Objective

Optic neuropathy (ON) refers to the irreversible degeneration of optic nerves and manifests in glaucoma or mitochondrial diseases as Leber’s hereditary ON. ON is caused by death of the specialized retina neurons whose axons collectively form the optic nerves, namely the retinal ganglion cells (RGC). ON predominantly leads to blindness. Its burden on society manifests in disability, co-morbidities, medical care and loss of employability. Efficient therapeutic solutions are still needed. The aim of this project is to implement a new model of human RGC on planar micro electrode arrays (MEA) for personalized ON modelling and candidate drug screening.

Methods

Induced Pluripotent Stem Cells from healthy donor were grown as monolayers and driven to differentiate into Retinal Progenitor Cells (RPC) upon inhibition of the BMP, TGF-beta and Wnt pathways. RPCs were in turn differenciated into RGC-like neurons (iRGC) by inhibiting the Notch pathway. Specific blends of growth factors or drugs were used depending on the differenciation stage. MEAs were manufactured using poly 3,4-ethylenedioxythiophene (PEDOT) and carbon nanofibers. We tested two passivation materials: the widely used SU8 and the emerging parylene, both coated with matrigel TM or EHS laminin.

Results

Our method relies on inhibiting specific pathways in the presence of factors critical for eye development, as IGF-1 or FGF-8. We successfully generated actively growing RPC which showed immunoreactivity to retinal progenitor markers (nestin, PAX6, RAX, SOX2) at day 22. RPC efficiently (>80%) differentiated into iRGC which showed immunoreactivity to RGC markers (Brn3A/B/C, CD90, RBPMS) alongside HUC/D and beta 3 tubulin, 14 days after the medium shift. iRGC self-organized in remarkable cluster-based networks, with up to 200-µm diameter dense cell foci connected to each other by up to 30-µm thick and 300-µm long projections. MEAs showed remarkably low impedance and no cytotoxicity on iRGCs. Both SU8 and parylene provided good support for iRGC differentiation and spreading. Electrophysiological studies are ongoing and will be presented in the poster.

Conclusion

Here we present the first step towards an innovative iRGC – MEA chip amenable to ON modelling, personalized drug screening or investigation of the outcomes of electrostimulation on diseased iRGC. In the future, the addition of retinal glial or epithelial cells to iRGC will pave the way for an integrated, connected retinal interface.