|Home | About | Journals | Submit | Contact Us | Français|
Asthma is characterized by appearance of eosinophils in the airway. Eosinophils purified from the airway 48 h after segmental antigen challenge are described as exhibiting greater adhesion to albumin-coated surfaces via an unidentified β2 integrin and increased expression of αMβ2 (CD11b/18) compared with purified blood eosinophils. We have investigated the determinants of this hyperadhesive phenotype. Airway eosinophils exhibited increased reactivity with the CBRM1/5 anti-αM activation-sensitive antibody as well as enhanced adhesion to VCAM-1 (CD106) and diverse ligands, including albumin, ICAM-1 (CD54), fibrinogen, and vitronectin. Purified blood eosinophils did not adhere to the latter diverse ligands. Enhanced adhesion of airway eosinophils was blocked by anti-αMβ2. Podosomes, structures implicated in cell movement and proteolysis of matrix proteins, were larger and more common on airway eosinophils adherent to VCAM-1 when compared with blood eosinophils. Incubation of blood eosinophils with IL-5 replicated the phenotype of airway eosinophils. That is, IL-5 enhanced recognition of αM by CBRM1/5; stimulated αMβ2-mediated adhesion to VCAM-1, albumin, ICAM-1, fibrinogen, and vitronectin; and increased podosome formation on VCAM-1. Thus, the hyperadhesion of airway eosinophils after antigen challenge is mediated by upregulated and activated αMβ2.
Asthma is an inflammatory syndrome of the airway accompanied by recurrent episodes of airway obstruction, mucus hypersecretion, and bronchial smooth muscle contraction (1). The prevalence of asthma worldwide is increasing, but the causes of the increase remain unclear (2–4). In individuals with allergic asthma, recruitment of eosinophils (EOS) into the airway lumen is believed to exacerbate asthmatic symptoms and lead to the chronic character of asthma (5, 6).
How EOS traffic to the airway is incompletely understood and of considerable importance to asthma. Heterodimeric integrin adhesion receptors, which are among the most functionally versatile of known cellular receptors, have generated particular interest as determinants of how EOS are recruited to the asthmatic airway (7, 8). Purified blood EOS express seven integrin heterodimers: α4β1 (CD49d/29), α6β1 (CD49f/29), αMβ2 (CD11b/18), αLβ2 (CD11a/18), αXβ2 (CD11c/18), αDβ2 (αD/18), and α4β7 (CD49d/β7) (9–11). Each kind of heterodimer may adopt several conformational states that regulate integrin-mediated adhesion and movement (12, 13). Integrins are thus believed to mediate rolling and arrest of EOS on endothelium, migration through endothelium and the underlying basement membrane in response to chemotactic stimuli, and traversing of bronchial epithelium into the airway lumen (14, 15).
EOS obtained from bronchoalveolar lavage (BAL) fluid after segmental antigen challenge represent cells that have transmigrated from blood to airway in response to inflammatory mediators. These EOS exhibit elevated adhesion to albumin mediated via an unidentified β2 integrin family member and enhanced migration compared with blood EOS (16). We have done further analyses of airway EOS to identify the β2 integrin responsible for the hyperadhesive phenotype and learn whether increased migration is associated with increased podosomes, transitory cell adhesive contacts that are found in migratory cells and recently have been demonstrated on cytokine-stimulated blood EOS (17). Our investigations led to the important findings that human eosinophils recruited to the airway in response to segmental antigen challenge express an allosterically active form of αMβ2 concomitant with enhanced αMβ2-mediated adhesion to diverse ligands and increased formation of podosomes. Treatment of blood EOS with IL-5 replicates the phenotype of airway EOS recovered after segmental antigen challenge.
Samples were from blood of unchallenged subjects or from blood or BAL fluid 48 h after segmental bronchoprovocation with antigen in challenge of normal subjects, subjects with allergic rhinitis, or subjects with allergic asthma. The sex, age, antigen type, antigen dose, and percentage of EOS in BAL fluid for each subject are listed in Table E1 in the online supplement. Antigen doses constituting 10–15% of the dose that provoked a 20% fall (Ag PD20) in forced expiratory volume (FEV1) as calculated from a dose–response curve were used for antigen challenges. Cat dander and house dust mite extract were from Hollister Stier (Spokane, WA), and ragweed extract was from Greer (Lenoir, NC). All extracts passed the Food and Drug Administration live animal safety testing in which animals must live and gain weight for at least 2 wk after antigen injection. Bronchoscopy and segmental bronchoprovocation were performed as previously described (18). Human EOS were isolated from peripheral blood by anti-CD16 magnetic bead selection (19, 20). For purification of BAL EOS, BAL fluid collected 48 h after segmental bronchoprovocation was layered over a double Percoll gradient of 1.085/1.100 g/ml density, and the cells at the interface were collected (21). The purity of eosinophils was at least 98% as determined by Diff-Quik staining. Viability was at least 99% as assessed by staining with propidium iodide and annexin V–FITC (BD Biosciences, San Jose, CA). Endotoxin in BAL fluid tested from pre- and post-challenge samples of 13 subjects was either undetectable or < 0.25 endotoxin units (EU)/ml according to the Limulus Amebocyte Lysate assay from BioWhittaker (Walkersville, MD). There were no systematic differences in endotoxin in BAL fluid obtained before and after antigen challenge. The University of Wisconsin Human Subjects Committee approved the study, and informed consent was obtained before participation of volunteers.
Eosinophilic leukemic cell lines, EoL-3 and AML14.3D10, were cultured as described previously (17). Jurkat T lymphocytic cells were provided by Laura Kiessling (University of Wisconsin–Madison, Madison, WI) and grown in RPMI 1640 with 10% FBS, 10 mM HEPES, 1 mM sodium pyruvate, 100 U/ml penicillin, and 100 μg/ml streptomycin and subcultured by dilution to 1 × 105/ml twice a week. Human fibroblasts (17), and human umbilical vein endothelial cells (22) were cultured as described.
The seven-module form of soluble vascular cell adhesion molecule-1 (VCAM-1) was produced in High 5 insect cells transfected with recombinant baculoviruses. The secreted protein contained a C-terminal histidine tag and was purified from cell medium by Ni-NTA beads (Qiagen, Valencia, CA) as described (17). Recombinant soluble human intercellular adhesion molecule-1 (ICAM-1) and human plasma fibrinogen were purchased from R&D Systems (Minneapolis, MN) and Sigma-Aldrich (St. Louis, MO), respectively. Human plasma fibronectin (FN) was purified in the absence of urea or other denaturant as described (23). Rat collagen type I was from Upstate Biotechnology (Lake Placid, NY), human vitronectin was purified in its native form as described (24), and entactin-free mouse laminin was purchased from Collaborative Biomedical Products (Bedford, MA). Bovine serum albumin (BSA) rendered free of fatty acids was purchased from Sigma-Aldrich. The purity and molecular weight of each ligand was verified by SDS-PAGE as shown in Figure E1.
Sources of the following mAbs have been described previously: anti-α4 mAb HP2/1 (mouse IgG1), anti-α4 mAb 9F10 (mouse IgG1), anti-αD mAb 240I (mouse IgG1), anti-β2 mAb TS1/18 (mouse IgG1), anti-β7 mAb Fib504 (rat IgG2a), isotype control mouse IgG1 (anti-keyhole limpet hemocyanin, clone A112–2), isotype control rat IgG2a (anti-keyhole limpet hemocyanin, clone A110–2), FITC-conjugated goat anti-mouse, anti-gelsolin mAb GS-2C4 (mouse IgG1), and rhodamine-phalloidin (17). The following mAbs were purchased from BD Biosciences: activation-specific mAbs against β1, HUTS-21 (mouse IgG1) and 9EG7 (mouse IgG1); anti-β1 mAb MAR4 (mouse IgG1), and mouse IgG2a (anti-keyhole limpet hemocyanin, clone G155–178). The latter mAb was used as an isotype control in adhesion studies. Anti-αL clone 38 (mouse IgG2a) and anti-αX clone 3.9 (mouse IgG1) were from Calbiochem (San Diego, CA). Anti-αM mAb 2LPM19c (mouse IgG1) was from Biomeda (Foster City, CA). Anti-β1 4B4 (mouse IgG1) was from Coulter (Miami, FL). The N29 activation-specific mAb against β1 (mouse IgG1) was from Chemicon (Temecula, CA). The mAb recognizing the activated form of αMβ2, CBRM1/5 (mouse IgG1), was purchased from BioLegend (San Diego, CA). Anti-αD mAb 217I (mouse IgG1) was a gift from ICOS (Bothell, WA). Eotaxin and RANTES were from Cell Sciences (Canton, MA).
Polystyrene 96-well non–tissue culture treated plates (Becton Dickinson, Franklin Lakes, NJ) were coated in triplicate with 100 μl per well of 10 μg/ml protein solution, unless otherwise noted, in TBS, pH 8.0, for 3 h at 37°C and then blocked with heat-inactivated FBS for 20 min at 37°C. The protein coating amount was determined to support maximum adhesion of fibroblasts or airway EOS. Endotoxin in wells coated with proteins was either undetectable or < 0.06 EU/well according to the Limulus Amebocyte Lysate assay (BioWhittaker). FBS was heat-inactivated by heating to 56°C for 30 min, mixed every 10 min, and then rapidly cooled in an ice water bath. For inhibition experiments, mAbs were added to cells at 10 μg/ml final concentration and incubated 5 min at room temperature before cell addition to wells. Cells (1 × 104 per well) were added to wells in volumes of 100 μl/well and allowed to adhere for 1 h at 37°C, then washed three times with TBS, pH 8.0. For quantitation of adherent EOS, 100 μl HBSS was added to wells with 100 μl of solution containing 0.1% Triton-X100, 1 mM o-phenylenediamine dihydrochloride (OPD; Sigma-Aldrich), 1 mM H2O2 and 55 mM Tris, pH 8.0 (25). After 30 min of color development at room temperature, 50 μl of 4 M H2SO4 was added to quench the reaction. Absorbances were read at 490 nm in an EL microplate reader (Bio-Tek Instruments, Winooski, VT). Percentages of adherent EOS were calculated by dividing the absorbance per experimental well by the absorbance from uncoated wells containing 100% of input EOS and multiplying by 100. Absorbance and signal values were within the linear range of the detector as determined by a standard curve. For inhibition studies, values are given as percentages of the absorbances obtained in the presence of respective isotype control mAbs.
EOS (1–5 × 105) were incubated in 0.25 ml HBSS along with primary mAb or isotype control for 30 min at 4°C, washed with 0.25 ml HBSS, resuspended in 0.25 ml HBSS, and then incubated with secondary mAb for 30 min at 4°C in the dark. Primary mAbs were added in final amounts of 0.5 μg. FITC-conjugated goat anti-mouse IgG was added at final concentrations of 20 μg/ml. After secondary mAb incubation, cell lines were pelleted and fixed by resuspension in 1% paraformaldehyde, 67.5 mM sodium cacodylate, 113 mM NaCl, pH 7.2, stored at 4°C in the dark, and within 1 wk washed with RPMI buffer and analyzed. EOS were always analyzed after fixation. Fluorescence measurements were collected within 1 h on a FACSCalibur (BD Biosciences; available through the Flow Cytometry Facility, Comprehensive Cancer Center, University of Wisconsin, Madison). Data were collected from 10,000 cells per condition and analyzed using the CellQuest software (BD Biosciences) or FlowJo software (Tree Star, Inc., Ashland, OR). Fixed cells were gated based on forward and side scatter. The specific geometric mean channel fluorescence (gMCF) is reported. The gMCF was obtained by subtracting the gMCF of the isotype control from the gMCF of the experimental sample according to the following equation, where gMF = geometric mean fluorescence: gMCF = [1,024 × (log gMF of experimental sample)] – [1,024 × (log gMF of isotype control sample)]. The percentages of EOS from different donors that reacted with mAb CBRM1/5 compared with isotype control mAb was calculated from histogram plots using the Overton cumulative histogram subtraction algorithm in FlowJo, which subtracts histograms on a channel-by-channel basis to provide a percent of positive cells (26).
Immunofluorescence was performed as previously described (17).
Results were analyzed by one-way or repeated measures ANOVA with Dunnett's or Tukey-Kramer multiple comparison post test or two-tailed t test on GraphPad Prism software (San Diego, CA). Results are the mean and SD or SEM of assays involving several different donors and multiple replicates as noted in the tables and figure legends.
It has been shown previously that airway EOS purified from antigen-challenged subjects exhibit elevated adhesion to surface-coated albumin via an unidentified β2 integrin (16). We hypothesized that airway EOS exhibit increased β2-dependent adhesion to diverse ligands expressed on or within airway endothelium and basement membrane. Blood EOS purified either before or after antigen challenge did not adhere specifically to albumin, ICAM-1, fibrinogen, fibronectin, laminin, collagen type I, or vitronectin (Figures 1A and 1B). In contrast, airway EOS purified from subjects after segmental antigen challenge with three different antigens (cat dander, ragweed, or house dust mite; Table E1) adhered to albumin, ICAM-1, fibrinogen, or vitronectin (Figure 1C). Both blood and airway EOS adhered to the seven-module form of soluble VCAM-1; adhesion of airway EOS was 1.6-fold greater (P < 0.05) and required a lesser coating of VCAM-1 (Figure 1D). The percentage of airway EOS that adhered to albumin, in contrast, was 6-fold greater than the percent adhesion of blood EOS to albumin. Airway EOS, like blood EOS, did not adhere specifically to fibronectin, laminin, or collagen type I (Figure 1), to which fibroblasts or endothelial cells adhered readily (not shown). Thus, airway EOS adhered specifically to a surprising spectrum of adhesive ligands and exhibited an adhesive profile not shared by blood EOS.
The adhesion of airway EOS to ICAM-1, fibrinogen, albumin, or vitronectin, all of which are potential targets of αMβ2 (27–29), prompted us to determine whether αMβ2 was involved in the adhesion to all the proteins. Antibody-blocking experiments identified αMβ2 as the integrin responsible for the specific adhesion of airway EOS to the four recognized αMβ2 ligands (Figure 2). That is, inhibitory mAbs recognizing αM and β2 subunits blocked adhesion of airway EOS to all four ligands, whereas inhibitory mAbs recognizing β1, αL, or αX did not inhibit adhesion (Figure 2). In experiments not shown, EoL-3 and AML14.3D10 eosinophilic leukemic cells, which do not express αMβ2 but do express αLβ2, did not adhere to albumin, ICAM-1, fibrinogen, or vitronectin, consistent with the conclusion that adhesion of airway EOS to these ligands is mediated by αMβ2. Adhesion of blood (not shown) and airway (Figure 3) EOS to VCAM-1 was mostly blocked by anti-α4 or anti-β1. In addition, there was partial inhibition of adhesion of airway EOS to VCAM-1 with an antibody to αM (Figure 3). These results identify αMβ2 as the integrin receptor that mediates the hyperadhesion of airway EOS to diverse ligands, including known αMβ2-reactive ligands—ICAM-1, fibrinogen, albumin, and vitronectin—and one that is not an already recognized αMβ2-ligand— VCAM-1.
The conformation-sensitive CBRM1/5 mAb, which recognizes an epitope in the ligand-binding “insert” (I) domain of the αM subunit (30), reacted 3- to 4-fold higher with airway EOS purified from antigen-challenged subjects compared with blood EOS purified before or after challenge (Figure 4 and Table 1). These results indicate that antigen challenge allosterically activates αMβ2 on airway EOS, consistent with the notion that the enhanced adhesion of airway EOS to diverse integrin ligands is mediated by αMβ2.
We assayed effects of segmental antigen challenge on the allosteric structure and expression level of a second integrin heterodimer, α4β1, an important adhesion receptor involved in recognition by EOS of VCAM-1 (9, 16, 31). We probed β1 conformation with three conformation-sensitive mAbs: N29, HUTS-21, and 9EG7 (32–35). The locations of these epitopes in various allosteric conformations assumed by β1, based on the αVβ3 and αIIbβ3 structural models, suggest that the mAbs recognize increasingly activated forms of β1 in the order N29 < HUTS-21 < 9EG7 (12, 36). HUTS-21 (Figure 3B and Table 1) and 9EG7 (Table 1) failed to recognize β1, and N29 exhibited low or undetectable recognition of purified blood or airway EOS (Table 1). All three mAbs reacted robustly with β1 on Jurkat cells (not shown). Thus, in contrast to αMβ2, the structure of α4β1 on purified EOS is not detectably altered in response to antigen challenge. Indeed, airway EOS expressed less total α4 subunit compared with purified blood EOS (Table E2) as reported previously (37), and β1 and β7 subunits were similarly decreased on purified airway EOS (Table E2).
Treatment of blood EOS with IL-5, especially when TNF-α is present, enhances the number and size of podosomes that form when adherent to VCAM-1 (17). Podosomes are transient assemblies of cytoskeletal and membrane proteins that are believed to be important in cell movement and localization of proteolytic machinery (38, 39). We analyzed whether airway EOS purified from antigen-challenged subjects, which are more migratory than unchallenged blood EOS (16), demonstrate enhanced podosome formation compared with blood EOS. Actin and gelsolin, markers of podosomes, were localized to punctate structures that were coarser in challenged airway EOS compared with blood EOS from unchallenged subjects (Figure 5A). Blood EOS purified after challenge exhibited podosomes that were variable in size and number compared with challenged airway EOS and unchallenged blood EOS. Gelsolin co-localized with actin in airway EOS adherent to VCAM-1 (not shown). By virtue of the punctate staining and co-localization by gelsolin and actin (38), the structures have the characteristics of podosomes. The percentage of airway EOS that contained podosomes was ~ 2- to 3-fold greater than the percentage of blood EOS that contained the podosome structure (Figure 5B). Podosome formation was also detected in airway EOS adherent to albumin, ICAM-1, fibrinogen, and vitronectin (not shown). There was no detectable change in β1 distribution of airway or blood EOS adherent to VCAM-1 in immunofluorescent analyses, consistent with the notion that α4β1 activity and distribution are not altered on purified EOS in response to segmental antigen challenge. These results indicate that EOS recruited to the airway in response to antigen challenge exhibit changes in the cytoskeleton compared with EOS in blood. That is, airway EOS display increases in formation of podosomes, which may partly explain the enhanced mobility of airway EOS in comparison to blood EOS (16).
IL-5 is a critical cytokine expressed in individuals with asthma and is important in EOS recruitment to airway. Specifically, IL-5 expression is increased in the airway mucosa after segmental antigen challenge and is associated with airway eosinophilia (18, 40). IL-5 induces enhanced adhesion of EOS to endothelial cells (41) and albumin (9, 16, 42) and formation of podosomes (17). We, therefore, determined whether IL-5 treatment of purified blood EOS mimicked the enhanced activity of αMβ2 and prominent podosomes of airway EOS. Blood EOS obtained from either normal or atopic subjects and treated with IL-5 exhibited increased reactivity with the CBRM1/5 anti-αM conformation-specific mAb (Figure 6A), concomitant with enhanced adhesion to albumin, ICAM-1, fibrinogen, vitronectin, or VCAM-1 (Figure 6B). Blood EOS treated with IL-5 did not adhere to fibronectin, laminin, or collagen type I, consistent with the adhesive phenotype of airway EOS purified from antigen-challenged subjects. Adhesion to albumin (Figure 6C), ICAM-1, fibrinogen, or vitronectin (not shown) after IL-5 stimulation was mediated by αMβ2, since adhesion was blocked by inhibitory mAbs against αM and β2 subunits. There was no detectable difference in adhesion of EOS treated with IL-5 alone compared with IL-5 and eotaxin or RANTES (not shown). As reported previously (17), activation of EOS by IL-5 was accompanied by enhanced formation of podosomes (Figure 6D). Thus, treatment of blood EOS with IL-5 causes the phenotype characteristic of airway EOS after segmental antigen challenge. That is, IL-5 induces the epitope of αM on EOS recognized by CBRM1/5; enhances EOS adhesion to VCAM-1, albumin, ICAM-1, fibrinogen, or vitronectin via an αMβ2-dependent mechanism; and stimulates formation of podosomes of EOS. These results are consistent with the proposal that IL-5 is likely responsible for the allosteric activation of αMβ2, αMβ2-mediated enhanced adhesion, and podosome formation of EOS recruited to the asthmatic airway.
In this article we report correlative immunochemical and functional studies of integrins of EOS isolated from blood and airway of human subjects in a clinical model of allergic asthma. We identified αMβ2 as the principal heterodimeric integrin adhesion receptor on purified EOS that is structurally and functionally altered in the antigen challenge model. Specifically, airway EOS purified after antigen challenge express an allosterically activated form of αMβ2 as ascertained with the CBRM1/5 activation-specific anti-αM mAb. The allosteric activation of αMβ2 is accompanied by enhanced adhesion of airway EOS to diverse ligands mediated by αMβ2 and by increased formation of podosomes, transitory adhesive contacts implicated in cell movement and proteolysis of matrix proteins. Treatment of blood EOS with IL-5 results in similar allosteric activation of αMβ2, increased αMβ2-dependent adhesion, and induction of podosome formation. IL-5, found in the airway after segmental antigen challenge (40) is, therefore, a likely effector of the enhanced adhesive, podosome, and migratory phenotype of airway EOS. We did not detect enhanced activity of α4β1 of EOS purified from airway as compared with blood, indicating that antigen challenge results in a hyperadhesive phenotype that is specifically as well as primarily αMβ2-dependent.
We initially hypothesized that antigen challenge causes activation of α4β1 and enhanced adhesion to VCAM-1. However, we did not detect an increase in specific adhesion to VCAM-1 that could be attributed to α4β1. These results are strengthened by results from conformation-sensitive mAbs to β1, N29, HUTS-21, and 9EG7, which reacted at either low or undetectable levels with β1 on purified blood or airway EOS. Indeed, total α4 and β1 subunit expression levels were decreased on airway EOS compared with blood EOS, and there was no detectable difference in distribution of β1 on either cell type adherent to VCAM-1. Interestingly, α4β1 on blood or airway EOS did not support adhesion to a second α4β1 ligand, plasma fibronectin. Even after incubation of EOS with potential activators of α4β1, including Mn2+ and PMA, EOS remained unable to recognize fibronectin (not shown). Adhesion of EOS to fibronectin can, however, be induced by incubation with the 8A2 stimulatory mAb (16), which binds to a regulatory region in the βA domain shared by the adhesion blocking mAbs, 4B4 and 13 (43). Incubation with 8A2 exposes the epitope in β1 of the K562 human erythroleukemic cell line recognized by the 9EG7 activation-sensitive mAb (44). Because 9EG7 did not recognize β1 on purified blood or airway EOS, β1 on these cells likely is in a conformational state that is not competent to support adhesion to fibronectin. Indeed, Jurkat cells, which we have found to react with 9EG7, adhered to fibronectin and adhered better to VCAM-1 than EOS (not shown). Blood EOS did not adhere to laminin or fibrinogen, indicating that α6β1, αMβ2, and αXβ2 on blood EOS from unchallenged or challenged subjects are in allosteric states that do not allow for such interactions. Our evidence that attributes the hyperadhesive phenotype of airway EOS solely to αMβ2 leaves unanswered the question of if and when the potential adhesive activities of other integrins on EOS come into play.
Considerable evidence indicates that αMβ2 activation may be critical in mediating trafficking of EOS as well as neutrophils to inflammatory foci. αMβ2 mediates transmigration of EOS through endothelial cell monolayers in vitro (45). Migration mediated by αMβ2 is facilitated by the chemokines RANTES, MCP-3, and C5a, which also induce expression of the activation-sensitive epitope of αMβ2 recognized by the CBRM1/5 conformation-sensitive mAb (46–49). Numbers of EOS in blood and airway of humans with asthma are reduced after administration of a humanized mAb against IL-5 (50, 51). Expression of αMβ2 is upregulated on purified human and mouse airway EOS after antigen challenge compared with blood EOS purified before challenge (37, 52–54). αMβ2 expression is increased on migratory EOS (45). Studies of mice null for αM indicate that αMβ2 may be functionally important in a variety of cellular functions. Thus, neutrophils isolated from such mice exhibit reduced respiratory burst (55), defects in degranulation (55), impaired phagocytosis (56), and diminished apoptosis (56). Mice that lack β2 expression are characterized by impaired emigration of neutrophils into inflamed or infected skin (57) and mimic the phenotype of leukocyte adhesion deficiency syndrome in humans (58). Mice that are null for expression of the β2-ligand, ICAM-1, display decreased neutrophil influx in peritonitis (59).
Podosomes are foci of proteolytic activity and play a role in cell locomotion (38). EOS expressing active αMβ2 adhere to diverse ligands, migrate, form podosomes, and remove VCAM-1 (17). One scenario linking functions of αMβ2 and podosomes is that enhanced adhesion mediated by activated αMβ2 nucleates formation of podosomes, leading to enhanced mobility and proteolysis of matrix proteins via clustering of membrane-associated proteases, including ADAM8 (17). αMβ2 interacting with diverse ligands, including some that may become better αMβ2 adhesive and migratory ligands as a result of proteolysis via podosomes, may modulate migration of EOS from blood to the airway wall and finally into the airway lumen. The hyperadhesive and hypermigratory phenotype of airway EOS characterizes most EOS purified from BAL fluid obtained after segmental antigen challenge, as assessed by the uniform increase in CBRM1/5 positivity (Figure 4). One can only speculate whether this phenotype represents the residua of changes required for EOS to move from blood into the airway wall and on into airway lumen or is important for the function of EOS in airway, for example, to allow movement between airway wall and lumen. In either case, an even more marked hyperadhesive phenotype may characterize EOS in the airway wall.
The authors thank Heather Gerbyshak, Kristyn Jansen, Anne Brooks, and Julie Sedgwick for eosinophil isolation; Ming Lye for help with immunofluorescence experiments; Lara Johansson for help with cell culture; Kathleen Schell, Joan Batchelder, Kristin Elmer, Janet Lewis, Joel Puchalski, and Pamela Whitney for help with flow cytometry data collection and analysis; JoAnn Meerschaert for TS1/18 purification; Pat Hoffman (ICOS), Lynn Allen-Hoffmann, Eva Gordon, Roy Lobb, Richard Lynch, Cassandra Paul, and Kenneth Yamada for providing reagents and cell lines; Ioana Oltean for BAL fluid samples; Becky Kelly and Cheri Swenson for advice; and Mary-Ellen Bates and Paul Bertics for manuscript comments and suggestions.
This work was supported by Specialized Center of Research (SCOR) grant HL56396, Hematology Training grant RTH T32 HL007899, and General Clinical Research Center (GCRC) grant M01 RR03186 from the National Institutes of Health.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1165/rcmb.2006-0027OC on April 6, 2006
Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.