de Menezes Renata Priscila Barros, de Assis Edileuza Bezerra, de Sousa Natália Ferreira, de Souza Jullyanne Maria Saraiva, da França Rodrigues Klinger Antônio, Scotti Luciana, Tavares Josean Fechine, da Silva Marcelo Sobral, Scotti Marcus Tullius
Programa de Pós-Graduação Em Produtos Naturais E Sintéticos Bioativos, Universidade Federal da Paraíba, João Pessoa, PB, 58051-900, Brazil.
Laboratório de Doenças Infecciosas, Universidade Federal Do Delta Do Parnaíba, Parnaíba, PI, 64202-020, Brazil.
Mol Divers. 2025 Apr 26. doi: 10.1007/s11030-025-11200-y.
Neglected tropical diseases such as Leishmaniasis and Chagas disease remain critical public health challenges. This study applied ligand-based virtual screening to a dataset of 4,150 secondary metabolites from the Lamiaceae family, aiming to identify multitarget molecules against four Leishmania species (L. infantum, L. donovani, L. amazonensis, and L. braziliensis) and Trypanosoma cruzi forms. Random forest models exhibited high accuracy (over 72%), leading to the identification of 82 molecules with potential multitarget activity across five of six predictive models. Nineteen prioritized molecules were subjected to molecular docking simulations targeting key enzymes-including sterol 14-alpha demethylase, glucose-6-phosphate dehydrogenase, dihydroorotate dehydrogenase, nucleoside diphosphate kinase, tryparedoxin peroxidase, and cruzain-with compounds 12, 18, and 19 exhibiting a high binding affinity across multiple targets. In vitro assays confirmed the predicted activity of selected molecules (3, 4, and 5) against Leishmania and T. cruzi. Importantly, these molecules represent novel findings, with antileishmanial or antitrypanosomal activities that have not been previously reported. The results highlight their potential as multitarget therapeutic candidates for neglected tropical diseases, paving the way for further biological evaluation and development.
利什曼病和恰加斯病等被忽视的热带病仍然是严峻的公共卫生挑战。本研究对唇形科4150种次生代谢产物的数据集应用基于配体的虚拟筛选，旨在鉴定针对四种利什曼原虫（婴儿利什曼原虫、杜氏利什曼原虫、亚马逊利什曼原虫和巴西利什曼原虫）和克氏锥虫形态的多靶点分子。随机森林模型表现出较高的准确性（超过72%），从而在六个预测模型中的五个模型中鉴定出82种具有潜在多靶点活性的分子。对19种优先分子进行了针对关键酶的分子对接模拟，这些关键酶包括甾醇14-α脱甲基酶、葡萄糖-6-磷酸脱氢酶、二氢乳清酸脱氢酶、核苷二磷酸激酶、锥虫硫氧还蛋白过氧化物酶和克鲁斯蛋白酶，化合物12、18和19在多个靶点上表现出高结合亲和力。体外试验证实了所选分子（3、4和5）对利什曼原虫和克氏锥虫的预测活性。重要的是，这些分子代表了新的发现，具有以前未报道过的抗利什曼原虫或抗锥虫活性。结果突出了它们作为被忽视热带病多靶点治疗候选药物的潜力，为进一步的生物学评估和开发铺平了道路。
