RIG-I signaling has been previously implicated as a driver of inflammation to the retinal pigment epithelium (RPE) during age-related macular degeneration (AMD). Double-stranded RNA (dsRNA) is known to initiate RIG-I signaling and lead to a type I interferon response. We show through shRNA knockdown that RIG-I is essential for initiating an interferon response in iPS-RPE in response to both synthetic dsRNA-mimetic 3p-hpRNA and the double-stranded retrotransposable element Alu . Analysis of human tissue from patients suffering from AMD show accumulation of dsRNA, peaking at the geographic atrophy (GA) stage. Using a new murine model of 3p-hpRNA subretinal challenge to RPE cells, we confirmed that accumulation of dsRNA initiates a type I interferon response, as well as RPE and photoreceptor degeneration. Although RPE response to synthetic dsRNA was acute, extensive leukocyte migration was observed. The results from this study verify the importance of RIG-I signaling in regulating inflammation in the subretinal space and implicates dsRNA accumulation as a driver of AMD pathogenesis.
Age-related macular degeneration (AMD) is typified by the formation of lipid-rich drusen under the retinal pigment epithelium (RPE) layer. Apolipoprotein E (APOE) is a known genetic risk factor for AMD and a substantial component of drusen, however, the mechanism by which APOE variants contribute to AMD pathology remains unclear. APOE is the primary cholesterol and lipid transport protein of the central nervous system, as well as a component circulating lipoproteins. To better understand how APOE-dependent lipid transport may impact AMD risk, we generated isogenic APOE iPS-RPE cells expressing each of the common human APOE isoforms, as well as an APOE knockout line. APOE knockout cells showed significant morphological and barrier function deficits, suggesting that APOE is essential for RPE health. Furthermore, we observed that APOE abundance is isoform-dependent in RPE cells and that lipid transport is deficient in APOE knockout RPE cells, as well as in RPE cells expressing APOE2, a variant associated with higher risk of AMD. Contrastingly, cells expressing APOE4 seem to respond strongly to lipid challenges by upregulating APOE to support efficient lipid transport. Our results suggest that disease associated APOE variants may impact lipid transport in RPE, contributing to the formation of drusen and impairing cellular function.### Competing Interest StatementAll authors are current or former employees of Novartis Biomedical Research
The death of retinal pigment epithelial (RPE) cells and photoreceptors (PR) is a hallmark of the progression of several degenerative ocular disorders. The precise molecular driver(s) behind RPE and PR cell death, however, remains unknown. Recent studies have suggested the involvement of ATP and purinergic signaling in the progression of age-related macular degeneration (AMD) and retinal degeneration. We have discovered that RPE cells release ATP when subjected to stress, which in turn exacerbates stress-related signaling via purinergic receptors that ultimately results in degeneration. Our findings demonstrate that blocking P2Y purinergic receptors using suramin can effectively prevent toxin-induced RPE cell death and dysfunction in vitro . Furthermore, we show efficacy of suramin in preventing photoreceptor degeneration in vivo using the RHO-P23H zebrafish model. This study reinforces the involvement of ATP and purinergic signaling in maintaining retinal health, and highlights the potential of purinergic receptor antagonism as a therapeutic strategy for retinal degeneration.### Competing Interest StatementAll authors for this manuscript, during the time of this study, are employees of Novartis Pharma AG, Cambridge, US.
Increasing evidence suggests that chronic inflammation plays an important role in the pathogenesis of age-related macular degeneration (AMD); however, the precise pathogenic stressors and sensors, and their impact on disease progression remain unclear. Several studies have demonstrated that type I interferon (IFN) response is activated in the retinal pigment epithelium (RPE) of AMD patients. Previously, we demonstrated that human RPE cells can initiate RNA-mediated type I IFN responses through RIG-I, yet are unable to directly sense and respond to DNA. In this study, we utilized a co-culture system combining primary human macrophage and iPS-derived RPE to study how each cell type responds to nucleic acids challenges and their effect on RPE barrier function in a homotypic and heterotypic manner. We find that DNA-induced macrophage activation induces an IFN response in the RPE, and compromises RPE barrier function via tight-junction remodeling. Investigation of the secreted cytokines responsible for RPE dysfunction following DNA-induced macrophages activation indicates that neutralization of macrophage-secreted TNFα, but not IFNβ, is sufficient to rescue RPE morphology and barrier function. Our data reveals a novel mechanism of intercellular communication by which DNA induces RPE dysfunction via macrophage-secreted TNFa, highlighting the complexity and potential pathological relevance of RPE and macrophage interactions.
Age-related macular degeneration (AMD) is one of the most common causes of visual impairment in the elderly, with a complex and still poorly understood etiology. Whole-genome association studies have discovered 34 genomic regions associated with AMD. However, the genes and cognate proteins that mediate the risk, are largely unknown. In the current study, we integrate levels of 4782 human serum proteins with all genetic risk loci for AMD in a large population-based study of the elderly, revealing many proteins and pathways linked to the disease. Serum proteins are also found to reflect AMD severity independent of genetics and predict progression from early to advanced AMD after five years in this population. A two-sample Mendelian randomization study identifies several proteins that are causally related to the disease and are directionally consistent with the observational estimates. In this work, we present a robust and unique framework for elucidating the pathobiology of AMD.