Our previous reports have shown that in the EAE model, MOG35–55 peptide-specific CD8+ T cells were antigen-specific and encephalitogenic, with IFN-γ production, but were weaker than their CD4+ counterparts. However, the subtype of CD8+ T cells that plays a critical role in the pathogenesis of EAE remained unclear. Currently, the major subsets of CD8+ T cells are Tc1, Tc2, Tc9, Tc17, and Treg cells.[23, 24] Among these subsets, Tc17 cells are strongly associated with autoimmune diseases in both human and animal models.[28–34]
In the present study, we used our already-established strategy for the preparation of the EAE model to study the role of MOG35–55-specific Tc17 cells in the pathogenesis of EAE and MS.[51, 53, 54, 56] Here, we focused on Tc17 cells in EAE and used Th17 cells (CD4+ counterparts of Tc17) and Tc1 cells (different subsets of CD8+) as controls. We found that Tc17 cells responded to MOG35–55 peptides and secreted IFN-γ and IL-17, as well as could induce tEAE in naïve mice models. However, we observed that the proliferation assay results, cytokine secretion, and encephalitogenic activities of Tc17 cells are weaker than those of their Th17 counterparts. Compared with MOG35–55-specific Tc1 cells which we previously explored, MOG35–55-specific Tc17 cells showed similar proliferation assay results, lower IFN-γ and higher IL-17 secretion, and weaker encephalitogenic. The results of this study are consistent with our previous findings on MOG35–55-specific Th1 and Tc1 cells in the EAE model, and interphotoreceptor retinoid-binding protein (IRBP)1–20-specific Th1 and Th17 cells in the EAU model, both of which were studied on female C57BL/6 mice.
As mentioned previously, similar functional profiles of CD4+ and CD8+ autoreactive T cells and their subsets were observed in closely related autoimmune disease models, such as MS, uveitis, and type I diabetes, all of which were also female-dominated animal models. Hence, future research into the common mechanisms of these autoimmune diseases, including thymus development,[57] local microenvironments,[57, 58] and estrogen relationships[59, 60] would be beneficial.
Wagner et al. suggested that classic EAE was dominated by CD4+ T cells, whereas atypical EAE was CD8+ T cells dominant, explaining why the classic EAE model did not mimic all features of MS very well.[5] Unfortunately, only a few studies have reported about CD8+ T cells dominant atypical EAE model, which uses myelin basic protein (MBP) on TCR-transgenic 8.8 mice.[5] This is a major drawback for research on CD8+ T cells, particularly Tc17 cells, in EAE. In addition, studies on different types of EAE models—such as myelin antigen-specific TCR-transgenic mouse EAE models,[5, 61–64] spontaneous EAE models,[61, 65, 66] humanized mouse EAE models,[63, 67–69] and non-human primate EAE models[70–72]—might provide new insights into the pathogenesis of the disease.
Although Tc17 cells have cytokine profiles similar to those of Th17 cells, there are still some differences between the two cell types.[24] In the skin, Tc17 and Th17 cells co-express retinoic acid receptor-related orphan nuclear receptor gamma (RORγt) and GATA-3 only under tissue challenge.[73] RORγt regulates IL-6 and IL-23 production via signal transducer and activator of transcription 3 (STAT3), and regulates IL-17A, IL-17F, IL-23R, and IL-21 production via epigenetic mechanisms.[74] In the thymus, T cell factor (TCF)-1 suppresses Tc17 cells only at the double-positive stage and suppresses Th17 cells before the stage of CD4 co-receptor expression.[75] Interferon regulator factor 3 (IRF3), via RORγt, inhibits Tc17 cell development much more than Th17 development.[76] Moreover, DMF affects IL-17 production in murine and human Tc17 cells, but to a much lesser extent than in Th17 cells. Tc17 and Th17 cells may have different AKT/mTOR signaling requirements, as well as divergent metabolic requirements.[77] Finally, the plasticity and related functions of Tc17 and Th17 cells are different.[24] Th17 cells can switch to a Th1-like phenotype, resulting in a more pathogenic profile, which has been confirmed by studies on pathological inflammation in the CNS[78] and intestine during bacterial infection or colitis.[79] In contrast, Tc17 cells shift to a Tc1-like phenotype only during CNS autoimmunity.[77] This suggests that the functional specificity of Tc17 and Th 17 cells differs in terms of plasticity towards the type 1 phenotype.[80]
Murine Tc17 cells exhibit phenotypic similarities to Th17 cells; however, the cytokine profile of human Tc17 cells remains unclear. A report has shown that some cytokines—such as IFN-γ, tumor necrosis factor-alpha (TNF-α), IL-21, IL-22, GM-CSF, RORγt, and its subfamily homolog RORα—are co-expressed with IL-17 in cultured Tc17 cells.[47] RORγt is a key factor in the development, maintenance, and function of IL-17-producing cells, and plays a role in regulating thymopoiesis. Clinical trials of several RORγ inhibitors in patients are ongoing. Other TFs, including STAT3, IRF3, and IRF4 also promote Tc17 cell differentiation.[47, 81]
Different subsets of CD8+ T cells are sometimes not stable, and differ between Tc1 and Tc17 phenotypes—a phenomenon called “plasticity of Tc17 cells”.[24] For example, the lack of CTLA-4 drives Tc17 cells to downregulate RORγt and IL-17 expression and shift towards Tc1.[82] In contrast, in a mouse model, Tc17 cells expressed a stable phenotype and mediated protection against fungi and bacteria.[83–85] Tc17 cells can also switch to Tc2 phenotype by increasing IL-5 and IL-13 cytokine production.[73] Thus, the plasticity of Tc17 cells can be an intriguing and promising target for future studies. TGF-β inhibits various functions of Tc1 cells, however, in presence of IL-6, TGF-β induces Tc17 cells.[86] In HKx31 (H3N2) influenza A virus-infected Il2−/−, Il2ra−/−, Il12−/−, Blimp1GFP, and Blimp1fl/fl mice, Tc17 cells were non-cytotoxic and downregulated the expression of T-bet and Eomes, compared with Tc1 cells.[48]
In summary, our results suggest that autoreactive Tc17 cells have a lower encephalitogenic function, but are unique and independent of the pathogenicity of EAE compared to their Th17 counterparts and the Tc1 subset. The pathogenesis of MS is still unclear, and many factors—such as different cells and their subsets, and different kinds of EAE models—need to be investigated to determine the underlying mechanisms of MS.