Abstract
Several studies have reported the computational study of platinum-based compounds for chemotherapeutic treatment of different cancers. However, there remains a need to study cisplatin and its derivatives to further understand properties such as the chemical reactivity, kinetic stability, and reactivity descriptors that can further elaborate why some compounds are active against cancer cells, while some are not. The present study aimed to perform molecular structure optimization calculations for the bond lengths and angles, NBO analysis, MESP, HOMO-LUMO and Fukui function analysis of selected platinum-based anti-cancer drugs using DFT techniques to evaluate the anti-cancer activities against the computed structural and electronic properties. The M06-2X functional was found to produce structures with the lowest energy conformations and closest replication of experimental findings in literature. The NBO analysis was carried out to understand the bonding interactions within the structures, showing that the second order perturbation energy was increased in the presence of π-electron delocalization within the structure. The MESP and Fukui reactivity indices revealed that the most nucleophilic atom within the platinum complexes was the chloride leaving group. The results for the HOMO-LUMO analysis showed that the frontier orbital energy gap is increased by an increase in alkyl chain length. The energy gap was also noticed to decrease for compounds having an electron-withdrawing group in the carrier ligand in the order: COOH<Br<F<OH<NH2. The global reactivity descriptors showed that the most reactive compound was C28 which had the lowest global hardness and highest global softness. 3D QSAR models that can be used to predict the activity of new molecules or independent datasets were generated using a set of structurally diverse compounds. Multiple linear regression 3D QSAR models were generated for the HOMO-LUMO energy gap and the electrophilicity index. A multiple regression QSAR model was also generated using multiple reactivity descriptors such as the molecular weight (MW), partition coefficient (AlogP), and the polar surface area (PSA). Radial and dendritic fingerprints were generated and compared for the 15 compounds, and the dendritic fingerprints showed a higher degree of linearity within the dataset...
M.Sc. (Chemistry)