Benchmarking Computational Methods for Modeling Cyclodextrin–Drug Systems

Pharmacokinetics plays a significant role in determining drug pharmacological effects, and exploring delivery systems that enhance these properties is crucial for drug development. Cyclodextrins (CDs) are a group of cyclic oligosaccharides characterized by a hydrophilic outer surface and a lipophilic central cavity that can form inclusion complexes with various drugs, often enhancing their aqueous solubility and making them suitable vehicles for drug delivery. In this work, we explore the adequacy of using Molecular Docking calculations, Molecular Dynamics (MD) simulations and Quantum Mechanical (QM) calculations to study the interactions between a set of drugs and CDs. Three cyclodextrin systems (HP[Formula: see text]CD, HP[Formula: see text]CD and HP[Formula: see text]CD) were studied in combination with 36 different drugs. For Molecular Docking calculations, a tweaked version of the Vina scoring function with a hydrophobic weight of −0[Formula: see text]3159 was used, yielding the best correlation with the experimental data (Pearson correlation coefficient, [Formula: see text]of 0.58). The MD simulations enabled the evaluation of the systems over time, and several analyses were performed. In particular, energy landscapes were determined to highlight the populations of different configurations, thereby allowing a detailed description of the drug’s behavior. Subsequently, Molecular Mechanics/Poisson–Boltzmann Surface Area (MM/PBSA) calculations were performed, which illustrated a clear limitation in the solvation description of this method. QM calculations were also performed, but were found to be computationally too expensive to sample the numerous systems and their various configurations. Overall, our study highlights the challenges of the tested methods in characterizing binding phenomena involving CDs, while also proposing possible strategies to mitigate these limitations.