Tamoxifen is the most frequently used anti-estrogen adjuvant treatment for estrogen receptor-positive breast cancer. However, it is associated with an increased risk of several serious side–effects, such as uterine cancer, stroke, and pulmonary embolism. The 2′,4′-dihydroxy-6-methoxy-3,5-dimethylchalcone (ChalcEA) from plant leaves of Eugenia aquea, has been found to inhibit the proliferation of MCF-7 human breast cancer cells in a dose-dependent manner, with an IC₅₀ of 74.5 μg/mL (250 μM). The aim of this work was to study the molecular interactions of new ChalcEA derivatives formed with the Estrogen Receptor α (ERα) using computer aided drug design approaches. Molecular docking using Autodock 4.2 was employed to explore the modes of binding of ChalcEA derivatives with ERα. The 3D structure-based pharmacophore model was derived using LigandScout 4.1 Advanced to investigate the important chemical interactions of the ERα-tamoxifen complex structure. The binding energy and the tamoxifen-pharmacophore fit score of the best ChalcEA derivative (HNS10) were −12.33 kcal/mol and 67.07 kcal/mol, respectively. The HNS10 interacted with Leu346, Thr347, Leu349, Ala350, Glu353, Leu387, Met388, Leu391, Arg394, Met421, and Leu525. These results suggest that the new ChalcEA derivatives could serve as the lead compound for potent ERα inhibitor in the fight against breast cancer.

The accurate prediction of molecular properties, such as lipophilicity and aqueous solubility, are of great importance and pose challenges in several stages of the drug discovery pipeline. Machine learning methods, such as graph-based neural networks (GNNs), have shown exceptionally good performance in predicting these properties. In this work, we introduce a novel GNN architecture, called directed edge graph isomorphism network (D-GIN). It is composed of two distinct sub-architectures (D-MPNN, GIN) and achieves an improvement in accuracy over its sub-architectures employing various learning, and featurization strategies. We argue that combining models with different key aspects help make graph neural networks deeper and simultaneously increase their predictive power. Furthermore, we address current limitations in assessment of deep-learning models, namely, comparison of single training run performance metrics, and offer a more robust solution.

Hsp90 C-terminal domain (CTD) inhibitors are promising novel agents for cancer treatment, as they do not induce the heat shock response associated with Hsp90 N-terminal inhibitors. One challenge associated with CTD inhibitors is the lack of a co-crystallized complex, requiring the use of predicted allosteric apo pocket, limiting structure-based (SB) design approaches. To address this, a unique approach that enables the derivation and analysis of interactions between ligands and proteins from molecular dynamics (MD) trajectories was used to derive pharmacophore models for virtual screening (VS) and identify suitable binding sites for SB design. Furthermore, ligand-based (LB) pharmacophores were developed using a set of CTD inhibitors to compare VS performance with the MD derived models. Virtual hits identified by VS with both SB and LB models were tested for antiproliferative activity. Compounds 9 and 11 displayed antiproliferative activities in MCF-7 and Hep G2 cancer cell lines. Compound 11 inhibited Hsp90-dependent refolding of denatured luciferase and induced the degradation of Hsp90 clients without the concomitant induction of Hsp70 levels. Furthermore, compound 11 offers a unique scaffold that is promising for the further synthetic optimization and development of molecules needed for the evaluation of the Hsp90 CTD as a target for the development of anticancer drugs.

The irreversible inhibitors of monoamine oxidases (MAO) slow neurotransmitter metabolism in depression and neurodegenerative diseases. After oxidation by MAO, hydrazines, cyclopropylamines and propargylamines form a covalent adduct with the flavin cofactor. To assist the design of new compounds to combat neurodegeneration, we have updated the kinetic parameters defining the interaction of these established drugs with human MAO-A and MAO-B and analyzed the required features. The K_i values for binding to MAO-A and molecular models show that selectivity is determined by the initial reversible binding. Common to all the irreversible inhibitor classes, the non-covalent 3D-chemical interactions depend on a H-bond donor and hydrophobic-aromatic features within 5.7 angstroms apart and an ionizable amine. Increasing hydrophobic interactions with the aromatic cage through aryl halogenation is important for stabilizing ligands in the binding site for transformation. Good and poor inactivators were investigated using visible spectroscopy and molecular dynamics. The initial binding, close and correctly oriented to the FAD, is important for the oxidation, specifically at the carbon adjacent to the propargyl group. The molecular dynamics study also provides evidence that retention of the allenyl imine product oriented towards FADH⁻ influences the formation of the covalent adduct essential for effective inactivation of MAO.

DNA gyrase is an important target for the development of novel antibiotics. Although ATP-competitive DNA gyrase (GyrB) inhibitors are a well-studied class of antibacterial agents, there is currently no representative used in therapy, largely due to unwanted off-target activities. Selectivity of GyrB inhibitors against closely related human ATP-binding enzymes should be evaluated early in development to avoid off-target binding to homologous binding domains. To address this challenge, we developed selective 3D-pharmacophore models for GyrB, human topoisomerase IIα (TopoII), and the Hsp90 N-terminal domain (NTD) to be used in in silico activity profiling paradigms to identify molecules selective for GyrB over TopoII and Hsp90, as starting points for hit expansion and lead optimization. The models were used to profile highly active GyrB, TopoII, and Hsp90 inhibitors. Selected compounds were tested in in vitro assays. GyrB inhibitors 1 and 2 were inactive against TopoII and Hsp90, while 3 and 4, potent Hsp90 inhibitors, displayed no inhibition of GyrB and TopoII, and TopoII inhibitors 5 and 6 were inactive at GyrB and Hsp90. The results provide a proof of concept for the use of target activity profiling methods to identify selective starting points for hit and lead identification.

(1) Background: SARS-CoV-2 RNA-dependent RNA polymerase (RdRp) is a promising therapeutic target to fight COVID-19, and many RdRp inhibitors nucleotide/nucleoside analogs, such as remdesivir, have been identified or are in clinical studies. However, the appearance of resistant mutations could reduce their efficacy. In the present work, we structurally evaluated the impact of RdRp mutations found at baseline in 39 patients treated with remdesivir and associated with a different degree of antiviral response in vivo. (2) Methods: A refined bioinformatics approach was applied to assign SARS-CoV-2 clade and lineage, and to define RdRp mutational profiles. In line with such a method, the same mutations were built and analyzed by combining docking and thermodynamics evaluations with both molecular dynamics and representative pharmacophore models. (3) Results: Clinical studies revealed that patients bearing the most prevalent triple mutant P323L+671S+M899I, which was present in 41% of patients, or the more complex mutational profile P323L+G671S+L838I+D738Y+K91E, which was found with a prevalence of 2.6%, showed a delayed reduced response to remdesivir, as confirmed by the increase in SARS-CoV-2 viral load and by a reduced theoretical binding affinity versus RdRp (ΔGbind_WT = −122.70 kcal/mol; ΔGbind_{P323L+671S+M899I} = −84.78 kcal/mol; ΔGbind_{P323L+G671S+L838I+D738Y+K91E} = −96.74 kcal/mol). Combined computational approaches helped to rationalize such clinical observations, offering a mechanistic understanding of the allosteric effects of mutants on the global motions of the viral RNA synthesis machine and in the changes of the interactions patterns of remdesivir during its binding.