BLIND DOCKING SIMULATIONS WITH AUTODOCK TOOLS AND VINA: A STUDY ON THE REPRODUCTION OF EXPERIMENTAL CRYSTAL STRUCTURES

Human serum albumin (HSA) is the main protein in the blood plasma. It performs several important functions, including the transport of poorly soluble drugs, a key aspect from a pharmacological point of view. HSA contains three main binding sites for this purpose: binding sites 1, 2, and 3 (DS1, DS2, and DS3). To understand the interaction between drugs and HSA, researchers use molecular docking techniques to simulate and analyze ligand-protein interactions. In this sense, this study aims to evaluate the effectiveness of Autodock4 (ATD4) and Autodock vina (ATDV) software in accurately reproducing the correct binding sites of crystallographic ligands in complexes with albumin. To this end, the blind docking methodology was applied to the following HSA complexes from the Protein Database: 2BXG, 1AO6, 1N5U, 1E7B, 2BXF, 2BXC, 2BXD, 2BX8, 4L9K, 4L9Q, and 4LB9, using three specific ligands for each drug interaction site, namely Ibuprofen, Phenylbutazone, and Camptothecin. The proteins were prepared by removing ligands and water molecules, adding Kollman charges and polar hydrogens, setting a box resolution of 0.5 Å, and a box size that encompassed the entire crystallographic structure. We used the Lamarckian Genetic Algorithm settings for AutoDock4. For AutoDock Vina, the same proteins previously prepared were used, but it was necessary to fill in a text file with information about the box size and center. Subsequently, we performed blind docking for each complex and evaluated the percentage of accuracy, which refers to the ability to return to the original binding site. Overall, ATD4 and ATDV were unable to reproduce the original binding sites accurately. For example, HSA: Phenylbutazone (DS1,2BXC): 11.7±4.22% (ATD4) and 0±0% (ATDV) return to DS1. HSA: Ibuprofen (DS2,2BXG): 28.7±2.44% (ATD4) and 17±0.4% (ATDV) return to DS2. HSA: Camptothecin (DS3,4L9K): 49.3±1.11% (ATD4) and 52±0.4% (ATDV) return to DS3. The results suggest that the blind docking methodology was not efficient in reproducing the crystallographic structures. Therefore, other methodologies are needed to understand the formation of the protein-ligand complex better.

Conselho Nacional De Desenvolvimento Científico e Tecnológico (CNPq) supported this work.