A Th17/Lymphotoxin axis enables cytokine production by brain meningeal stromal cells and propagates de novo CNS-intrinsic T cell responses during inflammation

Abstract

Background: What does it require to be at risk for multiple sclerosis (MS)? To investigate the mechanisms involved we apply experimental autoimmune encephalomyelitis (EAE) as a model of MS in rats. By cross-breeding the EAE-susceptible DA strain and the EAE-resistant PVG strain we identified several quantitative trait loci (QTL) governing the susceptibility to EAE. Approach and results: The R11R6 congenic strain has an EAEsusceptible DA background, but comprises a gene region on chromosome 4 from the EAE-resistant PVG and is itself resistant to active EAE immunization, but not passive transfer. The identified region could contain potentially eight genes encoding for C-type lectin (CLEC) family members: Clec4a, Clec4a2, Clec4a3, Clec4a1, Clec4b2, Clec4n, Cle4d and Clec4e. Using backcrossing strategies coupled with SNP genotyping we narrowed the difference to four CLECs from PVG origin: Clec4b2, Clec4n, Clec4d and Clec4e. Expression of these and other CLECs is associated to myeloid cells. In vitro-generated bone marrow-derived dendritic cells (BMDCs), monocyte-derived DCs (Mo-DCs) or microglia from R11R6 exhibited comparatively lower expression of Clec4d and Clec4e. Surprisingly, despite having dramatic resistance to EAE induction, the R11R6 congenic showed no defect in immunization per se: We observed neither differences in lymph node cell counts, nor frequencies of T-cells, B-cells or DCs. Ex-vivo re-stimulation with the immunogen (Myelin Oligodendrocyte Glycoprotein, MOG) yielded equivalent proliferation between DA and R11R6.We then assessed the inflammatory response in the CNS before onset, at onset and at the peak of the first relapse. We found few differences in the frequency of microglia, lymphocytes and granulocytes/macrophages (Gran/Ma) cells in the CNS. R11R6 rats had significantly more lymphocytes in the brain at onset and significantly less Gran/Ma at peak both in the brain and spinal cord. Therewas no significant difference in the frequency of CD3 T cells in the CNS. However, lymphocytes from R11R6 strain proliferated less and infiltrating CD4CD3 lymphocytes were less frequent in the spinal cord at peak. They also produced drastically less IL-17 at onset and peak in both brain and spinal cord. Furthermore, granulocyte-marker positive cells from the R11R6 strain expressed significantly lower levels of MHC II. Conclusions: Though R11R6 rat have no defect in mounting an immune response and generate CNS-reactive T-cells, they exhibit a dramatic confinement of the subsequent inflammatory response. As the only genetic difference in this congenic strain is the origin and expression of CLEC genes, we conclude that the expression of CLECs on myeloid cells governs the susceptibility to EAE in rats. As CLEC receptors are important for sensing pathogens and tissue damage, we assume their regulatory involvement in shaping the extent of inflammation during EAE. Currently, we are investigating the exact mechanisms in further detail.