The effects of NSAID on the matrix of human articular cartilages.

The present paper presents data obtained over a 12 year period, on the matrix synthesis and turnover in some 650 arthritic and 180 non-arthritic (N) human cartilages using a standardised in vitro method. When the relative metabolic (synthetic/repair activity) of these human cartilages was compared, it was demonstrated that in osteoarthritis (OA) and rheumatoid arthritis (RA) cartilages synthetic activity was diminished by approximately 50% as compared with N cartilages. However, the turnover rate of matrix was not significantly different between Non-arthritic and OA, but was very substantially increased in RA cartilages compatible with the activity of inflammatory cells and proteolytic enzymes released from pannus. The action of 13 NSAIDs was compared in terms of their effect on cartilage GAG synthesis. 3 of these NSAIDs were also studied in terms of their effect on cartilage collagen synthesis. Consideration of the results in this study and from published material, led to the suggestion that NSAIDs may be divided into 3 categories in respect of their in vitro action on the extracellular matrix of human arthritic cartilages: 1. Those such as Aceclofenac, Tenidap and Tolmetin which can stimulate matrix synthesis 2. Those such as Piroxicam, Tiaprofenic Acid and Aspirin which appear to be without significant effect on matrix synthesis and, 3. Those like Naproxen, Ibuprofen, Indomethacin, Nimezulide which significantly inhibit matrix synthesis. It is suggested that the stimulatory action of group 1 NSAID is due to inhibition of locally produced IL1 and consequent expression of growth factor activity. Other NSAIDs may also inhibit IL1 synthesis or release, but probably do not have a beneficial effect on chondrocyte synthetic activity as they have toxic effects on cartilage metabolism. These experiments led to the suggestion that NSAIDs such as Aceclofenac would be appropriate for long-term treatment of arthritic conditions provided that one is prepared to extrapolate between in vitro experiments on human cartilage and what may be happening in vivo.

Chondroprotection and the role of non steroidal anti-inflammatory drugs (NSAID) / O papel dos antiinflamatórios näo hormonais (AINHs) na condroproteçäo: um estudo do metabolismo da cartilagem humana

O termo condroproteçäo foi originalmente cunhado para descrever os possíveis efeitos da inibiçäo do processo catabólico imaginado responsável pela lesäo da matriz da cartilagem articular artrítica. Apesar de diversas evidências laboratoriais de possíveis inibidores da degradaçäo proteolítica do glicosaminoglicano da cartilagem (GAG) e colágeno, nunca houve validaçäo desses resultados em ensaios in vivo, possivelmente devido à complexidade dos mecanismos enzimáticos responsáveis pela degeneraçäo tecidual. Assim, o conceito condroproteçäo começou a ser encarado como uma meta inatingível. Estudos recentes, que têm demonstrado a importância da atividade sintética da cartilagem humana no prolongamento de sua integridade estrutural durante processos patológicos, têm levado a uma reavaliaçäo da possibilidade de condroproteçäo. É possível que alguns antiinflamatórios näo hormonais (AINHs) sejam capazes de diminuir a açäo da IL-1 na síntese da matriz, permitindo assim o reparo desta, enquanto outros possam diminuir de forma importante o metabolismo dos condrócitos. Os AINHs têm sido divididos em três categorias: grupo I, exemplificado pelo aceclofenaco, säo capazes de causar estímulo, auxiliados pela presença de outros fatores; grupo II, exemplificados pelo diclofenaco, em doses moderadas näo apresentam efeitos relevantes sobre o metabolismo da cartilagem, embora em doses maiores possam ser inibitórios; e o grupo III é exemplificado pela indometacina, que é altamente inibitória em termos de atividade sintética da matriz e provavelmente diminui o processo reparativo normal. Concluímos que é desejável que os clínicos considerem as possíveis propriedades protetoras ou destrutivas dos diversos AINHs ao decidir por seu uso em terapêutica prolongada

Metalloproteinase production by rabbit articular cartilage: comparison of the effects of interleukin-1 alpha in vitro and in vivo.

To assess the effects of interleukin-1 on intact To assess the effects of interleukin-1 on intact articular cartilage in vitro, explants from young and adult rabbits were cultured with interleukin-1 and the distributions of the matrix metalloproteinases and tissue inhibitor of metalloproteinases (TIMP-1) were investigated by indirect immunofluorescence microscopy. One to 2-week-old cartilage chondrocytes synthesized collagenase in response to pure or crude interleukin-1 (monocyte conditioned medium), with subarticular cells most responsive. Collagenase synthesis was not stimulated in adult articular chondrocytes when explants were treated with either pure or crude interleukin-1. Stromelysin, gelatinase and TIMP-1 could not be demonstrated within any zone of the cartilage, indicating that their synthesis was not stimulated by either pure or crude interleukin-1. The addition of fibroblast growth factors, either alone or in combination with interleukin-1, did not modify these responses. These results contrast markedly with observations on cultured chondrocyte monolayers, where interleukin-1 treatment induces near co-ordinate expression of metalloproteinases. To assess the effects of interleukin-1 in vivo, it was injected into adult rabbit knee joint spaces and the articular cartilage subsequently analysed for evidence of altered metalloproteinase production by immunocytochemistry. No significant increase in metalloproteinase or TIMP-1 synthesis by chondrocytes was detected, although the cartilage matrix showed a marked loss of toluidine blue metachromasia. We conclude that metalloproteinases are not involved in the rapid loss of proteoglycan from cartilage matrix in these situations.

Effect of tenidap on cartilage integrity in vitro.

OBJECTIVES: The maintenance of articular cartilage integrity during long term treatment with non-steroidal anti-inflammatory drugs (NSAIDs) is of clinical importance. These experiments were set up to test the action of tenidap, naproxen, and diclofenac on bovine and porcine cartilage, matrix synthesis, and catabolism. METHODS: Short term organ culture techniques were used to determine the effect of interleukin 1 (IL-1) on synthesis and degradation, and the action of tenidap and the other drugs on these parameters. The retention of glycosaminoglycans (GAGs) and the synthesis of GAGs by incorporation of sulphur-35 labelled sulphate was used to determine the chondrocyte metabolic activity. RESULTS: The action of human recombinant interleukin 1 alpha (hrIL-1 alpha) in increasing catabolic activity and inhibiting synthetic activity of the animal cartilages was confirmed. Tenidap was shown to give substantial and significant protection against the catabolic effects of hrIL-1 alpha and, to a lesser degree, against the inhibition of matrix synthesis by the cytokine. Neither diclofenac nor naproxen in doses expected to occur in the synovial fluid showed this action. Tenidap also inhibited the GAG loss from cocultures and, to a moderate degree, reversed the inhibition of synthesis by synovial tissue. Tenidap also stimulated cartilage repair activity during recovery from IL-1 treatment. The optimum concentration of the action against IL-1 was between 5 and 10 micrograms/ml. Above this concentration tenidap itself showed some inhibitory action on GAG synthesis. CONCLUSIONS: Bearing in mind the problems in extrapolating from in vitro work on animal cartilages to humans, it seems possible that tenidap may be useful in decreasing the deleterious action of cytokines such as IL-1 on cartilage integrity during arthritic disease and in stimulating chondrocyte repair processes.

Prostaglandins in human cartilage metabolism.

(1) Human cartilage, both non-arthritic (N) and arthritic, is extremely sensitive to inhibition of glycosaminoglycan (GAG) synthesis by low concentrations of interleukin 1 (IL1). Local episodic synthesis and secretion of sub-nanogram concentrates of the cytokine is considered to play a significant role in the pathogenesis of osteoarthritis (OA) by preventing matrix repair. (2) The synthesis of IL1 can be controlled by prostaglandins (PGs), which may explain why the inhibitory action can be at least partially overcome by the action of the PG analogue Misoprostol in the dose range 10-100 ng/ml. It is suggested that this action is due to the suppression of a positive feedback loop for local IL1 synthesis and secretion. (3) Certain non-steroidal anti-inflammatory drugs (NSAIDs), in particular Indomethacin, Ibuprofen and Naproxen, cause inhibition of GAG synthesis, and hence may diminish the potentiality for repair in arthritic cartilage. It is suggested that these NSAIDs induce IL1 synthesis by diminishing PG levels. Misoprostol is able to reverse this effect at least partially. (4) Some cartilages in the presence of other NSAIDs, such as Diclofenac, which do not greatly inhibit chondrocyte matrix metabolism, nevertheless respond to the presence of Misoprostol by increased GAG synthetic activity. (5) The low mean matrix synthetic activity of human OA cartilages was significantly increased by Misoprostol. (6) Taken together, these studies substantiate the suggestion that Misoprostol is able to increase the repair potentiality of human OA cartilage, particularly during treatment with NSAIDs.

The sensitivity of synthesis of human cartilage matrix to inhibition by IL-1 suggests a mechanism for the development of osteoarthritis.

The damage to articular cartilage, characteristic of arthritic disease, is usually ascribed to increased degradative activity by enzymes or free radicals from locally activated cells. We propose that inhibition of matrix synthesis, and consequential impairment of the natural repair process, may be at least as important in chronic joint disease.

Cartilage maintenance in osteoarthritis: interaction of cytokines, NSAID and prostaglandins in articular cartilage damage and repair.

The structural integrity of the matrix of human articular cartilage is maintained by a dynamic equilibrium between synthesis and degradation. In osteoarthritis (OA), synthesis may be inhibited by the presence of subnanogram quantities of the cytokine interleukin 1 (IL-1), leading in the longterm to loss of matrix and susceptibility to mechanical damage. IL-1 may also inhibit the potential for repair processes to take place in this cartilage if continued synthesis and secretion of the cytokine occurs. Evidence is presented that animal and human cartilages are sensitive to the action of certain nonsteroidal antiinflammatory drugs (NSAID) in inhibiting the synthesis of cartilage proteoglycan and also diminishing the repair activity of cartilage recovering after IL-1. In OA cartilage, the sensitivity to action of NSAID may depend on the state of the tissue in terms of glycosaminoglycan (GAG) turnover and GAG synthetic activity of the indigenous chondrocytes. Preliminary investigations of the prostaglandin analog misoprostol on the synthetic repair activities of animal and human cartilage in the presence of NSAID are reported.

In vivo studies of cartilage regeneration after damage induced by catabolin/interleukin-1.

The response of the rabbit knee joint to a brief episode of cytokine induced damage is described. After three intra-articular injections of catabolin/interleukin-1 all joint cartilages showed an immediate extensive loss of proteoglycan (glycosaminoglycan), which was gradually replaced over three to four weeks. Glycosaminoglycan biosynthesis (measured by 35SO4 uptake) was initially depressed, but at one week had almost doubled its rate as compared with the normal side. This increased synthetic activity was further maintained throughout the duration of the experiment (28 days), though the rate gradually fell. Histological cartilage metachromasia to toluidine blue mirrored the glycosaminoglycan changes. No disturbance of the articular cartilage collagen network was found. It is considered, therefore, that during treatment for arthritis the indigenous chondrocyte must continue to be capable of carrying out regenerative matrix repair and that antiarthritic agents should first be screened for interference with that process.

Expression of intercellular adhesion molecule 1 (ICAM-1) on human articular cartilage chondrocytes.

Monoclonal antibodies have been used to demonstrate the induction of intercellular adhesion molecule 1 (ICAM-1) on chondrocytes in human articular cartilage. ICAM-1 was found not to be constitutively expressed but could be induced by exogenous interleukin 1 alpha(IL1- alpha) at concentrations ranging from 0.01 to 20 ng/ml during in vitro culture. Maximum expression was observed with 2-5ng/ml. In time-course experiments ICAM-1 was not expressed after 4h in culture with IL1 alpha. Expression was induced by 16h and was sustained for a minimum of 6 days in the continued presence of the cytokine. The endothelial leukocyte adhesion molecule (ELAM-1) was not expressed on chondrocytes and was not induced by IL1-alpha.

Immunorecognition of chondrocytes in articular cartilage activated by synovial interleukin 1.

A polyclonal antiserum which recognizes surface epitopes on IL1-activated pig chondrocytes has been used to immunolocalize chondrocytes responding to IL1 produced during co-culture of pig synovium and articular cartilage. Activation of the chondrocytes by the cytokine was restricted to the articular and subarticular region of the cartilage adjacent to the synovium. Chondrocyte activation was also seen when human rheumatoid synovium was co-cultured with the cartilage. The presence of IL1 in some synovial cells was confirmed by immunolocalization using antisera specific for IL1 alpha and IL1 beta.

Immunohistological identification of interleukin-1 activated chondrocytes.

Interleukin-1 activated chondrocytes have been identified in pig articular cartilage by immunolocalisation using a polyclonal antiserum recognising cytokine induced surface epitopes. Although chondrocytes with membrane staining were noted in all zones of the cartilage, only a proportion of the cells were positive within the range of antiserum dilution used (1:100-1:2000). Maximum staining was found after four days' treatment of cultured cartilage with pig IL-1 alpha, though weak staining was just detectable after treatment for one day at high antiserum concentration. Immunoreactivity was also shown in cartilage which had been cultured with human recombinant IL-1 alpha.

The sulphated glycosaminoglycan levels in synovial fluid aspirates in patients with acute and chronic joint disease.

Proteoglycan levels were measured in a series of synovial fluid samples from patients with acute and chronic joint diseases using a modified chemical dye binding method. Levels found in 50 miscellaneous inflammatory arthritis fluids (mean = 173.2 +/- 90.9 micrograms/ml) were higher than found in either 50 with rheumatoid arthritis (96.3 +/- 31.3 micrograms/ml) or 50 with osteoarthritis (83.8 +/- 27.3 micrograms/ml). For comparison, proteoglycan levels were measured in 15 cadaver synovial fluids (98.9 +/- 44.2 micrograms/ml) and 12 synovial fluids from patients with sports injury (163.7 +/- 79.4 micrograms/ml). Patients were recruited into a trial where synovial fluid was aspirated as often as possible over a 6-month period during which the patients were followed using a number of well proven clinical parameters. No correlation was found between the degree of joint destruction as measured by X-ray damage and the concentration or total amount of proteoglycan in the synovial fluid. In addition, there was no correlation between the level or total amount of proteoglycan and any clinical parameter of disease activity.

Clearance of proteoglycan from joint cavities.

The rate of loss from the synovial cavity of proteoglycan subunit, a major constituent of cartilage, has been measured in rabbits with and without experimental arthritis. The effect of aggregation between proteoglycan and hyaluronic acid upon the rate of elimination has also been investigated. The results indicate that proteoglycan subunit has a clearance half life of around 12 hours in normal joints and that this rate is not significantly altered in the presence of an active synovitis. Neither injection of proteoglycan as an aggregate, nor in a form incapable of aggregation, had any significant effect on clearance rates. These data indicate that loss of proteoglycan from the joint is not retarded by its ability to aggregate with hyaluronic acid in the synovial fluid and, together with measurements of synovial fluid proteoglycan, may enable rates of cartilage breakdown in vivo to be calculated.

In vivo studies of articular tissue damage mediated by catabolin/interleukin 1.

A partially purified porcine synovial catabolin interleukin 1 (CF) preparation was injected intra-articularly into rabbit stifle (knee) joints. Radiolabelled CF was rapidly cleared from the joint (0.4 h). Repeated injections of CF caused a marked loss of articular cartilage glycosaminoglycan (GAG) and a great increase in synovial fluid GAG. 35SO4 uptake was inhibited. Time course experiments after a single injection produced similar loss of GAG from knee cartilages, which was maximal three days after injection. The above changes were significantly less with heat inactivated preparations. Loss of articular cartilage metachromasia was found histologically, and an acute synovitis occurred together with lymphocytic foci and plasma cell infiltration.

Production of cartilage degrading activity by human synovial tissues.

Human synovial tissues have been assayed for the production of cartilage degrading activity (CDA). This activity is thought to be homologous with catabolin/interleukin 1 (IL1) produced by porcine synovium and leucocytes and by human leucocytes. The CDA of 26 rheumatoid (RA) and 41 non-rheumatoid synovia was measured on a dry weight basis. The rheumatoid synovia showed a threefold increase in activity over the non-rheumatoid, but there was no significant overall difference on a deoxyribonucleic acid (DNA) basis. The rheumatoid synovia appeared to consist of two populations; in over half the samples CDA was not related to cellularity, but eight patients had a high CDA and a high cellularity.

The role of cytokines in arthritic diseases: in vitro and in vivo measurements of cartilage degradation.

The injection of partially purified porcine synovial catabolin/IL-1 alpha intra-articularly in rabbits resulted in a 50% loss of glycosaminoglycan (GAG) after 3 days. An increase in the synovial fluid content of GAG was found, and 35SO4 incorporation into proteoglycan was inhibited. Measurements were also made of the GAG loss from articular cartilage into the synovial fluid in human rheumatoid (RA) and osteoarthritic (OA) patients. Very high levels of GAG content in the synovial fluid was found, and calculations were made of the half-life of the cartilage proteoglycan during the active phases of the disease. Estimation of the synovial fluid GAG is believed to be a simple and quantitative method for monitoring the effectiveness of cartilage-"sparing" anti-arthritic drugs.

Porcine heart valves produce a protein that induces cell-mediated connective tissue degradation: II. Biochemical properties of the partially purified protein.

A cardiac catabolic factor (CCF) has been partially purified from serum-free medium conditioned by minced porcine heart valves. CCF was prepared by a series of chromatographic techniques and compared directly with porcine synovial catabolin purified by the same protocol. CCF displayed a somewhat higher molecular weight (Mr 21,000) and isoelectric point (pI 5.2) than did synovial catabolin (Mr 18,000 and pI 4.8), but the two factors clearly resemble one another closely. CCF stimulated the release of glycosaminoglycans from cultured cartilage and mitral valve and provoked porcine valves to degrade their own collagen extracellular matrix. The release of hydroxyproline was inhibited by corticosteroids, whereas proteoglycan breakdown was not. Partially pure preparations of CCF and synovial catabolin stimulated murine thymocyte proliferation; moreover, that activity was almost totally abolished by an antibody raised against pure porcine interleukin-1. These observations suggest that CCF may represent a catabolic factor that belongs to the interleukin-1 family and that it could potentially regulate the composition of valvular connective tissue.