Effects of copper and zinc on proteoglycan metabolism in articular cartilage

Co-Cultures of porcine articular cartilage and synovium or synovial conditioned medium were used as an in vitro model to mimic inflammatory events at the cartilage/synovial junction in degenerative joint disease. This model provides a useful tool to assess the anti-inflammatory and antiarthritic properties of pharmacological agents. In this study the effects of copper and zinc on (i) PG synthesis by cartilage and (ii) synovial-induced PG depletion have been investigated. Copper sulphate at a concentration of 0.01 mM did not stimulate PG synthesis significantly in cultured cartilage explants but completely abrogated the inhibitory effects of synovial tissue in co-culture experiments. This finding was supported by the histological demonstration of copper-dependent reversal of the PG depletion in cartilage exposed to synovial conditioned medium. Zinc sulphate at 0.01 mM had no effect on PG synthesis and was unable to protect cartilage against synovialinduced PG depletion. These results reveal possible mechanisms by which copper exerts its anti-inflammatory and anti-arthritic actions.


Introduction
Copper and zinc are essential trace minerals known to have important roles in immuneand inflammatory -6 processes. Adequate amounts of both minerals are required for normal maturation and function of the immune system, and deficiency, particularly of zinc, results in diminished immune responsiveness due to inhibition of T cell-mediated cytotoxicity r and antibody production. It is also well established that dietary deficiency of copper and zinc correlates with enhanced acute inflammatory responses. 6,9,1 Although the therapeutic anti-inflammatory properties of copper and zinc have been extensively studied, both in human disease and in animal models, 1surprisingly little is known about their mechanisms of action. Even less information is available concerning their anti-arthritic action. [14][15][16] In order to gain some insight into anti-arthritic mechanisms we used a co-culture model of cartilage and synovial tissue to mimic events occurring at the cartilage/synovial pannus junction, the site of the initial pathological process of rheumatoid joint disease. As a result of an inflammatory reaction the synovial membrane becomes infiltrated by immunocompetent cells to form the pannus which invades the cartilage surface and underlying bone. Concomitant triggering of the cytokine network and production of degradative enzymes leads to depletion of the extracellular matrix macromolecules proteoglycan and collagen, and eventually to joint destruction. Our in vitro co-culture model provides a useful tool for studying the mechanism of action of pharmacological agents on the cellular and mediator interactions which play a role in these destructive pathological processes.
The aim of the present study was to investigate the ability of copper and zinc to abrogate the degradative effects of synovial tissue on cartilage matrix proteoglycan. Materials   0.58 mg/ml glutamine, 120 l.tg/ml benzyl penicillin, 200 l.tg/ml streptomycin, in a 5% CO2 environment at 37C. In the first experiment cartilage explants were cultured for 48 h in the presence or absence of a range of concentrations (0.01-2.0mM) of copper sulphate or zinc sulphate in order to assess their effects on proteoglycan (PG) synthesis. From these data nontoxic concentrations were selected for further experiments in which cartilage explants were cocultured in direct contact with minced synovium in the presence or absence of copper and zinc prior to measurement of PG synthesis.
In another group of experiments synovial tissue was pre-cultured for 48 h in the presence or absence of a range of copper sulphate concentrations (0.001-0.1 mM). The tissue-free synovial conditioned medium was then cultured with cartilage explants for a further 48h, at which time the explants were removed, snap-frozen in OCT embedding medium (Miles Diagnostics) in liquid nitrogen and stored at-70C prior to sectioning for toluidine blue histology. Five cartilage explants were used for each experimental group.

Measurement of proteoglycan (PG) synthesis:
After culture in the presence of copper and zinc, or co-culture with synovial tissue, the explants were radiolabelled essentially as described previously 9 by placing fresh DMEM containing 20 l.tCi/ml (0.74 MBq/ml) of [5S]-sulphate and incubating for 2h at 37C. The explants were then digested in papain (300 l.tg/ml) for 2h at 65C and the PG synthesis rate determined as the 35SO42incorporation, and expressed as CPM/mg dry weight of cartilage. Toluidine blue histology: Frozen sections 5-8 l.tm thick were cryostat cut at-25-35C from cartilage explants exposed to synovial conditioned medium. The sections were placed on poly-lysine (Sigma) 8-well multitest slides (Flow Labs) and stained with toluidine blue for visual assessment of proteoglycan depletion.
Statistics: Proteoglycan synthesis measurements were performed in quintuplicate in every experiment. The dose-response experiments of copper and zinc on proteoglycan synthesis were done twice. The co-culture experiments were done three times with copper and twice with zinc, and the synovial conditioned medium experiment was done only once. Statistical evaluation of the data was done using Student's t-test.

Results
Effects of copper and zinc on PG synthesis by cartilage explants: In two separate experiments copper sulphate at concentrations of 0.1 mM and below had little effect on PG synthesis in 48h cartilage cultures. Since there was wide variation in basal PG synthesis between the two experiments the results were not combined. The results from a typical experiment are shown in Fig. I(A). Higher concentrations of copper (0.25-2.0mM) caused a dose-dependent decrease in PG synthesis. Zinc sulphate at 0.01 mM caused a slight but non-significant increase in PG synthesis ( Fig. I(B)). Concentrations Figure 3 illustrates typical staining patterns seen in frozen sections prepared from the explants from the five different experimental groups. Figure 3(A) shows uniform metachromatic staining throughout the matrix, with slight depletion at the tip, in cartilage cultured alone for 48h. Cartilage cultured in synovial conditioned medium shows characteristic extensive loss of metachromasia (Fig. 3(B)). Pre-treatment of synovium with 0.001 mM copper sulphate had c a ft. ca rt.
c ft. + syn little effect on the depletion of metachromasia alone + syn + 0.01 mM (Fig. 3(C)) but higher concentrations increased copoe sulpoe the intensity of staining (0.01mM, Fig. 3(D)) indicating markedly reduced loss of metachromasia induced by the copper-pretreatment conditioned B medium (0.1 mM, Fig. 3(E)). Experimental studies show that copper and zinc are involved in the expression and control of inflammation. How their metabolism changes in inflammation and how they can affect the development of inflammatory processes is unknown. In an attempt to gain insight into their possible anti-inflammatory and anti-arthritic mechanisms, we have investigated the effects of copper and zinc in an in vitro model of rheumatoid joint inflammation.
Our results presented here show that, co- Effects of copper and zinc on PG synthesis by The same inhibitory effect on PG synthesis has cartilage co-cultured with synovium.. Co-culture also been found by pre-culturing the synovium of cartilage with synovial tissue for 48h resulted for 2 days alone and then adding the synovial in marked reduction in PG synthesis. Figure2(A) conditioned medium (SCM) to the cartilage shows the results of a typical experiment in explants for 2 more days. In fact, cartilage culwhich synovial tissue caused a 40% reduction in tured in SCM alone showed a pronounced loss PG synthesis by cartilage. The presence of of toluidine blue staining, reflecting depletion of copper sulphate at a concentration of 0.01 mM matrix PG. This extensive reduction of metachroduring the 48h culture period abrogated the masia did not occur if the synovium had been inhibitory effects of synovium (p < 0.01).
pre-cultured in the presence of copper Zinc sulphate at a concentration of 0.01 mM (0.01 mM) and if copper was still present in SCM was unable to reverse the inhibitory effects of added to cartilage, demonstrating a coppersynovial tissue on PG synthesis (Fig. 2(B)). dependent reversal of PG depletion in cartilage exposed to SCM.
Effect of conditioned medium from copper-From these results we can conclude that (i) treated synovium on PG content in cultured the PG-degrading effect of synovium on cartilage does not necessarily require direct synovium-cartilage contact, but is likely attributable to mediators released from the synovium into the medium; (ii) copper is protective on cartilage matrix depletion either in co-culture or if the two tissues are not simultaneously present in the culture; and (iii) zinc is unable to reverse the PG depletion by synovium. Because copper was always present in the medium, these results cannot distinguish between an inhibitory effect of copper on synovium or a stimulatory effect on PG synthesis in cartilage. However, we have also shown that copper added to the medium at a concentration of 0.01 mM did not increase PG synthesis in cartilage cultured alone, supporting indirectly an action of copper via synovium inhibition.
The original studies of Fell and Jubb 17 in 1977 demonstrated the ability of porcine synovial tissue to promote loss of PG from cartilage, and it is now known that interleukin 1 (IL-1) is one of the synovial mediators responsible for this 23 24 degradative activity.
Interestingly, recent reports have shown that cotlPer is able to inhibit IL-1 activity 25 and secretion," findings which may explain our observations that copper protects cartilage from synovium-induced destruction. At present we have no direct evidence that copper inhibits IL-1 in our co-culture system. An alternative mechanism by which copper may exert its protective action is via inhibition of the free radical, nitric oxide (NO). Increasing evi-dence implicates NO in the pathophysiology of arthritis, 26 suggesting a role for NO in both the inflammation and the chondrodepletion that occurs in this disease. Recent data obtained by Taskiran and co-workers 27 demonstrate that endogenously synthesized NO acts as the mediator of IL-l-dependent PG synthesis suppression.
Analyses of human synovial fluids s.uegest that NO is synthesized locally within joints and that synoviocytes can potentially serve as a source of intra-articular NO. 29 In support of this we have preliminary data showing NO production by normal pig synovium (data not shown), and furthermore we have found a 50% reduction of NO content in SCM when synovium was cultured in the presence of non-toxic concentrations of copper (0.001-0.01 mM). The same SCM also abrogated the degradative effects of synovial tissue on cartilage, suggesting a connection between copper, NO reduction and cartilage protection. In the light of these results and because Scuderi 21 has recently shown that copper may play an important role in the homeostasis of cTtokines such as IL-1, 2 we could hypothesize an action of copper on PG metabolism as indicated in Fig. 4.
The question as to whether copper affects NO production directly or via modulation of IL-1 synthesis and/or secretion is under investigation.
In conclusion copper, in contrast to zinc, seems to be able to protect cartilage from the destructive effects of mediators released by synovium. Because loss of articular cartilage is the major pathological lesion common to all arthritides, copper chondroprotection may explain the therapeutic efficacy of this metal in maintaining cartilage integrity during the inflammation process.