Oxymatrine exerts protective effects on osteoarthritis via modulating chondrocyte homoeostasis and suppressing osteoclastogenesis

Abstract Osteoarthritis (OA) is a common degenerative disease characterized by the progressive destruction both articular cartilage and the subchondral bone. The agents that can effectively suppress chondrocyte degradation and subchondral bone loss are crucial for the prevention and treatment of OA. Oxymatrine (OMT) is a natural compound with anti‐inflammatory and antitumour properties. We found that OMT exhibited a strong inhibitory effect on LPS‐induced chondrocyte inflammation and catabolism. To further support our results, fresh human cartilage explants were treated with LPS to establish an ex vivo degradation model, and the results revealed that OMT inhibited the catabolic events of LPS‐stimulated human cartilage and substantially attenuated the degradation of articular cartilage ex vivo. As subchondral bone remodelling is involved in OA progression, and osteoclasts are a unique cell type in bone resorption, we investigated the effects of OMT on osteoclastogenesis, and the results demonstrated that OMT suppresses RANKL‐induced osteoclastogenesis by suppressing the RANKL‐induced NFATc1 and c‐fos signalling pathway in vitro. Further, we found that the anti‐inflammatory and anti‐osteoclastic effects of oxymatrine are mediated via the inhibition of the NF‐κB and MAPK pathways. In animal studies, OMT suppressed the ACLT‐induced cartilage degradation, and TUNEL assays further confirmed the protective effect of OMT on chondrocyte apoptosis. MicroCT analysis revealed that OMT had an attenuating effect on ACLT‐induced subchondral bone loss in vivo. Taken together, these results show that OMT interferes with the vicious cycle associated with OA and may be a potential therapeutic agent for abnormal subchondral bone loss and cartilage degradation in osteoarthritis.


| INTRODUCTION
Osteoarthritis (OA) is a common, heterogeneous arthritic disorder that causes a major public health burden all over the world. 1 According to previous studies, OA ranks 11th of the 291 diseases listed by the WHO and is one of the leading causes of global disability. 2 Despite the high prevalence of OA, the specific pathogenesis of the disease is not yet clear.
Over the last decade, our understanding of OA has improved dramatically. The disease is now considered an inflammatory disease, 3 characterized by a low-grade inflammatory state. 4 Pro-inflammatory cytokines not only stimulate their own expression but also activate the NF-jB and mitogen-activated protein kinase (MAPK) signalling pathway, the key mediators whose activation leads to the induction of cartilage matrix-degrading enzymes such as a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS), matrix metalloproteinases (MMPs) and aggrecanases. [5][6][7] Furthermore, OA is now considered an organ-level disease that involves not only the articular cartilage but also the subchondral bone, 8,9 characterized by a decrease in bone density that is later followed by abnormal bone deposition. Several studies have suggested that there is increased porosity in the subchondral plate during disease development, as evidenced by a mouse model of instability-induced knee OA. 10,11 Osteoclasts, as the main cells for bone resorption, have recently become a new target for the treatment of osteolytic diseases. [12][13][14][15][16] For this reason, treatment focusing on the prevention of cartilage degeneration and osteoclast-mediated bone loss is of great significance for the prevention and treatment of OA.
In recent years, there has been increasing interest in compounds extracted from traditional Chinese medicinal plants, which may play a significant role in the treatment of OA because of their potent anti-inflammatory effects. 17 Oxymatrine (OMT), which is a quinolizidine alkaloid extracted from the traditional Chinese herb Sophora flavescens Aiton, 18 has attracted much attention because of its low toxicity and wide pharmacological effects and is now often used to treat various diseases, including angina pectoris, myocardial infarction, stroke, cancer and inflammation. [19][20][21] The anticancer effects of OMT mainly through inhibiting cancer proliferation, invasion and metastasis. 18,21 Furthermore, OMT accelerates cancer cell apoptosis and reverses multidrug resistance when combined with other chemotherapeutic drugs. 22 The anti-inflammation effects of OMT appear to involve both inhibition of MAPK phosphorylation and reduce the activation of the NF-jB signalling pathway. [23][24][25][26] Furthermore, OMT exhibits substantial therapeutic potential for the treatment of myocardial diseases through modulating of the JAK2/STAT3 and TGF-b1/Smad3 signalling pathway, 27,28 but the effects of OMT on the onset and progression of OA have not been reported, to our knowledge.
In this study, we investigated the potential effect of OMT as a preventive treatment for OA in vitro and in vivo. We are committed to providing OMT as a novel potential alternative for the treatment of OA.

| Primary human chondrocyte isolation and culture
Osteoarthritis cartilage tissues were obtained from patients who underwent total knee arthroplasty. Cartilage was separated from the underlying bone and connective tissues, cut into 1 9 1 9 1 mm 3 pieces and washed 3 times with PBS. Afterwards, the joint cartilage pieces were digested with 0.25% trypsin-EDTA solution for 30 minutes at 37°C, followed by collagenase type II for 6 hours at 37°C.
Then, the samples were centrifuged at 300 g for 5 minutes, and the supernatant was discarded. The chondrocytes were collected and cultured in DMEM/F12 medium containing 10% FBS and 100 U/mL of penicillin-streptomycin at 37°C in a humidified atmosphere of 5% CO 2 . Only passages 2-4 were used in our study to avoid the pheno-

| Cell viability
The effect of OMT on cell viability was assessed using a cell counting kit 8 (CCK-8) assay (Dojindo, Kumamoto, Japan) according to the manufacturer's instructions. Human OA chondrocytes and BMMs were cultured in 96-well plates at a density of 5 9 10 3 cells per well and then pre-treated with or without different concentrations (0, 0.5, 1, 2 or 4 mg/mL) of OMT for the indicated time. After that, 10 lL of CCK-8 was added to each well and incubated at 37°C for 2 hours. The optical density was read at a wavelength of 450 nm with a microplate spectrophotometer (SpectraMax; Molecular Devices, Sunnyvale, CA).
Three independent experiments were carried out in triplicate.

| Real-time PCR analysis
After treatment at the indicated time-points, total RNA was extracted from chondrocytes using TRIzol reagent (Invitrogen, CA, USA) according to the manufacturer's instructions. For each sample, reverse transcription was accomplished on 1 ng of total RNA using the M-MLV reverse transcriptase (Invitrogen). PCR assays were performed in triplicate on a ViiA TM 7 real-time PCR system (Life Technology, USA) according to the manufacturer's instructions. The mRNA expression levels of target genes were normalized to that of b-actin, and the 2 ÀDDCt method was used to assess the relative expression of different candidate genes. The primers used for real-time PCR are listed in Table S1.

| Human cartilage explant ex vivo assay
Fresh articular cartilage was collected aseptically from patients who underwent total knee arthroplasty for primary knee OA. Explants were dissected from the articular surface without calcified cartilage layers; 2-mm-thick explants with a 4-mm diameter were obtained, washed in PBS 3 times and cultured in DMEM/high glucose supplemented with 10% FBS. The explants were exposed with 10 lg/mL

| In vitro osteoclast differentiation assay
For the in vitro osteoclastogenesis assay, bone marrow cells (BMMs) were isolated from the femurs and tibias of C57BL/6 mice and cultured in a-MEM supplemented with 10% FBS and macrophage colony-stimulating factor (M-CSF) (10 ng/mL) for 24 hours; the BMMs were allowed to adhere overnight and then cultured with mM-CSF (30 ng/mL) and Receptor Activator for Nuclear Factor-j B Ligand (RANKL) (50 ng/mL) in 96-well plates in the presence or absence of OMT for 6 days with media changed every 2 days. After treatment, cells were fixed with 4% paraformaldehyde (PFA) for 20 minutes at room temperature and stained for TRAP activity according to the manufacturer's instructions. TRAP-positive multinucleated cells with more than 3 nuclei were counted as osteoclasts.

| In vitro osteoblast differentiation assay
BMSCs between passages 3 and 5 were used for experiments. To induce osteogenic differentiation, BMSCs were cultured with a-MEM supplemented with 10% FBS, ascorbic acid (50 mg/mL) and bglycerophosphate (10 mmol/L) for up to 7 and 14 days, after treatment, the cells were fixed with 4% PFA for 20 minutes at room temperature and stained for Alizarin Red according to the manufacturer's instructions.

| Western blot
After treatment, total protein was extracted from chondrocytes using ice-cold RIPA lysis buffer. The protein concentration was determined using a BCA protein assay kit (Beyotime, NanJing, China). Total protein (20-50 lg) was resolved on 10% SDS-PAGE and transferred to PVDF membranes. Membranes were blocked with 5% non-fat milk in TBS con-

| Immunofluorescence microscopy
After treatment, cells were fixed with 4% paraformaldehyde (PFA) for 20-30 minutes at room temperature. The cells were permeabilized with 0.1% Triton X-100 for 15 minutes. The cells were blocked in 0.1% bovine serum albumin (BSA) for 30 minutes and then incubated with primary antibody at 4°C overnight. After washing 3 times for 5 minutes with a TBS-T solution, Alexa Fluor-conjugated secondary antibodies (1:400) were used for visualization. Cell nuclei were stained with DAPI for 3 minutes. After washing 3 times for 5 minutes, the dish was covered with PBS. The images were collected on an Olympus FluoView FV10i confocal microscope.

| Nuclear and cytoplasmic extraction
Human primary chondrocytes and BMMs were treated with OMT for 2 hours and then stimulated with LPS or RANKL for another 30 minutes. To separate the cytoplasmic and nuclear proteins, cell pellets were processed using a nuclear and cytoplasmic extraction kit (Beyotime, NanJing, China) according to the manufacturer's instructions.

| MicroCT analysis
After treatment, all hind limbs of the mice were dissected and fixed in 4% PFA for 24 hours before micro CT analysis. Axial scans were Model Index (SMI) using the manufacturer provided software.

| Statistical analysis
Statistical analysis was performed with one-way ANOVA. Data are shown as the means AE SD. Statistical calculations were performed by SPSS 17.0. P < .05 was considered to indicate a significant difference from the control.

| OMT inhibited LPS-induced pro-inflammatory cytokines and matrix metalloproteinase (MMP) production in human OA chondrocytes
To explore the effects of OMT ( Figure 1A) on chondrocytes, we first investigated the effect of OMT exposure on the viability of primary human chondrocytes. As shown in Figure 1B, cell viability decreased with OMT exposure up to 2 mg/mL, but this compound had no significant toxicity on chondrocyte viability.
Lipopolysaccharides has been used as a classical methodological approach to perform diverse in vitro studies and has been used to induce arthritis. 29 In our studies, we have investigated the effects of OMT on the production of pro-inflammatory cytokines and MMPs in LPS-stimulated human OA chondrocytes. As shown in Figure 1C F I G U R E 3 OMT dose-dependently inhibited osteoclastogenesis in vitro. A, The effect of OMT on mouse BMM differentiation was detected using TRAP staining. Scale bar, 100 lm. The number (B) and area (C) of TRAP-positive multinucleated (≥3 nuclei) osteoclasts were quantified. D, The effect of OMT on cell viability of mouse BMMs was measured by CCK-8 assay. E, The effects of OMT on apoptotic-related proteins were analysed using Western blotting; GAPDH was used as the loading control. F, After differentiation, BMMs were fixed and stained for F-actin ring. Scale bar, 100 lm. G, The number of osteoclasts per well with an intact actin ring was counted. H, The expression of OC marker genes (CTSK, NFATc1, TRAP, DC-STAMP and CTR) was determined using real-time PCR. The gene expression was normalized to that of b-actin. I, OMT suppresses the RANKL-induced activation of NFATc1 and c-Fos. GAPDH was used as the loading control. (right panel) The protein expression levels of NFATc1 and c-Fos were quantified using ImageJ. J, The effects of OMT on osteogenesis were detected using Alizarin Red staining. Scale bar, 100 lm. All experiments were performed at least 3 times. Values are expressed as the means AE SD. *P < .05 compared to the control

| OMT attenuates the LPS-induced degeneration of human articular cartilage explants
To investigate the protective effects of OMT on human articular cartilage, we then examined the LPS-induced degradation of human articular cartilage ex vivo. As shown in Figure 2A, Safranin O staining showed that OMT administration significantly attenuated the LPSinduced loss of proteoglycan. The results clearly showed that OMT was able to preserve the proteoglycan content in the ECM of ex vivo explants, as indicated by the red staining intensity being F I G U R E 4 OMT attenuates NF-jB and MAPK signalling pathway activation. A, The activity of NF-jB and MAPK signalling pathways in mouse BMMs (or chondrocyte) was determined using Western blotting; GAPDH was used as the loading control. B, The ratio of p-p65 to total p65 was quantified using ImageJ. The expression and nuclear translocation of NF-jB p65 in chondrocytes (C) and BMMs (D) were determined using immunofluorescence and a nuclear-cytoplasmic extraction assay. Scale bar, 10 lm. P65 protein expression levels in the nucleus and cytoplasm were analysed by Western blot. All experiments were performed 3 times. Values are expressed as the mean AE SD; *P < .05 compared to control similar to that of the control. Immunohistochemical staining results further revealed that LPS exposure inhibited the expression of collagen II, while in the OMT-treated group, the expression of collagen II was higher than that of the LPS exposure group. The differences in the overall staining intensity of the explant groups were assessed according to the Safranin O staining results ( Figure 2B). Based on the immunostaining intensities of 5 randomly selected areas of cartilage, the expression level of collagen II was calculated and analysed ( Figure 2C).
As the changes in LPS-induced chondrocytes lead to the up-regulation of MMPs, the GAG content released into the supernatant was interpreted as the degree of matrix degradation caused by MMPs. We measured the concentration of GAGs in the culture medium by the DMMB assay, and the results showed that at days 7 and 14, OMT treatment significantly decreased the release of GAGs from LPS-stimulated human articular cartilage tissues into the culture supernatant ( Figure 2D). Above all, these results demonstrated that OMT inhibited catabolic events in LPS-stimulated human cartilage and substantially attenuated the degradation of articular cartilage.

| OMT inhibited RANKL-induced osteoclast formation in vitro
To investigate the effect of OMT on osteoclastogenesis, we employed an in vitro osteoclast differentiation model; mouse BMMs were stimulated with M-CSF or RANKL. As shown in Figure  To evaluate the influence of OMT on osteoblast differentiation from progenitors, induction cultures of BMSCs were performed.
Determination of differentiation was based on the Alizarin Red staining. As shown in Figure 3J, OMT had no significant effects on osteogenic differentiation of BMSC.

| OMT suppresses the NF-jB and MAPK signalling pathway
The NF-jB signalling pathway is an important signal transducer involved in both LPS-induced inflammation and RANKL-induced osteoclastogenesis and has been implicated as a key regulator of cartilage destruction and bone remodelling in OA. 30,31 To determine the mechanism of OMT on chondrocyte inflammation, catabolism and osteoclastogenesis, BMMs (and human chondrocytes) were treated with RANKL (and LPS) in the presence or absence of 1 mg/mL OMT for 5, 15, 30 minutes or up to 1 hour, and the expression levels of p65, p-p65 and IjBa were assessed using Western blotting. As shown in Figure 4A, the ratio of p-p65 to p65 was higher in the absence of OMT at several time-points ( Figure 4B). At the same time, the immunofluorescence microscopy and nuclear-cytoplasmic extraction results further supported that OMT suppressed the nuclear translocation of p65 ( Figure 4C,D). The results indicated that OMT inhibited the activation of NF-jB signalling pathway and may offer value as a novel alternative for treating NF-jB related diseases.
MAPK is another signalling pathway involved in OA-related cartilage destruction and OC-mediated bone resorption. 32 In our study, both RANKL and LPS significantly promoted the phosphorylation of ERK, JNK and p38 in human chondrocytes and BMMs, while OMT abolished these effects at several time-point after stimulation (Figure 4A). Together, these data suggest that the inhibitory effect of OMT on chondrocyte inflammation and osteoclastogenesis may be due to the attenuation of the NF-jB and MAPK signalling cascades.

| OMT treatment rescues cartilage degeneration in a mouse ACLT OA model in vivo
To explore the potential protective effects of OMT on articular cartilage in vivo, we first performed a histological analysis using H&E and  Finally, we examined the systemic toxicity of OMT, the body weights of mice were measured once a week over the treatment period. As shown in Figure S1, OMT showed no obvious damage on mice.

| DISCUSSION
Osteoarthritis (OA) is a representative degenerative joint disease that affects not only articular cartilage but also subchondral bone. The crosstalk among osteochondral units is of great importance. 8,34,35 Although pathological OA might selectively target a single joint tissue during the early stage of disease, both components of the joint will ultimately be affected because of their intimate association. Subchondral bone also plays a major role in maintaining articular cartilage integrity, and any attempt to repair a cartilage lesion without sufficient support from intact subchondral bone will likely result in failure. 37 Because of the crosstalk between cartilage and subchondral bone, the integrity of articular cartilage has been proposed to depend on mechanical properties of the underlying bone. 38 Nevertheless, in some cases, researchers view subchondral bone degradation as the causative factor in the development of OA, and studies based on animal models showed that during the early stages of OA, the subchondral bone has a decrease in both volume and stiffness. 39 Osteoclasts, as the primary cells involved in bone resorption, were recently identified as an effective target for the prevention and treatment of osteolytic disorders. Based on these previous studies, we concluded that an effective treatment against OC-mediated  ACLT is one of the most common sporting injuries and is associated with an increased risk for developing post-traumatic osteoarthritis (PTOA). 40 The ACLT injury model has been shown to reliably reproduce the changes found in human osteoarthritic bone.
In our study, we employed this model to explore the effects of OMT on OA cartilage and subchondral bone in vivo. The results demonstrated an increased remodelling of the knee joint subchondral bone during early stage OA that led to substantial bone loss and increased porosity. With disease progression, compared to the sham controls, ACLT mice displayed a significant loss of cartilage proteoglycans and exhibited a significantly higher degree of destruction, which was observed down to subchondral bone.

CONFLI CT OF INTERESTS
All authors declare that they have no conflict of interests concerning this article. F I G U R E 7 Proposed mechanism of OMT interference with the OA-associated vicious cycle. OMT exerts dual effects on cartilage degradation and osteoclastmediated bone resorption, and those effects are imparted by inhibiting the activation of NF-jB signalling pathways F I G U R E 6 OMT protects against ACLT-induced bone loss in vivo. A, Representative microCT 3D reconstructed images were obtained for each group. Scale bar, 1 mm. B, The BV/TV, BMD, Tb.Sp, Tb.Th, Tb.N and SMI were measured. C, The OCs in the subchondral plate (SCP) and subchondral trabecular bone (STB) were detected using H&E and TRAP staining (2009). Scale bar, 100 lm. The number of osteoclasts (D) and the percentage area (E) stained by TRAP in sections were analysed. All bar graphs are expressed as the means AE SD. *P < .05