Distinct Myeloid Cell Subsets with Divergent Functions in Central Nervous System Autoimmunity

Multiple sclerosis (MS) is a disabling disorder of the central nervous system (CNS) characterized by motor, visual, and sensory deficits. It is believed to be an autoimmune process in which immune cells infiltrate the brain and spinal cord leading to demyelination and cell death. Myeloid cells, such as macrophages and dendritic cells, are a major component of the cellular infiltrates in the CNS. Initial studies described these myeloid cells as pathogenic, but recent evidence in other systems has revealed that myeloid cells may also be anti-inflammatory and even pro-regenerative. Little is known about the heterogeneity of myeloid cells in MS. We set out to investigate the diverse phenotypes and functional properties of myeloid cells in CNS autoimmunity. Pro-inflammatory and anti-inflammatory myeloid cells have been termed classically or alternatively activated myeloid cells (CAMC, AAMC), respectively. We began by identifying markers of CAMC and AAMC in MS lesions. We observed a spectrum of activation such that CAMC were present at the active lesion rim while AAMC were predominant in the relatively quiescent lesion core. Using the animal model experimental autoimmune encephalomyelitis, we identified similar CAMC and AAMC in the CNS with the markers iNOS and Arg1, respectively. Myeloid cell phenotypes evolved such that iNOS+ CAMC were present early in disease while Arg1+ AAMC were present at peak and late disease. We further demonstrated that individual cells were plastic and shifted from expression of iNOS to Arg1 over the course of disease. One function of myeloid cells is the activation of T cells. Activation is critical for the initiation of disease. We compared the functional properties of Arg1+ and Arg1- cells and found that Arg1+ cells were deficient in activating CD4+ T cells compared with Arg1- cells. Investigation of the mechanism of this defect led to the elucidation of additional myeloid cell subsets. We identified that the dendritic cell population was heterogenous, not only in activation state or phenotype, but also in developmental lineage. We identified cells of both the conventional (cDC) and monocyte-derived dendritic cell (moDC) lineages in the CNS during EAE and demonstrated functional specialization between these subsets. iNOS and Arg1 expression were restricted to the moDC subset, and this subset as a whole was deficient in T cell activation but proficient in phagocytosis. Conversely, cDC were highly efficient in activating T cells but weak phagocytes. It is not clear which antigen presenting cell activates T cells to initiate disease. We observed that cDC were present in the naïve brain and meninges and that these cells from the naïve CNS could activate myelin-specific T cells. We further demonstrated that specific depletion of cDC reduces the incidence of EAE. This identifies cDC as critical antigen presenting cells for the initiation of CNS autoimmunity. Together, these studies clarify the diversity of myeloid cells and demonstrate functional differences with implications for clinical disease. The results presented herein provide a foundation for the development of myeloid cell-targeted therapies for MS.PHDImmunology PhDUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttps://deepblue.lib.umich.edu/bitstream/2027.42/155089/1/dagiles_1.pd