Deep Learning in Drug Design: Protein-Ligand Binding Affinity Prediction.

Computational drug design relies on the calculation of binding strength between two biological counterparts especially a chemical compound, i.e., a ligand, and a protein. Predicting the affinity of protein-ligand binding with reasonable accuracy is crucial for drug discovery, and enables the optimization of compounds to achieve better interaction with their target protein. In this paper, we propose a data-driven framework named DeepAtom to accurately predict the protein-ligand binding affinity. With 3D Convolutional Neural Network (3D-CNN) architecture, DeepAtom could automatically extract binding related atomic interaction patterns from the voxelized complex structure. Compared with the other CNN based approaches, our light-weight model design effectively improves the model representational capacity, even with the limited available training data. We carried out validation experiments on the PDBbind v.2016 benchmark and the independent Astex Diverse Set. We demonstrate that the less feature engineering dependent DeepAtom approach consistently outperforms the other baseline scoring methods. We also compile and propose a new benchmark dataset to further improve the model performances. With the new dataset as training input, DeepAtom achieves Pearson's R=0.83 and RMSE=1.23 pK units on the PDBbind v.2016 core set. The promising results demonstrate that DeepAtom models can be potentially adopted in computational drug development protocols such as molecular docking and virtual screening.

Discovery of Amantamide, a Selective CXCR7 Agonist from Marine Cyanobacteria.

CXCR7 plays an emerging role in several physiological processes. A linear peptide, amantamide (1), was isolated from marine cyanobacteria, and the structure was determined by NMR and mass spectrometry. The total synthesis was achieved by solid-phase method. After screening two biological target libraries, 1 was identified as a selective CXCR7 agonist. The selective activation of CXCR7 by 1 could provide the basis for developing CXCR7-targeted therapeutics and deciphering the role of CXCR7 in different diseases.

Discovery, Synthesis, Pharmacological Profiling, and Biological Characterization of Brintonamides A-E, Novel Dual Protease and GPCR Modulators from a Marine Cyanobacterium.

Five novel modified linear peptides named brintonamides A-E (1-5) were discovered from a marine cyanobacterial sample collected from Brinton Channel, Florida Keys. The total synthesis of 1-5 in addition to two other structurally related analogues (6 and 7) was achieved, which provided more material to allow rigorous biological evaluation and SAR studies. Compounds were subjected to cancer-focused phenotypic cell viability and migration assays and orthogonal target-based pharmacological screening platforms to identify their protease and GPCR modulatory activity profiles. The cancer related serine protease kallikrein 7 (KLK7) was inhibited to similar extents with an IC50 near 20 µM by both representative members 1 and 4, which differed in the presence or lack of the N-terminal unit. In contrast to the biochemical protease profiling study, clear SAR was observed in the functional GPCR screens, where five GPCRs in antagonist mode (CCR10, OXTR, SSTR3, TACR2) and agonist mode (CXCR7) were modulated by compounds 1-7 to varying extents. Chemokine receptor type 10 (CCR10) was potently modulated by brintonamide D (4) with an IC50 of 0.44 µM. We performed in silico modeling to understand the structural basis underlying the differences in the antagonistic activity among brintonamides toward CCR10. Because of the significance of KLK7 and CCR10 in cancer progression and metastasis, we demonstrated the ability of brintonamide D (4) at 10 µM to significantly target downstream cellular substrates of KLK7 (Dsg-2 and E-cad) in vitro and to inhibit CCL27-induced CCR10-mediated proliferation and the migration of highly invasive breast cancer cells.