You are here

  • Home
  • Archive
  • Volume 67, Issue 12
  • Myeloid-related proteins S100A8/S100A9 regulate joint inflammation and cartilage destruction during antigen-induced arthritis

Article Text

Article menu

Download PDFPDF

Basic and translational research
Myeloid-related proteins S100A8/S100A9 regulate joint inflammation and cartilage destruction during antigen-induced arthritis
  1. P L E M van Lent1,
  2. L Grevers1,
  3. A B Blom1,
  4. A Sloetjes1,
  5. J S Mort2,
  6. T Vogl3,
  7. W Nacken3,
  8. W B van den Berg1,
  9. J Roth3
  1. 1
    Department of Rheumatology, Radboud University Medical Centre, Nijmegen, The Netherlands
  2. 2
    Shriners Hospital for Children and Department of Surgery, Montreal, Quebec, Canada
  3. 3
    Institute of Experimental Dermatology, University of Munster, Munster, Germany
  1. Dr P van Lent, Department of Rheumatology, UMC Nijmegen, Building 850, Geert Grooteplein 26–28, Nijmegen, The Netherlands; P.vanlent{at}reuma.umcn.nl

Abstract

Objective: To study the active involvement of Myeloid-related proteins S100A8 and S100A9 in joint inflammation and cartilage destruction during antigen-induced arthritis (AIA).

Methods: Joint inflammation and cartilage destruction was measured with 99mTc uptake and histology. The role of S100A8/A9 was investigated by inducing AIA in S100A9–/– mice that also lack S100A8 at protein level, or after intra-articular injection of rS100A8 in mouse knee joints. Cartilage destruction was measured using immunolocalisation of the neoepitope VDIPEN or NITEGE. mRNA levels of matrix metalloproteinases (MMPs) and cytokines were measured using reverse transcriptase (RT)-PCR.

Results: Immunisation of S100A9–/– mice with the antigen mBSA induced normal cellular and humoral responses, not different from wild type (WT) controls. However, joint swelling measured at day 3 and 7 after AIA induction was significantly lower (36 and 70%, respectively). Histologically, at day 7 AIA, cellular mass was much lower (63–80%) and proteoglycan depletion from cartilage layers was significantly reduced (between 50–95%). Cartilage destruction mediated by MMPs was absent in S100A9–/– mice but clearly present in controls. MMP3, 9 and 13 mRNA levels were significantly lowered in arthritic synovia of S100A9–/–. In vitro stimulation of macrophages by the heterodimer S100A8/A9 or S100A8 elevated mRNA levels of MMP3, 9 and in particular MMP13. Intra-articular injection of S100A8 caused prominent joint inflammation and depletion of proteoglycans at day 1. Significant upregulation of mRNA levels of S100A8/A9, cytokines (interleukin 1 (IL1)), MMPs (MMP3, MMP13 and a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS)4) was found in the synovium and correlated with strong upregulation of NITEGE neoepitopes within the cartilage layers.

Conclusions: S100A8/A9 regulate joint inflammation and cartilage destruction during antigen-induced arthritis.

https://doi.org/10.1136/ard.2007.077800

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    During rheumatoid arthritis (RA), large amounts of inflammatory cells mainly macrophages migrate into the synovial layer.1 A significant correlation was found between the number of activated macrophages and occurrence of cartilage destruction.2 3 In previous studies we showed that activated macrophages are crucial in regulating joint inflammation and cartilage destruction during experimental arthritis.4–6 Selective elimination of synovial lining macrophages prior to or during experimental arthritis inhibited synovial inflammation almost completely4 and diminished cartilage destruction.

    Cells of the innate immune system, especially macrophages, are activated by recognising invading microorganisms through pathogen associated molecular patterns (PAMPs) and by reacting to tissue damage recognised as damage associated molecular patterns (DAMPs). There is a growing body of evidence that DAMPs promote inflammatory responses after being released by activated cells.7 Most DAMPs exhibit a kind of double life: within cells they have a role in cell homeostasis, eg, as calcium-binding proteins, chaperones, or chromatin-stabilising molecules.8 After secretion into the extracellular compartment during cellular damage or activation they become danger signals, which activate leukocytes and endothelial cells.9 Due to their inflammatory activities the terms “endokines” or “alarmins” have been introduced for these molecules.10

    The most prominent proteins released by activated macrophages are the calgranulins, Myeloid-related proteins (MRP)8 (S100A8) and MRP14 (S100A9). Both proteins belong to the S100 family of calcium-binding proteins that is comprised of 22 members.11 S100A8 and S100A9 form non-covalently associated complexes that exhibit typical properties of DAMPs.12 The complex of S100A8/S100A9 modulates polymerisation of microtubules during migration of phagocytes integrating p38 mitogen-activated protein (MAP) kinase and calcium depending signalling.13 S100A8 has been shown to be the active component of this complex whereas S100A9 exhibits rather regulatory functions and protects S100A8 from degradation.13 However, inflammatory mediators such as interleukin (IL)1, tumour necrosis factor (TNF)α or interferon (IFN)γ stimulate macrophages to upregulate and secrete S100A8/S100A9 and it is now clear that S100A8/S100A9 induces proinflammatory responses in leucocytes and endothelial cells.13 14

    The nature of surface receptors for S100A8/S100A9 and their signalling pathways is a matter of debate.15 16 Recently it was found that Toll-like receptor (TLR)4 is the dominant receptor for S100A8 signalling.17 S100A9 or the heterodimer S100A8/A9 in the absence of lipopolysaccharide (LPS) fails to activate macrophages.17 Stimulation by S100A8/S100A9 leads to nuclear factor (NF)κB and MAP kinase (MAPK) signalling.15 16

    S100A8 and S100A9 and the heterodimer accumulate in inflammatory fluids and are postulated to be involved in the pathogenesis of rheumatoid arthritis.18 19 Levels of S100A8/S100A9 correlated well with severity of arthritis and levels were higher in synovial fluid than in blood.18 Moreover S100A8 and S100A9 are both expressed in macrophages present in lining and sublining layers but also in the inflamed synovial pannus that invades cartilage and bone20 and it was recently found that S100A8/S100A9 strongly correlated to radiological damage21 suggesting an active role in RA pathogenesis

    In the present study we investigated the functional role of S100A8/S100A9 in joint inflammation and cartilage destruction during antigen-induced arthritis using mice that were made deficient for S100A9. Targeted deletion of the S100A9 gene results in a loss of S100A8 and S100A9 proteins in mice probably due to a high turnover of isolated S100A8 in absence of its binding partner S100A9.21 Our results show that S100A8/9 regulate joint inflammation and may directly regulate induction of cartilage destruction.

    MATERIALS AND METHODS

    Animals

    S100A9–/– mice were generated as described previously.22 23 S100A9–/– mice and their control littermates were used in our experiments. All mice were housed under specified pathogen-free conditions during breeding and experiments. For each set of experiments mice used were age matched (10 to 20 weeks) and sex matched. All experiments were approved by the local authority Animal Care and Use Committee and were performed by personnel certified by the Dutch Ministry of WVC.

    S100A8/S100A9 proteins

    Recombinant murine S100A8 was expressed and purified as described previously.24 25 The identity of proteins was ascertained by amino acid sequencing and electrospray ionisation mass spectrometry. Endotoxin contaminations were excluded by blocking experiments using polymyxin B sulfate (Sigma, Deisenhofen, Germany). In addition, all protein preparations were tested by limulus amoebocyte lysate assay (BioWhittaker, Walkersville, Maryland, USA) and LPS content was not detectable (mean (SD) sensitivity ∼0.7 (0.5) pg LPS/μg protein), indicating that the maximal possible contamination of the S100 protein preparation was <1.2 pg LPS/μg protein. As control protein for injections, ovalbumin (Sigma Aldrich Corporation, St Louis, Missovri, USA) was used.

    Induction of AIA

    Mice were immunised with 100 μg of methylated bovine serum albumin (BSA) (mBSA, Sigma, Zwijndrecht, The Netherlands) emulsified in 100 μl of Freund complete adjuvant (CFA). Injections were divided over flanks and footpad of the front paws. At 1 week thereafter, two subcutaneous booster injections with 50 μg of mBSA/CFA were given in the neck region. At 2 weeks after these injections, arthritis was induced by intra-articular injection of 60 μg of mBSA in 6 μl of saline into the right knee joint, resulting in chronic arthritis.

    Humoral immunity against mBSA

    Antibodies of various isotypes (IgG, IgG1, IgG2a, IgG2b) directed against mBSA were measured in sera of individual mice with an ELISA. Antigen was coated on microtitre plates (Greiner, Alphen a/d Rijn, The Netherlands) at a concentration of 100 μg/ml. Antibody titres were assessed by twofold serial dilution of the sera followed by detection of bound mouse Ig with 1:500 diluted peroxidase-conjugated rabbit anti-mouse Ig (Miles Laboratories, Elkhart, Indiana, USA). O-Phenylenediamine (1 mg/ml; Sigma) was used as substrate for peroxidase, and the antibody titre was determined by using 50% of the maximal extinction as an endpoint.

    Cellular immunity against mBSA

    Mouse spleen cells were isolated and washed in RPMI supplemented with 10% foetal calf serum (FCS), glutamin (2 mM) and pyruvate (1 mM). Erythrocytes were lysed by treatment of the cells with an 0.16 M NH4Cl solution in 0.17 M Tris, pH 7.2 for 5 min. After two washes in RPMI, the cells were plated on plastic T flasks (75 mm2) from Falcon Plastics (Oxnard, California, USA). After 60 min of incubation at 37°C, the non-adherent cells were harvested by aspiration and two 4–5 ml RPMI washes of the adherent cells; 100 μl of RPMI containing 1×105 T cell enriched spleen cells were placed in each well of a sterile, U bottomed polystyrene microculture plate (Costar, Cambridge, Massachusetts, USA). Antigens or mitogens were added in another 100 μl to give a total volume of 200 μl, and final concentrations of antigen of 50, 25,12, 6 and 3 μg/ml. Cultures were maintained at 37°C in a humidified atmosphere of 5% CO2 and 98% air for 4 days. At 16 h before harvesting, 1 μCi of [3H]thymidine (6.7 Ci/mmol; New England Nuclear, Boston, Massachusetts, USA) was added in 25 μl of RPMI. Cultures were harvested with a cell harvester (Tomtec, Hamden, Connecticut, USA) and 3H thymidine incorporation was determined.

    Intra-articular injection of rS100A8

    A total of 6 μl of phosphate-buffered saline (PBS) containing 5 μg of recombinant S100A8 (together with polymyxin B (10 μg/ml)) was injected into the joint cavity of knees of normal C57Bl/6 mice. As controls, 6 μl of ovalbumin (5 μg), polymixin B (10 μg/ml) or PBS alone was injected. At 1 and 3 days thereafter, knee joints were processed for histology.

    99mTc uptake measurements

    Joint inflammation was measured by 99mTc pertechnetate uptake in the knee joint. Mice were injected intraperitoneally with 12 μCi 99mTc. Gamma radiation was assessed by use of a collimated NaI scintillation crystal with the knee in a fixed position. Arthritis was scored as the ratio of 99mTc uptake in the right and the left knee joint. Right/left ratios >1.1 were taken to indicate inflammation of the right knee joint.

    Histology of arthritic knee joints

    Total knee joints of mice were isolated 7 days after arthritis onset. Knee joints were decalcified, dehydrated and embedded in paraffin. Tissue sections (7 μm) were stained with haematoxylin and eosin (H&E). Histopathological changes were scored using the following parameters. Inflammation was graded on a scale from 0 (no inflammation) to 3 (severe inflamed joint) based on influx of inflammatory cells into synovium and joint cavity. Cartilage destruction was measured as chondrocyte death and surface erosion and was determined in various cartilage layers (medial and lateral femur and tibia) in total knee joint sections stained with H&E. Chondrocyte death was determined as the percentage of the area of the cartilage containing empty lacunae in relation to the total area. Cartilage surface erosion was determined as loss of cartilage in relation to the total cartilage area. All experiments were scored separately and independently from each other in a blinded way by two independent observers. There was agreement between scorers. Data were expressed as the mean of all cartilage layers measured.

    Immunohistochemical analysis of VDIPEN, NITEGE and Myeloid-related proteins S100A8 and S100A9

    Formalin-fixed sections were digested with proteinase-free chondroitinase ABC (0.25 units/ml in 0.1 M Tris-HCl, pH 8.0; Sigma) to remove the glycosaminoglycan side chains followed by incubation with affinity-purified rabbit anti-VDIPEN or anti-NITEGE IgG. The primary antibody was detected using biotinylated goat anti-rabbit IgG and avidin–streptravidin–peroxidase (Elite kit; Vector, Burlingame, California, USA). Counterstaining was with orange G (2%). Areas of immunostaining were expressed as percentage of the total cartilage surface. Rabbit antisera against recombinant murine S100A8 (αS100A8) and S100A9 (αS100A9) were produced as described previously.24 Monospecificity of antibodies was analysed by immunoreactivity against recombinant S100A8 and S100A9 and western blot analysis of lysates of granulocytes. Sections were stained using a final antibody concentration of 1 μg/ml. Primary antibodies were detected using peroxidase-conjugated second-stage antibodies against rabbit IgG (Dinova). Finally sections were counterstained with Mayer haematoxylin (Merck, Darmstadt, Germany).

    Stimulation of macrophages by recombinant S100A8 and S100A8/S100A9

    Bone marrow cells were differentiated into macrophages by culturing in Dulbecco modified Eagle medium (DMEM) containing 10% FCS and 10 ng/ml macrophage colony stimulating factor (MCSF) for 7 days. Macrophages (106 per ml) were stimulated with various concentrations of S100A8 or S100A8/S100A9 heterodimer for 24 h.

    Quantitative detection of MMP mRNA levels using reverse transcriptase (RT)-PCR

    RNA was isolated from well defined synovial specimens or bone marrow derived macrophages (C57BL/6). Synovial biopsies were taken with a biopsy punch (diameter 3 mm) from tissue adjacent to the suprapatellar ligament. RNA was isolated with 1 ml of TRIzol reagent (Life technologies, Breda, The Netherlands). Specific mRNA level for S100A8/A9, cytokines (IL1,TNFα and IL6), MMPs (MMP3, 9, 13 and 14 and a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS) 4 and 5 were detected using the ABI/PRISM 7000 Sequence Detection System (ABI/PE, Foster City, California, USA). Briefly, 1 μg of synovial RNA was used for RT-PCR. mRNA was reverse transcribed to cDNA using oligo dT primers 1/100 of the cDNA was used in one PCR amplification. PCR was performed in SYBR Green Master Mix using the following amplification protocol: 2 min at 50°C followed by 40 cycles of 15 s at 95°C and 1 min at 60°C with data collection in the last 30 s. Message for murine genes was amplified using specific primers (Biolegio, Malden, The Netherlands) at a final concentration of 300 nmol/l. Relative quantification of the PCR signals was performed by comparing the cycle threshold value (Ct) of the Fcγ receptor (FcγR) and MMP genes in the different samples after correction of the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) content for each individual sample to rule out confounding by variation in the RNA purification and reverse transcriptase steps.

    Statistical analysis

    Differences between experimental groups were assessed using the Wilcoxon rank test.

    RESULTS

    Myeloid-related proteins S100A8 and S100A9 are abundantly expressed in inflamed synovium during experimental arthritis

    We first investigated expression of S100A8 and S100A9 mRNA and proteins in inflamed synovium during antigen-induced arthritis. RNA was isolated from synovia derived from inflamed joints at several timepoints (0.16, 1, 2, 4 and 7 days) after induction of antigen-induced arthritis. S100A8 and S100A9 mRNA levels were both strongly upregulated (between 1000–2000 times) when compared to expression found in normal synovium. S100 A8 and S100A9 expression was already maximal at day 1 and remained high until day 7 after arthritis induction (fig 1A,B).

    Figure 1
    Figure 1 mRNA and protein expression of S100A8 and S100A9 were determined in inflamed synovium after induction of antigen-induced arthritis. mRNA levels were measured in inflamed synovia isolated at 0.16, 1, 2, 4 and 7 days after antigen-induced arthritis (AIA) induction using quantitative reverse transcriptase (RT)-PCR. Protein expression of S100A8 and S100A9 was measured using immunolocalisation on total knee joint sections at day 7 after AIA. Note that mRNA levels of S100A8 and S100A9 are significantly raised in inflamed synovium (A and B, S100A8 and S100A9, respectively). Protein is highly expressed in the synovium up to day 7 after AIA and expressed by polymorphonuclear cells and macrophages (micrographs C and D, S100A8 and S100A9, respectively). In non-inflamed synovium, no expression of S100A8 and S100A9 was found (data not shown). Original magnification 400×.

    S100 A8 and S100A9 protein expression was additionally investigated using immunolocalisation in formalin-fixed total knee joint at day 7 after AIA induction. Protein expression of S100A8 and S100A9 was predominantly and abundantly found within the inflammatory cellular mass and stained predominantly macrophages and polymorphonuclear cells (fig 1C,D).

    S100A8/S100A9 regulates joint inflammation during antigen-induced arthritis

    As S100A8 and S100A9 proteins were abundantly expressed in inflamed synovium during antigen-induced arthritis, we next investigated the role of these proteins in joint inflammation using mice that were made deficient for S100A9. Previous studies have shown that in addition to S100A9, S100A8 is also deficient in myeloid cells within these mice. Thus, targeted deletion of S100A9 leads to a complete lack of a functional S100A8/S100A9 complex in the mouse.21 22 As the absence of S100A8/A9 may alter the immunological response against methylated mBSA and thereby impairing the onset and course of arthritis, we first measured humoral and cellular immunity against the antigen mBSA, 4 weeks after immunisation. Humoral immunity was tested by ELISA. Total IgG, IgG1,IgG2a and IgG3 anti-mSBA levels (titres between 11 and 14) were not significantly different between immunised wild type (WT) and S100A9–/– mice. Cellular immunity, as measured by spleen lymphocyte stimulation (LST) against various concentrations of mBSA (SI between 2–4), was also not different between knockout and controls.

    Subsequently, AIA was induced and knee joint swelling was determined at various timepoints after induction. Joint inflammation as measured by 99mTc uptake was significantly lower in arthritic knee joints of S100A9 –/– mice at day 7 but not at day 3 after AIA induction (fig 2A). To further support these findings, histology of total arthritic knee joints was investigated. At day 7 after AIA induction, cellular mass in the joint cavity (exudate) or in the synovial layer (infiltrate) were both significantly lower (63 and 80%, respectively), suggesting that S100A8/A9 promotes cell influx fig 2B, and micrograph (fig 2D) vs WT control (fig 2C)).

    Figure 2
    Figure 2 Inflammation of the knee joint, expressing cellular activation, was determined as right/left ratios of 99mTc uptake at various days (3 and 7) after injection of 60 μg methylated bovine serum albumin (mBSA) in the right knee joint of mBSA-immunised S100A9–/– and wild type (WT) controls (A). Swelling was almost absent in knee joints of arthritic S100A9–/– at day 7 after antigen-induced arthritis (AIA). Frontal sections of whole knee joints, 7 days after induction of AIA in S100A9–/– (C) and their WT controls (D). The amount of cells present in the synovium (infiltrate) and in the knee joint cavity (exudate) was determined using an arbitrary scale from 0–3: 0, no cells; 1, minor; 2, moderate; 3,maximal (B). The amount of cells was determined independently by two blinded observers. Data are the mean of two different experiments. In each experiment seven mice per group were used. Significance was tested using the Wilcoxon rank test (*, p<0.05). Original magnification of the photographs is 250×. F, femur; JS, joint space; P, patella; S, synovium. Note the significantly lower cell mass in S100A9–/– knee joints when compared to WT controls (B and micrographs S100A9–/– D vs WT control C).

    S100A8/S100A9 regulates proteoglycan (PG) depletion, MMP-mediated cartilage destruction and chondrocyte death during antigen-induced arthritis

    Whether S100A8/S100A9 is involved in cartilage destruction was additionally investigated in various cartilage layers of the inflamed knee joint at day 7 after AIA induction. Loss of proteoglycans, a marker of early cartilage destruction was significantly reduced in inflamed joints of S100A9–/– animals (fig 3A and micrographs (fig 3E) vs WT arthritis (fig 3D)). Chondrocyte death, measured as percentage of empty lacunae, varied from 15 up to 35% in arthritic WT controls and was almost completely prevented in S100A9–/– mice (fig 3C). Metalloproteinases are essential in generation of severe breakdown of the cartilage matrix starting at day 5 after antigen-induced arthritis. MMP-mediated proteoglycan neoepitopes were measured in various cartilage layers using VDIPEN immunolocalisation. At day 7 after AIA, VDIPEN staining was moderately elevated and 15–25% of the cartilage surfaces expressed the neoepitope. In inflamed joints of S100A9–/–, VDIPEN staining was almost completely absent at that timepoint (fig 3B and micrographs (fig 3G) vs WT arthritis (fig 3F)).

    Figure 3
    Figure 3 Measurement of various parameters of cartilage destruction at day 7 after antigen-induced arthritis (AIA) induction in knee joints of S100A9–/– and their wild type (WT) controls. A. Loss of proteoglycans was measured as loss of safranin O staining from various cartilage layers in the knee joint (LF, lateral femur; LT, lateral tibia; MF, medial femur; MT, medial tibia) using an arbitrary score from 0 to 3. 0, no loss; 1, minor; 2, moderate; 3, maximal. Loss of staining was significantly lower in S100A9–/– in the lateral tibia (A and E vs WT arthritis (D). B. VDIPEN staining, expressing matrix metalloproteinase (MMP)-mediated degradation, was determined in various cartilage layers and expressed as percentage positive staining of the total cartilage area. VDIPEN staining was almost absent in arthritic knee joints of S100A9–/– when compared to WT controls (G) vs WT arthritis (F). C. In addition, chondrocyte death was also determined in various cartilage layers of the knee joint. Chondrocyte death was expressed as percentage of empty lacunae of the total cartilage area. Note that chondrocyte death was almost absent in arthritic S100A9–/– when compared to WT controls. Data express the mean (SD) of seven mice per group. Two independent experiments were performed. Significance was tested using the Wilcoxon rank test (*, p<0.05). LF, lateral femur; LT, lateral tibia; MF, medial femur; MT, medial tibia. mRNA levels of various MMPs in inflamed synovia of S100A9–/– and their WT controls at day 7 after antigen-induced arthritis. mRNA levels were measured using quantitative reverse transcriptase (RT)-PCR. Synovia of seven mice were determined separately. Note that various MMPs (MMP3, 9 and particularly MMP13) were significantly lower in S100A9–/– animals (H). mRNA levels of MMPs (MMP3, 9 and 13) were additionally determined in macrophages, 24 h after stimulation with various concentrations (0.2, 1 and 5 μg/ml) of rS100A8 or the complex S100A8/9; 5 μg of S100A8 significantly upregulated MMP3, 9 and particularly MMP13 (I).

    S100A8/S100A9 regulates MMP and ADAMTS expression in the inflamed synovium and peritoneal macrophages

    To investigate whether S100A8/A9 regulate MMP expression in the synovial layer, we isolated well defined synovial specimen from inflamed joints of S100A9–/– and control animals at day 7 after antigen-induced arthritis. mRNA levels of S100A8 and various MMPs (MMP3,9,13, 14) were measured using quantitative RT-PCR. mRNA levels of S100A8 were 64 times lower in inflamed synovia of S100A9–/– when compared to controls. S100A8 protein could not be detected by immunolocalisation in the inflamed synovia of S100A9–/– mice indicating that the S100A9–/– mouse also lacks S100A8. mRNA levels of MMP3, 9, particularly MMP13 and also ADAMTS4 and 5 were significantly lower (4, 8, 12, 4 and 3 times, respectively) in inflamed synovia of S100A9–/– when compared to their controls (fig 3H).

    As synovial macrophages are crucial cells in mediating cartilage destruction, we next investigated the direct effect of rS100A8 and rS100A8/S100A9 heterodimer on MMP production by macrophages. Macrophages were stimulated with various concentrations (0.2, 1 and 5 μg) of S100A8 or S100A8/S100A9. To exclude effects due to LPS contaminations all preparations of S100A8 and S100A9 were tested by limulus assay and LPS was not detectable (sensitivity ∼0.7 (0.5) pg LPS/μg protein). Interestingly, 5 μg of S100A8 and the complex S100A8/A9 upregulated mRNA levels of MMP2, 3, 9 and particularly MMP13 (MMP13: 16 and 10 times by S100A8 and S100A8/A9, respectively) (fig 3I) confirming that S100A8 is the active component of the S100A8/S100A9 complex. Moreover, blockade with polymixin B (10 μg/ml) did not alter the proinflammatory effects of S100A8/S100A9 (data not shown).

    Intra-articular injection of S100A8 causes joint inflammation and breakdown of proteoglycans and upregulates mRNA levels of S100A8/A9, cytokines and MMPs

    To investigate further whether S100 proteins are able to directly induce joint inflammation and/or cartilage destruction, 5 μg of S100A8 (in the presence of polymixin B) was injected into knee joints of normal C57BL/6 mice. At 1 and 3 days thereafter, total knee joints were isolated for histological examination. A moderate amount of inflammatory cells was found within the synovium (infiltrate) and joint cavity (exudate) (fig 4A). The lining layer became significantly thickened (fig 4C vs control, fig 4D). Maximal inflammation was observed at day 1 and was lower at day 3 (fig 4A,B). Polymixin B alone had no effect.

    Figure 4
    Figure 4 Joint inflammation, 1 (A) and 3 (B) days after bilateral intra-articular injection of 5 μg of rS100A8 in knee joints of C57Bl/6 mice. Total knee joint sections were stained with haematoxylin and eosin (H&E) and joint inflammation (infiltrate and exudate) was scored using an arbitrary scale from 0–3. (0 = no, 1 = low, 2 =  moderate, 3 = maximal cell influx. Note that moderate inflammation was observed at day 1 (A and micrographs C vs control D). F, femur; P, patella. Data represent the mean (SD) of six mice for each timepoint. Significance was tested using the Wilcoxon rank test (*, p<0.05). Original magnification 400×.

    In addition, the effect of intra-articular injection of S100A8 protein on synovial activation was investigated using quantitative RT-PCR. At day 1 after injection, a significant autoinduction of S100A8 but also upregulation of S100A9 was observed (fig 5A). Cytokines, in particular IL1 levels were significantly upregulated at day 1 (IL1β, IL6,TNFα, 96, 3 and 6 times, respectively) (fig 5B). MMPs and ADAMTS, crucial in mediating PG depletion, were also significantly upregulated at day 1 (MMP3, 9, 13 and 14, 24, 4, 12 and 4 times, respectively and ADAMTS4 and 5, 6 and 3 times, respectively ) (fig 5C). Interestingly particularly MMP14 expression remained high up to day 3 (fig 5C).

    Figure 5
    Figure 5 mRNA levels of S100 proteins (S100A8/S100A9) (A), cytokines (interleukin (IL)1β, IL6, tumour necrosis factor (TNF)α) (B) and matrix metalloproteinases (MMPs; MMP3, 9, 13, 14, a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS)4 and 5) (C) in inflamed synovia, 1 and 3 days after intra-articular injection of 5 μg of S100A8. Data represent the mean (SD) of six mice. Note that strong elevations of all genes are found at day 1 after intra-articular injection.

    In addition, loss of PG depletion (measured as loss of safranin O staining) was measured in various cartilage layers of the knee joint. Intra-articular injection of S100A8 caused PG loss varying from 20% in the medial femur and tibia up to 70% in the lateral tibia. Highest PG depletion was observed at day 1 after injection (fig 6A and micrograph fig 6B). Injection of 5 μg of ovalbumin or PBS only had no effect on synovial inflammation or PG depletion (fig 6C). To further analyse whether MMPs or aggrecanases were responsible for the observed PG loss, total knee joint sections were stained with anti-VDIPEN or anti-NITEGE antibodies. A single injection of S100A8 failed to induce MMP-mediated neoepitopes within the cartilage (data not shown). In contrast NITEGE epitopes were markedly present at day 1 after injection (fig 6D and micrographs (fig 6F) vs ovalbumin injected control joint (fig 6E)).

    Figure 6
    Figure 6 Depletion of proteoglycans from and induction of a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS) neoepitopes in various cartilage surfaces of the knee joint, 1 day after a single injection of 5 μg of rS100A8 or ovalbumin. Proteoglycan (PG) depletion was measured as loss of red staining from safranin-O stained knee joint sections. PG depletion was expressed as % of total cartilage layer. Note that moderate depletion was measured at day 1 (A and micrographs B vs control C). NITEGE was determined with immunolocalisation using specific antibodies. Note that S100A8 induces large scale NITEGE expression (D and micrograph F vs control E). Data represent the mean (SD) of six mice. Significance was tested using the Wilcoxon rank test (*, p<0.05). LF, lateral femur; LT, lateral tibia; MF, medial femur; MT, medial tibia.

    DISCUSSION

    In the present study we found that S100A8/S100A9 regulate joint inflammation and cartilage destruction. S100A8/S100A9 may regulate joint inflammation in various ways. Potentially S100A8/A9 proteins may be involved in the generation of antibodies and T cells obligatory for development of antigen-induced arthritis. However neither antibody titres nor antigen-specific T cell activation against methylated BSA were different between immunised S100A9–/– mice and their controls suggesting that the observed differences are not related to immunological response development but rather to proinflammatory effects on innate immune mechanisms by local production of S100 proteins within the inflamed joint.

    S100A8 and S100A9 were initially identified in the context of RA.26 Early infiltrating monocytes and granulocytes expressing S100A8/S100A9 have been found in the sublining and lining layer of inflamed synovial tissue and both proteins are also expressed in macrophage-like cells within the lining layer. The expression of S100A8 and S100A9 was found to be strongest at the cartilage–pannus junction, which is the prime site of cartilage destruction and bone erosion in arthritis.20 27 Concentrations of S100A8/S100A9 found in synovial fluid are more than 10-fold higher compared to concentrations in serum indicating release of these proteins from activated phagocytes within the synovium. Expression and serum concentrations of S100A8/S100A9 correlate very well with disease activity in RA, JRA and psoriatic arthritis28 29 but even more importantly these proteins correlate well to radiological damage.20 A novel inflammatory syndrome characterised by an extraordinarily high expression of these two calcium-binding molecules confirms a direct role of S100A8 and S100A9 in inflammatory processes in the joint since arthritis is a common symptom in most affected patients.30 In addition, previous data from experimental models of arthritis have already provided indirect evidence for a local role of S100A8/S100A9 in arthritis. Mice lacking FcγRs show a clear correlation with disease activity, cartilage destruction and activation of synovial macrophages, with expression of S100A8 and S100A9 in experimental arthritis.31–33 Secretion of S100A8/S100A9 is associated with activation of phagocytes.34 The complex of S100A8/S100A9 binds to endothelial cells by specific interaction with heparan sulfate proteoglycans and novel carboxylated glycans.35 S100A8/S100A9 induces a thrombogenic and inflammatory response in endothelial cells, which is caused by induction of proinflammatory chemokines and adhesion molecules as well as by a loss of cell–cell contacts increasing the endothelial permeability.36 The complex of S100A8/S100A9 increases binding activity of the integrin receptor CD11b/CD18 on neutrophils, which enhances adhesion of these cells to endothelial cells.37 Furthermore, S100A8/S100A9 proteins have recently been shown to activate macrophages to produce cytokines such as IL1 and TNF.17 These cytokines are crucial in activating synovial lining cells, thereby producing chemokines such as macrophage inflammatory protein (MIP)1α and IL8 involved in influx of monocytes and polymorphonuclear cells.

    Apart from joint inflammation, levels of cartilage destruction also significantly decreased in S100A9–/– mice. Cartilage destruction is particularly mediated by enzymes such as MMPs. As mentioned above S100A9 –/– mice lack S100A8 at protein level as well. S100A8 has been described earlier as the active component of the S100A8/S100A9 complex.13 17 We confirmed the active role of S100A8 in our model showing induction of different MMPs in macrophages by the S100A8/S100A9 complex and the isolated S100A8 chain.

    Early cartilage destruction in antigen-induced arthritis is characterised by loss of aggrecan and ADAMTS family members are thought to be involved in mediating this early aggrecan loss during antigen-induced arthritis.38 39 These proteins are glutamyl endopeptidases that cleave the aggrecan core protein at a series of specific sites. In particular ADAMTS4 and 5, have been implicated in tissue destruction during rheumatoid arthritis. Recently, Stanton et al clearly demonstrated that ablation of the ADAMTS5 gene protected against aggrecan loss and cartilage erosion in antigen-induced arthritis.40 In the present study we found that a single intra-articular injection of S100A8 leads to strong upregulation of ADAMTS4 and 5 within the synovial layer and marked production of ADAMTS generated neoepitopes (NITEGE) was found within cartilage layers of the knee joint. After intra-articular injection, S100A8 may bind to as yet unknown macrophage receptors resulting in activation and subsequent release of ADAMTS. ADAMTS4 and 5 are synthesised as latent precursor proteins that requires activation through removal of their prodomains before they can exert catalytic activity. Proprotein convertases efficiently remove the prodomain through cleavage in the case of ADAMTS4 at Arg212/Phe213 thereby generating the active enzyme.41

    MMPs are also crucial in mediating late cartilage destruction in AIA.36 Neoepitopes generated by MMPs (VDIPEN) did not occur earlier than 5 days after induction of AIA. MMP3 appeared to be the dominant MMP within this model since VDIPEN staining was completely inhibited in MMP3 knockout mice.38 VDIPEN neoepitopes colocalise with collagenase-induced neoepitopes suggesting that activation of MMPs eventually leads to cartilage erosion.38 MMP3 may thereby act by activating MMP13

    In the present study we found that synovial expression of MMPs, particularly MMP13 was significantly downregulated in S100A9–/– mice. A plausible explanation may be that the low expression of MMPs is due to the lower amount of inflammatory cells found within the synovia of these mice. The lower inflammatory activity and recruitment in S100A9–/– mice during AIA can either be due to an intracellular function in signal transduction or extracellular effects after release by activated phagocytes. We therefore investigated whether S100A8/S100A9 can directly stimulate macrophages to produce MMPs. Stimulation of macrophages with S100A8/S100A9 strongly upregulated MMPs 3,9 and particularly 13 mRNA levels. This stimulating effect was not due to LPS since adding polymixin B did not alter the results. In addition our data confirm earlier reports that S100A8 is the active component of the S100A8/S100A9 complex whereas S100A9 seems to function predominantly as the regulatory unit.13

    It may be therefore surprising to find protective effects during experimental arthritis in the S100A9 –/– mice. However, keeping in mind the fact that S100A8 protein levels are highly dependent on the presence of S100A9, probably due to degradation of the isolated S100A8 subunit in the absence of its binding partner S100A9, this finding can be easily explained.17 S100A8 is almost undetectable in mature granulocytes and monocytes of S100A9 –/– mice. Thus, targeted deletion of S100A9 leads to functional S100A8 –/– and S100A9 –/– phagocytes.20 21

    Apart from in vitro stimulation of macrophages, intra-articular injection of S100A8 into the knee joint caused strong upregulation of MMP3, 13 and 14 within the synovial layer. We also found that S100A8 was able to stimulate its own expression as well as that of S100A9 in mouse macrophages, which may reflect another positive feedback mechanism promoting inflammatory reactions. A single injection of S100A8 into murine knee joints failed to induce MMP-mediated VDIPEN expression in the cartilage layers. S100A8 is a 10 kDa protein with an electric point around 7. Due to these physicochemical characteristics, the protein will be cleared from the joint very rapidly. This may explain why larger concentrations are needed to induce significant amounts of heterodimers/tetramers mediating the effect.

    MMPs are released in a latent form and need an activation step in order to become active. Production of oxygen radicals may be involved in this activation step. Recent findings have shown that arthritis induced in mice with impaired oxygen radical production (P47phox–/–), show higher MMPs mRNA levels in the synovial layer but lower expression of VDIPEN neoepitopes in the cartilage layers suggesting that oxygen radicals are crucial in mediating MMP activation.42 S100A8 and S100A9 have been described to modulate turnover of oxygen radicals in the extracellular space, which could be the molecular mechanism by which these calcium-binding proteins modulate MMP activity.43 Our present findings suggest that S100A8/S100A9 proteins, which have been shown to be a reliable marker of disease activity and joint inflammation in rheumatic diseases, are key effectors/amplifiers of inflammation as well as cartilage destruction and may form new interesting targets for combating this crippling disease.

    REFERENCES

    1. ↵
      1. van den Berg WB,
      2. van Lent PL
      . The role of macrophages in chronic arthritis. Immunobiology 1996;195:614–23.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Mulherin D,
      2. Fitzgerald O,
      3. Bresnihan B
      . Synovial tissue macrophage populations and articular damage in rheumatoid arthritis. Arthritis Rheum 1996;39:115–24.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Bresnihan B
      . Pathogenesis of joint damage in rheumatoid arthritis. J Rheumatol 1999;26:717–9.
      OpenUrlPubMedWeb of Science
    4. ↵
      1. Van Lent PL,
      2. Holthuysen AE,
      3. Van Rooijen N,
      4. Van De Putte LB,
      5. Van Den Berg WB
      . Local removal of phagocytic synovial lining cells by clodronate-liposomes decreases cartilage destruction during collagen type II arthritis. Ann Rheum Dis 1998;57:408–13.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      1. van Lent PL,
      2. Holthuysen AE,
      3. van Rooijen N,
      4. van de Loo FA,
      5. van de Putte LB,
      6. van den Berg WB
      . Phagocytic synovial lining cells regulate acute and chronic joint inflammation after antigenic exacerbation of smouldering experimental murine arthritis. J Rheumatol 1998;25:1135–45.
      OpenUrlPubMedWeb of Science
    6. ↵
      1. Barrera P,
      2. Blom A,
      3. van Lent PL,
      4. van Bloois L,
      5. Beijnen JH,
      6. van Rooijen N,
      7. et al
      . Synovial macrophage depletion with clodronate-containing liposomes in rheumatoid arthritis. Arthritis Rheum 2000;43:1951–9.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Lotfi R,
      2. Lee JJ,
      3. Lotze MT
      . Eosinophilic granulocytes and damage-associated molecular pattern molecules (DAMPs): role in the inflammatory response within tumors. J Immunother 2007;30:16–28.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. van Eden W,
      2. Hauet-Broere F,
      3. Berlo S,
      4. Paul L,
      5. van der Zee R,
      6. de Kleer I,
      7. et al
      . Stress proteins as inducers and targets of regulatory T cells in arthritis. Int Rev Immunol 2005;24:181–97.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Viemann D,
      2. Schulze-Osthoff K,
      3. Roth J
      . Potentials and pitfalls of DNA array analysis of the endothelial stress response. Biochim Biophys Acta 2005;1746:73–84.
      OpenUrlPubMed
    10. ↵
      1. Oppenheim JJ,
      2. Yang D
      . Alarmins: chemotactic activators of immune responses. Curr Opin Immunol 2005;17:359–65.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Ravasi T,
      2. Hsu K,
      3. Goyette J,
      4. Schroder K,
      5. Yang Z,
      6. Rahimi F,
      7. et al
      . Probing the S100 protein family through genomic and functional analysis. Genomics, 2004;84:10–22.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Foell D,
      2. Wittkowski H,
      3. Vogl T,
      4. Roth J
      . S100 proteins expressed in phagocytes: a novel group of damage-associated molecular pattern molecules. J Leukoc Biol 2007;81:28–37.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Vogl T,
      2. Ludwig S,
      3. Goebeler M,
      4. Strey A,
      5. Thorey IS,
      6. Reichelt R,
      7. et al
      . MRP8 and MRP14 control microtubule reorganization during transendothelialmigration of phagocytes. Blood 2004;104:4260–8.
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Ryckman C,
      2. Vandal K,
      3. Rouleau P,
      4. Talbot M,
      5. Tessier PA
      . Proinflammatory activities of S100: proteins S100A8, S100A9, and S100A8/A9 induce neutrophil chemotaxis and adhesion. J Immunol 2003;170:3233–42.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Srikrishna G,
      2. Turovskaya O,
      3. Shaikh R,
      4. Newlin R,
      5. Foell D,
      6. Murch S,
      7. et al
      . Carboxylated glycans mediate colitis through activation of NF-κB. J Immunol 2005;175:5412–22.
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Hermani A,
      2. De Servi B,
      3. Medunjanin S,
      4. Tessier PA,
      5. Mayer D
      . S100A8 and S100A9 activate MAP kinase and NF-κB signaling pathways and trigger translocation of RAGE in human prostate cancer cells. Exp Cell Res 2006;312:184–97.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Vogl T,
      2. Tenbrock K,
      3. Ludwig S,
      4. Leukert N,
      5. Ehrhardt C,
      6. van Zoelen MA,
      7. et al
      . Mrp8 and Mrp14 are endogenous activators of Toll-like receptor 4, promoting lethal, endotoxin-induced shock. Nat Med 2007;13:1042–9.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Sunahori K,
      2. Yamamura M,
      3. Yamana J,
      4. Takasugi K,
      5. Kawashima M,
      6. Yamamoto H,
      7. et al
      . The S100A8/A9 heterodimer amplifies proinflammatory cytokine production by macrophages via activation of nuclear factor kappa B and p38 mitogen-activated protein kinase in rheumatoid arthritis. Arthritis Res Ther 2006;8:R69.
      OpenUrlCrossRefPubMed
    19. ↵
      1. Foell D,
      2. Roth J
      . Proinflammatory S100 proteins in arthritis and autoimmune disease. Arthritis Rheum 2004;50:3762–71.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Youssef P,
      2. Roth J,
      3. Frosch M,
      4. Costello P,
      5. Fitzgerald O,
      6. Sorg C,
      7. et al
      . Expression of Myeloid-related proteins (MRP) 8 and 14 and the MRP8/14 heterodimer in rheumatoid arthritis synovial membrane. J Rheumatol 1999;26:2523–8.
      OpenUrlPubMedWeb of Science
    21. ↵
      1. Hammer HB,
      2. Odegard S,
      3. Fagerhol MK,
      4. Landewe R,
      5. van der Heijde D,
      6. Uhlig T,
      7. et al
      . Calprotectin (a major leucocyte protein) is strongly and independently correlated to joint inflammation and dage in rheumatoid arthritis. Ann Rheum Dis 2007;66:1093–7.
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. Manitz MP,
      2. Horst B,
      3. Seeliger S,
      4. Strey A,
      5. Skryabin BV,
      6. Gunzer M,
      7. et al
      . Loss of S100A9 (MRP14) results in reduced interleukin-8-induced CD11b surface expression, a polarized microfilament system, and diminished responsiveness to chemoattractants in vitro. Mol Cell Biol 2003;23:1034–43.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Hobbs JA,
      2. May R,
      3. Tanousis K,
      4. McNeill E,
      5. Mathies M,
      6. Gebhardt C,
      7. et al
      . Myeloid cell function in MRP-14 (S100A9) null mice. Mol Cell Biol 2003;23:2564–76.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Goebeler M,
      2. Roth J,
      3. Burwinkel F,
      4. Vollmer E,
      5. Bocker W,
      6. Sorg C
      . Expression and complex formation of S100-like proteins MRP8 and MRP14 by macrophages during renal allograft rejection. Transplantation 1994;58:355–61.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Hunter MJ,
      2. Chazin WJ
      . High level expression and dimer characterization of the S100 EF-hand proteins, migration inhibitory factor-related proteins 8 and 14. J Biol Chem 1998;273:12427–35.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Odink K,
      2. Cerletti N,
      3. Bruggen J,
      4. Clerc RG,
      5. Tarcsay L,
      6. Zwadlo G,
      7. et al
      . Two calcium-binding proteins in infiltrate macrophages of rheumatoid arthritis. Nature 1987;330:80–2.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Kane D,
      2. Roth J,
      3. Frosch M,
      4. Vogl T,
      5. Bresnihan B,
      6. FitzGerald O
      . Increased perivascular synovial membrane expression of myeloid-related proteins in psoriatic arthritis. Arthritis Rheum 2003;48:1676–85.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Wulffraat NM,
      2. Haas PJ,
      3. Frosch M,
      4. De Kleer IM,
      5. Vogl T,
      6. Brinkman DM,
      7. et al
      . Myeloid related protein 8 and 14 secretion reflects phagocyte activation and correlates with disease activity in juvenile idiopathic arthritis treated with autologous stem cell transplantation. Ann Rheum Dis 2003;62:236–41.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Frosch M,
      2. Strey A,
      3. Vogl T,
      4. Wulffraat NM,
      5. Kuis W,
      6. Sunderkotter C,
      7. et al
      . Myeloid-related proteins 8 and 14 are specifically secreted during interaction of phagocytes and activated endothelium and are useful markers for monitoring disease activity in pauciarticular-onset juvenile rheumatoid arthritis. Arthritis Rheum 2000;43:628–37.
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Sampson B,
      2. Fagerhol MK,
      3. Sunderkotter C,
      4. Golden BE,
      5. Richmond P,
      6. Klein N,
      7. et al
      . Hyperzincaemia and hypercal-protectinaemia: a new disorder of zinc metabolism. Lancet 2002;360:1742–5.
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      1. van Lent PL,
      2. Grevers L,
      3. Lubberts E,
      4. de Vries TJ,
      5. Nabbe KC,
      6. Verbeek S,
      7. et al
      . Fcγ 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 2006;54:3868–77.
      OpenUrlCrossRefPubMedWeb of Science
      1. Nabbe KC,
      2. Boross P,
      3. Holthuysen AE,
      4. Sloetjes AW,
      5. Kolls JK,
      6. Verbeek S,
      7. et al
      . Joint inflammation and chondrocyte death become independent of Fcγ receptor type III by local overexpression of interferon-γ during immune complex-mediated arthritis. Arthritis Rheum 2005;52:967–74.
      OpenUrlCrossRefPubMed
    33. ↵
      1. van Lent P,
      2. Nabbe KC,
      3. Boross P,
      4. Blom AB,
      5. Roth J,
      6. Holthuysen A,
      7. et al
      . The inhibitory receptor FcγRII reduces joint inflammation and destruction in experimental immune complex-mediated arthritides not only by inhibition of FcγRI/III but also by efficient clearance and endocytosis of immune complexes. Am J Pathol 2003;163:1839–48.
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Rammes A,
      2. Roth J,
      3. Goebeler M,
      4. Klempt M,
      5. Hartmann M,
      6. Sorg C
      . Myeloid-related protein (MRP) 8 and MRP14, calcium-binding proteins of the S100 family, are secreted by activated monocytes via a novel, tubulin-dependent pathway. J Biol Chem 1997;272:9496–502.
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Robinson MJ,
      2. Tessier P,
      3. Poulsom R,
      4. Hogg N
      . The S100 family heterodimer, MRP-8/14, binds with high affinity to heparin and heparan sulfate glycosaminoglycans on endothelial cells. J Biol Chem 2002;277:3658–65.
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Viemann D,
      2. Strey A,
      3. Janning A,
      4. Jurk K,
      5. Klimmek K,
      6. Vogl T,
      7. et al
      . Myeloid-related proteins 8 and 14 induce a specific inflammatory response in human microvascular endothelial cells. Blood 2005;105:2955–62.
      OpenUrlAbstract/FREE Full Text
    37. ↵
      1. Newton RA,
      2. Hogg N
      . The human S100 protein MRP-14 is a novel activator of the β2 integrin Mac-1 on neutrophils. J Immunol 1998;160:1427–35.
      OpenUrlAbstract/FREE Full Text
    38. ↵
      1. van Meurs J,
      2. van Lent P,
      3. Stoop R,
      4. Holthuysen A,
      5. Singer I,
      6. Bayne E,
      7. et al
      . Cleavage of aggrecan at the Asn341-Phe342 site coincides with the initiation of collagen damage in murine antigen-induced arthritis: a pivotal role for stromelysin 1 in matrix metalloproteinase activity. Arthritis Rheum 1999;42:2074–84.
      OpenUrlCrossRefPubMedWeb of Science
    39. ↵
      1. van Meurs JB,
      2. van Lent PL,
      3. Holthuysen AE,
      4. Singer II,
      5. Bayne EK,
      6. van den Berg WB
      . Kinetics of aggrecanase- and metalloproteinase-induced neoepitopes in various stages of cartilage destruction in murine arthritis. Arthritis Rheum 1999;42:1128–39.
      OpenUrlCrossRefPubMedWeb of Science
    40. ↵
      1. Stanton H,
      2. Rogerson FM,
      3. East CJ,
      4. Golub SB,
      5. Lawlor KE,
      6. Meeker CT,
      7. et al
      . ADAMTS5 is the major aggrecanase in mouse cartilage in vivo and in vitro. Nature 2005;434:648–52.
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      1. Tortorella MD,
      2. Arner EC,
      3. Hills R,
      4. Gormley J,
      5. Fok K,
      6. Pegg L,
      7. et al
      . ADAMTS-4 (aggrecanase-1): N-terminal activation mechanisms. Arch Biochem Biophys 2005;444:34–44.
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      1. van Lent PL,
      2. Nabbe KC,
      3. Blom AB,
      4. Sloetjes A,
      5. Holthuysen AE,
      6. Kolls J,
      7. et al
      . NADPH-oxidase-driven oxygen radical production determines chondrocyte death and partly regulates metalloproteinase-mediated cartilage matrix degradation during interferon-γ-stimulated immune complex arthritis. Arthritis Res Ther 2005;7:R885–95.
      OpenUrlCrossRefPubMedWeb of Science
    43. ↵
      1. Harrison CA,
      2. Raftery MJ,
      3. Walsh J,
      4. Alewood P,
      5. Iismaa SE,
      6. Thliveris S,
      7. et al
      . Oxidation regulates the inflammatory properties of the murine S100 protein S100A8. J Biol Chem 1999;274:8561–9.
      OpenUrlAbstract/FREE Full Text

    Footnotes

    • Competing interests: None declared.

    • Funding: This study was supported by Autocure, PAD (ZonMW) and the Dutch Arthritis Association (grant 07-02-301).