Disease mechanism and therapies for retinal degeneration

National Eye Institute

Investigators

Linked publications & trials

Abstract

1. Disease modeling and therapeutic investigation in retinal organoid models Usher syndrome is the most common cause of combined deafness and blindness with retinal degeneration onset usually during childhood in Usher syndrome type 1 (USH1). Due to limited vision-related phenotype in existing USH1 rodent models, the underlying mechanisms of retinal degeneration in this syndrome remain largely unclear. To fill this gap in knowledge, we established an in vitro model for Usher type 1C syndrome by differentiating induced pluripotent stem cells from patients carrying USH1C mutations into retinal organoids (ROs). Patient-derived ROs exhibited compromised long-term maintenance of photoreceptor outer segment-like structures, accompanied by decreased expression of photoreceptor-specific genes at a late stage. Consistent with recent reports in the literature, USH1C was found to be primarily expressed by Mller glia in the human retina and human ROs, suggesting that compromised maintenance of photoreceptor cells in patient organoids is an indirect phenotype caused by Mller glia dysfunction. Consistent with this, Mller cell-related phenotypes, including mislocalization of Mller glia nuclei and upregulation of gliosis markers, were observed in USH1C patient ROs. AAV-mediated expression of USH1C specifically in Mller cells partially restored pathological phenotype in patient organoids, confirming an important function of Harmonin protein in this cell type and providing a promising proof of concept for future clinical applications (Regent et al., manuscript in preparation). 2. Improving the yield, reproducibility and quality of retinal organoid generation from iPSCs continue to receive considerable attention in our group. And we made two notable advances. 2a. Conventional retinal organoid generation from iPSCs requires an obligatory initial step of embryoid body formation. This costs extra time, cell numbers and skills, with the inevitable occasional failures. To further simplify this process, we developed a widely applicable and highly efficient differentiation protocol for the generation of the retinal organoids. By bypassing embryoid body formation, incorporating nicotinamide treatment, and starting differentiation from properly inter-spaced colonies, our method streamlines the process and accelerates differentiation by at least one week. Subsequent long-term cultures of the generated organoids yield large-scale production with stratified retinal organoids containing the seven major retinal cell types, as well as extended inner and outer segment-like structures. Having been validated in more than a dozen iPSC lines, this protocol has been shown to be robust and generally applicable. When combined with our earlier published improvements, our procedure offers an efficient and reliable alternative for researchers to generate retinal organoids, without the need for extensive specialized expertise. A manuscript is in preparation (Hwang et al., in preparation). 2b. Most current hPSC differentiation methods rely on static culture (SC) conditions, especially at the initial stage of retinal differentiation, where stem cells and emerging neural spheres and optic vesicles remain adhered to the bottom of culture vessels. During static culture, the oxygen consumption rate of growing tissues exceeds the rate at which atmospheric oxygen can diffuse through media volume leading to continuous decline in oxygen tension around the cells. To further exacerbate this situation, optic vesicles are pseudostratified such that cell bodies localized to the luminal side experience even greater starvation for oxygen. Insufficient access to oxygen is known to disrupt cellular fate determination and inhibit the formation of complex neural structures. To address this limitation, we have developed a novel, 3D-printed stirred bioreactor (SBR), which can be easily implemented in most labs, requiring only a 100mm Petri dish, magnetic stir plate, and a small stir bar. We evaluated our SBR using an adherent and embryoid body culture protocol with multiple iPSC lines. Once hPSCs reached 20-30% confluency in a 100mm Petri dish, differentiation was begun using neural-induction media. At D6, SBRs fitted with small stir bars were inserted and returned to incubation on a magnetic stir plate set to 300 RPM. Culture proceeded until days 1821 when sufficient maturation of optic vesicles was observed. Measurements of the partial pressure of oxygen in cell culture media were taken using contactless optical sensor spots placed on the bottom of the dish. Dishes containing SBRs maintained an average of 28.05 mmHg, while static conditions averaged 2.34 mmHg, representing a greater than 10-fold increase in available oxygen. Analysis of the number and size of hPSC-derived ROs performed on D24 demonstrated significantly higher yield and larger size in all cell lines cultured using SBRs compared to static conditions. IHC staining found optic vesicles in static conditions to manifest hypoxia and increased cell death, whereas SBR greatly improves the culture by these parameters. Our research suggests that the SBR platform maintains high oxygen concentrations during differentiation leading to greater size and yield of ROs compared to traditional static culture methods and demonstrated enhanced neural retina (NR) and pigmented epithelial domains. Our study identifies a previously unappreciated bottle neck during early 2D phase of retinal development in culture, and the SBR designed to relieve this bottle neck should lead to greater consistency and efficiency in retinal organoid generation. A manuscript is in preparation (Schwab et. al., in preparation). 3. A new effort in our group is to force retinal organoid to go apical in. Technical limitations in current organoid models hampers their ultimate usefulness in the study of RGCs: (1) RGCs in retinal organoids degenerate as the organoids age; (2) RGCs within retinal organoids cannot project axons to and synapse with targets without dramatically altering the 3D structure of the organoid, and (3) RGCs in retinal organoids lack maturity in electrophysiology, subtype development, and inner retinal wiring, which could be due in part to the early demise of RGCs. We hypothesize that these problems can be overcome by flipping its polarity so that RGC layers locate to the exterior (apical in). In this configuration, oxygen and nutrient availability would be greatly increased for the RGCs, and their ability to naturally extend axons out of the organoid would be unencumbered. Since nutrient/oxygen depletion and lack of trophic support from target cells are the two likely culprits for RGC degeneration in traditional organoids, RGCs in the now apical in organoids would be expected to develop more mature subtypes, extend axons toward and innervate downstream targets, and survive long term. We recently developed a biomaterial engineering strategy to reconfigure the classic retinal organoid from an apical out to an apical in configuration, in which the photoreceptors line the inner lumen of the organoid, while the RGCs form the outermost layer (Figure 1). In this configuration, RGCs are directly exposed to the liquid media environment and thus have greater oxygen and nutrient availability, as well as ready exposure to any added neurotrophic factors or signaling molecules, which might be introduced to affect outcomes. We are investigating RGC maturation, axon pathfinding and regeneration in this model system (Wyndham Batchelor et al, in progress).