Prospective Evaluation of Free Energy Calculations for the Prioritization of Cathepsin L Inhibitors

Discovering novel Cathepsin L inhibitors from natural products using artificial intelligence

Cathepsin L (CTSL) is a promising therapeutic target for metabolic disorders. Current pharmacological interventions targeting CTSL have demonstrated potential in reducing body weight gain, serum insulin levels, and improving glucose tolerance. However, the clinical application of CTSL inhibitors remains limited. In this study, we used a combination of artificial intelligence and experimental methods to identify new CTSL inhibitors from natural products. Through a robust deep learning model and molecular docking, we screened 150 molecules from natural products for experimental validation. At a concentration of 100 µM, we found that 36 of them exhibited more than 50 % inhibition of CTSL. Notably, 13 molecules displayed over 90 % inhibition and exhibiting concentration-dependent effects. The molecular dynamics simulation on the two most potent inhibitors, Plumbagin and Beta-Lapachone, demonstrated stable interaction at the CTSL active site. Enzyme kinetics studies have shown that these inhibitors exert an uncompetitive inhibitory effect on CTSL. In conclusion, our research identifies Plumbagin and Beta-Lapachone as potential CTSL inhibitors, offering promising candidates for the treatment of metabolic disorders and illustrating the effectiveness of artificial intelligence in drug discovery.

Alchemical Transformations and Beyond: Recent Advances and Real-World Applications of Free Energy Calculations in Drug Discovery

Computational methods constitute efficient strategies for screening and optimizing potential drug molecules. A critical factor in this process is the binding affinity between candidate molecules and targets, quantified as binding free energy. Among various estimation methods, alchemical transformation methods stand out for their theoretical rigor. Despite challenges in force field accuracy and sampling efficiency, advancements in algorithms, software, and hardware have increased the application of free energy perturbation (FEP) calculations in the pharmaceutical industry. Here, we review the practical applications of FEP in drug discovery projects since 2018, covering both ligand-centric and residue-centric transformations. We show that relative binding free energy calculations have steadily achieved chemical accuracy in real-world applications. In addition, we discuss alternative physics-based simulation methods and the incorporation of deep learning into free energy calculations.

Molecular Integrative Study on Inhibitory Effects of Pentapeptides on Polymerization and Cell Toxicity of Amyloid-β Peptide (1-42)

Alzheimer's Disease (AD) is a multifaceted neurodegenerative disease predominantly defined by the extracellular accumulation of amyloid-β (Aβ) peptide. In light of this, in the past decade, several clinical approaches have been used aiming at developing peptides for therapeutic use in AD. The use of cationic arginine-rich peptides (CARPs) in targeting protein aggregations has been on the rise. Also, the process of peptide development employing computational approaches has attracted a lot of attention recently. Using a structure database containing pentapeptides made from 20 L-α amino acids, we employed molecular docking to sort pentapeptides that can bind to Aβ42, then performed molecular dynamics (MD) analyses, including analysis of the binding stability, interaction energy, and binding free energy to screen ligands. Transmission electron microscopy (TEM), circular dichroism (CD), thioflavin T (ThT) fluorescence detection of Aβ42 polymerization, MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assay, and the flow cytometry of reactive oxygen species (ROS) were carried out to evaluate the influence of pentapeptides on the aggregation and cell toxicity of Aβ42. Two pentapeptides (TRRRR and ARRGR) were found to have strong effects on inhibiting the aggregation of Aβ42 and reducing the toxicity of Aβ42 secreted by SH-SY5Y cells, including cell death, reactive oxygen species (ROS) production, and apoptosis.

Impact of protein conformations on binding free energy calculations in the beta-secretase 1 system

In binding free energy calculations, simulations must sample all relevant conformations of the system in order to obtain unbiased results. For instance, different ligands can bind to different metastable states of a protein, and if these protein conformational changes are not sampled in relative binding free energy calculations, the contribution of these states to binding is not accounted for and thus calculated binding free energies are inaccurate. In this work, we investigate the impact of different beta-sectretase 1 (BACE1) protein conformations obtained from x-ray crystallography on the binding of BACE1 inhibitors. We highlight how these conformational changes are not adequately sampled in typical molecular dynamics simulations. Furthermore, we show that insufficient sampling of relevant conformations induces substantial error in relative binding free energy calculations, as judged by a variation in calculated relative binding free energies up to 2 kcal/mol depending on the starting protein conformation. These results emphasize the importance of protein conformational sampling and pose this BACE1 system as a challenge case for further method development in the area of enhanced protein conformational sampling, either in combination with binding calculations or as an endpoint correction.

The Open Force Field Initiative: Open Software and Open Science for Molecular Modeling

Force fields are a key component of physics-based molecular modeling, describing the energies and forces in a molecular system as a function of the positions of the atoms and molecules involved. Here, we provide a review and scientific status report on the work of the Open Force Field (OpenFF) Initiative, which focuses on the science, infrastructure and data required to build the next generation of biomolecular force fields. We introduce the OpenFF Initiative and the related OpenFF Consortium, describe its approach to force field development and software, and discuss accomplishments to date as well as future plans. OpenFF releases both software and data under open and permissive licensing agreements to enable rapid application, validation, extension, and modification of its force fields and software tools. We discuss lessons learned to date in this new approach to force field development. We also highlight ways that other force field researchers can get involved, as well as some recent successes of outside researchers taking advantage of OpenFF tools and data.

Identification of Novel Potent NSD2-PWWP1 Ligands Using Structure-Based Design and Computational Approaches

Dysregulation of histone methyl transferase nuclear receptor-binding SET domain 2 (NSD2) has been implicated in several hematological and solid malignancies. NSD2 is a large multidomain protein that carries histone writing and histone reading functions. To date, identifying inhibitors of the enzymatic activity of NSD2 has proven challenging in terms of potency and SET domain selectivity. Inhibition of the NSD2-PWWP1 domain using small molecules has been considered as an alternative approach to reduce NSD2-unregulated activity. In this article, we present novel computational chemistry approaches, encompassing free energy perturbation coupled to machine learning (FEP/ML) models as well as virtual screening (VS) activities, to identify high-affinity NSD2 PWWP1 binders. Through these activities, we have identified the most potent NSD2-PWWP1 binder reported so far in the literature: compound 34 (pIC50 = 8.2). The compounds identified herein represent useful tools for studying the role of PWWP1 domains for inhibition of human NSD2.

Discovery of (R)-4-(8-methoxy-2-methyl-1-(1-phenylethy)-1H-imidazo[4,5-c]quinnolin-7-yl)-3,5-dimethylisoxazole as a potent and selective BET inhibitor for treatment of acute myeloid leukemia (AML) guided by FEP calculation

The functions of the bromodomain and extra terminal (BET) family of proteins have been proved to be involved in various diseases, particularly the acute myeloid leukemia (AML). In this work, guided by free energy perturbation (FEP) calculation, a methyl group was selected to be attached to the 1H-imidazo[4,5-c]quinoline skeleton, and a series of congeneric compounds were synthesized. Among them, compound 10 demonstrated outstanding activity against BRD4 BD1 with an IC50 value of 1.9 nM and exhibited remarkable antiproliferative effects against MV4-11 cells. The X-ray cocrystal structure proved that 10 occupied the acetylated lysine (KAc) binding cavity and the WPF shelf of BRD4 BD1. Additionally, 10 displayed high selectivity towards BET family members, effectively inhibiting the growth of AML cells, promoting apoptosis, and arresting the cell cycle at the G0/G1 phase. Further mechanistic studies demonstrated that compound 10 could suppress the expression of c-Myc and CDK6 while enhancing the expression of P21, PARP, and cleaved PARP. Moreover, 10 exhibited remarkable pharmacokinetic properties and significant antitumor efficacy in vivo. Therefore, compound 10 may represent a new, potent and selective BET bromodomain inhibitor for the development of therapeutics to treat AML.

Targeting Cathepsin L in Cancer Management: Leveraging Machine Learning, Structure-Based Virtual Screening, and Molecular Dynamics Studies

Cathepsin L (CTSL) expression is dysregulated in a variety of cancers. Extensive empirical evidence indicates their direct participation in cancer growth, angiogenic processes, metastatic dissemination, and the development of treatment resistance. Currently, no natural CTSL inhibitors are approved for clinical use. Consequently, the development of novel CTSL inhibition strategies is an urgent necessity. In this study, a combined machine learning (ML) and structure-based virtual screening strategy was employed to identify potential natural CTSL inhibitors. The random forest ML model was trained on IC50 values. The accuracy of the trained model was over 90%. Furthermore, we used this ML model to screen the Biopurify and Targetmol natural compound libraries, yielding 149 hits with prediction scores >0.6. These hits were subsequently selected for virtual screening using a structure-based approach, yielding 13 hits with higher binding affinity compared to the positive control (AZ12878478). Two of these hits, ZINC4097985 and ZINC4098355, have been shown to strongly bind CTSL proteins. In addition to drug-like properties, both compounds demonstrated high affinity, ligand efficiency, and specificity for the CTSL binding pocket. Furthermore, in molecular dynamics simulations spanning 200 ns, these compounds formed stable protein-ligand complexes. ZINC4097985 and ZINC4098355 can be considered promising candidates for CTSL inhibition after experimental validation, with the potential to provide therapeutic benefits in cancer management.

Machine learning models based on residue interaction network for ABCG2 transportable compounds recognition

As the one of the most important protein of placental transport of environmental substances, the identification of ABCG2 transport molecules is the key step for assessing the risk of placental exposure to environmental chemicals. Here, residue interaction network (RIN) was used to explore the difference of ABCG2 binding conformations between transportable and non-transportable compounds. The RIN were treated as a kind of special quantitative data of protein conformation, which not only reflected the changes of single amino acid conformation in protein, but also indicated the changes of distance and action type between amino acids. Based on the quantitative RIN, four machine learning algorithms were applied to establish the classification and recognition model for 1100 compounds with transported by ABCG2 potential. The random forest (RF) models constructed with RIN presented the best and satisfied predictive ability with an accuracy of training set of 0.97 and the test set of 0.96 respectively. In conclusion, the construction of residue interaction network provided a new perspective for the quantitative characterization of protein conformation and the establishment of prediction models for transporter molecular recognition. The ABCG2 transport molecular recognition model based on residue interaction network provides a possible way for screening environmental chemistry transported through placenta.

The maximal and current accuracy of rigorous protein-ligand binding free energy calculations

Computational techniques can speed up the identification of hits and accelerate the development of candidate molecules for drug discovery. Among techniques for predicting relative binding affinities, the most consistently accurate is free energy perturbation (FEP), a class of rigorous physics-based methods. However, uncertainty remains about how accurate FEP is and can ever be. Here, we present what we believe to be the largest publicly available dataset of proteins and congeneric series of small molecules, and assess the accuracy of the leading FEP workflow. To ascertain the limit of achievable accuracy, we also survey the reproducibility of experimental relative affinity measurements. We find a wide variability in experimental accuracy and a correspondence between binding and functional assays. When careful preparation of protein and ligand structures is undertaken, FEP can achieve accuracy comparable to experimental reproducibility. Throughout, we highlight reliable protocols that can help maximize the accuracy of FEP in prospective studies.

Broadening the Scope of Binding Free Energy Calculations Using a Separated Topologies Approach

Binding free energy calculations predict the potency of compounds to protein binding sites in a physically rigorous manner and see broad application in prioritizing the synthesis of novel drug candidates. Relative binding free energy (RBFE) calculations have emerged as an industry-standard approach to achieve highly accurate rank-order predictions of the potency of related compounds; however, this approach requires that the ligands share a common scaffold and a common binding mode, restricting the methods' domain of applicability. This is a critical limitation since complex modifications to the ligands, especially core hopping, are very common in drug design. Absolute binding free energy (ABFE) calculations are an alternate method that can be used for ligands that are not congeneric. However, ABFE suffers from a known problem of long convergence times due to the need to sample additional degrees of freedom within each system, such as sampling rearrangements necessary to open and close the binding site. Here, we report on an alternative method for RBFE, called Separated Topologies (SepTop), which overcomes the issues in both of the aforementioned methods by enabling large scaffold changes between ligands with a convergence time comparable to traditional RBFE. Instead of only mutating atoms that vary between two ligands, this approach performs two absolute free energy calculations at the same time in opposite directions, one for each ligand. Defining the two ligands independently allows the comparison of the binding of diverse ligands without the artificial constraints of identical poses or a suitable atom-atom mapping. This approach also avoids the need to sample the unbound state of the protein, making it more efficient than absolute binding free energy calculations. Here, we introduce an implementation of SepTop. We developed a general and efficient protocol for running SepTop, and we demonstrated the method on four diverse, pharmaceutically relevant systems. We report the performance of the method, as well as our practical insights into the strengths, weaknesses, and challenges of applying this method in an industrial drug design setting. We find that the accuracy of the approach is sufficiently high to rank order ligands with an accuracy comparable to traditional RBFE calculations while maintaining the additional flexibility of SepTop.

Identifying and Overcoming the Sampling Challenges in Relative Binding Free Energy Calculations of a Model Protein:Protein Complex

Relative alchemical binding free energy calculations are routinely used in drug discovery projects to optimize the affinity of small molecules for their drug targets. Alchemical methods can also be used to estimate the impact of amino acid mutations on protein:protein binding affinities, but these calculations can involve sampling challenges due to the complex networks of protein and water interactions frequently present in protein:protein interfaces. We investigate these challenges by extending a graphics processing unit (GPU)-accelerated open-source relative free energy calculation package (Perses) to predict the impact of amino acid mutations on protein:protein binding. Using the well-characterized model system barnase:barstar, we describe analyses for identifying and characterizing sampling problems in protein:protein relative free energy calculations. We find that mutations with sampling problems often involve charge-changes, and inadequate sampling can be attributed to slow degrees of freedom that are mutation-specific. We also explore the accuracy and efficiency of current state-of-the-art approaches─alchemical replica exchange and alchemical replica exchange with solute tempering─for overcoming relevant sampling problems. By employing sufficiently long simulations, we achieve accurate predictions (RMSE 1.61, 95% CI: [1.12, 2.11] kcal/mol), with 86% of estimates within 1 kcal/mol of the experimentally determined relative binding free energies and 100% of predictions correctly classifying the sign of the changes in binding free energies. Ultimately, we provide a model workflow for applying protein mutation free energy calculations to protein:protein complexes, and importantly, catalog the sampling challenges associated with these types of alchemical transformations. Our free open-source package (Perses) is based on OpenMM and is available at https://github.com/choderalab/perses.

Mycobacterium tuberculosis PptT Inhibitors Based on Heterocyclic Replacements of Amidinoureas

4'-Phosphopantetheinyl transferase (PptT) is an essential enzyme for Mycobacterium tuberculosis (Mtb) survival and virulence and therefore an attractive target for a tuberculosis therapeutic. In this work, two modeling-informed approaches toward the isosteric replacement of the amidinourea moiety present in the previously reported PptT inhibitor AU 8918 are reported. Although a designed 3,5-diamino imidazole unexpectedly adopted an undesired tautomeric form and was inactive, replacement of the amidinourea moiety afforded a series of active PptT inhibitors containing 2,6-diaminopyridine scaffolds.

Identifying and overcoming the sampling challenges in relative binding free energy calculations of a model protein:protein complex

Relative alchemical binding free energy calculations are routinely used in drug discovery projects to optimize the affinity of small molecules for their drug targets. Alchemical methods can also be used to estimate the impact of amino acid mutations on protein:protein binding affinities, but these calculations can involve sampling challenges due to the complex networks of protein and water interactions frequently present in protein:protein interfaces. We investigate these challenges by extending a GPU-accelerated open-source relative free energy calculation package (Perses) to predict the impact of amino acid mutations on protein:protein binding. Using the well-characterized model system barnase:barstar, we describe analyses for identifying and characterizing sampling problems in protein:protein relative free energy calculations. We find that mutations with sampling problems often involve charge-changes, and inadequate sampling can be attributed to slow degrees of freedom that are mutation-specific. We also explore the accuracy and efficiency of current state-of-the-art approaches-alchemical replica exchange and alchemical replica exchange with solute tempering-for overcoming relevant sampling problems. By employing sufficiently long simulations, we achieve accurate predictions (RMSE 1.61, 95% CI: [1.12, 2.11] kcal/mol), with 86% of estimates within 1 kcal/mol of the experimentally-determined relative binding free energies and 100% of predictions correctly classifying the sign of the changes in binding free energies. Ultimately, we provide a model workflow for applying protein mutation free energy calculations to protein:protein complexes, and importantly, catalog the sampling challenges associated with these types of alchemical transformations. Our free open-source package (Perses) is based on OpenMM and available at https://github.com/choderalab/perses .

Highly selective preparation of N-terminus Horseradish peroxidase-DNA conjugate with fully retained enzymatic activity: HRP-DNA structure - activity relation

Within the last decade, the field of bio-nanoengineering has achieved significant advances allowing us to generate, e.g., nanoscaled molecular machineries with arbitrary shapes. To unleash the full potential of novel methods such as DNA origami technology, it is important to functionalise complex molecules and nanostructures precisely. Thus, considerable attention has been given to site-selective modifications of proteins allowing further incorporation of various functionalities. Here, we describe a method for the covalent attachment of oligonucleotides to the glycosylated horseradish peroxidase protein (HRP) with high N-terminus selectivity and significant yield while conserving the enzymatic activity. This two-step process includes a pH-controlled metal-free diazotransfer reaction using imidazole-1-sulfonyl azide hydrogen sulfate, which at pH 8.5 results in an N-terminal azide-functionalized protein, followed by the Cu-free click SPAAC reaction to dibenzocyclooctyne- (DBCO) modified oligonucleotides. The reaction conditions were optimised to achieve maximum yield and the best performance. The resulting protein-oligonucleotide conjugates (HRP-DNA) were characterised by electrophoresis and mass spectrometry (MS). Native-PAGE experiments demonstrated different migration patterns for HRP-DNA and the azido-modified protein allowing zymogram experiments. Structure-activity relationships of novel HRP-DNA conjugates were assessed using molecular dynamics simulations, characterising the molecular interactions that define the structural and dynamical properties of the obtained protein-oligonucleotide conjugates (POC).

Omicsynin B4 potently blocks coronavirus infection by inhibiting host proteases cathepsin L and TMPRSS2

The emergence of SARS-CoV-2 variants represents a major threat to public health and requires identification of novel therapeutic agents to address the unmet medical needs. Small molecules impeding viral entry through inhibition of spike protein priming proteases could have potent antiviral effects against SARS-CoV-2 infection. Omicsynin B4, a pseudo-tetrapeptides identified from Streptomyces sp. 1647, has potent antiviral activity against influenza A viruses in our previous study. Here, we found omicsynin B4 exhibited broad-spectrum anti-coronavirus activity against HCoV-229E, HCoV-OC43 and SARS-CoV-2 prototype and its variants in multiple cell lines. Further investigations revealed omicsynin B4 blocked the viral entry and might be related to the inhibition of host proteases. SARS-CoV-2 spike protein mediated pseudovirus assay supported the inhibitory activity on viral entry of omicsynin B4 with a more potent inhibition of Omicron variant, especially when overexpression of human TMPRSS2. Moreover, omicsynin B4 exhibited superior inhibitory activity in the sub-nanomolar range against CTSL, and a sub-micromolar inhibition against TMPRSS2 in biochemical assays. The molecular docking analysis confirmed that omicsynin B4 fits well in the substrate binding sites and forms a covalent bond to Cys25 and Ser441 in CTSL and TMPRSS2, respectively. In conclusion, we found that omicsynin B4 may serve as a natural protease inhibitor for CTSL and TMPRSS2, blocking various coronavirus S protein-driven entry into cells. These results further highlight the potential of omicsynin B4 as an attractive candidate as a broad-spectrum anti-coronavirus agent that could rapidly respond to emerging variants of SARS-CoV-2.

An approach combining deep learning and molecule docking for drug discovery of cathepsin L

OBJECTIVES

Cathepsin L (CTSL) is a promising therapeutic target for metabolic disorders and COVID-19. However, there are still no clinically available CTSL inhibitors. Our objective is to develop an approach for the discovery of potential reversible covalent CTSL inhibitors.

METHODS

The authors combined Chemprop, a deep learning-based strategy, and the Schrödinger CovDock algorithm to identify potential CTSL inhibitors. First, they used Chemprop to train a deep learning model capable of predicting whether a molecule would inhibit the activity of CTSL and performed predictions on ZINC20 in-stock librarie (~9.2 million molecules). Then, they selected the top-200 predicted molecules and performed the Schrödinger covalent docking algorithm to explore the binding patterns to CTSL (PDB: 5MQY). The authors then calculated the binding energies using Prime MM/GBSA and examined the stability between the best two molecules and CTSL using 100ns molecular dynamics simulations.

RESULTS

The authors found five molecules that showed better docking results than the well-known cathepsin inhibitor odanacatib. Notably, two of these molecules, ZINC-35287427 and ZINC-1857528743, showed better docking results with CTSL compared to other cathepsins.

CONCLUSION

Our approach enables drug discovery from large-scale databases with little computational consumption, which will save the cost and time required for drug discovery.

Prioritizing Small Sets of Molecules for Synthesis through in-silico Tools: A Comparison of Common Ranking Methods

Prioritizing molecules for synthesis is a key role of computational methods within medicinal chemistry. Multiple tools exist for ranking molecules, from the cheap and popular molecular docking methods to more computationally expensive molecular-dynamics (MD)-based methods. It is often questioned whether the accuracy of the more rigorous methods justifies the higher computational cost and associated calculation time. Here, we compared the performance on ranking the binding of small molecules for seven scoring functions from five docking programs, one end-point method (MM/GBSA), and two MD-based free energy methods (PMX, FEP+). We investigated 16 pharmaceutically relevant targets with a total of 423 known binders. The performance of docking methods for ligand ranking was strongly system dependent. We observed that MD-based methods predominantly outperformed docking algorithms and MM/GBSA calculations. Based on our results, we recommend the application of MD-based free energy methods for prioritization of molecules for synthesis in lead optimization, whenever feasible.

Pathfinder-Driven Chemical Space Exploration and Multiparameter Optimization in Tandem with Glide/IFD and QSAR-Based Active Learning Approach to Prioritize Design Ideas for FEP+ Calculations of SARS-CoV-2 PLpro Inhibitors

A global pandemic caused by the SARS-CoV-2 virus that started in 2020 and has wreaked havoc on humanity still ravages up until now. As a result, the negative impact of travel restrictions and lockdowns has underscored the importance of our preparedness for future pandemics. The main thrust of this work was based on addressing this need by traversing chemical space to design inhibitors that target the SARS-CoV-2 papain-like protease (PLpro). Pathfinder-based retrosynthesis analysis was used to generate analogs of GRL-0617 using commercially available building blocks by replacing the naphthalene moiety. A total of 10 models were built using active learning QSAR, which achieved good statistical results such as an R2 > 0.70, Q2 > 0.64, STD Dev < 0.30, and RMSE < 0.31, on average for all models. A total of 35 ideas were further prioritized for FEP+ calculations. The FEP+ results revealed that compound 45 was the most active compound in this series with a ΔG of −7.28 ± 0.96 kcal/mol. Compound 5 exhibited a ΔG of −6.78 ± 1.30 kcal/mol. The inactive compounds in this series were compound 91 and compound 23 with a ΔG of −5.74 ± 1.06 and −3.11 ± 1.45 kcal/mol. The combined strategy employed here is envisaged to be of great utility in multiparameter lead optimization efforts, to traverse chemical space, maintaining and/or improving the potency as well as the property space of synthetically aware design ideas.

An overview of the SAMPL8 host-guest binding challenge

The SAMPL series of challenges aim to focus the community on specific modeling challenges, while testing and hopefully driving progress of computational methods to help guide pharmaceutical drug discovery. In this study, we report on the results of the SAMPL8 host–guest blind challenge for predicting absolute binding affinities. SAMPL8 focused on two host–guest datasets, one involving the cucurbituril CB8 (with a series of common drugs of abuse) and another involving two different Gibb deep-cavity cavitands. The latter dataset involved a previously featured deep cavity cavitand (TEMOA) as well as a new variant (TEETOA), both binding to a series of relatively rigid fragment-like guests. Challenge participants employed a reasonably wide variety of methods, though many of these were based on molecular simulations, and predictive accuracy was mixed. As in some previous SAMPL iterations (SAMPL6 and SAMPL7), we found that one approach to achieve greater accuracy was to apply empirical corrections to the binding free energy predictions, taking advantage of prior data on binding to these hosts. Another approach which performed well was a hybrid MD-based approach with reweighting to a force matched QM potential. In the cavitand challenge, an alchemical method using the AMOEBA-polarizable force field achieved the best success with RMSE less than 1 kcal/mol, while another alchemical approach (ATM/GAFF2-AM1BCC/TIP3P/HREM) had RMSE less than 1.75 kcal/mol. The work discussed here also highlights several important lessons; for example, retrospective studies of reference calculations demonstrate the sensitivity of predicted binding free energies to ethyl group sampling and/or guest starting pose, providing guidance to help improve future studies on these systems.

In Silico Positional Analogue Scanning with Amber GPU-TI

Positional analogue scanning (PAS) is an accepted strategy for multiparameter lead optimization (MPO) in drug discovery. Small structural changes as introduced by PAS can lead to 10-fold changes in binding potency in ∼10-20% of cases, a significant parameter shift irrespective of other MPO objectives. Sometimes performing a complete PAS is challenging due to resource and time constraints, building block availability, or difficulty in synthesis. Calculating relative binding free energies (RBFEs) for all positions can contribute to prioritizing the most promising analogues for synthesis. We tested a well-established RBFE calculation method, Amber GPU-TI, for 20 positional analogue scans in 14 test systems (cyclin-dependent kinase 8 (CDK8), hepatitis C virus nonstructural protein 5B (HCV NS5B), tankyrase, RAC-α serine/threonine-protein kinase (Akt), phosphodiesterase 1B (PDE1B), orexin/hypocretin receptor type 1 (OX1R), orexin/hypocretin receptor type 2 (OX2R), histone acetyltransferase K (lysine) acetyltransferase 6A (KAT6A), peroxisome proliferator-activated receptor γ (PPARγ), extracellular signal-regulated kinases (ERK1/2), coactivator-associated arginine methyltransferase 1 (PRMT4), α_vβ₆, bromodomain 1 (BD1), human immunodeficiency virus-1 (HIV-1) entry) involving nitrogen, methyl, halogen, methoxy, and hydroxyl scans with at least four analogues per set. Among the 66 analogue positions explored, we found that in 18 cases Amber GPU-TI calculations predicted a more than 10-fold change in potency. In all of these cases, the experimentally observed direction of potency changes agreed with the predictions. In 16 cases, more than 10-fold changes in experimental potency were observed. Again, in all of these cases, Amber GPU-TI predicted the direction of the potency changes correctly. In none of these cases would a decision made for or against synthesis based on a 10-fold change in potency have resulted in missing an important analogue. Therefore, in silico RBFE calculations using Amber GPU-TI can meaningfully contribute to the prioritization of positional analogues before synthesis.

Insight into Thermodynamic and Kinetic Profiles in Small-Molecule Optimization

Structure-activity relationships (SARs) and structure-property relationships (SPRs) have been considered the most important factors during the drug optimization process. For medicinal chemists, improvements in the potencies and druglike properties of small molecules are regarded as their major goals. Among them, the binding affinity and selectivity of small molecules on their targets are the most important indicators. In recent years, there has been growing interest in using thermodynamic and kinetic profiles to analyze ligand-receptor interactions, which could provide not only binding affinities but also detailed binding parameters for small-molecule optimization. In this perspective, we are trying to provide an insight into thermodynamic and kinetic profiles in small-molecule optimization. Through a highlight of strategies on the small-molecule optimization with specific cases, we aim to put forward the importance of structure-thermodynamic relationships (STRs) and structure-kinetic relationships (SKRs), which could provide more guidance to find safe and effective small-molecule drugs.

Andrographolide induces anti-SARS-CoV-2 response through host-directed mechanism: an in silico study

Aim: Considering the present alarming situation of COVID-19 pandemic, we concentrated on evaluating the efficacy of a novel natural antiviral drug-candidate andrographolide against SARS-CoV-2 through an in silico model of study. Materials & methods: Interaction of andrographolide against the major host molecules that are responsible for SARS-CoV-2 pathogenesis were determined using bio-computational tools, in other words, molecular docking, molecular dynamics simulation and pharmacodynamics–pharmacokinetics analysis. Result: Computational findings represent that andrographolide efficiently interacts with the major human–host-associated putative drug-targets of viral-entry points like furin (-10.54 kcal/mol), TMPRSS-2 (-9.50 kcal/mol), ACE2 (-8.99 kcal/mol) and Cathepsin L (-8.98 kcal/mol). Moreover, it also blocks the inflammatory regulators including TLR4-MD2 and IL-6, which promote virus-induced inflammation leading to cytokine storm in the host body. Conclusion: This work elucidates that, the candidature of andrographolide can be utilized as a potent natural agent for the therapeutic intervention of SARS-CoV-2 through host-directed treatment.

Potential drug discovery for COVID-19 treatment targeting Cathepsin L using a deep learning-based strategy

Cathepsin L (CTSL), a cysteine protease that can cleave and activate the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein, could be a promising therapeutic target for coronavirus disease 2019 (COVID-19). However, there is still no clinically available CTSL inhibitor that can be used. Here, we applied Chemprop, a newly trained directed-message passing deep neural network approach, to identify small molecules and FDA-approved drugs that can block CTSL activity to expand the discovery of CTSL inhibitors for drug development and repurposing for COVID-19. We found 5 molecules (Mg-132, Z-FA-FMK, leupeptin hemisulfate, Mg-101 and calpeptin) that were able to significantly inhibit the activity of CTSL in the nanomolar range and inhibit the infection of both pseudotype and live SARS-CoV-2. Notably, we discovered that daptomycin, an FDA-approved antibiotic, has a prominent CTSL inhibitory effect and can inhibit SARS-CoV-2 pseudovirus infection. Further, molecular docking calculation showed stable and robust binding of these compounds with CTSL. In conclusion, this study suggested for the first time that Chemprop is ideally suited to predict additional inhibitors of enzymes and revealed the noteworthy strategy for screening novel molecules and drugs for the treatment of COVID-19 and other diseases with unmet needs.

Fast and Effective Prediction of the Absolute Binding Free Energies of Covalent Inhibitors of SARS-CoV-2 Main Protease and 20S Proteasome

The COVID-19 pandemic has been a public health emergency with continuously evolving deadly variants around the globe. Among many preventive and therapeutic strategies, the design of covalent inhibitors targeting the main protease (M^pro) of SARS-CoV-2 that causes COVID-19 has been one of the hotly pursued areas. Currently, about 30% of marketed drugs that target enzymes are covalent inhibitors. Such inhibitors have been shown in recent years to have many advantages that counteract past reservation of their potential off-target activities, which can be minimized by modulation of the electrophilic warhead and simultaneous optimization of nearby noncovalent interactions. This process can be greatly accelerated by exploration of binding affinities using computational models, which are not well-established yet due to the requirement of capturing the chemical nature of covalent bond formation. Here, we present a robust computational method for effective prediction of absolute binding free energies (ABFEs) of covalent inhibitors. This is done by integrating the protein dipoles Langevin dipoles method (in the PDLD/S-LRA/β version) with quantum mechanical calculations of the energetics of the reaction of the warhead and its amino acid target, in water. This approach evaluates the combined effects of the covalent and noncovalent contributions. The applicability of the method is illustrated by predicting the ABFEs of covalent inhibitors of SARS-CoV-2 M^pro and the 20S proteasome. Our results are found to be reliable in predicting ABFEs for cases where the warheads are significantly different. This computational protocol might be a powerful tool for designing effective covalent inhibitors especially for SARS-CoV-2 M^pro and for targeted protein degradation.

From computer-aided drug discovery to computer-driven drug discovery

Computational chemistry and structure-based design have traditionally been viewed as a subset of tools that could aid acceleration of the drug discovery process, but were not commonly regarded as a driving force in small molecule drug discovery. In the last decade however, there have been dramatic advances in the field, including (1) development of physics-based computational approaches to accurately predict a broad variety of endpoints from potency to solubility, (2) improvements in artificial intelligence and deep learning methods and (3) dramatic increases in computational power with the advent of GPUs and cloud computing, resulting in the ability to explore and accurately profile vast amounts of drug-like chemical space in silico. There have also been simultaneous advancements in structural biology such as cryogenic electron microscopy (cryo-EM) and computational protein-structure prediction, allowing for access to many more high-resolution 3D structures of novel drug-receptor complexes. The convergence of these breakthroughs has positioned structurally-enabled computational methods to be a driving force behind the discovery of novel small molecule therapeutics. This review will give a broad overview of the synergies in recent advances in the fields of computational chemistry, machine learning and structural biology, in particular in the areas of hit identification, hit-to-lead, and lead optimization.

Mechanism-Based and Computational-Driven Covalent Drug Design

Covalent drugs offer higher efficacy and longer duration of action than their noncovalent counterparts. Significant advances in computational methods for modeling covalent drugs are poised to shift the paradigm of small molecule therapeutics within the next decade. This viewpoint discusses the advantages of a two-state model for ranking reversible and irreversible covalent ligands and of more complex models for dissecting reaction mechanisms. The relation between these models highlights the complexity and diversity of covalent drug binding and provides opportunities for mechanism-based rational design.

Predicting the Relative Binding Affinity for Reversible Covalent Inhibitors by Free Energy Perturbation Calculations

Covalent inhibitors are assuming central importance in drug discovery projects, especially in this pandemic scenario. Many research groups have focused their attention on inhibiting viral proteases or human proteases such as cathepsin L (hCatL). The inhibition of these critical enzymes may impair viral replication. However, molecular modeling of covalent ligands is challenging since covalent and noncovalent ligand-bound states must be considered in the binding process. In this work, we evaluated the suitability of free energy perturbation (FEP) calculations as a tool for predicting the binding affinity of reversible covalent inhibitors of hCatL. Our strategy relies on the relative free energy calculated for both covalent and noncovalent complexes and the free energy changes have been compared with experimental data for eight nitrile-based inhibitors, including three new inhibitors of hCatL. Our results demonstrate that the covalent complex can be employed to properly rank the inhibitors. Nevertheless, a comparison of the free energy changes in both noncovalent and covalent states is valuable to interpret the effect triggered by the formation of the covalent bond on the interactions played by functional groups distant from the warhead. Overall, FEP can be employed as a powerful predictor tool in developing and understanding the activity of reversible covalent inhibitors.

Steps minimize dissipation in rapidly driven stochastic systems

Micro- and nanoscale systems driven by rapid changes in control parameters (control protocols) dissipate significant energy. In the fast-protocol limit, we find that protocols that minimize dissipation at fixed duration are universally given by a two-step process, jumping to and from a point that balances jump size with fast relaxation. Jump protocols could be exploited by molecular machines or thermodynamic computing to improve energetic efficiency, and implemented in nonequilibrium free-energy estimation to improve accuracy.

Optimization of New Catalytic Topoisomerase II Inhibitors as an Anti-Cancer Therapy

Simple Summary

DNA topoisomerase II (TOP2) is a drug target for many types of cancers. However, clinically used TOP2 inhibitors not only kill cancer cells, but also damage normal cells, and can even give rise to other types of cancers. To discover new TOP2 inhibitors to more effectively treat cancer patients, we have applied computer-aided drug design technology to develop several TOP2 inhibitors that can strongly inhibit cancer cell growth but exert low side effects. Results of one exemplary compound are presented in this study. It shows several promising drug-like properties that can be potentially developed into anticancer drugs.

Abstract

Clinically used topoisomerase II (TOP2) inhibitors are poison inhibitors that induce DNA damage to cause cancer cell death. However, they can also destroy benign cells and thereby show serious side effects, including cardiotoxicity and drug-induced secondary malignancy. New TOP2 inhibitors with a different mechanism of action (MOA), such as catalytic TOP2 inhibitors, are needed to more effectively control tumor growth. We have applied computer-aided drug design to develop a new group of small molecule inhibitors that are derivatives of our previously identified lead compound T60. Particularly, the compound T638 has shown improved solubility and microsomal stability. It is a catalytic TOP2 inhibitor that potently suppresses TOP2 activity. T638 has a novel MOA by which it binds TOP2 proteins and blocks TOP2–DNA interaction. T638 strongly inhibits cancer cell growth, but exhibits limited genotoxicity to cells. These results indicate that T638 is a promising drug candidate that warrants further development into clinically used anticancer drugs.

PyAutoFEP: An Automated Free Energy Perturbation Workflow for GROMACS Integrating Enhanced Sampling Methods

Free energy perturbation (FEP) calculations are now routinely used in drug discovery to estimate the relative FEB (RFEB) of small molecules to a biomolecular target of interest. Using enhanced sampling can improve the correlation between predictions and experimental data, especially in systems with conformational changes. Due to the large number of perturbations required in drug discovery campaigns, the manual setup of FEP calculations is no longer viable. Here, we introduce PyAutoFEP, a flexible and open-source tool to aid the setup of RFEB FEP. PyAutoFEP is written in Python3, and automates the generation of perturbation maps, dual topologies, system building and molecular dynamics (MD), and analysis. PyAutoFEP supports multiple force fields, incorporates replica exchange with solute tempering (REST) and replica exchange with solute scaling (REST2) enhanced sampling methods, and allows flexible λ values along perturbation windows. To validate PyAutoFEP, it was applied to a set of 14 Farnesoid X receptor ligands, a system included in the drug design data resource grand challenge 2. An 88% mean correct sign prediction was achieved, and 75% of the predictions had an error below 1.5 kcal/mol. Results using Amber03/GAFF, CHARMM36m/CGenFF, and OPLS-AA/M/LigParGen had Pearson's r values of 0.71 ± 0.13, 0.30 ± 0.27, and 0.66 ± 0.20, respectively. The Amber03/GAFF and OPLS-AA/M/LigParGen results were on par with the top grand challenge 2 submissions. Applying REST2 improved the results using CHARMM36m/CGenFF (Pearson's r = 0.43 ± 0.21) but had little impact on the other force fields. CHARMM36-YF and CHARMM36-WYF modifications did not yield improved predictions compared to CHARMM36m. Finally, we estimated the probability of finding a molecule 1 pK_i better than a lead when using PyAutoFEP to screen 10 or 100 analogues. The probabilities, when compared to random sampling, increased up to sevenfold when 100 molecules were to be screened, suggesting that PyAutoFEP would likely be useful for lead optimization. PyAutoFEP is available on GitHub at https://github.com/lmmpf/PyAutoFEP.

Scaffold Hopping Transformations Using Auxiliary Restraints for Calculating Accurate Relative Binding Free Energies

In silico screening of drug-target interactions is a key part of the drug discovery process. Changes in the drug scaffold via contraction or expansion of rings, the breaking of rings, and the introduction of cyclic structures from acyclic structures are commonly applied by medicinal chemists to improve binding affinity and enhance favorable properties of candidate compounds. These processes, commonly referred to as scaffold hopping, are challenging to model computationally. Although relative binding free energy (RBFE) calculations have shown success in predicting binding affinity changes caused by perturbing R-groups attached to a common scaffold, applications of RBFE calculations to modeling scaffold hopping are relatively limited. Scaffold hopping inevitably involves breaking and forming bond interactions of quadratic functional forms, which is highly challenging. A novel method for handling ring opening/closure/contraction/expansion and linker contraction/expansion is presented here. To the best of our knowledge, RBFE calculations on linker contraction/expansion have not been previously reported. The method uses auxiliary restraints to hold the atoms at the ends of a bond in place during the breaking and forming of the bonds. The broad applicability of the method was demonstrated by examining perturbations involving small-molecule macrocycles and mutations of proline in proteins. High accuracy was obtained using the method for most of the perturbations studied. The rigor of the method was isolated from the force field by validating the method using relative and absolute hydration free energy calculations compared to standard simulation results. Unlike other methods that rely on λ-dependent functional forms for bond interactions, the method presented here can be employed using modern molecular dynamics software without modification of codes or force field functions.

Assessment of select synthetic cannabinoid receptor agonist bias and selectivity between the type 1 and type 2 cannabinoid receptor

The first synthetic cannabinoid receptor agonists (SCRAs) were designed as tool compounds to study the endocannabinoid system's two predominant cannabinoid receptors, CB1R and CB2R. Unfortunately, novel SCRAs now represent the most rapidly proliferating novel psychoactive substances (NPS) of abuse globally. Unlike ∆9-tetrahydrocannabinol, the CB1R and CB2R partial agonist and the intoxicating constituent of Cannabis, many SCRAs characterized to date are full agonists of CB1R. Gaining additional insight into the pharmacological activity of these SCRAs is critical to assess and regulate NPSs as they enter the marketplace. The purpose of this study was to assess select SCRAs recently identified by Canadian police, border service agency, private companies and the illicit market as potential CB1R and CB2R agonists. To this end, fifteen SCRAs were screened for in vitro activity and in silico interactions at CB1R and CB2R. Several SCRAs were identified as being highly biased for cAMP inhibition or βarrestin2 recruitment and receptor subtype selectivity between CB1R and CB2R. The indazole ring and halogen-substituted butyl or pentyl moieties were identified as two structural features that may direct βarrestin2 bias. Two highly-biased SCRAs-JWH-018 2'-napthyl-N-(3-methylbutyl) isomer (biased toward cAMP inhibition) and 4-fluoro MDMB-BINACA (biased toward βarrestin2 recruitment) displayed unique and differential in vivo activity in mice. These data provide initial insight into the correlations between structure, signalling bias, and in vivo activity of the SCRAs.

Machine learning-accelerated quantum mechanics-based atomistic simulations for industrial applications

Atomistic simulations have become an invaluable tool for industrial applications ranging from the optimization of protein-ligand interactions for drug discovery to the design of new materials for energy applications. Here we review recent advances in the use of machine learning (ML) methods for accelerated simulations based on a quantum mechanical (QM) description of the system. We show how recent progress in ML methods has dramatically extended the applicability range of conventional QM-based simulations, allowing to calculate industrially relevant properties with enhanced accuracy, at reduced computational cost, and for length and time scales that would have otherwise not been accessible. We illustrate the benefits of ML-accelerated atomistic simulations for industrial R&D processes by showcasing relevant applications from two very different areas, drug discovery (pharmaceuticals) and energy materials. Writing from the perspective of both a molecular and a materials modeling scientist, this review aims to provide a unified picture of the impact of ML-accelerated atomistic simulations on the pharmaceutical, chemical, and materials industries and gives an outlook on the exciting opportunities that could emerge in the future.

Free energy perturbation in the design of EED ligands as inhibitors of polycomb repressive complex 2 (PRC2) methyltransferase

Free Energy Perturbation (FEP) calculations can provide high-confidence predictions of the interaction strength between a ligand and its protein target. We sought to explore a series of triazolopyrimidines which bind to the EED subunit of the PRC2 complex as potential anticancer therapeutics, using FEP calculations to inform compound design. Combining FEP predictions with a late-stage functionalisation (LSF) inspired synthetic approach allowed us to rapidly evaluate structural modifications in a previously unexplored region of the EED binding site. This approach generated a series of novel triazolopyrimidine EED ligands with improved physicochemical properties and which inhibit PRC2 methyltransferase activity in a cancer-relevant G401 cell line.

The phytochemical hyperforin triggers thermogenesis in adipose tissue via a Dlat-AMPK signaling axis to curb obesity

Stimulation of adipose tissue thermogenesis is regarded as a promising avenue in the treatment of obesity. However, pharmacologic engagement of this process has proven difficult. Using the Connectivity Map (CMap) approach, we identified the phytochemical hyperforin (HPF) as an anti-obesity agent. We found that HPF efficiently promoted thermogenesis by stimulating AMPK and PGC-1α via a Ucp1-dependent pathway. Using LiP-SMap (limited proteolysis-mass spectrometry) combined with a microscale thermophoresis assay and molecular docking analysis, we confirmed dihydrolipoamide S-acetyltransferase (Dlat) as a direct molecular target of HPF. Ablation of Dlat significantly attenuated HPF-mediated adipose tissue browning both in vitro and in vivo. Furthermore, genome-wide association study analysis indicated that a variation in DLAT is significantly associated with obesity in humans. These findings suggest that HPF is a promising lead compound in the pursuit of a pharmacological approach to promote energy expenditure in the treatment of obesity.

Structural Insights into Carboxylic Polyester-Degrading Enzymes and Their Functional Depolymerizing Neighbors

Esters are organic compounds widely represented in cellular structures and metabolism, originated by the condensation of organic acids and alcohols. Esterification reactions are also used by chemical industries for the production of synthetic plastic polymers. Polyester plastics are an increasing source of environmental pollution due to their intrinsic stability and limited recycling efforts. Bioremediation of polyesters based on the use of specific microbial enzymes is an interesting alternative to the current methods for the valorization of used plastics. Microbial esterases are promising catalysts for the biodegradation of polyesters that can be engineered to improve their biochemical properties. In this work, we analyzed the structure-activity relationships in microbial esterases, with special focus on the recently described plastic-degrading enzymes isolated from marine microorganisms and their structural homologs. Our analysis, based on structure-alignment, molecular docking, coevolution of amino acids and surface electrostatics determined the specific characteristics of some polyester hydrolases that could be related with their efficiency in the degradation of aromatic polyesters, such as phthalates.

Anti-MRSA Constituents from Ruta chalepensis (Rutaceae) Grown in Iraq, and In Silico Studies on Two of Most Active Compounds, Chalepensin and 6-Hydroxy-rutin 3',7-Dimethyl ether

Ruta chalepensis L. (Rutaceae), a perennial herb with wild and cultivated habitats, is well known for its traditional uses as an anti-inflammatory, analgesic, antipyretic agent, and in the treatment of rheumatism, nerve diseases, neuralgia, dropsy, convulsions and mental disorders. The antimicrobial activities of the crude extracts from the fruits, leaves, stem and roots of R. chalepensis were initially evaluated against two Gram-positive and two Gram-negative bacterial strains and a strain of the fungus Candida albicans. Phytochemical investigation afforded 19 compounds, including alkaloids, coumarins, flavonoid glycosides, a cinnamic acid derivative and a long-chain alkane. These compounds were tested against a panel of methicillin-resistant Staphylococcus aureus (MRSA) strains, i.e., ATCC 25923, SA-1199B, XU212, MRSA-274819 and EMRSA-15. The MIC values of the active compounds, chalepin (9), chalepensin (10), rutamarin (11), rutin 3′-methyl ether (14), rutin 7,4′-dimethyl ether (15), 6-hydroxy-rutin 3′,7-dimethyl ether (16) and arborinine (18) were in the range of 32–128 µg/mL against the tested MRSA strains. Compounds 10 and 16 were the most active compounds from R. chalepensis, and were active against four out of six tested MRSA strains, and in silico studies were performed on these compounds. The anti-MRSA activity of compound 16 was comparable to that of the positive control norfloxacin (MICs 32 vs 16 μg/mL, respectively) against the MRSA strain XU212, which is a Kuwaiti hospital isolate that possesses the TetK tetracycline efflux pump. This is the first report on the anti-MRSA property of compounds isolated from R. chalepensis and relevant in silico studies on the most active compounds.

A drug repurposing screen identifies hepatitis C antivirals as inhibitors of the SARS-CoV2 main protease

Effective SARS-CoV-2 antiviral drugs are desperately needed. The SARS-CoV-2 main protease (Mpro) appears as an attractive target for drug development. We show that the existing pharmacopeia contains many drugs with potential for therapeutic repurposing as selective and potent inhibitors of SARS-CoV-2 Mpro. We screened a collection of ~6,070 drugs with a previous history of use in humans for compounds that inhibit the activity of Mpro in vitro and found ~50 compounds with activity against Mpro. Subsequent dose validation studies demonstrated 8 dose responsive hits with an IC50 ≤ 50 μM. Hits from our screen are enriched with hepatitis C NS3/4A protease targeting drugs including boceprevir, ciluprevir. narlaprevir, and telaprevir. This work suggests previous large-scale commercial drug development initiatives targeting hepatitis C NS3/4A viral protease should be revisited because some previous lead compounds may be more potent against SARS-CoV-2 Mpro than boceprevir and suitable for rapid repurposing.

Protein–ligand free energies of binding from full-protein DFT calculations: convergence and choice of exchange–correlation functional

Quantum mechanical binding free energies based on thousands of full-protein DFT calculations are tractable, reproducible and converge well.

Skewed thermodynamic geometry and optimal free energy estimation

Free energy differences are a central quantity of interest in physics, chemistry, and biology. We develop design principles that improve the precision and accuracy of free energy estimators, which have potential applications to screening for targeted drug discovery. Specifically, by exploiting the connection between the work statistics of time-reversed protocol pairs, we develop near-equilibrium approximations for moments of the excess work and analyze the dominant contributions to the precision and accuracy of standard nonequilibrium free-energy estimators. Within linear response, minimum-dissipation protocols follow the geodesics of the Riemannian metric induced by the Stokes friction tensor. We find that the next-order contribution arises from the rank-3 supra-Stokes tensor that skews the geometric structure such that minimum-dissipation protocols follow the geodesics of a generalized cubic Finsler metric. Thus, near equilibrium, the supra-Stokes tensor determines the leading-order contribution to the bias of bidirectional free-energy estimators.

Affinity and Selectivity Assessment of Covalent Inhibitors by Free Energy Calculations

Covalent inhibitors have been gaining increased attention in drug discovery due to their beneficial properties such as long residence time, high biochemical efficiency, and specificity. Optimization of covalent inhibitors is a complex task that involves parallel monitoring of the noncovalent recognition elements and the covalent reactivity of the molecules to avoid potential idiosyncratic side effects. This challenge calls for special design protocols, including a variety of computational chemistry methods. Covalent inhibition proceeds through multiple steps, and calculating free energy changes of the subsequent binding events along the overall binding process would help us to better control the design of drug candidates. Inspired by the recent success of free energy calculations on reversible binders, we developed a complex protocol to compute free energies related to the noncovalent and covalent binding steps with thermodynamic integration and hybrid quantum mechanical/molecular mechanical (QM/MM) potential of mean force (PMF) calculations, respectively. In optimization settings, we examined two therapeutically relevant proteins complexed with congeneric sets of irreversible cysteine targeting covalent inhibitors. In the selectivity paradigm, we studied the irreversible binding of covalent inhibitors to phylogenetically close targets by a mutational approach. The results of the calculations are in good agreement with the experimental free energy values derived from the inhibition and kinetic constants (K_i and k_inact) of the enzyme-inhibitor binding. The proposed method might be a powerful tool to predict the potency, selectivity, and binding mechanism of irreversible covalent inhibitors.

Exploring the Mechanism of Covalent Inhibition: Simulating the Binding Free Energy of α-Ketoamide Inhibitors of the Main Protease of SARS-CoV-2

The development of reliable ways of predicting the binding free energies of covalent inhibitors is a challenge for computer-aided drug design. Such development is important, for example, in the fight against the SARS-CoV-2 virus, in which covalent inhibitors can provide a promising tool for blocking Mpro, the main protease of the virus. This work develops a reliable and practical protocol for evaluating the binding free energy of covalent inhibitors. Our protocol presents a major advance over other approaches that do not consider the chemical contribution of the binding free energy. Our strategy combines the empirical valence bond method for evaluating the reaction energy profile and the PDLD/S-LRA/β method for evaluating the noncovalent part of the binding process. This protocol has been used in the calculations of the binding free energy of an α-ketoamide inhibitor of Mpro. Encouragingly, our approach reproduces the observed binding free energy. Our study of covalent inhibitors of cysteine proteases indicates that in the choice of an effective warhead it is crucial to focus on the exothermicity of the point on the free energy surface of a peptide cleavage that connects the acylation and deacylation steps. Overall, we believe that our approach should provide a powerful and effective method for in silico design of covalent drugs.

Water Bridges Play a Key Role in Affinity and Selectivity for Malarial Protease Falcipain-2

Falcipain-2 (FP-2) is hemoglobinase considered an attractive drug target of Plasmodium falciparum. Recently, it has been shown that peptidomimetic nitriles containing a 3-pyridyl (3Pyr) moiety at P2 display high affinity and selectivity for FP-2 with respect to human cysteine cathepsins (hCats), outperforming other P2-Pyr isomers and analogs. Further characterization demonstrated that certain P3 variants of these compounds possess micromolar inhibition of parasite growth in vitro and no cytotoxicity against human cell lines. However, the structural determinants underlying the selectivity of the 3Pyr-containing nitriles for FP-2 remain unknown. In this work, we conduct a thorough computational study combining MD simulations and free energy calculations to decipher the bases of the selectivity of the aforementioned nitriles. Our results reveal that water bridges involving the nitrogen and one carboxyl oxygen of I85 and D234 of FP-2, respectively, and the nitrogen of the neutral 3Pyr moiety, which are either less prevalent or nonexistent in the other complexes, explain the experimental activity profiles. The presence of crystallographic waters close to the bridging water positions in the studied proteases strongly supports the occurrence of such interactions. Overall, our findings suggest that selective FP-2 inhibitors can be designed by promoting water bridge formation at the bottom of the S2 subsite and/or by introducing complementary groups that displace the bridging water.

Automated, Accurate, and Scalable Relative Protein-Ligand Binding Free-Energy Calculations Using Lambda Dynamics

Accurate predictions of changes to protein-ligand binding affinity in response to chemical modifications are of utility in small-molecule lead optimization. Relative free-energy perturbation (FEP) approaches are one of the most widely utilized for this goal but involve significant computational cost, thus limiting their application to small sets of compounds. Lambda dynamics, also rigorously based on the principles of statistical mechanics, provides a more efficient alternative. In this paper, we describe the development of a workflow to set up, execute, and analyze multisite lambda dynamics (MSLD) calculations run on GPUs with CHARMM implemented in BIOVIA Discovery Studio and Pipeline Pilot. The workflow establishes a framework for setting up simulation systems for exploratory screening of modifications to a lead compound, enabling the calculation of relative binding affinities of combinatorial libraries. To validate the workflow, a diverse data set of congeneric ligands for seven proteins with experimental binding affinity data is examined. A protocol to automatically tailor fit biasing potentials iteratively to flatten the free-energy landscape of any MSLD system is developed, which enhances sampling and allows for efficient estimation of free-energy differences. The protocol is first validated on a large number of ligand subsets that model diverse substituents, which shows accurate and reliable performance. The scalability of the workflow is also tested to screen more than 100 ligands modeled in a single system, which also resulted in accurate predictions. With a cumulative sampling time of 150 ns or less, the method results in average unsigned errors of under 1 kcal/mol in most cases for both small and large combinatorial libraries. For the multisite systems examined, the method is estimated to be more than an order of magnitude more efficient than contemporary FEP applications. The results thus demonstrate the utility of the presented MSLD workflow to efficiently screen combinatorial libraries and explore the chemical space around a lead compound and thus are of utility in lead optimization.

In Silico Identification of Potential Natural Product Inhibitors of Human Proteases Key to SARS-CoV-2 Infection

Presently, there are no approved drugs or vaccines to treat COVID-19, which has spread to over 200 countries and at the time of writing was responsible for over 650,000 deaths worldwide. Recent studies have shown that two human proteases, TMPRSS2 and cathepsin L, play a key role in host cell entry of SARS-CoV-2. Importantly, inhibitors of these proteases were shown to block SARS-CoV-2 infection. Here, we perform virtual screening of 14,011 phytochemicals produced by Indian medicinal plants to identify natural product inhibitors of TMPRSS2 and cathepsin L. AutoDock Vina was used to perform molecular docking of phytochemicals against TMPRSS2 and cathepsin L. Potential phytochemical inhibitors were filtered by comparing their docked binding energies with those of known inhibitors of TMPRSS2 and cathepsin L. Further, the ligand binding site residues and non-covalent interactions between protein and ligand were used as an additional filter to identify phytochemical inhibitors that either bind to or form interactions with residues important for the specificity of the target proteases. This led to the identification of 96 inhibitors of TMPRSS2 and 9 inhibitors of cathepsin L among phytochemicals of Indian medicinal plants. Further, we have performed molecular dynamics (MD) simulations to analyze the stability of the protein-ligand complexes for the three top inhibitors of TMPRSS2 namely, qingdainone, edgeworoside C and adlumidine, and of cathepsin L namely, ararobinol, (+)-oxoturkiyenine and 3α,17α-cinchophylline. Interestingly, several herbal sources of identified phytochemical inhibitors have antiviral or anti-inflammatory use in traditional medicine. Further in vitro and in vivo testing is needed before clinical trials of the promising phytochemical inhibitors identified here.

Hybrid Alchemical Free Energy/Machine-Learning Methodology for the Computation of Hydration Free Energies

A methodology that combines alchemical free energy calculations (FEP) with machine learning (ML) has been developed to compute accurate absolute hydration free energies. The hybrid FEP/ML methodology was trained on a subset of the FreeSolv database and retrospectively shown to outperform most submissions from the SAMPL4 competition. Compared to pure machine-learning approaches, FEP/ML yields more precise estimates of free energies of hydration and requires a fraction of the training set size to outperform standalone FEP calculations. The ML-derived correction terms are further shown to be transferable to a range of related FEP simulation protocols. The approach may be used to inexpensively improve the accuracy of FEP calculations and to flag molecules which will benefit the most from bespoke force field parametrization efforts.