DOCK 4.0: search strategies for automated molecular docking of flexible molecule databases

Comparative evaluation of eight docking tools for docking and virtual screening accuracy

Eight docking programs (DOCK, FLEXX, FRED, GLIDE, GOLD, SLIDE, SURFLEX, and QXP) that can be used for either single-ligand docking or database screening have been compared for their propensity to recover the X-ray pose of 100 small-molecular-weight ligands, and for their capacity to discriminate known inhibitors of an enzyme (thymidine kinase) from randomly chosen "drug-like" molecules. Interestingly, both properties are found to be correlated, since the tools showing the best docking accuracy (GLIDE, GOLD, and SURFLEX) are also the most successful in ranking known inhibitors in a virtual screening experiment. Moreover, the current study pinpoints some physicochemical descriptors of either the ligand or its cognate protein-binding site that generally lead to docking/scoring inaccuracies.

Binding mode prediction of cytochrome p450 and thymidine kinase protein-ligand complexes by consideration of water and rescoring in automated docking

The popular docking programs AutoDock, FlexX, and GOLD were used to predict binding modes of ligands in crystallographic complexes including X-ray water molecules or computationally predicted water molecules. Isoenzymes of two different enzyme systems were used, namely cytochromes P450 (n = 19) and thymidine kinases (n = 19) and three different "water" scenarios: i.e., docking (i) into water-free active sites, (ii) into active sites containing crystallographic water molecules, and (iii) into active sites containing water molecules predicted by a novel approach based on the program GRID. Docking accuracies were determined in terms of the root-mean-square deviation (RMSD) accuracy and, newly defined, in terms of the ligand catalytic site prediction (CSP) accuracy. Consideration of both X-ray and predicted water molecules and the subsequent pooling and rescoring of all solutions (generated by all three docking programs) with the SCORE scoring function significantly improved the quality of prediction of the binding modes both in terms of RMSD and CSP accuracy.

Comparison of automated docking programs as virtual screening tools

The performance of several commercially available docking programs is compared in the context of virtual screening. Five different protein targets are used, each with several known ligands. The simulated screening deck comprised 1000 molecules from a cleansed version of the MDL drug data report and 49 known ligands. For many of the known ligands, crystal structures of the relevant protein-ligand complexes were available. We attempted to run experiments with each docking method that were as similar as possible. For a given docking method, hit rates were improved versus what would be expected for random selection for most protein targets. However, the ability to prioritize known ligands on the basis of docking poses that resemble known crystal structures is both method- and target-dependent.

A structure-based strategy for discovery of small ligands binding to functionally unknown proteins: Combination ofin silico screening and surface plasmon resonance measurements

In the postgenomic era, many researchers and organizations have been engaged in structural and functional analyses of proteins. As a part of these efforts, searching for small organic compounds that bind specifically to target proteins is quite important. In this study, we have developed a rational strategy for ligand discovery based on the three-dimensional structures of target proteins, which were elucidated by X-ray crystallography and nuclear magnetic resonance spectroscopy. The strategy has three features: (i) rapid selection of candidate compounds by in silico screening, (ii) automated preparation of sample solutions with robotics, and (iii) reliable evaluation of the candidates with surface plasmon resonance. Applying the strategy to a protein, At2g24940 from Arabidopsis thaliana, we discovered four small ligands out of a commercially available library of about 150 000 compounds. Although these compounds had only weak affinities to the target protein, with dissociation constants ranging from 68 to 120 microM, they apparently possess common structural features. They would be leads for the development of specific inhibitors/drugs for At2g24940, and provide important clues toward elucidation of the protein function.

LigScore: a novel scoring function for predicting binding affinities

We present two new empirical scoring functions, LigScore1 and LigScore2, that attempt to accurately predict the binding affinity between ligand molecules and their protein receptors. The LigScore functions consist of three distinct terms that describe the van der Waals interaction, the polar attraction between the ligand and protein, and the desolvation penalty attributed to the binding of the polar ligand atoms to the protein and vice versa. Utilizing a regression approach on a data set of 118 protein-ligand complexes we have obtained a linear equation, LigScore2, using these three descriptors. LigScore2 has good predictability with regard to experimental pKi values yielding a correlation coefficient, r2), of 0.75 and a standard deviation of 1.04 over the training data set, which consists of a diverse set of proteins that span more than seven protein families.

A scoring function for docking ligands to low-resolution protein structures

We present a docking method that uses a scoring function for protein-ligand docking that is designed to maximize the docking success rate for low-resolution protein structures. We find that the resulting scoring function parameters are very different depending on whether they were optimized for high- or low-resolution protein structures. We show that this docking method can be successfully applied to predict the ligand-binding site of low-resolution structures. For a set of 25 protein-ligand complexes, in 76% of the cases, more than 50% of ligand-contacting residues are correctly predicted (using receptor crystal structures where the binding site is unspecified). Using decoys of the receptor structures having a 4 A RMSD from the native structure, for the same set of complexes, in 72% of the cases, we obtain at least one correctly predicted ligand-contacting residue. Furthermore, using an 81-protein-ligand set described by Jain, in 76 (93.8%) cases, the algorithm correctly predicts more than 50% of the ligand-contacting residues when native protein structures are used. Using 3 A RMSD from native decoys, in all but two cases (97.5%), the algorithm predicts at least one ligand-binding residue correctly. Finally, compared to the previously published Dolores method, for 298 protein-ligand pairs, the number of cases in which at least half of the specific contacts are correctly predicted is more than four times greater.

A low-molecular-weight compound discovered through virtual database screening inhibits Stat3 function in breast cancer cells

This study focused on the screening of small-molecule inhibitors that target signal transducers and activators of transcription 3 (Stat3) in human breast carcinoma. The constitutive activation of Stat3 is frequently detected in human breast cancer cell lines as well as clinical breast cancer specimens and may play an important role in the oncogenesis of breast carcinoma. Activated Stat3 may participate in oncogenesis by stimulating cell proliferation, promoting tumor angiogenesis, and resisting apoptosis. Because a variety of human cancers are associated with constitutively active Stat3, Stat3 represents an attractive target for cancer therapy. In this study, of the nearly 429,000 compounds screened by virtual database screening, chemical samples of top 100 compounds identified as candidate small-molecule inhibitors of Stat3 were evaluated by using Stat3-dependent luciferase reporter as well as other cell-based assays. Through serial functional evaluation based on our established cell-based assays, one compound, termed STA-21, was identified as the best match for our selection criteria. Further investigation demonstrated that STA-21 inhibits Stat3 DNA binding activity, Stat3 dimerization, and Stat3-dependent luciferase activity. Moreover, STA-21 reduces the survival of breast carcinoma cells with constitutive Stat3 signaling but has minimal effect on the cells in which constitutive Stat3 signaling is absent. Together, these results demonstrate that STA-21 inhibits breast cancer cells that express constitutively active Stat3.

Selectivity and specificity of substrate binding in methionyl-tRNA synthetase

The accuracy of in vivo incorporation of amino acids during protein biosynthesis is controlled to a significant extent by aminoacyl-tRNA synthetases (aaRS). This paper describes the application of the HierDock computational method to study the molecular basis of amino acid binding to the Escherichia coli methionyl tRNA synthetase (MetRS). Starting with the protein structure from the MetRS cocrystal, the HierDock calculations predict the binding site of methionine in MetRS to a root mean square deviation in coordinates (CRMS) of 0.55 A for all the atoms, compared with the crystal structure. The MetRS conformation in the cocrystal structure shows good discrimination between cognate and the 19 noncognate amino acids. In addition, the calculated binding energies of a set of five methionine analogs show a good correlation (R(2) = 0.86) to the relative free energies of binding derived from the measured in vitro kinetic parameters, K(m) and k(cat). Starting with the crystal structure of MetRS without the methionine (apo-MetRS), the putative binding site of methionine was predicted. We demonstrate that even the apo-MetRS structure shows a preference for binding methionine compared with the 19 other natural amino acids. On comparing the calculated binding energies of the 20 natural amino acids for apo-MetRS with those for the cocrystal structure, we observe that the discrimination against the noncognate substrate increases dramatically in the second step of the physical binding process associated with the conformation change in the protein.

Virtual screening of novel noncovalent inhibitors for SARS-CoV 3C-like proteinase

The SARS coronavirus 3C-like proteinase is considered as a potential drug design target for the treatment of severe acute respiratory syndrome (SARS). Owing to the lack of available drugs for the treatment of SARS, the discovery of inhibitors for SARS coronavirus 3C-like proteinase that can potentially be optimized as drugs appears to be highly desirable. We have built a "flexible" three-dimensional model for SARS 3C-like proteinase by homology modeling and multicanonical molecular dynamics method and used the model for virtual screening of chemical databases. After Dock procedures, strategies including pharmocophore model, consensus scoring, and "drug-like" filters were applied in order to accelerate the process and improve the success rate of virtual docking screening hit lists. Forty compounds were purchased and tested by HPLC and colorimetric assay against SARS 3C-like proteinase. Three of them including calmidazolium, a well-known antagonist of calmodulin, were found to inhibit the enzyme with an apparent K(i) from 61 to 178 microM. These active compounds and their binding modes provide useful information for understanding the binding sites and for further selective drug design against SARS and other coronavirus.

Integrating virtual screening in lead discovery

Target- and ligand-based virtual screening have emerged as resource-saving techniques that have been successfully applied to identify novel chemotypes in biologically active molecules. Eight confirmed virtual screening hits have recently been described and are discussed in this review, with focus on the workflow. These are then evaluated in the light of pharmacokinetics prediction (e.g. Caco-2 permeability, cytochrome P450 inhibition and hERG binding). We anticipate problems for five of these hits (e.g. cardiac toxicity), which warrant further experiments. Future challenges include dynamic tautomer/protonation treatment for both ligands and targets and improved pre- and post- virtual screening filters.

Development of KiBank, a database supporting structure-based drug design

KiBank is a database of inhibition constant (Ki) values with 3D structures of target proteins and chemicals. Ki values were accumulated from peer-reviewed literature searched via PubMed. The 3D structure files of target proteins were originally from Protein Data Bank (PDB), while the 2D structure files of the chemicals were collected together with the Ki values and then converted into 3D ones. In KiBank, the chemical and protein 3D structures with hydrogen atoms were optimized by energy minimization and stored in MDL MOL and PDB format, respectively. KiBank is designed to support structure-based drug design. It provides structure files of proteins and chemicals ready for use in virtual screening through automated docking methods, while the Ki values can be applied for tests of docking/scoring combinations, program parameter settings, and calibration of empirical scoring functions. Additionally, the chemical structures and corresponding Ki values in KiBank are useful for lead optimization based on quantitative structure-activity relationship (QSAR) techniques. KiBank is updated on a daily basis and is freely available at . As of August 2004, KiBank contains 8000 Ki values, over 6000 chemicals and 166 proteins covering the subtypes of receptors and enzymes.

Progress in cytochrome P450 active site modeling

Models capable of predicting the possible involvement of cytochromes P450 in the metabolism of drugs or drug candidates are important tools in drug discovery and development. Ideally, functional information would be obtained from crystal structures of all the cytochromes P450 of interest. Initially, only crystal structures of distantly related bacterial cytochromes P450 were available-comparative modeling techniques were used to bridge the gap and produce structural models of human cytochromes P450, and thereby obtain some useful functional information. A significant step forward in the reliability of these models came four years ago with the first crystal structure of a mammalian cytochrome P450, rabbit CYP2C5, followed by the structures of two human enzymes, CYP2C8 and CYP2C9, and a second rabbit enzyme, CYP2B4. The evolution of a CYP2D6 model, leading to the validation of the model as an in silico tool for predicting binding and metabolism, is presented as a case study.

The MPSim-Dock hierarchical docking algorithm: application to the eight trypsin inhibitor cocrystals

To help improve the accuracy of protein-ligand docking as a useful tool for drug discovery, we developed MPSim-Dock, which ensures a comprehensive sampling of diverse families of ligand conformations in the binding region followed by an enrichment of the good energy scoring families so that the energy scores of the sampled conformations can be reliably used to select the best conformation of the ligand. This combines elements of DOCK4.0 with molecular dynamics (MD) methods available in the software, MPSim. We test here the efficacy of MPSim-Dock to predict the 64 protein-ligand combinations formed by starting with eight trypsin cocrystals, and crossdocking the other seven ligands to each protein conformation. We consider this as a model for how well the method would work for one given target protein structure. Using as a criterion that the structures within 2 kcal/mol of the top scoring include a conformation within a coordinate root mean square (CRMS) of 1 A of the crystal structure, we find that 100% of the 64 cases are predicted correctly. This indicates that MPSim-Dock can be used reliably to identify strongly binding ligands, making it useful for virtual ligand screening.

Exploring biology with small organic molecules

Small organic molecules have proven to be invaluable tools for investigating biological systems, but there is still much to learn from their use. To discover and to use more effectively new chemical tools to understand biology, strategies are needed that allow us to systematically explore 'biological-activity space'. Such strategies involve analysing both protein binding of, and phenotypic responses to, small organic molecules. The mapping of biological-activity space using small molecules is akin to mapping the stars--uncharted territory is explored using a system of coordinates that describes where each new feature lies.

Docking and scoring in virtual screening for drug discovery: methods and applications

Computational approaches that 'dock' small molecules into the structures of macromolecular targets and 'score' their potential complementarity to binding sites are widely used in hit identification and lead optimization. Indeed, there are now a number of drugs whose development was heavily influenced by or based on structure-based design and screening strategies, such as HIV protease inhibitors. Nevertheless, there remain significant challenges in the application of these approaches, in particular in relation to current scoring schemes. Here, we review key concepts and specific features of small-molecule-protein docking methods, highlight selected applications and discuss recent advances that aim to address the acknowledged limitations of established approaches.

A molecular docking study of estrogenically active compounds with 1,2-diarylethane and 1,2-diarylethene pharmacophores

Numerous selective estrogen receptor modulators (SERMs) have been synthesized and assayed in recent years. The focus of this study is to apply coarse-grain molecular docking procedures coupled with fine-grain all-atom force field optimization strategies to shed light on the binding mechanisms of currently available estrogen receptor-active compounds. Although the mechanics of ligand binding in estrogen receptors is generally well understood, there is room for surprises. In this paper computational evidence corroborating the experimentally observed type I agonistic binding mode for estradiol (E2) and diethylstilbesterol (DES) and the type II antagonistic binding mode for 4-hydroxytamoxifen and raloxifen is presented. Included in this type I agonistic mode are the DES derivatives, transstilbene and 1,2-diaryldiaminoethane. In addition, a novel 'type II agonistic' binding mode for 2,3-diarylimidazolines, 4,5-diarylimidazoles, 2,3-diarylpiperazines is introduced. This mode is stabilized by suggesting alternative hydrogen bond anchor points in the ligand binding domain as potential leads for future drug design.

Studies of Binding Modes of (S)-Mephenytoin to Wild Types and Mutants of Cytochrome P450 2C19 and 2C9 Using Homology Modeling and Computational Docking

PURPOSE

This study investigated the structural features of CYP2C19 complexed with (S)-mephenytoin, using computational methods. In addition to wild-type CYP2C19 proteins (1A and 1B), which have selective 4-hydroxylase activities of (S)-mephenytoin, CYP2C19 mutants were also studied, together with a wild type and artificial mutants of CYP2C19.

METHODS

Three-dimensional structures of wild-type and mutant proteins of CYP2C19 and CYP2C9 were estimated from homology modeling using the crystal structure of rabbit CYP2C5 as a reference. The binding mode of (S)-mephenytoin to CYP2C19 was investigated using computational docking.

RESULTS

The results reproduced the specific bindings between (S)-mephenytoin and the wild types of CYP2C19. Our findings suggest that Asp293 of CYP2C19 plays an important role in the binding of (S)-mephenytoin, which was surrounded by Vall13 and Ala297, and points the phenyl ring at the heme iron. In addition the wild types of CYP2C19, the computational docking studies also accounted for the experimental activities of CYP2C19 mutants, and wild-type and mutant CYP2C19 proteins.

CONCLUSIONS

These results confirm that the predicted three-dimensional structure of the CYP2C19-(S)-mephenytoin complex is reasonable, and that this strategy is useful for investigating complex structures. Virtual screening for drug discovery can also be carried out using these methods.

Scoring functions for protein-ligand interactions: a critical perspective

Scoring functions play an essential role in structure-based virtual screening. They are required to guide the docking of candidate compounds to structures of receptor binding sites, to select probable binding modes, and to discriminate binders from non-binders. Although many scoring functions have successfully been used to identify novel ligands for a wide variety of targets, much work remains to be done to avoid incorrect prediction of binding modes and high numbers of false positives. This review gives an overview of the current state of the field and outlines key issues for the further development of scoring functions.:

Unraveling the interface of signal recognition particle and its receptor by using chemical cross-linking and tandem mass spectrometry

Among the methods used to unravel protein interaction surfaces, chemical cross-linking followed by identification of the cross-linked peptides by mass spectrometry has proven especially useful in dynamic and complex systems. During the signal recognition particle (SRP)-dependent targeting of proteins to the bacterial plasma membrane, the specific interaction between Ffh (the protein component of SRP) and FtsY (the SRP receptor) is known to be essential for the efficiency and fidelity of this process. In this work, we studied the Escherichia coli and Thermus aquaticus Ffh.FtsY complexes by using chemical cross-linking and tandem mass spectrometry to identify nine intermolecular cross-linked peptides. This information was used in conjunction with a previously undescribed model-building approach that combines geometric restraint optimization with macromolecular docking. The resulting model of the Ffh.FtsY complex is in good agreement with the crystal structure solved shortly thereafter. Intriguingly, four of the cross-linked pairs involve the M domain of Ffh, which is absent from the crystal structure, providing previously undocumented experimental evidence that the M domain is positioned in close proximity to the Ffh.FtsY interface in the complex.

Exploring the binding mode of semicarbazide-sensitive amine oxidase/VAP-1: identification of novel substrates with insulin-like activity

We previously reported that substrates of semicarbazide-sensitive amine oxidase in combination with low concentrations of vanadate exert potent insulin-like effects. Here we performed homology modeling of the catalytic domain of mouse SSAO/VAP-1 and searched through chemical databases to identify novel SSAO substrates. The modeling of the catalytic domain revealed that aromatic residues Tyr384, Phe389, and Tyr394 define a pocket of stable size that may participate in the binding of apolar substrates. We identified a number of amines as substrates of human, rat, and mouse SSAO. The compounds PD0119035, 2,3-dimethoxy-benzylamine, and C-naphthalen-1-yl-methylamine showed high affinity as substrates of rat SSAO. C-Naphthalen-1-yl-methylamine was the only substrate that showed high affinity for human SSAO. C-Naphthalen-1-yl-methylamine and 4-aminomethyl-benzenesulfonamide showed the highest capacity to stimulate glucose transport in isolated rat adipocytes. The impact of these findings on the development of new treatments for diabetes is discussed.

Design and synthesis of potent, orally active, inhibitors of carboxypeptidase U (TAFIa)

A series of 3-mercapto-propionic acid derivatives that function as reversible inhibitors of carboxypeptidase U have been prepared. We present a successful design strategy using cyclic, low basicity guanidine mimetics resulting in potent, selective and bioavailable inhibitors of carboxypeptidase U (TAFIa).

UCSF Chimera--a visualization system for exploratory research and analysis

The design, implementation, and capabilities of an extensible visualization system, UCSF Chimera, are discussed. Chimera is segmented into a core that provides basic services and visualization, and extensions that provide most higher level functionality. This architecture ensures that the extension mechanism satisfies the demands of outside developers who wish to incorporate new features. Two unusual extensions are presented: Multiscale, which adds the ability to visualize large-scale molecular assemblies such as viral coats, and Collaboratory, which allows researchers to share a Chimera session interactively despite being at separate locales. Other extensions include Multalign Viewer, for showing multiple sequence alignments and associated structures; ViewDock, for screening docked ligand orientations; Movie, for replaying molecular dynamics trajectories; and Volume Viewer, for display and analysis of volumetric data. A discussion of the usage of Chimera in real-world situations is given, along with anticipated future directions. Chimera includes full user documentation, is free to academic and nonprofit users, and is available for Microsoft Windows, Linux, Apple Mac OS X, SGI IRIX, and HP Tru64 Unix from http://www.cgl.ucsf.edu/chimera/.

G protein-coupled receptors: in silico drug discovery in 3D

The application of structure-based in silico methods to drug discovery is still considered a major challenge, especially when the x-ray structure of the target protein is unknown. Such is the case with human G protein-coupled receptors (GPCRs), one of the most important families of drug targets, where in the absence of x-ray structures, one has to rely on in silico 3D models. We report repeated success in using ab initio in silico GPCR models, generated by the predict method, for blind in silico screening when applied to a set of five different GPCR drug targets. More than 100,000 compounds were typically screened in silico for each target, leading to a selection of <100 "virtual hit" compounds to be tested in the lab. In vitro binding assays of the selected compounds confirm high hit rates, of 12-21% (full dose-response curves, Ki < 5 microM). In most cases, the best hit was a novel compound (New Chemical Entity) in the 1- to 100-nM range, with very promising pharmacological properties, as measured by a variety of in vitro and in vivo assays. These assays validated the quality of the hits as lead compounds for drug discovery. The results demonstrate the usefulness and robustness of ab initio in silico 3D models and of in silico screening for GPCR drug discovery.

High-throughput docking as a source of novel drug leads

Receptor-based virtual screening has become a viable source of novel leads in the pharmaceutical industry. The rapidly growing availability of structural information across protein families, the accessibility to increased computational power at affordable cost, as well as an improved understanding on how to effectively apply virtual screening technologies has contributed to their emergence. Nonetheless, continued improvement in the accuracy of scoring functions and a greater understanding of protein mobility is critical to advance the technology further.

Integrating cheminformatic analysis in combinatorial chemistry

Cheminformatic analysis of drug-related compound databases has enabled the identification of the physicochemical properties that have the greatest influence on determining the drug-like characteristics of a compound. This enables definition of the parameters and profiles used in constructing a high-quality combinatorial library. Awareness of the multi-objective nature of combinatorial library construction has also given rise to techniques designed to enhance the likelihood of including the best compounds in a given library.

The process of structure-based drug design

The field of structure-based drug design is a rapidly growing area in which many successes have occurred in recent years. The explosion of genomic, proteomic, and structural information has provided hundreds of new targets and opportunities for future drug lead discovery. This review summarizes the process of structure-based drug design and includes, primarily, the choice of a target, the evaluation of a structure of that target, the pivotal questions to consider in choosing a method for drug lead discovery, and evaluation of the drug leads. Key principles in the field of structure-based drug design will be illustrated through a case study that explores drug design for AmpC beta-lactamase.

Discovery of embelin as a cell-permeable, small-molecular weight inhibitor of XIAP through structure-based computational screening of a traditional herbal medicine three-dimensional structure database

The X-linked inhibitor of apoptosis (XIAP) is a promising new molecular target for the design of novel anticancer drugs aiming at overcoming apoptosis-resistance of cancer cells to chemotherapeutic agents and radiation therapy. Recent studies demonstrated that the BIR3 domain of XIAP where caspase-9 and Smac proteins bind is an attractive site for designing small-molecule inhibitors of XIAP. Through computational structure-based screening of an in-house traditional herbal medicine three-dimensional structure database of 8221 individual natural products, followed by biochemical testing of selected candidate compounds, we discovered embelin from the Japanese Ardisia herb as a small-molecular weight inhibitor that binds to the XIAP BIR3 domain. We showed that embelin binds to the XIAP BIR3 protein with an affinity similar to that of the natural Smac peptide using a fluorescence polarization-based binding assay. Our NMR analysis further conclusively confirmed that embelin interacts with several crucial residues in the XIAP BIR3 domain with which Smac and caspsase-9 bind. Embelin inhibits cell growth, induces apoptosis, and activates caspase-9 in prostate cancer cells with high levels of XIAP, but has a minimal effect on normal prostate epithelial and fibroblast cells with low levels of XIAP. In stably XIAP-transfected Jurkat cells, embelin effectively overcomes the protective effect of XIAP to apoptosis and enhances the etoposide-induced apoptosis and has a minimal effect in Jurkat cells transfected with vector control. Taken together, our results showed that embelin is a fairly potent, nonpeptidic, cell-permeable, small-molecule inhibitor of XIAP and represents a promising lead compound for designing an entirely new class of anticancer agents that target the BIR3 domain of XIAP.

Creating Artificial Binding Pocket Boundaries To Improve the Efficiency of Flexible Ligand Docking

Traditionally, algorithms for binding site characterization or identification focus on the problem of identifying atoms within a macromolecule that might be responsible for ligand binding. In this manuscript, we focus on the binding pocket problem from a different perspective as a challenge of calculating an artificial binding pocket boundary that is sufficient to isolate binding pocket volume. The approach involves the calculation of a macromolecule encapsulating surface (MES) that separates binding pocket volume from outside space. We show that the MES can be used to increase the efficiency of flexible docking as implemented in AutoDock 3.0. The most significant improvement in docking efficiency is seen when the entire protein is searched and results show additional support for the use of AutoDock, in and of itself, as a feasible tool for binding-site identification for cases in which a protein ligand is known.

Applying data mining techniques to library design, lead generation and lead optimization

Many data mining techniques have been applied to activity and ADMET datasets and the resulting models are being used to understand quantitative structure-activity relationships and design new libraries. This review summarizes data mining concepts and discuss their application to library design, lead generation (particularly for sequential screening) and lead optimization (specifically for generating and interpreting QSAR models). Also, this review discusses recent comparative studies between data mining techniques and draws some conclusions about the patterns emerging in the drug discovery data mining field.

Computer-Assisted Design of Selective Imidazole Inhibitors for Cytochrome P450 Enzymes

A modified version of the DOCK program has been used to predict inhibitors for cytochrome P450cam and its L244A mutant. A library of azole compounds was designed in silico and screened for binding to wild-type P450cam. Lead compounds were synthesized and found to inhibit wild-type P450cam. To test our approach to designing ligands that discriminate between closely related sites, the azole library was DOCKed into both the active sites of wild-type P450cam and its L244A mutant. The L244A active site is predicted to be slightly larger than that of wild-type P450cam. Ligands predicted to have a high affinity for the mutant alone were synthesized and assayed with the recombinant enzymes. All of the compounds showed inhibition of the L244A enzyme (IC(50) = 6-40 microM), and the compounds that were predicted to be too large to bind to the wild-type showed poor inhibition (IC(50) > or = 1 mM). The binding mode was shown to be similar to that predicted by our modified version of DOCK by spectroscopic analysis. A discrepancy between the IC(50) values and spectroscopic K(s) values indicates that the spectroscopic binding constants do not accurately estimate inhibitory activity. This study, the first report of computer-assisted ligand (drug) design for P450 enzymes in which the coordination bond between imidazole and the heme is explicitly considered in structural modeling, opens a promising design avenue because azole compounds are widely used as P450 enzyme inhibitors and drugs.

Virtual Screening Using Protein−Ligand Docking: Avoiding Artificial Enrichment

This study addresses a number of topical issues around the use of protein-ligand docking in virtual screening. We show that, for the validation of such methods, it is key to use focused libraries (containing compounds with one-dimensional properties, similar to the actives), rather than "random" or "drug-like" libraries to test the actives against. We also show that, to obtain good enrichments, the docking program needs to produce reliable binding modes. We demonstrate how pharmacophores can be used to guide the dockings and improve enrichments, and we compare the performance of three consensus-ranking protocols against ranking based on individual scoring functions. Finally, we show that protein-ligand docking can be an effective aid in the screening for weak, fragment-like binders, which has rapidly become a popular strategy for hit identification. All results presented are based on carefully constructed virtual screening experiments against four targets, using the protein-ligand docking program GOLD.

Finding More Needles in the Haystack: A Simple and Efficient Method for Improving High-Throughput Docking Results

The technology underpinning high-throughput docking (HTD) has developed over the past few years to where it has become a vital tool in modern drug discovery. Although the performance of various docking algorithms is adequate, the ability to accurately and consistently rank compounds using a scoring function remains problematic. We show that by employing a simple machine learning method (naïve Bayes) it is possible to significantly overcome this deficiency. Compounds from the Available Chemical Directory (ACD), along with known active compounds, were docked into two protein targets using three software packages. In cases where HTD alone was able to show some enrichment, the application of naïve Bayes was able to improve upon the enrichment. The application of this methodology to enrich HTD results can be carried out without a priori knowledge of the activity of compounds and results in superior enrichment of known actives compared to the use of scoring methods alone.

A detailed comparison of current docking and scoring methods on systems of pharmaceutical relevance

A thorough evaluation of some of the most advanced docking and scoring methods currently available is described, and guidelines for the choice of an appropriate protocol for docking and virtual screening are defined. The generation of a large and highly curated test set of pharmaceutically relevant protein-ligand complexes with known binding affinities is described, and three highly regarded docking programs (Glide, GOLD, and ICM) are evaluated on the same set with respect to their ability to reproduce crystallographic binding orientations. Glide correctly identified the crystallographic pose within 2.0 A in 61% of the cases, versus 48% for GOLD and 45% for ICM. In general Glide appears to perform most consistently with respect to diversity of binding sites and ligand flexibility, while the performance of ICM and GOLD is more binding site-dependent and it is significantly poorer when binding is predominantly driven by hydrophobic interactions. The results also show that energy minimization and reranking of the top N poses can be an effective means to overcome some of the limitations of a given docking function. The same docking programs are evaluated in conjunction with three different scoring functions for their ability to discriminate actives from inactives in virtual screening. The evaluation, performed on three different systems (HIV-1 protease, IMPDH, and p38 MAP kinase), confirms that the relative performance of different docking and scoring methods is to some extent binding site-dependent. GlideScore appears to be an effective scoring function for database screening, with consistent performance across several types of binding sites, while ChemScore appears to be most useful in sterically demanding sites since it is more forgiving of repulsive interactions. Energy minimization of docked poses can significantly improve the enrichments in systems with sterically demanding binding sites. Overall Glide appears to be a safe general choice for docking, while the choice of the best scoring tool remains to a larger extent system-dependent and should be evaluated on a case-by-case basis.

A target site for template-based design of measles virus entry inhibitors

Measles virus (MV) constitutes a principal cause of worldwide mortality, accounting for almost 1 million deaths annually. Although a live-attenuated vaccine protects against MV, vaccination efficiency of young infants is low because of interference by maternal antibodies. Parental concerns about vaccination safety further contribute to waning herd immunity in developed countries, resulting in recent MV outbreaks. The development of novel antivirals that close the vaccination gap in infants and silence viral outbreaks is thus highly desirable. We previously identified a microdomain in the MV fusion protein (F protein) that is structurally conserved in the paramyxovirus family and constitutes a promising target site for rationally designed antivirals. Here we report the template-based development of a small-molecule MV inhibitor, providing proof-of-concept for our approach. This lead compound specifically inhibits fusion and spread of live MV and MV glycoprotein-induced membrane fusion. The inhibitor induces negligible cytotoxicity and does not interfere with receptor binding or F protein biosynthesis or transport but prevents F protein-induced lipid mixing. Mutations in the postulated target site alter viral sensitivity to inhibition. In silico docking of the compound in this microdomain suggests a binding model that is experimentally corroborated by a structure-activity analysis of the compound and the inhibition profile of mutated F proteins. A second-generation compound designed on the basis of the interaction model shows a 200-fold increase in antiviral activity, creating the basis for novel MV therapeutics. This template-based design approach for MV may be applicable to other clinically relevant members of the paramyxovirus family.

Identification of novel estrogen receptor alpha antagonists

We have identified novel estrogen receptor alpha (ERalpha) antagonists using both cell-based and computer-based virtual screening strategies. A mammalian two-hybrid screen was used to select compounds that disrupt the interaction between the ERalpha ligand binding domain (LBD) and the coactivator SRC-3. A virtual screen was designed to select compounds that fit onto the LxxLL peptide-binding surface of the receptor, based on the X-ray crystal structure of the ERalpha LBD complexed with a LxxLL peptide. All selected compounds effectively inhibited 17-beta-estradiol induced coactivator recruitment with potency ranging from nano-molar to micromolar. However, in contrast to classical ER antagonists, these novel inhibitors poorly displace estradiol in the ER-ligand competition assay. Nuclear magnetic resonance (NMR) suggested direct binding of these compounds to the receptors pre-complexed with estradiol and further demonstrated that no estradiol displacement occurred. Partial proteolytic enzyme digestion revealed that, when compared with 17-beta-estradiol- and 4 hydroxy-tamoxifen (4-OHT) bound receptors, at least one of these compounds might induce a unique receptor conformation. These small molecules may represent new classes of ER antagonists, and may have the potential to provide an alternative for the current anti-estrogen therapy.

A practical approach to docking of zinc metalloproteinase inhibitors

Forty zinc-dependent metalloproteinase/ligand complexes with known crystal structures were re-docked using five docking/scoring approaches (DOCK, FlexX, DrugScore, GOLD, and AutoDock). Correct geometry of the coordination bonds between the ligand's zinc binding group (ZBG) and the catalytic zinc is important for docking accuracy and scoring reliability. More than 75% of docked poses with RMSD less than 2A were found to have appropriate ZBG binding, but for poor ZBG binding, about 95% of poses failed to dock correctly. Elimination of poses with inappropriate zinc binding resulted in better binding energy predictions that were further improved by dividing the ligands into subsets according to the ZBG (carboxylates, hydroxamates, and phosphorus containing groups). After a subset re-scoring using the regression functions obtained for individual subsets, DrugScore was able to explain 77% and the consensus scoring scheme X-CSCORE even 88% of variance in binding energies. The approach combining ZBG-based pose selection and subset re-scoring improved the hit rate in virtual screening for metalloproteinase inhibitors for all tested methods by 4-16%.

Shape signatures: a new approach to computer-aided ligand- and receptor-based drug design

A unifying principle of rational drug design is the use of either shape similarity or complementarity to identify compounds expected to be active against a given target. Shape similarity is the underlying foundation of ligand-based methods, which seek compounds with structure similar to known actives, while shape complementarity is the basis of most receptor-based design, where the goal is to identify compounds complementary in shape to a given receptor. These approaches can be extended to include molecular descriptors in addition to shape, such as lipophilicity or electrostatic potential. Here we introduce a new technique, which we call shape signatures, for describing the shape of ligand molecules and of receptor sites. The method uses a technique akin to ray-tracing to explore the volume enclosed by a ligand molecule, or the volume exterior to the active site of a protein. Probability distributions are derived from the ray-trace, and can be based solely on the geometry of the reflecting ray, or may include joint dependence on properties, such as the molecular electrostatic potential, computed over the surface. Our shape signatures are just these probability distributions, stored as histograms. They converge rapidly with the length of the ray-trace, are independent of molecular orientation, and can be compared quickly using simple metrics. Shape signatures can be used to test for both shape similarity between compounds and for shape complementarity between compounds and receptors and thus can be applied to problems in both ligand- and receptor-based molecular design. We present results for comparisons between small molecules of biological interest and the NCI Database using shape signatures under two different metrics. Our results show that the method can reliably extract compounds of shape (and polarity) similar to the query molecules. We also present initial results for a receptor-based strategy using shape signatures, with application to the design of new inhibitors predicted to be active against HIV protease.

Comparing performance of computational tools for combinatorial library design

In using computational tools for library design it is necessary to understand the performance and limitations of available methods. This letter reports systematic comparisons of applying ligand-based and structure-based tools across therapeutic project-derived data sets. Included are assessments of performance in real-world iterative design applications and the utility of target structural information. The results suggest that combining screening and target structure information is robust; further, a well-designed screening library can compensate for lacking structural information.

A Study on the Interaction between p60c-Src Receptor Tyrosine Kinase and Arylcarboxylic and Arylacetic Acid Derivatives Based on Docking Modes and in Vitro Activity

The fundamental role that receptor tyrosine kinases play in cancer and other proliferative diseases has provided the impetus for an extensive effort on the part of both academic and pharmaceutical laboratories to develop highly specific inhibitors. In this study, inhibitory activity of previously synthesized arylacetic and arylcarboxylic acid derivatives were examined against substrate of tyrosine kinase. It can be assumed that the activity of compounds becomes higher when the -CH(2) linkage exist between aromatic ring and the amide group of the side chain. In addition, when the R(1) and R(2) substitutents are methyl group in both series, the higher activity observed. The data obtained from docking study (DOCK4.0) indicated that compounds 2, 4, 7, 8, 11 render satisfactory interaction with the active site of enzyme, Lys295 of p60(c-Src) tyrosine kinase. Comparison of this interaction and the evaluation of biological data showed that compound 4 is the most active among the entire derivatives.

Molecular recognition and docking algorithms

Molecular docking is an invaluable tool in modern drug discovery. This review focuses on methodological developments relevant to the field of molecular docking. The forces important in molecular recognition are reviewed and followed by a discussion of how different scoring functions account for these forces. More recent applications of computational chemistry tools involve library design and database screening. Last, we summarize several critical methodological issues that must be addressed in future developments.

Identification of a novel class of inhibitor of human and Escherichia coli thymidine phosphorylase by in silico screening

Structure-based computational screening of the National Cancer Institute database of anticancer compounds identified novel non-nucleobase-derived inhibitors of human thymidine phosphorylase as candidates for lead optimization. The hierarchical in silico screening strategy predicted potentially strong low molecular weight ligands exhibiting a range of molecular scaffolds. Of the thirteen ligands assayed for activity, all displayed inhibitory activity against Escherichia coli thymidine phosphorylase. One compound, hydrazine carboxamide 2-[(1-methyl-2,5-dioxo-4-pentyl-4-imidazolidinyl)methylene], was found to inhibit E. coli thymidine phosphorylase with an IC(50) value of 20 microM and an IC(50) value of 77 microM against human thymidine phosphorylase. As this hydantoin derivative lacks the undesirable ionic sites of existing tight-binding nucleobase-derived inhibitors, such as 5-chloro-6-[(2-iminopyrrolidin-1-yl)methyl]uracil hydrochloride, it provides an opportunity for the design of potent thymidine phosphorylase inhibitors with improved pharmacokinetic properties.

Structure of tRNA pseudouridine synthase TruB and its RNA complex: RNA recognition through a combination of rigid docking and induced fit

RNA pseudouridine synthase, TruB, catalyzes pseudouridine formation at U55 in tRNA. This posttranscriptional modification is almost universally conserved and occurs in the T arm of most tRNAs. We determined the crystal structure of Escherichia coli TruB apo enzyme, as well as the structure of Thermotoga maritima TruB in complex with RNA. Comparison of the RNA-free and -bound forms of TruB reveals that this enzyme undergoes significant conformational changes on binding to its substrate. These conformational changes include the ordering of the "thumb loop," which binds right into the RNA hairpin loop, and a 10 degree hinge movement of the C-terminal domain. Along with the result of docking experiments performed on apo TruB, we conclude that TruB recognizes its RNA substrate through a combination of rigid docking and induced fit, with TruB first rigidly binding to its target and then maximizing the interaction by induced fit.

Virtual screening of virtual libraries

Virtual screening of virtual libraries (VSVL) is a rapidly changing area of research. Great efforts are being made to produce better algorithms, selection methods and infrastructure. Yet, the number of successful examples in the literature is not impressive, although the quality of work certainly is high. Why is this? One reason is that these methods tend to be applied at the lead generation stage and therefore there is a large lead-time before successful examples appear in the literature. However, any computational chemist would confirm that these methods are successful and there exists a glut of start-up companies specialising in virtual screening. Moreover, the scientific community would not be focussing so much attention on this area if it were not yielding results. Even so, the paucity of literature data is certainly a hindrance to the development of better methods. The VSVL process is unique within the discovery process, in that it is the only method that can screen the > 10(30) genuinely novel molecules out there. Already, some VSVL methods are evaluating 10(13) compounds, a capacity that high throughput screening can only dream of. There is a huge potential advantage for the company that develops efficient and effective methods, for lead generation, lead hopping and optimization of both potency and ADME properties. To do this, it requires more than the software, it requires confidence to exploit the methodology, to commit synthesis on the basis of it, and to build this approach into the medicinal chemistry strategy. It is a fact that these tools remain quite daunting for the majority of scientists working at the bench. The routine use of these methods is not simply a matter of education and training. Integration of these methods into accessible and robust end user software, without dilution of the science, must be a priority. We have reached a coincidence, where several technologies have the required level of maturity predictive computational chemistry methods, algorithms that manage the combinatorial explosion, high throughput crystallography and ADME measurements and the massive increase in computational horsepower from distributed computing. The author is confident that the synergy of these technologies will bring great benefit to the industry, with more efficient production of higher quality clinical candidates. The future is bright. The future is virtual!

Identification of small-molecule inhibitors of human angiogenin and characterization of their binding interactions guided by computational docking

Angiogenin (ANG) is a potent inducer of angiogenesis and an RNase A homologue whose ribonucleolytic activity is essential for its biological action. Recently, we reported the identification of small non-nucleotide inhibitors of the enzymatic activity of ANG by high-throughput screening (HTS) [Kao, R. Y. T., et al. (2002) Proc. Natl. Acad. Sci. U.S.A. 99, 10066-10071]. Two of the inhibitors that were obtained, National Cancer Institute compound NSC-65828 [8-amino-5-(4'-hydroxybiphenyl-4-ylazo)naphthalene-2-sulfonate] and ChemBridge compound C-181431 [4,4'-dicarboxy-3,3'-bis(naphthylamido)diphenylmethanone], were judged to be suitable for further development, and one of these (NSC-65828) was shown to possess antitumor activity in mice. Here we have used computational docking as a guide for the identification of available NSC-65828 and C-181431 analogues that bind more tightly to ANG, and for the characterization of inhibitor binding modes. Numerous analogues were found to have greater avidity than the HTS compounds or any small nucleotide inhibitors; four were considered to be of interest as potential leads (K(i) = 5-25 microM). Two of these analogues bind more tightly to ANG than to RNase A, and are the first small molecules shown to exhibit this selectivity. The predicted binding orientations of the HTS compounds and the new lead inhibitors were evaluated by determining the effects of ANG active site mutations on inhibitory potency. The results with ANG variants R5A, H8A, N68A, and des(121-123) are highly consistent with the docking models. Affinity changes observed with Q12A and Q117G reveal aspects of active site function that are not apparent from the free ANG crystal structure or from the modeled complexes. These findings should prove to be useful in the design of more effective and specific ANG antagonists.