Estudos da relação quantitativa estrutura-atividade (QSAR) de adutos de Morita-Baylis-Hillman bioativos contra Leishmania amazonensis

• Ano de defesa: 2012
• Autor(a) principal: Alencar Filho, Edilson Beserra de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraí­ba; BR; Química; Programa de Pós-Graduação em Química; UFPB
• Palavras-chave em Português: Adutos de Morita-Baylis- Hillman; Leishmania amazonensis; Morita-Baylis-Hillman Adducts; Density Functional Theory - DFT; Quantitative Structure-Activity Relationship-QSA; CIENCIAS EXATAS E DA TERRA::QUIMICA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/tede/7163

The Morita-Baylis-Hillman Adducts (MBHA) is a class of molecules studied by our research group on synthetic, theoretical and bioactivity aspects. In this work, we present Quantitative Structure-Activity Relationship (QSAR) models involving 32 aromatic MBHA. Initially, the most stable conformations of thirty-two MBHA were investigated by theoretical methods, which were used to construct models. For this study, were obtained potential energy curves using AM1 semi-empirical method, considering rotational degrees of freedom (sigma bonds). From these curves, the less energy conformation to each molecule was selected and optimized at B3LYP/6- 31+G(d) level, considering solvent effects through Polarizable Continuum Model (PCM). Proton Nuclear Magnetic Ressonance data are in agreement with the conformational study. Intramolecular Hydrogen Bonds (IHB) are presents in the most of the studied compounds, according to structural characterization and QTAIM calculations. Curiously, compounds that showed hydrogen bonds involving the nitro and hydroxyl groups have the best values of biological activity (IC50). An explanation is based on redox mechanism of action of nitrocompounds. NBO (Natural Bond Orbital) charges and LUKO (Lowest Unoccupied Kohn-Sham Orbitals) analysis at the ortho-nitro group are in agreement with these analyses. Considering quantum calculations and structural observations, four descriptors were selected a priori and submitted to a QSAR study using PLS (Partial Least Squares) and MLR (Multiple Linear Regression) modeling. A second QSAR approach was made from the another set of descriptors obtained through the online platform E-DRAGON, which were submitted to a variable selection method. The quality parameters obtained for models indicate that both are robust and predictive.