Abstract
Purpose :
Retinal ganglion cell (RGC) transplantation holds promise for vision restoration in optic neuropathies, such as glaucoma or neurofibromatosis type 1. Current animal studies show that only a tiny fraction of donor RGC can successfully integrate into the retina and recapitulate the typical developmental trajectory, which includes survival, migration, differentiation, maturation and specialization, neurite outgrowth, and synaptogenesis. This in silico study aims to identify critical factors to be modulated in transplantation by designing computational models for these functions.
Methods :
We use scRNA-seq data for human and mouse developing and adult retina (FD 59, 82, 125, hAS, and E14, E16, E18, mAS) and human retinal organoids at different days of differentiation (OD 45, 60, and 205). We have focused on several functions: a) survival; b) migration; c) synaptogenesis; d) integration. Our computational toolbox includes: 1) pseudo-time- and cell-fates-based approaches; 2) a pipeline based on synaptogenesis activity score, ligand-receptor communications, and multiple high-throughput filters 3) a mathematical model based on ligand-receptor communications with 256 possible patterns of expression changes for RGC vs. total retina; 4) timeline merging for GSEA analysis and differential expression score; 5) ForceAtlas2 3D (2D + Z-function of interest) approach for biologically relevant dimensionality reduction. We selected a unique combination from above based on the relevance for every cell function.
Results :
For each function, we identified unique and species-specific genes that can be modulated in transplantation. For synaptogenesis: APLP2, APP, HHIP, LINGO1, TUBB3, ARF6, DBN1, ERC1, and RAB3A. For migration, we confirmed that CXCR4 and DCC ligands could be used to guide RGCs in vivo. For survival and integration, we ranked 103 RGC-specific ligand-receptor pairs that can be used to enhance transplantation outcomes. These include GDNF-GFRA1 and BDNF-TRKB. The survival and migration were validated in RGC transplantation in vivo.
Conclusions :
We describe a new way of single-cell data bioinformatical analysis for every transplantation-related function of interest based on comparing cell populations and microenvironment in the retina during development and adult state. Our pipelines pinpoint genes that can be modulated to enhance RGC integration after in-cell replacement studies.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.