Abstract

Although there have been major advances in the treatment of rheumatoid arthritis with the advent of biological agents, the mechanisms that drive cytokine production and sustain disease chronicity remain unknown. Tenascin-C (encoded by Tnc) is an extracellular matrix glycoprotein specifically expressed at areas of inflammation and tissue damage in inflamed rheumatoid joints. Here we show that mice that do not express tenascin-C show rapid resolution of acute joint inflammation and are protected from erosive arthritis. Intra-articular injection of tenascin-C promotes joint inflammation in vivo in mice, and addition of exogenous tenascin-C induces cytokine synthesis in explant cultures from inflamed synovia of individuals with rheumatoid arthritis. Moreover, in human macrophages and fibroblasts from synovia of individuals with rheumatoid arthritis, tenascin-C induces synthesis of proinflammatory cytokines via activation of Toll-like receptor 4 (TLR4). Thus, we have identified tenascin-C as a novel endogenous activator of TLR4-mediated immunity that mediates persistent synovial inflammation and tissue destruction in arthritic joint disease.

Introduction

Rheumatoid arthritis is characterized by synovial inflammation and destruction of joint cartilage and bone mediated by persistent synthesis of proinflammatory cytokines and matrix metalloproteinases. Biological compounds that suppress the synthesis of key inflammatory cytokines, including tumor necrosis factor- (TNF-) and interleukin-6 (IL-6), are successful at treating rheumatoid arthritis in the short term. However, repeated treatments are required, rendering this an expensive therapeutic approach that does not lead to long-term remission. Total systemic suppression of cytokine function is not without inherent problems, such as increased infectious risk. Thus, despite advances in care, there remains an unmet need for an economical mode of treatment that is efficacious over the long term^{1, 2}. However, a major obstacle lies in the fact that the mechanisms that underpin disease chronicity remain unclear, and the factor(s) that drive the prolonged expression of inflammatory and destructive mediators are unknown.

TLRs have a key role in driving inflammation, and blockade of TLR function may yield considerable clinical benefit^{3, 4}. This receptor family mediates host defense against infection and injury by recognizing pathogen- and damage-associated molecular patterns (PAMPs and DAMPs)⁵. DAMPs are endogenous proinflammatory molecules generated upon tissue injury and include intracellular molecules released from necrotic cells, extracellular matrix fragments or extracellular matrix molecules upregulated upon injury^{6, 7}. Upon activation, TLRs promote innate and adaptive immune responses, including induction of proinflammatory cytokines and matrix metalloproteinases⁸. TLRs are highly expressed in synovial tissue from individuals with rheumatoid arthritis^{9, 10, 11, 12}, and mice with targeted deletions or loss-of-function mutations in Tlr4 are protected from experimental arthritis^{13, 14}. Furthermore, inhibitors of TLR4 have reduced destructive arthritis in mice¹⁵ and improved symptoms in patients with moderate to severe rheumatoid arthritis in a preliminary phase 1 trial¹⁶.

However, it is unclear which TLR ligand(s) are involved in disease pathogenesis. Tenascin-C is an extracellular matrix glycoprotein associated with tissue injury and repair. It is not normally expressed in most adult tissues but is specifically and transiently upregulated during acute inflammation and persistently expressed in chronic inflammation¹⁷. Little tenascin-C is expressed in normal human joints, but expression is increased in synovia^{18, 19, 20}, synovial fluid^{21, 22} and cartilage^{20, 21} from individuals with rheumatoid arthritis. It is a hexameric protein of 1.5 million Da that comprises an assembly domain, epidermal growth factor–like repeats (EGF-L), fibronectin type III–like repeats (TNIII) and a fibrinogen-like globe (FBG)²³. The elevated amounts of tenascin-C in rheumatoid arthritis and the homology of tenascin-C domains to other known DAMPs prompted us to examine whether this molecule is a TLR ligand that contributes to the progression of inflammatory joint disease.

Top of page

---

Results

Zymosan-induced inflammation is not sustained in Tnc^-/- mice

We used zymosan to induce acute synovitis in mice. Wild-type mice showed rapid paw swelling that reached maximal diameter by 24 h and that was maintained for a further 24 h (Fig. 1a). After 2 d, the paw diameter decreased, but paws remained swollen at 4 d (Fig. 1a). There was no significant difference in the paw swelling of Tnc^-/- mice compared to wild-type mice 24 h after injection (Fig. 1a). However, swelling subsided faster than in wild-type mice; paw diameter was significantly lower at 2 d and had declined to near basal levels by 4 d (Fig. 1a). By 4 d, the paws of wild-type mice were still visibly swollen and red, whereas the paws of Tnc^-/- mice resembled noninjected paws (Supplementary Fig. 1).

Figure 1: Accelerated resolution of acute inflammation in Tnc^-/- mice.

(a) Paw swelling in wild-type (+/+) and Tnc^-/- (-/-) mice over time after injection of zymosan. Data are shown as the mean increase in paw diameter compared to paw diameter before injection s.e.m. (n = 24 mice per genotype). ^**P < 0.01. (b–e) Representative sections of the ankle joints from wild-type (b,c) and Tnc^-/- (d,e) mice 4 d after zymosan injection, stained with H&E (b,d) and safranin-O (c,e). Boxes highlight the joint synovium (s) and cartilage proteoglycan (cp). Scale bar, 250 m. (f) Quantification of joint inflammation and chondrocyte death in knee joints 4 d after injection with zymosan from wild-type mice and Tnc^-/- mice. Data are expressed as the mean s.d. (n = 24 mice per genotype). ^*P < 0.05, ^**P < 0.01.

Full size image (57 KB)

Histologically, synovia of wild-type mice were substantially inflamed at 4 d, showing cellular infiltration and cartilage proteoglycan loss (Fig. 1b,c). In contrast, Tnc^-/- mice showed no synovitis, cellular infiltrate or cartilage proteoglycan loss (Fig. 1d,e). There was little cellular exudate in either wild-type or Tnc^-/- mice; however, cellular infiltration was significantly lower in Tnc^-/- mice (Fig. 1f). No erosion of cartilage or bone occurred in mice of either genotype (data not shown); however, chondrocyte death occurred in wild-type but not Tnc^-/- mice (Fig. 1f). Tenascin-C expression was elevated in the joints of wild-type mice 24 h after injection, and expression increased over time up to 4 d (Supplementary Fig. 2a). Induction of tenascin-C expression was specific to the zymosan-injected joint and was not observed in the contralateral, noninjected joint of the same mouse (data not shown). Thus, tenascin-C expression seems to promote the maintenance of acute inflammation.

Tnc^-/- mice are protected from joint destruction

To determine whether tenascin-C contributes to more destructive inflammatory joint disease, we induced erosive arthritis by immunization and intra-articular injection with methylated BSA (mBSA) (Fig. 2), a model that induces pathological changes similar to human rheumatoid arthritis²⁴. Cell infiltration and synovial thickening was apparent by 24 h in mice of both genotypes (Fig. 2c–f,h,i) compared to sham-injected (Fig. 2a,b,g) or noninjected (data not shown) mice.

Figure 2: Synovial inflammation is induced in Tnc^-/- mice upon injection of antigen.

(a,b) Representative sections of the knee joints of sham-injected wild-type mice. (c–f) Representative sections of the knee joints of wild-type (c,d) or Tnc^-/- (e,f) mice 24 h after intra-articular injection of mBSA. Inflammatory cell infiltration is indicated in the capsule (cap), meniscus (M) and the joint space (J) of both wild-type and Tnc^-/- mice. (g–i) Higher magnification images show the synovium of sham-injected wild-type mice (g), wild-type mice injected with mBSA (h) or Tnc^-/- mice injected with mBSA (i). 'S' indicates the healthy synovium of sham-injected mice that is no more than 1–3 cells thick along the entire bone surface, and 'ST' highlights the synovia of wild-type and Tnc^-/- mice, which are both considerably thickened. Sections are stained with H&E (a, c, e, g–i) and safranin-O (b,d,f). Scale bars, 250 m (a–f) and 50 m (g–i). n = 5 mice per genotype.

Full size image (97 KB)

However, inflammation did not persist in Tnc^-/- mice as it did in wild-type mice (Fig. 3). Wild-type mice showed increased inflammation of the meniscus and capsule, synovial hyperplasia, cell accumulation and fibrin deposits in the joint space, pannus formation, and localized cartilage proteoglycan loss 3 d after injection (Fig. 3a,b,f). In contrast, in Tnc^-/- mice, inflammation was limited to the capsule, synovial inflammation had subsided, and there were no fibrin or cell aggregates present in the joint space, as well as no pannus formation and no proteoglycan loss (Fig. 3c–e).

Figure 3: Synovial inflammation subsides rapidly in Tnc^-/- mice.

(a–f) Representative sections of the knee joints of wild-type (a,b,f) or Tnc^-/- (c,d,e) mice 3 d after intra-articular injection of mBSA. The line in a and c highlights increased inflammation of the capsule in wild-type mice compared to Tnc^-/- mice. 'cp' highlights increased cartilage proteoglycan loss in wild-type mice compared to Tnc^-/- mice. In e and f, substantial synovial hyperplasia is indicated by the line, cell and fibrin deposits in the joint space are indicated by the arrow, and pannus invasion is indicated by arrowheads. Sections are stained with H&E (a,c,e,f) and safranin-O (b,d). Scale bars, 250 m (a–d) and 100 m (e–f). n = 5 mice per genotype.

Full size image (95 KB)

By 7 d, wild-type mice showed persistent inflammatory cell infiltration and joint space exudate, extensive synovitis and pannus formation and destruction of articular cartilage and bone erosion (Fig. 4a,b). However, Tnc^-/- mice had healthy joints that showed only mild inflammatory cell infiltration, with no joint space exudate, synovitis, pannus formation, destruction of articular cartilage or bone erosion (Fig. 4c,d).

Figure 4: Tnc^-/- mice are protected from tissue destruction, and tenascin-C induces cytokine synthesis in primary human cells.

(a–d) Representative sections of the knee joints of wild-type (a,b) and Tnc^-/- (c,d) mice 7 d after intra-articular injection of mBSA, stained with H&E (a,c) and safranin-O (b,d). Scale bar, 250 m. n = 24 mice per genotype. Arrowhead indicates area of bone erosion. Arrow indicates pannus invasion into articular cartilage. 'J' indicates the joint space, and 'AC' indicates the intact articular cartilage. (e) Quantification of knee joint inflammation, chondrocyte death, cartilage surface erosion and bone erosion in wild-type mice (+/+) and Tnc^-/- mice (-/-). Data represent the mean s.d. n = 5 per genotype (24 h, 3 d) or 24 per genotype (7 d). Cartilage surface erosion and bone erosion are shown at 7 d only. (f,g) IL-6, IL-8 and TNF synthesis by primary human macrophages (f) and synovial fibroblasts (g) that were either unstimulated (no addition) or stimulated with LPS (1 ng ml^-1 (f) or 10 ng ml^-1 (g) or recombinant tenascin-C (1.0 M–1.0 nM) for 24 h. Data are shown as the mean of triplicate values s.d. from one of three representative experiments. (h) IL-6 synthesis by primary human macrophages that were either unstimulated (no addition) or stimulated with LPS (1 ng ml^-1) or recombinant tenascin-C (1.0 M) for 24 h. Cells were preincubated with medium alone (-) or with polymyxin B (P) before stimulation, or they were incubated with heat-denatured LPS or tenascin-C (H). Data are shown as the mean of triplicate values s.d. from one of three representative experiments. ^*P < 0.05, ^**P < 0.01.

Full size image (97 KB)

These histological data are reflected in the scoring of joint disease. Cellular infiltrate and exudate were not significantly different in wild-type and Tnc^-/- mice 24 h after injection (Fig 4e). Cellular mass continued to increase in wild-type mice over time but was attenuated in Tnc^-/- mice (Fig. 4e). Increasing chondrocyte death and cartilage and bone erosion occurred in wild-type mice but was not present in Tnc^-/- mice (Fig. 4e). These data indicate that whereas the initiation of joint inflammation is unaffected in Tnc^-/- mice, disease does not progress to tissue destruction and cell death. Tenascin-C expression was elevated in the joints of wild-type mice 24 h and 3 d after injection and was maintained up to 7 d (Supplementary Fig. 2b). Induction of tenascin-C expression was specific to the mBSA-injected joint and was not observed in the contralateral, noninjected joint of the same mouse data (not shown). These results indicate that expression of tenascin-C is required for persistent synovial inflammation and joint destruction in this model.

Tenascin-C induces cytokine synthesis in primary human cells

We next investigated whether tenascin-C can activate the innate immune response. Exogenously added tenascin-C stimulated proinflammatory cytokine synthesis in primary human macrophages and synovial fibroblasts in a cell type–specific manner that was markedly different from that of bacterial lipopolysaccharide (LPS) (Fig. 4f,g). Tenascin-C induced tumor necrosis factor- (TNF-), interleukin-6 (IL-6) and IL-8 production in a dose-dependent fashion in human macrophages (Fig. 4f). However, tenascin-C induced only IL-6 synthesis in synovial fibroblasts, whereas LPS induced both IL-6 and IL-8 (Fig. 4g). Neither LPS nor tenascin-C induced TNF- synthesis in fibroblasts (data not shown).

Tenascin-C–mediated stimulation of IL-6 (Fig. 4h), IL-8 and TNF- production by human macrophages and IL-6 by synovial fibroblasts (data not shown) was heat sensitive and unaffected by the LPS inhibitor polymyxin B. Together, these results provide strong evidence that cytokine induction by tenascin-C is not due to LPS contamination.

The FBG domain of tenascin-C mediates cell activation

Tenascin-C is a large molecule, each domain of which binds to a different cell surface receptor²³. To identify which domains are crucial for promoting cytokine production, we synthesized recombinant proteins comprising the various domains of the molecule (Supplementary Figs. 3 and 4). Each domain preparation contained <10 pg ml^-1 LPS. Only one domain of tenascin-C was active. The FBG domain stimulated TNF- (Fig. 5a), IL-6 and IL-8 (data not shown) synthesis in human macrophages and IL-6 synthesis in synovial fibroblasts (data not shown) to an equal extent to full-length tenascin-C. Like full-length tenascin-C, FBG did not induce IL-8 synthesis in synovial fibroblasts, whereas LPS did (data not shown). FBG-induced cytokine synthesis was also heat sensitive and unaffected by polymyxin B (data not shown).

Figure 5: The FBG domain of tenascin-C mediates stimulation of cytokine synthesis in vivo and in vitro.

(a) TNF synthesis by primary human macrophages that were either unstimulated (no addition) or stimulated with LPS (1 ng ml^-1), 1.0 M recombinant tenascin-C (TNC) or 1.0 M tenascin-C domains (assembly domain (TA), EGF-L, TNIII 1–5, TNIII 1–3, TNIII 3–5, TNIII 5–7, TNIII 6–8 and FBG) for 24 h. Data are shown as the mean of triplicate values s.d. from one of three representative experiments. (b) IL-6, IL-8 and TNF synthesis by synovial membrane cells that were either unstimulated (no addition) or stimulated with LPS (10 ng ml^-1) or recombinant FBG (1.0–0.01 M) for 24 h. Data are shown as the mean percentage change in cytokine abundance compared to unstimulated cells s.e.m. from five different subjects. (c–h) Representative sections of the knee joints of wild-type mice 3 d after intra-articular injection of PBS (c,e,g) or 1 g FBG (d,f,h). Sections are stained with H&E (c–f) or safranin-O (g,h). Scale bar, 250 m (c,d) and 100 m (e–h) (n = 5 mice per genotype). (i) Quantification of joint inflammation, bone erosion, cartilage surface erosion and chondrocyte death in the knee joints of wild-type mice 3 d after intra-articular injection of PBS (PBS) or 1 g FBG (FBG). Data represent the mean s.d. (n = 5 per genotype). ^*P < 0.05, ^**P < 0.01.

Full size image (97 KB)

We also investigated whether FBG could promote expression of inflammatory cytokines in synovial membranes from individuals with rheumatoid arthritis. This tissue model of rheumatoid arthritis (comprising a mixed population of all synovial cell types) spontaneously produces high amounts of IL-6, IL-8 and TNF-²⁵ (Fig. 5b). FBG further enhanced synthesis of all of these cytokines (Fig. 5b).

FBG induces joint inflammation in mice in vivo

To determine whether FBG could induce inflammation in vivo, we injected wild-type mice intra-articularly with FBG. We observed a dose-dependent stimulation of joint inflammation (Fig. 5). No inflammation or proteoglycan loss occurred in noninjected mice or in mice injected with PBS (Fig. 5c,e,g) or 100 ng FBG (data not shown). However, injection of 1 g FBG induced inflammatory cell infiltration (Fig. 5d), mild synovitis, pannus formation (Fig. 5f) and proteoglycan loss (Fig. 5h). We observed a similar response upon injection with 3 g FBG (data not shown).

Upon histological quantification, we found high levels of cellular infiltrate and exudate and chondrocyte death, together with a modest amount of cartilage and bone damage in mice injected with FBG (Fig. 5i).

FBG-mediated cytokine synthesis is dependent on Myd88

Many DAMPs have been shown to activate TLRs. Therefore, we investigated whether TLRs might also mediate tenascin-C–induced cytokine production. Myeloid differentiation factor-88 (MyD88) is required for signaling by all TLRs except TLR3 (ref. 4). Infection of synovial fibroblasts with adenovirus expressing dominant-negative MyD88 abolished FBG induction of IL-6 (Fig. 6a), suggesting that FBG action is dependent on functional MyD88. FBG action is not mediated by IL-1, as addition of IL-1 receptor antagonist did not inhibit induction of cytokines (data not shown).

Figure 6: FBG-mediated cytokine synthesis is MyD88 and TLR4 dependent.

(a) IL-6 synthesis by synovial fibroblasts that were either uninfected or infected with adenovirus expressing GFP (AdGFP) or dominant-negative MyD88 (AdMyD88dn). Cells were unstimulated, stimulated with LPS (10 ng ml^-1) or stimulated with FBG (1 M) for 24 h. (b) IL-6 synthesis by mouse embryonic fibroblasts isolated from wild-type (Myd88^+/+) or Myd88^-/- mice that were either unstimulated (-) or stimulated with PAM3 (100 ng ml^-1), LPS (100 ng ml^-1), TNF- (100 ng ml^-1), IL-1 (5 ng ml^-1) or FBG (1 M) for 24 h. (c) TNF synthesis by macrophages that were preincubated with medium alone, medium containing function-blocking antibodies to TLR2 (10 g ml^-1) or TLR4 (25 g ml^-1) or isotype control antibodies (25 g ml^-1) for 30 min before stimulation. Cells were unstimulated or stimulated with LPS (1 ng ml^-1), FBG (1 M) or PAM3 (10 ng ml^-1) for 24 h. (d) IL-6 synthesis by mouse embryonic fibroblasts isolated from wild-type, Tlr2^-/- or Tlr4^-/- mice that were either unstimulated or stimulated with PAM3 (100 ng ml^-1), LPS (100 ng ml^-1), IL-1 (5 ng ml^-1) and FBG (1 M) for 24 h. (e) IL-6 synthesis by bone marrow–derived macrophages isolated from wild-type, Tlr2^-/- or Tlr4^-/- mice that were either unstimulated or stimulated with PAM3 (100 ng ml^-1), LPS (100 ng ml^-1) or FBG (1 M) for 24 h. (f) TNF synthesis by human macrophages that were preincubated with no inhibitor, 1 g ml^-1 msbB LPS or 10 g ml^-1 antibodies to CD14 for 30 min before stimulation with LPS (1 ng ml^-1), FBG (1 M) or PAM3 (10 ng ml^-1) for 24 h. In all panels, data are shown as the mean of three independent experiments s.e.m. ^*P < 0.05, ^**P < 0.01.

Full size image (62 KB)

Confirming that FBG action is MyD88 dependent, we found that FBG does not stimulate cytokine synthesis in embryonic fibroblasts isolated from Myd88^-/- mice (Fig. 6b). The TLR2 ligand Pam3Cys-Ser-Lys4 (PAM3), the TLR4 ligand LPS and IL-1 all signal via MyD88. Stimulation with these ligands was also abolished in Myd88^-/- mouse embryonic fibroblasts (Fig. 6b). However, stimulation by exogenously added TNF-, which does not signal via MyD88, was unaffected in these cells (Fig. 6b). Retransfection of wild-type MyD88 restored the responsiveness of these cells to FBG, PAM3, LPS and IL-1 (data not shown).

FBG signals via TLR4

TLRs show specificity for endogenous ligands; proteins are recognized by TLR2, TLR4 or both⁴. Neutralizing antibodies to TLR4 inhibited FBG- and LPS-induced IL-6, IL-8 and TNF- synthesis in human macrophages and IL-6 synthesis in synovial fibroblasts but had no effect on stimulation with PAM3 (Fig. 6c). Antibodies to TLR2 inhibited PAM3-mediated cytokine synthesis but had no effect on LPS- or FBG-mediate cytokine synthesis (Fig. 6c). (TNF- synthesis by human macrophages is shown in Fig. 6c.) Confirming that FBG action is TLR4 dependent, we found that FBG does not stimulate cytokine synthesis in embryonic fibroblasts or macrophages isolated from Tlr4^-/- mice (Fig. 6d,e). Embryonic fibroblasts or macrophages isolated from Tlr2^-/- mice were unresponsive to PAM3 but were responsive to FBG, LPS and IL-1 (Fig. 6d,e). Cells isolated from Tlr4^-/- mice were unresponsive to LPS but did respond to PAM3 and IL-1 (Fig. 6d,e).

In addition, expression of TLR4 was required for the arthritogenic action of FBG in vivo; FBG induced joint inflammation in Tlr2^-/- mice but not in Tlr4^-/- mice (Supplementary Fig. 5). We also found that TLR4 is involved in the progression of antigen-induced arthritis. Tlr4^-/- mice show less joint inflammation and are protected from chondrocyte death and cartilage and bone erosion compared to wild-type mice (Supplementary Fig. 6).

Different co-receptor requirements for FBG and LPS

LPS signaling via TLR4 is mediated by a receptor complex including myeloid differentiation protein-2 (MD-2) and CD14 (ref. 26). We next examined whether CD14 and MD-2 are required for FBG activation of TLR4. As a positive control, here we examined the activity of smooth glycosylated LPS, which requires both MD-2 and CD14 for its action²⁷. LPS-mediated IL-6, IL-8 and TNF- synthesis by human macrophages and IL-6 synthesis by synovial fibroblasts was inhibited by antibodies to CD14 and an antagonistic LPS derived from the msbB strain of Escherichia coli, which competes for LPS binding to MD-2 (ref. 28) (Fig. 6f). Conversely, cytokine synthesis mediated by both PAM3, which does not require these co-receptors for activation of TLR2, and FBG was unaffected by antibodies to CD14 or msbB mutant LPS (Fig. 6f shows TNF- synthesis by human macrophages). These data suggest that neither CD14 nor MD-2 is required for FBG-mediated cytokine synthesis. Therefore, whereas LPS and FBG both signal via activation of TLR4, they may have different co-receptor requirements.

Note: Supplementary information is available on the Nature Medicine website.

Acknowledgments

This paper is dedicated to B. Foxwell, an inspirational colleague and a good friend, who will be sadly missed. We thank R. Best for processing tissue, D. Essex for immunohistology and C. ffrench-Constant (University of Edinburgh) for the gift of the Tnc^-/- mice. This work was funded by the Arthritis Research Campaign, The Kennedy Institute of Rheumatology Trustees and a UK Medical Research Council New Investigators Research Grant awarded to K.M. We are also grateful for support from the UK National Institute for Health Research Biomedical Research Centre funding scheme. G.O. was supported by Institut National du Cancer–Institut National de la Santé et de la Recherche Médicale and a Contrat d'interface with the Hospital Hautepierre.

Received 5 March 2009; Accepted 9 May 2009; Published online 28 June 2009.

References

1. Smolen, J.S. & Maini, R.N. Interleukin-6: a new therapeutic target. Arthritis Res. Ther. 8 Suppl 2, S5 (2006).
2. Williams, R.O., Paleolog, E. & Feldmann, M. Cytokine inhibitors in rheumatoid arthritis and other autoimmune diseases. Curr. Opin. Pharmacol. 7, 412–417 (2007). | Article | PubMed | ChemPort |
3. Brentano, F., Kyburz, D., Schorr, O., Gay, R. & Gay, S. The role of Toll-like receptor signalling in the pathogenesis of arthritis. Cell. Immunol. 233, 90–96 (2005). | Article | PubMed | ISI | ChemPort |
4. O'Neill, L.A. Primer: Toll-like receptor signaling pathways—what do rheumatologists need to know? Nat. Clin. Pract. Rheumatol. 4, 319–327 (2008). | Article | PubMed | ChemPort |
5. Matzinger, P. The danger model: a renewed sense of self. Science 296, 301–305 (2002). | Article | PubMed | ISI | ChemPort |
6. Bianchi, M.E. DAMPs, PAMPs and alarmins: all we need to know about danger. J. Leukoc. Biol. 81, 1–5 (2007). | Article | PubMed | ChemPort |
7. Gordon, S. Pattern recognition receptors: doubling up for the innate immune response. Cell 111, 927–930 (2002). | Article | PubMed | ISI | ChemPort |
8. Medzhitov, R. & Janeway, C.A. Jr. Decoding the patterns of self and nonself by the innate immune system. Science 296, 298–300 (2002). | Article | PubMed | ISI | ChemPort |
9. Radstake, T.R. et al. Expression of Toll-like receptors 2 and 4 in rheumatoid synovial tissue and regulation by proinflammatory cytokines interleukin-12 and interleukin-18 via interferon-. Arthritis Rheum. 50, 3856–3865 (2004). | Article | PubMed | ISI | ChemPort |
10. Roelofs, M.F. et al. The expression of Toll-like receptors 3 and 7 in rheumatoid arthritis synovium is increased and costimulation of Toll-like receptors 3, 4, and 7/8 results in synergistic cytokine production by dendritic cells. Arthritis Rheum. 52, 2313–2322 (2005). | Article | PubMed | ChemPort |
11. Sacre, S.M. et al. The Toll-like receptor adaptor proteins MyD88 and Mal/TIRAP contribute to the inflammatory and destructive processes in a human model of rheumatoid arthritis. Am. J. Pathol. 170, 518–525 (2007). | Article | PubMed | ISI | ChemPort |
12. Sacre, S.M. et al. Inhibitors of TLR8 reduce TNF production from human rheumatoid synovial membrane cultures. J. Immunol. 181, 8002–8009 (2008). | PubMed | ChemPort |
13. Choe, J.Y., Crain, B., Wu, S.R. & Corr, M. Interleukin 1 receptor dependence of serum transferred arthritis can be circumvented by Toll-like receptor 4 signaling. J. Exp. Med. 197, 537–542 (2003). | Article | PubMed | ISI | ChemPort |
14. Lee, E.K., Kang, S.M., Paik, D.J., Kim, J.M. & Youn, J. Essential roles of Toll-like receptor-4 signaling in arthritis induced by type II collagen antibody and LPS. Int. Immunol. 17, 325–333 (2005). | Article | PubMed | ChemPort |
15. Abdollahi-Roodsaz, S. et al. Inhibition of Toll-like receptor 4 breaks the inflammatory loop in autoimmune destructive arthritis. Arthritis Rheum. 56, 2957–2967 (2007). | Article | PubMed | ChemPort |
16. Vanags, D. et al. Therapeutic efficacy and safety of chaperonin 10 in patients with rheumatoid arthritis: a double-blind randomised trial. Lancet 368, 855–863 (2006). | Article | PubMed | ISI | ChemPort |
17. Chiquet-Ehrismann, R. & Chiquet, M. Tenascins: regulation and putative functions during pathological stress. J. Pathol. 200, 488–499 (2003). | Article | PubMed | ISI | ChemPort |
18. Cutolo, M., Picasso, M., Ponassi, M., Sun, M.Z. & Balza, E. Tenascin and fibronectin distribution in human normal and pathological synovium. J. Rheumatol. 19, 1439–1447 (1992). | PubMed | ChemPort |
19. McCachren, S.S. & Lightner, V.A. Expression of human tenascin in synovitis and its regulation by interleukin-1. Arthritis Rheum. 35, 1185–1196 (1992). | Article | PubMed | ISI | ChemPort |
20. Salter, D.M. Tenascin is increased in cartilage and synovium from arthritic knees. Br. J. Rheumatol. 32, 780–786 (1993). | Article | PubMed | ChemPort |
21. Chevalier, X., Groult, N., Larget-Piet, B., Zardi, L. & Hornebeck, W. Tenascin distribution in articular cartilage from normal subjects and from patients with osteoarthritis and rheumatoid arthritis. Arthritis Rheum. 37, 1013–1022 (1994). | Article | PubMed | ChemPort |
22. Hasegawa, M. et al. Expression of large tenascin-C splice variants in synovial fluid of patients with rheumatoid arthritis. J. Orthop. Res. 25, 563–568 (2007). | Article | PubMed | ChemPort |
23. Orend, G. Potential oncogenic action of tenascin-C in tumorigenesis. Int. J. Biochem. Cell Biol. 37, 1066–1083 (2005). | Article | PubMed | ISI | ChemPort |
24. Brackertz, D., Mitchell, G.F. & Mackay, I.R. Antigen-induced arthritis in mice. I. Induction of arthritis in various strains of mice. Arthritis Rheum. 20, 841–850 (1977). | Article | PubMed | ISI | ChemPort |
25. Brennan, F.M., Chantry, D., Jackson, A., Maini, R. & Feldmann, M. Inhibitory effect of TNF antibodies on synovial cell interleukin-1 production in rheumatoid arthritis. Lancet 2, 244–247 (1989). | Article | PubMed | ISI | ChemPort |
26. Fitzgerald, K.A., Rowe, D.C. & Golenbock, D.T. Endotoxin recognition and signal transduction by the TLR4/MD2-complex. Microbes Infect. 6, 1361–1367 (2004). | Article | PubMed | ISI | ChemPort |
27. Jiang, Z. et al. CD14 is required for MyD88-independent LPS signaling. Nat. Immunol. 6, 565–570 (2005). | Article | PubMed | ISI | ChemPort |
28. Coats, S.R., Do, C.T., Karimi-Naser, L.M., Braham, P.H. & Darveau, R.P. Antagonistic lipopolysaccharides block E. coli lipopolysaccharide function at human TLR4 via interaction with the human MD-2 lipopolysaccharide binding site. Cell. Microbiol. 9, 1191–1202 (2007). | Article | PubMed | ChemPort |
29. Siri, A. et al. Human tenascin: primary structure, pre-mRNA splicing patterns and localization of the epitopes recognized by two monoclonal antibodies. Nucleic Acids Res. 19, 525–531 (1991). | Article | PubMed | ISI | ChemPort |
30. Gondokaryono, S.P. et al. The extra domain A of fibronectin stimulates murine mast cells via toll-like receptor 4. J. Leukoc. Biol. 82, 657–665 (2007). | Article | PubMed | ChemPort |
31. Taylor, K.R. et al. Recognition of hyaluronan released in sterile injury involves a unique receptor complex dependent on Toll-like receptor 4, CD44, and MD-2. J. Biol. Chem. 282, 18265–18275 (2007). | Article | PubMed | ChemPort |
32. Kim, H.M. et al. Crystal structure of the TLR4–MD-2 complex with bound endotoxin antagonist Eritoran. Cell 130, 906–917 (2007). | Article | PubMed | ChemPort |
33. Schaefer, L. et al. The matrix component biglycan is proinflammatory and signals through Toll-like receptors 4 and 2 in macrophages. J. Clin. Invest. 115, 2223–2233 (2005). | Article | PubMed | ChemPort |
34. Foell, D., Wittkowski, H. & Roth, J. Mechanisms of disease: a 'DAMP' view of inflammatory arthritis. Nat. Clin. Pract. Rheumatol. 3, 382–390 (2007). | Article | PubMed | ChemPort |
35. Taniguchi, N. et al. High mobility group box chromosomal protein 1 plays a role in the pathogenesis of rheumatoid arthritis as a novel cytokine. Arthritis Rheum. 48, 971–981 (2003). | Article | PubMed | ChemPort |
36. Pullerits, R. et al. High mobility group box chromosomal protein 1, a DNA binding cytokine, induces arthritis. Arthritis Rheum. 48, 1693–1700 (2003). | Article | PubMed | ChemPort |
37. Kokkola, R. et al. Successful treatment of collagen-induced arthritis in mice and rats by targeting extracellular high mobility group box chromosomal protein 1 activity. Arthritis Rheum. 48, 2052–2058 (2003). | Article | PubMed | ISI | ChemPort |
38. Gutowski, N.J., Newcombe, J. & Cuzner, M.L. Tenascin-R and C in multiple sclerosis lesions: relevance to extracellular matrix remodelling. Neuropathol. Appl. Neurobiol. 25, 207–214 (1999). | Article | PubMed | ISI | ChemPort |
39. Amin, K. et al. Inflammation and structural changes in the airways of patients with primary Sjogren's syndrome. Respir. Med. 95, 904–910 (2001). | Article | PubMed | ChemPort |
40. Loots, M.A. et al. Differences in cellular infiltrate and extracellular matrix of chronic diabetic and venous ulcers versus acute wounds. J. Invest. Dermatol. 111, 850–857 (1998). | Article | PubMed | ISI | ChemPort |
41. Keystone, E.C., Schorlemmer, H.U., Pope, C. & Allison, A.C. Zymosan-induced arthritis: a model of chronic proliferative arthritis following activation of the alternative pathway of complement. Arthritis Rheum. 20, 1396–1401 (1977). | Article | PubMed | ISI | ChemPort |
42. van Lent, P.L. et al. Fc receptors directly mediate cartilage, but not bone, destruction in murine antigen-induced arthritis: uncoupling of cartilage damage from bone erosion and joint inflammation. Arthritis Rheum. 54, 3868–3877 (2006). | Article | PubMed | ChemPort |

Online methods

Materials. The sources of the reagents and mice that we used in this study, together with detailed methods, are given in the Supplementary Methods. All experimental animal procedures were approved by the Kennedy Institute of Rheumatology Ethics Committee and the UK Home Office.

Zymosan-induced arthritis. We induced arthritis with zymosan as previously described⁴¹. We dissolved 15 mg of zymosan (Saccharomyces cerevisiae) in 1 ml of sterile PBS and boiled the solution twice before sonicating it. We anesthetized mice by intraperitoneal injection of 150 l of Hypnorm diluted 1 in 10 in sterile water and then injected them with zymosan (10 l) into the right footpad (day 0). Control mice received an injection of 10 l PBS alone or were not injected. For macroscopic assessment of arthritis, we measured the thickness of each hind paw daily with microcalipers (Kroeplin), and we expressed the diameter as an average for each inflamed hind paw per mouse. After completion of the experiment (day 4), we killed the mice and fixed the hind paws in 10% (vol/vol) buffered formalin, decalcified them with 10% EDTA and embedded in paraffin.

Antigen-induced arthritis. We induced arthritis as described previously²⁴. Briefly, at day 0, we anesthetized mice by intraperitoneal injection of 150 l of Hypnorm diluted 1 in 10 in sterile water and then immunized them with 200 g of mBSA. We emulsified the mBSA in 0.2 ml of complete Freund's adjuvant and injected it intradermally at the base of the tail. At 7 d, we induced arthritis by intra-articular injection of mBSA (100 g in 10 l of sterile PBS) into the right knee joint using a sterile 33-gauge microcannula. We gave control mice an injection of 10 l PBS alone or did not inject them. At 8, 10 or 14 d, we killed the mice and excised the knee joints, fixed them in 10% (vol/vol) buffered formalin, decalcified them with 10% EDTA and processed them to paraffin.

Synthesis and purification of recombinant proteins. We synthesized and purified full-length tenascin-C and proteins corresponding to each domain of tenascin-C (assembly domain, EGF-L various TNIII repeats and FBG) as described in the Supplementary Methods and Supplementary Figures 1 and 2.

Injection of fibrinogen-like globe. We anesthetized mice by intraperitoneal injection of 150 l of Hypnorm diluted 1 in 10 in sterile water and then injected them with 100 ng, 1 g or 3 g FBG in 10 l of sterile PBS into the right knee joint using a sterile 33-gauge microcannula. Control mice received an injection of 10 l PBS alone or were not injected. At 3 d and 7 d, we kicked the mice, and we excised the knee joints, fixed them in 10% (vol/vol) buffered formalin, decalcified them with 10% EDTA and processed them to paraffin.

Histology of knee joints. We cut coronal tissue sections (4 m) at seven depths throughout the joint, 80 m apart and stained them with H&E or safranin-O to assess joint pathology. We scored histopathological changes using the following parameters as described previously⁴². We graded inflammation (the influx of inflammatory cells into synovium (infiltrate) and the joint cavity (exudate) with an arbitrary scale from 0 (no inflammation) to 3 (severe inflammation). We determined chondrocyte death as the percentage of cartilage area containing empty lacunae in relation to the total area. We determined cartilage surface erosion as the amount of cartilage lost in relation to the total cartilage area. We determined bone destruction in ten different areas of the total knee joint section and graded it on a scale of 0 (no damage) to 3 (complete loss of bone structure). Histological analysis was performed by an investigator who was blinded to the experimental groups. We calculated the mean score for each mouse in an experimental group by averaging the histopathological scores in at least five section depths per joint.

Human specimens. We isolated, cultured and stimulated human macrophages, rheumatoid arthritis membrane cells (representing a mixed population of all synovial cell types) and rheumatoid arthritis synovial fibroblasts as described in the Supplementary Methods. The study was approved by the Riverside National Health Service Research Committee, and tissue was obtained only after signed informed consent from the subject.

Cell culture and stimulation. We isolated mouse embryonic fibroblasts and bone marrow–derived macrophages and cultured and stimulated them and HEK293 cells as described in the Supplementary Methods.

Statistical analyses. We used GraphPad v3.0 (GraphPad Software) to calculate the mean, s.d., s.e.m. and perform statistical tests. We analyzed multiple group means by one-way analysis of variance, followed by the Dunnett multiple comparisons test, where appropriate. We used two-tailed, unpaired t test for experiments involving only two groups and two-tailed, paired t test to compare two paired groups. We used the Mann-Whitney U test to determine statistical significance between the histological scores of different groups. P values less than 0.05 were considered significant.

MORE ARTICLES LIKE THIS

These links to content published by NPG are automatically generated.

NPG Journals

by Subject Area

• Chemistry

  • Chemistry
  • Drug discovery
  • Biotechnology
  • Materials
  • Methods & Protocols

• Clinical Practice & Research

  • Cancer
  • Cardiovascular medicine
  • Dentistry
  • Endocrinology
  • Gastroenterology & Hepatology
  • Methods & Protocols
  • Pathology & Pathobiology
  • Urology

• Earth & Environment

  • Earth sciences
  • Evolution & Ecology

• Life sciences

  • Biotechnology
  • Cancer
  • Development
  • Drug discovery
  • Evolution & Ecology
  • Genetics
  • Immunology
  • Medical research
  • Methods & Protocols
  • Microbiology
  • Molecular cell biology
  • Neuroscience
  • Pharmacology
  • Systems biology

• Physical sciences

  • Physics
  • Materials

by A - Z Index