Project description:Purpose: The purpose of this study was to develop a framework for analyzing RPE expression profiles from zebrafish eye mutants. Methods: The fish model we used was smarca4 (SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily a, member 4), a retinal dystrophic mutant that the retinal phenotype and expression profiles were previously characterized. Histological and Affymetrix GeneChip analyses were conducted to define the RPE defects and underlying differential expression respectively. Results: Histological analysis indicates that smarca4 RPE was formed but its differentiation was abnormal. In particular, ultra-structural analysis of smarca4 RPE by transmission electron microscopy showed a number of defects in melanogenesis, suggesting that the cytoskeletal dynamics was impaired. To compare the expression profile of normal wild-type (WT) and smarca4 RPE, their retinas and RPE-attached retinas were microdissected and the gene expression values of these tissues measured by Affymetrix GeneChip analysis. The RPE expression values were then estimated from these samples using an approach previously established by us. A factorial analysis was conducted using the expression values of RPE, retinal as well as the whole-embryo samples. Specific rules (contrasts) were built using the coefficients of the resulting fitted model to select for three groups of genes: 1) Smarca4-regulated RPE genes, 2) Smarca4-regulated retinal genes, and 3) Smarca4-regulated RPE genes that are not differentially expressed in the retina. The latter group consists of 39 genes that are highly related to cytoskeletal dynamics, melanogenesis, paracrine and intracellular signal transduction. Conclusions: Our analytical framework can potentially identify genes in zebrafish mutants that both retina and RPE are affected by the underlying mutation. The gene expression levels of three independent replicates of WT whole embryos (WA52), WT retinas (WR52), WT RPE-attached retinas (WRR52), smarca4/yng retinas (YR52), smarca4 RPE-attached retinas (YRR52) and smarca4 whole embryos (YA52) were measured by Affymetrix Zebrafish Whole Genome Arrays. All samples were collected at 52 hpf. The probe-level data of these sample groups were background adjusted, normalized and summarized by a robust multiarray average (RMA) algorithm. RPE expression values were estimated from the comparison between the retinal and RPE-attached retinal samples of the same genotype based on a method we previously developed (Leung et al., Invest Ophthalmol Vis Sci. 2007; 48:881). Then, a 3x2 factorial design was used to analyze the expression values of three kinds of tissue (T): whole embryo, retina and RPE in two kinds of mutation (M) background: WT and smarca4. Using the coefficients from the fitted models, contrasts were built to identify candidate genes that fulfilled specific criteria. The false discovery rate (FDR) q-value cutoff for a contrast was 0.001. Several contrasts were ultimately combined to allow for selection of genes that were regulated by Smarca4 in the retina and RPE. These include 1) Smarca4-regulated RPE genes, 2) Smarca4-regulated retinal genes, and 3) Smarca4-regulated RPE genes that are not differentially expressed in the retina.

Project description:Purpose: The goals of this study were to identify cell specific expression patterns of the retinal pigment epithelium (RPE) and a variety of choroidal cell types originating from the macula and the periphery of human donor eyes, with a particular emphasis on identifying expression signatures of choroidal endothelial cells. Methods: Independent libraries were prepared for macular and peripheral samples of combined RPE/choroid from seven donors in two single cell sequencing experiments. In the first experiment, 8-mm macular and peripheral punches of RPE/choroid from Donors 1-3 were digested in papain prior to cryopreservation and subsequent GEM barcoding/library construction. Donors 1-2 had no noted ophthalmologic disease documented, while Donor 3 was diagnosed with age-related macular degeneration. In the second experiment, 12-mm macular and peripheral punches of RPE/choroid from Donors 4-7 were digested in collagenase II prior to cryopreservation, CD31-magentic bead enrichment and subsequent GEM barcoding/library construction. Donors 5-7 had no noted ophthalmologic disease documented, while Donor 4 was diagnosed with age-related macular degeneration. In both experiments, libraries were sequenced on a HiSeq4000. Sequenced reads were mapped to the human genome build hg19 will CellRanger(v3.0.1) and filters removed cells likely to be doublets or cells with a high proportion of mitochondrial reads. Clustering of cells with similar expression profiles was performed with Seurat (v3.0.2). Results: In the first experiment with unenriched RPE/choroid from donors 1-3, we recovered 4,355 cells post filtering with 40,177,477 corresponding reads. A total of 2,167 cells originated from the macula and 2,168 cells originated from the periphery. A total of 11 clusters were identified, and highly enriched genes in each cluster were used to classify clusters into their presumed dominant cell type. Differential expression analysis was performed between cells of macular and peripheral origin in each cluster and between donors with and without a history of age-related macular degeneration. Results: In the second experiment with CD31-enriched RPE/choroid from donors 4-7, we recovered 14,234 cells post filtering with 135,742,297 corresponding reads. A total of 7,647 cells originated from the macula and 6,587 cells originated from the periphery. A total of 8,521 of these cells were presumed endothelial cells. A total of 13 clusters were identified, and highly enriched genes in each cluster were used to classify clusters into their presumed dominant cell type(s). Differential expression analysis was performed between cells of macular and peripheral origin in each cluster and between donors with and without a history of age-related macular degeneration. Within each of the endothelial clusters (Clusters 5-8), differential expression analysis was performed between each cluster and the remaining endothelial cells. Conclusions: This study provides a large atlas of single-cell level gene expression patterns of the human retinal pigment epithelium and choroidal cell types. We identify expression patterns of most expected choroidal cell populations and describe macular versus peripheral regional differences. We additionally identify enriched transcripts in specific cell populations in age-related macular degeneration. We characterize gene expression patterns along the choroidal vascular tree, and identify populations of arterial, capillary, and venous endothelial cells. Our results show that single-cell sequencing can be performed on human RPE/choroid after cryopreservation, and that CD31 magnetic bead enrichment can be employed to enrich for endothelial cells for single-cell sequencing.

Dataset Information

RNAseq of human retinal pigment epithelial cells/choroid

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure