Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Interleukin-4 (IL-4) was initially described as a growth factor for B cells stimulated with anti-immunoglobulin M (IgM) antibodies (1). Since then, a plethora of other functions have been revealed, indicating the immunoregulatory importance of IL-4 in vivo (2). In allergy, an imbalance in the production of Th1- and Th2-like cytokines (IL-2 and interferon- [IFN-], and IL-4 and IL-5, respectively) in favor of Th2 has been postulated. Several findings support a pathogenic key role of Th2-derived cytokines. Indirect evidence includes typical features of most allergic patients, for example, increased serum IgE levels and blood eosinophilia. In humans, IL-4 is an essential cofactor in the induction of IgE synthesis, whereas IL-5 is essential for the terminal differentiation of eosinophils (3). More direct evidence has been provided by analysis of allergen-specific T-cell clones. These studies have shown that the great majority of allergen-specific T-cell clones derived from peripheral blood of atopic donors preferentially produce IL-4, IL-5, IL-6, IL-13, tumor necrosis factor- (TNF-), and granulocyte macrophage colony-stimulating factor (GM-CSF), whereas they produce little or no IFN- and relatively little IL-2 (6, 7). Moreover, allergen-specific T-cell clones derived from allergically inflamed tissue have been shown to express a Th2 phenotype (8, 9). In addition to T cells, mast cells and basophils may be an important source of IL-4 and may play an initiating role in the development of an allergic response (10). IL-4 is also involved in the selective tissue recruitment of eosinophils. Intradermal injection of IL-4 in mice resulted in marked eosinophil infiltration (14). Moreover, a murine model of asthma, based on the development of a Th2-type eosinophilic response to a soluble antigen preparation, showed that the IL-4 production correlated with eosinophil recruitment (15). When human umbilical vein endothelial cells (HUVEC) monolayers are used, it could be shown that IL-4 induces the expression of vascular cell adhesion molecule-1 on endothelial cells, by which it allows eosinophils, but not neutrophils, to attach to and migrate across the endothelium (16, 17). Recently, we demonstrated that in vitro IL-4 can also act directly on eosinophils from atopic donors by inducing a chemotactic response (18). In contrast, eosinophils from normal donors were unresponsive.

IL-4 mediates its biologic response through binding to the IL-4 receptor (IL-4R) complex, comprising an IL-4 binding chain and associated proteins. The diverse array of IL-4-mediated responses is reflected by a broad distribution of the IL-4R (2). Two subunits of the IL-4R have been identified: the IL-4R -chain (IL-4R) (19) and the -chain (c) of the IL-2R (20, 21). There is now substantial evidence indicating that the IL-4R is a common component of both the IL-4R and the IL-13R. A monoclonal antibody (mAb) against the IL-4R has been shown to block both IL-4- and IL-13-induced responses, indicating that the IL-4R is a component of both receptors (22, 23). Taken together, these findings suggest that the IL-4R can associate either with the c to form an IL-4R that does not bind IL-13, or with the IL-13R to form a functional IL-13 receptor. Whether this IL-13R also functionally binds IL-4 or only competes for the IL-4R has not yet been established. A consequence of IL-4 binding to the IL-4R would be signal transduction leading to a physiologic response. Engagement of the IL-4R complex results in the association and activation of signaling intermediates, such as 4PS, an insulin receptor substrate-1 (IRS-1)-like molecule (24- 26). Similar to IRS-1, 4PS shows a striking association with the p85 subunit of phosphoinositol-3 (PI-3) kinase after factor stimulation (27). We used this phenomenon to investigate functional responses of IL-4 and IL-13 in human eosinophils via the IL-4R.

In this study, we demonstrated the presence of an IL-4R on human peripheral blood eosinophils on the basis of fluorescent-activated cell sorter (FACS) and mRNA analysis. Furthermore, it was shown that both IL-4 and IL-13 induce activation of PI-3 kinase in eosinophils. Using a mAb against the IL-4R, we showed that the IL-4R is involved in both IL-4- and IL-13-induced PI-3 kinase activity. In addition, IL-4 was shown to increase eosinophil chemotaxis toward regulated on activation, normal T cells expressed and secreted (RANTES), but did not influence the chemotactic response toward platelet activating factor (PAF). Taken together, these findings indicate that human eosinophils express a functional IL-4R.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Blood Donors

All patients participating in this study had atopic dermatitis (AD), classified according to the criteria of Hanifin and Rajka (28). The patients all were allergic (i.e., they showed positive intracutaneous skin reactions to three or more different allergens), having elevated levels of total IgE and positive radio-allergosorbent tests for the relevant allergens. The patients had not taken oral steroids for at least 2 wk before the study. They only received small amounts of topical applied steroids. All other therapy was abandoned at least 2 wk before the study. At the time of blood collection most patients had moderate to mild eczema; locally active lesions were present. All patients had elevated blood eosinophil levels (over 4%). The normal healthy volunteers were not allergic, did not have increased blood eosinophilia, and did not take any kind of medication. All participating individuals gave their informed consent.

Reagents and Cytokines

Ficoll-paque and Percoll were obtained from Pharmacia (Uppsala, Sweden). PAF (1-O-hexadecyl-2-O-acetyl-sn-glycero-3-phosphorylcholine) was purchased from Sigma Chemical Co. (St. Louis, MO). Recombinant human (rh) GM-CSF (11.5 × 10⁴ U/mg) was a kind gift from Dr. G. Zenke (Novartis, Basel, Switzerland). rhIL-4 and rhIL-2 were kind gifts of Dr. F. Kalthoff (Novartis, Vienna, Austria). rhIL-13 was a kind gift of Dr. A. Minty (Sanofi, Labege, France). rhIL-5 was obtained from Amersham (Buckinghamshire, UK; 5 × 10⁶ U/mg). rhRANTES was obtained from PeproTech (Rocky Hill, NJ). Stock solutions of the cytokines were prepared in phosphate-buffered salt solution (PBS) supplemented with 0.1% purified human albumin and were stored at 70°C until use. The Escherichia coli bacteria, carrying a plasmid with the cDNA encoding the human IL-4R from the N-terminus (aa 1) up to aa 709 of the aa 800 receptor protein, was kindly provided by Dr. F. Kalthoff. All other chemicals were reagent grade.

Antibodies

Anti-CD16 mAb (CLB FcR gran 1) was purchased from the Central Laboratory of the Red Cross Blood Transfusion (Amsterdam, The Netherlands). Several mAbs against IL-4R were used. The mouse IgG1 anti-IL-4R mAb was obtained from Genzyme (Cambridge, UK). A second mouse IgG1 anti-IL-4R mAb, MR6, was a kind gift from Dr. E. P. Prens (Department of Immunology, EUR, Rotterdam, The Netherlands). A mouse IgG2a anti-IL-4R mAb and the anti-IL-2R mAb (mouse IgG1) were obtained from R&D Systems (Abingdon, UK). The isotype control (mouse IgG1) was purchased at Southern Biotechnology Associates (Birmingham, AL). The antiphosphotyrosine mAb (4G10, mouse IgG2bk) and the anti-p85 rabbit antiserum were obtained from UBI (Lake Placid, NY).

Cell Lines and Culture Conditions

Raji cells (B lymphoma) were cultured in RPMI 1640 containing 10% fetal calf serum (FCS), L-glutamine (2 mM), penicillin (100 IU/ml), and streptomycin (100 µg/ml). HUVEC were cultured in RPMI 1640 containing 20% heat- inactivated human serum, penicillin (100 IU/ml), streptomycin (100 µg/ml), and amphotericin B. The human microvascular endothelial cell line (HMEC-1) was cultured in basal endothelial medium (MCDB 131; Clonetics, San Diego, CA), supplemented with 10 ng/ml epidermal growth factor, 1 µg/ml hydrocortisone (Sigma), and 10% FCS. All cells were cultured at 37°C in a humidified atmosphere at 5% CO₂.

Isolation of Human Eosinophils

Blood was obtained from healthy volunteers or from AD patients. Eosinophils from the blood of normal donors were isolated from the buffy coat of 500 ml of blood, and eosinophils from the AD patients were isolated from 50 ml of blood anticoagulated with 0.4% (wt/vol) trisodium citrate (pH 7.4) as described previously (29). In short, the mononuclear cells were removed via separation of blood over isotonic Ficoll-paque (1.077 g/ml, pH 7.4). After isotonic lysis of the erythrocytes in an ice-cold solution containing 155 mM NH₄Cl, 10 mM KHCO₃, and 0.1 mM ethylenediaminetetraacetic acid (EDTA) (pH 7.2), the mixed granulocytes were washed and resuspended in RPMI supplemented with human serum albumin (1% wt/ vol) and incubated 30 min at 37°C to restore initial cell densities. After this incubation period, the cells were washed and resuspended in PBS solution supplemented with human serum albumin (1% wt/vol) and trisodium citrate (0.4% wt/vol). One milliliter of cell suspension (containing 80 × 10⁶ cells at the maximum) was layered on 4 ml of an isotonic Percoll solution (density: 1.082 g/ml). To prevent contamination of the cells with cell debris and remaining erythrocytes, 1 ml of Percoll (density: 1.100 g/ml) was brought under the Percoll 1.082 g/ml. After centrifugation (20 min, 1,000 g_max, room temperature), the eosinophil-rich fraction was collected from the interface. After washing with buffer, eosinophils were further purified from this eosinophil-rich fraction using the immunomagnetic bead method as described by Hansel and coworkers (30). In short, neutrophils present in the eosinophil-rich granulocyte preparation were coated with a monoclonal antibody against CD16 (1.2 µg/10⁷ cells/ml) during 30 min at 0°C. Hereafter, the cells were washed twice and subsequently coincubated head over head with beads (Dynal beads; Dynal A.S., Oslo, Norway) at a ratio of 1:2 (cells:beads) for 20 min at 4°C. The neutrophils were subsequently removed by a magnetic particle concentrator (MPC^TH-1; Dynal A.S.). The eosinophils were washed with buffer and suspended in the buffer solution to be used for consecutive experiments. Eosinophil purity was always > 95% and the viability was over 98%.

Stimulation of Eosinophils

Freshly isolated eosinophils were suspended in culture medium at a concentration of 1 to 2 × 10⁶ cells/ml. The complete medium for eosinophil cell culture consisted of RPMI 1640 (GIBCO, Grand Island, NY) supplemented with 10% (vol/vol) heat-inactivated FCS (GIBCO), penicillin (100 IU/ml), and streptomycin (100 µg/ml). Cells were cultured overnight in the absence or presence of different stimuli in 5% CO₂ in a humidified atmosphere at 37°C. The viability of the cultured eosinophils was > 90%, as assessed by trypan blue exclusion.

Flow Cytometric Analysis

Staining for flow cytometric analysis was performed using 2 × 10⁵ cells in a final volume of 50 µl PBS with 2% FCS and 0.05% NaN₃ (PBS-CFG). Cells were incubated with the indicated mAb for 20 min at 4°C. After the cells were washed three times in 200 ml PBS-CFG, they were further incubated with a 1:50 dilution of fluorescein isothiocyanate-conjugated rabbit antimouse immunoglobulins (DAKO, Glostrup, Denmark) in PBS-CFG for 20 min at 4°C. After the cells were washed three times, they were fixed in a 2% p-formaldehyde solution and subsequently analyzed for binding of the antibody by flow cytometry using a FACSstar plus (Becton Dickinson, Mountain View, CA). Nonspecific fluorescence was determined by incubation cells with mouse IgG of the same isotype but with irrelevant antigen specificity. Data analysis was done by the program PC Lysys (Becton Dickinson).

Total RNA Isolation

Total RNA was extracted from 50 (AD patients) or 20 (normal donor) × 10⁶ (Northern blot analysis), or 5 to 10 × 10⁶ reverse-transcriptase/polymerase chain reaction (RT-PCR) eosinophils using the following procedure: cells were lysed in 1 ml extraction buffer/10⁷ cells containing 0.2 M Tris-HCl (pH 8.5), 0.25 M NaCl, 0.05 M EDTA, p-aminosalicylic acid (48 g/liter), and tri-isopropylnaphtalene sulfonic acid (8 g/liter). DNA and protein were removed by phenol/chloroform/isoamylalcohol (50/48/2) extraction. The RNA was precipitated overnight by adding 0.1 part 3 M NaAc (pH 5.6) and 2 parts ice-cold absolute ethanol. The samples were dissolved in 20 µl 10 mM Tris buffer (pH 8.0) containing 0.1 mM EDTA, after which the total RNA content was measured spectrophotometrically. Finally, 40 U ribonuclease (RNase) inhibitor (Boehringer Mannheim, Almere, The Netherlands) was added and the samples were stored at 70°C until use.

Northern Blot Analysis

The total amount of RNA from each cell type was fractionated on a 1% agarose-formaldehyde gel for electrophoresis. The gel was blotted onto Hybond N filters (Amersham) and the RNA was crosslinked to the filter by exposure to ultraviolet light for 2 min. The filter was hybridized with a ³²P-labeled C-terminally truncated version of the full-length cDNA of the IL-4R. The filter was then washed twice at room temperature in 300 mM NaCl, 30 mM sodium citrate, 0.5% sodium pyrophosphate, and 1% sodium lauryl sarkosine (2× saline sodium citrate [SSC]), once at 55°C in 2× SSC, and twice at 63°C in 0.2× SSC with 2% sodium dodecyl sulfate. The hybridization was then visualized by autoradiography.

Reverse Transcriptase and Polymerase Chain Reaction

Aliquots of RNA were reverse transcribed in a final volume of 20 µl. The reverse transcription mix contained 20 U avian myeloblastosis virus reverse transcriptase (RT) (Promega), 2 µl 10× RT buffer (100 mM Tris/HCl, 500 mM KCl; pH 8.3), 4 µl MgCl₂ (25 mM), 2 µl dNTP mix (10 mM each), 25 U RNase inhibitor, 0.001% gelatine, and 1 mM oligo(dT) primer. The mix was incubated at room temperature for 10 min, followed by 42°C for 90 min. Following reverse transcription, 5 µl 10× PCR buffer (100 mM Tris/HCl, pH 8.9; 1M KCl; 15 mM MgCl₂; 0.5 mg/ml BSA; 0.5% [vol/vol] Tween 20), 2U Thermus thermophilus HB8 (Boehringer Mannheim), and 1 mM of each primer were added to each sample to a final volume of 50 µl. The IL-4R primer sequences were 5'-TCT CTA CTT GCG AGT GGA AGA TGA ATG GTC-3' (forward primer) and 5'-CCT GAG CAT CCT GGA TTA TTA TAG CCA CG-3' (reverse primer), which span a 729-bp region of the IL-4R cDNA. The samples were overlaid with 100 µl mineral oil and transferred to a DNA Thermal Cycler (Perkin Elmer, Branchburg, NJ). After an initial denaturation step at 94°C for 2 min, the cDNA was amplified with 35 cycles under the following conditions; 94°C for 1 min followed by 60°C for 1 min, and 72°C for 90 s. The final extension step was performed at 72°C for 7 min. A 20-µl aliquot of the PCR reaction product was separated by electrophoresis in a 1% agarose gel and transferred onto Hybond N⁺ filters (Amersham). The PCR product was detected by hybridization with a peroxidase-labeled C-terminally truncated version of the full-length cDNA of the IL-4R and chemiluminescence development according to the manufacturer's instructions (Boehringer Mannheim).

Measurement of IL-4- and IL-13-Induced PI-3 Kinase Activity Human Eosinophils

Eosinophils (4 × 10⁶/sample) were stimulated with a concentration range (10¹¹ to 10⁸ M) IL-4 or IL-13 for 5 min at 37°C. Reactions were stopped by adding two volumes of ice-cold incubation buffer containing 2 mM Na₃VO₄. Subsequently, the eosinophils were pelleted by centrifugation at 4°C. Hereafter, the cells were resuspended in lysis buffer (1% Triton X-100, 20 mM Tris/HCl, 100 mM NaCl, 10 mM Na₄P₂O₇, 2 mM EDTA, 50 mM NaF, 10% glycerol, 10 µg/ ml aprotinin, 10 µg/ml leupeptin, 10 µg/ml soybean tryptase inhibitor, 1 mM phenylmethylsulfonyl fluoride, and 1 mM Na₃VO₄, pH 8.0) for 30 min on ice. After 30 min, detergent-insoluble material was removed by centrifugation for 10 min at 14,000 rpm at 4°C. Lysates were treated for 1 h at 4°C with the antiphosphotyrosine mAb 4G10 (2.5 µg/ml). Hereafter, protein A sepharose (Pharmacia) was added for another hour and, subsequently, the protein sepharose beads were washed three times with lysis buffer and two times with 10 mM Tris-HCl, pH 7.4, containing 1 mM Na₃VO₄. PI-3 kinase activity was measured by adding 100 µg of sonicated PI and 20 µCi of (-³²P) adenosine triphosphate (ATP) (ICN, Costa Mesa, CA) in the presence of 200 µM adenosine to inhibit PI-4 kinase activity, 30 mM MgCl₂, and 35 µM ATP in a volume of 60 µl. Reactions were carried out for 20 min at room temperature and stopped by addition of 100 µl 1 M HCl and 200 µl chloroform:methanol (1:1 vol/vol). After centrifugation and removal of the upper layer, 80 µl methanol/HCl was added. After centrifugation, lipids were separated on thin layer chromatography (TLC) plates (Merck) using a solvent system of chloroform:methanol:ammoniumhydroxide (45:35: 10 vol/vol/vol). TLC plates were exposed to X-ray film at 80°C. Immunoprecipitation with polyclonal anti-p85 antibody was used as a positive control for PI-3 kinase activity.

Chemotaxis Assay

Migration was measured with a modified Boyden chamber assay using a 48-well microchemotaxis chamber (Neuroprobe, Cabin John, MD). Chemotaxins or N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid (Hepes) buffer (30 µl) were placed in the lower compartments. Two filters (cellulose nitrate) were placed between the lower and the upper compartments. The lower filter had a pore width of 0.45 µm (Millipore type HA; Millipore Corporation, Bedford, MA), and the upper filter had a pore width of 8 µm (Sartorius, SM 113; Sartorius AG, Göttingen, Germany). Before use, the filters were soaked in the Hepes buffer. Purified eosinophils were placed in the upper compartment (25 µl of 5 × 10⁶ cells/ml). The chemotaxis chambers were subsequently incubated for 2.5 h at 37°C, unless otherwise stated. Hereafter, the upper filters were removed, fixed in butanol/ethanol (20%/80%, vol/vol) for 10 min, and stained with Weigert solution (1% hematoxylin [vol/ vol] in 95% ethanol [vol/vol] and an acidic FeCl₃-solution [70 mM] mixed in a volume ratio of 1:1). The filters were dehydrated with ethanol, made transparant with xylene, and fixed upside down. The number of cells per 10 high-power fields (hpf) was determined with light microscopy (magnification ×400). In this way, the number of cells that had passed the upper filter was determined.

Statistical Analysis

All data are presented as means ± SEM. Student's t test for paired or unpaired data was applied. P values < 0.05 were considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Demonstration of Surface Expression of IL-4R by Flow Cytometry

Because a functional IL-4R can consist of an IL-4 specific -chain (IL-4R) and the IL-2R c, expression of both chains on human eosinophils was investigated using flow cytometric analysis. Figure 1A shows an illustrative example of IL-4R and c expression on eosinophils from a normal donor. All monoclonal antibodies against the IL-4R gave similar results (data not shown). The HUVEC and the microvascular endothelial cell line (HMEC-1) were used as control cells. As shown in Figure 1B, HUVEC only expressed detectable levels of the IL-4R. HMEC-1 cells expressed no detectable levels of either IL-4R or c (Figure 1C). No differences in IL-4R expression were observed between eosinophils from normal individuals and AD patients (Table 1). To investigate whether the IL-4R expression might be immunologically regulated, receptor expression was measured on eosinophils after overnight culture with medium or cytokines known to be able to activate eosinophils in vitro. The cytokines assessed for their effect on the IL-4R expression were IL-2 (1 nM), IL-4 (1 nM), IL-5 (10 pM), IL-3 (10 pM), GM-CSF (10 pM), and IFN- (1 nM). Neither of the cytokines tested had a significant effect on the expression of the IL-4R on eosinophils from normal individuals (Table 1), indicating a constitutional expression of the IL-4R.

View larger version (30K): [in this window] [in a new window]   Figure 1.   Surface expression of IL-4R and c on (A) normal eosinophils (illustrative example of eight donors), (B) HUVEC, and (C) the microvascular cell line HMEC-1. Histograms of flow cytometric analysis with an anti-IL-4R mAb (open histogram) as compared with an IgG1 isotype control (closed histogram).

Demonstration of mRNA for the IL-4R by Northern Blot Analysis and RT-PCR

To investigate further the presence of an IL-4R, the presence of mRNA for the IL-4R in eosinophils from both normal individuals and AD patients was examined by Northern blot analysis. Raji B lymphoma cells, known to express the IL-4R, were used as a positive control (Figure 2A, lane 1). Figure 2A shows the Northern blot analysis of RNA from freshly isolated eosinophils from two normal donors (20 × 10⁶ cells, lanes 2 and 3) and two AD patients (50 × 10⁶ cells, lanes 3 and 4). The experiments shown are illustrative for a total of five normal individuals, all of whom expressed low to undetectable levels of mRNA for the IL-4R, and six AD patients, all of whom expressed mRNA for the IL-4R. Because of the limited amount of eosinophils available from normal donors, the Northern blot analysis did not convincingly demonstrate the presence of mRNA for the IL-4R. Therefore, we also performed a RT-PCR using specific primers for the IL-4R. The specificity of the IL-4R fragment was confirmed by Southern blot analysis (Figure 2B). Raji B lymphoma cells were used as a positive control (lane 1). As shown in Figure 2B (lanes 2 and 3), eosinophils from normal donors express mRNA for the IL-4R. The experiments shown are representative for a total of six normal donors, all of whom expressed mRNA for the IL-4R.

View larger version (45K): [in this window] [in a new window]   Figure 2.   Expression of mRNA for the IL-4R by Raji B lymphoma cells and human eosinophils. (A) Total RNA extracted from Raji B lymphoma cells (10 × 106 cells, lane 1), eosinophils from normal individuals (20 × 106 cells, lanes 2 and 3), and eosinophils from AD patients (50 × 106 cells, lanes 4 and 5) were analyzed by Northern blot analysis and probed with a C-terminally truncated version of the full-length cDNA of the IL-4R. (B) Southern blot analysis, probed with a C-terminally truncated version of the full-length cDNA of the IL-4R, of RT-PCR products obtained from RNA extracted from Raji B lymphoma cells (5 × 106 cells, lane 1) and eosinophils from normal donors (10 × 106 cells, lanes 2 and 3). The results shown are an illustrative example of four donors.

Induction of PI-3 Kinase Activity by IL-4 and IL-13 in Human Eosinophils

Binding IL-4 to the IL-4R complex results in the association and activation of various signaling intermediates, such as the IRS-1 (31). 4PS, an IRS-1-related protein, has shown to be involved in both IL-4- and IL-13-mediated signal transduction and strongly associates with the p85 subunit of PI-3 kinase (32).

Here, we studied activation of PI-3 kinase in human eosinophils by IL-4 and IL-13. Incubation of human eosinophils with a concentration range (10⁸ to 10¹¹ M) of IL-4 or IL-13 resulted in a dose-dependent induction of PI-3 kinase activity (Figure 3). To investigate whether the IL-4R is involved in the induction of PI-3 kinase activity by IL-4 and possibly IL-13, eosinophils were incubated with an mAb against the IL-4R prior to activation. As shown in Figure 4, preincubation with the mAb resulted in a marked decrease of IL-4-induced PI-3 kinase activity compared with nontreated cells (lanes 2 and 3). Also, the IL-13- induced PI-3 kinase activity is greatly reduced by preincubation with the mAb (lanes 4 and 5). As a control, IL-5- induced PI-3 kinase activity in eosinophils was determined after preincubation with the mAb. As shown in Figure 4 (lanes 6 and 7), IL-5-induced PI-3 kinase activity was not affected by the anti-IL-4R mAb.

View larger version (54K): [in this window] [in a new window]   Figure 3.   The induction of PI-3 kinase activity in human eosinophils by IL-4 and IL-13. Eosinophils (4 × 106 cells/sample) were untreated (lane 1) or treated with a concentration range (1011 to 108 M) of IL-4 (lanes 2 to 5) or IL-13 (lanes 6 to 9) for 5 min at 37°C. Subsequently, antiphosphotyrosine immunoprecipitates were analyzed for PI-3 kinase activity in an in vitro kinase assay using phosphotidylinositol as a substrate and [-32P]ATP as a phosphate donor (see also MATERIALS AND METHODS). Immunoprecipitation with an anti-p85 antiserum of lysates from untreated cells was used as a positive control (lane 1). The results shown are an illustrative example of four donors.

View larger version (40K): [in this window] [in a new window]   Figure 4.   Involvement of the IL-4R in the induction of PI-3 kinase activity by both IL-4 and IL-13. Eosinophils (4 × 106 cells/ sample) were left untreated () or preincubated (+) with an anti-IL-4R mAb (25 µg/ml) for 30 min at 37°C prior to stimulation with IL-4 (109 M, lanes 2 and 3), IL-13 (109 M, lanes 4 and 5), or IL-5 (109 M, lanes 6 and 7) for 5 min at 37°C. Untreated and unstimulated eosinophils were used as a control (lane 1). After stimulation, cell lysis, and immunoprecipitation, PI-3 kinase activity was detected. The results shown are an illustrative example of three donors.

Priming of Chemotactic Responses by IL-4

To investigate whether IL-4 could preactivate the chemotactic responses toward PAF and RANTES, eosinophils were preincubated overnight with IL-4 (10⁹ M). As shown in Table 2, IL-4 significantly increased the chemotactic response toward RANTES. Interestingly, overnight preincubation with IL-4 did not influence the chemotactic response toward PAF (Table 2).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Human eosinophils express specific cell surface membrane receptors for a number of immunoregulatory cytokines. Receptors for the cytokine cluster IL-3, IL-5, and GM-CSF are the most extensively studied receptors expressed on eosinophils (33, 34), but receptors for IL-2 (CD25) (35), TNF- (36), and recently IFN- (37) have also been described. In the present study, we have examined IL-4R expression on human eosinophils. Our data demonstrate that human eosinophils express both IL-4R mRNA and membrane-associated IL-4R. The IL-4R appears to be expressed constitutively and receptor expression was not influenced by overnight incubation with cytokines such as IL-5, GM-CSF, IL-2, and IFN-. This is further substantiated by our finding that eosinophils from AD patients and normal donors express similar levels of IL-4R, although previous studies have shown that eosinophils from AD patients are preactivated in vivo (18, 38, 39). Furthermore, we have demonstrated that both IL-4 and IL-13 induce the activation of PI-3 kinase in human eosinophils in a concentration-dependent manner. When a mAb was used against the IL-4R, the observed effects of both IL-4 and IL-13 were markedly reduced. In contrast, IL-5-induced PI-3 kinase activity was not affected. These data are the first evidence for the expression of a functional IL-4R on human eosinophils.

The diversity of biologic effects mediated by IL-4 is reflected by the broad distribution of its receptor. A wide variety of hematopoietic and nonhematopoietic cell types expresses the IL-4R, ranging from a few hundred to several thousand binding sites per cell. The IL-4R was initially described as a single class of a high-affinity receptor complex comprising the IL-4 binding chain and associated proteins. The IL-4R alone is sufficient for IL-4 binding with high affinity (K_a = 10¹⁰ M¹). Crosslinking studies have shown that the IL-2 receptor c associates with the IL-4R after the binding of IL-4 (20). Association of the IL-4R with the c promotes only a modest increase of the affinity for IL-4. In addition to forming a heterodimer with the c, the IL-4R subunit may also be employed by other receptor complexes. For example, the IL-13R complex apparently consists of the IL-4R and an IL-13-specific component that binds IL-13 (40). Whether this receptor complex also has binding capacities for IL-4 has not yet been established. The IL-4R may also associate with another as yet unidentified subunit (41). In addition, a recent study has proposed a homomeric IL-4R complex in which no additional subunits are necessary (42). The exact nature of the IL-4R complex expressed on human eosinophils is far from being completely delineated. So far, we have not been able to demonstrate specific binding of [¹²⁵I]IL-4 to human eosinophils, probably because of low receptor expression (unpublished observation). However, our data show that eosinophils express the IL-4R, which is a crucial component for any IL-4 binding complex and signal transduction.

What signal transduction pathway is activated depends predominantly on the arrangement of the IL-4R complex. Recently, we have shown that several chemoattractants and both IL-4 and IL-5 induce PI-3 kinase activity and downstream PKB/cAkt in human eosinophils (43). Here, we extend these findings by showing that IL-13 also induces PI-3 kinase activity and that both IL-4 and IL-13 use the IL-4R for ligand binding. The IL-4R is known to associate with and activate several signaling intermediates. Engagement of the IL-4R complex, for example, results in the phosphorylation of the IL-4 protein substrate 4PS. 4PS has recently been cloned and named IRS-2. The IRS-2 shares functional and structural characteristics with the IRS-1, including a striking association with the p85 subunit of the PI-3 kinase after stimulation (24). In the present study, PI-3 kinase activity was demonstrated in antiphosphotyrosine immunoprecipitates derived from lysates from both IL-4- and IL-13-stimulated eosinophils. Because the association of p85 with IRS-2 does not necessarily involve tyrosine phosphorylation of p85 itself, recruitment of PI-3 kinase to signaling complexes, most likely IRS-2, was measured. Further studies may elucidate the exact structure of this signaling complex in human eosinophils. A second pathway known to be activated by IL-4 is the Janus kinases (JAK) signal transducers and activators of transcription (STAT) pathway. The IL-4R associates with and activates JAK1 upon ligand binding. In combination with the c, which is able to associate with JAK3, STAT6 is activated (44). Both components, JAK1 and JAK3, appear to be required for IL-4-induced activation of STAT6. That IL-13 can activate STAT6 without activation of JAK3 implies that involvement of the IL-4R is more important than the c and activation of JAK3. So far, we have not been able to demonstrate phosphorylation of JAK1 or STAT6 in human eosinophils after IL-4 stimulation, although JAK1 itself could be immunoprecipitated (unpublished observation). Whether this is due to insufficient amounts of eosinophils in combination with low phosphorylation levels of JAK1 needs to be investigated further.

Activation of eosinophils in vivo is an important phenomenon observed in patients with an allergic inflammation. It renders eosinophils from atopic donors more responsive to various stimuli than are eosinophils from normal donors (18, 38, 39). Eosinophils from atopic donors, for example, exhibit an increased chemotactic responsiveness to a range of chemoattractants, such as fMLP, PAF, and platelet factor 4, compared with eosinophils from normal individuals. Because increased responsiveness toward different stimuli could be induced in eosinophils from normal individuals by pretreatment with IL-3, IL-5, or GM-CSF, it has been proposed that these cytokines are responsible for the in vivo priming of eosinophils during the allergic inflammation. The results described here show that similar to these cytokines, IL-4 also has the capacity to increase functional responses in human eosinophils, indicated by increased chemotactic responsiveness toward RANTES. The priming phenomenon induced by IL-4 appears to be more restricted because, in contrast to IL-5, IL-4 did not influence the chemotactic response of eosinophils toward PAF.

The role of IL-4 in the allergic inflammatory reaction is well established, and there are numerous immunopathologic correlations between IL-4 and eosinophils. However, the immunologic role of the IL-4R on eosinophils is not yet clear. The association of elevated levels of IgE and increased eosinophil counts in the peripheral blood of patients with an allergic disorder suggest that eosinophils may be exposed to IL-4 in vivo. Recently, we (unpublished paper) and others (47) have shown that IL-4 is able to upregulate mRNA for the high-affinity IgE receptor (Fc RI) -chain in human eosinophils. Furthermore, it has been shown that IL-4 induces binding of IgA-coated beads, whereas the binding of IgG-coated beads is not influenced by IL-4 (48). In contrast, IL-5 increased the binding of both IgA- and IgG-coated beads. These data clearly show that IL-4 is able to induce several functional responses in eosinophils. In addition, IL-13 has recently been described to activate human eosinophils by the induction of CD69 (49). Another phenomenon in which IL-4-induced activation of human eosinophils may be important is the observation that, during the course of the atopy patch test, there is a shift in the ratio IL-4/IFN- in favor of IFN-. As a consequence, the Th phenotype seems to change from a Th2 phenotype to a Th1 phenotype. Recently, Grewe and colleagues have shown that IL-4-induced activation of eosinophils leads to secretion of considerable amounts of IL-12 (50). Because IL-12 induces the production of IFN- in Th1-like lymphocytes, eosinophils could modulate the immune response via IL-4-induced activation.

Taken together, our data show the presence of an IL-4R on human eosinophils and provide a mechanism by which eosinophils can be stimulated with IL-4 coordinately. This may be an important pathway through which the eosinophil modulates the immune response.