Single cell transcriptomics of the retinal vascular endothelium in health and diabetes

The endothelial cells that line retinal blood vessels play vital roles in angiogenesis and microvascular barrier formation and are supported by retinal mural cells. When these cells become dysfunctional during diabetes, they can lead to sight threatening complications such as diabetic retinopathy (DR) and diabetic macular oedema (DMO). Gene expression changes that underlie endothelial dysfunction in diabetes have been studied by PCR and bulk RNA sequencing methods. These methods give the impression of uniform expression across all cells within a sample; however, great variability has been found in gene expression between individual cells even within a relatively homogeneous population. To uncover this cell-to-cell variation a new technology known as single cell RNA sequencing (scRNA-seq) has been developed in recent years, which identifies gene expression within single cells. Therefore, the aim of my PhD project was to use scRNA-seq to better understand the biology of the retinal endothelium and retinal vasculature and how this becomes dysfunctional within a diabetic milieu. A single cell RNA-sequencing workflow was established to uncover novel insights into retinal vascular biology including the expression of novel alternative splice variants, endothelial heterogeneity, patterns of active transcription factor expression and development of a more complete model of the inner blood retinal barrier (iBRB). Key changes in gene expression triggered by high glucose (HG) were related to the upregulation of inflammatory and proliferation-related genes with many genes not previously linked to these processes within hyperglycaemia. We found that during experimental diabetes, retinal endothelial cells adopt an angiogenic phenotype even during very early diabetes in the absence of overt hypoxia and a neovascular response. Retinal mural cells were found to switch to a more migratory phenotype which is in agreement with recent publications that have shown that pericyte migration is a key event in DR progression. In addition, numerous novel high glucose and diabetes specific gene expression and TF changes were found which warrant further investigation. Current therapeutics for DR mainly focus on the end stages of the disease process and can be ineffective in many patients with some adverse side effects. Through this study, we have identified many novel potential therapeutic targets for the early-stage treatment of diabetic retinopathy that could potentially slow or prevent the development of the disease. This paves the way for numerous studies to further investigate these novel findings to develop more effective therapeutics against DR and DMO.