Early Stages Of Drug Discovery Research Articles

In Vitro Image-Based Assay for Trypanosoma cruzi Intracellular Forms.

The advances in development and popularization of automated fluorescence microscopes and pipetting robots allowed scientists to establish high-throughput compound screening using image-based assays for Trypanosoma cruzi intracellular forms, which are associated to chronic Chagas disease. An intracellular T. cruzi image-based assay is a valuable tool to early stage drug discovery for Chagas disease, because it allows scientists to assess a compound's efficacy and safety in the same experiment. During the last 10 years, several improvements have been incorporated into intracellular T. cruzi assay protocols to make them more predictable in what happens with parasites within an infected organism. In the present chapter, a protocol will be presented for an intracellular T. cruzi assay, but at a low-throughput scale, more compatible with facilities in many academic laboratories.

Rapid Syntheses of (-)-FR901483 and (+)-TAN1251C Enabled by Complexity-Generating Photocatalytic Olefin Hydroaminoalkylation.

We report a common strategy to facilitate the syntheses of the polycyclic alkaloids (-)-FR901483 (1) and (+)-TAN1251C (2). A divergent synthetic strategy provides access to both natural products through a pivotal spirolactam intermediate (3), which can be accessed on a gram-scale. A photocatalytic olefin hydroaminoalkylation brings together three readily available building blocks and forges the majority of the carbon framework present in 1 and 2 in a single operation, leading to concise total syntheses. The complexity-generating photocatalytic process also provides direct access to novel non-racemic spirolactam scaffolds that are likely to be of interest to early-stage drug discovery programs.

Biopartitioning micellar electrokinetic chromatography – Concept study of cationic analytes

In the early stages of drug discovery, beyond the biological activity screening, the physicochemical properties including lipophilicity, ionization, solubility as well as metabolic profiling and protein-binding affinity should be determined for each drug candidate. The aim of the presented investigation was the transfer of biopartitioning micellar chromatography (BMC) proposed by Medina-Hernández and co-workers to micellar electrokinetic chromatography (MEKC). Therefore, biopartitioning micellar electrokinetic chromatography (BMEKC) was demonstrated as an alternative tool for estimation of biological properties of drug candidates. To simplify the data processing of developed approach and increase the assay throughput, the application of apparent retention factor (kapp) was proposed. Based on introduced kapp, simple molecular descriptors, protein binding constant and biological properties like blood-brain barrier (BBB) permeability, quantitative retention–activity relationships (QRARs) models were established. These results indicated usefulness of BMEKC method. In comparison to BMC, the proposed BMEKC method offers significant reduction of reagents consumption which meet green analytical chemistry principles.

Development of an in silico prediction system of human renal excretion and clearance from chemical structure information incorporating fraction unbound in plasma as a descriptor

Prediction of pharmacokinetic profiles of new chemical entities is essential in drug development to minimize the risks of potential withdrawals. The excretion of unchanged compounds by the kidney constitutes a major route in drug elimination and plays an important role in pharmacokinetics. Herein, we created in silico prediction models of the fraction of drug excreted unchanged in the urine (fe) and renal clearance (CLr), with datasets of 411 and 401 compounds using freely available software; notably, all models require chemical structure information alone. The binary classification model for fe demonstrated a balanced accuracy of 0.74. The two-step prediction system for CLr was generated using a combination of the classification model to predict excretion-type compounds and regression models to predict the CLr value for each excretion type. The accuracies of the regression models increased upon adding a descriptor, which was the observed and predicted fraction unbound in plasma (fu,p); 78.6% of the samples in the higher range of renal clearance fell within 2-fold error with predicted fu,p value. Our prediction system for renal excretion is freely available to the public and can be used as a practical tool for prioritization and optimization of compound synthesis in the early stage of drug discovery.

Evaluation of colorectal cancer subtypes and cell lines using deep learning.

Colorectal cancer (CRC) is a common cancer with a high mortality rate and a rising incidence rate in the developed world. Molecular profiling techniques have been used to better understand the variability between tumors and disease models such as cell lines. To maximize the translatability and clinical relevance of in vitro studies, the selection of optimal cancer models is imperative. We have developed a deep learning-based method to measure the similarity between CRC tumors and disease models such as cancer cell lines. Our method efficiently leverages multiomics data sets containing copy number alterations, gene expression, and point mutations and learns latent factors that describe data in lower dimensions. These latent factors represent the patterns that are clinically relevant and explain the variability of molecular profiles across tumors and cell lines. Using these, we propose refined CRC subtypes and provide best-matching cell lines to different subtypes. These findings are relevant to patient stratification and selection of cell lines for early-stage drug discovery pipelines, biomarker discovery, and target identification.

Cyclodextrin Inclusion of Medicinal Compounds for Enhancement of their Physicochemical and Biopharmaceutical Properties.

Owing to their wide structural diversity and unique complexing properties, cyclodextrins (CDs) find manifold applications in drug discovery and development. The focus of this mini-review is on their uses as 'enabling excipients' both in the context of early drug discovery and in subsequent optimisation of drug performance. Features highlighted here include descriptions of the structures of CDs, synthetic derivatisation to fine-tune their properties, the nature of inclusion complexation of drugs within the CD cavity, methodology for the study of free and complexed hosts in the solid state and in solution, the inherent pharmacological activity of several CDs and its utility, novel CD-based drug delivery systems, and the role of CDs in drug discovery and optimisation. Illustrative examples are generally based on research reported during the last two decades. Application of CDs to the optimisation of the performance of established drugs is commonplace, but there are many opportunities for the intervention of CDs during the early stages of drug discovery, which could guide the selection of suitable candidates for development, thereby contributing to reducing the attrition rate of new molecular entities.

Prediction of toxicity of secondary metabolites

Abstract The prediction of toxicological endpoints has gained broad acceptance; it is widely applied in early stages of drug discovery as well as for impurities obtained in the production of generic or equivalent products. In this work, we describe methodologies for the prediction of toxicological endpoints compounds, with a particular focus on secondary metabolites. Case studies include toxicity prediction of natural compound databases with anti-diabetic, anti-malaria and anti-HIV properties.

A novel evaluation method for determining drug-induced hepatotoxicity using 3D bio-printed human liver tissue

Predicting drug-induced liver injury is important in early stage drug discovery; however, an accurate prediction with existing hepatotoxicity evaluation tools is difficult. Conventional monolayer (2D) cultures have short viabilities and are therefore inappropriate for performing long-term toxicity tests. Conventionally used 200-μm spheroids also have toxicity detection limits. The goal of this study was to develop a humanized liver tissue capable of evaluating long-term toxicity with high sensitivity. Spheroids consisting of co-cultured cryopreserved primary human hepatocytes and human hepatic stellate cells were developed using a 3D bio-printer. The “3D bio-printed liver tissue”, of ∼1 mm, was then used for long-term viability assessments (over 25 days) based on ATP, albumin, and urea levels. Hepatotoxicity evaluation was performed by analyzing the expression of genes involved in drug metabolism and transport over a 2-week drug exposure period. The 3D bio-printed liver tissue showed improved viability and enhanced gene expression of enzymes related to drug metabolism and transport, as compared to the controls. Additionally, the 3D bio-printed liver tissue demonstrated a high sensitivity for hepatotoxicity evaluation when combined with pathological evaluation and measurements for ATP production, and secretion of albumin and urea. In conclusion, the 3D bio-printed liver tissue was able to detect the toxicity of compounds that was, otherwise, undetected by 2D culture and conventionally used spheroids. These findings demonstrate a 3D bio-printed liver tissue with increased accuracy of hepatotoxicity prediction in the early stages of drug discovery, as compared to currently available methods.

Comparison Study of Computational Prediction Tools for Drug-Target Binding Affinities.

The drug development is generally arduous, costly, and success rates are low. Thus, the identification of drug-target interactions (DTIs) has become a crucial step in early stages of drug discovery. Consequently, developing computational approaches capable of identifying potential DTIs with minimum error rate are increasingly being pursued. These computational approaches aim to narrow down the search space for novel DTIs and shed light on drug functioning context. Most methods developed to date use binary classification to predict if the interaction between a drug and its target exists or not. However, it is more informative but also more challenging to predict the strength of the binding between a drug and its target. If that strength is not sufficiently strong, such DTI may not be useful. Therefore, the methods developed to predict drug-target binding affinities (DTBA) are of great value. In this study, we provide a comprehensive overview of the existing methods that predict DTBA. We focus on the methods developed using artificial intelligence (AI), machine learning (ML), and deep learning (DL) approaches, as well as related benchmark datasets and databases. Furthermore, guidance and recommendations are provided that cover the gaps and directions of the upcoming work in this research area. To the best of our knowledge, this is the first comprehensive comparison analysis of tools focused on DTBA with reference to AI/ML/DL.

Paper-based sol-gel thin films immobilized cytochrome P450 for enzyme activity measurement

Cytochrome P450 (CYP450), and in particular CYP3A4, is the most abundantly expressed CYP450 isozyme implicated in many drug-drug and medicinal plant-drug interactions. Therefore, incorporation of CYP3A4 enzyme screening at an early stage of drug discovery is preferable in order to avoid enzymatic interactions. Here we present for the first time a paper-based CYP3A4 immobilized sol-gel-derived a platform using resorufin benzyl ether as a fluorogenic enzyme substrate used to investigate enzyme activity. The fluorescence intensity of the product can be simply quantified by using a handheld digital microscope and an image analysis software. The limit of quantitation was 0.35 μM with good precision (RSDs < 4.1%). Furthermore, the assay of CYP3A4 activity on the developed paper-based device provided comparable results with those obtained from conventional well-plates (p > 0.05), while offering simplicity and lower cost. Kinetic parameters of the immobilized CYP3A4 in sol-gel coated paper were calculated from the Lineweaver-Burk plot, including Michaelis constant (Km) and maximum velocity (Vmax), which were 2.71 ± 0.35 μM and 0.43 ± 0.05 μM/min, respectively. Moreover, a functional test of these devices was conducted by assessments of known CYP3A4 inhibitors (i.e. ketoconazole, itraconazole) and inducers (i.e. phenytoin, carbamazepine). To further demonstrate the broad range of uses, the devices were utilized to assay plant extracts i.e. Areca catechu seeds, Camellia sinensis leaves, Eclipta prostrata aerial part, providing results in good agreement with previous studies. Furthermore, the sol-gel immobilized enzyme stored at 4 °C can increase storage stability, offering the activity of 86.3 ± 0.4% after 3-weeks storage, equivalent to the activity of the free enzyme solution after 1-week storage. The developed paper-based devices offer versatility, portability and low-cost.

A Miniaturized Pump Out Method for Characterizing Molecule Interaction with ABC Transporters.

Characterizing interaction of newly synthetized molecules with efflux pumps remains essential to improve their efficacy and safety. Caco-2 cell line cultivated on inserts is widely used for measuring apparent permeability of drugs across biological barriers, and for estimating their interaction with efflux pumps such as P-gp, BCRP and MRPs. However, this method remains time consuming and expensive. In addition, detection method is required for measuring molecule passage across cell monolayer and false results can be generated if drugs concentrations used are too high as demonstrated with quinidine. For this reason, we developed a new protocol based on the use of Caco-2 cell directly seeded on 96- or 384-well plates and the use of fluorescent substrates for efflux pumps. We clearly observed that the new method reduces costs for molecule screening and leads to higher throughput compared to traditional use of Caco-2 cell model. This accelerated model could provide quick feedback regarding the molecule design during the early stage of drug discovery and therefore reduce the number of compounds to be further evaluated using the traditional Caco-2 insert method.

CapsCarcino: A novel sparse data deep learning tool for predicting carcinogens

Determining chemical carcinogenicity in the early stages of drug discovery is fundamentally important to prevent the adverse effect of carcinogens on human health. There has been a recent surge of interest in developing computational approaches to predict chemical carcinogenicity. However, the predictive power of many existing approaches is limited, and there is plenty of room for improvement. Here, we develop a new deep learning architecture, termed CapsCarcino, to distinguish between carcinogens and noncarcinogens. CapsCarcino is constructed based on a dynamic routing algorithm that requires less data, extracts more comprehensive information, and does not require feature selection. We find that CapsCarcino provides a significantly improved predictive and generalization ability over, and outperforms five other machine learning models. Specifically, the best model of CapsCarcino achieves an accuracy of 85.0% on an external validation dataset. In addition, we discover that the enhanced predictive capability of CapsCarcino over that of the other methods is robust and can be achieved using sparse datasets. Training on merely 20% of the dataset, CapsCarcino performs comparably to the other methods based on the full training dataset. Further mechanism analysis indicates that CapsCarcino could efficiently learn the characteristics of carcinogens even if structural alerts are insufficiently represented. The results indicate that CapsCarcino should be helpful for carcinogen risk assessment.

ADMET Evaluation in Drug Discovery. 19. Reliable Prediction of Human Cytochrome P450 Inhibition Using Artificial Intelligence Approaches.

Adverse effects induced by drug-drug interactions may result in early termination of drug development or even withdrawal of drugs from the market, and many drug-drug interactions are caused by the inhibition of cytochrome P450 (CYP450) enzymes. Therefore, the accurate prediction of the inhibition capability of a given compound against a specific CYP450 isoform is highly desirable. In this study, three ensemble learning methods, including random forest, gradient boosting decision tree, and eXtreme gradient boosting (XGBoost), and two deep learning methods, including deep neural networks and convolutional neural networks, were used to develop classification models to discriminate inhibitors and noninhibitors for five major CYP450 isoforms (1A2, 2C9, 2C19, 2D6, and 3A4). The results demonstrate that the ensemble learning models generally give better predictions than the deep learning models for the external test sets. Among all of the models, the XGBoost models achieve the best classification capability (average prediction accuracy of 90.4%) for the test sets, which even outperform the previously reported model developed by the multitask deep autoencoder neural network (88.5%). The Shapley additive explanation method was then used to interpret the models and analyze the misclassified molecules. The important molecular descriptors given by our models are consistent with the structural preferences for inhibitors of different CYP450 isoforms, which may provide valuable clues to detect potential drug-drug interactions in the early stage of drug discovery.

A Quantum-Inspired Method for Three-Dimensional Ligand-Based Virtual Screening.

Measuring similarity between molecules is an important part of virtual screening (VS) experiments deployed during the early stages of drug discovery. Most widely used methods for evaluating the similarity of molecules use molecular fingerprints to encode structural information. While similarity methods using fingerprint encodings are efficient, they do not consider all the relevant aspects of molecular structure. In this paper, we describe a quantum-inspired graph-based molecular similarity (GMS) method for ligand-based VS. The GMS method is formulated as a quadratic unconstrained binary optimization problem that can be solved using a quantum annealer, providing the opportunity to take advantage of this nascent and potentially groundbreaking technology. In this study, we consider various features relevant to ligand-based VS, such as pharmacophore features and three-dimensional atomic coordinates, and include them in the GMS method. We evaluate this approach on various datasets from the DUD_LIB_VS_1.0 library. Our results show that using three-dimensional atomic coordinates as features for comparison yields higher early enrichment values. In addition, we evaluate the performance of the GMS method against conventional fingerprint approaches. The results demonstrate that the GMS method outperforms fingerprint methods for most of the datasets, presenting a new alternative in ligand-based VS with the potential for future enhancement.

Current Progress on the Genomics of Schistosomiasis for Drug Discovery and Diagnostics.

For a number of decades, schistosomiasis has remained a public threat and an economic burden in a number of countries, directly impacting over 200 million people. The past 15 years have seen tremendous progress in the development of high-throughput methods for targeting or compound selection that are vital to early-stage schistosome drug discovery research. Genomewide approaches to analyze gene expression at the transcriptional and other -omic levels have helped immensely for gaining insight into the pathways and mechanisms involved in the schistosomiasis and it is expected to revolutionize the drug discovery as well as related diagnostics. This review discusses the most recent progress of pharmacology and genomics concerning schistosomiasis with a focus on drug discovery and diagnostic tools. It also provides chemical structural insights of promising targets along with available in vitro and/or in vivo data. Although significant research has been done to identify new molecules for the treatment and new methods for diagnosis, the necessity of new options for the sustainable control of schistosomiasis remains a great challenge.

Protein Preparation Automatic Protocol for High-Throughput Inverse Virtual Screening: Accelerating the Target Identification by Computational Methods.

Structure-based virtual screening is highly used in the early stages of drug discovery to identify new putative lead compounds for a given target. However, when a small molecule elicits a biological effect, but its target is unknown, or the side effects it causes arise from its undesired interaction with unknown counterparts, the identification of its interacting targets represents an indispensable task. The computational procedure named inverse virtual screening, which relies on docking a molecule (or a small set of compounds) against panels of target proteins to select the most promising complexes, could be useful to overcome these issues. Panels can contain thousands of proteins, and they must be correctly prepared to assure the best docking performance. Therefore, the preparation of panels of proteins collected in the Protein Data Bank ( www.rcsb.org ), if manually performed, may be costly in terms of time and efforts, and this can limit the applicability of this approach in high-throughput virtual screening workflows. We here show an automated workflow to speed up panel preparation and development, and to test its performance, this protocol was initially applied to a panel of 628 viral proteins and, afterward, to a panel of transferase proteins (2789 entries) to perform a large inverse virtual screening study, testing a small set of compounds synthesized in our laboratory. Tankyrase 2 (PARP 5b) was selected as their preferred target of interaction, and the predicted binding was validated by means of surface plasmon resonance experiments. This protocol is useful for the rapid identification of the interacting target for a bioactive compound; accordingly, it facilitates the re-evaluation of the pharmacological activity of known active compounds, addressing the repurposing and the polypharmacology concepts.

PASS-based prediction of metabolites detection in biological systems

ABSTRACTMetabolite identification is an essential part of the drug discovery and development process. Experimental methods allow identifying metabolites and estimating their relative amount, but they require cost-intensive and time-consuming techniques. Computational methods for metabolite prediction are devoid of these shortcomings and may be applied at the early stage of drug discovery. In this study, we investigated the possibility of creating SAR models for the prediction of the qualitative metabolite yield (‘major’, ‘minor’, ”trace” and ”negligible”) depending on species and biological experimental systems. In addition, we have created models for prediction of xenobiotic excretion depending on its administration route for different species. The prediction is based on an algorithm of naïve Bayes classifier implemented in PASS software. The average accuracy of prediction was 0.91 for qualitative metabolite yield prediction and 0.89 for prediction of xenobiotic excretion. The created models were included as a component of MetaTox web application, which allows predicting the xenobiotic metabolism pathways (http://www.way2drug.com/mg).

Prediction of hERG K+ channel blockage using deep neural networks

Human ether-a-go-go-related gene (hERG) K+ channel blockage may cause severe cardiac side-effects and has become a serious issue in safety evaluation of drug candidates. Therefore, improving the ability to avoid undesirable hERG activity in the early stage of drug discovery is of significant importance. The purpose of this study was to build predictive models of hERG activity by deep neural networks. For each combination of sampling methods and descriptors, deep neural networks with different architectures were implemented to build classification models. The optimal model M15 with three hidden layers, undersampling method, and 2D descriptors yielded the prediction accuracy of 0.78 and F1 score of 0.75 on the test set as well as accuracy of 0.77 and F1 score of 0.34 on the external validation set, outperforming the other 35 models including 9 random forest models. Particularly, the optimal model M15 achieved the highest F1 score and the second highest accuracy when compared with other five methods from four groups using different machine learning algorithms with the same external validation set. It can be believed that this model has powerful capability on prediction of hERG toxicity, which is of great benefit for developing novel drug candidates.

Engineering a surrogate human heteromeric α/β glycine receptor orthosteric site exploiting the structural homology and stability of acetylcholine-binding protein.

Protein-engineering methods have been exploited to produce a surrogate system for the extracellular neurotransmitter-binding site of a heteromeric human ligand-gated ion channel, the glycine receptor. This approach circumvents two major issues: the inherent experimental difficulties in working with a membrane-bound ion channel and the complication that a heteromeric assembly is necessary to create a key, physiologically relevant binding site. Residues that form the orthosteric site in a highly stable ortholog, acetylcholine-binding protein, were selected for substitution. Recombinant proteins were prepared and characterized in stepwise fashion exploiting a range of biophysical techniques, including X-ray crystallography, married to the use of selected chemical probes. The decision making and development of the surrogate, which is termed a glycine-binding protein, are described, and comparisons are provided with wild-type and homomeric systems that establish features of molecular recognition in the binding site and the confidence that the system is suited for use in early-stage drug discovery targeting a heteromeric α/β glycine receptor.

Three-Compartment Model Analysis with Minimal Sampling Points in the Caco-2 Permeability Assay.

The purpose of this study was to establish a modified method of three-compartment model analysis that minimized the sampling frequency. A Caco-2 permeability assay was performed on ten structurally diverse compounds with passive diffusion. A three-compartment model was analyzed by a conventional method and a method with fewer sampling points, called the simplified method, using concentration-time profiles in the donor, intracellular, and receiver compartments. The concentration-time profiles in all compartments were well described by the conventional method. The calculated unbound fraction of intracellular (fu2) and apparent permeability coefficient (Papp) were 0.0107-1.22 and 0.886-146 × 10-6 cm/s, respectively. The simplified method also described the concentration profiles in the compartments of all ten compounds except one, ibuprofen. The difference in values calculated by the simplified method compared to the conventional method was between -7 and 7% for fu2 and between -6 and 42% for Papp. These results suggested that the parameter values from the simplified method were comparable with those from the conventional method. The simplified method may be a promising approach to improve the throughput of three-compartment model analyses of Caco-2 permeability assays in the early stages of drug discovery.