Interleukin-4 (IL-4) enhances and soluble interleukin-4 receptor (sIL-4R) inhibits histamine release from peripheral blood basophils and mast cells in vitro and in vivo

The aim of the study was to analyse the effect of interleukin-4 (IL-4) on allergen and anti-IgE mediated histamine release from basophils and human skin mast cells and to assess whether soluble recombinant interleukin-4 receptor (sIL4R) can inhibit these effects. Anti-IgE stimulated histamine release from peripheral blood basophils and mast cells of atopic donors was enhanced after preincubation with IL-4, whereas after preincubation with sIL-4R it was inhibited. These effects were even more pronounced when samples were stimulated with a clinically relevant allergen. In IL-4 preincubated skin mast cells, there was a similar enhancement of anti-IgE stimulated histamine release, which could again be inhibited by sIL-4R. The effects of IL-4 and sIL4R were dose- and time-dependent. Mice sensitized to ovalbumin and treated with soluble recombinant murine sIL-4R showed significantly reduced immediate-type cutaneous hypersensitivity responses compared with untreated mice. These in vivo effects were IgE independent, since there were no significant differences in total and allergen specific IgE/IgG1 antibody titres between treated and untreated mice. This indicates that IL4 exerts priming effects on histamine release by effector cells of the allergic response and that these effects are potently antagonized by soluble IL-4R both in vitro and in vivo.


Introduction
Interleukin-4 (IL-4) plays a central role in allergic in ammation. 1 It acts as an important immunoglobulin isotype switch factor for IgE synthesis and is required for initiation and promotion of a TH2 T-cell response. 2 5 IL-4 stimulates expression of endothelial cell adhesion molecules on lymphocytes, 6 eosinophils, basophils and mast cells. 7 9 IL-4 production is increased in atopic dermatitis patients 10 and IgE synthesis induced by IL-4 is enhanced during seasonal allergen exposure. 11 While there is clear evidence of IL-4 production by T-lymphocytes, some reports raise the possibility of IL-4 production by human mast cells 12,13 and basophils. 14 Furthermore, human mast cells express IL-4 receptors. 15 Mast cells and basophils represent important effector cells of the immediate-type allergic response which is characterized by the release of histamine and other mediators upon cross-linking of receptor bound IgE molecules by allergens. The expression of IL-4 receptors on these cells suggests that mast cells and basophils are sensitive to this cytokine. Since little is known about the interaction of IL-4 with effector cells, we examined the potential role of IL-4 on allergen and anti-IgE triggered histamine release from human basophils and skin mast cells. We examined further whether immunomodulation with an IL-4 antagonist would inhibit the IL-4 mediated effects. It has recently been shown that a recombinant soluble IL-4 receptor can inhibit many of the IL-4 mediated cell functions both in vitro and in vivo. 16,17 To examine the biological relevance of these ndings, allergen sensitized mice were treated with a murine recombinant soluble IL-4 receptor and the immediate skin response was assessed.

Patients
For experiments on basophils, blood was drawn from 11 adult atopic patients (six women, ve men), aged 20± 44 years (mean age 33 years).
Total IgE ranged from 45 to 6650 kU/l (median 429 kU/l). Patients were sensitive to birch pollen (Betul a verrucos a) (Bet v) and/or Derm atoph agoides pteronyssinus (Der p). Sensitization was proven by speci c IgE to birch and/ or house-dust mite in serum (Pharmacia CAPsystem, Uppsala, Sweden). Blood was taken over all the year except during the seasons April/May for birch-and September/October for house dust mite-sensitive patients. All patients suffered from allergic rhinoconjunctivitis, one patient from atopic dermatitis as well. All patients were without any anti-allergic medication for at least 4 weeks before drawing blood for the experiments. No patient had any infection during 2 weeks before the study.
Mast cells were prepared from normal human breast skin of six healthy female donors, aged 21± 45 years (mean age 29 years), with no history of atopy who were undergoing plastic surgery for cosmetic reasons. All patients gave their informed consent. The study was approved by the local ethics committee.

Human and murine recombinant soluble IL-4 receptor
Soluble IL-4 receptor was prepared as previously described. 18 Brie y, using cDNA for the extracellular region of human and murine IL-4R, 19,20 recombinant soluble monomeric forms of IL-4R (human, rhuIL-4R and murine, muIL-4R) were constructed. The murine and human forms of rIL-4R were expressed in BHK cells.
The proteins were puri ed by af nity chromatography using speci c antibodies. Bioactivity, glycosylation and pharmacokinetics in mice were controlled for each lot of the molecules to achieve an uniform reactivity in vitro and in vivo.

Preparation of basophil cultures from peripheral blood
Basophils were enriched and cultivated as previously described. 21 Brie y, peripheral blood leukocytes were obtained by spontaneous sedimentation for 60 min at room temperature. The percentage of viable basophils within the leukocyte preparation ranged between 6% and 11%. Cells were washed twice with Pipes A/EDTA and centrigued for 10 min. 1 3 10 7 leukocytes were placed in 4 ml plastic tubes (Sarstedt, Nü mbrecht, Germany) in Pipes EDTA buffer. Cells were preincubated with IL-4 (100 to 5000 IU/ml), sIL-4R (1 to 200 mg/ml) or buffer in a water bath at 378 C for 2.5 h. After decanting supernatants and washing cells in Pipes A, Pipes ACMwas added and cells were challenged with anti-IgE (30 and 100 IU/ml), Der p (30 and 100 SQ/ml) or Bet v (22 and 220 ng/ml). After a 30 min incubation at 378 C, cells were centrifuged for 15 min and supernatants were stored at 208 C until determination of histamine. All measurements were performed in duplicate.

Preparation of human skin mast cells cultures
Mast cells were isolated by enzymatic dispersion, with slight modi cation as previously described. 22 Brie y, skin was cut into small squares (1± 2 mm 2 ), dispersed in two 1-h cycles with collagenase and hyaluronidase (15 mg and 7.5 mg/g tissue respectively, 2.5 ml medium/g tissue). Undissociated tissue was removed by ltration with nylon gauze (150 mm) and thereafter the cells were washed twice in Pipes A. Viability of mast cells was determined by trypan blue staining and mast cell purity was assessed by toluidine blue staining. Viability of mast cells was . 95%. This procedure yielded 6 8 3 10 5 mast cells/g wet tissue with 3± 5%mast cells of total nucleated cells. After collection of the dissociated cells, these were washed twice with Pipes A and then passively sensitized with IgE (1 mg/10 6 cells) at room temperature overnight. The cells were then incubated in Pipes ACM with 10± 1000 ng/ml IL-4 for 5± 15 min at 378 C in a water bath, or overnight in medium in an incubator. Aliquots of 10 4 mast cells in 0.4 ml pipes ACM were challenged in duplicate with anti-IgE (1000, 2000 and 4000 IU/ml, with 0.417 IU precipitating 1 ng IgE WHO standard).

Measurement of histamine release
Spontaneous histamine release, which was determined by the addition of Pipes ACM instead of stimuli, ranged between 5% and 10% of total histamine content. Total cellular histamine content was assayed by lysis of cell aliquots with 2%perchloric acid. After a 20 min incubation at 378 C, cells were centrifuged and supernatants were stored at 208 C until analysed.
Histamine was determined with an automated uorometric analyser (Technicon, New York, USA) using a procedure previously described. 23 Results were only considered where spontaneous histamine release was less than 10% of total histamine content. More than 80% of our experiments ful lled the criteria. For each result, spontaneous histamine release (incubation in Pipes ACM buffer) was subtracted and calculated as a percentage of total histamine content ( 100%) for each culture condition. The measurements were performed in duplicate.
Allergic sensitization of mice BALB/c mice (aged 6± 10 weeks, from Bomholtgard, Denmark) were sensitized to OVA by nebulization of 1% ovalbumin (OVA) diluted in PBS. Sensitization was performed by exposure for 20 min every seventh day. As previously shown 24 and con rmed in this study, this procedure stimulated production of allergenspeci c IgE/IgG1. After 4 weeks, mice were treated ve times every other day by 20 min nebulization of 1 mg sIL-4R diluted in 7 ml PBS. Skin testing was performed 24 h after the last treatment. The study was approved by the local Animal Ethics Committee.

Assessment of immediate-type skin tests response in mice
Intradermal skin tests (ST) were performed as previously described. 25 Brie y, abdominal skin was shaved and 50 ml of test solution was injected intradermally. PBS was used as negative control and compound 48/80 (Sigma, Mü nchen, Germany) as positive control. Injection points were separated by at least 1.5 cm in order to avoid con uence of solutions. Allergen concentration was 50 mg/ml for ovalbumin. Resultant wheal formations were scored in a blinded fashion after 15± 20 min by semiquantitative assessment of skin test reactivity. Reactions with a diameter < 1 mm were scored with 0 points, 2± 4 mm with 1 point and reactions > 5 mm with 2 points. All animals developed positive wheal formations to compound 48/80, none reacted to PBS.
Measurement of total and allergen-speci® c IgE/IgG1 antibody titres in serum samples Analysis of total IgE and anti-OVA IgE/IgG1 antibody titres were determined by ELISA. Brie y, 96-well U-bottom, polystyrene microtitre plates (Greiner, Nürtingen, Germany) were coated either with anti-mouse IgE (10 mg/ml) (Pharmingen, San Diego, USA) or ovalbumin (20 mg/ml) Sigma, Deisenhofen, Germany) diluted in 0.05 M carbonate coating buffer, pH 9.6. Plates were incubated overnight, washed three times in washing buffer (PBS/0.05% plus Tween 20, pH 8.2) and blocked for 2 h at 378 C with blocking solution (2% BSA in PBS). After three washes with washing buffer, samples diluted in PBS/0.4%BSA were incubated for 2 h at 378 C. Washed plates were incubated with biotin-conjugated anti-mouse IgE/IgG1 monoclonal antibodies for 2 h at 378 C (2.5 mg/ml) (Pharmingen, San Diego, USA) followed by incubation with alkaline phosphate-conjugated streptavidin (1:500) (Jackson Immuno Research Laboratories, West Grove, USA) for 1 h at 378 C. Plates were developed with p-NPP substrate (0.5 mg/ml) (Sigma, Deisenhofen, Germany) and absorbance was measured at 405 nm.

Statistical analysis
For statistical analysis, we used the Wilcoxon test for paired data, and Mann-Whitney U-test. For all numbers, standard error of mean is given. P values , 0 05 were considered as statistically signi cant.

Modulation of histamine release by human basophils
In order to investigate the effect of IL-4 on histamine release from human basophils, cells were stimulated with anti-IgE or the clinically relevant allergens. After preincubation with buffer alone, both anti-IgE and allergen enhanced histamine release in a dose-dependent manner, with a peak release at 100 IU/ml for anti-IgE (percentage of total histamine over background release 32 8% 4%). The optimal concentration of allergen ranged at 22 ng/ml for Bet v (percentage of total histamine over background release 40 8% 4%) and 100 standard quantity units (SQU)/ml for Der p (percentage of total histamine over background release 27 0% 3%). Using concentrations lower than 22 ng/ml Bet v or 100 SQU/ml Der p, less pronounced histamine release was measured, which was not signi cantly different from controls (data not shown).
Cells were preincubated with IL-4 for 2.5 h followed by stimulation with optimal doses of anti-IgE and allergen. IL-4 enhanced anti-IgE triggered histamine release by 9 5% 1 4% (Fig.  1). However, IL-4 by itself was unable to induce histamine release in the absence of any stimulus. A more pronounced effect was observed on allergen-triggered mediator release. Increases of histamine release by up to 35%were measured, with a percentage total histamine over background release of 19 0% 6%. The reason why we converted the data into values of percentage stimulation/inhibition was the interindividual variability of the patients. We did not nd major differences between Bet v and Der p effects in our culture system. For this reason, and because the individual groups would have been too small, we decided to present the results of both allergens together in one gure.
Since histamine release by peripheral blood basophils was analysed in the presence of cell subsets capable of secreting IL-4, and since enhanced IL-4 production has been demonstrated by peripheral blood lymphocytes from atopic donors, 10,11,26 it was necessary to determine whether intrinsic ongoing IL-4 production contributed to the priming effect observed in the assay. A soluble recombinant human IL-4 receptor (sIL-4R) was added to cultures prior to stimulation with anti-IgE or allergen. Histamine release in peripheral blood basophils stimulated with anti-IgE was inhibited after preincubation with sIL-4R by 7 9 6 5% (Fig. 1). An even more pronounced reduction was measured when cells were stimulated with allergen (Der p): sIL-4R inhibited histamine release by up to 20%, with a mean reduction of 11 3 3 2% (Fig. 1). This data provides evidence that intrinsic ongoing IL-4 secretion by peripheral blood leukocytes exerts priming effects on histamine release by basophils.
Dose-response experiments showed that the effects of IL-4 ( Fig. 2A) and sIL-4R (Fig. 2B) on both anti-IgE and allergen induced histamine release were dependent on the concentration. However, despite a clear tendency, results failed to reach statistical signi cance due to small numbers in some concentrations. Maximal augmentation of histamine release was found following preincubation with more than 1000 U/ml IL-4; sIL-4R at a concentration of . 100 mg ml resulted in optimal inhibition of histamine release. To assess the speci city of these effects, cells were preincubated with a murine recombinant soluble IL-4R that does not bind to human IL-4. Incubation of leukocytes with murine sIL-4R in the same dose range was without any effects on anti-IgE or allergen-induced mediator release (data not shown).
Optimal effects were achieved when cells were preincubated with IL-4 and sIL-4R for about 2.5 h. Preincubation for a shorter period decreased the priming effect of IL-4 (data not shown). When cells were preincubated for a period longer than 3.5 h, spontaneous histamine release signi cantly (P , 0 05) increased and interfered with the assessment of IL-4/sIL-4R effects.

Modulation of histamine release from human skin mast cells
To analyse whether IL-4 would also exert priming effects on other effector cells of the allergic reaction, we examined the effect of IL-4 on mast cells prepared from human non-atopic skin. Mast cells were presensitized with human IgE and in vitro histamine release was measured following stimulation with anti-IgE. There was a signi cant ( P , 0 05) enhancement of anti-IgEtriggered histamine release (from 4 5% 1 9%to 8 3% 1 9%) following preincubation with IL-4 (Fig. 3). This could be blocked by simultaneous addition of sIL-4R (n 2, data not shown).

Inhibition of immediate-type cutaneous hypersensitivity responses by treatment with sIL-4R
The previous results indicate a priming effect of IL-4 and an inhibitory effect of sIL-4R on in vitro histamine release. The biological relevance of these ndings was evaluated using an animal model of allergic sensitization and IgE production. 27 BALB/c mice were locally sensitized to ovalbumin via the airways. It has been shown that this procedure results in an increase of allergen-speci c IgE/IgG1, development of positive skin test responses and enhanced airway responsiveness. 25 The mice were sensitized to OVA for 4 weeks followed by a brief period of treatment with murine sIL-4R also via the airways. The development of allergen speci c cutaneous hypersensitivity responses was then assessed by intradermal injection of allergen as described above. As shown in Fig. 4, mice sensitized to OVA via airways and treated by the same route with sIL-4R developed signi cantly smaller wheal responses than mice that were  not treated with the drug (P , 0 07). Inhibition of in vivo mediator release was selective, since wheal formations stimulated by administration of compound 48/80 were not affected. Furthermore, these effects were not related to inhibition of IgE/IgG1 production by sIL-4R, since total and allergen speci c IgE/IgG1 concentrations were the same in untreated OVA-sensitized animals ( Table 1).

Discussion
This study provides evidence that IL-4 primes effector cells of the allergic response to release histamine upon stimulation with allergen and anti-IgE. With allergen as the trigger factor for histamine release, mediator release was found to increase by up to 35%. These ndings were supported by a similar and signi cant effect on the development of in vivo immediate-type cutaneous hypersensitivity responses. These results point to a novel function of IL-4. IL-4 interferes with the allergic response at several levels of regulation: it is associated with the development of Th2 type T-lymphocytes and acts as a growth factor for Th2 cells; 2 5 it stimulates IgE synthesis 3 5 and is involved in mast cell activation. 7 9 IL-4 is produced by Tlymphocytes and effector cells including basophils and mast cells. 12,13 The results of this study indicate that the functional role of IL-4 in allergic in ammatory responses is not restricted to the development of pro-allergic T-and B-cell functions. It also interferes during the effector phase of the allergic response by priming mediator secretion from basophils and mast cells. We measured histamine release by these effector cells, since it represents an important and clinically relevant mediator of the early phase of the immediate-type allergic response. It is likely that secretion of other mediators which are released together with the prototypic mediator histamine is also enhanced under the in uence of IL-4.
Since there is increasing evidence for the secretion of IL-4 by mast cells and basophils, 28,29 one can envisage a multidirectional activity of IL-4 during the development of immediate-type I allergic reactions. If IL-4 is released upon crosslinking of membrane-bound IgE antibodies, this source of IL-4 may prime neighbouring effector cells to release higher amounts of histamine (and other mediators). Therefore, IL-4 would potentiate the development of the allergic response in an autocrine and paracrine fashion. In our experiments it is possible that preactivated mast-cells and/or basophils were the source of IL-4 production.
Exactly how IL-4 primes effector cells requires further clari cation. It is possible that it triggers the release of preformed mediators which are stored in cytoplasmic vesicles. Since IL-4 acts relatively quickly (the priming effects reached optimal levels within 2.5 h) it is unlikely that IL-4 increases histamine synthesis by regulation of transscriptional and/or translational events.
In skin mast cells, histamine releasability is generally lower than in basophils. 30 The enhancement of anti-IgE induced histamine release observed here was in the range of 4.5% mean histamine release before and 8.3%after pretreatment with IL-4. Although these differences may appear low, this represents an 80% increase of histamine release, which could be biologically relevant considering that the differences observed in anti-IgE induced basophil histamine release between normal controls and asthmatics appear low as well (median histamine release 20%vs. 26%). 31 The effects of IL-4 were inhibited in vitro and in vivo by a recombinant soluble IL-4 receptor. We have shown in a previous study that treatment with sIL-4R can inhibit many IL-4 mediated functions under certain experimental conditions. 32 For example, when BALB/c mice were treated with sIL-4R simultaneously to allergen sensitization, it inhibited the development of IgE/IgG1 production and therefore the induction of immediate-type allergic reactions, including postive immediate-type skin test responses, and also increased airway responsiveness. We concluded from that study that the effects of sIL-4R treatment were related to a functional interference at the level of T± B cell interaction. To study whether sIL-4R would also act directly on the effector phase of the allergic response, we modi ed the sensitization and treatment protocol. Sensitized mice were treated with the receptor via nebulization for a short time prior to analysis of skin test reactivity. In the present study, the drug was delivered via the airways since it has recently been shown that local administration was superior to systemic treat- ment in terms of the inhibitory effects of immediate hypersensitivity responses. 33 It is unlikely that the in vivo effects of sIL-4R in our study design were due to down-regulation of IgE/IgG1 production, because mice were already sensitized at the time the treatment was started and because the treatment was carried out over a short period of time. In addition, it is likely that Ig was already bound to IgE receptors on effector cells and it is suggested that receptor bound IgE has a long half-life time. In fact, measurement of IgE/IgG1 antibody titres indicated that there were no differences between sIL-4R treated and untreated mice. It is therefore likely that the effects observed in treated mice were related to a down-regulation of histamine release. SIL-4R treated mice still developed wheal formations immediately following allergen application into the skin, but the wheals were signi cantly smaller to a degree that re ects the results of the in vitro experiments. However, it should also be considered that IgE-mediated allergic sensitization to systemic allergen can occur in the absence of circulating IgE. In addition, a direct correlation exist between increased IgE production and mast cell activation.
In conclusion, this study provides evidence for a new functional activity of IL-4 as a cytokine which primes effector cells of the allergic response to release histamine upon stimulation with allergen. This activity underscores the important and multidirectional activities of this cytokine in the regulation of many aspects of the immunopathogenesis of allergic in ammatory responses. IL-4 not only controls the development of pro-allergic T-cell functions and triggers production of IgE, but also acts on effector cells of the allergic response. The biological relevance of these functions was demonstrated in sensitized mice which were treated with recombinant murine sIL-4R. The treatment study elucidated two important aspects: rstly, the development of in vivo immediate cutaneous hypersensitivity responses is directly controlled by IL-4, and, secondly, sIL-4R represents a potent immuno-pharmacological inhibitor for this (and other) functional activities of IL-4.