Neutral proteases in human osteoarthritic synovium

Chondroprotective drugs in degenerative joint diseases

Catabolic cytokine and anabolic growth factor pathways control destruction and repair in osteoarthritis (OA). A unidirectional TNF-alpha/IL-1-driven cytokine cascade disturbs the homeostasis of the extracellular matrix of articular cartilage in OA. Although chondrocytes in OA cartilage overexpress anabolic insulin-like growth factor (IGF) and its specific receptor (IGFRI) autocrine TNF-alpha released by apoptotic articular cartilage cells sets off an auto/paracrine IL-1-driven cascade that overrules the growth factor activities that sustain repair in degenerative joint disease. Chondroprotection with reappearance of a joint space that had disappeared has been documented unmistakably in peripheral joints of patients suffering from spondyloarthropathy when treated with TNF-alpha-blocking agents that repressed the unidirectional TNF-alpha/IL-1-driven cytokine cascade. A series of connective tissue structure-modifying agents (CTSMAs) that directly affect IL-1 synthesis and release in vitro and down-modulate downstream IL-1 features, e.g. collagenase, proteoglycanase and matrix metalloproteinase activities, the expression of inducible nitric oxide synthase, the increased release of nitric oxide, and the secretion of prostaglandin E(2), IL-6 and IL-8, have been shown to possess disease-modifying OA drug (DMOAD) activities in experimental models of OA and in human subjects with finger joint and knee OA. Examples are corticosteroids, some sulphated polysaccharides, chemically modified tetracyclines, diacetylrhein/rhein, glucosamine and avocado/soybean unsaponifiables.

Physiological levels of hydrocortisone maintain an optimal chondrocyte extracellular matrix metabolism

OBJECTIVE

To investigate the effects of physiological doses of hydrocortisone on synthesis and turnover of cell associated matrix (CAM) by human chondrocytes obtained from normal articular cartilage.

METHODS

Human articular cartilage cells were obtained from visually intact cartilage of the femoral condyles of five donors and maintained in culture for one week to reach equilibrium in accumulated CAM compounds. 0, 0.05, 0.20, and 1.0 micro g/ml hydrocortisone was added to the nutrient media during the entire culture period. Cells were liberated and levels of CAM aggrecan, type II collagen, and fibronectin, of intracellular IGF-1, IL1alpha and beta, and of their respective plasma membrane bound receptors IGFR1, IL1RI, and the decoy receptor IL1RII, were assayed by flow cytometry.

RESULTS

In comparison with controls, hydrocortisone treated chondrocytes, at all concentrations, expressed significantly higher plasma membrane bound IGFR1. Intracellular IGF-1 levels remained unchanged. Together with these changes, reflecting an increased ability to synthesise extracellular matrix (ECM) macromolecules, hydrocortisone treated cells expressed significantly higher amounts of the plasma membrane bound decoy IL1RII. Concurrently, intracellular IL1alpha and beta levels and membrane bound IL1RI were down regulated. Levels of CAM aggrecan, type II collagen, and fibronectin were significantly up regulated in the chondrocytes treated with hydrocortisone.

CONCLUSION

0.05 micro g/ml hydrocortisone treated chondrocytes had decreased catabolic signalling pathways and showed an enhanced ability to synthesise ECM macromolecules. Because IL1 activity was decreased and the expression of IL1RII decoy receptor enhanced, more of the ECM macromolecules produced remained accumulated in the CAM of the chondrocytes. The effects were obtained at doses comparable with physiological plasma levels of hydrocortisone in humans.

C-reactive protein as a biomarker of emergent osteoarthritis

OBJECTIVE

We evaluated C-reactive protein (C-RP), a quantitative marker of the acute phase response, as a potential biomarker of prevalent and incident osteoarthritis of the knee (OAK).

METHODS

Serum C-reactive protein concentrations were characterized with ultrasensitive rate nephelometry in a population-based sample of 1025 women (318 African-American and 707 Caucasian) who are enrollees in a study of musculoskeletal conditions at the mid-life. Assignment of OAK was based on Kellgren-Lawrence (K-L) scores of 2 or more on radiographs. Prevalent OAK was based on the baseline (1996) score while the classification of incident OAK was based on a score of 2 or greater at the follow-up examination 2.5 years later amongst those with a baseline K-L scores of 0 or 1.

RESULTS

At baseline, the prevalence of radiographic OAK was 12% in participants who were aged 27-53 years and 18% in the subgroup of women aged 40-53 years. The mean C-RP value was 2.31 mg/L, with values ranging from below detection (0.3 mg/L) to 47.4 mg/L. Higher C-RP concentrations were associated with both prevalent and incident OAK (P < 0.0001, and P < 0.0001, respectively). For each K-L score increase from 0 to 3, there was a significantly higher mean C-RP value. Compared to women without incident OAK, women who developed OAK in the 2.5-year follow-up had significantly higher baseline C-RP concentrations. Women with bilateral OAK had higher C-RP concentrations than women with unilateral OAK (6.65 mg/L +/- 0.56 vs 3.63 mg/L +/- 0.42, P < 0.007). BMI was highly correlated with C-RP (r = 0.58) and obesity was an effect modifier with respect to OAK and C-RP concentrations. When stratified according presence or absence of OAK and obesity (BMI > 30 kg/m2), mean C-RP values were: obesity and OAK, 6.3 +/- 0.4 mg/L; obesity but not OAK, 4.3 mg/L +/- 0.2; no obesity but OAK, 1.7 mg/L +/- 0.8; and neither obesity nor OAK, 1.3 mg/L +/- 0.2 mg/L. These stratum means were significantly different from each other, indicating a higher C-RP with OAK after accounting for obesity.

CONCLUSION

C-RP, as a measure of an acute phase response and inflammation, is highly associated with OAK; however, its high correlation with obesity limits its utility as an exclusive marker for OAK.

The Novartis-ILAR Rheumatology Prize 2001 Osteoarthritis: from molecule to man

During our careers, we have developed new and innovative concepts pertaining to the pathophysiology of osteoarthritis which have assisted in the development of new therapeutic approaches. Moreover, our laboratory has long sought to develop protective agents for osteoarthritic structural joint tissues. The most significant concepts that have originated from our lab are briefly outlined in this commentary.

Ro 32-3555, an orally active collagenase selective inhibitor, prevents structural damage in the STR/ORT mouse model of osteoarthritis

OBJECTIVE

To determine the efficacy of a selective inhibitor of collagenases in an animal model of osteoarthritis (OA).

METHODS

Ro 32-3555, an orally active collagenase selective inhibitor, was administered to STR/ORT mice. Microfocal x-ray-generated images of the hind limbs were visually scored for joint space narrowing, osteophyte formation, and calcification of tendons. Histologic sections of the knees were scored for cartilage changes including loss of surface matrix, fibrillation, and eburnation.

RESULTS

Significant inhibition of joint space narrowing and osteophyte formation was achieved in groups of animals treated with 10-50 mg/kg(-1) of Ro 32-3555. These effects were confirmed histologically in the same groups of animals: histologic analysis revealed that Ro 32-3555 protected cartilage from degradative changes.

CONCLUSION

Ro 32-3555, a collagenase selective inhibitor, inhibits both the cartilage and bone changes in this mouse model of OA, and thus shows great potential as a treatment of OA in humans.

Ultrastructural characteristics of the synovial membrane in osteoarthritic temporomandibular joints

PURPOSE

This study analyzed the ultrastructural characteristics of the synovial membrane in various stages of osteoarthritis (OA) of the temporomandibular joint (TMJ), and developed a classification of this involvement based on these morphologic characteristics.

PATIENTS AND METHODS

Synovial membrane biopsies were performed during unilateral arthroscopy in 40 patients. Thirty-one TMJs constituted the OA group; nine TMJs that were not involved by OA constituted the control group. During light microscopic (LM) examination, various variables were recorded and related to the duration of clinical signs and symptoms. Ten synovial membranes from osteoarthritic joints showing histologically visible pathologic changes in various stages and one control synovial membrane were selected for electron microscopic examination.

RESULTS

The initial, early, and intermediate stages of synovial membrane involvement in TMJ OA were characterized by intima hyperplasia. In the initial and early stages, active and hypertrophic intimal cells are found. In the intermediate stage, an increased number of both intracytoplasmic and extracellular filaments was predominant. Fibrosis of the subintimal tissue was initiated by an increased number of active fibroblasts. The late stage of synovial membrane involvement in TMJ OA was characterized by a relatively normal synovial intima of normal thickness, whereas extensive fibrosis was seen in the subintimal tissues.

CONCLUSIONS

Synovial membrane involvement in TMJ OA is characterized by an early proliferative phase with probable growth factor-mediated increases in the cellular activity of the synovial intima cells (resulting in hyperplasia and hypertrophy), of fibroblasts (resulting in increased production of collagen fibrils and fibrosis), and of endothelial cells (resulting in blood vessel growth and hypervascularity). The late phase is characterized by extensive fibrosis of the subintimal tissue, whether caused by sustained production of growth factors or by chronic venous insufficiency, with normal or little cellular activity.

Acute-phase proteins in osteoarthritis

The joint destruction of osteoarthritis (OA) comprises loss of articular cartilage resulting from an imbalance of enzyme-catalized cartilage breakdown and regeneration. OA is thought to derive from defective chondrocyte metabolism and thus to inherently lack the large-scale systemic response that is the hallmark of rheumatoid arthritis (RA). Because of the apparent absence of systemic inflammation in OA, acute-phase response proteins have not been as extensively studied in OA as they have been in RA. The diagnosis of OA almost always involves radiographic assessment of joint damage, which is useful only after the disease process has been underway for several months. Radiographic evaluation cannot give a good assessment of current disease activity and is a relatively insensitive indicator of prognosis. Cartilage breakdown products can potentially serve as direct surrogate markers of OA disease activity, but have not been extensively used because of their limited sensitivity and the technical difficulties associated with their measurement. Markers of disease activity in RA are indirect and are derived from the acute-phase response, a cycle of temporal changes in cellular and metabolic function. The early part of the acute-phase response involves the local action and production of cytokines such as interleukin-1 (IL-1), tumor necrosis factor (TNF-alpha) and IL-6. In the late acute-phase response, these cytokines can effect many systemic changes, including increased production of acute-phase proteins (APP). Three valuable surrogate markers of disease activity in RA are provided by the acute-phase response: the time-honored erythrocyte sedimentation rate (ESR) and the newer APPs C-reactive protein (CRP) and serum amyloid A (SAA). As in RA, the joint destruction of OA involves IL-1, TNF-alpha, and IL-6; however, OA can be viewed as an indolent stimulus of the later (systemic) acute-phase response. Recent studies of the acute-phase response in OA suggest that the concentrations of CRP and SAA are elevated in OA, but to a lesser extent than in RA. In the future, long-term monitoring of CRP concentrations in the blood may permit the earlier detection and more effective treatment of OA.

In vitro effects of NSAIDs and corticosteroids on the synthesis and secretion of interleukin 1 by human osteoarthritic synovial membranes

The mechanism(s) of action of NSAIDs and corticosteroids in arthritic diseases has been the subject of intensive investigation in recent years. Although NSAIDs and corticosteroids have many effects, their possible ability to modify the disease course in patients has not been fully documented. In an attempt to characterize the mechanism(s) involved in the effect of some NSAIDs in joint diseases, we investigated the effect of three concentrations within the pharmacological (260 micrograms/ml) and therapeutic (26 and 2.6 micrograms/ml) ranges of tiaprofenic acid in the synthesis and release of interleukin (IL-1) alpha and beta in human OA synovial membranes. The effect of tiaprofenic acid was compared to the effect of two other NSAIDs, sodium salicylate (160 micrograms/ml) and indomethacin (1.5 micrograms/ml), and to a corticosteroid, hydrocortisone (0.725 and 7.25 micrograms/ml). This study was carried out using human OA synovium explants incubated in the presence or absence of LPS. In the absence of LPS and at therapeutic concentration, tiaprofenic acid decreased both the synthesis and release of IL-1 beta. A less marked effect of the drug was noted under LPS treatment, and inhibition of the production/secretion of IL-1 beta was found only at pharmacological concentration. Sodium salicylate and indomethacin did not share this action, and demonstrated either no effect or enhancement of IL-1 beta synthesis, respectively, in the presence of LPS. As expected, hydrocortisone demonstrated a marked decrease on IL-1 alpha and IL-1 beta, both in the presence and absence of LPS. These results bring forth new information on the action of these drugs and their effects on the OA pathophysiological process.

Immunological analysis of proteoglycan structural changes in the early stage of experimental osteoarthritic canine cartilage lesions

Specific modifications of the proteoglycan (PG) structure of osteoarthritic (OA) dog cartilage in relation to endogenous metalloprotease activity were examined using murine anti-proteoglycan monoclonal antibodies (MoAbs). OA lesions were induced over a period of 8 weeks in crossbred dogs (Pond-Nuki model). The articular cartilage was removed and homogenized in a Tris buffer, pH 7.5, and then divided into four groups: direct PG extraction, no addition, presence of 1 mM p-aminophenyl mercuric acetate (APMA), and presence of 1 mM APMA and 10 mM o-phenanthroline, incubated for 42 h at 37 degrees C followed by PG extraction. MoAbs reactive with PG protein and carbohydrate epitopes included 1C6, 3B3, 5D4, D1B2, and m4D6. The results showed marked alterations induced by APMA activation of the endogenous metalloproteases. PG changes were apparent at at least three sites: one was either in the hyaluronic acid-binding region or between the hyaluronic-binding region and the G2 globular domain, another was between the keratan-sulfate-rich domain and the chondroitin sulfate-attachment domain, and a third was in the chondroitin sulfate-attachment domain. Constitutive metalloprotease activity resulted in less marked PG alterations with preservation of functional PG aggregation to hyaluronan.

Are cytokines involved in osteoarthritic pathophysiology?

The putative role and mechanism of action of cytokines in the progression of arthritic diseases such as osteoarthritis (OA) has received particular attention because of the important interaction between articular cartilage and synovium in the pathophysiology of the diseased state. Maintaining matrix homeostasis in the normal adult cartilage phenotype requires normal turnover of matrix components, principally collagen and proteoglycan. Chondrocytes and synovial fibroblasts are targeted, via specific cell-surface receptors, by cytokines like interleukin 1 (IL-1) and tumor necrosis factor alpha (TNF-alpha) to produce matrix proteases and to suppress the synthesis of collagen and proteoglycan. Thus, cytokines not only favor tissue destruction, but also inhibit tissue repair. A structurally heterogeneous group of factors capable of directly antagonizing cytokine action is described, which acts either by blocking cytokine-receptor binding, inhibiting local cytokine synthesis, or complexing the cytokine into an inactive form. Furthermore, many growth factors, such as transforming growth factor-beta (TGF-beta), can counteract the net effect of cytokines by stimulating the synthesis of matrix components or natural inhibitors of cartilage degrading enzymes.

Morphologic and morphometric changes in synovial membrane associated with mechanically induced osteoarthrosis

We sought to determine whether synovial leukocytic inflammation is a primary event in mechanically induced osteoarthrosis. Repetitive impulse loading (50 ms duration at 60 Hz for 40 minutes each day) was applied to the right hindlimbs of 24 New Zealand white rabbits for 3, 6, or 9 weeks. The synovial membrane from the medial suprapatellar area was examined qualitatively using transmission electron microscopy and quantitatively using light microscopic morphometry. The results indicate that synovial inflammation is not a primary event in this mechanically induced osteoarthrosis, but synovial hyperplasia occurs prior to histologically evident cartilage destruction at 6-9 weeks.

Imbalance between the mechanisms of activation and inhibition of metalloproteases in the early lesions of experimental osteoarthritis

Levels of tissue inhibitor of metalloproteases (TIMP) and plasminogen activator (PA)/plasmin were measured and the distribution of PA was studied by immunohistochemical techniques in cartilage and synovium samples from dogs subjected to sectioning of the anterior cruciate ligament of their right knees and sham operation of their left knees (controls). Twenty-three animals were divided into 3 groups and killed at 2, 4, or 8 weeks after surgery. The levels of PA and plasmin were found to be significantly elevated in the osteoarthritic (OA) knee cartilage and synovium at all times after surgery, except for levels of PA in the OA cartilage at 2 weeks. There was a positive correlation between the levels of PA and plasmin in the synovial membrane (r = 0.64, P less than 0.001). In OA knees, the presence of high levels of total and active collagenase was detected in cartilage and in synovium. The levels of these 2 forms of collagenase showed a positive correlation both in cartilage (r = 0.65, P less than 0.001) and in synovium (r = 0.77, P less than 0.001). The levels of TIMP in cartilage from OA and sham operated knees were similar. Although the TIMP level was increased in the OA synovium, it was found only in trace amounts in cartilage. Immunohistochemical studies revealed that both forms of PA, urokinase-type PA and tissue-type PA, and TIMP were present in OA tissues. In the synovium, they were found mainly in monocyte/macrophages, synovial lining cells, and blood vessel cells. In OA cartilage, PA was present only at the superficial level in chondrocytes and in cartilage matrix, whereas TIMP was present in chondrocyte lacunae throughout the full thickness of the cartilage. TIMP was also detected in the superficial level of cartilage from sham operated knees. The results of this study indicate that in OA tissues, there are conditions that favor the synthesis and activation of metalloproteases. PA and plasmin are likely to play an important role in the physiologic activation of metalloproteases, although they are probably not the only system involved in this process. The lack of increased TIMP levels in the OA cartilage, in the presence of increased metalloprotease activity, is also a possible contributing factor in the enzymatic degradation of this tissue.

Cathepsin B and cysteine protease inhibitors in human osteoarthritis

The aim of this study was to determine the involvement of cathepsin B and its inhibitors in the proteolytic degradation of human osteoarthritic (OA) tissue. The characteristics of the cathepsin B found in both normal and OA cartilage and synovium were similar to those of the lysosomal cathepsin B. Two inhibitors of cysteine proteases were found with a molecular weight of 67,000 and 16,000 Da. The cartilage cathepsin B level of OA specimens (54.8 +/- 7.3 units/micrograms of DNA) was greater than the controls (39.8 +/- 3.2 units/micrograms of DNA). Mild-moderate graded samples (78.1 +/- 12.0 units/micrograms of DNA) had significantly higher levels of enzyme activity than the severely graded ones (31.4 +/- 3.9 units/micrograms of DNA, p less than 0.001) and controls (p less than 0.01). Compared to controls (2.3 +/- 0.4 units/mg of tissue w.w.), cysteine protease inhibitory activity in OA cartilage was decreased in specimens with severe lesions (1.5 +/- 0.2 units/mg of tissue). This was particularly noted in patients who had not received steroid injections (1.2 +/- 0.3 units/mg of tissue, p less than 0.05). In OA synovia, the cathepsin B level was greater (40.7 +/- 7.4 units/mg of tissue w.w., p less than 0.02) than in the controls (13.6 +/- 3.7 units/mg of tissue). The cysteine protease inhibitory activity was similar in OA synovium (1.7 +/- 0.2 units/mg of tissue w.w.) and in controls (1.5 +/- 0.3 units/mg of tissue). This data demonstrated an imbalance between the levels of cathepsin B and cysteine protease inhibitors in OA tissue. A decrease of specific inhibitors could be an important contributing factor, particularly in more severe lesions.

Protective effects of corticosteroids on cartilage lesions and osteophyte formation in the Pond-Nuki dog model of osteoarthritis

The in vivo effects of corticosteroids on osteoarthritic (OA) lesions were examined in 12 dogs in which the anterior cruciate ligament had been sectioned. Six were treated with oral prednisone and 6 were treated with intraarticular (IA) injections of triamcinolone hexacetonide (TH), at surgery and 4 weeks later. Twelve other operated dogs received no treatment. All dogs were killed 8 weeks postsurgery. Four of 15 normal control dogs received IA TH injections. Operated untreated dogs developed significant cartilage lesions on the femoral condyles and tibial plateaus with prominent osteophytes. Operated dogs treated orally or with IA injections had a significant reduction in osteophyte size. Cartilage erosions on femoral condyles were observed in 25% of the untreated dogs, 8% of the dogs receiving oral prednisone, and none of the dogs receiving IA TH. In both groups of treated dogs, the size of the tibial plateau lesions was significantly reduced compared with the operated untreated dogs. Histologically, corticosteroids significantly reduced the severity of OA structural changes of the cartilage on both medial and lateral femoral condyles and tibial plateaus in operated animals, with the exception of the lateral plateaus of those treated orally. In electron microscopy studies, we found no evidence of increased cell degeneration or death associated with steroids. IA corticosteroids had no deleterious effects on normal articular cartilage. These results indicate that glucocorticoids administered orally or intraarticularly are effective against the development of OA lesions in this model.

The biochemical and histological effects of artificial ligament wear particles: in vitro and in vivo studies

Growing evidence suggests that biochemical mechanisms play a role in the pathogenesis of arthritis. Cartilaginous wear particles have been shown to induce destructive enzymes and cytokines. To assess the biocompatibility of artificial ACL replacements, the effects of wear particles from the following ligaments were analyzed biochemically and histologically: GORETEX, Stryker Dacron Ligament Prosthesis, Versigraft carbon, Kennedy LAD, Xenograft, Leeds-Keio, and human patellar tendon allograft. Ligaments were frozen and ground to produce wear particles similar to those seen clinically and were added to lapine synovial cell cultures. The resulting conditioned medium was analyzed for collagenase, gelatinase, and chondrocyte activating factor (CAF) production. All of the ligaments induced significantly elevated enzyme and CAF production by the synoviocytes, with Xenograft and carbon inducing significantly higher enzyme levels than those of the other five ligaments. Five milligrams of wear particles were injected into the knees of 4 kg to 5 kg rabbits that were analyzed histologically after 14 weeks. Wear particles accumulated in the periarticular synovial folds and induced modest to severe macrophage infiltration in the synovium. A hypothetical model explaining the role of artificial ligament wear particles in the pathogenesis of arthritis is presented.

Proteoglycan-degrading acid metalloprotease activity in human osteoarthritic cartilage, and the effect of intraarticular steroid injections

Cartilage samples from both the immediate and remote lesion areas were obtained from the tibial plateaus of 21 patients with osteoarthritis, and were subjected to histologic and enzymatic study. There was a frequent loss of pericellular metachromatic staining in the OA cartilage. Seven patients had received intraarticular injections of steroids, and in 21% of those cartilage samples, a pericellular halo was seen. This halo was seen in 71% of patients who had not received steroid injections. The total acid metalloprotease activity was increased more than twofold in specimens from OA lesions and in those samples graded moderate, as compared with age-matched control cartilages. These differences were greater when the specimens from patients who had received steroid therapy were excluded from the data. The cartilage specimens from steroid-treated patients were not significantly different from those of controls with respect to the enzyme activity in the lesions or in cartilage with moderate disease. The active form of the protease was suppressed by steroids. In samples from patients who did not receive steroid injections and who had a moderate grade of OA, a significantly elevated level of the active protease was present, as compared with control samples. Those samples graded moderate which came from patients who received steroid treatments showed no difference in the active protease level versus that of controls. Our results are consistent with the hypothesis that acid metalloprotease activity is involved in the degradation of the cartilage matrix in OA. Since the protease retains a significant fraction (40%) of its activity at neutral pH, its physiologic role might occur either at acid pH or at neutral pH.(ABSTRACT TRUNCATED AT 250 WORDS)