Abstract
Purpose :
To identify the cellular origins of vascular endothelial growth factors (VEGF) and potential new therapeutic targets for ocular angiogenic disorders, we performed single-cell transcriptomic analysis on retinal, retinal pigment epithelium (RPE), and choroidal tissues from mouse eyes after laser-induced choroidal neovascularization (CNV).
Methods :
Eight to ten laser CNV spots were created in wild-type adult C57/BL6 mouse eyes. One week after laser injury, retinal and RPE-choroidal tissues were dissected from the CNV lesions from 9 animals for single-cell preparation. Single-cell RNA sequencing libraries were prepared using Chromium technology, and barcoded libraries were pooled and sequenced at 300 million reads per library using the Illumina HiSeq 4000 platform.
Results :
A total of 16,321 cells (14,966 from retina and 1,355 from RPE-choroid) were recovered and approximately 1200 gene were identified from each cell. Nine major transcriptionally distinct cell populations were identified from unsupervised clustering (Figure 1), including rod/cone photoreceptors, cholinergic/GABAergic/glycinergic amacrine cells, rod/cone bipolar cells, and RPE. One major cluster was further subdivided into Muller glia, horizontal, retinal ganglion cells, endothelial cells, stromal cells, and melanocytes. Among these cell types, we found that VEGFa and VEGFb were expressed in all major cell types except rod and cone photoreceptors, with the highest levels of expression found in Muller glia and RPEs. By contrast, VEGFc and VEGFd were minimally expressed in all cell types.
Conclusions :
Single-cell transcriptomic analysis demonstrate differential expression of VEGF family members across different retinal and choroidal cell types after laser-induced CNV in mouse eyes, providing critical guidance for cell-specific targeting of anti-angiogenesis retinal gene therapies. Analyzing differentially expressed genes may help identify other novel therapeutic targets for ocular angiogenic disorders.
This is a 2021 ARVO Annual Meeting abstract.