A deep learning approach for the discovery of tumor-targeting small organic ligands from DNA-Encoded Chemical Libraries

Abstract

AbstractDNA-Encoded Chemical Libraries (DELs) emerged as efficient and cost-effective ligand discovery tools, which enable the generation of protein-ligand interaction data of unprecedented size. In this article, we present an approach that combines DEL screening and instance-level deep learning modeling to identify tumor-targeting ligands against Carbonic Anhydrase IX (CAIX), a clinically validated marker of hypoxia and clear cell Renal Cell Carcinoma. We present a new ligand identification and HIT-to-LEAD strategy driven by Machine Learning (ML) models trained on DELs, which expand the scope of DEL-derived chemical motifs. CAIX screening datasets obtained from three different DELs were used to train ML models for generating novel HITs, dissimilar to elements present in the original DELs. Out of the 152 novel potential HITs that were identified with our approach and screened in anin vitroenzymatic inhibition assay, 70% displayed submicromolar activities (IC50< 1 μM). Based on the first HIT set, the model was further used to prioritize and generate LEAD compounds with nanomolar affinity forin vivotumor-targeting applications. Three LEAD candidates showed accumulation on the surface of CAIX-expressing tumor cells in cellular binding assays. The best compound displayedin vitroKDof 5.7 nM and selectively targeted tumors in mice bearing human Renal Cell Carcinoma lesions. Our results demonstrate the synergy between DEL and machine learning for the identification of novel HITs and for the successful translation of LEAD candidates forin vivotargeting applications.Graphical Abstracts