<i>Ab initio</i> evaluation of complexation energies for cyclodextrin–drug inclusion complexes

Within the framework of density functional theory (DFT), an appropriate incorporation of van der Waals (vdW) into the exchange correlation (XC) functional is important to accurately predict complexation energies of noncovalent systems. This point has not been widely verified in large host-drug systems. Here we benchmarked various DFT to evaluate complexation energies of host-guest complexes, three types of β-cyclodextrins and plumbagin (known as anti-cancer drug). We modeled these systems as follows: (1) DFT geometry optimization of individual guest and hosts, (2) their docking conformation search based on genetic algorithm with semi-empirical simulations, and (3) DFT geometry optimization of the docking system. Then single-point energy calculations were performed to evaluate complexation energies. Furthermore, we applied DMC (diffuse Monte Carlo) to validate the DFT results. We found M06-2X-D3 and CAM-B3LYP-D3 give almost the same energies for all the cases and they are both consistent with DMC within the error bar. From these results, we concluded that a proper incorporation of vdW and long-range exchange corrections into XC functionals is essential for describing the cyclodextrin-plumbagin systems.