Current Drug Metabolism (v.15, #5)

Editorial (Thematic Issue: Chemoinformatics in Metabolomics, From Molecular Mechanics, Dynamics, and Docking to Complex Metabolic Networks, Part 2) by Humberto Gonzalez-Diaz, Alejandro Speck-Planche, Maria Natalia Dias Soeiro Cordeiro (489-489).

State of the Art and Development of a Drug-Drug Interaction Large Scale Predictor Based on 3D Pharmacophoric Similarity by Santiago Vilar, Eugenio Uriarte, Lourdes Santana, Carol Friedman, Nicholas P. Tatonetti (490-501).
Co-administration of drugs is a primary cause of Adverse Drug Reactions (ADRs) and a drain on the health care industry costingbillions of dollars and reducing quality of life. Drug-Drug Interactions (DDIs) account for as much as 30% of all ADRs. Unfortunately,DDIs are not systematically explored pre-clinically and are difficult to detect in post-marketing drug surveillance. For this reason,the detection and prediction of DDIs is an important problem in both drug development and pharmacovigilance. The comparison of the3D drug structures provides a powerful tool for DDI prediction. In this article, we present the first large scale model for predicting DDIsusing the drug's 3D molecular structure. In addition to identifying putative drug interactions we can also isolate the pharmacological orclinical effect associated with the predicted interactions. The model has good performance in two different hold-out validations and in externaltest sets. We found that the top scored drug pairs were significantly enriched for known clinically relevant interactions and that 3Dstructure data is providing significantly independent information from other approaches, including 2D structure (p=0.003). We demonstratedthe usefulness of the proposed methodology to systematically identify pharmacokinetic and pharmacodynamic interactions, providedan exploratory tool that can be used for patient safety and pre-clinical toxicity screening, and reviewed the state of the art methodsused to detect DDIs.

As a kind of monooxygenase with the function of catalyzing many reactions involved in drug metabolism and synthesis ofcholesterol, steroids and other lipids, CYP2J2 is an important member of the cytochrome P450 superfamily. Located at the endoplasmicreticulum, CYP2J2 is responsible for epoxidation of endogenous arachidonic acid in cardiac tissue to produce cis-epoxyeicosatrienoic acids(EETs), which have anti-inflammatory and antifibrinolytic properties, and can protect endothelial cells from ischemic or hypoxic injuries.Some polymorphisms, e.g., CYP2J2 with mutation T143A, R158C, I192N or N404Y, could significantly reduce the metabolism ofthe arachidonic acid, causing or deteriorating the coronary artery disease. However, so far the detailed mechanism for the mutationinduceddysfunction of arachidonic metabolism is still unknown. To reveal its mechanism, a 3D (three-dimensional) structure for humanCYP2J2 was developed, followed by docking the arachidonic acid ligand into the active site of the receptor. It was observed based on thebinding mode thus found that Gly486 and Leu378 in the active site of the receptor played a key role in recognizing and positioning thecarboxyl group of the ligand via hydrogen bonding interactions, and that any of the aforementioned five mutations might have, eitherdirectly or indirectly, impact to their role and hence causing the mutation-induced dysfunction of CYP2J2-mediated arachidonic acidmetabolism. It is anticipated that the findings as reported in this review article may stimulate new strategy for finding novel therapeuticapproaches to treat coronary artery disease.

In silico Prediction of Drug Metabolism by P450 by Carolina H. Andrade, Diego C. Silva, Rodolpho C. Braga (514-525).
In the drug discovery cascade, metabolism studies should be performed as early as possible to allow an early evaluation of themetabolism profiles of drug candidates. To help design new drug candidates with improved pharmacokinetics, the knowledge of the siteof metabolism is necessary. Computational or in silico metabolism approaches can be broadly classified into (i) ligand-based methods,and (ii) structure-based methods. This review highlight tools used to predict P450-mediated metabolism including ligand-based andstructure-based approaches. Some examples of successful application of an integrated in silico approach for the prediction of Phase I metabolismfor some flavonoids and lead compounds are presented. Moreover, an integrated in silico approach for the prediction of P450-mediated metabolism is described.

The M2 proton channel is translated by the M gene segment of influenza viruses, and has been adopted as an attractive targetfor influenza A viruses, on which a series of adamantane-based drugs act. However, recently epidemic influenza viruses have had strongresistant effects against the adamantane-based drugs. In this paper, we combined evolutionary analyses, linkage disequilibrium as well asmolecular dynamics simulations to explore the drug resistance of the M2 proton channel, with an aim of providing an in-depth understandingof the resistant mechanism for adamantane-based drugs. We collected 2746 coding sequences for swine, avian, and human M2proteins. After evolutionary and linkage disequilibrium analyses, we found that the some residues in the C-terminal were associated withthe famed resistant mutation S31N. Subsequently, we constructed the 3D structures of the swine, avian as well as human M2 channel,and performed MD simulations on these channels with a typical adamantane-based drug rimantadine. From the simulation trajectories,we found that the resistance against the adamantane-based drugs for the M2 channel from 2009 A(H1N1) viruses was derived from thestructural allostery in the transmembrane and C-terminal regions. The helices in the transmembrane region were irregular in formationand employed larger distances between the adjacent 2 helices, which can weaken the interactions between the adjacent 2 helices and destabilizethe helix-helix assembly, resulting in a comparatively loosely structure. The helices in the C-terminal region show a disorderedconfiguration, giving chances for solvent molecules to enter into the channel pore.

Plants and their natural components sophisticated with the cornerstone of traditional conventional medicinal system throughoutthe globe for many years and extend to furnish mankind with latest remedies. Natural Products act as lead molecules for the synthesisof various potent drugs. In the current research a study is conducted on herbal small molecule and their potential binding chemical affinityto the effect or molecules of major diseases such as pancreatic cancer. Clinical studies demonstrate correlation between Cyclin-Dependent Kinase 4 (CDK4) and malignant progression of Pancreatic Cancer. Using Bioruby Gem's we were able to analyze better characteristicsof the target protein. VegaZZ and NAMD were used to minimize the energy of the target protein. Therefore identification ofeffective, well- tolerated targets was analyzed. Further the target protein was subjected to docking with the anti cancer inhibitors whichrepresents a rational chemo preventive strategy using AutoDock Vina. Later using the dock score top ranked phytochemicals were analyzedfor Toxicity Analysis. Using the BioRuby gem we were able to measure the distance between the amino acid. Various R scriptinglibraries were used to hunt the best leads, as in this case the phytochemicals. Phytochemicals such as Wedelolactones and Catechin wereanalyzed computationally. This study has presented the various effects of naturally occurring anti pancreatic cancer compounds Catechin,Wedelolactones that inhibits Cyclin Dependent Kinase 4. The study results reveal that compounds use less binding energy to CDK4 andinhibit its activity. Future investigation of other various wet lab studies such as cell line studies will confirm results of these two herbalchemical formulations potential ones for treating Pancreatic Cancer.

Review and Research on Feature Selection Methods from NMR Data in Biological Fluids. Presentation of an Original Ensemble Method Applied to Atherosclerosis Field by Nabil Semmar, Cecile Canlet, Bernadette Delplanque, Pascale Le Ruyet, Alain Paris, Jean-Charles Martin (544-556).
Metabolic pools of biological matrices can be extensively analyzed by NMR. Measured data consist of hundreds of NMR signalswith different chemical shifts and intensities representing different metabolites' types and levels, respectively. Relevant predictiveNMR signals need to be extracted from the pool using variable selection methods. This paper presents both a review and research on thismetabolomics field. After reviews on discriminant potentials and statistical analyses of NMR data in biological fields, the paper presentsan original approach to extract a small number of NMR signals in a biological matrix A (BM-A) in order to predict metabolic levels inanother biological matrix B (BM-B). Initially, NMR dataset of BM-A was decomposed into several row-column homogeneous blocks usinghierarchical cluster analysis (HCA). Then, each block was subjected to a complete set of Jackknifed correspondence analysis (CA) byremoving separately each individual (row). Each CA condensed the numerous NMR signals into some principal components (PCs). Thedifferent PCs representing the (n - 1) active individuals were used as latent variables in a stepwise multi-linear regression to predictmetabolic levels in BM-B. From the built regression model, metabolite level in the outside individual was predicted (for next model validation).From all the PCs-based regression models resulting from all the jackknifed CA applied on all the individuals, the most contributiveNMR signals were identified by their highest absolute contributions to PCs. Finally, these selected NMR signals (measured in BMA)were used to build final population and sub-population regression models predicting metabolite levels in BM-B.

Galvez-Markov Network Transferability Indices: Review of Classic Theory and New Model for Perturbations in Metabolic Reactions by Jorge Vergara-Galicia, Francisco J. Prado-Prado, Humberto Gonzalez-Diaz (557-564).
Topological Indices (TIs) are numerical parameters useful to carry out Quantitative Structure-Property Relationships (QSPR)analysis and predict the effect of perturbations in many types of Complex Networks. This work, focuses on a very powerful class of TIscalled Galvez charge transfer indices. First, we review the classic concept and some applications of these indices. Next, we review theGalvez-Markov TIs of order k (GMk), a recent generalization to these TIs introduced by us. We also reviewed some previous examples ofcalculation of GMk values for different classes of networks, including metabolic networks. Here, we also demonstrated that Galvez-Markov TIs are useful to predict perturbations and the transferability of biochemical patterns forms metabolic networks of species to others.We report a linear QSPR-Perturbation theory model that predicts more than 300,000 perturbations in metabolic networks with 85 -99% of good classification in training and validation series.