Age-related macular degeneration (AMD) is the leading cause of vision loss in the US. It affects more than 9 million people and these numbers are likely to reach epidemic proportions in the coming decade (Rein et al., 2009 Arch. Ophthal.). It is thought that the disease originates in a monolayer of pigmented epithelium cells located in the back of the eye - the retinal pigment epithelium (RPE) (Datta et al., 2017 Prog Retin Eye Res; Ambati and Fowler 2012 Neuron; Swaroop et al 2009 Ann. Rev. Genomics Hum. Genet.) RPE performs multiple functions that are fundamentally important for the health and integrity of photoreceptors, including regulation of nutrient and metabolite flow, regeneration of visual pigment, and phagocytosis of shed photoreceptor outer segments. Consequently, RPE dysfunction or atrophy leads to photoreceptor cell death and vision loss, as seen in AMD patients (Bharti et al., 2006 Pig. Cell Mel. Res.; Bharti et al 2010 Pig. Cell Mel. Res.; Berber et al., 2017 Mol. Diagn. Ther.) Genome-wide association studies have identified multiple (approximately 40) risk alleles associated with AMD and these risk-alleles affect different RPE signaling pathways whose dysregulation cause AMD progression (Fritsche et. al., Nat. Gen, 2016).  Thus we seek a more precise description of AMD disease pathobiology by using AMD genetic risk factors as sentinels for specific RPE signaling pathways and for changes in RPE physiology.  We propose this approach as a guide to the development of personalized therapies for AMD.


We will utilize an existing repository of AMD patient clinical data that includes 5-10 years of extensive clinical imaging and medical history, combined with genomic data to generate a comprehensive set of iPSC-derived RPE for disease pathway analysis and genotype-phenotype association. In vitro analysis of iPSC-RPE can be used to link RPE signaling pathways to AMD risk alleles and thus to various stages of AMD. This allows making a direct link between RPE signaling pathways, AMD risk alleles, and AMD phenotype. To help accomplish this goal we have established a collaboration with the New York Stem Cell Foundation (NYSCF) who will generate 100 AMD iPSC-derived RPE and, in some few cases, twin controls and provide storage and distribution of these cell lines to the entire scientific community.  

Specific example
Recent literature (Pandey et al, 2017 Nature; Calippe et al., 2017 Immunity) links non-canonical activation of complement (specifically C5a and C3a – complement anaphylatoxins) to AMD initiation. Recently, in vitro activation of non-canonical complement in RPE cultures has been linked to changes in RPE structure, increased “drusen” formation, reduced autophagy, increased activation of inflammasome and inflammatory cytokines, and increased activation of NFƙb signaling pathway (through CD88 and TLR4 receptors) (Pilgrim et al., 2017 IOVS; (Celkova, et al., 2015 J. Clin. Med.). Downstream activation of miR-155, under circadian control, as well as RPE physiology changes have been determined (NEI IRP data). Based on these findings, high-throughput assays will be performed on iPSC-RPE.  To mimic the initiation of AMD pathogenesis in vitro we will treat iPSC-RPE monolayers with C5a/C3a competent human serum and quantitatively analyze:

  • RPE shape, pigmentation, and autofluorescent changes using live-imaging (software developed at NEI IRP). These findings will be correlated to in vivo changes in RPE shape and pigmentation and analyzed longitudinally in patients using adaptive optics.
  • Formation of drusen-like particles and their polarized secretion by iPSC-derived RPE cells (Pilgrim et al., 2017, IOVS) to determine a direct relationship between AMD genetics and drusen or pseudoreticular drusen formation.
  • Specific intracellular autophagic pathways using fluorescent reporters (Barmada et al., 2014) to identify potentially druggable targets to slow down disease initiation or progression (The HD iPSC Consortium, Nat. Neuroscience, epub March 2017 – Analysis of iPSC-neurons derived from Huntington disease patients demonstrates that mutant huntingtin impairs neurodevelopmental pathways that could disrupt synaptic homeostasis and increase vulnerability to the pathologic consequence of expanded polyglutamine repeats over time. Some of these phenotypes could be reversed by the treatment of isoxazole-9).
  • The polarized secretion by human RPE inflammatory cytokines - IL6, Il1-beta, IL18, and VEGF as well as the expression of miRNAs (eg miRs-155, 204) that have recently been shown to activate RPE inflammasome activity and AMD pathophysiology (Shi et al., 2008, IOVS; Wang et al., 2010 Hum. Gen.; Ambati et al., 2013 Nat. Rev. Immunology; Campbell, 2014 Adv. Exp. Med. Biol; Celkova, et al., 2015 J. Clin. Med; Berber et al., 2017 Mol. Diagn. Ther.).