Modelagem matemático-computacional para análise de dados espectrofluorimétricos e localização de sítios de ligação em albumina sérica

Abstract

The aim of this work was to elaborate a computational mathematician model to analyze data from spectrofluorimetric experiments, directing, in particular, for the localization of binding sites for human (HSA) and bovine (BSA) albumins. The model was based on the geometry related to the positioning of the site in relation to the residues of tryptophan present in the protein molecule, besides considering the similarities/structural differences between these two proteins. To test the model, we used data obtained from fluorescence quenching experiments of three antipsychotic drugs: risperidone, chlorpromazine and haloperidol, and two pesticides: methyl parathion and glyphosate. According to our model, the values found for mean distances between the tryptophan 212 residue and the binding sites for the five substances tested were: 2.59 (± 0.92) nm, 2.65 (± 1.06 ) nm, 2.52 (± 0.97) nm, 2.89 (± 1.49) nm, 3.15 (± 2.00) nm. For the average distances of the tryptophan 134 residue at the same sites were: 2.25 (± 0.80) nm, 2.37 (± 0.94) nm, 2.74 (± 1.05) nm, 3.01 (± 1.55) nm, 3.02 (± 1.97) nm respectively. These values are compatible with the dimensions of the protein domains and thus the project developed presents effective results in this advance of the search of location of primary binding sites in the interaction of drugs to the proteins under study.