Abstract
Purpose :
Until now, methods for identifying compounds or genetic manipulations that alter the homeostasis of retinal proteins of interest (POI) have been limited mainly to non-physiologic constitutive overexpression of large fluorescent/luminescent reporter tags fused to that POI. Here we describe an attractive alternative strategy that allows for easy monitoring endogenously expressed POIs from retinal cells for unbiased drug/gene discovery.
Methods :
As an initial proof-of-principle, we used a CRISPR/Cas9 ribonucleoprotein and ssODN HDR template to genome engineer immortalized RPE cells to express a 2xFLAG HiBiT tag (29 total residues) on fibulin-3 (F3, EFEMP1), a protein associated with Malattia Leventinese/Doyne Honeycomb Retinal Dystrophy and glaucoma. To detect the HiBiT tag, conditioned media or lysates were supplemented with LgBiT and substrate to generate NanoBiT (HiBiT + LgBiT), a bright and stable luciferase. Cells were used to efficiently screen ~10,000 compounds from the UT Southwestern chemical library. Toxic compounds and NanoBiT inhibitors were identified in counterscreens. Hit compounds were verified in dose-response, repurchased, and validated in differentiated/polarized primary porcine RPE cells to confirm their activity on F3 secretion/intracellular levels.
Results :
Genome engineering of F3 resulted in the formation of a single 2xFLAG-tagged HiBiT species as detected by western blotting. LgBiT complementation of 2xFLAG HiBiT F3 in media enabled detection in sub-microliter volumes. Compound screening identified both F3 enhancers and reducers. Surprisingly, a series of drugs typically used for treatment of glaucoma significantly increased F3 expression whereas a set of compounds that can regulate epithelial to mesenchymal transition (EMT) signaling significantly reduced both secreted and intracellular F3.
Conclusions :
These findings support the use of the versatile 2xFLAG HiBiT tag in retinal drug discovery. Such an approach can also be used to enrich (via the 2xFLAG peptide) endogenous POI interacting partners. We speculate that this same general approach can also be applied to other retinal degeneration genes and may ultimately be useful in identifying mutant-specific drugs or genetic manipulators.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.