Dynamic compression counteracts IL-1beta induced iNOS and COX-2 activity by human chondrocytes cultured in agarose constructs

*NO and PGE2 are inflammatory mediators derived from the inducible iNOS and COX enzymes and are potentially important pharmacological targets in OA. Both mechanical loading and IL-1beta will influence the release of *NO and PGE2. Accordingly, the current study examines the effect of dynamic compression on *NO and PGE2 release by human chondrocytes cultured in agarose constructs in the presence and absence of selective iNOS and COX-2 inhibitors. The current data demonstrate that IL-1beta induced nitrite and PGE2 release and inhibited [3H]-thymidine and 35SO4 incorporation. Inhibitor experiments indicate that 1400W and NS-398 either partially reversed or abolished IL-1beta induced nitrite and PGE2 release. IL-1beta induced inhibition of cell proliferation and proteoglycan synthesis was partially reversed with 1400W but was not influenced by NS-398. For the dynamic loading experiments, 1400W and NS-398 either reduced or abolished the compression-induced inhibition of *NO and PGE2 release in the presence of IL-1beta. The IL-1beta induced inhibition of cell proliferation was not influenced by 1400W or NS-398 whereas strain-induced stimulation of proteoglycan synthesis in the presence of IL-1beta was enhanced by 1400W. The data obtained using human chondrocytes demonstrate that IL-1beta induced *NO and PGE2 release via an iNOS-driven-COX-2 inter-dependent pathway. This response could be reversed by dynamic compression. These data indicate interactions exist between the NOS and COX pathways, a finding which will provide new insights in the development of pharmacological or biophysical treatments for cartilage disorders such as OA.     

Glutathione depletion inhibits IL-1 beta-stimulated nitric oxide production by reducing inducible nitric oxide synthase gene expression

L-buthionine-S,R-sulfoximine (BSO), an inhibitor of GSH synthesis, decreased IL-1 beta-induced nitrite release in rat islets and purified rat beta cells, nitrite formation and iNOS gene promoter activity in insulinoma cells, and iNOS mRNA expression in rat islets. The thiol depletor diethyl maleate (DEM) and an inhibitor of glutathione reductase 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) reduced IL-1 beta-stimulated nitrite release in islets. We conclude that GSH regulates IL-1 beta-induced NO production in islets, purified beta cells and insulinoma cells by modulation of iNOS gene expression.     

Dynamic compression inhibits fibronectin fragment induced iNOS and COX-2 expression in chondrocyte/agarose constructs

Mechanical loading and the fibronectin fragments (FN-fs) are known to stimulate the anabolic and catabolic processes in articular cartilage, possible through pathways mediated by ^·NO. This study examined the combined effects of dynamic compression and the NH₂-hep I or COOH-hep II FN-fs on the expression levels of iNOS and COX-2 and production of ^·NO and PGE₂ release. Both types of fragments induced iNOS and COX-2 expression and stimulated the production of ^·NO release. This response was inhibited by dynamic compression. Inhibitor experiments indicated that both dynamic compression and the iNOS inhibitor were important in restoring cell proliferation and proteoglycan synthesis in the presence of the FN-fs. This is the first study which demonstrates a downregulation of the FN-f-induced iNOS and COX-2 expression by dynamic compression. The combination of mechanical and pharmacological interventions makes this study a powerful tool to examine further the interactions of biomechanics and cell signalling in osteoarthritis.     

Mechanical compression influences intracellular Ca2+ signaling in chondrocytes seeded in agarose constructs.

Ca2+ signaling forms part of a possible mechanotransduction pathway by which chondrocytes may alter their metabolism in response to mechanical loading. In this study, a well-characterized model system utilizing bovine articular chondrocytes embedded in 4% agarose constructs was used to investigate the effect of physiological mechanical compressive strain applied after 1 and 3 days in culture. The intracellular Ca2+ concentration was measured by use of the ratiometric Ca2+ indicator indo 1-AM and confocal microscopy. A positive Ca2+ response was defined as a percent increase in Ca2+ ratio above a preset threshold. A significantly greater percentage of cells exhibited a positive Ca2+ response in strained constructs compared with unstrained controls at both time points. In strained constructs, treatment with either Ga3+ or EGTA significantly reduced the number of positive Ca2+ responders compared with untreated controls. These results represent an important step in understanding the physiological role of intracellular Ca2+ in chondrocytes under mechanical compression.     

Dynamic compression influences interleukin-1beta-induced nitric oxide and prostaglandin E2 release by articular chondrocytes via alterations in iNOS and COX-2 expression

Interleukin-1beta (IL-1beta) induces the release of nitric oxide (.NO) and prostaglandin E2 (PGE2) by chondrocytes and this effect can be reversed with the application of dynamic compression. Previous studies have indicated that integrins may play a role. In addition, IL-1beta upregulates the expression of iNOS and COX-2 mRNA via upstream activation of p38 MAPK. The current study examines the involvement of these pathways in mediating .NO and PGE2 release in IL-1beta stimulated bovine chondrocytes subjected to dynamic compression. Bovine chondrocytes were seeded in agarose constructs and cultured with 0 or 10 ng.ml(-1) IL-1beta with or without the application of 15% dynamic compressive strain at 1 Hz. Selected inhibitors were used to interrogate the role of alpha5beta1 integrin signalling and p38 MAPK activation in mediating the release of .NO and PGE2 in response to both IL-1beta and dynamic compression. The relative expression levels of iNOS and COX-2 were assessed using real-time quantitative PCR. Nitrite, a stable end product of .NO, was measured using the Griess assay and PGE2 release was measured using an enzyme immunoassay. IL-1beta enhanced .NO and PGE2 release and this effect was reversed by the application of dynamic compression. Co-incubation with an integrin binding peptide (GRGDSP) abolished the compression-induced effect. Real-time quantitative PCR analysis revealed that IL-1beta enhanced iNOS and COX-2 mRNA levels, with the maximum expression at 6 or 12 hours. Dynamic compression reduced this effect via a p38 MAPK sensitive pathway. These results suggest that dynamic compression acts to abrogate of .NO and PGE2 release by directly influencing the expression levels of iNOS and COX-2.     

Green tea polyphenol epigallocatechin-3-gallate inhibits the IL-1 beta-induced activity and expression of cyclooxygenase-2 and nitric oxide synthase-2 in human chondrocytes

We have previously shown that green tea polyphenols inhibit the onset and severity of collagen II-induced arthritis in mice. In the present study, we report the pharmacological effects of green tea polyphenol epigallocatechin-3-gallate (EGCG), on interleukin-1 beta (IL-1 beta)-induced expression and activity of cyclooxygenase-2 (COX-2) and inducible nitric oxide synthase (iNOS) in human chondrocytes derived from osteoarthritis (OA) cartilage. Stimulation of human chondrocytes with IL-1 beta (5 ng/ml) for 24 h resulted in significantly enhanced production of nitric oxide (NO) and prostaglandin E(2) (PGE(2)) when compared to untreated controls (p <.001). Pretreament of human chondrocytes with EGCG showed a dose-dependent inhibition in the production of NO and PGE(2) by 48% and 24%, respectively, and correlated with the inhibition of iNOS and COX-2 activities (p <.005). In addition, IL-1 beta-induced expression of iNOS and COX-2 was also markedly inhibited in human chondrocytes pretreated with EGCG (p <.001). Parallel to these findings, EGCG also inhibited the IL-1 beta-induced LDH release in chondrocytes cultures. Overall, the study suggests that EGCG affords protection against IL-1 beta-induced production of catabolic mediators NO and PGE(2) in human chondrocytes by regulating the expression and catalytic activity of their respective enzymes. Furthermore, our results also indicate that ECGC may be of potential therapeutic value for inhibiting cartilage resorption in arthritic joints.     

Divergent stress responses to IL-1beta, nitric oxide, and tunicamycin by chondrocytes

As the only cell in cartilage responsible for matrix synthesis, the chondrocyte's viability is crucial to healthy tissue. It must tolerate stresses from both mechanical and cellular sources. This study examines the endoplasmic reticulum (ER) stress response in chondrocytes after exposure to IL-1beta, nitric oxide, or tunicamycin in order to determine whether this form of stress causes cell death. Cultures of the immortalized human juvenile costal chondrocyte cell line, C-28/I2, were treated with IL-1beta, S-nitroso-N-acetylpenicillamine (SNAP), and tunicamycin. Increasing intracellular nitric oxide levels by SNAP treatment or inhibiting protein folding in the ER lumen by tunicamycin induced the ER stress response as evidenced by increased protein and gene expression of GADD153 as well as PERK and eIF2-alpha phosphorylation, and resulted in apoptosis. IL-1beta treatment induced PERK and eIF2-alpha phosphorylation, but not GADD153 expression or apoptosis. The ER stress signaling pathway of IL-1beta involved iNOS because blocking its expression, inhibited ER stress gene expression. Therefore, inducing the ER stress response in chondrocytes results in divergent responses depending on the agent used. Even though IL-1beta, a common proinflammatory cytokine, induces the ER stress response, it is not proapoptotic to chondrocytes. On the other hand, exposure to high levels of intracellular nitric oxide induce chondrocyte apoptosis as part of the ER stress response.     

Acetylcholine inhibits nitric oxide (NO) synthesis in the gastropod nervous system

Acetylcholine (ACh) is one of the main signals regulating nitric oxide synthase (NOS) expression and nitric oxide (NO) biosynthesis in mammals. However, few comparative studies have been performed on the role of ACh on NOS activity in non-mammalian animals. We have therefore studied the cholinergic control of NOS in the snail Helix pomatia and compared the effects of ACh on NO synthesis in the enteric nervous system of the snail and rat. Analyses by the NADPH-diaphorase reaction, immunocytochemistry, purification with ion-exchange chromatography, Western-blot, and quantitative polymerase chain reaction have revealed the expression of neuronal NOS in the rat intestine and of a 60-kDa subunit of NOS in the enteric nerve plexus of H. pomatia. In H. pomatia, quantification of the NO-derived nitrite ions has established that NO formation is confined to the NOS-containing midintestine. Nitrite production can be elevated by L-arginine but inhibited by N_ω-nitro-L-arginine. In rats, ACh moderately elevates nitrite production, whereas ACh, the nicotinic receptor agonists (nicotine, acetyl thiocholine iodide, metacholine) and the cholinesterase inhibitor eserine reduce enteric nitrite formation in snails. The nicotinic receptor antagonist tubocurarine also provokes nitrite liberation, whereas the muscarinic receptor agonists or antagonists have no significant effect in snails. In the presence of EDTA or tetrodotoxin, ACh fails to inhibit nitrite production. In pharmacological studies, we have found that ACh contracts the midintestinal muscles and, in snails, simultaneously reduces the antagonistic muscle relaxant effect of L-arginine. Our experiments provide the first evidence for an inhibitory regulation of neuronal NO synthesis by ACh in an invertebrate species. This article is dedicated to Dr. Gábor Hollósi on the 50th anniversary of his graduation and being a teacher at the University of Debrecen.     

Role of nitric oxide in D-galactosamine-induced cell death and its protection by PGE1 in cultured hepatocytes.

Prostaglandin E(1) (PGE(1)) reduces cell death in experimental and clinical manifestations of liver dysfunction. Nitric oxide (NO) has been shown to exert a protective or noxious effect in different experimental models of liver injury. The aim of the present study was to investigate the role of NO during PGE(1) protection against D-galactosamine (D-GalN) citotoxicity in cultured hepatocytes. PGE(1) was preadministered to D-GalN-treated hepatocytes. The role of NO in our system was assessed by iNOS inhibition and a NO donor. Different parameters related to apoptosis and necrosis, NO production such as nitrite+nitrate (NO(x)) release, iNOS expression, and NF-kappaB activation in hepatocytes were evaluated. The inhibition of iNOS reduced apoptosis induced by D-GalN in hepatocytes. PGE(1) protection against D-GalN injury was associated with its capacity to reduce iNOS expression and NO production induced by D-GalN. Nevertheless, iNOS inhibition showed that protection by PGE(1) was also mediated by NO. Low concentrations of a NO donor reduced D-GalN injury with a decrease in the extracellular NO(x) concentration. High concentrations of the NO donor enhanced NO(x) concentration and increased cell death by D-GalN. The present study suggests that low NO production induced by PGE(1) preadministration reduces D-GalN-induced cell death through its capacity to reduce iNOS expression and NO production caused by the hepatotoxin.     

Modulation of PGE(2) and TNFalpha by nitric oxide and LPS-activated RAW 264.7 cells

Prostaglandins (PGs), the arachidonic acid (AA) metabolites of the cyclooxygenase (COX) pathway, and the cytokine TNFalpha play major roles in inflammation and they are synthesised mainly by macrophages. Their syntheses have been shown to be regulated by several factors, including nitric oxide, a further important macrophage product. Since both positive and negative regulations of PGs and TNFalpha synthesis by NO have been reported, we sought to understand the mechanisms underlying these opposite NO effects by using a recent class of NO releasing compounds, the NONOates, which have been shown to release NO in a controlled fashion. To this aim, we analysed the effect of NO released from PAPA/NO (t1/2 15 min) and DETA/NO (t1/2 20 h) in RAW 264.7 cells. Both NONOates were used at the same concentrations allowing the cell cultures to be exposed either at high levels of NO for brief time (PAPA/NO) or at low levels of NO for long time (DETA/NO). We found that the two NONOates had opposite effect on basal TNFalpha release, being increased by PAPA/NO and decreased by DETA/NO, while they did not affect the release stimulated by LPS. At variance, both NONOates increased the basal PGE(2) production, while the LPS-stimulated production was slightly increased only by PAPA/NO. The modulation of PGE(2) synthesis was the result of the distinct effects of the two NO-donors on either arachidonic acid (AA) release or cyclooxygense-2 (COX-2) expression, the precursor and synthetic enzyme of PGs, respectively. Indeed, in resting cultures AA release was enhanced only by PAPA/NO whereas COX-2 expression was moderately upregulated by both donors. In LPS activated cells, both NONOates induced AA release, although with different kinetics and potencies, but only DETA/NO significantly increased COX-2 expression. In conclusion, by comparing the activities of these two NONOates, our observations indicate that level and time of exposure to NO are both crucial in determining the molecular target and the final result of the interactions between NO and inflammatory molecules.     

Interleukin-10 (IL-10) inhibits the induction of nitric oxide synthase by interferon-gamma in murine macrophages.

A murine macrophage cell line, J774, expresses high levels of the enzyme nitric oxide synthase (NOS) and produces large amounts of nitric oxide (NO) when activated with recombinant interferon (IFN)-gamma and a low concentration of LPS (10 ng/ml). Both the expression of NOS and the production of NO were inhibited by recombinant IL-10 in a dose-dependent manner. The inhibition was effective only when the cells were pretreated with IL-10; addition of IL-10 at the same time or after IFN-gamma activation was without effect. These results demonstrate that IL-10, a product of Th2 (helper T lymphocyte 2) cells, can antagonise the function of IFN-gamma, a product of Th1 cells, by modulating the mechanism of synthesis of nitric oxide in the macrophages.     

Anti-inflammatory effects of IL-4 and dynamic compression in IL-1beta stimulated chondrocytes

Mechanical loading can counteract inflammatory pathways induced by IL-1beta by inhibiting *NO and PGE2, catabolic mediators known to be involved in cartilage degradation. The current study investigates the potential of dynamic compression, in combination with the anti-inflammatory cytokine, IL-4, to further abrogate the IL-1beta induced effects. The data presented demonstrate that IL-4 alone can inhibit nitrite release in the presence and absence of IL-1beta and partially reverse the IL-1beta induced PGE2 release. When provided in combination, IL-4 and dynamic compression could further abrogate the IL-1beta induced nitrite and PGE2 release. IL-1beta inhibited [3H]thymidine incorporation and this effect could be reversed by IL-4 or dynamic strain alone or both in combination. By contrast, 35SO4 incorporation was not influenced by IL-4 and/or dynamic strain in IL-1beta stimulated constructs. IL-4 and mechanical loading may therefore provide a potential protective mechanism for cartilage destruction as observed in OA.     

Hyaluronan fragments activate nitric oxide synthase and the production of nitric oxide by articular chondrocytes

Chondrocyte CD44 receptors anchor hyaluronan to the cell surface, enabling the assembly and retention of proteoglycan aggregates in the pericellular matrix. Hyaluronan-CD44 interactions also provide signaling important for maintaining cartilage homeostasis. Disruption of chondrocyte-hyaluronan contact alters CD44 occupancy, initiating alternative signaling cascades. Treatment with hyaluronan oligosaccharides is one approach to uncouple CD44 receptors from its native ligand, hyaluronan. In bovine articular chondrocytes, treatment with hyaluronan oligosaccharides or purified hyaluronan hexasaccharides induced the production of nitric oxide that mirrored nitric oxide production following interleukin-1 treatment. In contrast, 120 and 1,260 kDa hyaluronan did not induce production of nitric oxide. Human chondrocytes responded similarly to treatment with hyaluronan or hyaluronan oligosaccharides. Nitric oxide production from chondrocytes was mediated by activation of the inducible nitric oxide synthase, as confirmed by mRNA expression and inhibition of nitric oxide production by diphenyleneiodonium. Co-treatment of chondrocytes with hyaluronan oligosaccharides and interleukin-1 did not demonstrate additive effects. Blocking interleukin-1 receptors with an antagonist did not abolish the production of nitric oxide induced by treatment with hyaluronan oligosaccharides. Moreover, only COS-7 following transfection with a pCD44, not the CD44-null parental cells, responded to treatment with hyaluronan oligosaccharides by releasing nitric oxide. This study demonstrates a novel signaling potential by hyaluronan fragments, in lieu of endogenous hyaluronan-chondrocyte interactions, resulting in the activation of inducible nitric oxide synthase.     

Effect of pericellular matrix formation by chondrocytes cultured in agarose gel on the viscoelastic properties of agarose gel matrix

The viscoelasticity of chondrocyte-seeded agarose gel (AGC0) and that of chondrocyte-seeded agarose gel after 21 days of cultivation (AGC3) were investigated. In AGC3, pericellular matrix (PCM)-like material around each chondrocyte was found to be constructed, which was confirmed by an optical micrograph in conjunction with toluidine blue staining. The relaxation modulus of each of the chondrocyte-agarose gel composite systems was measured by a non-constrained indentation method. Stress-strain curves for all of the specimens examined had a toe region followed by a linear region terminated by specimen fracture. The slope of the linear region of AGC0 was smaller than that of AG, while the SS curve of AGC0 was indistinguishable from that of AGC3. All of the relaxation curves studied were typical of gels, having a fast relaxation process up to 10³ s followed by a plateau. The relaxation modulus of AGC0 was smaller than that of agarose gel (AG), the decrement in relaxation modulus from AG to AGC0 being attributed to the seeding of chondrocytes that have a smaller modulus than that of agarose gels. However, the relaxation modulus of AGC3 was increased at the early viscoelastic region in particular, as compared with that of AGC0. The increments in the relaxation modulus in AGC3 were attributed to the PCM-like material produced by chondrocytes, where the produced material may provide crosslink points and reinforce the agarose gel.