Structure based drug design and machine learning approaches for identifying natural inhibitors against the human αβIII tubulin isotype.

Microtubules (MTs) play a crucial role in mitosis and are composed of α-/β-tubulin heterodimeric subunits. In eukaryotes, eight α-tubulin and ten β-tubulin isotypes have been reported, each displaying tissue-specific expression patterns. Among them, the β-tubulin isotype is significantly overexpressed in various cancers and is closely associated with resistance to anticancer agents, making it an attractive target for cancer therapies. This study employed a comprehensive approach integrating structure-based drug design, machine learning, ADME-T and PASS biological property evaluations, molecular docking, and molecular dynamics simulations to identify potential natural compounds targeting the 'Taxol site' of the αβ-tubulin isotype. Screening of 89,399 compounds from the ZINC natural compound database yielded 1,000 initial hits based on binding energy. Further, refinement using machine learning classifiers narrowed down these to 20 active natural compounds, of which four - ZINC12889138, ZINC08952577, ZINC08952607, and ZINC03847075 exhibited exceptional ADME-T properties and notable anti-tubulin activity. Molecular docking revealed significant binding affinities of these compounds to the 'Taxol site' of the αβ-tubulin isotype. Molecular dynamics simulations evaluated using RMSD, RMSF, Rg, and SASA analysis, revealed that these compounds significantly influenced the structural stability of the αβ-tubulin heterodimer compared to the apo form of the αβ-tubulin isotype. Moreover, binding energy calculations showed a decreasing order of binding affinity for αβ-tubulin; ZINC12889138 > ZINC08952577 > ZINC08952607 > ZINC03847075. In conclusion, this study identified natural compounds against drug resistant αβ-tubulin isotype. These findings offer a promising foundation for developing novel therapeutic strategies targeting carcinomas associated with β-tubulin overexpression.