Back

Leveraging crystallographic fragment screening, algorithmic merging and digital chemistry to target NCS-1 protein-protein interactions for treatment of neurological disorders

Munoz-Reyes, D.; Fieseler, K. K.; Winokan, M.; Golding, M.; Capkin, E.; Ferla, M.; Perez-Suarez, S.; Tomlinson, C. W. E.; Marples, P. G.; Miro-Rodriguez, C.; Mansilla, A.; Fearon, D.; Thompson, W.; von Delft, F.; Sanchez-Barrena, M. J.

2025-07-22 biophysics
10.1101/2025.06.18.660085 bioRxiv
Show abstract

Efficient drug discovery relies on workflows that integrate structural insights with rapid and cost-effective exploration of chemical space. Here, we present a data-driven fragment-based lead discovery approach to target Neuronal Calcium Sensor 1 (NCS-1) protein-protein interactions (PPIs). This study represents a complete implementation of a single high-value design-make-test-analyze cycle that directly yields compounds with micromolar affinity with the potential to modulate NCS-1 interactions with key targets, including the G-protein chaperone Ric-8A and the dopamine D2 and cannabinoid CB1 receptors. X-ray crystallographic fragment screening (CFS) revealed diverse interaction patterns within the NCS-1 hydrophobic crevice. Algorithmically guided fragment evolution and automated synthesis enabled the rapid generation of over 250 derivatives, with biophysical validation using LC-MS and Grating-coupled interferometry. Structural analyses highlighted key pharmacophores, with selected compounds exhibiting favorable drug-like properties and potential blood-brain barrier penetration, making them promising candidates for neurodegenerative and neurodevelopmental disorders. Our results demonstrate the feasibility of accelerated hit-to-lead development at synchrotrons, demonstrating a robust, scalable platform for PPI-targeting drug discovery. The generated chemically diverse scaffolds provide a strong foundation for future therapeutic optimization.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.