Genetic, Cellular and Molecular Mechanisms in Autoimmunity to Retina

National Eye Institute

Investigators

Linked publications & trials

Abstract

Dendritic cells (DC) are important Ag presenting cells (APC) in induction of immune and autoimmune responses. We developed a new model EAU model where disease is induced with in vitro-matured, Ag pulsed DC obtained from mice injected hydrodynamically with an expression plasmid of Flt-3L, that expands these cells. When matured (by culture with LPS and anti-CD40), pulsed with a uveitogenic IRBP peptide and injected into naive syngeneic mice, these DC elicit ocular inflammation. Interestingly, DC-induced EAU differs from "traditional" EAU induced by immunization in terms of clinical manifestations, the type of effector response elicited and the nature of the inflammatory infiltrate in the eye (granulocytic vs. monocytic). This shows (i) that distinct forms of EAU follow exposure to the same Ag presented to the immune system in a different context, (ii) provide an alternative EAU model representing types of uveitis not well represented by the traditional EAU model, and (iii) help explain the heterogeneity of clinical uveitis in the face of recognition directed predominantly at the same Ag (retinal arrestin = S-Ag) (5).[unreadable] [unreadable] Compared to classical EAU in wild type mice, a distinct type of disease histologically (more severe and granulocyte dominated) and immunologically (Th2/Th17 profile as opposed to Th1/Th17 profile) is induced by uveitogenic immunization of IFN-gamma deficient mice. We examined the expression of chemokines in uveitic eyes and found that Th1-associated chemokines, CXCL10, CXCL9, CCL5, CXCL11 were highly expressed in WT mice, whereas Th2-associated chemokines CCL11, CCL17, CCL1 and the Th17-associated chemokines CCL22 and CXCL2 were upregulated in GKO mice. These chemokine profiles may explain the differential recruitment of leukocyte subpopulation into uveitic eyes in the absence, vs. presence, of an IFN-gamma response. Importantly, as indicated by depletion studies, we found that granulocytes are a necessary cellular element for pathogenesis in both GKO and WT genotypes (4).[unreadable] [unreadable] AIRE deficient mice, which lack expression of retinal (and other) Ags in the thymus and have deficient central tolerance, develop spontaneous EAU as they age. We collaborated with M. Anderson of UCSD and G. Liou of the Medical College of Georgia to examine the impact of lack of central tolerance on ocular autoimmunity. Using AIRE deficient mice and IRBP KO mice (developed by G. Liou), we demonstrated that IRBP is the only Ag targeted, and that in the absence of IRBP no other Ag serves as target for EAU, even though central tolerance to other retinal Ags is also deficient as well. The preference of mice for IRBP as an auto-Ag may help explain the finding that uveitis patients preferentially respond to retinal arrestin (S-Ag) rather than to other retinal Ags (2).[unreadable] [unreadable] Our previous studies suggested that EAU susceptibility may be enhanced because peripheral tolerance to antigens (Ags) from the retina is deficient, due to their relative sequestration as part of ocular immune privilege. We hypothesized that revoking the immune privileged status of a uveitogenic Ag by expressing it systemically will protect from EAU. We therefore expressed IRBP in extraocular tissues by using a hydrodynamic IV delivery of a DNA expression plasmid encoding a uveitogenic IRBP fragment. Mice given this protocol expressed IRBP in the liver within 8 h of vaccination and were protected from EAU. Regulatory cells that are Ag specific, anergic and suppress by contact were found to be an important part of the mechanism of protection (3).[unreadable] [unreadable] We are dissecting the respective roles in EAU of the Th1 and the Th17 pathway, driven by IL-12 and IL-23, respectively. We used monoclonal antibodies and mutant mice deficient in various components of these pathways. We found that autoimmunity to retina can be either Th17 or Th1 driven. We further dissected the conditions that determine which effector type will predominate. The conditions that lead to a Th17 driven disease are strong TLR stimulation (CFA), but when disease is induced not in the context of CFA, Th1 is dominant and Th17 is either insufficient or dispensable. (manuscript in revision).[unreadable] [unreadable] As part of our studies on the Th17 pathway we found that mouse as well as human NKT cells make a very rapid IL-17 response upon ligation of TCR and IL-23 Receptor. The TCR and IL-23R pathways are independent and additive. NKT have constitutive expression of IL-23R and RORgt and their IL-17 response is independent of IL-6 and probably independent of IL-21. Unlike IFN-gamma and IL-4, IL-17 is made by a minority of NKT cells, but all known NKT[unreadable] subtypes are able to make an IL-17 response. The relevance, if any, of this innate IL-17 response to autoimmunity and to EAU, is unknown (manuscript in revision). [unreadable] [unreadable] NKT cells can, however, inhibit induction of EAU when pharmacologically triggered in vivo with their invariant TCR ligand alpha-GalCer at the time of uveitogenic immunization. This is doubly interesting in view of the reported role of NKT as participants in the process of eye-induced tolerance, contributing to ocular immune privilege. In contrast to other autoimmune disease models, which by and large connected the protective effect of NKT to IL-4 and induction of a Th2 response, we observe a dependence on NKT-produced IFN-gamma. This was causally related to inhibition down the line of the adaptive Th1 and Th17 responses, and constitutes the first report connecting the protective effect of NKT to inhibition of the IL-17 effector response (manuscript in revision).[unreadable] [unreadable] Continuing our efforts to better characterize the mouse EAU model, we were able to identify several new pathogenic peptides for the C57BL/6 (H-2b) and B10.RIII (H-2r) strains. Some of these represent specificities recognized by IRBP KO mice, that are reduced or eliminated through central tolerance in WT mice of the respective strains. Many of the pathogenic epitopes represented either overlapping peptides, or homologous sequences in different repeats. These findings will be useful for immunological studies that require knowledge of multiple epitopes and will help to understand the pathogenic potential of IRBP as an auto-antigen (manuscript in revision).[unreadable] [unreadable] Within the scope of our studies on genetic susceptibility to uveitis in the rat EAU model using an F2 cross of EAU-susceptible and resistant strains, we have defined a number of genes that may be involved in susceptibility. The methods use combined approaches of classical genetics, microarray analysis of expressed genes and in silico analysis (the latter in collaboration with Roche Pharmaceuticals). Some susceptibility regions are unique, and could be mapped to specific genes using microarray and in silico approaches, and some are shared with other autoimmune and non-autoimmune diseases, among them EAE, arthritis and type 2 diabetes. This indicates that uveitis shares common pathways with other autoimmune and inflammatory diseases, and may suggest use of common therapeutic approaches (2 manuscripts submitted).[unreadable] [unreadable] Ocular immune privilege, composed in part of inhibition of inflammatory processes in the eye, plays an important role in uveitis. We have previously shown that retinal glial Muller cells inhibit proliferation of T cells in an Ag-independent fashion by a contact dependent mechanism. Using primary cultures of murine Muller cells and genetically manipulated mice, we have identified thrombospondin-1 and TGF-beta as partly responsible for the suppressive effect of Muller cells. T cells that have interacted with Muller cells may themselves become regulatory and able to themselves inhibit activation and proliferation of fresh T cells (manuscript in preparation).