Selective eosinophil recruitment is the result of orchestrated events involving cell adhesion molecules, chemokines, and their receptors. The mechanisms by which chemokines regulate eosinophil adhesion and migration via integrins are not fully understood. In our study, we examined the effect of CCR3-active chemokines on eosinophil adhesion to VCAM-1 and BSA under both static and flow conditions. When eotaxin-2 or other CCR3-active chemokines were added to adherent eosinophils, it induced rapid and sustained eosinophil detachment from VCAM-1 in a concentration-dependent manner. Adhesion was detectably reduced within 3 min and was further reduced at 10–60 min. Simultaneously, eotaxin-2 enhanced eosinophil adhesion to BSA. Preincubation of eosinophils with the CCR3-blocking mAb 7B11 completely prevented chemokine-induced changes in adhesion to VCAM-1 and BSA. Using a different protocol, pretreatment of eosinophils with chemokines for 0–30 min before their use in adhesion assays resulted in inhibition of VCAM-1 adhesion and enhancement of BSA adhesion. By flow cytometry, expression of α4 integrins and a β1 integrin activation epitope on eosinophils was decreased by eotaxin-2. In a flow-based adhesion assay, eotaxin-2 reduced eosinophil accumulation and the strength of attachment to VCAM-1. These results show that eotaxin-2 rapidly reduced α4 integrin function while increasing β2 integrin function. These findings suggest that chemokines facilitate migration of eosinophils by shifting usage away from β1 integrins toward β2 integrins.

The migration of cells from blood vessels into tissues involves cell adhesion molecules, chemokines, and chemokine receptors participating in sequence (1, 2). During the earliest stage of this process, circulating leukocytes undergo rolling on the endothelial surface, an event involving selectins and their carbohydrate-containing counterligands (3). Rolling of leukocytes on the endothelium may be followed by the induction of firm adhesion. The currently accepted paradigm is that chemokines and other chemoattractants enhance leukocyte adhesion to their endothelial ligands and contribute to firm adhesion (4, 5, 6). Subsequent leukocyte migration into tissues is thus the result of coordinated events involving adhesion and de-adhesion.

Selective leukocyte recruitment is felt to occur based on the pattern of adhesion molecules and chemokines expressed at the inflammatory site. For example, eosinophil recruitment during allergic inflammation is a complex process initiated by the interaction of leukocyte adhesion molecules with counterligands on vascular endothelial cells (7). In particular, CCR3-active chemokines selectively induce eosinophil chemotaxis and transendothelial migration (8), both of which are primarily mediated by β2 integrins (9). How chemokines affect integrins in eosinophils remains perplexing. It has been reported that 15-min incubation with monocyte chemoattractant protein-3 or RANTES induces the enhancement of eosinophil adhesion to VCAM-1, while 30-min incubation reduces adhesion to VCAM-1 (6). However, Kuijpers et al. (10) reported that artificial potentiation of β1 integrin function with an activating mAb blocks transendothelial migration, presumably by inducing a state of hyperadhesion. Other investigators reported that the extent of eosinophil migration across cytokine-treated human pulmonary microvascular endothelial cells (HPMEC)3 varied inversely with VCAM-1 expression on HPMEC and eosinophil adhesion to HPMEC (11). In addition, it has been reported that anti-VCAM-1 mAb enhanced eosinophil transendothelial migration across IL-1β-activated human endothelial monolayers under static conditions (12) and an α4 integrin-blocking mAb did not inhibit chemokine-induced eosinophil transendothelial migration across IL-1β-activated human endothelial cells (9 , 13). These data suggest that stimulation of β1 integrins can inhibit migration and that chemokines appear to down-regulate β1 integrin function during eosinophil migration.

We hypothesized that in order for chemokines to induce eosinophil accumulation in tissues, they must stimulate their de-adhesion from luminal endothelial counterligands such as VCAM-1. To investigate this hypothesis, we examined the effect of CCR3-active chemokines on eosinophil adhesion to VCAM-1, an event known to be mediated by α4 integrins (14, 15, 16). Our data demonstrate rapid and sustained reductions in α4 integrin function and VCAM-1 adhesion induced by chemokines. These changes occur concomitantly with increases in β2 integrin function.

Human eotaxin-2 was kindly provided by Dr. John White (SmithKline Beecham Pharmaceuticals, King of Prussia, PA). Human soluble recombinant VCAM-1 (sVCAM-1), IL-5, eotaxin, and RANTES were purchased from R&D Systems (Minneapolis, MN). BSA, platelet-activating factor (PAF), cytochalasin D, and colchicine were purchased from Sigma (St. Louis, MO).

The CCR3-blocking mAb 7B11 (IgG2a) was kindly provided by Dr. Walter Newman (LeukoSite, Cambridge, MA). Mouse IgG1 mAbs recognizing αd integrin (169A) and blocking αd integrin (240I) were generously provided by Dr. Pat Hoffman (ICOS, Bothell, WA). The mouse IgG2a mAb recognizing HLA class I (W6/32) was purchased from Accurate Chemicals & Scientific (Westbury, NY). Blocking mouse IgG1 mAbs recognizing β1 integrin were used; 33B6 was kindly provided by Dr. Bradley McIntyre (Texas Medical Center, Houston, TX) and 4B4 was purchased from Coulter-Immunotech (Hialeah, FL). Blocking mouse IgG1 mAbs recognizing α4 (HP2/1), β2 (7E4), and CD11b (clone 44) integrins were purchased from Coulter-Immunotech. The β1 integrin activation epitope detecting mAb 15/7 (17, 18) was kindly provided by Dr. Ted Yednock (Elan Pharmaceuticals, San Francisco, CA).

Human eosinophils were isolated from EDTA-anticoagulated venous blood of donors with mild allergic rhinitis or asthma by 1.090 g/ml Percoll density gradient centrifugation at room temparature and removal of CD16-positive cells (neutrophils) using immunomagnetic beads exactly as previously described (12, 19). Eosinophil purity (based on the examination of Diff-Quik-stained cytocentrifugation preparations) was >96%, and viability (by erythrosin B dye exclusion) was nearly 100%. Eosinophils were labeled with 51Cr and resuspended at 3–4 × 106 cells/ml in Dulbecco’s PBS containing 1 mM CaCl2, and MgCl2, as well as 1% BSA (PBS-BSA).

Expression of integrins on eosinophils was tested by indirect immunofluorescence and flow cytometry as previously described (19). Freshly isolated eosinophils were prewarmed for 5 min at 37°C and incubated with 0–30 nM eotaxin-2 for 10 min at 37°C. Then eosinophils were incubated for 30 min at 4°C in PBS solution containing 0.2% BSA (Sigma) and 4 mg/ml human IgG (Sigma) with saturating concentrations of mAb or an equivalent concentration of irrelevant IgG control mAb. Cells were washed and incubated with PE-conjugated F(ab′)2 goat anti-mouse IgG Ab (Biosource, Cammarillo, CA) for another 30 min at 4°C. After fixation in 1% paraformaldehyde in PBS, 5000 cells were evaluated using a FACScalibur flow cytometer (Becton Dickinson, San Jose, CA).

For these adhesion assays, 96-well plates (Nunc, Maxi-sorb Immunoplates; PGC Scientific, Gaithersburg, MD) were coated overnight at 4°C with 50 μl aliquots of 4 μg/ml sVCAM-1 diluted in PBS containing 1 mM CaCl2 and MgCl2 (19). In some experiments, 30 nM eotaxin-2 and sVCAM-1 were coimmobilized overnight. The wells were then blocked with PBS-BSA for at least 2 h at room temperature to reduce adherence to plastic. Control adherence was measured in wells coated with PBS alone and blocked with PBS-BSA. 51Cr-labeled eosinophils (2 × 105 in 50 μl) were added to the wells in duplicate and incubated for up to 30 min at 37°C. Next, 0–30 nM chemokine (eotaxin, eotaxin-2, or RANTES), 10 ng/ml IL-5 (R&D Systems), or 1 μM PAF was added to the appropriate wells, and cells were incubated for another 3–60 min. In some experiments, eosinophils were preincubated with chemokines, 1 μM drugs (cytochalasin D, colchicine), or 1:2500 dilution of DMSO (an identical concentration for both drugs) for 10 min at 37°C, or with saturating concentrations of blocking mAbs before adding them to the wells. At the end of the adhesion assay, nonadherent cells were removed by rinsing with 100 μl PBS-BSA twice and adherent cells were lysed. The radioactivity of adherent cell lysates was determined with a gamma-counter, and percent adherence was calculated by comparing the radioactivity of adherent cell lysates to that of separate 50-μl aliquots of cell suspension.

The assembled parallel plate flow assay system consisted of a transparent polycarbonate block, a silicon rubber gasket the thickness of which determines the channel height, with a cutout in the form of channel (178 μm channel depth, 0.5 cm channel width), and a 35-mm tissue-culture plate (Corning, Corning, NY) coated with sVCAM-1 overnight at 4°C (20, 21). The apparatus was held together by vacuum and was mounted on an inverted-stage microscope (Nikon TE300) equipped with ×10 phase objective and a ×0.55 projection lens (Nikon, Melville, NY). Once assembled, the chamber and plate were placed on the microscope stage and the flow of cells was initiated by the syringe pump attached to the outlet port so that cells were drawn (rather than pushed) through the chamber. The wall shear stress (τ), assuming Newtonian fluid behavior and constant density and viscosity, was calculated by the following formula: τ = 6Qμ/wh (2), where Q is the volumetric flow rate, μ is the viscosity of the flowing fluid, h is the channel height, and w is the channel width (21). Eosinophils (106/ml) were drawn at a constant flow controlled by a syringe pump through a parallel plate laminar flow chamber. Surfaces of the plate were preblocked with BSA (Sigma) before use. Buffer or eotaxin-2 (3 nM) was added to cells immediately before infusion. Cells were infused under a shear force of 0.5 dyn/cm2 for 3 min. After cells had tethered at 0.5 dyn/cm2 for 3 min, detachment assays were performed. This was accomplished by doubling the shear stresses every 15 s to a maximum of 32 dyn/cm2. Interactions between cells and plate were visualized in real time with video microscopy using phase contrast optics (Nikon TE300 and a CCD camera (Dage-MTI, Michigan City, IN). A single field of view (×10; 0.55 mm2) was monitored on a black and white high resolution Sony monitor (Tokyo, Japan) for the entire perfusion period and videotaped on a JVC recorder for later analysis. To quantify the number of adherent cells that remained bound at the end of each 15-s interval, images were digitized from the videotape recorder using a Scion frame grabber and a personal computer, and processed with OPTIMAS6.5 software (Agris-Schoen Vision System, Alexandria, VA). In the detachment assay, data were expressed as the percentage of initially bound cells remaining adherent (22). During all experiments, the entire flow system was maintained at 37°C in a warm air box surrounding the microscope.

All results were shown as mean ± SEM. Statistical analyses were performed using ANOVA with a Fisher posthoc t test. The level of significance was set at p < 0.05.

Eosinophils were incubated on sVCAM-1-coated wells for 30 min and then 0–30 nM eotaxin-2 was added for another 10 min. Spontaneous eosinophil adhesion to plate-bound VCAM-1 was 45.8 ± 4.0% (Fig. 1). The inhibition of eosinophil adhesion to VCAM-1 was dependent on the concentration of eotaxin-2; the lowest eosinophil adhesion to VCAM-1 was observed in the presence of 3 nM and 30 nM eotaxin-2 (24.4 ± 1.85% and 22.1 ± 2.0%, respectively). The lowest concentration of eotaxin-2 to significantly inhibit eosinophil adhesion to VCAM-1 was 0.3 nM. In contrast, spontaneous eosinophil adhesion to BSA-coated plates was 6.0 ± 1.0%, and adhesion was enhanced by eotaxin-2 in a concentration-dependent manner. Eotaxin-2 at 30 nM increased eosinophil adhesion to plate-bound BSA to 17.0 ± 2.3%. Next, we examined the kinetics of the effect of eotaxin-2 on adhesion of eosinophils to VCAM-1 and BSA. Eosinophils were incubated on sVCAM-1-coated wells for 30 min and then were stimulated with 3 nM eotaxin-2 for another 3, 5, 10, 30, or 60 min (Fig. 2,A). Even at 3 min, eosinophil adhesion to VCAM-1 or BSA was already markedly reduced or enhanced, respectively, by eotaxin-2. The effect then remained stable for the entire 60-min period. We also compared this effect of eotaxin-2 to that of a different eosinophil activator, namely IL-5. As shown in Fig. 2 B, IL-5 had a similar effect, except that the magnitude of enhancement of attachment to BSA was greater.

FIGURE 1.

Effect of eotaxin-2 on eosinophil adhesion to VCAM-1 (•) and BSA (○). 51Cr-labeled eosinophils were added to the VCAM-1- or BSA-coated plate and incubated for 30 min at 37°C. Then eotaxin-2 was added to the appropriate wells and cells were incubated another 10 min. At the end of the incubation, nonadherent cells were rinsed away and adherent cells were lysed. Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

FIGURE 1.

Effect of eotaxin-2 on eosinophil adhesion to VCAM-1 (•) and BSA (○). 51Cr-labeled eosinophils were added to the VCAM-1- or BSA-coated plate and incubated for 30 min at 37°C. Then eotaxin-2 was added to the appropriate wells and cells were incubated another 10 min. At the end of the incubation, nonadherent cells were rinsed away and adherent cells were lysed. Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

Close modal
FIGURE 2.

A, Kinetics of effect of 3 nM eotaxin-2 on adhesion of eosinophils to VCAM-1 and BSA. 51Cr-labeled eosinophils were incubated in the VCAM-1- (squares) or BSA- (circles) coated plate for 30 min at 37°C. Next, buffer (open symbols) or eotaxin-2 (filled symbols) was added to appropriate wells and cells were incubated for an additional 3–60 min. Data are shown as mean ± SEM of three separate experiments; ∗, p < 0.05. B, Comparison of the kinetics of effect of 3 nM eotaxin-2 and 10 ng/ml IL-5 on adhesion of eosinophils to VCAM-1 and BSA. 51Cr-labeled eosinophils were incubated in the VCAM-1- (filled symbols) or BSA- (open symbols) coated plate for 30 min at 37°C. Next, buffer (circles), eotaxin-2 (squares), or IL-5 (triangles) was added to appropriate wells and cells were incubated for an additional 10–60 min. Data are shown as mean ± SEM of two separate experiments.

FIGURE 2.

A, Kinetics of effect of 3 nM eotaxin-2 on adhesion of eosinophils to VCAM-1 and BSA. 51Cr-labeled eosinophils were incubated in the VCAM-1- (squares) or BSA- (circles) coated plate for 30 min at 37°C. Next, buffer (open symbols) or eotaxin-2 (filled symbols) was added to appropriate wells and cells were incubated for an additional 3–60 min. Data are shown as mean ± SEM of three separate experiments; ∗, p < 0.05. B, Comparison of the kinetics of effect of 3 nM eotaxin-2 and 10 ng/ml IL-5 on adhesion of eosinophils to VCAM-1 and BSA. 51Cr-labeled eosinophils were incubated in the VCAM-1- (filled symbols) or BSA- (open symbols) coated plate for 30 min at 37°C. Next, buffer (circles), eotaxin-2 (squares), or IL-5 (triangles) was added to appropriate wells and cells were incubated for an additional 10–60 min. Data are shown as mean ± SEM of two separate experiments.

Close modal

To this point, eotaxin-2 was added to the adherent eosinophils. Preincubation of eosinophils with eotaxin-2 for up to 30 min, or coincubation of eosinophils with eotaxin-2 during the VCAM-1 adhesion assay, inhibited adhesion to a similar extent as in the initial assay (Table I). Coimmobilization of eotaxin-2 with VCAM-1 reduced the adhesion to a similar degree as observed with VCAM-1 alone (data not shown). These data indicate that eotaxin-2 causes rapid and sustained inhibition of eosinophil adhesion to VCAM-1 and enhanced adhesion to BSA.

Table I.

Effect of eotaxin-2 on eosinophil adhesion to VCAM-1 using three different treatment protocolsa

Timing of Stimulation of Eotaxin-2 (3 nM)
PreincubationCoincubationPostadhesion
No chemokine 50.6 ± 7.0 50.4 ± 3.0 47.9 ± 3.2 
Eotaxin-2 27.8 ± 5.6 23.1 ± 2.9 32.8 ± 4.5 
Timing of Stimulation of Eotaxin-2 (3 nM)
PreincubationCoincubationPostadhesion
No chemokine 50.6 ± 7.0 50.4 ± 3.0 47.9 ± 3.2 
Eotaxin-2 27.8 ± 5.6 23.1 ± 2.9 32.8 ± 4.5 
a

Mean ± SEM for percent adhesion: n = 3.

Among receptors for chemokines, eosinophils express CCR1, CCR3, and CCR6 on their surface (8, 23, 24). Because eotaxin-2 induces eosinophil chemotaxis via CCR3 (25, 26), we next examined whether CCR3 mAb inhibited chemokine-induced changes in adhesion. The CCR3-blocking mAb, 7B11 (8), had no effect on spontaneous eosinophil adhesion to BSA or VCAM-1, but reduced the effect of eotaxin-2 on eosinophil adhesion (Fig. 3). We also examined the effect of other CCR3-active chemokines, such as eotaxin and RANTES, on eosinophil adhesion to BSA or VCAM-1. Eotaxin and RANTES, like eotaxin-2, reduced adhesion to VCAM-1 and enhanced adhesion to BSA. These effects were completely inhibited by CCR3 mAb. In contrast, the HLA class I mAb, W6/32, used as a control (12, 16), had no effect on chemokine-induced changes in adhesion to BSA or VCAM-1 (data not shown).

FIGURE 3.

Effect of CCR3-active chemokines and CCR3-blocking mAb on eosinophil adhesion to BSA and VCAM-1. 51Cr-labeled eosinophils were incubated with (▪) or without (□) anti-CCR3 mAb for 30 min before adding them to the wells. Eosinophils were then incubated on the VCAM-1- or BSA-coated plates for 30 min at 37°C and 3 nM of the indicated chemokine was added as indicated. Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

FIGURE 3.

Effect of CCR3-active chemokines and CCR3-blocking mAb on eosinophil adhesion to BSA and VCAM-1. 51Cr-labeled eosinophils were incubated with (▪) or without (□) anti-CCR3 mAb for 30 min before adding them to the wells. Eosinophils were then incubated on the VCAM-1- or BSA-coated plates for 30 min at 37°C and 3 nM of the indicated chemokine was added as indicated. Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

Close modal

We also examined the effect of another chemoattractant, PAF (1 μM), on eosinophil adhesion to VCAM-1 and BSA. Eosinophil adhesion to VCAM-1 was decreased (from 43% to 25%, n = 4) and adhesion to BSA was increased (from 10% to 16%, n = 4). These data indicated that CCR3 agonists and PAF have reciprocal affects on eosinophil adhesion to VCAM-1 and BSA, and, for the chemokines, this is completely dependent on CCR3.

We observed that eotaxin-2 reduced eosinophil adhesion to VCAM-1. Integrins on eosinophils exist in high- and low-affinity states (27). We therefore tested the effect of eotaxin-2 on levels of VCAM-1 ligands including α4, αd, β1, and β7 integrins. Eosinophils were incubated with eotaxin-2 for 10 min and then labeled with Ab for flow cytometry. Levels of α4 integrins were decreased by eotaxin-2 in a concentration-dependent manner (Fig. 4), but only slightly. Total expression of β1 integrins, as recognized by mAb 4B4, was also decreased (∼20% reduction, data not shown). Although levels of activated β1 integrin (15/7) were low, the expression of this activation epitope was decreased by more than 50% by eotaxin-2 treatment. In contrast, levels of αd integrin, another ligand for VCAM-1 (15), were slightly increased. Finally, the expression of β2 and β7 integrins was not changed by eotaxin-2 (data not shown). Given the pronounced effect of eotaxin-2 on adhesion seen in Figs. 1 and 2, it appears likely that changes in integrin function rather than expression are responsible for decreased VCAM-1 adhesion.

FIGURE 4.

Expression of α4 (HP2/1), activated β1 (15/7), and αd (169A) integrins on eosinophils. Eosinophils were incubated with eotaxin-2 for 10 min at 37°C and then labeled with Ab. Net mean fluorescence intensity (MFI) was determined after subtracting values for IgG1 controls (3.7 ± 0.5). Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

FIGURE 4.

Expression of α4 (HP2/1), activated β1 (15/7), and αd (169A) integrins on eosinophils. Eosinophils were incubated with eotaxin-2 for 10 min at 37°C and then labeled with Ab. Net mean fluorescence intensity (MFI) was determined after subtracting values for IgG1 controls (3.7 ± 0.5). Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

Close modal

To examine the specific ligands used by freshly isolated and eotaxin-2-treated eosinophils for adhesion to BSA or VCAM-1, assays were performed in the presence of integrin-blocking mAb. As expected, in the absence of chemokine, adhesion of freshly isolated eosinophils to VCAM-1 was effectively inhibited by α4 and β1 integrin mAb (Fig. 5) and not inhibited by control mAb (CD11b or isotype-matched irrelevant mAb, data not shown). Adhesion was also significantly inhibited by β2 integrin mAb, with the rank order of efficacy with each mAb being α4 > β1 > β2. Surprisingly, adhesion was not inhibited by αd mAb. After eotaxin-2 treatment, α4, β1, and β2 integrin mAb inhibited adhesion to VCAM-1, but now the rank order of efficacy with each mAb was α4 > β2 ∼ β1. Again, neither αd nor control mAbs inhibited adhesion (Fig. 5 and data not shown). In contrast to VCAM-1 adhesion, only the β2 integrin mAb significantly inhibited eotaxin-2-induced enhancement of eosinophil adhesion to BSA (Fig. 5; data not shown).

FIGURE 5.

Effect of integrin-blocking mAb on eosinophil adhesion to BSA and VCAM-1 in the presence (▪) or absence (□) of 3 nM eotaxin-2. 51Cr-labeled eosinophils were incubated with mAb for 30 min before adding them to the wells. Eosinophils were incubated on the VCAM-1- or BSA-coated plates for 30 min at 37°C and as indicated 3 nM of eotaxin-2 was added for another 10 min. Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

FIGURE 5.

Effect of integrin-blocking mAb on eosinophil adhesion to BSA and VCAM-1 in the presence (▪) or absence (□) of 3 nM eotaxin-2. 51Cr-labeled eosinophils were incubated with mAb for 30 min before adding them to the wells. Eosinophils were incubated on the VCAM-1- or BSA-coated plates for 30 min at 37°C and as indicated 3 nM of eotaxin-2 was added for another 10 min. Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

Close modal

To test the roles of microtubules or microfilaments, eosinophils were preincubated with colchicine or cytochalasin D, allowed to adhere to VCAM-1-coated wells for 10 min, then stimulated with 3 nM eotaxin-2 for 30 min. As shown in Fig. 6, pretreatment of eosinophils with cytochalasin D did not alter their spontaneous adhesion to VCAM-1 or BSA, but completely prevented the decrease in adhesion to VCAM-1 and the increase in adhesion to BSA induced by eotaxin-2. In contrast, colchicine inhibited spontaneous VCAM-1 adhesion as well as eotaxin-2 effects on adhesion to BSA and VCAM-1.

FIGURE 6.

Effect of cytochalasin D or colchicine on eosinophil adhesion to BSA and VCAM-1 in the presence (▪) or absence (□) of eotaxin-2. 51Cr-labeled eosinophils were preincubated with cytochalasin D, 1 μM colchicine, buffer alone, or an equivalent dilution of DMSO in buffer for 10 min at 37°C before adding them to the wells. Eosinophils were incubated on the VCAM-1- or BSA-coated plates for 30 min at 37°C and, as indicated, 3 nM of eotaxin-2 was added for another 10 min. Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

FIGURE 6.

Effect of cytochalasin D or colchicine on eosinophil adhesion to BSA and VCAM-1 in the presence (▪) or absence (□) of eotaxin-2. 51Cr-labeled eosinophils were preincubated with cytochalasin D, 1 μM colchicine, buffer alone, or an equivalent dilution of DMSO in buffer for 10 min at 37°C before adding them to the wells. Eosinophils were incubated on the VCAM-1- or BSA-coated plates for 30 min at 37°C and, as indicated, 3 nM of eotaxin-2 was added for another 10 min. Data are shown as mean ± SEM of three separate experiments. ∗, p < 0.05

Close modal

Up to this point, all adhesion assays were performed under static conditions, with effects on adhesion determined after rinsing, where detachment force could not be reliably determined. To better quantify the change in adhesive strength induced by chemokines, we next examined the effect of eotaxin-2 on eosinophil adhesion under controlled shear flow conditions. Eosinophils, with or without eotaxin-2 pretreatment, were perfused through a flow chamber coated with BSA or VCAM-1. Under low flow rates (0.5 dyn/cm2), eosinophils did not roll on or adhere to BSA, but did accumulate on VCAM-1 (Fig. 7,A; data not shown). On VCAM-1, the number of attached eosinophils after 3 min of perfusion at 0.5 dyn/cm2 was measured. Approximately twice as many untreated eosinophils attached compared with eotaxin-2-treated eosinophils (Figs. 7, A and B, left panels). As shear forces were step-wise increased, more untreated eosinophils accumulated, and the accumulation reached a plateau at 2 dyn/cm2 (Fig. 7,A). In contrast, the accumulation of eotaxin-2-treated eosinophils did not increase (Fig. 7,A), such that by 32 dyn/cm2, many more unstimulated eosinophils remained attached than eotaxin-2-treated eosinophils (Fig. 7, A and B). Thus, when the number of eosinophils attached at 0.5 dyn/cm2 were specifically monitored as the shear force was increased, very few unstimulated eosinophils detached, while ∼20% of the eotaxin-2-treated eosinophils detached (Fig. 7 C). Therefore, both the extent of attachment and the strength of attachment to VCAM-1 were inhibited by eotaxin-2.

FIGURE 7.

Eosinophil accumulation on VCAM-1-coated plates under flow conditions. A, Total number of accumulated cells was evaluated at every shear stress condition. Untreated eosinophil suspensions (○) were perfused at 0.5 dyn/cm2 for 3 min and the shear was then increased every 15 s. In parallel experiments, 3 nM eotaxin-2 was added to the eosinophil suspensions just before perfusion (•). B, Photomicrographs demonstrate adherent eosinophils at 0.5 and 32 dyn/cm2 with or without 3 nM eotaxin-2. C, Cells remaining after each shear step were calculated as the percentage of cells still adherent compared with those that accumulated at 0.5 dyn/cm2. These data are shown as mean ± SEM of four separate experiments. ∗, p < 0.05

FIGURE 7.

Eosinophil accumulation on VCAM-1-coated plates under flow conditions. A, Total number of accumulated cells was evaluated at every shear stress condition. Untreated eosinophil suspensions (○) were perfused at 0.5 dyn/cm2 for 3 min and the shear was then increased every 15 s. In parallel experiments, 3 nM eotaxin-2 was added to the eosinophil suspensions just before perfusion (•). B, Photomicrographs demonstrate adherent eosinophils at 0.5 and 32 dyn/cm2 with or without 3 nM eotaxin-2. C, Cells remaining after each shear step were calculated as the percentage of cells still adherent compared with those that accumulated at 0.5 dyn/cm2. These data are shown as mean ± SEM of four separate experiments. ∗, p < 0.05

Close modal

In this study, the regulation of integrin-mediated eosinophil adhesiveness was examined. Chemokines including eotaxin-2 and other stimuli such as IL-5 or PAF, caused rapid sustained reductions of eosinophil adhesion to VCAM-1 in static adhesion assays (Figs. 1 and 2). These changes occurred at physiologic concentrations of stimuli, and for the chemokines, the effects were mediated via CCR3, because a blocking mAb completely prevented these changes ( Figs. 1–3). Identical results were obtained with chemokine activation before, during, or after the initiation of adhesion, regardless of whether the chemokine was in solution or coimmobilized with VCAM-1 (Table I), further strengthening the validity of our findings. In contrast, CCR3-active chemokines enhanced eosinophil adhesion to BSA with identical kinetics and concentration dependence as that seen with VCAM-1. We also tested the effect of eotaxin-2 on eosinophil accumulation on immobilized VCAM-1 under shear flow conditions, because integrins can mediate leukocyte tethering and rolling on VCAM-1 under physiological flow conditions (28, 29, 30, 31, 32, 33). Freshly isolated eosinophils accumulated on VCAM-1 at shear forces between 0.5 and 1 dyn/cm2 as previously reported (34) (Fig. 7, A and B). Preincubation of eosinophils with eotaxin-2 just before the flow assay reduced their accumulation on VCAM-1, consistent with our observation in the static assays. As the shear forces were increased, eotaxin-2-treated eosinophils, but not untreated eosinophils, detached (Fig. 7 C). Significant cell detachment occurred at levels >8 dyn/cm2, suggesting that eotaxin-2 decreased the adhesive strength for VCAM-1. These data indicate that the detachment from VCAM-1 by eotaxin-2 was due to reductions in the strength of adhesiveness.

Our results seem to be in partial conflict with some of the currently proposed paradigms that hypothesize that chemoattractants, including chemokines, stimulate cell adhesion to their ligand on endothelial cells. For example, it has been demonstrated that FMLP and IL-8 can enhance β2 integrin-mediated adhesion to ICAM-1 (3, 35). Our data demonstrated that CCR3-active chemokines enhanced β2 integrin-mediated eosinophil adhesion to BSA, consistent with a previous report (6). However, it was reported that RANTES, monocyte chemoattractant protein-3, and C5a transiently stimulated α4β1 integrin function followed by subsequent inhibition of function (6). In other studies, eotaxin enhanced the strength of eosinophil adhesiveness for VCAM-1 (4, 5). Furthermore, IL-5 acted like chemokines to rapidly inhibit adhesion to VCAM-1 while simultaneously enhancing adhesion to BSA, which differs from a study using GM-CSF, another eosinophil-activating cytokine (36). In the present studies, we never observed enhanced adhesion to VCAM-1 in either the static or flow assays. Although the reasons for these discrepancies are not entirely clear, methodologic differences, such as the use of nonphysiologic preincubation of eosinophils with chemokines on ice and rapid warming during the adhesion assay in one of these previous reports (6) may in part be responsible. In support of our observations, it has been reported that chemokines induce monocyte transendothelial migration under flow within 12 min (37), suggesting that detachment from VCAM-1 can be quite rapid.

The mechanisms responsible for concomitant reductions in adhesion to VCAM-1 with enhanced adhesion to BSA were explored using mAb and pharmacologic agents. Using a blocking mAb, most of the effects of chemokines on eosinophil adhesion to VCAM-1 were dependent on α4β1 integrins, while for BSA adhesion, these changes were shown to be β2 integrin dependent (Fig. 5), consistent with previous reports (19, 38, 39). These findings were further confirmed by examining the effect of eotaxin-2 on eosinophils obtained from a patient with leukocyte adhesion deficiency type 1, where adhesion to BSA was not enhanced by eotaxin-2 even though decreases in VCAM-1 adhesion were seen (H. Tachimoto and B. S. Bochner, unpublished observations). Taken together, these findings show that stimulation by eotaxin-2 causes reciprocal, simultaneous changes in certain β1 and β2 integrin functions. However, this paradigm does not appear to apply to the αd integrin subunit, because blocking mAb failed to significantly inhibit attachment even after eotaxin-2 treatment enhanced expression (Figs. 4 and 5). In contrast to our previous report, where ∼20% inhibition was seen using αd integrin mAb to block adhesion of freshly isolated eosinophils to VCAM-1 (15), the inhibition seen in the present studies (8.2 ± 6.1% inhibition, as calculated from Fig. 5) did not reach statistical significance. This was probably due to variability in effects seen among the six different eosinophil preparations (range of inhibition, 23.2% to −11.6%; data not shown). Because other β2 integrin α subunit-specific mAb had no effect on adhesion to VCAM-1, it is possible that eotaxin-2 exposure may preferentially alter the β2 subunit.

Previous studies demonstrated that α4β1 integrins on eosinophils exist in a state of partial activation, and can be maximally activated for adhesion to ligands such as fibronectin and VCAM-1 after exposure to Mn2+ without affecting the total cell surface expression of β1 integrin (27, 40). Data in Fig. 4 show that expression of α4 integrins and a β1 integrin activation epitope recognized by mAb 15/7 were decreased by eotaxin-2. However, it has been reported that PAF treatment of eosinophils increases Mac-1 expression but does not alter very late Ag-4 (VLA-4) expression (41). Nevertheless, reduced levels of α4 integrins on eosinophils obtained from bronchoalveolar lavage after allergen challenge compared with peripheral blood eosinophils in the same patient have been reported (42). However, the reduction in expression of β1 integrin activation epitopes was greater in magnitude than the more subtle reductions in total α4 integrins, suggesting that the decrease of VCAM-1 adhesion induced by eotaxin-2 involves mainly a reduction in activation of these integrins. The finding that pretreatment of eosinophils with cytochalasin D before stimulation with eotaxin-2 prevented their decreased binding to VCAM-1 (Fig. 6) suggests that actin polymerization and perhaps integrin avidity is also being affected. Similar conclusions can be reached regarding the enhancement in β2 integrin function although no change in expression of β2 integrins was detected. While recent studies implicate intracellular signaling pathways involving Rho in mediating cross-talk between chemokine receptors and integrins (43, 44), the precise mechanisms responsible for the reciprocal regulation of β1 and β2 integrins by CCR3-active chemokines will require additional investigation.

Our findings confirm the important role of integrins and chemokines in selectively regulating eosinophil adhesive responses. In vivo, if CCR3-active chemokines including eotaxin-3 (45, 46) are displayed on or released by activated endothelial cells and selectively promote attachment under flow conditions (47), we speculate that reductions of β1 integrin function and activation of β2 integrin adhesiveness may accompany the migration process. Treatment with a β1 integrin-activating mAb prevented eosinophil transendothelial migration in response to chemoattractants (10) and mAb blockade of α4 integrins either had no effect or enhanced chemokine-induced eosinophil transendothelial migration (9, 12, 13). Taken together, these findings suggest that chemokines may be necessary to facilitate detachment from luminal VCAM-1 and facilitate the process of diapedesis by shifting integrin usage in these cells away from β1 integrin-dominated interactions with VCAM-1 toward β2 integrin-dominated interactions with other ligands.

We thank Drs. Walter Newman, Pat Hoffman, Bradley McIntyre, and Ted Yednock for providing valuable reagents, Dr. Howard Lederman for providing blood samples from the leukocyte adhesion deficiency patient, Dr. Robert Schleimer for helpful discussions, Carol Bickel and Owen J. T. McCarty for technical support, and Bonnie Hebden for assistance in the preparation of this manuscript.

1

This work was supported by National Institutes of Health Grant AI41472 (to B.S.B.), the Uehara Memorial Foundation (H.T.), National Science Foundation Graduate Fellowship (M.M.B.), and Whitaker Foundation Grant RG98005 (to K.K.).

3

Abbreviations used in this paper: HPMEC, human pulmonary microvascular endothelial cells; sVCAM-1, human soluble recombinant VCAM-1; PAF, platelet-activating factor.

1
Springer, T. A..
1994
. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm.
Cell
76
:
301
2
Butcher, E. C..
1991
. Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity.
Cell
67
:
1033
3
Lawrence, M. B., T. A. Springer.
1991
. Leukocytes roll on a selectin at physiologic flow rates: distinction from and prerequisite for adhesion through integrins.
Cell
65
:
859
4
Kitayama, J., R. C. Fuhlbrigge, K. D. Puri, T. A. Springer.
1997
. P-selectin, L-selectin, and α4 integrin have distinct roles in eosinophil tethering and arrest on vascular endothelial cells under physiological flow conditions.
J. Immunol.
159
:
3929
5
Kitayama, J., C. R. Mackay, P. D. Ponath, T. A. Springer.
1998
. The C-C chemokine receptor CCR3 participates in stimulation of eosinophil arrest on inflammatory endothelium in shear flow.
J. Clin. Invest.
101
:
2017
6
Weber, C., J. Katayama, T. A. Springer.
1996
. Differential regulation of β1 and β2 integrin avidity by chemoattractants in eosinophils.
Proc. Natl. Acad. Sci. USA
93
:
10939
7
Bochner, B. S., R. P. Schleimer.
1994
. The role of adhesion molecules in human eosinophil and basophil recruitment.
J. Allergy Clin. Immunol.
94
:
427
8
Heath, H., S. Qin, P. Rao, L. Wu, G. LaRosa, N. Kassam, P. D. Ponath, C. R. Mackay.
1997
. Chemokine receptor usage by human eosinophils: the importance of CCR3 demonstrated using an antagonistic monoclonal antibody.
J. Clin. Invest.
99
:
178
9
Ebisawa, M., T. Yamada, C. Bickel, D. Klunk, R. P. Schleimer.
1994
. Eosinophil transendothelial migration induced by cytokines. III. Effect of the chemokine RANTES.
J. Immunol.
153
:
2153
10
Kuijpers, T. W., E. P. Mul, M. Blom, N. L. Kovach, F. C. Gaeta, V. Tollefson, M. J. Elices, J. M. Harlan.
1993
. Freezing adhesion molecules in a state of high-avidity binding blocks eosinophil migration.
J. Exp. Med.
178
:
279
11
Yamamoto, H., J. B. Sedgwick, W. W. Busse.
1998
. Differential regulation of eosinophil adhesion and transmigration by pulmonary microvascular endothelial cells.
J. Immunol.
161
:
971
12
Ebisawa, M., B. S. Bochner, S. N. Georas, R. P. Schleimer.
1992
. Eosinophil transendothelial migration induced by cytokines. I. Role of endothelial and eosinophil adhesion molecules in IL-1 β-induced transendothelial migration.
J. Immunol.
149
:
4021
13
Shahabuddin, S., P. Ponath, R. P. Schleimer.
2000
. Migration of eosinophils across endothelial cell monolayers: interactions among IL-5, endothelial-activating cytokines, and C-C chemokines.
J. Immunol.
164
:
3847
14
Walsh, G. M., F. A. Symon, A. L. Lazarovils, A. J. Wardlaw.
1996
. Integrin α4β7 mediates human eosinophil interaction with MAdCAM-1, VCAM-1 and fibronectin.
Immunology
89
:
112
15
Grayson, M. H., M. Van der Vieren, S. A. Sterbinsky, W. Michael Gallatin, P. A. Hoffman, D. E. Staunton, B. S. Bochner.
1998
. αdβ2 integrin is expressed on human eosinophils and functions as an alternative ligand for vascular cell adhesion molecule 1 (VCAM-1).
J. Exp. Med.
188
:
2187
16
Bochner, B. S., F. W. Luscinskas, M. J. Gimbrone, W. Newman, S. A. Sterbinsky, A. C. Derse, D. Klunk, R. P. Schleimer.
1991
. Adhesion of human basophils, eosinophils, and neutrophils to interleukin 1-activated human vascular endothelial cells: contributions of endothelial cell adhesion molecules.
J. Exp. Med.
173
:
1553
17
Picker, L. J., J. R. Treer, M. Nguyen, L. W. Terstappen, N. Hogg, T. Yednock.
1993
. Coordinate expression of β1 and β2 integrin “activation” epitopes during T cell responses in secondary lymphoid tissue.
Eur. J. Immunol.
23
:
2751
18
Yednock, T. A., C. Cannon, C. Vandevert, E. G. Goldbach, G. Shaw, D. K. Ellis, C. Liaw, L. C. Fritz, L. I. Tanner.
1995
. α4β1 integrin-dependent cell adhesion is regulated by a low-affinity receptor pool that is conformationally responsive to ligand.
J. Biol. Chem.
270
:
28740
19
Matsumoto, K., S. A. Sterbinsky, C. A. Bickel, D. F. Zhou, N. L. Kovach, B. S. Bochner.
1997
. Regulation of α4 integrin-mediated adhesion of human eosinophils to fibronectin and vascular cell adhesion molecule-1.
J. Allergy Clin. Immunol.
99
:
648
20
Konstantopoulos, K., S. Kukreti, C. W. Smith, L. V. McIntire.
1997
. Endothelial P-selectin and VCAM-1 each can function as primary adhesive mechanisms for T cells under conditions of flow.
J. Leukocyte Biol.
61
:
179
21
Konstantopoulos, K., S. Kukreti, L. V. McIntire.
1998
. Biomechanics of cell interactions in shear fields.
Adv. Drug Delivery Rev.
33
:
141
22
Diacovo, T. G., S. J. Roth, J. M. Buccola, D. F. Bainton, T. A. Springer.
1996
. Neutrophil rolling, arrest, and transmigration across activated, surface-adherent platelets via sequential action of P-selectin and the β2-integrin CD11b/CD18.
Blood
88
:
146
23
Rot, A., M. Krieger, T. Brunner, S. C. Bischoff, T. J. Schall, C. A. Dahinden.
1992
. RANTES and macrophage inflammatory protein 1α induce the migration and activation of normal human eosinophil granulocytes.
J. Exp. Med.
176
:
1489
24
Sullivan, S. K., D. A. McGrath, F. Liao, S. A. Boehme, J. M. Farber, K. B. Bacon.
1999
. MIP-3α induces human eosinophil migration and activation of the mitogen-activated protein kinases (p42/p44 MAPK).
J. Leukocyte Biol.
66
:
674
25
White, J. R., C. Imburgia, E. Dul, E. Appelbaum, K. O’Donnell, D. J. O’Shannessy, M. Brawner, J. Fornwald, J. Adamou, N. A. Elshourbagy, et al
1997
. Cloning and functional characterization of a novel human CC chemokine that binds to the CCR3 receptor and activates human eosinophils.
J. Leukocyte Biol.
62
:
667
26
Forssmann, U., M. Uguccioni, P. Loetscher, C. A. Dahinden, H. Langen, M. Thelen, M. Baggiolini.
1997
. Eotaxin-2, a novel CC chemokine that is selective for the chemokine receptor CCR3, and acts like eotaxin on human eosinophil and basophil leukocytes.
J. Exp. Med.
185
:
2171
27
Werfel, S. J., T. A. Yednock, K. Matsumoto, S. A. Sterbinsky, R. P. Schleimer, B. S. Bochner.
1996
. Functional regulation of β1 integrins on human eosinophils by divalent cations and cytokines.
Am. J. Respir. Cell Mol. Biol.
14
:
44
28
Lichtman, A. H., H. Ding, L. Henault, G. Vachino, R. Camphausen, D. Cumming, F. W. Luscinskas.
1997
. CD45RARO+ (memory) but not CD45RA+RO (naive) T cells roll efficiently on E- and P-selectin and vascular cell adhesion molecule-1 under flow.
J. Immunol.
158
:
3640
29
Berlin, C., R. F. Bargatze, J. J. Campbell, U. H. von Andrian, M. C. Szabo, S. R. Hasslen, R. D. Nelson, E. L. Berg, S. L. Erlandsen, E. C. Butcher.
1995
. α4 integrins mediate lymphocyte attachment and rolling under physiologic flow.
Cell
80
:
413
30
Alon, R., P. D. Kassner, M. W. Carr, E. B. Finger, M. E. Hemler, T. A. Springer.
1995
. The integrin VLA-4 supports tethering and rolling in flow on VCAM-1.
J. Cell Biol.
128
:
1243
31
Sriramarao, P., R. G. DiScipio, R. R. Cobb, M. Cybulsky, G. Stachnick, D. Castaneda, M. Elices, D. H. Broide.
2000
. VCAM-1 is more effective than MAdCAM-1 in supporting eosinophil rolling under conditions of shear flow.
Blood
95
:
592
32
Sriramarao, P., U. H. von Andrian, E. C. Butcher, M. A. Bourdon, D. H. Broide.
1994
. L-selectin and very late antigen-4 integrin promote eosinophil rolling at physiological shear rates in vivo.
J. Immunol.
153
:
4238
33
Van der Vieren, M., D. T. Crowe, D. Hoekstra, L. Adams, R. Vazeux, M. H. Grayson, B. S. Bochner, D. E. Staunton.
1999
. The leukocyte integrin αdβ2 binds VCAM-1: evidence for a binding interface between I domain and VCAM-1.
J. Immunol.
163
:
1984
34
Patel, K. D..
1998
. Eosinophil tethering to interleukin-4-activated endothelial cells requires both P-selectin and vascular cell adhesion molecule-1.
Blood
92
:
3904
35
Rot, A..
1992
. Endothelial cell binding of NAP-1/IL-8: role in neutrophil emigration.
Immunol. Today
13
:
291
36
Sung, K. L. P., L. Yang, M. Elices, G. Jin, P. Sriramarao, D. H. Broide.
1997
. Granulocyte-macrophage colony-stimulating factor regulates the functional adhesive state of very late antigen-4 expressed by eosinophils.
J. Immunol.
158
:
919
37
Allport, J. R., W. A. Muller, F. W. Luscinskas.
2000
. Monocytes induce reversible focal changes in vascular endothelial cadherin complex during transendothelial migration under flow.
J. Cell Biol.
148
:
203
38
Kita, H..
1997
. Regulation of eosinophil mediator release by adhesion molecules. B. S. Bochner, ed.
Adhesion Molecules in Allergic Diseases
227
Marcel Dekker, New York.
39
Kita, H., S. Horie, G. J. Gleich.
1996
. Extracellular matrix proteins attenuate activation and degranulation of stimulated eosinophils.
J. Immunol.
156
:
1174
40
Tomioka, K., D. W. MacGlashan, Jr, L. M. Lichtenstein, B. S. Bochner, R. P. Schleimer.
1993
. Granulocyte-macrophage colony stimulating factor regulates human eosinophil responses to F-Met peptide and platelet activating factor.
J. Immunol.
151
:
4989
41
Neeley, S. P., K. J. Hamann, S. R. White, S. L. Baranowski, R. A. Burch, A. R. Leff.
1993
. Selective regulation of expression of surface adhesion molecules Mac-1, L-selectin, and VLA-4 on human eosinophils and neutrophils.
Am. J. Respir. Cell Mol. Biol.
8
:
633
42
Georas, S. N., M. C. Liu, W. Newman, W. D. Beall, B. A. Stealey, B. S. Bochner.
1992
. Altered adhesion molecule expression and endothelial activation accompany the recruitment of human granulocytes to the lung following segmental antigen challenge.
Am. J. Respir. Cell Mol. Biol.
7
:
261
43
Laudanna, C., J. J. Campbell, E. C. Butcher.
1996
. Role of Rho in chemoattractant-activated leukocyte adhesion through integrins.
Science
271
:
981
44
Laudanna, C., J. J. Campbell, E. C. Butcher.
1997
. Elevation of intracellular cAMP inhibits RhoA activation and integrin-dependent leukocyte adhesion induced by chemoattractants.
J. Biol. Chem.
272
:
24141
45
Kitaura, M., N. Suzuki, T. Imai, S. Takagi, R. Suzuki, T. Nakajima, K. Hirai, H. Nomiyama, O. Yoshie.
1999
. Molecular cloning of a novel human CC chemokine (eotaxin-3) that is a functional ligand of CC chemokine receptor 3.
J. Biol. Chem.
274
:
27975
46
Shinkai, A., H. Yoshisue, M. Koike, E. Shoji, S. Nakagawa, A. Saito, T. Takeda, S. Imabeppu, Y. Kato, N. Hanai, et al
1999
. A novel human CC chemokine, eotaxin-3, which is expressed in IL-4-stimulated vascular endothelial cells, exhibits potent activity toward eosinophils.
J. Immunol.
163
:
1602
47
Patel, K. D..
1999
. Mechanisms of selective leukocyte recruitment from whole blood on cytokine-activated endothelial cells under flow conditions.
J. Immunol.
162
:
6209