Abstract EN:

Multiple Sclerosis (MS) is a multifocal demyelinating disease with progressive neurodegeneration caused by an autoimmune response to self-antigens. Clinical symptoms vary based on the site of neurologic lesions and often correlate with invasion of inflammatory cells across the blood-brain barrier with resulting demyelination and oedema. Development of MS is the result of genetic predisposition, post-genomic modifications and environmental triggers. The antigenic targets involved in MS pathogenesis are unknown although molecular mimicry between viral and neural antigens has been reported. Shaping of the TCR repertoire in the thymus through negative selection might be a major early step determining susceptibility to MS. The paradigm that each T cell can recognize a continuum of related ligands, defined as T-cell polyspecificity, implies that multiple antigens can tolerize T cells specific for a given self-antigen. We previously showed in C57BL/6 mice that part of the CD4 T-cell repertoire specific for the encephalitogenic antigen myelin oligodendrocyte glycoprotein (MOG) 35-55 also recognizes the neuronal antigen neurofilament medium (NF-M) 15-35. Such bi-specific autoreactive CD4 T cells are frequent, produce inflammatory cytokines after stimulation and are essential to progression of experimental autoimmune encephalomyelitis (EAE), an animal model of MS. Since T cells recognizing two self-antigens would be expected to be more likely to be tolerized, we were prompted to study how tolerance to bi-specific T cells is imparted. We found that similar to MOG, NF-M is expressed in the thymus by medullary thymic epithelial cells, i.e. in a potentially tolerogenic setting. Nevertheless, we found that the frequency, pro-inflammatory phenotype, and capacity to transfer experimental encephalomyelitis (EAE) of MOG35-55-reactive CD4 T cells were increased in MOG-deficient but not in NF-M-deficient mice. This observation led us to investigate the efficiency of NF-M15-35 presentation by antigen presenting cells, which we found of short duration, suggesting unstable MHC class II binding. Consistent with this hypothesis, introducing an MHC-anchoring residue into NF-M15-35 (NF-M15-35T20Y) increased its immunogenicity, and activated a repertoire able to induce EAE. Our results show that in C57BL/6 mice the expression of MOG, but not of NF-M, restrains bi-specific autoreactive CD4 T cells, but not to a level sufficient as to prevent EAE induction. Besides, we investigated the contribution of genetic factors implicated in MS physiopathology. Genome wide association studies (GWAS) identified more than 150 variants associated with MS susceptibility. Each variant alone is responsible for only a small amount of the risk but it is possible to cluster variants depending on which biological process they participate in, notably a signaling cascade. Several variants are located within the IL12/23 pathway (IL12B, IL23R, STAT3, JAK3, and TYK2) and the cumulative effect of several variants in a same pathway might result in dysfunctional aberrant signalling. To study the functional effect of such accumulation we biobanked PBMCs from heathly controls bearing none, one or several susceptibility variants. We realized an immunophenotyping of the different immune subsets by mass cytometry (CyTOF). As GWAS only characterized common variants with a frequency above 5% we realized a genetic association study focussing on rare (frequency < 5%) coding variants and identified a protective variant located in the HDAC7 locus, a major epigenetic regulator of CD4 T cells function. The protective variant HDAC7 R166H impacts Treg function by increasing IL-10 secretion and Foxp3 expression. To explore the underlying mechanisms of this tolerogenic phenotype, we realized a RNAseq analysis of Treg expressing the protective and the risk variant of HDAC7.