bioRxiv2025
Visual information from the retina is sent to diverse targets throughout the brain by different retinal ganglion cells (RGCs). Much of our knowledge about the different RGC types and how they are routed to these brain targets is based on mice, largely due to the extensive library of genetically modified mouse lines. To alleviate the need for using genetically modified animal models for studying retinal projections, we developed a high-throughput approach called projection targeting with phototagging that combines retrograde viral labeling, optogenetic identification, functional characterization using multi-electrode arrays, and morphological analysis. This method enables the simultaneous investigation of projections, physiology, and structure-function relationships across dozens to hundreds of cells in a single experiment. We validated this method in rats by targeting RGCs projecting to the superior colliculus, revealing multiple functionally defined cell types that align with prior studies in mice. By integrating established techniques into a scalable workflow, this framework enables comparative investigations of visual circuits across species, expanding beyond genetically tractable models. MotivationVisual information from the retina is distributed to diverse targets throughout the brain. Much of our knowledge about how visual information is processed and routed to these brain targets is based on mice because of the large library of genetically modified mouse lines. For most other species, such libraries are not available. Therefore, we were motivated to develop an approach for characterizing diverse retinal projections into the brain that can be applied to other species. We aimed to achieve projection targeting with retrograde viral vectors, followed by identification of circuit-specific retinal ganglion cells (RGCs) with optogenetics, functional characterization with multi-electrode array (MEA) recordings, and morphological description with in situ and confocal microscopy. The resulting high-throughput approach permits functional and morphological investigation of dozens to hundreds of cells in individual experiments. By replacing the need for genetically modified animals with a suite of standard techniques, this approach should improve cross-species comparisons of the initial stages of visual processing. SummaryUnderstanding the structure-function relationships across neurons is challenging, particularly when circuits are composed of dozens of distinct cell types. We refined an approach, called projection targeting with phototagging, that allows simultaneous elucidation of the projections, morphology, and visual response properties of diverse RGC types in the mammalian retina. The approach combines retrograde virally mediated phototagging of RGCs, microscopy, and large-scale MEA measurements. Importantly, the approach does not rely on transgenic animals and thus is generalizable across species. We validated this approach in rats by targeting retinal projections to the superior colliculus (SC). We showed that multiple RGC types project to the SC and that these results in rats align well with prior findings from transgenic mouse studies. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=194 HEIGHT=200 SRC="FIGDIR/small/661576v1_ufig1.gif" ALT="Figure 1"> View larger version (49K): org.highwire.dtl.DTLVardef@1003310org.highwire.dtl.DTLVardef@1881cdorg.highwire.dtl.DTLVardef@f1dba1org.highwire.dtl.DTLVardef@8a539f_HPS_FORMAT_FIGEXP M_FIG C_FIG HighlightsO_LIThe viral vector, AAV2retro, is effective for delivering opsins to retinal ganglion cells (RGCs) for phototagging and projection mapping. C_LIO_LIPhototagging allows matching RGC visual responses to morphology, while also identifying their central projections. C_LIO_LIThe approach does not rely on genetically modified animals and thus is generalizable across species. C_LI