Ying He, Ying Zhang, Mengying Wang Meng ZhangDan Zhang Ying Zhang#, Zhuocheng JiangFeng Wu, Jinghong Chen＊

1Institute of Endemic Diseases, School of Public Health,

2Graduate Students Teaching Experiment Center, Xi’an Jiaotong University College of Medicine, Xi’an 710061, China

KASHIN-BECK disease (KBD) is an endemic and degenerative osteoarthropathy, mainly prevalent in China, the Siberian provinces of Russia, and North Korea.1As symptoms of the disease include joint deformities and joint pain, most KBD patients suffer both physical and psychological pain. Apart from significantly reducing the patients’ quality of life, KBD also bestows a heavy medical and financial burden onto society.2Various environmental etiologic hypotheses for KBD have been proposed, including consumption of water contaminated with organic compounds, selenium deficiency, and mycotoxin cereal contamination.3-4While these studies helped identify effective approaches for preventing KBD, and although the disease has been studied for more than one hundred years, its precise pathogenesis remains unclear, a fact that constitutes the bottleneck in developing effective therapies for KBD.

Exposure to T-2 toxin as well as selenium deficiency have been shown to cause chondronecrosis in the deep zone of the knee joint articular cartilage in rats, indicating that both these factors may be etiological agents of KBD.5T-2 toxin is representative of fungal toxins that can promote the production of reactive oxygen species (ROS) and reactive nitrogen free radicals, causing oxidative damage to cartilage cells.6Supplementation with selenium salts can partly prevent KBD.7Selenium is a free radical scavenger,so the low selenium diet of children in the regions where KBD is prevalent makes them vulnerable to oxidative damage from free radicals. Continuous attacks from pathogenic factors combined with the lack of the protective factor (selenium) may lead to the development of KBD.

Our previous study8showed that the levels of thiobarbituric acid reactive substances (TBARS) were significantly higher in the serum of KBD children,while the levels of antioxidants, such as superoxide dismutase (SOD) and catalase (CAT), as well as the total antioxidant capacity (T-AOC) of the serum were significantly lower in KBD and healthy children from KBD-endemic regions compared to healthy children from other regions. CAT activity was also increased in osteoarthritis (OA) patients compared to healthy controls.9The rise in the levels of antioxidants might be an adaptive response to counter the effect of oxidative stress in both KBD and normal children.Overexpression of two oxidative damage biomarkers,8-hydroxydeoxyguanisine (8-OHdG) and 4-hydroxy-2-nonenal (4-HNE), is observed in both OA and rheumatoid arthritis (RA).10-11There was also an accumulation of 8-OHdG and 4-HNE in articular cartilage, especially in the deep zone necrotic chondrocytes of KBD children.10

The feature of KBD distinguishing it from other degenerative joint diseases such as OA and RA is the localization of necrosis in deep-zone cartilage near the adjacent subchondral bone.10Thus, we speculated that excessive free radicals that are produced by hypertrophic chondrocytes of the deep zone cartilage and lead to cartilage oxidative damage may be the triggering factor of KBD. Research on the disease should focus more on hypertrophic chondrocytes. In addition, hydrogen peroxide (H2O2)has been used as an oxygen free radical donor in many studies.12-13It can increase intracellular ROS levels and cause the death of human mesenchymal stem cells (MSCs).14Considering the above aspects,we chose H2O2as the oxygen free radical donor for this study, to assess effects of H2O2on hypertrophic chondrocytes.

MATERIALS AND METHODS

Cell culture and establishment of hypertrophic chondrocyte model

Murine chondrogenitor cells (ATDC5) purchased from the European Collection of Cell Cultures (Salisbury,UK), were cultured in complete DMEM/F-12 medium(Hyclone, USA). ATDC5 were induced into hypertrophy by Insulin-Transferrin-Selenium (ITS, BD, USA; 1:100 diluted in DMEM/F12) differentiation medium and the medium were changed every second day. Total RNA was isolated at 0-, 7-, 14-, and 21-day post-ITS addition. The concentration and purity of total RNA were determined. One microgram of total RNA from each group was reversely transcribed according to Revert Aid? First cDNA Synthesis Kit, and then the obtained cDNAs were diluted by 5-fold with nucleasefree water. Type X collagen (Col X) was used as a marker of hypertrophy.15-16Therefore, mRNA level of Col X was admeasured during the induction of ATDC5 cells with qRT-PCR (Table 1). GAPDH mRNA was used as an endogenous control.

MTT assay to detect viability of cells treated with H2O2

ATDC5 were planted in a 96-well plate at a density of 1 × 103cells for each well, and differentiation medium was used from the next day up to 21 days. Cells were incubated with 0, 25, 50, 100, 200 and 400 μmol/L H2O2for 24 hours, following with further incubation for another 4 hours in medium containing 0.5 mg/ml 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, MP) at 37°C. The supernatant was discarded and the intracellular purple formazan in each well was dissolved in 150 μl of DMSO. The purple crystal was determined by measuring the absorbance at a wavelength of 490 nm in a microplate reader.The cell survival rate was calculated according to the following equation: cell survival rate (%)=(absorbance of each well treated with H2O2/average absorbance of the control group) × 100%, and control group was not treated with H2O2.

RNA preparation for PCR array

Total RNA was extracted respectively from hypertrophic chondrocytes stimulated by 0 (control) and 200 μmol/L H2O2for 24 hours, using RNeasy Plus Mini Kit (Qiagen,74134). The concentration and purity of total RNA were determined, and the quality and integrity were evaluated by 2% agarose gel electrophoresis. cDNA of each group was synthesized from 1 μg of total RNA,using RT2First Strand Kit (Qiagen, 330401) according to manufacturer's instructions. It is noteworthy that genomic DNA elimination step was essential for obtaining optimal real-time gene expression profiles. The mixed sample was incubated at 42°C for 60 minutes and 95°C for 5 minutes. Each sample (including 20 μl synthetic cDNA and 92 μl RNase-free water) was mixed by pipetting up and down for several times. Mixtures were stored at –20°C until for processing.

RT2 Profiler PCR Arrays?

The 102 μl above mixture was mixed with 1248 μl of RNase-free water and 1350 μl of RT2qPCR SYBR green Mastermix (Qiagen, 330522). An aliquot of 25 ml of freshdy prepared mixtrue was used for mouse Osteogenesis RT2Profiler PCR Array (Qiagen, PAMM-026Z). Real time-PCR was performed in the Thermal Cycler Dice Real Time System (Takara TP-800, Japan)with SYBR green detection system. The thermal profile was as following: step 1 (one cycle): 95°C for 10 minutes; step 2 (40 cycles): 95°C for 15 seconds,55°C for 40 seconds, and 72°C for 30 seconds; step 3(dissociation curve): 95°C for 15 seconds, 60°C for 30 seconds, and 95°C for 15 seconds. Five housekeeping genes (Actb, B2m, Gapdh, Gusb, and Hsp90ab1) were used as internal controls. It was considered as no genomic DNA contamination when Ct value of Mouse Genomic DNA Contamination (MGDA) control was higher than 30. All data was analyzed by SABiosciences PCR Array Data Analysis web-based software.17Fold change more than two times was considered as a statistically significant difference.

Real time PCR validation

Quantitative RT-PCR was conducted to validate PCR array data, using five down-regulated genes ［Smad2,Smad4, transforming growth factor β receptor1(TGF-βr1), TGFβr3 and matrix metalloproteinase 10(MMP10)］. Primers for RCR are showed in Table 1.Total RNA was isolated using the RNeasy Plus Mini Kit(Qiagen, 74104). One microgram of total RNA was reversely transcribed into cDNA, and finally diluted five times with RNase-free water. Single-stranded cDNAwas amplified by using QuantiFast SYBR Green PCR Kit(Qiagen, 204054). GAPDH expression was also used as an endogenous control.

Table 1. Primers used in RT-PCR experiments

Statistical analysis

All data were expressed as mean ± standard deviation(SD) and analyzed by Student’s t-test using SPSS Version 13.0 software (SPSS Inc, USA). Likewise, P-values less than 0.05 were considered as statistically significant.

RESULTS

Hypertrophic chondrocyte model

During the induction of ATDC5 by using an insulincontaining medium, we measured mRNA levels of gene encoding Col X, which is a well-known marker of chondrocyte hypertrophy. We found that expression of Col X mRNA was significantly increased on day 21 (Fig. 1), signifying that cells had passed onto the hypertrophic stage. Hence, ATDC5 after an induction period of 21 days were regarded as a model of hypertrophic chondrocytes in all subsequent experiments.

Effects of H2O2 on cell viability

The results of MTT assay demonstrated that 25 μmol/L of H2O2had a weak proliferative effect on hypertrophic chondrocytes; in contrast, higher concentrations of H2O2,such as 100, 200, and 400 μmol/L reduced cell viability(Fig. 2). H2O2of 200 μmol/L caused a reduction in cell survival rate of about 50% (Fig. 2). These results, combined with previous reports demonstrating that 200 μmol/L of H2O2induced the death of hypertrophic chondrocytes,led us to adopt a treatment for 24 hours with 200 μmol/L H2O2for our experiments.

Figure 1. mRNA expression of type X collagen (Col X)was observed by qRT-PCR during the induction of ATDC5 differentiation. GAPDH was used as a reference gene.The mRNA levels of non-induced cultures were used as a control. Data are presented as the mean±SD from three independent experiments; *P＜0.05 compared with the contols.

PCR array results

Figure 2. Cell viability of hypertrophic chondrocytes treated with H2O2. Cells were treated with 0, 25, 50, 100, 200, and 400 μmol/L H2O2 for 24 hours. Data are presented as the mean±SD (n=3); *P＜0.05 compared with 0 μmol/L H2O2 group.

Figure 3. Scatter plot of 84 genes related to osteogenesis.Group 1 was treated with 200 μmol/L H2O2; Control Group was not exposed to H2O2. Genes whose expression levels exhibited 2-fold or greater changes were deemed to be differentially expressed. Up-regulated genes are marked with a red circle, while down-regulated genes are marked with a green circle. Black circles represent genes whose expression was unchanged.

The expressions of 84 genes were compared between hypertrophic chondrocytes treated with 200 μmol/L H2O2and ones without H2O2exposure. The scatter plots revealed that the mRNA expression levels of 21 genes in hypertrophic chondrocytes exposed to H2O2exhibited two-fold or greater changes (Fig. 3).Specifically, 12 genes were down-regulated (Table 2),mainly encoding extracellular matrix (ECM) proteases(MMP2, 9, and 10), transcription factors (Smad2 and 4), ossification protein (Chordin and Cathepsin K), collagen chain (Col4a1), etc. Nine genes were up-regulated (Table 3), primarily encoding ECM molecules (Biglycan and Phex), growth factors associated with bone regeneration (vascular endothelial growth factor β), collagen chain (Col14a1)and so on.

Real time PCR validation

To confirm the PCR array results, we selected five down-regulated genes (Smad2, Smad4, TGF-βr1,TGF-βr3, and MMP10) for validation by qRT-PCR.The results confirmed that treatment of hypertrophic chondrocytes with 200 μmol/L H2O2caused significant reductions in all five genes (Fig. 4). The consistency between these results and the former PCR array-based findings provides confidence in the reliability of the array data.

DISCUSSION

Even though KBD has been studied for more than 160 years, it still afflicts a large number of patients.During the last decade, research on the disease was directed towards the cellular and molecular level, into areas such as gene polymorphisms,15susceptibility genes,16associated signaling pathways,18and so on. Nevertheless, relatively little is known about the molecular mechanisms of KBD. One of the common symptoms of KBD are observed in the hypertrophic zone of the epiphyseal growth plate. Taking the above conditions into consideration, alterations in the hypertrophic layer are probably closely related to KBD.

Table 2. Significantly down-regulated genes in hypertrophic chondrocytes treated with H2O2

Table 3. Significantly up-regulated genes in hypertrophic chondrocytes treated with H2O2

Figure 4. mRNA expression levels of five selected genes in hypertrophic chondrocytes stimulated by H2O2. GAPDH was used as an endogenous control. Data were expressed as the mean ± SD from three independent experiments;*P＜0.05 compared with 0 μmol/L group.

In this study, we identified expression levels of several genes in hypertrophic chondrocytes after treatment with H2O2changed; these genes may provide new insights for uncovering the underlying causes of KBD.

Transforming growth factors: TGF-β2, TGF-βr1 and TGF-βr3

Transforming growth factor-beta (TGF-β), highly expressed in cartilages and bones, plays a critical role in cell proliferation, apoptosis, and differentiation. The canonical TGF-β signaling pathway is activated by three isoforms (TGF-β1, 2, and 3) by binding to the type ⅡTGF receptor (TGF-βr2), followed by phosphorylation of type Ⅰ transmembrane serine-threonine kinase receptors (ALKs).19TGF-βr2 reportedly facilitates terminal chondrocyte differentiation.20The PCR array was performed in this study to investigate the expressions of TGF-β isoforms and their receptors.Resultsshowed that H2O2decreased the mRNA levels of TGF-β2, TGF-βr1, and TGF-βr3 in hypertrophic chondrocytes. The ratio of TGF-βr1 (ALK5) to another receptor, ALK1, is crucial to MMP-13 expression in osteoarthritis, thus loss of signaling via ALK5 would disturb cartilage homeostasis and integrity.21TGF-β2 knockout mice exhibit a wide range of developmental defects, including craniofacial, cardiac, limb, urogenital,and lung defects, while TGF-β3 null mice only display defective palatogenesis and pulmonary development delay.22Moreover, with respect to the role of agerelated alterations in chondrocyte signaling pathways,TGF-β is considered as one of the two major factors.23

Transcription factors: Smad2 and Smad4

Smad proteins, being transcriptional co-modulators that shuttle from cytoplasm to nucleus, are critical for TGF-β signaling pathways. TGF-β/Smad pathway has been proved to play a vital role in the development of osteoarthritis.19Smad3 is considered a risk factor of osteoarthritis; in a European patient cohort, a single nucleotide polymorphism (SNP) in an intron of the Smad3 gene was found to be related to hip and knee osteoarthritis.24In this study, we found that H2O2down-regulated Smad2 and Smad4 in hypertrophic chondrocytes. The expression of Smad2 and Smad3 is reportedly decreased in the cartilage of rats with osteoarthritis, whereas administration of pilose antler increased their expression in both mRNA and protein level.25Smad2 and Smad3 have some overlapping functions; loss of Smad3 in early development can be compensated by Smad2 or through other mechanisms.26Unlike Smad3, Smad2 is indispensable for the formation of definitive endoderm.27Smad2 knockout mice die due to failure of mesoderm formation.28Global knockout of Smad4 also leads to early embryonic lethality because of defects in gastrulation.29Smad4 expression was found to be decreased in an osteoarthritis model, whereas miR-146a induced vascular endothelial growth factor and chondrocyte apoptosis by targeting Smad4.30

Extracellular matrix proteases: MMP2, MMP9,MMP10, Ctsk, and Phex

In recent years, growing evidence indicates the importance of extracellular matrix (ECM). It has been reported that ECM may play a role in wound healing, aging,bone and cartilage regeneration,31-32etc. The family of MMPs, which as a whole possess the ability to degrade all the components of ECM, have been confirmed as the primary factors in the degradation of connective tissue macromolecules.33In this study, a down-regulation of expression of MMP2, MMP9, and MMP10 was observed, indicating that H2O2may alter ECM microenvironment of hypertrophic chondrocytes. The expression of MMP2 and MMP9 was reportedly elevated in osteoarthritis,34whereas the activity of MMP2 and MMP9 was found to be significantly enhanced by activated protein C in human osteoarthritic cartilage chondrocytes.35MMP-10 was reported to play an important role during wound repair, mainly with respect to remodeling activity of macrophages and for reduction of scar formation.36Moreover, we observed an increase in Phex expression and a decrease in Ctsk expression in this study. Phex gene (phosphate-regulating gene with homologies to endopeptidase on X chromosome)encodes a metallopeptidase, and its expression was shown to increase in early experimental osteoarthritis.37Cathepsin K (CTSK) is a cysteine protease produced predominantly by osteoclasts, mainly for degrading triple helix collagen.38It has been proposed that CTSK-positive chondrocytes and synovial cells might be suitable targets for preventing OA progression.39

In conclusion, to our knowledge, this study shows the genes associated with osteogenesis are abnormally expressed in hypertrophic chondrocytes treated with H2O2. In other words, these genes take part in the oxidative stress reaction of murine hypertrophic chondrocytes. Based on our results and previous studies, these oxidative stress-dependent genes may be involved in the chondronecrosis that takes place in the deep-zone cartilage of KBD patients. Further exploration of the molecular mechanisms underlying oxidative damage in hypertrophic chondrocytes is likely to contribute to the development of new prevention strategies and treatments for KBD and other similar types of osteoarthritis.

1. Duan C, Guo X, Zhang XD, et al. Comparative analysis of gene expression profiles between primary knee osteoarthritis and an osteoarthritis endemic to Northwestern China, Kashin-Beck disease. Arthritis Rheumatol 2010; 62(3):771-80. doi: 10.1002/art.27282.

2. Zhang F, Guo X, Wang W, et al. Genome-wide gene expression analysis suggests an important role of hypoxia in the pathogenesis of endemic osteochondropathy Kashin-Beck Disease. PLoS One 2011; 6(7):e22983. doi: 10.1371/journal.pone.0022983.

3. Zou K, Liu G, Wu T, et al. Selenium for preventing Kashin-Beck osteoarthropathy in children: a metaanalysis. Osteoarthritis Cartilage 2009; 17(2):144-51.doi: 10.1016/j.joca.2008.06.011.

4. Zhang WH, Neve J, Xu JP, et al. Selenium, iodine and fungal contamination in Yulin District (People’s Republic of China) endemic for Kashin-Beck disease.Int Orthop 2001; 25(3):188-90.

5. Guan F, Li S, Wang ZL, et al. Histopathology of chondronecrosis development in knee articular cartilage in a rat model of Kashin-Beck disease using T-2 toxin and selenium deficiency conditions.Rheumatol Int 2013; 33(1):157-66. doi: 10.1007/s00296-011-2335-7.

6. Chen JH, Xue S, Li S, et al. Oxidant damage in Kashin-Beck disease and a rat Kashin-Beck disease model by employing T-2 toxin treatment under selenium deficient conditions. J Orthop Res 2012; 30:(8):1229-37. doi: 10.1002/jor.22073.

7. Allander E. Kashin-Beck disease. An analysis of research and public health activities based on a bibliography. Scand J Rheumatol Suppl 1994; 99:1-36.

8. Wang W, Wei S, Luo M, et al. Oxidative stress and status of antioxidant enzymes in children with Kashin-Beck disease. Osteoarthritis Cartilage 2013;21(11):1781-9. doi: 10.1016/j.joca.2013.08.002.

9. Olszewska-Slonina DM, Matewski D, Drewa G,et al. Oxidative equilibrium in the prophylaxis of degenerative joint changes: an analysis of pre and postoperative activity of antioxidant enzymes in patient with hip and knee osteoarthritis. Med Sci Monit 2010; 16(5):CR238-45.

10. Biniecka M, Kennedy A, Fearon U, et al. Oxidative damage in synovial tissue is associated with in vivo hypoxic status in the arthritic joint. Ann Rheum Dis 2010; 69(6):1172-8. doi: 10.1136/ard.2009.111211.

11. Grigolo B, Roseti L, Fiorini M, et al. Enhanced lipid peroxidation in synoviocytes from patients with osteoarthritis. J Rheumatol 2003; 30(2):345-7.

12. Liochev SI. Reactive oxygen species and the free radical theory of aging. Free Radic Biol Med 2013;60:1-4. doi: 10.1016/j.freeradbiomed.2013.02.011.

13. Gao J, Deng Y, Yin C, et al. Icariside Ⅱ, a novel phosphodiesterase 5 inhibitor, protects against H2O2-induced PC12 cells death by inhibiting mitochondriamediated autophagy. J Cell Mol Med 2017; 21(2):375-86. doi: 10.1111/jcmm.12971.

14. Lee JH, Jung HK, Han YS, et al. Antioxidant effects of Cirsium setidens extract on oxidative stress in human mesenchymal stem cells. Mol Med Rep 2016;14(4):3777-84. doi: 10.3892/mmr.2016.5706.

15. Boone DR, Micci MA, Taglialatela IG, et al. Pathwayfocused PCR array profiling of enriched populations of laser capture microdissected hippocampal cells after traumatic brain injury. PLoS One 2015;10(5):e0127287. doi: 10.1371/journal.pone.0127287.

16. Zhang F, Dai L, Lin W, et al. Exome sequencing identified FGF12 as a novel candidate gene for Kashin-Beck disease. Funct Integr Genomics 2016; 16(1):13-7. doi: 10.1007/s10142-015-0462-z.

17. Boone DR, Micci MA, Taglialatela IG, et al. Pathwayfocused PCR array profiling of enriched populations of laser capture microdissected hippocampal cells after traumatic brain injury. PLoS One 2015;10(5):e0127287. doi: 10.1371/journal.pone.0127287.

18. Zhang F, Wen Y, Guo X, et al. Genome-wide association study identifies ITPR2 as a susceptibility gene for Kashin-Beck disease in Han Chinese. Arthritis Rheumatol 2015; 67(1):176-81. doi: 10.1002/art.38898.

19. Shen J, Li S, Chen D. TGF-β signaling and the development of osteoarthritis. Bone Res 2014; 2. pii:14002. doi: 10.1038/boneres.2014.2.

20. Serra R, Johnson M, Filvaroff EH, et al. Expression of a truncated, kinase-defective TGF-beta type Ⅱreceptor in mouse skeletal tissue promotes terminal chondrocyte differentiation and osteoarthritis. J Cell Biol 1997; 139(2):541-52.

21. Blaney Davidson EN, Remst DF, Vitters EL, et al.Increase in ALK1/ALK5 ratio as a cause for elevated MMP-13 expression in osteoarthritis in humans and mice. J Immunol 2009; 182(12):7937-45. doi:10.4049/jimmunol.0803991.

22. Dünker N, Krieglstein K. Targeted mutations of transforming growth factor-beta genes reveal important roles in mouse development and adult homeostasis. Eur J Biochem 2000; 2677(24):6982-8.

23. van der Kraan P, Matta C, Mobasheri A. Age-related alterations in signaling pathways in articular chondrocytes: implications for the pathogenesis and progression of osteoarthritis―a mini-review. Gerontology 2016; 63(1):29-35. doi: 10.1159/000448711.

24. Valdes AM, Spector TD, Tamm A, et al. Genetic variation in the SMAD3 gene is associated with hip and knee osteoarthritis. Arthritis Rheum 2010;62(8):2347-52. doi: 10.1002/art.27530.

25. Niu W, Sun ZT, Cao XW, et al. Regulation of single herb pilose antler on the expression of Smad2 and Smad3 in the cartilage of OA rats: an experimental research.Zhongguo Zhong Xi Yi Jie He Za Zhi 2014; 34(2):209-13.

26. Song B, Estrada KD, Lyons KM. Smad signaling in skeletal development and regeneration. Cytokine Growth Factor Rev 2009; 20(5-6):379-88. doi:10.1016/j.cytogfr.2009.10.010.

27. Tremblay KD, Hoodless PA, Bikoff EK, et al.Formation of the definitive endoderm in mouse is a Smad2-dependent process. Development 2000;127(14):3079-90.

28. Weinstein M, Yang X, Li C, et al. Failure of egg cylinder elongation and mesoderm induction in mouse embryos lacking the tumor suppressor smad2.Proc Natl Acad Sci USA 1998; 95(16):9378-83. doi:10.1073/pnas.95.16.9378.

29. Sirard C, de la Pompa JL, Elia A, et al. The tumor suppressor gene Smad4/Dpc4 is required for gastrulation and later for anterior development of the mouse embryo. Genes Dev 1998; 12(1):107-19.

30. Li J, Huang J, Dai L, et al. miR-146a, an IL-1β responsive miRNA, induces vascular endothelial growth factor and chondrocyte apoptosis by targeting Smad4. Arthritis Res Ther 2012; 14(2):R75. doi:10.1186/ar3798.

31. Hughes OB, Rakosi A, Macquhae F, et al. A review of cellular and acellular matrix products: indications,techniques, and outcomes. Plast Reconstr Surg 2016; 138(3 Suppl):138S-47S. doi: 10.1097/PRS.0000000000002643.

32. Pak J, Lee JH, Park KS, et al. Regeneration of cartilage in human knee osteoarthritis with autologous adipose tissue-derived stem cells and autologous extracellular matrix. Biores Open Access 2016; 5(1):192-200. doi:10.1089/biores.2016.0024.

33. Murphy G, Kn?uper V, Atkinson S, et al. Matrix metalloproteinases in arthritic disease. Arthritis Res 2002; 4 Suppl 3:S39-49. doi: 10.1186/ar572.

34. Zeng GQ, Chen AB, Li W, et al. High MMP-1, MMP-2,and MMP-9 protein levels in osteoarthritis. Genet Mol Res 2015; 14(4):14811-22. doi: 10.4238/2015.

35. Jackson MT, Moradi B, Smith MM, et al. Activation of matrix metalloproteinases 2, 9, and 13 by activated protein C in human osteoarthritic cartilage chondrocytes. Arthritis Rheumatol 2014; 66(6):1525-36. doi: 10.1002/art.38401.

36. Rohani MG, McMahan RS, Razumova MV, et al. MMP-10 regulates collagenolytic activity of alternatively activated resident macrophages. J Invest Dermatol 2015; 135(10):2377-84. doi: 10.1038/jid.2015.167.

37. Appleton CT, Pitelka V, Henry J, et al. Global analyses of gene expression in early experimental osteoarthritis. Arthritis Rheum 2007; 56(6):1854-68.doi: 10.1002/art.22711.

38. Dahlberg L, Billinghurst RC, Manner P, et al. Selective enhancement of collagenase-mediated cleavage of resident type Ⅱ collagen in cultured osteoarthritic cartilage and arrest with a synthetic inhibitor that spares collagenase1 (matrix metalloproteinase 1). Arthritis Rheum 2000; 43(3):673-82. doi:10.1002/1529-0131(200003)43: 3＜673::AIDANR25＞3.0.CO;2-8

39. Kozawa E, Nishida Y, Cheng XW, et al. Osteoarthritic change is delayed in a Ctsk-knockout mouse model of osteoarthritis. Arthritis Rheum 2012; 64(2):454-64.doi: 10.1002/art.33398.