Genetic, Cellular and Molecular Mechanisms in Autoimmunity to Retina

National Eye Institute

Investigators

Linked publications & trials

Abstract

IMMUNITY AND AUTOIMMUNITY INVOLVING THE NEUORORETINA AND THE ROLE OF THE GUT MICROBIOME For most of these studies we utilize mouse models of autoimmune uveitis developed in our laboratory, (i) experimental autoimmune uveitis (EAU) induced in mice by immunization with the retinal antigen (Ag) IRBP, and (ii) spontaneous uveitis in R161H mice engineered to express a transgenic T cell receptor (TCR) specific for IRBP peptide 161-180. 1) Laquinimod is an aryl hydrocarbon receptor (AHR) agonist. We found that it has strong protective effects in the induced model of EAU by modulating multiple inflammatory and anti-inflammatory cytokines. Laquinimod strongly inhibited EAU in the setting of active immunization and also had moderate effects in the setting of adoptive transfer. Data indicated that the major effect was exerted via myeloid cells, possibly by inhibiting antigen presentation and/or effector function during inflammation. Notably, it had inhibitory effect on immune response to recall antigens in human PBMC, suggesting potential clinical utility (B. Xu et al, in preparation) 2) Ocular immune privilege: We have previously shown that nave retina-specific T cells in the eye are primed locally, and after 7 days a proportion of them convert into Foxp3+ T-regulatory cells (Tregs). However, it was not clear what differentiation pathway was adopted by cells that were not converted to Tregs. Using single-cell transcriptomics of cells of fresh retina-specific lymphocytes injected into the healthy eye found that although all cells recognized their antigen, inflammation was not induced. About 30% of the cells adopted the profile of T-regulatory cells, actively secreting anti-inflammatory mediators TGF- and IL-10, whereas the rest were divided along a spectrum expressing stereotypic gene induction profiles consistent with non-pathogenic effector Th17 cells, exhausted, and anergic cells. IFN--expressing cells were notably absent. RNA velocity and trajectory indicated that these cell fates were largely independent and developed in parallel, rather than in a sequential fashion. This is the first molecular representation of the immune privilege phenomenon in the living eye and promises to shed new light how the eye controls inflammation. We are also examining the transcriptome of eye-induced Tregs and how it differs from Tregs primed at other tissue sites or in vitro, (Peng et al., in preparation). 3) Commensal flora is necessary for development of spontaneous uveitis, in part due to molecular mimicry of IRBP161-180. However, innate mechanisms also play a role, as evidenced by the finding that antibiotics (ABX) starting a week before immunization for EAU (where the antigenic stimulus is provided), afforded measurable protection. That protection, however, was temporary and disappeared on prolonged ABX. Results suggest that prolonged ABX depletes intraepithelial lymphocytes (IEL), which are known to be microbiota-dependent, We hypothesize that IEL may have an inhibitory role in uveitis, an effect which is reversed after long-term ABX due to IEL depletion. (Salvador R et al., in revision). 4) Human gut flora and uveitis: We reconstituted germ free R161H mice with gut microbes from 3 healthy human donors and established 3 humanized flora mouse lines Results showed that human gut flora supports development of spontaneous uveitis. Mice adopted only a subset of the human flora, but preserved the microbial footprint of the donor. Mice with high disease scores appeared to harbor more diverse gut flora than those with low scores. Verrucomicrobia, Actinobacteria, and Fusobacteria were enriched in mice with high disease scores, whereas Firmicutes appeared enriched in mice with low disease scores and were associated with higher content of short chain fatty acids in the feces. In-vivo association studies are underway to examine the ability of candidate microbes to modulate autoimmune uveitis. This may instruct interventional strategies as an approach to regulating uveitis through directed manipulation of the microbiome (Horai, Zhou, Murphy et al, in preparation). 5) in collaboration with the group of Feng Lin, Cleveland Clinic, demonstrated that CDCP1 regulates retinal pigmented epithelial barrier integrity for the development of experimental autoimmune uveitis (ref.2) 6) Collaborated with the group of Dan Kastner and colleagues on the consequence of the Alpk1 mutation on susceptibility to EAU, since this mutation results in autoinflammatory disease, including uveitis, in humans. Demonstrated that the human mutant alpk1 knock-in mice, otherwise unperturbed, do not develop spontaneous ocular inflammation (ref.4) MUCOSAL IMMUNE RESPONSES TO COMMENSALS AT THE OCULAR SURFACE Mucosal sites such as the intestine, oral cavity, nasopharynx, and vagina all have associated commensal flora. The surface of the eye is also a mucosal site, but presence of ocular surface microbiome was contentious. Previously, we isolated and purified a candidate ocular commensal, Corynebacterium mastitidis (C. mast). This organism elicits a commensal-specific IL-17 response from T cells in the ocular mucosa, tuning local host defense to afford protection from infectious pathogenic organisms. 1) We are examining the molecular sensors of C. mast in T cells. Data suggest that Vg4 T cells respond to C. mast mainly via their TCR, whereas Vg6 T cells respond through innate receptors such as TLR2, and are highly dependent on IL-1. TLR2 appears to be required not only in DC (where it contributes to IL-1 induction), but also in T cells. The intrinsic TLR2 stimulation in V6 T cells is upregulates the transcription factor IAb, which in turn is required for metabolic support of IL-17A production in these cells through fatty acid oxidation. This study highlights the importance of intrinsic TLR2 signaling in driving the production of IL-17A in microbiome-specific gd T cells. (Z. Zhu & X. Xu et al., and X. Xu et al., in preparation). 2) We also examined commensal-elicited ocular surface responses in an immunologically abnormal host, mouse as well as human. In collaboration with the group of Dr. Warren Strober (NIAID) who developed knock-in mice expressing a gain-of-function mutation in the NLRP3 inflammasome gene, and Dr. Raphaela Goldbach-Mansky who treats patients with NLRP3 inflammasome related diseases, we obtained evidence to suggest that an ocular surface commensal can elicit ocular surface inflammation in an immunologically perturbed host, thus acting as a pathobiont. We are characterizing the local ocular immune response and the ocular surface microbiome in the mouse model and in patients with NLRP3 inflammasome mutations at the single-cell level (Siak, Mattapallil, St. Leger et al., in preparation). This will have implications for understanding and treatment of the characteristic ocular inflammation in patients with NLRP3-related diseases. (Ref. 5). 3) A collaborative study with the lab of Roland Laulau, INSERM, France, in an in vitro co-culture system identified perforin as an essential molecule for killing of blood brain barrier endothelial cells (BBB-EC) by CD8 T cells. We then found that short-term pharmacologic inhibition of perforin commencing after disease onset restored motor function and inhibited the neuropathology. Perforin inhibition resulted in preserved BBB-EC viability, maintenance of the BBB, and reduced CD8 T-cell accumulation in the brain and retina. 4) In collaboration with S. Fleiszig and D Evans of UC Berkeley, we demonstrated that Contact lens-induced corneal parainflammation involves Ly6G+ cell infiltration and requires IL-17A and T cells (ref. 6).