Targeted mast cell silencing protects against joint destruction and angiogenesis in experimental arthritis in mice.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 56, No. 6, June 2007, pp 1806–1816 DOI 10.1002/art.22602 © 2007, American College of Rheumatology Targeted Mast Cell Silencing Protects Against Joint Destruction and Angiogenesis in Experimental Arthritis in Mice Manfred Kneilling,1 Lothar Hültner,2 Bernd J. Pichler,3 Reinhard Mailhammer,2 Lars Morawietz,4 Samuel Solomon,5 Martin Eichner,1 Joseph Sabatino,2 Tilo Biedermann,1 Veit Krenn,4 Wolfgang A. Weber,6 Harald Illges,7 Roland Haubner,8 and Martin Röcken1 mice, congenic mast cell–deficient KitW/KitW-v mice, or mast cell–deficient KitW/KitW-v mice reconstituted with mast cells, either by intraperitoneal or selective intraarticular injection. Angiogenesis was quantified in vivo by measuring activated ␣v␤3 integrin using 18F– galacto-RGD and positron emission tomography. In addition, staining of joint tissue with hematoxylin and eosin, Giemsa, ␤3, and ␣-actin was performed. The effect of mast cell stabilization by treatment with cromolyn or salbutamol was investigated in C57BL/6 or BALB/c mice. Results. Comparing wild-type mice, mast cell– deficient KitW/KitW-v mice, and mast cell–reconstituted KitW/KitW-v mice, we first showed that intraarticular and intraperitoneal mast cell engraftment fully restores susceptibility to antibody-induced arthritis, angiogenesis, and ␣v␤3 integrin activation. Importantly, selective mast cell silencing with either salbutamol or cromolyn prevented ␣v␤3 integrin activation, angiogenesis, and joint destruction. Conclusion. Mast cell engraftment fully restores susceptibility to ␣v␤3 integrin activation, angiogenesis, and joint destruction in GPI antibody–induced arthritis. Importantly, selective mast cell stabilization prevents ␣v␤3 integrin activation, angiogenesis, and joint destruction. Objective. Induction of arthritis with autoantibodies against glucose-6-phosphate isomerase (GPI) is entirely independent of T cells and B cells but is strictly dependent on the presence of mast cells. Here, we used this disease model to analyze whether exclusive intraarticular mast cell reconstitution is sufficient for disease induction and whether targeted mast cell silencing can prevent neoangiogenesis and joint destruction, 2 hallmarks of rheumatoid arthritis. Methods. Ankle swelling and clinical index scores were determined after injection of either K/BxN mouse– derived serum or control serum in wild-type Kitⴙ/Kitⴙ Supported by the DFG (grant Ro764/8), SFB 685, the Wilhelm Sander-Stiftung Foundation (grant 2005.043.I), and the European Union (grant MRTN-CT-2004-005693). Mr. Sabatino’s work was supported by a Fulbright stipend. Drs. Kneilling, Hültner, and Pichler contributed equally to this work. 1 Manfred Kneilling, MD, Martin Eichner, PhD, Tilo Biedermann, MD, Martin Röcken, MD: Eberhard Karls University, Tübingen, Germany; 2Lothar Hültner, VMD, MD, Reinhard Mailhammer, PhD, Joseph Sabatino: GSF⫺Institute of Clinical Molecular Biology and Tumor Genetics, Munich, Germany; 3Bernd J. Pichler, PhD: Technical University, Munich, and Eberhard Karls University, Tübingen, Germany; 4Lars Morawietz, MD, Veit Krenn, MD: Charite University Medicine Berlin, Berlin, Germany; 5Samuel Solomon, PhD: University of Konstanz, Konstanz, Germany; 6Wolfgang A. Weber, MD: Technical University, Munich, Germany; 7Harald Illges, PhD: University of Konstanz, Konstanz, and Fachhochschule Bonn-RheinSieg, Immunology and Cell Biology, Rheinbach, Germany; 8Roland Haubner, PhD: Technical University, Munich, Germany, and Universitätsklinik für Nuklearmedizin, Medizinische Universität Innsbruck, Innsbruck, Austria. Dr. Röcken has received consultancies, speaking fees, and/or honoraria (less than $10,000 each) from Hermal and Schering Berlin, and from Novartis (more than $10,000). Address correspondence and reprint requests to Martin Röcken, MD, Department of Dermatology, Eberhard Karls University, Liebermeisterstrasse 25, 72076 Tübingen, Germany. E-mail: email@example.com. Submitted for publication October 18, 2006; accepted in revised form February 13, 2007. Disease models of severe inflammatory autoimmune diseases reveal that neutrophil infiltration into sites of local inflammation and tissue destruction is critically dependent on mast cells. Thus, mast cells are involved in the initiation of experimental models of autoimmune encephalomyelitis (1), contact hypersensitivity (2), rheumatoid arthritis (RA) (3), or bacterial infection (4). These observations seem to be of great 1806 MAST CELL SILENCING IN RA relevance in human diseases, because large numbers of activated mast cells infiltrate tissue in the corresponding human diseases, such as allergic contact dermatitis, psoriasis, or RA (5–10). However, the relative contribution of local mast cells to the initiation of these diseases is enigmatic. Consequently, it remains undetermined whether therapeutic mast cell silencing can directly prevent tissue damage in these diseases. To address these questions, we used a disease model of RA induced in mice by anti–glucose-6phosphate isomerase (GPI) antibodies, because this RA model is entirely independent of T cells and B cells but is strictly dependent on mast cells (3,11,12). RA, one of the most common chronic autoimmune diseases, is estimated to affect 1% of the world population. RA is a destructive polyarthritis (13–15) that ultimately results in proliferation of synovial cells and severe damage of the synovial architecture, with pannus formation and destruction of cartilage and bone (16). K/BxN mice in which arthritis develops are the F1 generation of C57BL/6 mice, bearing a transgenic T cell receptor (TCR) specific for the bovine RNase peptide presented by I-Ak (KRN) crossed with diabetes-prone NOD mice. Between 3 and 5 weeks of age, K/BxN mice spontaneously develop a progressive joint-specific autoimmune disease that shares striking similarities with human RA, including spontaneous development, primarily in the distal joints. K/BxN mice have increased levels of messenger RNA (mRNA) expression of proinflammatory cytokines and hypergammaglobulinemia, and they develop autoreactive antibodies and succumb to a severe arthritis that leads to pronounced expansion of the synovium, pannus formation, and destruction of cartilage and bone. In transgenic mice, initiation of the disease depends on transgenic T cells. Subsequently, the disease becomes T cell–independent as adoptive transfer of immunoglobulins from diseased K/BxN mice induces an otherwise indistinguishable arthritis in healthy C57BL/6 mice. In the context of the NOD mouse–derived I-Ag7 class II major histocompatibility complex molecule, KRN T cells react with the ubiquitously expressed GPI, an enzyme that is involved in glycolysis (17–20). Low concentrations of GPI, released from apoptotic cells or from viable cells via an unknown secretory mechanism, can be detected in the serum of humans, mice, and other mammals (21). Adoptive transfer of GPI-specific IgG monoclonal antibodies (mAb) can also induce arthritis, confirming the critical role of autoantibodies in disease development (22). Thus, this model of RA shares not only the clinical features of RA in humans, because 1807 striking therapeutic benefits are obtained with antiCD20 mAb in humans (23,24). Similar to human RA, K/BxN mouse serum–induced arthritis also is strictly dependent on autoantibodies for disease development and progression (17,18). Analysis of the local factors involved in this T cell– and B cell–independent autoimmune disease revealed the involvement of mast cells (3) and macrophages (25). However, critical questions remain. Thus, it must be clarified whether mast cells selectively present in joints are necessary and sufficient for the induction of inflammation, and whether mast cells restore only early inflammation or are also involved in angiogenesis, pannus formation, and joint destruction. Importantly, the data raise the question of whether selective mast cell silencing is capable of preventing arthritis, including pathologic angiogenesis or joint destruction. We first showed that selective mast cell reconstitution inside the joint is both necessary and sufficient to establish sensitivity for induction of arthritis, including angiogenesis, joint destruction, and pannus formation. Positron emission tomography (PET) revealed that mast cells are required even for ␣v␤3 integrin activation, one of the earliest signs of angiogenesis. Based on these findings, we then analyzed the potential of targeted mast cell silencing to prevent GPI antibody–induced arthritis and observed that mast cell stabilization with either the cAMP-inducing compound salbutamol or with cromolyn, a molecule that selectively prevents mast cell degranulation through direct membrane stabilization, efficiently protects against angiogenesis, joint destruction, and pannus formation. MATERIALS AND METHODS Mice. Female genetically mast cell–deficient KitW/ Kit mice and congenic normal WBB6F1⫹/⫹ (Kit⫹/Kit⫹) mice were bred under specific pathogen–free conditions at the GSF⫺National Research Center (Munich, Germany). Adult KitW/KitW-v mice have ⬍1.0% of the number of tissue mast cells of congenic wild-type mice. Female C57BL/6 or BALB/c mice (Charles River, Sulzbach, Germany) were used for the therapy studies. All mice were ages 8–12 weeks. The studies were conducted according to approved animal use and care protocols. Reagents. Pooled serum from K/BxN mice (ages 1–6 months) was obtained by tail bleeding. Control serum was obtained from C57BL/6 mice (ages 1–6 months). Sera were diluted at a ratio of 1:1 (volume/volume) with physiologic saline before injection. Histologic and immunohistochemical analyses. The mouse extremities were fixed in 10% formalin for 2 days and W-v 1808 then decalcified with EDTA at 56°C for 10 days. Tissue samples were paraffin embedded, and 1–2-m microsections were hematoxylin and eosin (H&E) or Giemsa stained according to standard procedures. Sections selected for immunostaining were deparaffinized with alcohol. For detection of blood vessels, the mAb against human smooth muscle actin (CBL171, clone asm-1; BioTech Trade and Service, St. LeonRot, Germany) was used at a 1:10 (volume/volume) dilution, and the mAb against CD61 (␤3; Becton Dickinson, San Jose, CA) was used. Monoclonal antibody CBL171 binds also to rodent smooth muscle actin in pericytes. As control, we used an isotype (IgG2a) non-sense mAb. Immunohistochemical analysis was carried out with the ARK (Animal Research Kit, K 3954; Dako, Glostrup, Denmark). Counterstaining was performed with hemalaun. Serum transfer and assessment of arthritis. We injected C57BL/6, BALB/c, Kit⫹/Kit⫹, or KitW/KitW-v mice with 5 l/gm of either K/BxN mouse serum or control serum, intraperitoneally. We measured ankle thickness with an Oditest micrometer (Kroeplin, Munich, Germany) before and on the indicated days after K/BxN mouse serum transfer. Arthritis was assessed visually as the clinical index score, where 1 ⫽ 1 ankle affected, 2 ⫽ 2 ankles affected, 3 ⫽ 3 ankles affected, and 4 ⫽ all 4 ankles affected. Histopathologic analysis of joints was performed with H&E-stained microsections of the tibiocalcaneus joints, hind foot joints, front limbs, and elbows. Two criteria were assessed and graded semiquantitatively: inflammatory infiltrate (I) with granulocytes and lymphocytes, and hyperplasia of the synovial stroma (S). Grades ranged from 0 to 3 and were defined as follows: I0 ⫽ no inflammatory infiltrate, I1 ⫽ slight inflammatory infiltrate, I2 ⫽ moderate inflammatory infiltrate, I3 ⫽ strong inflammatory infiltrate, S0 ⫽ no synovial hyperplasia, S1 ⫽ slight synovial hyperplasia, S2 ⫽ moderate synovial hyperplasia, and S3 ⫽ strong synovial hyperplasia or destruction of cartilage overlying bone (pannus formation). Synthesis of 18F–galacto-RGD. 18F–galacto-RGD was synthesized and radiolabeled as previously described (26). The linear peptide DfKRG was assembled on solid support by standard FMOC protocols and cyclized under high-dilution conditions and then conjugated with FMOC-protected galactose-based sugar amino acid FMOC-SAA2-OH (26,27). 18 F–galacto-RGD was labeled using 4-nitrophenyl-2-18Ffluoropropionate. The final product had a radiochemical purity of ⬎98% and a specific activity of ⬎40 TBq/mmole. In vivo examinations. In vivo high-resolution PET images were acquired with the Munich Avalanche Diode PET (MADPET) system, a prototype small-animal PET system (28). The axial field of view (FOV) of the MADPET scanner is 3.7 mm, and the transaxial FOV is 50 mm. The spatial resolution in reconstructed PET images is 2.5 mm. List-mode data were reconstructed by applying a statistical iterative ordered-subsets expectation-maximization algorithm. One hour after tail-vein injection of 5,550 kBq (150 Ci) of 18 F–galacto-RGD on day 6 after serum transfer, we scanned mice for 10 minutes in one bed position. The limited axial FOV of 3.7 mm allowed only a single-slice PET scan through the tibiocalcaneus joints. During tracer uptake and image acquisition, mice were kept anesthetized by intraperitoneal injection of 5 mg/kg xylazine and 100 mg/kg ketamine. In one experi- KNEILLING ET AL ment, animals were pretreated with 18 mg/kg of unlabeled c(RGDfV) peptide 10 minutes prior to 18F–galacto-RGD injection to prove specificity of RGD peptide binding. Reconstructed images were normalized to the injected activity and used to draw regions of interest (ROIs) around the imaged joints. The mean measured counts in the defined ROI were calculated. Experiments included 3–5 mice per group. Cell cultures. Femoral bone marrow cells obtained from Kit⫹/Kit⫹ mice were cultured in the presence of murine recombinant interleukin-3 (IL-3) and c-kit ligand (2) for intraarticular reconstitution or with IL-3 alone for intraperitoneal reconstitution. Mast cell reconstitution. Local mast cell reconstitution of KitW/KitW-v mice was performed by injecting ankle (tibiocalcaneus) and hind foot (calcaneotarsal, tarsometatarsal, and metatarsophalangeal) joints intraarticularly with 7.5 ⫻ 105 bone marrow–derived mast cells 5 weeks before K/BxN mouse serum transfer. Systemic mast cell reconstitution was performed by injecting 6 ⫻ 105 bone marrow–derived mast cells intraperitoneally at 3 days of age, 10 weeks before K/BxN mouse serum transfer. Treatment protocol. C57BL/6 mice were treated by intraperitoneal injection of salbutamol (25 g/gm) or cromolyn (25 g/gm). Control mice received physiologic saline (placebo). Salbutamol was given 2 days before the K/BxN mouse serum transfer. Cromolyn or placebo was administered 3 days before K/BxN mouse or control serum transfer, and therapy was repeated every 24 hours. BALB/c mice received salbutamol (25 g/gm), cromolyn (25 g/gm), or saline twice, 1 hour before and 24 hours after serum transfer. Statistical analysis. All results are presented as the mean ⫾ SEM. We performed a Wilcoxon test to compare the increase in ankle thickness in KitW/KitW-v mice versus mast cell–reconstituted KitW/KitW-v mice and in KitW/KitW-v mice versus wild-type mice. Dunnett’s test was applied to compare the increase in ankle thickness in placebo-treated mice with that in either salbutamol- or cromolyn-treated mice. The decadic logarithm of RGD peptide uptake in the tibiocalcaneus joints was compared using the Student’s 2-tailed t-test. P values less than 0.05 were considered significant. RESULTS Requirement of intraarticular mast cells for development of arthritis. Reconstitution of mast cell– deficient KitW/KitW-v mice with mast cells restores their susceptibility to both T cell– and B cell–dependent and –independent inflammation, including arthritis induced by anti-GPI antibodies (2,3). Importantly, some diseases such as GPI antibody–induced arthritis are entirely independent of T cells and B cells and thus are especially suitable for selective analysis of mast cell function in vivo (3,11,12). Thus, injection of K/BxN mouse serum containing anti-GPI antibodies into wild-type Kit⫹/Kit⫹ mice led to severe arthritis within 6 days, characterized by severe edema and functional impairment of the paws MAST CELL SILENCING IN RA (Figure 1A). In contrast, injection of K/BxN mouse serum into mast cell–deficient KitW/KitW-v did not cause clinical signs of arthritis, and the mice remained clinically healthy and showed no signs of reduced motility (Figure 1B). To determine whether the selective presence of intraarticular mast cells is sufficient for development of arthritis or whether systemic mast cell repopulation is required, we injected anti-GPI antibody–containing K/BxN mouse serum into 4 groups of mice: wild-type Kit⫹/Kit⫹ mice, congenic KitW/KitW-v mice, KitW/KitW-v mice with systemic mast cell reconstitution by intraperitoneal injection of mast cells, or KitW/KitW-v mice with selective intraarticular mast cell reconstitution into the ankle and hind foot joints. K/BxN mouse serum induced severe arthritis within 6 days in Kit⫹/Kit⫹ mice (3), with a rapid increase in ankle thickness (mean ⫾ SEM 400 ⫾ 225 m) (Figure 1C). In mast cell–deficient KitW/KitW-v mice, ankle thickness increased only marginally (mean ⫾ SEM 38 ⫾ 56 m) and the increase in clinical index score was minimal (Figures 1C and D). In 6 independent experiments, arthritis developed in only 1 of 23 K/BxN serum–injected KitW/KitW-v mice. Following intraperitoneal mast cell engraftment, mast cells repopulated joints within 10 weeks and restored susceptibility to seruminduced arthritis (Figures 1C and D), with a mean ⫾ SEM increase in ankle thickness of 480 ⫾ 264 m, 6 days after K/BxN mouse serum transfer. To determine whether systemic mast cell distribution is needed or whether exclusive intraarticular mast cell reconstitution is sufficient for arthritis induction, we injected mast cells selectively into the ankle (tibiocalcaneus) and the hind foot (calcaneotarsal, tarsometatarsal, and metatarsophalangeal) joints. Intraarticular mast cell injection established complete susceptibility to arthritis induction, but exclusively in those joints in which mast cells had been reconstituted. In the same mouse, the mast cell–deficient joints of the front limb remained unaffected. After local mast cell reconstitution, ankle thickness and the clinical index increased to the same extent as after intraperitoneal administration (mean ⫾ SEM 350 ⫾ 291 m) (Figures 1E and F). Both intraperitoneal and intraarticular mast cell reconstitution were repeatedly confirmed by Giemsa staining of ankle tissue (Figures 1G–J). Mast cell–dependent activation of endothelia and angiogenesis. Pannus formation and aberrant angiogenesis, 2 hallmarks of RA (13,29), also predominate in joints from wild-type Kit⫹/Kit⫹ mice (3). Abundant angiogenesis was visualized by ␣-actin staining of peri- 1809 cytes (Figure 2A). Mice receiving control serum (results not shown) or KitW/KitW-v mice receiving K/BxN mouse serum (Figure 2B) had neither pannus formation nor aberrant angiogenesis, while intraperitoneally mast cell– reconstituted KitW/KitW-v mice (Figure 2C) developed abundant angiogenesis and severe pannus, indistinguishable from those in wild-type Kit⫹/Kit⫹ mice, following injection of K/BxN mouse serum. Angiogenesis is characterized by intense expression of activated ␣v␤3 integrin on endothelia (30,31). Following K/BxN mouse serum transfer, ␤3 integrin was detectable by immunohistologic analysis in wild-type Kit⫹/Kit⫹ mice but not in KitW/KitW-v mice (results not shown). Expression of activated ␣v␤3 integrin can be imaged by PET with 18F–galacto-RGD, an 18F-labeled RGD peptide that selectively binds activated ␣v␤3 integrin heterodimer (32). One hour prior to imaging, we injected mice with 5,550 kBq (150 Ci) of 18F–galactoRGD. When compared with mice receiving control serum (Figures 2E and I), 18F–galacto-RGD uptake increased ⬃2-fold in joints of K/BxN serum–injected wild-type Kit⫹/Kit⫹ mice (Figures 2F and I), while 18 F–galacto-RGD uptake remained at background levels in mast cell–deficient KitW/KitW-v mice (Figures 2G and I). In intraperitoneally mast cell–reconstituted mice, activated ␣v␤3 integrin was expressed at least as strongly as in wild-type Kit⫹/Kit⫹ mice (Figures 2H and I). Quantitative analysis of relative tracer accumulation revealed an ⬃2-fold increase in 18F–galacto-RGD uptake in the presence of mast cells (Figures 2F, H, and I), while 18F–galacto-RGD uptake in mast cell–deficient KitW/KitW-v mice (Figures 2G and I) was identical to that in mice injected with control serum (Figures 2E and I) (P ⫽ 0.04). Prevention of angiogenesis by targeted mast cell silencing. This prominent role of mast cells in tissue destruction and angiogenesis suggests mast cell stabilization as an important target for disease prevention. Because in vivo mast cell membranes can be stabilized and activation-induced degranulation can be prevented with cAMP-inducing beta-mimetics or with cromolyn (33–35), we treated C57BL/6 mice with either salbutamol or cromolyn, from day 2 (salbutamol) or day 3 (cromolyn) to day 9 after K/BxN mouse serum injection. Both modes of mast cell stabilization protected against joint swelling, arthritis, and joint destruction. Ankle swelling was reduced to 16% (P ⫽ 0.03) in salbutamol-treated mice and to 30% (P ⫽ 0.02) in cromolyn-treated mice on day 6 after K/BxN mouse serum transfer, compared with placebo-treated mice (Figures 3A and C). Both groups of mice that received 1810 KNEILLING ET AL Figure 1. Effects of intraperitoneal (IP) and intraarticular (IA) mast cell (MC) reconstitution. A, K/BxN mouse–derived serum induced severe arthritis within 6 days in wild-type Kit⫹/Kit⫹ mice, characterized by severe edema and functional impairment of the paws. B, K/BxN mouse serum did not cause clinical signs of arthritis in mast cell–deficient KitW/KitW-v, mice which developed no ankle swelling and showed no signs of reduced motility. C–F, Ankle thickness (C and E) and clinical index scores (D and F) in wild-type Kit⫹/Kit⫹ mice (circles), mast cell–deficient KitW/KitW-v mice (shaded squares), mast cell–reconstituted KitW/KitW-v mice (open squares), and control wild-type Kit⫹/Kit⫹ mice (triangles). Values are the mean ⫾ SEM. Significant differences in ankle thickness in intraperitoneal mast cell reconstitution experiments (C) were as follows: P ⫽ 0.03, KitW/KitW-v mice (n ⫽ 5) versus mast cell–reconstituted KitW/KitW-v mice (n ⫽ 3) on day 6; P ⫽ 0.017, KitW/KitW-v mice versus wild-type Kit⫹/Kit⫹ mice (n ⫽ 5) on day 6. Significant differences in ankle thickness in intraarticular mast cell reconstitution experiments (E) were as follows: P ⫽ 0.008, KitW/KitW-v mice (n ⫽ 3) versus mast cell–reconstituted KitW/KitW-v mice (n ⫽ 3) on day 6; P ⫽ 0.002, KitW/KitW-v mice versus wild-type K⫹/Kit⫹ mice (n ⫽ 3) on day 6. G–J, Giemsa-stained ankle tissue from mast cell–reconstituted KitW/KitW-v mice. G and I, Overview (original magnification ⫻ 400). H and J, Higher-magnification views of G and I, respectively. Arrows indicate reconstituted mast cells. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org. MAST CELL SILENCING IN RA 1811 Figure 2. Susceptibility to ␣v␤3 integrin–associated angiogenesis and joint destruction in response to K/BxN mouse serum. A–D, Anti–␣-actin staining of pericytes (arrows) in sagittal ankle sections from K/BxN mouse serum–injected syngenic wild-type Kit⫹/Kit⫹ mice (A), KitW/KitW-v mice (B), or mast cell (MC)–reconstituted KitW/KitW-v mice (C) 6 days after serum transfer, and control staining with non-sense primary antibody IgG2a (D) (original magnification ⫻ 200). E–H, Quantification of activated ␣v␤3 integrin on day 6 by positron emission tomographic imaging of ankles from control serum–injected wild-type Kit⫹/Kit⫹ mice (E), K/BxN mouse serum–injected wild-type Kit⫹/Kit⫹ mice (F), mast cell–deficient KitW/KitW-v mice (G), or intraperitoneally (IP) mast cell–reconstituted KitW/KitW-v mice (H). I, Quantification of RGD peptide uptake ratio from in vivo–scanned mice. Bars show the mean and SEM results of 3–4 animals per group. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org. 1812 KNEILLING ET AL Figure 3. Effects of targeted mast cell silencing with intraperitoneal salbutamol or cromolyn in C57BL/6 mice. Ankle thickness (A and C) and clinical index scores (B and D) were determined in K/BxN mouse serum–injected mice treated with physiologic saline (circles), salbutamol (25 g/gm; solid squares), or cromolyn (25 g/gm; shaded squares), and control serum–injected mice (triangles). On day 6, ankle swelling was significantly reduced in salbutamol-treated mice (P ⫽ 0.03) and cromolyn-treated mice (P ⫽ 0.02) compared with placebo-treated mice. Values are the mean ⫾ SEM results from 4 mice per group. mast cell stabilizers had minor arthritis followed by almost complete remission until day 9, while placebotreated mice continued to have the maximal clinical index score (Figures 3B and D). Histologic analysis of C57BL/6 mice on day 9 after K/BxN mouse serum injection confirmed severe arthritis with synovitis, pannus formation, periostitis, and destruction of the corticalis in both untreated and placebo-treated mice (Figure 4A). In sharp contrast, joints of salbutamol-treated mice had only minor signs of synovitis and blood vessel proliferation (Figures 4B–D). Importantly, mast cell stabilization during disease initiation protected even arthritis-prone BALB/c mice; significant protection was also achieved by administering treatment twice, 1 hour before and 24 hours after K/BxN mouse serum injection (data not shown). Because examination of H&E-stained sections suggested that mast cell silencing also prevented blood vessel formation, with only a few new vessels at selected sites on day 9 after K/BxN serum injection (Figure 4D), we investigated the role of mast cells and mast cell silencing on angiogenesis in more detail, by characterizing the pericytes of mature vessels with ␣-actin immunohistology and by quantifying angiogenesis with 18F– galacto-RGD and PET in vivo. We performed those experiments on day 6 after K/BxN mouse serum transfer, because C57BL/6 mice already showed an increased number of ␣-actin–staining cells that characterize mature pericytes, together with severe arthritis as characterized by synovitis, pannus formation, periostitis, and destruction of the corticalis (Figures 4E and F), as compared with control animals (Figures 4G and H). Importantly, salbutamol almost completely prevented the generation of ␣-actin–expressing pericytes (Figures 4C and D). To assess the effect of salbutamol and cromolyn on ␣v␤3 integrin expression, we measured 18 F–galacto-RGD binding in joints of either placebotreated, salbutamol-treated, or cromolyn-treated C57BL/6 mice on day 6 after injection of K/BxN mouse serum. We determined tracer uptake in joints 1 hour after a 5,550-kBq (150 Ci) injection of 18F–galactoRGD. In placebo-treated mice, K/BxN mouse serum led MAST CELL SILENCING IN RA Figure 4. Effect of targeted mast cell silencing with salbutamol or cromolyn on arthritis, angiogenesis, and joint destruction. A and B, Hematoxylin and eosin– stained ankle sections from placebo-treated wild-type mice (A) (arrows indicate pannus formation) and salbutamol-treated wild-type mice (B) (arrows indicate normal synovium) after K/BxN mouse serum transfer. C–H, Blood vessel staining (arrows) with monoclonal antibody to ␣-actin in K/BxN mouse serum–injected and salbutamol-treated (C and D) C57BL/6 mice or in placebo-treated (E and F) and control serum–injected (G and H) C57BL/6 mice. D, Higher magnification view of boxed area in C. (Original magnification ⫻ 50 in A, B, C, E, and G; ⫻ 200 in D, F, and H.) 1813 1814 KNEILLING ET AL Figure 5. Effect of targeted mast cell silencing with salbutamol or cromolyn on arthritis, ␣v␤3 integrin–associated angiogenesis, and joint destruction. A–D, In vivo positron emission tomography imaging of ankles from control serum–injected mice treated with placebo (A) and K/BxN mouse serum–injected wild-type mice treated with placebo (B), salbutamol (C), or cromolyn (D). E, Quantification of the RGD peptide uptake ratio in in vivo–scanned K/BxN serum–injected mice treated with placebo (solid column), salbutamol (open column), or cromolyn (shaded column) or control serum– injected mice treated with placebo (hatched column). Values are the mean and SEM results from 3–4 mice per group. ROI ⫽ region of interest. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org. again to strong 18F–galacto-RGD uptake (Figures 5A and B), while in mice receiving either salbutamol (Figure 5C) or cromolyn (Figure 5D), 18F–galacto-RGD uptake was suppressed to the levels observed in mice injected with control serum (Figure 5E). Quantitative analysis showed that 18F–galacto-RGD binding was again significantly increased in positive controls but remained at background levels in salbutamol- or cromolyn-treated mice (Figure 5E) (P ⫽ 0.01, salbutamol versus placebo; P ⫽ 0.04, cromolyn versus placebo). In salbutamol- or cromolyn-treated mouse joints, 18F– galacto-RGD uptake was indistinguishable from that in mast cell–deficient KitW/KitW-v mice (Figure 5E). Thus, mast cell silencing efficiently prevented MAST CELL SILENCING IN RA angiogenesis, pannus formation, and joint destruction, all of which are hallmarks of RA. DISCUSSION This is the first study to show that mast cell reconstitution is both necessary and sufficient to establish sensitivity for the initiation of GPI-induced arthritis, including angiogenesis, pannus formation, and tissue destruction. In vivo measurement of activated ␣v␤3 integrin in K/BxN mouse serum–induced arthritis using 18 F–galacto-RGD and PET revealed that mast cells are required for ␣v␤3 integrin activation, which is one of the earliest signs of angiogenesis. Importantly, mast cell stabilization with either the cAMP-inducing compound salbutamol or with cromolyn, a molecule that selectively prevents mast cell degranulation through direct membrane stabilization, efficiently protected against ␣v␤3 integrin activation, angiogenesis, pannus formation, and joint destruction. We previously reported that mast cells are involved in the inflammation mediated by Th1 cells, and that mast cells are the primary source of 2 important proangiogenic factors, IL-8 and tumor necrosis factor. This seems to be relevant for Th1 cell–mediated skin inflammation in mice and in humans (2,8,36). Deviating Th1 responses into Th2 responses improved Th1 cell– mediated inflammation and protected against inflammation-induced angiogenesis in mice and humans (36–38). Most importantly, Th1 cells seem to coevolve simultaneously with IL-17–producing Th17 cells (39); IL-17 is a cytokine with strong proangiogenic effects (40,41). Because Th1 cell–associated skin and joint inflammation are associated with both Th17 cells and mast cells, the relative contribution of mast cells to inflammation-induced angiogenesis remains to be determined. To separate the effects of mast cells on joint destruction and angiogenesis from the effects of T cell–produced IL-17, we used GPI antibody–induced arthritis that is entirely independent of B cell and T cell activation, because it can be induced even in Rag-2⫺/⫺ mice (3,11,12). Importantly, injection of GPI antibodies induces arthritis more rapidly than T cells differentiate into distinct functional phenotypes. Here, we first observed that activation of mast cells inside a single joint was sufficient for disease induction, angiogenesis, and joint destruction in response to GPI antibodies. Moreover, we observed that mast cell activation with GPI antibodies activates ␣v␤3 integrin, which is expressed during early angiogenesis, 1815 and induces pericyte proliferation that occurs during differentiation of mature blood vessels. Selective mast cell silencing with the cAMPinducing agent salbutamol or cromolyn strongly suppressed K/BxN mouse serum–induced arthritis in C57BL/6 and BALB/c mice. The protective effects of cAMP induction by salbutamol and other beta-mimetics was, until now, attributed to the deviation of Th1 cells into Th2 cells, and the deviation of proinflammatory macrophages into a phenotype that has largely lost the capacity to cause inflammation-induced tissue destruction in diseases such as collagen-induced arthritis (33– 35). One of the effects of salbutamol on macrophages may be the suppression of IL-12 (33). Such effects are highly unlikely in the arthritis that we studied here, because mast cells exclusively at the site of inflammation are required for disease induction. Moreover, treatment with cromolyn or salbutamol during the first 24 hours after serum injection was sufficient for the suppression of GPI-induced arthritis, a time frame that is too short to attenuate any T cell or long-lasting macrophage action. Thus, the data presented here provide an entirely novel tool for quantitative in vivo evaluation of angiogenesis that may allow quantification of disease activity in human autoimmune diseases such as RA or psoriasis. By showing that mast cell silencing significantly attenuates inflammation-induced angiogenesis, pannus formation, and even joint destruction, the data strongly suggest mast cell silencing as a novel and safe approach for the prevention of major T cell–mediated autoimmune diseases such as RA, psoriasis, chronic obstructive bronchitis, multiple sclerosis, or inflammatory bowel disease. ACKNOWLEDGMENTS We appreciate the technical support of U. Bamberg, C. Bodenstein, D. Dick, and G. Fernahl. AUTHOR CONTRIBUTIONS Dr. Röcken 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. Kneilling, Hültner, Pichler, Solomon, Biedermann, Weber, Haubner, Röcken. Acquisition of data. Kneilling, Hültner, Pichler, Mailhammer, Morawietz, Solomon, Sabatino, Haubner, Röcken. Analysis and interpretation of data. Kneilling, Hültner, Pichler, Mailhammer, Morawietz, Solomon, Sabatino, Biedermann, Krenn, Weber. Manuscript preparation. Kneilling, Hültner, Pichler, Morawietz, Biedermann, Krenn, Röcken. Statistical analysis. Solomon, Eichner. Synthesis and labeling of galacto-RGD. Haubner. 1816 KNEILLING ET AL REFERENCES 1. Secor VH, Secor WE, Gutekunst CA, Brown MA. Mast cells are essential for early onset and severe disease in a murine model of multiple sclerosis. J Exp Med 2000;191:813–22. 2. Biedermann TM, Kneilling R, Mailhammer K, Maier CA, Sander G, Kollias SL, et al. Mast cells control neutrophil recruitment during T cell-mediated delayed-type hypersensitivity reactions through tumor necrosis factor and macrophage inflammatory protein 2. J Exp Med 2000;192:1441–52. 3. 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. 4. Echtenacher B, Mannel DN, Hultner L. Critical protective role of mast cells in a model of acute septic peritonitis. Nature 1996;381: 75–7. 5. Jiang WY, Chattedee AD, Raychaudhuri SP, Raychaudhuri SK, Farber EM. Mast cell density and IL-8 expression in nonlesional and lesional psoriatic skin. Int J Dermatol 2001;40:699–703. 6. Charlesworth EN. The role of basophils and mast cells in acute and late reactions in the skin. Allergy 1997;52:31–43. 7. Crisp AJ, Chapman CM, Kirkham SE, Schiller AL, Krane SM. Articular mastocytosis in rheumatoid arthritis. Arthritis Rheum 1984;7:845–51. 8. Rocken M, Hultner L. Heavy functions for light chains. Nat Med 2002;8:668–70. 9. Godfrey HP, Ilardi C, Engber W, Graziano FM. Quantitation of human synovial mast cells in rheumatoid arthritis and other rheumatic diseases. Arthritis Rheum 1984;27:852–6. 10. Malone DG, Wilder RL, Saavedra-Delgado AM, Metcalfe DD. Mast cell numbers in rheumatoid synovial tissues: correlations with quantitative measures of lymphocytic infiltration and modulation by antiinflammatory therapy. Arthritis Rheum 1987;30: 130–7. 11. Shin K, Gurish MF, Friend DS, Pemberton AD, Thornton EM, Miller HR, et al. Lymphocyte-independent connective tissue mast cells populate murine synovium. Arthritis Rheum 2006;54: 2863–71. 12. Binstadt BA, Patel PR, Alencar H, Nigrovic PA, Lee DM, Mahmood U, et al. Particularities of the vasculature can promote the organ specificity of autoimmune attack. Nat Immunol 2006;7: 284–92. 13. Feldmann M, Brennan FM, Maini RN. Rheumatoid arthritis. Cell 1996;85:307–10. 14. Brennan FM, Maini RN, Feldmann M. Role of pro-inflammatory cytokines in rheumatoid arthritis. Springer Semin Immunopathol 1988;20:133–47. 15. Firestein GS. Evolving concepts of rheumatoid arthritis. Nature 2003;423:356–61. 16. Zvaifler NJ, Firestein GS. Pannus and pannocytes: alternative models of joint destruction in rheumatoid arthritis. Arthritis Rheum 1994;37:783–9. 17. Kouskoff V, Korganow AS, Duchatelle V, Degott C, Benoist C, Mathis D. Organ-specific disease provoked by systemic autoimmunity. Cell 1996;87:811–22. 18. Korganow AS, Ji H, Mangialaio S, Duchatelle V, Pelanda R, Martin T, et al. From systemic T cell self-reactivity to organspecific autoimmune disease via immunoglobulins. Immunity 1999;10:451–61. 19. Matsumoto I, Staub A, Benoist C, Mathis D. Arthritis provoked by linked T and B cell recognition of a glycolytic enzyme. Science 1999;286:1732–5. 20. Schaller M, Burton DR, Ditzel HJ. Autoantibodies to GPI in rheumatoid arthritis: linkage between an animal model and human disease. Nat Immunol 2001;2:746–53. 21. Zimmerman HJ, Schwartz MA, Boley LE, West M. Comparative serum enzymology. J Lab Clin Med 1965;66:961–72. 22. Maccioni M, Zeder-Lutz G, Huang H, Ebel C, Gerber P, Hergueux J, et al. Arthritogenic monoclonal antibodies from K/BxN mice. J Exp Med 2002;195:1071–7. 23. Eisenberg R, Albert D. B-cell targeted therapies in rheumatoid arthritis and systemic lupus erythematosus. Nat Clin Pract Rheumatol 2006;2:20–7. 24. Eisenberg R, Looney RJ. The therapeutic potential of anti-CD20 “what do B-cells do?”. Clin Immunol 2005;117:207–13. 25. Solomon S, Rajasekaran N, Jeisy-Walder E, Snapper SB, Illges H. A crucial role for macrophages in the pathology of K/BxN serum-induced arthritis. Eur J Immunol 2005;35:3064–73. 26. Haubner R, Kuhnast B, Mang C, Weber WA, Kessler H, Wester HJ, et al. [18F]Galacto-RGD: synthesis, radiolabeling, metabolic stability, and radiation dose estimates. Bioconjug Chem 2004;15: 61–9. 27. Haubner R, Wester HJ, Weber WA, Mang C, Ziegler SI, Goodman SL, et al. Noninvasive imaging of ␣(v)␤3 integrin expression using 18F-labeled RGD-containing glycopeptide and positron emission tomography. Cancer Res 2001;61:1781–5. 28. Ziegler SI, Pichler BJ, Boening G, Rafecas M, Pimpl W, Lorenz E, et al. A prototype high-resolution animal positron tomograph with avalanche photodiode arrays and LSO crystals. Eur J Nucl Med 2001;28:136–43. 29. Koch AE. Angiogenesis as a target in rheumatoid arthritis. Ann Rheum Dis 2003;62 Suppl 2:ii60–7. 30. Cheresh DA, Stupack DG. Integrin-mediated death: an explanation of the integrin-knockout phenotype? Nat Med 2002;8:193–4. 31. Koning GA, Schiffelers RM, Wauben MH, Kok RJ, Mastrobattista E, Molema G, et al. Targeting of angiogenic endothelial cells at sites of inflammation by dexamethasone phosphate–containing RGD peptide liposomes inhibits experimental arthritis. Arthritis Rheum 2006;54:1198–208. 32. Pichler BJ, Kneilling M, Haubner R, Braumuller H, Schwaiger M, Rocken M, et al. Imaging of delayed-type hypersensitivity reaction by PET and 18F-galacto-RGD. J Nucl Med 2005;46:184–9. 33. Malfait AM, Malik AS, Marinova-Mutafchieva L, Butler DM, Maini RN, Feldmann M. The ␤2-adrenergic agonist salbutamol is a potent suppressor of established collagen-induced arthritis: mechanisms of action. J Immunol 1999;162:6278–83. 34. Lichtenstein LM, Gillespie E. Inhibition of histamine release by histamine controlled by H2 receptor. Nature 1973;244:287–8. 35. Viscardi RM, Hasday JD, Gumpper KF, Taciak V, Campbell AB, Palmer TW. Cromolyn sodium prophylaxis inhibits pulmonary proinflammatory cytokines in infants at high risk for bronchopulmonary dysplasia. Am J Respir Crit Care Med 1997;156:1523–9. 36. Ghoreschi K, Thomas P, Breit S, Dugas M, Mailhammer R, van Eden W, et al. Interleukin-4 therapy of psoriasis induces Th2 responses and improves human autoimmune disease. Nat Med 2003;9:40–6. 37. Biedermann T, Mailhammer R, Mai A, Sander C, Ogilvie A, Brombacher F, et al. Reversal of established delayed type hypersensitivity reactions following therapy with IL-4 or antigen-specific Th2 cells. Eur J Immunol 2001;31:1582–91. 38. Biedermann T, Zimmermann S, Himmelrich H, Gumy A, Egeter O, Sakrauski AK, et al. IL-4 instructs TH1 responses and resistance to Leishmania major in susceptible BALB/c mice. Nat Immunol 2001;2:1054–60. 39. 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. 40. Numasaki M, Fukushi J, Ono M, Narula SK, Zavodny PJ, Kudo T, et al. Interleukin-17 promotes angiogenesis and tumor growth. Blood 2003;101:2620–7. 41. Ryu S, Lee JH, Kim SI. IL-17 increased the production of vascular endothelial growth factor in rheumatoid arthritis synoviocytes. Clin Rheumatol 2006;25:16–20.