Abstract:The highly polymorphic human cytochrome P450 2D6 (CYP2D6) metabolizes about 25% of currently used drugs. In this study, we have explored the interaction of a large number of substrates (n = 120) with wild-type and mutated CYP2D6 by molecular docking using the CDOCKER module. Before we conducted the molecular docking and virtual mutations, the pharmacophore and QSAR models of CYP2D6 substrates were developed and validated. Finally, we explored the interaction of a traditional Chinese herbal formula, Fangjifuling decoction, with CYP2D6 by virtual screening. The optimized pharmacophore model derived from 20 substrates of CYP2D6 contained two hydrophobic features and one hydrogen bond acceptor feature, giving a relevance ratio of 76% when a validation set of substrates were tested. However, our QSAR models gave poor prediction of the binding affinity of substrates. Our docking study demonstrated that 117 out of 120 substrates could be docked into the active site of CYP2D6. Forty one out of 117 substrates (35.04%) formed hydrogen bonds with various active site residues of CYP2D6 and 53 (45.30%) substrates formed a strong pi-pi interaction with Phe120 (53/54), with only carvedilol showing pi-pi interaction with Phe483. The active site residues involving hydrogen bond formation with substrates included Leu213, Lys214, Glu216, Ser217, Gln244, Asp301, Ser304, Ala305, Phe483, and Phe484. Furthermore, the CDOCKER algorithm was further applied to study the impact of mutations of 28 active site residues (mostly non-conserved) of CYP2D6 on substrate binding modes using five probe substrates including bufuralol, debrisoquine, dextromethorphan, sparteine, and tramadol. All mutations of the residues examined altered the hydrogen bond formation and/or aromatic interactions, depending on the probe used in molecular docking. Apparent changes of the binding modes have been observed with the Glu216Asp and Asp301Glu mutants. Overall, 60 compounds out of 130 from Fangjifuling decoction matched our pharmacophore model for CYP2D6 substrates. Fifty four out of these 60 compounds could be docked into the active site of CYP2D6 and 24 of 54 compounds formed hydrogen bonds with Glu216, Asp301, Ser304, and Ala305 in CYP2D6. These results have provided further insights into the factors that determining the binding modes of substrates to CYP2D6. Screening of high-affinity ligands for CYP2D6 from herbal formula using computational models is a useful approach to identify potential herb-drug interactions.

Prediction of Sites of Metabolism of CYP3A4 Substrates Utilizing Docking-Derived Geometric Features

Structure-Based Site of Metabolism (SOM) Prediction of Ligand for CYP3A4 Enzyme: Comparison of Glide XP and Induced Fit Docking (IFD)

Assessment of CYP2C9 Structural Models for Site of Metabolism Prediction

Prediction of Cytochrome P450-Mediated Metabolism Using a Combination of QSAR Derived Reactivity and Induced Fit Docking

Effects of Protein Flexibility on the Site of Metabolism Prediction for Cyp2a6 Substrates

Computational Prediction of Site of Metabolism for UGT-catalyzed Reactions.

Effects of protein flexibility and active site water molecules on the prediction of sites of metabolism for cytochrome P450 2C19 substrates.

A Comprehensive Analysis of Human CYP3A4 Crystal Structures as a Potential Tool for Molecular Docking-Based Site of Metabolism and Enzyme Inhibition Studies

Pharmacophore, Qsar, and Binding Mode Studies of Substrates of Human Cytochrome P450 2d6 (cyp2d6) Using Molecular Docking and Virtual Mutations and an Application to Chinese Herbal Medicine Screening

Unraveling the Ligand-Binding Sites of CYP3A4 by Molecular Dynamics Simulations with Solvent Probes

Classification models for predicting cytochrome p450 enzyme-substrate selectivity

Three-Dimensional Quantitative Structure Activity Relationship for Cyp2d6 Substrates

Metabolic Behavior Prediction of Pazopanib by Cytochrome P450 (CYP) 3A4 by Molecular Docking

Prediction of Cytochrome P450 Inhibition Using a Deep Learning Approach and Substructure Pattern Recognition

Machine Learning Models to Predict Cytochrome P450 2B6 Inhibitors and Substrates

Computational prediction for the metabolism of human UDP-glucuronosyltransferase 1A1 substrates

Survey of Machine Learning Techniques for Prediction of the Isoform Specificity of Cytochrome P450 Substrates.

Improved Prediction of Michaelis Constants in CYP450-Mediated Reactions by Resilient Back Propagation Algorithm.

Establishment and evaluation of prediction models for the discovery of CYP1A2 inhibitors

Admet Evaluation in Drug Discovery. 13. Development of in Silico Prediction Models for P-Glycoprotein Substrates

Computational methods and tools to predict cytochrome P450 metabolism for drug discovery