Desenho computacional de metabólitos secundários de annonaceae: seleção e atividades antiparasitárias

Abstract

Plants are rich sources of natural products, which in turn are a potential source of bioactive substances for the development of new drugs. The Annonaceae family is extremely rich in secondary metabolites, with great chemical diversity, presenting a vast variety of biological potential. The use of computational techniques for the discovery of new drugs has become increasingly common and necessary since it leads to a reduction in research costs and time. Computer-assisted drug development allows the exploration of large chemical databases, reducing these banks to sets of molecules with high potential for biological activity, a process known as virtual screening. Therefore, this study aims to perform computational studies to obtain promising molecules with biological activity for neglected diseases, Chagas disease and leishmaniasis from the Annonaceae secondary metabolite database. In addition to investigating the leishmanicidal potential of extracts from four Annona species (Annona glabra, Annona mucosa, Annona sylvatica and Annona dolabripetala) through a metabolomics approach using multivariate statistical analysis (PCA and PLS) and LC-MS data, to correlate spectroscopic data with leishmanicidal activity, seeking to suggest compounds or groups of compounds responsible for the biological activity. In chapter 1, a review of the biological activity studies conducted with species of the Annonaceae family was conducted, aiming to show how versatile and promising this family is in the search for new drugs. In chapter 2, a review was conducted on machine learning applied to QSAR, written in Portuguese, to approach the subject in a simple and didactic way for students and researchers who are starting in this very promising and important area. In chapter 3, based on the construction of a database with secondary metabolites already isolated from the Annonaceae family between 1970 and 2019, a chemotaxonomic analysis was performed using the class of diterpenes. Through this chemotaxonomic study it was possible to separate the Annoneae, Xylopieae and Miliuseae tribes according to the morphological and taxonomic separation of the family. This phenomenon makes it possible to predict the location of a particular diterpene in the Annoneae, Xylopiieae and Miliuseae tribes of the Annonaceae and to search for these secondary metabolites and their biological potential more effectively. In chapters 4 and 5, virtual screening studies based on ligand are conducted in search of molecules with potential antichagasic and leishmanicidal activity using the Annonaceae secondary metabolite database, consisting of 1860 molecules. The predictive models created for L. amazonensis and T. cruzi obtained an accuracy above 72%. For the two protozoa it was possible to identify potentially active molecules, select some of them and perform the in vitro test. For T. cruzi, 13-epicupressic acid was the most promising, as it was predicted as an active compound in the in silico study against the amastigote form of T. cruzi, in addition to having in vitro activity against the epimastigote form. As for L. amazonensis, the triterpene lupeol showed the best activity in in silico and in vitro biological assays for the promastigote form, in addition to having a probability of active potential greater than 77% against the amastigote form. In chapter 6, a metabolomic analysis was performed using multivariate statistical analysis (PCA and PLS) and LCMS data, to correlate spectroscopic data with leishmanicidal activity, seeking to suggest compounds or groups of compounds responsible for the biological activity of 4 Annona species.