With Fibrinogen Activation of Mast Cells Through Interaction Induces

The Journal of Immunology

␣ ␤ Integrin IIb 3 Induces the Adhesion and Activation of Mast Cells through Interaction with Fibrinogen1

Toshihiko Oki,* Jiro Kitaura,* Koji Eto,† Yang Lu,* Mari Maeda-Yamamoto,‡ Naoki Inagaki,§ Hiroichi Nagai,§ Yoshinori Yamanishi,* Hideaki Nakajina,* Hidetoshi Kumagai,* and Toshio Kitamura2*

␣ Integrin IIb, a well-known marker of megakaryocyte-platelet lineage, has been recently recognized on hemopoietic progenitors. ␣ ␤ We now demonstrate that integrin IIb 3 is highly expressed on mouse and human mast cells including mouse bone marrow- derived mast cells, peritoneal mast cells, and human cord blood-derived mast cells, and that its binding to extracellular matrix proteins leads to enhancement of biological functions of mast cells in concert with various stimuli. With exposure to various stimuli, including cross-linking of Fc⑀RI and stem cell factor, mast cells adhered to extracellular matrix proteins such as fibrinogen ␣ ␤ and von Willebrand factor in an integrin IIb 3-dependent manner. In addition, the binding of mast cells to fibrinogen enhanced proliferation, cytokine production, and migration and induced uptake of soluble fibrinogen in response to stem cell factor stim- ␣ ␤ ulation, implicating integrin IIb 3 in a variety of mast cell functions. In conclusion, mouse and human mast cells express ␣ ␤ functional integrin IIb 3. The Journal of Immunology, 2006, 176: 52–60.

ast cells, derived from circulating CD34ϩ hemopoietic Under both IgE-dependent and -independent pathological con- progenitor cells, differentiate and proliferate in vascu- ditions, cell-extracellular matrix (ECM) interactions mediated by larized tissues. These steps are critically regulated by integrins play crucial roles in a variety of mast cell functions such M 3 stem cell factor (SCF), the ligand for c-kit, which is a receptor as histamine release (11), cytokine production (3, 5), survival (5), tyrosine kinase expressed on the surface of mast cells (1, 2) as well growth (12), and migration (13, 14). as immature hemopoietic cells. Integrins are heterodimeric type I transmembrane receptors It is widely accepted that mast cells play a critical role in IgE- ␣ ␤ ␣ composed of two subunits ( and ). Integrin IIb, recently proven mediated immune reactions such as immediate hypersensitivity; to be a marker for early hemopoietic progenitors (15), was con- mast cells are activated by cross-linking of Ag-specific IgE bound sidered to be expressed exclusively on megakaryocyte-platelet lin- to Fc⑀RI with a multivalent specific Ag. Mast cells secrete pre- ␤ ␣ ␤ eage as a complex with integrin 3 to form integrin IIb 3, also formed and newly synthesized proinflammatory mediators such as ␤ called glycoprotein IIb-IIIa or CD41-CD61, whereas integrin 3 is histamine, lipids, and cytokines (1–3). In addition, some sets of ␣ present on many types of cells as a complex with integrin V.In monomeric IgE molecules, termed high cytokinergenic (HC) IgE, ␣ ␤ platelets, integrin IIb 3 works as a receptor for fibrinogen, von can also activate mast cells and induce a similar response without Willebrand factor (vWF), vitronectin (VN), fibronectin (FN), specific Ags (1, 4, 5). ␣ ␤ CD40L, and others (16). Integrin IIb 3 links platelets to the in- In contrast, mast cells also participate in a wide variety of patho- jured sites of vessels through interaction with fibrinogen and vWF, logical processes independent of IgE, including innate immune together with an aggregate to form a platelet plug, leading to he- response (1, 2, 6), tissue repair (1, 2, 7), acute inflammatory re- mostasis (16–21). Eventually, fibrinogen is internalized and stored sponse to implanted biomaterials (8), atherosclerosis (9), and cer- in granules of megakaryocytes and platelets (22–25). tain autoimmune disorders (10). The activation of bone marrow-derived mast cell (BMMC) with various stimuli, including IgE and Ag (this mode of stimulation *Division of Cellular Therapy and Division of Hematopoietic Factors, Advanced termed IgE plus Ag) (1, 2, 11, 18, 26), monomeric IgE (1, 5, 27), Clinical Research Center, and †Laboratory of Stem Cell Therapy, Center for Exper- imental Medicine, Institute of Medical Science, University of Tokyo, Tokyo, Japan; SCF (28, 29), and thrombin (30) induces the adhesion to FN pre- ‡National Institute of Vegetable and Tea Science, National Agriculture Research Or- ␣ ␤ ␤ dominantly via integrin 5 1. BMMC stimulated by TGF- adhere ganization, Shizuoka, Japan; and §Department of Pharmacology, Gifu Pharmaceutical ␣ University, Gifu, Japan to laminin-1 via integrin 7 (31), whereas adhesion of BMMC to VN is mediated by integrin ␣ ␤ (12). IgE plus Ag-stimulated Received for publication May 20, 2005. Accepted for publication October 10, 2005. V 3 peritoneal mast cells (PMC), but not BMMC, adhere to type I The costs of publication of this article were defrayed in part by the payment of page ␣ ␤ charges. This article must therefore be hereby marked advertisement in accordance collagen via integrin 2 1 (32). However, functions of integrin with 18 U.S.C. Section 1734 solely to indicate this fact. ␣ ␤ IIb 3 have never been addressed in mast cells. 1 ␣ ␤ This study was supported by the Ministry of Education, Science, Technology, Sports In the present work, we demonstrate that integrin IIb 3 is and Culture and the Ministry of Health and Welfare, Japan. The Division of Hema- topoietic Factors is supported by the Chugai Pharmaceutical Company, Tokyo, Japan. highly expressed on mast cells using mouse BMMC, PMC, and human cord blood- derived mast cells. We also found that mast 2 Address correspondence and reprint requests to Dr. Toshio Kitamura, Division of Cellular Therapy, Advanced Clinical Research Center, Institute of Medical Science, cells stimulated by various stimuli adhere to ECM protein such as University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8039, Japan. E-mail ␣ ␤ fibrinogen and vWF in an integrin IIb 3-dependent manner. address: [email protected] ␣ ␤ Moreover, the interaction of integrin IIb 3 to fibrinogen enhances 3 Abbreviations used in this paper: SCF, stem cell factor; HC, high cytokinergenic; ECM, extracellular matrix; BMMC, bone marrow-derived mast cell; PMC, peritoneal mast cell functions and induces uptake of fibrinogen into mast mast cell; vWF, von Willebrand factor; VN, vitronectin; FN, fibronectin. cells. Considering that several drugs regulate the function of integrin

␣ ␤ ␣ ␤ IIb 3 (33), these results might provide a feasible way to overcome V 3 Ab (2C9.G2). Immunoprecipitates were run on 7.5% SDS-PAGE mast cell-mediated disorders. gels and transferred to Immobilon membranes. The membranes were blot- ted using HRP-conjugated streptavidin. Materials and Methods Adhesion assay Abs and other materials Adhesion assay was done as described previously (28). In brief, the 96-well ␣ ␤ ␮ The source of Abs are as described. Anti-mouse integrin IIb 3 mAb (1B5) plates (no. 3631; Corning) were coated with 20 or 50 g/ml VN, vWF, or was a gift from Dr. B. S. Coller (Rockefeller University, New York, NY) fibrinogen in PBS for1hat37°C, and washed three times with PBS, ␣ ␤ (34). Anti-mouse integrin V (8B3) and anti-mouse integrin 3 (8B11) followed by blocking with 3% BSA/PBS for1hat37°C and washing three mAbs were gifts from Drs. D. J. Gerber and S. Tonegawa (Picower Center, times with RPMI 1640 containing 10 mM HEPES and 0.03% BSA (the Massachusetts Institute of Technology, Boston, MA) (35). Anti-human in- assay medium). BMMC, washed four times in the assay medium, were ␣ ␤ ϫ 5 tegrin IIb 3 mAb (2G12) was a gift from Dr. V. L. Woods (University of resuspended at 5 10 cells/ml in the assay medium and transferred into California, San Diego, CA) (36). Anti-DNP IgE (SPE-7) was from Sigma- coated wells (100 ␮l/well) with or without stimulant including IgE (SPE- ␣ Aldrich. Anti-TNP IgE (C38-2), anti-mouse integrin IIb (MWReg30), anti- 7), DNP-BSA (for IgE plus Ag), SCF, and thrombin with the indicated ␣ ␣ mouse integrin V (RMV-7 and H9.2B8), anti-mouse integrin 4 (9C10), concentration for1hat37°C. For stimulation with IgE plus Ag, BMMC ␣ ␮ anti-mouse integrin 5 (5H10-27), anti-mouse LFA1 (M17/4), anti-mouse were pretreated with 0.5 g/ml IgE (C38-2) overnight at 37°C. After wash- ␤ ␤ integrin 1 (Ha2/5), anti-mouse integrin 2 (GAME-46), and anti-mouse ing, cell adhesion was quantitated using CellTiter-Glo (Promega) and a ␣ ␤ integrin V 3 (2C9.G2) mAbs, and others were from BD Biosciences. Micro Lumat Plus luminometer (EG&G Berthold), according to the man- Cytokines such as mouse IL-3 and SCF were obtained from R&D Systems. ufacturer’s instructions. Thrombin, bovine serum VN, and DNP-BSA were from Sigma-Aldrich, In assays using blocking Abs, BMMC were preincubated with 10–50 and human plasma fibrinogen and vWF were from Chemicon International. ␮g/ml Abs for 1 h before adding the cells to the plate. Human fibrinogen labeled with Alexa Fluor 488 was purchased from Mo- lecular Probes. Migration assay Cells Migration assays were conducted, as previously described (13), using 24- well Transwell chambers with 5-␮m polycarbonate filters (Corning). To generate BMMC with 90% purity (c-kitϩ/Fc⑀RIϩ by flow cytometry), Briefly, the underside of each insert was treated with 30 ␮g/ml FN or 50 bone marrow cells from 6-wk-old male BALB/c or C57BL/6 or CBA mice ␮g/ml fibrinogen for1hat37°C, blocked with 3% BSA/PBS for1hat (Charles River Laboratories) were cultured for 4–8 wk in RPMI 1640/10% 37°C, then placed back in the Transwell lower chamber containing 0.5 ml FCS supplemented with 2-ME and 10 ng/ml IL-3 (BMMC culture of the assay medium, with or without stimulants such as SCF. The washed medium). cells resuspended at 7.5 ϫ 107 cells/ml in 0.2 ml of the assay medium were PMC were generated from peritoneal cells, as described previously (37, transferred to the insert. After incubating them for8hat37°C, the cells 38). In brief, mononuclear cells collected from peritoneal lavage were cul- migrated to lower chambers were counted using a hematocytometer. tured for 10–14 days in the presence of 10 ng/ml IL-3 and 30 ng/ml SCF, and PMC with 90% purity were obtained. Cytokine assay (ELISA) Skin-derived mast cells were generated from day 16 fetal skin of mice The cells were transferred into FN- or fibrinogen-coated 96-well plates as described (39, 40). Briefly, excised trunk skin was treated with 0.25% of (1 ϫ 104 cells/well) with or without stimulants. After incubating for 12 h trypsin in HBSS for 30 min at 37°C, and the dispersed cells were cultured at 37°C, the supernatant of each well was collected, and the concentration in the medium containing 10% FCS, 10 ng/ml IL-3, and 30 ng/ml SCF for of IL-6 was quantified by ELISA with OptiEIA for IL-6 (BD Pharmingen). 14 days. Lung-derived mast cells were generated from lung tissue samples of Growth assay adult mice as described (41). Briefly, the samples were cut into fragments 5 and digested with collagenase and hyaluronidase. The single cell suspen- The cells were resuspended at 3 ϫ 10 cells/ml in a BMMC culture me- sion were prepared and cultured in the medium containing 10% of FCS, 10 dium with IL-3, and transferred into fibrinogen-coated 24-well plates with ng/ml IL-3, and 30 ng/ml SCF for 14 days. or without SCF. After incubation for the indicated time period at 37°C, the Human cord blood-derived mast cells were generated from CD34ϩ cord cells were collected and counted using a hematocytometer. blood cells, as described (42). Mast cells with 90% purity were generated by 8 wk of culture in the medium containing 10% FCS, 100 ng/ml human Induced uptake of fibrinogen into SCF-stimulated mast cells SCF, and 50 ng/ml human IL-6. Megakaryocytes were generated, as de- BMMC were suspended at 5 ϫ 105 cells/ml in a Tyrode’s buffer with 10 scribed previously (43). Animal and human studies were approved by the ␮ ␣ ␤ g/ml of a control Ab or the anti-integrin IIb 3 Ab (1B5). Cell were then animal care committee and the ethical committee of the Institute of Med- incubated for 30 min at room temperature, and incubated with 20 ␮g/ml ical Science, University of Tokyo (Tokyo, Japan). fibrinogen labeled with Alexa Fluor 488 in the presence of 100 ng/ml SCF Bone marrow-derived basophils were prepared as described (44). In or PBS for 30 min at room temperature. After washing, the cells were brief, bone marrow cells were cultured for 10 days in the presence of 10 ϩ Ϫ analyzed using FACS or a confocal microscope (Olympus Tokyo). For ng/ml IL-3. Approximately 30% of the cells were Fc⑀RI /c-kit popula- confocal microscopy, the cells attached to microscope slides were fixed tion, which were mainly composed of basophils. with 4% paraformaldehyde, stained with 4Ј,6Ј-diamidino-2-phenylindole, FACS analysis and then visualized. Cells (3 ϫ 105 cells) were suspended in 2% FCS/PBS, blocked with mouse Results FcBlock (BD Biosciences), washed with 2% FCS/PBS, incubated with 20 ␣ ␤ BMMC and PMC express integrin IIb 3 ␮g/ml primary Abs for 20 min at 4°C, washed twice, incubated with 20 ␮g/ml secondary Abs for 20 min, washed twice again, then resuspended in We first tested the expression of integrins on two different types of 2% FCS/PBS. The samples were then analyzed using a FACSCalibur flow mast cells, BMMC and PMC (Fig. 1, A and B), both of which were cytometer (BD Biosciences). Ͼ90% Fc⑀RIϩ/c-kitϩ using flow cytometric analysis. We con- ␣ ␣ RT-PCR firmed the expression of integrin 4 (11, 13, 30, 46, 47), 5 (5, 11, ␣ ␣ ␤ ␤ ␤ 26, 28, 30), V (12), L (48, 49), 1 (11, 26), 2 (50), and 3 (12), Total RNA was prepared from freshly isolated cells with TRIzol (Invitro- ␣ gen Life Technologies). Total RNA from each sample was reverse tran- as reported (Fig. 1A). Interestingly, the expression of integrin IIb scribed with SuperScript Reverse Transcriptase kit (Qiagen) and oligo(dT) was evident on both BMMC and PMC in FACS and RT-PCR (Fig. ␣ primer. The primer used for integrin IIb in this study was as described 1, A and B), the expression level being higher in the former. The previously (45) expression was not due to megakaryocyte progenitors contami- Immunoprecipitation assay nated in our BMMC preparation because we did not detect a megakaryocyte-specific marker, GPIb-a, GPIb-b, or GPV, in our Ten million cells of BMMC and BW1457 were surface biotinylated, collected, and lysed in a radioimmunoprecipitation assay buffer. The lysates BMMCs (Fig. 1A and data not shown). We also confirmed the were immunoprecipitated with a control Ab (hamster IgG or rat IgG) or expression on other types of mast cells, such as skin- and lung- ␣ ␣ anti-mouse integrin IIb Abs (1B5 or MWReg30) or an anti-mouse integrin derived mast cells (Fig. 1C). As integrin IIb heterodimerized ␣ ␤ 54 INTEGRIN IIb 3 EXPRESSION ON MAST CELLS

␣ ␤ Ͼ ⑀ ϩ ϩ FIGURE 1. Surface expression of integrin IIb 3 was detected on mouse mast cells. A, BMMC with 90% purity (Fc RI /c-kit ) expressed integrin ␣ ␣ ␣ ␣ ␣ ␤ ␤ ␤ ␣ ␤ IIb as well as integrins V, 4, 5, L, 1, 2, 3 on BMMC, but not platelet-specific markers GPIb- or GPIb- . The x-axis indicates fluorescence intensity 0 4 ⑀ ϩ ϩ ␣ as a log scale ranging from 10 to 10 . The y-axis indicates the number of the cells. B, PMC with 90% purity (Fc RI /c-kit ) also expressed integrin IIb ␤ ␣ and 3. RT-PCR confirmed the expression of integrin IIb in BMMC, PMC, and P815, a mast cell line, but not in EL-4, a T cell line. C, Skin- and ␣ ␣ ␤ ␣ ␤ lung-derived mast cells expressed integrin IIb. D, Both integrin IIb 3 and integrin V 3 were expressed on the surface of BMMC, whereas only integrin ␣ ␤ V 3 was expressed on BW5147. The cell lysates of surface biotinylated BMMC and BW5147 were immunoprecipitated with control Ab (lanes 1, 3, 5, ␣ ␤ ␣ ␤ and 7) or the Abs against integrin V 3 (lanes 2 and 6)or IIb 3 (lanes 4 and 8), then analyzed with HRP-conjugated streptavidin. E, The expression of ␣ integrin IIb on bone marrow-derived cells after different times of culture in the presence of IL-3 was also analyzed using flow cytometry. Dot plot shows ⑀ ϩ ϩ ␣ ϩ ϩ ␣ Fc RI /c-kit and IIb /c-kit cells on day 28. F, The expression of integrin IIb was observed on bone marrow-derived basophils. G, The expression of ␣ ϩ ϩ ϩ ϩ integrin IIb on B220 splenocytes, CD3 splenocytes, Gr-1 bone marrow cells, and Mac-1 bone marrow cells was not detected in FACS analysis.

␤ ␣ ␤ solely with integrin 3, integrin IIb 3 was expected to be ex- Adhesion of mast cells to immobilized fibrinogen- and ␣ ␤ ␣ ␤ pressed on mouse mast cells. To verify integrin IIb 3 expression, vWF-mediated by integrin IIb 3 BMMC and BW5147 expressing only integrin ␣ ␤ were surface- ␣ ␤ V 3 As integrin IIb 3 works as a receptor for ECM proteins, such as labeled by biotin and immunoprecipitated with a specific Ab fibrinogen, vWF, VN, and FN in megakaryocytes and platelets, we ␣ ␤ ␣ ␤ against integrin IIb 3 (1B5). Western blots developed with HRP- examined whether mast cells expressing integrin IIb 3 could ad- conjugated streptavidin gave two bands corresponding to integrin here to these ECM proteins. ␣ ␤ IIb and 3 only in the precipitates from BMMC but not from Although nonstimulated BMMC did not significantly bind to ␣ ␤ BW5147, suggesting that mouse mast cells express integrin IIb 3 any ECM proteins, BMMC stimulated by monomeric HC IgE on the cell surface (Fig. 1D). (SPE-7) adhered to all of the ECM proteins (Fig. 2, A and B). ␣ Integrin ␣ ␤ on platelets could interact with all these ECM pro- Next, the expression levels of integrin IIb were examined on IIb 3 BMMC in the course of culture with IL-3, that is, in the varying teins, but several studies revealed that the adhesion of mast cells to ϩ ␣ ␤ ␣ FN and VN was mainly mediated by integrin 5 1 (5, 11, 28–31) developmental stages toward BMMC. The percentage of IIb /c- ϩ and ␣ ␤ (12), respectively. kit cells gradually increased during the culture in parallel with V 3 ϩ ϩ To clarify the involvement of each integrin in mast cell adhesion increase in the percentage of Fc⑀RI /c-kit cells indicative of to ECM proteins, we used specific blocking Abs against integrin mast cells or mast cell progenitors, and reached ϳ90% after 28 ␣ ␤ (anti-integrin ␣ ␤ Ab), 1B5 and ␣ ␤3 (anti-integrin days (Fig. 1E). IIb 3 IIb 3 V ␣ ␤ Ab), 2C9.G2. A hamster mAb, 1B5, raised against mouse Interestingly, Fc⑀RIϩ/c-kitϪ cells, which were mainly com- V 3 integrin ␣ ␤ , specifically recognizes mouse integrin ␣ ␤ and ␣ IIb 3 IIb 3 posed of bone marrow-derived basophils, expressed integrin IIb at completely blocks platelet-fibrinogen interaction, thereby inhibit- the levels comparable to mast cells (Fig. 1F). By contrast, no or ing platelet aggregation (34). On the contrary, 2C9.G2, a hamster ␣ negligible expression of integrin IIb was observed on T cells ␣ ␤ ␤ ϩ ϩ mAb against mouse integrin V 3, reacts with integrin 3 in FACS (CD3 splenocytes), B cells (B220 splenocytes), granulocytes analysis and blocks the adhesion via integrin ␣ ␤ (11, 51). ϩ ϩ V 3 (Gr-1 bone marrow cells), macrophages (Mac-1 bone marrow BMMC were pretreated with specific blocking Abs, anti- cells, bone marrow-derived macrophages, and peritoneal macro- ␣ ␤ ␣ ␤ ␤ integrin IIb 3 Ab, anti-integrin V 3 Ab, or anti-integrin 1 Ab, phages), and dendritic cells (bone marrow-derived dendritic cells) Ha2/5, before the adhesion assay. As shown in Fig. 2, A and B, ␣ ␤ (Fig. 1G and data not shown). These results suggest that the ex- anti-integrin IIb 3 Ab inhibited 90% of the adhesion of mast ␣ pression of integrin IIb on cells other than megakaryocytes/plate- cells to fibrinogen and 70% of adhesion to vWF, whereas it only lets was limited to mast cells. weakly inhibited of the adhesion to VN. On the contrary, the The Journal of Immunology 55

␣ ␤ FIGURE 2. Integrin IIb 3 on stimulated BMMC mediated their adhesion to fibrinogen and vWF. A and B, Adhesiveness of BMMC stimulated with or without 3 ␮g/ml HC IgE (SPE-7) for1hat37°C was analyzed on plates coated with or without 20–50 ␮g/ml ECM proteins (fibrinogen, FN, VN, vWF) by calculating the percentage of the adherent cells (A) or by microscope (magnification, ϫ200 or ϫ400) (B). The results of the functional blockage of ␮ ␤ ␮ ␣ ␤ ␮ ␣ ␤ integrins with specific anti-integrin Abs (50 g/ml anti- 1 Ab, 50 g/ml anti-integrin V 3 Ab, 10 g/ml anti-integrin IIb 3 Ab) are also shown. C, The ␣ ␤ ␮ interaction between integrin IIb 3 and fibrinogen was observed in response to stimuli, such as 3 g/ml IgE (SPE-7), 10 ng/ml DNP-BSA (IgE plus Ag), ␮ ␣ ␤ ␮ 100 ng/ml SCF, 1 mM MnCl2, and 10 U/ml thrombin, and was blocked by 10 g/ml anti-integrin IIb 3 Ab. D, Interaction blocked by 100 g/ml RGD p Ͻ 1% (vs the control Ab-treated sample ,ءء .peptides but not by RGE peptides. The results shown are the average Ϯ SD of three independent experiments determined by Student’s t test).

␣ ␤ ␣ ␤ anti-integrin V 3 Ab profoundly inhibited the adhesion of As was the case of mast cell adhesion via integrin 5 1 (5, 11, 29), ␣ ␣ mast cells to VN, whereas it inhibited only 20–30% of the mast there was no change in the cell surface levels of integrins IIb, V, ␤ cell adhesion to vWF, and did not significantly inhibit that to and 3 following treatment with these reagents (data not shown). ␤ fibrinogen. The anti-integrin 1 Ab only partially inhibited mast Collectively, these results indicate that adhesion of mast cells to cell adhesion to VN and vWF, and did not significantly affect fibrinogen requires inside-out signaling followed by increase in the ␤ ␣ ␤ that to fibrinogen, indicating that the involvement of integrin 1 affinity/avidity of integrin IIb 3. in mast cell adhesion to these ECM proteins was limited. How- ever, binding to FN was inhibited by anti-integrin ␤ Ab but not ␣ ␤ 1 SCF induced integrin IIb 3-directed migration of mast cells by anti-integrin ␣ ␤ Ab or anti-integrin ␣ ␤ Ab, as reported IIb 3 V 3 Accumulating evidence revealed that several kinds of integrins (4, 5, 11, 28–31). Similar results were obtained in experiments were involved in the migration of mast cells (13, 37, 47, 50), and using PMC (data not shown). Collectively, these results indi- that the interaction between SCF and c-kit regulated the migration cate that the interaction of mast cell with fibrinogen was spe- of mast cells via ECM proteins, especially FN (13, 14, 37, 47, 50, cifically mediated by integrin ␣ ␤ and that the interaction of IIb 3 52). We examined SCF-induced migration of mast cells via fibrin- mast cells with VN was mainly mediated by integrin ␣ ␤ ,as V 3 ogen, and then tested whether integrin ␣ ␤ -fibrinogen interac- reported (12), but that the efficient adhesion to vWF required IIb 3 tion would induce migration. both integrin ␣ ␤ and ␣ ␤ . V 3 IIb 3 As shown in Fig. 3A, the number of migrated BMMC were Characterization of mast cell adhesion to fibrinogen 5-fold higher via the fibrinogen-treated Transwell membrane surface than the BSA-treated surface in the presence of SCF in the Next, we asked whether other stimuli could induce the same lower wells, and this enhancement was significantly blocked by adhesion, and found that BMMC bound to fibrinogen-coated anti-integrin ␣ ␤ Ab, indicating that SCF enhances migration plates upon stimulation with IgE plus Ag, SCF, thrombin, or IIb 3 via fibrinogen in an integrin ␣ ␤ -dependent manner. Mn2ϩ (Fig. 2C) IIb 3 ␣ ␤ The interaction of integrin IIb 3 on mast cells to fibrinogen was RGD ((Arg-Gly-Asp)-dependent (Fig. 2D) as noted in platelets, Effects of mast cell-fibrinogen interaction on mast cell functions and the kinetics of the adhesion induced by HC IgE (SPE-7) and Adhesion of mast cells to FN was reported to enhance mast cell SCF to fibrinogen showed that the adhesion reached a plateau level functions, such as production of IL-6 or TNF-␣ (28), proliferation from 30 to 60 min after the simulation at 37°C (data not shown). (12), survival (28), and histamine release (5). We tested whether ␣ ␤ 56 INTEGRIN IIb 3 EXPRESSION ON MAST CELLS

␣ ␤ FIGURE 3. Integrin IIb 3 enhanced mast cell function. A, BMMC in upper wells were attracted by 100 ng/ml SCF in lower wells, much more efficiently through fibrinogen or FN-treated Transwell than a BSA- treated surface in an 8-h assay. This enhancement was ␮ ␣ ␤ blocked by 10 g/ml anti-integrin IIb 3. B, BMMC stimulated with 100 ng/ml SCF for 12 h produced IL-6 on FN- or fibrinogen-treated plates. The concentration of IL-6 in culture supernatants was quantified by ELISA. IL-6 production was enhanced on fibrinogen- coated plates, and the enhancement was significantly ␮ ␣ ␤ blocked by 10 g/ml anti-integrin IIb 3 Ab. C, The number of BMMC after 7 days of culture in the presence of 10 ng/ml IL-3 and 100 ng/ml SCF was augmented on fibrinogen-coated plates and significantly blocked with ␮ ␣ ␤ 10 g/ml anti-integrin IIb 3 Abs. The results shown are the average Ϯ SD of three independent experiments. p Ͻ 1% (vs the cells incubated in ,ءء p Ͻ 5% and ,ء BSA-coated plates or the control Ab-treated cells, respectively, as determined by Student’s t test).

mast cell adhesion to fibrinogen had any effects on mast cell func- Second, growth and survival of mast cells was examined. The tions: IL-6 production, growth, and survival and histamine release. number of BMMC on fibrinogen-coated plates was 45% higher First, IL-6 production was measured using ELISA. BMMC than that on BSA-coated plates after 1 wk of culture, and this ␣ ␤ attached to fibrinogen-coated plates produced 50% larger concen- increment was also dependent on integrin IIb 3 (Fig. 3C). How- trations of IL-6 than those on BSA-coated plates in response to ever, we could not find the effects of the binding to fibrinogen on SCF or thrombin, and this enhancement of IL-6 production was the differentiation of BMMC or bone marrow cells cultured with ␣ ␤ clearly blocked by anti-integrin IIb 3 Ab, indicating the interac- IL-3 plus SCF (data not shown). ␣ ␤ tion between integrin IIb 3 and fibrinogen enhances IL-6 produc- SCF-induced survival was estimated by measuring the percent- tion of mast cells (Fig. 3B). In contrast, histamine release from age of apoptotic cells, induced by the withdrawal of IL-3. How- BMMC was not enhanced by the interaction with fibrinogen via ever, no significant difference was found between BMMC on fi- the integrin (data not shown). brinogen-coated and BSA-coated plates in the presence of SCF The Journal of Immunology 57

(data not shown). Thus, the adhesion to fibrinogen enhanced the proliferation mast cells in the presence of SCF but did not prevent the apoptosis of mast cells in the absence of cytokines. ␣ ␤ Endocytosis of fibrinogen into mast cells via integrin IIb 3 ␣ ␤ We analyzed the interaction of integrin IIb 3 to soluble fibrinogen labeled with Alexa Fluor 488 using FACS. BMMC bound soluble ␣ ␤ fibrinogen in an integrin IIb 3-dependent manner (Fig. 4A) when they were activated by HC IgE, IgE plus Ag, SCF, thrombin, or Mn2ϩ. It was previously shown that activated platelets bound to soluble fibrinogen and internalized it (24). To examine whether fibrinogen internalization occurs in mast cells as well, we analyzed the localization of soluble fibrinogen labeled with Alexa Fluor 488 in activated BMMC using a confocal microscope. Fibrinogen labeled with Alexa Fluor 488 was bound and internalized into the mast cells (Fig. 4B). This phenomenon was blocked by anti- ␣ ␤ integrin IIb 3 Ab, indicating that the activated mast cells can ␣ ␤ internalize fibrinogen via integrin IIb 3. ␣ ␤ Human mast cells expressed integrin IIb 3 and mediated adhesion to fibrinogen ␣ ␤ To confirm whether integrin IIb 3 is also expressed and is functional on human mast cells, human cord blood-derived mast cells (Fig. 5A) were generated as reported (42). Toluidine blue and tryptase staining confirmed that the purity of the mast cells ex-

FIGURE 5. Human cord blood-derived mast cells express integrin ␣ ␤ ␣ ␤ IIb 3 and adhered to ﬁbrinogen in an integrin IIb 3-dependent manner. A, Typical mast cells developed in the presence of SCF and IL-6 were stained with May-Gru¨nwald Giemsa. B, Surface expression of integrin ␣ ␤ IIb 3 on human cord blood-derived mast cells was analyzed using ﬂow cytometry. C, Adhesion of human cord blood-derived mast cells was in- ␮ duced by 100 ng/ml human SCF or 1 mM MnCl2, and blocked by 60 g/ml ␣ ␤ anti-integrin IIb 3 Ab (2G12).

ceeded 90%. FACS analysis showed that these cells expressed ␣ ␤ integrin IIb 3 (Fig. 5B). In addition, SCF- or MnCl2-stimulated human cord blood-derived mast cells adhered to fibrinogen in an ␣ ␤ integrin IIb 3-dependent manner, as observed in the case of mouse BMMC and PMC (Fig. 5C). An experiment using the blocking Ab (2G12) confirmed that SCF-induced migration of the human mast cells was also enhanced via fibrinogen through inte- ␣ ␤ grin IIb 3 (data not shown)

Discussion ␣ ␤ Expression of integrin IIb 3 on mast cells ␣ Expression of integrin IIb was considered to be restricted to platelets/megakaryocytes. As recently reported, it turned out to be a marker for early hemapoietic progenitors as well (12–16) ␣ ␤ In the present study, we demonstrate that integrin IIb 3 is expressed on several types of mouse mast cells, BMMC, PMC, skin- and lung-derived mast cells, and human cord blood-derived mast cells. There are two different types of mast cells, consisting of connective tissue and mucosal mast cells (1, 2). PMC and skin- derived mast cells represent connective tissue mast cells. BMMCs have characteristics similar to mucosal and connective tissue mast FIGURE 4. Uptake of soluble fibrinogen in SCF-stimulated BMMC in cells, depending on which factor their culture medium includes ␣ ␤ an integrin IIb 3-dependent manner. A and B, BMMC were incubated ␮ (IL-3 and IL-3 plus SCF, respectively). No remarkable change was with 20 g/ml soluble fibrinogen labeled with Alexa Fluor 488 (FB-Alexa) ␣ ␤ in the presence of 100 ng/ml SCF. A, In flow cytometry, SCF-stimulated observed in the expression level of integrin IIb 3 on BMMC un- ␣ ␤ BMMC bound soluble fibrinogen labeled with Alexa Fluor 488. B,Inthe der both culture conditions (data not shown). Thus, integrin IIb 3 analysis using a confocal laser microscope, the fibrinogen labeled with is thought to be expressed on both types of matured mast cells. We ␣ Alexa Fluor 488 was incorporated in the cells. This phenomenon was observed a low level expression of integrin IIb only on a small ␮ ␣ ␤ blocked by 10 g/ml anti-integrin IIb 3 Ab. population of bone marrow cells before the culture (data not ␣ ␤ 58 INTEGRIN IIb 3 EXPRESSION ON MAST CELLS

␣ ϩ ϩ ␣ ␤ shown), but the percentage of integrin IIb /c-kit cells and the Effects of attachment via integrin IIb 3 on mast cell functions expression level of integrin ␣ gradually increased in culture IIb Attachment of mast cells to ECM proteins has been reported to along with the increase of Fc⑀RIϩ/c-kitϩ cells, mast cell-lineage enhance various mast cell functions. We demonstrate the evidence cells (44). that the interaction between integrin ␣ ␤ and fibrinogen regu- Next, we examined the expression of integrin ␣ on basophils. IIb 3 IIb lates mast cell functions in vivo. Recent studies have revealed that However, it was difficult to confirm the expression on freshly iso- SCF and adhesion molecules like integrins are involved in mast lated basophils from peripheral blood, because of the binding of platelet-derived microparticles, which transfer several kinds of cell-associated diseases. In addition, the role of fibrinogen and its ␣ degradation product, fibrin, in inflammatory reactions has been platelet Ags like integrin IIb CD62 to other circulating blood cells (53). In contrast, BMMC Fc⑀RIϩ/c-kitϪ, mainly composed of ba- given focus. Importantly, SCF is known to critically define the sophils and basophil precursors (44, 54), express comparable lev- accumulation of mast cells at the site of inflammation (1, 2, 37, 47, els of integrin ␣ .AsFc⑀RIϩ/c-kitϪ cells were generated after 10 60) including atherosclerotic plaques (9). Gurish et al. (37) showed IIb ␣ ␤ days culture, the amount of platelet-derived microparticles binding that integrin 4 7 is responsible for tissue-specific homing of mast to these cells was supposed to be negligible, indicating that bone cell progenitors to the small intestine during a helminth infection. ␣ marrow-derived basophils express integrin IIb. However it re- Extravascular fibrinogen and fibrin, which also interact with inte- ␣ ␤ mains to be elucidated whether basophils circulating in peripheral grin IIb 3, are abundant at the site of inflammation, such as ar- ␣ blood express functional integrin IIb. thritis (61), transplant rejection (62), bacterial infection (63–65), ␣ By contrast, no or negligible expression of integrin IIb was and atherosclerosis (66), where immune cells are recruited and ␣ ␤ observed on splenic T cells, splenic B cells, granulocytes, macro- activated (66). Collectively, the interaction of integrin IIb 3 with phages, and dendritic cells. These results show that high expression of fibrinogen may be involved in mast cell-associated pathological ␣ integrin IIb on mature cells is restricted to mast cells and bone conditions, especially where SCF is highly produced. marrow-derived basophils, except for platelets/megakaryocytes. Another interesting aspect is that mast cells are known to ex- Although the precise mechanisms of the regulation of integrin press a profibrinolytic phenotype and contain fibrinolytic enzymes ␣ IIb expression remain unknown, various transcription factors and like tissue plasminogen activator and heparin in their granules their cofactors involved in megakaryocyte development, such as (67). Moreover, mast cells are a major source of these enzymes GATA-1, GATA-2, FOG-1, SCL, NF-E2, AML-1, Fli-1, Gfi-1b, (67). In accordance with these observations, mast cell deficiency MafB, Ets-1, and Ets-2 were implicated in the regulation of inte- leads to experimentally induced-thrombus formation and enhances grin ␣ gene (55, 56). The expression of all these molecules was IIb thrombosis-associated mortality (67). It is tempting to assume that detected by RT-PCR both in megakaryocytes and mast cells, but the adhesion of mast cell to fibrinogen augments such profibrino- not in mature cells of other lineages (data not shown). This result ␣ lytic phenotype by enhancing the effects of these fibrinolytic en- may illustrate, in part, the reason why integrin IIb is expressed on mast cells, which is in accordance with a previous report that mast zymes to lyse fibrinogen. The amount of fibrinogen was thought to cell progenitors and erythroid/megakaryocyte progenitors are be regulated by certain members of integrins using different mech- closely related (57). Further investigation is underway to under- anisms; human monocytes internalized and degraded fibrinogen, stand the regulatory mechanism of integrin expression. independently of plasmin activity via interaction with integrin ␣ ␤ M 2 (Mac-1) to clear fibrinogen and fibrin at injured or inflammatory sites (68). Activated platelets internalized fibrinogen via ␣ ␤ Adhesion of mast cells to ECM proteins via integrin IIb 3 ␣ ␤ integrin IIb 3 to modulate the coagulation process (22–25). In- Various integrins are implicated in mast cell adhesion. Our evi- terestingly, as we have presented in this paper, activated BMMC ␣ ␤ dence shows that integrin IIb 3 on mouse and human mast cells can also bind and internalize soluble fibrinogen in an integrin ␣ ␤ was functionally potent and mediated mast cell adhesion to fibrin- IIb 3-dependent manner (Fig. 4B). These results suggest that up- ␣ ␤ ogen and vWF. This finding was observed only when these mast take of fibrinogen and fibrin via integrin IIb 3 leads to the clear- cells were activated by some stimuli, including IgE plus Ag, HC ance of fibrinogen and fibrin in inflammatory sites. IgE, and SCF. In contrast, the blockade of integrin ␣ ␤ by a ␣ ␤ IIb 3 In conclusion, our novel findings show that integrin IIb 3 is specific Ab did not alter the adhesion of mast cells to VN and FN. expressed on mouse and human mast cells, and mediated adhesion This result is consistent with the previous result that the adhesion to fibrinogen and vWF, resulting in the enhancement of mast cell ␣ ␤ of mast cells to VN and FN is mainly mediated by integrins V 3 functions in concert with SCF. A drug that regulates the function ␤ and 1, respectively (5, 11, 28–31, 56). of integrin ␣ ␤ (33) may control the accumulation and activation Suehiro et al. (58) reported that purified integrin ␣ ␤ molecule IIb 3 V 3 of mast cells, leading to new therapeutic approaches forthcoming had a higher affinity to VN than did integrin ␣ ␤ . Kieffer et al. IIb 3 for mast cell-mediated diseases. (59) showed that integrin ␣ ␤ expressed on a melanoma cell line IIb 3 During our research, Berlanga et al. (69) reported the expression and HEL selectively bound to fibrinogen, not to VN, whereas in- of integrin ␣ ␤ on BMMC. Unlike that report, we detected integrin ␣ ␤ on these cell lines mediated the adhesion to VN. Col- IIb 3 V 3 tegrin ␣ ␤ both on BMMC and PMC. It might result from the lectively, the integrin that is used in the interaction to VN, fibrin- IIb 3 ogen, and vWF could be determined by the different affinity and difference in the affinity of the Abs against the integrin or in the expression levels of integrins ␣ ␤ and ␣ ␤ . mouse strains used. In fact, the expression levels of several inte- IIb 3 V 3 ␣ ␣ grins including integrin IIb on both types of mast cells were dif- Interestingly, bone marrow cells of integrin IIb knockout mice are reported to show decreased adhesive capacity to FN caused by ferent between BALB/c and C57BL/6 mice (T. Oki, J. Kitaura, Y. ␣ ␤ ␣ ␤ Yamanishi, and T. Kitamura, unpublished observation). They sug- impaired adhesive function of integrin 4 1 and 5 1, indicating ␣ gested the in vivo function of integrin ␣ ␤ expressed on BMMC cross-modulation of integrin IIb (15). This mode of modulation IIb 3 ␣ by integrin IIb was not detected in our study using specific block- by showing that the adhesion to VN was increased in BMMC ␣ ␣ ing Abs, implying that the loss of the expression of integrin IIb derived from the integrin IIb knockout mice. However this alter- ␣ ␤ and functional blockage of integrin IIb 3 might lead to distinct ation was probably due to the compensatory enhanced expression ␣ ␣ results. of integrin V on BMMC of the integrin IIb knockout mice, The Journal of Immunology 59 which we observed and later described; thus the function of inte- 17. Tronik-Le Roux, D., V. Roullot, C. Poujol, T. Kortulewski, P. Nurden, and ␣ ␤ G. Marguerie. 2000. Thrombasthenic mice generated by replacement of the in- grin IIb 3 was not directly addressed. In contrast, we character- ␣ ␣ ␤ tegrin IIb gene: demonstration that transcriptional activation of this megakaryo- ized functions of integrin IIb 3 on BMMC using a neutralizing cytic locus precedes lineage commitment. Blood 96: 1399–1408. Ab in multiple assays. In addition, Berlanga et al. (69) studied the 18. Tropel, P., V. Roullot, M. Vernet, C. Poujol, H. Pointu, P. Nurden, G. Marguerie, and D. Tronik-Le Roux. 1997. A 2.7-kb portion of the 5Ј flanking region of the mast cell functions in the absence of stimulation, whereas we in- ␣ murine glycoprotein IIb gene is transcriptionally active in primitive hematopoi- vestigated those in the presence and absence of the stimulation etic progenitor cells. Blood 90: 2995–3004. because the inside-out signaling is required for maximal adhesion 19. Ody, C., P. Vaigot, P. Que´re´, B. A. Imhof, and C. Corbel. 1999. Glycoprotein ␣ IIb-IIIa is expressed on avian multilineage hematopoietic progenitor cells. Blood of mast cells. Finally, using BMMC from integrin IIb knockout 93: 2898–2906. mice, a gift from Dr. J. Frampton (Institute of Biomedical Re- 20. Debili, N., C. Robin, V. Schiavon, R. Letestu, F. Pflumio, M. T. Mitjavila-Garcia, search, Birmingham University, Edgbaston, Birmingham, U.K.), L. Coulombel, and W. Vainchenker. 2001. Different expression of CD41 on human lymphoid and myeloid progenitors from adults and neonates. Blood 97: we confirmed the following phenomena: the adhesion of these 2023–2030. BMMC to VN was increased due to the enhanced expression of 21. Mitjavila-Garcia, M. T., M. Cailleret, I. Godin, M. M. Nogueira, K. Cohen-Solal, ␣ ␤ V. Schiavon, Y. Lecluse, F. Le Pesteur, A. H. Lagrue, and W. Vainchenker. 2002. integrin V 3, whereas their adhesion to fibrinogen was signifi- Expression of CD41 on hematopoietic progenitors derived from embryonic he- ␣ cantly diminished because of the integrin IIb deficiency (data not matopoietic cells. Development 129: 2003–2013. shown). These results are in accordance with our results obtained 22. Cramer, E. M., G. F. Savidge, W. Vainchenker, M. C. Berndt, D. Pidard, J. P. Caen, J. M. Masse, and J. Breton-Gorius. 1990. ␣-Granule pool of glyco- from the study using blocking Abs, which indicate that the inter- protein IIb-IIIa in normal and pathologic platelets and megakaryocytes. Blood 75: action of BMMC to fibrinogen and VN was mainly mediated by 1220–1227. integrin ␣ ␤ and ␣ ␤ , respectively. 23. Handagama, P., R. M. Scarborough, M. A. Shuman, and D. F. Bainton. 1993. IIb 3 V 3 Endocytosis of fibrinogen into megakaryocyte and platelet ␣-granules is mediated ␣ ␤ by IIb 3 (glycoprotein IIb-IIIa). Blood 82: 135–138. Acknowledgments 24. Wencel-Drake, J. D., C. Boudignon-Proudhon, M. G. Dieter, A. B. Criss, and We greatly thank Drs. B. S. Coller, V. L. Woods, D. J. Gerber, and L. V. Parise. 1996. Internalization of bound fibrinogen modulates platelet aggre- S. Tonegawa for providing Abs, Dr. J. Frampton for providing the knock- gation. Blood 87: 602–612. 25. Ni, H., J. M. Papalia, J. L. Degen, and D. D. Wagner. 2003. Control of thrombus out mice, and Dr. S. J. Shattil for careful reading of the manuscript and for ␣ ␤ embolization and fibronectin internalization by integrin IIb 3 engagement of the useful discussion. We also thank M. Ohara for language assistance. fibrinogen ␥ chain. Blood 102: 3609–3614. 26. Kinashi, T., T. Asaoka, R. Setoguchi, and K. Takatsu. 1999. Affinity modulation Disclosures of very late antigen-5 through phosphatidylinositol 3-kinase in mast cells. J. Im- munol. 162: 2850–2857. The authors have no financial conflict of interest. 27. Kitaura, J., K. Eto, T. Kinoshita, Y. Kawakami, M. Leitges, C. A. Lowell, and T. Kawakami. 2005. Regulation of highly cytokinergic IgE-induced mast cell References adhesion by Src, Syk, Tec, and protein kinase C family kinases. J. Immunol. 174: 4495–4504. 1. Kawakami, T., and S. J. Galli. 2002. Regulation of mast-cell and basophil func- kit tion and survival by IgE. Nat. Rev. Immunol. 2: 773–786. 28. Kinashi, T., and T. A. Springer. 1994. Steel factor and c- regulate cell-matrix Blood 2. Metcalfe, D. D., D. Baram, and Y. A. Mekori. 1997. Mast cells. Physiol. Rev. 77: adhesion. 83: 1033–1038. 1033–1079. 29. Dastych, J., and D. D. Metcalfe. 1994. Stem cell factor induces mast cell adhesion 3. Gonzalez-Espinosa, C., S. Odom, A. Olivera, J. P. Hobson, M. E. Martinez, to fibronectin. J. Immunol. 152: 213–219. A. Oliveira-Dos-Santos, L. Barra, S. Spiegel, J. M. Penninger, and J. Rivera. 30. Vliagoftis, H. 2002. Thrombin induces mast cell adhesion to fibronectin: evi- 2003. Preferential signaling and induction of allergy-promoting lymphokines dence for involvement of protease-activated receptor-1. J. Immunol. 169: upon weak stimulation of the high affinity IgE receptor on mast cells. J. Exp. Med. 4551–4558. 197: 1453–1465. 31. Rosbottom, A., C. L. Scudamore, H. von der Mark, E. M. Thornton, S. H. Wright, ␤ 4. Kitaura, J., J. Song, M. Tsai, K. Asai, M. Maeda-Yamamoto, A. Mocsai, and H. R. Miller. 2002. TGF- 1 regulates adhesion of mucosal mast cell homo- ␣ Y. Kawakami, F. T. Liu, C. A. Lowell, B. G. Barisas, et al. 2003. Evidence that logues to laminin-1 through expression of integrin 7. J. Immunol. 169: IgE molecules mediate a spectrum of effects on mast cell survival and activation 5689–5695. via aggregation of the Fc⑀RI. Proc. Natl. Acad. Sci. USA 100: 12911–12916. 32. Edelson, B. T., Z. Li, L. K. Pappan, and M. M. Zutter. 2004. Mast cell-mediated ␣ ␤ 5. Lam, V., J. Kalesnikoff, C. W. Lee, V. Hernandez-Hansen, B. S. Wilson, inflammatory responses require the 2 1 integrin. Blood 103: 2214–2220. J. M. Oliver, and G. Krystal. 2003. IgE alone stimulates mast cell adhesion to 33. Topol, E. J., T. V. Byzova, and E. F. Plow. 1999. Platelet GPIIb-IIIa blockers. fibronectin via pathways similar to those used by IgE ϩ antigen but distinct from Lancet 353: 227–231. those used by Steel factor. Blood 102: 1405–1413. 34. Lengweiler, S., S. S. Smyth, M. Jirouskova, L. E. Scudder, H. Park, T. Moran, 6. Marshall, J. S., and D. M. Jawdat. 2004. Mast cells in innate immunity. and B. S. Coller. 1999. Preparation of monoclonal antibodies to murine platelet ␣ ␤ J. Allergy Clin. Immunol. 114: 21–27. glycoprotein IIb/IIIa ( IIb 3) and other proteins from hamster-mouse interspecies 7. Puxeddu, I., A. M. Piliponsky, I. Bachelet, and F. Levi-Schaffer. 2003. Mast cells hybridomas. Biochem. Biophys. Res. Commun. 262: 167–173. in allergy and beyond. Int. J. Biochem. Cell Biol. 35: 1601–1607. 35. Gerber, D. J., P. Pereira, S.Y. Huang, C. Pelletier, and S. Tonegawa. 1996. Ex- ␣ ␤ 8. Tang, L., T. A. Jennings, and J. W. Eaton. 1998. Mast cells mediate acute in- pression of v and 3 integrin chains on murine lymphocytes. Proc. Natl. Acad. flammatory responses to implanted biomaterials. Proc. Natl. Acad. Sci. USA 95: Sci. USA 93: 14698–14703. 8841–8846. 36. Woods, V. L., Jr., E. H. Oh, D. Mason, and R. McMillan. 1984. Autoantibodies 9. Kelley, J. L., D. S. Chi, W. Abou-Auda, J. K. Smith, and G. Krishnaswamy. against the platelet glycoprotein IIb/IIIa complex in patients with chronic ITP. 2000. The molecular role of mast cells in atherosclerotic cardiovascular disease. Blood 63: 368–375. Mol. Med. Today 6: 304–308. 37. Gurish, M. F., H. Tao, J. P. Abonia, A. Arya, D. S. Friend, C. M. Parker, and ␤ ␣ ␤ 10. Lee, D. M., D. S. Friend, M. F. Gurish, C. Benoist, D. Mathis, and M. B. Brenner. K. F. Austen. 2001. Intestinal mast cell progenitors require CD49d 7 ( 4 7 in- 2002. Mast cells: a cellular link between autoantibodies and inflammatory arthri- tegrin) for tissue-specific homing. J. Exp. Med. 194: 1243–1252. tis. Science 297: 1689–1692. 38. Asai, K., J. Kitaura, Y. Kawakami, N. Yamagata, M. Tsai, D. P. Carbone, 11. Yasuda, M., Y. Hasunuma, H. Adachi, C. Sekine, T. Sakanishi, H. Hashimoto, F. T. Liu, S. J. Galli, and T. Kawakami. 2001. Regulation of mast cell survival C. Ra, H. Yagita, and K. Okumura. 1995. Expression and function of fibronectin by IgE. Immunity 14: 791–800. binding integrins on rat mast cells. Int. Immunol. 7: 251–258. 39. Yamada, N., H. Matsushima, Y. Tagaya, S. Shimada, and S. I. Katz. 2003. Gen- 12. Bianchine, P. J., P. R. Burd, and D. D. Metcalfe. 1992. IL-3-dependent mast cells eration of a large number of connective tissue type mast cells by culture of murine attach to plate-bound vitronectin: demonstration of augmented proliferation in fetal skin cells. J. Invest. Dermatol. 121: 1425–1432. response to signals transduced via cell surface vitronectin receptors. J. Immunol. 40. Matsushima, H., N. Yamada, H. Matsue, and S. Shimada. 2004. TLR3-, TLR7-, 149: 3665–3671. and TLR9-mediated production of proinflammatory cytokines and chemokines 13. Tan, B. L., M. N. Yazicioglu, D. Ingram, J. McCarthy, J. Borneo, D. A. Williams, from murine connective tissue type skin-derived mast cells but not from bone ␣ and R. Kapur. 2003. Genetic evidence for convergence of c-Kit- and 4 integrin- marrow-derived mast cells. J. Immunol. 173: 531–541. mediated signals on class IA PI-3kinase and the Rac pathway in regulating in- 41. Oskeritzian, C. A., W. Zhao, A. L. Pozez, N. M. Cohen, M. Grimes, and tegrin-directed migration in mast cells. Blood 101: 4725–4732. L. B. Schwartz. 2004. Neutralizing endogenous IL-6 renders mast cells of the 14. Kitaura, J., T. Kinoshita, M. Matsumoto, S. Chung, Y. Kawakami, M. Leitges, MCT type from lung, but not the MCTC type from skin and lung, susceptible to D. Wu, C. A. Lowell, and T. Kawakami. 2005. IgE- and IgEϩAg-mediated mast human recombinant IL-4-induced apoptosis. J. Immunol. 172: 593–600. cell migration in an autocrine/paracrine fashion. Blood 105: 3222–3229. 42. Saito, H., M. Ebisawa, N. Sakaguchi, T. Onda, Y. Iikura, M. Yanagida, 15. Emambokus, N. R., and J. Frampton. 2003. The glycoprotein IIb molecule is H. Uzumaki, and T. Nakahata. 1995. Characterization of cord-blood-derived hu- expressed on early murine hematopoietic progenitors and regulates their numbers man mast cells cultured in the presence of Steel factor and interleukin-6. Int. in sites of hematopoiesis. Immunity 19: 33–45. Arch. Allergy Immunol. 107: 63–65. 16. Shattil, S. J., and P. J. Newman. 2004. Integrins: dynamic scaffolds for adhesion 43. Eto, K., R. Murphy, S. W. Kerrigan, A. Bertoni, H. Stuhlmann, T. Nakano, and signaling in platelets. Blood 104: 1606–1615. A. D. Leavitt, and S. J. Shattil. 2002. Megakaryocytes derived from embryonic ␣ ␤ 60 INTEGRIN IIb 3 EXPRESSION ON MAST CELLS

stem cells implicate CalDAG-GEFI in integrin signaling. Proc. Natl. Acad. Sci. 56. Sevinsky, J. R., A. M. Whalen, and N. G. Ahn. 2004. Extracellular signal-reg- USA 99: 12819–12824. ulated kinase induces the megakaryocyte GPIIb/CD41 gene through MafB/Kre- 44. Nakanishi, K., T. Yoshimoto, H. Tsutsui, and H. Okamura. 2001. Interleukin-18 isler. Mol. Cell. Biol. 24: 4534–4545. is a unique cytokine that stimulates both Th1 and Th2 responses depending on its 57. Migliaccio, A. R., R. A. Rana, M. Sanchez, R. Lorenzini, L. Centurione, cytokine milieu. Cytokine Growth Factor Rev. 12: 53–72. L. Bianchi, A. M. Vannucchi, G. Migliaccio, and S. H. Orkin. 2003. GATA-1 as 45. Gaines, P., J. N. Geiger, G. Knudsen, D. Seshasayee, and D. M. Wojchowski. a regulator of mast cell differentiation revealed by the phenotype of the GATA- 2000. GATA-1- and FOG-dependent activation of megakaryocytic ␣IIB gene 1low mouse mutant. J. Exp. Med. 197: 281–296. expression. J. Biol. Chem. 275: 34114–34121. 58. Suehiro, K., J. W. Smith, and E. F. Plow. 1996. The ligand recognition specificity ␤ 46. Boyce, J. A., E. A. Mellor, B. Perkins, Y. C. Lim, and F. W. Luscinskas. 2002. of 3 integrins. J. Biol. Chem. 271: 10365–10371. Human mast cell progenitors use ␣ -integrin, VCAM-1, and PSGL-1 E-selectin 59. Kieffer, N., L. A. Fitzgerald, D. Wolf, D. A. Cheresh, and D. R. Phillips. 1991. 4 ␤ for adhesive interactions with human vascular endothelium under flow condi- Adhesive properties of the 3 integrins: comparison of GP IIb-IIIa and the vi- tions. Blood 99: 2890–2896. tronectin receptor individually expressed in human melanoma cells. J. Cell Biol. 47. Pennock, J. L., and R. K. Grencis. 2004. In vivo exit of c-kitϩ/CD49dhigh/␤7ϩ 113: 451–461. mucosal mast cell precursors from the bone marrow following infection with the 60. Fukao, T., T. Yamada, M. Tanabe, Y. Terauchi, T. Ota, T. Takayama, T. Asano, intestinal nematode Trichinella spiralis. Blood 103: 2655–2660. T. Takeuchi, T. Kadowaki, J. Hata Ji, and S. Koyasu. 2002. Selective loss of 48. Yoshikawa, H., Y. Nakajima, and K. Tasaka. 1999. Glucocorticoid suppresses gastrointestinal mast cells and impaired immunity in PI3K-deficient mice. Nat. autocrine survival of mast cells by inhibiting IL-4 production and ICAM-1 ex- Immunol. 3: 295–304. pression. J. Immunol. 162: 6162–6170. 61. Busso, N., V. Peclat, K. Van Ness, E. Kolodziesczyk, J. Degen, T. Bugge, and A. So. 1998. Exacerbation of antigen-induced arthritis in urokinase-deficient 49. Inamura, N., Y. A. Mekori, S. P. Bhattacharyya, P. J. Bianchine, and mice. J. Clin. Invest. 102: 41–50. D. D. Metcalfe. 1998. Induction and enhancement of Fc⑀RI-dependent mast cell 62. Zhang, J., R. Munda, P. Glas-Greenwalt, M. A. Weiss, V. E. Pollak, and degranulation following coculture with activated T cells: dependency on J. W. Alexander. 1983. Prolongation of survival of a heart xenograft by defi- ICAM-1- and leukocyte function-associated antigen (LFA)-1-mediated hetero- brination with ancrod. Transplantation 35: 620–622. typic aggregation. J. Immunol. 160: 4026–4033. 63. Koch, T., H. P. Duncker, R. Axt, H. G. Schiefer, K. Van Ackern, and H. Neuhof. 50. Rosenkranz, A. R., A. Coxon, M. Maurer, M. F. Gurish, K. F. Austen, 1993. Effects of hemorrhage, hypoxia, and intravascular coagulation on bacterial D. S. Friend, S. J. Galli, and T. N. Mayadas. 1998. Impaired mast cell develop- clearance and translocation. Crit. Care Med. 21: 1758–1764. ment and innate immunity in Mac-1 (CD11b/CD18, CR3)-deficient mice. J. Im- 64. Koch, T., C. Annuss, H. G. Schiefer, K. van Ackern, and H. Neuhof. 1997. munol. 161: 6463–6467. Impaired bacterial clearance after activation of the complement and coagulation 51. Castells, M. C., L. B. Klickstein, K. Hassani, J. A. Cumplido, M. E. Lacouture, systems. Shock 7: 42–48. ␣ ␤ K. F. Austen, and H. R. Katz. 2001. gp49B1- v 3 interaction inhibits antigen- 65. Goguen, J. D., T. Bugge, and J. L. Degen. 2000. Role of the pleiotropic effects induced mast cell activation. Nat. Immunol. 2: 436–442. of plasminogen deficiency in infection experiments with plasminogen-deficient 52. Ruoslahti, E. 1988. Fibronectin and its receptors. Annu. Rev. Biochem. 57: mice. Methods 21: 179–183. 375–413. 66. Szaba, F. M., and S. T. Smiley. 2002. Roles for thrombin and fibrin(ogen) in 53. Janowska-Wieczorek, A., M. Majka, J. Kijowski, M. Baj-Krzyworzeka, R. Reca, cytokine/chemokine production and macrophage adhesion in vivo. Blood 99: A. R. Turner, J. Ratajczak, S. G. Emerson, M. A. Kowalska, and M. Z. Ratajczak. 1053–1059. 2001. Platelet-derived microparticles bind to hematopoietic stem/progenitor cells 67. Bankl, H. C., and P. Valent. 2002. Mast cells, thrombosis, and fibrinolysis: the and enhance their engraftment. Blood 98: 3143–3149. emerging concept. Thromb. Res. 105: 359–365. 54. Dvorak, A. M., R. A. Seder, W. E. Paul, E. S. Morgan, and S. J. Galli. 1994. 68. Simon, D. I., A. M. Ezratty, S. A. Francis, H. Rennke, and J. Loscalzo. 1993. Effects of interleukin-3 with or without the c-kit ligand, stem cell factor, on the Fibrin(ogen) is internalized and degraded by activated human monocytoid cells survival and cytoplasmic granule formation of mouse basophils and mast cells in via Mac-1 (CD11b/CD18): a nonplasmin fibrinolytic pathway. Blood 82: vitro. Am. J. Pathol. 144: 160–170. 2414–2422. 55. Elagib, K. E., F. K. Racke, M. Mogass, R. Khetawat, L. L. Delehanty, and 69. Berlanga, O., N. Emambokus, and J. Frampton. 2005. GPIIb (CD41) integrin is A. N. Goldfarb. 2003. RUNX1 and GATA-1 coexpression and cooperation in expressed on mast cells and influences their adhesion properties. Exp. Hematol. megakaryocytic differentiation. Blood 101: 4333–4341. 33: 403–412.