RESUMO
Breast cancer is a highly heterogeneous disease influenced by the hormonal microenvironment and the most common malignancy in women worldwide. Some phytoestrogens and mycoestrogens have been epidemiologically linked as risk factors or protectors, however their mechanisms of action are complex and not fully understood. The aim of this study was to predict the potential of 36 natural xenoestrogens to interact with 189 breast cancer proteins using AutoDock Vina. In order to validate our protocol, an in silico docking pose and binding site determination was compared with the crystallographic structure and the power of prediction to distinguish between ligand and decoys was evaluated through a receiver operating characteristic curve (ROC) of the resultant docking affinities and in vitro data. The best affinity score was obtained for glyceollin III interacting with the sex hormone binding globulin (-11.9 Kcal/mol), a plasma steroid transport protein that regulates sex steroids bioavailability. Other natural xenoestrogens such as beta-carotene, chrysophanol 8-O-beta-d-glucopyranoside and glyceollin I, also presented good affinity for proteins related to this disease and the validation was successful. This study may help to prioritize compounds for toxicity tests or drug development from natural scaffolds, and to elucidate their mechanisms of action.