Estudo in silico de flavonoides e compostos análogos pertencentes à família Asteraceae contra a doença de Alzheimer

Abstract

Alzheimer's disease is a neurodegenerative disorder that affects more than 20 million people worldwide. Currently, the drugs available for the treatment have several side effects and only reduce the clinical manifestations of the disease, not being able to prevent it from progressing. Hence, the search for new therapeutic alternatives that delay the development of the disorder is essential. One of the strategies used in the drug discovery process against Alzheimer's disease is to carry out an in silico study. Thus, it is possible to predict the biological activities and physicochemical properties of molecules and contribute to reducing the time and financial costs of drug research. Flavonoids are plant secondary metabolites that have beneficial effects in preventing diseases, including neurodegenerative diseases. Based on this, the aim of this work was to develop in silico analyzes and perform a virtual screening of flavonoids and analogues of the Asteraceae family against Alzheimer's disease targets. In this study, 886 flavonoids were selected to initiate the virtual screening and the following computational methods were used: construction of QSAR (Quantitative Structure-Activity Relationship) models, analysis of cytotoxicity risks, analysis of violations of Lipinski's rule of five, calculations of physicochemical parameters related to antioxidant activity, consensus principal component analysis (CPCA), principal component analysis (PCA), molecular docking and molecular dynamics simulations. The QSAR models exhibited satisfactory statistical parameters and indicated molecules possibly active against acetylcholinesterase (AChE), amyloid β peptide (Aβ) and glycogen synthase kinase-3β (GSK-3β), which probably cross the blood-brain barrier. These molecules were submitted to cytotoxicity risk analysis and among those that did not present risks, flavonoids with probable multitarget activity were selected, obtaining: FL319, FL339, FL466, FL511, FL775 and FL807. Predicted metabolites for these flavonoids also showed no cytotoxicity risks, with the exception of only one metabolite formed by the interaction of FL511 with CYP2D6. The six flavonoids analyzed showed at most 1 violation of Lipinski's rule of five, demonstrating physicochemical properties favorable to oral bioavailability. Furthermore, all flavonoids were more reactive and had a greater ability to lose an electron and neutralize free radicals compared to the antioxidant ascorbic acid, suggesting that they have antioxidant properties. The CPCA and PCA results revealed that all flavonoids, with the exception of FL319, form highly favorable hydrophobic interactions with other molecules, and that FL466 and FL511 have a better solubility profile. It was also observed that the studied flavonoids have hydrophobic and hydrophilic regions, hydrogen bond acceptor groups and hydrogen bond donor groups. Molecular docking procedures provided data on the possible mechanisms of action of flavonoids, in which it was noted that FL466, FL511, FL775 and FL807 interacted with all amino acid residues from the peripheral anionic site of AChE. The six flavonoids formed interactions with the fundamental residues for the aggregation of Aβ, which are the residues of the central hydrophobic core of the peptide. FL319, FL339 and FL775 also participated in interactions with the residues in the adenosine triphosphate (ATP) binding site of GSK-3β. In molecular dynamics simulations, the flavonoids were able to stabilize the targets for a few nanoseconds. Therefore, it is believed that the flavonoids investigated in this study are promising molecules against the targets of Alzheimer's disease and have the potential to be tested experimentally.