* Departments of Pediatrics and Pharmacology, University of Wisconsin, Madison; and
Imperial Cancer Research Fund, Lincoln’s Inn Fields, London, United Kingdom
Correspondence: Anna Huttenlocher, Departments of Pediatrics and Pharmacology, University of Wisconsin Medical School, 3780 MSC, 1300 University Ave., Madison, WI 53706. E-mail: huttenlocher{at}facstaff.wisc.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: fibronectin • VCAM • motility • p38 MAPK
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Integrins are a family of cell-surface adhesion receptors that play a critical role in regulating the adhesive and migratory properties of cells. Recent studies have demonstrated that integrin-mediated adhesion regulates intracellular signaling pathways central for cell migration, including the activity of the Rho family of GTPases [1 , 2 ]. Many cell types exhibit a biphasic relationship between adhesion and migration rate, and optimum speed occurs at an intermediate, cell-substratum adhesiveness [3 ]. Several parameters may potentially contribute to the adhesiveness between a cell and its environment, including adhesive ligand concentration, receptor number, receptor-ligand affinity, and the strength of receptor-cytoskeletal interactions [3 ]. The effects of some of these parameters on migration have been tested in cell-culture systems, confirming a relationship between adhesion strength and migration speed. For example, modulation of integrin ligand-binding affinity or the strength of integrin-cytoskeletal linkages can alter migration speeds [4 , 5 ], and by changing the substrate concentration in fibroblasts, these effects can be compensated [5 ]. However, these adhesion-dependent parameters have not been carefully tested in more rapidly moving cell types such as eosinophils. In this study, we investigate how the density of fibronectin and vascular cell adhesion molecule (VCAM) regulates the polarization, migration, and signaling of eosinophils.
Eosinophils are a major component of the inflammatory response during persistent allergic disease and are considered to be key participants in the development of airway inflammation in patients with asthma [6 ]. The recruitment and retention of eosinophils are an adhesion-dependent process involving the coordinated actions of cell-surface receptors and chemoattractants [7 , 8 ]. Integrins that participate in eosinophil extravasation and recruitment include the ß2 integrins, lymphocyte function-associated antigen-1 and Mac-1, and the ß1 integrin, very late antigen (VLA)-4 [9 10 11 12 13 ]. Integrin ligands include the counter receptors, intercellular adhesion molecule and VCAM, expressed on the surface of endothelium, and bind to ß2 and ß1 integrins, respectively, and the ECM component fibronectin, which binds to VLA-4 through its CS-1 domain. Fibronectin and VCAM-1 are adhesive ligands for eosinophils, which affect eosinophil function. Although fibronectin and VCAM-1 play a role in modulating eosinophil function, there have been few studies that have examined how the density of these ligands regulates eosinophil signal transduction and migration.
The levels of fibronectin isolated from bronchoalveolar lavage fluids of patients with asthma increase significantly by 48 h after antigen challenge, and the concentrations of fibronectin correlate closely with the number of eosinophils [14 ]. The increase in airway fibronectin may contribute to eosinophil persistence by modulating the survival, recruitment, and retention of lung eosinophils [15 , 16 ]. Despite recent progress in understanding how eosinophil survival may be affected by adhesion to fibronectin [15 , 16 ], we still have a limited understanding of how fibronectin regulates eosinophil migration and signal transduction.
To address the role of fibronectin and VCAM-1 during eosinophil migration, we have examined the effects of ligand density on eosinophil migration, adhesion, and signaling by the p38 mitogen-activated protein kinase (MAPK) and extracellular regulated kinase (ERK) pathways. Our findings demonstrate that fibronectin is inhibitory to eosinophil polarization, migration, and signaling, and VCAM-1 is not. The results support the hypothesis that fibronectin modifies cell signaling and the migratory capacity of eosinophils.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Isolation of human eosinophils from peripheral blood
Eosinophils were isolated from the peripheral blood of subjects with allergic disease, such as allergic rhinitis and mild asthma, and eosinophils comprised between 3 and 20% of the peripheral blood leukocytes [11
, 17
]. Subjects ranged from 22 to 60 years, and gender distribution was equal. Eosinophils were purified from the peripheral blood using previously described methods [11
, 17
]. In brief, after centrifugation through a Percoll solution (density 1.090 g/mL) and after lysis of erythrocytes by hypotonic shocks, neutrophils were depleted by incubation with anti-CD16-conjugated microbeads and exposure to a magnetic field. Eosinophils were resuspended in Hanks’ balanced salt solution (HBSS) supplemented with 1% albumin. The preparations were 95–99% eosinophils by Wright stain.
Preparation of cell lysates and immunoblotting
Eosinophils were plated on nontissue culture plates from Becton Dickinson (Lincoln Park, NJ), coated with 0.1, 1, 10, and 100 µg/mL fibronectin or 1, 10, and 50 µg/mL recombinant VCAM-1 or as indicated. Cells were plated on coated plates in the hybridoma media CCM1 from Hyclone (Logan, UT). Fifteen minutes after plating, cells were treated with 1 x 10-7 M fMLP or eotaxin for 5 min, and then extracts were collected in cold lysis buffer using previously described methods [18
]. Cells in suspension were also treated with fMLP or eotaxin as indicated. The lysis buffer was a modified radio immunoprecipitation assay buffer (20 mM Tris, pH 7.4, 1.0% Triton X-100, 0.25% sodium deoxycholate, 150 mM NaCL, 2 mM EDTA, 2 mM EGTA) with protease inhibitors (20 µg/mL leupeptin, 0.7 mg/mL pepstatin, 1 mM phenanthroline, 2 mM phenylmethylsulfonyl chloride, 0.05 units aprotonin) and a phosphatase inhibitor (1.0 mM sodium orthovanadate). Protein content of the lysates was determined by a Pierce Chemical Co. (Rockford, IL) bicinchoninic acid assay with bovine serum albumin (BSA) as the standard. For the ERK and p38 MAPK blots, 3 µg lysates were separated on 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels and were transferred to nitrocellulose membranes. Membranes were blocked in 1% cooked BSA in 10 mM Tris-HCL, pH 7.5, 100 mM NaCl, 0.1% Tween 20. Blots were visualized using chemiluminescence (Pierce Chemical Co.). Densitometry was performed using Scion image software. Data are representative results from a minimum of three separate experiments.
Cell migration assays
Time-lapse videomicroscopy and transwell assays were performed as described previously [5
]. Briefly, cells were plated in HBSS and pretreated with inhibitors and/or 1 ng/mL IL-5 as indicated for 15 min in suspension. The plates and the upper and lower surfaces of the transwell filters (Costar, Cambridge, MA) were coated with fibronectin or VCAM-1 at the concentrations indicated for 1 h at 37°C. The filters were allowed to dry, and chemotactic assays were performed by placing chemotactic factor, 20 nM eotaxin, 25 nM IL-8, or 1 nM fMLP in the bottom well. Cells (2x105) were placed in the upper well, and assays were run for 1 h. Cells were then fixed with methanol and stained with methylene blue (Hema 3 kit, Fisher Scientific, Pittsburgh, PA), and absolute cell number migrated was determined per high power field. On average, 50–100 cells migrated on noncoated surfaces, and the migration for each experiment is reported as migration relative to control migration (noncoated surfaces). Each experiment was performed a minimum of three times.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (37K): [in a new window] |
Figure 1. Effect of fibronectin-coating concentration on eosinophil chemotaxis and morphology in response to fMLP. (A) Chemotaxis was examined toward 10 nM fMLP on transwell filters coated with increasing fibronectin (0–100 µg/mL). Migration was compared with filter alone (100%). Error bars show standard deviation. (B) Effect of fibronectin-coating concentration on eosinophil morphology in the presence of fMLP. Cell morphology was examined on 1 (a) and 100 (b) µg/mL fibronectin. Original bar, 20 µm. Representative results from a minimum of three separate experiments are shown.
|
Fibronectin inhibits p38 MAPK phosphorylation more than ERK phosphorylation in the presence of fMLP
The inhibitory effects of high fibronectin density on eosinophil spreading and migration suggest that fibronectin may be modulating intracellular signaling pathways that regulate cell spreading and migration. To determine if fibronectin regulates intracellular signaling, we studied the effects of fibronectin density on signaling via the ERK and p38 MAPK pathways (Fig. 2
). p38 MAPK and ERK have been implicated in cell migration [19
20
21
22
23
24
25
26
27
]. We found that p38 MAPK phosphorylation was reduced at increasing fibronectin-coating concentrations with undetectable p38 MAPK phosphorylation at high fibronectin density. In contrast, there was only a 30% inhibition of ERK phosphorylation at high-density fibronectin in the presence of fMLP. To determine the dose-response effect of fibronectin, we examined p38 MAPK phosphorylation in suspended cells in the presence of fMLP as compared with cells plated on plastic alone and surfaces coated with 1, 10, and 100 µg/mL fibronectin. It is interesting that adhesion to plastic surfaces alone promoted p38 MAPK phosphorylation over suspended cells treated with fMLP. There was no detectable inhibition of phosphorylation at 1 µg/mL fibronectin, but significant inhibition in phosphorylation was detected at intermediate-density fibronectin (10 µg/mL). Together, these findings demonstrate that increased fibronectin concentration is inhibitory to adhesion-induced phosphorylation of p38 MAPK.
 View larger version (29K): [in a new window] |
Figure 2. Effect of fibronectin density on p38 MAPK and ERK phosphorylation (phospho) in the presence of fMLP. Eosinophils were plated for 15 min, and adherent and suspended (susp) cells were treated with 10 nM fMLP for 5 min before collecting extracts as described in Materials and Methods. (A) Phosphorylation of p38 MAPK (P-p38) with cells plated on 0.1 and 100 µg/mL fibronectin (FN) as indicated. Equivalent loading is demonstrated by reblotting for total p38 MAPK. Quantification by densitometry shows phosphorylation of p38 MAPK divided by total p38 MAPK. (B) Phosphorylation of ERK (P-ERK) with cells plated as above. Equivalent loading is demonstrated by reblotting for total ERK. (C) Phosphorylation of p38 MAPK with eosinophils plated on different densities of fibronectin, plate alone, 1, 10, and 100 µg/mL fibronectin. Quantification by densitometry from three separate experiments is shown. Error bars show standard error of the mean.
|
 View larger version (42K): [in a new window] |
Figure 3. Effect of fibronectin-coating concentration on eosinophil chemotaxis and morphology in response to eotaxin. (A) Chemotaxis was examined toward 10 nM eotaxin on transwell filters coated with increasing fibronectin (0–100 µg/mL). Migration was compared with filter alone (100%). Error bars show standard deviation. (B) Effect of fibronectin-coating concentration on eosinophil morphology in the presence of eotaxin. Cell morphology was examined on 1 (a) and 100 (b) µg/mL fibronectin. Original bar, 20 µm.
|
 View larger version (41K): [in a new window] |
Figure 4. Effect of fibronectin density on p38 MAPK and ERK phosphorylation (phospho) in the presence of eotaxin. Eosinophils were plated for 15 min and suspended (susp), and adherent cells were treated with 100 nM fMLP for 5 min before collecting extracts as described in Materials and Methods. (A) Phosphorylation of p38 MAPK (P-p38) with cells plated on 0.1 and 100 µg/mL fibronectin (FN) as indicated. Equivalent loading is demonstrated by reblotting for total p38 MAPK. Quantification by densitometry shows phosphorylation of p38 MAPK divided by total p38 MAPK. (B) Phosphorylation of ERK (P-ERK) with cells plated as above. Equivalent loading is demonstrated by reblotting for total ERK. Quantification by densitometry from three separate experiments is shown. Error bars show standard error of the mean.
|
 View larger version (57K): [in a new window] |
Figure 5. Effect of VCAM density on eosinophil chemotaxis and morphology. (A) Chemotaxis was examined toward fMLP (open bars) or eotaxin (hatched bars) on transwell filters coated with increasing VCAM-1-coating concentration (0–50 µg/mL). Migration was compared with filter alone (100%). Error bars show standard deviation. (B) Effect of VCAM-coating concentration on eosinophil morphology in the presence of fMLP (a, b) and eotaxin (c, d). Cell morphology was examined on 1 (a, c) and 100 (b, d) µg/mL VCAM-1. Original bar, 20 µm.
|
50% reduction in ERK phosphorylation in the presence of eotaxin at high-density VCAM-1. However, although high-density VCAM-1 reduces ERK phosphorylation, no reduction in eosinophil chemotaxis to fMLP was observed at high-density VCAM-1. The findings demonstrate that adhesive ligands modulate the effects of chemoattractant on eosinophil activation and phosphorylation of p38 MAPK and ERK. Together, the results suggest that cross-talk between adhesive ligand and activator is a critical regulator of eosinophil signal transduction.
 View larger version (49K): [in a new window] |
Figure 6. Effect of VCAM-1 density on ERK and p38 MAPK phosphorylation (phospho). Eosinophils were plated for 15 min and suspended (susp), and adherent cells were treated with fMLP (A, B) or eotaxin (C, D) for 5 min before collecting extracts as described in Materials and Methods. (A, C) Phosphorylation of p38 MAPK (P-p38) with cells plated on 1, 10, and 50 µg/mL VCAM-1 as indicated. Equivalent loading is demonstrated by reblotting for total p38 MAPK. Quantification by densitometry shows phosphorylation of p38 MAPK divided by total p38 MAPK. (B, D). Phosphorylation of ERK (P-ERK) with cells plated as above. Equivalent loading is demonstrated by reblotting for total ERK. Quantification by densitometry from three separate experiments is shown. Error bars show standard error of the mean.
|
 View larger version (69K): [in a new window] |
Figure 7. Effect of the p38 MAPK inhibitor SB203580 and ERK inhibitor U0126 on eosinophil chemotaxis to fMLP. Dose-dependent inhibition of eosinophil chemotaxis was observed in the presence of SB203580 (A) and U0126 (B). Migration is relative to control migration in the absence of inhibitor (100%). Data show average from three separate experiments, and error bars show standard deviation.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It is known that many cell types including fibroblasts and carcinomas exhibit a biphasic dependence of cell migration speed on ligand density. We have recently reported that high fibronectin density induces a stop signal that inhibits Chinese hamster ovary cell migration by modulating intracellular signaling pathways via the Rho family of GTPases [18 ]. We found that even at low-coating concentrations of fibronectin, there was an inhibition of eosinophil migration in comparison with the transwell membrane alone, suggesting that fibronectin is inhibitory to eosinophil migration in a dose-dependent manner. This is in distinct contrast from other cell types, where fibronectin at low concentrations is required for their migration and is only inhibitory at high-coating concentrations [18 ]. Despite the inhibition of cell spreading and migration on high fibronectin density, there is no difference between eosinophil adhesion to fibronectin and VCAM-1 at high density (data not shown), consistent with previously published reports [10 ].
The physiological relevance of these findings may include a functional role for fibronectin in the airway of asthmatics as a mechanism to retain eosinophils and inhibit their migration out of the airway. It has recently been demonstrated that the concentrations of fibronectin are increased in the airway of asthmatics (240–697 ng/mL), and fibronectin density correlates with eosinophil numbers in the airway [14 ]. This effect may result from increased eosinophil survival and from fibronectin effects on eosinophil signaling, polarization, and migration. In contrast, VCAM-1 supports eosinophil migration and influx [12 ].
The effects of ligand density on eosinophil activation of the p38 MAPK and ERK signaling pathways demonstrate cross-talk between ECM and chemoattractants. We found that adhesion activates p38 MAPK and ERK in the presence of fMLP and eotaxin. It is interesting that in the presence of eotaxin but not fMLP, we found an inhibition of ERK signaling on high-density VCAM-1, demonstrating that there are distinct effects on intracellular signaling in the presence of different activators. This is in accordance with reported findings that demonstrate cross-talk between chemoattractant and integrin in regulating neutrophil migration [29 ].
In summary, we have found that fibronectin inhibits eosinophil spreading and migration. Fibronectin, but not VCAM-1, inhibited eosinophil polarization and spreading at intermediate-to-high ligand density. Accordingly, high-density fibronectin but not VCAM-1 inhibited activation of p38 MAPK more than ERK signaling, implicating p38 MAPK in the fibronectin density-dependent regulation of eosinophil chemotaxis. In support of this possibility, we found that p38 MAPK inhibitors reduced eosinophil polarization and chemotaxis, comparable with the effects of high fibronectin density on eosinophil function. Together, these findings suggest that fibronectin density regulates intracellular signaling pathways critical for eosinophil spreading and chemotaxis.
Received May 31, 2002; revised January 13, 2003; accepted January 16, 2003.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. Liu, B. Schwartz, Y. Tsubota, E. Raines, H. Kiyokawa, K. Yonekawa, J. M. Harlan, and L. M. Schnapp Cyclin-Dependent Kinase Inhibitors Block Leukocyte Adhesion and Migration J. Immunol., February 1, 2008; 180(3): 1808 - 1817. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Barthel, M. W. Johansson, D. M. McNamee, and D. F. Mosher Roles of integrin activation in eosinophil function and the eosinophilic inflammation of asthma J. Leukoc. Biol., January 1, 2008; 83(1): 1 - 12. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Kazi, S. Lotfi, E. A. Goncharova, O. Tliba, Y. Amrani, V. P. Krymskaya, and A. L. Lazaar Vascular endothelial growth factor-induced secretion of fibronectin is ERK dependent Am J Physiol Lung Cell Mol Physiol, March 1, 2004; 286(3): L539 - L545. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
