Deconstructing Retinal Organoids: Single Cell RNA-Seq Reveals the Cellular Components of Human Pluripotent Stem Cell-Derived Retina

Abstract

The rapid improvements in single cell sequencing technologies and analyses afford greater scope for dissecting organoid cultures composed of multiple cell types and create an opportunity to interrogate these models to understand tissue biology, cellular behavior and interactions. To this end, retinal organoids generated from human embryonic stem cells (hESCs) were analyzed by single cell RNA‐sequencing (scRNA‐Seq) at three time points of differentiation. Combinatorial data from all time points revealed the presence of nine clusters, five of which corresponded to key retinal cell types: retinal pigment epithelium (RPE), retinal ganglion cells (RGCs), cone and rod photoreceptors, and Müller glia. The remaining four clusters expressed genes typical of mitotic cells, extracellular matrix components and those involved in homeostasis. The cell clustering analysis revealed the decreasing presence of mitotic cells and RGCs, formation of a distinct RPE cluster, the emergence of cone and rod photoreceptors from photoreceptor precursors, and an increasing number of Müller glia cells over time. Pseudo‐time analysis resembled the order of cell birth during retinal development, with the mitotic cluster commencing the trajectory and the large majority of Müller glia completing the time line. Together, these data demonstrate the feasibility and potential of scRNA‐Seq to dissect the inherent complexity of retinal organoids and the orderly birth of key retinal cell types. Stem Cells 2019;37:593–598

Investigating cone photoreceptor development using patient-derived NRL null retinal organoids

“…To better understand the identity of the photoreceptor populations that develop in the absence of NRL, we performed scRNAseq to characterize the cell populations of WT organoids compared to L75Pfs organoids. Previous studies have analyzed retinal organoid development using scRNAseq; however, the findings were limited by an inability to resolve cell clusters, the loss of INL cell populations, or incomplete characterization of photoreceptor populations [26][27][28] . Using transcriptomics, our analyses verified that retinal organoids generate all major neuronal cell types present in in vivo retina (Fig.…”

“…More recently, scRNAseq studies with developing and adult retinal tissue have offered insight into in vivo human retinal cell populations [23][24][25] . While previous studies have utilized scRNAseq to identify cell types of developing retinal organoids, they have not discerned distinct photoreceptor sub-populations [26][27][28] . Additionally, while transcriptomics have been used to characterize Nrl loss in murine photoreceptors, these analyses were not performed at a single-cell level, thus limiting the potential to identify and characterize sub-populations [29][30][31] .…”

Single cell RNA sequencing reveals cellular diversity of trisomy 21 retina

“…In this study, we conducted scRNA-seq of the retina from a trisomy 21 fetal to dissect the heterogeneity of retina, providing the first retinal cell atlas under T21 condition. We also studied the influence of redundant chromosome 21 in the process of retinogenesis Finally, two retinal cell types, amacrine cells and horizontal cells, were found in another single cell human fetal study 11 and several other mammalians retinal study 14 , but absent in a retinal organoids study 16 speculatively due to the deficiency of sampled cells number. We assumed that the absence of these two cell types in our data might indicate the general delay of the development of trisomy 21 retina or the severe abnormity of highly chimera of trisomy 21.…”

“…C4 was considered to be retina ganglia cells due to the enrichment of RBPMS. C7 and C9 were identified as cone photoreceptor, because of the specific expression of LMOD1 16 (Figure 2c, d, e, Figure S2). C3, C5 and C6 expressed three different Müller glia cell markers respectively, CLU, GPX3 and HES1 17 .…”

“…C2 was characterized by the high expression of ANGPT1. C8, expressing markers of retina pigment cells, such as TYRP1 and RMEL 16 , was defined as retina pigment cells.…”