Bushenhuoxue formula attenuates cartilage degeneration in an osteoarthritic mouse model through TGF-β/MMP13 signaling

Background Articular cartilage degeneration plays a key role in the pathogenesis of osteoarthritis (OA). Bushenhuoxue formula (BSHXF) has been widely used in the treatment of OA in clinics. However, the molecular mechanisms responsible for the chondroprotective effect of BSHXF remain to be elucidated. The purpose of this study was to explore the effects of BSHXF on OA mice model. Methods In this study, we investigated the effects of BSHXF on destabilization of the medial meniscus (DMM)-induced chondrocyte degradation in OA mice model. At 12 weeks post-surgery, the joints were harvested for tissue analyses, including histology, histomorphometry, TUNEL, OARSI scoring, micro-CT and immunohistochemistry for COL2, TGFBR2, pSMAD2 and MMP13. Additionally, we also evaluated the effects of BSHXF on Mmp13 mRNA and protein expression in chondrogenic ATDC5 cells through real-time PCR and Western blot respectively. Moreover, we investigated the chondroprotective effect of BSHXF on mice with Tgfbr2 conditional knockout (Tgfbr2Col2ER mice) in chondrocyte, including the relative experiments mentioned above. We transfected Tgfbr2 siRNA in ATDC5 to further evaluate the changes of Mmp13 mRNA and protein expression followed by BSHXF treatment. Results Amelioration of cartilage degradation and chondrocyte apoptosis were observed in DMM-induced mice, with increases in cartilage area and thickness, proteoglycan matrix, COL2 content and decreases in OARSI score at 12 weeks post surgery. Moreover, the elevated TGFBR2 and pSMAD2, and reduced MMP13 positive cells were also revealed in DMM-induced mice treated with BSHXF. Besides, decreased Mmp13 mRNA and protein expression were observed inchondrogenic ATDC5 cells culture in serum containing BSHXF. As expected, Tgfbr2Col2ER mice exhibited significant OA-like phenotype. Interestingly, obvious improvement in articular cartilage structure was still observed in Tgfbr2Col2ER mice after BSHXF treatment via up-regulated pSMAD2 and down-regulated MMP13 expressional levels in articular cartilage. Conclusions BSHXF could inhibit cartilage degradation through TGF-β/MMP13 signaling, and be considered a good option for the treatment of OA.


Background
Osteoarthritis (OA) which is characterized by abnormal extracellular matrix content (EMC) as well as articular surface erosion, is the most common disabling disease of the joint tissue influenced by interactions among age, genetic and mechanical factors [1]. Knee and hip OA has led to one of the principal causes of global disability [2]. There are a variety of treatment options for OA, but much emphasis in its early stage has been placed on pain relief through medication. Major classes of OA pharmaceuticals include non-steroidal anti-inflammatory drugs (NSAIDs), non-specific cyclooxygenase inhibitors, selective cyclooxygenase-2 (COX-II) inhibitors, opiates and non-opioid oral analgesics [3]. All above painkillers are extensively used in the management of pain associated with OA. Unfortunately, the long-term uses of these drugs may induce prominent side effects. Besides, current OA therapies could not ameliorate all the symptoms and are not considered to be disease modifying approaches [4].
Cartilage degradation is a prominent feature of OA and it may be an important pathological event contributing to the disease pathogenesis. While adult mammal's articular cartilage appears to show a poor ability of repairing itself throughout life [5,6]. Both human clinical and animal researches have demonstrated that matrix metalloproteinase 13 (MMP13) plays a key role during articular cartilage degeneration. Clinical study showed that osteoarthritic patients with articular cartilage damage have high MMP13 expression in chondrocyte [7]. Animal studies have also revealed that overexpression of Mmp13 in transgenic mice's articular cartilage develop a spontaneous OA-like phenotype [8].
The transforming growth factor beta (TGF-β) signaling which is responsible for regulating the synthesis and degradation of extracellular matrix proteins, for controlling the proliferation and differentiation of chondrocytes and for inhibiting it hypertrophy and maturation [9][10][11]. Dys-regulation of TGF-β signaling in articular cartilage promotes development of OA [12]. Recently, related study has found that inactivation of TGF-β signaling via deletion of TGF-β receptor type II (Tgfbr2) gene in chondrocytes leads to the up-regulated expression Mmp13 which contributes to the cartilage degradation [13], suggesting that Mmp13 are critical downstream target gene in the TGF-β signaling during the development of OA. Therefore, MMP13 inhibitor such as ALS 1-0635 [14] would be a potential therapeutic strategy for the OA treatment. However, most of these compounds have failed for toxicity, including skin rash and musculoskeletal side effects characterized by joint stiffness and pain. Currently, no MMP inhibitor has been used in clinics [15].
Traditional Chinese medicine (TCM) acts as the promising alternative medicine as a therapeutic option for OA [16]. Bushenhuoxue formula (BSHXF) orally administered as decoction or granule is a traditional Chinese drug that has been clinically used for many years in the first affiliated hospital of Zhejiang Chinese Medical University. Our previous studies have demonstrated that BSHXF could inhibit cartilage degradation in vivo [17]. Although BSHXF has been proved effective in OA treatment, the precise mechanism and drug action target is still poorly understood.
Thus, we hypothesized that the chondroprotective effect of BSHXF based on inhibition of MMP13 through TGF-β/MMP13 signaling in chondrocyte. In this study, we firstly sought to determine the effect of high-performance liquid chromatography (HPLC) standardized BSHXF. Furthermore, we determined the impact of BSHXF in osteoarthritic cartilage in mouse model. We also used Tgfbr2 conditional knockout (cKO) mice in articular cartilage and determined the changes of articular cartilage structure followed by BSHXF treatment.

Preparation of BSHXF
Herb pieces of BSHXF were offered from pharmacy department of the first affiliated hospital of Zhejiang Chinese Medical University (Hangzhou, China). The extraction process includes two parts: aqueous and ethanol extracts. Carthamus tinctorius (L.), Rehmannia glutinosa (Libosch.), Eucommia ulmoides (Oliv.), Aconitum carmichaelii (Debx.), Lycium barbarum (L.), Cornus officinalis (Sieb.), Dioscorea opposite (Thunb.) and Glycyrrhiza uralensis (Fisch.) were mixed as a proportion of 1:3:2:2:2:1:2:1 (w/w) and the total dry weight was 42 kg for aqueous extracts. The mixture of herbs above was soaked in 12 volumes of distilled water for 1 h, extracted for 3 times by the reflux method and 1.5 h each time. Prunus persica (Batsch.) and Cinnamomum cassia (Presl.) were mixed in the ratio of 2:1 (w/w) and the total dry weight was 9 kg for ethanol extracts. These pieces were immersed in 10 volumes of 60% ethanol for 1 h, extracted for 3 times and 1.5 h each time. Then, two types of extract were thoroughly mixed to obtain 15 L medicated solution. The solution was concentrated to 2 g crude drug/mL for further usage according to the daily dose and kept in − 20 °C to preserve.

Animals
Both C57BL/6J mice for modeling and Sprague Dawley (SD) rats for serum acquisition were obtained from Shanghai Laboratory Animal Center of Chinese Academy of Science (Grade SPF II, SCXK 2012-0002). Tgfbr2 gene was specifically knocked out in chondrocyte of mice (C57BL/6J background) based on a Cre/loxP conditional knockout (cKO) allele of the Tgfbr2 gene. Col2-CreER mice were crossed with Tgfbr2 fx/fx mice to generate Col2-CreER; Tgfbr2 fx/fx mice (Tgfbr2 Col2ER mice) ( Table 1) and all original mice were donated from Rush University (Chicago, IL, USA). Genotyping was performed by polymerase chain reaction (PCR) using a DNA extraction kit (Sigma, St. Louis, MO, USA) tail biopsy tissues with a Eppendorf 5333 PCR System (Eppendorf AG, Hamburg, Germany). PCR primer sequences for genotyping were as follows: primer sequences of Cre, forward primer 5′-ATT GCT GTC ACT TGG TCG TGGC-3′ and reverse primer 5′-GAA AAT GCT TCT GTC CGT TTGC-3′ (200-base-pair PCR product); primer sequences of Tgfbr2 loxP, forward primer 5′-TAA ACA AGG TCC GGA GCC CA-3′ and reverse primer 5′-ACT TCT GCA AGA GGT CCC CT-3′ (Wild-type, 420-base-pair PCR product; homozygotic type, 540-base-pair PCR product). Cre-negative littermates were used as controls. After identification of genotypes, Tgfbr2 Col2ER mice were induced by intraperitoneal injection (i.p.) with tamoxifen (1 mg/10 g body weight/ day, for 5 days) (Sigma, St. Louis, MO, USA) at age of 2 weeks [13]. All mice were housed at 21 ± 2 °C with a relative humidity of 40 ± 5%. The room was maintained on a 12 h light/dark cycle with free access to water and lab chow. All studies were approved by the Committee on the Ethics of Animal Experiments of Zhejiang Chinese Medical University.

Surgical preparation
Knee osteoarthritic model was established when the WT mice were 10-week-old by destabilization of the medial meniscus (DMM) and challenge as described previously [18]. Briefly, DMM surgery was performed on the right hind limb of WT mice as follows: (1) after a 3 mm longitudinal incision was made on the medial part of the knee under anesthesia, blunt dissection of the knee extensor muscles and patellar ligament was performed to expose the medial meniscotibial ligament (MMTL); (2) The MMTL was transected to give destabilization of the medial meniscus (DMM); (3) The medial joint capsule was sutured, and the skin was closed. The sham surgery was also performed by a similar surgical approach without manipulating the joint tissue in WT mice.

Drug administration and experimental groups
All mice were randomly divided into six groups: the sham group, the model group, the BSHXF group, the Cre-negative group, the Tgfbr2 Col2ER group and the Tgf-br2 Col2ER + BSHXF group (n = 10 in each group). BSHXF was orally administered to both WT and Tgfbr2 Col2ER mice once a day after DMM surgery and tamoxifen inducement respectively for 12 consecutive weeks with dose of 0.2 mL/10 g body weight. The dose administered was determined according to the following formula: as detailed in The methodology of pharmacological experiment. D M and D H are doses for mice and humans, and W M and W H are body weights of mice and humans, respectively. R is the coefficient of 0.0026 for human mouse equivalent dosage conversion [19]. The mice in sham and Cre-negative group were fed with an equal dosage of 0.9% normal saline. Knee joints samples were harvested after drug intervention for the follow-up experiments.

Histological analysis
Knee joints harvested for histological analysis were successively fixed in 4% paraformaldehyde for 3 days, decalcified with 14% EDTA solution for 14 days and embedded in paraffin. Three micrometre thick sections were cut longitudinally from the medial compartment of the joints. The sections were stained with Alcian Blue Hematoxylin/Orange G and Toluidine Blue for analysis of gross structural changes. Histomorphometric analysis was performed using OsteoMeasure software (OsteoMetrics, Inc., Atlanta, GA, USA). Cartilage structure degeneration was scored by two blinded observers according to Osteoarthritis Research Society International (OARSI) recommendations [20].

Micro-CT evaluation
The representative knee joint images were obtained from micro-CT equipment (Skyscan 1176, Bruker microCT N.V., Kontich, Belgium). The pictures were taken with a resolution of 4000 × 2672 pixels and an isotropic voxel size of 9 μm. In addition to the visual assessment of structural pictures, quantitative morphometry indexes were determined from micro-tomographic data based on the 3-D morphometry [21]. The following indexes were determined: (i) bone volume fraction (BV/TV, %), (ii) average trabecular thickness (Tb. Th, mm) and (iii) average trabecular separation (Tb. Sp, mm).

Immunohistochemistry (IHC) analysis
Three micrometre deparaffinized sections were treated with 0.3% hydrogen peroxide to reduce endogenous peroxidase activity. Then the sections were incubated for 20 min at 95 °C with 0.1 mol/L citrate buffer as antigen retrieval. Non-specific staining was blocked by incubation of normal goat serum (diluted 1:20) (Invitrogen, MD, USA) for 20 min at room temperature. Subsequently, the sections were treated with collagen II (COL2) (diluted 1:1000), matrix metallopeptidase 13 (MMP13) (diluted 1:100), transforming growth factor beta receptor II (TGFBR2) (diluted 1:100) or phosphorylated protein mothers against decapentaplegic homolog 2 (pSMAD2) (diluted 1:100) primary antibodies (Abcam, MA, USA) and further incubated overnight at 4 °C. Secondary biotinylated goat anti-mouse antibody (diluted 1:1000) (Invitrogen, MD, USA) was added for 30 min on the second day. Positive staining of sections was detected by diaminobenzidine solution (Invitrogen, MD, USA) followed by counterstaining with hematoxylin. The primary antibody was not added to negative controls. The nuclei were considered positive for TGFBR2, pSMAD2 and MMP13 labeling if their immunostains were equal or larger than 50% of the nuclear area. Weak brown stains were excluded from the counting. The rate of TGFBR2, pSMAD2 and MMP13 positive cells were determined using Image-Pro Plus 6.0 (Media Cybernetics, Silver Spring, USA).

Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) analysis
TUNEL staining was performed using a staining kit to detect chondrocyte apoptosis. The protocol of TUNEL staining was based on the manufacturer's instruction (Roche, IN, USA).

Preparation of serum containing BSHXF
Fifty eight-week-old female SD rats weighted 200 ± 20 g were randomly divided into two equal groups (n = 25 in each group). The rats in BSHXF group were orally administrated with BSHXF (1 mg/10 g body weight/day [19], for 7 days), and normal control group were orally administrated with 0.9% saline at the same dose. Then the blood was collected separately and the serum was reserved via centrifugation at 3000 rpm/min for 10 min. After filtered and inactivated at 56 °C for 30 min, the serum of two groups was stored at − 80 °C for subsequent use.

Small interfering RNA (siRNA) transfection
Tgfbr2-siRNA and non-silencing siRNA were purchased from Ribo (RiboBio, Guangzhou, China). Active siRNA against Tgfbr2 used in the in vitro studies had sequences 5′-CCU GUU GCC UGU GUG ACU U-3′ (sense) and 3′-GGA CAA CGG ACA CAC UGA A-5′ (antisense). Tgfbr2-siRNA transfection of siRNA was performed using lipofectamine transfection regent 2000 (Invitrogen, Carlsbad, CA), according to the manufacturer's protocol. Control cells were mock transfected with lipofectamine transfection regent 2000 only. Chondrogenic ATDC5 cells were seeded in a 6-well plate until cells reached 50% confluence. The full culture medium was changed with serum-free and antibiotic-free medium at 20 min before transfection. The cells were incubated with transfection mixtures containing Tgfbr2-siRNA or non-silencing siRNA for 6 h, and then the mixtures were replaced with complete culture medium.

Real-time polymerase chain reaction (PCR) analysis
Total RNA was extracted from differentiated chondrogensis respectively cultured in control serum and serum containing BSHXF using TRIzol (Invitrogen, CA, USA), and cDNA was generated from 2 μg RNA using Rever-tAid First Strand cDNA Synthesis Kit (Invitrogen, CA, USA) following the manufacturer's instruction. Realtime PCR was performed for target gene using the SYBR Premix Ex Taq ™ II (Takara, Dalian, China) with a QuantStudio ™ 7 Flex Real-Time PCR System (Thermo Scientific, MA, USA) according to the manufacturer's instruction. The specific primers are as followed: primer sequences of Mmp13, forward primer 5′-TTT GAG AAC ACG GGG AAG A-3′ and reverse primer 5′-ACT TTG TTG CCA ATT CCA GG-3′; primer sequences of Actb, forward primer 5′-GGA GAT TAC TGC CCT GGC TCCTA-3′ and reverse primer 5′-GAC TCA TCG TAC TCC TGC TTG CTG -3′. Relative Mmp13 gene expression values were analysed using the comparative 2 −∆∆Ct method for relative quantification, which is implemented in the QuantStudio ™ 6 and 7 Flex Software (Thermo Scientific, MA, USA). Actb was used as an endogenous control gene for PCR normalization concerning the amount of RNA added to the reverse transcription reactions.

Western blot analysis
Total proteins were collected immediately in lysis buffer containing protease and phosphatase inhibitors, incubated on ice for 30 min and separated on a 12% SDS-PAGE gel. The samples were then transferred onto polyvinylidene fluoride membranes and blocked in 5% milk for 1 h. Immunoblotting was subsequently performed by incubating the membranes overnight at 4 °C with rabbit primary antibodies targeting MMP13 (diluted 1:1000; Abcam, MA, USA). The membranes were incubated with goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (diluted 1:5000; Abcam, MA, USA) at room temperature. The densities of the bands were determined using the Tocan 190 protein assay system (Bio-Rad, CA, USA) and normalized to Actb.

Date analysis
All the data were expressed as mean ± standard deviation. One-way analysis of variance (ANOVA) test was used to compare the means among groups. And the multiple comparisons were performed with least significant difference (LSD) test. A P value of less than 0.05 was considered to be statistically significant. The statistical analysis was performed using SPSS 18.0 software (SPSS Inc., Chicago, IL, USA).

BSHXF decelerates DMM-induced OA progression
Alcian Blue Hematoxylin/Orange G and Toluidine Blue staining revealed that prominent focal cartilage defects, early osteophyte formation and increases in subchondral bone mass was observed in DMM-induced mice compared to sham group. Additionally, the results of IHC showed lower COL2 content in DMM-induced mice. However, BSHXF dramatically inhibited the cartilage degeneration, COL2 content loss and osteophyte formation (Fig. 2a, e). Histomorphometric analysis showed a significant increased cartilage area (Fig. 2b) and thickness (Fig. 2c) in mice treated with BSHXF (both, P < 0.01). We also evaluated the cartilage degradation through OARSI grading, and we found that mice in BSHXF group had significantly lower scores compared to model mice (P < 0.01), with significantly reduced OARSI scores after treatment with BSHXF (P < 0.01) (Fig. 2d). Consistent with the histological analysis, the result of COL2 IHC also showed an increased COL2 content in DMM-induced mice treated with BSHXF (P = 0.023) (Fig. 2a, e). Moreover, the area of trabecular bone in the proximal metaphysis of tibia was evaluated by 3D data analysis software (CTAn, Bruker micro-CT ® ). As shown  Fig. 2f, BSHXF prominently reduced DMM-induced osteophyte formation. Compared to sham mice, up-regulated BV/TV, Tb. Th and down-regulated Tb. Sp were exhibited in model group (all, P < 0.01). After BSHXF treatment, the BV/TV and Tb. Th measurements were significantly lower than those of DMM-induced mice (P = 0.034, P = 0.040) (Fig. 2g, h). Conversely, the Tb. Sp measurements of the BSHXF group were significantly higher than those of DMM-induced mice (P = 0.016) (Fig. 2i), suggesting that BSHXF has a positive effect on bone health in DMM-induced mice.

DMM-induced up-regulation of MMP13 is reduced by BSHXF
A key link in articular cartilage erosion and chondrocytes apoptosis is the imbalance of synthesis and degeneration of the extracellular matrix (ECM), an event marked by the expression of proteolytic enzymes such as MMP13, and the important molecular of TGF-β/MMP13 signaling, TGFBR2 and pSMAD2. Accordingly, it is possible that the chondroprotective action of BSHXF documented in Fig. 3 is associated with TGF-β/MMP13 signaling. To address this question, IHC was performed to evaluate TGFBR2, pSMAD2 and MMP13 protein expressional levels in the articular cartilage. As expected, DMM-induced mice had reduced protein expressional level of TGFBR2 and pSMAD2, and elevated levels of MMP13 in cartilage compared to sham mice (both, P < 0.01) (Fig. 3a-d). Mice treated with BSHXF exhibited decreased MMP13 positive cells (P = 0.024) (Fig. 3d), suggesting that maintaince of cartilage architecture in BSHXF group could be due to inhibition of matrix degradation. Remarkably, elevated TGFBR2 and pSMAD2 positive cells were also observed in BSHXF group (P = 0.028, P < 0.01) (Fig. 3b, c). The mechanism of BSHXF regulation of Mmp13 expression was investigated through an in vitro experiment. Consistent with the IHC analysis, we found reduction of Mmp13 mRNA and protein levels in chondrogenic ATDC5 cells cultured with serum containing BSHXF through realtime PCR and Western blot analysis respectively (Fig. 3e). To investigate chondrocytes apoptosis, we performed TUNEL staining on DMM-induced mice. DMM-induction of broad chondrocyte apoptosis was reduced followed by BSHXF (P = 0.011) (Fig. 3f, g). These results indicated that BSHXF may have chondroprotective effects on DMM-induced mice in part due to reduced MMP13 production by chondrocytes residing in cartilage through TGF-β/MMP13 signaling.

BSHXF prevents OA-like phenotype observed in Tgfbr2 Col2ER mice
The marked preservation of ECM produced by chondrocytes post-DMM in BSHXF treated mice needed for subsequent investigation of the impact of BSHXF on OA progression in Tgfbr2 Col2ER mice. Result of histological showed significantly OA-like clefting and cartilage loss appeared in Tgfbr2 Col2ER mice. However, most notably cartilage defects were ameliorated with BSHXF treatment (Fig. 4a). Consistent with the histopathologic analysis, reduced tibial cartilage area and thickness in Tgfbr2 Col2ER mice were attenuated significantly by BSHXF treatment (P < 0.01, P = 0.033) (Fig. 4b, c). Furthermore, the impact of BSHXF on the articular cartilage in Tgfbr2 Col2ER mice was also significant difference based on OARSI scoring (P < 0.01) (Fig. 4d). Consistent with histological analysis, the result of COL2 IHC revealed a significantly elevated COL2 positive staining in Tgfbr2 Col2ER mice followed by BSHXF treatment (P = 0.023) (Fig. 4a, e). In addition, remarkable osteophyte formation observed in Tgfbr-2 Col2ER mice was reduced followed by BSHXF treatment via micro-CT (Fig. 4f ). Moreover, BV/TV and Tb. Th were increased, and Tb. Sp was decreased in Tgfbr2 Col2ER mice, was rescued by BSHXF (BV/TV, P = 0.035; area, P = 0.013; thickness, P = 0.010) (Fig. 4g-i).

Discussion
In the present study, we have investigated the effect of BSHXF on articular cartilage degradation in DMMinduced OA murine model. We also have observed in TGF-β/MMP13 signaling mechanism through Tgfbr-2 Col2ER mice. Our data provided molecular evidences that BSHXF decelerates osteoarthritic cartilage degeneration performed as a MMP13 inhibitor via regulation of TGF-β/MMP13 signaling.
Reproducing characteristics of OA in animals is critical to have a better understanding of disease mechanism. DMM surgical model is one of the most common models of OA in laboratory animals, and strongly associated with clinical process [22,23]. DMM model provided extremely good reproducibility and considered to be a gold standard in the field [24]. Histology demonstrated the cartilage damage and subchondral bone sclerosis developed 8 weeks after surgery and osteophyte formation was observed 12 weeks after surgery [18,25,26]. Consistent with the previous studies [18,25,26], DMM-induced mice in our research exhibited remarkable localized damage of articular cartilage, early osteophyte formation and increases in subchondral bone mass In OA disease, the precisely controlled balance of ECM synthesis and degeneration is disrupted, resulting in progressive erosion of articular cartilage [27]. The healthy chondrocytes play a key role for maintaining articular cartilage homeostasis partly through producing ECM components, including COL2, proteoglycan, etc. MMP13 is a major enzyme that not only degrades COL2 in cartilage, but also targets proteoglycan, osteonectin and perlecan in cartilage for degeneration [28]. Abnormal chondrocyte in osteoarthritic cartilage induced downregulated COL2 expression in the superficial layer and up-regulated MMP13 expression [29,30]. Notably, one of the pivotal regulatory regions of the MMP13 promoter is the RUNX2 binding site, which has been implicated in the TGF-β-inducible MMP13 expression in articular cartilage [13,31]. Moreover, related study reported that Tgfbr2 cKO induces development of OA via the up-regulation of RUNX2 and MMP13 in chondrocyte [13]. Consistent with the previous studies, we found reduced protein expressional level of TGFBR2 and elevated MMP13 in DMM-induced mice. These observations strongly support the notion that elevated MMP13 expression in chondrocyte through TGF-β/MMP13 signaling may play a critical role during the process of DMMinduced OA.
Such novel findings indicate that MMP13 expression in chondrocyte may represent a potential therapeutic target in OA. In preclinical testing, MMP inhibitors have inhibited the degradation of articular cartilage in animal models of OA [32,33]. However, most clinical trials have not been continued due to concerns of dose-limiting toxicity. Although a number of hypotheses have been proposed for the toxicity, including broad-spectrum MMP inhibitors, the pharmacological basis for side effects remains poorly understood [34,35]. Chinese medicine acts as the promising alternative medicine as a therapeutic option for OA and might be easily accepted by clinicians and patients [16]. BSHXF, a traditional Chinese Herbal formula, is composed of Prunus persica (Batsch.), Carthamus tinctorius (L.), Rehmannia glutinosa (Libosch.), Eucommia ulmoides (Oliv.), Aconitum carmichaelii (Debx.), Lycium barbarum (L.), Cinnamomum cassia (Presl.), Cornus officinalis (Sieb.), Dioscorea opposite (Thunb.) and Glycyrrhiza uralensis (Fisch.). In our study, the constituents of BSHXF were separated and defined by HPLC analysis, and six components were determined as major active material basis. Amygdalin and loganin showed highest concentrations, followed by liquiritin, pinoresinol diglucoside, hydroxysafflor yellow A and cinnamaldehyde. Our previous study had demonstrated that BSHXF may decelerate osteoarthritic cartilage degeneration [17]. Niu et al. [36] reported that combination of Amygdalin and hydroxysafflor yellow A was shown to have the potential to inhibit the degeneration of chondrocytes and better than the single use of amygdalin or hydroxysafflor yellow A, which fully embodied the TCM strengths and may partly explain the protective effect of BSHXF on articular cartilage.
In this study, significant chondroprotective effects were detected in DMM-induced mice treated with BSHXF. Our data showed that BSHXF could elevate protein expressional level of TGFBR2 and pSMAD2, and reduce MMP13 in cartilage of DMM-induced mice. To further determine the mechanism of BSHXF inhibited articular cartilage degeneration in OA disease, we investigated the efficacy of BSHXF in reversing the OA-like phenotype of Tgfbr2 Col2ER mice. As reported in previous study, TGF-β signaling regulates the expression of genes that are important in the maintenance of ECM [13]. Inactivation of TGF-β signaling promotes Mmp13 gene transcription in a RUNX2-dependent manner [13]. In addition to mutation of Tgfbr2 gene, potential causes of TGF-β signaling inactivation in articular cartilage include mutation of other genes responsible for mediating TGF-β signaling, such as Smad3 and p38 [31]. In our study, the loss of articular cartilage observed in Tgfbr2 Col2ER mice was rescued by BSHXF treatment through its suppressive effect on the expression of MMP13. However, we found that BSHXF treatment slightly elevated the TGFBR2 expression level although Tgfbr2 gene was specifically knocked out in articular cartilage. One possible reason for this finding is that about 80% TGFBR2 proteins were blocked in articular cartilage of Tgfbr2 Col2ER mice. And the Cre recombination efficiency mediated by Col2-CreER is close to the previous study [13]. In other words, TGFBR2 activity in the other 20% of chondrocytes remains normal in Tgfbr2 Col2ER mice. Thus, the rescue of BSHXF treatment on chondroprotective effects was incomplete. So we reasoned that it may be compensated for by regulation of the downstream gene expression in TGF-β signaling. Interestingly, Tgfbr2 Col2ER mice supplemented with BSHXF had significant elevated protein expression level of pSMAD2 in chondrocytes compared to control Tgf-br2 Col2ER mice. And this finding was consistent with the previous study [31]. Overall, these discoveries suggested that preservation of cartilage architecture in BSHXF treatment groups could be due to reduced MMP13 production by chondrocytes via activating TGF-β/MMP13 signaling.

Conclusions
In conclusion, our study revealed that BSHXF attenuates osteoarthritic articular cartilage degradation as a natural MMP13 inhibitor through TGF-β/MMP13 signaling in articular cartilage. These findings serve to underscore the potential value of TCM in the treatment of chronic OA disease and encourage further research into this therapeutic modality.