PROJECT SUMMARY Autosomal dominant optic atrophy (DOA) is the most commonly diagnosed inherited optic neuropathy. Mutations in the OPA1 gene, which encodes a mitochondrial dynamin like GTPase, account for 60-70% of all DOA cases. Although OPA1 is expressed throughout the body, secondary to dysfunctional mitochondria, patients with DOA associated OPA1 mutations exhibit loss of retinal ganglion cells (RGCs) specifically. Despite intensive study and the availability of mouse models of DOA, critical questions regarding how OPA1 mutations lead to specific loss of human RGCs in DOA patients remain unanswered and there are currently no treatments for this condition. A human RGC model would greatly facilitate the study of disease mechanisms as well as drug discovery efforts. Obtaining RGCs from DOA patient samples is not feasible, however, due to the rarity of DOA donor eyes, the sparsity of RGCs in the human retina, and poor RGC viability upon isolation. The proposed studies will address this unmet need by developing and characterizing in detail three human pluripotent stem cell (hPSC) models of DOA that track disease progress from stem cell differentiation to RGC degeneration. An important feature of our stem cell models is that they make use of techniques that produce large quantities of highly purified RGCs that display long term survival, features important for biochemical, functional, morphological, and transcriptomic analyses. We combined this protocol with CRISPR/Cas9 genome-editing to model OPA1 haploinsufficiency and developed an inducible CRISPR inference (CRISPRi) DOA model to control the timing of OPA1 loss of function. We propose to use these two complementary models together with RGCs derived from patient iPSCs to study the role of OPA1 in RGC differentiation and degeneration. In the future, these well-characterized stem cell models could be used for large-sale functional genomics studies and high throughput screening for neuroprotective and regenerative agents.