Development of a protocol for the virtual screening of a series of smallmolecules ligands of the tubulin colchicine binding site

Microtubules are cytoskeletal polymers composed of αβ-tubulin heterodimers, an extensive network of filaments that span the cell interior. [1] Interrupting the correct formation of microtubules lead to cell apoptosis, which is widely recognized as an attractive approach for the design of potent anticancer agents. The chemotherapeutic class of tubulin-binding agents (TBA), such as like taxanes, vinca alkaloids and epothilones, are well-known binders of β-tubulin able to suppress the dynamics of spindle microtubules, disturbing the chromosomal segregation and inducing cell death. In humans, microtubules are composed of eight α-tubulin isotypes and seven β-tubulin isotypes in mixed combinations. These isotypes have high degree of structural homology, but a vastly divergent sequence at their carboxy-terminal tail. Tubulin isotypes play an important role in microtubule dynamics. [2] Some isotypes present different binding affinities for a variety of anticancer agents. However, the reasons underlying the magnitude of the binding affinities are not well understood at the molecular level. Therefore, we have developed a protocol to perform a virtual screening to investigate the binding-affinity of a series of small-molecule ligands (acridone derivatives) to the tubulin colchicine site. First, we performed a molecular dynamics simulation to evaluate an average structure, using cluster analysis. Second, using the resulting structure, we executed a molecular docking strategy with a library of ligands with known IC50 values. Finally, we calculated the affinity energy using the Free Energy Perturbation method. Our results clearly show that this protocol succeeds in predicting new candidate molecules with high binding affinity to the tubulin colchicine binding site.