Abstract
Purpose :
Current therapies for neovascular age-related macular degeneration (nAMD) target extracellular cytokines, but developing more precise, cell-specific therapies require knowledge of the cellular origins of pathogenic factors and pathways. Here, we characterize the transcriptional landscape of two distinct and well-characterized mouse models of macular neovascularization–laser-induced choroidal neovascularization (CNV) and retinal neovascularization 3 (rnv3) mutant mice – using single-cell transcriptomic analysis.
Methods :
We collected posterior eye cups from 7-week-old rnv3 mutants and wild-type littermate controls with and without laser-induced CNV. Retinal and RPE-choroid tissues were dissected for single-cell dissociation, processed using 10x Genomics pipeline, sequenced on Illumina NovaSeq, and analyzed using Seurat and R packages. We compared differentially expressed genes and performed gene ontology (GO) and functional enrichment analysis, focusing on angiogenesis, inflammatory, and fibrosis-related gene sets.
Results :
We collectively recovered 22,769 cells from retina and 45,769 cells from RPE-choroid tissues from rnv3, laser CNV, and control eyes. Retinal cell populations were similar between the 3 groups, while RPE-choroid tissues showed marked expansion of some endothelial cell populations and a cell clusters exhibiting both RPE and fibroblast markers, suggestive of epithelial-mesenchymal transition (EMT) that was more prominent in CNV than rnv3 eyes. Retinal tissues showed the greatest proportion of DEGs in astrocytes specifically in rnv3 eyes, including metallotheioneins (Mt1-3), complement signals (Serping1/a3, C4b), and other markers of astroglial activation (Nupr1, GFAP). RPE-choroid tissues showed the greatest proportion of DEGs in endothelial cell populations in both rnv3 and CNV, with top GO terms that mostly include transcriptional regulators. Established angiogenic signals such as Hif1a and Vegfa/b are expressed in many retinal and choroidal cells, and are similarly upregulated in both disease models.
Conclusions :
Cellular level transcriptional profiling reveals distinct pathogenic processes in mouse models of neovascularization– laser CNV induces more EMT and rnv3 demonstrate more astroglial activation, although many angiogenic factors are similarly altered in both models.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.