Transcriptional Landscape of Retinal Endothelial cells reveals novel molecular signatures in Diabetic Retinopathy

Abstract

Purpose : The cell-specific transcriptomic analysis provides an opportunity for a deeper understanding of complex interactions among the “neurovascular unit” (NVU) of the retina in diabetes. In this study, we performed comprehensive transcriptomic profiling of single cell murine retinal endothelial population to understand the gene expression alterations in the diabetic retinal vascular endothelial cell ecosystem.

Methods : Retinal tissue from diabetic and non-diabetic C57bl/6j mice (n=10 eyes in each group) were digested with collagenase-D to obtain a single cell suspension. Enriched endothelial cell population from the retina was obtained through fluorescent activated cell sorting (FACS) using CD105-PECy7 and CD31-FITC (ebioscience) staining. RNA sequencing of the samples was done using paired-end Illumina sequencing (NovaSeq 6000 System). Also, quantification of mRNA expression from human retinal endothelial cells treated with high glucose (25.5mM) for 48 hrs was performed using Human Endothelial Cell Biology RT Profiler PCR Array (Qiagen). Alteration of the critical signaling pathway in activated retinal endothelial cells was identified using differentially expressed genes (DEGs) set through gene set enrichment analysis (GSEA).

Results : We classified 1016 DEGs from endothelial cell population from diabetic retina compared to non-diabetic retinas. mRNA from diabetic cells were mostly upregulated (60%) compared to control. We detected an augmented expression of genes linked to endothelial cell metabolism particularly genes linked to non-oxidative glucose metabolism. We also observed an upregulation of genes linked to activation of Rho GTPase, mTOR pathway and integrins. Retinal endothelial cells also displayed an increased Kdr expression indicating a possible role for Vegf induced endothelial cell activation in diabetes. In HRECs cells treated with high glucose, we found an increased expression of Ang-2, CCl2 and IL-6. Interestingly, diabetic retinal endothelial cell population expressed genes relevant to IL-6 activation (N=8).

Conclusions : Our analysis of single cell population indicates the potential candidate pivotal genes and key signaling pathways involved in retinal endothelial dysfunction in diabetes. This approach could significantly improve our understanding of complex interactions among retinal NVU to identify novel therapeutic targets for the precise treatment of DR.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.