Discovery of Selective Small-Molecule Ligands of SV2C by AI-Enhanced Virtual Screening and Experimental Validation.
Discovery of Selective Small-Molecule Ligands of SV2C by AI-Enhanced Virtual Screening and Experimental Validation.
Where did the research take place?
The study site has not been established. Author addresses may differ from where the research occurred.
Palo Alto, US · Author affiliation
SandboxAQ, Palo Alto, California, United States.Location evidence
Pittsburgh, US · Author affiliation
Department of Structural Biology, University of Pittsburgh, Pittsburgh, Pennsylvania, United States.Location evidence
Toronto, CA · Author affiliation
University of Toronto, Toronto, Ontario, Canada.Location evidence
New York City, US · Author affiliation
Columbia University Irving Medical Center, New York, New York, United States.Location evidence
Publication status: preprint
A plain-language reading has not been prepared for this paper yet.
Original abstract
Synaptic vesicle glycoprotein 2C (SV2C) is enriched in dopaminergic neurons and implicated in Parkinson's disease, but no selective small-molecule probes exist for it. Lacking a full-length SV2C structure, we built a homology model from SV2A cryo-EM templates and used molecular dynamics to characterize its conformational landscape. An AI-enhanced virtual screening pipeline, validated on a curated SV2A benchmark, was applied to 5.96 million commercial compounds, prioritizing 94 candidates for experimental testing. Of 71 compounds tested in an orthogonal biophysical assay cascade, 22 were active (31% hit rate), and five advanced to isoform-selectivity profiling. Compounds 36 and 56 emerged as leads, with SV2C K i values of 24.6 μM and 3.25 μM and greater than 10-fold selectivity over SV2A. An unpublished SV2A cryo-EM structure independently confirmed the predicted binding mode. This AI-driven pipeline delivered selective SV2C ligands from a general chemical library, providing tools to probe SV2C biology in Parkinson's disease.