Mining and visualizing large anticancer drug discovery databases

In order to find more effective anticancer drugs, the U.S. National Cancer Institute (NCI) screens a large number of compounds in vitro against 60 human cancer cell lines from different organs of origin. About 70,000 compounds have been tested in the program since 1990, and each tested compound can be characterized by a vector (i.e., "fingerprint") of 60 anticancer activity, or -[log(GI50)], values. GI50 is the concentration required to inhibit cell growth by 50% compared with untreated controls. Although cell growth inhibitory activity for a single cell line is not very informative, activity patterns across the 60 cell lines can provide incisive information on the mechanisms of action of screened compounds and also on molecular targets and modulators of activity within the cancer cells. Various statistical and artificial intelligence methods, including principal component analysis, hierarchical cluster analysis, stepwise linear regression, multidimensional scaling, neural network modeling, and genetic function approximation, among others, can be used to analyze this large activity database. Mining the database can provide useful information: (a) for the development of anticancer drugs; (b) for a better understanding of the molecular pharmacology of cancer; and (c) for improvement of the drug discovery process.     

Mass spectrometry based tools to investigate protein-ligand interactions for drug discovery

The initial stages of drug discovery are increasingly reliant on development and improvement of analytical methods to investigate protein-protein and protein-ligand interactions. For over 20 years, mass spectrometry (MS) has been recognized as providing a fast, sensitive and high-throughput methodology for analysis of weak non-covalent complexes. Careful control of electrospray ionization conditions has enabled investigation of the structure, stability and interactions of proteins and peptides in a solvent free environment. This critical review covers the use of mass spectrometry for kinetic, dynamic and structural studies of proteins and protein complexes. We discuss how conjunction of mass spectrometry with related techniques and methodologies such as ion mobility, hydrogen-deuterium exchange (HDX), protein footprinting or chemical cross-linking can provide us with structural information useful for drug development. Along with other biophysical techniques, such as NMR or X-ray crystallography, mass spectrometry provides a powerful toolbox for investigation of biological problems of medical relevance (204 references).     

Use of drug discovery tools in rational organometallic catalyst design

A computational procedure is detailed where techniques common in the drug discovery process-2D- and 3D-quantitative structure-activity relationships (QSAR)-are applied to rationalize the catalytic activity of a synthetically flexible, Ti-N=P ethylene polymerization catalyst system. Once models relating molecular properties to catalyst activity are built with the two QSAR approaches, two database mining approaches are used to select a small number of ligands from a larger database that are likely to produce catalysts with high activity when grafted onto the Ti-N=P framework. The software employed throughout this work is freely available, is easy to use, and was applied in a "black box" approach to highlight areas where the drug discovery tools, designed to address organic molecules, have difficulty in addressing issues arising from the presence of a metal atom. In general, 3D-QSAR offers an efficient way to screen new potential ligands and separate those likely to lead to poor catalysts from those that are likely to contribute to highly active catalysts. The results for 2D-QSAR appear to be quantitatively unreliable, likely due to the presence of a metal atom; nonetheless, there is evidence that qualitative predictions from different models may be reliable. Pitfalls in the database mining techniques are identified, none of which are insurmountable. The lessons learned about the potential uses and drawbacks of the techniques described herein are readily applicable to other catalyst frameworks, thereby enabling a rational approach to catalyst improvement and design.     

Bigger data,collaborative tools and the future of predictive drug discovery

Over the past decade we have seen a growth in the provision of chemistry data and cheminformatics tools as either free websites or software as a service commercial offerings. These have transformed how we find molecule-related data and use such tools in our research. There have also been efforts to improve collaboration between researchers either openly or through secure transactions using commercial tools. A major challenge in the future will be how such databases and software approaches handle larger amounts of data as it accumulates from high throughput screening and enables the user to draw insights, enable predictions and move projects forward. We now discuss how information from some drug discovery datasets can be made more accessible and how privacy of data should not overwhelm the desire to share it at an appropriate time with collaborators. We also discuss additional software tools that could be made available and provide our thoughts on the future of predictive drug discovery in this age of big data. We use some examples from our own research on neglected diseases, collaborations, mobile apps and algorithm development to illustrate these ideas.     

Heavy tools for genome mining

In this issue of Chemistry & Biology, Gross et al. report development of a novel genome mining method for isolating products of orphan biosynthetic gene clusters, and the application of this method to the isolation of orfamide A, a novel cyclic lipopeptide.     

Data mining the NCI60 to predict generalized cytotoxicity

Elimination of cytotoxic compounds in the early and later stages of drug discovery can help reduce the costs of research and development. Through the application of principal components analysis (PCA), we were able to data mine and prove that approximately 89% of the total log GI 50 variance is due to the nonspecific cytotoxic nature of substances. Furthermore, PCA led to the identification of groups of structurally unrelated substances showing very specific toxicity profiles, such as a set of 45 substances toxic only to the Leukemia_SR cancer cell line. In an effort to predict nonspecific cytotoxicity on the basis of the mean log GI 50, we created a decision tree using MACCS keys that can correctly classify over 83% of the substances as cytotoxic/noncytotoxic in silico, on the basis of the cutoff of mean log GI 50 = -5.0. Finally, we have established a linear model using least-squares in which nine of the 59 available NCI60 cancer cell lines can be used to predict the mean log GI 50. The model has R (2) = 0.99 and a root-mean-square deviation between the observed and calculated mean log GI 50 (RMSE) = 0.09. Our predictive models can be applied to flag generally cytotoxic molecules in virtual and real chemical libraries, thus saving time and effort.     

Web-based cheminformatics and molecular property prediction tools supporting drug design and development at Novartis

Web-based tools offer many advantages for processing chemical information, most notably ease of use and high interactivity. Therefore more and more pharmaceutical companies are using web technology to deliver sophisticated molecular processing tools directly to the desks of their chemists, to assist them in the process of designing and developing new drugs. In this paper, the web-based cheminformatics system developed at Novartis and currently used by more than thousand users is described. The system allows various molecular modeling and molecular processing tasks, including the calculation of molecular and substituent properties, property-based virtual screening, visualization of molecules, bioisosteric design, diversity analysis, and support of combinatorial chemistry. The methodology to calculate various molecular properties relevant to drug design is described, including the prediction of intestinal absorption, blood-brain barrier penetration, efflux, and water solubility. Information about the web technology used is also provided.     

Comparison of the NCI open database with seven large chemical structural databases.

Eight large chemical databases have been analyzed and compared to each other. Central to this comparison is the open National Cancer Institute (NCI) database, consisting of approximately 250 000 structures. The other databases analyzed are the Available Chemicals Directory ("ACD," from MDL, release 1.99, 3D-version); the ChemACX ("ACX," from CamSoft, Version 4.5); the Maybridge Catalog and the Asinex database (both as distributed by CamSoft as part of ChemInfo 4.5); the Sigma-Aldrich Catalog (CD-ROM, 1999 Version); the World Drug Index ("WDI," Derwent, version 1999.03); and the organic part of the Cambridge Crystallographic Database ("CSD," from Cambridge Crystallographic Data Center, 1999 Version 5.18). The database properties analyzed are internal duplication rates; compounds unique to each database; cumulative occurrence of compounds in an increasing number of databases; overlap of identical compounds between two databases; similarity overlap; diversity; and others. The crystallographic database CSD and the WDI show somewhat less overlap with the other databases than those with each other. In particular the collections of commercial compounds and compilations of vendor catalogs have a substantial degree of overlap among each other. Still, no database is completely a subset of any other, and each appears to have its own niche and thus "raison d'être". The NCI database has by far the highest number of compounds that are unique to it. Approximately 200 000 of the NCI structures were not found in any of the other analyzed databases.     

Engineered chimeric enzymes as tools for drug discovery: generating reliable bacterial screens for the detection, discovery, and assessment of estrogen receptor modulators

Engineered protein-based sensors of ligand binding have emerged as attractive tools for the discovery of therapeutic compounds through simple screening systems. We have previously shown that engineered chimeric enzymes, which combine the ligand-binding domains of nuclear hormone receptors with a highly sensitive thymidylate synthase reporter, yield simple sensors that report the presence of hormone-like compounds through changes in bacterial growth. This work describes an optimized estrogen sensor in Escherichia coli with extraordinary reliability in identifying diverse estrogenic compounds and in differentiating between their agonistic/antagonistic pharmacological effects. The ability of this system to assist the discovery of new estrogen-mimicking compounds was validated by screening a small compound library, which led to the identification of two structurally novel estrogen receptor modulators and the accurate prediction of their agonistic/antagonistic biocharacter in human cells. Strong evidence is presented here that the ability of our sensor to detect ligand binding and recognize pharmacologically critical properties arises from allosteric communication between the artificially combined protein domains, where different ligand-induced conformational changes in the receptor are transmitted to the catalytic domain and translated to distinct levels of enzymic efficiency. To the best of our knowledge, this is one of the first examples of an engineered enzyme with the ability to sense multiple receptor conformations and to be either activated or inactivated depending on the nature of the bound effector molecule. Because the proposed mechanism of ligand dependence is not specific to nuclear hormone receptors, we anticipate that our protein engineering strategy will be applicable to the construction of simple sensors for different classes of (therapeutic) binding proteins.     

Chemical data mining of the NCI human tumor cell line database

The NCI Developmental Therapeutics Program Human Tumor cell line data set is a publicly available database that contains cellular assay screening data for over 40 000 compounds tested in 60 human tumor cell lines. The database also contains microarray assay gene expression data for the cell lines, and so it provides an excellent information resource particularly for testing data mining methods that bridge chemical, biological, and genomic information. In this paper we describe a formal knowledge discovery approach to characterizing and data mining this set and report the results of some of our initial experiments in mining the set from a chemoinformatics perspective.     

Activity-based profiling for drug discovery

Activity-based protein profiling (ABPP) is emerging as a game-changing tool for drug discovery, target validation, and basic biology. In this issue, Chang et?al. (2011) report the ABPP-facilitated discovery of JW480, a highly selective potent and orally bioavailable inhibitor of monoalkylglycerol ether hydrolase KIAA1363 that dramatically impairs in?vivo growth of human prostate cancer cell lines.     

Structure-based drug discovery for Plasmodium falciparum

X-ray crystallography is a technique which is finding increasing utility in the effort to find new antimalarial drugs. This is in spite of the serious difficulties often encountered in obtaining sufficient quantities of protein to crystallize. This review provides an overview of the Plasmodium falciparum proteins which have been crystallized with bound inhibitors and the methodology employed in the heterologous expression of these proteins. Lactate dehydrogenase, plasmepsin II, and triosphosphate isomerase are the most advanced targets of structure-based drug design, but nine other P. falciparum proteins have been crystallized with inhibitors as well, and this is clearly an area which is moving very quickly. Some consideration will also be given to the limitations of structure-based drug discovery with respect to known antimalarial drugs.     

Carrier polymers for cisplatin-type anticancer drug models

Following a brief discussion of the concept of polymer–drug conjugation and the use of platinum drugs in cancer therapy, the paper presents recent results in the synthesis of water-soluble polymeric carriers designed for the binding of antineoplastic coordination compounds of the cisplatin type. The target polymers, specifically, are linear aliphatic polyamides comprising the ethylenediamine ligand system in the main chain as the potential metal binding site. With solubility in aqueous media a key requirement for intravenously injectable conjugates, the polymers also contain hydrosolubilizing oligo(ethylene oxide) units in the chain, which serve the additional purpose of imparting resistance to serum protein binding and capture by the reticuloendothelial system. The synthesis methods include interfacial polymerization, high-temperature solution polycondensation in polyphosphoric acid and Michael addition polymerization, with 1,2-bis(2-aminoethylamino)ethane and 1,2-bis(3-aminopropylamino)ethane used as the amine comonomers providing the ethylenediamine ligand segment. The target polymers, crudely fractionated by dialysis in 25,000 molecular-mass cult-off tubing, are isolated by freeze-drying as water-soluble solids possessing inherent viscosities of 10–20 ml/g. A selected carrier polymer is converted to the corresponding water-soluble cis-diaminedichloroplatinum(II) conjugate by treatment with tetrachloroplatinate(II) anion in aqueous solution.     

A knowledge-based approach in designing combinatorial or medicinal chemistry libraries for drug discovery. 1. A qualitative and quantitative characterization of known drug databases

The discovery of various protein/receptor targets from genomic research is expanding rapidly. Along with the automation of organic synthesis and biochemical screening, this is bringing a major change in the whole field of drug discovery research. In the traditional drug discovery process, the industry tests compounds in the thousands. With automated synthesis, the number of compounds to be tested could be in the millions. This two-dimensional expansion will lead to a major demand for resources, unless the chemical libraries are made wisely. The objective of this work is to provide both quantitative and qualitative characterization of known drugs which will help to generate "drug-like" libraries. In this work we analyzed the Comprehensive Medicinal Chemistry (CMC) database and seven different subsets belonging to different classes of drug molecules. These include some central nervous system active drugs and cardiovascular, cancer, inflammation, and infection disease states. A quantitative characterization based on computed physicochemical property profiles such as log P, molar refractivity, molecular weight, and number of atoms as well as a qualitative characterization based on the occurrence of functional groups and important substructures are developed here. For the CMC database, the qualifying range (covering more than 80% of the compounds) of the calculated log P is between -0.4 and 5.6, with an average value of 2.52. For molecular weight, the qualifying range is between 160 and 480, with an average value of 357. For molar refractivity, the qualifying range is between 40 and 130, with an average value of 97. For the total number of atoms, the qualifying range is between 20 and 70, with an average value of 48. Benzene is by far the most abundant substructure in this drug database, slightly more abundant than all the heterocyclic rings combined. Nonaromatic heterocyclic rings are twice as abundant as the aromatic heterocycles. Tertiary aliphatic amines, alcoholic OH and carboxamides are the most abundant functional groups in the drug database. The effective range of physicochemical properties presented here can be used in the design of drug-like combinatorial libraries as well as in developing a more efficient corporate medicinal chemistry library.     

NMR spectroscopy tools for structure-aided drug design

Biomolecular NMR spectroscopy has expanded dramatically in recent years and is now a powerful tool for the study of structure, dynamics, and interactions of biomolecules. Previous limitations with respect to molecular size are no longer a primary barrier, and systems as large as 900 kDa were recently studied. NMR spectroscopy is already well-established as an efficient method for ligand screening. A number of recently developed techniques show promise as aids in structure-based drug design, for example, in the rapid determination of global protein folds, the structural characterization of ligand-protein complexes, and the derivation of thermodynamic parameters. An advantage of the method is that all these interactions can be studied in solution--time-consuming crystallization is not necessary. This Review focuses on recent developments in NMR spectroscopy and how they might be of value in removing some of the current "bottlenecks" in structure-based drug discovery.     

Mechanism for ribonucleotide reductase inactivation by the anticancer drug gemcitabine

Gemcitabine (2',2'-difluoro-2'-deoxycytidine, dFdC) is a very promising anticancer drug, already approved for clinical use in three therapeutic indications. It is metabolized intracellularly to 5'-diphosphate (dFdCDP), which is known to be a potent inhibitor of ribonucleotide reductase (RNR). Although several nucleotide analogs show in vitro capacity of RNR inactivation, none has shown the in vivo efficacy of dFdCDP. Accordingly, the experimental data suggests that its mechanism of inhibition is different from the other known RNR suicide inhibitors. Enzyme inhibition in the absence of reductive species leads to complete loss of the essential radical in subunit R2, and formation of a new nucleotide-based radical. Interestingly, however, the presence of the reductants does not prevent inhibition--the radical is not lost but the targeted subunit of RNR becomes R1, which is inactivated possibly by alkylation. We have conducted a theoretical study, which led us to the first proposal of a possible mechanism for RNR inhibition by dFdCDP in the absence of reductants. This mechanism turned out to be very similar to the natural substrate reduction pathway and only deviates from the natural course after the formation of the well-known disulphide bridge. This deviation is caused precisely by the F atom in the beta-face, only present in this inhibitor. The essential radical in R2 is lost, and so is the enzyme catalytic activity. The nucleotide-based radical that constitutes the end product of our mechanism has been suggested in the literature as a possible candidate for the one detected experimentally. In fact, all experimental data available has been reproduced by the theoretical calculations performed here.     

Maximum-score diversity selection for early drug discovery

Diversity selection is a common task in early drug discovery. One drawback of current approaches is that usually only the structural diversity is taken into account, therefore, activity information is ignored. In this article, we present a modified version of diversity selection, which we term Maximum-Score Diversity Selection, that additionally takes the estimated or predicted activities of the molecules into account. We show that finding an optimal solution to this problem is computationally very expensive (it is NP-hard), and therefore, heuristic approaches are needed. After a discussion of existing approaches, we present our new method, which is computationally far more efficient but at the same time produces comparable results. We conclude by validating these theoretical differences on several data sets.     

Nanoscale coordination polymers for platinum-based anticancer drug delivery

Pt-containing nanoscale coordination polymer (NCP) particles with the formula of Tb2(DSCP)3(H2O)12 (where DSCP represents disuccinatocisplatin), NCP-1, were precipitated from an aqueous solution of Tb3+ ions and DSCP bridging ligands via the addition of a poor solvent. SEM and TEM images showed that as-synthesized NCP-1 exhibited a spherical morphology with a DLS diameter of 58.3 +/- 11.3 nm. NCP-1 particles were stabilized against rapid dissolution in water by encapsulation in shells of amorphous silica. The resulting silica-coated particles NCP-1' exhibited significantly longer half-lives for DSCP release from the particles (a t1/2 of 9 h for NCP-1' with 7 nm silica coating vs t1/2 of 1 h for as-synthesized NCP-1). In vitro cancer cell cytotoxicity assays with the human colon carcinoma cell line (HT-29) showed that internalized NCP-1' particles readily released the DSCP moieties which were presumably reduced to cytotoxic Pt(II) species to give the Pt-containing NCPs anticancer efficacy superior to the cisplatin standard. The generality of this degradable nanoparticle formulation should allow for the design of NCPs as effective delivery vehicles for a variety of biologically and medically important cargoes such as therapeutic and imaging agents.     

NMR-based methods and strategies for drug discovery

Nuclear Magnetic Resonance (NMR) spectroscopy has long been a favourite tool of chemists interested in host-guest systems because it permits access to a wealth of information about the molecular recognition reaction. NMR has evolved dramatically in the last 15 years and, in parallel with the development of NMR methods for the determination of protein structure, a variety of tools aimed at detecting protein ligand interactions have been proposed and are being now used both in industrial and academic laboratories as valuable tools for structure-based drug discovery. Very recent developments have considerably increased the fraction of therapeutic targets that can be tackled by NMR and significantly reduced the amount of sample required for analysis; in this tutorial review we outline the essential NMR-based techniques and describe some examples of their implementation as part of drug discovery programmes.