Signal-Regulated Kinase-Linked Cascade Reactive Oxygen Species

The Journal of Immunology

Transepithelial Migration of Neutrophils in Response to

Leukotriene B4 Is Mediated by a Reactive Oxygen Species-Extracellular Signal-Regulated Kinase-Linked Cascade1

Chang-Hoon Woo,* Min-Hyuk Yoo,† Hye-Jin You,† Sung-Hoon Cho,† Yeung-Chul Mun,‡ Chu-Myong Seong,‡ and Jae-Hong Kim2*

The epithelial cells that form a barrier lining the lung airway are key regulators of neutrophil trafﬁcking into the airway lumen

in a variety of lung inﬂammatory diseases. Although the lipid mediator leukotriene B4 (LTB4) is known to be a principal che- moattractant for recruiting neutrophils to inﬂamed sites across the airway epithelium, the precise signaling mechanism involved

remains largely unknown. In the present study, therefore, we investigated the signaling pathway through which LTB4 induces

transepithelial migration of neutrophils. We found that LTB4 induces concentration-dependent transmigration of DMSO-differentiated HL-60 neutrophils and human polymorphonuclear neutrophils across A549 human lung epithelium. This effect was

mediated via speciﬁc LTB4 receptors and was inhibited by pretreating the cells with N-acetylcysteine (NAC), an oxygen free radical scavenger, with diphenylene iodonium (DPI), an inhibitor of NADPH oxidase-like ﬂavoproteins, or with PD98059, an

extracellular signal-regulated kinase (ERK) inhibitor. Consistent with those ﬁndings, LTB4-induced ERK phosphorylation was

completely blocked by pretreating cells with NAC or DPI. Taken together, our observations suggest LTB4 signaling to transepithelial migration is mediated via generation of reactive oxygen species, which leads to downstream activation of ERK. The

physiological relevance of this signaling pathway was demonstrated in BALB/c mice, in which intratracheal instillation of LTB4

led to acute recruitment of neutrophils into the airway across the lung epithelium. Notably, the response to LTB4 was blocked by

NAC, DPI, PD98059, or CP105696, a speciﬁc LTB4 receptor antagonist. The Journal of Immunology, 2003, 170: 6273–6279.

3 eukotriene B4 (LTB4) is a key mediator of inflammatory LTB4 contributes to inflammatory responses. In particular, the in- processes, immune responses, and host defense against flux of neutrophils into the airway is a common feature of several L infection (1, 2) and is known to stimulate chemotaxis, inflammatory diseases, including acute bronchitis, adult respira- degranulation, release of lysosomal enzymes, and reactive oxygen tory distress syndrome (ARDS), and chronic obstructive pulmo- species (ROS) generation (3, 4). Although the precise signaling nary disease (COPD) (8, 9), as well as severe asthma (10, 11). pathway along which the biological activities of LTB are trans- Once leukocytes are beyond the endothelial barrier, chemotactic

tein-coupled receptors: LTB4 receptor (BLT1), a high affinity re- lial cells and in the lumen of the airway. Airway epithelial cells ceptor first isolated from HL-60 cells, and BLT2, a low affinity may thus be important regulators not only of retention and acti- receptor that shares ϳ45% amino acid identity with BLT1 (5Ð7). vation of leukocytes, but also of their passage into the airway

Neutrophils are the primary targets of LTB4, and their recruit- itself. It has therefore been suggested that LTB4-induced transep- ment to sites of inﬂammation is one of the principal ways in which ithelial migration of neutrophils plays a crucial role in the patho- genesis of ARDS, COPD, and severe asthma. Yet despite this im-

plication, the signaling pathway via which LTB4 mediates *Graduate School of Biotechnology, Korea University, Anam-dong, Seoul, Korea; http://classic.jimmunol.org transepithelial migration of neutrophils remains largely unknown. †Department of Life Science, Kwangju Institute of Science and Technology, Kwangju, Korea; and ‡Department of Oncology/Hematology, Ewha Woman’s Uni- In the present study, therefore, we investigated the signaling versity, College of Medicine, Seoul, Korea pathway by which LTB4 induces transepithelial migration of neu- Received for publication November 15, 2002. Accepted for publication April trophil-like granulocytes (differentiated HL-60 neutrophils (dHL- 14, 2003. 60)) differentiated from HL-60 cells, which have been shown to be The costs of publication of this article were defrayed in part by the payment of page a valid model system for the analysis of human neutrophilic migration charges. This article must therefore be hereby marked advertisement in accordance and chemotaxis (12), and by human polymorphonuclear neutrophils Downloaded from with 18 U.S.C. Section 1734 solely to indicate this fact. (PMNs) isolated from healthy volunteers. Our results indicate that 1 This work was supported by grants from the Basic Research Program (RO2-2002- 000-00029-0), the Science Research Center program (Aging and Apoptosis Research LTB4-induced transepithelial migration is accomplished via an ROS- Center), 2002 from the Korea Science and Engineering Foundation, and the Frontier extracellular signal-regulated kinase (ERK)-linked pathway. In addi- 21 Programs (Proteomics to J.H.K. and Stem Cell to C.M.S.) from the Korea Ministry tion, the physiological relevance of this pathway was validated in vivo of Science and Technology. by demonstrating that ROS generation and ERK activation are both 2 Address correspondence and reprint requests to Dr. Jae-Hong Kim, Graduate School of Biotechnology, Korea University, 5-1 Anam-dong, Sungbuk-gu, Seoul 136-701, required for transmigration of peripheral blood neutrophils across the Korea. E-mail address: [email protected] airway epithelium of BALB/c mice. 3 Abbreviations used in this paper: LTB4, leukotriene B4; ARDS, adult respiratory distress syndrome; BAL, bronchoalveolar lavage; BLT, LTB4 receptor; COPD, Materials and Methods chronic obstructive pulmonary disease; cPLA2, cytosolic phospholipase A2; DCFDA, 2Ј,7Ј-dichloroﬂuorescin diacetate; dHL-60, differentiated HL-60 neutrophils; DPI, di- Chemicals phenylene iodonium; ERK, extracellular signal-regulated kinase; NAC, N-acetylcysteine; PMN, polymorphonuclear neutrophil; PI 3-kinase, phosphatidylinositol 3-ki- 2Ј,7Ј-Dichloroﬂuorescin diacetate (DCFDA) was purchased from Molec- nase; PKC, protein kinase C; ROS, reactive oxygen species. ular Probes (Eugene, OR). LTB4APA, LY1718, genistein, and AACOCF3

were obtained from Biomol (Plymouth Meeting, PA). LTB4 and U75302 dene difluoride membranes using a wet transfer unit (NOVEX, San Diego, were purchased from Cayman Chemical Co. (Ann Arbor, MI). BSA, CA;for2hat100V).Themembranes were then blocked for 1 h with TBS DMSO, PMA, GF109203X, wortmannin, diphenylene iodonium (DPI), containing 0.05% (v/v) Tween 20 plus 5% (w/v) nonfat dry milk and in- and N-acetylcysteine (NAC) were obtained from Sigma-Aldrich (St. Louis, cubated, first for 2 h with the primary Ab (1/2000 dilutions of anti- MO). FBS, DMEM, phenol red-free DMEM, gentamicin, and nonessential phospho-ERK or anti-p38) in TBS containing 0.05% (v/v) Tween 20 plus amino acids were purchased from Life Technologies (Gaithersburg, MD). 3% (w/v) BSA, and then for 1 h with HRP-conjugated secondary Ab. The All other chemicals were obtained from standard sources and were molec- bands were developed using an ECL kit and were visualized on XAR-5 ular biology grade or higher. film (Eastman Kodak, Rochester, NY).

Cell culture and differentiation RNA isolation and RT-PCR Human promyelocytic HL-60 cells obtained from American Type Culture Total cellular RNA was extracted from dHL-60 cells using RNAzol B Collection (Manassas, VA; CCL-240) were grown in RPMI 1640 medium (Tel-Test, Friendswood, TX), dissolved in diethylpyrocarbonate-treated supplemented with 10% heat-inactivated FBS, 0.1 mM nonessential amino water, and quantified by UV scanning. One microgram of the extracted acids, 50 U/ml penicillin, and 50 ␮g/ml streptomycin at 37¡C under a RNA was then reverse transcribed by incubating it for 60 min at 37¡Cin humidified 95%/5% (v/v) mixture of air and CO2. To induce cells to un- 20 ␮l in buffer containing 10 mM Tris (pH 8.3); 50 mM KCl; 5 mM dergo differentiation along the neutrophil lineage, aliquots of HL-60 cell MgCl2; 1 mM each of dATP, dCTP, dGTP, and dTTP; and oligo(dT) suspension (5 ϫ 105 cells/ml) were seeded onto dishes and grown for 5 primers. Hot-start PCR was used to increase the specificity of the ampli- days in the absence or the presence of 1.25% DMSO. Cell differentiation fication with specific primers for human BLT1 (sense, 5Ј-TATGTCTGCG was then evaluated using four criteria (13Ð15): 1) morphological changes GAGTCAGCATGTACGC; antisense, 5Ј-CCTGTAGCCGACGCCCTAT (nuclear segmentation and granules formation) were analyzed microscop- GTCCG), human BLT2 (sense, 5Ј-AGCCTGGAGACTCTGACCGCTTT ically using H&E cytochemical staining; 2) cell proliferation was assayed CG; antisense, 5Ј-GACGTAGAGCACCGGGTTGACGCTA), and human with 0.4% trypan blue exclusion; 3) the capacity to reduce nitro blue tet- GAPDH (15). The PCR protocol entailed 25 cycles of denaturation at 94¡C razolium, a property associated with myeloid cell maturation, was deter- for 30 s and elongation at 68¡C for 2 min in a Mygenie Thurmal Block mined spectrophotometrically at 540 nm; and 4) surface expression of dif- from Bioneer (Taejon, Korea). The amplified products were subjected to ferentiation-related Ags was evaluated by flow cytometry using FITC- electrophoresis on 1.8% agarose gels, after which the bands were visual- conjugated mAbs against CD11b, CD16, and CD33. A549 human lung ized by ethidium bromide staining and photographed using Polaroid 667 epithelial cells were grown in RPMI containing 10% FBS, 0.1 mM non- film (Cambridge, MA). essential amino acids, 50 U/ml penicillin, and 100 ␮g/ml streptomycin. Human PMNs were isolated from blood samples collected from healthy Transient transfection and luciferase assay volunteers in heparin sodium salt (100 U; Sigma-Aldrich; H-3149). PMNs were separated on a Ficoll-Paque density gradient (1.077 Ϯ 0.001 g/ml), fol- Transient transfection was conducted by plating ϳ2 ϫ 105 cells in 60-mm lowed by 2% dextran sedimentation to remove the majority of RBCs (Amer- dishes and then adding Lipofectamine Plus/DNA complex prepared with sham Pharmacia Biotech, Uppsala, Sweden). Any remaining RBCs were lysed 1.8 ␮g of DNA/dish. To control for variations in cell number and trans- in 0.15 M ammonium chloride solution. The remaining PMNs were suspended fection efficiency, all clones were cotransfected with 0.4 ␮g of pSV40- in RPMI and used for transepithelial migration assays. ␤GAL, a eukaryotic expression vector containing the Escherichia coli ␤-galactosidase (lacZ) structural gene under transcriptional control of the Transepithelial migration assays SV40 promoter. The amount of DNA used in each transfection was held constant (1.8 ␮g) by adding sonicated calf thymus DNA. To measure Transepithelial migration assays were conducted as previously described ERK’s kinase activity using a PathDetect trans-reporting system (catalog ␮ (8). Briefly, a collagen-coated, polycarbonate filter insert (5- m pore size) no. 219005; Stratagene, La Jolla, CA), Elk1 fused to trans-activator plas- was placed inside each well of several Transwell 12-well tissue culture mid was cotransfected with pFR-Luc reporter plasmid according to the plates to create a well within a well, after which 5.2 ml of medium was

by guest on October 1, 2021. Copyright 2003 Pageant Media Ltd. manufacturer’s protocol. After incubating for 3 h with the Lipofectamine added to completely submerge the insert. Using sterilized forceps, the in- Plus/DNA complex, the cells were rinsed with serum-free medium and sert was then inverted, ensuring that no air bubbles were trapped under it, incubated in fresh RPMI 1640 supplemented with 10% FBS for an addi- and exactly 0.8 ml of medium was removed from each well. The insert was tional 24 h. The cells in each dish were then treated with LTB4 or control then moved to the center of its well to avoid an overlapping meniscus at the buffer for 6 h, after which they were harvested, lysed in 0.1 ml of lysis ϫ 6 edge. Each inverted insert was then gently seeded with a total of 0.5 10 solution (0.2 M Tris-Cl (pH 7.6) and 0.1% Triton X-100), and spun for 1 ␮ A549 cells in two 50- l aliquots of cell suspension, yielding a conﬁgura- min. The resultant supernatants were assayed for protein and ␤-galactosi- tion in which the cells were growing on the microporous membrane with dase activity. Luciferase activity was assayed in 10-␮l samples of extract. air above and medium below. The plates were then incubated at 37¡C under Luciferase luminescence was counted in a luminometer (Turner Design, a5%CO2 atmosphere for 4Ð5 days, at which time the inserts were returned Sunnyvale, CA; TD-20/20) and normalized to cotransfected ␤-galactosi- to their original orientation for the transmigration experiments. For exper- dase activity (16). Transfection experiments were performed in triplicate

http://classic.jimmunol.org imentation, HL-60 cells or PMNs were seeded into the upper chambers of using two independently isolated sets of cells, and the results were ϫ 5 the Transwell system to a density of 5 10 /well, and the indicated con- averaged. centrations of LTB4 or saline were added to the lower chambers. After incubating at 37¡C in 95%/5% (v/v) mixture of air and CO2 for 6 h, the LTB -challenged mouse model amount of basal to apical transepithelial migration of dHL-60 or human 4 PMNs was determined by counting the migrated cells on the underside of Pathogen-free female BALB/c mice (8 wk old) were purchased from the the insert. When assessing the effects of inhibitors, cells were pretreated Korea Research Institute of Chemistry Technology (Daejon, Korea). The

with the inhibitor of interest for 20 min before the addition of LTB4. mice were housed throughout the experiments in a laminar flow cabinet Downloaded from and were provided with standard laboratory chow and water ad libitum. For Measurement of intracellular H2O2 intratracheal administration, mice were anesthetized by i.p. injection of 150 ␮l of saline containing 13 mg/ml ketamine HCl (Yuhan, Seoul, Korea) and Intracellular H O was measured as a function of DCF fluorescence using 2 2 88.6 ␮g/ml xylazine (Bayer, Seoul, Korea). LTB (Cayman Chemicals, a fluorometer. Briefly, HL-60 cells were grown for 5 days in the presence 4 Ann Arbor, MI) was prepared in PBS containing 20% ethanol to a final of 1.25% DMSO and then were serum-starved for6hinphenol-red free concentration of 20 ␮g/ml, and 50 ␮l was injected intratracheally. After RPMI supplemented 0.5% FBS. To measure intracellular H O , cells were 2 2 6 h, bronchoalveolar lavage (BAL) fluid was collected from the lung space then incubated for 10 min with the H O -sensitive fluorophore DCFDA (5 2 2 as previously described (17). Briefly, the chest cavity was exposed to allow ␮g/ml), which, when taken up, fluorescently labels intracellular H O with 2 2 for expansion, after which the trachea was carefully intubated, and the DCF. After incubation with DCFDA, cells were exposed to LTB or PMA 4 catheter was secured with ligatures. Eight hundred microliters of pre- for 5 min and immediately observed under a fluorometer from Molecular warmed 0.9% NaCl solution was slowly infused into the lung and with- Devices (SpectraMAX GeminiXS; Sunnyvale, CA). DCF fluorescence was drawn. The aliquots were then pooled and then kept at 4¡C. To obtain excited at 488 nm, and the evoked emission was filtered with a 515-nm differential cell counts, the BAL fluid was cytospun and stained with Diff- long-pass filter. Quik (Dade Diagnostics, Aguada, Puerto Rico). Two independent, blinded SDS-PAGE and immunoblot analysis investigators then counted the cells using a hemocytometer. Approximately 400 cells were counted in each of four randomly selected locations. The Protein samples were heated at 95¡C for 5 min and then subjected by percentage of neutrophils was multiplied by the total cell number to obtain SDS-PAGE on 10% acrylamide gels, followed by transfer to polyvinyli- the neutrophil count. To evaluate lung histology, the lungs were removed The Journal of Immunology 6275

from mice 6 h after LTB4 challenge, cut into sections, and stained with H&E (Sigma-Aldrich) (18). Data analysis and statistics Data are expressed as mean percentages of control Ϯ SD. Statistical com- parisons between groups were made using Student’s t tests. Values of p Ͻ 0.01 were considered signiﬁcant. Results

LTB4-induced transepithelial migration of dHL-60 cells Human promyelocytic HL-60 cells induced with DMSO to differ- entiate toward neutrophil-like granulocytes (dHL-60) have been FIGURE 2. Role of BLTs in LTB -induced transepithelial migration of shown to be a valid model system for the analysis of human neu- 4 dHL-60 cells. A, RT-PCR analysis of BLT1 and BLT2 was conducted trophil migration and chemotaxis, (12, 19, 20). In the present study using normalized ﬁrst-strand cDNAs from parental (day 0) and DMSO- LTB4 induced migration of dHL-60 cells, but not the parental differentiated HL-60 cells (3 and 5 days) as templates. The data are rep- HL-60 cells, across a monolayer of A549 human type II alveolar resentative from three independent experiments with similar results. B,

epithelial cells (Fig. 1). The effect was concentration dependent, Effect of BLT antagonists on LTB4-induced transepithelial migration of with 2.3-fold more migration occurring in the presence of 100 nM dHL-60 cells. The cells were pretreated with control buffer, U75302 (1 ␮ ␮ ␮ LTB4 than control buffer. M), LTB4APA (0.1 M), or LY1718 (0.1 M) for 20 min before adding Previous reports have shown that the number of BLTs increases 100 nM LTB4 to the lower Transwell chamber. After 6 h the transmigrated during DMSO-induced differentiation of HL-60 cells (6, 20, 21). cells were stained with 0.4% trypan blue and counted. Data are expressed as Ϯ Consistent with those reports, RT-PCR showed little or no expres- the mean percentage of control SD from three independent experiments. sion of BLT1 or -2 in parent HL-60 cells, but their expression increased with maturation toward the neutrophil lineage; levels of whether ROS is similarly involved in the transepithelial migration BLT2, in particular, were dramatically increased during matura- of dHL-60 cells induced by LTB4, we examined the effect on LTB4- tion (Fig. 2A). mediated transepithelial migration of ROS inhibition by DPI, an in- With that in mind, we next tested whether the ability of dHL-60 hibitor of NADPH oxidase-like flavoenzymes, or by NAC, a ROS cells to migrate across lung epithelium was a specific result of the scavenger. We found that transmigration induced by 100 nM LTB4 increased expression of BLTs by examining the effects of selective was significantly diminished by pretreating dHL-60 cells with DPI BLT antagonists. As shown in Fig. 2B, pretreatment with either and NAC (Fig. 3A). In a similar fashion, pretreatment with DPI or U75302 (5), a selective BLT1 antagonist, or LTB4APA, a nonspe- NAC also diminished transmigration induced by PMA. To determine cific BLT antagonist, significantly diminished LTB4 (100 nM)- whether LTB4 in fact triggers intracellular ROS generation in dHL-60 induced transepithelial migration of dHL-60 cells, and in combi- cells, we assessed H2O2 levels using DCF. As shown in Fig. 3B, nation, they abolished the response. That LTB4APA exerted a LTB4 elicited a concentration-dependent increase in the level of ROS, more profound inhibitory effect than U75302 suggests that the ef- by guest on October 1, 2021. Copyright 2003 Pageant Media Ltd. and this effect was abolished by pretreating the cells with DPI or NAC fect was mediated primarily by BLT2. Consistent with that idea, (Fig. 3C). LTB4 showed biological activity at concentrations of 100Ð500 nM, which falls within the reported optimal range for BLT2 and is LTB4-induced transepithelial migration of dHL-60 cells requires 2 orders of magnitude higher than the optimal range for BLT1. On ERK activation the other hand, pretreatment with LY1718, a specific cysteinyl LT We also observed that in dHL-60 cells LTB4 (100 nM) induced antagonist, had no inhibitory effect (Fig. 2B). phosphorylation of ERK (Fig. 4A), but not p38 or c-Jun amino- ROS is critical for the transepithelial migration of dHL-60 cells terminal kinase (not shown). To further evaluate ERK stimulation by LTB4, ERK’s kinase activity was assessed using an Elk-lucif-

http://classic.jimmunol.org Recently, a ROS-dependent pathway was shown to mediate LTB4- erase trans-reporter system. We found that LTB4 (100 nM) clearly induced chemotaxis of Rat-2 ﬁbroblasts (22). To determine enhanced ERK/Elk-luciferase activity in differentiated dHL-60 cells, but not in wild-type HL-60 cells (Fig. 4B). Moreover, pretreating the cells with 10 or 20 ␮M PD98059, a speciﬁc mitogen- activated protein kinase kinase inhibitor that blocks ERK activation, dose-dependently inhibited the transepithelial migration

Downloaded from induced by 100 nM LTB4 (Fig. 4C). Pretreatment with genistein (10 ␮M), a nonspeciﬁc tyrosine kinase inhibitor, had a similar inhibitory effect. In a subsequent experiment aimed at determining whether ERK activation is situated upstream or downstream of ROS generation

in the LTB4 signaling pathway, we tested the effects of DPI and NAC on LTB4-induced ERK phosphorylation. As shown in Fig. 4D, pretreatment with either NAC or DPI virtually abolished FIGURE 1. LTB -induced migration of dHL-60 neutrophil-like granu- 4 LTB -induced ERK phosphorylation. Apparently, LTB signaling locytes across a monolayer of A549 human lung epithelial cells. Parental 4 4 to transmigration of dHL-60 cells is transduced via ERK, most and differentiated HL-60 cells were seeded into the upper chambers of a Transwell system to a density of 5 ϫ 105/well, and the indicated concen- likely acting downstream of ROS. We further investigated the mediators contributing to LTB4 sig- trations of LTB4 were added to the lower chambers. After 6 h the transmigrated cells were stained with 0.4% trypan blue and counted. The results naling by testing the effects of inhibitors of phosphatidylinositol shown are representative of three independent experiments with similar 3-kinase (PI 3-kinase), protein kinase C (PKC), and cytosolic

results. phospholipase A2 (cPLA2) on transepithelial migration and ERK 6276 LTB4-INDUCED TRANSEPITHELIAL MIGRATION VIA ROS-DEPENDENT PATHWAY

␮ FIGURE 3. Role of ROS in LTB4-induced transepithelial migration. A, dHL-60 cells were pretreated with 5 M DPI or 2 mM NAC for 20 min at room temperature before adding 100 nM LTB4 to the lower Transwell chamber. Transmigrated cells were counted, and the results are expressed as the mean Ϯ percentage of control SD. B, Serum-starved dHL-60 cells were incubated for 5 min with the indicated concentration of LTB4 or 20 ng/ml PMA, after which DCF fluorescence was quantified. Data are expressed as the mean percentage of control Ϯ SD from three independent experiments. C, Serum-starved ␮ dHL-60 cells were exposed to 100 nM LTB4 for 5 min in the presence or the absence of 5 M DPI or 2 mM NAC. Inhibitors were added 20 min before Ϯ the addition of LTB4. DCF fluorescence was quantified as described in Materials and Methods. Data are expressed as the mean percentage of control SD from three independent experiments.

activation. We found that LTB4-induced transmigration was com- obtained from peripheral blood of healthy volunteers. As in ␮ pletely blocked by pretreatment with 0.1 M wortmannin, a PI dHL-60 cells, LTB4 induced significant transepithelial migration 3-kinase inhibitor, or 0.1 ␮M GF109203X, a PKC inhibitor (Fig. of human PMNs, and the effect was significantly diminished by 5A), and there was a corresponding decrease in the level of ERK pretreatment with NAC, DPI, or PD98059 (Fig. 6). activation (Fig. 5B), which suggests that both PI 3-kinase and PKC Roles of ROS and ERK in migration of neutrophils across are situated upstream of ERK in the LTB4 signaling pathway. By ␮ airway epithelium in LTB4-challenged mice contrast, AACOCF3 (10 M), a specific cPLA2 inhibitor, had no effect on either transmigration or ERK activation (Fig. 5). To investigate the physiological function of ROS-ERK-linked signaling in vivo, BALB/c mice were subjected to intratracheal in- A ROS-ERK-linked cascade is essential for LTB4-induced stillation of LTB4. Analysis of BAL fluid revealed that very few transepithelial migration of PMNs neutrophils were present in the airways following intratracheal in- The functional importance of this ROS-ERK-linked pathway was stillation of saline (Fig. 7); mostly alveolar macrophages were de- ␮ ␮ further confirmed by investigating its presence in human PMNs tected. Following instillation of 1 gofLTB4 in 50 l of saline, by guest on October 1, 2021. Copyright 2003 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from

FIGURE 4. Role of ERK activation in LTB4-induced transepithelial migration. A, HL-60 (day 0) or dHL-60 cells (days 3 and 5) were treated with 100 nM LTB4 for 5 min, after which the lysates were immunoblotted with anti-phospho-ERK Ab. The blot was then stripped and reprobed with anti-total ERK Ab. The results shown are representative of at least three independent experiments. B, ERK kinase activity was measured using the Elk-luciferase trans-reporter system as described in Materials and Methods. HL-60 or dHL-60 cells were transiently cotransfected with pFR-Luc reporter plasmid and Ϯ then treated with 100 nM LTB4 or control buffer for 6 h. The lysates were assayed for Elk-luciferase activity. Data are expressed as the mean SD of control values from three independent experiments. C, Effect of ERK inhibition on transepithelial migration of dHL-60 cells. For the inhibitor experiments ␮ dHL-60 cells were pretreated for 20 min with control buffer, 10 M genistein, or the indicated concentration of PD98059 before adding 100 nM LTB4 to the lower Transwell chamber. After 6 h the transmigrated cells were stained with 0.4% trypan blue and counted. Data are expressed as the mean percentage Ϯ of control SD from three independent experiments. D, dHL-60 cells were treated with 100 nM LTB4 for 5 min in the presence or the absence of DPI (5 ␮M) or NAC (2 mM) and then lysed. The lysates were immunoblotted with anti-phospho-ERKs Ab, after which the blot was stripped and reprobed with anti-total ERKs Ab. The results shown are representative of at least three independent experiments. The Journal of Immunology 6277

FIGURE 5. Effect of inhibiting PI 3-kinase or PKC on LTB4 signaling and transepithelial migration. A, LTB4-induced transepithelial migration of dHL-60 cells was evaluated in the absence or the presence of control buffer, 0.1 ␮M wortmannin, 0.1 ␮M GF109203X, or 20 ␮M AACOCF3. Cells were

preincubated for 20 min with or without the indicated inhibitor before adding 100 nM LTB4 to the lower Transwell chamber. After 6 h the transmigrated cells were stained with 0.4% trypan blue and counted. Data are expressed as the mean percentage of control Ϯ SD from three independent experiments. ␮ ␮ ␮ B, dHL-60 cells were treated with 100 nM LTB4 for 5 min in the presence or the absence of 0.1 M wortmannin, 0.1 M GF109203X, or 20 M AACOCF3 and then lysed. The lysates were immunoblotted with anti-phospho-ERK Ab, after which the blot was stripped and reprobed with anti-total ERK Ab. The results shown are representative of at least three independent experiments.

however, the number of neutrophils in the BAL ﬂuid increased (27Ð30). In the context of the present study it is notable that a

signiﬁcantly within 6 h (Fig. 7, A and B). This effect was blocked ROS-linked cascade was recently shown to mediate LTB4-induced by prior i.p. administration of NAC, DPI, PD98059, or the BLT chemotaxis and proliferation of Rat-2 ﬁbroblasts (22), that oxidant

inhibitor CP105696 (Fig. 7C), suggesting that in vivo a ROS- stress generated by LTB4 reportedly induces transendothelial mi- ERK-linked signaling pathway governs LTB4-mediated neutrophil gration of monocytes (31, 32), and that ROS reportedly contribute trafﬁcking into the airway. to integrin-dependent cell adhesion (33, 34). Given the inhibitory

effect of DPI on LTB4-induced cell migration and ERK activation, Discussion we suggest that NADPH oxidase is the primary source of the ROS

The epithelial cells that form a barrier lining the airway of the lung generated in response to LTB4. Consistent with that idea, several

are key regulators of neutrophil trafﬁcking into the airway lumen. studies have shown that LTB4 promotes robust, receptor-mediated In that regard, LTB4 is known to be among the most potent of activation of NADPH oxidase (35Ð37), which in eosinophils mediates

endogenous chemoattractants of PMNs (along with IL-8 and C5a) activation of PKC, phospholipases C and D, and cPLA2 (35, 38). and to stimulate adhesion of PMNs to airway epithelial cells and Neutrophils function in the front line of cellular host defense transmigration. However, despite its role as a major mediator of by guest on October 1, 2021. Copyright 2003 Pageant Media Ltd. against microorganisms. This function relies in part on their ability Ϫ PMN transepithelial migration and neutrophil-induced lung injury, to generate large amounts of O2 and related ROS through acti- little is known about the LTB4 signaling cascade involved in these vation of NADPH oxidase on the plasma membrane, a phenome- processes. The major observation of the present study is that a non known as respiratory bursting (14, 36, 39). Thus, ROS gen-

ROS-ERK-linked signaling pathway mediates LTB4-induced eration by NADPH oxidase appears to have a dual function, host transepithelial migration of neutrophils both in vitro and in vivo. defense and intracellular signaling, although it remains unknown A growing body of evidence suggests that oxygen-derived rad- how the two functions are respectively regulated. The fact that the Ϫ icals (e.g., O2 and H2O2) serve as intracellular signaling mole- level of ROS generated by LTB4 is substantially lower than those ␣ cules for a variety of agonists, including LTB4, TNF- (16, 23, seen during respiratory bursts suggests that the regulatory mech- ␤ http://classic.jimmunol.org 24), IL-1 (25), and TGF- 1 (27), as well as various growth factors anisms are quite different. We were able to demonstrate that ERK mediates ROS signaling to transepithelial migration of neutrophils both in vitro and in vivo (Figs. 4 and 7). In vivo, the process of transepithelial neutrophil migration occurs through a series of steps orchestrated by the interaction of adhesion molecules on the surface of the neutrophils

Downloaded from and proteins secreted by the bronchial epithelial cells. Inhibition of ROS or ERK may interfere with any one or more of these steps. Recently, we and others reported that PI 3-kinase acts as an upstream mediator of ROS generation, leading to chemotaxis and ERK activation (40, 41). Herein we showed that pretreating dHL-60 cells with wortmannin or GF109203X, inhibitors of PI

3-kinase and PKC, respectively, diminished both LTB4-induced transmigration and ERK phosphorylation, suggesting that PI 3-kinase and PKC act upstream of ERK to regulate LTB signaling. FIGURE 6. Roles of ROS and ERK in the LTB -induced transepithelial 4 4 Consistent with that idea, PMA, an activator of several PKC iso- migration of human PMNs. Cells were pretreated for 20 min with control forms, stimulated transepithelial migration of dHL-60 cells, and buffer, 5 ␮M DPI, 2 mM NAC, or 10 ␮M PD98059 before adding 100 nM the effect was signiﬁcantly diminished by pretreatment with LTB4 to the lower Transwell chamber. After 6 h the transmigrated cells were stained with 0.4% trypan blue and counted. Data are expressed as the antioxidants. mean percentage of control Ϯ SD from three independent experiments ROS generation in response to external stimuli has been related (p Ͻ 0.01). to the activation of transcription factors AP-1 and NF-␬B (34), 6278 LTB4-INDUCED TRANSEPITHELIAL MIGRATION VIA ROS-DEPENDENT PATHWAY

FIGURE 7. Effects of ROS and ERK inhibitors on

LTB4-induced neutrophil trafficking into the lung airway in vivo. BALB/c mice were challenged with in- ␮ ␮ tratracheal administration of 1 gofLTB4 in 50 lof saline. Six hours later, samples of lung tissue and BAL fluids were collected. The lung samples were sec- tioned and stained with H&E, and the total and differential cellular components present in the BAL fluid were assessed. Representative H&E-stained lung sections (A) and BAL fluid smears (B) are shown. Each sample was assayed in duplicate, and the assays were repeated three times. Neutrophils in the BAL fluid were counted (C). In some experiments 10 ␮g/20 g body weight of the indicated inhibitor was adminis- trated i.p. 2 h before intratracheal administration of

LTB4. Data are expressed as the mean percentage of control Ϯ SD (p Ͻ 0.01).

which appear to be involved in the stimulation of the integrins 7. Yokomizo, T, K. Kato, H Hagiya, T. Izumi, and T. Shimizu. 2000. Hydroxyei- LFA-1 (CD11a/CD18) and Mac-1 (CD11b/CD18). Although neu- cosanoids bind to and activate the low affinity leukotriene B4 receptor, BLT2. J. Biol. Chem. 276:12454. trophil adherence and subsequent migration out of postcapillary 8. Kidney, J. C., and D. Proud. 2000. Neutrophil transmigration across human air- venules during systemic inflammation are largely dependent on way epithelial monolayers. Am. J. Respir. Cell Mol. Biol. 23:389. 9. Sibille, Y., and H. Y. Reynolds. 1990. Macrophages and polymorphonuclear CD18, the existence of CD18-independent neutrophil migration neutrophils in lung defense and injury. Am. Rev. Respir. Dis. 141:471. has also been demonstrated (42Ð45). For example, when Doers- 10. Sur, S., T. B. Crotty, G. M. Kephart, B. A. Hyma, T. V. Colby, C. E. Reed, chuk and co-workers (43) compared neutrophil migration in the L. W. Hunt, and G. J. Gleich. 1993. Sudden-onset fatal asthma: a distinct entry with few eosinophils and relatively more neutrophils in the airway submucosa? systemic and pulmonary circulations of rabbits exposed to HCl, Am. Rev. Respir. Dis. 148:713. Streptococcus pneumonia, E. coli endotoxin, or PMA, they found 11. Wenzel, S. E., S. J. Szefler, D. Y. M. Leung, S. I. Sloan, M. D. Rex, and that migration into the abdominal wall was always CD18 depen- R. J. Martin. 1997. Bronchoscopic evaluation of severe asthma: persistent inflammation associated with high dose corticosteroids. Am. J. Respir. Crit. Care dent, but migration into the lung was not. Similarly, analysis of Med. 156:737.

by guest on October 1, 2021. Copyright 2003 Pageant Media Ltd. neutrophil migration into the inﬂamed joints of rats with adjuvant 12. Hauert, A. B., S. Martinelli, C. Marone, and V. Niggli. 2002. Differentiated HL-60 cells are a valid model system for the analysis of human neutrophils arthritis showed that migration could only be completely blocked migration and chemotaxis. Int. J. Biochem. Cell. Biol. 34:838. ␤ by combining an anti-CD18 Ab with an Ab recognizing 1 very 13. Falanga, A., R. Consonni, M. Marchetti, G. Locatelli, E. Garattini, C. Gambacorti late Ag 4 (CD49d/CD29). Likewise, monocytes are capable of Passenrini, S. G. Gordon, and T. Barbui. 1998. Cancer procoagulant and tissue factor are differently modulated by all-trans-retinoic acid in acute promyelocytic CD18-independent migration mediated in part by members of the leukemia cells. Blood 92:143. ␤ 1 integrin family, very late Ag 4 in particular (44, 45). In addition 14. Hua, J., T. Hasebe, A. Someya, S. Nakamura, K. Sugimoto, and I. Nagaoka. to integrins, chemokines and proteases, such as elastase, would 2000. Evaluation of the expression of NADPH oxidase components during maturation of HL-60 cells to neutrophils lineage. J. Leukocyte Biol. 68:216. also seem to be likely downstream targets of ROS-ERK-linked 15. Yokomizo, T., T. Izumi, and T. Shimizu. 2001. Co-expression of two LTB4 signaling. With further study we anticipate complete clariﬁcation receptors in human mononuclear cells. Life Sci. 68:2207.

http://classic.jimmunol.org 16. Woo, C. H., Y. W. Eom, M. H. Yoo, H. J. You, H. J. Han, W. K. Song, Y. J. Yoo, of the molecular linkage between the ROS-ERK-linked pathway C. S. Chun, and J. H. Kim. 2000. Tumor necrosis factor-␣ generates reactive and its various potential targets. Considering a critical role of oxygen species via a cytosolic phospholipase A2-linked cascade. J. Biol. Chem. LTB as a major mediator of PMN transepithelial migration and 275:32357. 4 17. Jeong, D. W., M. H. Yoo, T. S. Kim, J. H. Kim, and I. Y. Kim. 2002. Protection neutrophil-induced lung injury, specific blockade of the ROS- of mice from allergen-induced asthma by selenite: prevention of eosinophil in- ERK-linked pathway may have therapeutic value in the treatment filtration by inhibition of NF-␬B activation. J. Biol. Chem. 277:17871. of inflammatory diseases such as acute bronchitis, ARDS, COPD, 18. Yang, L., L. Cohn, D. H. Zhang, R. Homer, A. Ray, and P. Ray. 1998. Essential role of nuclear factor ␬B in the induction of eosinophilia in allergic airway in- Downloaded from and severe asthma. flammation. J. Exp. Med. 188:1739. 19. Collins, S. J. 1987. The HL-60 promyelocytic leukemia cell line: proliferation, differentiation, and cellular oncogene expression. Blood 70:1233. References 20. Patry, C., E. Muller, J. Laporte, M. Rola-Pleszczynski, P. Sirois, and 1. Brock, T. G., R. W. McNish, and M. Peters-Golden. 1995. Translocation and A. J. de Brum-Fernandes. 1996. Leukotriene receptors in HL-60 cells differen- leukotriene synthetic capacity of nuclear 5-lipoxygenase in rat basophilic leuke- tiated into eosinophils, monocytes and neutrophils. Prostaglandins Leukotrienes mia cells and alveolar macrophages. J. Biol. Chem. 270:21652. Essent. Fatty Acids 54:361. 2. Yokomizo, T, T. Izumi, and T. Shimizu. 2001. Leukotriene B4: metabolism and 21. Pettersson, A., A. Boketoft, A. Sabirsh, N. E. Nilsson, K. Kotarsky, B. Olde, and signal transduction. Arch. Biochem. Biophys. 385:231. C. Owman. 2000. First-generation monoclonal antibodies identifying the human 3. Palmer, R. M., R. J. Stepney, G. A. Higgs, and K. E. Eakins. 1980. Chemokinetic leukotriene B4 receptor-1. Biochem. Biophys. Res. Commun. 279:520. activity of arachidonic and lipoxygenase products on leukocytes of different spe- 22. Woo, C. H., H. J. Cho, Y. W. Eom, J. S. Chun, Y. J. Yoo, and J. H. Kim. 2002. cies. Prostaglandins 20:411. Leukotriene B4 stimulate Rac-ERK cascade to generate reactive oxygen species 4. Busse, W. W. 1998. Leukotrienes and inflammation. Am. J. Respir. Crit. Care. that mediates chemotaxis. J. Biol. Chem. 277:8572. Med. 157:S210. 23. Lo, Y. Y. and T. F. Cruz. 1995. Reactive oxygen species mediate cytokine ac- 5. Yokomizo, T., K. Kato, K. Terawaki, T. Izumi, and T. Shimizu. 2001. A second tivation of c-jun N-terminal kinases. J. Biol. Chem. 270:11727. leukotriene B4 receptor, BLT2: a new therapeutic target in inflammation and 24. Li, Y., A. Ferrante, A. Poulos, and D. P. Harvey. 1996. Neutrophil oxygen radical immulogical disorders. J. Exp. Med. 192:421. generation: synergistic responses to tumor necrosis factor and mono/polyunsat- 6. Yokomizo, T., T. Izumi, K. Chang, Y. Takuwa, and T. Shimizu. 1997. A G- urated fatty acids. J. Clin. Invest. 97:1605. protein-coupled receptor for leukotriene B4 that mediates chemotaxis. Nature 25. Meier, B., H. H. Radeke, S. Selle, G. G. Habermehl, K. Resch, and H. Sies. 1990. 387:620. Human fibroblasts release low amounts of reactive oxygen species in response to The Journal of Immunology 6279

the potent phagocyte stimulants, serum-treated zymosan, N-formyl-methionyl- 36. Subramanian, N. 1992. Leukotriene B4 induced steady state calcium rise and leucyl-phenylalanine, leukotriene B4 or 12-O-tetradecanoylphorbol 13-acetate. superoxide anion generation in guinea pig eosinophils are not related events. Biol. Chem. Hoppe-Seyler 371:1021. Biochem. Biophys. Res. Commun. 187:670. 26. Ohba, M., M. Shibanuma, T. Kuroki, and K. Nose. 1994. Production of hydrogen 37. Lindsay, M. A., E. B. Haddad, J. Rousell, M. M. Teixeira, P. G. Hellewell, peroxide by transforming growth factor-␤1 and its involvement in induction of P. J. Barnes, and M. A. Giembycz. 1998. Role of the mitogen-activated protein egr-1 in mouse osteoblastic cells. J. Cell Biol. 126:1079. kinases and tyrosine kinases during leukotriene B4-induced eosinophil activation. 27. Sundaresan, M., Z. X. Yu, V. J. Ferrans, K. Irani, and T. Finkel. 1995. Require- J. Leukocyte Biol. 64:555. 38. Lindsay, M. A., R. S. Perkins, P. J. Barnes, and M. A. Giembycz. 1997. Leuko- ment for generation of H2O2 for platelet-derived growth factor signal transduction. Science 270:296. triene B4 activates the NADPH oxidase in eosinophils by a pertussis toxin-sensitive mechanism that is largely independent of arachidonic acid mobilization. 28. Bae, Y. S., S. W. Kang, M. S. Seo, I. C. Baines, E.Tekle, P. B. Chock, and J. Immunol. 160:4526. S. G. Rhee. 1997. Epidermal growth factor (EGF)-induced generation of hydro- 39. Dang, P. M., A. R. Cross, and B. M. Babior. 2001. Assembly of the neutrophil gen peroxide: role in EGF receptor-mediated tyrosine phosphorylation. J. Biol. respiratory burst oxidase: direct interaction between p67PHOX and cytochrome Chem. 272:217. b558. Proc. Natl. Acad. Sci. USA 13:98. 29. Krieger-Brauer, H. I., and H. Kather. 1995. Antagonistic effects of different mem- 40. Kim, B. C., M. N. Lee, J. Y. Kim, S. S. Lee, J. D. Chang, S. S. Lee, S. Y. Lee, bers of the fibroblast and platelet-derived growth factor families in adipose con- and J. H. Kim. 1999. Roles of phosphatidylinositol 3-kinase and Rac in the version and NADPH-dependent H2O2 generation in 3T3 L1-cells. Biochem. J. nuclear signaling by tumor necrosis factor-␣ in rat-2 fibroblasts. J. Biol. Chem. 307:549. 274:24372. 30. Rhee, S. G. 1999. Redox signaling: hydrogen peroxide as intracellular messenger. 41. Bianca, V. D., S. Dusi, E. Bianchini, I. D. Pra, and F. Rossi. 1999. ␤-Amyloid Ϫ Exp. Mol. Med. 31:53. activates the O2 forming NADPH oxidase in microglia, monocytes, and neu- 31. Rattan, V., C. Sultana, Y. Shen, and V. K. Kalra. 1997. Oxidant stress-induced trophils: a possible inflammatory mechanism of neuronal damage in Alzheimer’s transendothelial migration of monocytes is linked to phosphorylation of PE- disease. J. Biol. Chem. 274:15493. CAM-1. Am. J. Physiol. 273:E453. 42. Mackarel, A. J., K. J. Russell, C. S. Brady, M. X. FitzGerald, and C. M. 32. Sultana, C., Y. Shen, V. Rattan, C. Johnson, and V. K. Kalra. 1998. Interaction O’Connor. 2000. Interleukin-8 and leukotriene-B4, but not formylmethionyl of sickle erythrocytes with endothelial cells in the presence of endothelial cell leucylphenylalanine, stimulate CD18-independent migration of neutrophils conditioned medium induces oxidant stress leading to transendothelial migration across human pulmonary endothelial cells in vitro. Am. J. Respir. Cell. Mol. Biol. of monocytes. Blood 92:3924. 23:154. 43. Doerschuk, C. M., R. K. Winn, H. O. Coxson, and J. M. Harlan. 1990. CD18- 33. Blouin, E., L. Halbwachs-Mecarelli, and P. Rieu. 1999. Redox regulation of ␤ dependent or -independent mechanisms of neutrophil adherence in the pulmonary 2-integrin CD11b/CD18 activation. Eur. J. Immunol. 29:3419. and systemic microvasculature of rabbits. J. Immunol. 114:2327. 34. Guo, R.-F., and P. A. Ward. 2002. Mediators and regulation of neutrophils ac- ␤ ␤ 44. Shang, X. Z., and A. C. Issekutz. 1997. 2 (CD18) and 1 (CD29) integrin cumulation in inflammatory responses in lung: insights from the IgG immune mechanisms in migration of human polymorphonuclear leucocytes and mono- complex model. Free Radical Biol. Med. 33:303. cytes through lung fibroblast barriers: shared and distinct mechanisms. Immunol- 35. Perkins, R. S., M. A. Lindsay, P. J. Barnes, and M. A. Giembycz. 1995. Early ogy 92:527. signaling events implicated in leukotriene B4-induced activation of NADPH ox- 45. Li, X. C., M. Miyasaka, and T. B. Issekutz. 1998. Blood monocyte migration to idase in eosinophils: role of Ca2ϩ, protein kinase C and phospholipase C and D. acute lung inflammation involves both CD18 and very late activation antigen-4- Biochem. J. 310:795. dependent and independent pathways. J. Immunol. 161:6258. by guest on October 1, 2021. Copyright 2003 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from