-10 in IFN-treated T cells may include both IL-10 single positive T cells and IL-10/IL-17 double positive T cells.
In summary, our data suggest that IFNβ-mediated IL-10 production from Th17 cells may contribute to the therapeutic efficacy of IFNβ. This result is also correlated with clinical observations that IFN treatment leads to increased production of IL-10 in MS patients. Furthermore, the co-transfer of T cells treated with IFNβ decreased the severity of EAE compared with transfer of untreated cells. These results imply that IFN-treated T cells may have regulatory effects in vivo. Despite efficacy of IFNβ in treating multiple sclerosis, the very short-half life and side effects limit its use. Our results suggest that in vitro induction of IL-10-producing T cells might provide an alternative strategy for the treatment of Th17-associated inflammatory diseases, such as EAE and MS. In addition to inducing IL-10 production, IFNβ is able to induce a number of signaling pathways and downstream genes. It would be important to elucidate the complexity and interaction of IFNβ-induced multiple genes and signaling pathways in regulating Th17 differentiation and autoimmune diseases.