Increased degradation and alteered tissue distribution of cartilage oligomeric matrix protein in human rheumatoid and osteoarthritic cartilage

We investigated the degradation and tissue distribution of cartilage oligomeric matrix protein in normal, osteoarthritic, and rheumatoid arthritic articular cartilage of the human knee. Cartilage was subjected to sequential extractions with buffers containing neutral salt, with EDTA, and finally with guanidine/HCl and then was analyzed by Western blotting with a polyclonal antiserum to human cartilage oligomeric matrix protein. Western blots of the nine neutral salt extracts from normal cartilage revealed mostly intact pentameric molecules of cartilage oligomeric matrix protein, in contrast to the 13 osteoarthritic and five rheumatoid arthritic cartilage samples that demonstrated marked degradation of cartilage oligomeric matrix protein as noted by a predominance of reduction-sensitive bands at approximately 150 kDa and nonreduction-sensitive bands in the 67–94 kDa range. The EDTA and guanidine/HCl extracts from all groups were similar and showed mostly intact molecules of cartilage oligomeric matrix protein, with smaller amounts of degraded cartilage oligomeric matrix protein identical to those resolved by the Western blots of the neutral salt extracts. Western blots of matched pairs of synovial fluid and cartilage extracts demonstrated cartilage oligomeric matrix protein fragments of the same molecular mass. Competitive enzyme-linked immunosorbent assay revealed significantly less cartilage oligomeric matrix protein in rheumatoid articular cartilage than in either normal or osteoarthritic cartilage. In contrast to normal cartilage, where cartilage oligomeric matrix protein was predominately localized to the interterritorial matrix throughout all zones of the matrix, with increased staining in the deeper cartilaginous zones, the most intense staining in osteoarthritic cartilage was in the superficial zones of fibrillated cartilage, with little to no immunostaining in the midzones and relatively poor staining in the deeper cartilaginous zones. This distribution was the inverse of that for proteoglycans, as demonstrated by toluidine blue staining, where proteoglycans were depleted primarily from the superficial fibrillated cartilage. In mild to moderately affected rheumatoid cartilage, the tissue distribution of cartilage oligomeric matrix protein was similar to the distribution of proteoglycans, with relatively uniform staining of the interterritorial and territorial matrices. In more severely affected rheumatoid cartilage, the superficial zones demonstrated punctate immunostaining for cartilage oligomeric matrix protein in the interterritorial and territorial matrices, and staining was restricted to the territorial matrix in the deep cartilaginous zones. It is evident from this study that (a) noncollagenous proteins such as cartilage oligomeric matrix protein are greatly affected in arthritis. (b) degradation fragments released from the matrix into the synovial fluid reflect the processes occurring within the matrix, and (c) different zones of the articular cartilage are susceptible to degradation of cartilage oligomeric matrix protein in the different disease processes.     

Localization and expression of cartilage oligomeric matrix protein by human rheumatoid and osteoarthritic synovium and cartilage.

Synovium and cartilage from patients with osteoarthritis or rheumatoid arthritis were analyzed for expression of cartilage oligomeric matrix protein. Immunostaining of synovium with antiserum to cartilage oligomeric matrix protein demonstrated positive staining in both diseases. In osteoarthritis, there was positive staining within the synovial cells and immediately subjacent connective tissue, with less intense staining in the deeper connective tissue. In rheumatoid arthritis, there was less intense staining within the synovial cells and marked intense staining in the deeper connective tissue. In situ hybridization performed with an antisense digoxigenin-labeled riboprobe to human cartilage oligomeric matrix protein confirmed the presence of cartilage oligomeric matrix protein mRNA in the cells of the synovial lining in both types of synovium. Quantitative polymerase chain reaction with a cartilage oligomeric matrix protein MIMIC demonstrated increased cartilage oligomeric matrix protein mRNA in rheumatoid cartilage and synovium as compared with osteoarthritic cartilage and synovium, respectively; mRNA levels in rheumatoid synovium were similar to those from osteoarthritic chondrocytes. As a result of the high expression of cartilage oligomeric matrix protein from rheumatoid synovium, inflammatory synovium should be considered as a potential tissue source of cartilage oligomeric matrix protein in any investigation of biological markers of cartilage metabolism. The upregulated expression of cartilage oligomeric matrix protein in inflammatory tissues suggests its in vivo regulation by cytokines.     

Studies on YKL-40 in knee joints of patients with rheumatoid arthritis and osteoarthritis. Involvement of YKL-40 in the joint pathology

OBJECTIVE: The presence of YKL-40 (human cartilage glycoprotein 39) in synovium, cartilage and synovial fluid (SF) from knee joints of patients with rheumatoid arthritis and osteoarthritis (OA) were related to histopathological changes in synovium and cartilage and to serum YKL-40 and other biochemical markers. METHODS: The localization of YKL-40 in synovium and cartilage was determined by immunohistochemistry. Synovial inflammation was estimated histologically and by magnetic resonance imaging (MRI). Biochemical markers of inflammation, neutrophil activation and cartilage metabolism were analysed. YKL-40 concentrations in serum and SF were determined by RIA and ELISA. RESULTS: In the synovium YKL-40 positive cells were found in lining and stromal cells (macrophages) and the number of YKL-40 positive cells was related to the degree of synovitis. In arthritic cartilage, YKL-40 was located to chondrocytes. YKL-40 levels in SF were higher in RA patients with moderate/severe or none/slight synovitis of the knee joint compared to OA patients with moderate/severe or none/slight synovitis. SF YKL-40 correlated with the synovial membrane and the joint effusion volumes determined by magnetic resonance imaging (MRI) and with other biochemical markers of intercellular matrix metabolism. SF YKL-40 was higher than serum YKL-40, and a relationship existed between the YKL-40 levels in SF and serum. Intraarticular glucocorticoid injection was followed by clinical remission and a decrease in serum YKL-40, which increased again at clinical relapse. CONCLUSIONS: YKL-40 in SF is derived from cells in the inflamed synovium, chondrocytes and SF neutrophils. Joint derived YKL-40 influences serum YKL-40. YKL-40 may be involved in the pathophysiology of the arthritic processes and reflect local disease activity.     

The role of synovitis in pathophysiology and clinical symptoms of osteoarthritis

Osteoarthritis (OA), one of the most common rheumatic disorders, is characterized by cartilage breakdown and by synovial inflammation that is directly linked to clinical symptoms such as joint swelling, synovitis and inflammatory pain. The gold-standard method for detecting synovitis is histological analysis of samples obtained by biopsy, but the noninvasive imaging techniques MRI and ultrasonography might also perform well. The inflammation of the synovial membrane that occurs in both the early and late phases of OA is associated with alterations in the adjacent cartilage that are similar to those seen in rheumatoid arthritis. Catabolic and proinflammatory mediators such as cytokines, nitric oxide, prostaglandin E(2) and neuropeptides are produced by the inflamed synovium and alter the balance of cartilage matrix degradation and repair, leading to excess production of the proteolytic enzymes responsible for cartilage breakdown. Cartilage alteration in turn amplifies synovial inflammation, creating a vicious circle. As synovitis is associated with clinical symptoms and also reflects joint degradation in OA, synovium-targeted therapy could help alleviate the symptoms of the disease and perhaps also prevent structural progression.     

p38 MAPK and COX2 inhibition modulate human chondrocyte response to TGF-beta.

These studies compare actions of p38 MAPK inhibition and COX2 inhibition to modulate human arthritic chondrocyte responses to TGF-beta and FCS under basal and IL-1 activated conditions. Chondrocytes isolated from arthritic human femoral condyle cartilage obtained at total knee replacement were grown to 80% confluence. Proteoglycan synthesis and proliferation were measured with and without IL-1 activation in the presence and absence of growth factors and with and without inhibition of p38 MAPK or COX2 activity. Experiments to evaluate TIMP-1 production under these conditions were done using cartilage organ cultures. Neither p38 MAPK inhibitors nor COX2 inhibition affected basal proliferation. However both inhibitors enhanced the proliferative response to TGF-beta and FCS in IL-1 activated chondrocytes. TGF-beta stimulated proteoglycan synthesis was decreased by p38 MAPK inhibition, however COX2 inhibition restored the response to TGF-beta in IL-1 activated cells. In contrast, COX2 inhibition did not modulate TIMP-1 production while p38 MAPK inhibitors potentiated TGF-beta stimulated production of TIMP-1 in IL-1 activated cartilage. p38 MAPK inhibition and COX2 inhibition have unique and similar abilities to counteract some of the effects of IL-1 on human chondrocyte/cartilage metabolism. Both will partially restore the proliferative response to growth factors. p38 MAPK inhibition blunts TGF-beta stimulation of proteoglycan synthesis, but increases TIMP-1 synthesis. COX2 inhibition can restore the proteoglycan synthetic response to TGF-beta, but has no effect on cartilage TIMP-1 production. Use of these inhibitors to minimize cartilage damage in arthritic and mechanically stressed joints should reflect these characteristics.     

Do arachidonic acid and its metabolites, secreted by rheumatoid and osteoarthritic synovial tissue, account for the strong inhibition of DNA synthesis in cultured human articular chondrocytes? A novel approach to the mechanism of tissue damage

AIMS: To further characterize factors secreted in vitro by osteoarthritic and rheumatoid arthritis synovial membranes that inhibit DNA synthesis by cultured human articular chondrocytes, and extend these findings to synovial fluid. MATERIAL AND METHODS: Synovial tissue, synovial fluid and articular cartilage were obtained at surgery from two patients suffering rheumatoid arthritis and two other patients suffering from osteoarthritis. Synovial tissue was incubated in DMEM, then condition media and synovial fluids were extracted with methanol. Methanol extracts and extracted residues (hyaluronic acid, proteins) were assayed for their capacity to inhibit DNA synthesis in articular chondrocytes. Methanol extracts were also fractionated by thin layer chromatography on silica-coated plates and recovered fractions similarly tested. RESULTS: All extracts exhibited strong and concentration-dependent inhibition of [3H]-thymidine incorporation. The most potent inhibition was obtained with the extracts from rheumatoid joints and the least potent inhibition was with synovial fluids. The removal of active substances with methanol leaves an inactive residue. Methanol extraction does not alter the mitogenic activity of five exogenous growth factors and two cytokines, thus suggesting that such activity is entirely due to lipids. The bulk of anti-mitotic factors extracted by methanol co-migrate when fractionated by thin layer chromatography on silica-coated plates with arachidonic acid and its lipo-oxygenase metabolites. IN CONCLUSION: Inflamed synovium produces and releases lipids, most probably arachidonic acid metabolites that inhibit cell proliferation thus limiting inflammation and pannus formation in arthritis.     

Synovial factors and chondrocyte-mediated breakdown of cartilage: inhibition by hydrocortisone

Cartilage-synovium interactions were explored in a model culture system. Bovine nasal-cartilage discs were cocultured with minced rheumatoid synovium or synovium-conditioned media (SCM) in the presence or absence of hydrocortisone. Cartilage breakdown was assessed by the release of proteoglycan (PG) and hydroxyproline, and matrix biosynthesis by [35S]sulfate incorporation during pulse labeling. Chondrocyte-dependent breakdown in response to synovial factors (i.e., "catabolin" activity) was assessed by the difference in PG release between living and dead cartilages. Short-term contact with minced synovial membrane or exposure to its products released at a distance was sufficient to induce cartilage degradation in coculture; continued exposure was not required for breakdown to persist. Conditioned media from short-term synovial culture were similarly potent, and the induced breakdown was chondrocyte dependent. Matrix biosynthesis was inhibited in exposed cartilage but could be rapidly restored to normal on synovium removal despite the persistence of cartilage breakdown. Early hydrocortisone treatment suppressed the initiation of cartilage breakdown in cocultures and largely abolished the appearance of inductive factors in SCM. Later applications had little effect either in cocultures or in catabolin assays. We conclude that synovium-induced breakdown is an early event and that chondrocyte catabolic mechanisms once they have been activated are sufficient to maintain breakdown at a high level. Hydrocortisone, as well as limiting proteolysis, inhibits early tissue interactions at the level of synovial catabolin production or release.     

Up-regulation and differential expression of the hyaluronan-binding protein TSG-6 in cartilage and synovium in rheumatoid arthritis and osteoarthritis

OBJECTIVE: TSG-6 [the product of tumor necrosis factor (TNF)-stimulated gene-6] is a hyaluronan-binding protein that is present in the synovial fluids of arthritis patients and is secreted by cells of articular joints (e.g. chondrocytes and synoviocytes). This study examines the pattern of TSG-6 expression in normal and diseased cartilage and synovium using immunohistochemical techniques. DESIGN: A polyclonal antibody was raised against recombinant Link module from human TSG-6 and used to detect the protein in tissue sections taken from osteoarthritis (OA) and rheumatoid arthritis (RA) patients and controls. RESULTS: There was no TSG-6 detected in normal tissues. In all OA synovium there was intense TSG-6 expression throughout the intimal layer, whereas in RA staining in this region was generally less pronounced and was absent at the synovial surface in tissues exhibiting significant inflammation. In RA TSG-6 was also expressed by infiltrating leukocytes and by cells at the cartilage-synovium pannus junction. TSG-6 immunoreactivity was present in the tunica intima of blood vessels in OA subintima, being particularly noticeable in the thickened smooth muscle of inflamed vessel walls, but was mostly confined to the lumen of the vessel in RA. In cartilage the majority of chondrocytes expressed TSG-6 in both OA and RA, usually with extensive staining in the surrounding matrix. CONCLUSION: TSG-6 is present within synovium and cartilage of arthritic joints, but not normal controls, and is synthesized by the resident cells. The pattern of TSG-6 expression is consistent with its proposed roles in extracellular matrix (ECM) remodeling and cellular proliferation.     

Effects of pathological synovial fluids on the metabolism of chondrocytes

The purpose of this study was to clarify the mechanism of joint destruction in rheumatoid arthritis and osteoarthritis by observing the effects of the pathological synovial fluids on proteoglycan (PG) and collagen metabolism of chondrocytes. When chondrocytes from chick embryo were cultured with pathological synovial fluids, especially RA synovial fluid, biosynthesis of both proteoglycan and collagen of chondrocytes were found to increase in proportion to the amounts of pathological synovial fluids applied to the culture medium. Chondrocytes began to synthesize PG of small molecular size, approximately 70,000 in addition to PG of normal molecular size. It is noteworthy that the PG of small molecular size shows a shortening of the glycosaminoglycan link and core protein on undersulfation. These findings indicate that pathological synovial fluids disturb the biochemical regulation of the articular cartilage matrix by altering both PG and collagen metabolisms of chondrocytes.     

The synthetic processes of articular cartilage

The cells of cartilage are constantly remodeling the matrix in which they are suspended. The stimulus to initiate remodeling is probably the chondrocyte's response to physical and or chemical changes in the environment. Heat, pressure, friction, load, pH, and growth are examples of such factors, which, if altered, would have a dramatic effect on the cell's state of health. The mode of response by the chondrocyte is specific for a given stimulus. Elevated temperature, for example, switches on a set of genes, the heat shock genes, in chondrocytes. This results in the synthesis of a series of cellular protein that presumably in turn protects the cell from the injurious effects of heat. Load and pressure affect both the synthetic rate of matrix protein and the degradation rates of preexisting matrix. A number of low-molecular-weight proteins appear to be involved in anabolic and catabolic processes of cartilage. A protein recently isolated from synovium stimulates the synthesis of degradative enzymes in cartilage. This factor is probably involved in the remodeling process under normal physiologic conditions. More recently, it has been found in elevated levels under pathologic conditions such as in the synovial fluid of patients with rheumatoid and osteoarthritis. The mechanism by which this factor turns on the degradative pathway appears complex and is under investigation.     

L'acide arachidonique et ses métabolites, sécrétés par le tissu synovial rhumatoïde et arthrosique, interviennent-ils dans l'inhibition de la synthèse d'ADN par les chondrocytes articulaires humains en culture ? Une nouvelle approche d'analyse des mécanismes des lésions tissulaires

Aims. – To further characterize factors secreted in vitro by osteoarthritic and rheumatoid arthritis synovial membranes that inhibit DNA synthesis by cultured human articular chondrocytes, and extend these findings to synovial fluid.Material and methods. – Synovial tissue, synovial fluid and articular cartilage were obtained at surgery from two patients suffering rheumatoid arthritis and two other patients suffering osteoarthritis. Synovial tissue was incubated in DMEM, then condition media and synovial fluids were extracted with methanol. Methanol extracts and extracted residues (hyaluronic acid, proteins) were assayed for their capacity to inhibit DNA synthesis in articular chondrocytes. Methanol extracts were also fractionated by thin layer chromatography on silica-coated plates and recovered fractions similarly tested.Results. – All extracts exhibited strong and concentration-dependent inhibition of [³H]-thymidine incorporation. The most potent inhibition was obtained with the extracts from rheumatoid joints and the least potent inhibition was with synovial fluids. The removal of active substances with methanol leaves an inactive residue. Methanol extraction does not alter the mitogenic activity of five exogenous growth factors and two cytokines, thus suggesting that such activity is entirely due to lipids. The bulk of antimitotic factors extracted by methanol comigrate when fractionated by thin layer chromatography on silica-coated plates with arachidonic acid and its lipooxygenase metabolites.In conclusion. – Inflamed synovium produces and releases lipids, most probably arachidonic acid metabolites that inhibit cell proliferation thus limiting inflammation and pannus formation in arthritis.     

Indomethacin-induced augmentation of insulin release

The effect of indomethacin, a potent prostaglandin synthetase inhibitor upon portal venous glucose, glucagon, and insulin, was examined in eight glucose-loaded, nondiabetic adult male mongrel dogs, by examining the response of each animal to a standard enteral glucose challenge with and without indomethacin pretreatment. Mean portal venous insulin rose after glucose loading with the rise after indomethacin pretreatment being significantly greater (P < 0.001) and more prolonged than that seen in the control group. Mean basal glucagon was increased with indomethacin pretreatment relative to controls (P < 0.01) and remained so throughout the measurement period. The glucose tolerance curves generated in control and indomethacin pretreatment groups were of normal contour and magnitude, and indistinguishable from one another. The effects seen are presumably the consequence of prostaglandin synthetase inhibition and imply that prostaglandins may play a suppressor role in modulating the insulin response to a glucose challenge.     

Inhibition of adenosine kinase attenuates interleukin-1- and lipopolysaccharide-induced alterations in articular cartilage metabolism

OBJECTIVE: To investigate the effect of adenosine kinase inhibition on interleukin (IL)-1beta- and lipopolysaccharide (LPS)-induced cartilage damage. DESIGN: Articular cartilage was obtained from the metacarpophalangeal joints of 10 young adult horses. Following a stabilization period, weighed cartilage explants were exposed to IL-1beta (10 ng/ml) or LPS (50 microg/ml) to induce cartilage degradation. To test the potential protective effects of adenosine, these explants were simultaneously exposed to adenosine (100 microM), the adenosine kinase inhibitor 5'iodotubercidin (ITU, 1 microM) or to both adenosine and ITU. After 72 h in culture, conditioned medium was collected for evaluation of glycosaminoglycan (GAG), nitric oxide (NO), prostaglandin E2 (PGE2) and matrix metalloproteinase (MMP)-3 release. RESULTS: IL-1beta and LPS stimulated significant release of GAG, NO, PGE2 and MMP-3. Incubation with ITU significantly inhibited both IL-1beta- and LPS-induced GAG release, but did not alter MMP-3 production. Exposure to ITU also reduced IL-1beta-induced PGE2 release and LPS-induced NO production. Direct adenosine supplementation did not attenuate the effects of IL-1beta or LPS, and the addition of adenosine or ITU in the absence of IL-1beta or LPS did not have any detectable effect on cartilage metabolism in this model. CONCLUSIONS: The adenosine kinase inhibitor ITU attenuated experimentally induced cartilage damage in an in vitro cartilage explant model. Release of adenosine from chondrocytes may play a role in the cellular response to tissue damage in arthritic conditions and modulation of these pathways in the joint may have potential for treatment of arthropathies.     

Sulindac and indomethacin suppress the diuretic action of furosemide in patients with cirrhosis and ascites: evidence that sulindac affects renal prostaglandins

Nonsteroidal anti-inflammatory drugs (NSAID) suppress prostaglandin-dependent renal blood flow and furosemide-induced diuresis in patients with cirrhosis and ascites. Since sulindac may selectively spare inhibition of renal prostaglandins, we evaluated the interactions of acute administration of sulindac or indomethacin with furosemide in 15 patients with cirrhosis and ascites. Prior to furosemide, indomethacin reduced creatinine clearance (by 55%), urinary volume (by 82%), sodium (by 93%), and prostaglandin E2 (by 87%) (all P less than 0.05), whereas sulindac had no effect. However, both drugs reduced furosemide-induced diuresis. Indomethacin appeared slightly more potent in reducing the diuresis (55% v 38%), natriuresis (67% v 52%), and prostaglandin E2 (PGE2) release (81% v 74%). In a similar protocol in healthy subjects, furosemide-induced diuresis and natriuresis were also blunted by both drugs. Thus, under conditions of enhanced prostaglandin activity from furosemide, sulindac does affect renal function. These data suggest that renal function should be monitored in patients with cirrhosis and ascites who receive sulindac as well as other NSAID.     

Cartilaginous transdifferentiation of rat tenosynovial cells under the influence of bone morphogenetic protein in tissue culture

In an in vitro system cartilage tissue was formed from the patellar ligament of young adult rats in the crevices of bone matrix gelatin (BMG) under the influence of bone morphogenetic protein (BMP). The rate of cartilage induction was 75% (six of eight) for the explants using patellar ligaments of rat knee joints. Almost the same rate was observed, 67% (four of six), for explants of ligament with fatty tissue. Tenosynovial cells and synovial cells from the surface of ligaments are potent responders to BMP, and this property may be applied to accelerate the biologic attachment of excised tendons and ligaments in clinical practice.     

Production of cartilage oligomeric matrix protein (COMP) by cultured human dermal and synovial fibroblasts

OBJECTIVE: Cartilage oligomeric matrix protein (COMP) is a large disulfide-linked pentameric protein. Each of its five subunits is approximately 100,000 Da in molecular weight. COMP was originally identified and characterized in cartilage and it has been considered a marker of cartilage metabolism because it is currently thought not to be present in other joint tissues, except for tendon. To confirm the tissue specificity of COMP expression we examined cultured human dermal fibroblasts, human foreskin fibroblasts, and normal human synovial cells for the synthesis of COMP in culture. METHOD: Normal synovial cells and normal human dermal foreskin fibroblasts were isolated from the corresponding tissues by sequential enzymatic digestions and cultured in media containing 10% fetal bovine serum until confluent. During the final 24 h of culture, the cells were labeled with 35S-methionine and 35S-cysteine in serum- and cysteine/methionine-free medium. The newly synthesized COMP molecules were immunoprecipitated from the culture media with a COMP-specific polyclonal antiserum, or with monoclonal antibodies or affinity-purified COMP antibodies. The immunoprecipitated COMP was analyzed by electrophoresis in 5.5% polyacrylamide gels. For other experiments, synovial cells cultured from the synovium of patients with rheumatoid arthritis (RA) and osteoarthritis (OA) were similarly examined. RESULTS: A comparison of the amounts of COMP produced by each cell type (corrected for the DNA content) revealed that synovial cells produced > or = 9 times more COMP than chondrocytes or dermal fibroblasts. COMP could be easily detected by immunoprecipitation in all cell types. Electrophoretic analysis revealed a distinct band with an apparent MW of 115-120 kDa in samples from each of the three cell types, regardless of the antibody used. COMP expression in cultures of synoviocytes derived from OA and RA patients showed that OA and RA synovial cells produced similar amounts of monomeric COMP of identical size to those COMP monomers produced by normal synovial cells. The addition of TGF-beta to these cultures resulted in an increase in COMP production in normal, OA and RA synovial cells (45, 116 and 115% respectively). CONCLUSION: These studies demonstrate that substantial amounts of COMP are produced by several mesenchymal cells including synoviocytes and dermal fibroblasts. These findings raise important concerns regarding the utility of measurements of COMP levels in serum or in synovial fluid as markers of articular cartilage degradation because of the likelihood that a substantial proportion of COMP or COMP fragments present in serum or synovial fluid may be produced by cells other than articular chondrocytes.     

The cleavage of biglycan by aggrecanases

OBJECTIVE: Aggrecanase-1 [a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS)-4] and aggrecanase-2 (ADAMTS-5) have been named for their ability to degrade the proteoglycan aggrecan. While this may be the preferred substrate for these enzymes, they are also able to degrade other proteins. The aim of this work was to determine whether the aggrecanases could degrade biglycan and decorin. METHODS: Biglycan, decorin and aggrecan were purified from human and bovine cartilage and subjected to degradation by recombinant aggrecanase-1 or aggrecanase-2. In vitro degradation was assessed by sodium dodecyl sulfate/polyacrylamide gel electrophoresis (SDS/PAGE) and immunoblotting, and the cleavage site in biglycan was determined by N-terminal amino acid sequencing. SDS/PAGE and immunoblotting were also used to assess in situ degradation in both normal and arthritic human articular cartilage. RESULTS: Both aggrecanase-1 and aggrecanase-2 are able to cleave bovine and human biglycan at a site within their central leucine-rich repeat regions. Cleavage occurs at an asparagine-cysteine bond within the fifth leucine-rich repeat. In contrast, the closely related proteoglycan decorin is not a substrate for the aggrecanases. Analysis of human articular cartilage from osteoarthritic (OA) and rheumatoid arthritic (RA) joints showed that a biglycan degradation product of equivalent size is present in the extracellular matrix. No equivalent degradation product was, however, detectable in normal adult human articular cartilage. CONCLUSION: Biglycan, which is structurally unrelated to aggrecan, can act as a substrate for aggrecanase-1 and aggrecanase-2, and these proteinases may account for at least part of the biglycan degradation that is present in arthritic cartilage.     

Effects of isolated rheumatoid synovial cells on cartilage degradation in vitro

Rheumatoid synovium in coculture with cartilage has been shown to release a factor(s) that stimulates the depletion of glycosaminoglycans (GAG) from cartilage matrix. Human rheumatoid synovium was enzymatically disaggregated and the isolated cells were subjected to a variety of mechanical and immunological treatments. Synovial cell conditioned media (SCCM) were prepared and analyzed for their ability to stimulate GAG depletion. SCCM prepared from increasing concentrations of isolated synovial cells demonstrated cartilage degradative activity in a dose-dependent manner. This activity was characterized as interleukin-1 like and was found mostly within the adherent cell population where the synovial macrophages retained significant degradative ability. T cells alone were found to have no direct degradative effect on cartilage, but their presence appeared to augment the response of the adherent cells. The techniques described here provide a quantitative model for examining the degradative factors from synovium as well as the cellular interactions that promote their release.     

Studies on degradation of cartilage proteoglycan by rheumatoid synovial tissue. Part I: On the degraded products of proteoglycan (author's transl)

The degradation of MgCl2-extracted proteoglycans (PG) from bovine nasal and articular cartilage was performed by using human rheumatoid synovial tissue extract, and the products were compared with that of being degraded by different commercially available proteases. 2. Degradation of PG by crude synovial extract occurred under the wide range of pH, and the degradation products showed the changes as follows; 1) a decrease of viscosity, 2) a decrease of hydrodynamic size, 3) an increase of electrophoretic mobility, 4) no change of the length of glycosaminoglycan side chains, 5) an increase of chondroitin sulfate relative to keratan sulfate in precipitate with 1% cetylpyridinium chloride (CPC), and 6) an increase of serine content in precipitate with 1% CPC. 3. The average hydrodynamic size of degradation products by crude synovial extract was larger than that of degradation products by pronase or papain. 4. The results indicate that the proteases play the main role on the PG-degrading activity of rheumatoid synovial tissue extract but the glycosidase play no significant role.