From Complex Natural Products to Simple Synthetic Mimetics by Computational De Novo Design

We present the computational de novo design of synthetically accessible chemical entities that mimic the complex sesquiterpene natural product (?)‐Englerin A. We synthesized lead‐like probes from commercially available building blocks and profiled them for activity against a computationally predicted panel of macromolecular targets. Both the design template (?)‐Englerin A and its low‐molecular weight mimetics presented nanomolar binding affinities and antagonized the transient receptor potential calcium channel TRPM8 in a cell‐based assay, without showing target promiscuity or frequent‐hitter properties. This proof‐of‐concept study outlines an expeditious solution to obtaining natural‐product‐inspired chemical matter with desirable properties.     

Deorphaning Pyrrolopyrazines as Potent Multi‐Target Antimalarial Agents

The discovery of pyrrolopyrazines as potent antimalarial agents is presented, with the most effective compounds exhibiting EC₅₀ values in the low nanomolar range against asexual blood stages of Plasmodium falciparum in human red blood cells, and Plasmodium berghei liver schizonts, with negligible HepG2 cytotoxicity. Their potential mode of action is uncovered by predicting macromolecular targets through avant‐garde computer modeling. The consensus prediction method suggested a functional resemblance between ligand binding sites in non‐homologous target proteins, linking the observed parasite elimination to IspD, an enzyme from the non‐mevalonate pathway of isoprenoid biosynthesis, and multi‐kinase inhibition. Further computational analysis suggested essential P. falciparum kinases as likely targets of our lead compound. The results obtained validate our methodology for ligand‐ and structure‐based target prediction, expand the bioinformatics toolbox for proteome mining, and provide unique access to deciphering polypharmacological effects of bioactive chemical agents.     

De Novo Fragment Design for Drug Discovery and Chemical Biology

Automated molecular de novo design led to the discovery of an innovative inhibitor of death‐associated protein kinase 3 (DAPK3). An unprecedented crystal structure of the inactive DAPK3 homodimer shows the fragment‐like hit bound to the ATP pocket. Target prediction software based on machine learning models correctly identified additional macromolecular targets of the computationally designed compound and the structurally related marketed drug azosemide. The study validates computational de novo design as a prime method for generating chemical probes and starting points for drug discovery.     

A Computational Method for Unveiling the Target Promiscuity of Pharmacologically Active Compounds

Drug discovery is governed by the desire to find ligands with defined modes of action. It has been realized that even designated selective drugs may have more macromolecular targets than is commonly thought. Consequently, it will be mandatory to consider multitarget activity for the design of future medicines. Computational models assist medicinal chemists in this effort by helping to eliminate unsuitable lead structures and spot undesired drug effects early in the discovery process. Here, we present a straightforward computational method to find previously unknown targets of pharmacologically active compounds. Validation experiments revealed hitherto unknown targets of the natural product resveratrol and the nonsteroidal anti‐inflammatory drug celecoxib. The obtained results advocate machine learning for polypharmacology‐based molecular design, drug re‐purposing, and the “de‐orphaning” of phenotypic drug effects.     

Optimization of TRPV6 Calcium Channel Inhibitors Using a 3D Ligand‐Based Virtual Screening Method

Herein, we report the discovery of the first potent and selective inhibitor of TRPV6, a calcium channel overexpressed in breast and prostate cancer, and its use to test the effect of blocking TRPV6‐mediated Ca²⁺‐influx on cell growth. The inhibitor was discovered through a computational method, xLOS, a 3D‐shape and pharmacophore similarity algorithm, a type of ligand‐based virtual screening (LBVS) method described briefly here. Starting with a single weakly active seed molecule, two successive rounds of LBVS followed by optimization by chemical synthesis led to a selective molecule with 0.3 μM inhibition of TRPV6. The ability of xLOS to identify different scaffolds early in LBVS was essential to success. The xLOS method may be generally useful to develop tool compounds for poorly characterized targets.     

Deorphaning the Macromolecular Targets of the Natural Anticancer Compound Doliculide

The cyclodepsipeptide doliculide is a marine natural product with strong actin‐polymerizing and anticancer activities. Evidence for doliculide acting as a potent and subtype‐selective antagonist of prostanoid E receptor 3 (EP3) is presented. Computational target prediction suggested that this membrane receptor is a likely macromolecular target and enabled immediate in vitro validation. This proof‐of‐concept study demonstrates the in silico deorphanization of phenotypic screening hits as a viable concept for future natural‐product‐inspired chemical biology and drug discovery efforts.     

Chemoproteomic Approach to Explore the Target Profile of GPCR ligands: Application to 5‐HT1A and 5‐HT6 Receptors

Determination of the targets of a compound remains an essential aspect in drug discovery. A complete understanding of all binding interactions is critical to recognize in advance both therapeutic effects and undesired consequences. However, the complete polypharmacology of many drugs currently in clinical development is still unknown, especially in the case of G protein‐coupled receptor (GPCR) ligands. In this work we have developed a chemoproteomic platform based on the use of chemical probes to explore the target profile of a compound in biological systems. As proof of concept, this methodology has been applied to selected ligands of the therapeutically relevant serotonin 5‐HT_1A and 5‐HT₆ receptors, and we have identified and validated some of their off‐targets. This approach could be extended to other drugs of interest to study the targeted proteome in disease‐relevant systems.     

An evaluation of experimental design in QSAR modelling utilizing the k‐medoid clustering

A reliable selection of a representative subset of chemical compounds has been reported to be crucial for numerous tasks in computational chemistry and chemoinformatics. We investigated the usability of an approach on the basis of the k‐medoid algorithm for this task and in particular for experimental design and the split between training and validation set. We therefore compared the performance of models derived from such a selection to that of models derived using several other approaches, such as space‐filling design and D‐optimal design. We validated the performance on four datasets with different endpoints, representing toxicity, physicochemical properties and others. Compared with the models derived from the compounds selected by the other examined approaches, those derived with the k‐medoid selection show a high reliability for experimental design, as their performance was constantly among the best for all examined datasets. Of all the models derived with all examined approaches, those derived with the k‐medoid approach were the only ones that showed a significantly improved performance compared with a random selection, for all datasets, the whole examined range of selected compounds and for each dimensionality of the search space. Copyright © 2012 John Wiley & Sons, Ltd.     

From conventional to unusual enzyme inhibitor scaffolds: the quest for target specificity

The tremendous challenge presented by the specific molecular recognition of single biomacromolecular targets within complex biological systems demands novel and creative design strategies. This Minireview discusses some conventional and unusual approaches for the design of target-selective enzyme inhibitors with a focus on the underlying chemical scaffolds. These include complicated natural-product-like organic molecules, stable octahedral metal complexes, fullerenes, carboranes, polymetallic clusters, and even polymers. Thus the whole repertoire of organic, inorganic, and macromolecular chemistry can be applied to tackle the problem of target-specific enzyme inhibition.     

Chemoproteomic Evaluation of the Polyacetylene Callyspongynic Acid

Polyacetylenes are a class of alkyne‐containing natural products. Although potent bioactivities and thus possible applications as chemical probes have already been reported for some polyacetylenes, insights into the biological activities or molecular mode of action are still rather limited in most cases. To overcome this limitation, we describe the application of the polyacetylene callyspongynic acid in the development of an experimental roadmap for characterizing potential protein targets of alkyne‐containing natural products. To this end, we undertook the first chemical synthesis of callyspongynic acid. We then used in situ chemical proteomics methods to demonstrate extensive callyspongynic acid‐mediated chemical tagging of endoplasmic reticulum‐associated lipid‐metabolizing and modifying enzymes. We anticipate that an elucidation of protein targets of natural products may serve as an effective guide to the development of subsequent biological assays that aim to identify chemical phenotypes and bioactivities.     

Customizable de novo design strategies for DOCK: Application to HIVgp41 and other therapeutic targets

De novo design can be used to explore vast areas of chemical space in computational lead discovery. As a complement to virtual screening, from‐scratch construction of molecules is not limited to compounds in pre‐existing vendor catalogs. Here, we present an iterative fragment growth method, integrated into the program DOCK, in which new molecules are built using rules for allowable connections based on known molecules. The method leverages DOCK's advanced scoring and pruning approaches and users can define very specific criteria in terms of properties or features to customize growth toward a particular region of chemical space. The code was validated using three increasingly difficult classes of calculations: (1) Rebuilding known X‐ray ligands taken from 663 complexes using only their component parts (focused libraries), (2) construction of new ligands in 57 drug target sites using a library derived from ∼13M drug‐like compounds (generic libraries), and (3) application to a challenging protein‐protein interface on the viral drug target HIVgp41. The computational testing confirms that the de novo DOCK routines are robust and working as envisioned, and the compelling results highlight the potential utility for designing new molecules against a wide variety of important protein targets. © 2017 Wiley Periodicals, Inc.     

Selection of DNA‐Encoded Small Molecule Libraries Against Unmodified and Non‐Immobilized Protein Targets

The selection of DNA‐encoded libraries against biological targets has become an important discovery method in chemical biology and drug discovery, but the requirement of modified and immobilized targets remains a significant disadvantage. With a terminal protection strategy and ligand‐induced photo‐crosslinking, we show that iterated selections of DNA‐encoded libraries can be realized with unmodified and non‐immobilized protein targets.     

Lessons To Be Learned: The Molecular Basis of Kinase‐Targeted Therapies and Drug Resistance in Non‐Small Cell Lung Cancer

The treatment of non‐small cell lung cancer (NSCLC) is currently experiencing a revolution. Over the last decade, the knowledge gained about the biochemical features of biomarkers and their predictive abilities has led to the development of targeted small‐molecule inhibitors that present an alternative to harsh chemotherapy. The use of these new therapies has improved the quality of life and increased the survival of patients. The occurrence of inevitable drug resistance requires the constant development of precision medicine. The detailed understanding of the target biology and the search for innovative chemical approaches has encouraged investigations in this field. Herein, we review selected aspects of the molecular targets and present an overview of current topics and challenges in the rational development of small molecules to target NSCLC.     

Structure activity relationship by NMR and by computer: a comparative study

There has recently been considerable interest in using NMR spectroscopy to identify ligand binding sites of macromolecules. In particular, a modular approach has been put forward by Fesik et al. (Shuker, S. B.; Hajduk, P. J.; Meadows, R. P.; Fesik, S. W. Science 1996, 274, 1531-1534) in which small ligands that bind to a particular target are identified in a first round of screening and subsequently linked together to form ligands of higher affinity. Similar strategies have also been proposed for in silico drug design, where the binding sites of small chemical groups are identified, and complete ligands are subsequently assembled from different groups that have favorable interactions with the macromolecular target. In this paper, we compare experimental and computational results on a selected target (FKBP12). The binding sites of three small ligands ((2S)1-acetylprolinemethylester, 1-formylpiperidine, 1-piperidinecarboxamide) in FKBP12 were identified independently by NMR and by computational methods. The subsequent comparison of the experimental and computational data showed that the computational method identified and ranked favorably ligand positions that satisfy the experimental NOE constraints.     

Dynamic Combinatorial Chemistry: A New Methodology Comes of Age

Dynamic combinatorial chemistry (DCC) has repeatedly proven to be an effective approach to generate directed ligand libraries for macromolecular targets. In the absence of an external stimulus, a dynamic library forms from reversibly reacting building blocks and reaches a stable thermodynamic equilibrium. However, upon addition of a macromolecular host which can bind and stabilize certain components of the library, the equilibrium composition changes and induces an evolution-like selection and enrichment of high-affinity ligands. A valuable application of this so-called target-directed DCC (tdDCC) is the identification of potent ligands for pharmacologically relevant targets. Over time, the term tdDCC has been applied to describe a number of different experimental setups, leading to some ambiguity concerning its definition. This article systematically classifies known procedures for tdDCC and related approaches, with a special focus on the methods used for analysis and evaluation of experiments.     

Multi‐Objective Molecular De Novo Design by Adaptive Fragment Prioritization

We present the development and application of a computational molecular de novo design method for obtaining bioactive compounds with desired on‐ and off‐target binding. The approach translates the nature‐inspired concept of ant colony optimization to combinatorial building block selection. By relying on publicly available structure–activity data, we developed a predictive quantitative polypharmacology model for 640 human drug targets. By taking reductive amination as an example of a privileged reaction, we obtained novel subtype‐selective and multitarget‐modulating dopamine D₄ antagonists, as well as ligands selective for the sigma‐1 receptor with accurately predicted affinities. The nanomolar potencies of the hits obtained, their high ligand efficiencies, and an overall success rate of 90 % demonstrate that this ligand‐based computer‐aided molecular design method may guide target‐focused combinatorial chemistry.     

Repositioning of Quinazolinedione-Based Compounds on Soluble Epoxide Hydrolase (sEH) through 3D Structure-Based Pharmacophore Model-Driven Investigation

The development of new bioactive compounds represents one of the main purposes of the drug discovery process. Various tools can be employed to identify new drug candidates against pharmacologically relevant biological targets, and the search for new approaches and methodologies often represents a critical issue. In this context, in silico drug repositioning procedures are required even more in order to re-evaluate compounds that already showed poor biological results against a specific biological target. 3D structure-based pharmacophoric models, usually built for specific targets to accelerate the identification of new promising compounds, can be employed for drug repositioning campaigns as well. In this work, an in-house library of 190 synthesized compounds was re-evaluated using a 3D structure-based pharmacophoric model developed on soluble epoxide hydrolase (sEH). Among the analyzed compounds, a small set of quinazolinedione-based molecules, originally selected from a virtual combinatorial library and showing poor results when preliminarily investigated against heat shock protein 90 (Hsp90), was successfully repositioned against sEH, accounting the related built 3D structure-based pharmacophoric model. The promising results here obtained highlight the reliability of this computational workflow for accelerating the drug discovery/repositioning processes.     

Novel Big Data-Driven Machine Learning Models for Drug Discovery Application

Most contemporary drug discovery projects start with a ‘hit discovery’ phase where small chemicals are identified that have the capacity to interact, in a chemical sense, with a protein target involved in a given disease. To assist and accelerate this initial drug discovery process, ’virtual docking calculations’ are routinely performed, where computational models of proteins and computational models of small chemicals are evaluated for their capacities to bind together. In cutting-edge, contemporary implementations of this process, several conformations of protein targets are independently assayed in parallel ‘ensemble docking’ calculations. Some of these protein conformations, a minority of them, will be capable of binding many chemicals, while other protein conformations, the majority of them, will not be able to do so. This fact that only some of the conformations accessible to a protein will be ’selected’ by chemicals is known as ’conformational selection’ process in biology. This work describes a machine learning approach to characterize and identify the properties of protein conformations that will be selected (i.e., bind to) chemicals, and classified as potential binding drug candidates, unlike the remaining non-binding drug candidate protein conformations. This work also addresses the class imbalance problem through advanced machine learning techniques that maximize the prediction rate of potential protein molecular conformations for the test case proteins ADORA2A (Adenosine A2a Receptor) and OPRK1 (Opioid Receptor Kappa 1), and subsequently reduces the failure rates and hastens the drug discovery process.     

Isoxazolone Reactivity Explained by Computed Electronic Structure Analysis

Summary of main observation and conclusion lsoxazol(2H)-5-one was chosen as a model molecule to study the structural features(a,p angles and carbonyl bond length)regarding lactone moiety in the isoxazolone species by computational calculation.DFT method with B3LYP 6-311++G(2df,2p)basis set was used to carry out the optimization on a series(A and B families)of isoxazolones suitably substituted on double bond.     

Probing the range of applicability of structure‐ and energy‐adjusted QM/MM link bonds

The hydrogen‐capping method is one of the most popular and widely used coupling‐schemes for quantum mechanics/molecular mechanics (QM/MM)‐molecular dynamics simulations of macromolecular systems. This is mostly due to the fact that it is fairly convenient to implement and parametrize, thus providing an excellent compromise between accuracy and computational effort. In this work, a viable and straight‐forward approach to optimize the placing of the link atom on a suitable distance ratio between the frontier atoms is discussed. To further increase the accuracy, instead of global parameters for all amino acids, different parameter sets for each type of amino acid are derived. The dependency of the link bond parameters on the chemical environment and the used QM‐method is probed to assess the range of applicability of the parametrization. Suitable sets of parameters for RI‐MP2, B3LYP, (RI)‐B3LYP‐D3, and RI‐BLYP‐D3 at triple‐zeta level for all relevant proteinogenic amino acids are presented. Furthermore, the scope and range of the perturbation, stemming from the introduction of link bonds is evaluated through application of the presented QM/MM scheme in calculations of the active site of 15S‐lipoxygenase. © 2015 Wiley Periodicals, Inc.