Minimizing false positives in kinase virtual screens

In spite of recent improvements in docking and scoring methods, high false-positive rates remain a common issue in structure-based virtual screening. In this study, the distinctive features of false positives in kinase virtual screens were investigated. A series of retrospective virtual screens on kinase targets was performed on specifically designed test sets, each combining true ligands and experimentally confirmed inactive compounds. A systematic analysis of the docking poses generated for the top-ranking compounds highlighted key aspects differentiating true hits from false positives. The most recurring feature in the poses of false positives was the absence of certain key interactions known to be required for kinase binding. A systematic analysis of 444 crystal structures of ligand-bound kinases showed that at least two hydrogen bonds between the ligand and the backbone protein atoms in the kinase hinge region are present in 90% of the complexes, with very little variability across targets. Closer inspection showed that when the two hydrogen bonds are present, one of three preferred hinge-binding motifs is involved in 96.5% of the cases. Less than 10% of the false positives satisfied these two criteria in the minimized docking poses generated by our standard protocol. Ligand conformational artifacts were also shown to contribute to the occurrence of false positives in a number of cases. Application of this knowledge in the form of docking constraints and post-processing filters provided consistent improvements in virtual screening performance on all systems. The false-positive rates were significantly reduced and the enrichment factors increased by an average of twofold. On the basis of these results, a generalized two-step protocol for virtual screening on kinase targets is suggested.     

Pocket detection and interaction-weighted ligand-similarity search yields novel high-affinity binders for Myocilin-OLF,a protein implicated in glaucoma

Traditional structure and ligand based virtual screening approaches rely on the availability of structural and ligand binding information. To overcome this limitation, hybrid approaches were developed that relied on extraction of ligand binding information from proteins sharing similar folds and hence, evolutionarily relationship. However, they cannot target a chosen pocket in a protein. To address this, a pocket centric virtual ligand screening approach is required. Here, we employ a new, iterative implementation of a pocket and ligand-similarity based approach to virtual ligand screening to predict small molecule binders for the olfactomedin domain of human myocilin implicated in glaucoma. Small-molecule binders of the protein might prevent the aggregation of the protein, commonly seen during glaucoma. First round experimental assessment of the predictions using differential scanning fluorimetry with myoc-OLF yielded 7 hits with a success rate of 12.7%; the best hit had an apparent dissociation constant of 99 nM. By matching to the key functional groups of the best ligand that were likely involved in binding, the affinity of the best hit was improved by almost 10,000 fold from the high nanomolar to the low picomolar range. Thus, this study provides preliminary validation of the methodology on a medically important glaucoma associated protein.     

Rapid filtration assays for protein kinase C activity and phorbol ester binding using multiwell plates with fitted filtration discs.

In the conventional approach protein kinase activity and phorbol ester binding associated with protein kinase C (PKC) are measured by initially incubating samples in either test tubes or multiwell plates, followed by filtration of the terminated reaction mixture using either a manifold filtration device or a cell harvester. Here we report a method in which both the incubations and filtrations necessary for the determination of either protein kinase activity or phorbol ester binding are carried out in the same multiwell plate with fitted filtration discs made of polyvinylidene difluoride (Durapore membrane). Due to the very low binding of protein to these filters, there is no interference caused by these filters during the incubation period of the assays. The drawback with these filters compared to commonly used cellulose acetate membrane filters is that they retain less of the phosphate acceptor substrate histone H1 (only 15%) if filtered and washed with standard 5% trichloroacetic acid. However, this can be overcome by increasing the trichloroacetic acid concentration to 25% during filtration. For phorbol ester binding determinations, the samples are incubated with [3H]phorbol 12,13-dibutyrate in the microwells, the ligand bound PKC is adsorbed onto DEAE-Sephadex beads, and the beads then are filtered and washed in the same microwells. Furthermore, this multiwell filtration approach can also be adopted to previously described cytosolic phorbol ester receptor assays, which have the broader conditions for optimal binding to receptors. Durapore membrane filters are found to work well for punching into scintillation vials and there is complete recovery of the radioactivity retained with the filters. In the protein kinase assay the background radioactivity is very low (< 200 cpm) and in the phorbol ester binding assay the nonspecific binding is less than 1%. Thus, these low background values result in at least a fourfold increase in sensitivity for these assays. Since the incubations and filtrations are carried out in the same well without any transfer of the sample, the coefficient of variation in multiple determinations is found to be low. Furthermore, this method is rapid and more convenient for analyzing a larger number of samples than conventional methods which use test tubes, and it is less expensive to set up compared to the automated methods that use a cell harvester.     

Inhibitors of human 2,3-oxidosqualene cyclase (OSC) discovered by virtual screening

Five human 2,3-oxidosqualnene cyclase (OSC) inhibitors were discovered using the combination of a virtual screening based on a docking study and an isotope-free assay system. All of these inhibitors were revealed to have activities comparable or superior to that of LDAO, a known OSC inhibitor. The computational study of the newly identified inhibitors suggested that CH/π interactions with F444 and W581 contribute to the binding, and these interactions are candidates for additional structural filters in the next generation of virtual screening.     

A rapid filtration assay for cAMP

The receptor-binding assay for cAMP was improved by using polyethylenimine-treated glass filters. A polyethylenimine-treated glass filter has high protein binding capacity. This high capacity allows an increase in the amount of protein per assay tube and the use of a crude preparation, such as a beef heart extract, as specific binding protein instead of a purified protein, which has been used in the classical filtration assays involving cellulose ester filters. Since the time required for the separation of the protein-cAMP complex and the free nucleotide can be shortened by the use of polyethylenimine-treated filters, the dissociation of the bound ligand during the separation procedure, which is a serious problem with other modified assay methods involving charcoal adsorption, is minimized. Filtration through polyethylenimine-treated glass filters also gives low blanks and prevents the loss of protein or ligand due to breakage of the filters, which is often observed with fragile cellulose ester membranes. In consequence, this simple and rapid filtration assay allows more accurate and reproducible determinations.     

Structure‐guided screening of chemical database to identify NS3‐NS2B inhibitors for effective therapeutic application in dengue infection

Dengue infection is the most common arthropod‐borne disease caused by dengue viruses, predominantly affecting millions of human beings annually. To find out promising chemical entities for therapeutic application in Dengue, in the current research, a multi‐step virtual screening effort was conceived to screen out the entire “screening library” of the Asinex database. Initially, through “Lipinski rule of five” filtration criterion almost 0.6 million compounds were collected and docked with NS3‐NS2B protein. Thereby, the chemical space was reduced to about 3500 compounds through the analysis of binding affinity obtained from molecular docking study in AutoDock Vina. Further, the “Virtual Screening Workflow” (VSW) utility of Schrödinger suite was used, which follows a stepwise multiple docking programs such as ‐ high‐throughput virtual screening (HTVS), standard precision (SP), and extra precision (XP) docking, and in postprocessing analysis the MM‐GBSA based free binding energy calculation. Finally, five potent molecules were proposed as potential inhibitors for the dengue NS3‐NS2B protein based on the investigation of molecular interactions map and protein‐ligand fingerprint analyses. Different pharmacokinetics and drug‐likeness parameters were also checked, which favour the potentiality of selected molecules for being drug‐like candidates. The molecular dynamics (MD) simulation analyses of protein‐ligand complexes were explained that NS3‐NS2B bound with proposed molecules quite stable in dynamic states as observed from the root means square deviation (RMSD) and root means square fluctuation (RMSF) parameters. The binding free energy was calculated using MM‐GBSA method from the MD simulation trajectories revealed that all proposed molecules possess such a strong binding affinity towards the dengue NS3‐NS2B protein. Therefore, proposed molecules may be potential chemical components for effective inhibition of dengue NS3‐NS2B protein subjected to experimental validation.     

How to choose relevant multiple receptor conformations for virtual screening: a test case of Cdk2 and normal mode analysis

Better treatment of protein flexibility is essential in structure-based drug design projects such as virtual screening and protein-ligand docking. Diversity in ligand-binding mechanisms and receptor conformational changes makes it difficult to treat dynamic features of the receptor during the docking simulation. Thus, the use of pregenerated multiple receptor conformations is applied today in virtual screening studies. However, generation of a small relevant set of receptor conformations remains challenging. To address this problem, we propose a new protocol for the generation of multiple receptor conformations via normal mode analysis and for the selection of several receptor conformations suitable for docking/virtual screening. We validated this protocol on cyclin-dependent kinase 2, which possesses a binding site located at the interface between two subdomains and is known to undergo significant conformational changes in the active site region upon ligand binding. We believe that the suggested rules for the choice of suitable receptor conformations can be applied to other targets when dealing with in silico screening on flexible receptors.     

Identification of chemically diverse Chk1 inhibitors by receptor-based virtual screening

Inhibition of the Chk1 kinase by small molecules is of great therapeutic interest for oncology and in understanding the cellular regulation of the G2/M checkpoint. We report how computational docking of a large electronic catalogue of compounds to an X-ray structure of the Chk1 ATP-binding site allowed prioritisation of a small subset of these compounds for assay. This led to the discovery of 10 novel Chk1 inhibitors, distributed among nine new and clearly different chemical scaffolds. Several of these scaffolds have promising lead-like properties. All these ligands act by competitive binding to the targeted ATP site. The crystal structures of four of these compounds bound to this site are presented, and reasonable modelled docking modes are suggested for the 5 other scaffolds. This structural context is used to assess the potential of these scaffolds for further medicinal chemistry efforts, suggesting that several of them could be elaborated to make additional interactions with the buried part of the ATP site. Some unusual interactions with the conserved kinase backbone motif are pointed out. The ligand-binding modes are also used to discuss their medicinal chemistry potential with respect to undesirable chemical functionalities, whether these functionalities bind directly to the protein or not. Overall, this work illustrates how virtual screening can identify a diverse set of ligands which bind to the targeted site. The structural models for these ligands in the Chk1 ATP-binding site will facilitate further medicinal chemistry efforts targeting this kinase.     

Statistical potentials and scoring functions applied to protein-ligand binding

In virtual screening, small-molecule ligands are docked into protein binding sites and their binding affinity is predicted. Knowledge-based, regression-based and first-principle-based methods have been developed to rank computer-generated binding modes. As a result of still existing deficiencies, a best compromise might be the combination of several scoring schemes into a consensus scoring approach.     

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.     

Ensemble docking of multiple protein structures: considering protein structural variations in molecular docking

One approach to incorporate protein flexibility in molecular docking is the use of an ensemble consisting of multiple protein structures. Sequentially docking each ligand into a large number of protein structures is computationally too expensive to allow large-scale database screening. It is challenging to achieve a good balance between docking accuracy and computational efficiency. In this work, we have developed a fast, novel docking algorithm utilizing multiple protein structures, referred to as ensemble docking, to account for protein structural variations. The algorithm can simultaneously dock a ligand into an ensemble of protein structures and automatically select an optimal protein structure that best fits the ligand by optimizing both ligand coordinates and the conformational variable m, where m represents the m-th structure in the protein ensemble. The docking algorithm was validated on 10 protein ensembles containing 105 crystal structures and 87 ligands in terms of binding mode and energy score predictions. A success rate of 93% was obtained with the criterion of root-mean-square deviation <2.5 A if the top five orientations for each ligand were considered, comparable to that of sequential docking in which scores for individual docking are merged into one list by re-ranking, and significantly better than that of single rigid-receptor docking (75% on average). Similar trends were also observed in binding score predictions and enrichment tests of virtual database screening. The ensemble docking algorithm is computationally efficient, with a computational time comparable to that for docking a ligand into a single protein structure. In contrast, the computational time for the sequential docking method increases linearly with the number of protein structures in the ensemble. The algorithm was further evaluated using a more realistic ensemble in which the corresponding bound protein structures of inhibitors were excluded. The results show that ensemble docking successfully predicts the binding modes of the inhibitors, and discriminates the inhibitors from a set of noninhibitors with similar chemical properties. Although multiple experimental structures were used in the present work, our algorithm can be easily applied to multiple protein conformations generated by computational methods, and helps improve the efficiency of other existing multiple protein structure(MPS)-based methods to accommodate protein flexibility.     

A strategy for generating monoclonal antibodies against recombinant baculovirus-produced proteins: application to the Bcl-2 oncoprotein.

A strategy is described for production of monoclonal antibodies against recombinant proteins that are produced using the baculovirus expression system and that requires no prior purification of the protein of interest. Crude lysates prepared from cultured Sf9 insect cells infected with recombinant or control baculoviruses are absorbed to nitrocellulose filters and used in a dot-immunobinding assay for screening hybridomas. The monoclonal antibody-producing hybridomas are derived by immunization of mice with a synthetic peptide corresponding to a hydrophilic region in the recombinant protein of interest. By using the baculovirus-produced recombinant protein as the screening antigen and by comparing antibody binding to filters containing control Sf9 lysates, hybridomas are identified that produce monoclonal antibodies with specific reactivity for the recombinant protein of interest and that can then subsequently be used to assist in the large-scale purification of the recombinant protein from baculovirus-infected cells. We applied this method to recombinant 26-kDa human Bcl-2 (B-cell lymphoma/leukemia-2), an integral membrane oncoprotein that regulates programmed cell death ("apoptosis") in hematolymphoid cells through unknown mechanisms. Two mouse monoclonal antibodies were produced that specifically bound the recombinant Bcl-2 baculoprotein in both solution and solid-phase assays.     

Identification of PDZ ligands by docking-based virtual screening for the development of novel analgesic agents

Disrupting the interaction between the PDZ protein, PSD-95, and its target ligands (such as the glutamate NMDA receptor or the serotonin 5-HT_2A receptor) was found to reduce hyperalgesia in various models of neuropathic pain. Here, we set out to identify lead molecules which would interact with PSD-95, and hence, would potentially display analgesic activity. We describe the virtual screening of the Asinex and Cambridge databases which together contain almost one million molecules. Using three successive docking filters and visual inspection, we identified three structural classes of molecules and synthesized a potential lead compound from each class. The binding of the molecules with the PDZ domains of PSD-95 was assessed by ¹H–¹⁵N HSQC NMR experiments. The analgesic activity of the best ligand, quinoline 2, was evaluated in vivo in a model of neuropathic pain and showed promising results.     

Uncovering false positives on a virtual screening search for cruzain inhibitors

Some unexpected promiscuous inhibitors were observed in a virtual screening protocol applied to select cruzain inhibitors from the ZINC database. Physical-chemical and pharmacophore model filters were used to reduce the database size. The selected compounds were docked into the cruzain active site. Six hit compounds were tested as inhibitors. Although the compounds were designed to be nucleophilically attacked by the catalytic cysteine of cruzain, three of them showed typical promiscuous behavior, revealing that false positives are a prevalent concern in VS programs.     

A competitive receptor binding radioassay for beta-1 and beta-2 adrenergic agents

A rapid and sensitive competitive receptor binding assay for beta-1 and beta-2 adrenergic binding for adrenergic agents has been developed. The steps that are critical for the success of the assay are given in detail so that the assay can be set up in any routine laboratory with relative ease. The rationale behind the use of specific reagents is discussed. The assay requires microgram quantities of test compound, a radiolabeled specific beta adrenergic antagonist [3H]dihydroalprenolol (DHA), and turkey erythrocyte beta-1 and rat erythrocyte beta-2 receptor membranes. Serial dilutions of sample are incubated with appropriate receptor membranes and DHA for 1 hr at room temperature. After equilibrium is attained, the bound radioligand is separated by rapid filtration under vacuum through Whatman GF/B filters. The amount of bound DHA trapped on the filter is inversely proportional to the degree of beta-1 or beta-2 adrenergic binding of the sample. Separation of bound from free radioligand by filtration permits rapid determination of a large number of samples. This assay quantitates and differentiates beta-1 and beta-2 adrenergic binding of synthetic adrenergic agents.     

Developing novel approaches to improve binding energy estimation and virtual screening: a PARP case study

Poly-(ADP-ribose)-polymerase (PARP) is a promising anti-cancer target as it plays a crucial role in the cellular reparation and survival mechanisms. However, the development of a robust and cost effective experimental technique to screen PARP inhibitors is still a scientific challenge owing to the difficulties in quantitative detection of the enzyme activity. In this work we demonstrate that the computational chemistry tools including molecular docking and scoring can perform on par with the experimental studies in assessing binding constants and in the recovery of active compounds in virtual screening. Using the recently introduced Lead Finder software we were able to dock a set of 142 well characterized PARP inhibitors and obtain a good correlation between the calculated and experimentally measured binding energies with the rmsd of 1.67 kcal mol⁻¹. Additionally, fine-tuning of the energy scaling coefficients within the Lead Finder scoring function has further decreased rmsd to the value of 0.88 kcal mol⁻¹. Moreover, we were able to reproduce the selectivity of ligand binding between the two isoforms of the enzyme-PARP1 and PARP2-suggesting that the Lead Finder software can be used to design isoform-selective inhibitors of PARP. An impressive enrichment was obtained in the virtual screening experiment, in which the mentioned set of PARP inhibitors was mixed with a commercial library of 300,000 compounds. We also demonstrate that the virtual screening performance can be significantly improved by an additional structural filtration of the docked ligand poses through detection of the crucial hydrogen bonding interactions with the enzyme.     

Electrostatic contributions to protein-protein interactions: fast energetic filters for docking and their physical basis

The methods of continuum electrostatics are used to calculate the binding free energies of a set of protein-protein complexes including experimentally determined structures as well as other orientations generated by a fast docking algorithm. In the native structures, charged groups that are deeply buried were often found to favor complex formation (relative to isosteric nonpolar groups), whereas in nonnative complexes generated by a geometric docking algorithm, they were equally likely to be stabilizing as destabilizing. These observations were used to design a new filter for screening docked conformations that was applied, in conjunction with a number of geometric filters that assess shape complementarity, to 15 antibody-antigen complexes and 14 enzyme-inhibitor complexes. For the bound docking problem, which is the major focus of this paper, native and near-native solutions were ranked first or second in all but two enzyme-inhibitor complexes. Less success was encountered for antibody-antigen complexes, but in all cases studied, the more complete free energy evaluation was able to identify native and near-native structures. A filter based on the enrichment of tyrosines and tryptophans in antibody binding sites was applied to the antibody-antigen complexes and resulted in a native and near-native solution being ranked first and second in all cases. A clear improvement over previously reported results was obtained for the unbound antibody-antigen examples as well. The algorithm and various filters used in this work are quite efficient and are able to reduce the number of plausible docking orientations to a size small enough so that a final more complete free energy evaluation on the reduced set becomes computationally feasible.     

Binding of [3H] Cocaine in Mouse Brain: Kinetics and Saturability

Abstract

Kinetic experiments indicate that association of [³H]- cocaine to its binding site in brain occurs rapidly (seconds). Dissociation of membrane-bound cocaine is also rapid, with a dissociation half-life in seconds; this raises the question of whether membrane-bound cocaine is released during the time required for rapid filtration of tissue-containing filters. Results from experiments with increasing numbers of filterwashes indicate that there is no significant loss of [³H]-cocaine saturably bound to brain membranes within the timescale of the rapid filtration procedure. In addition, saturation analysis of binding data obtained with the filtration procedure and with the centrifugation method give similar estimates of the affinity (dissociation constant: 0.8 μM) and the maximal binding (5 pmol/mg of protein) of cocaine. However, the nonspecific binding and the experimental error in the saturable binding are considerably greater in centrifugation assays than in filtration assays.     

Pharmacophore modeling and structure-based virtual screening to identify potent inhibitors targeting LuxP of Vibrio harveyi

The main aim of the study is to identify molecules that can disrupt quorum sensing (QS) system of Vibrio harveyi and therefore perhaps the production of toxins. Recently, a novel class of dioxazaborocane derivatives has been found to block AI-2 QS by targeting LuxPQ, but the mechanism of protein inhibition is still unclear. In order to investigate the possible binding modes, all the derivatives were docked into the binding site of LuxP using induced fit docking (IFD). The computed binding affinity is in good agreement with the experimental data. Resultant protein–ligand complexes were simulated using Desmond module and the result revealed better binding of ligands in the binding site of LuxP. Both pharmacophore- and structure-based virtual screening was performed to identify novel hits against LuxP. A filtering protocol, including lipinski filters, number of rotatable bonds and three levels of docking precisions were used for the selection of hits with specific properties. The virtual screening results were then combined and analyzed, which retrieved six hits with significant Glide score, binding affinity toward LuxP. The pharmacokinetic properties of the retrieved hits are in the acceptable range. Enrichment calculation was performed to validate the final hits, to discriminate the active compounds from the inactive compounds. The identified hits could serve as a base for further drug development against LuxP of Vibrio harveyi.