A Comprehensive QSAR Study on Antileishmanial and Antitrypanosomal Cinnamate Ester Analogues

A QSAR Study for Antileishmanial 2-Phenyl-2,3-dihydrobenzofurans †

Leishmaniasis, a parasitic disease that represents a threat to the life of millions of people around the globe, is currently lacking effective treatments. We have previously reported on the antileishmanial activity of a series of synthetic 2-phenyl-2,3-dihydrobenzofurans and some qualitative structure-activity relationships within this set of neolignan analogues. Therefore, in the present study, various quantitative structure-activity relationship (QSAR) models were created to explain and predict the antileishmanial activity of these compounds. Comparing the performance of QSAR models based on molecular descriptors and multiple linear regression, random forest, and support vector regression with models based on 3D molecular structures and their interaction fields (MIFs) with partial least squares regression, it turned out that the latter (i.e., 3D-QSAR models) were clearly superior to the former. MIF analysis for the best-performing and statistically most robust 3D-QSAR model revealed the most important structural features required for antileishmanial activity. Thus, this model can guide decision-making during further development by predicting the activity of potentially new leishmanicidal dihydrobenzofurans before synthesis.

Quantitative Structure-Activity Relationship (QSAR) Study Predicts Small-Molecule Binding to RNA Structure

The diversity of RNA structural elements and their documented role in human diseases make RNA an attractive therapeutic target. However, progress in drug discovery and development has been hindered by challenges in the determination of high-resolution RNA structures and a limited understanding of the parameters that drive RNA recognition by small molecules, including a lack of validated quantitative structure-activity relationships (QSARs). Herein, we develop QSAR models that quantitatively predict both thermodynamic- and kinetic-based binding parameters of small molecules and the HIV-1 transactivation response (TAR) RNA model system. Small molecules bearing diverse scaffolds were screened against TAR using surface plasmon resonance. Multiple linear regression (MLR) combined with feature selection afforded robust models that allowed direct interpretation of the properties critical for both binding strength and kinetic rate constants. These models were validated with new molecules, and their accurate performance was confirmed via comparison to ensemble tree methods, supporting the general applicability of this platform.

Development of Predictive Classification Models for Whole Cell Antimycobacterial Activity of Benzothiazinones

Nitrobenzothiazinones (BTZs) are a very potent class of antibiotics against Mycobacterium tuberculosis. However, relationships between their structural properties and whole cell activity remain poorly predictable. Herein, we present the synthesis and antimycobacterial evaluation of a diverse set of BTZs. High potency was predominantly achieved by piperidine and piperazine substitutions, whereupon three compounds were identified as promising candidates, showing preferable metabolic stability. Lack of correlation between potency and calculated binding energies suggested that target inhibition is not the only requirement to obtain suitable antimycobacterial agents. In contrast, prediction of whole cell activity class was successfully accomplished by extensively validated machine learning models. The performance of the superior model was further verified by >70% correct class predictions for a large set of reported BTZs. Our generated model is thus a key prerequisite to streamline lead optimization endeavors, particularly regarding the improvement of overall hit rates in whole cell antimycobacterial assays.

Optimization of physicochemical properties is a strategy to improve drug-likeness associated with activity: novel active and selective compounds against Trypanosoma cruzi

Trypanosoma cruzi is the causing agent of Chagas disease, a parasitic infection without efficient treatment for chronic patients. Despite the efforts, no new drugs have been approved for this disease in the last 60 years. Molecular modifications based on a natural product led to the development of a series of compounds (LINS03 series) with promising antitrypanosomal activity, however previous chemometric analysis revealed a significant impact of excessive lipophilicity and low aqueous solubility on potency of amine and amide derivatives. Therefore, this work reports different modifications in the core structure to achieve adequate balance of the physicochemical properties along with biological activity. A set of 34 analogues were designed considering predicted properties related to lipophilicity/hydrosolubility and synthesized to assess their activity and selective toxicity towards the parasite. Results showed that this strategy contributed to improve the drug-likeness of the series while considerable impacts on potency were observed. The rational analysis of the obtained data led to the identification of seven active piperazine amides (28-34, IC₅₀ 8.7 to 35.3 µM against intracellular amastigotes), devoid of significant cytotoxicity to mammalian cells. The addition of water-solubilizing groups and privileged substructures such as piperazines improved the physicochemical properties and overall drug-likeness of these compounds, increased potency and maintained selectivity towards the parasite. The obtained results brought important structure-activity relationship (SAR) data and new lead structures for further modifications were identified to achieve improved antitrypanosoma compounds.

A Novel Automated Framework for QSAR Modeling of Highly Imbalanced Leishmania High-Throughput Screening Data

In silico prediction of antileishmanial activity using quantitative structure-activity relationship (QSAR) models has been developed on limited and small datasets. Nowadays, the availability of large and diverse high-throughput screening data provides an opportunity to the scientific community to model this activity from the chemical structure. In this study, we present the first KNIME automated workflow to modeling a large, diverse, and highly imbalanced dataset of compounds with antileishmanial activity. Because the data is strongly biased toward inactive compounds, a novel strategy was implemented based on the selection of different balanced training sets and a further consensus model using single decision trees as the base model and three criteria for output combinations. The decision tree consensus was adopted after comparing its classification performance to consensuses built upon Gaussian-Naı̈ve-Bayes, Support-Vector-Machine, Random-Forest, Gradient-Boost, and Multi-Layer-Perceptron base models. All these consensuses were rigorously validated using internal and external test validation sets and were compared against each other using Friedman and Bonferroni-Dunn statistics. For the retained decision tree-based consensus model, which covers 100% of the chemical space of the dataset and with the lowest consensus level, the overall accuracy statistics for test and external sets were between 71 and 74% and 71 and 76%, respectively, while for a reduced chemical space (21%) and with an incremental consensus level, the accuracy statistics were substantially improved with values for the test and external sets between 86 and 92% and 88 and 92%, respectively. These results highlight the relevance of the consensus model to prioritize a relatively small set of active compounds with high prediction sensitivity using the Incremental Consensus at high level values or to predict as many compounds as possible, lowering the level of Incremental Consensus. Finally, the workflow developed eliminates human bias, improves the procedure reproducibility, and allows other researchers to reproduce our design and use it in their own QSAR problems.

Virtual Screening Based on QSAR and Molecular Docking of Possible Inhibitors Targeting Chagas CYP51

Chagas is a neglected tropical disease caused by the parasite Trypanosoma cruzi with no effective treatment in all its forms. There is a need to find more effective therapeutic alternatives with reduced toxicity. In this contribution, multiple linear regression models were used to identify the molecular descriptors that best describe the inhibitory activity of 52 fenarimol analogues against Trypanosoma cruzi. The topological, physicochemical, thermodynamic, electronic, and charge descriptors were evaluated to cover a wide range of properties that frequently encode biological activity. A model with high predictive value was obtained based on geometrical descriptors and descriptors encoding hydrophobicity and London dispersion forces as necessary for the inhibition of Trypanosoma cruzi-CYP51. Docking methodology was implemented to evaluate molecular interactions in silico. The virtual screening results in this study can be used for rational design of new analogues with improved activity against Chagas disease.

Discovery of new TLR7 agonists by a combination of statistical learning-based QSAR, virtual screening, and molecular dynamics

Search for new antiviral medications has surged in the past two years due to the COVID-19 crisis. Toll-like receptor 7 (TLR7) is among one of the most important TLR proteins of innate immunity that is responsible for broad antiviral response and immune system control. TLR7 agonists, as both vaccine adjuvants and immune response modulators, are among the top drug candidates for not only our contemporary viral pandemic but also other diseases. The agonists of TLR7 have been utilized as vaccine adjuvants and antiviral agents. In this study, we hybridized a statistical learning-based QSAR model with molecular docking and molecular dynamics simulation to extract new antiviral drugs by drug repurposing of the DrugBank database. First, we manually curated a dataset consisting of TLR7 agonists. The molecular descriptors of these compounds were extracted, and feature engineering was done to restrict the number of features to 45. We applied a statistically inspired modification of the partial least squares (SIMPLS) method to build our QSAR model. In the next stage, the DrugBank database was virtually screened structurally using molecular docking, and the top compounds for the guanosine binding site of TLR were identified. The result of molecular docking was again screened by the ligand-based approach of QSAR to eliminate compounds that do not display strong EC₅₀ values by the previously trained model. We then subjected the final results to molecular dynamics simulation and compared our compounds with imiquimod (an FDA-approved TLR7 agonist) and compound 1 (the most active compound against TLR7 in vitro, EC₅₀ = 0.2 nM). Our results evidently demonstrate that cephalosporins and nucleotide analogues (especially acyclic nucleotide analogues such as adefovir and cidofovir) are computationally potent agonists of TLR7. We finally reviewed some publications about cephalosporins that, just like pieces of a puzzle, completed our conclusion.

Alkyl and Aryl Derivatives Based on p-Coumaric Acid Modification and Inhibitory Action against Leishmania braziliensis and Plasmodium falciparum

In low-income populations, neglected diseases are the principal cause of mortality. Of these, leishmaniasis and malaria, being parasitic, protozoan infections, affect millions of people worldwide and are creating a public health problem. The present work evaluates the leishmanicidal and antiplasmodial action of a series of twelve p-coumaric acid derivatives. Of the tested derivatives, eight presented antiparasitic activities 1–3, 8–12. The hexyl p-coumarate derivative (9) (4.14 ± 0.55 μg/mL; selectivity index (SI) = 2.72) showed the highest leishmanicidal potency against the Leishmania braziliensis amastigote form. The results of the molecular docking study suggest that this compound inhibits aldehyde dehydrogenase (ALDH), mitogen-activated kinase protein (MPK4), and DNA topoisomerase 2 (TOP2), all of which are key enzymes in the development of Leishmania braziliensis. The data indicate that these enzymes interact via Van der Waals bonds, hydrophobic interactions, and hydrogen bonds with phenolic and aliphatic parts of this same compound. Of the other compounds analyzed, methyl p-coumarate (64.59 ± 2.89 μg/mL; IS = 0.1) demonstrated bioactivity against Plasmodium falciparum. The study reveals that esters presenting a p-coumarate substructure are promising for use in synthesis of derivatives with good antiparasitic profiles.