For the purpose of minimizing animal suffering, the research was conducted as per the guiding principles and protocols established by the Animal Care and Use Committee for Teaching and Research at the Tongji Medical College of Huazhong University of Science and Technology.

FKA (T3S0737, purity: 99.41%) and 3-Methyladenine (3 MA) were acquired from TOPSCIENCE (Shanghai, China), and the mouse IL-1β cytokine was obtained from the R&D system (501-RL-010, United States). Safranin O solution were provided by Solarbio (Beijing, China). The primary antibodies for iNOS, COX2, Col2, Aggrecan, Sox9, and immunofluorescence secondary antibodies were acquired from Abcam (Shanghai, China). Atg12, Beclin, LC3I/II, P21, P16, Bax, Bcl2, GAPDH, and antibodies for all pathways were purchased from CST (Beverly, MA, United States). Corresponding primary antibodies for MMP3/13 and Nrf2/HO-1/NQO1 were supplied by the Proteintech Group (Wuhan, Hubei, China). ADAMTS5 primary antibody, secondary antibody, phosphate-buffered saline (PBS), trypsin, collagenase type II, the CCK8 assay kit, bovine serum albumin (BSA), and protein extraction kit were ordered and acquired from Boster Biological Technology (Wuhan, Hubei, China). RFP-GFP-LC3-adenovirus was purchased from Hanbio (Shanghai, China). The Senescence β-galactosidase staining kit and the Annexin V-FITC apoptosis detection kit were supplied by Beyotime (Shanghai, China).

In our study, five-day-old male C57BL/6J mice were sacrificed to obtain chondrocytes. The cartilage was obtained from around the knee joint region of the mice and was cut into granules after removing the surrounding synovial tissue. The granules were digested with trypsin for half an hour. Later, 0.2% type II collagenase was added for digestion for a period of 6 h. Subsequently, the cell suspension was concentrated by the process of centrifugation at 300g for a period of 5 min to obtain primary chondrocytes. Moreover, the incubation of primary chondrocytes was performed at a temperature of 37°C, with 5% CO₂ in DMEM/F12 (1:1) medium containing penicillin/streptomycin solution (1%) and FBS (10%). The cells of the third passage were utilized for subsequent experiments.

For the measurement of the viability of cells, the CCK8 assay was employed. Mouse chondrocytes were seeded at a density of 1×10⁴ cells/well in a 96-well plate containing the required culture medium. Following cell adhesion, 0.1 ml of medium with or without IL-1β (5 ng/ml) was added at various concentrations of FKA (0, 5, 10, 20, and 40 μM) for a period of 24 h. Subsequently, an addition of 10 μl of CCK 8 solution was carried out in all the wells, and incubation was performed for a period of 1 h. Thereafter, we employed a microplate reader (Bio-Rad, Richmond, CA, United States) to measure the viability of cells by measuring the absorbance at 450 nm.

Seeding of chondrocytes was performed in 24-well plates. When chondrocytes reached 80% confluency, FKA with a concentration of 40 μM was administered alone or in combination with IL-1β (5 ng/ml). Cells were incubated for 24 h and washed with PBS. Cells were then fixed with 4% paraformaldehyde, and after discarding the cell fixative, they were stained with 0.5% Safranin O reagent for 2 h. The dye was finally removed and washed once again with distilled water.

The protein extraction was performed by first washing the mouse chondrocytes seeded in a 6-well plate using PBS thrice, followed by the addition of 100 μl RIPA lysate buffer, 1 μl phosphatase inhibitor, and protease inhibitor to each well. Additionally, the protein concentration was measured by employing the Bicinchoninic Acid Assay. Moreover, 25 μg of protein samples were added to each well in the SDS-PAGE gel (8.0–12.5%) for electrophoresis. The electroblotting method was employed for the separation of protein bands and transferring onto a polyvinylidene membrane, and blocking of the membrane was performed at room temperature using BSA (5%) for a period of 1 h, and overnight incubation of the target band was carried out in the corresponding primary antibody at a temperature of 4°C. Washing of the membrane was conducted in TBST three times, and a second incubation was performed in the corresponding secondary antibody for a period of 1 h, then rinsed thrice with TBST again. The Image Lab software (Bio-Rad) was used to obtain clear western blotting bands.

Laser confocal culture dish (34.8 mm diameter) was utilized to seed the chondrocyte cells at a density of 1 × 10⁴ cells/well. 5 ng/ml of IL-1β was added for treating the cells for 24 h in the absence or presence of 40 µM FKA. Furthermore, PBS was utilized for washing (three times); cell fixation was carried out using 4% paraformaldehyde for a period of 15 min; cell permeabilization was performed by employing 0.5% Triton X-100 for a period of 15 min, and 5% BSA was employed for blocking. Overnight incubation of the chondrocytes was done with primary antibody against MMP13 (18165-1-AP, 1:50) and Aggrecan (ab36861, 1:100) at 4°C. After that, they were rinsed using PBS three times, and incubation was performed again with fluorescent secondary antibodies (ab150083, 1:200) for a period of 60 min. In addition, 0.5 μg/ml DAPI was utilized for the nuclei staining. The cell imaging was carried out with the aid of a confocal microscopy (Leica, United States).

The senescent β-galactosidase staining was utilized to check the degree of cellular senescence. The washing of cells was conducted using PBS, and fixation was carried out for 15 min. After that, the staining solution was added for overnight incubation at 37°C in the absence of carbon dioxide. A microscope was employed for the observation and recording of the number of blue spots in the field of view.

Chondrocytes were collected using 0.1% trypsin following the instructions provided by the manufacturer; PBS was used for washing (twice), and cell resuspension was performed in a binding buffer. Apoptotic cells were assessed by flow cytometry (Beckman Coulter, Inc. United States) after performing staining using propidium iodide and Annexin V (Sun et al., 2021).

RFP-GFP-LC3-adenovirus was employed for the detection of autophagic flux. When the cells reached about 50% confluency, the cells were infected with adenovirus vectors for a day. Then use IL-1β (5 ng/ml) pretreated cells in the presence or absence of FKA (40 μM) for 24 h. Due to acid sensitivity, green fluorescence is extinguished after autolysosome formation, and only red fluorescence can be detected (Pei et al., 2021). The number of autophagosomes and autolysosomes was then observed by confocal microscopy (Leica, United States).

Through the PharmMapper and miRWalk2.0 databases, the FKA and OA-related pathways were identified, respectively. The prediction of FKA-target genes was performed based on the PharmMapper database, which was subsequently imported into the Database for Annotation, Visualization, and Integrated Discovery (DAVID) to carry out the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. Moreover, in order to obtain FKA-related OA pathways, the incorporation of the KEGG pathways was conducted with OA-related pathways acquired using the miRWalk 2.0 database. Additionally, the Venn diagram (Venny2.1, http://bioinfogp.cnb.csic.es/tools/venny/index.html) was utilized for the visualization of various intersected pathways related to FKA. Furthermore, Cytoscape 3.8.2 was employed for the screening of hub genes based on the degree, betweenness, and closeness centrality of FKA-target genes. The Pathway Builder Tool 2.0 (https://www.proteinlounge.com) was employed to conduct KEGG pathway enrichment analysis, and the leading KEGG pathways were selected from the crossover pathways. Finally, the proximity of FKA-target genes in the network were illustrated by constructing a circos plot with the aid of the circlize R package, which also included the location of the genes on the chromosome and their connections.

Eighteen specific pathogen-free C57BL/6J male mice were acquired from the Experimental Animal Center of Tongji Hospital. Mouse DMM OA model was established as previously described (Lu et al., 2021). The mouse were anesthetized by intraperitoneal injection of pentobarbital (100 μl/10g body weight). Right knee surgery of all the mice was carried out and randomly classified into 3 groups. FKA group mice (n = 6) underwent DMM surgery and were intra-articularly injected with 50 mg/kg FKA weekly. Sham-operated group mice (n = 6) underwent sham surgery (capsulotomy and suturing) and were injected with saline (equal volume). OA group mice (n = 6) underwent DMM surgery and were injected intra-articularly with saline (equal volume). After Eight weeks of surgery, all mice were sacrificed with overdose anesthetics, and sample collection and fixation were performed using 4% paraformaldehyde for 24 h.

Decalcification of the fixed samples was performed for 30 days using the EDTA-Decalcifying-fluid, and 5-μm-thick fragments of the right knee joint were cut. In addition, hematoxylin and eosin (H&E), safranin-o-fast green as well as toluidine blue were used for the staining of these fragments (Pan et al., 2022). The expression of Aggrecan and Col2 in mouse cartilage was assessed by immunohistochemistry. The assessment of the changes in mouse articular cartilage was carried out as per the Osteoarthritis Research Society International (OARSI) scoring system (Glasson et al., 2010). Scoring was done by two independent investigators who have no knowledge of the grouping.

We used a small animal micro-CT (Scanco Medical AG, Bassersdorf, Switzerland) for the purpose of scanning the right knee joint of the mice. After obtaining the image, the internal software of the microCT system was used for 3D reconstruction. Statistical analysis was then performed using the trabecular numbers (Tb.N), trabecular separation (Tb.Sp), and bone volume/tissue volume (BV/TV) of the samples.

Independent repetition of each experiment was conducted at least thrice, and analysis of data was carried out by employing the GraphPad prism V.8 software (GraphPad Inc., La Jolla, CA, United States). Additionally, all the data were represented by mean +standard deviation (SD), and relevant analyses were conducted using one-way analysis of variance (ANOVA) followed by Turkey’s post-hoc test. Statistical significance was determined by p ≤ 0.05.

We used the CCK 8 kit to determine the impact of various concentrations of FKA on the viability of chondrocytes. The molecular structure of FKA can be seen in Figure 1A. The illustrations in Figures 1C,D reveal that the presence or absence of IL-1β (5 ng/ml) and different concentrations of FKA had no effect on cell viability. Based on current and previous findings (Kwon et al., 2013), we used 20 μM and 40 μM of FKA for subsequent experiments. In addition, Safranin O staining showed that the morphology of chondrocytes did not change after FKA intervention (Figure 1B).

The western blot was utilized to assess the impact of FKA on cartilage inflammation and metabolism. Figure 2(A–D) illustrates that IL-1β elevated the expression levels of chondrocyte inflammatory cytokines (COX2 and iNOS), MMPs (MMP3/13), and ADAMTS5, whereas FKA lowered their expression levels in a concentration-dependent manner. Additionally, FKA also reversed the impacts of IL-1β on the anabolism of chondrocytes. The expression of Col2, Aggrecan, and Sox9 in chondrocytes showed a significant downward trend under the influence of IL-1β, but the effect was reversed under the effect of FKA (Figures 2E,F). The results of immunofluorescence were consistent with the western blot data (Figures 2G–I).

The chondrocyte cells during OA exhibited elevated levels of various senescence markers, for instance, β-galactosidase, as well as the increased expression of senescence-related genes (McCulloch et al., 2017). Figures 3A,B show that the levels of expression of P16 and P21 proteins were elevated in chondrocytes under the influence of IL-1β, whereas FKA suppressed their expression. Additionally, the senescence β-galactosidase staining kit was employed to test the β-galactosidase activity in different groups. Under elevated IL-1β levels, the activity of β-galactosidase was increased (more blue spots), and FKA decreased its activity (Figures 3C,D).

Chondrocyte apoptosis has a close correlation with the severity of cartilage damage in OA. In the current study, we treated chondrocytes with FKA in the IL-1β presence or absence for a period of 24 h. Figures 4A,B indicate that treatment with IL-1β markedly upregulated Bax and downregulated Bcl2 expression in chondrocytes. However, FKA alleviated these trends. In order to determine the apoptosis status of chondrocytes, we employed the Annexin V-FITC/PI Apoptosis Detection kit. Figures 4C,D illustrate that FKA treatment led to a remarkable decrease in IL-1β-induced apoptosis of chondrocytes.

Numerous pieces of evidence indicate that a crucial role is played by autophagy in the progression of OA (Musumeci et al., 2015; Liu et al., 2021; Yan et al., 2022). Therefore, we analyzed the impacts of FKA on autophagy in IL-1β-treated mouse chondrocyte cells. Figures 5A,B reveal that IL-1β treatment remarkably suppressed the ATG12, LC3Ⅱ/LC3, and Beclin, expression levels in mouse chondrocytes. However, the administration of 40 μM FKA significantly reversed these changes. Moreover, RFP-GFP-LC3-adenovirus was employed to further verify the impact of FKA on autophagy. We observed that the number of autophagolysosomes and autophagosomes decreased in chondrocyte cells treated with IL-1β, and FKA pretreatment reversed this trend as illustrated in Figure 5E. Furthermore, the autophagy inhibitor, 3MA, attenuated the promoting effect of FKA on IL-1β-induced chondrocyte autophagy (Figures 5C,D).

A crucial role is exerted by oxidative stress in the advancement of OA (Feng et al., 2019). According to a previous study, FKA alters the expression of antioxidant genes in primary splenocytes (Yang et al., 2020). Therefore, we assessed whether FKA could alter the IL-1β-induced expression of antioxidant genes, including NQO1, Nrf2, and HO-1 in chondrocytes. The results in Figures 6A,B illustrate that FKA substantially increased the expression of antioxidant genes in chondrocytes.

The PI3K/Akt/mTOR and MAPK pathways exert a significant function in the autophagy, inflammation, and progression of OA (Park et al., 2018). In this study, treatment of chondrocyte cells was performed with IL-1β only or in combination with 20 μM or 40 μM FKA for 30 min, and protein expression in chondrocytes was assessed using the western blot technique. As shown in (Figure 8), FKA pretreatment partially inhibited IL-1β-stimulated phosphorylation of the genes involved in the MAPK and PI3K pathways.

Furthermore, the effects of FKA on OA onset and progression were detected in the mouse models. The histological analysis revealed remarkably less cartilage damage in the FKA group in comparison to the OA model group (Figures 9A,B). Moreover, immunohistochemical results revealed that the FKA group showed significantly increased expression of Aggrecan and Col2 in comparison to the OA model group (Figures 9C–E).

A micro-CT was conducted for the evaluation of the impact of FKA on subchondral bone remodeling in the OA models. In Figure 10, we can observe that the bone volume fraction (BV/TV) and the trabecular number (Tb.N) in the OA model group were evidently increased in comparison to the sham group, and the degree of trabecular separation (Tb.Sp) was significantly reduced. However, after FKA treatment, these trends were significantly reversed. These findings suggest that FKA had a promising amelioration impact on subchondral bone remodeling in OA mice.