NRL- and CRX-guided gene network modulates photoreceptor presynapse size and positioning during retinal development

Abstract

Unique morphologies of rod and cone photoreceptor presynaptic terminals permit the formation of synapses onto interneurons during retina development. We integrated multiple \"omics\" datasets of developing rod and S-cone-like photoreceptors and identified 719 genes that are regulated by NRL and CRX, critical transcriptional regulators of rod differentiation, as candidates for controlling presynapse morphology. In vivo knockdown of 72 candidate genes in the developing retina uncovered 26 genes that alter size and/or positioning of rod spherules in the outer plexiform layer. Co-expression of seven cDNAs with their cognate shRNAs rescued the rod presynapse phenotype. Loss of function of four genes in germline or by an AAV-mediated CRISPR/Cas9 strategy validated RNAi screen findings. A protein interaction network analysis of the 26 positive effectors revealed additional candidates in the NRL/CRX-regulated presynapse morphology-associated gene network. Follow-up knockdowns of two novel candidates support the proposed network. Our studies demonstrate a requirement of multiple components in a modular network for rod presynapse morphogenesis and provide a functional genomic framework for deciphering genetic determinants of morphological specification during development.\n\nAuthor SummaryThe relationship between neuronal morphology and function has been recognized for over 100 years. However, we still have poor understanding of genes and proteins that control morphogenesis of a specific neuron. In the current study, we address this connection between gene expression and neural morphology by identifying and knocking down a subset of expressed genes in rod photoreceptors. We ascertained a number of candidate genes controlling photoreceptor pre-synaptic terminal morphology, which is necessary for its connection with second-order neurons in the retinal circuit. Furthermore, we have curated a more plausible network of genes, either identified in our study or predicted, that are enriched for processes underlying photoreceptor morphogenesis. We suggest that our work will provide a framework for dissecting genetic basis of neuronal architecture and assist in better design of cell replacement therapies for retinal degeneration.