Computational Unveiling of the Mechanism of Action of Selected Steroidal Saponins from Trillium tschonoskii as Potential Anti-hepatocarcinoma Agents

Hepatocellular carcinoma imposes a high global disease burden and remains challenging to manage because current modalities can be limited by suboptimal efficacy and treatment-related toxicity. Since apoptosis evasion is a hallmark of this malignancy, inhibition of anti-apoptotic Bcl-2 represents a rational strategy to restore cell-death signaling and improve therapeutic sensitivity. Four selected steroidal saponins from Trillium tschonoskii including saponin Tg (CPD1), pennogenin 3-O-β-chacotrioside (CPD2), pennogenin 3-O-α-L-rhamnopyranosyl-(1 → 2)-β-D-glucopyranoside (CPD3), and pennogenin 3-O-β-D-glucopyranoside (CPD4), were therefore evaluated as potential Bcl-2 inhibitors (PDB: 6GL8) using molecular docking, 100 ns molecular dynamics, MMGBSA binding free-energy estimation, ADMET prediction, and DFT descriptors, with Simvastatin as reference. Docking prioritized CPD2 (–9.71 kcal/mol), exceeding CPD1-CPD4 and Simvastatin (–7.72 kcal/mol), with hydrogen bonds involving Asp111, Gln118, and Arg129 plus extensive hydrophobic contacts. Molecular dynamics supported stable complex formation, and CPD2 exhibited a richer hydrogen-bond profile and slightly reduced solvent exposure relative to Simvastatin. MMGBSA favored CPD2 (ΔGtot = –35.87 ± 6.28 kcal/mol) over Simvastatin (–27.21 ± 2.88 kcal/mol). ADMET indicated improved solubility and lower clearance for CPD2, but weaker intestinal absorption and limited BBB/CNS permeation. Both ligands were predicted to be CYP3A4 substrates, Ames and hepatotoxicity negative, with a hERG II flag. DFT suggested slightly greater electronic stability and electrophilicity for CPD2.