Human oxidosqualene cyclase (OSC) is one key enzyme in the biosynthesis of cholesterol. It can catalyze the linear-chain 2,3-oxidosqualene to form lanosterol, the tetracyclic (6–6–6–5 members for A–B–C–D rings) cholesterol precursor. It also has been treated as a novel antihyperlipidemia target. In addition, the structural diversity of cyclic terpenes in plants originates from the cyclization of 2,3-oxidosqualene. The enzyme catalytic mechanism is considered to be one of the most complicated ones in nature, and there are a lot of controversies about the mechanism in the past half a century. Herein, state-of-the-art ab initio QM/MM MD simulations are employed to investigate the detailed cyclization mechanism of C-ring and D-ring formation. Our study reveals that the C and D rings are formed near-synchronously from a stable “6–6–5” ring intermediate. Interestingly, the transition state of this concerted reaction presents a “6–6-6” structure motif, while this unstable “6–6-6” structure in our simulations is thought to be a stable intermediate state in most previous hypothetical mechanisms. Furthermore, as the tailed side chain of 2,3-oxidosqualene shows a β conformation while it is α conformation in lanosterol, finally, it is observed that the rotatable “tail” chain prefers to transfer β conformation to α conformation at the “6–6–5” intermediate state.

Terpenes (isoprenoids) represent the most functionally and structurally diverse group of natural products. Terpenes are assembled from two building blocks, isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP or DPP), by prenyltransferases (PTSs). Geranyl pyrophosphate synthase (GPPS) is the enzyme that assembles DPP and IPP in the first step of chain elongation during isoprenoid biosynthesis. The mechanism by which GPPS assembles the terpene precursor remains unknown; elucidating this mechanism will help in development of new technology to generate novel natural product-like scaffolds. With classic and QM/MM MD simulations, an “open-closed” conformation change of the catalytic pocket was observed in the GPPS active site at its large subunit (LSU), and a critical salt bridge between Asp91(in loop 1) and Lys239(in loop 2) was identified. The salt bridge is responsible for opening or closing the catalytic pocket. Meanwhile, the small subunit (SSU) regulates the size and shape of the hydrophobic pocket to flexibly host substrates with different shapes and sizes (DPP/GPP/FPP, C5/C10/C15). Further QM/MM MD simulations were carried out to explore the binding modes for the different substrates catalyzed by GPPS. Our simulations suggest that the key residues (Asp91, Lys239, and Gln156) are good candidates for site-directed mutagenesis and may help in protein engineering.

Shape comparing technologies based on Gaussian functions have been widely used in virtual screening of drug discovery. For efficiency, most of them adopt the First Order Gaussian Approximation (FOGA), in which the shape density of a molecule is represented as a simple sum of all individual atomic shape densities. In the current work, the effectiveness and error in shape similarity calculated by such an approximation are carefully analyzed. A new approach, which is called the Weighted Gaussian Algorithm (WEGA), is proposed to improve the accuracy of the first order approximation. The new approach significantly improves the accuracy of molecular volumes and reduces the error of shape similarity calculations by 37% using the hard-sphere model as the reference. The new algorithm also keeps the simplicity and efficiency of the FOGA. A program based on the new method has been implemented for molecular overlay and shape-based virtual screening. With improved accuracy for shape similarity scores, the new algorithm also improves virtual screening results, particularly when a shape-feature combo scoring function is used.

Human intestinal carboxyl esterase (hiCE) is a drug target for ameliorating irinotecan-induced diarrhea. By reducing irinotecan-induced diarrhea, hiCE inhibitors can improve the anti-cancer efficacy of irinotecan. To find effective hiCE inhibitors, a new virtual screening protocol that combines pharmacophore models derived from the hiCE structure and its ligands has been proposed. The hiCE structure has been constructed through homology techniques using hCES1’s crystal structure. The hiCE structure was optimized via molecular dynamics simulations with the most known active hiCE inhibitors docked into the structure. An optimized pharmacophore, derived from the receptor, was then generated. A ligand-based pharmacophore was also generated from a larger set of known hiCE inhibitors. The final hiCE inhibitor predictions were based upon the virtual screening hits from both ligand-based and receptor-based pharmacophore models. The hit rates from the ligand-based and receptor-based pharmacophore models are 88% and 86%, respectively. The final hit rate is 94%. The two models are highly consistent with one another (85%). This proves that both models are reliable.

High cholesterol levels contribute to hyperlipidemia. Liver X receptors (LXRs) are the drug targets. LXRs regulate the cholesterol absorption, biosynthesis, transportation, and metabolism. Novel agonists of LXR, especially LXRβ, are attractive solutions for treating hyperlipidemia. In order to discover novel LXRβ agonists, a three-dimensional pharmacophore model was built based upon known LXRβ agonists. The model was validated with a test set, a virtual screening experiment, and the FlexX docking approach. Results show that the model is capable of predicting a LXRβ agonist activity. Ligand-based virtual screening results can be refined by cross-linking by structure-based approaches. This is because two ligands that are mapped in the same way to the same pharmacophore model may have significantly different binding behaviors in the receptor’s binding pocket. This paper reports our approach to identify reliable pharmacophore models through combining both ligand- and structure-based approaches.

Glycogen synthase kinase 3β (GSK-3β) is a potential therapeutic target for cancer, type-2 diabetes, and Alzheimer’s disease. This paper proposes a new lead identification protocol that predicts new GSK-3β ATP competitive inhibitors with topologically diverse scaffolds. First, three-dimensional quantitative structure–activity relationship (3D QSAR) models were built and validated. These models are based upon known GSK-3β inhibitors, benzofuran-3-yl-(indol-3-yl) maleimides, by means of comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA). Second, 28 826 maleimide derivatives were selected from the PubChem database. After filtration via Lipinski’s rules, 10 429 maleimide derivatives were left. Third, the FlexX-dock program was employed to virtually screen the 10 429 compounds against GSK-3β. This resulted in 617 virtual hits. Fourth, the 3D QSAR models predicted that from the 617 virtual hits, 93 compounds would have GSK-3β inhibition values of less than 15 nM. Finally, from the 93 predicted active hits, 23 compounds were confirmed as GSK-3β inhibitors from literatures; their GSK-3β inhibition ranged from 1.3 to 480 nM. Therefore, the hits rate of our virtual screening protocol is greater than 25%. The protocol combines ligand- and structure-based approaches and therefore validates both approaches and is capable of identifying new hits with topologically diverse scaffolds.

Recently, reversible antagonists of the P2Y12 receptor have been reported. However, the mechanisms of binding have not been elucidated. To this end, a number of homology models were built by means of three programs from four templates. A consensus model was derived from those initial models. The final model was created by refining the consensus model with molecular dynamics simulations. The agonist and antagonists of P2Y12 have been docked in the final model. For the agonist, the Arg256, Lys280, and Phe252 are “hot” residues. For the antagonists, the Lys280 and Phe252 are “hot” residues that have hydrogen bonding contacts and π–π interactions, respectively. These results can explain the observations of mutation experiments and can guide the design of new inhibitors.Research highlights► A homology model for P2Y12 receptor was created based on a consensus approach following with molecular dynamics simulations. ► The mechanisms of binding for the reversible antagonists of the P2Y12 receptor are elucidated. ► For the agonist, the Arg256, Lys280, and Phe252 were found as “hot” residues. ► For the antagonists, the Lys280 and Phe252 had hydrogen bonding contacts and π-π interactions.

Covering: up to 2010

Aurora kinases (A–C) belong to the serine/threonine protein kinase family. In recent years, the constitutive or elevated expression of Aurora kinases has been found in cancer cells and oncogene transfected cells. In this review, we summarize the common binding modes of Aurora-A kinase inhibitors, the hot spot residues in the binding sites and the privileged inhibitor structures. Our review of the reported chemical scaffolds of Aurora-A kinase inhibitors and their binding modes could provide a useful framework from which new design strategies for inhibitors might be assessed or developed.