Indexed in SCIE, Scopus, PubMed & PMC
pISSN 1226-4512 eISSN 2093-3827


home Article View

Original Article

Korean J Physiol Pharmacol 2024; 28(1): 59-72

Published online January 1, 2024

Copyright © Korean J Physiol Pharmacol.

Mechanism of Wenshen Xuanbi Decoction in the treatment of osteoarthritis based on network pharmacology and experimental verification

Hankun You1,2,#, Siyuan Song1,2,#, Deren Liu1,2, Tongsen Ren1,2, Song Jiang Yin1,2, Peng Wu1,2, and Jun Mao1,2,*

1Department of Orthopedics, Affiliated Hospital of Nanjing University of Chinese Medicine, 2Department of Orthopedics, Jiangsu Provincial Hospital of Chinese Medicine, Nanjing 210029, Jiangsu, China

Correspondence to:Jun Mao

#These authors contributed equally to this work.

Author contributions: J.M. designed the research. H.Y. analyzed the data and wrote the paper. S.S., D.L., T.R., P.W., and S.J.Y. processed the data. All authors read and approved the submitted version.

Received: October 18, 2023; Revised: November 7, 2023; Accepted: November 14, 2023

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

To investigate the mechanism of Wenshen Xuanbi Decoction (WSXB) in treating osteoarthritis (OA) via network pharmacology, bioinformatics analysis, and experimental verification. The active components and prediction targets of WSXB were obtained from the TCMSP database and Swiss Target Prediction website, respectively. OA-related genes were retrieved from GeneCards and OMIM databases. Protein-protein interaction and functional enrichment analyses were performed, resulting in the construction of the Herb-Component-Target network. In addition, differential genes of OA were obtained from the GEO database to verify the potential mechanism of WSXB in OA treatment. Subsequently, potential active components were subjected to molecular verification with the hub targets. Finally, we selected the most crucial hub targets and pathways for experimental verification in vitro. The active components in the study included quercetin, linolenic acid, methyl linoleate, isobergapten, and beta-sitosterol. AKT1, tumor necrosis factor (TNF), interleukin (IL)-6, GAPDH, and CTNNB1 were identified as the most crucial hub targets. Molecular docking revealed that the active components and hub targets exhibited strong binding energy. Experimental verification demonstrated that the mRNA and protein expression levels of IL-6, IL-17, and TNF in the WSXB group were lower than those in the KOA group (p < 0.05). WSXB exhibits a chondroprotective effect on OA and delays disease progression. The mechanism is potentially related to the suppression of IL-17 and TNF signaling pathways and the down-regulation of IL-6.

Keywords: Inflammation, Molecular docking simulation, Network pharmacology, Osteoarthritis

Osteoarthritis (OA) is the most prevalent chronic degenerative disease of the osteoarticular system in adults, characterized by its intractability to cure even after protracted treatment and its propensity for recurrence. Empirical research indicates that approximately 60% of individuals aged 65 and above exhibit degenerative changes in their hand joints, with 33% of these patients experiencing knee osteoarthritis (KOA) to varying degrees [1,2]. Modern research has demonstrated a close correlation between inflammatory factors and the incidence and progression of OA. Inflammatory mediators facilitate the degradation of the chondrocyte matrix, resulting in the degeneration of OA chondrocytes, and ultimately contributing to OA development and disease progression [3].

The current mainstream medical interventions for OA primarily consist of oral and topical non-steroidal anti-inflammatory drugs (NSAIDs). However, a lack of evidence exists to support the notion that NSAIDs can effectively prevent the progression of OA. Furthermore, NSAIDs are associated with notable adverse effects on cardiovascular and cerebrovascular conditions, as well as the upper digestive system [4].

Several studies have reported that traditional Chinese medicine demonstrates a positive impact on the treatment of OA [5]. Wenshen Xuanbi Formula is a famous prescription created by famous traditional Chinese medicine practitioners in the country and the Department of Orthopedics and Traumatology, Jiangsu Provincial Hospital of Traditional Chinese Medicine. Wenshen Xuanbi Decoction (WSXB) consists of Fuzi, Du Zhong, Bai Zhu, Guizhi, Fuling, Yiyiren, Duhuo, Muxiang, Gancao, and Zexie [6]. Modern pharmacology has shown that aconitine contained in Fuzi can activate opioid receptors, thereby exerting a significant analgesic effect, and simultaneously reducing the level of prostaglandin E2 (PGE2) in inflammatory tissues and its damage to the synovial membrane of joints [7]. Cinnamic aldehyde, the main component of Guizhi, inhibits the NFκB signaling pathway by interfering with the expression of Toll-like receptor 4, thus exerting anti-inflammatory and analgesic effects [8]. The compatibility of Guizhi and Fuzi can significantly increase the pain threshold of rats and protect the articular cartilage [9]. The impact of Duzhong on the LPS-induced inflammatory response of abdominal macrophages in mice was investigated in a previous study. The results demonstrated that Duzhong was capable of inhibiting the production of LPS-induced tumor necrosis factor-α (TNF-α) and interleukin (IL)-6, as well as reducing the increments in COX-2 level and the generation of PGE2 and nitric oxide [10]. Modified WSXB can inhibit chondrocyte degeneration by resisting mitochondrial apoptosis to treat OA [11]. Basic studies have suggested that the mechanism of WSXB in the treatment of OA is related to inflammation-related pathway signals such as the NF-kB signaling pathway [12].

Although clinical and basic experiments have corroborated the efficacy of WSXB in ameliorating OA, the specific mechanism underlying its therapeutic action remains obscure. Network pharmacology, as a rational and efficacious approach to studying traditional Chinese medicine formulations, transcends the erstwhile paradigm of single-component, single-target, and disease-centric strategies. It can serve as a vital instrument for investigating the multi-component, multi-target, and multi-biological functions of Chinese medicine [13]. It can predict the active components and disease targets of compounds at the system, locks on key targets related to drugs and diseases, and establishes multi-level networks such as drug-component-target-disease [14]. With the rapid development of bioinformatics, more and more studies have shown that the occurrence and development of OA are related to a series of genes and signaling pathways [15].

Based on a comprehensive network pharmacology approach, integrating bioinformatics analysis and experimental verification, this study delved into the mechanism of active components in WSXB for the treatment of OA. The findings provide a solid theoretical basis for clinical research. The detailed protocol of our study procedures is presented in Fig. 1.

Figure 1. The protocol of our study procedures.
WSXB, Wenshen Xuanbi Decoction; TCMSP, Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform; OA, osteoarthritis; DEGs, differential genes; PPI, protein-protein interaction; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; qRT-PCR: quantitative real-time polymerase chain reaction.

This study does not involve animal or human experiments, therefore, no additional ethical approval or consent is required.

Network pharmacology analysis

Screening the main active compounds of WSXB: The active components of WSXB were retrieved by the Traditional Chinese Medicine System Pharmacology analysis platform (TCMSP) [16] ( In this study, the active components were screened as much as possible based on the consideration of sample size and data complexity, oral bioavailability ≥ 30% [17], and drug-like properties (drug-likeness) ≥ 0.05 [18] were used as the screening criteria.

Screening targets of WSXB: The active components were converted into SMILES structures by the PubChem database ( [19], and the SMILES structures were imported into the Swiss Target Prediction website ( [20] for prediction of all potential targets of WSXB.

Screening of targets related to OA: The keyword “osteoarthritis” was used for retrieval in the GeneCards database ( [21] and the OMIM database ( [22], and the retrieved potential OA targets were summarized and the duplicate targets were excluded.

Construction of “Herb-Component-Target” network: In this study, the probability TOP50 for each SMILES structural prediction target was used as the standard for screening. The predicted targets of active components and OA-related targets were intersected to establish a target set for WSXB in OA treatment. Relevant data were input into Cytoscape 3.7.2 to generate a “Herb-Component-Target” network. The larger the degree, the greater the importance of the node within the network.

Construction of protein-protein interaction (PPI) network: The target set was imported into the STRING ( [23] to construct a protein-protein network, and “Organism” was set as “Homo sapiens”, the appropriate PPI protein-protein network, and its tsv file were downloaded and visualized by Cytoscape software 3.7.2 [24]. The network was analyzed by the Network Analyzer plug-in, and the hub targets in the PPI network were determined according to the Degree values.

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis: To elucidate the genetic functions and pathways of WSXB for the treatment of OA, we conducted GO and KEGG [25] enrichment analysis through the DAVID database ( [26]. In this study, p < 0.05 was used as the screening standard for GO analysis. p < 0.01 was used as the screening standard for KEGG analysis. Visualize TOP10 using the R software 3.1.08 ( package.

Molecular docking: The hub targets were imported into the PDB database ( [27] and the appropriate protein structures were selected. The SDF files of the significant active components were downloaded from the PubChem database ( [28], the file was obtained by PyMOL visualization software, and the rotatable hydrogen atoms were activated by AutoDock and stored as pdbqt file. Before docking, ligands and receptors need to minimize energy. The ligands and water molecules were removed using PyMOL visualization, and then hydrogenated and charged using AutoDock software to establish a rigid box. The receptor proteins were set as rigid docking, and a genetic algorithm was selected with the maximum eval number as the medium. The docking results were obtained by running autogrid4 and autodock4, revealing the binding energy. The target protein was used as the receptor and the active component was used as the ligand for molecular docking by the molecular docking software AutoDock Vina 1.5.6 [29]. The relatively stable results of molecular docking were selected to draw a 3D map with PyMol software [30].

Bioinformatics analysis

Acquisition of gene expression profiles of OA: GSE55235, GSE55457, and GSE55584 data sets were obtained from the GEO database [31] ( using the GPL96 platform ([HG ‐ U133A] Affymetrix Human Genome U 133ARRAY).

Acquisition and analysis of differential genes (DEGs): We first merged the three data sets using the R software package inSilicoMerging [32]. Then the R sva package ComBat function was used to remove the batch effect. The difference was analyzed using the limma package. In this study, to identify DEGs with significant differences across the three data sets, a stringent threshold of p < 0.01 was established for the screening process. Furthermore, to enrich the number of DEGs, we adopted a more lenient criterion of |logFC| > 0.5 in our analysis [33]. Therefore, the criteria for differential genes are set with p < 0.01 and | logFC | > 0.5.

Functional enrichment analysis of co-DEGs: A venn diagram was used to obtain co-DEGs from 3 data sets and functional enrichment analysis was performed.

Experimental verification in vitro

Preparation of freeze-dried powder of WSXB: According to the clinical dosage of WSXB, Fuzi, Du Zhong, Bai Zhu, Guizhi, Fuling, Yiyiren, Duhuo, Muxiang, Gancao, and Zexie 10 g each. The above medicinal materials were weighed and mixed, soaked in 400 ml of water for 1 h, then boil with strong fire, decoct twice, and combine with the liquid medicines. Finally, concentrate the liquid medicine to 100 ml with slow fire, and the drug concentration is 1.2 g crude drug/ml. Freeze-dry the liquid medicine for 24 h by vacuum freeze-drying method to obtain dried WSXB formula freeze-dried powder. The concentration of WSXB was set according to the previous experimental results of the research group [34].

Culture of fibroblast-like synovial cells (FLS): FLSs were isolated and cultured from the synovial membrane of rat knee joints. The synovial membrane tissue was washed three times in phosphate-buffered saline (PBS), cut into pieces, and digested with 5 mg/ml type I collagenase at 37°C for 4 h, filtered and centrifuged, and then suspended in 100 mm cell culture dish based on Dulbecco's Modified Eagle Medium culture medium containing 10% fetal bovine serum and 1% cyan-streptomycin double antibody, cultured overnight at 37°C in a 5% CO2 cell incubator, and the medium was changed after 24 h. The third to sixth generations of FLS cells were taken for subsequent experiments.

Cytotoxicity test (CCK-8): FLSs were inoculated into 96-well plates (1 × 105 cells/well) and cultured for 24 h. According to the principle of randomized controlled experiment, the experiment was divided into blank group (culture medium and CCK-8 reagent), control group (FLS, culture medium and CCK-8 reagent) and WSXB group (WSXB concentration was 450 μg/ml, 500 μg/ml, 550 μg/ml, 600 μg/ml, 650 μg/ml, 700 μg/ml, 750 μg/ml, and 800 μg/ml), with 3 holes in each group. According to the standard operation of CCK-8 kit, the drug concentration of WSXB was obtained by screening.

FLSs were inoculated into 6-well plates and randomly divided into 3 groups. Blank group: no drug intervention. KOA group: 10 ng ml-1 TGF-β1 was added for 24 h to simulate the synovial fibrosis environment of KOA [35]. WSXB group: After 24 h of TGF-β1 modeling, WSXB freeze-dried powder was given for 24 h.

Quantitative real-time polymerase chain reaction (qRT-PCR): According to the results of network pharmacology and bioinformatics, we selected AKT1, TNF, IL-6, and IL-17 to explore the mechanism of WSXB in the treatment of OA. FLS cells were inoculated into 6-well plates (1 × 105 cells/well) and cultured for 24 h. The blank group, KOA group, and WSXB group were set. The KOA group and the WSXB group were treated with LPS stimulation at the dose of 5 μg/ml for 24 h to induce KOA inflammatory cell model. Then, the WSXB group was added with 2 ml of complete medium containing the concentration was 690.7 μg/ml of WSXB. The blank group was only added with sterile PBS of equal volume and cultured for 24 h. The total RNA of the cells was extracted with RNA extraction reagent and reversely transcribed into cDNA. PCR amplification was performed using cDNA as the template. qRT-PCR was performed using an Applied Biosystems PCR system. The 2-ΔΔ Ct method was used to calculate the relative mRNA expression of the target genes. The primer sequences used for qRT-PCR are listed in Table 1.

Real-time polymerase chain reaction primers
GeneSequence (5′-3′)

Western blotting: Experimental grouping and dosing methods are the same as the qRT-PCR. The intracellular protein concentration of each group was measured by the BCA kit and the standard curve was drawn. After the protein was boiled and denatured, sample electrophoresis was performed and the sample was transferred to the PVDF membrane by wet transfer method. The samples were blocked with 5% skimmed milk powder at room temperature, and the primary antibody was added in a ratio of 1:1,000 and incubated overnight in a shaking table at 4°C. After membrane washing on the next day, the secondary antibody (1:10,000) was added for incubation at room temperature and the color solution was exposed. The relative expression level of the target protein was determined by the ratio of the gray values of the bands of the target protein and internal reference protein using ImageJ software.

Statistical analysis: Statistical analysis was performed using GraphPad Prism Version 8.0. One-way analysis of variance was used for comparison among multiple groups of data with normal distribution and homogeneity of variance, and the Kruskal–Wallis H test was used for comparison among multiple groups. p < 0.05 indicated that the difference was statistically significant.

Network pharmacology analysis

Screening of active components and targets: A total of 253 active components in WSXB were retrieved from the TCMSP database, including 9 Fuzi (Aconiti Lateralis Radix Praeparata), 26 Duzhong (Eucommiae Cortex), 14 Baizhu (Atractylodes macrocephala Koidz), 40 Guizhi (Cinnamomi Ramulus), 11 Fuling (Poria Cocos), 16 Yiyiren (Coicis Semen), 17 Duhuo (Radix Angelicae Biseratae), 28 Muxiang (Aucklandiae Radix), 83 Gancao (Licorice), 9 Zexie (Alisma Orientale) (Supplementary Table 1). The search results obtained from the GeneCards and OMIM databases were combined to obtain 3,807 OA-related targets. The Swiss Target Prediction website was used to predict a total of 1,251 targets of WSXB. After the intersection of the WSXB prediction targets and OA-related targets, 520 potential targets of WSXB for the treatment of OA were obtained (Fig. 2A).

Figure 2. Screening of targets and network construction.
(A) Venn diagram of WSXB and OA targets. (B) Network of Drug-Component-Target. Circle nodes represent the drugs and components in WSXB, and the blue diamond represents the targets of WSXB. WSXB, Wenshen Xuanbi Decoction; OA, osteoarthritis.

Construction of “Herb-Component-Target” network: The active components of WSXB and the relationship between the corresponding targets and common genes of OA were integrated and analyzed by Cytoscape software, which constituted the Network of “Herb-Component-Target” (Fig. 2B). According to the degree value, the TOP5 included quercetin, linolenic acid, methyl linoleate, isobergapten, and beta-sitosterol.

Construction of PPI network: The PPI network consisted of 518 target protein nodes, with a total of 11,192 interaction lines (Fig. 3). The nodes that were more in contact with other nodes in the PPI were the hub targets, and the ones with higher degree values were AKT1 (degree = 275), TNF (degree = 266), IL-6 (degree = 248), GAPDH (degree = 252), and CTNNB1 (degree = 220). These targets interweaved with other targets and played an important role in the PPI network.

Figure 3. The PPI network.
(A) The PPI network constructed in the STRING database. (B) The PPI network visualized by Cytoscape software. Node size and color were set to reflect the degree value. The greater the degree value is, the redder the color is. PPI, protein-protein interaction.

GO and KEGG enrichment analysis: The targets of WSXB in the treatment of OA were analyzed in the DAVID database, and the biological process (BP) included cell population proliferation, and protein phosphorylation (Fig. 4A). Molecular function (MF) included enzyme binding, and protein kinase activity (Fig. 4B). Cellular component (CC) included cell surface, and cell-substrate junction (Fig. 4C). The results of the TOP10 GO enrichment analysis were visualized using the R software package, which showed that WSXB could regulate a variety of BPs and thus played an important role in treating OA.

Figure 4. Bubble diagram for GO and KEGG enrichment analysis.
(A) Biological processes for the targets of WSXB in the treatment of OA. (B) Molecular function for the targets of WSXB in the treatment of OA. (C) Cellular components for the targets of WSXB in the treatment of OA. (D) represents KEGG for the targets of WSXB in the treatment of OA. The bubble size represents the Gene Count, and the bubble color represents the p-value. GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; WSXB, Wenshen Xuanbi Decoction; OA, osteoarthritis.

The results of KEGG enrichment analysis showed that the targets were mainly enriched in the phosphatidylinositol 3-kinase/protein kinase B (PI3K-Akt), MAPK, Ras, cAMP, and TNF signaling pathway. The TOP10 signaling pathways were plotted into bubbles (Fig. 4D) and circle diagrams (Fig. 5).

Figure 5. Circle diagram for KEGG enrichment analysis.
(A) KEGG enrichment analysis of TOP10 (first 5). (B) KEGG enrichment analysis TOP10 (last 5). KEGG, Kyoto Encyclopedia of Genes and Genomes.

Molecular docking: Molecular docking verification was performed between the hub targets (AKT1, TNF, IL-6, GAPDH, CTNNB1) in the PPI network and the key active components. The results showed that the key active components in WSXB had a good binding affinity with hub targets for the treatment of OA (Fig. 6). It could form the optimal complex, indicating that WSXB could play a role by regulating AKT1, TNF, IL-6, GAPDH, and CTNNB1. We selected Quercetin with relatively a stable binding force for the drawing (Fig. 7), and the docking interaction information was shown in Supplementary Table 2.

Figure 6. Heatmap of binding affinity.
The yellower the color is, the more stable the binding force. A represents AKT1, B represents TNF, C represents IL-6, D represents GAPDH, and E represents CTNNB1. 1 represents quercetin, 2 represents linolenic acid, 3 represents methyl linoleate, 4 represents isobergapten, and 5 represents beta-sitosterol.

Figure 7. Schematic diagram of docking between quercetin and hub targets.
Molecular models of quercetin binding to the targets (A) AKT1, (B) TNF, (C) IL-6, (D) GAPDH, (E) CTNNB1.

Bioinformatics analysis

Acquisition of DEGs: Three OA data sets were downloaded from the GEO database, a total of 20 control groups and 27 OA group samples. After the inter-batch difference was removed using the R sva package ComBat function, boxplot, and UMAP plots before and after batch effect removal was drawn (Fig. 8). Limma package was used for screening using p < 0.01, | logFC | > 0.5 as the standard. A total of 4,389 DEGs were obtained, of which 2,334 were significantly up-regulated and 2,055 were significantly down-regulated, heat map and volcano map were plotted (Fig. 9).

Figure 8. Boxplot and UMAP plots before and after batch effect removal.
(A, B) Boxplot plots before and after batch effect removal. (C, D) UMAP plots before and after batch effect removal.

Figure 9. Heat map and volcano map of databases in the GEO.
(A) Heat map and volcano map of GSE55235. (B) Heat map and volcano map of GSE55457. (C) Heat map and volcano map of GSE55584.

Acquisition of co-DEGs and functional enrichment analysis: We combined the DEGs from the 3 datasets to obtain 146 co-DEGs (Fig. 10). And GO and KEGG enrichment analysis was conducted. As suggested from the results, the co-DEGs displayed tight relations to the BP, which included cell migration, leukocyte migration, and response to chemokine. CCs included inflammasome complex, and receptor complex. MF included chemokine activity, and cytokine receptor binding (Fig. 11A–C).

Figure 10. Venn diagram of co-DEGs.
DEGs, differential genes.

Figure 11. Bubble diagram and bubble diagram for KEGG enrichment analysis of co-DEGs.
(A) Biological processes of co-DEGs. (B) Cellular components of co-DEGs. (C) Molecular function of co-DEGs. (D) Bubble diagram for KEGG enrichment analysis of co-DEGs. KEGG, Kyoto Encyclopedia of Genes and Genomes; DEGs, differential genes.

Comparing with the pathways enriched in the WSXB therapeutic OA targets. The co-DEGs were mainly enriched in the chemokine, NF-kappa B signaling pathway, and Th17 cell differentiation (Fig. 11D). Their common pathways of enrichment included the IL-17, TNF, MAPK, JAK-STAT, and PI3K-Akt signaling pathway.

Experimental verification in vitro

Effect of WSXB on the survival rate of FLS: The results of CCK-8 showed that compared with the blank group, the IC50 of WSXB Decoction for inhibiting the growth of FLS was 690.7 μg/ml (Fig. 12A). We selected this concentration of WSXB Decoction for the subsequent experiments.

Figure 12. Results of experimental verification in vitro.
(A) Effect of WSXB on the survival rate of FLS cells. (B) Effects of WSXB on expression of IL-17, IL-6, AKT1, TNF mRNA in FLS cells. (C) Effects of WSXB on expression of IL-17, IL-6, AKT1, TNF proteins in FLS cells. (a) Images of IL-17, IL-6, AKT1, and TNF proteins in FLS cells; (b) Relative protein expression of IL-17, IL-6, AKT1, and TNF in FLS cells. Values are presented as mean ± SD. WSXB, Wenshen Xuanbi Decoction; FLS, fibroblast-like synovial cells; KOA, knee osteoarthritis. Compared with the blank group, * represents p < 0.05; compared with the KOA group, # represents p < 0.05.

Effects of WSXB on expression of IL-17, IL-6, AKT1, TNF mRNA and proteins in FLS: qRT-PCR results showed that compared with the blank group, the mRNA relative expression levels of IL-17, IL-6, AKT1, and TNF in the KOA group were significantly increased (p < 0.05), and compared with the KOA group, the mRNA relative expression levels of IL-17, IL-6, AKT1, and TNF in the WSXB group were significantly reduced (p < 0.05) (Fig. 12B).

Western blotting results showed that compared with the KOA group, the relative protein expression levels of IL-6, IL-17 and TNF in the in the KOA group were significantly increased (p < 0.05), and the relative protein expression levels of IL-6, IL-17 and TNF in the WSXB group were significantly reduced (p < 0.05), while the relative protein expression levels of AKT in the WSXB group were also reduced, but there was no statistical difference (Fig. 12C).

Through network pharmacology research and bioinformatics analysis, we found that quercetin, linolenic acid, methyl linoleate, isobergapten, and beta-sitosterol were considered as the active components. Studies have shown that quercetin can inhibit the apoptosis induced by oxidative stress in rat chondrocytes by inhibiting SIRT1/AMPK-mediated endoplasmic reticulum stress, to achieve the effect of reducing articular cartilage degeneration [36]. Linolenic acid and methyl linoleate are polyunsaturated fatty acids that participate in immune regulation, cell growth, and apoptosis [37]. They can inhibit the expression of tissue factors. Linolenic acid promotes wound healing and reduces the number of inflammatory cells and the levels of inflammatory factors during wound healing [38]. Isobergapten can improve the activity of alkaline phosphatase in osteoblasts and thus has a pro-differentiation effect on osteoblasts [39]. Beta-sitosterol can not only directly promote osteogenesis, but also effectively inhibit osteoclasts to achieve the effect of bone protection [40].

AKT1, TNF, IL-6, GAPDH, and CTNNB1 were considered as the hub targets. AKT1 can regulate neutrophils, and high AKT1 will cause a strong inflammatory reactions [41]. AKT1 is the most strongly expressed AKT subtype in chondrocytes, and it is a kinase that mediates upstream signal transduction, which promotes cartilage calcification by inhibiting the accumulation of pyrophosphate in chondrocytes, thereby controlling bone growth and endochondral ossification by osteophyte formation in OA [42]. TNF-α is an important inflammatory factor in the inflammatory response of chondrocytes [43]. TNF promotes the progression of OA by preventing chondrocytes from synthesizing protein-binding proteoglycans and stimulating chondrocytes to produce matrix metalloproteinases [44]. As a type of cytokine, IL-6 can directly cause damage to the articular cartilage and cartilage matrix [7]. GAPDH is an enzyme in glycolysis reaction, and it is widely distributed in cells of various tissues. The expression of SOX9 in normal cartilage is quite high, reaching approximately 20% of GAPDH levels [45]. There is no report about the relationship between GAPDH and OA in the literature. CTNNB1 can regulate cell apoptosis and proliferation [46]. The high expression of CTNNB1 in synovial tissue of OA not only causes pathological changes in synovial fibroblasts, but also inhibits the activation of autophagy of chondrocytes, leading to the development of OA [47].

KEGG analysis found that the common pathways of enrichment included the IL-17, TNF, MAPK, JAK-STAT, and PI3K-Akt signaling pathway. IL-17, one of many inflammatory cytokines, is present in the synovial membrane and chondrocytes of joints. IL-17 levels in serum and synovial fluid were elevated in patients with OA and positively correlated with the severity of the lesion as seen on X-ray images [48]. Pro-inflammatory TNF-α can stimulate the release of nerve growth factor (NGF), and NGF can regulate the release of TNF-α [35]. TNF-α and NGF in OA patients’ joints interact through molecular cascade, which leads to the aggregation of pro-inflammatory factors and joint swelling. Activation of the MAPK pathway can increase the content of IL-1 and TNF-α in cartilage and synovial fluid [49], and effectively control the progression of OA by inhibiting the downstream reaction caused by the MAPK pathway [50]. The JAK-STAT pathway regulates cell proliferation, differentiation, and apoptosis [51]. The research conducted by Yang indicated that CXCL8 and CXCL11 might enhance the expression of inflammatory cytokines, promote the apoptosis of chondrocytes and inhibit the proliferation by affecting the JAK-STAT signaling pathway. The results suggest that JAK/STAT signaling pathway plays an important role in the pathogenesis of OA [52]. The PI3K/Akt signaling pathway is a very classical signaling pathway that can regulate cell growth, proliferation, differentiation, and other life processes. miRNA-103 inhibits the activity of the PI3K/Akt signaling pathway by reducing the activity of SPHK1, thereby affecting the process of OA and promoting the activity of OA chondrocytes. As a result, PI3K/Akt signaling pathway has a protective effect on OA chondrocytes [53,54].

qRT-PCR results showed that compared with the blank group, the mRNA relative expression levels of IL-17, IL-6, AKT1, and TNF in the KOA group were significantly increased (p < 0.05), and compared with the KOA group, the mRNA relative expression levels of IL-17, IL-6, AKT1, and TNF in the WSXB group were significantly reduced (p < 0.05). Western blotting results showed that compared with the KOA group, the relative protein expression levels of IL-6, IL-17 and TNF in the in the KOA group were significantly increased (p < 0.05), and the relative protein expression levels of IL-6, IL-17 and TNF in the WSXB group were significantly reduced (p < 0.05), while the relative protein expression levels of AKT in the WSXB group were also reduced, but there was no statistical difference. The above results were consistent with our study and previous studies.

Based on the aforementioned analysis, it is speculated that the WSXB treatment for OA primarily targets key player IL-6, via the IL-17 and TNF signaling pathways. This elicits effects that modulate inflammatory responses and regulate immune functions, ultimately curbing the onset and progression of diseases. This reflects the attributes of multifaceted active components, multiple targets, and multiple pathways.

In summary, WSXB exhibits a chondroprotective effect on OA and delays disease progression. The mechanism is potentially related to the suppression of IL-17 and TNF signaling pathways and the down-regulation of IL-6.

However, there are some limitations in this study. Firstly, Chinese medicine is not a simple addition of chemical components. Secondly, the content and concentration of traditional Chinese medicine components affect the efficacy. Thirdly, there are various types of action between Chinese medicine components and targets, and some targets may have multiple states, such as activated state and inactive state. Fourthly, most of the online pharmacological data come from experiments, in view of the development of biomedicine, the number of drug-target effects that have been verified by experiments is limited, and their complete pharmacological effects cannot be revealed. Fifthly, we only carried out experiments in vitro on the predicted signal pathways and targets we selected, there is no further experimental verification of the active components (quercetin, linolenic acid, methyl linoleate, isobergapten, and beta-sitosterol) and the pathways (IL-17, TNF, MAPK, JAK-STAT, and PI3K-Akt) identified in this paper, which is the direction of our future research group, and further verification experiments in vivo are needed.

The third batch of peak academic talent project of Jiangsu Provincial Hospital of Traditional Chinese Medicine (second level) (No.: y2021rc20).

  1. McHugh D, Gil J. Senescence and aging: causes, consequences, and therapeutic avenues. J Cell Biol. 2018;217:65-77.
    Pubmed KoreaMed CrossRef
  2. Prieto-Alhambra D, Judge A, Javaid MK, Cooper C, Diez-Perez A, Arden NK. Incidence and risk factors for clinically diagnosed knee, hip and hand osteoarthritis: influences of age, gender and osteoarthritis affecting other joints. Ann Rheum Dis. 2014;73:1659-1664.
    Pubmed KoreaMed CrossRef
  3. Chen FP, Chang CM, Hwang SJ, Chen YC, Chen FJ. Chinese herbal prescriptions for osteoarthritis in Taiwan: analysis of National Health Insurance dataset. BMC Complement Altern Med. 2014;14:91.
    Pubmed KoreaMed CrossRef
  4. Salvo F, Fourrier-Réglat A, Bazin F, Robinson P, Riera-Guardia N, Haag M, Caputi AP, Moore N, Sturkenboom MC, Pariente A; Investigators of Safety of Non-Steroidal Anti-Inflammatory Drugs: SOS Project. Cardiovascular and gastrointestinal safety of NSAIDs: a systematic review of meta-analyses of randomized clinical trials. Clin Pharmacol Ther. 2011;89:855-866.
    Pubmed CrossRef
  5. Mei ZG, Cheng CG, Zheng JF. Observations on curative effect of high-frequency electric sparkle and point-injection therapy on knee osteoarthritis. J Tradit Chin Med. 2011;31:311-315.
    Pubmed CrossRef
  6. Liu W, Wu YH, Liu XY, Xue B, Shen W, Yang K. Metabolic regulatory and anti-oxidative effects of modified Bushen Huoxue decoction on experimental rabbit model of osteoarthritis. Chin J Integr Med. 2013;19:459-463.
    Pubmed CrossRef
  7. Wu JJ, Guo ZZ, Zhu YF, Huang ZJ, Gong X, Li YH, Son WJ, Li XY, Lou YM, Zhu LJ, Lu LL, Liu ZQ, Liu L. A systematic review of pharmacokinetic studies on herbal drug Fuzi: implications for Fuzi as personalized medicine. Phytomedicine. 2018;44:187-203.
    Pubmed CrossRef
  8. Yoo SR, Kim Y, Lee MY, Kim OS, Seo CS, Shin HK, Jeong SJ. Gyeji-tang water extract exerts anti-inflammatory activity through inhibition of ERK and NF-κB pathways in lipopolysaccharide-stimulated RAW 264.7 cells. BMC Complement Altern Med. 2016;16:390.
    Pubmed KoreaMed CrossRef
  9. Yang M, Ji X, Zuo Z. Relationships between the toxicities of Radix Aconiti Lateralis Preparata (Fuzi) and the toxicokinetics of its main diester-diterpenoid alkaloids. Toxins (Basel). 2018;10:391.
    Pubmed KoreaMed CrossRef
  10. Kim MC, Kim DS, Kim SJ, Park J, Kim HL, Kim SY, Ahn KS, Jang HJ, Lee SG, Lee KM, Hong SH, Um JY. Eucommiae cortex inhibits TNF-α and IL-6 through the suppression of caspase-1 in lipopolysaccharide-stimulated mouse peritoneal macrophages. Am J Chin Med. 2012;40:135-149.
    Pubmed CrossRef
  11. Feng C, Zhao M, Jiang L, Hu Z, Fan X. Mechanism of Modified Danggui Sini Decoction for knee osteoarthritis based on network pharmacology and molecular docking. Evid Based Complement Alternat Med. 2021;2021:6680637.
    Pubmed KoreaMed CrossRef
  12. Xu Y, Li H, He X, Huang Y, Wang S, Wang L, Fu C, Ye H, Li X, Asakawa T. Identification of the key role of NF-κB signaling pathway in the treatment of osteoarthritis with Bushen Zhuangjin Decoction, a verification based on network pharmacology approach. Front Pharmacol. 2021;12:637273.
    Pubmed KoreaMed CrossRef
  13. Zhang X, Gao R, Zhou Z, Sun J, Tang X, Li J, Zhou X, Shen T. Uncovering the mechanism of Huanglian-Wuzhuyu herb pair in treating nonalcoholic steatohepatitis based on network pharmacology and experimental validation. J Ethnopharmacol. 2022;296:115405.
    Pubmed CrossRef
  14. Ye J, Li L, Hu Z. Exploring the molecular mechanism of action of Yinchen Wuling powder for the treatment of hyperlipidemia, using network pharmacology, molecular docking, and molecular dynamics simulation. Biomed Res Int. 2021;2021:9965906.
    Pubmed KoreaMed CrossRef
  15. Chen D, Shen J, Zhao W, Wang T, Han L, Hamilton JL, Im HJ. Osteoarthritis: toward a comprehensive understanding of pathological mechanism. Bone Res. 2017;5:16044.
    Pubmed KoreaMed CrossRef
  16. Ru J, Li P, Wang J, Zhou W, Li B, Huang C, Li P, Guo Z, Tao W, Yang Y, Xu X, Li Y, Wang Y, Yang L. TCMSP: a database of systems pharmacology for drug discovery from herbal medicines. J Cheminform. 2014;6:13.
    Pubmed KoreaMed CrossRef
  17. Li J, Zhao P, Li Y, Tian Y, Wang Y. Systems pharmacology-based dissection of mechanisms of Chinese medicinal formula Bufei Yishen as an effective treatment for chronic obstructive pulmonary disease. Sci Rep. 2015;5:15290.
    Pubmed KoreaMed CrossRef
  18. Xu X, Zhang W, Huang C, Li Y, Yu H, Wang Y, Duan J, Ling Y. A novel chemometric method for the prediction of human oral bioavailability. Int J Mol Sci. 2012;13:6964-6982.
    Pubmed KoreaMed CrossRef
  19. Cincilla G, Thormann M, Pons M. Structuring chemical space: similarity-based characterization of the PubChem database. Mol Inform. 2010;29:37-49.
    Pubmed CrossRef
  20. Gfeller D, Grosdidier A, Wirth M, Daina A, Michielin O, Zoete V. SwissTargetPrediction: a web server for target prediction of bioactive small molecules. Nucleic Acids Res. 2014;42:W32-W38.
    Pubmed KoreaMed CrossRef
  21. Lin Y, Hu Z. Bioinformatics analysis of candidate genes involved in ethanol-induced microtia pathogenesis based on a human genome database: GeneCards. Int J Pediatr Otorhinolaryngol. 2021;142:110595.
    Pubmed CrossRef
  22. van Triest HJ, Chen D, Ji X, Qi S, Li-Ling J. PhenOMIM: an OMIM-based secondary database purported for phenotypic comparison. Annu Int Conf IEEE Eng Med Biol Soc. 2011;2011:3589-3592.
    Pubmed CrossRef
  23. Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, Simonovic M, Doncheva NT, Morris JH, Bork P, Jensen LJ, Mering CV. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 2019;47:D607-D613.
    Pubmed KoreaMed CrossRef
  24. Reimand J, Isserlin R, Voisin V, Kucera M, Tannus-Lopes C, Rostamianfar A, Wadi L, Meyer M, Wong J, Xu C, Merico D, Bader GD. Pathway enrichment analysis and visualization of omics data using g:Profiler, GSEA, Cytoscape and EnrichmentMap. Nat Protoc. 2019;14:482-517.
    Pubmed KoreaMed CrossRef
  25. Kanehisa M, Goto S. KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28:27-30.
    Pubmed KoreaMed CrossRef
  26. Sherman BT, Huang da W, Tan Q, Guo Y, Bour S, Liu D, Stephens R, Baseler MW, Lane HC, Lempicki RA. DAVID Knowledgebase: a gene-centered database integrating heterogeneous gene annotation resources to facilitate high-throughput gene functional analysis. BMC Bioinformatics. 2007;8:426.
    Pubmed KoreaMed CrossRef
  27. Liu Z, Li Y, Han L, Li J, Liu J, Zhao Z, Nie W, Liu Y, Wang R. PDB-wide collection of binding data: current status of the PDBbind database. Bioinformatics. 2015;31:405-412.
    Pubmed CrossRef
  28. Kim S, Chen J, Cheng T, Gindulyte A, He J, He S, Li Q, Shoemaker BA, Thiessen PA, Yu B, Zaslavsky L, Zhang J, Bolton EE. PubChem in 2021: new data content and improved web interfaces. Nucleic. Acids Res. 2021;49:D1388-D1395.
    Pubmed KoreaMed CrossRef
  29. Seeliger D, de Groot BL. Ligand docking and binding site analysis with PyMOL and Autodock/Vina. J Comput Aided Mol Des. 2010;24:417-422.
    Pubmed KoreaMed CrossRef
  30. Mooers BHM. Shortcuts for faster image creation in PyMOL. Protein Sci. 2020;29:268-276.
    Pubmed KoreaMed CrossRef
  31. Toro-Domínguez D, Martorell-Marugán J, López-Domínguez R, García-Moreno A, González-Rumayor V, Alarcón-Riquelme ME, Carmona-Sáez P. ImaGEO: integrative gene expression meta-analysis from GEO database. Bioinformatics. 2019;35:880-882.
    Pubmed CrossRef
  32. Taminau J, Meganck S, Lazar C, Steenhoff D, Coletta A, Molter C, Duque R, de Schaetzen V, Weiss Solís DY, Bersini H, Nowé A. Unlocking the potential of publicly available microarray data using inSilicoDb and inSilicoMerging R/Bioconductor packages. BMC Bioinformatics. 2012;13:335.
    Pubmed KoreaMed CrossRef
  33. Huang Y, Yang DD, Li XY, Fang DL, Zhou WJ. ZBP1 is a significant pyroptosis regulator for systemic lupus erythematosus. Ann Transl Med. 2021;9:1773.
    Pubmed KoreaMed CrossRef
  34. Fan Y, Xia T, Shen J. The study of protective efficacy of Wenshenxuanbi Decoction on chondrocyte of knee osteoarthritis in rats. Chin J Med & Orth. 2021;6:5-8.
  35. Ganesan R, Doss HM, Rasool M. Majoon ushba, a polyherbal compound, suppresses pro-inflammatory mediators and RANKL expression via modulating NFкB and MAPKs signaling pathways in fibroblast-like synoviocytes from adjuvant-induced arthritic rats. Immunol Res. 2016;64:1071-1086.
    Pubmed CrossRef
  36. Feng K, Chen Z, Pengcheng L, Zhang S, Wang X. Quercetin attenuates oxidative stress-induced apoptosis via SIRT1/AMPK-mediated inhibition of ER stress in rat chondrocytes and prevents the progression of osteoarthritis in a rat model. J Cell Physiol. 2019;234:18192-18205.
    Pubmed CrossRef
  37. Luo JS, Kammerer R, von Kleist S. Comparison of the effects of immunosuppressive factors from newly established colon carcinoma cell cultures on human lymphocyte proliferation and cytokine secretion. Tumour Biol. 2000;21:11-20.
    Pubmed CrossRef
  38. Park HG, Heo W, Kim SB, Kim HS, Bae GS, Chung SH, Seo HC, Kim YJ. Production of conjugated linoleic acid (CLA) by Bifidobacterium breve LMC520 and its compatibility with CLA-producing rumen bacteria. J Agric Food Chem. 2011;59:984-988.
    Pubmed CrossRef
  39. Zhang HL, Zheng L, Bian YZ, Zhang JF, Wang JH. Effects of 8-methoxypsoralen on proliferation and differentiation of cultured osteoblasts in vitro. Nat Prod Res Dev. 2011;5:927-930.
  40. Ueland T, Yndestad A, Øie E, Florholmen G, Halvorsen B, Frøland SS, Simonsen S, Christensen G, Gullestad L, Aukrust P. Dysregulated osteoprotegerin/RANK ligand/RANK axis in clinical and experimental heart failure. Circulation. 2005;111:2461-2468.
    Pubmed CrossRef
  41. Liu G, Bi Y, Wang R, Shen B, Zhang Y, Yang H, Wang X, Liu H, Lu Y, Han F. Kinase AKT1 negatively controls neutrophil recruitment and function in mice. J Immunol. 2013;191:2680-2690.
    Pubmed CrossRef
  42. Fukai A, Kawamura N, Saito T, Oshima Y, Ikeda T, Kugimiya F, Higashikawa A, Yano F, Ogata N, Nakamura K, Chung UI, Kawaguchi H. Akt1 in murine chondrocytes controls cartilage calcification during endochondral ossification under physiologic and pathologic conditions. Arthritis Rheum. 2010;62:826-836.
    Pubmed CrossRef
  43. Saklatvala J. Tumour necrosis factor alpha stimulates resorption and inhibits synthesis of proteoglycan in cartilage. Nature. 1986;322:547-549.
    Pubmed KoreaMed CrossRef
  44. Séguin CA, Bernier SM. TNFalpha suppresses link protein and type II collagen expression in chondrocytes: role of MEK1/2 and NF-kappaB signaling pathways. J Cell Physiol. 2003;197:356-369.
    Pubmed CrossRef
  45. Haag J, Gebhard PM, Aigner T. SOX gene expression in human osteoarthritic cartilage. Pathobiology. 2008;75:195-199.
    Pubmed CrossRef
  46. Li Z, Chen S, Chen S, Huang D, Ma K, Shao Z. Moderate activation of Wnt/β-catenin signaling promotes the survival of rat nucleus pulposus cells via regulating apoptosis, autophagy, and senescence. J Cell Biochem. 2019;120:12519-12533.
    Pubmed CrossRef
  47. Wang J, Wang Y, Zhang H, Gao W, Lu M, Liu W, Li Y, Yin Z. Forkhead box C1 promotes the pathology of osteoarthritis by upregulating β-catenin in synovial fibroblasts. FEBS J. 2020;287:3065-3087.
    Pubmed CrossRef
  48. Chen B, Deng Y, Tan Y, Qin J, Chen LB. Association between severity of knee osteoarthritis and serum and synovial fluid interleukin 17 concentrations. J Int Med Res. 2014;42:138-144.
    Pubmed CrossRef
  49. Chen Y, Shou K, Gong C, Yang H, Yang Y, Bao T. Anti-inflammatory effect of geniposide on osteoarthritis by suppressing the activation of p38 MAPK signaling pathway. Biomed Res Int. 2018;2018:8384576. Erratum in: Biomed Res Int. 2022;2022:9814323.
    Pubmed KoreaMed CrossRef
  50. Shi J, Zhang C, Yi Z, Lan C. Explore the variation of MMP3, JNK, p38 MAPKs, and autophagy at the early stage of osteoarthritis. IUBMB Life. 2016;68:293-302.
    Pubmed CrossRef
  51. Morris R, Kershaw NJ, Babon JJ. The molecular details of cytokine signaling via the JAK/STAT pathway. Protein Sci. 2018;27:1984-2009.
    Pubmed KoreaMed CrossRef
  52. Leisengang S, Gluding D, Hörster J, Peek V, Ott D, Rummel C, Schmidt MJ. Expression of adipokines and adipocytokines by epidural adipose tissue in cauda equina syndrome in dogs. J Vet Intern Med. 2022;36:1373-1381.
    Pubmed KoreaMed CrossRef
  53. Chen H, Wu G, Sun Q, Dong Y, Zhao H. Hyperbaric oxygen protects mandibular condylar chondrocytes from interleukin-1β-induced apoptosis via the PI3K/AKT signaling pathway. Am J Transl Res. 2016;8:5108-5117.
    Pubmed KoreaMed
  54. Zhang Q, Lai S, Hou X, Cao W, Zhang Y, Zhang Z. Protective effects of PI3K/Akt signal pathway induced cell autophagy in rat knee joint cartilage injury. Am J Transl Res. 2018;10:762-770.
    Pubmed KoreaMed