Computational Exploration of a Diverse Flavonoid Library for Targeted Allosteric Inhibition of AKT1 in Cancer Therapy.

Abstract

Background/aim: AKT serine/threonine kinase 1 (AKT1) is an established therapeutic target in cancer therapy due to its role in promoting cell survival and proliferation. This study aimed to identify potential allosteric inhibitors of AKT1 from a large flavonoid library using computational methods.

Materials and methods: A computational screening of a comprehensive flavonoid library to identify novel allosteric inhibitors targeting the AKT1 allosteric site was performed. Molecular docking identified compounds with favorable binding interactions and the top 10 were selected for binding pose analysis. Molecular dynamics simulations for 200 ns were further employed to assess the stability of the highest-ranked compound.

Results: The study proposed 10 flavonoids as potential allosteric AKT1 inhibitors. The docking analysis highlighted critical interactions between the 10 flavonoids and AKT1, with residues such as Trp-80, Ile-84, Tyr-272, Arg-273 and Asp-292 playing significant roles in binding stability. Trp-80 emerged as a pivotal residue, consistently forming the highest number of non-bonding contacts across most compounds, corresponding with prior studies that identified it as essential for allosteric inhibition. The highest-ranked flavonoid, CID 108790283, demonstrated the strongest binding affinity, with a binding energy of -10.64 kcal/mol, 56 non-bonding contacts, and a hydrogen bond. Molecular dynamics simulations further confirmed the stability of this flavonoid within the allosteric site, exhibiting minimal conformational fluctuation throughout the 200-ns simulation.

Conclusion: This computational investigation identified 10 flavonoids with strong interaction profiles and stable binding within the allosteric site of AKT1, suggesting their potential as novel AKT1 inhibitors. These findings provide a basis for further experimental studies to validate their efficacy in cancer treatment.