Triagem virtual de alcaloides isolados da família Euphorbiaceae frente a Doença de Chagas

• Ano de defesa: 2023
• Autor(a) principal: Araujo, Igor Mikael Alves de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Farmacologia; Programa de Pós-Graduação em Produtos Naturais e Sintéticos Bioativos; UFPB
• Palavras-chave em Português: Doença de Chagas; Triagem virtual; Produtos naturais - Banco de dados; Alcaloides; Moléculas multitarget; Chagas disease; Virtual screening; Natural products - Database; Alkaloids; Multitarget molecules; CNPQ::CIENCIAS BIOLOGICAS::FARMACOLOGIA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/30247

Chagas disease is an infection caused by the Trypanosoma cruzi parasite and is endemic in several Latin American countries. Treatment for Chagas disease is limited to a few medications, which also have limitations, including significant side effects and variable effectiveness at different stages of the disease. Because of this there is an urgent need for new drugs to treat the evolving forms of the parasite. Natural products, such as those derived from plants of the Euphorbiaceae family, are considered a potential source of new treatments. These plants are widely used in traditional medicine and contain several secondary metabolites. The main objective was to identify compounds with potential biological activity against Chagas disease through virtual screening. The focus was isolated alkaloids from the Euphorbiaceae family. Thus, a database of alkaloids isolated from the Euphorbiaceae family was built, aiming at the elaboration of a chemical profile. A prediction model was built in KNIME. For this, the data set was obtained from the ChEMBL database, and the compounds were classified according to the pIC50 values and the calculation of descriptors was performed using the Volsurf software. Molecular docking was carried out using the Molegro Virtual Docker software, along with the analysis of interactions for four proteins obtained from the Protein Data Bank. In addition, the Moldock score, PLANTS score and Rerank score functions were used, and the binding energies were consensually evaluated. The developed predictive model classified molecules with a probability above 70% for the T. cruzi tripromastigotes, resulting in the identification of 21 molecules with potential activity. Molecular docking analyzes were positive, indicating interactions of the selected compounds with the target enzymes, evidenced by negative energies. As for absorption, the compounds have demonstrated greater than 55% absorption orally, with good availability, usually with only one rule violation. In the toxicity analysis, only nine compounds showed signs of toxicity in one or two parameters. After calculating the combined probability values (molecular anchorage and prediction model), molecule 149 was selected because it presented a higher percentage in all enzymes, thus demonstrating a multitarget potential. The results of molecular dynamics simulations demonstrate that the RMSD of the complex for the test compound molecule 149 (Magnoflorina) remains stable. Furthermore, it can be suggested that through calculations of RMSF and interaction energy, the test compound Magnoflorina interacts with this target, enabling interaction, flexibility and stability.