No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Drug profiling: Knowing where it hits
Abstract
Off-target hits of drugs can lead to serious adverse effects or, conversely, to unforeseen alternative medical utility. Selectivity profiling against large panels of potential targets is essential for the drug discovery process to minimize attrition and maximize therapeutic utility. Lately, it has become apparent that drug promiscuity (polypharmacology) goes well beyond target families; therefore, lowering the profiling costs and expanding the coverage of targets is an industry-wide challenge to improve predictions. Here, we review current and promising drug profiling alternatives and commercial solutions in these exciting emerging fields.
... However, a biological target may regulate more than one biological pathway, only one of which may be disease related. If this is the case, then altering the function of a biological target by a drug can lead to unintended results of disruption of healthy pathways [8]. ...
... The FMCM framework for selecting therapeutic drug compound consisted of two segments, selection of functional modules of predicted cancer genes based on the GSToP procedure [26] (steps 1-5 below), and multiple queries, one for each functional module, of the CMap for drug identification (steps [6][7][8]. Steps in the selection procedure ( Figure 1 [17] to obtain for each function two lists respectively for predicted beneficial (ES ,20.5) and harmful (ES .0.3) drugs (see above for requirement on randomization p-value). (7) Function-drug association map (FDAM). ...
... Include in FDAM only drugs that have at least one beneficial link. (8) Construct from FDAM all predicted drug compounds, where a compound is minimum set of purely beneficial drugs that covered all functions. ...
Drug repurposing has become an increasingly attractive approach to drug development owing to the ever-growing cost of new drug discovery and frequent withdrawal of successful drugs caused by side effect issues. Here, we devised Functional Module Connectivity Map (FMCM) for the discovery of repurposed drug compounds for systems treatment of complex diseases, and applied it to colorectal adenocarcinoma. FMCM used multiple functional gene modules to query the Connectivity Map (CMap). The functional modules were built around hub genes identified, through a gene selection by trend-of-disease-progression (GSToP) procedure, from condition-specific gene-gene interaction networks constructed from sets of cohort gene expression microarrays. The candidate drug compounds were restricted to drugs exhibiting predicted minimal intracellular harmful side effects. We tested FMCM against the common practice of selecting drugs using a genomic signature represented by a single set of individual genes to query CMap (IGCM), and found FMCM to have higher robustness, accuracy, specificity, and reproducibility in identifying known anti-cancer agents. Among the 46 drug candidates selected by FMCM for colorectal adenocarcinoma treatment, 65% had literature support for association with anti-cancer activities, and 60% of the drugs predicted to have harmful effects on cancer had been reported to be associated with carcinogens/immune suppressors. Compounds were formed from the selected drug candidates where in each compound the component drugs collectively were beneficial to all the functional modules while no single component drug was harmful to any of the modules. In cell viability tests, we identified four candidate drugs: GW-8510, etacrynic acid, ginkgolide A, and 6-azathymine, as having high inhibitory activities against cancer cells. Through microarray experiments we confirmed the novel functional links predicted for three candidate drugs: phenoxybenzamine (broad effects), GW-8510 (cell cycle), and imipenem (immune system). We believe FMCM can be usefully applied to repurposed drug discovery for systems treatment of other types of cancer and other complex diseases.
... Second, target-based screens are typically performed in vitro by inspecting individual cellular components such as purified enzymes or ligand-receptor complexes. In such experiments, a small chemical molecule producing a strong phenotype in vitro may have side effects on the cell, owing to its binding of different proteins within the cell 7 . In addition, the premise upon which the target was selected is often based on incomplete knowledge at the systems level. ...
... In addition, the premise upon which the target was selected is often based on incomplete knowledge at the systems level. This implies that the effect observed in vivo may be different from the expected phenotype, even in the case of validated targets 7 . These factors have a key role in the high rate of attrition in drug discovery. ...
... In contrast, silencing of a subunit of a complex can lead to disruption and even degradation of the entire complex. Third, both a compound and a siRNA can have multiple 'off-targets' in the cell 7 . Thus, even if the compound and the siRNA have the same molecular target, the off-target effects need not be conserved, resulting in differences in the multiparametric profiles produced. ...
Cell-based high-content screens are increasingly used to discover bioactive small molecules. However, identifying the mechanism of action of the selected compounds is a major bottleneck. Here we describe a protocol consisting of experimental and computational steps to identify the cellular pathways modulated by chemicals, and their mechanism of action. The multiparametric profiles from a 'query' chemical screen are used as constraints to select genes with similar profiles from a 'reference' genetic screen. In our case, the query screen is the intracellular survival of mycobacteria and the reference is a genome-wide RNAi screen of endocytosis. The two disparate screens are bridged by an 'intermediate' chemical screen of endocytosis, so that the similarity in the multiparametric profiles between the chemical and the genetic perturbations can generate a testable hypothesis of the cellular pathways modulated by the chemicals. This approach is not assay specific, but it can be broadly applied to various quantitative, multiparametric data sets. Generation of the query system takes 3-6 weeks, and data analysis and integration with the reference data set takes an 3 additional weeks.
... locks) that are associated with a desired clinical effect and that might not be from the same target family. As discussed by Merino et al., it is important to emphasize that there are different levels of polypharmacology (or promiscuity) that could lead to positive or negative effects [11]. Such levels will largely depend on the dose of the drug. ...
... It might be that, at therapeutic doses, a drug has a positive clinical effect owing to its interaction with multiple targets. However, depending on the dose, interactions of the drug with antitargets will lead to adverse effects [11]. ...
... In fact, many drugs have been identified through the initial in vivo testing of either natural product mixtures or individual compounds. Interestingly, several approved drugs were developed based on favorable animal model data without a clear understanding of the mechanism of action of the drug at the molecular level [11]. In fact, as pointed out by Merino et al., 'despite the broad scientific knowledge around human drug targets, today's pharmacopoeia still includes many drugs that are being prescribed with unknown Mechanism of Action' [11]. ...
Increasing evidence that several drug compounds exert their effects through interactions with multiple targets is boosting the development of research fields that challenge the data reductionism approach. In this article, we review and discuss the concepts of drug repurposing, polypharmacology, chemogenomics, phenotypic screening and high-throughput in vivo testing of mixture-based libraries in an integrated manner. These research fields offer alternatives to the current paradigm of drug discovery, from a one target-one drug model to a multiple-target approach. Furthermore, the goals of lead identification are being expanded accordingly to identify not only 'key' compounds that fit with a single-target 'lock', but also 'master key' compounds that favorably interact with multiple targets (i.e. operate a set of desired locks to gain access to the expected clinical effects).
... Studying interactions between bioactive compounds or drugs and biological targets, experimentally and/or computationally, has become a major topic in pharmaceutical research (1)(2)(3)(4)(5)(6)(7)(8). Compound promiscuity is rationalized as the ability of small molecules to specifically interact with multiple targets (but does not imply non-specific binding effects). ...
... Compound promiscuity is rationalized as the ability of small molecules to specifically interact with multiple targets (but does not imply non-specific binding effects). The increasing notion of compound promiscuity and ensuing polypharmacology is beginning to change specificity-centric drug discovery paradigms, at least in some therapeutic areas such as oncology (5,6). In addition, target promiscuity is also responsible for undesired (or desired) side effects of drugs (9,10). ...
Compound promiscuity refers to the ability of small molecules to specifically interact with multiple targets, which represents the origin of polypharmacology. Promiscuity is thought to be a widespread characteristic of pharmaceutically relevant compounds. Yet, the degree of promiscuity among active compounds from different sources remains uncertain. Here, we report a thorough analysis of compound promiscuity on the basis of more than 1,000 PubChem confirmatory bioassays, which yields an upper-limit assessment of promiscuity among active compounds. Because most PubChem compounds have been tested in large numbers of assays, data sparseness has not been a limiting factor for the current analysis. We have determined that there is an overall likelihood of ∼50% of an active PubChem compound to interact with two or more targets. The probability to interact with more than five targets is reduced to 7.6%. On average, an active PubChem compound was found to interact with ∼2.5 targets. Moreover, if only activities consistently detected in all assays available for a given target were considered, this ratio was further reduced to ∼2.3 targets per compound. For comparison, we have also analyzed high-confidence activity data from ChEMBL, the major public repository of compounds from medicinal chemistry, and determined that an active ChEMBL compound interacted on average with only ∼1.5 targets. Taken together, our results indicate that the degree of compound promiscuity is lower than often assumed.
... This concept was coined in analogy to the seminal lock and key model, proposed by Erlich in 1894. Furthermore, there is a thin line between the therapeutic and toxic effects of a drug, which reinforces the complexity that underlies the interaction with the biological system [7,8]. The applicability of the PAINS filter should be restricted to libraries used in biochemical assays (molecular targetbased), and the latter is just one of the two main drug screening strategies currently adopted. ...
... Privileged structures can act as molecular master keys (see Glossary) to open several locks (receptors) to reach the desirable clinical effect [5,6]. The level of promiscuity with antitargets will lead the adverse effects, so a selectivity barrier must be overtaken [8]. ...
... However, mutagenicity and geno74 toxicity of NZL remain unclear to date. Determining the potential 75 risks of NZL is important to establish NZL as a safe drug (Jena 76 et al., 2002;Li, 2004;Merino et al., 2010). ...
... Owing to the ubiquitous nature of halogens, medicinal chemists have been introducing halogens into drug candidates for increasing binding affinity and/or membrane permeability, facilitating the blood-brain barrier crossing, prolonging the lifetime of drugs, and so on [12,[27][28][29]. Moreover, halogen bonding has been further discovered to render its essential roles in multiple biological processes through regulating the molecular recognition, modulating the selectivity of small compounds, and driving biomolecular conformation, etc. [2,15,17,[30][31][32][33][34][35][36][37][38][39][40][41]. ...
A quantum mechanics-based scoring function for halogen bonding interaction, namely XBScore(QM), is developed based on 18,135 sets of geometrical and energetical parameters optimized at M06-2X/aug-cc-pVDZ level. Applying the function on typical halogen bonding systems from Protein Data Bank demonstrates its strong ability of predicting halogen bonding as attractive interaction with strength up to -4 kcal mol(-1). With a diverse set of proteins complexed with halogenated ligands, a systematic evaluation demonstrates the integrative advantage of XBScore(QM) over 12 other scoring functions on halogen bonding in four aspects, viz. pseudo docking power, ranking power, scoring power, and genuine docking power. Thus, this study not only provides a practicable scoring function of halogen bonding for high throughput virtual screening, but also serves as a benchmark for evaluating the performance of current scoring functions on characterizing halogen bonding.
... The previous discussion explains why multi-target drugs are being pursued today and also sets the logic ground for drug repositioning. However, it should be underlined that selectively non-selective drugs and promiscuous drugs are not exactly equivalent concepts in a drug repositioning scenario: while a certain, convenient degree of promiscuity may be desirable, an excess would certainly represent serious safety issues [67, 68]. Network-based approaches focused on drug repositioning may prove helpful to select candidates with an adequate degree of polypharmacology depending on the pursued new indication. ...
Drug repurposing/reprofiling has attracted considerable attention during the last decade. The object of such approach is to discover second or further medical uses of known chemicals, i. e. targeting existing, withdrawn or abandoned drugs, or yet to be pursued clinical candidates to new disease areas. Recently (2011-2012), the US and UK governments launched public-private joint initiatives towards finding new uses of previously shelved compounds (drug rescue). While in the past repurposing emerged from serendipitous findings and/or from rational exploitation of drug side-effects (e.g. sildenafil, aspirin), the current tendency in the drug development field focuses on knowledge-based drug repurposing, particularly, computer-aided repositioning approaches. The present chapter reviews different cheminformatic and bioinformatic applications, as well as high-throughput literature analysis, oriented to the discovery of new medical uses of known drugs. Applications of such strategies to the discovery of innovative medications for neglected or rare diseases are discussed. Finally, we also review publicly available resources (e.g. chemical libraries) valuable for reprofiling.
... The last decade has witnessed an increasing awareness that drugs often bind to more than one molecular target, exhibiting polypharmacology [74][75][76][77][78][79][80][81]. Accordingly, drugs designed to selectively bind one specific protein may show unanticipated activities on additional targets. ...
Inhibition of members of the ADP-ribosyltransferases with diphtheria toxin homology (ARTD), widely known as the poly(ADP-ribose) polymerase (PARP) family, is a strategy under development for treatment of various conditions, including cancers and ischemia. Here, we give a brief summary of ARTD enzyme functions and the implications for their potential as therapeutic targets. We present an overview of the PARP inhibitors that have been enrolled in clinical trials. Finally, we summarize recent insights from structural biology, and discuss the molecular aspects of PARP inhibitors in terms of broad-range vs. selective inhibition of ARTD-family enzymes. This article is protected by copyright. All rights reserved.
... Although drug candidates can be promiscuous, the level seems to vary according to the data set used. [21][22][23] Our work did not address pharmacological effects per se, but rather clinically effective plasma concentrations and their relative value to in vitro potency of the claimed primary target. Many of the drugs examined are also considered selective for their primary target, particularly taking the unbound exposures into consideration. ...
The in vitro affinity of a compound for its target is an important feature in drug discovery, but what remains is how predictive in vitro properties are of in vivo therapeutic drug exposure. We assessed the relationship between in vitro potency and clinically efficacious concentrations for marketed small molecule drugs (n=164) and how they may differ depending on therapeutic indication, mode‐of‐action, receptor type, target localisation and function. Approximately 70 % of compounds had a therapeutic unbound plasma exposure lower than in vitro potency; the median Ratio of exposure in relation to in vitro potency was 0.32, and 80% had Ratios within 0.007 and 8.7. We identified differences in the in vivo‐to‐in vitro potency Ratio between indications, mode‐of‐action, target type and matrix localisation, and whether or not the drugs had active metabolites. The in vitro‐assay variability contributions appeared to be the smallest, within the same drug target and mode of action the within‐variability was slightly broader, but both were substantially less compared to the overall distribution of Ratios. These data suggest that in vitro potency conditions, estimated in vivo potency, required level of receptor occupancy and target turnover are key components for further understanding the link between clinical drug exposure and in vitro potency.
... Polypharmacology has emerged as a new theme in drug discovery, especially for complex diseases such as AD that involve functional modulation of multiple proteins such as AD. [3][4][5] Polypharmacology focuses on the fact that one drug can hit multiple targets. Prediction of polypharmacology for known drugs is highly useful for finding new indication and explaining the molecular mechanism of action. ...
To determine chemical-protein interactions (CPI) is costly, time-consuming, and labor-intensive. In silico prediction of CPI can facilitate the target identification and drug discovery. Though many in silico target prediction tools have been developed, few of them could predict active molecules against multi-target for a single disease. In this investigation, naive Bayesian (NB) and recursive partitioning (RP) algorithms were applied to construct classifiers for predicting the active molecules against 25 key targets toward Alzheimer’s disease (AD) using multitarget-quantitative structure-activity relationships (mt-QSAR) method. Each molecule was initially represented with two kinds of fingerprint descriptors (ECFP6 and MACCS). 100 classifiers were constructed and their performance was evaluated and verified with internally five-fold cross-validation and external test set validation. The range of the area under the receiver operating characteristic curve (ROC) for the test sets was from 0.741 to 1.0, with an average of 0.965. In addition, the important fragments for multi-target against AD given by NB classifiers were also analyzed. Finally, the validated models were employed to systematically predict the potential targets for 6 approved anti-AD drugs, and 19 known active compounds related to AD. The prediction results were confirmed by reported bioactivity data and our in vitro experimental validation, resulting in several multi-target-directed ligands (MTDLs) against AD, including 7 acetylcholinesterase (AChE) inhibitors ranging from 0.442 to 72.26 μM and 4 histamine receptor 3 (H3R) antagonists ranging from 0.308 to 58.6 μM. To be exciting, the best MTDL DL0410 was identified as an dual cholinesterase inhibitor with IC50 values of 0.442 μM (AChE) and 3.57 μM (BuChE) as well as a H3R antagonist with an IC50 of 0.308 μM. This investigation is the first report using mt-QASR approach to predict chemical-protein interaction for a single disease and discovering highly potent MTDLs. This protocol may be useful for in silico multi-target prediction of other diseases.
... Due to the increase in the understanding of the commonalities in enzyme mechanisms and of enzyme "families", assay panels of proteins have become available in order to examine the selectivity of candidate inhibitor drugs. For example, see [1]. ...
Most enzymes act on more than a single substrate. There is frequently a need to block the production of a single pathogenic outcome of enzymatic activity on a substrate but to avoid blocking others of its catalytic actions. Full blocking might cause severe side effects because some products of that catalysis may be vital. Substrate selectivity is required but not possible to achieve by blocking the catalytic residues of an enzyme. That is the basis of the need for "Substrate Selective Inhibitors" (SSI), and there are several molecules characterized as SSI. However, none have yet been designed or discovered by computational methods. We demonstrate a computational approach to the discovery of Substrate Selective Inhibitors for one enzyme, Prolyl Oligopeptidase (POP) (E.C 3.4.21.26), a serine protease which cleaves small peptides between Pro and other amino acids. Among those are Thyrotropin Releasing Hormone (TRH) and Angiotensin-III (Ang-III), differing in both their binding (Km) and in turnover (kcat). We used our in-house "Iterative Stochastic Elimination" (ISE) algorithm and the structure-based "Pharmacophore" approach to construct two models for identifying SSI of POP. A dataset of ~1.8 million commercially available molecules was initially reduced to less than 12,000 which were screened by these models to a final set of 20 molecules which were sent for experimental validation (five random molecules were tested for comparison). Two molecules out of these 20, one with a high score in the ISE model, the other successful in the pharmacophore model, were confirmed by in vitro measurements. One is a competitive inhibitor of Ang-III (increases its Km), but non-competitive towards TRH (decreases its Vmax).
... These predictive studies consider the effects of many compounds on one (or a small number) of targets in order to identify promising compounds for further development. However, it is not uncommon in drug development for previously unknown effects to be discovered after significant investment in a potential drug, resulting in relatively high attrition rates in later phases or even after drug release [8]. These side effects are not discovered earlier because screening is for desired effects of compounds on a single target protein without considering whether compounds have undesired effects on other targets. ...
Background
Drug discovery and development has been aided by high throughput screening methods that detect compound effects on a single target. However, when using focused initial screening, undesirable secondary effects are often detected late in the development process after significant investment has been made. An alternative approach would be to screen against undesired effects early in the process, but the number of possible secondary targets makes this prohibitively expensive.
Results
This paper describes methods for making this global approach practical by constructing predictive models for many target responses to many compounds and using them to guide experimentation. We demonstrate for the first time that by jointly modeling targets and compounds using descriptive features and using active machine learning methods, accurate models can be built by doing only a small fraction of possible experiments. The methods were evaluated by computational experiments using a dataset of 177 assays and 20,000 compounds constructed from the PubChem database.
Conclusions
An average of nearly 60% of all hits in the dataset were found after exploring only 3% of the experimental space which suggests that active learning can be used to enable more complete characterization of compound effects than otherwise affordable. The methods described are also likely to find widespread application outside drug discovery, such as for characterizing the effects of a large number of compounds or inhibitory RNAs on a large number of cell or tissue phenotypes.
... Over the past decade it has been increasingly recognized that many pharmaceutically relevant compounds are promiscuous in nature [1–3] and that many drugs elicit their therapeutic effects -and undesired side effects- through polypharmacology [4, 5]. For a number of drugs that were originally considered to be target-selective or -specific, high degrees of promiscuity and ensuing polypharmacology have been shown to be responsible for their efficacy, with protein kinase inhibitors applied in oncology being a prime example [6]. ...
Compound promiscuity is rationalized as the specific interaction of a small molecule with multiple biological targets (as opposed to non-specific binding events) and represents the molecular basis of polypharmacology, an emerging theme in drug discovery and chemical biology. This concise review focuses on recent studies that have provided a detailed picture of the degree of promiscuity among different categories of small molecules. In addition, an exemplary computational approach is discussed that is designed to navigate multi-target activity spaces populated with various compounds.
... This problem is manifest when trying to determine the effects of potential drugs on complex systems, since drugs with desired effects often have undesired side effects. It has been argued that these constitute the greatest component of risk in drug development since unforeseen deleterious behaviors are costly to correct [4,5]. The only way to be sure that a drug does not have side effects is to measure its effect in assays for all potential targets. ...
High throughput and high content screening involve determination of the effect of many compounds on a given target. As currently practiced, screening for each new target typically makes little use of information from screens of prior targets. Further, choices of compounds to advance to drug development are made without significant screening against off-target effects. The overall drug development process could be made more effective, as well as less expensive and time consuming, if potential effects of all compounds on all possible targets could be considered, yet the cost of such full experimentation would be prohibitive. In this paper, we describe a potential solution: probabilistic models that can be used to predict results for unmeasured combinations, and active learning algorithms for efficiently selecting which experiments to perform in order to build those models and determining when to stop. Using simulated and experimental data, we show that our approaches can produce powerful predictive models without exhaustive experimentation and can learn them much faster than by selecting experiments at random.
... Drug Discov. (2014) 9 (2) mechanism of action at the molecular level [36]. Examples of lead compounds identified using in vivo HTS are reviewed elsewhere [34]. ...
Introduction:
The concept of chemical space has broad applications in drug discovery. In response to the needs of drug discovery campaigns, different approaches are followed to efficiently populate, mine and select relevant chemical spaces that overlap with biologically relevant chemical spaces.
Areas covered:
This paper reviews major trends in current drug discovery and their impact on the mining and population of chemical space. We also survey different approaches to develop screening libraries with confined chemical spaces balancing physicochemical properties. In this context, the confinement is guided by criteria that can be divided in two broad categories: i) library design focused on a relevant therapeutic target or disease and ii) library design focused on the chemistry or a desired molecular function.
Expert opinion:
The design and development of chemical libraries should be associated with the specific purpose of the library and the project goals. The high complexity of drug discovery and the inherent imperfection of individual experimental and computational technologies prompt the integration of complementary library design and screening approaches to expedite the identification of new and better drugs. Library design approaches including diversity-oriented synthesis, biological-oriented synthesis or combinatorial library design, to name a few, and the design of focused libraries driven by target/disease, chemical structure or molecular function are more efficient if they are guided by multi-parameter optimization. In this context, consideration of pharmaceutically relevant properties is essential for balancing novelty with chemical space in drug discovery.
... It is important to emphasize that there are different aspects related to polypharmacology that can lead to desirable or undesirable effects. 22 Moreover, in many instances, polypharmacology can be conceived of as a "normal" scenario, for example, in odor perception. 23 Knowledge of polypharmacology can also be used as a means not only for repurposing drugs but also in the repositioning of chemical compounds initially designed by man, or found to occur in nature, for other purposes. ...
The explosion in the generation of information parallels the explosion of computational resources. The use of computers to collect, store and manipulate chemical information is at the heart of chemoinformatics. These methodologies, whose main target thus far has been the pharmaceutical field, are general and can be applied to other types of chemical data sets, such as those containing food chemicals. While the use of chemical information methodologies to address food-related challenges is still in its infancy, interest is growing and will continue to do so as the methods prove useful, particularly for providing practical solutions to food industry challenges. Foodinformatics gives an overview of basic concepts, applications, tools and perspectives of the emerging field of foodinformatics. The book is an important addition to the literature and will be of interest of food chemists, nutritionists, informaticians and scientists of related fields. © 2014 Springer International Publishing Switzerland. All rights are reserved.
... During the last decade, the notion that drugs often bind to more than one drug target has elbowed its way into the scientific community, ushering "out" the paradigm of one drug -one target from drug discovery, and "in" the premise of polypharmacology [34][35][36][37][38]. ...
Poly(ADP-ribose)polymerases (PARPs) catalyze a post-transcriptional modification of proteins, consisting in the attachment of mono, oligo or poly ADP-ribose units from NAD+ to specific polar residues of target proteins. The scientific interest in members of this superfamily of enzymes is continuously growing since they have been implicated in a range of diseases including stroke, cardiac ischemia, cancer, inflammation and diabetes. Despite some inhibitors of PARP-1, the founder member of the superfamily, have advanced in clinical trials for cancer therapy, and other members of PARPs have recently been proposed as interesting drug targets, challenges exist in understanding the polypharmacology of current PARP inhibitors as well as developing highly selective chemical tools to unravel specific functions of each member of the superfamily. Beginning with an overview on the molecular aspects that affect polypharmacology, in this article we discuss how these may have an impact on PARP research and drug discovery. Then, we review the most selective PARP inhibitors hitherto reported in literature, giving an update on the molecular aspects at the basis of selective PARP inhibitor design. Finally, some outlooks on current issues and future directions in this field of research are also provided.
... Lots of launched drugs are halogenated compounds, and the halogen atoms are now intentionally introduced in new bioactive entities, owing to the ubiquitous nature of halogens, such as increasing the binding affinity and membrane permeability, facilitating the blood-brain barrier crossing and prolonging the lifetime of the drug, and so on [12,[24][25][26]. Dobes et al. reported that semiempirical quantum mechanical method PM6-DH2X described the geometry and energetics of CK2-inhibitor complexes involving halogen bonds well, while the Amber empirical potentials failed [27]. ...
Halogen bonding, a non-covalent interaction between the halogen σ-hole and Lewis bases, could not be properly characterized by majority of current scoring functions. In this study, a knowledge-based halogen bonding scoring function, termed XBPMF, was developed by an iterative method for predicting protein-ligand interactions. Three sets of pairwise potentials were derived from two training sets of protein-ligand complexes from the Protein Data Bank. It was found that two-dimensional pairwise potentials could characterize appropriately the distance and angle profiles of halogen bonding, which is superior to one-dimensional pairwise potentials. With comparison to six widely used scoring functions, XBPMF was evaluated to have moderate power for predicting protein-ligand interactions in terms of "docking power", "ranking power" and "scoring power". Especially, it has a rather satisfactory performance for the systems with typical halogen bonds. To the best of our knowledge, XBPMF is the first halogen bonding scoring function that is not dependent on any dummy atom, and is practical for high-throughput virtual screening. Therefore, this scoring function should be useful for the study and application of halogen bonding interactions like molecular docking and lead optimization.
... Accordingly, methods for rapid selection of compounds which are structurally unrelated to AMP and can directly interact with specific binding site of AMPK are urged. However, novel drug development usually takes a decade or even longer before marketing approval can be granted (Merino et al., 2010). Such demanding process generally consume more than $2 billion USD for investment (DiMasi et al., 2016) which account for most of the failing cases of new drug development (Hernandes et al., 2010). ...
Adenosine 5′-monophsphate-activated protein kinase (AMPK) is a crucial energy sensor for maintaining cellular homeostasis. Targeting AMPK may provide an alternative approach in treatment of various diseases like cancer, diabetes, and neurodegenerations. Accordingly, novel AMPK activators are frequently identified from natural products in recent years. However, most of such AMPK activators are interacting with AMPK in an indirect manner, which may cause off-target effects. Therefore, the search of novel direct AMPK modulators is inevitable and effective screening methods are needed. In this report, a rapid and straightforward method combining the use of in silico and in vitro techniques was established for selecting and categorizing huge amount of compounds from chemical library for targeting AMPK modulators. A new class of direct AMPK modulator have been discovered which are anilides or anilide-like compounds. In total 1,360,000 compounds were virtually screened and 17 compounds were selected after biological assays. Lipinski’s rule of five assessment suggested that, 13 out of the 17 compounds are demonstrating optimal bioavailability. Proton acceptors constituting the structure of these compounds and hydrogen bonds with AMPK in the binding site appeared to be the important factors determining the efficacy of these compounds.
... A single compound with MTDL or multifunctional activity would circumvent the limitations associated with combining two or more drugs with potentially different degrees of bioavailability, pharmacokinetics, and metabolism [7] as well as simplify the dosing regimen (an advantage that could be especially important from a compliance perspective). Drug repurposing (also called drug repositioning), is another (increasingly popular) approach designed to increase the speed of the drug discovery process by identifying a new clinical use for an existing approved drug [6,8,9]. There are numerous cases in the past where clinical observations have led to the eventual use of drugs for purposes other than what they were originally approved for. ...
... Statistical quantitative structure-activity relationship (QSAR) models, proteome-wide molecular docking, and chemoinformatics are needed to predict potential adverse off-target non-homologous binding. Merlot (2010), Xie et al. (2011), Merino et al. (2010), and Schmidt et al. (2014 have extensively reviewed the existing methods for predicting off-target binding. We also plan to focus on improving the quality and expanding the scope of systems models, which are one of the key bottlenecks to all in silico approaches to drug discovery. ...
... The enormous costs and time investments required to develop new drugs (i.e., from the discovery phase to clinical approval) combined with frequent failures in clinical trials has resulted in several large pharmaceutical companies abandoning or severely restricting their research and development programs for neuropsychiatric disorders (Kaitin and DiMasi 2011;Riordan and Cutler 2012). This dilemma has led to an increased interest in alternative approaches to the drug discovery process such as drug repurposing or repositioning (i.e., identifying new clinical uses for existing approved drugs, see Ashburn and Thor 2004;Merino et al. 2010;Medina-Franco et al. 2013). Drug repurposing can potentially reduce the time, costs, and safety risks associated with clinical approval for a new indication, since most repurposed candidates have already been assessed in phase I or II clinical trials for their original indications (Ashburn and Thor 2004). ...
Rationale
Due to the rising costs of drug development especially in the field of neuropsychiatry, there is increasing interest in efforts to identify new clinical uses for existing approved drugs (i.e., drug repurposing).
Objectives
The purpose of this work was to evaluate in animals the smoking cessation agent, varenicline, a partial agonist at α4β2 and full agonist at α7 nicotinic acetylcholine receptors, for its potential as a repurposed drug for disorders of cognition.
Methods
Oral doses of varenicline ranging from 0.01 to 0.3 mg/kg were evaluated in aged and middle-aged monkeys for effects on the following: working/short-term memory in a delayed match to sample (DMTS) task, distractibility in a distractor version of the DMTS (DMTS-D), and cognitive flexibility in a ketamine-impaired reversal learning task.
Results
In dose-effect studies in the DMTS and DMTS-D tasks, varenicline was not associated with statistically significant effects on performance. However, individualized “optimal doses” were effective when repeated on a separate occasion (i.e., improving DMTS accuracy at long delays and DMTS-D accuracy at short delays by approximately 13.6 and 19.6 percentage points above baseline, respectively). In reversal learning studies, ketamine impaired accuracy and increased perseverative responding, effects that were attenuated by all three doses of varenicline that were evaluated.
Conclusions
While the effects of varenicline across the different behavioral tasks were modest, these data suggest that varenicline may have potential as a repurposed drug for disorders of cognition associated with aging (e.g., Alzheimer’s disease), as well as those not necessarily associated with advanced age (e.g., schizophrenia).
... The use of chemically similar but target-inactive (or less active) "control" compounds is very helpful here. The shape of the cell viability curve can also reveal off-target effects, as excessively steep or shallow slopes have been associated with polypharmacology (Anighoro, Bajorath, & Rastelli, 2014;Merino, Bronowska, Jackson, & Cahill, 2010), population response heterogeneity or nonspecific toxicity (Fallahi-Sichani, Honarnejad, Heiser, Gray, & Sorger, 2013). The third scenario is one in which the cell death occurs at much higher concentrations than target inhibition (Fig. 5C), and most often this kind of response is caused by nonspecific compound toxicity at high concentrations. ...
Chemical biology approaches are a powerful means to functionally characterize epigenetic regulators such as histone modifying enzymes. We outline experimental protocols and best practices for the cellular characterization and use of “chemical probes” that selectively inhibit protein methyltransferases, many of which methylate histones to regulate heritable gene expression patterns. We describe biomarker assays to validate the probes in specific cellular systems, and provide guidelines for their use in functional characterization of methyltransferases including detailed protocols, examples, and controls. Together these techniques enable precision manipulation of cellular epigenomes and the exploration of the therapeutic potential of epigenetic targets in human disease.
... Different levels of polypharmacology could lead to better efficacy and/or safety or adverse effects. These differences in effectiveness may be due to the interaction of a compound with different therapeutic and/or nontherapeutic targets and/or dose dependence [70]. As mentioned earlier, polypharmacologybased drug repurposing refers to the reinvestigation of existing drugs for novel therapeutic purpose [13][14][15][16][17]. ...
Better the drugs you know than the drugs you do not know. Drug repurposing is a promising, fast, and cost effective method that can overcome traditional de novo drug discovery and development challenges of targeting neuropsychiatric and other disorders. Drug discovery and development targeting neuropsychiatric disorders are complicated because of the limitations in understanding pathophysiological phenomena. In addition, traditional de novo drug discovery and development are risky, expensive, and time-consuming processes. One alternative approach, drug repurposing, has emerged taking advantage of off-target effects of the existing drugs. In order to identify new opportunities for the existing drugs, it is essential for us to understand the mechanisms of action of drugs, both biologically and pharmacologically. By doing this, drug repurposing would be a more effective method to develop drugs against neuropsychiatric and other disorders. Here, we review the difficulties in drug discovery and development in neuropsychiatric disorders and the extent and perspectives of drug repurposing.
... INTRODUCTION Polypharmacology, conceptualized by the compound's activities against multiple targets, has emerged as a new paradigm in drug discovery and design. 1,2 In contrast to traditional concept of targeting single therapeutic target with high selectivity, designing therapeutic agents that are able to bind multiple drug targets enables controls over therapeutic efficacy, prevention of drug resistance, and reduction of therapeutic-target-related adverse effects. 3,4 Many multi-target drugs like multi-kinase inhibitors are used in the treatment of complex diseases such as cancer and psychiatric diseases and acquired unprecedented efficacy. ...
PharmMapper is a web server for drug targets identification by reversed pharmacophore matching the query compound against an annotated pharmacophore model database, which provides a computational polypharmacology prediction approach for drug repurposing and side effects risk evaluation. But due to the inherent non-discriminative feature of the simple fit scores used for prediction results ranking, the signal/noise ratio of the prediction results is high, posing a challenge for predictive reliability. In this paper, we improved the predictive accuracy of PharmMapper by generating a ligand-target pairwise fit scores matrix from profiling all the annotated pharmacophore models against corresponding ligands in the original complex structures that were used to extract these pharmacophore models. The matrix reflects the noise baseline of fit scores distribution of the background database, thus enables estimating the probability of finding a given target by random with the calculated ligand-pharmacophore fit score. Two retrospective tests were performed and confirmed the probability-based ranking score outperformed the simple fit score in terms of identification of both known drug targets and adverse drug reactions (ADRs) related off-targets.
... The previous discussion explains why multi-target drugs are being pursued today and also sets the logic ground for drug repositioning. However, it should be underlined that selectively non-selective drugs and promiscuous drugs are not exactly equivalent concepts in a drug repositioning scenario: while a certain, convenient degree of promiscuity may be desirable, an excess would certainly represent serious safety issues [67, 68]. Network-based approaches focused on drug repositioning may prove helpful to select candidates with an adequate degree of polypharmacology depending on the pursued new indication. ...
Drug repurposing/reprofiling has attracted considerable attention during the last decade. The object of such approach is to discover second or further medical uses of known chemicals, i. e. targeting existing, withdrawn or abandoned drugs, or yet to be pursued clinical candidates to new disease areas. Recently (2011-2012), the US and UK governments launched public-private joint initiatives towards finding new uses of previously shelved compounds (drug rescue). While in the past repurposing emerged from serendipitous findings and/or from rational exploitation of drug side-effects (. e.g. sildenafil, aspirin), the current tendency in the drug development field focuses on knowledge-based drug repurposing, particularly, computer-aided repositioning approaches. The present chapter reviews different cheminformatic and bioinformatic applications, as well as high-throughput literature analysis, oriented to the discovery of new medical uses of known drugs. Applications of such strategies to the discovery of innovative medications for neglected or rare diseases are discussed. Finally, we also review publicly available resources (. e.g. chemical libraries) valuable for reprofiling. © 2015 Bentham Science Publisher Ltd. Published by Elsevier Inc. All rights reserved.
... Such insights depart from the single-target-specificity paradigm that has long governed drug discovery strategies. Multitarget drug activities give rise to polypharmacological drug behavior, which is thought to be of critical relevance for therapeutic effects of many (but not all) drugs (5,6). Current estimates are that a drug might on average interact with about six different targets (7). ...
Given the increasing notion of target promiscuity of bioactive compounds and polypharmacological drug behavior, a detailed analysis of publicly available compound activity data from medicinal chemistry sources was carried out to determine and quantify the degree of promiscuity of active compounds across all known human target families. The results are surprising. Approximately 62% of currently available compounds with high-confidence activity data are only annotated with a single biological target, whereas 36% are known to act against multiple targets within the same family (i.e., closely related targets). However, only ∼2% of bioactive compounds are promiscuous across different target families. Thus, despite general data sparseness, these findings indicate that highly promiscuous bioactive compounds only rarely occur. Because pharmaceutically relevant active compounds represent the pool from which drug candidates emerge, one might extrapolate from these results and conclude that there is a low statistical probability to obtain drugs that act against multiple targets belonging to distinct families.
... Small molecule drugs are rarely selective enough to interact solely with their designated targets. Known drugs have, on average, six molecular targets on which they exhibit activity [1,2], usually resulting in unexpected side effects or toxicity [3][4][5][6]. On the other hand, the ability of small molecules to interact with multiple proteins also provides the basis to develop multitarget drugs [7,8]. ...
Target fishing often relies on the use of reverse docking to identify potential target proteins of ligands from protein database. The limitation of reverse docking is the accuracy of current scoring funtions used to distinguish true target from non-target proteins. Many contemporary scoring functions are designed for the virtual screening of small molecules without special optimization for reverse docking, which would be easily influenced by the properties of protein pockets, resulting in scoring bias to the proteins with certain properties. This bias would cause lots of false positives in reverse docking, interferring the identification of true targets. In this paper, we have conducted a large-scale reverse docking (5000 molecules to 100 proteins) to study the scoring bias in reverse docking by DOCK, Glide, and AutoDock Vina. And we found that there were actually some frequency hits, namely interference proteins in all three docking procedures. After analyzing the differences of pocket properties between these interference proteins and the others, we speculated that the interference proteins have larger contact area (related to the size and shape of protein pockets) with ligands (for all three docking programs) or higher hydrophobicity (for Glide), which could be the causes of scoring bias. Then we applied the score normalization method to eliminate this scoring bias, which was effective to make docking score more balanced between different proteins in the reverse docking of benchmark dataset. Later, the Astex Diver Set was utilized to validate the effect of score normalization on actual cases of reverse docking, showing that the accuracy of target prediction significantly increased by 21.5% in the reverse docking by Glide after score normalization, though there was no obvious change in the reverse docking by DOCK and AutoDock Vina. Our results demonstrate the effectiveness of score normalization to eliminate the scoring bias and improve the accuracy of target prediction in reverse docking. Moreover, the properties of protein pockets causing scoring bias to certain proteins we found here can provide the theory basis to further optimize the scoring functions of docking programs for future research.
... Over 50% of drugs make an interaction with at least five proteins, and only 15% are reported to target only one protein [2]. Spotting of offtargets in early phase of drug discovery process is important to decline the failure rate in clinical trials and for speeding the process [4,5]. Several protein kinases are key drug targets in oncology as they play a vital and central roles in signal transduction cascades [6][7][8] and cellular processes such as cell division, metabolism, survival and apoptosis. ...
Potential drug target identification and mechanism of action is an important step in drug discovery process, which can be achieved by biochemical methods, genetic interactions or computational conjectures. Sometimes more than one approach is implemented to mine out the potential drug target and characterize the on-target or off-target effects. A novel anticancer agent RH1 is designed as pro-drug to be activated by NQO1, an enzyme overexpressed in many types of tumors. However, increasing data show that RH1 can affect cells in NQO1-independent fashion. Here, we implemented the bioinformatics approach of modeling and molecular docking for search of RH1 targets among protein kinase species. We have examined 129 protein kinases in total where 96 protein kinases are in complexes with their inhibitor, 11 kinases were in the unbound state with any ligand and for 22 protein kinases 3D structure were modeled. Comparison of calculated free energy of binding of RH1 with indigenous kinase inhibitors binding efficiency as well as alignment of their pharmacophoric maps let us predict and ranked protein kinases such as KIT, CDK2, CDK6, MAPK1, NEK2 and others as the most prominent off-targets of RH1. Our finding opens new avenues in search of protein targets that might be responsible for curing cancer by new promising drug RH1 in NQO1-independent way.
... morphine, cocaine, penicillin, taxols…) and are also good lead compounds suitable for further modification during drug development. Introducing a new compound on the market is time consuming and cost-intensive process [6,7], in particular for natural products, so that strategies allowing time saving are welcomed. ...
Bacteria and fungi use a set of enzymes called nonribosomal peptide synthetases to provide a wide range of natural peptides displaying structural and biological diversity. So, nonribosomal peptides (NRPs) are the basis for some efficient drugs. While discovering new NRPs is very desirable, the process of identifying their biological activity to be used as drugs is a challenge. In this paper, we present a novel peptide fingerprint based on monomer composition (MCFP) of NRPs. MCFP is a novel method for obtaining a representative description of NRP structures from their monomer composition in fingerprint form. Experiments with Norine NRPs database and MCFP show high prediction accuracy (>93 %). Also a high recall rate (>82 %) is obtained when MCFP is used for screening NRPs database. From this study it appears that our fingerprint, built from monomer composition, allows an effective screening and prediction of biological activities of NRPs database.
Figure optionsView in workspaceDownload full-size imageDownload as PowerPoint slide
The study of compound promiscuity is a hot topic in medicinal chemistry and drug discovery research. Promiscuous compounds are increasingly identified, but the molecular basis of promiscuity is currently only little understood. Utilizing the matched molecular pair formalism, we have analyzed patterns of compound promiscuity in a publicly available small molecule microarray data set. On the basis of our analysis, we introduce "promiscuity cliffs" as pairs of structural analogs with single-site substitutions that lead to large-magnitude differences in apparent compound promiscuity involving between 50 and 97 unrelated targets. No substructures or substructure transformations have been detected that are generally responsible for introducing promiscuity. However, within a given structural context, small chemical replacements were found to lead to dramatic promiscuity effects. On the basis of our analysis, promiscuity is not an inherent feature of molecular scaffolds but can be induced by small chemical substitutions. Promiscuity cliffs provide immediate access to such modifications.
Bioactive food compounds can be both therapeutically and nutritionally relevant. As in pharmaceutical settings, screening strategies are employed to identify bioactive compounds from edible plants. Flavor additives contained in the so-called FEMA GRAS (Generally Recognized as Safe) list of approved flavoring ingredients is an additional source of potentially bioactive compounds. In this work, we used the principles of molecular similarity to identify compounds with potential mood modulating properties. The ability of certain GRAS molecules to inhibit histone deacetylase-1 (HDAC1), proposed as an important player in mood modulation, was assayed. Two GRAS chemicals were identified as HDAC1 inhibitors in the micromolar range, results similar to what was observed for the structurally related mood prescription drug, valproic acid. Additional studies on bioavailability, toxicity at higher concentrations and off-target effects are warranted. The methodology described in this work could be employed to identify potentially bioactive flavor chemicals present in the FEMA GRAS list.
A large number of research articles are describing novel methodologies of docking and/or scoring methods. An even larger number of publications report the successful use of these methods in the identification of novel hit molecules. What is less documented is the application of docking methods in other areas. We review herein the application of docking methods to not only hit identification but also to de novo drug design, fragment-based drug discovery, lead optimization, metabolism prediction, off-target binding, selectivity, protein structure prediction and drug-drug interaction.
A novel approach in molecular docking was successfully used to reproduce protein-ligand experimental geometries. When dealing with halogenated compounds the correct description of halogen bonds between the ligand and the protein is shown to be essential. Applying a simple molecular mechanistic model for halogen bonds improved the protein-ligand geometries as well as halogen bond features, which makes it a promising tool for future computer-aided drug development.
Label-free is a broad term used to describe a number of cutting-edge biosensor technologies that have attracted considerable attention in the area of drug discovery for seven-transmembrane G protein-coupled receptors (GPCRs). Label-free biosensors resolve receptor-mediated responses noninvasively in real time and living cells and do so with high textural information and broad signaling-pathway coverage. They should facilitate studies of the receptor's integrated signal transduction biology intractable to classical assays with single pathway focus. Label-free occupies a privilege niche with respect to mechanistic studies in human native cells-healthy or disease-relevant-and the probing of context-dependent pharmacology in relation to whole biological system efficacy. It is expected that implementation of label-free approaches into the drug discovery process will improve clinical predictability of drug candidates at early stages of discovery research by their exquisite capability to sense whole cellular responses akin to tissue bioassays. Here, we present an overview of promises and challenges this rapidly evolving technology offers to drug screening and we also discuss the prospect of advancing drug discovery.
High-throughput compound screening is time and resource consuming, and considerable effort is invested into screening compound libraries, profiling, and selecting the most promising candidates for further testing. Active-learning methods assist the selection process by focusing on areas of chemical space that have the greatest chance of success while considering structural novelty. The core feature of these algorithms is their ability to adapt the structure-activity landscapes through feedback. Instead of full-deck screening, only focused subsets of compounds are tested, and the experimental readout is used to refine molecule selection for subsequent screening cycles. Once implemented, these techniques have the potential to reduce costs and save precious materials. Here, we provide a comprehensive overview of the various computational active-learning approaches and outline their potential for drug discovery.
Copyright © 2014. Published by Elsevier Ltd.
Current trends in computational de novo design provide a fresh approach to ‘scaffold-hopping’ in drug discovery. The methodological repertoire is no longer limited to receptor-based methods, but specifically ligand-based techniques that consider multiple properties in parallel, including the synthetic feasibility of the computer-generated molecules and their polypharmacology, provide innovative ideas for the discovery of new chemical entities. The concept of fragment-based and virtual reaction-driven design enables rapid compound optimization from scratch with a manageable complexity of the search. Starting from known drugs as a reference, such algorithms suggest drug-like molecules with motivated scaffold variations, and advanced mathematical models of structure-activity landscapes and multi-objective design techniques have created new opportunities for hit and lead finding.
Preclinical ResearchChemoinformatic approaches have an essential role in the systematic description and visualization of the chemical space for drug discovery projects. These methods enable the quantitative comparison of general screening collections and the systematic classification of approved drugs and databases annotated with biological activity to define biologically and medicinally relevant chemical spaces. Profiling of chemical diversity, molecular complexity, and physicochemical properties of compound libraries using chemoinformatic approaches provide a solid basis to generate hypothesis of how to interrogate novel areas of chemical space for enhanced drug discovery. This commentary is focused on the application of chemoinformatic approaches to mine, and to navigate the chemical space of compound collections. The discussion is centered on the concept of chemical space, types of compound libraries used in drug discovery programs, applications of chemical space mining and visualization using chemoinformatic methods, and strategies to expand the pharmaceutical relevant chemical space with emphasis on the notion of molecular complexity.
The non-covalent halogen bonding could be attributed to the attraction between the positively charged σ-hole and a nucleophile. Quantum mechanics (QM) calculation indicated that the negatively charged organohalogens have no positively charged σ-hole on their molecular surface, leading to a postulation of repulsion between negatively charged organohalogens and nucleophile in vaccum. However, PDB survey revealed that 24% of the ligands with halogen bonding geometry could be negatively charged. Moreover, 36% of ionizable drugs in CMC (Comprehensive Medicinal Chemistry) are possibly negatively charged at pH 7.0. QM energy scan showed that the negatively charged halogen bonding is probably meta-stable in the vacuum. However, the QM calculated bonding energy turned negative in various solvents, suggesting that halogen bonding with negatively charged donors should be stable in reality. Indeed, QM/MM calculation on three crystal structures with negatively charged ligands revealed that the negatively charged halogen bonding was stable. Hence, we concluded that halogen bonding with negatively charged donors is unstable or meta-stable in the vacuum but stable in protein environment, and possesses similar geometric and energetic characteristics as conventional halogen bonding. Therefore, negatively charged organohalogens are still effective halogen bonding donors for medicinal chemistry and other applications.
Accurately predicting relative binding affinities and biological potencies for ligands that interact with proteins remains a significant challenge for computational chemists. Most evaluations of docking and scoring algorithms have focused on enhancing ligand affinity for a protein by optimizing docking poses and enrichment factors during virtual screening. However, there is still relatively limited information on the accuracy of commercially available docking and scoring software programs for correctly predicting binding affinities and biological activities of structurally related inhibitors of different enzyme classes. Presented here is a comparative evaluation of eight molecular docking programs (Autodock Vina, Fitted, FlexX, Fred, Glide, GOLD, LibDock, MolDock) using sixteen docking and scoring functions to predict the rank-order activity of different ligand series for six pharmacologically important protein and enzyme targets (Factor Xa, Cdk2 kinase, Aurora A kinase, COX-2, pla2g2a, β Estrogen receptor). Use of Fitted gave an excellent correlation (Pearson 0.86, Spearman 0.91) between predicted and experimental binding only for Cdk2 kinase inhibitors. FlexX and GOLDScore produced good correlations (Pearson > 0.6) for hydrophilic targets such as Factor Xa, Cdk2 kinase and Aurora A kinase. By contrast, pla2g2a and COX-2 emerged as difficult targets for scoring functions to predict ligand activities. Albeit possessing a high hydrophobicity in its binding site, β Estrogen receptor produced reasonable correlations using LibDock (Pearson 0.75, Spearman 0.68). These findings can assist medicinal chemists to better match scoring functions with ligand-target systems for hit-to-lead optimization using computer-aided drug design approaches.
Given the rapidly increasing amounts of compound activity data that become publicly available, structure–activity relationships (SARs) can be explored on a large scale. The concept of molecular scaffolds is widely applied to aid in the rationalization of SARs. Scaffolds are derived from bioactive compounds by removal of R groups and thus represent core structures of compound series. In computational compound screening, scaffold hopping is a popular exercise. This term refers to the detection of compounds with similar activity having different core structures. In our laboratory, we are systematically analyzing scaffolds contained in compounds from medicinal chemistry resources and are studying compound–scaffold–activity relationships from different points of view. For this purpose, a variety of data mining strategies have been designed to identify scaffolds with different SAR characteristics. For example, we have detected sets of scaffolds with apparent selectivity for communities of closely related targets or promiscuity across different target families. Moreover, global scaffold hopping potential has been assessed across current pharmaceutical targets. Exemplary studies originating from our laboratory are presented herein in their scientific context.
Pharmacological profiling comprises the estimation and prediction of a compound's pharmacokinetic and toxicological properties. This chapter focuses mainly on mechanism-based toxicity and investigation of adverse drug effects caused by off-target binding. Bioactivity profiling, no matter if in the sense of determining pharmacokinetic and toxicologic properties or exploring the target spectrum of a certain compound, is an important strategy that improves the drug discovery and development process in many ways. In vitro assays are much less costly and time-consuming than in vivo testing, and allow for general investigations as well as for specific proof of principles on isolated targets. Therefore, they are widely applied in research and development. Understanding the molecular mechanisms underlying the biological effects enables a rational approach and the in silico prediction of a compound's behavior in a complex biological system.
In the thesis of my "Habilitation à diriger des recherches", I'm presenting the pioneer work on computational biology for nonribosomal peptides (NRPs). Those researches began in 2006 in Lille and lead to the unique plate-form dedicated to computational biology analysis of NRPs called Norine, of which I am a founder member. Nonribosomal peptides are small molecules produced by micro-organisms, bacteria and fungi, to colonize their environment. Those peptides have the advantage of presenting a large range of structures. They can be linear, but also can contain cycles and/or branches, and consist of more than 500 different building blocks. This variety comes from their synthesis done by huge enzymatic complexes, called nonribosomal peptide synthetases (NRPSs). They select amino acids and other compounds, called monomers, and then link them by peptide and other bonds. So, nonribosomal peptides cover a large range of activities such as antibiotic, anti-tumor or immuno-suppressor. Some, as penicillin, are commonly used as drugs. In the first part, I present the synthetases by associating the peptides' characteristics to the enzymatic functions required to accomplish them. Then, I describe the main steps needed to design a tool that analyse the NRPS protein sequences by specifying the characteristics of the existing tools. Then, I present my own contribution to the investigation of the production of NRPs from genomic or protein sequences by participating to the design of bioinformatics protocols and to genome annotation. In the second part, I start by specifying the contributions of the Norine resource to the knowledge of the nonribosomal peptides diversity, supplemented by a study of the chemistry of those molecules. Next, I present the few databases and tools related to those peptides that are developed elsewhere. Then, I describe my own contribution to the Norine resource and suggest modernisation of the process to collect data and expansion of the query functionalities for structure search. I finish by suggesting new prospects: cheminformatics dedicated to nonribosomal peptides with the goal to predict one or several synthetases able to produce a peptide having a given activity.
In recent years, most new candidate antiparasitic drugs have been found by screening huge numbers of compounds for their ability to kill parasites, followed by counterscreening for toxicity to mammalian cells. Several public–private initiatives have supported this, yielding many hits each for Plasmodia and Kinetoplastids. From these, candidates are selected for further investigation. Although knowledge of the precise mode of action is not necessary for successful development, detailed understanding of the drug's uptake, activation, and target can be very useful in guiding medicinal chemistry, toxicology, and pharmacology. Knowledge of the target can also provide information for further drug discovery studies and in choosing partner drugs in combinations. A multiplicity of complementary approaches can be applied to investigate the drug mode of action. Examples include selecting drug-resistant parasites and identifying the resistance-causing mutations, reverse genetics to find genes required for drug susceptibility, metabolomics, and biochemical approaches such as affinity purification. Here, we review the myriad possibilities, including numerous examples.
This chapter presents a general introduction to polypharmacology in the context of drug discovery and development. The relationship between polypharmacology and other major concepts such as polypharmacy, drug repurposing, combination of drugs, and in vivo testing is discussed. Modern applications of polypharmacology and polypharmacy in modern epigenetic and antiviral drug development are described as representative examples. An introductory overview of modern methodologies to design and develop multitarget ligands is presented with a special focus on computational-based methods. These approaches include, but are not limited to, target fishing, proteochemometric modeling, data mining of side effects of drugs, and computer-aided drug repurposing.
The process of developing an old drug for new indications is now a widely accepted strategy of shortening drug development time, reducing drug costs, and improving drug availability, especially for rare and neglected diseases. In this mini-review, we highlighted the impact of drug delivery systems in the fulfillment of crucial aspects of drug repurposing such as (i) maximizing the repurposed drug effects on a new target, (ii) minimizing off-target effects, (iii) modulating the release profiles of drug at the site of absorption, (iv) modulating the pharmacokinetics/in vivo biodistribution of the repurposed drug, (v) targeting/modulating drug retention at the sites of action, and (vi) providing a suitable platform for therapeutic application of combination drugs.
We discovered a constitutively activating mutation (CAM) V308E for the neurotensin NT1 receptor. Molecular dynamics (MD) performed for the CAM NT1-V308E exhibiting a high spontaneous activity, and for the wild-type NT1 without basal activity, show dramatic conformational changes for the CAM. To test if the two MD models could be valuable active and inactive templates for building molecular models for other class-A GPCR, supposed active and inactive models were built by homology for the cholecystokinin CCK1 receptor. Virtual screening of a corporate library with 250 000 compounds was performed with the two CCK1 models, and a differential virtual screening analysis (DVS), led us to isolate 250 predicted agonists and 250 predicted antagonists. The two sets were merged and the compounds were tested in CCK1 agonist and antagonist cellular assays. An excellent correlation was obtained between predictions and biological results. The effective profiling provided by DVS with active and inactive molecular models, opens new perspectives for finding agonists and antagonists for other class-A GPCR, notably for orphan GPCRs for which no ligands are known.
We present a strategy for identifying off-target effects and hidden phenotypes of drugs by directly probing biochemical pathways that underlie therapeutic or toxic mechanisms in intact, living cells. High-content protein-fragment complementation assays (PCAs) were constructed with synthetic fragments of a mutant fluorescent protein ('Venus', EYFP or both), allowing us to measure spatial and temporal changes in protein complexes in response to drugs that activate or inhibit particular pathways. One hundred and seven different drugs from six therapeutic areas were screened against 49 different PCA reporters for ten cellular processes. This strategy reproduced known structure-function relationships and also predicted 'hidden,' potent antiproliferative activities for four drugs with novel mechanisms of action, including disruption of mitochondrial membrane potential. A simple algorithm identified a 25-assay panel that was highly predictive of antiproliferative activity, and the predictive power of this approach was confirmed with cross-validation tests. This study suggests a strategy for therapeutic discovery that identifies novel, unpredicted mechanisms of drug action and thereby enhances the productivity of drug-discovery research.
Although drugs are intended to be selective, at least some bind to several physiological targets, explaining side effects and efficacy. Because many drug-target combinations exist, it would be useful to explore possible interactions computationally. Here we compared 3,665 US Food and Drug Administration (FDA)-approved and investigational drugs against hundreds of targets, defining each target by its ligands. Chemical similarities between drugs and ligand sets predicted thousands of unanticipated associations. Thirty were tested experimentally, including the antagonism of the beta(1) receptor by the transporter inhibitor Prozac, the inhibition of the 5-hydroxytryptamine (5-HT) transporter by the ion channel drug Vadilex, and antagonism of the histamine H(4) receptor by the enzyme inhibitor Rescriptor. Overall, 23 new drug-target associations were confirmed, five of which were potent (<100 nM). The physiological relevance of one, the drug N,N-dimethyltryptamine (DMT) on serotonergic receptors, was confirmed in a knockout mouse. The chemical similarity approach is systematic and comprehensive, and may suggest side-effects and new indications for many drugs.
The anticonvulsant properties of VPA (valproic acid), a branched short-chain fatty acid, were serendipitously discovered in 1963. Since then, therapeutic roles of VPA have increased to include bipolar disorder and migraine prophylaxis, and have more recently been proposed in cancer, Alzheimer's disease and HIV treatment. These numerous therapeutic roles elevate VPA to near 'panacea' level. Surprisingly, the mechanisms of action of VPA in the treatment of many of these disorders remain unclear, although it has been shown to alter a wide variety of signalling pathways and a small number of direct targets. To analyse the mechanism of action of VPA, a number of studies have defined the structural characteristics of VPA-related compounds giving rise to distinct therapeutic and cellular effects, including adverse effects such as teratogenicity and hepatotoxicity. These studies raise the possibility of identifying target-specific novel compounds, providing better therapeutic action or reduced side effects. This short review will describe potential therapeutic pathways targeted by VPA, and highlight studies showing structural constraints necessary for these effects.
G protein-coupled receptors (GPCRs) represent the largest class of targets in drug discovery, one-third of all marketed drugs are active at GPCRs and drugs targeted at GPCRs are marketed in virtually every therapeutic area. GPCRs can be classified by virtue of their coupling to second messenger signaling systems. In the last decade functional evaluation of Galphaq-coupled GPCRs has been enabled by advances in fluorescence dye-based methodologies and detection instrumentation. Investigations into the bioluminescence of jelly fish in the early 1960s isolated the photoprotein aequorin that required only the addition of calcium to generate a luminescent signal. The recent development of sensitive detection platforms with integrated fluidics for liquid handling has revived interest in bioluminescence as an alternative to chemical fluorophore-based detection for characterizing the pharmacology of this target class. In this chapter we describe a detailed methodology for the development and execution of bioluminescence apoprotein aequorin-based screens for hit identification and structure-activity relationship compound profiling and highlight the opportunities and challenges associated with this technique.
The inadvertent activation of the Abelson tyrosine kinase (Abl) causes chronic myelogenous leukemia (CML). A small-molecule
inhibitor of Abl (STI-571) is effective in the treatment of CML. We report the crystal structure of the catalytic domain of
Abl, complexed to a variant of STI-571. Critical to the binding of STI-571 is the adoption by the kinase of an inactive conformation,
in which a centrally located “activation loop” is not phosphorylated. The conformation of this loop is distinct from that
in active protein kinases, as well as in the inactive form of the closely related Src kinases. These results suggest that
compounds that exploit the distinctive inactivation mechanisms of individual protein kinases can achieve both high affinity
and high specificity.
To ascertain the current burden of adverse drug reactions (ADRs) through a prospective analysis of all admissions to hospital.
Prospective observational study.
Two large general hospitals in Merseyside, England.
18 820 patients aged > 16 years admitted over six months and assessed for cause of admission.
Prevalence of admissions due to an ADR, length of stay, avoidability, and outcome.
There were 1225 admissions related to an ADR, giving a prevalence of 6.5%, with the ADR directly leading to the admission in 80% of cases. The median bed stay was eight days, accounting for 4% of the hospital bed capacity. The projected annual cost of such admissions to the NHS is 466m pounds sterling (706m Euros, 847m dollars). The overall fatality was 0.15%. Most reactions were either definitely or possibly avoidable. Drugs most commonly implicated in causing these admissions included low dose aspirin, diuretics, warfarin, and non-steroidal anti-inflammatory drugs other than aspirin, the most common reaction being gastrointestinal bleeding.
The burden of ADRs on the NHS is high, accounting for considerable morbidity, mortality, and extra costs. Although many of the implicated drugs have proved benefit, measures need to be put into place to reduce the burden of ADRs and thereby further improve the benefit:harm ratio of the drugs.
Genomic and post-genomic biological research has provided fine-grain insights into the molecular processes of life, but also
threatens to drown biomedical researchers in data. Moreover, as new high-throughput technologies are developed, the types
of data that are gathered en masse are diversifying. The need to collect, store and curate all this information in ways that allow its efficient retrieval and
exploitation is greater than ever. The European Bioinformatics Institute's (EBI's) databases and tools have evolved to meet
the changing needs of molecular biologists: since we last wrote about our services in the 2003 issue of Nucleic Acids Research, we have launched new databases covering protein–protein interactions (IntAct), pathways (Reactome) and small molecules (ChEBI).
Our existing core databases have continued to evolve to meet the changing needs of biomedical researchers, and we have developed
new data-access tools that help biologists to move intuitively through the different data types, thereby helping them to put
the parts together to understand biology at the systems level. The EBI's data resources are all available on our website at
http://www.ebi.ac.uk.
A major goal in contemporary drug design is to develop new ligands with high affinity of binding toward a given protein receptor. Pharmacophore, which is the three-dimensional arrangement of essential features that enable a molecule to exert a particular biological effect, is a very useful model for achieving this goal. If the three dimensional structure of the receptor is known, pharmacophore is a complementary tool to standard techniques, such as docking. However, frequently the structure of the receptor protein is unknown and only a set of ligands together with their measured binding affinities towards the receptor is available. In such a case, a pharmacophore based strategy is one of the few applicable tools. Here we present a broad, yet concise guide to pharmacophore identification and review a sample of applications for drug design. In particular, we present the framework of the algorithms, classify their modules and point out their advantages and challenges.
This paper replicates the drug development cost estimates of Joseph DiMasi and colleagues ("The Price of Innovation"), using their published cost estimates along with information on success rates and durations from a publicly available data set. For drugs entering human clinical trials for the first time between 1989 and 2002, the paper estimated the cost per new drug to be 868 million dollars. However, our estimates vary from around 500 million dollars to more than 2,000 million dollars, depending on the therapy or the developing firm.
Much of drug discovery today is predicated on the concept of selective targeting of particular bioactive macromolecules by low-molecular-mass drugs. The binding of drugs to their macromolecular targets is therefore seen as paramount for pharmacological activity. In vitro assessment of drug-target interactions is classically quantified in terms of binding parameters such as IC(50) or K(d). This article presents an alternative perspective on drug optimization in terms of drug-target binary complex residence time, as quantified by the dissociative half-life of the drug-target binary complex. We describe the potential advantages of long residence time in terms of duration of pharmacological effect and target selectivity.
Although cyclooxygenase (COX)-2 inhibitors (coxibs) are effective in controlling inflammation, pain, and tumorigenesis, their use is limited by the recent revelation of increased adverse cardiovascular events. The mechanistic basis of this side effect is not well understood. We show that the metabolism of endocannabinoids by the endothelial cell COX-2 coupled to the prostacyclin (PGI(2)) synthase (PGIS) activates the nuclear receptor peroxisomal proliferator-activated receptor (PPAR) delta, which negatively regulates the expression of tissue factor (TF), the primary initiator of blood coagulation. Coxibs suppress PPARdelta activity and induce TF expression in vascular endothelium and elevate circulating TF activity in vivo. Importantly, PPARdelta agonists suppress coxib-induced TF expression and decrease circulating TF activity. We provide evidence that COX-2-dependent attenuation of TF expression is abrogated by coxibs, which may explain the prothrombotic side-effects for this class of drugs. Furthermore, PPARdelta agonists may be used therapeutically to suppress coxib-induced cardiovascular side effects.
In addition to maintaining the GenBank(R) nucleic acid sequence database, the National Center for Biotechnology Information
(NCBI) provides analysis and retrieval resources for the data in GenBank and other biological data available through NCBI's
web site. NCBI resources include Entrez, the Entrez Programming Utilities, My NCBI, PubMed, PubMed Central, Entrez Gene, the
NCBI Taxonomy Browser, BLAST, BLAST Link, Electronic PCR, OrfFinder, Spidey, Splign, RefSeq, UniGene, HomoloGene, ProtEST,
dbMHC, dbSNP, Cancer Chromosomes, Entrez Genome, Genome Project and related tools, the Trace, Assembly, and Short Read Archives,
the Map Viewer, Model Maker, Evidence Viewer, Clusters of Orthologous Groups, Influenza Viral Resources, HIV-1/Human Protein
Interaction Database, Gene Expression Omnibus, Entrez Probe, GENSAT, Database of Genotype and Phenotype, Online Mendelian
Inheritance in Man, Online Mendelian Inheritance in Animals, the Molecular Modeling Database, the Conserved Domain Database,
the Conserved Domain Architecture Retrieval Tool and the PubChem suite of small molecule databases. Augmenting the web applications
are custom implementations of the BLAST program optimized to search specialized data sets. These resources can be accessed
through the NCBI home page at www.ncbi.nlm.nih.gov.
Drug repositioning is the process of developing new indications for existing drugs or biologics. Increasing interest in drug repositioning has occurred due to sustained high failure rates and costs involved in attempts to bring new drugs to market, It has been estimated that it may cost more than USD 800 million to develop a new drug de novo. In addition, due to regulatory requirements regarding safety, efficacy and quality, the time required to develop a new drug de novo has been estimated to be 10 to 17 years. De novo drug discovery has failed to efficiently supply pharmaceutical company pipelines. A rational approach to drug repositioning may include a cross-disciplinary focus on the elucidation of the mechanisms of disease, allowing matching of disease pathways with appropriately targeted therapeutic agents. Repurposed drugs or biologics have the advantage of decreased development costs and decreased time to Launch due to previously collected pharmacokinetic, toxicology and safety data. For these reasons, repurposing should be a primary strategy in drug discovery for every broadly focused, research-based pharmaceutical company. Copyright © 2009 Prous Science, S.A.U. or its licensors. All rights reserved.
Far from the traditional view of selective drug-target interactions, the recent accumulation of large amounts of interaction data for small-molecule drugs and protein targets requires innovative visualisation and analysis tools that are able to deal with what has become a truly complex system. In this context, network theory offers both a robust and illustrative framework to investigate drug-target connections and has been swiftly and widely embraced by the chemical biology and molecular informatics communities. A survey of the most recent applications of drug-target networks to detect cross-pharmacology relationships among targets and to identify new targets for known drugs is provided. Finally, some of the current limitations are also discussed, including the actual completeness of interaction data and the information loss intrinsically associated with the one-mode projection of drug-target networks.
Because of its construction and parametrization for more than 80 elements, the semiempirical quantum chemical PM6 method is superior to other similar methods. Despite its advantages, however, the PM6 method fails for the description of noncovalent interactions, specifically the dispersion energy and H-bonding. Upon inclusion of correction terms for dispersion and H-bonding, the performance of the method was found to be dramatically improved. The former correction included two parameters in the damping function that were parametrized to reproduce the benchmark interaction energies [CCSD(T)/complete basis set (CBS) limit] of the dispersion-bonded complexes from the S22 data set. The latter correction was parametrized on an extended set of H-bonded stabilization energies determined at the MP2/cc-pVTZ level. The resulting PM6-DH method was tested on the S22 data set, for which chemical accuracy (error < 1 kcal/mol) was achieved, and also on the JSCH2005 set, for which significant improvement over the original PM6 method was also obtained. Implementation of analytical gradients allows very efficient geometry optimization, which, for all complexes, provides better agreement with the benchmark data. Excellent results were also achieved for small peptides, and here again, chemical accuracy was obtained (i.e., the error with respect to CCSD(T)/CBS results was smaller than 1 kcal/mol). The performance of the technique was finally demonstrated on extended complexes, namely, the porphine dimer and various graphene models with DNA bases and base pairs, where the PM6-DH stabilization energies agree very well with available benchmark data obtained with DFT-D, SCS-MP2, and MP2.5 methods. The PM6-DH calculations are very efficient and can be routinely applied for systems of up to 1000 atoms. For nonaromatic systems, the use of a linear scaling version of the SCF procedure based on localized orbitals speeds up the method significantly and allows one to investigate systems with several thousand atoms. The method can thus replace force fields, which face basic problems for the description of quantum effects, in many applications.
Phosphodiesterase 4 (PDE4), the primary cAMP-hydrolyzing enzyme in cells, is a promising drug target for a wide range of conditions. Here we present seven co-crystal structures of PDE4 and bound inhibitors that show the regulatory domain closed across the active site, thereby revealing the structural basis of PDE4 regulation. This structural insight, together with supporting mutagenesis and kinetic studies, allowed us to design small-molecule allosteric modulators of PDE4D that do not completely inhibit enzymatic activity (I(max) approximately 80-90%). These allosteric modulators have reduced potential to cause emesis, a dose-limiting side effect of existing active site-directed PDE4 inhibitors, while maintaining biological activity in cellular and in vivo models. Our results may facilitate the design of CNS therapeutics modulating cAMP signaling for the treatment of Alzheimer's disease, Huntington's disease, schizophrenia and depression, where brain distribution is desired for therapeutic benefit.
Computational methods that reliably predict the biological activities of compounds have long been sought. The validation of one such method suggests that in silico predictions for drug discovery have come of age.
The study of protein-protein interactions that are involved in essential life processes can largely benefit from the recent upraising of computational docking approaches. Predicting the structure of a protein-protein complex from their separate components is still a highly challenging task, but the field is rapidly improving. Recent advances in sampling algorithms and rigid-body scoring functions allow to produce, at least for some cases, high quality docking models that are perfectly suitable for biological and functional annotations, as it has been shown in the CAPRI blind tests. However, important challenges still remain in docking prediction. For example, in cases with significant mobility, such as multidomain proteins, fully unrestricted rigid-body docking approaches are clearly insufficient so they need to be combined with restraints derived from domain-domain linker residues, evolutionary information, or binding site predictions. Other challenging cases are weak or transient interactions, such as those between proteins involved in electron transfer, where the existence of alternative bound orientations and encounter complexes complicates the binding energy landscape. Docking methods also struggle when using in silico structural models for the interacting subunits. Bringing these challenges to a practical point of view, we have studied here the limitations of our docking and energy-based scoring approach, and have analyzed different parameters to overcome the limitations and improve the docking performance. For that, we have used the standard benchmark and some practical cases from CAPRI. Based on these results, we have devised a protocol to estimate the success of a given docking run.
Drug promiscuity is one of the key issues in current drug development. Many famous drugs have turned out to behave unexpectedly due to their propensity to bind to multiple targets. One of the primary reasons for this promiscuity is that drugs bind to multiple distinctive target environments, a feature that we call multi-modal binding. Accordingly, investigations into whether multi-modal binding propensities can be predicted, and if so, whether the features determining this behavior can be found, would be an important advance. In this study, we have developed a structure-based classifier that predicts whether small molecules will bind to multiple distinct binding sites. The binding sites for all ligands in the Protein Data Bank (PDB) were clustered by binding site similarity, and the ligands that bind to many dissimilar binding sites were identified as multi-modal binding ligands. The mono-binding ligands were also collected, and the classifiers were built using various machine-learning algorithms. A 10-fold cross-validation procedure showed 70-85% accuracy depending on the choice of machine-learning algorithm, and the different definitions used to identify multi-modal binding ligands. In addition, a quantified importance measurement for global and local descriptors was also provided, which suggests that the local features are more likely to have an effect on multi-modal binding than the global ones. The interpretable global and local descriptors were also ranked by their importance. To test the classifier on real examples, several test sets including well-known promiscuous drugs were collected by a literature and database search. Despite the difficulty in constructing appropriate testable sets, the classifier showed reasonable results that were consistent with existing information on drug behavior. Finally, a test on natural enzyme substrates and artificial drugs suggests that the natural compounds tend to exhibit a broader range of multi-modal binding than the drugs.
To be an effective medicine a drug has to possess many attributes to ensure target potency and specificity, lack of toxicity, bioavailability and duration of action. Discovering a compound with these properties is invariably an evolutionary process. Fragment based drug discovery sets out to identify a starting compound by screening a library of small molecules representing fragments which cover the chemical space of drug like matter. Fragment based screening is increasingly used in the pharmaceutical industry in the early stages of lead identification and optimization. We will provide an introduction into this approach and discuss a number of examples which show how fragment based drug discovery has been used in the discovery of starting points for drug discovery programs and in their optimization.
The (poly-)pharmacological activities of a drug can only be understood if its interactions with cellular components are comprehensively characterized. Mass spectrometry-based chemical proteomics approaches have recently emerged as powerful tools for the characterization of drug-target interactions in samples from cell lines and tissues. At the same time, off-target activities can be identified. This information can contribute toward optimization of candidate drug molecules and reduction of side effects. In this review, we describe recent advances in chemical proteomics and outline potential applications in drug discovery.
Current GPCR cell-based assays often rely on the measurement of a loaded fluorescent dye, fluorescently tagged targets, or the expression of a reporter. These manipulations may alter the cellular physiology of the target GPCR, and the measurements may be subject to off-target interference of compounds. Label-free optical biosensor-based technologies that provide a noninvasive methodology to study GPCR activation and signaling have been developed. These technologies enable the evaluation of drug effects on various GPCRs that couple to different signal transduction pathways using only one assay platform. This technology is highly sensitive and detects inverse agonism, therefore providing a convenient tool to study the pharmacology of drugs. Furthermore, its real-time kinetic measurements give researchers additional information about the biological responses induced by the drug. This assay platform when applied in early drug discovery efforts can provide valuable information on the mechanism of action and pharmacology profiles of drug candidates.
Naïve Bayesian classifiers are a relatively recent addition to the arsenal of tools available to computational chemists. These classifiers fall into a class of algorithms referred to broadly as machine learning algorithms. Bayesian classifiers may be used in conjunction with classical modeling techniques to assist in the rapid virtual screening of large compound libraries in a systematic manner with a minimum of human intervention. This approach allows computational scientists to concentrate their efforts on their core strengths of model building. Bayesian classifiers have an added advantage of being able to handle a variety of numerical or binary data such as physicochemical properties or molecular fingerprints, making the addition of new parameters to existing models a relatively straightforward process. As a result, during a drug discovery project these classifiers can better evolve with the needs of the projects from general models in the lead finding stages to increasingly precise models in the lead optimization stages that are of particular interest to a specific medicinal chemistry team. Although other machine learning algorithms abound, Bayesian classifiers have been shown to compare favorably under most working conditions and have been shown to be tolerant of noisy experimental data.
G protein-coupled receptors (GPCRs) represent 50-60% of the current drug targets. There is no doubt that this family of membrane proteins plays a crucial role in drug discovery today. Classically, a number of drugs based on GPCRs have been developed for such different indications as cardiovascular, metabolic, neurodegenerative, psychiatric, and oncologic diseases. Owing to the restricted structural information on GPCRs, only limited exploration of structure-based drug design has been possible. Much effort has been dedicated to structural biology on GPCRs and very recently an X-ray structure of the beta2-adrenergic receptor was obtained. This breakthrough will certainly increase the efforts in structural biology on GPCRs and furthermore speed up and facilitate the drug discovery process.
The initial stage of drug development is the hit (active) compound search from a pool of millions of compounds; for this process, in silico (virtual) screening has been successfully applied. One of the problems of in silico screening, however, is the low hit ratio in relation to the high computational cost and the long CPU time. This problem becomes serious in structure-based in silico screening. The major reason is the low accuracy of the estimation of protein-compound binding free energy. The problem of ligand-based in silico screening is that the conventional quantitative structure-activity relationship (QSAR) approach is not effective at predicting new hit compounds with new scaffolds. Recently, machine-learning approaches have been applied to in silico drug screening to overcome the above problems. We review here machine-learning approaches for both structure-based and ligand-based drug screening. Machine learning is used to improve database enrichment in two ways, namely by improving the docking score calculated by the protein-compound docking program and by calculating the optimal distance between the feature vectors of active and inactive compounds. Both approaches require compounds that are known to be active with respect to the target protein. In structure-based screening, the former approach is mainly used with a protein-compound affinity matrix. In ligand-based screening, both the former and latter approaches are used, and the latter approach can be applied to various kinds of descriptors, such as 1D/2D descriptors/fingerprints and the affinity fingerprint given by the protein-compound affinity matrix.
Although recognized in small molecules for quite some time, the implications of halogen bonding in biomolecular systems are only now coming to light. In this study, several systems of proteins in complex with halogenated ligands have been investigated by using a two-layer QM/MM ONIOM methodology. In all cases, the halogen-oxygen distances are shown to be much less than the van der Waals radius sums. Single-point energy calculations unveil that the interaction becomes comparable in magnitude to classical hydrogen bonding. Furthermore, we found that the strength of the interactions attenuates in the order H approximately I > Br > Cl. These results agree well with the characteristics discovered within small model halogen-bonded systems. A detailed analysis of the interactions reveals that halogen bonding interactions are responsible for the different conformation of the molecules in the active site. This study would help to establish such interaction as a potential and effective tool in the context of drug design.
Protein kinases catalyse key phosphorylation reactions in signalling cascades that affect every aspect of cell growth, differentiation and metabolism. The kinases have become prime targets for drug intervention in the diseased state, especially in cancer. There are currently 10 drugs that have been approved for clinical use and many more in clinical trials. This review summarises the structural basis for protein kinase inhibition and discusses the mode of action for each of the approved drugs in the light of structural results. All but one of the approved compounds target the ATP binding site on the kinase. Both the active and inactive conformations of protein kinases have been used in strategies to produce potent and selective compounds. Targeting the inactive conformation can give high specificity. Targeting the active conformation is favourable where the diseased state has arisen from activating mutations, but such inhibitors generally target several protein kinases. Drug resistance mutations are a potential risk for both conformational states, where drug-binding regions are not directly involved in catalysis. Imatinib (Glivec), the most successful of protein kinase inhibitors, targets the inactive conformation of ABL tyrosine kinase. Newer compounds, such as dasatinib, which targets the ABL active state, have been developed to increase potency and have proved effective for some, but not all, drug-resistant mutations. The first epidermal growth factor receptor (EGFR) inhibitors in clinical use [gefitinib (Iressa) and erlotinib (Tarceva)] targeted the active form of the kinase, and this proved advantageous for patients whose cancer was caused by mutations that resulted in a constitutively active EGFR kinase domain. Newer approved compounds, such as lapatinib (Tykerb), target the inactive conformation with high potency. A further compound that forms a covalent attachment to the kinase has been found to overcome one of the major drug resistance mutations, where the effectiveness of the drug in vivo is dependent on its ability to compete successfully in the presence of cellular concentrations of ATP. Inhibitors of vascular endothelial growth factor receptor (VEGFR) kinase against cancer angiogenesis show the advantage of some relaxation in specificity. Sorafenib, originally developed as RAF inhibitor, is now in clinical use as a VEGFR inhibitor. Temsirolimus (a derivative of rapamycin) is the only example of a drug in clinical use that does not target the kinase ATP site. Instead rapamycin, when in complex with the protein FKBP12, effectively targets mTOR kinase at a site located on a domain, the FRB domain, that appears to be involved in localisation or substrate docking.
The term "pharmacological promiscuity" describes a compound's pharmacological activity at multiple targets. Pharmacological promiscuity is undesired in typical drug discovery projects, which focus on the "one drug-one target" paradigm. Off-target activity can lead to adverse drug reactions, or can obscure pharmacodynamic effects in animal models. Therefore, advanced lead compounds, pharmacological tool compounds, and drug candidates are usually screened against panels of safety-relevant targets to detect unwanted pharmacological activities. To identify determinants of pharmacological promiscuity, we compared the panel screening outcomes of 213 recent Roche compounds with their molecular properties. Pronounced promiscuity was not observed below a threshold Clog P value of 2. For basic compounds, the propensity for weak off-target activity was found to increase with calculated basicities, whereas the potential for strong off-target activity depends more qualitatively on the presence of a positive charge at physiological pH. Compounds originating from projects with an aminergic receptor or transporter as a therapeutic target are particularly prone to promiscuity; the promiscuity of such compounds is mainly caused by their activity at other aminergic targets in the screening panel.
Molecular recognition between proteins and their interacting partners underlies the biochemistry of living organisms. Specificity in this recognition is thought to be essential, whereas promiscuity is often associated with unwanted side effects, poor catalytic properties and errors in biological function. Recent experimental evidence suggests that promiscuity, not only in interactions but also in the actual function of proteins, is not as rare as was previously thought. This has implications not only for our fundamental understanding of molecular recognition and how protein function has evolved over time but also in the realm of biotechnology. Understanding protein promiscuity is becoming increasingly important not only to optimize protein engineering applications in areas as diverse as synthetic biology and metagenomics but also to lower attrition rates in drug discovery programs, identify drug interaction surfaces less susceptible to escape mutations and potentiate the power of polypharmacology.
The applicability of the newly developed PM6 method for modeling proteins is investigated. In order to allow the geometries of such large systems to be optimized rapidly, three modifications were made to the conventional semiempirical procedure: the matrix algebra method for solving the self-consistent field (SCF) equations was replaced with a localized molecular orbital method (MOZYME), Baker's Eigenfollowing technique for geometry optimization was replaced with the L-BFGS function minimizer, and some of the integrals used in the NDDO set of approximations were replaced with point-charge and polarization functions. The resulting method was used in the unconstrained geometry optimization of 45 proteins ranging in size from a simple nonapeptide of 244 atoms to an importin consisting of 14,566 atoms. For most systems, PM6 gave structures in good agreement with the reported X-ray structures. Some derived properties, such as pKa and bulk elastic modulus, were also calculated. The applicability of PM6 to model transition states was investigated by simulating a hypothetical reaction step in the chymotrypsin-catalyzed hydrolysis of a peptide bond. A proposed technique for generating accurate protein geometries, starting with X-ray structures, was examined.
There have been a number of highly publicized safety-based drug withdrawals in the United States in recent years. We conducted a review of drugs withdrawn since 1993 and examined trends in drug withdrawals. Our objective was to determine the frequency and characteristics of withdrawn drugs and trends since 1993, and to discuss the implications of the findings. We found that a mean of 1.5 drugs per year have been withdrawn since 1993, and that the number of withdrawals has not increased over time. However, some recent drug withdrawals have impacted large numbers of people. The rate of withdrawals alone is not an adequate measure of the status of drug safety in the US, and there is a serious dearth of data that can be used to examine the impact of drug withdrawals. Although drug withdrawals are an important issue to address, drug safety policies need to be developed within the broader context of drug safety and effectiveness. A comprehensive approach will be needed to address the improvement of drug safety. We propose improvements to the evidence base to increase drug safety and assess how new scientific evidence can be incorporated into drug safety efforts.
DrugBank is a freely available web-enabled database that combines detailed drug data with comprehensive drug-target and drug-action information. It was specifically designed to facilitate in silico drug-target discovery, drug design, drug-metabolism prediction, drug-interaction prediction, and general pharmaceutical education. One of the most unique and useful components of the DrugBank database is the information it contains on drug metabolism, drug-metabolizing enzymes and drug-target polymorphisms. As pharmacogenomics is fundamentally concerned with the role of genes and genetic variation of how an individual responds to a drug, DrugBank is able to offer a convenient venue to explore pharmacogenomic questions in silico. This paper provides a brief overview on DrugBank and how it can facilitate pharmacogenomic research.
Most biologists now conduct sequence searches as a matter of course. But how do we know that a relationship predicted by a homology search is a true, rather than false, hit with the same score? Many biologists design their own experiments with exquisite care yet still assume that results from programs with more than 20 adjustable parameters are 100% reliable. This article explains some of the key steps in getting the most from PSI-Blast, one of the most popular and powerful homology search programs currently available.
The pharmaceutical industry faces considerable challenges, both politically and fiscally. Politically, governments around the world are trying to contain costs and, as health care budgets constitute a very significant part of governmental spending, these costs are the subject of intense scrutiny. In the United States, drug costs are also the subject of intense political discourse. This article deals with the fiscal pressures that face the industry from the perspective of R&D. What impinges on productivity? How can we improve current reduced R&D productivity?
Biotechnology is evolving at a tremendous rate. Although drug discovery is now heavily focused on high throughput and miniaturized screening, the application of these advances to the toxicological assessment of chemicals and chemical products has been slow. Nevertheless, the impending surge in demands for the regulatory toxicity testing of chemicals provides the impetus for the incorporation of novel methodologies into hazard identification and risk assessment. Here, we review the current and likely future value of these new technologies in relation to toxicological evaluation and the protection of human health.
Forget drugs carefully designed to hit one particular molecule - a better way of treating complex diseases such as cancer may be to aim for several targets at once, says Simon Frantz.
The current revision of the European policy for the evaluation of chemicals (REACH) has lead to a controversy with regard to the need of additional animal safety testing. To avoid increases in animal testing but also to save time and resources, alternative in silico or in vitro tests for the assessment of toxic effects of chemicals are advocated. The draft of the original document issued in 29th October 2003 by the European Commission foresees the use of alternative methods but does not give further specification on which methods should be used. Computer-assisted prediction models, so-called predictive tools, besides in vitro models, will likely play an essential role in the proposed repertoire of "alternative methods". The current discussion has urged the Advisory Committee of the German Toxicology Society to present its position on the use of predictive tools in toxicology. Acceptable prediction models already exist for those toxicological endpoints which are based on well-understood mechanism, such as mutagenicity and skin sensitization, whereas mechanistically more complex endpoints such as acute, chronic or organ toxicities currently cannot be satisfactorily predicted. A potential strategy to assess such complex toxicities will lie in their dissection into models for the different steps or pathways leading to the final endpoint. Integration of these models should result in a higher predictivity. Despite these limitations, computer-assisted prediction tools already today play a complementary role for the assessment of chemicals for which no data is available or for which toxicological testing is impractical due to the lack of availability of sufficient compounds for testing. Furthermore, predictive tools offer support in the screening and the subsequent prioritization of compound for further toxicological testing, as expected within the scope of the European REACH program. This program will also lead to the collection of high-quality data which will broaden the database for further (Q)SAR approaches and will in turn increase the predictivity of predictive tools.
The US National Cancer Institute (NCI) 60 human tumour cell line anticancer drug screen (NCI60) was developed in the late 1980s as an in vitro drug-discovery tool intended to supplant the use of transplantable animal tumours in anticancer drug screening. This screening model was rapidly recognized as a rich source of information about the mechanisms of growth inhibition and tumour-cell kill. Recently, its role has changed to that of a service screen supporting the cancer research community. Here I review the development, use and productivity of the screen, highlighting several outcomes that have contributed to advances in cancer chemotherapy.
What is a drug target? And how many such targets are there? Here, we consider the nature of drug targets, and by classifying known drug substances on the basis of the discussed principles we provide an estimation of the total number of current drug targets.
Conventional similarity searching of molecules compares single (or multiple) active query structures to each other in a relative framework, by means of a structural descriptor and a similarity measure. While this often works well, depending on the target, we show here that retrieval rates can be improved considerably by incorporating an external framework describing ligand bioactivity space for comparisons ("Bayes affinity fingerprints"). Structures are described by Bayes scores for a ligand panel comprising about 1000 activity classes extracted from the WOMBAT database. The comparison of structures is performed via the Pearson correlation coefficient of activity classes, that is, the order in which two structures are similar to the panel activity classes. Compound retrieval on a recently published data set could be improved by as much as 24% relative (9% absolute). Knowledge about the shape of the "bioactive chemical universe" is thus beneficial to identifying similar bioactivities. Principal component analysis was employed to further analyze activity space with the objective to define orthogonal ligand bioactive chemical space, leading to nine major (roughly orthogonal) activity axes. Employing only those nine activity classes, retrieval rates are still comparable to original Bayes affinity fingerprints; thus, the concept of orthogonal bioactive ligand chemical space was validated as being an information-rich but low-dimensional representation of bioactivity space. Correlations between activity classes are a major determinant to gauge whether the desired multitarget activity of drugs is (on the basis of current knowledge) a feasible concept because it measures the extent to which activities can be optimized independently, or only by strongly influencing one another.
It is generally recognized that drug discovery and development are very time and resources consuming processes. There is an ever growing effort to apply computational power to the combined chemical and biological space in order to streamline drug discovery, design, development and optimization. In biomedical arena, computer-aided or in silico design is being utilized to expedite and facilitate hit identification, hit-to-lead selection, optimize the absorption, distribution, metabolism, excretion and toxicity profile and avoid safety issues. Commonly used computational approaches include ligand-based drug design (pharmacophore, a 3D spatial arrangement of chemical features essential for biological activity), structure-based drug design (drug-target docking), and quantitative structure-activity and quantitative structure-property relationships. Regulatory agencies as well as pharmaceutical industry are actively involved in development of computational tools that will improve effectiveness and efficiency of drug discovery and development process, decrease use of animals, and increase predictability. It is expected that the power of CADDD will grow as the technology continues to evolve.
Surface plasmon resonance (SPR) sensing has long been used to study biomolecular binding events and their kinetics in a label-free way. This approach has recently been extended to SPR microscopy, which is an ideal tool for probing large microarrays of biomolecules for their binding interactions with various partners and the kinetics of such binding. Commercial SPR microscopes now make it possible to simultaneously monitor binding kinetics on >1300 spots within a protein microarray with a detection limit of approximately 0.3 ng/cm(2), or <50 fg per spot (<1 million protein molecules) with a time resolution of 1s, and spot-to-spot reproducibility within a few percent. Such instruments should be capable of high-throughput kinetic studies of the binding of small ( approximately 200 Da) ligands onto large protein microarrays. The method is label free and uses orders of magnitude less of the precious biomolecules than standard SPR sensing. It also gives the absolute bound amount and binding stoichiometry.
Paradigms in drug design and discovery are changing at a significant pace. Concomitant to the sequencing of over 180 several genomes, the high-throughput miniaturization of chemical synthesis and biological evaluation of a multiple compounds on gene/protein expression and function opens the way to global drug-discovery approaches, no more focused on a single target but on an entire family of related proteins or on a full metabolic pathway. Chemogenomics is this emerging research field aimed at systematically studying the biological effect of a wide array of small molecular-weight ligands on a wide array of macromolecular targets. Since the quantity of existing data (compounds, targets and assays) and of produced information (gene/protein expression levels and binding constants) are too large for manual manipulation, information technologies play a crucial role in planning, analysing and predicting chemogenomic data. The present review will focus on predictive in silico chemogenomic approaches to foster rational drug design and derive information from the simultaneous biological evaluation of multiple compounds on multiple targets. State-of-the-art methods for navigating in either ligand or target space will be presented and concrete drug design applications will be mentioned.
British Journal of Pharmacology (2007) 152, 38–52; doi:10.1038/sj.bjp.0707307
Within recent years, a paradigm shift from traditional receptor-specific studies to a cross-receptor view has taken place within pharmaceutical research to increase the efficiency of modern drug discovery. Receptors are no longer viewed as single entities but grouped into sets of related proteins or receptor families that are explored in a systematic manner. This interdisciplinary approach attempting to derive predictive links between the chemical structures of bioactive molecules and the receptors with which these molecules interact is referred to as chemogenomics. Insights from chemogenomics are used for the rational compilation of screening sets and for the rational design and synthesis of directed chemical libraries to accelerate drug discovery.
British Journal of Pharmacology (2007) 152, 5–7; doi:10.1038/sj.bjp.0707308
Cardiovascular diseases (CDs) are among the most encountered pathologies in western countries; with obesity reaching pandemic proportions, they are soon to become a worldwide problem. High blood pressure is the main risk factor for CDs, and its tight control is an imperative for the treatment of complications such as renal diseases, heart failure, and atherosclerosis. Blood homeostasis and vascular tone are regulated through at least 3 major closely interrelated pathways in which zinc metallopeptidases modulate the concentration of vasoactive mediators. Those extensively studied vasopeptidases were therefore rapidly targeted with specific inhibitors in order to control the levels of vasoconstrictors [angiotensin II (AII) and endothelin-1 (ET-1)] and vasodilators [bradykinin (BK) and atrial natriuretic peptide (ANP)], thereby controlling blood pressure. The first class of inhibitors to be developed were against angiotensin-converting enzyme (ACE), recently followed by dual inhibitors of ACE/neprylisin (NEP), NEP/endothelin-converting enzyme (ECE), and finally triple ACE/NEP/ECE inhibitors. The dual and triple inhibitors are defined as vasopeptidase inhibitors (VPI). In addition to their ability to effectively lower blood pressure in hypertensive patients, drugs targeting these enzymes also displayed antiinflammatory and antifibrotic activities. The major point emerging from recent studies undertaken to improve the management of CDs is that the combined action of different therapeutic strategies (ie, simultaneous modulation of several neurohumoral mediators) shows better results than conservative therapeutic approaches. In this review, we historically present the advances made in the comprehension of the different mechanisms of blood pressure regulation and some of the drugs that arose from this understanding.
Recognition of some of the limitations of target-based drug discovery has recently led to the renaissance of a more holistic approach in which complex biological systems are investigated for phenotypic changes upon exposure to small molecules. The subsequent identification of the molecular targets that underlie an observed phenotypic response--termed target deconvolution--is an important aspect of current drug discovery, as knowledge of the molecular targets will greatly aid drug development. Here, the broad panel of experimental strategies that can be applied to target deconvolution is critically reviewed.