Identification of Bioactive Phytoconstituents from Michelia champaca L. Flowers and Their Interactions with CXCR4 and PTEN Targets: A Comprehensive ADMET,Molecular Docking, and Molecular Dynamics Study

Abstract

The identification of bioactive phytoconstituents and their molecular interactions with therapeutic targets provides a rational framework for early-stage drug discovery from medicinal plants. The present study investigates the chemical composition of Michelia champaca L. flower extract using GC—MS and evaluates the potential binding affinity of the prioritized compounds against PTEN (PDB ID: 1D5R) and CXCR4 (PDB ID: 3ODU) via a co-crystal validated in silico workflow. GC—MS profiling putatively identified 42 distinct constituents, with a clear predominance of long-chain fatty acids, primarily n-hexadecanoic acid (44.7%) and octadecanoic acid (26.6%). Following Lipinskis rule of five and ADMET filtering, site-specific molecular docking was executed using a protocol verified by native ligand redocking. Against the PTEN phosphatase active site, the minor polar constituent 4H-pyran-4-one, 2,3-dihydro-3,5-dihydroxy-6-methyl- exhibited the highest predicted structural affinity, with an exceptional docking score of -7.179 kcal/mol, followed by dl-mevalonic acid lactone (-5.553 kcal/mol). For the CXCR4 transmembrane pocket, Cholesta-4,6-dien-3-ol, (3.beta.)- (-5.280 kcal/mol) and acetamide, N-(4-ethoxy-3-hydroxyphenyl)- (-5.156 kcal/mol) showed the strongest predicted binding interactions. A subsequent 100 ns all-atom molecular dynamics simulation and MM-GBSA free energy analysis of the top-ranked PTENcomplex confirmed strong physical and thermodynamic stability, driven by favorable van der Waals (-30.8 to -36.9 kcal/mol) and electrostatic interactions. These predictive findings suggest that specific M. champaca flower constituents possess strong targeted binding potential against PTEN and CXCR4 nodes, establishing a validated computational foundation that warrants downstream in vitro and in vivo functional experimental validation.