It is well accepted that increased age is a key risk factor for osteoarthritis in the joints typically affected including hands, hips, knees, and spine. The rare cases of osteoarthritis diagnosed in people under the age of 25–30 years are almost always due to mutations in matrix genes that cause significant structural abnormalities and/or joint deformities. Osteoarthritis is a prototypic ‘degenerative’ condition of aging that in the past had been attributed to an accumulation of ‘wear and tear’ type damage to joint tissues. As both general aging research and osteoarthritis-specific research have progressed, it has become increasingly clear that conditions associated with aging, such as osteoarthritis, are much more complex than an accumulation of damage over time, and that the term degenerative disease is a vast oversimplification. This review focuses on research published over the past 18 months in the field of aging and osteoarthritis that has contributed to the advancement in our knowledge of this complex and multifactorial disease. Although intervertebral disc aging likely contributes to osteoarthritis in the spine, this area of research is not covered here.
Osteoarthritis in humans usually becomes symptomatic after the age of about 50 years that is also when the radiographic changes of osteoarthritis become more common. Although radiographic signs of osteoarthritis, such as osteophytes and joint space narrowing, do not always correlate well with symptoms, epidemiologic studies of large cohorts commonly depend on radiographs to define osteoarthritis. A recent study [1▪] from the Chingford Women's study, which is a community-based cohort living in the United Kingdom, showed that at baseline, when the median age was 53 years, 13.7% of the patients had radiographic knee osteoarthritis. Over the 14 subsequent years of the study, the annual cumulative incidence was 2.3%. The affect of age was apparent when the annual rate between baseline and year 5 (2.3%) was compared with the rate between years 10 and 15 (3.3%). Demonstrating that age is not the only risk factor for osteoarthritis, just over half of the entire cohort (52.2%) remained free of radiographic knee osteoarthritis over the 14 years of follow-up when the median age had reached 68 years, but obese individuals had an almost 20% higher cumulative incidence of knee osteoarthritis, compared with both normal and overweight categories.
Atherosclerosis is another common age-related multifactorial disease. In the Rotterdam population of men and women aged 55 years and older, radiographic osteoarthritis and atherosclerosis (measured as carotid intima–media thickness or plaque) were associated, particularly for hand osteoarthritis in the distal interphalangeal joints and carotid plaque in women (OR 1.4, CI 1.19–1.65) . The risk for hand osteoarthritis has been thought to be more associated with systemic or metabolic factors than knee osteoarthritis, which has a greater association with joint injury. Systemic inflammation, which is associated with aging, could play a role in both diseases. Previous studies have associated systemic markers of inflammation, such as high sensitivity C-reactive protein (CRP) with radiographic osteoarthritis and atherosclerosis, and a recent study [3▪] of patients reported an association of serum levels of high sensitivity CRP, as well as tumor necrosis factor α, with worsening knee pain.
Because articular cartilage lacks a nerve supply, joint pain is thought to be due to osteoarthritis-related changes in the synovium, the bone (including osteophytes), the joint capsule, the ligaments, and, in the knee, the meniscus . Age-related processes in these tissues and their role in the development of osteoarthritis have been less well studied than in the articular cartilage. However, as will be reviewed below, recent studies demonstrate age-related changes in these tissues that may be important to the development of osteoarthritis. Most of this work is at the histological level, whereas recent studies of aging in cartilage have focused more on cellular changes that will be reviewed as well.
AGING CHANGES IN TISSUES OTHER THAN CARTILAGE THAT MAY CONTRIBUTE TO OSTEOARTHRITIS
Injury to the meniscus is an important risk factor for knee osteoarthritis. MRI studies have shown that meniscal lesions are common in people with knee osteoarthritis, even without a known history of joint injury [5▪], and meniscal lesions are associated with knee pain . A pathological study of menisci from deceased human tissue donors by Pauli et al.[7▪▪] revealed age-related changes including increased Safranin-O staining, decreased cell density with associated appearance of acellular zones, mucoid degeneration, and loss of collagen fiber organization. These changes are similar to those seen in other tissues with aging, such as articular cartilage. The exception is the increased Safranin-O staining. Increased staining would suggest a focal increase in proteoglycans that would be difficult to explain unless cells in the aged meniscus were producing more proteoglycans. An alternative is that loss of some other matrix protein, such as collagen, allowed increased dye binding. As expected, menisci taken from knees that exhibited significant cartilage disorder showed more severe matrix disruption, areas of calcification, and large areas of cell loss accompanied by focal areas of cell clusters, similar to the clusters seen in osteoarthritis cartilage.
A study  that compared normal bovine meniscus obtained from fetal, juvenile, and adult animals demonstrated developmental changes that result in increased size of the meniscus and maturation of the mechanical properties of the tissue but also found the repair capacity of the tissue, when studied in vitro after full-thickness defects were created, was reduced with age. In a human study [9▪], expression of a panel of proinflammatory and matrix degrading genes was evaluated using RNA from meniscal tissue obtained during partial meniscectomy in patients with either an isolated meniscal tear or one accompanying an ACL tear. An age-related difference in gene expression was noted such that samples from patients under the age of 40 years exhibited higher interleukin-1β, a disintegrin and metalloproteinase with thrombospondin motif, matrix metalloproteinase (MMP) 1, MMP-9, MMP-13, and nuclear factor kappa B. This finding suggests that younger patients exhibit a greater inflammatory and catabolic response after injury that would seem to contradict the notion that aging promotes a more proinflammatory state. However, an active response to acute injury could be important to the body's attempt at wound repair and aging may be more associated with a chronic low-grade proinflammatory state that contributes to degenerative-type meniscal tears.
In addition to aging studies in the human meniscus, the anterior [10▪▪] and posterior  cruciate ligaments (ACL and PCL, respectively) have also been studied. Similarly to the meniscus work noted above by Pauli et al.[7▪▪], the same group used human donor tissue and compared changes in the ACL and PCL with those in cartilage from the same joints. Even though the patients did not have a known history of joint trauma, ligamentous changes were common in older adults and included ACL rupture in 10% of the knees but no ruptures of the PCL. The correlations between histologic scores of degeneration in the ligaments and cartilage were also much stronger for the ACL than the PCL and, unlike the ACL, the PCL changes were not significantly correlated with donor age. A difference from the meniscus study, in which inflammatory cell infiltration was not apparent, was the finding of ligament sheath inflammation that was much more common in the ACL than PCL and increased with age. These studies suggest that aging changes, particularly in the ACL, likely contribute to age-related osteoarthritis.
In addition to the soft tissues discussed above, bone is affected in osteoarthritis with osteophyte formation, subchondral plate thickening, and focal areas of necrosis and remodeling that appear as bone marrow lesions on MRI in areas of greatest mechanical stress (reviewed in ). The bone lesions could be attributed to aging processes in the osteocytes that are important regulators of the bone remodeling that occurs in response to mechanical stress. An autopsy study  using normal human femoral bone from 32 tissue donors with ages in a range between the 1st and 9th decade noted an age-dependent decrease in total osteocyte lacunar number and hypermineralization of the osteocyte lacunae that were present that has been called micropetrosis. These changes could reduce the capacity of bone to respond to mechanical stress and make the bone more brittle. Consistently with a role in osteoarthritis, similar findings of a reduction in osteocyte lacunar number and an increase in micropetrosis were noted in a study [13▪] of bone from patients with hip osteoarthritis.
RODENT MODELS OF AGING AND OSTEOARTHRITIS
Because of the difficulties studying cellular and molecular processes in human joint tissue in vivo, animal models are necessary. A comprehensive histological study  of knee joint tissues from male C57BL/6 mice demonstrated that lesions induced by a commonly used surgical procedure (destabilized medial meniscus or DMM model) were similar to those that developed naturally with age. This study supported previous work demonstrating the importance of chondrocyte death, with both aging and osteoarthritis as well as an age-related decline in cartilage thickness. The lesions were most prominent in the medial tibial plateau, which is also the most common site for human osteoarthritis. The findings were consistent with a histological study  of young and old male Wistar rats that also had a reduction in cell number and in noncalcified cartilage thickness in the medial plateau.
In a study [16▪] that compared the severity of histologic osteoarthritis and changes in gene expression between young adult (12-week old) and middle-aged (older) adult (12-month old) male C57BL/6 mice, the articular cartilage lesions developing after DMM surgery in the older animals were twice as severe as those in the young. Significant differences in the patterns of gene expression were noted when RNA isolated from the joint organ (cartilage, subchondral bone, meniscus, and joint capsule with synovium) was analyzed by microarray. More genes were upregulated in the older mice relative to the young and differentially expressed genes included muscle structure and development, and immunoglobulin domain genes. In order to study which genes changed with age, independently of the DMM surgery to induce osteoarthritis, the sham control joints were compared. This revealed an age-related decrease in matrix gene expression and an increase in immune and defense response genes. Of interest was that many of the genes in these categories that went down with age were found to be re-expressed in the older DMM joints.
Studies using transgenic and knockout mice continue to provide information on specific genes that may play a role in the osteoarthritis process . Most commonly, investigators induce osteoarthritis in very young animals around the ages of 8–12 weeks. The differences in gene expression between 12-week-old and 12-month-old mice noted above suggest that caution will be needed when interpreting studies only done in young mice that are the equivalent of a teenaged human. However, two recent studies have examined both naturally occurring osteoarthritis with age and surgically induced osteoarthritis and found results that were similar in the two models. The role of glycosphingolipid synthesis in osteoarthritis was investigated by studying the effect of knocking out a key enzyme required for glycosphingolipid synthesis and then examining the development of both naturally occurring osteoarthritis in aged animals as well as surgically induced osteoarthritis in young animals . They showed worse osteoarthritis developed in the knockouts in both models. Another study [19▪] examined conditional deletion of the fibroblast growth factor receptor 1 (FGFR1) in both aging-associated spontaneous osteoarthritis at 12 months of age and in the DMM model with surgery performed on 10-week-old animals. They noted proteoglycan loss at 12 months of age was less severe in the FGFR1 knockouts as were cartilage lesions at 8 weeks after surgery in the DMM model using younger mice. This study contrasts with a previous study  in which FGFR3 knockout mice were found to develop more severe osteoarthritis with age. Together, these studies suggest an anabolic/joint protective function for FRGFR3 and a catabolic/joint destructive function for FGFR1.
Dual functions for a growth factor in cartilage homeostasis and age-related osteoarthritis that depend on differences in signals generated by specific receptors have also been suggested by studies of the activin receptor-like kinases (ALKs) activated by transforming growth factor-beta. ALK5 activation has been shown to be proanabolic and ALK1 activation is procatabolic . With aging and in osteoarthritis, the ratio of ALK1 to ALK5 is increased promoting the development of osteoarthritis and similar changes may occur with the ratio of FGFR1 to FGFR3 .
There has been a growing interest in the aging field about the role epigenetics plays in age-related conditions including osteoarthritis. Epigenetic regulation of gene expression includes DNA methylation, histone acetylation and methylation, and microRNA (miRNA). Sirtuins are a family of NAD+-dependent deacetylases that have been linked to aging and more recently shown to be involved in osteoarthritis through regulation of cellular energy and metabolism . The sirtuin SirT1 promotes chondrocyte survival and matrix gene expression. TNFα was found to cleave and inactivate SirT1, and thereby contribute to reduced matrix gene expression . Heterozygous SirT1 (+/−) mice with a significant decrease in SirT1 were found to develop premature osteoarthritis-like changes at 9 months of age that was associated with increased chondrocyte apoptosis [25▪].
Recent evidence for altered DNA methylation in osteoarthritis was provided by a study [26▪] that found that a site within the MMP-13 promoter was demethylated in osteoarthritis chondrocytes resulting in increased MMP-13 expression mediated by cAMP response element binding at the demethylated site. Histone methylation has been implicated in the age-dependent expression of the transcription factor nuclear factor of activated T cells 1 that promotes cartilage homeostasis . Finally, several studies have recently reported on the role of miRNAs including work, showing changes with age. In one study , miR-199a-3p and miR-193b were upregulated with age, whereas miR-320c was decreased. The two upregulated miRNAs were found to reduce collagen and aggrecan expression in vitro suggesting they may be antianabolic and involved in the age-related decrease in matrix gene expression.
The study of cell senescence continues to be an active area in osteoarthritis and aging research. Telomere shortening and/or telomere damage have been noted to occur with aging and have been suggested to contribute to osteoarthritis. Rose et al.[29▪] compared normal cartilage from autopsy specimens and osteoarthritis cartilage from knee replacements but did not find evidence of telomere shortening. However, another study  of human osteoarthritis knee cartilage that used a unique assay for ultra-short single telomeres found increased numbers of ultra-short telomers per cell in osteoarthritis cartilage lesions but without a reduction in mean telomere length. In another study , passaged human bone marrow mesenchymal stem cells were shown to exhibit replicative senescence with telomere shorting that was prevented by culturing in the presence of estrogen. This work, if replicated, could be relevant to the use of mesenchymal cells for cartilage repair that may be limited by the age of the donor.
The study by Rose et al.[29▪] was successful in demonstrating DNA damage in osteoarthritis cartilage and that agents that cause oxidative stress or genotoxic stress can promote DNA damage and senescence-associated β-galactosidase activity in vitro. They also used immunostaining for two proteins that should be found in most chondrocytes, vimentin and S-100 protein and found heterogeneity in cells double labeled for both proteins in osteoarthritis cartilage and in normal chondrocytes with induced DNA damage. They concluded that these findings are consistent with stress-induced senescence resulting in a chaotic gene activation pattern. Stress-induced senescence, as opposed to replicative senescence, does seem more likely in a tissue such as articular cartilage in which cell division is rare. More studies with additional genes will be needed to support the concept that this leads to discoordinated or chaotic gene expression.
One of the genetic conditions that results in premature aging is Hutchinson–Gilford progeria syndrome resulting from mutations in lamin A. Patients with this syndrome have premature osteoporosis but osteoarthritis has not been well documented . However, Attur et al. found increased expression of lamin A in osteoarthritis chondrocytes relative to normal and when overexpressed in chondrocytes in vitro, lamin A promoted expression of the senescence marker p21WAF1 but not p16INK4A. The latter makes the significance of the association between lamin A and chondrocyte senescence unclear. However, overexpression of lamin A did increase caspase-3 activation and terminal deoxynucleotidyl transferase dUTP nick end labeling staining for apoptosis that could be relevant to cell death in osteoarthritis.
An important feature of cell senescence that could contribute to joint tissue destruction in osteoarthritis is the development of the senescence-associated secretory phenotype (SASP). The SASP is characterized by increased production of proinflammatory mediators and matrix-degrading enzymes . The ‘secretome’ of cartilage has been studied using a proteomics approach [35,36], but studies to date have not focused on age-related differences. With aging, there is evidence for increased MMP-mediated cartilage matrix damage, and recently, both collagenases and cathepsin K have been implicated . Because of the central role of matrix-degrading enzymes in joint tissue destruction in osteoarthritis, future studies should determine whether cell senescence contributes to an increased release of these factors and determine the mechanisms involved.
Aging changes in the cartilage matrix and chondrocyte senescence are key contributors to the aging processes that promote age-associated osteoarthritis. However, recent work has begun to elucidate aging changes in the other joint tissues involved in osteoarthritis, including meniscus, ligaments, bone, and perhaps synovium. Studies in these tissues are mainly at the descriptive level, but mechanistic studies should be forthcoming. Because aging is a systemic process, it is likely that mechanisms can be found that apply to most, if not all, the joint tissues. Discovery of the common mechanisms will lead to novel interventions to slow the aging process, and thus slow the development of osteoarthritis.
Conflicts of interest
Dr R.F.L's. research was supported by grants from the National Institute of Arthritis, Musculoskeletal and Skin Diseases (R37AR049003–11), the Arthritis Foundation, the National Institute of Aging (P30 AG021332) and the Dorothy Rhyne and Willard Duke Kimbrell Professorship.
There are no relevant conflicts of interest related to this work.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
Additional references related to this topic can also be found in the Current World Literature section in this issue (pp. 153–154).
1▪. Leyland KM, Hart DJ, Javaid MK, et al. The natural history of radiographic knee osteoarthritis
: a fourteen-year population-based cohort study. Arthritis Rheum 2012; 64:2243–2251.
One of the longest natural history studies of knee osteoarthritis, this work nicely presents cummulative indcidence by age.
2. Hoeven TA, Kavousi M, Clockaerts S, et al.
Association of atherosclerosis with presence and progression of osteoarthritis
: the Rotterdam Study. Ann Rheum Dis 2012.
3▪. Stannus OP, Jones G, Blizzard L, et al.
Associations between serum levels of inflammatory markers and change in knee pain over 5 years in older adults: a prospective cohort study. Ann Rheum Dis 2012.
Demonstrates a potential link between inflammation and pain in knee osteoarthritis.
4. Loeser RF, Goldring SR, Scanzello CR, et al. Osteoarthritis
: a disease of the joint as an organ. Arthritis Rheum 2012; 64:1697–1707.
5▪. Englund M, Felson DT, Guermazi A, et al. Risk factors for medial meniscal pathology on knee MRI in older US adults: a multicentre prospective cohort study. Ann Rheum Dis 2011; 70:1733–1739.
One of several recent studies that emphasize the important link between meniscal damage and osteoarthritis of the knee and reports the factors involved.
6. Kim HA, Kim I, Song YW, et al. The association between meniscal and cruciate ligament damage and knee pain in community residents. Osteoarthritis
Cartilage 2011; 19:1422–1428.
7▪▪. Pauli C, Grogan SP, Patil S, et al. Macroscopic and histopathologic analysis of human knee menisci in aging and osteoarthritis
Cartilage 2011; 19:1132–1141.
One of the first and the largest study to date on aging changes in the meniscus that are likely relevant to osteoarthritis.
8. Ionescu LC, Lee GC, Garcia GH, et al. Maturation state-dependent alterations in meniscus integration: implications for scaffold design and tissue engineering. Tissue Eng Part A 2011; 17:193–204.
9▪. Brophy RH, Rai MF, Zhang Z, et al. Molecular analysis of age and sex-related gene expression in meniscal tears with and without a concomitant anterior cruciate ligament tear. J Bone Joint Surg Am 2012; 94:385–393.
A demonstration of proinflammatory and matrix-degrading gene expression in the injured meniscus that provides new insight on how meniscal injury could promote damage in neighboring joint tissues.
10▪▪. Hasegawa A, Otsuki S, Pauli C, et al. Anterior cruciate ligament changes in the human knee joint in aging and osteoarthritis
. Arthritis Rheum 2012; 64:696–704.
This study provides needed data on how aging affects a key ligament in the knee that plays a role in the development of osteoarthritis.
11. Levy YD, Hasegawa A, Patil S, et al.
Histopathological changes in the human posterior cruciate ligament during aging and osteoarthritis
: correlations with anterior cruciate ligament and cartilage changes. Ann Rheum Dis 2012.
12. Busse B, Djonic D, Milovanovic P, et al. Decrease in the osteocyte lacunar density accompanied by hypermineralized lacunar occlusion reveals failure and delay of remodeling in aged human bone. Aging Cell 2010; 9:1065–1075.
13▪. Carpentier VT, Wong J, Yeap Y, et al. Increased proportion of hypermineralized osteocyte lacunae in osteoporotic and osteoarthritic human trabecular bone: implications for bone remodeling. Bone 2012; 50:688–694.
Extends the important findings from reference  to osteoarthritis.
14. McNulty MA, Loeser RF, Davey C, et al. Histopathology of naturally occurring and surgically induced osteoarthritis
in mice. Osteoarthritis
Cartilage 2012; 20:949–956.
15. Moriyama H, Kanemura N, Brouns I, et al. Effects of aging and exercise training on the histological and mechanical properties of articular structures in knee joints of male rat. Biogerontology 2012; 13:369–381.
16▪. Loeser RF, Olex A, McNulty MA, et al. Microarray analysis reveals age-related differences in gene expression during the development of osteoarthritis
in mice. Arthritis Rheum 2012; 64:705–717.
Demonstrates the importance of age in the development of osteoarthritis and its affects on gene expression in a commonly used mouse model.
17. Little CB, Zaki S. What constitutes an ‘animal model of osteoarthritis
’: the need for consensus? Osteoarthritis
Cartilage 2012; 20:261–267.
18. Seito N, Yamashita T, Tsukuda Y, et al. Interruption of glycosphingolipid synthesis enhances osteoarthritis
development in mice. Arthritis Rheum 2012; 64:2579–2588.
19▪. Weng T, Yi L, Huang J, et al.
Genetic inhibition of FGFR1 in cartilage attenuates articular cartilage degeneration in adult mice. Arthritis Rheum 2012; doi: 10.1002/art.34645.
In conjunction with reference , it helps to explain the paradoxical role of bFGF in cartilage by demonstrating that the anabolic versus catabolic effects of bFGF are receptor dependent.
20. Valverde-Franco G, Binette JS, Li W, et al. Defects in articular cartilage metabolism and early arthritis in fibroblast growth factor receptor 3 deficient mice. Hum Mol Genet 2006; 15:1783–1792.
21. van der Kraan PM, Goumans MJ, Blaney Davidson E, et al. Age-dependent alteration of TGF-beta signalling in osteoarthritis
. Cell Tissue Res 2012; 347:257–265.
22. Yan D, Chen D, Cool SM, et al. Fibroblast growth factor receptor 1 is principally responsible for fibroblast growth factor 2-induced catabolic activities in human articular chondrocytes. Arthritis Res Ther 2011; 13:R130.
23. Gabay O, Sanchez C Epigenetics, sirtuins and osteoarthritis
. Joint Bone Spine 2012.
24. Dvir-Ginzberg M, Gagarina V, Lee EJ, et al. Tumor necrosis factor alpha-mediated cleavage and inactivation of SirT1 in human osteoarthritic chondrocytes. Arthritis Rheum 2011; 63:2363–2373.
25▪. Gabay O, Oppenhiemer H, Meir H, et al. Increased apoptotic chondrocytes in articular cartilage from adult heterozygous SirT1 mice. Ann Rheum Dis 2012; 71:613–616.
First study to show premature osteoarthritis develops in mice deficient in SirT1 that is an important regulator of energy homeostasis. SirT1 is a target of compounds designed to increase its activity in aging to slow development of age-related conditions.
26▪. Bui C, Barter MJ, Scott JL, et al. cAMP response element-binding (CREB) recruitment following a specific CpG demethylation leads to the elevated expression of the matrix metalloproteinase 13 in human articular chondrocytes and osteoarthritis
. FASEB J 2012; 26:3000–3011.
Adds to a growing body of evidence that altered methylation at promoter sites of matrix-degrading enzymes could be a key factor in osteoarthritis.
27. Rodova M, Lu Q, Li Y, et al. Nfat1 regulates adult articular chondrocyte function through its age-dependent expression mediated by epigenetic histone methylation. J Bone Miner Res 2011; 26:1974–1986.
28. Ukai T, Sato M, Akutsu H, et al.
MicroRNA-199a-3p, microRNA-193b, and microRNA-320c are correlated to aging and regulate human cartilage metabolism. J Orthop Res 2012.
29▪. Rose J, Soder S, Skhirtladze C, et al. DNA damage, discoordinated gene expression and cellular senescence in osteoarthritic chondrocytes. Osteoarthritis
Cartilage 2012; 20:1020–1028.
Presents solid evidence of a link between cell senescence and osteoarthritis and the potential role of DNA damage.
30. Harbo M, Bendix L, Bay-Jensen AC, et al. The distribution pattern of critically short telomeres in human osteoarthritic knees. Arthritis Res Ther 2012; 14:R12.
31. Breu A, Sprinzing B, Merkl K, et al. Estrogen reduces cellular aging in human mesenchymal stem cells and chondrocytes. J Orthop Res 2011; 29:1563–1571.
32. Merideth MA, Gordon LB, Clauss S, et al. Phenotype and course of Hutchinson-Gilford progeria syndrome. N Engl J Med 2008; 358:592–604.
33. Attur M, Ben-Artzi A, Yang Q, et al. Perturbation of nuclear lamin A causes cell death in chondrocytes. Arthritis Rheum 2012; 64:1940–1949.
34. Rodier F, Campisi J. Four faces of cellular senescence. J Cell Biol 2011; 192:547–556.
35. Polacek M, Bruun JA, Johansen O, et al. Differences in the secretome of cartilage explants and cultured chondrocytes unveiled by SILAC technology. J Orthop Res 2010; 28:1040–1049.
36. Clutterbuck AL, Smith JR, Allaway D, et al. High throughput proteomic analysis of the secretome in an explant model of articular cartilage inflammation. J Proteomics 2011; 74:704–715.
37. Dejica VM, Mort JS, Laverty S, et al. Increased type II collagen cleavage by cathepsin K and collagenase activities with aging and osteoarthritis
in human articular cartilage. Arthritis Res Ther 2012; 14:R113.