Modulation of IL-1-induced cartilage injury by NO synthase inhibitors: a comparative study with rat chondrocytes and cartilage entities.

Nitric oxide (NO) is produced in diseased joints and may be a key mediator of IL-1 effects on cartilage. Therefore, we compared the potency of new [aminoguanidine (AG), S-methylisothiourea (SMT), S-aminoethylisothiourea (AETU)] and classical [Nomega-monomethyl-L-arginine (L-NMMA), Nomega-nitro-L-arginine methyl ester (L-NAME)] NO synthase (NOS) inhibitors on the inhibitory effect of recombinant human interleukin-1beta (rhIL-1beta) on rat cartilage anabolism. Three different culture systems were used: (1) isolated chondrocytes encapsulated in alginate beads; (2) patellae and (3) femoral head caps. Chondrocyte beads and cartilage entities were incubated in vitro for 48 h in the presence of rhIL-1beta with a daily change of incubation medium to obtain optimal responses on proteoglycan synthesis and NO production. Proteoglycan synthesis was assessed by incorporation of radiolabelled sodium sulphate [Na2(35)SO4] and NO production by cumulated nitrite release during the period of study. Chondrocytes and patellae, as well as femoral head caps, responded concentration-dependently to IL-1beta challenge (0 to 250 U ml(-1) and 0 to 15 U ml(-1) respectively) by a large increase in nitrite level and a marked suppression of proteoglycan synthesis. Above these concentrations of IL-1beta (2500 U ml(-1) and 30 U ml(-1) respectively), proteoglycan synthesis plateaued whereas nitrite release still increased thus suggesting different concentration-response curves. When studying the effect of NOS inhibitors (1 to 1000 microM) on NO production by cartilage cells stimulated with IL-1beta (25 U ml(-1) or 5 U ml(-1)), we observed that: (i) their ability to reduce nitrite level decreased from chondrocytes to cartilage samples, except for L-NMMA and AETU; (ii) they could be roughly classified in the following rank order of potency: AETU > L-NMMA > or = SMT > or = AG > or = L-NAME and (iii) AETU was cytotoxic when used in the millimolar range. When studying the effect of NOS inhibitors on proteoglycan synthesis by cartilage cells treated with IL-1beta, we observed that: (i) they had more marked effects on proteoglycan synthesis in chondrocytes than in cartilage samples; (ii) they could be roughly classified in the following rank order of potency: L-NAME > or = L-NMMA > > AG > SMT > > AETU and (iii) potentiation of the IL-1 effect by AETU was consistent with cytotoxicity in the millimolar range. D-isomers of L-arginine analog inhibitors (1000 microM) were unable to correct nitrite levels or proteoglycan synthesis in IL-1beta treated cells. L-arginine (5000 microM) tended to reverse the correcting effect of L-NMMA (1000 microM) on proteoglycan synthesis, thus suggesting a NO-related chondroprotective effect. However, data with L-NAME and SMT argued against a general inverse relationship between nitrite level and proteoglycan synthesis. Dexamethasone (0.1 to 100 microM) (i) failed to inhibit NO production in femoral head caps and chondrocytes beads whilst reducing it in patellae (50%) and (ii) did not affect or worsened the inhibitory effect of IL-1beta on proteoglycan synthesis. Such results suggested a corticosteroid-resistance of rat chondrocyte iNOS. Data from patellae supported a possible contribution of subchondral bone in NO production. In conclusion, our results suggest that (i) NO may account only partially for the suppressive effects of IL-1beta on proteoglycan synthesis, particularly in cartilage samples; (ii) the chondroprotective potency of NOS inhibitors can not be extrapolated from their effects on NO production by joint-derived cells and (iii) L-arginine analog inhibitors are more promising than S-substituted isothioureas for putative therapeutical uses.