The main protease enzyme (M^pro) of SARS-CoV-2 is one of the most promising targets for COVID-19 treatment. Accordingly, in this work, a structure-based virtual screening of 3.8 million ligand libraries was carried out. After rigorous filtering, docking, and post screening assessments, 78 compounds were selected for biological evaluation, 3 of which showed promising inhibition of the M^pro enzyme. The obtained hits (CB03, GR04, and GR20) had reasonable potencies with K_i values in the medium to high micromolar range. Interestingly, while our most potent hit, GR20, was suggested to act via a reversible covalent mechanism, GR04 was confirmed as a noncompetitive inhibitor that seems to be one of a kind when compared to the other allosteric inhibitors discovered so far. Moreover, all three compounds have small sizes (~300 Da) with interesting fittings in their relevant binding sites, and they possess lead-like characteristics that can introduce them as very attractive candidates for the future development of COVID-19 treatments.

The need for new antibiotics has become a major worldwide challenge as bacterial strains keep developing resistance to the existing drugs at an alarming rate. Enoyl-acyl carrier protein reductases (FabI) play a crucial role in lipids and fatty acid biosynthesis, which are essential for the integrity of the bacterial cell membrane. Our study aimed to discover small FabI inhibitors in continuation to our previously found hit MN02. The process was initially started by conducting a similarity search to the NCI ligand database using MN02 as a query. Accordingly, ten compounds were chosen for the computational assessment and antimicrobial testing. Most of the compounds showed an antibacterial activity against Gram-positive strains, while RK10 exhibited broad-spectrum activity against both Gram-positive and Gram-negative bacteria. All tested compounds were then docked into the saFabI active site followed by 100 ns MD simulations (Molecular Dynamics) and MM-GBSA (Molecular Mechanics with Generalised Born and Surface Area Solvation) calculations in order to understand their fitting and estimate their binding energies. Interestingly, and in line with the experimental data, RK10 was able to exhibit the best fitting with the target catalytic pocket. To sum up, RK10 is a small compound with leadlike characteristics that can indeed act as a promising candidate for the future development of broad-spectrum antibacterial agents.

ERK5, a member of the MAP kinase family, has been implicated in several cancer types due to its role in regulating cell proliferation, survival, and migration. In this study, structure-based virtual screening was employed, followed by cell assays, and molecular dynamics simulations to identify novel ERK5 inhibitors. A commercially available library of 1.6 million compounds was subjected to a three-stage docking process (HTVS, SP, and XP), using the docking module in Schrodinger Maestro, yielding 40 candidates with superior docking scores compared to the co-crystallized ligand. These compounds were then tested for antiproliferative activity using an MTT assay in A549 and H292 lung cancer cell lines. Among the hits, STK038175, STK300222, and GR04 showed significant activity with IC₅₀ values of ranging from 10 to 25 µM. Western blot analysis revealed that STK300222 at 50 µM reduced the phosphorylation of ERK5 downstream targets similarly to a known inhibitor, while wound healing assays confirmed a dose-dependent decrease in cell migration. Molecular dynamics simulations of 200 ns further demonstrated that all three compounds form stable complexes with ERK5 that are comparable to the co-crystallized ligand in 5BYZ. The MD simulations also revealed strong electrostatic and solvation interactions observed for STK300222 and GR04 particularly. Furthermore, by calculating the MM-GB/SA scores from the MD trajectories, the binding affinities of the three hits, along with the co-crystallized ligand in 5BYZ, were re-scored. Although the co-crystallized ligand had the highest MM-GB/SA score at −38.96 Kcal mol⁻¹, STK300222 had a comparable score of −35.45 Kcal mol⁻¹. These results highlight STK300222 and GR04 as promising candidates for further optimization and in vivo validation as ERK5 inhibitors.

The papain-like protease (PLᵖʳᵒ) plays a critical role in the SARS-CoV-2 life cycle by regulating both viral replication and immune evasion, making it an attractive target for antiviral drug development. In this study, a virtual screening strategy combining pharmacophore modeling, molecular docking, enzymatic evaluation, and molecular dynamics (MD) simulations was employed to identify novel PLᵖʳᵒ inhibitors. From this process, 30 top ranked candidates were selected for experimental enzymatic evaluation. Among them, a quinolone-based scaffold compound ST50940260 demonstrated the highest inhibitory activity with an IC₅₀ of 69.53 ± 4.4 µM, exhibiting favourable potency and lead like properties, with a docking score of −10.215 kcal/mol. Notably, ST50940260 also showed inhibitory activity against Mᵖʳᵒ, suggesting dual target potential with an IC₅₀ of 96.7 µM. To further explore structure activity relationships, a 2D similarity search was conducted, resulting in the identification and testing of seven structural derivatives, among which Z30231927 emerged as the most promising analogue with an IC₅₀ of 124.8 µM against PLᵖʳᵒ. Molecular dynamics simulations of 100 ns revealed that the lead compound forms a stable complex with PLᵖʳᵒ, exhibiting interaction patterns comparable to those of the co-crystallized ligand. Furthermore, MM-GB/SA binding free energies were calculated from the MD trajectories for the lead (ST50940260), the derivative (Z30231927), and the co-crystallized ligand in 7JN2. The lead demonstrated favourable binding (−27.33 kcal/mol) within a comparable range, while the derivative showed the most favourable binding (−30.26 kcal/mol), followed closely by the co-crystallized reference (−29.82 kcal/mol). Overall, these findings highlight ST50940260 as a promising lead candidate for further optimization, with potential dual inhibitory activity against both proteases.

Pharmacological inhibition of the nucleoside transporter hENT1 is a promising therapeutic target across a range of diseases, including cardiovascular disorders, neurodegenerative conditions, and cancer. However, current inhibitors lack drug-like properties, necessitating the development of new inhibitors with improved pharmacological profiles. We employed a dual-pharmacophore virtual screening protocol to identify putative hENT1 inhibitors from a library of over 2 million compounds, followed by structure-based molecular docking. To validate the inhibition effect of the lead compounds, we established a functional assay using gemcitabine (GEM)-induced cytotoxicity as a readout of hENT transport activity using eight cancer cell lines. H292 was the optimal cancer cell line for the validation assay based on its high GEM sensitivity (IC50 = 28 nM) and the concentration-dependent cytotoxicity inhibition of the reference inhibitor NBTI, a hENT1 inhibitor. Of the 19 candidate compounds, two leads (compounds 2 and 3) demonstrated potency comparable to NBTI, increasing GEM IC50 values by 2.2- and 2.9-fold at 5 µM, respectively. Both compounds were non-cytotoxic to normal fibroblasts, exhibited favorable ADME properties, displayed superior docking scores of −12.63 and −12.49 kcal/mol compared to NBTI (−9.06 kcal/mol), and displayed a novel vertical binding orientation within the hENT1 binding pocket distinct from NBTI’s horizontal mode. This study established a validated non-radioactive, gemcitabine-based functional assay for hENT inhibitor discovery and identified two putative inhibitors with therapeutic potential for cancer chemosensitization, pain management, and cardio- and neuroprotection. The non-radioactive functional assay overcomes the limitations of traditional radiolabeled methods, enabling scalable, broader screening applications.