Comparative study between deep learning and QSAR classifications for TNBC inhibitors and novel GPCR agonist discovery.

Machine learning is a well-known approach for virtual screening. Recently, deep learning, a machine learning algorithm in artificial neural networks, has been applied to the advancement of precision medicine and drug discovery. In this study, we performed comparative studies between deep neural networks (DNN) and other ligand-based virtual screening (LBVS) methods to demonstrate that DNN and random forest (RF) were superior in hit prediction efficiency. By using DNN, several triple-negative breast cancer (TNBC) inhibitors were identified as potent hits from a screening of an in-house database of 165,000 compounds. In broadening the application of this method, we harnessed the predictive properties of trained model in the discovery of G protein-coupled receptor (GPCR) agonist, by which computational structure-based design of molecules could be greatly hindered by lack of structural information. Notably, a potent (~ 500 nM) mu-opioid receptor (MOR) agonist was identified as a hit from a small-size training set of 63 compounds. Our results show that DNN could be an efficient module in hit prediction and provide experimental evidence that machine learning could identify potent hits in silico from a limited training set.

Epitope resurfacing on dengue virus-like particle vaccine preparation to induce broad neutralizing antibody.

Dengue fever is caused by four different serotypes of dengue virus (DENV) which is the leading cause of worldwide arboviral diseases in humans. Virus-like particles (VLPs) containing flavivirus prM/E proteins have been demonstrated to be a potential vaccine candidate; however, the structure of dengue VLP is poorly understood. Herein VLP derived from DENV serotype-2 were engineered becoming highly matured (mD2VLP) and showed variable size distribution with diameter of ~31 nm forming the major population under cryo-electron microscopy examination. Furthermore, mD2VLP particles of 31 nm diameter possess a T = 1 icosahedral symmetry with a groove located within the E-protein dimers near the 2-fold vertices that exposed highly overlapping, cryptic neutralizing epitopes. Mice vaccinated with mD2VLP generated higher cross-reactive (CR) neutralization antibodies (NtAbs) and were fully protected against all 4 serotypes of DENV. Our results highlight the potential of 'epitope-resurfaced' mature-form D2VLPs in inducing quaternary structure-recognizing broad CR NtAbs to guide future dengue vaccine design.

Discovery, structure-activity relationship studies, and anti-nociceptive effects of N-(1,2,3,4-tetrahydro-1-isoquinolinylmethyl)benzamides as novel opioid receptor agonists.

µ-Opioid receptor (MOR) agonists are analgesics used clinically for the treatment of moderate to severe pain, but their use is associated with severe adverse effects such as respiratory depression, constipation, tolerance, dependence, and rewarding effects. In this study, we identified N-({2-[(4-bromo-2-trifluoromethoxyphenyl)sulfonyl]-1,2,3,4-tetrahydro-1-isoquinolinyl}methyl)cyclohexanecarboxamide (1) as a novel opioid receptor agonist by high-throughput screening. Structural modifications made to 1 to improve potency and blood-brain-barrier (BBB) penetration resulted in compounds 45 and 46. Compound 45 was a potent MOR/KOR (κ-opioid receptor) agonist, and compound 46 was a potent MOR and medium KOR agonist. Both 45 and 46 demonstrated a significant anti-nociceptive effect in a tail-flick test performed in wild type (WT) B6 mice. The ED50 value of 46 was 1.059 mg/kg, and the brain concentrations of 45 and 46 were 7424 and 11696 ng/g, respectively. Accordingly, compounds 45 and 46 are proposed for lead optimization and in vivo disease-related pain studies.