Modeling and Rescue Strategies for Ciliopathies using Stem Cells Ciliopathies include a spectrum of phenotypes with defects in primary cilia biogenesis and/or function. Many retinal diseases are characterized by defects in cilia including, but not limited to LCA and JSRD. CEP290 is a gene which is important for cilia biogenesis and transport. Cilia functions of patients can be differentially affected due to distinct mutations in this gene, leading to CEP290-related ciliopathies. Therefore, understanding the molecular mechanisms underlying the retinal development in JSRD and LCA patients is the utmost need for the development of treatments for these diseases. To elucidate the mechanisms, we have generated human iPSC lines from fibroblasts of LCA and JSRD patients and control individuals. We have generated retinal organoids from these iPSC lines and collected neural retina at different stages of differentiation for immunohistochemistry and transcriptome analysis. We have recently shown that iPSC-derived optic vesicles from CEP290-LCA patients harbor less developed photoreceptor cilia compared to controls. Our studies thus recapitulate the pathologic changes in CEP290-LCA-specific human iPSC-derived 3-D retina and should serve as a useful model to test treatment strategies (Shimada H et al, Cell reports, 2017). Drug Discovery and Small Molecule Screening Using Retinal Organoids In attempt to find candidate drugs to rescue ciliopathy-related photoreceptor dysfunction, we developed in vitro screening platforms using mouse embryonic fibroblasts (MEFs) and human pluripotent stem cell derived-retinal organoid cells with mutations in CEP290. Defects in ciliogenesis resembling in vivo mouse retina was observed in both MEFs and human retinal orgainoid cells. Mutant MEFs displayed a higher expression of CEP290 and shorter cilia compared to non-mutant wild type MEFs. Photoreceptors in retinal organoids had more severe phenotypes, including missing ciliary rootlets and reduced ciliation. In collaboration with colleagues at NEI, we screened potential gene therapies and demonstrated reduced expression of CEP290 with longer cilia in MEFs. Photoreceptors from retinal organoids also showed improvements in number of ciliated photoreceptors and cilia length. We performed a high-throughput screening of candidate drugs in collaboration with the National Center for Advancing Translational Sciences (NCATS) in an attempt to rescue ciliopathies from retinal organoid-derived photoreceptors. Our primary screening resulted in 12 positive hits to maintain cell survival and secondary screening identified several compounds that were able to recapitulate the effect demonstrated with the initial gene therapy screening. Modelling of Mutations within Transcription Factors for LCA Photoreceptor dysfunction characteristic of patients with photoreceptor genetic discordes can be also caused by mutations in genes unrelated to cilia. One causal mutation of LCA is noted in CRX which encodes an essential transcription factor, cone-rod homeobox protein (CRX), for cone and rod development and function. To better understand the disease phenotype, we have derived human iPSCs from CRX-LCA patients and familial controls and differentiated them toward photoreceptors. To date, retinal organoids of both patient and control lines have been formulated. We have observed lamination similar to positive control retinal organoids indicating that differentiation is proceeding successfully. Future work will compare the transcriptome of CRX-LCA retinal organoids and controls to healthy embryonic tissue. Using a similar high-throughput system outlined above, candidate drugs to rescue disease phenotype will be screened. Stabilizing In Vitro 3-D Models of Retinal Diseases with RPE Modelling of retinogenesis in vitro has been hampered by limited development of outer segments essential to detection of light by photoreceptors. In collaboration with Tiansen Li, PhD, we hypothesize that a retinal pigmented epithelial (RPE) layer is necessary to form mature connecting cilia and functional outer segments. We have used an embryonic stem (ES) cell line (H9) and patient induced pluripotent stem cell lines (hiPSCs) to differentiate stem cells into RPE-like cells with 60 days of induction. Our differentiated RPE cells correctly express proteins similar to mature RPE cells, have polarity typical of RPE cells, and display distinct RPE morphology. This indicates that we can successfully differentiate stem cells lines toward RPE phenotypes with our protocol in a reliable manner. Our lab has also developed an electrospun poly-caprolactone scaffold that mimics the properties of the supporting structure of RPE, the Bruchs Membrane. Stem cells have been differentiated into RPE on these electrospun scaffolds and maintained similar characteristics of cells differentiated with our standard protocol. Initial results from our lab show stem cells differentiated on this electrospun scaffolds develop markers for RPE, express necessary RPE proteins, and undergo morphological changes such as: hexagonal shape formation and polarization typical of RPE. We hypothesize that differentiating toward RPE cells on Bruchs Membrane mimics will increase efficiency and efficacy. Future work will test this hypothesis by comparing the transcriptome to embryonic human samples. Due to limited outer segement and cilia formation in vitro, functional assays for testing photoreceptor retina viability and functionality are unable to be performed. This limits the potential for patient-derived retinal organoids in patient-specific drug screening in vitro. We have begun to develop a co-culture model system with optic vesicles and differentiated RPE on electrospun scaffolds. With RPE support, we hypothesize that we can enhance photoreceptor maturation and viability, ciliagenesis, and outer segment formation, leading to functional photoreceptors in vitro. Bioengineering Strategies to Increase Scale and Efficiency of Organoid formation Current protocols require long-term cell culture (one month for mouse and >200 days for human retinal organoids) to form structures similar to outer segments that would be required for functionality. Combined with high variability in initial organoid structure and size, variability with in vitro organoid models is high. In order to establish an efficient and scalable in vitro platform for modeling retinogenesis and development treatments, we optimized the previously established serum-free floating cultures of embryoid body-like aggregates with quick reaggregation (SFEBq) protocol to differentiate mouse pluripotent stem cells and developed the novel high efficiency hypoxia induced generation of photoreceptors in retinal organoids (HIPRO) protocol. By applying hypoxic condition to early stage of organoid differentiation, a significantly higher efficiency of optic vesicles and optic cups formation could be achieved. At the end stage of differentiation, neural retina formed polarized and laminated structure similar to native tissue, with all major cell types with advanced structure formation including cilia and synapses. Transcriptome analysis of flow-sorted rod photoreceptors indicated that day 35 rod-like cells in organoids correlated to postnatal day 6 mouse retina, indicating that the in vivo phenotype could be recapitulated. We also optimized creation of photoreceptor differentiated from retinal organoids with the use of commercially-available rotating-wall bioreactors. Organoids cultured in bioreactors demonstrated higher production of photoreceptor-related genes and accelerated differentiation. Transcriptome analysis of neural retina revealed similar maturity among different conditions, suggesting that additional factors may be required for functional maturation of the retina in vitro. Future work will extend these
Showing the most recent 10 out of 18 publications
