P21 deficiency exhibits delayed endochondral ossification during fracture healing.

INTRODUCTION: Endochondral ossification is a complex biological phenomenon involving a variety of factors and cells. Cyclin-dependent kinase inhibitor 1 (p21) inhibits cell cycle progression and is affected by external stress. We recently reported that embryonic endochondral ossification is unaffected by endogenous p21 deficiency. In this study, we evaluated whether p21 expression affects endochondral ossification during fracture healing. METHODS: Tibial fractures were introduced into p21 knockout (p21-/-) (n = 24) and wild-type C57BL/6 (p21+/+) (n = 24) mice at age 10 weeks. Fracture healing was evaluated using radiological, histological, and immunohistochemical (IHC) analyses. The effect of p21 small interfering RNA (siRNA) on ATDC5 cells was assessed in vitro. RESULTS: The Allen score for fracture healing was lower in p21-/- mice than in p21+/+ mice. In addition, p21-/- mice exhibited larger calluses and lower bone mineral density. IHC analyses showed that p21-/- mice exhibited delayed endochondral ossification via the Ihh-Runx2-Osterix pathway in vivo. Down-regulation of p21 expression in ATDC5 cells delayed endochondral ossification in vitro. CONCLUSIONS: p21 deficiency leads to delayed endochondral ossification by attenuating the Ihh-Runx2-Osterix pathway in vivo, and p21 deficiency in hypertrophic chondrocytes causes delayed differentiation of hypertrophic chondrocytes in vitro. p21 plays a role in endochondral ossification during fracture healing.

A Novel Model of Hip Femoroacetabular Impingement in Immature Rabbits Reproduces the Distinctive Head-Neck Cam Deformity.

BACKGROUND: Femoroacetabular impingement (FAI) is a leading cause of hip pain in young adults and often leads to degenerative osteoarthritis (OA). A small animal model of hip deformities is crucial for unraveling the pathophysiology of hip OA secondary to FAI. PURPOSES: To (1) characterize a new minimally invasive surgical technique to create a proximal femoral head-neck deformity in a skeletally immature rabbit model and (2) document the effect of an injury to the medial proximal femoral epiphysis on head-neck morphology at 28 days after the injury. STUDY DESIGN: Controlled laboratory study. METHODS: Six-week-old New Zealand White rabbits (n = 10) were subjected to right hip surgery, with the left hip used as a control. An epiphyseal injury in the medial femoral head was created using a 1.6-mm drill. Hips were harvested bilaterally at 28 days after surgery. Alpha and epiphyseal shaft angles were measured on radiographs. Alpha angles at the 1- and 3-o'clock positions were measured on the oblique axial plane of micro-computed tomography images. Bone bar formation secondary to growth plate injuries was confirmed using alcian blue hematoxylin staining. RESULTS: All hips in the study group showed a varus-type head-neck deformity, with lower epiphyseal shaft angles on anteroposterior radiographs versus those in the control group (133°± 8° vs 142°± 5°, respectively; P = .022) and higher epiphyseal shaft angles on lateral radiographs (27°± 12° vs 10°± 7°, respectively; P < .001). The mean alpha angles in the study group were higher at both the 1- (103°± 14° vs 46°± 7°, respectively; P < .002) and 3-o'clock (99°± 18° vs 35°± 11°, respectively; P < .002) positions than those in the control group. Alcian blue hematoxylin staining of all hips in the study group indicated that the injured physis developed a bony bar, leading to growth plate arrest on the medial femoral head. CONCLUSION: The proposed model led to growth arrest at the proximal femoral physis, resulting in a femoral head-neck deformity similar to human FAI. CLINICAL RELEVANCE: Our novel small animal model of a femoral head-neck deformity is a potential platform for research into the basic mechanisms of FAI disease progression and the development of disease-modifying therapies.

Patients' Characteristics Can Predict Clinical Outcomes Following Hip Arthroscopy by Reflecting the Patterns of Labral Tears: A Retrospective Observational Study.

PURPOSE: To evaluate the relationship between morphological differences in labral tears and clinical features of the hip joint in patients who underwent hip arthroscopy. MATERIALS AND METHODS: We retrospectively analyzed data from patients who underwent arthroscopic surgery for the treatment of labral tears. Hip labral tears were morphologically classified as longitudinal peripheral tears (group L), radial fibrillated tears (group FI), radial flaps (group FL), and an unstable labrum (group U). Radiographically, the center-edge angle, acetabular roof obliquity, vertical-center-anterior angle, alpha angle, femoral head-neck offset ratio, and crossover sign were evaluated and compared among the groups. The relationship between labral morphology and these radiographic findings, as well as clinical findings, such as age, gender, preoperative range of hip motion, and the clinical outcomes using modified Harris Hip Score (mHHS) were also examined. RESULTS: This study included fifty patients. Groups L and FI were often observed in late middle-aged patients with relatively shallow acetabular coverage. Group FL tears were frequently observed in young males with radiographic features, such as femoroacetabular impingement (FAI), compared to the other groups. Group U comprised mostly young females with relatively shallow acetabular coverage compared to the other groups. For the postoperative mHHS, group FL showed the best score among all groups, with a significant difference between groups FL and FI (p = 0.034). CONCLUSIONS: Our study revealed that morphologically, different labral tears were associated with different clinical features and radiological findings. Especially, our study can provide predictive findings for hip arthroscopists that younger males with FAI show better clinical outcomes when compared to middle-aged females with shallow acetabulum, which is indicative of degenerative hip labral tears. LEVEL OF EVIDENCE: IV case series.

Posterior Ankle Arthroscopy for Flexor Hallucis Longus Entrapment: A Case Report.

INTRODUCTION: Stenosing tenosynovitis is a chronic disorder frequently observed in finger triggering of a digit. Regarding the toes, although entrapment of the flexor hallucis longus (FHL) has already been reported in a few cases among sports players, the clinical condition is uncommon. Besides, the case without any specific causes is particularly rare. CASE REPORT: We report the case of a 26-year-old male with FHL entrapment. Even though he was unaware of any cause, he felt tenderness on the posteromedial side of his left ankle, and his great toe was locked in the flex position. Magnetic resonance imaging indicated effusion in the tendon sheath of the FHL and the possibility of a partial tear of the FHL. We hypothesized that the scar tissue secondary to the partial tear of the FHL may have been irritated at the retrotalar pulley below the sustentaculum tali, where the FHL glides. Therefore, posterior ankle arthroscopy was performed for the treatment of the FHL entrapment. CONCLUSION: Orthopedic surgeons should list this pathology as a differential diagnosis of posterior ankle pain, even in non-athletes.

The Hip Arthroscopy Post-less Procedure Impingement (HAPPI) Technique: Achieving Distraction With Standard Hip Tables at Zero Additional Cost.

Hip arthroscopy is becoming a more common treatment for femoroacetabular impingement, labral tears, and a variety of other hip pathologies. Unlike arthroscopy of the shoulder and knee, hip arthroscopy requires a significant amount of traction to gain access into the joint. Historically, traction has been achieved with the use of a perineal post. The use of a perineal post in hip arthroscopy can cause several avoidable complications such as neuropraxias (i.e., the pudendal nerve), vaginal or scrotal injuries, and perineal skin injuries. Several articles have been proposed using post-less techniques to obtain hip traction, but many of these techniques are expensive and require the purchase of new equipment. The purpose of this Technical Note is to describe a post-less technique for hip arthroscopy, the HAPPI technique (hip arthroscopy post-less procedure impingement), which is more affordable and does not require the addition of any special equipment. In addition, we will review some pearls and pitfalls, as well as advantages and disadvantages of the proposed technique.

Distinct Pattern of Inflammation of Articular Cartilage and the Synovium in Early and Late Hip Femoroacetabular Impingement.

BACKGROUND: The molecular mechanism of how femoroacetabular impingement (FAI) morphology leads to hip osteoarthritis (OA) is yet to be determined. The expression and location of inflammation-related molecules during early- and late-stage FAI have not been previously described. Moreover, the characterization of intra-articular inflammation away from the cam deformity as well as the nature of adjacent synovial tissue have also not been extensively reported. HYPOTHESIS: Early-stage FAI has a similar expression of inflammation-related markers in the head-neck and acetabular cartilage but less synovitis than late-stage FAI. STUDY DESIGN: Controlled laboratory study. METHODS: Head-neck cartilage, acetabular cartilage, and synovial samples were obtained from patients undergoing hip preservation surgery for the treatment of symptomatic cam FAI (early FAI group; n = 15) and advanced OA secondary to cam FAI (late FAI group; n = 15). Samples procured from healthy young adult donors served as the control group (n = 7). Cartilage degeneration was assessed by histology, and the expression of inflammation-related proteins (interleukin-1 beta [IL-1ß], matrix metalloproteinase-13 [MMP-13], a disintegrin and metalloproteinase with thrombospondin motifs-4 [ADAMTS-4], type II collagen [COL2], and aggrecan neoepitope [NITEGE]) was measured by immunostaining. Synovial samples in the early and late FAI groups were examined for synovitis and the expression of IL-1ß. RESULTS: Head-neck cartilage in the early FAI group showed significantly more degeneration than the control group and an increased expression of inflammation-related proteins (IL-1ß: 69.7% ± 18.1% vs 20.2% ± 4.9%, respectively; MMP-13: 79.6% ± 12.6% vs 25.3% ± 9.5%; ADAMTS-4: 83.9% ± 12.2% vs 24.3% ± 11.1%; NITEGE: 89.7% ± 7.7% vs 39.8% ± 20.5%) (P < .001). Head-neck and acetabular cartilage in the early and late FAI groups showed a similar degree of degeneration. Moreover, a similar expression of inflammation-related proteins was observed between the early and late FAI groups for head-neck cartilage (IL-1ß: 69.7% ± 18.1% vs 72.5% ± 13.2%; MMP-13: 79.6% ± 12.6% vs 71.4% ± 18.8%; ADAMTS-4: 83.9% ± 12.2% vs 82.6% ± 12.5%; COL2: 93.6% ± 3.9% vs 92.5% ± 5.8%; NITEGE: 89.7% ± 7.7% vs 95.7% ± 4.7%) and acetabular cartilage (IL-1ß: 83.3% ± 24.8% vs 80.7% ± 15.6%; MMP-13: 94.3% ± 9.7% vs 85.2% ± 12.3%; ADAMTS-4: 98.5% ± 2.3% vs 98.4% ± 3.4%; COL2: 99.8% ± 0.7% vs 99.7% ± 1.1%; NITEGE: 96.7% ± 6.7% vs 99.2% ± 2.2%). In contrast, synovitis was minimal with a low expression of IL-1ß in the early FAI group compared with the late FAI group. CONCLUSION: Hip cartilage exhibited an OA phenotype in patients with early-stage FAI, similar to what was observed in hip OA secondary to FAI. Severe synovitis was only evident with late-stage FAI. CLINICAL RELEVANCE: This study supports the concept that early hip impingement is associated with cartilage degeneration and catabolism.

Inflammatory Response of Articular Cartilage to Femoroacetabular Impingement in the Hip.

BACKGROUND: Femoroacetabular impingement (FAI) has been proposed as an etiologic factor in up to 50% of hips with osteoarthritis (OA). Inflammation is thought to be one of the main initiators of OA, yet little is known about the origin of intra-articular inflammation in FAI hips. HYPOTHESIS: Articular cartilage from the impingement zone of patients with FAI has high levels of inflammation, reflecting initial inflammatory process in the hip. STUDY DESIGN: Controlled laboratory study. METHODS: Head-neck cartilage samples were obtained from patients with cam FAI (cam FAI, early FAI; n = 15), advanced OA secondary to cam FAI (FAI OA, late FAI; n = 15), and advanced OA secondary to developmental dysplasia of the hip (DDH OA, no impingement; n = 15). Cartilage procured from young adult donors (n = 7) served as control. Safranin O-stained sections were assessed for cartilage abnormality. Tissue viability was detected by TUNEL assay. Immunostaining of interleukin 1ß (IL-1ß), catabolic markers (matrix metalloproteinase 13 [MMP-13], a disintegrin and metalloproteinase with thrombospondin motif 4 [ADAMTS-4], aggrecan antibody to C-terminal neoepitope [NITEGE]), and an anabolic marker (type II collagen [COL2]) was performed to evaluate molecular inflammation and metabolic activity. The average percentage of immunopositive cells from the total cell count was calculated. Kruskal-Wallis test followed by Steel-Dwass post hoc test was used for multiple comparisons. RESULTS: Microscopic osteoarthritic changes were more prevalent in cartilage of cam FAI and FAI OA groups compared with DDH OA and control groups. Cartilage in cam FAI and FAI OA groups, versus the DDH group, had higher expression of inflammatory molecules IL-1ß (69.7% ± 18.1% and 72.5% ± 13.2% vs 32.7% ± 14.4%, respectively), MMP-13 (79.6% ± 12.6% and 71.4% ± 18.8% vs 38. 5% ± 13.3%), ADAMTS-4 (83.9% ± 12.2% and 82.6% ± 12.5% vs 45.7% ± 15.5%), and COL2 (93.6% ± 3.9% and 92.5% ± 5.8% vs 53.3% ± 21.0%) (P < .001). Expression of NITEGE was similar among groups (cam FAI, 89.7% ± 7.7%; FAI OA, 95.7% ± 4.7%; DDH OA, 93.9% ± 5.2%; P = .0742). The control group had minimal expression of inflammatory markers. Inflammatory markers were expressed in all cartilage zones of early and late FAI but only in the superficial zone of the no impingement group. CONCLUSION: Cartilage from the impingement zone in FAI is associated with a high expression of inflammatory markers, extending throughout all cartilage zones. CLINICAL RELEVANCE: Inflammation associated with FAI likely has a deleterious effect on joint homeostasis. Further clinical and translational studies are warranted to assess whether and how surgical treatment of FAI reduces molecular inflammation.

Deletion of p21 expression accelerates cartilage tissue repair via chondrocyte proliferation.

Articular cartilage tissue has a poor healing potential, and when subjected to traumatic damage this tissue undergoes cartilage degeneration and osteoarthritis. The association between the regulation of cell cycle checkpoints and tissue regeneration has been previously investigated, and p21 was initially identified as a potent inhibitor of cell cycle progression. However, the effects of p21 defects on damaged tissue remain controversial. Therefore, the aim of the present study was to evaluate the effects of p21 deficiency on cartilage repair. A mouse model of articular cartilage repair was generated by inducing a patellar groove scratch in 8­week­old p21­knockout (KO) mice and C57Bl/6 wild­type (WT) mice. Mice were sacrificed at 4 and 8 weeks post­surgery. The present study also investigated the effect of p21 deficiency on cartilage differentiation in ATDC5 cells in vitro. Safranin O staining results indicated that cartilage repair initially occurred in p21 KO mice. In addition, immunohistochemical analysis demonstrated that p21 KO upregulated proliferating cell nuclear antigen and increased cell proliferation. However, type II collagen and Sox9 expression levels remained unchanged in p21 KO and WT mice. Moreover, it was identified that p21 downregulation did not affect Sox9 and type II collagen expression levels in vitro. Furthermore, p21 deficiency promoted healing of articular cartilage damage, which was associated with cell proliferation in vivo, and increased chondrocyte proliferation but not differentiation in vitro. Therefore, the present results suggested that p21 does not affect Sox9 or type II collagen expression levels during cartilage differentiation in the repair of cartilage defects.

Surgical Treatment of Gluteus Medius Tears Augmented With Allograft Human Dermis.

Greater trochanteric pain syndrome can be caused by gluteus medius and minimus tendinopathy/tears and chronic trochanteric bursitis. Specifically, moderate-to-severe abductor tendon tears can cause severe lateral hip pain, limp, and abnormal gait. A variety of open and endoscopic techniques to treat glut abductors hip tears have been described. The use of scaffolds, such as acellular human dermal allograft, to augment tendon repair, already has been successfully reported in rotator cuff repairs of the shoulder. Still, the use of acellular human dermal allograft in the hip has been limited. However, there are some clinical scenarios in which augmentation of abductors hip tendon repair with scaffold is indicated. Chronic or massive gluteus tears or revision cases may benefit from augmentation with a scaffold. The purpose of this technical note and accompanying video is to describe our indications, pearls, and pitfalls of repair of moderate to severe gluteus tears via a minimally invasive technique augmented with acellular human dermal allograft.

Downregulation of aquaporin 9 decreases catabolic factor expression through nuclear factor­κB signaling in chondrocytes.

Aquaporins (AQPs) are small integral membrane proteins that are essential for water transport across membranes. AQP9, one of the 13 mammalian AQPs (including AQP0 to 12), has been reported to be highly expressed in hydrarthrosis and synovitis patients. Given that several studies have identified signal transduction as an additional function of AQPs, it is hypothesized that AQP9 may modulate inflammatory signal transduction in chondrocytes. Therefore, the present study used a model of interleukin (IL)­1ß­induced inflammation to determine the mechanisms associated with AQP9 functions in chondrocytes. Osteoarthritis (OA) and normal cartilage samples were subjected to immunohistological analysis. In addition, matrix metalloproteinase (MMP)3, MMP13 and a disintegrin and metalloproteinase with thrombospondin motifs 5 (ADAMTS­5) mRNA and protein analysis was conducted in normal human articular chondrocytes from the knee (NHAC­Kn) stimulated with IL­1ß by reverse transcription­polymerase chain reaction (RT­qPCR) and western blotting, respectively. AQP9 knockdown was also performed by transfection of AQP9­specific small interfering RNA using Lipofectamine. AQP1, 3, 7, 9 and 11 mRNA expression levels were detected in OA human chondrocytes and in IL­1ß­treated normal human chondrocytes. The levels of AQP9, MMP­3, MMP­13 and ADAMTS­5 mRNA were increased by treatment with 10 ng/ml IL­1ß in a time­dependent manner, while knockdown of AQP9 expression significantly decreased the mRNA levels of the MMP3, MMP13 and ADAMTS­5 genes, as well as the phosphorylation of IκB kinase (IKK). Treatment with a specific IKK inhibitor also significantly decreased the expression levels of MMP­3, MMP­13 and ADAMTS­5 in response to IL­1ß stimulation. Furthermore, immunohistochemical analysis demonstrated that AQP9 and inflammatory markers were highly expressed in OA cartilage. In addition, the downregulation of AQP9 in cultured chondrocytes decreased the catabolic gene expression in response to IL­1ß stimulation through nuclear factor­κB signaling. Therefore, AQP9 may be a promising target for the treatment of OA.

Depletion of aquaporin 1 decreased ADAMTS­4 expression in human chondrocytes.

Inflammation serves an important role in the progression of osteoarthritis (OA), and IL­1ß may act as a catabolic factor on cartilage, reducing the synthesis of primary cartilage components type II collagen and aggrecan. Aquaporin 1 (AQP1) is a 28­kDa water channel formed of six transmembrane domains on the cell membrane. AQP1 is highly expressed in the anus, gallbladder and liver, and is moderately expressed in the hippocampus, ependymal cells of the central nervous system and articular cartilage. It was hypothesized that AQP1 may be highly expressed in OA cartilage and that it may increase the expression of catabolic factors during inflammatory OA progression. Therefore, the present study evaluated AQP1 functions in human OA articular chondrocytes. Primary chondrocytes were isolated from human hip and knee cartilage tissues, cultured and transfected with AQP1­specific small interfering RNA with or without subsequent IL­1ß treatment. In vitro explant culture from hip cartilages were also prepared. Reverse transcription­polymerase chain reaction (RT­PCR) was performed to assess the expression of AQP genes in human articular cartilage, AQP1 immunohistochemistry of the cartilages and explant culture, as well as RT­quantitative PCR, western blotting and immunocytochemistry/immunofluorescence of OA chondrocytes to evaluate the expression of AQP1, and catabolic and anabolic factors. RT­PCR results demonstrated that AQP0, 1, 3, 7, 9, and 11 were expressed in OA chondrocytes. Immunohistochemistry revealed that AQP1 was highly expressed in the superficial to middle zones of OA articular cartilages. Additionally, AQP1 mRNA was significantly higher in OA cartilage and IL­1ß treatment significantly increased AQP1 expression in hip explant cartilage. Furthermore, AQP1 downregulation decreased a disintegrin and metalloprotease with thrombospondin motifs (ADAMTS)­4 expression in OA chondrocytes, though it did not affect other associated genes. Immunofluorescence showed that AQP1 and ADAMTS­4 were co­localized. These findings indicated that AQP1 depletion may decrease ADAMTS­4 expression in human OA chondrocytes. Therefore, regulating AQP1 expression may be a strategy to suppress catabolic factors during OA progression.

Cyclin-Dependent Kinase Inhibitor-1-Deficient Mice are Susceptible to Osteoarthritis Associated with Enhanced Inflammation.

Osteoarthritis (OA) is a multifactorial disease, and recent data suggested that cell cycle-related proteins play a role in OA pathology. Cyclin-dependent kinase (CDK) inhibitor 1 (p21) regulates activation of other CDKs, and recently, we reported that p21 deficiency induced susceptibility to OA induced by destabilization of the medial meniscus (DMM) surgery through STAT3-signaling activation. However, the mechanisms associated with why p21 deficiency led to susceptibility to OA by the STAT3 pathway remain unknown. Therefore, we focused on joint inflammation to determine the mechanisms associated with p21 function during in vitro and in vivo OA progression. p21-knockout (p21-/- ) mice were used to develop an in vivo OA model, and C57BL/6 (p21+/+ ) mice with the same background as the p21-/- mice were used as controls. Morphogenic changes were measured using micro-CT, IL-1ß serum levels were detected by ELISA, and histological or immunohistological analyses were performed. Our results indicated that p21-deficient DMM-model mice exhibited significant subchondral bone destruction and cartilage degradation compared with wild-type mice. Immunohistochemistry results revealed p21-/- mice susceptibility to OA changes accompanied by macrophage infiltration and enhanced MMP-3 and MMP-13 expression through IL-1ß-induced NF-κB signaling. p21-/- mice also showed subchondral bone destruction according to micro-CT analysis, and cathepsin K staining revealed increased numbers of osteoclasts. Furthermore, p21-/- mice displayed increased serum IL-1ß levels, and isolated chondrocytes from p21-/- mice indicated elevated MMP-3 and MMP-13 expression with phosphorylation of IκB kinase complex in response to IL-1ß stimulation, whereas treatment with a specific p-IκB kinase inhibitor attenuated MMP-3 and MMP-13 expression. Our results indicated that p21-deficient DMM mice were susceptible to alterations in OA phenotype, including enhanced osteoclast expression, macrophage infiltration, and MMP expression through IL-1ß-induced NF-κB signaling, suggesting that p21 regulation may constitute a possible therapeutic strategy for OA treatment. © 2017 American Society for Bone and Mineral Research.

P21 deficiency delays regeneration of skeletal muscular tissue.

The potential relationship between cell cycle checkpoint control and tissue regeneration has been indicated. Despite considerable research being focused on the relationship between p21 and myogenesis, p21 function in skeletal muscle regeneration remains unclear. To clarify this, muscle injury model was recreated by intramuscular injection of bupivacaine hydrochloride in the soleus of p21 knockout (KO) mice and wild type (WT) mice. The mice were sacrificed at 3, 14, and 28 days post-operation. The results of hematoxylin-eosin staining and immunofluorescence of muscle membrane indicated that muscle regeneration was delayed in p21 KO mice. Cyclin D1 mRNA expression and both Ki-67 and PCNA immunohistochemistry suggested that p21 deficiency increased cell cycle and muscle cell proliferation. F4/80 immunohistochemistry also suggested the increase of immune response in p21 KO mice. On the other hand, both the mRNA expression and western blot analysis of MyoD, myogenin, and Pax7 indicated that muscular differentiation was delayed in p21KO mice. Considering these results, we confirmed that muscle injury causes an increase in cell proliferation. However, muscle differentiation in p21 KO mice was inhibited due to the low expression of muscular synthesis genes, leading to a delay in the muscular regeneration. Thus, we conclude that p21 plays an important role in the in vivo healing process in muscular injury.

Oxidative stress-induced apoptosis and matrix loss of chondrocytes is inhibited by eicosapentaenoic acid.

Eicosapentaenoic acid (EPA) is an antioxidant and n-3 polyunsaturated fatty acid that reduces the production of inflammatory cytokines. We evaluated the role of EPA in chondrocyte apoptosis and degeneration. Normal human chondrocytes were treated with EPA and sodium nitroprusside (SNP). Expression of metalloproteinases (MMPs) was detected by real-time polymerase chain reaction (PCR) and that of apoptosis-related proteins was detected by western blotting. Chondrocyte apoptosis was detected by flow cytometry. C57BL/6J mice were used for the detection of MMP expression by immunohistochemistry and for investigation of chondrocyte apoptosis. EPA inhibited SNP-induced chondrocyte apoptosis, caspase 3 and poly(ADP-ribose) polymerase cleavage, phosphorylation of p38 MAPK and p53, and expression of MMP3 and MMP13. Intra-articular injection of EPA prevented the progression of osteoarthritis (OA) by inhibiting MMP13 expression and chondrocyte apoptosis. EPA treatment can control oxidative stress-induced OA progression, and thus may be a new approach for OA therapy.