Computational modeling of selected phytochemicals from Reevesia formosana: mechanistic insights into their potential against lung cancer
TL;DR
Integrated modeling prioritizes CPD1 as a Bcl-2-targeting scaffold for lung cancer-relevant studies, supporting structure-guided optimization and experimental verification.
Abstract
Background: Lung cancer remains a dominant cause of cancer mortality, and impaired apoptosis mediated by anti-apoptotic Bcl-2 family proteins contributes to persistence and treatment resistance. Phytochemicals from Reevesia formosana display reported anti-NSCLC (Non-Small Cell Lung Cancer) activity, yet target-level mechanisms remain incompletely defined. Objectives: To prioritize Reevesia formosana phytochemicals as putative Bcl-2 (PDB ID: 6GL8) inhibitors and to derive mechanistic insight through multiscale computation. Materials and methods: Six phytochemicals and the reference Tivantinib were docked to 6GL8, and the top candidate (CPD1) was evaluated by 100 ns molecular dynamics simulations alongside Tivantinib. MMGBSA binding free energies were computed from 125 snapshots spanning 20-100 ns. ADMET properties were predicted using pkCSM. Frontier-orbital and global reactivity descriptors were obtained by DFT. Results: Docking ranked CPD1 highest (-10.15 kcal/mol) relative to Tivantinib (-8.83 kcal/mol), with hydrogen bonding to Tyr108 and Arg129 and extensive pocket complementarity. MD trajectories indicated a more confined CPD1-6GL8 complex (RMSD mainly 0.18–0.22 nm) than Tivantinib-6GL8 (0.21–0.26 nm, occasional 0.27-0.28 nm), with comparable compactness (Rg of 1.44 nm) and bounded SASA. Hydrogen-bond counts supported intermittent polar anchoring for CPD1 (0-3) and higher early sampling for Tivantinib. MMGBSA favored CPD1 (ΔTOTAL -32.87 ± 4.28 kcal/mol) over Tivantinib (-17.27 ± 2.81 kcal/mol), supported by stronger ΔVDWAALS (-40.76 vs -31.34 kcal/mol) and a smaller ΔG_SOLV (20.28 vs 24.14 kcal/mol). ADMET prediction indicated high intestinal absorption with limited solubility, restricted BBB/CNS permeability, fewer CYP inhibition flags, and no hepatotoxicity alert for CPD1, while hERG II inhibition was flagged for both ligands. DFT showed similar EHOMO (eV) but a narrower ΔE (eV) and higher ω (eV) for CPD1 than for Tivantinib. Conclusions: Integrated modeling prioritizes CPD1 as a Bcl-2-targeting scaffold for lung cancer-relevant studies, supporting structure-guided optimization and experimental verification.