Acta Biochimica et Biophysica Sinica•2025•DOI: 10.3724/abbs.2024138
Chemoresistance is the primary reason for poor prognosis in patients with pancreatic cancer (PC). Recent studies have indicated that ferroptosis may improve chemoresistance, but the underlying mechanisms remain unclear. In this study, significant upregulation of heat shock protein 90α (Hsp90α) expression is detected in the peripheral blood and tissue samples of patients with chemoresistant PC. Further studies reveal that Hsp90α promotes the proliferation, migration, and invasion of a chemoresistant pancreatic cell line (Panc-1-gem) by suppressing ferroptosis. Hsp90α competitively binds to Kelch-like ECH-associated protein 1 (Keap1), liberating nuclear factor erythroid 2-related factor 2 (Nrf2) from Keap1 sequestration. Nrf2 subsequently translocates into the nucleus and activates the glutathione peroxidase 4 (GPX4) pathway, thereby suppressing ferroptosis. This process further worsens the chemoresistance of PC cells. This study provides valuable insight into potential molecular targets to overcome chemoresistance in PC. It sheds light on the intricate mechanisms linking Hsp90α and ferroptosis to chemoresistance in PC and provides a theoretical foundation for the development of novel therapeutic strategies.
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: 10.12307/2026.21395
BACKGROUND: With the widespread application of total knee arthroplasty in China, non-infectious prosthesis loosening has become one of the main reasons for postoperative revision. For complex loosening cases, traditional revision techniques are relatively complex and difficult. The application of artificial intelligence-assisted preoperative planning combined with Sleeve extension rods and mobile bearing tray prostheses provides a new solution for precisely reconstructing joint stability and mechanical alignment, which is expected to improve the long-term outcomes of revision surgeries. OBJECTIVE: To explore the mid-and early-term clinical efficacy of revision surgery for non-infectious total knee prosthesis loosening using Sleeve extension rods combined with mobile bearing tray prostheses under the assistance of artificial intelligence-assisted preoperative planning. METHODS: A retrospective analysis was conducted on 17 patients with non-infectious prosthesis loosening after total knee arthroplasty in Department of Orthopedics and Traumatology, Jiangsu Provincial Hospital of Traditional Chinese Medicine from January 2021 to September 2024. There were 6 males and 11 females, aged 59-81 years (mean 72.06±6.10 years). The affected side was left in 8 cases and right in 9 cases. The duration of prosthesis use ranged from 2 to 22 years (mean 10.53±4.60 years). All cases were revisions after primary arthroplasty. Revision reasons included periprosthetic osteolysis with liner wear in 15 cases, femoral condyle old fracture causing loosening in 1 case, and tibial plateau prosthesis fracture in 1 case. According to AORI classification, there were 13 cases of type IIB and 4 cases of type IIA. The artificial intelligence-designed prosthesis sizes were recorded and compared with the actual intraoperative sizes. Visual analog scale (VAS) score, American Knee Society knee score, hip-knee-ankle angle, and knee range of motion were compared preoperatively, at 1 week, 6 months, and 12 months postoperatively to evaluate surgical efficacy. RESULTS AND CONCLUSION: (1) Except for one patient with poor incision healing at 1 month postoperatively, all other patients recovered well without adverse events such as deep vein thrombosis, infection, periprosthetic fracture, or prosthesis loosening. (2) The follow-up period ranged from 6 to 41 months (mean 23.63±12.50 months). At the last follow-up, 2 patients had slight soreness and discomfort after activity, and 1 patient had obvious pain during activity. (3) At the last follow-up after revision, resting pain, exercise pain VAS scores, affected side range of motion, hip-knee-ankle angle, and American Knee Society knee score were significantly improved compared with preoperative values (P < 0.01). (4) The matching rate of artificial intelligence preoperative design for femoral condyle and tibial plateau prostheses was 85%, and the matching rate for other components was 62%. (5) The use of Sleeve+ extension rod combined with MBT prosthesis for non-infectious knee revision can effectively correct joint alignment, fill bone defects, improve pain and knee range of motion, and enhance patients' quality of life, with good early and mid-term efficacy. Artificial intelligence preoperative planning generally helps improve surgical accuracy, reduce revision difficulty, minimize risks, and promote postoperative recovery.