Evaluation of the Xtb Semiempirical Method for the Prediction of Antioxidant Properties in Alzheimer’s Disease: Salen-Type Ligands

Abstract

Alzheimer’s disease (AD) stands as the predominant form of dementia, accounting for up to 70% of all cases worldwide. AD is a complex disease with various contributing factors. Evidence suggests that the metalliccomplexes formed by the β-amyloid peptide (Aβ) and extraneuronal copper can catalyze the generation ofreactive oxygen species, consequently increasing oxidative stress and contributing to the decline of neurons. This interaction underscores the significance of bioavailable copper as a crucial redox-active target in exploring protocols for multifunctional agents in AD treatment. In the field of computational chemistry, density functional theory (DFT) is widely accepted as a standard method across different disciplines. Despite this, DFT presents computational challenges, particularly in screening extensive molecular sets during the initial phases of drug research. Recent advances in semiempirical quantum mechanical methods (SQM) offer a promising alternative, providing rapid molecular geometry optimization and approximate estimation of thermodynamical properties, being at least two orders of magnitude faster than traditional DFT calculations. In this work, we present an evaluation of the GFNn-xTB SQM methods in the rapid screening of antioxidant properties in AD, performed on a set of salen ligands by calculating the standard reduction potentials of their copper complexes as key property. Results show that the implementation of GFNn-xTB SQM calculations before DFT evaluations is