Intercellular adhesion molecule-1 mediates the inhibitory effects of hyaluronan on interleukin-1beta-induced matrix metalloproteinase production in rheumatoid synovial fibroblasts via down-regulation of NF-kappaB and p38

OBJECTIVE: In rheumatoid arthritis (RA), it is well known that rheumatoid synovial fibroblasts (RSF) produce matrix metalloproteinases (MMPs) when stimulated with proinflammatory cytokines such as interleukin-1beta (IL-1beta), which causes joint destruction. We have previously shown that hyaluronan (HA) inhibits IL-1beta actions in RSF via CD44, the principal HA receptor. However, CD44 mediates HA effects only partially, and intracellular events after the HA binding to its receptors remain unclear. We investigated the role of intercellular adhesion molecule-1 (ICAM-1), another cell surface receptor for HA, and the intracellular signalling pathways in the actions of HA. METHODS: RSF were isolated from rheumatoid synovial tissues by enzymatic digestion and cultured in monolayers. The confluent cells were incubated for 48 h with IL-1beta, IL-1beta in the presence of HA, or IL-1beta in the presence of HA with pretreatment with anti-ICAM-1 antibody. Secretion of MMP-1 and MMP-3 was analysed by immunoblotting and immunofluorescence cytochemistry. Immunofluorescence cytochemistry was also performed to evaluate binding of HA to ICAM-1. The phosphorylation of nuclear factor (NF)-kappaB and mitogen-activated protein kinases (MAPKs) was analysed by immunoblotting. RESULTS: Production of MMP-1 and MMP-3 by RSF was stimulated by IL-1beta. HA at > or =2 mg/ml significantly inhibited MMP production induced by IL-1beta in a dose-dependent manner. Moreover, pretreatment with anti-ICAM-1 antibody at 50 mug/ml significantly blocked the effects of HA on the actions of IL-1beta on RSF, as shown by immunoblotting and immunofluorescence cytochemistry. Another immunofluorescence cytochemistry study demonstrated that HA bound RSF via ICAM-1. Inhibition studies revealed the requirement of NF-kappaB, p38 and c-jun NH2-terminal kinase (JNK) for IL-1beta-induced MMP production. IL-1beta activated all three pathways, whereas HA down-regulated their phosphorylation. Pretreatment with anti-ICAM-1 antibody reversed the inhibitory effects of HA on the activation of NF-kappaB and p38 without affecting JNK. CONCLUSION: HA suppresses IL-1beta-enhanced MMP-1 and MMP-3 synthesis in RSF via ICAM-1 through down-regulation of NF-kappaB and p38. Intra-articular injection of HA of high molecular weight may work through such a mechanism in RA joints.     

Inhibition of interleukin-1beta-stimulated production of matrix metalloproteinases by hyaluronan via CD44 in human articular cartilage

OBJECTIVE: To investigate the mechanism of the inhibitory action of hyaluronan (HA) on interleukin-1beta (IL-1beta)-stimulated production of matrix metalloproteinases (MMPs) in human articular cartilage. METHODS: IL-1beta was added to normal and osteoarthritic (OA) human articular cartilage in explant culture to stimulate MMP production. Articular cartilage was incubated or preincubated with a clinically used form of 800-kd HA to assess its effect on IL-1beta-induced MMPs. Levels of secreted MMPs 1, 3, and 13 in conditioned media were detected by immunoblotting; intracellular MMP synthesis in chondrocytes was evaluated by immunofluorescence microscopy. Penetration of HA into cartilage tissue and its binding to CD44 were analyzed by fluorescence microscopy using fluoresceinated HA. Blocking experiments with anti-CD44 antibody were performed to investigate the mechanism of action of HA. RESULTS: Treatment and pretreatment with 800-kd HA at 1 mg/ml resulted in significant suppression of IL-1beta-stimulated production of MMPs 1, 3, and 13 in normal and OA cartilage explant culture. Fluorescence histocytochemistry revealed that HA penetrated cartilage tissue and localized in the pericellular matrix around chondrocytes. HA-binding blocking experiments using anti-CD44 antibody demonstrated that the association of HA with chondrocytes was mediated by CD44. Preincubation with anti-CD44 antibody, which suppressed IL-1beta-stimulated MMPs, reversed the inhibitory effect of HA on MMP production that was induced by IL-1beta in normal and OA cartilage. CONCLUSION: This study demonstrates that HA effectively inhibits IL-1beta-stimulated production of MMP-1, MMP-3, and MMP-13, which supports the clinical use of HA in the treatment of OA. The action of HA on IL-1beta may involve direct interaction between HA and CD44 on chondrocytes.     

Hyaluronan inhibits IL-1β-stimulated collagenase production via down-regulation of phosphorylated p38 in SW-1353 human chondrosarcoma cells

We investigated the intracellular mechanism for the inhibitory effects of hyaluronan (HA) on interleukin-1beta (IL-1beta)-stimulated collagenase-1 and -3 (matrix metalloproteinases (MMPs)-1 and -13) production in a human chondrosarcoma cell line, SW-1353. MMPs-1 and -13 were induced by IL-1beta at 2 ng/ml in SW-1353 cells for 48 h. HA of 800 kDa, which is used clinically, significantly suppressed IL-1beta-stimulated production of MMPs-1 and -13 by immunoblotting. SW-1353 cells express the standard form of CD44 (CD44H), and immunofluorescent cytochemistry demonstrated the association of HA with CD44 on SW-1353 cells. Phosphorylated p38 (Phos-p38) mitogen-activated protein kinase was stimulated in SW-1353 cells by IL-1beta but not by HA alone. SB203580, a p38 MAPK inhibitor, partially blocked the MMP-1 and -13 production stimulated by IL-1beta. 800-kDa HA suppressed IL-1beta-activated Phos-p38 in a dose-dependent manner. CD44 blocking significantly reversed the inhibitory effects of HA on IL-1beta-activated Phos-p38 production. The present study clearly suggests that HA binds CD44 and inhibits IL-1beta-induced MMP-1 and -13 expression via down-regulation of Phos-p38 in SW-1353 cells.     

Effects of RWJ 67657, a p38 mitogen activated protein kinase (MAPK) inhibitor, on the production of inflammatory mediators by rheumatoid synovial fibroblasts

OBJECTIVE: To investigate the effect of the p38 mitogen activated protein kinase (MAPK) inhibitor RWJ 67657 on inflammatory mediator production by rheumatoid synovial fibroblasts (RSF). METHODS: RSF were pretreated with RWJ 67657 and stimulated with TNF alpha and/or IL-1 beta. Protein levels and mRNA expression of MMP-1, MMP-3, TIMP-1, IL-6, and IL-8 were determined, as was mRNA expression of COX-2 and ADAMTS-4. RESULTS: MMP-3 production was significantly inhibited at 1 microM RWJ 67657 and MMP-1 production at 10 microM, while TIMP-1 production was not inhibited. Inhibition of IL-6 and IL-8 protein production was seen at 0.1 microM RWJ 67657. Expression profiles of mRNA were in accordance with protein production. Inhibition of COX-2 mRNA expression occurred at 0.01 microM RWJ 67657. CONCLUSIONS: RWJ 67657 inhibits major proinflammatory mediator production in stimulated RSF at pharmacologically relevant concentrations. These findings could have important relevance for the treatment of rheumatoid arthritis.     

IL-1-mediated expression of membrane type matrix-metalloproteinase in rheumatoid osteoblasts

OBJECTIVE: To determine whether the activation of metalloproteinases can be achieved by the interaction of the inflammatory cytokine, IL-1 beta, with rheumatoid osteoblasts. RESULTS: MMP-2 is not secreted by rheumatoid osteoblasts as a proenzyme; however, IL-1 beta stimulation induced the secretion of MMP-2 as an active form from rheumatoid osteoblasts. This MMP-2 activating activity was detected significantly in IL-1 beta-stimulated rheumatoid osteoblasts. In support of this result, IL-1 beta stimulation induced the expression of membrane type matrix-metalloproteinase (MT-MMP), a newly-identified MMP-2 specific activator, in rheumatoid osteoblasts. CONCLUSION: These results suggest that IL-1 beta induces MMP-2 activation in part by up-regulating MT-MMP expression and represents a new mechanism for cytokine-mediated articular destruction in RA.     

Suppressive effects of hyaluronan on MMP-1 and RANTES production from chondrocytes

The purpose of the study was to examine the effects of hyaluronan (HA) on human chondrocytes in terms of production of MMP-1 and RANTES. Chondrocytes were obtained from patients with osteoarthritis (OA) or femoral neck fracture (control). Chondrocytes in monolayer culture were treated with various molecular weights (1.2, 50, 800 and 1,900 kD) of HA and then stimulated with IL-1. Production and expression of MMP-1 and RANTES were quantified by ELISA and real-time polymerase chain reaction (PCR). The response was blocked by anti-CD44 antibody. Production of MMP-1 was significantly suppressed by both 800- and 1,900-kD HA, while production of RANTES was suppressed by 1,900-kD HA. Expression of MMP-1 and RANTES mRNA was inhibited by 1,900-kD HA. Suppressive effects of HA on production of MMP-1 were canceled by treatment of anti-CD44 antibody. Higher CD44 expression was found in OA chondrocytes than in those of control. High-molecular-weight HA suppressed MMP-1 and RANTES production, mediated partly by CD44–HA interaction.     

Hyaluronan inhibits IL-1β-stimulated collagenase production via down-regulation of phosphorylated p38 in SW-1353 human chondrosarcoma cells

Abstract

We investigated the intracellular mechanism for the inhibitory effects of hyaluronan (HA) on interleukin-1β (IL-1β)-stimulated collagenase-1 and -3 (matrix metalloproteinases (MMPs)-1 and -13) production in a human chondrosarcoma cell line, SW-1353. MMPs-1 and -13 were induced by IL-1β at 2 ng/ml in SW-1353 cells for 48 h. HA of 800 kDa, which is used clinically, significantly suppressed IL-1β-stimulated production of MMPs-1 and -13 by immunoblotting. SW-1353 cells express the standard form of CD44 (CD44H), and immunofluorescent cytochemistry demonstrated the association of HA with CD44 on SW-1353 cells. Phosphorylated p38 (Phos-p38) mitogen-activated protein kinase was stimulated in SW-1353 cells by IL-1β but not by HA alone. SB203580, a p38 MAPK inhibitor, partially blocked the MMP-1 and -13 production stimulated by IL-1β. 800-kDa HA suppressed IL-1β-activated Phos-p38 in a dose-dependent manner. CD44 blocking significantly reversed the inhibitory effects of HA on IL-1β-activated Phos-p38 production. The present study clearly suggests that HA binds CD44 and inhibits IL-1β-induced MMP-1 and -13 expression via down-regulation of Phos-p38 in SW-1353 cells.     

The effects of interferon-beta treatment of synovial inflammation and expression of metalloproteinases in patients with rheumatoid arthritis

OBJECTIVE: Interferon-beta (IFNbeta) treatment is emerging as a potentially effective form of therapy in various immune-mediated conditions. This study evaluated the effects of IFNbeta therapy on the cell infiltrate, cytokine profile, and expression of metalloproteinase 1 (MMP-1) in synovial tissue from patients with rheumatoid arthritis (RA). To further assess the mechanism of action, in vitro experiments were conducted to determine the effects of IFNbeta on the production of MMP-1, MMP-3, tissue inhibitor of metalloproteinases 1 (TIMP-1), and prostaglandin E2 (PGE2) by human fibroblast-like synoviocytes (FLS). METHODS: Eleven patients were treated for 12 weeks with purified natural fibroblast IFNbeta (Frone; Ares-Serono, Geneva, Switzerland) subcutaneously 3 times weekly with the following dosages: 6 million IU (n = 4), 12 million IU (n = 3), and 18 million IU (n = 4). Synovial biopsy specimens were obtained by needle arthroscopy at 3 time points: directly before and at 1 month and 3 months after initiation of treatment. Immunohistologic analysis was performed using monoclonal antibodies specific for the following phenotypic markers and mediators of joint inflammation and destruction: CD3, CD38, CD68, CD55, tumor necrosis factor alpha (TNFalpha), interleukin-1beta (IL-1beta), IL-6, MMP-1, and TIMP-1. In addition, we measured the production of MMP-1, MMP-3, TIMP-1, and PGE2 by RA FLS and dermal fibroblasts in the presence and absence of IFNbeta. RESULTS: A statistically significant reduction in the mean immunohistologic scores for CD3+ T cells and the expression of MMP-1 and TIMP-1 at 1 month, CD38+ plasma cells and the expression of IL-6 at 3 months, and the expression of IL-1beta at both 1 and 3 months was observed in synovial tissue after IFNbeta treatment. The scores for CD68+ macrophages and TNFalpha expression also tended to decrease, but the differences did not reach statistical significance. The in vitro experiments revealed inhibition of MMP-1, MMP-3, and PGE2 production by RA FLS, whereas TIMP-1 production was only slightly decreased. These effects were more consistent in RA FLS than in dermal fibroblasts. CONCLUSION: The changes in synovial tissue after IFNbeta treatment and the in vitro data support the view that IFNbeta therapy has immunomodulating effects on rheumatoid synovium and might help to diminish both joint inflammation and destruction. Larger well-controlled studies are warranted to show the efficacy of IFNbeta treatment for RA.     

Establishment of nurse-like stromal cells from bone marrow of patients with rheumatoid arthritis: indication of characteristic bone marrow microenvironment in patients with rheumatoid arthritis.

OBJECTIVE: To investigate the microenvironment of bone marrow (BM) of patients with rheumatoid arthritis (RA). METHODS: Nurse cell-like BM stromal cell lines were established from BM mononuclear cells of patients with RA. We examined the various characteristics of these cell lines, including morphology, pseudoemperipolesis activity, cell surface markers, cytokine production and hyaluronan (HA) production. RESULTS: These RA BM nurse cell-like lines (RA-BMNC) were of mesenchymal origin and positive for CD44, CD54 and HLA-DR. They were defined as nurse cells because of pseudoemperipolesis activity that allowed lymphocytes to migrate underneath. RA-BMNC lines produced HA and multiple cytokines including interleukin (IL)-6, IL-7, IL-8 and granulocyte-macrophage colony-stimulating factor (GM-CSF). HA production by BM stromal cells was correlated with pseudoemperipolesis activity. RA-BMNC produced significantly higher levels of IL-6, IL-8 and GM-CSF by co-culture with lymphocytes. The cells also produced IL-1beta, G-CSF and tumour necrosis factor only when co-cultured with lymphocytes. The RA-BMNC maintained the growth of CD14+ myeloid cells unique to severe RA. CONCLUSION: The present results both indicate that RA-BMNC are nurse cells and suggest that they may play an important role in the pathogenesis of RA.     

Platelet-released growth factors enhance the secretion of hyaluronic acid and induce hepatocyte growth factor production by synovial fibroblasts from arthritic patients

OBJECTIVES: Autologous platelet-secreted growth factors (GFs) may have therapeutic effects in osteoarthritis (OA) capsular joints via multiple mechanisms. Our aim was to examine the effect of a platelet-derived preparation rich in growth factors (PRGFs) in OA synovial cell biology. METHODS: Synovial cells were isolated from 10 osteoarthritic patients and cultured in serum-free media (basal conditions) and exposed to either a platelet-poor preparation or PRGF for 72 h. Cells activated with interleukin-1beta (IL-1beta) for 48 h were also exposed to PRGF. Changes in several events relevant to joint homeostasis including (i) hyaluronic acid (HA) secretion, (ii) the balance between metalloproteinase-1, -3 and -13 (MMP-1, MMP-3 and MMP-13) and tissue inhibitor-1 (TIMP-1) and (iii) the secretion of transforming growth factor-beta1(TGF-beta1), vascular endothelial growth factor (VEGF) and hepatocyte growth factor (HGF), were all assessed. RESULTS: PRGF significantly enhanced HA secretion compared with platelet-poor preparations, P < 0.05; at the same time release of TIMP-1, MMP-1, MMP-3 and MMP-13 were not affected. An increased HGF production was observed (P < 0.05) but VEGF and TGF-beta1 levels remained unchanged. PRGF significantly enhanced the secretion of HA induced by IL-1beta activation, P < 0.05, but it did not modify the IL-1beta-induced rise in MMP-1, MMP-3 and VEGF. In contrast, PRGF-induced HGF production was abolished by the presence of IL-1beta during PRGF treatment, P < 0.05. CONCLUSIONS: Intra-articular administration of PRGF might be beneficial in restoring HA concentration and switching angiogenesis to a more balanced status but does not halt the effects of IL-1beta on synovial cells.     

Effects of hyaluronate on CD44 expression of infiltrating cells in exudate of rat air pouch, induced by sensitization with lipopolysaccharide.

OBJECTIVE: To investigate the effects of hyaluronate (HA) on CD44 expression of infiltrating cells in vivo. METHODS: Intra-air pouch injection of 10 microg lipopolysaccharide (LPS) with 0.4, 4.0, or 40 mg HA, 40 mg carboxymethylcellulose (CMC), or saline was performed on rats immunized with LPS. The percentage of CD44+ cells in the exudate of the air pouch was measured by flow cytometry, and the concentrations of interleukin 1beta (IL-1beta) and tumor necrosis factor-alpha (TNF-alpha) in the air pouch exudate were measured by ELISA. IL-1beta and TNF-alpha in the air pouch lining layer were stained by immunohistochemistry. RESULTS: The percentage of CD44+ cells in air pouch exudate was greater in the presence of HA, with a dose dependent increase (0.4 mg, 9.4+/-2.6%, n = 4; 4.0 mg, 13.8+/-2.9%, n = 4; 40 mg, 24.9+/-6.3%, n = 3; p < 0.05), while it was 4.9+/-1.2% (n = 4) in the presence of 40 mg CMC. The concentration of IL-1beta was lower in the presence of 40 mg HA (251.0+/-61.4 pg/ml, n = 4) or 40 mg CMC (168.2+/-43.5 pg/ml, n = 4; p < 0.05) than in saline (403.0+/-60.5 pg/ml, n = 4). The concentration of TNF-alpha was lower in the presence of 40 mg HA (14.0+/-6.7 pg/ml, n = 4) or 40 mg CMC (7.04+/-7.0 pg/ml, n = 4) than in saline (38.2+/-12.2 pg/ml, n = 4). Extensively stained lining cells in superficial layer of the air pouch with IL-1beta and TNF-alpha were observed in rats inoculated with 0.4 mg HA. CONCLUSION: These findings suggest that HA might affect the inflammatory process through modifying CD44 expression on infiltrating cells in air pouch exudate.     

Interleukin-4 inhibits interleukin-11 production by rheumatoid synovial cells

OBJECTIVE: To examine the effect of interleukin-4 (IL-4) on IL-11 production by rheumatoid synovial cells. METHODS: Freshly isolated rheumatoid synovial cells (FRS) were obtained by collagenase digestion of rheumatoid arthritis (RA) synovial tissue specimens taken at the time of operation. Rheumatoid synovial cells at four to eight passages were used as cultured rheumatoid synovial fibroblasts (RSF). IL-11 concentration was measured by ELISA. RESULTS: IL-4 inhibited the production of IL-11 by FRS in a dose-dependent manner. This inhibition was observed in FRS obtained from six patients, and the mean inhibition was 46.5%. The inhibitory effect of IL-4 on IL-11 production was cancelled by the addition of anti-IL-4 antibody. IL-4 also inhibited IL-11 production by IL-1alpha-stimulated cultured RSF. CONCLUSION: IL-4 inhibited IL-11 production by rheumatoid synovial cells. IL-4 has a protective effect on bone resorption. On the contrary, IL-11 participates in bone resorption via osteoclastogenesis. Therefore, IL-4 may exert its protective effect on bone resorption, at least in part, via inhibition of IL-11 production in rheumatoid joints.     

Hyaluronate-enhanced hematopoiesis: two different receptors trigger the release of interleukin-1beta and interleukin-6 from bone marrow macrophages.

The glycosaminoglycan hyaluronate (HA) is part of the extracellular environment in bone marrow. We show here that HA activates signal transduction cascades important for hemopoiesis. In myeloid and lymphoid long-term bone marrow cultures (LTBMC), treatment with hyaluronidase (HA'ase) results in reduced production of both progenitor and mature cells. Exogeneous HA added to LTBMC had the opposite effect: it enhanced hematopoiesis. The effect of HA is mediated through two different HA receptors on bone marrow macrophage-like cells, one of which is CD44 while the other is unknown. HA induces bone marrow macrophages to secrete IL-1beta (CD44-dependent) and IL-6 (CD44-independent). The two receptors address different signal transduction pathways: CD44 links to a pathway activating p38 protein kinase while the other yet unknown receptor induces Erk activity. There was no difference of the effect of HA and HA'ase on hematopoiesis in LTBMC and on cytokine production by macrophages in CD44-deficient mice compared with wild-type mice, indicating that the CD44 hyaluronate receptor and its signal transduction can be compensated for. Our data suggest a regulatory role for the extracellular matrix component HA in hematopoiesis and show the induction of signal transduction by HA receptors.     

Anti-interleukin-1 and anti-CD44 interventions producing significant inhibition of cartilage destruction in an in vitro model of cartilage invasion by rheumatoid arthritis synovial fibroblasts

OBJECTIVE: To establish an in vitro model for the investigation of destructive processes in rheumatoid arthritis (RA), to study the interaction between fibroblasts, macrophages, and chondrocytes, and to evaluate strategies to inhibit joint destruction in RA. METHODS: Human and bovine chondrocytes cultured in sponges pretreated with native bovine embryonic extracellular matrix produced a cartilaginous matrix reflected by the incorporation of 35S into proteoglycans. The 3-dimensional culture system was optimized for the number of chondrocytes (10(5) cells/sponge), the timing of 35S incorporation (day 21 after chondrocyte isolation), and the medium (20% fetal calf serum). RA synovial fibroblasts (RASF; 10(5)) were added, and the matrix destruction mediated by these RASF was monitored by the release of 35S. The system was modulated by the addition of monocytes (U937 cells), cytokines (interleukin-1beta [IL-1beta] and tumor necrosis factor alpha [TNFalpha]), interleukin-1 receptor antagonist (IL-1Ra), and monoclonal antibodies against IL-1beta and CD44. RESULTS: RASF destroyed the bovine cartilaginous matrix within 2 weeks (days 5-12) and the human cartilaginous matrix within 3 weeks (days 10-18). Compared with the effect of RASF alone (mean +/- SD 948 +/-180 counts per minute/week), the addition of U937 cells (a monocytic cell line), IL-1beta, or TNFalpha to the incubation medium increased the destruction of human cartilaginous matrix by at least 71% up to 90% (ranging from 1,618+/-204 cpm/week to 1,802+/-307 cpm/week). IL-1Ra and anti-IL-1beta monoclonal antibodies reduced the destruction of human matrix by 45% and 35%, respectively; this was partially reversed by the addition of U937 cells. The pretreatment of RASF with anti-CD44 monoclonal antibody (an adhesion molecule and receptor for hyaluronic acid) inhibited the destruction of the cartilaginous matrix by an average of 41% over 3 weeks. CONCLUSION: This model is envisioned to study distinct aspects of human destructive joint diseases under in vitro conditions and to replace and/or supplement animal experiments in basic research and drug testing. Based on the fact that proinflammatory cytokines enhance destruction whereas IL-1Ra and antibodies against IL-1beta and CD44 inhibit the process, it is concluded that anti-IL-1- and anti-CD44-directed therapies may help prevent cartilage destruction in RA.