α2β1 Integrin Regulates Th17 Cell Activity and Its Neutralization Decreases the Severity of Collagen-Induced Arthritis

This information is current as Mohammed-Amine El Azreq, Marc Boisvert, Annabelle of September 24, 2021. Cesaro, Nathalie Pagé, Lionel Loubaki, Isabelle Allaeys, Jamila Chakir, Patrice E. Poubelle, Philippe A. Tessier and Fawzi Aoudjit J Immunol 2013; 191:5941-5950; Prepublished online 15

November 2013; Downloaded from doi: 10.4049/jimmunol.1301940 http://www.jimmunol.org/content/191/12/5941

References This article cites 52 articles, 20 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/191/12/5941.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 24, 2021

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2013 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology a2b1 Integrin Regulates Th17 Cell Activity and Its Neutralization Decreases the Severity of Collagen-Induced Arthritis

Mohammed-Amine El Azreq,*,† Marc Boisvert,*,† Annabelle Cesaro,*,† Nathalie Page´,*,† Lionel Loubaki,‡,x Isabelle Allaeys,*,† Jamila Chakir,‡,x Patrice E. Poubelle,†,x Philippe A. Tessier,*,† and Fawzi Aoudjit*,†

Th17 cells play a critical role in the pathogenesis of rheumatoid arthritis (RA), but the mechanisms by which these cells regulate the development of RA are not fully understood. We have recently shown that a2b1 integrin, the of type I collagen, is the major collagen-binding integrin expressed by human Th17 cells. In this study, we examined the role of a2b1 integrin in Th17- mediated destructive arthritis in the murine model of collagen-induced arthritis (CIA). We found that a2b1 integrin is expressed Downloaded from on synovial Th17 cells from CIA mice and its neutralization with a specific mAb significantly reduced inflammation and cartilage degradation, and protected the mice from bone erosion. Blockade of a2b1 integrin led to a decrease in the number of Th17 cells in the joints and to a reduction of IL-17 levels in CIA mice. This was associated with an inhibition of receptor activator of NF-kB ligand levels and osteoclast numbers, and reduction of bone loss. We further show that a2b1 integrin is expressed on synovial Th17 cells from RA patients, and that its ligation with collagen costimulated the production of IL-17 by polarized human Th17 cells by enhancing the expression of retinoic acid receptor–related orphan receptor C through ERK and PI3K/AKT. Our findings http://www.jimmunol.org/ provide the first evidence, to our knowledge, that a2b1 integrin is an important pathway in Th17 cell activation in the patho- genesis of CIA, suggesting that its blockade can be beneficial for the treatment of RA and other Th17-associated autoimmune diseases. The Journal of Immunology, 2013, 191: 5941–5950.

heumatoid arthritis (RA) is an char- cytokine is critical for the development of osteoclasts, the cells acterized by a massive infiltration of inflammatory cells to responsible for bone erosion associated with RA (13). the joints, leading to synovitis, cartilage erosion, and bone Integrins are a/b membrane receptors that mediate cell–cell

R by guest on September 24, 2021 damage (1). Th17 cells are a recently described subpopulation of interactions and cell adhesion to the surrounding extracellular Th cells that are specialized in the production of IL-17. Evidence matrix (ECM). T cells express several ECM receptors, among from animal and human studies indicate that RA is predominantly which are the very late activating Ags (VLA-1 to VLA-6) that a Th17-driven disease (2–8). IL-17 stimulates the production of constitute the b1 subfamily of integrins (14, 15). After ac- inflammatory cytokines and chemokines by macrophages, syno- tivation, these receptors coordinate T cell adhesion and migration vial fibroblasts, and chondrocytes leading to the recruitment of through basement membranes and interstitial tissue to reach the additional Th17 cells and other inflammatory cells such as neu- inflammatory sites (15, 16). b1 integrins could also regulate T cell trophils (9–12). IL-17 also stimulates fibroblasts and osteoblasts to activation in the inflammatory sites because in these sites, T cells produce the receptor activator of NF-kB ligand (RANKL). This are in contact with ECM. This is in contrast with lymphoid organs such as lymph nodes and spleen where the ECM is not accessible to T cells because it is surrounded by reticular fibroblasts (16, 17). *Axe de Recherche sur les Maladies Infectieuses et Immunitaires, Centre de Recherche du Centre Hospitalier Universitaire de Que´bec, Ville de Que´bec, Que´bec Emerging evidence suggests that the collagen-binding integrin G1V 4G2, Canada; †De´partement de Microbiologie-Infectiologie et d’Immunologie, a2b1 (VLA-2) can be a major regulator of T cell activation. a2b1 Faculte´ de Me´decine de l’Universite´ Laval, Ville de Que´bec, Que´bec G1V 0A6, integrin is the receptor for collagen type I (Coll I) on T cells and is Canada; ‡Centre de Recherche de l’Institut de Cardiologie et Pneumologie de Que´bec, Ville de Que´bec, Que´bec G1V 4G5, Canada; and xDe´partement de Me´decine, expressed only on effector T cells associated with inflammation in Faculte´ de Me´decine de l’Universite´ Laval, Ville de Que´bec, Que´bec G1V 0A6, extravascular tissues (15, 16, 18). It is found on human CD4+ and Canada CD8+ effector T cells generated after in vitro activation of pe- Received for publication July 22, 2013. Accepted for publication October 8, 2013. ripheral blood T cells (18, 19), as well as on virus-activated mouse This work was supported by the Arthritis Society of Canada. CD4+ and CD8+ T cells (20, 21), although the expression of a2b1 Address correspondence and reprint requests to Dr. Fawzi Aoudjit, Centre de Re- integrin is transient on CD8+ T cells. We and others have found cherche du Centre Hospitalier Universitaire de Que´bec, 2705 Boulevard Laurier, local T1-49, Ville de Que´bec, Que´bec G1V 4G2, Canada. E-mail address: fawzi. that a2b1 integrin protects human effector T cells from Fas-mediated [email protected] apoptosis (19, 22). In addition, we have recently reported that a2b1, Abbreviations used in this article: BV, bone volume; CIA, collagen-induced arthritis; but not a1b1, integrin is the major collagen-binding integrin ex- Coll I, collagen type I; EAE, experimental autoimmune encephalomyelitis; ECM, pressed by human Th17 cells (23). a2b1 integrin mediates Th17 extracellular matrix; micro-CT, microcomputerized tomography; qRT-PCR, quanti- tative RT-PCR; RA, rheumatoid arthritis; RANKL, receptor activator of NF-kB li- cell adhesion to collagen, which costimulated the production of gand; RORc, retinoic acid receptor–related orphan receptor C; S/FG, safranin and IL-17 (23). fast green; Tb.N, trabecular number; Tb.pf, trabecular pattern factor; TRAP, tartrate- Currently, the mechanisms regulating the development of RA are resistant acid phosphatase; TV, tissue volume; VLA, very late activating Ag. not fully understood. Given the role of Th17 cells in RA and the Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 abundance of collagen in the joints (24), we hypothesized that Th17 www.jimmunol.org/cgi/doi/10.4049/jimmunol.1301940 5942 a2b1 INTEGRIN REGULATES Th17 CELL ACTIVITY IN ARTHRITIS cell interactions with collagen via a2b1 integrin could represent an Mice important pathway in the pathogenesis of RA. In this study, we Eight-week-old female DBA/1J mice (Jackson Laboratories) were used. found that a2b1 integrin is expressed on synovial Th17 cells and its Water and food (Purina Mouse Chow 5015) were provided ad libitum. All neutralization decreases the severity of collagen-induced arthritis procedures involving animals were conducted according to the requirements (CIA; Th17 cell–driven RA animal model) by reducing synovial of and with the approval of the Laval University animal protection com- Th17 cell numbers and IL-17 levels. We further show that a2b1 mittee. integrin is expressed on synovial Th17 cells from RA patients and Induction, treatment, clinical assessment, and histopathology its ligation with collagen costimulated the production of IL-17 by of CIA polarized human Th17 cells by enhancing through ERK and PI3K/ AKT the expression of retinoic acid receptor–related orphan re- Mice were immunized s.c. (day 0) at the base of the tail with chicken Coll II ceptor C (RORc), the master regulator of IL-17 gene transcription (100 mg/mouse) emulsified in CFA. On day 28 after initial immunization, and to synchronize disease onset, animals were injected i.p. with LPS (25 and Th17 differentiation. These data indicate for the first time, mg/mouse) (25). Animals start to develop arthritis within 24 h after LPS to our knowledge, that a2b1 integrin is an important pathway of injection. The anti-a2b1 integrin blocking mAb (Ha1/29) and isotypic Th17 cell activation in the pathogenesis of CIA, suggesting that control mAb (Ha4/8; 100 mg/mouse) were injected i.p. 24 h before LPS its blockade can be beneficial for the treatment of RA and other challenge (day 27) and at days 30, 32, and 34 thereafter (2, 4, and 6 d after Th17-associated autoimmune diseases. LPS injection, respectively). Mice were closely monitored and scored daily for clinical symptoms of arthritis in a blinded manner by two observers until the animals were sacrificed at day 36 (8 d after LPS injection). A Materials and Methods scoring scale from 0 to 4 per paw (for a maximum disease score of 16 per

Abs and reagents mouse) was used as previously described (26, 27): 0, normal; 1, mild, but Downloaded from definite redness and swelling of the ankle or wrist, or apparent redness and Hamster anti-mouse a2b1 integrin blocking mAb (clone Ha1/29) and its swelling limited to individual digits, regardless of the number of affected isotypic control hamster IgG (clone Ha4/8), anti-mouse CD4-FITC (clone digits; 2, moderate redness and swelling of ankle or wrist; 3, severe redness RM4-5), anti-mouse IL-17–Alexa 647 (clone TC11-18H10), anti-human and swelling of the entire paw including digits; and 4, maximally inflamed IL-17–Alexa 647 (clone N49-653), anti-CD3 (OKT3), anti-human a2 integrin- limb with involvement of multiple joints. At day 36, mice were sacrificed, PE (12F1), anti-mouse a2 integrin-PE (clone HM), and control isotypic Abs and blood samples and paws were recovered for analysis. were from BD Bioscience (San Diego, CA). CD3/CD28 Dynabeads were from Forepaws were rapidly fixed in 4% paraformaldehyde, decalcified in Invitrogen Dynal AS (Oslo, Norway). Chicken collagen II, LPS from E. coli TBD-2 Thermo-Shandon solution (Thermo Fisher Scientific) for 10 d, and http://www.jimmunol.org/ (serotype 0111:B4), and CFA were purchased from Chondrex. Cytokines embedded in paraffin. Sections (5 mm thickness) were prepared and stained (TGF-b,IL-1b, IL-6, IL-23) and ELISA kits (IL-17, RANKL) were from with H&E (Thermo Fisher Scientific) to assess cell infiltration and bone R&D Systems. degradation as previously described (26). Cell infiltration was evaluated BSA, collagenase D, PMA, Ionomycin, and Coll I were from Sigma following a 0 to 2 scale (0, normal; 1, mild infiltration; 2, severe infiltra- (St. Louis, MO). The MEK1/2 inhibitor U0126, the JNK/MAPK inhibitor tion), and bone degradation was scored using the following 0 to 3 scale: SP600125, the p38/MAPK inhibitor SB203580, and PI3K/AKT inhibitor 0, no bone erosion; 1, mild surface erosion; 2, moderate surface erosion; LY294002 were from Calbiochem (San Diego, CA). The anti–phospho- and 3, strong surface erosion. Paw sections were also stained with safranin/ ERK1/2 mAb (E-4), anti-ERK2 Ab (C-14), and anti-actin (C-2) were from fast green (S/FG; VWR International) to assess cartilage degradation by Santa Cruz Biotechnology (Santa Cruz, CA). The anti–phospho-JNK1/2 Ab evaluating the presence and the intensity of red staining (corresponding to (G9), anti–phospho-p38 (#9216) Ab, anti–phospho-AKT (#9271) Ab, and cartilage proteoglycans) using the following 0 to 2 scale: 0, normal; 1, mild by guest on September 24, 2021 anti-AKT (#9272) were from Cell Signaling Technologies (Beverly, MA). loss; and 2, complete loss (26). Histopathological quantifications were made Patients, isolation of arthritic synovial T cells, and human by two observers in a blinded manner from images of three fields repre- Th17 cell differentiation senting the distal interphalangeal joint, the proximal interphalangeal joint, and joints in the carpal region. Synovial fluids from five RA patients with active disease based on the American College of Rheumatology criteria (three had established RA and Flow cytometry of cell-surface Ag and IL-17 expression by two were diagnosed as early RA) were collected by arthrocentesis during joint and spleen cells a clinical procedure. The three patients with established RA were between 31 and 52 y old with disease duration between 1 and 7 y. Two patients were Hind paws were freshly dissected to remove the skin and were rapidly treated with , Plaquenil, and Salazopyrin, and one patient incubated in PBS containing collagenase D (2 mg/ml) for 2 h at 37˚C with received and prednisone. The two patients with early RA were 21 gentle agitation. The resulting material was passed through 40-mm cell and 26 y old with disease duration between 3 and 6 mo and were without strainers (Sigma Aldrich) to remove the undigested tissues. The cell sus- treatment. pensions were washed three times with PBS and resuspended in culture Neutrophils were depleted from the synovial fluids by positive selection medium. Spleens were directly minced through a 40-mmcellstrainer.RBCs using an EasySep Neutrophil Enrichment Kit (StemCell Technologies). were lysed by incubating the preparations in a buffer containing 0.83% CD4 T cells were then isolated by negative selection using magnetic beads NH4Cl, 0.1% KHCO3, and 0.1 mM EDTA for 3 min on ice. For IL-17 (StemCell Technologies, Vancouver, BC) according to the manufacturer’s detection, the cells were stimulated (PMA/Ionomycin) as described earlier, instructions. To obtain polarized human Th17 cells, we isolated naive CD4 first stained with anti–CD4-FITC and anti-a2 integrin-PE Abs, washed, T cells from peripheral blood of healthy adult volunteers as we previously and permeabilized (CytoFix/CytoPerm ). The cells were then stained described (23). In both cases, .95% of the isolated cells were CD3/CD4 with anti–IL-17–Alexa 647 and analyzed by flow cytometry. Cells stained double positive. Human polarized Th17 cells were then generated by ac- with isotypic Abs were used as a control. tivating naive CD4+ T cells for 6 d in the presence of CD3/CD28 beads and cytokines (TGF-b,IL-1b, IL-6, and IL-23) as we previously described (23). ELISA assays of IL-17 and RANKL in sera and joint cell All RA patients and healthy volunteers signed an informed consent form, and the ethical committee of Laval University approved the study. lysates Flow cytometry analysis of human Th17 cells Blood was collected by cardiac puncture in heparinized tubes and centri- fuged to recover the serum fractions. Hind paws were removed and ho- To determine the association of a2 integrin with IL-17, we activated human mogenized on ice in a tissue grinder (Polytron, Kinematica AG) in PBS Th17 cells with PMA (50 ng/ml) and Ionomycin (1 mM) for 6 h in the containing a mixture of protease inhibitors (Roche). Samples were then presence of 5 mg/ml GolgiPlug containing brefeldin A (BD Biosciences). sonicated (Branson Sonifier 450) on ice for 2 s at power level 1, and the Subsequently, the cells were washed and first stained with anti-a2 integrin- lysates were cleared by two successive centrifugations at 4000 3 g (5 min PE or isotypic-PE. The cells were then washed, permeabilized with a at 4˚C). The supernatants were recovered and levels were assessed CytoFix/CytoPerm kit (BD Biosciences), stained for intracellular IL-17 by Bradford protein assay. Lysates with equal amounts of protein were using an anti–IL-17-Alexa 647, and analyzed using a BD FACSCalibur. then subjected to ELISA assay for the detection of IL-17 and RANKL Cells stained with isotypic Abs were used as a control. using specific ELISA kits (R&D Systems). The Journal of Immunology 5943

Microcomputerized tomography analysis and tartrate-resistant acid phosphatase assay Femurs were fixed in 4% paraformaldehyde and scanned with high-resolution microcomputerized tomography (micro-CT) scanner (SkyScan 1072; Ant- werp, Belgium). In brief, image acquisition was performed at 45 kV/222 mA at resolution of 5.63 mm per pixel with step rotation of 0.9 degree for 180 degrees and step exposure of 2.87 s. Three-dimensional reconstructions were enabled by the three-dimensional creator software (NRecon [v1.6.1.3] and CT-Analyzer [v1.8.1.2]) supplied with the instrument. Trabecular bone in- dices such as bone volume/tissue volume (BV/TV), trabecular number (Tb.N), and trabecular pattern factor (Tb.pf) were measured in the distal femur, 2.25 mm starting from growth plate. Femurs were then decalcified in 10% EDTA and were embedded in paraffin and sliced at equivalent sections through the center of the bone. Sections were prepared and stained for tartrate-resistant acid phosphatase (TRAP) activity using the Sigma-Aldrich kit. The sections were counter- stained with methyl green and examined histologically. Ten fields from each distal femur area were randomly chosen and observed microscopically at 3400 magnification, and the related pictures were analyzed by two ob- servers in a blinded manner. Osteoclasts were identified by the red staining corresponding to TRAP activity. The number of TRAP+ cells per section was calculated by counting the osteoclasts present in the 10 fields analyzed for Downloaded from each femur. The mean values of TRAP+ cells per section were then calcu- lated for the control IgG and anti-a2b1–treated groups. These experiments (micro-CT and TRAP) were carried out in the McGill Bone Center platform (McGill University, Montre´al, QC).

Assessment of human IL-17 and RORc levels

These experiments were carried out as we previously described (23). http://www.jimmunol.org/ Human Th17 cells (2 3 105 in 100 ml X-Vivo 15 medium) were activated in wells coated with anti-CD3 mAb and Coll I. The wells were first coated with 1 mg/ml anti-CD3 mAb overnight at 4˚C. The wells were then washed three times with PBS and coated with 2 mg/ml Coll I or BSA (used as a control) for 2 h at 37˚C, after which time they were washed three times with PBS. After activation, IL-17 production was measured by ELISA in cell-free supernatants from activated cells using specific ELISA kit (R&D Systems) according to the manufacturer’s instructions. IL-17 and RORc mRNA levels expressed by activated Th17 cells were evaluated by quan- titative RT-PCR (qRT-PCR) as we previously described (23). by guest on September 24, 2021 Immunoblotting and activation of MAPKs and AKT Human Th17 cells (2 3 105 in 100 ml X-Vivo 15 medium) were stimulated for 1 h, and activation of ERK, JNK, p38, and AKT were determined by immunoblot analysis as we previously described (19, 28). Statistical analysis Statistical analysis was performed by the Student t test (two-tailed, two-sample equal variance). Results with p values ,0.05 were considered significant.

Results FIGURE 1. Blockade of a2b1 integrin reduces the severity of CIA. (A) a2b1 integrin blockade inhibits the development of CIA Arthritic scores of control IgG- and a2b1 mAb–treated CIA mice (n =8) were evaluated by two blinded observers every day from days 27 (D27) To examine the role of a2b1 integrin in Th17 cells and in ex- to 36 (D36). (B) Histological assessment of CIA in forepaws from mice perimental RA, we first evaluated the effect of anti-a2b1 mAb treated with control IgG and anti-a2b1 mAb. Representative H&E-stained targeting the a2 integrin chain (Ha1/29) on CIA in mice, a model sections of proximal interphalangeal joint tissue at 3100 original magni- of RA known to be driven by Th17 cells (11). The Ha1/29 mAb fication (upper panel). Representative S/FG-stained sections of a joint in has previously been shown to have potent blocking activity in vivo the carpal region at 3400 original magnification (lower panel). (C) (29, 30). Mice treated with the anti-a2b1 mAb showed a marked Quantification of cell infiltration, cartilage destruction, and bone destruc- decrease in arthritis severity compared with the control mAb tion as scaled by two blinded observers (n = 8 for each experimental A C 6 , (IgG)-treated mice (Fig. 1A). By day 36, the arthritic clinical score group). Data shown in ( ) and ( ) are mean values SEM. *p 0.05. B, Bone; C, cartilage; I, synovial infiltration; J, joint. was 64.2% inferior in the anti-a2b1 mAb–treated group compared with the control IgG-treated group (Fig. 1A). H&E staining revealed a significant decrease in cellular infil- (Fig. 1C). Taken together, these results indicate that neutralization tration and bone destruction in paws of anti-a2b1 mAb–treated of a2b1 integrin with a specific mAb decreases arthritis severity animals in comparison with the control mAb-treated group (Fig. in the CIA murine model. 1B, upper panel). Similarly, S/FG staining of joint tissues indicated a b that anti-a2b1 integrin mAb also reduced cartilage destruction, 2 1 integrin is associated with Th17 cells from CIA mice, and because the red staining is more intense in anti-a2b1 mAb–treated its neutralization reduces Th17 cell numbers and activity in the animals (Fig. 1B, lower panel). Quantification analysis indicated joints that anti-a2b1 integrin mAb treatment reduced inflammation by We have previously shown that a2b1 integrin is the major collagen- 62%, cartilage degradation by 56%, and bone destruction by 74% binding integrin expressed by human Th17 cells (23). To examine 5944 a2b1 INTEGRIN REGULATES Th17 CELL ACTIVITY IN ARTHRITIS whether the protective effect of a2b1 integrin neutralization on the of CD4 T cells, by 50% the number of IL-17+ cells, and by severity of CIA occurs via Th17 cells, we first evaluated whether a2 39% the number of CD4/IL-17+ cells present in the joints (Fig. integrin can be associated with pathogenic Th17 cells in the CIA 3B). model. As shown in Fig. 2A and 2B, ∼40% of all CD4 T cells from We then tested whether anti-a2b1 mAb affected IL-17 protein the joints of CIA mice express a2 integrin. Furthermore, a signifi- levels. As expected, IL-17 production was reduced by 51% in the cant proportion of Th17 cells (60%) also express a2 integrin (Fig. serum and by 65% in the paw joints of CIA mice treated with anti- 2A, 2B). This is in contrast with CD4 T cells isolated from the a2b1 mAb in comparison with CIA mice treated with a control spleen where only 10% of the total CD4 T cell population expressed IgG (Fig. 3C). Together, these results show that a2b1 integrin a2 integrin (Fig. 2A, 2B). In addition, the proportion of IL-17+ cells neutralization inhibits Th17 cell numbers in the joints and reduces expressing a2 integrin is higher on those isolated from the joints IL-17 levels in CIA mice. than from the spleen (Fig. 2B, right panel). The Th17 levels a b detected in the joints in our model are comparable with those re- 2 1 integrin blockade reduces bone damage in CIA mice ported by other investigators (27, 31). These results indicate that a2 The results in Fig. 1 indicated that a2b1 integrin regulated bone integrin is associated with CD4 T cells from CIA mice and that the destruction associated with CIA. Th17 cells have a major role CD4/a2 integrin cell population is enriched in the joint tissue. This in bone degradation associated with RA. IL-17 stimulates the de- is in agreement with previous studies showing that collagen-binding velopment of osteoclasts, the cells responsible for bone resorp- integrins are expressed at higher levels in effector CD4 T cells tion, by increasing the levels of RANKL by osteoblasts and other found in extralymphoid tissues rather than those found in lym- stromal cells (13, 32). Thus, we reasoned that if a2b1 integrin phoid organs and blood circulation (15, 16, 18). neutralization affected Th17 cell activity, it would also protect Downloaded from We then examined the effect of anti-a2b1 mAb on Th17 cell CIA mice from bone damage. Accordingly, we found that treat- activity in CIA mice. To this end, total cell suspensions (total ment of CIA mice with anti-a2b1 mAb inhibited by ∼45 and 65% cells) from joint tissues of control IgG and anti-a2b1 mAb-treated the production of RANKL in sera and in the paw joints, respec- animals were characterized by flow cytometry. Double-staining tively, in comparison with CIA mice treated with control IgG (Fig. analysis indicates that anti-a2b1 mAb treatment reduces the num- 4A). This is consistent with the inhibitory effect of anti-a2b1 + ber of Th17 cells (CD4/IL-17 ) in the joints of CIA mice (Fig. 3A). mAb on IL-17 levels (Fig. 3C). In addition, TRAP staining of http://www.jimmunol.org/ Treatment with anti-a2b1 mAb reduced by ∼40% the number distal femur sections indicated that CIA mice treated with anti- by guest on September 24, 2021

FIGURE 2. Th17 cells from CIA mice express a2b1 integrin. Total cells were recovered from hind paws (joints) and spleen of CIA mice, and stimulated with PMA+Ionomycin in the presence of brefeldin and stained with FITC-, Alexa 647– and PE-conjugated isotypic (Iso) Abs or with FITC–anti-CD4, Alexa 647–anti–IL-17, and PE–anti-a2 integrin mAbs as described in Materials and Methods.(A) Representative flow cytometry profiles of total CD4 T cells and of a2/IL-17–expressing cells gated on the CD4 T cell population from the joints and spleen. (B) Mean percentage (6 SEM) of synovial or splenic CD4/a2 integrin–positive cells (left panel) and of a2 integrin–positive Th17 cells (n =5)(right panel). *p , 0.05. The Journal of Immunology 5945

FIGURE 3. Blockade of a2b1 integ- rin decreased Th17 cells in arthritic joints. Total cell suspensions from hind- paw joints of control IgG- and anti-a2b1 mAb–treated CIA mice were prepared, stimulated with PMA+Ionomycin in the presence of brefeldin, and stained with conjugated isotypic (Iso) control mAbs or with anti–CD4-FITC and anti–IL-17– Alexa 647 as described in Materials and Methods.(A)Representativeflowcytom- etry profiles from each experimental group are shown. (B) Mean percentage of CD4, IL-17, and CD4/IL-17 cells in control IgG- or anti-a2b1mAb–treatedCIAmice (n =5).(C) Assessment of IL-17 levels in sera and hind-paw joints from CIA mice Downloaded from treated with control IgG or with anti-a2b1 mAb (n = 8). Hind paws from CIA mice treated with control IgG or anti-a2b1mAb were homogenized and sonicated on ice, and lysates were normalized for protein concentration as described in Materials and Methods. IL-17 levels were mea- http://www.jimmunol.org/ sured by ELISA assay. Data shown in (B) and (C) are the mean values 6 SEM. *p , 0.05.

a2b1 mAb had fewer osteoclasts (red staining) than CIA mice a2b1 integrin costimulates IL-17 via ERK and p38 MAPKs treated with control IgG (Fig. 4B, upper panel). Quantification and PI3K/AKT pathways by guest on September 24, 2021 analysis indicates that anti-a2b1 mAb treatment reduced by 47% To gain more insights into how a2b1 integrin regulates Th17 cell the number of osteoclasts (Fig. 4B, lower panel). activation, we investigated the signaling pathways involved in Bone quality was also monitored by micro-CT analysis. Three- IL-17 production downstream of a2b1 integrin. We could not use dimensional reconstructions of trabecular BV in the distal femurs human RA synovial Th17 cells because of the limited quantities of indicated that anti-a2b1 mAb reduced trabecular bone damage synovial fluid, and thus because of the limited numbers of Th17 associated with CIA (Fig. 4C, upper panel). This concurred with cells. We therefore used the model of human Th17 cells polarized an increase in BV/TV ratio, in Tb.N, and with a decrease in Tb.pf from peripheral blood T cells in which we have shown that a2b1 observed in the anti-a2b1 mAb–treated mice (Fig. 4C, lower integrin is the major collagen-binding integrin expressed and where panel). Together, these results showed that, in addition to tissue its engagement with collagen enhanced TCR/CD3-mediated IL-17 inflammation and cartilage degradation, a2b1 integrin also reg- ulates bone destruction associated with CIA. production (23). Integrin signaling is known to activate the MAPK pathways; a2b1 integrin is associated with synovial Th17 cells from RA therefore, we tested the implication of ERK, p38, and JNK in patients collagen costimulation of IL-17. To this end, we used specific inhibitors for each MAPK (U0126 for ERK; SB203580 for p38 and The expression of a2b1 in autoimmune diseases and association SP600125 for JNK). Treatment of Th17 cells with U0126 and with Th17 cells has not been previously addressed. Thus, to fur- ther examine the importance of a2b1 integrin in Th17 cells and in SP600125 almost completely abolished TCR/CD3 (anti-CD3)- RA, we determined whether it is expressed on human RA synovial induced increase of IL-17 mRNA (Fig. 6A) and protein levels Th17 cells. We found that synovial CD4 T cells from RA patients (Fig. 6B). Similarly, both U0126 and SP600125 inhibitors abol- express significant levels of a2b1 integrin. Flow cytometry anal- ished the production of IL-17 in cells activated with anti-CD3+ ysis of a representative sample shows that a substantial percentage collagen. However, treatment of cells with the p38 inhibitor (30%) of the synovial CD4 T cells were positive for a2 integrin SB203580 had no effect on anti-CD3–induced IL-17 production chain (Fig. 5A). In addition, 50% of the total IL-17+ cells express but significantly reduced the costimulatory effect of collagen (Fig. a2 integrin (Fig. 5A). Analysis of five different samples indicates 6A, 6B). No differences in cell viability and cell proliferation were that between 26.5 and 76.8% of the total Th17 cells express a2 noticed among the different samples after 24 h of treatment (data integrin chain (Fig. 5B). The entire synovial CD4 T cell samples not shown). Immunoblot analysis showed that collagen alone in- tested expressed (95–99% positive cells) the b1 integrin chain creases ERK phosphorylation by 2.7-fold and also increases ERK (data not shown). Together, these results demonstrate that a2b1 phosphorylation in anti-CD3–activated Th17 cells by almost 2- integrin is expressed on RA synovial Th17 cells, further sup- fold (Fig. 6C, upper left panel). In contrast, collagen had no effect porting its function in Th17 cells and RA development. on JNK activation and did not costimulate anti-CD3–induced JNK 5946 a2b1 INTEGRIN REGULATES Th17 CELL ACTIVITY IN ARTHRITIS Downloaded from

FIGURE 5. a2b1 integrin is expressed on RA synovial Th17 cells. CD4 synovial T cells were activated with PMA+Ionomycin for 6 h in the presence of brefeldin and stained with control isotypic Abs or with anti-a2

integrin mAb followed with intracellular staining with anti–IL-17 mAb as http://www.jimmunol.org/ described in Materials and Methods. The cells were then analyzed by flow cytometry, and the percentage of cells in each quadrant is noted. (A) Flow cytometry analysis of a representative sample. (B) Numbers of a2 integrin/ IL-17+ cells (%) from five different RA samples.

phosphorylation, it did increase by 2-fold its activity in anti-CD3– activated Th17 cells (Fig. 6C, lower panel). Together, these results indicate that JNK and ERK are essential for anti-CD3–induced IL- 17 production, and that a2b1 integrin costimulates IL-17 pro- by guest on September 24, 2021 duction by increasing the ERK and p38 activities. Similarly, we evaluated the implication of the PI3K/AKT path- way in the production of human IL-17. We found that the specific inhibitor LY294002 completely blocks the expression and produc- tion of IL-17 elicited by anti-CD3 and by anti-CD3+collagen (Fig. 7A, 7B). Collagen did not activate AKT but increased its phos- phorylation in anti-CD3–activated Th17 cells (Fig. 7C), indicating that collagen could also costimulate IL-17 production via the PI3K/AKT pathway. Collagen increases the levels of RORc in Th17 cells via ERK and PI3K/AKT pathways RORc is the master regulator of IL-17 gene transcription and Th17 differentiation (33). Activation of Th17 cells increases RORc ex- FIGURE 4. Blockade of a2b1 integrin inhibits RANKL production, the pression, which then enhances IL-17 gene transcription. We there- number of osteoclasts, and trabecular bone loss. (A) RANKL levels in the fore sought to examine the regulation of RORc in human Th17 cells sera and hind-paw lysates (joints) of control IgG and anti-a2b1–treated by collagen/a2b1 integrin signaling. A significant 1.5-fold increase B CIA mice (n = 8) were measured by ELISA. ( ) Representative TRAP is observed after 2 h of treatment and reaches a 2-fold increase staining (corresponding to osteoclasts) of distal femur sections from con- between 4 and 8 h of treatment with anti-CD3 mAb (Fig. 8A). trol IgG- and anti-a2b1–treated CIA mice at 3400 magnification (upper panel). Quantification of osteoclast numbers in control IgG- and anti- Collagen had no effect on RORc mRNA levels but significantly a2b1–treated CIA mice (n =5)(lower panel). (C) Representative three- costimulated the anti-CD3 effect. The strongest effect of collagen dimensional reconstruction of micro-CT images of trabecular BV from (2-fold) is observed at 2–4 h and to a lesser extent at 8 h of anti-CD3 control IgG- and anti-a2b1 mAb–treated CIA mice (upper panel). His- treatment (Fig. 8A). The costimulatory effect of collagen is blocked tomorphometric analysis of trabecular BV from control IgG- and anti- by an anti-a2b1 integrin mAb (data not shown), which is consistent a2b1–treated mice (n = 5). BV/TV, Tb.N, and Tb.pf (lower panels). Data with our previous study showing that the blocking anti-a2b1integrin are the mean 6 SEM. *p , 0.05. mAb inhibits the costimulatory effect of collagen on IL-17 produc- tion (23). The ERK and PI3K/AKT inhibitors abolished the ex- phosphorylation (Fig. 6C, upper right panel). Finally, we found pression of RORc elicited by anti-CD3 and anti-CD3+collagen (Fig. that the TCR/CD3 complex is a weak inducer of p38 phosphor- 8B). However, the JNK inhibitor, which also blocked IL-17 ex- ylation, and although collagen by itself did not stimulate p38 pression, and the p38 inhibitor, which blocked the costimulatory The Journal of Immunology 5947 Downloaded from http://www.jimmunol.org/

FIGURE 7. Collagen costimulation of IL-17 involves the PI3K/AKT by guest on September 24, 2021 pathway. Th17 cells were stimulated with 1 mg/ml coated anti-CD3 mAb in the presence or absence of 2 mg/ml coated collagen (Coll I). The LY294002 (LY), a specific inhibitor of PI3K/AKT, was added at 10 mM1h before cell stimulation. (A) IL-17 expression was determined by qRT-PCR. (B) IL-17 production was determined by ELISA assay. The results in (A) and (B) represent mean values (6 SD) from five different experiments performed in triplicate with T cells isolated from five different donors. (C) AKT activation was evaluated by immunoblot analysis using an Ab rec- ognizing the phosphorylated form of AKT. The membrane was stripped and reprobed with an Ab directed against total AKT to ensure equal loading. Representative immunoblots from three independent experiments are shown. FIGURE 6. Collagen costimulates IL-17 expression via ERK and p38 The numbers under the bands represent fold increase of the ratio between MAPKs. Th17 cells were stimulated with 1 mg/ml coated anti-CD3 mAb total phospho-AKT and total AKT. *p , 0.05. NT, Nontreated cells. in the presence or absence of 2 mg/ml coated collagen (Coll I). Specific inhibitors of MAPKs (U0126 [U] for the ERK inhibitor; SP600125 [SP] for the JNK inhibitor and SB203580 [SB] for the p38 inhibitor) were added effect of collagen on IL-17 expression, had no effect (Fig. 8B). at 10 mM for 1 h before cell stimulation. (A) IL-17 mRNA levels were Together, these results show that the ERK/MAPK and PI3K/AKT determined after 6 h of activation by qRT-PCR, and the results are pre- RORc sented as fold increase normalized to b-actin levels. (B) The IL-17 protein pathways are essential for expression, and that collagen levels were determined after 24 h of activation by ELISA assay. The results enhances RORc mRNA levels via the ERK and AKT pathways in (A) and (B) represent mean values (6 SD) from five different experi- and independently from the p38 MAPK. ments performed in triplicate with T cells isolated from five different blood donors. (C) Collagen activation of ERK, JNK, and p38 MAPKs. The cells Discussion were activated with coated anti-CD3 mAb, collagen I (Coll I), or anti-CD3+ Th17 cells are major effector cells in the pathogenesis of RA and collagen for 1 h. The cells were lysed, and activation of ERK (left upper other autoimmune diseases. However, the mechanisms regulating panel), JNK (right upper panel), and p38 (lower panel) was determined by their activity are not fully understood. Growing evidence suggests immunoblot analysis using specific Abs recognizing the phosphorylated that cell–ECM interactions could be an important factor in the forms. The blots were stripped and reprobed with anti–ERK-2 and anti–b- actin Abs as indicated to ensure equal loading. Representative immuno- regulation of T cell–mediated immunity and inflammation (15, blots from three independent experiments are shown. The numbers under 16). In this study, we have shown that targeting the type I collagen the bands represent fold increase of the ratio between total phospho-ERK1/ receptor, a2b1 integrin, reduces the severity of arthritis in the CIA 2 and ERK-2, phospho-JNK1/2 and b-actin, and phospho-p38 and b-actin. murine model of RA by affecting Th17 cell activity. Our results *p , 0.05. NT, Nontreated cells. demonstrate that a2b1 integrin is expressed on synovial Th17 5948 a2b1 INTEGRIN REGULATES Th17 CELL ACTIVITY IN ARTHRITIS

This could be explained by three nonexclusive mechanisms. First, a2b1 integrin can be important for the migration and retention of Th17 cells in the joint tissue. In the influenza infection animal model, a2b1 integrin was not found to be important for effector CD4 T cell migration to the lungs (35). However, a2b1 integrin seems to be necessary for the homing of CD4 effector/memory T cells to the bone marrow (36). These studies suggest that a2b1 integrin can regulate the migration of effector CD4 T cells to the tissues that are rich in collagen such as the bone marrow and the synovium. Accordingly, inhibition of Th17 cell migration to the joints of CIA mice is likely to represent one mechanism contrib- uting to the function of a2b1integrininTh17cellsandinCIA pathogenesis. Inhibition of apoptosis is a hallmark in autoimmune disease and RA. We and others have previously demonstrated that a2b1 integrin engagement with collagen inhibits Fas-induced ap- optosis of effector T cells (19, 22), suggesting the possibility that a2b1 integrin could promote Th17 cell survival in the arthritic joint tissue. Finally, a2b1 integrin could also costimulate Th17 cells; thus, its blockade could lead to the reduction of IL-17 levels ob- Downloaded from served in anti-a2 b1 mAb–treated animals. This is supported by the fact that a2b1 integrin ligation with collagens, which are abundant in the synovium, significantly enhances the production of IL-17 by human Th17 cells (23). The fact that the anti-a2 integrin blocking Ab also reduced serum IL-17 levels in CIA mice is likely to be the consequence of the observed reduction of IL-17 in the joints. http://www.jimmunol.org/ FIGURE 8. Collagen increases the expression levels of RORc mRNA A recent study using the a2b1 integrin knockout mice has via ERK/MAPK and PI3K/AKT. (A) Th17 cells were stimulated for 1, 2, shown that a2b1 integrin also regulates synovial fibroblast pro- 4, and 8 h with or without anti-CD3 mAb in the presence or absence of duction of MMP-3 and subsequent cartilage destruction in RA collagen (Coll I). RORc expression was evaluated by qRT-PCR, and the animal models (37). IL-17 is known to promote cartilage degra- data are presented as fold increase normalized to b-actin expression levels. dation by enhancing the production by synovial fibroblasts and (B) Th17 cells were stimulated for 4 h with anti-CD3 mAb or with anti- CD3+collagen in the presence or absence of specific inhibitors of MAPK chondrocytes of various metalloproteinases involved in matrix break- or PI3K/AKT as indicated, and the expression levels of RORc mRNA were down (38, 39). Thus, a2b1integrinislikelytocontributetocar- 6 determined by qRT-PCR. The results represent mean values ( SD) from tilage destruction associated with arthritis by enhancing Th17 cell by guest on September 24, 2021 five different experiments performed in triplicate with T cells isolated from activity (this study) and through activation of synovial fibroblasts, five different blood donors. *p , 0.05. further emphasizing the importance of the a2b1 integrin pathway in RA pathogenesis. In addition to cartilage degradation, our study showed that a2b1 integrin promotes bone loss associated with RA, cells from CIA mice, and its neutralization with a specific mAb reduces the number of Th17 cells infiltrating the joints and the which is a major factor in increased fractures observed in RA levels of IL-17 in CIA mice. Blockade of a2b1 integrin led to patients. Neutralization of a2b1 integrin reduced RANKL levels reduction of synovial inflammation, cartilage destruction, and a and osteoclast numbers, and protected the CIA mice from trabecular reduction of RANKL levels, osteoclast numbers, and bone loss. bone loss. Th17 cells are critical in the development of osteoclasts Th17 cells play a central role in orchestrating the inflammatory and bone erosion (13, 32). IL-17 increases the levels of RANKL, response and tissue destruction in the joints, and the model of CIA a crucial cytokine involved in the development of osteoclasts, is a Th17-driven disease (6, 10, 11, 13). Therefore, our results which are responsible for bone erosion. Therefore, it is likely that suggest that the observed inhibition of Th17 cell activity is likely the function of a2b1 in bone degradation occurs, at least in part, at to represent a major mechanism underlying the protective effect of the level of Th17 cell activation. In support, a2b1integrinex- anti-a2b1 mAb treatment on the severity of CIA. pression is not associated with neutrophils or the monocytic/mac- We have recently reported that a2b1integrinisthemajor rophage lineage (40–42), which constitutes the main source of collagen-binding integrin expressed on human Th17 cells polar- osteoclast precursors. In addition, previous studies and recent ized from naive CD4 T cells, and that its ligation with collagen evidence from the a2b1 integrin–deficient mice indicated that enhanced the production of IL-17 (23). A previous study using although osteoclasts may express a2b1 integrin, it seems that their immunoblot analysis has reported that a2b1 can be expressed function depends more on avb3 integrin rather than on a2b1 and on human RA synovial T cells (34). In this study, and although other integrins (43, 44). In contrast, a2b1 integrin has been shown we were limited in RA samples and synovial fluid quantities, we to regulate mast cell activation (45, 46), and mast cells have been further show that a2b1 integrin is expressed on human RA sy- associated with the pathogenesis of arthritis (47). Although mast novial Th17 cells. This study also provides evidence for a patho- cells are not required in the model of CIA used in this study (48), physiological role of a2b1integrininTh17cellactivityassociated the earlier findings suggest that a2 integrin blockade can also with a murine model of RA. Taken together, these results strongly affect mast cell activity during the course of arthritis and, there- support an important role for a2b1 integrin in Th17 cell activation fore, contribute to the reduction of arthritis severity. Thus, al- and suggest that it can represent an important pathogenic pathway though a2b1 integrin can contribute to RA by regulating other cell in RA. types such as fibroblasts but given the critical role of Th17 cells in Our results showed that a2b1 integrin controls both the number RA and in the CIA model, our results indicate that the functional of Th17 cells in the joints and the levels of IL-17 in CIA mice. association of a2b1 integrin with Th17 cells is a major pathway The Journal of Immunology 5949 involved in inflammation and tissue destruction associated with 4. Leipe, J., M. Grunke, C. Dechant, C. Reindl, U. Kerzendorf, H. Schulze-Koops, and A. Skapenko. 2010. Role of Th17 cells in human autoimmune arthritis. arthritis. Arthritis Rheum. 62: 2876–2885. Blockade of a2b1 integrin using the same Ab as we did in this 5. Murphy, C. A., C. L. Langrish, Y. Chen, W. Blumenschein, T. McClanahan, study also reduced the symptoms of experimental autoimmune R. A. Kastelein, J. D. Sedgwick, and D. J. Cua. 2003. Divergent pro- and anti- inflammatory roles for IL-23 and IL-12 in joint autoimmune inflammation. J. Exp. encephalomyelitis (EAE) (29), an animal model of multiple Med. 198: 1951–1957. sclerosis. However, that study did not investigate the mechanisms 6. Nakae, S., A. Nambu, K. Sudo, and Y. Iwakura. 2003. Suppression of immune underlying the protective effect of anti-a2b1 mAb treatment. induction of collagen-induced arthritis in IL-17-deficient mice. J. Immunol. 171: 6173–6177. Because EAE and multiple sclerosis in humans are Th17-based 7. Nakae, S., S. Saijo, R. Horai, K. Sudo, S. Mori, and Y. Iwakura. 2003. IL-17 diseases, it is possible that blockade of a2b1 integrin could have production from activated T cells is required for the spontaneous development of destructive arthritis in mice deficient in IL-1 receptor antagonist. Proc. Natl. blocked EAE symptoms by blocking Th17 cell activity as in the Acad. Sci. USA 100: 5986–5990. CIA model (this study). 8. Zizzo, G., M. De Santis, S. L. Bosello, A. L. Fedele, G. Peluso, E. Gremese, Finally, to further understand the function of a2b1 integrin in B. Tolusso, and G. Ferraccioli. 2011. Synovial fluid-derived T helper 17 cells correlate with inflammatory activity in arthritis, irrespectively of diagnosis. Clin. Th17 cells, we undertook in vitro experiments to decipher the Immunol. 138: 107–116. signaling pathways activated by a2b1 integrin. We found that 9. Dardalhon, V., T. Korn, V. K. Kuchroo, and A. C. Anderson. 2008. Role of Th1 ligation of a2b1 integrin with collagen promotes IL-17 production and Th17 cells in organ-specific autoimmunity. J. Autoimmun. 31: 252–256. 10. Lubberts, E. 2010. Th17 cytokines and arthritis. Semin. Immunopathol. 32: 43– by enhancing the activation of ERK and p38 MAPKs and PI3K/ 53. AKT pathways. Collagen also increased the expression of RORc, 11. Lubberts, E., L. A. Joosten, B. Oppers, L. van den Bersselaar, C. J. Coenen-de Roo, J. K. Kolls, P. Schwarzenberger, F. A. van de Loo, and W. B. van den Berg. which occurred through ERK and AKT pathways independently 2001. IL-1-independent role of IL-17 in synovial inflammation and joint de- from the p38 MAPK. struction during collagen-induced arthritis. J. Immunol. 167: 1004–1013. Downloaded from Interestingly, ERK, p38, and AKT have been involved with the 12. Pelletier, M., L. Maggi, A. Micheletti, E. Lazzeri, N. Tamassia, C. Costantini, L. Cosmi, C. Lunardi, F. Annunziato, S. Romagnani, and M. A. Cassatella. 2010. development of Th17 cells in mice, and inhibition of p38 and ERK Evidence for a cross-talk between human neutrophils and Th17 cells. Blood 115: pathways attenuated autoimmune diseases such as EAE and colitis, 335–343. respectively (49–53). Together, these studies further support our 13. Lubberts, E., L. van den Bersselaar, B. Oppers-Walgreen, P. Schwarzenberger, C. J. Coenen-de Roo, J. K. Kolls, L. A. Joosten, and W. B. van den Berg. 2003. findings in human Th17 cells and reinforce the importance of a2b1 IL-17 promotes bone erosion in murine collagen-induced arthritis through loss of integrin in Th17 cell activation and in the development of CIA. The the receptor activator of NF-kappa B ligand/osteoprotegerin balance. J. Immunol. http://www.jimmunol.org/ 170: 2655–2662. fact that a2b1 integrin has been shown to regulate the development 14. Hemler, M. E. 1990. VLA in the integrin family: structures, functions, of EAE (29) suggests that it can do so by increasing the production and their role on leukocytes. Annu. Rev. Immunol. 8: 365–400. of IL-17 through ERK, p38, and AKT pathways, which may also 15. Pribila, J. T., A. C. Quale, K. L. Mueller, and Y. Shimizu. 2004. Integrins and T cell-mediated immunity. Annu. Rev. Immunol. 22: 157–180. operate downstream of a2b1 integrin in CIA. Although additional 16. Dustin, M. L., and A. R. de Fougerolles. 2001. Reprogramming T cells: the role work is needed to understand how these signaling pathways regulate of extracellular matrix in coordination of T cell activation and migration. Curr. Th17 cells and autoimmune diseases, our findings with human po- Opin. Immunol. 13: 286–290. 17. Baje´noff, M., J. G. Egen, L. Y. Koo, J. P. Laugier, F. Brau, N. Glaichenhaus, and larized Th17 cells argue in favor of targeting these pathways along R. N. Germain. 2006. Stromal cell networks regulate entry, migra- with a2b1 integrin for the treatment of RA and other Th17- tion, and territoriality in lymph nodes. Immunity 25: 989–1001.

18. Krivacic, K. A., and A. D. Levine. 2003. Extracellular matrix conditions T cells by guest on September 24, 2021 dependent diseases. However, although our study shows the im- for adhesion to tissue interstitium. J. Immunol. 170: 5034–5044. portance of a2b1 integrin in the development of arthritis, it is 19. Gendron, S., J. Couture, and F. Aoudjit. 2003. Integrin alpha2beta1 inhibits Fas- unclear whether its neutralization will provide clinical benefit in mediated apoptosis in T by protein phosphatase 2A-dependent activation of the MAPK/ERK pathway. J. Biol. Chem. 278: 48633–48643. arthritis and other inflammatory diseases. Given the complexity of 20. Richter, M., S. J. Ray, T. J. Chapman, S. J. Austin, J. Rebhahn, T. R. Mosmann, ECM and their receptors, it is possible that additional b1orb3 H. Gardner, V. Kotelianski, A. R. deFougerolles, and D. J. Topham. 2007. integrin pathways might compensate for its inhibition. Collagen distribution and expression of collagen-binding alpha1beta1 (VLA-1) and alpha2beta1 (VLA-2) integrins on CD4 and CD8 T cells during influenza In summary, to our knowledge, our study provides the first evi- infection. J. Immunol. 178: 4506–4516. dence that a2b1 integrin is expressed on human RA synovial Th17 21. Andreasen, S. O., A. R. Thomsen, V. E. Koteliansky, T. I. Novobrantseva, A. G. Sprague, A. R. de Fougerolles, and J. P. Christensen. 2003. Expression and cells and that it regulates the development of inflammatory diseases functional importance of collagen-binding integrins, alpha 1 beta 1 and alpha 2 (CIA) by targeting Th17 cells, suggesting that a2b1integrincan beta 1, on virus-activated T cells. J. Immunol. 171: 2804–2811. represent an important pathogenic pathway in RA. Further un- 22. Lin, Y. P., C. C. Su, J. Y. Huang, H. C. Lin, Y. J. Cheng, M. F. Liu, and B. C. Yang. 2009. Aberrant integrin activation induces p38 MAPK phosphory- derstanding of the mechanisms by which a2b1 integrin regulates lation resulting in suppressed Fas-mediated apoptosis in T cells: implications for Th17 cell activation and RA is likely to lead to new insights in au- rheumatoid arthritis. Mol. Immunol. 46: 3328–3335. toimmunity and RA pathogenesis, and to the development of new 23. Boisvert, M., N. Chetoui, S. Gendron, and F. Aoudjit. 2010. Alpha2beta1 integrin is the major collagen-binding integrin expressed on human Th17 cells. therapeutic avenues. Eur. J. Immunol. 40: 2710–2719. 24. Chan, M. W., B. Hinz, and C. A. McCulloch. 2010. Mechanical induction of gene expression in connective tissue cells. Methods Cell Biol. 98: 178–205. Disclosures 25. Wen, T., Y. Li, M. Wu, X. Sun, X. Bao, Y. Lin, J. Hao, L. Han, G. Cao, Z. Wang, The authors have no financial conflicts of interest. et al. 2012. Therapeutic effects of a novel tylophorine analog, NK-007, on collagen-induced arthritis through suppressing tumor necrosis factor a produc- tion and Th17 cell differentiation. Arthritis Rheum. 64: 2896–2906. 26. Cesaro, A., N. Anceriz, A. Plante, N. Page´, M. R. Tardif, and P. A. Tessier. 2012. References An inflammation loop orchestrated by S100A9 and calprotectin is critical for 1. Boissier, M. C., L. Semerano, S. Challal, N. Saidenberg-Kermanac’h, and development of arthritis. PLoS ONE 7: e45478. G. Falgarone. 2012. Rheumatoid arthritis: from autoimmunity to synovitis and 27. Po¨llinger, B., T. Junt, B. Metzler, U. A. Walker, A. Tyndall, C. Allard, S. Bay, joint destruction. J. Autoimmun. 39: 222–228. R. Keller, F. Raulf, F. Di Padova, et al. 2011. Th17 cells, not IL-17+ gd T cells, 2. Genovese, M. C., F. Van den Bosch, S. A. Roberson, S. Bojin I. M. Biagini, drive arthritic bone destruction in mice and humans. J. Immunol. 186: 2602– P. Ryan, and J. Sloan-Lancaster. 2010. LY2439821, a humanized anti-interleukin- 2612. 17 , in the treatment of patients with rheumatoid arthritis: 28. Naci, D., M. A. El Azreq, N. Chetoui, L. Lauden, F. Sigaux, D. Charron, R. Al- a phase I randomized, double-blind, placebo-controlled, proof-of-concept study. Daccak, and F. Aoudjit. 2012. a2b1 integrin promotes chemoresistance against Arthritis Rheum. 62: 929–939. doxorubicin in cancer cells through extracellular signal-regulated (ERK). 3. Kirkham, B. W., M. N. Lassere, J. P. Edmonds, K. M. Juhasz, P. A. Bird, C. S. Lee, J. Biol. Chem. 287: 17065–17076. R. Shnier, and I. J. Portek. 2006. Synovial membrane cytokine expression is pre- 29. Tsunoda, I., E. J. Terry, B. J. Marble, E. Lazarides, C. Woods, and dictive of joint damage progression in rheumatoid arthritis: a two-year prospective R. S. Fujinami. 2007. Modulation of experimental autoimmune encephalomy- study (the DAMAGE study cohort). Arthritis Rheum. 54: 1122–1131. elitis by VLA-2 blockade. Brain Pathol. 17: 45–55. 5950 a2b1 INTEGRIN REGULATES Th17 CELL ACTIVITY IN ARTHRITIS

30. Jafri, M., B. Donnelly, S. Allen, A. Bondoc, M. McNeal, P. D. Rennert, 42. Ammon, C., S. P. Meyer, L. Schwarzfischer, S. W. Krause, R. Andreesen, and P. H. Weinreb, R. Ward, and G. Tiao. 2008. Cholangiocyte expression of M. Kreutz. 2000. Comparative analysis of integrin expression on monocyte- alpha2beta1-integrin confers susceptibility to rotavirus-induced experimental bili- derived macrophages and monocyte-derived dendritic cells. Immunology 100: ary atresia. Am. J. Physiol. Gastrointest. Liver Physiol. 295: G16–G26. 364–369. 31. Pineda, M. A., M. A. McGrath, P. C. Smith, L. Al-Riyami, J. Rzepecka, 43. Uemura, T., Y. K. Liu, and Y. Kuboki. 2000. Preliminary communication. J. A. Gracie, W. Harnett, and M. M. Harnett. 2012. The parasitic helminth MRNA expression of MT1-MMP, MMP-9, cathepsin K, and TRAP in highly product ES-62 suppresses pathogenesis in collagen-induced arthritis by targeting enriched osteoclasts cultured on several matrix proteins and ivory surfaces. the interleukin-17-producing cellular network at multiple sites. Arthritis Rheum. Biosci. Biotechnol. Biochem. 64: 1771–1773. 64: 3168–3178. 44. Stange, R., D. Kronenberg, M. Timmen, J. Everding, H. Hidding, B. Eckes, 32. Okamoto, K., and H. Takayanagi. 2011. Osteoclasts in arthritis and Th17 cell U. Hansen, M. Holtkamp, U. Karst, T. Pap, and M. J. Raschke. 2013. Age-related development. Int. Immunopharmacol. 11: 543–548. bone deterioration is diminished by disrupted collagen sensing in integrin a2b1 33. Ivanov, I. I., B. S. McKenzie, L. Zhou, C. E. Tadokoro, A. Lepelley, J. J. Lafaille, deficient mice. Bone 56: 48–54. D. J. Cua, and D. R. Littman. 2006. The orphan nuclear receptor RORgammat 45. Edelson, B. T., T. P. Stricker, Z. Li, S. K. Dickeson, V. L. Shepherd, directs the differentiation program of proinflammatory IL-17+ T helper cells. S. A. Santoro, and M. M. Zutter. 2006. Novel collectin/C1q receptor mediates Cell 126: 1121–1133. mast cell activation and innate immunity. Blood 107: 143–150. 34. Hemler, M. E., D. Glass, J. S. Coblyn, and J. G. Jacobson. 1986. Very late ac- 46. McCall-Culbreath, K. D., Z. Li, Z. Zhang, L. X. Lu, L. Orear, and M. M. Zutter. tivation antigens on rheumatoid synovial fluid T lymphocytes. Association with 2011. Selective, a2b1 integrin-dependent secretion of il-6 by connective tissue stages of T cell activation. J. Clin. Invest. 78: 696–702. mast cells. J. Innate Immun. 3: 459–470. 35. Kassiotis, G., D. Gray, Z. Kiafard, J. Zwirner, and B. Stockinger. 2006. Func- 47. Kenna, T. J., and M. A. Brown. 2013. The role of IL-17-secreting mast cells in tional specialization of memory Th cells revealed by expression of integrin inflammatory joint disease. Nat. Rev. Rheumatol. 9: 375–379. CD49b. J. Immunol. 177: 968–975. 48. Pitman, N., D. L. Asquith, G. Murphy, F. Y. Liew, and I. B. McInnes. 2011. 36. Tokoyoda, K., S. Zehentmeier, A. N. Hegazy, I. Albrecht, J. R. Gru¨n, Collagen-induced arthritis is not impaired in mast cell-deficient mice. Ann. M. Lo¨hning, and A. Radbruch. 2009. Professional memory CD4+ T lymphocytes Rheum. Dis. 70: 1170–1171. preferentially reside and rest in the bone marrow. Immunity 30: 721–730. 49. Brereton, C. F., C. E. Sutton, S. J. Lalor, E. C. Lavelle, and K. H. Mills. 2009. 37. Peters, M. A., D. Wendholt, S. Strietholt, S. Frank, N. Pundt, A. Korb-Pap, Inhibition of ERK MAPK suppresses IL-23- and IL-1-driven IL-17 production L. A. Joosten, W. B. van den Berg, G. Kollias, B. Eckes, and T. Pap. 2012. The and attenuates autoimmune disease. J. Immunol. 183: 1715–1723. Downloaded from loss of a2b1 integrin suppresses joint inflammation and cartilage destruction in 50. Kurebayashi, Y., S. Nagai, A. Ikejiri, M. Ohtani, K. Ichiyama, Y. Baba, T. Yamada, mouse models of rheumatoid arthritis. Arthritis Rheum. 64: 1359–1368. S. Egami, T. Hoshii, A. Hirao, et al. 2012. PI3K-Akt-mTORC1-S6K1/2 axis 38. Koshy, P. J., N. Henderson, C. Logan, P. F. Life, T. E. Cawston, and A. D. Rowan. controls Th17 differentiation by regulating Gfi1 expression and nuclear trans- 2002. Interleukin 17 induces cartilage collagen breakdown: novel synergistic effects location of RORg. Cell Rep 1: 360–373. in combination with proinflammatory cytokines. Ann. Rheum. Dis. 61: 704–713. 51. Liu, H., S. Yao, S. M. Dann, H. Qin, C. O. Elson, and Y. Cong. 2013. ERK 39. Lubberts, E., L. A. Joosten, F. A. van de Loo, L. A. van den Gersselaar, and differentially regulates Th17- and Treg-cell development and contributes to the W. B. van den Berg. 2000. Reduction of interleukin-17-induced inhibition of pathogenesis of colitis. Eur. J. Immunol. 43: 1716–26.

chondrocyte proteoglycan synthesis in intact murine articular cartilage by in- 52. Noubade, R., D. N. Krementsov, R. Del Rio, T. Thornton, V. Nagaleekar, http://www.jimmunol.org/ terleukin-4. Arthritis Rheum. 43: 1300–1306. N. Saligrama, A. Spitzack, K. Spach, G. Sabio, R. J. Davis, et al. 2011. Acti- 40. Edelson, B. T., Z. Li, L. K. Pappan, and M. M. Zutter. 2004. Mast cell-mediated vation of p38 MAPK in CD4 T cells controls IL-17 production and autoimmune inflammatory responses require the alpha 2 beta 1 integrin. Blood 103: 2214–2220. encephalomyelitis. Blood 118: 3290–3300. 41. Mambole, A., S. Bigot, D. Baruch, P. Lesavre, and L. Halbwachs-Mecarelli. 53. Wan, Q., L. Kozhaya, A. ElHed, R. Ramesh, T. J. Carlson, I. M. Djuretic, 2010. Human neutrophil integrin alpha9beta1: up-regulation by cell activation M. S. Sundrud, and D. Unutmaz. 2011. Cytokine signals through PI-3 kinase and synergy with beta2 integrins during adhesion to endothelium under flow. J. pathway modulate Th17 cytokine production by CCR6+ human memory T cells. Leukoc. Biol. 88: 321–327. J. Exp. Med. 208: 1875–1887. by guest on September 24, 2021