Calculation of interaction energies with different levels of computational chemistry methods

Abstract

In this thesis, both quantum mechanics and molecular mechanics methods are used to find solutions to two different real life problems. In the first part of the thesis, the interaction energies of the complexes between drugs and drug delivery systems were calculated by quantum mechanical methods, in the second part, potential inhibitor investigation studies for the Ape1 enzyme, which was defined as a target in antibiotic-resistant bacterial infections, were carried out by molecular mechanical methods. One of the important problems in classical microsurgery operations is the spasm of the vessels during the operation. In order to prevent this problem, lidocaine and/or papaverine solutions are applied directly to the veins at certain intervals throughout the operation. Pectin is a linear heteropolysaccharide found in the cell walls and is mainly formed by galacturonic acid residue. Divalent and trivalent cationic salts of pectin are weakly soluble or insoluble depending on the cation. Low methoxy pectin gels form the egg box structure, which becomes more stable with neighboring pectin chains via van der Waals interactions. Pectin is used as a drug delivery system in biomedical applications is becoming widespread due to its highly biocompatible, biodegradable, and non-toxic feature. The 3D gel structures can be developed which may serve as a carrier system for vasodilator drugs (papaverine and/or lidocaine) against vasospasm that may occur in vessels during the microsurgery operation. In the first part of the thesis, quantum mechanical calculations based on Density Functional Theory were carried out to predict the release rates of papaverine and lidocaine molecules. Mutual interactions are examined both for the same and the different kinds of molecules at different pH environments. In order to take into account the dispersion effects encountered in the QM calculations of the complexes formed by van der Waals interactions the wB97xD functional and 6-311++G(d,p) basis set was used. Basis Set Superposition Error is another problem arising during the calculation of complexation energies, where the basis functions of interacting molecules affect each other and create more stable complexes artificially, was overcome with Counterpoise Correction. The implicit solvent medium was defined with water to mimic the physiological environment. All calculations was performed with Gaussian16 programme. Calculations carried out simultaneously with the experiment were performed at low pH values, as it was understood that the physiological environment had a negative effect on the gelation of pectin. In the light of the findings, it can be stated lidocaine could be released faster from the drug carrier pectin, while papaverine, which formed a more stable complex, would be released in a more controlled manner. The second chapter of the thesis includes the study of enzyme drug interactions. Antibiotic resistance is one of the common problems all over the world. Although there has been an increase in research on gram-negative bacteria in recent years, the amount of observable information is still limited. This restricts understanding of the nature of resistance of gram-negative bacteria and new drug therapy research. Peptidoglycan O-acetylesterase (Ape1) which belongs to the class of SGNH esterase was reported as a potential drug target since it plays an important role for the survival of the gram-negative bacteria. It catalyzes the de-O-acetylation of peptidoglycan the main substance of the bacterial cell wall and is one of the important biomolecules for the life of the bacteria. Different levels of computational drug design approaches have been used to identify potential inhibitors for Ape1. A virtual screening library of 2552 FDA-approved drugs was studied for this purpose. First, flexible docking studies were performed to predict interactions between drugs and active site of Ape1. The results obtained were ranked and the Prime-MMGBSA binding energy calculation was performed to rescore compounds up to 15% of the best score. To further validate the stability of four selected compounds giving binding free energy values below -50.0 kcal/mol Molecular Dynamics simulations was performed. Post-simulation MMGBSA calculations were performed with trajectories from MD simulation for three compounds that remained in the active site. The drugs with the best binding energies, namely Rifaximin, Ulipristal and Dihydroergotamine were selected as candidates for in-vitro studies of Ape1 inhibition.