Estudo in silico de potenciais alvos proteicos para moléculas citotóxicas em linhagens de células leucêmicas humanas.

Abstract

Leukemias are malignant neoplasms characterized by the proliferation of immature cells of the hematopoietic system. Acute Myeloid Leukemia (AML) is a malignant neoplasm characterized by the accumulation of myeloblasts in the bone marrow and peripheral blood, being one of the most prevalent leukemias in adults. The genomic heterogeneity of the disease makes it difficult to establish treatment protocols. On the other hand, Chronic Myeloid Leukemia (CML) is a rare myeloproliferative disease characterized by the accumulation of myeloid cells and the presence of the Philadelfia (Ph) chromosome. Tyrosine kinase inhibitors are the main chemotherapeutic agents available, but there are patients who do not respond adequately to treatment. Therefore, the search for new compounds with antitumor activity is justified. In this work, an in silico study was carried out to search for protein targets for nine compounds with known antitumor activity against the THP-1 and K562 strains. Three compounds are triterpenes (1 to 3) and six are alkaloid analogs of 3-alkylpyridine (4, 5, 6, 7, 8 and 9). The target fishing methodology was used to search for targets, based on the principle of similarity in which similar molecules can bind to the same targets. The ChEMBL database was used to search for molecules with cytotoxic activity against the THP-1 and K562 strains, to then be compared to the compounds of interest by calculating the Tanimoto coefficient (Tc), which assesses the similarity between molecules. The Tc was calculated using the MACCS and PubChem fingerprints. Compounds with Tc values greater than or equal to 0.75 were considered for the bibliographic search of targets already described in the literature. Another strategy was the use of servers for the prediction of molecular targets. The proteins 6- phosphofruct-2-kinase/fructose-2,6-bisphosphatase 3 (PFKFB3) and nicotinamide phosphoribosyltransferase (NAMPT) were indicated as possible targets for compound 7 and phase M inducing phosphatase 2 (CDC25B) for the compounds 1 and 3. The crystallographic structures of these potential targets were obtained using the Protein Data Bank (PDB). Compounds 7, 1 and 3 were submitted to docking simulations with the GOLD program. The in silico study was successful in the search for molecular targets for compound 7, since important interactions described in the literature with the binding site of the PFKFB3 and NAMPT proteins were observed. On the other hand, compounds 1 and 3 did not bind at the CDC25B protein binding site favorably.