Chinese Traditional and Herbal Drugs•2026•DOI: 10.7501/j.issn.0253-2670.2026.16.20261605
Property prediction of traditional Chinese medicine (TCM) molecules remains constrained by their complex ring systems and conformational flexibility. Conventional graph neural networks (GNNs) treat molecules as topological graphs, discarding bond length and angle strain information critical for accurate property estimation. This study introduces the flexibility-ring enhanced graph neural network (FRGNN), which augments the graph isomorphism network (GIN) with edge descriptors encoding bond length strain and angle strain, and incorporates multiple favorable conformations to construct multi-graph data. The model was evaluated on two TCM databases comprising 37,822 molecules across seven key molecular properties, benchmarked against three state-of-the-art (SOTA) GNN models and two basic GNN models. FRGNN achieved an average root mean square error (RMSE) reduction of 8.63% relative to the second-best model across all seven tasks. For molecules containing polycyclic and macrocyclic structures, the RMSE reduction reached 10.04%. These results demonstrate that FRGNN outperforms existing SOTA small-molecule property prediction models on TCM compounds, offering a robust computational approach for complex natural product characterization. The incorporation of flexibility and ring-specific descriptors addresses a critical gap in molecular representation learning, enabling more accurate predictions for structurally diverse TCM constituents.
Chinese Journal of Tissue Engineering Research•2026•DOI: 10.12307/2026.21492
BACKGROUND: Previous studies from our group have shown that triptolide exerts protective effects on nerve cells and alleviates symptoms of neurodegenerative diseases. However, whether it acts by improving mitochondrial dynamic abnormalities requires further investigation. OBJECTIVE: To explore the effect and mechanism of triptolide in regulating the mitochondrial fusion-fission balance to mitigate hydrogen peroxide (H₂O₂)-induced apoptosis in SH-SY5Y cells. METHODS: Human neuroblastoma SH-SY5Y cells were cultured and divided into three groups: control group, model group (200 μmol/L H₂O₂), and triptolide group (2.5 nmol/L triptolide + 200 μmol/L H₂O₂). After 24 hours of intervention, oxidative stress markers (superoxide dismutase activity and malondialdehyde levels), mitochondrial membrane potential, and apoptosis levels were measured. Western blot was used to detect the expression of apoptosis-related proteins, mitochondrial dynamics-related proteins, and respiratory chain-related proteins. Immunofluorescence staining was used to detect the expression of phosphorylated dynamin-related protein 1, optic atrophy protein 1, cytochrome C oxidase 1, and ATP synthase F1 subunit alpha. RESULTS AND CONCLUSION: Compared with the control group, the model group showed significantly decreased superoxide dismutase activity, mitochondrial membrane potential, anti-apoptotic protein Bcl-2, mitochondrial fusion proteins 1 and 2, optic atrophy protein 1, and oxidative phosphorylation complex proteins (NADH dehydrogenase [ubiquinone] iron-sulfur protein 8, ubiquinol-cytochrome c reductase core protein 2, cytochrome c oxidase 1, succinate dehydrogenase B, ATP synthase F1 subunit alpha) (P < 0.05). Meanwhile, malondialdehyde levels, pro-apoptotic proteins Bax and Caspase-3, mitochondrial fission protein 1, phosphorylated dynamin-related protein 1 expression, and apoptosis rate were significantly increased (P < 0.05). Compared with the model group, triptolide intervention reduced malondialdehyde levels, increased superoxide dismutase activity and mitochondrial membrane potential, promoted fusion protein expression, inhibited fission protein expression, increased oxidative phosphorylation complex protein levels, and decreased apoptosis rate (P < 0.05). These results confirm that triptolide can regulate mitochondrial dynamic imbalance to alleviate H₂O₂-induced apoptosis in SH-SY5Y cells.