Abstract
Purpose :
Due to well-known limitations of current retinal models, an in vitro 3D model of the human retina that is reproducible and accurately predicts in vivo outcomes is highly desirable. In this study, we have compared the gene and protein expression profile in several batches of human iPSC-derived retinal organoids (ROs) to understand the level of variability found in ROs produced at large scale.
Methods :
Gene expression of 20 known retinal genes and representative markers of key cell types were quantified every 30 days by real-time PCR and immunofluorescence, respectively. RNA and 10 µm-cryosections were extracted from pools of ROs selected at random at different stages of differentiation and maturation from at least 4 biological replicates. iPSCs were used as a baseline control for relative quantification of gene expression. Quantification of Z-stack sum projections for marker expression was done using Fiji. One-way ANOVA and Turkey’s multiple comparison test of gene and protein expression data was performed.
Results :
Expression of photoreceptors (PRs), bipolar cells, müller glia, retinal ganglion cells (RGCs), horizontal cells (HCs), amacrine (AC) and retinal pigmented epithelium cells was increased in all ROs compared to iPSCs. The expression of different cell types fluctuates across RO development, which resembles in vivo development. Interestingly, the expression of RGC markers (MATH5 and BRN3) peaked at day 60, whereas cone (OPN1SW, OPN1MW and OPN1LW) and rod (RHO) PR genes are expressed from day 120 achieving highest expression at day 210 in culture. In terms of cell populations, RCVRN was observed on average in 20% of the cells in day 150 ROs. This average was consistent across different batches of ROs (p>0.05). This expression level plateaus at day 210 with nearly 30% RCVRN positive cells present in several ROs.
Conclusions :
We have analysed the gene and protein expression profile of several batches of human iPSC-derived ROs across different stages of development. We observed that some key retinal cell types, such as PR, HC and AC, appear at consistent levels between batches. This data set provides crucial information for pre-clinical studies in ROs with application in drug discovery, disease modelling and gene therapy.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.