Discovery of a Novel DCAF1 Ligand Using a Drug-Target Interaction Prediction Model: Generalizing Machine Learning to New Drug Targets

DCAF1 functions asa substrate recruitment subunit for the RING-typeCRL4(DCAF1) and the HECT family EDVPDCAF1 E3 ubiquitinligases. The WDR domain of DCAF1 serves as a binding platform forsubstrate proteins and is also targeted by HIV and SIV lentiviraladaptors to induce the ubiquitination and proteasomal degradationof antiviral host factors. It is therefore attractive both as a potentialtherapeutic target for the development of chemical inhibitors andas an E3 ligase that could be recruited by novel PROTACs for targetedprotein degradation. In this study, we used a proteome-scale drug-targetinteraction prediction model, MatchMaker, combined with cheminformaticsfiltering and docking to identify ligands for the DCAF1 WDR domain.Biophysical screening and X-ray crystallographic studies of the predictedbinders confirmed a selective ligand occupying the central cavityof the WDR domain. This study shows that artificial intelligence-enabledvirtual screening methods can successfully be applied in the absenceof previously known ligands.

Automated summary

DCAF1 serves as a subunit for RING-type CRL4(DCAF1) and HECT family EDVPDCAF1 E3 ubiquitin ligases in substrate recruitment. The WDR domain of DCAF1 acts as a binding platform for substrate proteins and is targeted by HIV and SIV lentiviral adaptors. This study used a proteome-scale drug-target interaction prediction model to identify ligands for the DCAF1 WDR domain through biophysical screening and X-ray crystallography, confirming the selective binding of a predicted ligand. The findings demonstrate the successful application of artificial intelligence-enabled virtual screening methods in the absence of previously known ligands.