Exploring natural products as multi-target-directed drugs for Parkinson’s disease: an in-silico approach integrating QSAR, pharmacophore modeling, and molecular dynamics simulations

Abstract

Parkinson’s disease is a neurodegenerative disorder characterized by the progressive loss of dopaminergic neurons in the midbrain. Current treatments provide limited symptomatic relief without halting disease progression. A multi-targeting approach has shown potential benefits in treating neurodegenerative diseases. In this study, we employed in silico approaches to explore the COCONUT natural products database and identify novel drug candidates with multi-target potential against relevant Parkinson’s disease targets. QSAR models were developed to screen for potential bioactive molecules, followed by a hybrid virtual screening approach involving pharmacophore modeling and molecular docking against MAO-B, AA2AR, and NMDAR. ADME evaluation was performed to assess drug-like properties. Our findings revealed 22 candidates that exhibited the desired pharmacophoric features. Particularly, two compounds: CNP0121426 and CNP0242698 exhibited remarkable binding affinities, with energies lower than −10 kcal/mol and promising interaction profiles with the chosen targets. Furthermore, all the ligands displayed desirable pharmacokinetic properties for brain-targeted drugs. Lastly, molecular dynamics simulations were conducted on the lead candidates, belonging to the dihydrochalcone and curcuminoid class, to evaluate their stability over a 100 ns timeframe and compare their dynamics with reference complexes. Our findings revealed the curcuminoid CNP0242698 to have an overall better stability with the three targets compared to the dihydrochalcone, despite the high ligand RMSD, the curcuminoid CNP0242698 showed better protein stability, implying ligand exploration of different orientations. Similarly, AA2AR exhibited higher stability with CNP0242698 compared to the reference complex, despite the high initial ligand RMSD due to the bulkier active site. In NMDAR, CNP0242698 displayed good stability and less fluctuations implying a more restricted conformation within the smaller active site of NMDAR. These results may serve as lead compounds for the development and optimization of natural products as multi-target disease-modifying natural remedies for Parkinson’s disease patients. However, experimental assays remain necessary to validate these findings. Communicated by Ramaswamy H. Sarma.