The study, published in the Journal of Cellular and Molecular Medicine, builds on previous work showing that topical Sanse Powder patches alleviate synovial pathology in knee osteoarthritis patients. However, long-term topical use carries drawbacks such as local skin damage and allergic reactions. To refine the therapy, the team prepared an in vitro release medium using a fully automated transdermal diffusion system held at 32°C to mimic human skin surface temperature, then analysed what actually passes through the release apparatus. Ultra-high performance liquid chromatography tandem mass spectrometry identified 172 compounds in the original powder, dominated by phenolic acids, flavonoids and alkaloids, but only 54 compounds in the release medium, mainly flavonoids, phenolic acids and coumarins. Among the top four constituents, casticin stood out for further investigation.

To predict what casticin might do, the researchers turned to network pharmacology, a computational approach that maps the interactions between drug components and disease targets. Casticin contributed 26 drug targets, while the other major components, isoorientin, chlorogenic acid and salvianolic acid B, contributed 20, 3 and 45 targets respectively, yielding 75 unique drug targets after deduplication. Cross-referencing these against 1886 osteoarthritis-related targets collected from GeneCards, DrugBank, MalaCards, PharmGKB and the Therapeutic Targets Database produced 52 intersecting targets. Protein-protein interaction network analysis revealed 14 genes tied to the PI3K-Akt signalling pathway, and molecular docking showed that casticin binds strongly to AKT1, BCL2L1, BCL2, MTOR, GSK3B and CDKN1A, hinting at a role in cell survival and cell cycle control.

The team then validated these predictions in rats. Sixty male Sprague-Dawley rats underwent anterior cruciate ligament transection to establish knee osteoarthritis, and were treated for 28 consecutive days with either the Sanse Powder release medium or casticin at low, medium and high doses of 0.1, 0.2 and 0.4 mg/kg/day. ELISA assays of peripheral blood showed that both interventions significantly reduced the elevated levels of IL-1β, IL-6, CXCL1 and CXCL3 that accompany disease progression. Histopathological staining with haematoxylin and eosin, Masson and Sirius Red revealed less fibrosis and inflammatory cell infiltration in the synovium, with lower Krenn scores. Western blotting and quantitative PCR confirmed that casticin reduced protein and mRNA expression of the fibrosis markers MMP3, MMP9, TGF-β and collagen I in a dose-dependent manner.

Crucially, the researchers also measured cellular senescence, the state in which cells stop dividing but secrete inflammatory factors, proteases and pro-fibrotic molecules, a phenomenon known as the senescence-associated secretory phenotype. In the synovial tissue of treated rats, casticin lowered the expression of the senescence markers p53 and p21/CDKN1A while upregulating the cell cycle proteins CDC2 and CCNA1. This pattern suggests that casticin does not merely suppress inflammation downstream but may actually reverse the senescent state of synovial fibroblasts, the cells that drive much of the chronic inflammation and fibrosis in knee osteoarthritis.

To dissect the mechanism further, the team isolated primary rat synovial fibroblast-like cells and exposed them to lipopolysaccharide to mimic the osteoarthritic environment. Casticin at 2.5, 5 and 10 μg/mL, concentrations that did not impair cell viability, reduced the LPS-induced upregulation of IL-1β, IL-6, CXCL1 and CXCL3 in a dose-dependent fashion. Fluorescent probes revealed that LPS treatment increased reactive oxygen species, decreased mitochondrial membrane potential and accelerated cellular senescence, but casticin reversed all three effects, reducing ROS accumulation, repairing mitochondrial dysfunction and lowering the proportion of senescence-positive cells.

Transcriptomic sequencing provided a genome-wide view of these changes. Principal component analysis showed that casticin-treated cells clustered closer to healthy controls than to LPS-treated cells. In the osteoarthritic state, 3921 genes were upregulated and 1904 downregulated, whereas casticin intervention upregulated 2045 and downregulated 5058 genes. KEGG enrichment analysis pointed to focal adhesion, ECM-receptor interaction, PI3K-Akt signalling, TNF signalling, p53 signalling and TGF-beta signalling as the pathways most affected by casticin, with genes at the FAK and PAK nodes predominantly downregulated after treatment, partially reversing the transcriptional changes seen in disease.

Western blotting then confirmed that the ratios of phosphorylated to total FAK, PI3K and AKT were significantly elevated in both osteoarthritic synovial tissue and LPS-treated fibroblasts, but casticin markedly reduced these ratios. Molecular docking of casticin into the FAK kinase domain yielded a predicted binding affinity of −8.4 kcal/mol, with conventional hydrogen bonds to Glu430 and Thr503 at distances of 2.5 and 2.9 Å, alongside van der Waals, alkyl and π-sigma interactions. The researchers caution that docking alone does not establish direct binding or inhibition of FAK kinase activity, and an indirect effect through upstream integrin-FAK signalling cannot be excluded.

To test whether the FAK/PI3K/AKT axis is genuinely required for casticin’s effects, the team used two pharmacological tools that push the pathway in opposite directions. VSLRGDTRG acetate, an integrin-binding peptide, maintained FAK phosphorylation and weakened casticin’s ability to dephosphorylate FAK, PI3K and AKT, restoring fibrosis markers, senescence proteins, ROS levels and inflammatory cytokines. Conversely, Y-15, a FAK phosphorylation inhibitor, enhanced casticin’s effects, further suppressing the pathway and amplifying the reductions in senescence and inflammation. These gain-of-function and loss-of-function experiments provide strong evidence that casticin acts through this signalling axis rather than through an unrelated mechanism.

The findings carry real translational promise. By identifying casticin as the principal active component of a traditional herbal preparation and clarifying its mechanism of action, the study offers a path toward more targeted and better-tolerated therapies for knee osteoarthritis, a condition for which there is currently no cure and where treatment relies on symptom management or, in severe cases, joint replacement. The authors note that future work should include validation in different animal models and human-derived samples, systematic pharmacokinetic and safety assessments, and exploration of how the FAK/PI3K/AKT pathway interacts with other signalling networks, using gene editing approaches to clarify the roles of key molecules. If those steps succeed, a compound drawn from a centuries-old herbal formula could become the basis for a modern, mechanism-driven treatment for one of the world’s most burdensome joint diseases.

Subject of Research: Casticin from Sanse Powder inhibits FAK/PI3K/AKT phosphorylation to reduce synovial senescence and inflammation in knee osteoarthritis

Article Title: Casticin From Sanse Powder‐ In Vitro Release Medium Inhibits FAK/PI3K/AKT Phosphorylation Reducing Synovial Senescence and Alleviates Inflammation in Knee Osteoarthritis

Article References: Liu, J., Wu, P., Jiang, R., Jie, L., Shi, L., Fu, H., Wei, Y., Liu, D., Wang, P., Yin, S., & Mao, J. (2026). Casticin From Sanse Powder‐ In Vitro Release Medium Inhibits FAK / PI3K / AKT Phosphorylation Reducing Synovial Senescence and Alleviates Inflammation in Knee Osteoarthritis. Journal of Cellular and Molecular Medicine, 30(19), Article e71391. https://doi.org/10.1111/jcmm.71391

Image Credits: AI Generated

DOI: 10.1111/jcmm.71391

Keywords: knee osteoarthritis, casticin, Sanse Powder, synovial inflammation, cellular senescence, FAK/PI3K/AKT signalling, traditional Chinese medicine, network pharmacology, Sanse, Powder, Vitro, Release