Neutral proteinases from articular chondrocytes in culture. 2. Metal-dependent latent neutral proteoglycanase, and inhibitory activity

Inhibition of interleukin 1 beta induced rat and human cartilage degradation in vitro by the metalloproteinase inhibitor U27391

Interleukin 1 induced proteoglycan loss from cartilage in vitro was prevented by a biochemical inhibitor of metalloproteinase activity. The inhibitor also partially relieved the inhibition of proteoglycan synthesis caused by interleukin 1. The loss of glycosaminoglycan by rat and human femoral head cartilage in response to human recombinant interleukin 1 beta (rhIL-1 beta) was established, and the modulation of this loss by the metalloproteinase inhibitor U27391 was investigated. Rat femoral head cartilage consistently lost glycosaminoglycan in response to rhIL-1 beta whereas only a proportion (30%) of normal human femoral head cartilage did so. Concentrations of 10-100 mumol/l U27391 inhibited the action of rhIL-1 beta on rat femoral head cartilage, reversing both the loss of glycosaminoglycan and the inhibition of glycosaminoglycan synthesis. U27391 also prevented the reduction in glycosaminoglycan content of those human femoral head cartilage explants responsive to rhIL-1 beta. Metalloproteinase inhibition therefore prevents rhIL-1 beta induced glycosaminoglycan loss by rat and human femoral head cartilage, suggesting that inhibitors of such enzymes may prove to be of therapeutic benefit in erosive diseases in humans.

Damage of cultured chondrocytes by hydrogen peroxide derived from polymorphonuclear leukocytes: a possible mechanism of cartilage degradation

To study the mechanisms of chondrocyte damage, chondrocyte cytotoxicity as shown by chromium-51 release induced by polymorphonuclear leukocytes (PMNLs) was examined. PMNLs significantly enhanced chondrocyte cytotoxicity in the presence of phorbol dibutyrate. This chondrocyte damage was abolished by the addition of catalase, whereas superoxide dismutase and scavengers of hydroxyl radicals and protease inhibitors failed to reverse it. When cartilage matrix components such as hyaluronic acid and various proteoglycans were added to the PMNL-chondrocyte cultures, these components failed to affect the chromium-51 release. These results suggest that the increase in chondrocyte cytotoxicity is due to hydrogen peroxide generated by the PMNLs, and that cartilage matrix components do not prevent it. Hydrogen peroxide from PMNLs may therefore play an important role in cartilage degradation through direct damage of chondrocytes during inflammatory process.

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.

Low molecular weight serine proteinase inhibitors of human articular cartilage. Isolation, characterization, and biosynthesis

The major low molecular weight serine proteinase inhibitor of human articular cartilage was purified to homogeneity as determined by single-peak elution with 4 high resolution techniques. The purified protein was found to be a potent inhibitor of human leukocyte elastase and cathepsin G, as well as the native serine proteinases derived from human articular cartilage and intervertebral disc. The inhibitor and lysozymes were synthesized by human articular cartilage in vitro. These properties and the ability of this cationic inhibitor to bind to cartilage matrix components suggest a possible role in the modulation of matrix catabolism in normal and pathologic states.

Septic arthritis. Staphylococcal induction of chondrocyte proteolytic activity

We report herein that cartilage proteolytic activity increased in bovine and rabbit articular cartilage after treatment with a purified staphylococcal culture medium or intraarticular infection with Staphylococcus aureus. Staphylococcal culture medium increased the release of gelatinolytic, collagenolytic, and caseinolytic activity into the medium of isolated chondrocytes or cartilage organ culture. The proteolytic activities were determined in assays using radiolabeled substrate and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Staphylococcal culture medium was proteolytically inactive by both assay techniques. RNA synthesis of isolated chondrocytes was unaffected by staphylococcal culture medium, whereas overall protein synthesis was inhibited by 84%. An analysis of extracts of Staphylococcus aureus-infected rabbit knee cartilage by substrate gels showed increased gelatinolytic and caseinolytic activity compared with extracts of uninfected knee cartilage. Our data suggest that the rapid loss of proteoglycan and persistent degradation of cartilage in staphylococcal septic arthritis is due to the production and activation of chondrocyte proteases.

In vitro and in vivo effects of metal chelators on cartilage metabolism

The effect of metallic chelating agents (EDTA, EGTA) on cartilage metabolism was studied both in vitro, on calf cartilage, and in vivo, in rabbits. The question asked was whether it was possible to affect neutral protease activity, and not also inhibit beneficial synthetic systems. In vitro, EDTA suppressed anabolic processes, while EGTA had no effect. However, EDTA in vivo did not suppress glycosaminoglycan or RNA production, but paradoxically stimulated them. At the same time, EDTA inhibited neutral protease activity in normal animals.

Drug-induced inhibition of proteoglycanase activity in the Hulth-Telhag model

Evidence has accumulated that there are increased degradative enzymes in osteoarthritis responsible for joint destruction. These enzymes are metal-dependent, and inhibited by EDTA. EDTA was administered intra-articularly to rabbits in an experimental model of osteoarthritis. There was a 25% reduction in neutral proteoglycanase activity and 75% of the animals had a reduction in the severity of the arthritis as measured on a histologic-histochemical grading system. It is suggested that future chemotherapeutic studies on arthritis might focus on enzyme inhibition.

Activation of neutral metalloprotease in human osteoarthritic knee cartilage: evidence for degradation in the core protein of sulphated proteoglycan

The neutral, metal dependent, proteoglycan degrading enzymes (NMPEs) in human osteoarthritic knee cartilage homogenates were activated by p-aminophenylmercuric acetate (APMA). The resultant effect on the structure of newly synthesised and already existing sulphated proteoglycan was measured. Newly synthesised and already existing proteoglycan aggregated to hyaluronic acid was reduced (p less than 0.01, p less than 0.05 respectively) when measured by chromatography on Sepharose CL-2B eluted with associative buffer. The APMA activated enzyme affected both the newly synthesised and already existing proteoglycan aggregate similarly (r = 0.79, p less than 0.001). Treatment of cartilage homogenates with APMA and 1,10-phenanthroline (10 mM) showed that the amount of aggregated proteoglycan was at the control level. The hydrodynamic size of the proteoglycan monomer (A1D1) was also reduced by treatment of cartilage homogenates with APMA. Reaggregation experiments with fraction A1D1 and exogenous hyaluronic acid and link protein showed a similar defect in forming proteoglycan aggregates. These data showed that activation of the NMPEs altered the structure of proteoglycan in two ways. The most consistent change was a reduction in the ability of proteoglycan to form aggregates with hyaluronic acid. This was likely to have occurred via a cleavage of the core protein in or around the hyaluronic acid binding globular domain. A second alteration probably includes a limited proteolytic cleavage in the remainder of the core protein.

Characterisation of chondrocyte activation in response to cytokines synthesised by a synovial cell line

The lapine, synovial cell line, HIG-82, secretes 'chondrocyte activating factors' (CAF) which induce the synthesis of collagenase (EC 3.4.24.7), gelatinase, caseinase and prostaglandin E2 (PGE2) by confluent, monolayer cultures of lapine, articular chondrocytes. Partially purified CAF increased the production of PGE2 by chondrocytes within 3 h; in certain cultures this occurred in as little as 1 h. Increased levels of the three neutral metalloproteinases, in contrast, were only measurable in the conditioned medium after a delay of 9-18 h. After removal of the CAF, the synthesis of PGE2 reverted to basal levels within 1-4 h, but synthesis of the three proteinases remained high for an additional 4 days. Indomethacin, at concentrations which completely inhibited PGE2 synthesis, had no effect upon the coordinate induction of collagenase, gelatinase and caseinase. However, cycloheximide, alpha-amanitin and 5,6-dichlororibosylbenzimidazole (DRB) suppressed induction of these proteinases suggesting that CAF derepressed the genes coding for these enzymes. Once the chondrocytes had been activated by CAF, the inhibitors of transcription had a much weaker effect on the production of the neutral proteinases, indicating that their mRNAs may be relatively stable. In the presence of CAF, inhibition under these conditions was weaker still, possibly due to stabilisation of these mRNA molecules. Experiments with a number of compounds which modulate cellular Ca2+, cAMP or cGMP failed to support a straightforward role for these mediators in the induction of neutral metalloproteinases in chondrocytes. High concentrations of phorbol myristate acetate (PMA) provoked only a slight synthesis of these enzymes.

Effect of cytokines and growth factors on the expression of elastase activity by human synoviocytes, dermal fibroblasts and rabbit articular chondrocytes

Human synoviocytes, rabbit articular chondrocytes and human skin fibroblasts in culture were examined for their ability to express elastase activity. Latent enzyme activity degrading insoluble elastin was detected in the culture media of the three cell types and was completely abolished by metal chelating agents. Triton X-100 cell extracts were found to degrade a synthetic elastase substrate, N Succinyl-(Ala)3p-nitroanilide (SANA). The SANA-degrading activity of cell extracts could be attributed to a metalloprotease for fibroblasts and synoviocytes (100%) and to a metalloprotease associated with a cysteine protease for chondrocytes (70 and 30% respectively). This SANA-degrading activity was partly due to the combined action of an endo and an exopeptidase. Tumor Necrosis Factor-alpha (TNF-alpha) and Interferon-gamma (IFN-gamma) significantly enhanced the elastin degrading activity present in the culture media of both synoviocytes and chondrocytes. Interleukin-1 beta significantly increased the secretion of elastase by chondrocytes. By contrast, Transforming Growth Factor-beta (TGF-beta) reduced by 80 per cent the secretion of elastinolytic activity by chondrocytes but had not effect on other cell types.

Cartilage degradation by neutral metalloproteases in experimental collagen-like syndrome

In this study, we sought to determine the role of neutral proteases in cartilage matrix proteoglycan degradation, which occurs during the experimental hydralazine-induced collagen-like syndrome (c-l-s) in rats. The digestion of endogenous proteoglycans by neutral proteases in homogenates of cartilage from rats with c-l-s has been measured and compared with that of normal age-matched controls. Cartilage specimens from the tibial plateau were analysed for DNA and proteoglycan content, and neutral proteoglycan-degrading activity, No significant difference in cartilage DNA concentration was observed among the rats with c-l-s and the controls. Total neutral proteoglycan-degrading metallo-enzyme activity, determined by direct tissue assay, was significantly higher in c-l-s cartilage than that in control cartilage. Serine protease activity on proteoglycans was much lower than that of metalloprotease. The results of this study are consistent with the hypothesis that the neutral metalloproteases of cartilage are involved in the degradation of proteoglycans in c-l-s.

Metalloproteinases in endochondral bone formation: appearance of tissue inhibitor-resistant metalloproteinases

Dissected embryonic chick limbs release neutral metalloproteinases during endochondral bone development. These enzymes degrade cartilage proteoglycan and gelatin in culture medium. We found the enzymes active in the medium conditioned by explants of the region adjacent to the bone marrow cavity (cavity-surround). These enzymes degrade proteoglycan (PG) and/or gelatin. These spontaneously active enzymes are resistant to serum and tissue proteinase inhibitors, alpha 2-macroglobulin, and cartilage metalloproteinase inhibitor (TIMP). The other enzymes secreted from tarsus and bone marrow explants are mostly latent in the culture medium. Activated tarsus enzymes (PG degrading and gelatinolytic) are blocked by the above inhibitors. Activated marrow enzyme does not degrade PG but is resistant to those inhibitors. Cavity-surround enzymes may play an important role in embryonic osteogenesis of long bones because of their resistance to tissue and serum inhibitors. The in vivo mechanisms by which cavity-surround enzymes are activated are yet to be determined.

Interleukin-1 and osteoarthritis

Il-1, a multifunctional monokine, can stimulate both synoviocytes and articular chondrocytes to release neutral proteases and prostaglandin E2. It is also capable of promoting bone resorption. Therefore, this molecule (or family of molecules) is likely to play an important role in the mechanism of articular cartilage destruction that occurs in degenerative arthropathies. The synovial tissue itself can produce Il-1 (Catabolin) in some conditions, such as a slight traumatism, so that the presence of local inflammation is not necessary for "Il-1-cartilage" interaction to occur. Fundamental macromolecules of cartilage (collagens, proteoglycans) exert a stimulatory effect on Il-1 production, either as such or in the form of immune complexes. Some activated complement fractions (C3a and C5a) may also be actively involved. Studies on the mechanisms which regulate Il-1 synthesis and release, as well as investigations on the response of target cells to Il-1, are presently fascinating goals that could lead to new strategies in therapeutic research.

The immunologic detection and characterization of cartilage proteoglycan degradation products in synovial fluids of patients with arthritis

Antibodies were used in radioimmunoassays with gel chromatography to detect the hyaluronic acid-binding region, core protein, and keratan sulfate of human cartilage proteoglycan in the synovial fluids of patients with rheumatoid arthritis, juvenile rheumatoid arthritis, and osteoarthritis. All fluids contained proteoglycan that was mainly included on Sepharose CL-4B; this result indicates cleavage of proteoglycan (which is normally excluded). The hyaluronic acid-binding region was the smallest and most commonly detected fragment. It was relatively free of keratan sulfate and core protein, and it could sometimes bind to hyaluronic acid. Other larger fragments containing core protein and/or keratan sulfate were detected in every fluid.

Modulation of cartilage destruction by select nonsteroidal antiinflammatory drugs. In vitro effect on the synthesis and activity of catabolism-inducing cytokines produced by osteoarthritic and rheumatoid synovial tissue

Non-enzymatic factors produced by synovial tissue can potentially mediate cartilage destruction by inducing the synthesis and release of matrix-degrading proteinases from chondrocytes. Pharmacologic control of this process is of potential clinical relevance. The in vitro effect of therapeutic concentrations of select nonsteroidal antiinflammatory drugs on the synthesis and activity of catabolism-inducing cytokines produced by 6-day explant cultures of osteoarthritic and rheumatoid synovial tissue was studied. Piroxicam regularly suppressed such factor synthesis by both types of tissue without significantly affecting total protein synthesis. This did not occur using sodium salicylate or indomethacin in osteoarthritis tissue cultures and was observed only occasionally in rheumatoid arthritis cultures. None of the nonsteroidal antiinflammatory drugs studied consistently blocked catabolism-inducing activity in osteoarthritis tissue, whereas piroxicam more consistently inhibited activity produced by rheumatoid arthritis tissue. Results suggest that the catabolism-inducing factors produced by the 2 tissue sources may differ.

Rabbit chondrocytes maintained in serum-free medium. I. Synthesis and secretion of hydrodynamically-small proteoglycans

The biosynthesis of sulfated proteoglycan in vitro by rabbit articular chondrocytes in first passage monolayer culture maintained in fetal bovine serum (FBS) or in serum-free conditions was compared. Neosynthesized proteoglycan in the culture medium in the most dense fraction of an associative CsCl density gradient (fraction dAl) declined with increasing time under serum-free conditions, but not when cells were maintained in the presence of serum. After one day, the major peak of incorporated 35SO4 in medium fraction dAl eluted as a retarded peak (Kav 0.28) on Sepharose CL-2B, whether cells were maintained under serum-free or serum-containing conditions. The hydrodynamic size of proteoglycan monomer fraction dAlDl obtained after one day of exposure to serum-free culture media was smaller than dAlDl from serum-containing cultures. The hydrodynamic size of dAlDl obtained from serum-free culture media became even progressively smaller after 2 and 3 days' exposure to these conditions. Hydrodynamically small sulfated proteoglycans were identified in the cell-associated dAlDl fraction as early as one day after switching chondrocytes from serum-containing to serum-free medium. Culture medium fraction dAlDl from serum-free culture medium aggregated poorly when incubated with human hyaluronic acid (HA) in the presence of bovine link protein or when dialysed against bovine nasal cartilage proteoglycan aggregate. Proteoglycan monomer from serum-containing medium reaggregated more efficiently under both conditions. No change in the size of glycosaminoglycan chains was seen in the smaller proteoglycan subpopulations, nor was there any indication of marked changes in the glycosaminoglycan types.

Pathogenesis of osteoarthritis

This article reviews the etiology and pathogenesis of osteoarthritis, particularly one of several current concepts concerning the possible central mechanisms regulating degradation of cartilage. According to this theory, degradation involves diffuse or focal exposure of the extracellular matrix to active neutral metalloproteinases, which then results in injury as well as initiation of repair processes. Diffuse matrix exposure is probably not a physiologic aberrancy but rather a pathologic result of either physical injury to local chondrocytes or inflammatory mediators.

An assay for proteoglycanase (PGase) and it's application to articular chondrocyte cultures

The role of PGase in the degradation of cartilage proteoglycans (PG) is still unclear. Although several assays are available, there still exists some problems in the sensitivity, specificity and simplicity of these procedures. We therefore decided to develop an improved assay for PGase in order to detect small quantities of enzyme activity in the culture medium of articular chondrocytes. As substrates of PGase, we prepared fluorescein isothiocyanate labelled core protein (FITC-PGC) and fluorescein amine labelled hyaluronic acid-proteoglycan subunit complex (FAHA-PGS) from new born calf costal cartilages. PGase activity was estimated from the change of elution profile of Sepharose CL-4B chromatography of fluorescein labelled substrates after incubation with the activated enzyme solution. For measurement of PGase in large numbers of samples, we developed the microultrafiltration method using ultrafiltration membrane (MPS-I kit, Amicon). The concentrated media of articular chondrocyte cultures was used for the investigation of the nature of PGase(s) derived from articular chondrocytes. It was found that APMA, CoCl2, and CaCl2 were needed for activation of the enzyme. Sephadex-G75 gel chromatography of the crude enzyme solution suggests that the molecular weight of PGase from rabbit articular chondrocytes was approximately 5 X 10(4) daltons. From the results of gel chromatography of the digested FITC-PGC and FAHA-PGS it was concluded that one of the sites of attack of the chondrocyte secreted PGase might be near the hyaluronic acid binding region of core protein. The effects of various agents on the production of PGase were also studied. Interleukin-I was found to stimulate PGase production in cultured articular chondrocytes.

Cartilage degradation by neutral proteoglycanases in experimental osteoarthritis. Suppression by steroids

In this study, we sought to determine the role of neutral proteases in cartilage matrix proteoglycan degradation, which occurs during the early stages of experimental osteoarthritis. The anterior cruciate ligament was transected in the right knees of 33 dogs. Their left knees served as sham operated controls. The animals were killed 2, 4, 8, and 12 weeks after surgery. Six dogs were treated with oral prednisone and then killed 4 weeks after surgery. Cartilage specimens from medial and lateral tibial plateaus were analyzed for DNA, proteoglycan content, and neutral proteoglycan degrading activity. No significant differences in cartilage DNA and proteoglycan content were observed among the dogs that had surgery, the controls, and the prednisone-treated animals. Total neutral metalloproteoglycan-degrading enzyme (NMPE) activity, determined by direct tissue assay, was significantly higher at all time points in osteoarthritic cartilage than in control cartilage. The active form of NMPE was significantly higher in osteoarthritic cartilage than in control cartilage at 2, 4, and 8 weeks in lateral plateaus and at 2 and 4 weeks in medial plateaus. Treatment for 4 weeks with prednisone (0.20-0.25 mg/kg/day) blocked the increased NMPE activity in osteoarthritic cartilage. The increase in the total and active neutral proteoglycanases supports the hypothesis that these enzymes are involved in early osteoarthritis. The synthesis of NMPE appears to be controlled by stimulating factors released by the synovium.

Secretion of an articular cartilage proteoglycan-degrading enzyme activity by murine T lymphocytes in vitro

Destruction of articular cartilage is the hallmark of inflammatory arthritides. Enzymes elaborated by mononuclear cells infiltrating the synovium mediate, in part, the degradation of the cartilage extracellular matrix. Since mononuclear cells are the dominant cell type found in chronic inflammatory synovitis, we investigated whether interaction of immune mononuclear cells with antigen initiated the synthesis and secretion of a proteoglycan-degrading enzyme activity. Proteoglycan-degrading enzyme activity was monitored by the capacity of murine spleen cell conditioned medium to release [3H]serine/35SO4 incorporated into rabbit cartilage proteoglycan monomer fraction (A1D1), and by the relative change in specific viscosity of bovine nasal cartilage proteoglycan monomer. The results demonstrated that both virgin and immune mononuclear cells spontaneously generated proteoglycan-degrading enzyme activity and that cellular activation and proliferation induced by the antigen keyhole limpet hemocyanin or the mitogen phytohemagglutinin was not required. Kinetic studies demonstrated stable release of the enzyme activity over 72 h. Cell separation studies showed that T lymphocytes, a thymoma line, and macrophages separately produced proteoglycan-degrading enzyme activity. The enzyme activity has been partially characterized and appears to belong to a class of neutral pH metal-dependent proteinases. These observations, the first to demonstrate that T lymphocytes secrete an enzyme capable of degrading cartilage proteoglycan, raise the possibility that this enzyme activity contributes to cartilage extracellular matrix destruction in vivo. Moreover, these data support the conclusion that production of this enzyme by T lymphocytes is independent of an antigen-specific stimulus.

Human rheumatoid arthritic cartilage and its neutral proteoglycan-degrading proteases. The effects of antirheumatic drugs

Measurements were made of the neutral proteoglycan-digesting protease activity in the cartilage matrix breakdown observed in the rheumatoid arthritic process. Normal knee (tibial plateau) cartilage specimens were obtained from 7 fresh cadavers and 29 cartilage specimens were obtained from 23 patients diagnosed as having rheumatoid arthritis (RA). The total neutral metalloproteoglycan-degrading enzyme (NMPE) activity in RA cartilages exhibited roughly an eightfold elevation over that of control subjects. The active form of the NMPE for diseased cartilage was higher than that observed for normal cartilage, but was not statistically different. A very low level of activity was detected for serine proteases and no variation was observed between normal and diseased cartilages. Data obtained from RA cartilages were also analyzed with respect to the relationship between enzyme activities and the patients' medications. Four groups of patients were then selected according to their drug treatments: S + G patients received steroid and gold therapy; S patients received steroids only; NS + NG patients did not receive steroid or gold therapy; G patients received gold therapy alone. The total NMPE activity for each of these groups remained at a very high level. The active enzyme activity measured in S + G and S patients was decreased to a level not different from that of normal controls. Specimens from NS + NG patients presented a significantly higher level of the active form of the enzyme (P less than 0.05) when compared with either normal controls, S + G, or S patients. No significant difference was noted in the level of serine protease activity between the RA cartilage and normal cartilage.(ABSTRACT TRUNCATED AT 250 WORDS)

Kinetics of extraction of proteoglycans from human cartilage

Fifty-one uniformly sliced human patellar cartilage specimens were extracted with physiologic buffer containing inhibitors. Galactosamine-rich proteoglycans were more easily extracted than glucosamine-rich proteoglycans. Extraction occurred in 2 phases. There was a large proteoglycan extraction during the first 30 minutes, followed by a steady loss proportional to the square root of time. The amount of proteoglycans extracted during phase 1 was approximately half that of phase 2. Extraction of proteoglycans varied with age, structural integrity, and cartilage thickness. Intact osteoarthritic cartilage had normal proteoglycan extraction.

Nonsteroidal anti-inflammatory drugs and modulation of cartilaginous changes in osteoarthritis and rheumatoid arthritis. Clinical implications

Nonsteroidal anti-inflammatory drugs have a potential for modifying the complex pathophysiologic events leading to cartilage destruction in various forms of arthritis. Following an evaluation of basic mechanisms in the pathogenesis of cartilaginous destructive lesions, the effects of nonsteroidal anti-inflammatory drugs on normal chondrocyte metabolism are discussed. Their capacity to modulate cartilage and bone lesions in experimental forms of arthritis is addressed, as is the manner in which they may modify the pathophysiology of cartilage destruction in human forms of arthritis. Different classes of nonsteroidal anti-inflammatory drugs produce different effects in certain in vivo or in vitro settings.

Neutral proteases capable of proteoglycan digesting activity in osteoarthritic and normal human articular cartilage

Proteases have been postulated to account for the progressive disappearance of matrix proteoglycans in osteoarthritic (OA) cartilage. The digestion of endogenous proteoglycans by neutral proteases in human OA cartilage homogenates has been measured and compared with that of normal age-matched controls. Cartilage was obtained from 16 patients at the time of knee arthroplasty and from 7 accident victims. Tissue blocks were cut from the tibial plateau; part was used for histologic grading of the severity of OA and part was homogenized for the quantification of neutral metallo- and serine protease activities, based on the release of digested products from endogenous proteoglycans. Total metalloprotease activity (latent plus active forms) was elevated 3- to 10-fold in all diseased cartilage. This elevation was already significant in mild disease, but was greatest in samples of moderate to severe disease. The active form of the enzyme was highest at the center of erosions and decreased in the margins of the plateau. The digestion of proteoglycans, as distinct from their mere release from the tissue, was demonstrated by chromatography on Sepharose-CL2B and by large pore electrophoresis. Serine protease activity on proteoglycans was much lower than that of metalloprotease. The mean activity was highest in mild disease and declined in the severe disease samples, but the difference between these 2 groups and the controls was not statistically significant. The results of this study are consistent with the hypothesis that the neutral metalloproteases of cartilage are involved in the degradation of proteoglycans in osteoarthritis.

Cytokine modulation of chondrocyte proteinase release

Nonenzymatic, trypsin sensitive cytokines derived from lectin stimulated normal human mononuclear cells have been shown to induce release of proteoglycan and collagen degrading proteinase activity from chondrocytes in cartilage organ and isolated suspension culture systems. Active chondrocyte protein and RNA synthesis were required to induce activity. Cytokines responsible were of both monocyte and T cell origin. Direct monokine catabolic induction and monokine/lectin-triggered lymphokine inducing activity could be demonstrated. Cyclooxygenase inhibitors and direct or indirect modulation of mononuclear cell or chondrocyte cAMP levels had no effect on factor synthesis or activity. Hydrocortisone abrogated the effect. Cytokines responsible were heat labile, unaffected by reduction/alkylation or neuraminidase exposure, and stable over a pH range of 3-10.

Effects of retinol and dexamethasone on cytokine-mediated control of metalloproteinases and their inhibitors by human articular chondrocytes and synovial cells in culture

Human articular chondrocytes in culture produced large amounts of specific mammalian collagenase, gelatinase and proteoglycanase when exposed to dialysed supernatant medium derived from cultured human blood mononuclear cells (mononuclear cell factor) or to conditioned medium, partially purified by fractionation with ammonium sulphate (60-90% fraction), from cultures of human synovial tissue (synovial factor). Human chondrocytes and synovial cells also released into culture medium an inhibitor of collagenase of apparent molecular weight about 30 000, which appeared to be similar to the tissue inhibitor of metalloproteinases synthesised by tissues in culture. The amounts of free collagenase inhibitor were reduced in culture media from chondrocytes or synovial cells exposed to mononuclear cell factor or synovial factor. While retinol inhibited the production of collagenase brought about by mononuclear cell factor or synovial factor, it restored the levels of inhibitor, which were reduced in the presence of mononuclear cell factor or synovial factor. Dexamethasone markedly reduced the production of collagenase by synovial cells, while only partially inhibiting factor-stimulated collagenase production by chondrocytes. Addition of puromycin as an inhibitor of protein synthesis reduced the amounts of both collagenase and inhibitor to control or undetectable levels.

Characterization of latent and active forms of cartilage proteinases produced by normal immature rabbit articular cartilage in tissue culture

Cultured tissue slices from normal immature rabbit articular cartilage released latent neutral metalloproteinases into serum-free medium. On activation with 4-aminophenylmercuric acetate, these metalloproteinases could degrade collagen, proteoglycan, and gelatin. Also produced were an acid proteinase with the properties of cathepsin D and an inhibitor of the neutral metalloproteinases. The appearance of both the proteinases and the inhibitor in the culture medium could be prevented by incubation of cultures with cycloheximide. The active and latent forms of the proteinases were characterized using Ultrogel AcA 54 chromatography.

'Gelatinase-like' activity from articular chondrocytes in monolayer culture

In addition to releasing collagenase and proteoglycanase activity, rabbit articular chondrocytes in monolayer culture released into the culture medium, latent, neutral enzyme activity which when activated by p-aminophenylmercuric acetate degraded fluorescein-labeled polymeric rat tail tendon Type I collagen and the tropocollagen TCA and TCB fragments of human Type II collagen into smaller peptides at 37 degrees C. Enzyme activity was abolished if p-aminophenylmercuric acetate-activated culture medium was preincubated with 1.10-phenanthroline, a metal chelator. Thus, articular chondrocytes in monolayer culture are capable of producing neutral proteinases which acting together can result in complete degradation of tendon and cartilage collagen to small peptides.

Biosynthesis of proteoglycan in vitro by cartilage from human osteochondrophytic spurs

Proteoglycan biosynthesis by human osteochondrophytic spurs (osteophytes) obtained from osteoarthritic femoral heads at the time of surgical joint replacement was studied under defined culture conditions in vitro. Osteophytes were primarily present in two anatomic locations, marginal and epi-articular. Minced tissue slices were incubated in the presence of [(35)S]sulphate or [(14)C]glucosamine. Osteophytes incorporated both labelled precursors into proteoglycan, which was subsequently characterized by CsCl-isopycnic-density-gradient ultracentrifugation and chromatography on Sepharose CL-2B. The material extracted with 0.5m-guanidinium chloride showed 78.1% of [(35)S]sulphate in the A1 fraction after centrifugation. Only 23.0% of the [(35)S]sulphate in this A1 fraction was eluted in the void volume of Sepharose CL-2B under associative conditions. About 60-80% of the [(35)S]sulphate in the tissue 4m-guanidinium chloride extract was associated with monomeric proteoglycan (fraction D1). The average partition coefficient (K(av.)) of the proteoglycan monomer on Sepharose CL-2B was 0.28-0.33. Approx. 12.4% of this monomer formed stable aggregates with high-molecular-weight hyaluronic acid in vitro. Sepharose CL-2B chromatography of fractions with lower buoyant densities (fractions D2-D4) demonstrated elution profiles on Sepharose CL-2B substantially different than that of fraction D1, indicative of the polydisperse nature of the newly synthesized proteoglycan. Analysis of the composition and chain size of the glycosaminoglycans showed the following: (1) preferential elution of both [(35)S]sulphate and [(14)C]glucosamine in the 0.5m-LiCl fraction on DEAE-cellulose; (2) the predominant sulphated glycosaminoglycan was chondroitin 6-sulphate (60-70%), with 9-11% keratan sulphate in the monomer proteoglycan; (3) K(av.) values of 0.38 on Sephadex G-200 and 0.48 on Sepharose CL-6B were obtained with papain-digested and NaBH(4)-treated D1 monomer respectively. A comparison of the synthetic with endogenous glycosaminoglycans indicated similar types. These studies indicated that human osteophytes synthesized in vitro sulphated proteoglycans with some characteristics similar to those of mature human articular cartilage, notably in the size of their proteoglycan monomer and predominance of chondroitin 6-sulphate. They differed from articular cartilage primarily in the lack of substantial quantities of keratan sulphate and aggregation properties associated with monomer interaction with hyaluronic acid.

Further characterization of a neutral metalloprotease isolated from human articular cartilage

The neutral metalloprotease extracted from 1,200 gm of human articular cartilage was purified 1,400- to 2,400-fold by diethylaminoethyl- and carboxymethyl-Sephadex chromatography. Disc electrophoresis and an isoelectric focusing method resolved the neutral enzyme activity into 4 bands. All bands had a similar amino acid analysis and a similar molecular weight by sodium dodecylsulfate electrophoresis and gel filtration: 24,000-27,000 daltons. The enzyme degraded proteoglycan subunit and proteoglycan aggregate to products with a sedimentation coefficient of 3S, but at low dilutions the enzyme produced 19.3S fragments. It is postulated that this protease may contribute to the development of osteoarthritis from within the cartilage matrix.

The properties of the neutral proteinase released by primary chondrocyte cultures and its action on proteoglycan aggregate

The enzymatic mechanism of proteoglycan breakdown is of major interest, since it has been proposed that osteoarthritis involves increased proteolytic breakdown of proteoglycans. This paper describes the properties of the proteoglycan-degrading enzymes released into the extracellular milieu by chondrocyte cultures that produce cartilage-specific type II collagen but no detectable type I collagen. Attention has been focused on enzymes active at neutral pH, since the pH of the extracellular matrix is around neutrality. Biogel P-60 chromatography of concentrated culture medium showed a major peak of enzyme activity on proteoglycan monomer entrapped in polyacrylamide beads as well as on native proteoglycan aggregates. The enzyme yields a specific limit digestion peptide from the aggregate of approximately 55,000 daltons (in the presence of SDS). This limit peptide is probably derived from the hyaluronic acid-binding region of proteoglycan. The proteolytic enzyme is latent but can be activated by aminophenylmercuric acetate or trypsin. The molecular weight of both the active and latent forms, determined by gel filtration, is approximately 33,000. The activity is not inhibited by phenylmethylsulfonyl fluoride or pepstatin but is completely inhibited by o-phenanthroline; the activity is restored by Zn or Co ions in the presence of calcium chloride. Removal of calcium by dialysis results in a reversible loss of activity. The release of such a metalloproteinase by chondrocytes into the extracellular milieu, its activity at physiological pH and its ability to degrade native proteoglycans are consistent with a role of the enzyme in proteoglycan metabolism.

Correlation of the biosynthesis of prostaglandin and cyclic AMP in monolayer cultures of rabbit articular chondrocytes

We have utilized ionophores to test whether stimulation of chondrocyte prostaglandin biosynthesis is accompanied by an increase in cyclic nucleotide levels in these cells. Radioimmunoassay of prostaglandin E2, 6-oxo-prostaglandin F1 alpha (the stable metabolite of prostaglandin I2) and prostaglandin F2 alpha showed that synthesis of each was stimulated by the divalent-cation ionophore, A23187 after short-term incubation (1-7 min) in serum-free medium. No stimulation of thromboxane B2 was detected. Two monovalent ionophores, lasalocid and monensin failed to stimulate prostaglandin biosynthesis after short-term incubation. Ionophore A23187-stimulated prostaglandin biosynthesis was variably and partially inhibited by sodium meclofenamate, indomethacin and aspirin, but not by sodium salicylate. Ionophore A23187-stimulated prostaglandin biosynthesis was accompanied by a 7.5-fold increase in cyclic AMP levels after 15 min. Sodium meclofenamate, indomethacin and aspirin which inhibited prostaglandin E2 biosynthesis also reduced cyclic AMP levels. Exogenous prostaglandin E2 (1 microgram/ml) stimulated cyclic AMP biosynthesis, which was not inhibited by aspirin. These results indicated that prostaglandins can be considered as one of the local effectors controlling cyclic AMP production in articular cartilage.