Computational study of interactions of anti-cancer drug eribulin with human tubulin isotypes.

Abstract

Microtubules (MTs) are widely targeted for the treatment of various types of cancer due to their essential role in cell division. MTs are polymers made of αβ-tubulin heterodimers. These α- and β-tubulins have 8 and 10 different isotypes, respectively. It is known that a few tubulin isotypes have anti-cancer drug resistance properties, especially βIII, which shows poor sensitivity to many potent anti-cancer drugs such as eribulin. However, the molecular-level understanding of drug-resistance due to tubulin isotype variation is poorly understood. This paper presents the study of differential binding affinities of different tubulin isotypes with the potent anti-cancer drug eribulin. Eribulin (MT destabilizer) binds at the inter-dimer interface of MTs near the vinca site and induces a lattice deformation, which results in catastrophic events in MT dynamics. In this study, sequence analysis has been done throughway and the binding sites and based on that 2α-tubulin isotypes (αI and αVIII) and 7β tubulin isotypes (βI, βIIa, βIII, βIVa, βVI, βVII and βVIII) were selected. In total, 14 combinations were prepared after building homology models of these selected isotypes. Molecular docking and molecular dynamics simulations were performed to deeply understand the binding mode of eribulin at different MT compositions. RMSD, RMSF, radius of gyration, SASA, ligand-protein interactions, and calculations of binding free energy were performed to investigate the eribulin binding variations to tubulin isotypes and it was found that αIβII showed the maximum binding affinity among all 14 systems to eribulin. The βIII-tubulin isotype, which shows low sensitivity to eribulin in experimental results, had the least binding affinity in the system αVIIIβIII complex and the average binding affinity in the system αIβIII among all 14 systems. Additionally, we performed steered MD simulations and DynDom analysis of the systems with the lowest binding energy (αIβII) and the highest binding energy (αVIIIβIII) and extracted force, displacement, and H-bonding profiles during the pulling simulations to get a better insight.