In silico study identifies two benzofuran candidates for selective MAO-B inhibition

A multilayered computational study of 37 previously synthesized benzofuran derivatives identified compounds 10 and 16 as the strongest candidates for further investigation as selective monoamine oxidase B inhibitors. The researchers combined molecular docking, induced-fit docking, structural inspection, molecular mechanics/generalized Born surface area calculations, re-docking, 200-nanosecond molecular dynamics simulations, density functional theory calculations and in silico pharmacokinetic predictions. These findings are model-based and have not been validated in laboratory, animal or human studies.

Monoamine oxidase B is a flavin adenine dinucleotide-containing enzyme on the outer mitochondrial membrane that oxidizes dopamine and other arylalkylamine neurotransmitters. Its larger bipartite active site contains an entrance cavity and a substrate cavity separated by Ile199, while Tyr398 and Tyr435 form an aromatic cage in front of the FAD cofactor. The corresponding MAO-A site is a smaller, single hydrophobic cavity. Differences in cavity architecture and residues—including replacement of MAO-B Ile199 and Tyr326 by Phe208 and Ile335 in MAO-A—provide a structural basis for investigating isoform selectivity.

Initial Glide extra-precision docking produced valid MAO-B poses for 13 of the 37 derivatives. Compound 33 obtained the best XP score, −11.31 kcal/mol, followed by compound 16 at −10.14 kcal/mol; the reference drug safinamide scored −15.25 kcal/mol. Only compounds 12, 11 and 3 also produced valid poses in MAO-A, so the researchers did not carry them forward as candidate selective inhibitors. Thirteen compounds remained after this first exclusion step, rather than two candidates being identified directly by docking alone.

The investigators then used induced-fit docking and visual inspection to evaluate whether predicted poses occupied the MAO-B cavity appropriately and contacted the aromatic cage, FAD and other key residues. Compounds 2 and 8 had unsuitable orientations, while compounds 33 and 12 lacked terminal aromatic groups capable of occupying the cage. Compound 3 positioned its 5-bromobenzofuran fragment 9.44 Å from the FAD N5 atom. Compounds 13 and 37 also failed one or more structural criteria, and compounds 11 and 12 had been excluded for their MAO-A poses. Five derivatives—1, 5, 10, 14 and 16—passed the structural filters and proceeded to MM/GBSA rescoring and re-docking with GOLD 5.3.

MM/GBSA calculations ranked compound 14 as having the most favorable estimated binding energy at −76.52 kcal/mol, followed by compound 1 at −72.87 kcal/mol. The final GOLD ranking differed: compound 10 achieved the highest ChemPLP fitness score at 161.60, compound 16 followed at 159.45, and safinamide scored 165.77. Compounds 10 and 16 were therefore selected for molecular dynamics and density functional theory analyses because they were the only compounds that ranked highly under two scoring approaches after passing the structural filters.

The predicted binding modes involved contacts with the features that distinguish MAO-B. For compound 10, the N,N-diethylamino group faced the FAD cofactor, its oxadiazole ring formed a predicted π–π interaction with Tyr326 and a π–sulfur interaction with Cys172, and it contacted Ile198, Ile199 and Tyr398. Compound 16 placed its 2-chlorophenyl group near the aromatic cage, with predicted π–π interactions involving Tyr398, Tyr435 and FAD, plus contacts with Leu171, Ile199 and Ile316. These are modeled interactions rather than measurements of inhibition or selectivity.

In 200-nanosecond GROMACS simulations at 310 K, compound 10 maintained a mean ligand root-mean-square deviation of approximately 0.3 nm after fitting to the protein, consistent with retention in the active site. It predominantly formed two or three protein–ligand hydrogen bonds. Compound 16 fluctuated more extensively during the first 60 ns and continued to show broader variation, although its trajectory converged with compound 10’s after approximately 60 ns and the ligand remained in the active site. Compound 16 formed mostly one or two hydrogen bonds, with intermittent higher counts later in the trajectory.

Density functional theory calculations at the B3LYP/6-311++G(d,p) level produced similar HOMO–LUMO energy gaps for the compounds: 4.473 eV for compound 10 and 4.462 eV for compound 16. The study interpreted these values as indicating similar chemical stability and low predicted electron-transfer propensity, consistent with the non-covalent binding modes modeled by docking and simulation. Frontier orbitals were concentrated mainly around the benzofuran and oxadiazole regions in both compounds, while their terminal substituents differed in orientation and interaction pattern.

SwissADME predictions placed all six principal physicochemical descriptors for compound 10 within the tool’s optimal ranges. Compound 16 met the ranges for lipophilicity, molecular size, polarity, solubility and flexibility but had a fraction of sp3-hybridized carbons of 0.06, below the cited threshold of 0.25; its value was 0.31 for compound 10. Overall, the authors nominated compound 10, and to a lesser extent compound 16, as benzofuran-based candidates for in vitro and in vivo evaluation. The computational workflow provides hypotheses for further testing but does not establish inhibitory potency, isoform selectivity, safety or clinical benefit.

Subject of Research: Computational screening of benzofuran derivatives as potential selective MAO-B inhibitors

Article Title: Exploring benzofuran derivatives as selective MAO-B inhibitors: a multilayered in silico study

Article References: Mateev, E., Irfan, A., Kostov, S., Muhammed, M. T., & Zlatkov, A. (2026). Exploring benzofuran derivatives as selective MAO-B inhibitors: a multilayered in silico study. Pharmacia, 73, Article e206275. https://doi.org/10.3897/pharmacia.73.e206275

Image Credits: AI Generated

DOI: 10.3897/pharmacia.73.e206275

Keywords: MAO-B inhibitors, benzofuran derivatives, in silico screening, molecular docking, molecular dynamics, density functional theory, drug discovery, ADME prediction, Exploring, benzofuran, derivatives, selective