Blockade of Focal Clustering and Active Conformation in β2-Integrin-Mediated Adhesion of Eosinophils to Intercellular Adhesion Molecule-1 Caused by Transduction of HIV TAT-Dominant Negative Ras1

We transduced dominant negative (dn) HIV TAT-Ras protein into mature human eosinophils to determine the signaling pathways and mechanism involved in integrin-mediated adhesion caused by cytokine, chemokine, and chemoattractant stimulation. Transduction of TAT-dnRas into nondividing eosinophils inhibited endogenous Ras activation and extracellular signal-regulated kinase (ERK) phosphorylation caused by IL-5, eotaxin-1, and fMLP. IL-5, eotaxin-1, or fMLP caused 1) change of Mac-1 to its active conformation and 2) focal clustering of Mac-1 on the eosinophil surface. TAT-dnRas or PD98059, a pharmacological mitogen-activated protein/ERK kinase inhibitor, blocked both focal surface clustering of Mac-1 and the change to active conformational structure of this integrin assessed by the mAb CBRM1/5, which binds the activation epitope. Eosinophil adhesion to the endothelial ligand ICAM-1 was correspondingly blocked by TAT-dnRas and PD98059. As a further control, we used PMA, which activates ERK phosphorylation by postmembrane receptor induction of protein kinase C, a mechanism which bypasses Ras. Neither TAT-dnRas nor PD98059 blocked eosinophil adhesion to ICAM-1, up-regulation of CBRM1/5, or focal surface clustering of Mac-1 caused by PMA. In contrast to β2-integrin adhesion, neither TAT-dnRas nor PD98059 blocked the eosinophil adhesion to VCAM-1. Thus, a substantially different signaling mechanism was identified for β1-integrin adhesion. We conclude that H-Ras-mediated activation of ERK is critical for β2-integrin adhesion and that Ras-protein functions as the common regulator for cytokine-, chemokine-, and G-protein-coupled receptors in human eosinophils.

T rafficking of leukocytes from the vasculature to target tissues during inflammation occurs by a multistep process consisting of rolling, firm adhesion, and diapedesis. These steps are regulated by the sequential activation of adhesive proteins and their ligands on both eosinophils and endothelial cells (1)(2)(3)(4). Although leukocytes share a number of recruitment pathways, their responses to chemotactic and inflammatory signals are affected by their qualitative and quantitative expression of adhesion molecules (5,6).
Recent studies indicate that cytosolic phospholipase A 2 (cPLA 2 ) phosphorylation is essential for integrin-mediated eosinophil adhesion (11,12). We also have reported previously that pharmacological inhibition of cPLA 2 prevents Ag and IL-5-induced eosinophil migration and airway hyperresponsiveness in immune-stimulated guinea pigs (13). However, the process by which cPLA 2 is activated to cause integrin-mediated adhesion remains unknown. H-Ras has been reported to cause attenuation of IL-5-mediated survival of eosinophils in vitro at 16 h (14). H-Ras also has been reported to enhance both ␤ 1 -and ␤ 2 -integrin-mediated adhesion in nondifferentiated cell systems (15,16). By contrast, H-Ras has been reported to dowregulate ␤ 1 -, ␤ 2 -, and ␤ 3 -integrin adhesion (17,18).
To assess the mechanism of ␤ 2 -mediated integrin adhesion in human eosinophils, we transduced dominant negative (dn) H-Ras protein using an HIV-TAT protein vector into fully differentiated human eosinophils and examined the mechanism of chemokine, cytokine, and chemoattractant peptides in the induction of adhesion. To examine the role of Ras in eosinophil adhesion, we constructed a dnRas plasmid (pTAT-dnRas) containing six His residues, 11 amino acids of TAT, an N17 dn H-Ras, and purified TAT-dnRas fusion protein (Fig. 1A). The role of H-Ras in adhesion was then determined in TAT-dnRas-transduced eosinophils stimulated by the cytokine IL-5, the chemokine eotaxin-1, or the chemoattractant fMLP. We further determined the involvement of H-Ras in the regulation of Mac-1 expression, conformational change to the activated state, and lateral clustering of integrin on the eosinophil surface. Our data demonstrate that Ras is a common regulator of extracellular signal-regulated kinase (ERK) phosphorylation, which changes ␤ 2 -integrin into its active conformation and activates lateral movement of integrin to achieve critical clustering for adhesion mediated by cytokines, chemokines, and chemoattractant substances.

Generation of TAT fusion protein
A cDNA fragment encoding dn H-Ras was amplified by PCR from the H-Ras cDNA (17N mutant) in pUSEamp with the following primers (sense, 5Ј-ATC ATG ACA GAA TAC AAG CTT GTG GTG GT-3Ј; antisense, 5Ј-AGG AGA GCA CAC ACT TGC AGC TCA TG-3Ј). The amplified PCR products were inserted into a pcDNA3.1/CT-GFP-TOPO vector. After insertion of H-Ras 17N cDNA, the end of dnRas cDNA had a TAA stop codon. Site-directed mutagenesis was conducted to create an AgeI cutting site in pcDNA3.1-dnRas. The mutagenized primers were as follows (mutagenized bases are identified by lowercase letters): 5Ј-GCT CGG ATC CAC TAG TCC AGa ccG GTG GAA TTG CCC TTA TC-3Ј (sense) and 5Ј-GAT AAG GGC AAT TCC ACC ggt CTG GAC TAG TGG ATC CGA GC-3Ј (antisense). The mutated plasmids were digested with AgeI/SphI and ligated into an AgeI/SphI digested pTAT vector (kindly provided by Dr. S. Dowdy) using T4 ligase.
Purification of TAT fusion proteins was performed by a modified method described by Nagahara et al. (19). TAT-dnRas was purified by sonication of high-expressing BL21(DE3)pLysS cells in 10 ml of buffer Z (8 M urea, 20 mM HEPES, pH 8.0, 100 mM NaCl). Cellular lysates were resolved by centrifugation, loaded onto 5-ml Ni-NTA columns in buffer Z, washed, and eluted with imidazole in buffer Z sequentially in 100, 250, and 500 mM concentrations. Urea and imidazole were removed from the resultant protein solution using a low-volume 10,000 MWCO Slide-A-Lyzer dialysis cassette (Pierce). Each fusion protein was flash-frozen at Ϫ80°C.

Cell adhesion assay
Cell adherence was assessed as residual eosinophil peroxidase (EPO) activity of adherent cells. Fifty microliters of soluble human ICAM-1 (10 g/ml) or VCAM-1 (5 g/ml) dissolved in 0.05 M NaHCO 3 coating buffer (15 mM NaHCO 3 and 35 mM Na 2 CO 3 , pH 9.2) was added to flat-bottom 96-well microtiter plates and incubated at 4°C overnight. ICAM-1 or VCAM-1 was decanted, and 100 l/well neat FBS was added to coated wells. After 60-min incubation at 37°C, the wells were decanted and washed with HBSS before the addition of eosinophils. Cells (1 ϫ 10 4 /100 l HBSS/0.1% gelatin) were added to each well of ICAM-1-or VCAM-1-coated microplates with or without stimulators such as 10 ng/ml IL-5, 100 ng/ml eotaxin-1, 1 M fMLP, or 1 nM PMA and were allowed to settle for 10 min on ice. Plates were rapidly warmed to 37°C and incubated for 30 min. After wash with HBSS, 100 l of HBSS/0.1% gelatin was added to the reaction wells, and serial dilutions of the original cell suspension were added to the empty wells to generate a standard curve. A total of 100 l of EPO substrate (1 mM H 2 O 2 , 1 mM o-phenylenediamine, and 0.1% Triton X-100 in Tris buffer, pH 8.0) was then added to the wells. After a 30-min incubation at room temperature, 50 l of 4 M H 2 SO 4 was added to stop the reaction. Absorbance was measured at 490 nM in a microplate reader (Thermomax; Molecular Devices, Menlo Park, CA). All assays were performed in duplicate. Data storage and analysis were facilitated by the use of computer software interfaced with the reader (Softmax; Molecular Devices). The detection of EPO by this assay was linear between concentrations of 1.0 ϫ 10 3 and 1.5 ϫ 10 4 cells/well, as determined by a standard curve. The effect of TAT-dnRas or PD98059 on integrin-mediated adhesion was tested by preincubation of TAT-dnRas or PD98059 with cells for 30 min at 37°C.

Isolation of human eosinophils
Eosinophils were isolated by a method modified from Hansel et al. (20). The method is based on Percoll centrifugation (density 1.089 g/ml) to isolate granulocytes, hypotonic lysis of RBCs, and, finally, immunomagnetic depletion of neutrophils by a magnetic cell separation system using anti-CD16-coated MACS particles. Eosinophil purity of 99% was routinely obtained, as assessed by Wright-Giemsa staining. Cells were kept on ice until use.

Immunoblot analysis of ERK1/2
Eosinophils (2 ϫ 10 6 /group) were stimulated with 10 ng/ml IL-5 for 10 min, 100 ng/ml eotaxin-1 for 1 min, 1 nM fMLP for 1 min, or 1 nM PMA for 5 min, and the reaction was stopped by centrifugation at 12,000 ϫ g for 30 s. The pellets were then lysed in 80 l of lysis buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM Na 2 EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM Na 3 VO 4 , 1 g/ml leupeptin, and 1 mM PMSF). After 20 min on ice, the samples were centrifuged at 12,000 ϫ g for 2 min to remove nuclear and cellular debris. The supernatants were then mixed with 14 l of 6ϫ sample buffer and boiled for 5 min. The samples were collected and saved at Ϫ70°C.
Samples were subjected to SDS-PAGE, using 10% acrylamide gels under reducing condition (15 mA/gel). Electrotransfer of proteins from the gels to polyvinylidene fluoride membrane was achieved using a semidry system (400 mA, 60 min). The membrane was blocked with 1% BSA for 60 min and then incubated with 1/5000 anti-phosphorylation-specific ERK1/2 Ab or 1/1000 anti-ERK1/2 Ab diluted in TBST overnight. The membranes were then washed three times for 20 min with TBST. Donkey anti-rabbit IgG conjugated with HRP was diluted 1/3000 in TBST and incubated with polyvinylidene fluoride membrane for 60 min. The membrane was again washed three times with TBST and assayed by an ECL system (Amersham).

Analysis of surface integrin expression by immunofluorescence flow cytometry and confocal microscopy
Eosinophils were preincubated with TAT-dnRas or PD98059 for 30 min at 37°C and then stimulated by 10 ng/ml IL-5, 100 ng/ml eotaxin-1, 1 M fMLP, or 1 nM PMA for 30 min. Thereafter, cells were centrifuged at 400 ϫ g for 10 min, and the pellets were resuspended in PBS with 1% BSA for FACScan analysis or in PBS with 0.1% BSA for confocal microscope analysis. Aliquots of 5 ϫ 10 5 cells were incubated with 10 g/ml of mAb directed against Mac-1, CBRM1/5, or isotype-matched control Ab for 30 min at 4°C. After two washes, the cells were incubated with an excess of FITC-conjugated goat anti-mouse Ig for 20 min at 4°C. The cells were washed twice, resuspended in 1% paraformaldehyde, and kept at 4°C until analyzed. Flow cytometry was performed by FACScan (BD Biosciences). Fluorescence intensity was determined on at least 5000 cells from each sample. The results were expressed as the specific mean fluorescence intensity (sMFI) (control Ab fluorescence subtracted).

Ras activation assay
The activated Ras was affinity precipitated following the manufacturer's instructions (Upstate Biotechnology). Next, the eosinophils (5 ϫ 10 6 / group) were incubated with 100 nM TAT-dnRas for 30 min and then stimulated with 10 ng/ml IL-5 for 10 min, 100 ng/ml eotaxin-1 for 1 min, 1 nM fMLP for 1 min, or 1 nM PMA for 1 min. The reaction was stopped by centrifuging at 12,000 ϫ g for 30 s, and the cell pellets were immediately lysed in Ras affinity precipitation lysis buffer (25 mM HEPES, pH 7.5, 150 mM NaCl, 10 mM MgCl 2 , 2 mM EGTA, 10% glycerol, 1% Nonidet P-40, 5 g/ml aprotinin, 10 g/ml leupeptin, 5 g/ml pepstatin, 50 mM NaF, 2 mM Na 3 VO 4 , and 1 mM PMSF). After 20 min on ice, the sample was centrifuged at 16,000 ϫ g for 10 min at 4°C to remove nuclear and cellular debris. The supernatants were then incubated with GST-RBD for 30 min at 4°C with rocking. The GST-RBD beads were washed three times with lysis buffer. Bound proteins were eluted by boiling in 2ϫ Laemmli sample buffer, separated by SDS-PAGE, and processed for immunoblot analysis using an anti-Ras mAb (clone RAS10).

Statistical analysis
All measurements were expressed as mean Ϯ SEM. Variation between two groups was tested using Student's t test. Variation among more than two groups was tested using ANOVA followed by Fisher's protected least significant difference. A value of p Ͻ 0.05 was accepted as statistically significant.

Transduction of TAT-dnRas into eosinophils
To access the efficacy of protein transduction, eosinophils were incubated with 100 nM TAT-dnRas (Fig. 1A) for Յ30 min, and samples were analyzed by Western blot. TAT-dnRas in cell lysates was detected within 5 min of incubation; maximal detection occurred within 20 -30 min (Fig. 1B). Consequently, a 30-min incubation time was used in the subsequent experiments. Eosinophils were 95% viable during the course of these experiments, as indicated by trypan blue exclusion. TAT-dnRas transduction into eosinophils inhibited IL-5-induced ERK1/2 phosphorylation in a time-dependent fashion. ERK1/2 phosphorylation was reduced at 10 min and blocked completely after 15 min (Fig. 1C, upper panel). TAT-dnRas caused concentration-dependent blockade of IL-5-stimulated ERK1/2 phosphorylation. ERK1/2 phosphorylation was reduced at 30 nM and blocked completely with 100 -300 nM (upper panel). TAT vehicle control (Fig. 1D, second from right) did not block ERK phosphorylation. TAT-dnRas alone had no effect on ERK phosphorylation (Fig. 1D, right). These results demonstrated that TAT-dnRas functionally suppressed ERK phosphorylation after transduction into eosinophils using HIV-TAT protein.

Effect of TAT-dnRas on adhesion of eosinophils to ICAM-1 caused by IL-5, eotaxin-1, fMLP, and PMA
We then examined the role of H-Ras in IL-5-, eotaxin-1-, or fMLPinduced eosinophil adhesion to plated ICAM-1. Adhesion of eosinophils to ICAM-1 was previously confirmed to be mediated by ␤ 2 -integrin (Mac-1) in our system (12). To examine the potential role of ERK in eosinophil adhesion, we first used a pharmacological inhibitor of mitogen-activated protein/ERK kinase (MEK), PD98059. This compound inhibits ERK1/2 phosphorylation and activation (21,22). In preliminary studies, we determined the concentrations of IL-5, eotaxin-1, fMLP, and PMA causing maximal adhesion of eosinophils to ICAM-1 (see Materials and Methods). PD98059 blocked adhesion to eosinophils activated by IL-5, eotaxin-1, and fMLP at their most efficacious concentrations in a concentration-dependent manner ( Fig. 2A).

Involvement of H-Ras in Mac-1 clustering
We further determined the mechanism of the Mac-1 clustering in eosinophil adhesion using TAT-dnRas. Confocal immunofluorescence microscopy demonstrated that IL-5, eotaxin-1, and fMLP all caused focal clustering of Mac-1 on the eosinophil surface, which was blocked by pretreatment with TAT-dnRas (Fig. 6) or PD98059 (data not shown). In contrast, PMA caused Mac-1 clustering and subsequent adhesion to ICAM-1 that was not inhibited by either TAT-dnRas or PD98059. These data indicate that Mac-1 up-regulation is not the mechanism by which H-Ras causes adhesion of eosinophils to ICAM-1. Rather, Ras-mediated adhesion corre-sponds to active conformational change and focal clustering of ␤ 2 -integrin on the eosinophil surface.

Discussion
In this study, we used HIV-TAT transduction of dnRas into fully mature human eosinophils to determine the role and mechanism by which Ras protein regulates ␤ 2 -integrin adhesion to the endothelial counterligand ICAM-1. By transducing dnRas directly into fully mature human eosinophils using an HIV-TAT fusion protein, we were able to determine the common pathway for induction of eosinophil adhesion to Mac-1 caused by cytokine (IL-5), chemokine (eotaxin-1), and chemoattractant (fMLP) stimulation. TAT-dnRas blocked endogenous Ras activation and subsequent ERK phosphorylation, as predicted from pharmacological studies using an inhibitor of MEK (Fig. 4). For IL-5, eotaxin-1, and fMLP, this substantially attenuated Mac-1 adhesion to ICAM-1 (Fig. 2).
These results indicate that Ras may play a critical role in eosinophil adhesion caused by inside-out signaling from cytokine, chemokine, and chemoattractant.
Blockade of ERK phosphorylation caused by dnRas had no effect on up-regulation of Mac-1 after cellular activation. However, dnRas blocked both the induction of active conformation of Mac-1 and the focal lateral clustering of Mac-1 (Fig. 6) as assessed by confocal microscopy and immunofluorescent staining after activation by IL-5, eotaxin-1, and fMLP. This is the first demonstration of a common activation pathway for both induction of the active conformation of Mac-1 and the clustering distribution of adhesion molecules in eosinophils. The relative importance of each of these two mechanisms was not defined in this investigation. It is important to note, however, that the binding of mature human peripheral  blood eosinophils by VLA-4 to VCAM-1 is regulated independently of Ras (Fig. 3) and that our data demonstrate unique and separate mechanisms for cell membrane adhesion of ␤ 2 -and ␤ 1integrin to their respective counterligands, ICAM-1 and VCAM-1.
We have shown previously that both ␤ 2 -and ␤ 1 -integrin-mediated adhesion depend critically upon the phosphorylation of the 85-kDa cPLA 2 (12) and that eosinophil-mediated adhesion to fibronectin is critically dependent upon ERK phosphorylation (11). However, at the cell membrane, neither ERK inhibition nor dnRas caused any blockade of adhesion of mature human eosinophils to VCAM-1. Thus, a separate mechanism exists at the cell surface for regulation adhesion of VLA-4, which is constitutively expressed in its active state, and for Mac-1 adhesion, which must be activated from its dormant state. Although the mechanism potentially regulating Mac-1 adhesion appears to involve Ras activation as a critical pathway, these studies do not elucidate the mechanism by which VLA-4 adhesion to VCAM-1 is regulated.
It is important to recognize some limitations of these findings. Adhesion in vivo occurs during a state of constant flow and is a multistep process. The extent to which investigations using plated ligands replicate events in the living state, under conditions of constant flow and shear stress, cannot be assessed for individually plated ligands. However, we have shown previously that inhibition of cPLA 2 activity, which is activated by ERK1/2, prevents eosinophil migration and airway hyperresponsiveness in guinea pigs (13). We also note that blockade back to baseline control levels of ␤ 2 -integrin-FIGURE 6. Distribution of Mac-1 expressed on the eosinophil surface. Aliquots of eosinophils were preincubated with TAT-dnRas or TAT-vehicle control for 30 min and then were stimulated by IL-5, eotaxin-1, fMLP, or PMA for 30 min. Eosinophils were incubated with anti-CD11b mAb followed by FITC-labeled goat anti-mouse IgG. Mac-1 clustering was assessed by confocal microscopy. Images show cell populations (A) and individual representative cells having optimal confocal imaging (B). Mac-1 is localized in clusters (capping) in IL-5-, eotaxin-1-, or fMLP-treated cells, whereas it is dispersed on nonstimulated control cells (data not shown) or cells pretreated with 100 nM TAT-dnRas. Unlike IL-5, eotaxin-1, or fMLP, PMA caused Mac-1 clustering in the presence of TAT-dnRas. The results shown are representative of three different experiments. ICAM-1 adhesion was complete only for eotaxin-1 (Fig. 2). This may be a consequence of necessity for examining singly the binding of individual integrin/Ig-supergene reactions, or it may imply that Ras mediation of Mac-1 adhesion to ICAM-1 has additional regulatory components.
The sequential events of eosinophil migration and, particularly, the regulation of integrin avidity still are not fully elucidated, and our study has not outlined each step of Ras-mediated Mac-1 adhesion. However, we have demonstrated for the first time two potentially different mechanisms for regulation of eosinophil adhesion by ␤ 2 -and ␤ 1 -integrin in fully mature human eosinophils by HIV-TAT transduction of a dnRas into primary isolates of peripheral blood eosinophils. We demonstrate that Ras-induced adhesion is highly specific for Mac-1 and that VLA-4-mediated adhesion is regulated in a substantially different manner at the cell surface. Our data demonstrate that receptor-mediated activation of ␤ 2 -integrin may be regulated substantially and uniquely by Ras protein and that integrin surface clustering and conformational change rather than increased surface expression of Mac-1 are the critical steps for adhesion of activated eosinophils. Our findings suggest that Ras may be the common regulator for Mac-1-mediated adhesion of eosinophils induced by cytokines, chemokines, and membrane-activating chemoattractants.