Kinase Rho-Associated Coiled-Coil-Forming Protein Mitogen-Activated Protein Kinase and Agen

Home , Protein kinase

Sensitized Mast Cells Migrate Toward the Agen: A Response Regulated by p38 Mitogen-Activated Protein Kinase and Rho-Associated Coiled-Coil-Forming Protein Kinase1

Tamotsu Ishizuka,2* Fumikazu Okajima,‡ Mitsuteru Ishiwara,‡§ Kunihiko Iizuka,* Isao Ichimonji,* Tadayoshi Kawata,* Hideo Tsukagoshi,* Kunio Dobashi,* Tsugio Nakazawa,† and Masatomo Mori*

Although mast cells accumulate within the mucosal epithelial layer of patients with allergic rhinitis and bronchial asthma, the responsible chemotactic factors are undefined. We investigated whether mast cells sensitized with Ag-specific IgE migrate toward the Ag. MC/9 mast cells sensitized with anti-DNP IgE migrated toward DNP-conjugated human serum albumin. This migration was directional, and the degree was stronger than that induced by stem cell factor. IL-3 and stem cell factor-dependent cultured mast cells derived from mouse bone marrow also migrated toward the Ag. Subsequent migration mediated by the Fc⑀RI was significantly inhibited by incubating the cells with Y-27632, a Rho-associated coiled-coil-forming protein kinase inhibitor, or with SB203580, a p38 mitogen-activated protein kinase (MAPK) inhibitor. Both p38 MAPK and MAPK-activated protein kinase (MAPKAPK)2 were activated following Fc⑀RI aggregation, and activation of MAPKAPK2 was almost completely inhibited by 10␮M SB203580. Wortmannin or a low concentration of SB203580 partially inhibited MAPKAPK2, but did not block mast cell migration. In contrast, Y-27632 did not affect the activation of MAPKAPK2. These results indicate that Ag works not only as a stimulant for allergic mediators from IgE-sensitized mast cells, but also as a chemotactic factor for mast cells. Both p38 MAPK activation and Rho-dependent activation of Rho-associated coiled-coil-forming protein kinase may be required for Fc⑀RI-mediated cell migration. The Journal of Immunology, 2001, 167: 2298–2304.

ast cells play a central role in allergic reactions. The several factors, including stem cell factor (SCF),3 have chemotac- multivalent binding of Ag to receptor-bound IgE and tic activity for mast cells (4, 5). Mast cells are believed to encoun- M the subsequent aggregation of Fc⑀RI triggers the se- ter Ag by chance. However, if mast cells defend the host against cretion of several chemical mediators, such as histamine, and de invaders such as parasites through IgE and Fc⑀RI, mast cell mi- novo synthesized lipid mediators (1). Allergic inflammation is gration toward the Ag seems plausible. Mouse bone marrow-de- characterized by tissue infiltration with a wide variety of inflam- rived cultured mast cells migrate in response to monocyte chemo- matory cells including mast cells, eosinophils, and lymphocytes tactic protein-1 and RANTES primarily on specific matrices such (2). Directional migration of inflammatory cells is presumably as vitronectin- and laminin-coated filters. The migration of bone based on locally produced chemotactic factors. Besides during al- marrow-derived cultured mast cells in response to chemokines is lergy, mast cells accumulate during many pathologic conditions augmented following Fc⑀RI-mediated activation (6, 7). However, including parasitic infections, growth of some solid tumors, and whether the Ag itself induces the directional migration of IgE- chronic inflammatory conditions such as interstitial cystitis and sensitized mast cells has never been reported. rheumatoid arthritis (2, 3). Little is known about the chemotactic Here we show that mast cells sensitized with Ag-specific IgE factors involved in mast cell recruitment in disease states, although migrate toward the Ag. Our results suggest that IgE-sensitized mast cells can enter sites harboring a specific Ag and that the cells release chemical mediators such as histamine and generate lipid * First Department of Internal Medicine, Gunma University School of Medicine, mediators and cytokines. To define the intracellular mechanism of †Faculty of Health Sciences, and ‡Laboratory of Signal Transduction, Institute for Molecular and Cellular Regulation, Gunma University, Maebashi, Japan; and ¤Central mast cell migration, we investigated mitogen-activated protein ki- Research Laboratories, Dainippon Ink and Chemicals, Sakura, Japan nase (MAPK)-activated protein kinase (MAPKAPK)2, which is Received for publication February 14, 2001. Accepted for publication June 13, 2001. phosphorylated and activated by p38 MAPK (8Ð10) and Rho-as- The costs of publication of this article were defrayed in part by the payment of page sociated coiled-coil-forming protein kinase (ROCK) (11). Our re- charges. This article must therefore be hereby marked advertisement in accordance sults demonstrate that IgE-sensitized mast cells migrate toward a with 18 U.S.C. Section 1734 solely to indicate this fact. specific Ag and that this migration is regulated by p38 MAPK 1 This work was partly supported by the Ministry of Education, Science, and Culture and ROCK. of Japan (11670434). 2 Address correspondence and reprint requests to Dr. Tamotsu Ishizuka, First Department of Internal Medicine, Gunma University School of Medicine, 3-39-15 Showa-machi, Maebashi 371-8511, Japan. E-mail address: tamotsui@showa. Materials and Methods gunma-u.ac.jp Cells and reagents 3 Abbreviations used in this paper: SCF, stem cell factor; MAPK, mitogen-activated pro- The MC/9 mouse mast cell clone, obtained from the American Type Cul- tein kinase; MAPKAPK, MAPK-activated protein kinase; IL-3/SCF-BMMC, IL-3- and ture Collection (Manassas, VA), was maintained by passage in DMEM SCF-dependent mouse bone marrow-derived cultured mast cells; DNP-HSA, DNP-con- (Life Technologies, Grand Island, NY) supplemented with 10% FBS (Life jugated human serum albumin; JNK, c-Jun amino-terminal kinase; ERK, extracellular ␮ signal-regulated kinase; MEK, MAPK/ERK kinase; MKK, MAPK kinase; ROCK, Rho- Technologies), 50 M 2-ME (Life Technologies), and 5 ng/ml mouse 2ϩ associated coiled-coil-forming protein kinase; HSP, heat shock protein; [Ca ]i, cytoplas- rIL-3 (provided by KIRIN Brewery, Yokohama, Japan). Mouse bone mar- mic free Ca2ϩ. row obtained from the femurs of female BALB/c mice was cultured in

DMEM supplemented with 10% FBS, 50 ␮M 2-ME, 100 ␮g/ml strepto- tified by counting the number of mast cells in the bottom well using a mycin, 100 U/ml penicillin, 0.5 ␮g/ml amphotericin B, 5 ng/ml IL-3, and Fuchus Rosenthal calculator (Erma, Tokyo, Japan). IL-3/SCF-BMMC 10 ng/ml recombinant mouse SCF (PeproTech, London, U.K). After 4 wk were cultured with 500 ng/ml anti-DNP IgE in complete culture medium of culture, Ͼ95% of nonadherent cells contained granules that stained pos- for 18 h. The cell viability and characteristics of IL-3/SCF-BMMC were itively with toluidine blue. These cells are referred to as IL-3- and SCF- maintained by complete culture medium. Thereafter, cells were washed dependent mouse bone marrow-derived cultured mast cells (IL-3/SCF- three times with 0.1% BSA-DMEM. IL-3/SCF-BMMC suspended at 1 ϫ BMMC). Bovine myelin basic protein, rabbit anti-MAPKAPK2 polyclonal 106 cells/200 ␮l in 0.1% BSA-DMEM were placed in the top wells, and Ab, and MAPKAPK2 substrate peptide (TTYADFIASGRTGR) were ob- DNP-HSA (10 ng/ml) was applied to the bottom wells. Cells that had tained from Upstate Biotechnology (Lake Placid, NY). Polyclonal anti- migrated were counted after 6 h. extracellular signal-regulated kinase (ERK)2 (C-14) agarose conjugate was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti- Measurement of cytoplasmic free Ca2ϩ ([Ca2ϩ] ) phospho-p38 MAPK polyclonal Ab was purchased from Cell Signaling i Technology (Beverly, MA). Recombinant protein G agarose and protein A MC/9 cells were incubated for 20 min with 1 ␮M fura 2 acetoxymethyl 2ϩ agarose were purchased from Zymed Laboratories (San Francisco, CA). ester in 0.1% BSA-DMEM. [Ca ]i was estimated as a change in the flu- Mouse monoclonal anti-DNP IgE was obtained from Seikagaku Kogyo orescence of the fura 2-loaded cells, as previously described (12). (Tokyo, Japan). DNP-conjugated human serum albumin (DNP-HSA), DNP-conjugated lysine (DNP-lysine), and actinomycin D were obtained Kinase assay of MAPKAPK2 from Sigma (St. Louis, MO). Y-27632 was kindly provided by WelFide (Osaka, Japan). SB203580 and wortmannin were purchased from Calbio- MC/9 cells (3 ϫ 106) were lysed in buffer (50 mM Tris-HCl (pH 7.5), 1 ␥ 32 chem (San Diego, CA), and [ - P]ATP was purchased from ICN Bio- mM EDTA, 1 mM EGTA, 1 mM Na3VO4, 0.1% (v/v) 2-ME, 1% Triton medicals (Costa Mesa, CA). X-100, 5 mM sodium pyrophosphate, 10 mM ␤-glycerophosphate, 50 mM NaF, 1 mM PMSF, 10 ␮g/ml aprotinin, and 5 ␮g/ml leupeptin). The lysates Passive sensitization and stimulation of MC/9 cells were incubated with 5 ␮g/ml rabbit anti-MAPKAPK2 Ab for2hat4¡C. Recombinant protein G agarose was added to the lysates and incubated for MC/9 cells cultured with 500 ng/ml anti-DNP IgE in DMEM containing 1hat4¡C. The immunoprecipitates were washed twice with lysis buffer 0.1% BSA (0.1% BSA-DMEM) for 2 h were washed three times with and once with kinase buffer (20 mM MOPS (pH 7.2), 25 mM ␤-glycero- medium and incubated in fresh medium for an additional 2 h. DNP-HSA phosphate, 5 mM EGTA, 1 mM Na VO , and 1 mM DTT). After the final dissolved in PBS was the stimulant, and PBS was the control vehicle. In 3 4 wash, 30 ␮l kinase assay buffer containing 10 ␮Ci [␥-32P]ATP, 150 ␮M some experiments, MC/9 cells were incubated with 0.1% BSA-DMEM for cold ATP, 23 mM MgCl , and 10 ␮g substrate peptide (KKLNRTLSVA) 2 h, and then SCF was added to the medium. 2 was added per sample. The samples were incubated for 30 min at 30¡C, and ␮ Mast cell migration then the reaction was stopped by adding 30 l of excessive phosphoric acid. Samples (25 ␮l) were loaded onto phosphocellulose paper (Whatman, MC/9 cell migration was quantified by a modification of the Boyden cham- Maidstone, U.K.), and phosphorylation of the substrate peptide was deter- ber technique. Briefly, MC/9 cells sensitized with anti-DNP IgE were sus- mined by liquid scintillation counting. pended at 1 ϫ 106 cells/200 ␮l in 0.1% BSA-DMEM and placed in the top well. The bottom well was filled with 600 ␮l of 0.1% BSA-DMEM and Kinase assay of c-Jun amino-terminal kinase (JNK) was separated from the top well by a 3-␮m-pore polycarbonate filter (Che-

motaxicell; Kurabou, Osaka, Japan). The chambers were incubated for 1Ð6 GST-c-Jun1Ð79 fusion protein was prepared, and kinase activity was mea- hat37¡C in a moist 5% CO2 atmosphere. Mast cell movement was quan- sured as previously described (13Ð16).

FIGURE 1. Mast cells migrate toward Ag and SCF. A, MC/9 cells sensitized with anti-DNP IgE (1 ϫ 106 cells) were placed in the upper wells and separated from the lower wells by polycarbonate ﬁlters. Ag (0Ð100 ng/ml DNP-HSA) was added to the lower wells. Unsensitized MC/9 cells were also

applied to the upper wells, and 10 ng/ml DNP-HSA was added to the lower wells. Chambers were incubated for4hat37¡C under moist 5% CO2. Mast cell movement was quantiﬁed by counting numbers of cells that migrated to the lower wells. B, Unsensitized MC/9 cells (1 ϫ 106 cells) were placed in the top wells, and SCF (0Ð100 ng/ml) was applied to the bottom wells. After a 4-h incubation, migrated cells were counted. C, Sensitized or unsensitized MC/9 cells (1 ϫ 106 cells) were placed in the upper wells. Ag (10 ng/ml DNP-HSA) or SCF (10 ng/ml) was added to the lower wells. Chambers were incubated for 1Ð6 h, and migrated cells were counted. D, IL-3/SCF-BMMC sensitized with anti-DNP IgE (1 ϫ 106 cells) were applied to the top wells, and Ag (0Ð100 ng/ml DNP-HSA) was added to the lower wells. Migrated cells were counted after a 6-h incubation. 2300 SENSITIZED MAST CELLS MIGRATE TOWARD THE Ag

Kinase assay of p38 MAPK The activity of p38 kinase was assayed as previously described using ac- tivating transcription factor-2 as the substrate (14Ð16), and anti-phospho- p38 MAPK (1/100 dilution) was used for immunoprecipitation. Kinase assay of ERK2 The kinase activity of ERK2 was assayed in vitro as previously described (13Ð16) using myelin basic protein as the substrate, and anti-ERK2 agarose conjugate was used for immunoprecipitation. Results Sensitized mast cells migrate toward the Ag Sensitized MC/9 cells migrated toward the bottom wells in a dose- dependent manner after 4 h. Cell migration reached maximal levels at 10 ng/ml DNP-HSA. In contrast, unsensitized MC/9 cells did not migrate toward the bottom wells containing 10 ng/ml DNP- HSA (Ͻ200/well) (Fig. 1A). The degree of cell migration de- creased with excess Ag. Unlike when the bottom wells contained Ag, few sensitized MC/9 cells migrated toward the bottom wells when DNP-HSA was added to the top wells (1,000 vs 81,200 migrated cells/well). These results suggest that the migration was directional. Ag-induced cell migration was significantly increased after 4 h with a 1- to 2-h lag (Fig. 1C). To exclude the possibility that an unknown factor generated from MC/9 cells following Fc⑀RI aggregation acts as a chemotactic factor for mast cells, the supernatant from Ag-stimulated MC/9 cells was applied to the bottom wells. MC/9 cells sensitized with anti-DNP IgE were stimulated by DNP-HSA (10 ng/ml) for 1, 2, or 4 h, and the supernatant was obtained by centrifugation and Ͻ FIGURE 2. Effect of excess monovalent hapten and high dose of acti- filtration. Unsensitized MC/9 cells did not migrate at all ( 200/ nomycin D on Ag-induced mast cell migration. A, MC/9 cells sensitized well) to the bottom wells containing supernatants from cultures with anti-DNP IgE (5 ϫ 106 cells/ml) were incubated in the presence or stimulated for 1, 2, and 4 h with DNP-HSA (10 ng/ml). Incubating absence of DNP-lysine (10Ð1000 ␮M) for 15 min. Cells (1 ϫ 106) were sensitized MC/9 cells for 15 min with a large excess of monovalent then placed in the upper wells, and Ag (10 ng/ml DNP-HSA) was applied hapten (DNP-lysine) disrupted multivalent binding and inhibited to the the bottom wells. Migrated cells were counted after 4 h. The number subsequent cell migration induced by multivalent Ag (DNP-HSA) of migrated cells was standardized by comparison with the number of in a dose-dependent manner. Cell migration was completely migrated cells in the absence of DNP-lysine. Data are shown as means Ϯ ϭ ϫ 6 blocked by 1 mM DNP-lysine (Ͻ200/well) (Fig. 2A). Supernatants SD (n 4). B, MC/9 cells sensitized with anti-DNP IgE (5 10 cells/ml) ␮ derived from Ag-stimulated culture incubated for 1, 2, and 4 h with were incubated with 5 g/ml actinomycin D or control vehicle (0.1% ethanol) for 30 min. Thereafter, cells (1 ϫ 106) were placed in the upper wells, DNP-HSA (10 ng/ml) were placed in the bottom wells. The top and Ag (10 ng/ml DNP-HSA) was applied to the the bottom wells. Acti- wells contained sensitized MC/9 cells in the presence of 1 mM nomycin D (5 ␮g/ml) or control vehicle was added to both top and bottom Ͻ DNP-lysine. Sensitized MC/9 cells hardly migrated ( 200/well) to wells. Migrated cells were counted after 1, 2, and 4 h. The number of the bottom wells under these conditions. To further confirm that migrated cells was standardized by comparison with the number of mi- Ags can cause migration in the absence of chemotactic factors in grated cells in control vehicle after 4 h. Data are shown as means Ϯ SD the medium, we examined the chemotactic ability of sensitized (n ϭ 4). MC/9 cells to Ag in the presence of a high dose of actinomycin D (5 ␮g/ml). Sensitized MC/9 cells were incubated with actinomycin D for 30 min and placed in the top wells. Actinomycin D (5 ␮g/ml) and a control vehicle (0.1% ethanol) were added to both wells, and Few sensitized IL-3/SCF-BMMC stimulated by DNP-HSA in the DNP-HSA was added to the bottom wells. Ag-induced cell migra- top wells migrated toward the bottom wells (640 vs 11,200 mi- tion was not inhibited by actinomycin D after incubation for 1, 2, grated cells/well). and 4 h (Fig. 2B). Another unique receptor on mast cells is c-Kit, a transmembrane Effects of wortmannin, a MAPK/ERK kinase (MEK)1 inhibitor, ⑀ tyrosine kinase receptor (17, 18). The cognate ligand for c-Kit, PD98059, and the p38 MAPK inhibitor, SB203580, on Fc RI- SCF, promotes the growth, differentiation, and activation of mast or c-Kit-mediated cell migration cells (19, 20). SCF has chemotactic activity for mast cells (4, 5) We showed that three members of the MAPK family, JNK, p38 and, when applied to the bottom wells, also induced cell migration. MAPK, and ERK, are activated by Fc⑀RI aggregation in mast cells Migrated cells reached a maximal level at 10 ng/ml after 4 h (Fig. (13Ð16). Therefore, we investigated whether MAPKs are involved 1B). However, the degree of cell migration was weaker than that in the intracellular mechanism of mast cell migration. Fc⑀RI-me- induced by the Ag (Fig. 1C). diated activation of JNK and p38 MAPK (to a lesser extent) is The MC/9 mast cell line has been used as a model of mast cells inhibited by wortmannin, whereas ERK activation was resistant to to investigate intracellular signaling and function (13Ð15, 21, 22). wortmannin (13Ð16). The present study showed that wortmannin To examine the generality of this phenomenon, we investigated marginally inhibited Ag-induced mast cell migration (Fig. 3A). whether IL-3/SCF-BMMC migrate toward Ag as well as MC/9 Therefore, we examined the effects of wortmannin on the Fc⑀RI- cells. Sensitized IL-3/SCF-BMMC migrated toward the bottom or c-Kit-mediated activation of JNK, p38 MAPK, and ERK in the wells in a dose-dependent manner after a 6-h incubation (Fig. 1D). serum-free medium that had been used to study cell migration. The Journal of Immunology 2301

FIGURE 4. Effects of wortmannin on Fc⑀RI- or c-Kit-mediated activation of JNK and p38 MAPK in MC/9 cells. MC/9 cells sensitized with anti-DNP IgE were incubated with 0.01% DMSO (Ϫ Wortmannin) or 100 nM wortmannin (ϩ Wortmannin) for 15 min. Cells were then incubated with control vehicle (PBS) or 10 ng/ml DNP-HSA for 15 min (JNK) or 5 min (p38 MAPK). Kinase activities in cells were measured as described in Materials and Methods.Fc⑀RI-mediated JNK activation was inhibited by 100 nM wortmannin (A), whereas p38 MAPK was partially inhibited (B). The autoradiograph is representative of two independent experiments. MC/9 cells were incubated with 0.01% DMSO (Ϫ Wortmannin) or 100 nM wortmannin (ϩ Wortmannin) for 15 min. Cells were then incubated with control vehicle (PBS) or 10 ng/ml SCF for 15 min (JNK) or 5 min (p38 MAPK). Kinase activities in cells were measured as described in Materials and Methods. Wortmannin did not inhibit SCF-induced activation of JNK (C) or p38 MAPK (D). The autoradiograph is representative of two independent experiments.

Effects of the p38 MAPK inhibitor, SB203580, on MAPKAPK2 FIGURE 3. Effects of wortmannin, a MEK1 inhibitor, PD98059, and a activation p38 MAPK inhibitor, SB203580 on mast cell migration. MC/9 cells sensitized with anti-DNP IgE or unsensitized MC/9 cells (5 ϫ 106 cells/ml) In contrast to SB203580, wortmannin slightly blocked mast cell were incubated with 100 nM wortmannin (15 min) (A), 30 ␮M PD98059 migration and partially inhibited the Fc⑀RI-mediated activation of for 30 min (B), 0.1Ð10 ␮M SB203580 (30 min) (C), or control vehicle p38 MAPK. Therefore, we further examined how p38 MAPK reg- (0.01Ð0.1% DMSO). Thereafter, cells (1 ϫ 106 cells) were placed in the ulates mast cell migration. Activation of p38 MAPK results in the top wells, and Ag (10 ng/ml DNP-HSA) or SCF (10 ng/ml) was added to phosphorylation of heat shock protein (HSP)25/27, which may the bottom wells. SB203580, PD98059, wortmannin, or control vehicle modulate F-actin polymerization in several types of cells (23Ð25). was added to both top and bottom wells. Migrated cells were counted after Because SB203580 is a reversible inhibitor of p38 MAPK but not 4 h. The number of migrated cells in the presence of SB203580, PD98059, and wortmannin was standardized by that in control vehicle. Data are of the phosphorylation of p38 MAPK regulated by MAPK kinase shown as means Ϯ SD (n ϭ 6). (MKK)3 or MKK6, it is difficult to evaluate the degree of inhibitory effect of SB203580 on p38 MAPK activity in intact cells using lysates (26). Therefore, we assessed the activation of MAP- KAPK2 because this enzyme is phosphorylated and activated by p38 MAPK (8Ð10). MC/9 cells sensitized with anti-DNP IgE were Wortmannin completely inhibited Fc⑀RI-mediated JNK activation challenged by DNP-HSA or SCF. MAPKAPK2 activity reached (Ͼ90%), but only partially inhibited Fc⑀RI-mediated p38 MAPK maximum levels at 5 min. MAPKAPK2 activation was mediated activation (Fig. 4, A and B). This suggests that Fc⑀RI-mediated more by Fc⑀RI than by c-Kit (Fig. 5A). Fig. 5B shows that MAP- JNK activation is not required for cell migration (Fig. 3A). In con- KAPK2 activation through either Fc⑀RI or c-Kit was inhibited by trast to signaling through Fc⑀RI, wortmannin failed to alter the SB203580 in a dose-dependent manner, and that susceptibility c-Kit-mediated activation of JNK and p38 MAPK (Fig. 4, C and through either receptor did not differ. SB203580 inhibited MAP- ⑀ D). The MEK1 inhibitor, PD98059, did not affect Fc RI- or c-Kit- KAPK2 activation in MC/9 cells almost completely at 10 ␮M and mediated migration of MC/9 cells (Fig. 3B), but significantly in- considerably (60Ð70%) at 1 ␮M. hibited ERK activation (data not shown), in agreement with our To exclude the possibility of SB203580 cytotoxicity, we com- previous findings (15). These results suggested that ERK activa- 2ϩ pared the [Ca ]i increase in sensitized MC/9 cells stimulated by tion is not required for mast cell migration. In contrast, the p38 DNP-HSA after a 4.5-h incubation with 10 ␮M SB203580 to that MAPK inhibitor, SB203580, inhibited Fc⑀RI- and c-Kit-mediated 2ϩ with a control vehicle. The difference in the [Ca ]i increase be- migration in a dose-dependent manner. SB203580 (10 ␮M) sig- tween the two conditions was not significant (control vs nificantly inhibited Fc⑀RI- and c-Kit-mediated migration (81 and SB203580, 1055 Ϯ 233 nM vs 1005 Ϯ 242 nM; mean Ϯ SD; n ϭ 76%, respectively). The inhibitory effect of SB203580 on cell mi- 6). These results indicate that almost complete inhibition of p38 gration was more significant at 10 ␮M than at 1 ␮M (Fig. 3C). MAPK and MAPKAPK2 activation is required to block mast cell 2302 SENSITIZED MAST CELLS MIGRATE TOWARD THE Ag

FIGURE 6. Y-27632 inhibits Fc⑀RI- or c-Kit-mediated migration but not MAPKAPK2 activation in MC/9 cells. A, MC/9 cells sensitized with anti-DNP IgE or not (5 ϫ 106 cells/ml each) were incubated with 10 ␮M Y-27632 or control vehicle (distilled water) for 15 min. Cells (1 ϫ 106 cells) were then placed in the upper wells, and Ag (10 ng/ml DNP-HSA) or SCF (10 ng/ml) was applied to the bottom wells. Y-27632 or control vehicle was added to both top and bottom wells. Migrated cells were counted after 4 h. The number of migrated cells in the presence of Y-27632 was standardized by comparison with that in the presence of control vehicle. B, MC/9 cells sensitized or not with anti-DNP IgE (3 ϫ 106 cells each) were incubated for 15 min with Y-27632 or control vehicle and were ⑀ FIGURE 5. MAPKAPK2 is activated following aggregation of Fc RI then stimulated for 5 min by 10 ng/ml DNP-HSA or 10 ng/ml SCF. MAP- or ligation of c-Kit in MC/9 cells. A, MC/9 cells sensitized with anti-DNP KAPK2 activities in the cells were measured as described in Materials and ϫ 6 IgE or unsensitized MC/9 cells (3 10 ) were stimulated by 10 ng/ml Methods and standardized to that in the presence of control vehicle. DNP-HSA or 10 ng/ml SCF. MAPKAPK2 activities were measured at 0, 1, 5, 15, 30, and 60 min after adding DNP-HSA or SCF. Fold increases in MAPKAPK2 activity (means Ϯ SD; n ϭ 4) are shown. B, MC/9 cells Effects of Y-27632 on Fc⑀RI- or c-Kit-mediated migration sensitized with anti-DNP IgE (3 ϫ 106 cells) or unsensitized MC/9 cells (3 ϫ 106 cells) were stimulated for 5 min by 10 ng/ml DNP-HSA or 10 The ROCK inhibitor, Y-27632, signiﬁcantly inhibited both the ng/ml SCF after a 30-min incubation with SB203580 or control vehicle Fc⑀RI- and c-Kit-mediated migration of MC/9 cells at 10 ␮M (Fig. (0.1% DMSO). SB203580 dose dependently inhibited both Fc⑀RI- and SCF-mediated MAPKAPK2 activation. SB203580 (10 ␮M) completely blocked MAPKAPK2 activation. Kinase activities were measured as increased substrate phosphorylation in the presence of SB203580 or control vehicle and were standardized by comparison with kinase activity in the presence of control vehicle. Results from four independent experiments shown as means Ϯ SD. C, MC/9 cells sensitized with anti-DNP IgE or not (3 ϫ 106 cells each) were stimulated for 5 min by 10 ng/ml DNP-HSA or 10 ng/ml SCF after a 15-min incubation with 100 nM wortmannin or control vehicle (0.01% DMSO). Kinase activities were measured as increased substrate phosphorylation in the presence of 100 nM wortmannin com- pared with control vehicle. Results from four independent experiments shown as means Ϯ SD.

migration, and that partial activation of these kinases is sufficient for MC/9 cells to migrate. Wortmannin, as well as 1 ␮M SB203580, weakly inhibited mast cell migration as described FIGURE 7. Putative intracellular mechanism of Fc⑀RI-mediated mast ⑀ above, although wortmannin partially inhibited the Fc⑀RI-medi- cell migration. Aggregation of Fc RI in mast cells activated p38 MAPK ated activation of p38 MAPK and MAPKAPK2 (Figs. 4B and 5C). and MAPKAPK2/3 followed by phosphorylation of HSP25 (HSP27). Such aggregation also activated Rho/ROCK followed by phosphorylation of my- These findings support the notion that the partial inhibition of p38 osin light chain (MLC) independently of p38 MAPK activation. Activation MAPK and MAPKAPK2 is not sufficient to block mast cell of both p38 MAPK and ROCK seems to be required for mast cell migra- migration. tion. MP, Myosin phosphatase. The Journal of Immunology 2303

6A). To investigate the mechanism of this action, we measured blocked by the C3 exoenzyme in permeabilized mast cells (45). MAPKAPK2 activation following ligation of Fc⑀RI or c-Kit in the Several proteins have been identified as Rho effectors, including presence or absence of Y-27632. Subsequent Fc⑀RI- or c-Kit-me- ROCK (11), which plays a key role in focal adhesion and stress diated MAPKAPK2 activation was not affected by Y-27632 (Fig. fiber formation and in the Ca2ϩ sensitization of smooth muscle 6B). Y-27632 should inhibit cell migration by inhibiting Rho-de- (46). Y-27632 was originally identified as an inhibitor of two pendent ROCK activation. Although the activation of Rho/ROCK ROCK isoforms and its antihypertensive effect is achieved by pre- is required for mast cell migration, the Rho-dependent intracellular venting ROCK from inhibiting smooth muscle protein phosphatase signaling pathway seems to be independent of the p38 MAPK- 1M, the major myosin phosphatase of this tissue. This decreases dependent pathway (Fig. 7). the phosphorylation of myosin, increases arterial smooth muscle relaxation, and hence increases the dilation of blood vessels (47). Discussion Y-27632 also inhibits Rho-mediated cell transformation, tumor Allergic reaction is initiated by Fc⑀RI aggregation in mast cells cell invasion, and chemotaxis in many types of cells (48Ð52). Both following the binding of a multivalent Ag to IgE. It is believed that Fc⑀RI- and c-Kit-mediated migration of mast cells were inhibited such Ags bind to IgE on mast cells by chance. Mast cells accu- by Y-27632 in the present investigation. A Rho/ROCK activation mulate at sites of inflammation and wound healing (2, 3) and in the pathway seems to be independent of the p38 MAPK/MAPKAPK2 epithelial layer of the mucosa of patients with allergic rhinitis and activation pathway because Y-27632 did not affect MAPKAPK2 bronchial asthma after natural exposure to pollen or a provocation activation. test (27Ð32). This area of the mucosa normally has very few, if In summary, we demonstrated that the Ag itself induces direc- any, mast cells. Such epithelial mast cells bear surface IgE (33). tional mast cell migration, the process of which is regulated by The chemotactic factors responsible for mast cells in an allergic both p38 MAPK/MAPKAPK2 and Rho/ROCK activation path- reaction are undefined, although many chemokines and cytokines, ways. In allergic conditions such as bronchial asthma and allergic including SCF, RANTES, eotaxin, monocyte chemotactic pro- rhinitis, mast cell accumulation in a lesion is important in the de- tein-1, and TGF-␤, can elicit mast cell migration (4Ð7, 34, 35). In velopment of disease. Our findings provide novel evidence of Ag- the present study, we found that the Ag itself has chemotactic induced mast cell migration and should support the development activity for IgE-sensitized mast cells, which suggests that mast of a therapeutic approach to this phenomenon in allergic states. cells sensitized with Ag-specific IgE migrate to sites of optimal Ag concentration. This phenomenon is reasonable if mast cells pri- Acknowledgments marily work as host defense cells against parasites and other or- We thank the WelFide Corporation for providing Y-27632. ganisms through Fc⑀RI, although it also may result in the development of allergic states. References 1. Metcalfe, D. D., D. Baram, and Y. A. Mekori. 1997. Mast cells. Physiol. Rev. The mammalian MAPK family consists of ERK, JNK, and p38 77:1033. MAPK. We recently showed that Fc⑀RI aggregation or c-Kit li- 2. Wersil, B. K., Y. A. Mekori, and S. J. Galli. 1989. The contribution of mast cells gation activates all three (13Ð16). These kinases should regulate to immunological responses with IgE and/or T cell-mediated components. In Mast Cell and Basophil Differentiation and Function in Health and Disease. gene expression by phosphorylating transcription factors. We also S.J. Galli and K.F. Austen, eds. Raven, New York, p. 229. showed that phosphatidylinositol 3-kinase and MEK kinases, es- 3. Galli, S. J., and A. M. Dvorak. 1984. What do mast cells have to do with delayed pecially MEK kinase 2, play an important role in cytokine pro- hypersensitivity? Lab. Invest. 50:365. 4. Meninger, C. J., H. Yano, R. Rottapel, A. Bernstein, K. M. Zsebo, and duction by Ag-stimulated mast cells, mainly through JNK activa- B. R. Zetter. 1992. The c-kit receptor ligand functions as a mast cell chemoat- tion (36). The role of p38 MAPK in mast cells is obscure, although tractant. Blood 79:958. ⑀ 5. Nilsson, G., J. H. Butterfield, K. Nilsson, and A. Siegbahn. 1994. Stem cell factor Fc RI-mediated activation of p38 MAPK as well as JNK is reg- is a chemotactic factor for human mast cells. J. Immunol. 153:3717. ulated by phosphatidylinositol 3-kinase and calcineurin (14Ð16). 6. Thompson, H. L., P. D. Burbelo, Y. Yamada, H. K. Kleinman, and MAPKAPK2, which was originally designated HSP25/27 ki- D. D. Metcalfe. 1989. Mast cells chemotax to laminin with enhancement after IgE-mediated activation. J. Immunol. 143:4188. nase, is phosphorylated and activated by p38 MAPK (8Ð10). 7. Taub, D., J. Dastych, N. Inamura, J. Upton, D. Kelvin, D. Metcalfe, and HSP25/27 phosphorylation by MAPKAPK2/3 induces actin fiber J. Oppenheim. 1995. Bone marrow-derived murine mast cells migrate, but do not polymerization and is required for the directional migration of degranulate, in response to chemokines. J. Immunol. 154:2393. 8. Stokoe, D., D. G. Campbell, S. Nakielny, H. Hidaka, S. J. Leevers, C. Marshall, many kinds of cells, including human bronchial smooth muscle and P. Cohen. 1992. MAPKAP kinase-2; a novel protein kinase activated by and endothelial cells (24, 25, 37Ð41). The p38 MAPK inhibitor, mitogen-activated protein kinase. EMBO J. 11:3985. 9. Stokoe, D., K. Engel, D. G. Campbell, P. Cohen, and M. Gaestel. 1992. Identi- SB203580, inhibits cell migration in response to platelet-derived fication of MAPKAP kinase 2 as a major enzyme responsible for the phosphor- growth factor, sphingosine 1-phosphate, IL-1␤, and TGF-␤ (42, ylation of the small mammalian heat shock proteins. FEBS Lett. 313:307. 43). The activated mutant of MKK6, which activates p38 MAPK, 10. Rouse, J., P. Cohen, S. Trigon, M. Morange, A. Alonso-Llamazares, D. Zamanillo, T. Hunt, and A. R. Nebreda. 1994. A novel kinase cascade trig- increases cell migration, and either the dominant negative p38 gered by stress and heat shock that stimulates MAPKAP kinase-2 and phosphor- MAPK or an HSP27 phosphorylation mutant blocks cell migration ylation of the small heat shock proteins. Cell 78:1027. in human bronchial smooth muscle cells (37). The present study 11. Matsui, T., M. Amano, T. Yamamoto, K. Chihara, M. Nakafuyu, M. Ito, T. Nakano, K. Okawa, A. Iwamatsu, and K. Kaibuchi. 1996. Rho-associated shows that MAPKAPK2 was significantly activated following kinase, a novel serine/threonine kinase, as a putative target for small GTP binding Fc⑀RI or c-Kit ligation, and its activation was abolished by 10 ␮M protein Rho. EMBO J. 15:2208. ⑀ 12. Okjajima, F., K. Sato, M. Nazarea, K. Sho, and Y. Kondo. 1989. A permissive SB203580. Although Fc RI-mediated activation of p38 MAPK role of pertussis toxin substrate G-protein in P2-purinergic stimulation of phos- and MAPKAPK2 was significantly inhibited by wortmannin, it phoinositide turnover and arachidonate release in FRTL-5 thyroid cells: cooper- was not sufficient to block mast cell migration. The inhibitory ef- ative mechanism of signal transduction systems. J. Biol. Chem. 264:13029. ␮ 13. Ishizuka, T., A. Oshiba, N. Sakata, N. Terada, G. L. Johnson, and E. W. Gelfand. fect of SB203580 on mast cell migration was weaker at 1 M than 1996. Aggregation of the Fc⑀RI on mast cells stimulates c-Jun amino-terminal at 10 ␮M. These results suggested that MAPKAPK2 must be com- kinase activity: a response inhibited by wortmannin. J. Biol. Chem. 271:12762. pletely inhibited to block mast cell migration. 14. Ishizuka, T., N. Terada, P. Gerwins, E. Hamelmann, A. Oshiba, G. R. Fanger, G. L. Johnson, and E. W. Gelfand. 1997. Mast cell tumor necrosis factor-␣ The small GTPase, Rho, controls cell adhesion and motility production is regulated by MEK kinases. Proc. Natl. Acad. Sci. USA 94:6358. through reorganization of the actin cytoskeleton (44). Rho in mast 15. Ishizuka, T., H. Kawasome, N. Terada, K. Takeda, P. Gerwins, G. M. Keller, G. L. Johnson, and E. W. Gelfand. 1998. Stem cell factor augments Fc⑀RI- cells promotes cortical F-actin disassembly in addition to control- mediated TNF␣ production and stimulates MAP kinases via a different pathway ling secretion and actin polymerization, and all of these effects are in MC/9 mast cells. J. Immunol. 161:3624. 2304 SENSITIZED MAST CELLS MIGRATE TOWARD THE Ag

16. Ishizuka, T., K. Chayama, K. Takeda, E. Hamelmann, N. Terada, G. M. Keller, 33. Bachert, C., P. Prohaska, and U. Pipkorn. 1990. IgE-positive mast cells on the G. L. Johnson, and E. W. Gelfand. 1999. Mitogen-activated protein kinase acti- human nasal mucosal surface in response to allergen exposure. Rhinology 28:149. vation through Fc⑀ receptor I and stem cell factor receptor is differentially reg- 34. Olsson, N., E. Piek, P. ten Dijke, and G. Nilsson. 2000. Human mast cell mi- ulated by phosphatidylinositol 3-kinase and calcineurin in mouse bone marrow- gration in response to members of the transforming growth factor-␤ family. derived mast cells. J. Immunol. 162:2087. J. Leukocyte Biol. 67:350. 17. Yarden, Y., J. A. Escobedo, W.-J. Kuang, T. L. Yang-Feng, T. O. Daniel, 35. Mattori, S., V. Ackerman, E. Vittori, and M. Marini. 1995. Mast cell chemotactic P. M. Tremble, E. Y. Chen, M. E. Ando, R. N. Harkins, U. Francke, et al. 1986. activity of RANTES. Biochem. Biophys. Res. Commun. 209:316. Structure of the receptor for platelet-derived growth factor helps define a family 36. Garrington, T. P., T. Ishizuka, P. J. Papst, K. Chayama, S. Webb, T. Yujiri, of closely related growth factor receptors. Nature 323:226. W. Sun, S. Sather, D. M. Russell, S. B. Gibson, et al. 2000. MEKK2 gene 18. Qui, F., P. Ray, K. Brown, P. E. Barker, S. Jhanwar, F. H. Ruddle, and P. Besmer. disruption causes loss of cytokine production in response to IgE and c-Kit ligand 1988. Primary structure of c-kit: relationship with the CSF-1/PDGF receptor ki- stimulation of ES cell-derived mast cells. EMBO J. 19:5387. nase family—oncogenic activation of v-kit involves deletion of extracellular do- 37. Hedges, J. C., M. A. Dechert, I. A. Yamboliev, J. L. Martin, E. Hickey, main and C terminus. EMBO J. 7:1003. L. A. Weber, and W. T. Gerthoffer. 1999. A role for p38 (MAPK)/HSP27 path- 19. Tsai, M., T. Takeishi, H. Thompson, K. E. Langley, K. M. Zsebo, D. D. Metcalfe, way in smooth muscle cell migration. J. Biol. Chem. 274:24211. E. N. Geissler, and S. J. Galli. 1991. Induction of mast cell proliferation, matu- 38. Larsen, J. K., I. A. Yamboliev, L. A. Weber, and W. T. Gerthoffer. 1997. Phos- ration, and heparin synthesis by the rat c-kit ligand, stem cell factor. Proc. Natl. phorylation of the 27-kDa heat shock protein via p38 MAP kinase and MAPKAP Acad. Sci. USA 88:6382. kinase in smooth muscle. Am. J. Physiol. 273:L930. 20. Tsai, M., L.-S. Shih, G. F. J. Newlands, T. Takeishi, K. E. Langley, K. M. Zsebo, H. R. P. Miller, E. N. Geissler, and S. J. Galli. 1991. The rat c-kit ligand, stem 39. Piotrowicz, R. S., and E. G. Levin. 1997. Basolateral membrane-associated 27- cell factor, induces the development of connective tissue-type and mucosal mast kDa heat shock protein and microfilament polymerization. J. Biol. Chem. 272: cells in vivo: analysis by anatomical distribution, histochemistry, and protease 25920. phenotype. J. Exp. Med. 174:125. 40. Hout, J., F. Houle, F. Marceau, and J. Landry. 1997. Oxidative stress-induced 21. Galli, S. J., A. M. Dvorak, J. A. Marcum, T. Ishizaka, G. Nabel, H. Der Simonian, actin reorganization mediated by the p38 mitogen-activated protein kinase/heat K. Pyne, J. M. Goldin, R. D. Rosenberg, H. Cantor, and H. F. Dvorak. 1982. Mast shock protein 27 pathway in vascular endothelial cells. Circ. Res. 80:383. cell clones: a model for the analysis of cellular maturation. J. Cell Biol. 95:435. 41. Razandi, M., A. Pedram, and E. R. Levin. 2000. Estrogen signals to the preser- 22. Nabel, G., S. J. Galli, A. M. Dvorak, H. F. Dvorak, and H. Cantor. 1981. Inducer vation of endothelial cell form and function. J. Biol. Chem. 275:38540. T lymphocytes synthesize a factor that stimulates proliferation of cloned mast 42. Heuertz, R. M., S. M. Tricomi, U. R. Ezekiel, and R. O. Webster. 1999. C-re- cells. Nature 291:332. active protein inhibits chemotactic peptide-induced p38 mitogen-activated pro- 23. Rousseau, S., F. Houle, J. Landry, and J. Hout. 1997. p38 MAP kinase activation tein kinase activity and human neutrophil movement. J. Biol. Chem. 274:17968. by vascular endothelial growth factor mediates actin reorganization and cell mi- 43. Kimura, T., T. Watanabe, K. Sato, J. Kon, H. Tomura, K. Tamama, A. Kuwabara, gration in human endothelial cells. Oncogene 15:2169. T. Kanda, I. Kobayashi, H. Ohta, M. Ui, and F. Okajima. 2000. Sphingosine 24. Guay, J., H. Lambert, G. Gingras-Breton, J. N. Lavoie, J. Hout, and J. Landry. 1-phosphate stimulates proliferation and migration of human endothelial cells 1997. Regulation of actin filament dynamics by p38 MAP kinase-mediated phos- possibly through the lipid receptors, Edg-1 and Edg-3. Biochem. J. 348:71. phorylation of heat shock protein 27. J. Cell. Sci. 110:357. 44. Narumiya, S., T. Ishizaki, and N. Watanabe. 1997. Rho effectors and reorgani- 25. Rousseau, S., F. Houle, H. Kotanides, L. Witte, J. Waltenberger, J. Landry, and zation of actin cytoskeleton. FEBS Lett. 410:68. J. Huot. 2000. Vascular endothelial growth factor (VEGF)-driven actin-based 45. Sullivant, R., L. S. Price, and A. Koffer. 1999. Rho controls cortical F-actin motility is mediated by VEGFR2 and requires concerted activation of stress- disassembly in addition to, but independently of, secretion in mast cells. J. Biol. activated protein kinase 2 (SAPK2/p38) and geldanamycin-sensitive phosphory- Chem. 274:38140. lation of focal adhesion kinase. J. Biol. Chem. 275:10661. 46. Iizuka, K., A. Yoshii, K. Samizo, H. Tsukagoshi, T. Ishizuka, K. Dobashi, 26. Cuenda, A., J. Rouse, Y. N. Doza, R. Meier, P. Cohen, T. F. Gallagher, T. Nakazawa, and M. Mori. 1999. A major role for the Rho-associated coiled coil P. R. Young, and J. C. Lee. 1995. SB203580 is a specific inhibitor of a MAP protein kinase in G-protein-mediated Ca2ϩ sensitization through inhibition of kinase homologue which is stimulated by cellular stresses and interleukin-1. myosin phosphatase in rabbit trachea. Br. J. Pharmacol. 128:925. FEBS Lett. 364:229. 47. Uehata, M., T. Ishizaki, H. Satoh, T. Ono, T. Kawahara, T. Morishita, 27. Slater, A., L. A. Smallman, and A. B. Drake-Lee. 1996. Increase in epithelial H. Tamakawa, K. Yamagami, J. Inui, M. Maekawa, and S. Narumiya. 1997. mast cell numbers in the nasal mucosa of patients with perennial allergic rhinitis. Calcium sensitization of smooth muscle mediated by a Rho-associated protein J. Laryngol. Otol. 110:929. kinase in hypertension. Nature 389:990. 28. Gibson, P. G., C. J. Allen, J. P. Yang, B. J. Wong, J. Dolovich, J. Denburg, and F. E. Hargreave. 1993. Intraepithelial mast cells in allergic and nonallergic asth- 48. Itoh, K., K. Yoshioka, H. Akedo, M. Uehata, T. Ishizaki, and S. Narumiya. 1999. ma: assessment using bronchial brushing. Am. Rev. Respir. Dis. 148:80. An essential part for Rho-associated kinase in the transcellular invasion of tumor 29. Laitinen, L. A., A. Laitinen, and T. Haahtela. 1993. Airway mucosal inflamma- cells. Nat. Med. 5:221. tion even in patients with newly diagnosed asthma. Am. Rev. Respir. Dis. 147: 49. Somlyo, A. V., D. Bradshaw, S. Ramos, C. Murphy, C. E. Myers, and 697. A. P. Somlyo. 2000. Rho-kinase inhibitor retards migration and in vivo dissem- 30. Kawabori, S., M. Okuda, T. Unno, and A. Nakamura. 1985. Dynamics of mast ination of human prostate cancer cells. Biochem. Biophys. Res. Commun. 269: cell degranulation in human allergic nasal epithelium after provocation with al- 652. lergen. Clin. Allergy 15:509. 50. Seasholtz, T. M., M. Majumdar, D. D. Kaplan, and J. H. Brown. 1999. Rho and 31. Viegas, M., E. Gomez, J. Brooks, and R. J. Davies. 1987. Changes in mast cell Rho kinase mediate thrombin-stimulated vascular smooth muscle cell DNA syn- numbers in and out of the pollen season. Int. Arch. Allergy Appl. Immunol. 82: thesis and migration. Circ. Res. 84:1186. 275. 51. Niggi, V. 1999. Rho-kinase in human neutrophils: a role in signaling for myosin 32. Pesci, A., A. Foresi, G. Bertorelli, A. Chetta, and D. Olivieri. 1993. Histochem- light chain phosphorylation and cell migration. FEBS Lett. 445:69. ical characteristics and degranulation of mast cells in epithelium and lamina 52. Kawada, N., S. Seki, T. Kuroki, and K. Kaneda. 1999. ROCK inhibitor Y-27632 propria of bronchial biopsies from asthmatic and normal subjects. Am. Rev. Re- attenuates stellate cell contraction and portal pressure increase induced by endo- spir. Dis. 147:684. thelin-1. Biochem. Biophys. Res. Commun. 266:296.