Improving Docking Results via Reranking of Ensembles of Ligand Poses in Multiple X-ray Protein Conformations with MM-GBSA

Druggable cavities and allosteric modulators of the cell division cycle 7 (CDC7) kinase

Cell division cycle 7 kinase (CDC7) has been found overexpressed in many cancer cell lines being also one of the kinases involved in the nuclear protein TDP-43 phosphorylation in vivo. Thus, inhibitors of CDC7 are emerging drug candidates for the treatment of oncological and neurodegenerative unmet diseases. All the known CDC7 inhibitors are ATP-competitives, lacking of selectivity enough for success in clinical trials. As allosteric sites are less conserved among kinase proteins, discovery of allosteric modulators of CDC7 is a great challenge and opportunity in this field.Using different computational approaches, we have here identified new druggable cavities on the human CDC7 structure and subsequently selective CDC7 inhibitors with allosteric modulation mainly targeting the pockets where the interaction between this kinase and its activator DBF4 takes place.

Unearthing phytochemicals as natural inhibitors for pantothenate synthetase in Mycobacterium tuberculosis: A computational approach

Pantothenate synthetase protein plays a pivotal role in the biosynthesis of coenzyme A (CoA), which is a crucial molecule involved in a number of cellular processes including the metabolism of fatty acid, energy production, and the synthesis of various biomolecules, which is necessary for the survival of Mycobacterium tuberculosis (Mtb). Therefore, inhibiting this protein could disrupt CoA synthesis, leading to the impairment of vital metabolic processes within the bacterium, ultimately inhibiting its growth and survival. This study employed molecular docking, structure-based virtual screening, and molecular dynamics (MD) simulation to identify promising phytochemical compounds targeting pantothenate synthetase for tuberculosis (TB) treatment. Among 239 compounds, the top three (rutin, sesamin, and catechin gallate) were selected, with binding energy values ranging from -11 to -10.3 kcal/mol, and the selected complexes showed RMSD (<3 Å) for 100 ns MD simulation time. Furthermore, molecular mechanics generalized Born surface area (MM/GBSA) binding free energy calculations affirmed the stability of these three selected phytochemicals with binding energy ranges from -82.24 ± 9.35 to -66.83 ± 4.5 kcal/mol. Hence, these identified natural plant-derived compounds as potential inhibitors of pantothenate synthetase could be used to inhibit TB infection in humans.

From Proteome to Potential Drugs: Integration of Subtractive Proteomics and Ensemble Docking for Drug Repurposing against Pseudomonas aeruginosa RND Superfamily Proteins

Pseudomonas aeruginosa (P. aeruginosa) poses a significant threat as a nosocomial pathogen due to its robust resistance mechanisms and virulence factors. This study integrates subtractive proteomics and ensemble docking to identify and characterize essential proteins in P. aeruginosa, aiming to discover therapeutic targets and repurpose commercial existing drugs. Using subtractive proteomics, we refined the dataset to discard redundant proteins and minimize potential cross-interactions with human proteins and the microbiome proteins. We identified 12 key proteins, including a histidine kinase and members of the RND efflux pump family, known for their roles in antibiotic resistance, virulence, and antigenicity. Predictive modeling of the three-dimensional structures of these RND proteins and subsequent molecular ensemble-docking simulations led to the identification of MK-3207, R-428, and Suramin as promising inhibitor candidates. These compounds demonstrated high binding affinities and effective inhibition across multiple metrics. Further refinement using non-covalent interaction index methods provided deeper insights into the electronic effects in protein-ligand interactions, with Suramin exhibiting superior binding energies, suggesting its broad-spectrum inhibitory potential. Our findings confirm the critical role of RND efflux pumps in antibiotic resistance and suggest that MK-3207, R-428, and Suramin could be effectively repurposed to target these proteins. This approach highlights the potential of drug repurposing as a viable strategy to combat P. aeruginosa infections.

Hit Identification Driven by Combining Artificial Intelligence and Computational Chemistry Methods: A PI5P4K-β Case Study

Computer-aided drug design (CADD), especially artificial intelligence-driven drug design (AIDD), is increasingly used in drug discovery. In this paper, a novel and efficient workflow for hit identification was developed within the ID4Inno drug discovery platform, featuring innovative artificial intelligence, high-accuracy computational chemistry, and high-performance cloud computing. The workflow was validated by discovering a few potent hit compounds (best IC50 is ∼0.80 μM) against PI5P4K-β, a novel anti-cancer target. Furthermore, by applying the tools implemented in ID4Inno, we managed to optimize these hit compounds and finally obtained five hit series with different scaffolds, all of which showed high activity against PI5P4K-β. These results demonstrate the effectiveness of ID4inno in driving hit identification based on artificial intelligence, computational chemistry, and cloud computing.

Predicting Protein-Peptide Interactions: Benchmarking Deep Learning Techniques and a Comparison with Focused Docking

The accurate prediction of protein structures achieved by deep learning (DL) methods is a significant milestone and has deeply impacted structural biology. Shortly after its release, AlphaFold2 has been evaluated for predicting protein-peptide interactions and shown to significantly outperform RoseTTAfold as well as a conventional blind docking method: PIPER-FlexPepDock. Since then, new AlphaFold2 models, trained specifically to predict multimeric assemblies, have been released and a new ab initio folding model OmegaFold has become available. Here, we assess docking success rates for these new DL folding models and compare their performance with our state-of-the-art, focused peptide-docking software AutoDock CrankPep (ADCP). The evaluation is done using the same dataset and performance metric for all methods. We show that, for a set of 99 nonredundant protein-peptide complexes, the new AlphaFold2 model outperforms other Deep Learning approaches and achieves remarkable docking success rates for peptides. While the docking success rate of ADCP is more modest when considering the top-ranking solution only, it samples correct solutions for around 62% of the complexes. Interestingly, different methods succeed on different complexes, and we describe a consensus docking approach using ADCP and AlphaFold2, which achieves a remarkable 60% for the top-ranking results and 66% for the top 5 results for this set of 99 protein-peptide complexes.

Virtual screening and molecular dynamics simulations provide insight into repurposing drugs against SARS-CoV-2 variants Spike protein/ACE2 interface

After over two years of living with Covid-19 and hundreds of million cases worldwide there is still an unmet need to find proper treatments for the novel coronavirus, due also to the rapid mutation of its genome. In this context, a drug repositioning study has been performed, using in silico tools targeting Delta Spike protein/ACE2 interface. To this aim, it has been virtually screened a library composed by 4388 approved drugs through a deep learning-based QSAR model to identify protein-protein interactions modulators for molecular docking against Spike receptor binding domain (RBD). Binding energies of predicted complexes were calculated by Molecular Mechanics/Generalized Born Surface Area from docking and molecular dynamics simulations. Four out of the top twenty ranking compounds showed stable binding modes on Delta Spike RBD and were evaluated also for their effectiveness against Omicron. Among them an antihistaminic drug, fexofenadine, revealed very low binding energy, stable complex, and interesting interactions with Delta Spike RBD. Several antihistaminic drugs were found to exhibit direct antiviral activity against SARS-CoV-2 in vitro, and their mechanisms of action is still debated. This study not only highlights the potential of our computational methodology for a rapid screening of variant-specific drugs, but also represents a further tool for investigating properties and mechanisms of selected drugs.

Molecular Screening and Toxicity Estimation of 260,000 Perfluoroalkyl and Polyfluoroalkyl Substances (PFASs) through Machine Learning

Perfluoroalkyl and polyfluoroalkyl substances (PFASs) are a class of chemicals widely used in industrial applications due to their exceptional properties and stability. However, they do not readily degrade in the environment and are linked to contamination and adverse health effects in humans and wildlife. To find alternatives for the most commonly used PFAS molecules that maintain their desirable chemical properties but are not adverse to biological lifeforms, a novel approach based upon machine learning is utilized. The machine learning model is trained on an existing set of PFAS molecules to generate over 260,000 novel PFAS molecules, which we dub PFAS-AI-Gen. Using molecular descriptors with known relationships to toxicity and industrial suitability followed by molecular docking and molecular dynamics simulations, this set of molecules is screened. In this manner, increasingly complex calculations are performed only for candidate molecules that are most likely to yield the desired properties of low binding affinity toward two selected protein receptors, the human pregnane x receptor (hPXR) and peroxisome proliferator-activated receptor γ (PPAR-γ), and high industrial suitability, defined by critical micelle concentration (CMC). The selection criteria of low binding affinity and high industrial suitability are relative to the popular PFAS alternative GenX. hPXR and PPAR-γ are selected as they are PFAS targets and facilitate a variety of functions, such as drug metabolism and glucose regulation, respectively. Through this approach, 22 promising new PFAS substitutes that may warrant experimental investigation are identified. This integrated approach of molecular screening and toxicity estimation may be applicable to other chemical classes.

Natural Products as Mcl-1 Inhibitors: A Comparative Study of Experimental and Computational Modelling Data

The human myeloid leukemia cell differentiation protein (hMcl-1) is an anti-apoptotic multi-partner protein, belonging to the B-cell lymphoma-2 (Bcl-2) family of proteins. Studies have linked hMcl-1 alleviated expression with resistance to hemopoietic chemotherapeutics, which makes it a key drug target in blood cancers. However, most of the developed small- to medium-sized hMcl-1 inhibitors have typical off-target activity towards other members of the Bcl-2 family. To improve the hMcl-1 inhibitor design, especially exploring a suitable scaffold with pharmacophoric features, we focused on natural hMcl-1 inhibitors. To date, seven classes of natural compounds have been isolated, which display a low micromolar affinity for hMcl-1 and have limited biophysical studies. We screened hMcl-1 co-crystal structures, and identified nine co-crystal structures of hMcl-1 protein, which were later evaluated by multiple receptor conformations (which indicates that the differences between hMcl-1 in crystal structures are low (RMSD values between 0.52 and 1.13 Å, average RMSD of 0.638–0.888 Å, with a standard deviation of 0.102–0.185Å)), and multiple ligand conformations (which led to the selection of the PDB structure, 3WIX (RMSD value = 0.879 Å, standard deviation 0.116 Å), to accommodate various Mcl-1 ligands from a range of co-crystal PDB files) methods. Later, the three adopted docking methods were assessed for their ability to reproduce the conformation bound to the crystal as well as predict trends in Ki values based on calculated RMSD and docking energies. Iterative docking and clustering of the docked pose within ≤1.0 Å was used to evaluate the reproducibility of the adopted docking methods and compared with their experimentally determined hMcl-1 affinity data.

Structure-Activity Relationship Development Efforts towards Peripherally Selective Analogs of the Cannabinoid Receptor Partial Agonist BAY 59-3074

Selective modulation of peripheral cannabinoid receptors (CBRs) has potential therapeutic applications in medical conditions, including obesity, diabetes, liver diseases, GI disorders and pain. While there have been considerable efforts to produce selective antagonists or full agonists of CBRs, there has been limited reports on the development of partial agonists. Partial agonists targeting peripheral CBRs may have desirable pharmacological profiles while not producing centrally mediated dissociative effects. Bayer reported that BAY 59-3074 is a CNS penetrant partial agonist of both CB1 and CB2 receptors with efficacy in rat models of neuropathic and inflammatory pain. In this report, we demonstrate our efforts to synthesize analogs that would favor peripheral selectivity, while maintaining partial agonism of CB1. Our efforts led to the identification of a novel compound, which is a partial agonist of the human CB1 (hCB1) receptor with vastly diminished brain exposure compared to BAY 59-3074.

Dynamic Docking of Macrocycles in Bound and Unbound Protein Structures with DynaDock

Macrocycles are interesting molecules with unique features due to their conformationally constrained yet flexible ring structure. This characteristic poses a difficult challenge for computational modeling studies since they rely on accurate structural descriptions. In particular, molecular docking calculations suffer from the lack of ring flexibility during pose generation, which is often compensated by using pregenerated ligand conformer ensembles. Moreover, receptor structures are mainly treated rigidly, which limits the use of many docking tools. In this study, we optimized our previous molecular dynamics-based sampling and docking pipeline specifically designed for the accurate prediction of macrocyclic compounds. We developed a dihedral classification procedure for in-depth conformational analysis of the macrocyclic rings and extracted structural ensembles that were subsequently docked in both bound and unbound protein structures employing a fully flexible approach. Our results suggest that including a ring conformer close to the bound state in the starting ensemble increases the chance of successful docking. The bioactive conformations of a diverse set of ligands could be predicted with high and decent accuracy in bound and unbound protein structures, respectively, due to the incorporation of full molecular flexibility in our approach. The remaining unsuccessful docking calculations were mainly caused by large flexible substituents that bind to surface-exposed binding sites, rather than the macrocyclic ring per se and could be further improved by explicit molecular dynamics simulations of the docked complex.

Ensemble learning from ensemble docking: revisiting the optimum ensemble size problem

Despite considerable advances obtained by applying machine learning approaches in protein–ligand affinity predictions, the incorporation of receptor flexibility has remained an important bottleneck. While ensemble docking has been used widely as a solution to this problem, the optimum choice of receptor conformations is still an open question considering the issues related to the computational cost and false positive pose predictions. Here, a combination of ensemble learning and ensemble docking is suggested to rank different conformations of the target protein in light of their importance for the final accuracy of the model. Available X-ray structures of cyclin-dependent kinase 2 (CDK2) in complex with different ligands are used as an initial receptor ensemble, and its redundancy is removed through a graph-based redundancy removal, which is shown to be more efficient and less subjective than clustering-based representative selection methods. A set of ligands with available experimental affinity are docked to this nonredundant receptor ensemble, and the energetic features of the best scored poses are used in an ensemble learning procedure based on the random forest method. The importance of receptors is obtained through feature selection measures, and it is shown that a few of the most important conformations are sufficient to reach 1 kcal/mol accuracy in affinity prediction with considerable improvement of the early enrichment power of the models compared to the different ensemble docking without learning strategies. A clear strategy has been provided in which machine learning selects the most important experimental conformers of the receptor among a large set of protein–ligand complexes while simultaneously maintaining the final accuracy of affinity predictions at the highest level possible for available data. Our results could be informative for future attempts to design receptor-specific docking-rescoring strategies.

Molecular docking and dynamics studies on propolis sulabiroin-A as a potential inhibitor of SARS-CoV-2

Molecular docking and dynamics simulations were conducted to investigate the antiviral activity of Propolis Sulabiroin-A to inhibit the SARS-CoV-2 virus with quercetin, hesperidin, and remdesivir as control ligands. The parameters calculated were docking score and binding energy/molecular mechanics-generalized born surface area (MMGBSA), root mean square displacement (RMSD), and root mean square fluctuation (RMSF). Docking and MMGBSA scores showed that all the ligands demonstrate an excellent candidate as an inhibitor, and the order of both scores is hesperidin, remdesivir, quercetin, and sulabiroin-A. The molecular dynamics simulation showed that all the ligands are good candidates as inhibitors. Although the fluctuation of Sulabiroin-A is relatively high, it has less protein-ligand interaction time than other ligands. Overall, there is still a good possibility that sulabiroin-A can be used as an alternative inhibitor if a new structure of receptor SARS-CoV-2 is used.

Improving Structure-Based Virtual Screening with Ensemble Docking and Machine Learning

One of the main challenges of structure-based virtual screening (SBVS) is the incorporation of the receptor's flexibility, as its explicit representation in every docking run implies a high computational cost. Therefore, a common alternative to include the receptor's flexibility is the approach known as ensemble docking. Ensemble docking consists of using a set of receptor conformations and performing the docking assays over each of them. However, there is still no agreement on how to combine the ensemble docking results to obtain the final ligand ranking. A common choice is to use consensus strategies to aggregate the ensemble docking scores, but these strategies exhibit slight improvement regarding the single-structure approach. Here, we claim that using machine learning (ML) methodologies over the ensemble docking results could improve the predictive power of SBVS. To test this hypothesis, four proteins were selected as study cases: CDK2, FXa, EGFR, and HSP90. Protein conformational ensembles were built from crystallographic structures, whereas the evaluated compound library comprised up to three benchmarking data sets (DUD, DEKOIS 2.0, and CSAR-2012) and cocrystallized molecules. Ensemble docking results were processed through 30 repetitions of 4-fold cross-validation to train and validate two ML classifiers: logistic regression and gradient boosting trees. Our results indicate that the ML classifiers significantly outperform traditional consensus strategies and even the best performance case achieved with single-structure docking. We provide statistical evidence that supports the effectiveness of ML to improve the ensemble docking performance.

A practical guide to large-scale docking

Structure-based docking screens of large compound libraries have become common in early drug and probe discovery. As computer efficiency has improved and compound libraries have grown, the ability to screen hundreds of millions, and even billions, of compounds has become feasible for modest-sized computer clusters. This allows the rapid and cost-effective exploration and categorization of vast chemical space into a subset enriched with potential hits for a given target. To accomplish this goal at speed, approximations are used that result in undersampling of possible configurations and inaccurate predictions of absolute binding energies. Accordingly, it is important to establish controls, as are common in other fields, to enhance the likelihood of success in spite of these challenges. Here we outline best practices and control docking calculations that help evaluate docking parameters for a given target prior to undertaking a large-scale prospective screen, with exemplification in one particular target, the melatonin receptor, where following this procedure led to direct docking hits with activities in the subnanomolar range. Additional controls are suggested to ensure specific activity for experimentally validated hit compounds. These guidelines should be useful regardless of the docking software used. Docking software described in the outlined protocol (DOCK3.7) is made freely available for academic research to explore new hits for a range of targets.

Ligand Strain Energy in Large Library Docking

While small molecule internal strain is crucial to molecular docking, using it in evaluating ligand scores has remained elusive. Here, we investigate a technique that calculates strain using relative torsional populations in the Cambridge Structural Database, enabling fast precalculation of these energies. In retrospective studies of large docking screens of the dopamine D4 receptor and of AmpC β-lactamase, where close to 600 docking hits were tested experimentally, including such strain energies improved hit rates by preferentially reducing the ranks of strained high-scoring decoy molecules. In a 40-target subset of the DUD-E benchmark, we found two thresholds that usefully distinguished between ligands and decoys: one based on the total strain energy of the small molecules and another based on the maximum strain allowed for any given torsion within them. Using these criteria, about 75% of the benchmark targets had improved enrichment after strain filtering. Relying on precalculated population distributions, this approach is rapid, taking less than 0.04 s to evaluate a conformation on a standard core, making it pragmatic for precalculating strain in even ultralarge libraries. Since it is scoring function agnostic, it may be useful to multiple docking approaches; it is openly available at http://tldr.docking.org.

Promising inhibitors of main protease of novel corona virus to prevent the spread of COVID-19 using docking and molecular dynamics simulation

Coronavirus disease-2019 (COVID-19) is a global health emergency and the matter of serious concern, which has been declared a pandemic by WHO. Till date, no potential medicine/ drug is available to cure the infected persons from SARS-CoV-2. This deadly virus is named as novel 2019-nCoV coronavirus and caused coronavirus disease, that is, COVID-19. The first case of SARS-CoV-2 infection in human was confirmed in the Wuhan city of the China. COVID-19 is an infectious disease and spread from man to man as well as surface to man . In the present work, in silico approach was followed to find potential molecule to control this infection. Authors have screened more than one million molecules available in the ZINC database and taken the best two compounds based on binding energy score. These lead molecules were further studied through docking against the main protease of SARS-CoV-2. Then, molecular dynamics simulations of the main protease with and without screened compounds were performed at room temperature to determine the thermodynamic parameters to understand the inhibition. Further, molecular dynamics simulations at different temperatures were performed to understand the efficiency of the inhibition of the main protease in the presence of the screened compounds. Change in energy for the formation of the complexes between the main protease of novel coronavirus and ZINC20601870 as well ZINC00793735 at room temperature was determined on applying MM-GBSA calculations. Docking and molecular dynamics simulations showed their antiviral potential and may inhibit viral replication experimentally. Communicated by Ramaswamy H. Sarma.

Discovery of the first chemical tools to regulate MKK3-mediated MYC activation in cancer

The transcription master regulator MYC plays an essential role in regulating major cellular programs and is a well-established therapeutic target in cancer. However, MYC targeting for drug discovery is challenging. New therapeutic approaches to control MYC-dependent malignancy are urgently needed. The mitogen-activated protein kinase kinase 3 (MKK3) binds and activates MYC in different cell types, and disruption of MKK3-MYC protein-protein interaction may provide a new strategy to target MYC-driven programs. However, there is no perturbagen available to interrogate and control this signaling arm. In this study, we assessed the drugability of the MKK3-MYC complex and discovered the first chemical tool to regulate MKK3-mediated MYC activation. We have designed a short 44-residue inhibitory peptide and developed a cell lysate-based time-resolved fluorescence resonance energy transfer (TR-FRET) assay to discover the first small molecule MKK3-MYC PPI inhibitor. We have optimized and miniaturized the assay into an ultra-high-throughput screening (uHTS) 1536-well plate format. The pilot screen of ~6,000 compounds of a bioactive chemical library followed by multiple secondary and orthogonal assays revealed a quinoline derivative SGI-1027 as a potent inhibitor of MKK3-MYC PPI. We have shown that SGI-1027 disrupts the MKK3-MYC complex in cells and in vitro and inhibits MYC transcriptional activity in colon and breast cancer cells. In contrast, SGI-1027 does not inhibit MKK3 kinase activity and does not interfere with well-known MKK3-p38 and MYC-MAX complexes. Together, our studies demonstrate the drugability of MKK3-MYC PPI, provide the first chemical tool to interrogate its biological functions, and establish a new uHTS assay to enable future discovery of potent and selective inhibitors to regulate this oncogenic complex.

Rational design of cannabinoid type-1 receptor allosteric modulators: Org27569 and PSNCBAM-1 hybrids

Allosteric modulation offers an alternate approach to target the cannabinoid type-1 receptor (CB₁) for therapeutic benefits. Examination of the two widely studied prototypic CB₁ negative allosteric modulators (NAMs) Org27569 and PSNCBAM-1 revealed structural resemblance and similar structure-activity relationships (SARs). In silico docking and dynamics simulation studies using the crystal structure of CB₁ co-bound with CP55,940 and Org27569 suggested that Org27569 and PSNCBAM-1 occupied the same binding pocket and several common interactions were present in both series with the CB₁ receptor. A new scaffold was therefore designed by merging the key structural features from the two series and the hybrids retained these binding features in the in silico docking studies. In addition, one such hybrid displayed similar functions to Org27569 in dynamic simulations by preserving a key R214^3.50-D338^6.30 salt bridge and maintaining an antagonist-like Helix3-Helix6 interhelical distance. Based on these results, a series of hybrids were synthesized and assessed in calcium mobilization, [³⁵S]GTPγS binding and cAMP assays. Several compounds displayed comparable potencies to Org27569 and PSNCBAM-1 in these assays. This work offers new insight of the SAR requirement at the allosteric site of the CB₁ receptor and provides a new scaffold that can be optimized for the development of future CB₁ allosteric modulators.

Virtual screening of epalrestat mimicking selective ALR2 inhibitors from natural product database: auto pharmacophore, ADMET prediction and molecular dynamics approach

Epalrestat is the only effective aldose reductase (ALR2) inhibitor available in the market for the treatment of diabetic neuropathy. Clinical effectiveness of epalrestat in diabetic neuropathy encouraged us to develop some more ALR2 inhibitors with a better therapeutic profile. Herein, we utilized the pharmacophoric features of epalrestat to search some novel ALR2 inhibitors from an InterBioScreen database of natural compounds. ADME and PAINS filters were applied to provide drug-likeness and to remove toxicophores from the screened hits. The pharmacophoric features of 4-hydroxy-2-nonenal (HNE), a well-known substrate of ALR1, were also explored to identify selective ALR2 inhibitors. The structure-based analysis was then adopted to find out the molecules showing interactions with ALR2 which are crucial for their therapeutic activity. These interaction patterns and binding modes were compared with that of epalrestat. Molecular dynamics (MD) analysis was also carried out to get more insight into the interactions of screened hits in the catalytic domain of ALR2. Additionally, the top hits were docked and simulated with aldehyde reductase (ALR1) to determine their selectivity for ALR2 over ALR1. Overall, five hits including STOCKIN-44771, STOCKIN-46041, STOCKIN-59369, STOCKIN-69620 and STOCKIN-88220 were found to possess a good therapeutic profile in terms of key interactions, binding energies and drug-likeness. Two hits, STOCKIN-46041 and STOCKIN-59369, were identified as the most selective ALR2 inhibitors when assessed their selectivity profile.Communicated by Ramaswamy H. Sarma.

Removal and toxicity evaluation of a diverse group of drugs from water by a cyclodextrin polymer/pulsed light system

The presence of pharmaceutical compounds (PhCs) in the effluents of wastewater treatment plants (WWTPs) is an ecological concern. The issue could be alleviated by trapping those substances by cyclodextrin (CD) polymers or photolyzing them by pulsed light (PL). Consequently, a sequential CD polymer/PL system was tested for the removal of PhCs. Firstly, a survey detected the presence of recurrent PhCs in the effluents of local WWTPs. Then, pure water was spiked with 21 PhCs, 100 μg/L each one. The three-dimensional network provides amphiphilic features to the CD polymer that reduced the pollutant concentration by 77 %. Sorption involves a plead of physical and chemical mechanisms hindering the establishment of a general removal model for all compounds. The performed simulations hint that the retention capacity mainly correlates with the computed binding energies, so that theoretical models are revealed as valuable tools for further improvements. The complementary action of PL rose the elimination to 91 %. The polymer can be reused at least 10 times for ibuprofen (model compound) removal, and was able to eliminate the ecotoxicity of an ibuprofen solution. Therefore, this novel sequential CD polymer/PL process seems to be an efficient alternative to eliminate PhCs from wastewater.

Multibasin Quasi-Harmonic Approach for the Calculation of the Configurational Entropy of Small Molecules in Solution

Entropy is a key thermodynamic property governing most biomolecular processes, including binding. Nonetheless, quantification of the configurational entropy of a single molecule in solution remains a grand challenge. Here, we present an original approach for the calculation of absolute molecular entropies based on the analysis of converged molecular dynamics (MD) simulations. Our method, named quasi-harmonic multibasin (QHMB), relies on a multibasin decomposition of the simulated trajectory by root-mean-square deviation clustering and subsequent quasi-harmonic analysis (QHA) of extracted sub-trajectories. Last, the entropy of the landscape is evaluated using the Gibbs formula. Because of the nature of QHA, this method is directly applicable to explicit-solvent simulations to access configurational entropies in solution. When compared with calorimetric data from NIST, QHMB is shown to predict absolute entropies in the gas phase for 23 small molecules with a root-mean-squared error of 0.36 kcal/mol from the experiments. In addition, the introduction of a QHMB correction in MM/GBSA calculations to account for the ligand configurational entropy loss on binding is shown to improve the correlation between calculated and experimental binding affinities with R² increasing from 0.67 to 0.78. Because this entropy correction penalizes large and flexible ligands more strongly, it might be useful to reduce the false-positive rate in virtual screening. The availability of an automatic procedure to compute QHMB entropies makes it a new available tool in the field of drug discovery.

Identification of novel human USP2 inhibitor and its putative role in treatment of COVID-19 by inhibiting SARS-CoV-2 papain-like (PLpro) protease

Human ubiquitin carboxyl-terminal hydrolase-2 (USP2) inhibitors, such as thiopurine analogs, have been reported to inhibit SARS-CoV papain-like proteases (PLpro). The PLpro have significant functional implications in the innate immune response during SARS-CoV-2 infection and considered an important antiviral target. Both proteases share strikingly similar USP fold with right-handed thumb-palm-fingers structural scaffold and conserved catalytic triad Cys-His-Asp/Asn. In this urgency situation of COVID-19 outbreak, there is a lack of in-vitro facilities readily available to test SARS-CoV-2 inhibitors in whole-cell assays. Therefore, we adopted an alternate route to identify potential USP2 inhibitor through integrated in-silico efforts. After an extensive virtual screening protocol, the best compounds were selected and tested. The compound Z93 showed significant IC50 value against Jurkat (9.67 μM) and MOTL-4 cells (11.8 μM). The binding mode of Z93 was extensively analyzed through molecular docking, followed by MD simulations, and molecular interactions were compared with SARS-CoV-2. The relative binding poses of Z93 fitted well in the binding site of both proteases and showed consensus π-π stacking and H-bond interactions with histidine and aspartate/asparagine residues of the catalytic triad. These results led us to speculate that compound Z93 might be the first potential chemical lead against SARS-CoV-2 PLpro, which warrants in-vitro evaluations.

Targeting the Initiator Protease of the Classical Pathway of Complement Using Fragment-Based Drug Discovery

The initiating protease of the complement classical pathway, C1r, represents an upstream and pathway-specific intervention point for complement-related autoimmune and inflammatory diseases. Yet, C1r-targeted therapeutic development is currently underrepresented relative to other complement targets. In this study, we developed a fragment-based drug discovery approach using surface plasmon resonance (SPR) and molecular modeling to identify and characterize novel C1r-binding small-molecule fragments. SPR was used to screen a 2000-compound fragment library for binding to human C1r. This led to the identification of 24 compounds that bound C1r with equilibrium dissociation constants ranging between 160–1700 µM. Two fragments, termed CMP-1611 and CMP-1696, directly inhibited classical pathway-specific complement activation in a dose-dependent manner. CMP-1611 was selective for classical pathway inhibition, while CMP-1696 also blocked the lectin pathway but not the alternative pathway. Direct binding experiments mapped the CMP-1696 binding site to the serine protease domain of C1r and molecular docking and molecular dynamics studies, combined with C1r autoactivation assays, suggest that CMP-1696 binds within the C1r active site. The group of structurally distinct fragments identified here, along with the structure–activity relationship profiling of two lead fragments, form the basis for future development of novel high-affinity C1r-binding, classical pathway-specific, small-molecule complement inhibitors.

Rigorous Free Energy Simulations in Virtual Screening

Virtual high throughput screening (vHTS) in drug discovery is a powerful approach to identify hits: when applied successfully, it can be much faster and cheaper than experimental high-throughput screening approaches. However, mainstream vHTS tools have significant limitations: ligand-based methods depend on knowledge of existing chemical matter, while structure-based tools such as docking involve significant approximations that limit their accuracy. Recent advances in scientific methods coupled with dramatic speedups in computational processing with GPUs make this an opportune time to consider the role of more rigorous methods that could improve the predictive power of vHTS workflows. In this Perspective, we assert that alchemical binding free energy methods using all-atom molecular dynamics simulations have matured to the point where they can be applied in virtual screening campaigns as a final scoring stage to prioritize the top molecules for experimental testing. Specifically, we propose that alchemical absolute binding free energy (ABFE) calculations offer the most direct and computationally efficient approach within a rigorous statistical thermodynamic framework for computing binding energies of diverse molecules, as is required for virtual screening. ABFE calculations are particularly attractive for drug discovery at this point in time, where the confluence of large-scale genomics data and insights from chemical biology have unveiled a large number of promising disease targets for which no small molecule binders are known, precluding ligand-based approaches, and where traditional docking approaches have foundered to find progressible chemical matter.

Tetrahydroquinoline-Isoxazole/Isoxazoline Hybrid Compounds as Potential Cholinesterases Inhibitors: Synthesis, Enzyme Inhibition Assays, and Molecular Modeling Studies

A series of 44 hybrid compounds that included in their structure tetrahydroquinoline (THQ) and isoxazole/isoxazoline moieties were synthesized through the 1,3-dipolar cycloaddition reaction (1,3-DC) from the corresponding N-allyl/propargyl THQs, previously obtained via cationic Povarov reaction. In vitro cholinergic enzymes inhibition potential of all compounds was tested. Enzyme inhibition assays showed that some hybrids exhibited significant potency to inhibit acetylcholinesterase (AChE) and butyrylcholinesterase (BChE). Especially, the hybrid compound 5n presented the more effective inhibition against AChE (4.24 µM) with an acceptable selectivity index versus BChE (SI: 5.19), while compound 6aa exhibited the greatest inhibition activity on BChE (3.97 µM) and a significant selectivity index against AChE (SI: 0.04). Kinetic studies were carried out for compounds with greater inhibitory activity of cholinesterases. Structure–activity relationships of the molecular hybrids were analyzed, through computational models using a molecular cross-docking algorithm and Molecular Mechanics/Generalized Born Surface Area (MM/GBSA) binding free energy approach, which indicated a good correlation between the experimental inhibition values and the predicted free binding energy.

Comparison of affinity ranking using AutoDock-GPU and MM-GBSA scores for BACE-1 inhibitors in the D3R Grand Challenge 4

Molecular docking has been successfully used in computer-aided molecular design projects for the identification of ligand poses within protein binding sites. However, relying on docking scores to rank different ligands with respect to their experimental affinities might not be sufficient. It is believed that the binding scores calculated using molecular mechanics combined with the Poisson-Boltzman surface area (MM-PBSA) or generalized Born surface area (MM-GBSA) can predict binding affinities more accurately. In this perspective, we decided to take part in Stage 2 of the Drug Design Data Resource (D3R) Grand Challenge 4 (GC4) to compare the performance of a quick scoring function, AutoDock4, to that of MM-GBSA in predicting the binding affinities of a set of [Formula: see text]-Amyloid Cleaving Enzyme 1 (BACE-1) ligands. Our results show that re-scoring docking poses using MM-GBSA did not improve the correlation with experimental affinities. We further did a retrospective analysis of the results and found that our MM-GBSA protocol is sensitive to details in the protein-ligand system: (i) neutral ligands are more adapted to MM-GBSA calculations than charged ligands, (ii) predicted binding affinities depend on the initial conformation of the BACE-1 receptor, (iii) protonating the aspartyl dyad of BACE-1 correctly results in more accurate binding affinity predictions.

AMPA Receptor Noncompetitive Inhibitors Occupy a Promiscuous Binding Site

Noncompetitive inhibitors of AMPA receptors have attracted interest in recent years as antiepileptic drugs. However, their development is hindered by a lack of detailed understanding of the protein-inhibitor interaction mechanisms. Recently, structures of AMPA receptor complexes with the structurally dissimilar, noncompetitive, small-molecule inhibitors pyridone perampanel (PMP), GYKI 53655 (GYKI), and CP 465022 (CP) were resolved, revealing that all three share a common binding site. However, due to the low resolution of the ligands, their exact binding modes and protein-ligand interactions remain ambiguous and insufficiently detailed. We carried out molecular dynamics (MD) simulations on X-ray-resolved and docked AMPA receptor complexes, including thermodynamic integration (TI) to compute ligand binding constants, in order to investigate the inhibitor binding modes in detail and identify key protein-ligand interaction mechanisms. Our analysis and simulations show that the ligand binding pocket at the interface of the receptor's transmembrane domain exhibits features also found in the binding pockets of the multidrug-resistance proteins. The inhibitors bind to such promiscuous pockets by forming multiple weak contacts, while the large, flexible pocket undergoes adjustments to accommodate structurally different ligands in different orientations. TI was able to identify a specific more favorable binding mode for GYKI, while PMP, which has a symmetric ring structure, produced several comparable poses indicating that it may bind in several orientations.

Computational Molecular Modeling of Pin1 Inhibition Activity of Quinazoline, Benzophenone, and Pyrimidine Derivatives

Pin1 (peptidyl-prolyl cis-trans isomerase NIMA-interacting 1) is directly involved in cancer cell-cycle regulation because it catalyses the cis-trans isomerization of prolyl amide bonds in proteins. In this sense, a modeling evaluation of the inhibition of Pin1 using quinazoline, benzophenone, and pyrimidine derivatives was performed by using multilinear, random forest, SMOreg, and IBK regression algorithms on a dataset of 51 molecules, which was divided randomly in 78% for the training and 22% for the test set. Topological descriptors were used as independent variables and the biological activity (pIC50) as a dependent variable. The most robust individual model contained 9 features, and its predictive capability was statistically validated by the correlation coefficient for adjusting, 10-fold cross validation, test set, and bootstrapping with values of 0.910, 0.819, 0.841, and 0.803, respectively. In order to improve the prediction of the pIC50 values, the aggregation of the individual models was performed through the construction of an ensemble, and the most robust one was constructed by two individual models (LR3 and RF1) by applying the IBK algorithm, and a substantial improvement in predictive performance is reflected in the values of R2ADJ = 0.982, Q2CV = 0.962, and Q2EXT = 0.918. Mean square errors <0.165 and good fitting between calculated and experimental pIC50 values suggest a robustness on the prediction of pIC50. Regarding the docking simulation, a binding affinity between the molecules and the active site for the Pin1 inhibition into the protein (3jyj) was estimated through the calculation of the binding free energy (BE), with values in the range of −5.55 to −8.00 kcal/mol, implying a stabilizing interaction molecule receptor. The ligand interaction diagrams between the drugs and amino acid in the binding site for the three most active compounds denoted a good wrapper of these organic compounds into the protein mainly by polar amino acids.

Predicting loop conformational ensembles

Motivation

Protein function is often facilitated by the existence of multiple stable conformations. Structure prediction algorithms need to be able to model these different conformations accurately and produce an ensemble of structures that represent a target's conformational diversity rather than just a single state. Here, we investigate whether current loop prediction algorithms are capable of this. We use the algorithms to predict the structures of loops with multiple experimentally determined conformations, and the structures of loops with only one conformation, and assess their ability to generate and select decoys that are close to any, or all, of the observed structures.

Results

We find that while loops with only one known conformation are predicted well, conformationally diverse loops are modelled poorly, and in most cases the predictions returned by the methods do not resemble any of the known conformers. Our results contradict the often-held assumption that multiple native conformations will be present in the decoy set, making the production of accurate conformational ensembles impossible, and hence indicating that current methodologies are not well suited to prediction of conformationally diverse, often functionally important protein regions.

Contact

marks@stats.ox.ac.uk.

Supplementary information

Supplementary data are available at Bioinformatics online.

Selection of protein conformations for structure-based polypharmacology studies

Several drugs exert their therapeutic effect through the modulation of multiple targets. Structure-based approaches hold great promise for identifying compounds with the desired polypharmacological profiles. These methods use knowledge of the protein binding sites to identify stereoelectronically complementary ligands. The selection of the most suitable protein conformations to be used in the design process is vital, especially for multitarget drug design in which the same ligand has to be accommodated in multiple binding pockets. Herein, we focus on currently available techniques for the selection of the most suitable protein conformations for multitarget drug design, compare the potential advantages and limitations of each method, and comment on how their combination could help in polypharmacology drug design.

Selectivity Challenges in Docking Screens for GPCR Targets and Antitargets

To investigate large library docking's ability to find molecules with joint activity against on-targets and selectivity versus antitargets, the dopamine D2 and serotonin 5-HT2A receptors were targeted, seeking selectivity against the histamine H1 receptor. In a second campaign, κ-opioid receptor ligands were sought with selectivity versus the μ-opioid receptor. While hit rates ranged from 40% to 63% against the on-targets, they were just as good against the antitargets, even though the molecules were selected for their putative lack of binding to the off-targets. Affinities, too, were often as good or better for the off-targets. Even though it was occasionally possible to find selective molecules, such as a mid-nanomolar D2/5-HT2A ligand with 21-fold selectivity versus the H1 receptor, this was the exception. Whereas false-negatives are tolerable in docking screens against on-targets, they are intolerable against antitargets; addressing this problem may demand new strategies in the field.

Blocking Alcoholic Steatosis in Mice with a Peripherally Restricted Purine Antagonist of the Type 1 Cannabinoid Receptor

Type 1 cannabinoid receptor (CB1) antagonists have demonstrated promise for the treatment of obesity, liver disease, metabolic syndrome, and dyslipidemias. However, the inhibition of CB1 receptors in the central nervous system can produce adverse effects, including depression, anxiety, and suicidal ideation. Efforts are now underway to produce peripherally restricted CB1 antagonists to circumvent CNS-associated undesirable effects. In this study, a series of analogues were explored in which the 4-aminopiperidine group of compound 2 was replaced with aryl- and heteroaryl-substituted piperazine groups both with and without a spacer. This resulted in mildly basic, potent antagonists of human CB1 (hCB1). The 2-chlorobenzyl piperazine, 25, was found to be potent ( K_i = 8 nM); to be >1000-fold selective for hCB1 over hCB2; to have no hERG liability; and to possess favorable ADME properties including high oral absorption and negligible CNS penetration. Compound 25 was tested in a mouse model of alcohol-induced liver steatosis and found to be efficacious. Taken together, 25 represents an exciting lead compound for further clinical development or refinement.

Ensemble docking to difficult targets in early-stage drug discovery: Methodology and application to fibroblast growth factor 23

Ensemble docking is now commonly used in early-stage in silico drug discovery and can be used to attack difficult problems such as finding lead compounds which can disrupt protein-protein interactions. We give an example of this methodology here, as applied to fibroblast growth factor 23 (FGF23), a protein hormone that is responsible for regulating phosphate homeostasis. The first small-molecule antagonists of FGF23 were recently discovered by combining ensemble docking with extensive experimental target validation data (Science Signaling, 9, 2016, ra113). Here, we provide a detailed account of how ensemble-based high-throughput virtual screening was used to identify the antagonist compounds discovered in reference (Science Signaling, 9, 2016, ra113). Moreover, we perform further calculations, redocking those antagonist compounds identified in reference (Science Signaling, 9, 2016, ra113) that performed well on drug-likeness filters, to predict possible binding regions. These predicted binding modes are rescored with the molecular mechanics Poisson-Boltzmann surface area (MM/PBSA) approach to calculate the most likely binding site. Our findings suggest that the antagonist compounds antagonize FGF23 through the disruption of protein-protein interactions between FGF23 and fibroblast growth factor receptor (FGFR).

Structural insights into the binding mode of flavonols with the active site of matrix metalloproteinase-9 through molecular docking and molecular dynamic simulations studies

Cartilage degradation in rheumatoid arthritis is mediated principally by the collagenases and gelatinases. Gelatinase B (also called matrix metalloproteinase 9 - MMP-9), is a valid target molecule which is known to participate in cartilage degradation as well as angiogenesis associated with the disease and inhibition of its activity shall prevent cartilage damage and angiogenesis. The focus of this study is to investigate the possibilities of MMP-9 inhibition by flavonol class of bioflavonoids by studying their crucial binding interactions at the active site of MMP 9 using molecular docking (Glide XP and QPLD) and further improvisation by post-docking MM-GBSA and molecular dynamic (MD) simulations. The results show that flavonols can convincingly bind to active site of MMP-9 as demonstrated by their stable interactions at the S1' specificity pocket and favourable binding energies. Gossypin has emerged as a promising candidate with a docking score of -14.618 kcal/mol, binding energy of -79.97 kcal/mol and a stable MD pattern over 15 ns. In addition, interaction mechanisms with respect to catalytic site zinc are also discussed. Further, the drug-like characters of the ligands were also analysed using ADME analysis.

Side Fenestrations Provide an "Anchor" for a Stable Binding of A1899 to the Pore of TASK-1 Potassium Channels

A1899 is a potent and selective inhibitor of the two-pore domain potassium (K_2P) channel TASK-1. It was previously reported that A1899 acts as an open-channel blocker and binds to residues of the P1 and P2 regions, the M2 and M4 segments, and the halothane response element. The recently described crystal structures of K_2P channels together with the newly identified side fenestrations indicate that residues relevant for TASK-1 inhibition are not purely facing the central cavity as initially proposed. Accordingly, the TASK-1 binding site and the mechanism of inhibition might need a re-evaluation. We have used TASK-1 homology models based on recently crystallized K_2P channels and molecular dynamics simulation to demonstrate that the highly potent TASK-1 blocker A1899 requires binding to residues located in the side fenestrations. Unexpectedly, most of the previously described residues that interfere with TASK-1 blockade by A1899 project their side chains toward the fenestration lumina, underlining the relevance of these structures for drug binding in K_2P channels. Despite its hydrophobicity, A1899 does not seem to use the fenestrations to gain access to the central cavity from the lipid bilayer. In contrast, binding of A1899 to residues of the side fenestrations might provide a physical "anchor", reflecting an energetically favorable binding mode that after pore occlusion stabilizes the closed state of the channels.

The Performance of Several Docking Programs at Reproducing Protein-Macrolide-Like Crystal Structures

The accuracy of five docking programs at reproducing crystallographic structures of complexes of 8 macrolides and 12 related macrocyclic structures, all with their corresponding receptors, was evaluated. Self-docking calculations indicated excellent performance in all cases (mean RMSD values ≤ 1.0) and confirmed the speed of AutoDock Vina. Afterwards, the lowest-energy conformer of each molecule and all the conformers lying 0–10 kcal/mol above it (as given by Macrocycle, from MacroModel 10.0) were subjected to standard docking calculations. While each docking method has its own merits, the observed speed of the programs was as follows: Glide 6.6 > AutoDock Vina 1.1.2 > DOCK 6.5 >> AutoDock 4.2.6 > AutoDock 3.0.5. For most of the complexes, the five methods predicted quite correct poses of ligands at the binding sites, but the lower RMSD values for the poses of highest affinity were in the order: Glide 6.6 ≈ AutoDock Vina ≈ DOCK 6.5 > AutoDock 4.2.6 >> AutoDock 3.0.5. By choosing the poses closest to the crystal structure the order was: AutoDock Vina > Glide 6.6 ≈ DOCK 6.5 ≥ AutoDock 4.2.6 >> AutoDock 3.0.5. Re-scoring (AutoDock 4.2.6//AutoDock Vina, Amber Score and MM-GBSA) improved the agreement between the calculated and experimental data. For all intents and purposes, these three methods are equally reliable.

Exploring the stability of ligand binding modes to proteins by molecular dynamics simulations

The binding mode prediction is of great importance to structure-based drug design. The discrimination of various binding poses of ligand generated by docking is a great challenge not only to docking score functions but also to the relatively expensive free energy calculation methods. Here we systematically analyzed the stability of various ligand poses under molecular dynamics (MD) simulation. First, a data set of 120 complexes was built based on the typical physicochemical properties of drug-like ligands. Three potential binding poses (one correct pose and two decoys) were selected for each ligand from self-docking in addition to the experimental pose. Then, five independent MD simulations for each pose were performed with different initial velocities for the statistical analysis. Finally, the stabilities of ligand poses under MD were evaluated and compared with the native one from crystal structure. We found that about 94% of the native poses were maintained stable during the simulations, which suggests that MD simulations are accurate enough to judge most experimental binding poses as stable properly. Interestingly, incorrect decoy poses were maintained much less and 38-44% of decoys could be excluded just by performing equilibrium MD simulations, though 56-62% of decoys were stable. The computationally-heavy binding free energy calculation can be performed only for these survived poses.

A fragment-based docking simulation for investigating peptide–protein bindings

We developed a fragment-based docking strategy for long peptide docking simulations, which separates a long peptide into halves for docking, and then recombined to rebuild whole-peptide docking conformations. With further screening, optimizations and MM/GBSA scoring, our method was capable of efficiently predicting the near-native peptide binding conformations.

Investigating Protein-Peptide Interactions Using the Schrödinger Computational Suite

The Schrödinger software suite contains a broad array of computational chemistry and molecular modeling tools that can be used to study the interaction of peptides with proteins. These include molecular docking using Glide and Piper, relative binding free energy predictions with FEP+, conformational searches using MacroModel and Desmond, and structural refinement using Prime and PrimeX. In this review we provide a comprehensive overview of these tools and describe their potential application in the identification and optimization of peptide ligands for proteins.

Prediction of luciferase inhibitors by the high-performance MIEC-GBDT approach based on interaction energetic patterns

The MIEC-GBDT model can be used as a powerful tool to identify potential interference compounds in luciferase-based high-throughput screening.

Effect of Substitution on the Aniline Moiety of the GPR88 Agonist 2-PCCA: Synthesis, Structure-Activity Relationships, and Molecular Modeling Studies

GPR88, an orphan receptor richly expressed in the striatum, is implicated in a number of basal ganglia-associated disorders. In order to elucidate the functions of GPR88, an in vivo probe appropriate for CNS investigation is required. We previously reported that 2-PCCA was able to modulate GPR88-mediated cAMP production through a Gα_i-coupled pathway. Early structure-activity relationship (SAR) studies suggested that the aniline moiety of 2-PCCA is a suitable site for diverse modifications. Aimed at elucidating structural requirements in this region, we have designed and synthesized a series of analogues bearing a variety of substituents at the phenyl ring of the aniline moiety. Several compounds (e.g., 5j, 5o) showed improved or comparable potency, but have lower lipophilicity than 2-PCCA (clogP 6.19). These compounds provide the basis for further optimization to probe GPR88 in vivo functions. Computational studies confirmed the SAR trends and supported the notion that 4'-substituents on the biphenyl ring exit through a largely hydrophobic binding site to the extracellular loop.