Integrated Structure- and Ligand-Based Pharmacoinformatic Protocol for the Modulation of Monoamine Oxidase B (MAO-B) in Parkinson’s disease

Abstract

Oxidative stress is a phenomenon associated with various neurodegenerative diseases. Death of dopaminergic neurons in the substantia nigra is a major implication of Parkinson disease which occurs as a consequence of oxidative stress owing to high free radical production. In our work, stochastic petri net simulations were performed to probe the mechanistic of dopamine metabolism in oxidative stress by reactive oxygen species (ROS). Our results depicted the involvement of higher monoamine oxidase-B levels in producing oxidative stress in Parkinson’s disease. Moreover, the effect of MAO-B inhibitors was also significant on lowering ROS production. However, inhibition of MAO-A which shares 70% sequence similarity with MAO-B, can lead to tyramine induced toxicity. Therefore, in our project we adopted an integrated structure and ligand based pharmacoinformatic protocol to explore the binding hypothesis of MAO-A and MAO-B. Our protocol includes molecular docking, physiochemical and 3D descriptors guided 2D/3D QSAR and pharmacophore models for the prediction of inhibitors of MAO-A and MAO-B. Molecular docking studies revealed that π-π interaction with Phe352 and Trp441 was common among top three highly active compounds in the data set with highest binding scores. Similarly, in MAO-B, hydrogen bond interaction with Thr201 and Tyr326 were the most common. However, π-π and hydrogen bonding interactions were both found to be important for binding with MAO-A and MAO-B. GRIND (Grid Independent Descriptors) based studies depicted that two hydrogen bond donor groups at a distance of 1.6-2 Å and a hydrogen bond donor and hydrophobic group at a mutual distance of 9.6-10 Å were favourable for activity against MAO-A. These virtual site features were complemented by actual residues of the binding site residues Tyr407 and Phe208 that guided in developing ligand based pharmacophore model. Moreover, interactions with these residues were also observed in docking analysis of active compounds. Similarly, in case of MAO-B, hydrogen bond donor and hydrophobic group at a distance of 1.6-2.0 Å and H-bond donor and acceptor group at a distance of 6.40-6.80 Å were also present in active compounds. Moreover, in MAO-B, Tyr326 of the actual binding site observed in docking interaction of active compounds complemented with hydrogen bond donor of the virtual receptor site. Overall, we developed GRIND guided pharmacophore models with some additional features for both MAO-A and MAO-B which elucidated the liability of new chemical entities to produce a selective inhibitory action against monoamine oxidase-B by virtual screening of drug bank data