Development of a Coarse-Grained Molecular Dynamics (CG-MD) approach to explore tubulin-ligand binding

Microtubules (MTs) are highly dynamic polymers that regulate numerous cellular functions, such as intracellular cargo trafficking and cell proliferation [1]. Tubulin αβ-heterodimer is the fundamental structural unit of MTs and small molecules’ binding to this protein can regulate dynamically growth and shrinkage of MTs, with important therapeutic implications [2]. Indeed, over the last decades many ligands with diverse chemical scaffolds have been investigated and eight binding sites have been experimentally characterized [3]. Further advances in the therapeutic development for this strategic target require the assistance of computational methods. Current in-silico investigations, such as docking or All-Atom Molecular Dynamics (AA-MD), are limited by either the conformational complexity of the investigated ligand or by the timescales that can be effectively simulated. An alternative approach is offered by Coarse-Grained Molecular Dynamics (CG-MD). By grouping several atoms into a smaller number of beads, this method allows to reduce the cost of simulation by three orders of magnitude [4]. In particular, the Martini 3 model was shown to be a good compromise between reduction of complexity and maintenance of chemical accuracy [5]. The objective of this ongoing Ph.D. project is to develop a computational approach with the Martini 3 model to investigate the binding of small molecules to the tubulin αβ-heterodimer. In particular, the first target is to elucidate the binding pathway of the alkaloid drug colchicine to its experimentally known binding site. Preliminary results show how unbiased CG-MD simulations of colchicine molecules can identify long-lived interaction sites near the known binding pocket. Current efforts focus on the investigation of these poses with enhanced sampling methods to elucidate the complete binding pathway. If this in-silico approach will be validated, it could provide a novel and multiscale approach to investigate the binding of innovative ligand candidates without a priori knowledge of their binding site, leveraging the reduced cost of CG-MD to efficiently identify meaningful starting points for more refined and computationally intensive methods. References [1] Janke, C., Journal of Cell Biology 2014, 206(4): 461-472 [doi: 10.1016/j.tcb.2018.05.001] [2] Steinmetz, M. O., et al, Trends in Cell Biology 2018, 28(10): 776–792 [doi: 10.1016/j.tcb.2018.05.001] [3] Pérez‐Peña, H., et al, Biomolecules 2023, 13(2): 285 [doi: 10.3390/biom13020285] [4] Ingólfsson, H.I., et al, WIREs Comput Mol Sci 2014, 4: 225-248 [doi: 10.1002/wcms.1169] [5] Souza, P.C.T., et al, Nat Commun 2020, 11: 3714 [doi: 10.1038/s41467-020-17437-5]