Inhibition of interleukin-33 signaling attenuates the severity of experimental arthritis.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 60, No. 3, March 2009, pp 738–749 DOI 10.1002/art.24305 © 2009, American College of Rheumatology Inhibition of Interleukin-33 Signaling Attenuates the Severity of Experimental Arthritis Gaby Palmer,1 Dominique Talabot-Ayer,1 Céline Lamacchia,1 Dean Toy,2 Christian A. Seemayer,3 Sébastien Viatte,1 Axel Finckh,1 Dirk E. Smith,2 and Cem Gabay1 Objective. Interleukin-33 (IL-33; or, IL-1F11) was recently identified as the ligand of the IL-1 family receptor T1/ST2. The aim of this study was to examine IL-33 production in human and mouse joints and to investigate the role of IL-33 and T1/ST2 in experimental arthritis. Methods. IL-33 expression was examined in human synovial tissue, rheumatoid arthritis (RA) synovial fibroblasts, and arthritic mouse joints. Mice with collagen-induced arthritis (CIA) were treated with blocking anti-ST2 antibody or control antibody beginning at the onset of disease. Arthritis severity was assessed by clinical and histologic scoring. Draining lymph node (LN) cell responses were examined ex vivo, and joint messenger RNA (mRNA) was used for expression profiling. Results. IL-33 was highly expressed in human RA synovium. In cultured synovial fibroblasts, IL-33 expression was strongly induced by IL-1␤ and/or tumor necrosis factor ␣. Furthermore, IL-33 mRNA was detected in the joints of mice with CIA and increased during the early phase of the disease. Administration of a blocking anti-ST2 antibody at the onset of disease attenuated the severity of CIA and reduced joint destruction. Anti-ST2 antibody treatment was associated with a marked decrease in interferon-␥ production as well as with a more limited reduction in IL-17 production by ex vivo–stimulated draining LN cells. Finally, RANKL mRNA levels in the joint were reduced by anti-ST2 treatment. Conclusion. IL-33 is produced locally in inflamed joints, and neutralization of IL-33 signaling has a therapeutic effect on the course of arthritis. These observations suggest that locally produced IL-33 may contribute to the pathogenesis of joint inflammation and destruction. Interleukin-33 (IL-33; or, IL-1F11) was recently identified as a ligand for the orphan IL-1 family receptor T1/ST2. IL-33 is produced as a 30-kd propeptide and, like IL-1␤ and IL-18, is cleaved by caspase 1 to generate mature 18-kd IL-33, at least in vitro (1). Mature IL-33 has been reported to mediate its biologic effects via T1/ST2 binding by activating NF-B and MAP kinases (1). T1/ST2-dependent IL-33 responses resemble classic IL-1–like signaling, and several recent studies identified the IL-1 receptor accessory protein as the coreceptor involved in IL-33 signaling (2–4). A number of studies have established T1/ST2 (also referred to as ST2L) as a selective marker of both murine and human Th2 lymphocytes (for review, see ref. 5). In addition, T1/ST2 is highly expressed on mast cells (6). Several recent reports describe a stimulatory effect of IL-33 on cytokine and chemokine secretion in mast cells (3,7–10), suggesting that, in addition to promoting Th2 responses, IL-33 exhibits proinflammatory potential by inducing the production of a number of inflammatory mediators in mast cells. T1/ST2 also exists as a soluble isoform (sST2) obtained by differential messenger RNA (mRNA) processing. Soluble ST2 is identical with the extracellular region of the long T1/ST2 isoform except Dr. Palmer’s work was supported by a grant from the de Reuter Foundation. Dr. Gabay’s work was supported by the Swiss National Science Foundation (grant 3200-107592/1). 1 Gaby Palmer, PhD, Dominique Talabot-Ayer, MS, Céline Lamacchia, MS, Sébastien Viatte, MD, PhD, Axel Finckh, MD, MS, Cem Gabay, MD: University Hospital of Geneva, and University of Geneva School of Medicine, Geneva, Switzerland; 2Dean Toy, MS, Dirk E. Smith, MS: Amgen, Inc., Seattle, Washington; 3Christian A. Seemayer, MD: University Hospital of Geneva, Geneva, Switzerland. Mr. Toy and Mr. Smith own stock or stock options in Amgen, Inc. Dr. Gabay has received consulting fees, speaking fees, and/or honoraria from Roche, Novartis, Wyeth, Merck, Bristol-Myers Squibb, Serono, and Abbott (less than $10,000 each). Address correspondence and reprint requests to Cem Gabay, MD, Division of Rheumatology, University Hospital of Geneva, 26 Avenue Beau-Séjour, 1211 Geneva 14, Switzerland. E-mail: cem. firstname.lastname@example.org. Submitted for publication April 24, 2008; accepted in revised form November 7, 2008. 738 ROLE OF IL-33 IN ARTHRITIS for 9 additional amino acids, and it was recently demonstrated to act as an antagonistic decoy receptor for IL-33 (11). Serum concentrations of sST2 are elevated in patients with various inflammatory disorders (5). Increased levels of sST2 have also been observed in the synovial fluid of patients with rheumatoid arthritis (RA) as compared with patients with osteoarthritis (12). Several studies have described a role of T1/ST2 and sST2 in regulating inflammatory responses (13–16). Interestingly, mast cells have been recognized as important mediators of the pathogenesis of arthritis (17,18), suggesting that IL-33–mediated mast cell activation might play a role in joint inflammation. The injection of an anti-T1/ST2 antibody had previously been reported to exacerbate collagen-induced arthritis (CIA), but this result was suggested to be due to complementdependent Th2 clone lysis rather than to an effect of the antibody on T1/ST2 signaling (19). Indeed, another study indicated that administration of sST2 decreased CIA (20). At the time, the mechanism proposed to explain this effect was direct inhibition of macrophage activation by sST2 via a putative sST2 receptor expressed at the macrophage surface. However, the identification of IL-33 as the bona fide ligand for T1/ST2 suggests neutralization of IL-33 by sST2 (11) as an alternative explanation. Expression of IL-33 mRNA has previously been detected by in situ hybridization in endothelial cells (ECs) in human RA synovium, suggesting that ECs constitute a major cellular source of IL-33 in inflamed synovial tissue (21). Interestingly, proIL-33 had previously been described in high endothelial venule (HEV) ECs as a nuclear protein and was thus called nuclear factor from HEVs (22). Like proIL-1␣ (23), nuclear proIL-33 may exert unique biologic activities independent of caspase 1 cleavage and cell surface receptor binding (21,22). In the present study, we further examined IL-33 production in human and mouse joints and investigated the role of IL-33 and T1/ST2 in experimental arthritis. The results described herein demonstrate expression of the IL-33 protein in human RA synovium. Furthermore, we show for the first time that, in addition to ECs, synovial fibroblasts produce IL-33 and that its expression is increased by proinflammatory stimuli. Consistent with these findings, IL-33 is produced locally in inflamed mouse joints. Finally, the administration of antibodies that block T1/ST2 signaling attenuated the severity of CIA, thus suggesting that IL-33 plays a pathogenic role in arthritis. 739 MATERIALS AND METHODS Mice. Male DBA/1 mice were obtained from Elevage Janvier (Le Genest St. Isle, France). BALB/c mice were supplied by The Jackson Laboratory (Bar Harbor, ME). The mice were housed under conventional conditions. Animal experiments were approved by the Geneva (Switzerland) Veterinarian Office or by the Amgen (Seattle, WA) Animal Care and Use Committee. Human synovial tissue. Formalin-fixed paraffinembedded tissue samples were obtained from an Amgen tissue collection and from the Department of Clinical Pathology, University Hospital of Geneva. RA synovial biopsy samples were obtained from a 72-year-old woman undergoing joint replacement surgery and from a 47-year-old man undergoing knee arthroscopy. Normal control synovial tissue was obtained at autopsy from a 52-year-old woman who died of an unrelated illness. Immunohistochemistry. Formalin-fixed paraffinembedded tissue samples were stained for IL-33 protein expression using an immunoperoxidase protocol. Briefly, sections were deparaffinized, rehydrated, and heated in a steamer with antigen recovery buffer (DIVA buffer; BioCare Medical, Concord, CA) for 60 minutes. Slides were then blocked and incubated with a polyclonal goat anti-human IL-33 antibody (AF3625; R&D Systems, Minneapolis, MN). Subsequently, slides were washed, blocked for endogenous peroxidase activity, incubated with secondary antibody–peroxidase complexes (ABC Elite; Vector, Burlingame, CA), and developed with diaminobenzidine. In additional experiments, staining was performed using a monoclonal anti-human IL-33 antibody (IL33305B; Alexis, Lausen, Switzerland), which yielded an identical staining pattern (data not shown). To characterize IL-33–positive cells, double immunostaining was performed with anti-CD34 and anti–D2-40 antibodies (M7165 and M3619, respectively; Dako, Baar, Switzerland) to stain vascular and lymphatic ECs or with an anti-CD68 antibody (M0814; Dako) to identify cells of the myeloid lineage and developed using the EnVision G兩2 alkaline phosphatase system with permanent red (Dako). To assess staining specificity, negative controls were performed using unrelated isotype-matched control antibodies. Culture of human synovial fibroblasts. Cell culture reagents were obtained from Invitrogen Life Technologies (Basel, Switzerland). Synovium was obtained from RA patients undergoing knee or hip joint replacement. Synovial fibroblasts were isolated by collagenase digestion, cultured as described previously (24), and used between passages 3 and 13. Reverse transcriptase–polymerase chain reaction (RT-PCR). Human synovial fibroblasts were grown to confluence in 6-well plates and stimulated with IL-1␤ (1 ng/ml) and tumor necrosis factor ␣ (TNF␣) (10 ng/ml) for 6 hours. Cells were solubilized in TRIzol reagent (Gibco Life Technologies, Basel, Switzerland), and RNA was isolated according to the manufacturer’s instructions. Total RNA (2 g) was digested by DNase I and reverse-transcribed, and IL-33 levels were examined by quantitative PCR using the iCycler iQ Real-Time PCR Detection System, iQ SYBR Green Supermix (both from Bio-Rad, Philadelphia, PA), and the following primers: 5⬘-GGTGTTACTGAGTTACTATGAG-3⬘ (forward) and 5⬘-GGAGCTCCACAGAGTGTTCCTTG-3⬘ (reverse) (GenBank 740 accession no. AY905581). IL-33 expression was normalized to expression of the housekeeping gene ␤-glucuronidase (␤-glu) using the following human ␤-glu primers: 5⬘-CCTCTGATGTTCACTGAAG-3⬘ (forward) and 5⬘-CTCTCGTCGGTGACTG-3⬘ (reverse) (GenBank accession no. NM_000181.1). The annealing temperature was 60°C. IL-33 mRNA expression was corrected for ␤-glu mRNA levels, and the IL-33:␤-glu ratios were normalized to the maximal value observed in each experiment. Non–reverse-transcribed RNA samples and water were included as negative controls. The identity of the amplified products was confirmed by sequencing. Western blotting. Human synovial fibroblasts were grown to confluence in 6-well plates and stimulated with IL-1␤ (1 ng/ml) and TNF␣ (10 ng/ml) for 24 hours before lysis in 500 l Laemmli buffer. For each sample, 40 l of total cell lysate was fractionated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to a Porablot membrane (Macherey-Nagel, Düren, Germany). The membrane was blocked in phosphate buffered saline (PBS), 0.05% Triton X-100, and 5% bovine serum albumin (BSA), and immunoblotting was performed with a biotinylated Nessy-1 monoclonal anti-human IL-33 antibody (1:1,000 dilution; Alexis) and visualized by enhanced chemiluminescence (Amersham Pharmacia Biotech, Dübendorf, Switzerland) using streptavidin– horseradish peroxidase (1:20,000 dilution; R&D Systems, Abingdon, UK). Immunofluorescence. Human synovial fibroblasts were plated on chamber slides and stimulated with IL-1␤ (1 ng/ml) and TNF␣ (10 ng/ml) for 48 hours. Cells were fixed for 10 minutes (with PBS, 4% paraformaldehyde, and RT), permeabilized for 10 minutes (with PBS and 0.1% Triton X-100), and blocked for 1 hour (with PBS, 1% BSA, and 10% normal goat serum) before staining with the Nessy-1 antibody (1:100 dilution) and detection with a phycoerythrin-labeled anti-mouse IgG1 secondary antibody. To assess staining specificity, negative controls were performed in the absence of the primary antibody. Slides were mounted using a mounting medium containing 4⬘,6-diamidino-2-phenylindole (Vector). Assessment of binding and in vitro blocking activity of monoclonal anti-ST2 antibody. A rat monoclonal anti-murine ST2 antibody directed against the extracellular domain of recombinant murine ST2 was produced by Amgen. P815 cells were cultured in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum (FBS) and penicillin/streptomycin. To investigate anti-ST2 antibody binding specificity, P815 cells were stained in PBS and 1% FBS with biotinylated anti-ST2 antibody (10 g/ml) in the presence or absence of unlabeled anti-ST2 antibody (10-fold excess). To evaluate the blocking activity of the anti-ST2 antibody, P815 cells were stimulated with recombinant murine IL-33 (10 ng/ml; R&D Systems) for 72 hours before IL-6 production was assessed by enzymelinked immunosorbent assay (ELISA; R&D Systems). The monoclonal anti-ST2 antibody or an isotype-matched control antibody was added to the cells at the indicated concentrations 30 minutes before stimulation with IL-33. Assessment of in vivo blocking activity of monoclonal anti-ST2 antibody. Female BALB/c mice (age ⬃4 months) were injected intraperitoneally (IP) with 250 g of monoclonal anti-ST2 antibody, mouse sST2-Fc (produced at Amgen), or control rat IgG (5 mice/group). Two hours later, mice were challenged intranasally with 200 ng of recombinant murine PALMER ET AL IL-33 or mouse serum albumin (MSA) as a control protein. Mice were killed 24 hours later, and IL-5 levels in the bronchoalveolar lavage (BAL) fluid were assessed by ELISA (R&D Systems). Induction and assessment of CIA. CIA was induced in male DBA/1 mice as described elsewhere (25), except that for the booster injection on day 21, 100 g type II collagen (CII) in PBS was given IP. Monoclonal rat IgG1 anti-murine ST2 antibody or isotype-matched control antibody M139 (rat IgG1 anti–keyhole limpet hemocyanin; Amgen) was injected (150 g/mouse IP) every 2–3 days starting on day 21. From this time point onward, arthritis severity in each paw was assessed by clinical scoring on a 3-point scale (25). Mice were killed on day 42 after the first immunization. RNase protection assay and RNA expression profiling of arthritic mouse joints. Ankle joints from mice with CIA were homogenized in TRIzol reagent for extraction of total RNA. IL-33 and GAPDH mRNA expression was analyzed by RNase protection assay. A probe for murine IL-33 corresponding to bp 1–264 of the IMAGE3593927 expressed sequence tag complementary DNA (cDNA) clone (GenBank accession no. BC003847) was cloned into Bluescript SK⫹ (Stratagene, La Jolla, CA). The probe for murine GAPDH has been described previously (26). Linearized probes were transcribed with T7 RNA polymerase (Promega, Madison, WI) to synthesize 32Plabeled riboprobes complementary to IL-33 and GAPDH mRNA. The riboprobes were purified, and 2 ⫻ 105 counts per minute of each probe was hybridized to 10 g of total RNA in 80% formamide, 40 mM PIPES (pH 6.7), 400 mM NaCl, and 1 mM EDTA before digestion with RNases A and T1. The RNase-protected probes were resolved on denaturing polyacrylamide gels, imaged by autoradiography, and quantified by phosphorimaging (Cyclone Storage Phosphor System; PerkinElmer, Zaventem, Belgium). IL-33 levels were normalized to GAPDH expression. Quantitative PCR profiling of cDNA synthesized from joint tissue mRNA was performed using a custom low-density array and standard TaqMan reagents (Applied Biosystems, Foster City, CA). Expression of 62 genes (see Supplementary Table 1, available on the Arthritis & Rheumatism web site at http://www3.interscience.wiley.com/journal/76509746/home) was measured simultaneously, and expression levels were normalized to murine hypoxanthine guanine phosphoribosyltransferase. Histologic grading of arthritis. Knee joints were fixed in 10% buffered formalin, decalcified in 15% EDTA, dehydrated, and embedded in paraffin. Sagittal sections (5 m) were stained with hematoxylin and eosin or toluidine blue and graded by a pathologist (CAS) in a blinded manner. Cartilage erosion, joint destruction, and inflammation including “pannus” formation were scored on a scale of 0–4 (0 ⫽ normal, 1 ⫽ minimal, 2 ⫽ moderate, 3 ⫽ severe, and 4 ⫽ very severe) (27). The amount of polymorphonuclear neutrophils was scored on a scale of 0–4 (0 ⫽ no neutrophils present and 4 ⫽ maximal neutrophil infiltration). T cell proliferation assay. Draining lymph node (LN) cells were harvested and seeded at 4 ⫻ 105/well in 96-well plates in 200 l RPMI 1640 medium containing 100 IU/ml of penicillin, 100 g/ml of streptomycin, 5 ⫻ 10 ⫺5 M ␤-mercaptoethanol, and 1% heat-inactivated mouse serum. Cells were incubated for 72 hours without or with 100 g/ml ROLE OF IL-33 IN ARTHRITIS 741 Figure 1. Analysis of interleukin-33 (IL-33) expression in human synovium. A and B, Expression of the IL-33 protein was assessed by immunohistochemistry in human rheumatoid arthritis (RA) (A) and normal (B) synovium. IL-33 staining (brown) was detected in endothelial cells (EC) both in inflamed and in normal synovium. In addition, in RA synovium, IL-33 immunoreactivity was observed in high endothelial venule (HEV) ECs in lymphoid aggregates (asterisks), in synovial lining cells (LC), in cells morphologically consistent with fibroblasts (solid arrowheads), and in cells resembling mononuclear inflammatory cells (open arrowheads) in the sublining. C, IL-33–expressing cells were further characterized by double staining. Left, Human RA synovium was stained for IL-33 (brown) and for vascular and lymphatic EC markers (red). Some examples of double-positive IL-33–expressing ECs (solid arrowheads) and of nonendothelial IL-33–positive cells (open arrowheads) are indicated. Right, Human RA synovium was stained for IL-33 (brown) and for CD68 to label inflammatory cells (red). Some examples of double-positive IL-33–expressing inflammatory cells (solid arrowheads), CD68-negative IL-33–expressing cells (open arrowheads), and IL-33–negative inflammatory cells (asterisks) are indicated. Staining specificity was assessed using isotype-matched antibodies as negative controls. Original magnifications are indicated. 742 PALMER ET AL CII or 5 g/ml of concanavalin A (Con A; Amersham Pharmacia Biotech). During the final 18 hours of incubation, 3 H-thymidine was added at 1 Ci/well. Determination of cytokine and antibody production. Draining LN cells were harvested at the time mice were killed and cultured for 72 hours without or with 100 g/ml of CII or 5 g/ml of Con A. Culture supernatants were harvested, and levels of interferon-␥ (IFN␥), IL-17, TNF␣, and IL-10 were quantified by ELISA using DuoSet ELISA Development Systems (R&D Systems). The detection limits were 16 pg/ml for IL-17 and TNF␣ and 31 pg/ml for IFN␥ and IL-10. Serum levels of anti-CII total IgG, IgG2a, and IgG1 were measured as described previously (25). Statistical analysis. Significance of differences was calculated by one-way analysis of variance with Fisher’s protected least significant difference test or Dunnett’s multiple comparison test, as indicated. To assess differences in the longitudinal evolution of arthritis outcomes, we used generalized linear mixed models for repeated measures (28). The tests were conducted at a 2-sided alpha level of 0.05 and performed with Stata version 9.2 for Windows (StataCorp, College Station, TX). A difference between experimental groups was considered significant when the P value was less than 0.05. RESULTS IL-33 is expressed in human synovial tissue, in cultured human RA fibroblasts, and in arthritic mouse joints. Expression of the IL-33 protein was assessed by immunohistochemistry in human RA and normal synovium (Figures 1A and B). IL-33 staining was detected in ECs both in inflamed and in normal synovial tissue. However, the RA sample contained an increased number of blood vessels displaying strong IL-33 expression in EC nuclei. The RA sample also showed an increased number of cells with immunoreactive nuclei and cytoplasm that were morphologically consistent with fibroblasts, as well as immunoreactive cells resembling mononuclear inflammatory cells. Consistently, double staining of RA synovial sections for IL-33 and EC markers clearly demonstrated the presence of other IL-33– expressing cells in addition to ECs (Figure 1C). Furthermore, double staining for IL-33 and CD68 identified a subset of CD68-positive inflammatory cells as an additional source of IL-33 (Figure 1C). Finally, synovial lining cells in the RA sample also showed a light cytoplasmic staining, but it was only slightly darker than the isotype control and was therefore equivocal. Human RA synovial fibroblasts constitutively expressed low levels of IL-33 mRNA and protein. IL-33 mRNA and protein expression were strongly increased after treatment of the cells with IL-1␤ and/or TNF␣ (Figures 2A and B). In cell lysates, predominantly Figure 2. Interleukin-33 (IL-33) expression in human synovial fibroblasts and arthritic mouse joints. A, IL-33 mRNA levels in synovial fibroblasts that were left unstimulated (open bar) or stimulated for 6 hours with 1 ng/ml IL-1␤ (gray bar), 10 ng/ml tumor necrosis factor ␣ (TNF␣; hatched bar), or IL-1␤ and TNF␣ (solid bar) were analyzed by real-time reverse transcriptase–polymerase chain reaction. Values are the mean and SEM of data obtained using 3 independent cultures. ⴱ ⫽ P ⬍ 0.05 versus unstimulated synovial fibroblasts. B, Synovial fibroblasts were left unstimulated or were stimulated with IL-1␤ and TNF␣ for 24 hours before IL-33 protein expression in cell lysates was analyzed by Western blotting. Lane 1, Recombinant mature human IL-33; lane 2, unstimulated synovial fibroblast lysate; lane 3, IL-1␤/TNF␣–stimulated synovial fibroblast lysate. C, Localization of IL-33 in unstimulated (left) or IL-1␤/TNF␣–stimulated (right) synovial fibroblasts was analyzed by immunostaining. Phycoerythrin-labeled anti–IL-33 staining (red) was detected mainly in cell nuclei, as visualized with 4⬘,6-diamidino2-phenylindole (blue) (original magnification ⫻ 200). D, IL-33 mRNA expression in the paws of naive and arthritic mice during early (days 1–8) or late (days 14–23) collagen-induced arthritis (CIA) was quantified in 1 paw from a naive mouse, 4 unaffected paws from arthritic mice, 5 arthritic paws from mice with early CIA, and 5 arthritic paws from mice with late CIA. Values are the mean and SEM. ⴱ ⫽ P ⬍ 0.05 versus unaffected paws and versus arthritic paws of mice with late CIA, by one-way analysis of variance with Fisher’s protected least significant difference test. GUSB ⫽ ␤-glucuronidase; AU ⫽ arbitrary units. ROLE OF IL-33 IN ARTHRITIS unprocessed 30-kd proIL-33 was detected by Western blotting (Figure 2B). Immunocytochemistry showed that this proIL-33 localized predominantly in the cell nucleus (Figure 2C). As a negative control, no staining was detected in IL-1␤– and TNF␣-stimulated synovial fibroblasts in the absence of the anti–IL-33 antibody (data not shown). In contrast to our findings in synovial fibroblasts, we could not detect IL-33 expression in primary human articular chondrocytes (data not shown). We also investigated whether human resident joint cells express the T1/ST2 receptor, which would make them responsive to IL-33. Human synovial fibroblasts expressed mRNA encoding short sST2, but not the long signaling T1/ST2 isoform (data not shown). Consistently, treatment with recombinant mature IL-33 enhanced neither IL-6 secretion nor IL-33 mRNA expression in synovial fibroblasts (data not shown). Similarly, we could detect neither T1/ST2 expression nor responsiveness to IL-33 in human articular chondrocytes. We next used an RNase protection assay to investigate the expression of IL-33 mRNA in mice with CIA. IL-33 expression was increased during the early, inflammatory phase of CIA (Figure 2D). A blocking antibody against the T1/ST2 receptor decreases the severity of CIA. To investigate the role of IL-33 and T1/ST2 in experimental arthritis, we tested a potential therapeutic effect of a blocking anti-ST2 antibody during CIA. The anti-ST2 antibody was generated by immunizing rats with the extracellular domain of recombinant mouse ST2. The resulting monoclonal antibody recognizes recombinant ST2 protein in platebased assays (data not shown), as well as cell surface ST2 as assessed by flow cytometry of ST2-positive P815 cells (Figure 3A). The potency of this blocking anti-ST2 antibody was first assessed in vitro by examining its ability to inhibit IL-6 secretion induced by IL-33 in P815 mastocytoma cells (Figure 3B). In this assay, anti-ST2 effectively inhibited the biologic activity of IL-33, with an estimated 50% inhibition concentration of 470 pM. The blocking ability of the anti-ST2 antibody was comparable with that of a well-described monoclonal antiST2 antibody DJ8 (data not shown) (29). Next, we evaluated the ability of anti-ST2 to inhibit the biologic effects of recombinant mature IL-33 in vivo. Mice were treated IP with anti-ST2, sST2-Fc, or control rat IgG prior to pulmonary challenge with IL-33 or MSA (Figure 3C). Intranasal instillation of IL-33 led to increased IL-5 levels in BAL fluid, and treatment of mice with anti-ST2 antibody or sST2-Fc efficiently in- 743 Figure 3. Characterization of a blocking monoclonal anti-ST2 antibody. A, Binding of the anti-ST2 antibody to cell surface ST2 as assessed by flow cytometry on P815 cells in the presence or absence of unlabeled anti-ST2 antibody. An unrelated isotype-matched antibody was used as a negative control. B, Potency of the anti-ST2 antibody in vitro. P815 cells were stimulated with 10 ng/ml recombinant mature interleukin-33 (IL-33) for 72 hours before IL-6 production was assessed in culture supernatants. Monoclonal anti-ST2 antibody (Ab) or an isotype-matched control antibody was added to the cells at the indicated final concentrations 30 minutes prior to stimulation with IL-33. Experiments were performed in triplicate for each condition; results are representative of 3 independent experiments. C, Efficacy of the anti-ST2 antibody in vivo. IL-5 levels were assessed in bronchoalveolar lavage fluid obtained from mice 24 hours after intranasal instillation of recombinant mature IL-33 (200 ng/mouse). BALB/c mice (n ⫽ 5 per group) were injected intraperitoneally (250 g/mouse) with an isotype-matched control antibody 2 hours before intranasal instillation of mouse serum albumin (MSA) or with an isotypematched control antibody, the monoclonal anti-ST2 antibody, or soluble ST2 (sST2)–Fc 2 hours before intranasal instillation of IL-33. ⴱⴱ ⫽ P ⬍ 0.01; ⴱⴱⴱ ⫽ P ⬍ 0.001 versus isotype control plus IL-33, by one-way analysis of variance with Dunnett’s multiple comparison test. IC50 ⫽ 50% inhibition concentration. hibited the effect of IL-33 in this experimental system. Taken together, these results demonstrate that the 744 PALMER ET AL Figure 4. Treatment with anti-ST2 antibody decreases the severity of collagen-induced arthritis. A, Incidence of arthritis in anti-ST2 antibody– treated mice (n ⫽ 10) and isotype-matched control antibody–treated mice (n ⫽ 10). The onset of arthritis was significantly retarded in anti-ST2 antibody–treated mice (P ⬍ 0.05 by longitudinal model for binomial data). B, Clinical severity scores in anti-ST2 antibody–treated mice and control antibody–treated mice. Values are the mean and SEM. Anti-ST2 antibody treatment significantly decreased the evolution of severity scores (P ⫽ 0.01 by mixed model for repeated measures) and disease severity at the end of the experiment (ⴱ ⫽ P ⬍ 0.05 versus control antibody treatment). C, Number of affected paws in anti-ST2–treated mice and control antibody–treated mice. Values are the mean and SEM. The number of affected paws tended to be lower in anti-ST2 antibody–treated mice (P ⫽ 0.11 by longitudinal model for ordinal data). D, Severe inflammation, pannus formation, cartilage infiltration, and joint destruction in an isotype-matched control antibody–treated mouse, but not in an anti-ST2 antibody– treated mouse (hematoxylin and eosin stained; original magnifications are indicated). E, Histologic scores for inflammation, cartilage erosion, and neutrophil infiltration. Scores in anti-ST2 antibody–treated mice (n ⫽ 10) (solid bars) and control antibody–treated mice (n ⫽ 10) (open bars) are the mean and SEM. ⴱ ⫽ P ⬍ 0.05 versus control antibody–treated mice, by one-way analysis of variance with Fisher’s protected least significant difference test. ROLE OF IL-33 IN ARTHRITIS monoclonal anti-ST2 antibody is an effective inhibitor of IL-33 activity in vitro and in vivo. We then tested the effect of the anti-ST2 antibody in CIA. Treatment of mice with blocking anti-ST2 antibody, initiated at the onset of arthritis, slightly delayed the incidence of the disease (Figure 4A). In addition, the evolution of clinical severity scores was significantly reduced in anti-ST2 antibody–treated mice (P ⫽ 0.01) (Figure 4B), and the clinical severity at the end of the experiment was significantly lower in anti-ST2 antibody–treated mice compared with control-treated mice. The number of affected paws also tended to be lower in the anti-ST2 antibody–treated group, although the difference was not statistically significant (P ⫽ 0.11) (Figure 4C). Histologic scoring of the joints at the end of the experiment confirmed decreased severity of arthritis and indicated in particular decreased inflammation and cartilage erosion in anti-ST2 antibody–treated mice compared with isotype control–treated mice (Figures 4D and E). Neutrophil infiltration also tended to be reduced, although the reduction was not statistically significant. A blocking antibody against the T1/ST2 receptor decreases T cell IFN␥ production during CIA. The in vitro proliferative response of draining LN cells to CII was slightly, but significantly, reduced in cells isolated from anti-ST2 antibody–treated mice compared with isotype-matched control antibody–treated mice (Figure 5A). Interestingly, despite this rather modest effect on cell proliferation, IFN␥ production was almost completely abrogated in draining LN cells isolated from anti-ST2 antibody–treated mice (Figure 5B). IL-17, TNF␣, and IL-10 production by draining LN cells was only slightly reduced in cells isolated from anti-ST2 antibody–treated mice. The effect of anti-ST2 treatment on IFN␥ production was antigen specific, since Con A–induced IFN␥ production was not impaired in draining LN cells isolated from anti-ST2 antibody–treated mice (mean ⫾ SEM IFN␥ concentration 11.02 ⫾ 0.62 ng/ml versus 15.87 ⫾ 0.30 ng/ml in cells from isotypematched control antibody–treated mice versus those from anti-ST2 antibody–treated mice, respectively; P ⬍ 0.05). Furthermore, this inhibitory effect could not be reproduced when the anti-ST2 antibody (50 g/ml) was added in vitro during restimulation with antigen to draining LN cells isolated from immunized mice that had not been treated with the anti-ST2 antibody in vivo (data not shown). Finally, anticollagen IgG titers (total IgG, IgG2a, and IgG1) were not different in anti-ST2 antibody–treated mice and control antibody–treated mice (data not shown). 745 Figure 5. Anti-ST2 antibody treatment decreases interferon-␥ (IFN␥) production by draining lymph node (LN) cells. A, Proliferation of draining LN cells isolated from anti-ST2 antibody–treated mice (n ⫽ 10) and isotype-matched control antibody–treated mice (n ⫽ 10). Cells were stimulated in vitro with 100 g/ml type II collagen (CII) or were left unstimulated. Values are the mean and SEM. Proliferation was increased by stimulation with CII (ⴱ ⫽ P ⬍ 0.05 versus unstimulated cultures). The proliferative response to CII was slightly reduced in cells from anti-ST2 antibody–treated mice (& ⫽ P ⬍ 0.05 versus cells from control antibody–treated mice with similar stimulation). B, IFN␥, interleukin-17 (IL-17), tumor necrosis factor ␣ (TNF␣), and IL-10 production by draining LN cells isolated from anti-ST2 antibody–treated mice (n ⫽ 10) and control antibody–treated mice (n ⫽ 10). Cells were stimulated in vitro with 100 g/ml CII or were left unstimulated. Values are the mean and SEM. Cytokine production was increased by stimulation with CII (ⴱ ⫽ P ⬍ 0.05 versus unstimulated cultures). The IFN␥ response to CII was markedly inhibited in cells from anti-ST2 antibody–treated mice. IL-17, IL-10, and TNF␣ responses were also slightly reduced (& ⫽ P ⬍ 0.05 versus cells from control antibody– treated mice with similar stimulation, by one-way analysis of variance with Fisher’s protected least significant difference test). A blocking antibody against the T1/ST2 IL-33 receptor decreases joint RANKL mRNA expression during CIA. We analyzed the expression of 62 different transcripts (see Supplementary Table 1, available on the 746 PALMER ET AL Figure 6. Treatment with anti-ST2 antibody decreases joint RANKL mRNA expression. Expression of 62 different inflammation-related gene transcripts was assessed in the ankle joints of anti-ST2 antibody–treated mice (n ⫽ 10) and isotype-matched control antibody–treated mice (n ⫽ 10), as described in Materials and Methods. Expression of RANKL, tumor necrosis factor ␣ (TNF␣), matrix metalloproteinase 9 (MMP-9), and inducible nitric oxide synthase (iNOS) was reduced in anti-ST2–treated animals. Each symbol represents a single mouse. Horizontal lines show the mean. RANKL mRNA expression was significantly reduced in anti-ST2 antibody–treated mice. Differences in expression of TNF␣, MMP-9, and iNOS did not reach statistical significance. All P values were assessed by one-way analysis of variance with Dunnett’s multiple comparison test. Arthritis & Rheumatism web site at http://www3. interscience.wiley.com/journal/76509746/home) by quantitative PCR of the total RNA isolated from ankle joints of anti-ST2 antibody–treated mice and isotype control antibody–treated mice. Among the transcripts examined, we observed a significant decrease in expression of RANKL as well as a trend toward decreased expression of inflammation-related genes, such as those for TNF␣, matrix metalloproteinase 9 (MMP-9), and inducible nitric oxide synthase (iNOS), although the differences between the groups did not reach statistical significance (Figure 6). DISCUSSION We report that IL-33, a novel member of the IL-1 family of cytokines, is expressed in human RA synovium. Consistent with previous observations, ECs appear to be a major source of IL-33 in the inflamed synovium (21). Interestingly, we also observed expression of nuclear IL-33 in ECs in normal synovium, suggesting constitutive IL-33 production by this cell type. Moreover, IL-33 staining indicated that cells resembling synovial fibroblasts, as well as a subset of CD68-positive mononuclear inflammatory cells, constitute additional cellular sources of IL-33 in the rheumatoid synovium. Consistently, we observed that cultured synovial fibroblasts produced IL-33 and that its expression was increased in response to IL-1␤ and/or TNF␣, 2 common proinflammatory mediators present in the rheumatoid synovial tissue and fluid. IL-33 mRNA was also detected in the joints of mice with CIA during the early phase of the inflammatory process. The kinetics of IL-33 mRNA production are similar to those found previously for other cytokines showing that, for example, IL-1␤ levels peak during the ROLE OF IL-33 IN ARTHRITIS early inflammatory phase of the disease, while expression of antiinflammatory cytokines predominates later in the disease process (30). The administration of a monoclonal antibody that block ST2 signaling, which was initiated at the time of arthritis onset, attenuated the severity of arthritis and the signs of structural damage. In addition, anti-ST2 antibody treatment was associated with a marked decrease in IFN␥ production as well as with a significant, but more limited, reduction in IL-17 production by ex vivo–stimulated draining LN cells. Finally, anti-ST2 treatment also decreased the expression of RANKL mRNA in the joints. These findings indicate that IL-33 is produced locally in inflamed joints and that neutralization of IL-33 signaling has a therapeutic effect on the course of arthritis. Overall, these results are consistent with those of a previous study demonstrating a protective effect of sST2 in CIA (20). Although the previous study explored a different hypothesis and did not address the role of IL-33 specifically, in retrospect, the biologic effects of sST2 could be attributed to IL-33 neutralization. The authors found that administration of an sST2-Fc fusion protein at disease onset significantly attenuated the severity of arthritis. In addition, cellular infiltration, synovial hyperplasia, and joint erosion were markedly reduced in the joints of sST2-Fc–treated mice compared with controls. Similar to our findings, treatment with sST2-Fc also modulated the immune response, decreasing spleen cell IFN␥, TNF␣, IL-6, and IL-12 production upon restimulation with CII in vitro. Finally, sST2 treatment was reported to down-regulate serum levels of IL-6, IL-12, and TNF␣. In the present experiment, we tried to assess serum levels of IFN␥ and IL-17 at the time that mice were killed; however, the levels of both cytokines were below the detection limits of the ELISAs. Taken together, the results of these 2 studies strongly suggest a role for IL-33 in the pathogenesis of arthritis. ST2 is known as a selective marker of Th2 lymphocytes (5), and studies using anti-ST2 antibodies or sST2-Fc demonstrated an important role for this receptor in Th2 responses, although there were conflicting results regarding the contribution of this receptor to Th2 effector functions in ST2-deficient mice (31–35). Systemic administration of recombinant IL-33 to mice induced prototypical Th2 responses, with increased levels of IL-4, IL-5, and IL-13, enhanced IgE and IgA production, and pathologic changes in the lungs and digestive tract (1). However, our findings of a marked decrease in IFN␥ production by draining LN cells from 747 anti-ST2 antibody–treated mice, as well as the decreased production of IFN␥ and TNF␣ by ex vivo–stimulated spleen cells reported after administration of ST2-Fc in mice with CIA, suggest that IL-33 is also involved in the development of Th1 responses (20). In addition, IL-33 was recently reported to promote the production of IFN␥ by natural killer (NK) cells in a model of parasitic infection in SCID mice (36). Furthermore, in human Th2, NK, and invariant NK T cells, IL-33 stimulated the production of IFN␥ in an antigen-dependent and -independent manner (37). Taken together, these results indicate that IL-33 can induce both Th1 and Th2 responses according to the stimulation, the cell type, and the cytokine environment involved. These findings are reminiscent of previous observations concerning IL-18, a cytokine originally described as a potent inducer of IFN␥ by Th1 cells, which is also able to stimulate Th2 responses depending on the system examined (38). IL-17 production by draining LN cells was also somewhat reduced in anti-ST2–treated mice, suggesting that, like other IL-1 cytokine family members, IL-33 can enhance Th17 responses (39–41). However, it is unknown whether IL-33 is able to stimulate the production of IL-17 directly or whether IL-33 acts indirectly through the induction of cytokines such as IL-1 and IL-6. Interestingly, in separate studies we have observed that systemic administration of IL-33 leads to increases in both IL-1 and IL-6 levels in C57BL/6 and BALB/c mice (Toy D, Smith DE: unpublished observations). RANKL mRNA expression was reduced in the joints of anti-ST2–treated mice, suggesting that anti-ST2 treatment affects local as well as systemic responses and may beneficially affect RANK-mediated bone erosions. In addition, expression of several inflammation-related genes, such as TNF␣, MMP-9, and iNOS, also tended to be lower in anti-ST2–treated mice. Like other members of the IL-1 family, such as IL-1␣, IL-1␤, and IL-18, IL-33 is expressed as a propeptide and lacks a typical leader sequence for protein secretion. Our results indicate that, like ECs, synovial fibroblasts express predominantly the pro form of IL-33 and that it is primarily detected in cell nuclei (21,22). Potential nuclear functions of IL-33 in these cells are as yet unclear. ProIL-1␣ is also mainly detected in cell nuclei, where it has been shown to exert several activities independent of receptor binding, such as regulation of cell proliferation, senescence, motility, apoptosis, and inflammatory responses (23,42–44). IL-33 may thus exert different biologic functions both inside and outside the cells, as previously shown for proIL-1␣ or for high 748 PALMER ET AL mobility group box chromosomal protein 1 (45). The ability of an antireceptor antibody to ameliorate signs of disease suggests that at least some of the inflammation is driven by extracellular IL-33. In conclusion, we report that IL-33, a novel member of the IL-1 family of cytokines, is expressed in human RA synovium and synovial fibroblasts and in the joints of mice with CIA. Administration of a blocking anti-ST2 antibody attenuated the severity of CIA and was associated with a marked decrease in IFN␥ production by draining LN cells. Our findings indicate that IL-33 is produced locally in inflamed joints and that neutralization of IL-33 signaling has a therapeutic effect on the course of arthritis. ACKNOWLEDGMENTS We would like to thank Philippe Henchoz and David Swart for their expert technical assistance, Jim Rottman and Kim Merrimam for human immunohistochemistry data, and Thomas McKee and Patricia Gindre for their help with double immunostainings. AUTHOR CONTRIBUTIONS Dr. Gabay had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Palmer, Gabay. Acquisition of data. Palmer, Talabot-Ayer, Lamacchia, Toy, Seemayer, Viatte. Analysis and interpretation of data. Palmer, Talabot-Ayer, Lamacchia, Toy, Seemayer, Viatte, Finckh, Smith, Gabay. Manuscript preparation. Palmer, Finckh, Smith, Gabay. Statistical analysis. Palmer, Toy, Finckh. REFERENCES 1. Schmitz J, Owyang A, Oldham E, Song Y, Murphy E, McClanahan TK, et al. IL-33, an interleukin-1-like cytokine that signals via the IL-1 receptor-related protein ST2 and induces T helper type 2-associated cytokines. Immunity 2005;23:479–90. 2. Chackerian AA, Oldham ER, Murphy EE, Schmitz J, Pflanz S, Kastelein RA. IL-1 receptor accessory protein and ST2 comprise the IL-33 receptor complex. J Immunol 2007;179:2551–5. 3. Ali S, Huber M, Kollewe C, Bischoff SC, Falk W, Martin MU. IL-1 receptor accessory protein is essential for IL-33-induced activation of T lymphocytes and mast cells. Proc Natl Acad Sci U S A 2007;104:18660–5. 4. Palmer G, Lipsky BP, Smithgall MD, Meininger D, Siu S, TalabotAyer D, et al. The IL-1 receptor accessory protein (AcP) is required for IL-33 signaling and soluble AcP enhances the ability of soluble ST2 to inhibit IL-33. Cytokine 2008;42:358–64. 5. Trajkovic V, Sweet MJ, Xu D. T1/ST2—an IL-1 receptor-like modulator of immune responses. Cytokine Growth Factor Rev 2004;15:87–95. 6. Moritz DR, Rodewald HR, Gheyselinck J, Klemenz R. The IL-1 receptor-related T1 antigen is expressed on immature and mature mast cells and on fetal blood mast cell progenitors. J Immunol 1998;161:4866–74. 7. Allakhverdi Z, Smith DE, Comeau MR, Delespesse G. Cutting edge: The ST2 ligand IL-33 potently activates and drives maturation of human mast cells. J Immunol 2007;179:2051–4. 8. Iikura M, Suto H, Kajiwara N, Oboki K, Ohno T, Okayama Y, et al. IL-33 can promote survival, adhesion and cytokine production in human mast cells. Lab Invest 2007;87:971–8. 9. Ho LH, Ohno T, Oboki K, Kajiwara N, Suto H, Iikura M, et al. IL-33 induces IL-13 production by mouse mast cells independently of IgE-FcRI signals. J Leukoc Biol 2007;82:1481–90. 10. Moulin D, Donze O, Talabot-Ayer D, Mezin F, Palmer G, Gabay C. Interleukin (IL)-33 induces the release of pro-inflammatory mediators by mast cells. Cytokine 2007;40:216–25. 11. Hayakawa H, Hayakawa M, Kume A, Tominaga SI. Soluble ST2 blocks IL-33 signaling in allergic airway inflammation. J Biol Chem 2007;282:26369–80. 12. Fraser A, Moore M, Jongbloed S, Gracie A, McInnes IB. Elevated soluble ST2 and cytokine levels in synovial fluids of patients with inflammatory synovitis. Ann Rheum Dis 2006;65 Suppl 1:A10. 13. Brint EK, Xu D, Liu H, Dunne A, McKenzie AN, O’Neill LA, et al. ST2 is an inhibitor of interleukin 1 receptor and Toll-like receptor 4 signaling and maintains endotoxin tolerance. Nat Immunol 2004;5:373–9. 14. Sweet MJ, Leung BP, Kang D, Sogaard M, Schulz K, Trajkovic V, et al. A novel pathway regulating lipopolysaccharide-induced shock by ST2/T1 via inhibition of Toll-like receptor 4 expression. J Immunol 2001;166:6633–9. 15. Yin H, Huang BJ, Yang H, Huang YF, Xiong P, Zheng F, et al. Pretreatment with soluble ST2 reduces warm hepatic ischemia/ reperfusion injury. Biochem Biophys Res Commun 2006;351: 940–6. 16. Fagundes CT, Amaral FA, Souza AL, Vieira AT, Xu D, Liew FY, et al. ST2, an IL-1R family member, attenuates inflammation and lethality after intestinal ischemia and reperfusion. J Leukoc Biol 2007;81:492–9. 17. Lee DM, Friend DS, Gurish MF, Benoist C, Mathis D, Brenner MB. Mast cells: a cellular link between autoantibodies and inflammatory arthritis. Science 2002;297:1689–92. 18. Nigrovic PA, Binstadt BA, Monach PA, Johnsen A, Gurish M, Iwakura Y, et al. Mast cells contribute to initiation of autoantibody-mediated arthritis via IL-1. Proc Natl Acad Sci U S A 2007;104:2325–30. 19. Xu D, Chan WL, Leung BP, Huang F, Wheeler R, Piedrafita D, et al. Selective expression of a stable cell surface molecule on type 2 but not type 1 helper T cells. J Exp Med 1998;187:787–94. 20. Leung BP, Xu D, Culshaw S, McInnes IB, Liew FY. A novel therapy of murine collagen-induced arthritis with soluble T1/ST2. J Immunol 2004;173:145–50. 21. Carriere V, Roussel L, Ortega N, Lacorre DA, Americh L, Aguilar L, et al. IL-33, the IL-1-like cytokine ligand for ST2 receptor, is a chromatin-associated nuclear factor in vivo. Proc Natl Acad Sci U S A 2007;104:282–7. 22. Baekkevold ES, Roussigne M, Yamanaka T, Johansen FE, Jahnsen FL, Amalric F, et al. Molecular characterization of NF-HEV, a nuclear factor preferentially expressed in human high endothelial venules. Am J Pathol 2003;163:69–79. 23. Werman A, Werman-Venkert R, White R, Lee JK, Werman B, Krelin Y, et al. The precursor form of IL-1␣ is an intracrine proinflammatory activator of transcription. Proc Natl Acad Sci U S A 2004;101:2434–9. 24. Klareskog L, Forsum U, Malmnas Tjernlund UK, Kabelitz D, Wigren A. Appearance of anti-HLA-DR-reactive cells in normal and rheumatoid synovial tissue. Scand J Immunol 1981;14:183–92. 25. Palmer G, Chobaz V, Talabot-Ayer D, Taylor S, So A, Gabay C, et al. Assessment of the efficacy of different statins in murine collagen-induced arthritis. Arthritis Rheum 2004;50:4051–9. 26. Gabay C, Porter B, Fantuzzi G, Arend WP. Mouse IL-1 receptor antagonist isoforms: complementary DNA cloning and protein ROLE OF IL-33 IN ARTHRITIS 27. 28. 29. 30. 31. 32. 33. 34. 35. expression of intracellular isoform and tissue distribution of secreted and intracellular IL-1 receptor antagonist in vivo. J Immunol 1997;159:5905–13. Camps M, Ruckle T, Ji H, Ardissone V, Rintelen F, Shaw J, et al. Blockade of PI3K␥ suppresses joint inflammation and damage in mouse models of rheumatoid arthritis. Nat Med 2005;11:936–43. Rabe-Hesketh S, Skrondal A. Higher level models and nested random effects. In: Multilevel and longitudinal modeling using Stata. 1st ed. College Station (TX): Stata Press; 2005. Moritz DR, Gheyselinck J, Klemenz R. Expression analysis of the soluble and membrane-associated forms of the interleukin-1 receptor-related T1 protein in primary mast cells and fibroblasts. Hybridoma 1998;17:107–16. Gabay C, Marinova-Mutafchieva L, Williams RO, Gigley JP, Butler DM, Feldmann M, et al. Increased production of intracellular interleukin-1 receptor antagonist type I in the synovium of mice with collagen-induced arthritis: a possible role in the resolution of arthritis. Arthritis Rheum 2001;44:451–62. Coyle AJ, Lloyd C, Tian J, Nguyen T, Erikkson C, Wang L, et al. Crucial role of the interleukin 1 receptor family member T1/ST2 in T helper cell type 2-mediated lung mucosal immune responses. J Exp Med 1999;190:895–902. Walzl G, Matthews S, Kendall S, Gutierrez-Ramos JC, Coyle AJ, Openshaw PJ, et al. Inhibition of T1/ST2 during respiratory syncytial virus infection prevents T helper cell type 2 (Th2)- but not Th1-driven immunopathology. J Exp Med 2001;193:785–92. Hoshino K, Kashiwamura S, Kuribayashi K, Kodama T, Tsujimura T, Nakanishi K, et al. The absence of interleukin 1 receptorrelated T1/ST2 does not affect T helper cell type 2 development and its effector function. J Exp Med 1999;190:1541–8. Townsend MJ, Fallon PG, Matthews DJ, Jolin HE, McKenzie AN. T1/ST2-deficient mice demonstrate the importance of T1/ST2 in developing primary T helper cell type 2 responses. J Exp Med 2000;191:1069–76. Mangan NE, Dasvarma A, McKenzie AN, Fallon PG. T1/ST2 expression on Th2 cells negatively regulates allergic pulmonary inflammation. Eur J Immunol 2007;37:1302–12. 749 36. Humphreys NE, Xu D, Hepworth MR, Liew FY, Grencis RK. IL-33, a potent inducer of adaptive immunity to intestinal nematodes. J Immunol 2008;180:2443–9. 37. Smithgall MD, Comeau MR, Yoon BR, Kaufman D, Armitage R, Smith DE. IL-33 amplifies both Th1- and Th2-type responses through its activity on human basophils, allergen-reactive Th2 cells, iNKT and NK cells. Int Immunol 2008;20:1019–30. 38. Hoshino T, Kawase Y, Okamoto M, Yokota K, Yoshino K, Yamamura K, et al. Cutting edge: IL-18-transgenic mice: in vivo evidence of a broad role for IL-18 in modulating immune function. J Immunol 2001;166:7014–8. 39. Acosta-Rodriguez EV, Napolitani G, Lanzavecchia A, Sallusto F. Interleukins 1␤ and 6 but not transforming growth factor-␤ are essential for the differentiation of interleukin 17-producing human T helper cells. Nat Immunol 2007;8:942–9. 40. Sutton C, Brereton C, Keogh B, Mills KH, Lavelle EC. A crucial role for interleukin (IL)-1 in the induction of IL-17-producing T cells that mediate autoimmune encephalomyelitis. J Exp Med 2006;203:1685–91. 41. Weaver CT, Harrington LE, Mangan PR, Gavrieli M, Murphy KM. Th17: an effector CD4 T cell lineage with regulatory T cell ties. Immunity 2006;24:677–88. 42. Pollock AS, Turck J, Lovett DH. The prodomain of interleukin 1␣ interacts with elements of the RNA processing apparatus and induces apoptosis in malignant cells. FASEB J 2003;17:203–13. 43. Palmer G, Trolliet S, Talabot-Ayer D, Mezin F, Magne D, Gabay C. Pre-interleukin-1␣ expression reduces cell growth and increases interleukin-6 production in SaOS-2 osteosarcoma cells: Differential inhibitory effect of interleukin-1 receptor antagonist (icIL1Ra1). Cytokine 2005;31:153–60. 44. Merhi-Soussi F, Berti M, Wehrle-Haller B, Gabay C. Intracellular interleukin-1 receptor antagonist type 1 antagonizes the stimulatory effect of interleukin-1␣ precursor on cell motility. Cytokine 2005;32:163–70. 45. Lotze MT, Tracey KJ. High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal. Nat Rev Immunol 2005;5:331–42.