hPSC-derived Photoreceptor Axon Extension and Growth Cone Motility Data Sets and Code

Published: 17 March 2022| Version 1 | DOI: 10.17632/cpy5j23354.1
Sarah Rempel


Raw data and scripts used for data processing and analysis of research on human stem cell-derived photoreceptors (PRs). This data is deposited as a companion to the Cell Reports paper by Sarah K. Rempel, Madalynn J. Welch, Allison L. Ludwig, M. Joseph Phillips, Yochana Kancherla, Donald J. Zack, David M. Gamm, and Timothy M. Gomez. See the associated paper for more information. Summary: This data is from microscopy experiments of dissociated retinal organoids differentiated from a WA09 CRX+/tdtomato line, which marks all post-mitotic photoreceptors. The included code consists of R scripts (as .R files) and ImageJ Macro scripts (as .txt files) that were used largely to process and transform the data. The raw data from running the images (not included) through the ImageJ Macro scripts (or manual analysis in ImageJ) is included as .csv files. While we expected PRs to extend axons through classical growth cone-mediated mechanisms, we instead found that they only had a very brief period in which they made growth cones with the classical cytoskeletal arrangements and robust protrusion and retraction that are characteristic of projection neuron growth cones. While day 40 PRs had dynamic growth cones that appeared capable of autonomous axon extension, by day 80 of development, PR terminals were instead stationary and highly adhered to the substratum. By live imaging photoreceptors at day 40-44 and day 80-82 with the immortalized Muller Glia line ImM10 (MG), we found that MG appeared to pull PRs to extend out their axons. Day 40 PRs that were in the MG co-culture dishes but not touching MG during the imaging period had higher terminal displacement rates than cell body displacement rates, consistent with autonomous growth cone-mediated axon extension, while day 80 PRs had little to no cell body or terminal displacement rates. Both cell body and terminal displacement rates were much higher among PRs that were contacting MG during the imaging period, and these events often looked as if the highly motile MG were pulling the PRs away from their adherent terminals, stretching the axon behind it. Additionally, we found that PRs only exhibit substrate-dependent effects on axon length when they are contacting other cells, suggesting these effects are non-cell autonomous. PRs also lose F-actin content from day 40 to day 80, including losing the tightly bundled F-actin characteristic of filopodia and the ordered mesh characteristic of lamellipodia. We also found that the synaptic markers Synaptic Vesicle Protein 2A and Synaptophysin-1 strongly localized to the adherent terminals of day 80 photoreceptors. Finally, we live imaged PRs in situ in 3D retinal organoids and found that they were less likely to have dynamic terminals with age.


Steps to reproduce

The original data was collected by confocal microscopy, multi-photon imaging, and interference reflection microscopy. See the Cell Reports paper for details on the microscopes, cell culture, immunocytochemistry, etc. The included CSV files were those generated from the ImageJ macro scripts or from manual analysis in ImageJ. Those files were then processed and transformed with the included R scripts to generate files that could be easily imported into Graphpad Prism, which is where statistical analysis was done.


University of Wisconsin Madison


Growth Cone, Photoreceptor, Retinal Development, Axon Guidance